From prvs=80177c78bd=anil.agarwal@viasat.com Fri Nov 1 05:32:05 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B089D21E81DB for ; Fri, 1 Nov 2013 05:32:05 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0q9CS1QRoK79 for ; Fri, 1 Nov 2013 05:32:00 -0700 (PDT) Received: from mta-us-west-01.viasat.com (mta-us-west-01.viasat.com [199.106.191.81]) by ietfa.amsl.com (Postfix) with ESMTP id 3DF6A21E81AC for ; Fri, 1 Nov 2013 05:31:49 -0700 (PDT) Received: from pps.filterd (VCASPAM01.hq.corp.viasat.com [127.0.0.1]) by VCASPAM01.hq.corp.viasat.com (8.14.5/8.14.5) with SMTP id rA1CVeaT017325 for ; Fri, 1 Nov 2013 05:31:49 -0700 From: "Agarwal, Anil" To: "tcpm@ietf.org" Thread-Topic: TCP mismatched sequence numbers issue Thread-Index: Ac7W+9eDU92jjK1PT/yb8pBTMNqsPQ== Date: Fri, 1 Nov 2013 12:31:47 +0000 Message-ID: <7A2801D5E40DD64A85E38DF22117852C7073722E@wdc1exchmbxp05.hq.corp.viasat.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.10.8794, 1.0.431, 0.0.0000 definitions=2013-11-01_03:2013-11-01, 2013-11-01, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 kscore.is_bulkscore=5.55111512312578e-17 kscore.compositescore=0 circleOfTrustscore=0 compositescore=0.994415628185148 urlsuspect_oldscore=0.994415628185148 suspectscore=0 recipient_domain_to_sender_totalscore=0 phishscore=0 bulkscore=0 kscore.is_spamscore=0 recipient_to_sender_totalscore=0 recipient_domain_to_sender_domain_totalscore=0 rbsscore=0.994415628185148 spamscore=0 recipient_to_sender_domain_totalscore=0 urlsuspectscore=0.9 adultscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1305240000 definitions=main-1311010078 Subject: [tcpm] TCP mismatched sequence numbers issue X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 01 Nov 2013 12:32:05 -0000 Hi All, Is the following behavior of TCP, relating to mismatched sequence numbers b= etween TCP peers, correct? Suppose host A opens a TCP connection to host B. After the connection is successfully opened, let's suppose that the TCP sequence numbers are as follows - A: snd_una =3D 1, snd_nxt =3D 1, rcv_nxt =3D 1 B: snd_una =3D 1, snd_nxt =3D 1, rcv_nxt =3D 1 Now suppose, that TCP connection A, accidentally or maliciously, changes it= s sequence numbers to the following values - A: snd_una =3D -4000, snd_nxt =3D -4000, rcv_nxt =3D +4000 (The send sequence number is retracted, the receive sequence number is adva= nced). The TCP states at A and B are now - A: snd_una =3D -4000, snd_nxt =3D -4000, rcv_nxt =3D 4000 B: snd_una =3D 1, snd_nxt =3D 1, rcv_nxt =3D 1 Next, A sends an ACK segment to B (or a data segment with < 4000 bytes of d= ata) A->B segment: seq =3D -4000, ack =3D 4000, datalen < 4000 B, on receiving the segment, rejects the segment data since it lies outside its receive window and sends an ACK segment back to A. B->A segment: seq =3D 1, ack =3D 1 Note that B does not change rcv_nxt or snd_una. Even if B processes the ack field (4000), it will reject it as invalid and not update snd_una. A ,on receiving the segment, rejects the segment data since it lies outside its receive window and sends an ACK segment back to B. A->B segment: seq =3D -4000, ack =3D 4000 Note that A does not change rcv_nxt or snd_una. Even if A processes the ack field (1), it will reject it as invalid and not update snd_una. And a vicious cycle is started. Each side sends an ACK packet to each other, at the rate of 1 segment per R= TT. A malicious host A might send a few extra ACK segments, for good measure, and get multiple segments into the cycle. The connection terminates after 15-20 minutes. I have tried this against Linux 2.6.18 and 2.4.20 servers and have observed the above behavior. Hosts A and B on a gigabit LAN merrily exchange 40000 ACK segments per seco= nd. Even if A changes either the send or the receive sequence numbers, it results in the above cycle. A: snd_una =3D -4000, snd_nxt =3D -4000, rcv_nxt =3D 1 or A: snd_una =3D 1, snd_nxt =3D 1, rcv_nxt =3D 4000 This behaviour can also occur due to a middle-box C that "spoofs" TCP conne= ctions and TCP sequences numbers. C may use a locally generated iss towards A whil= e B uses its own iss value. Subsequently, if A and B communicate directly due to a midd= le-box failure, or a route change, similar conditions can arise. Is the above behavior expected or well-known or did I get something wrong? If the behavior is expected, then is there a need to detect and break the c= ycle? Thanks, Anil Agarwal ViaSat Inc. From prvs=0018549097=anna.brunstrom@kau.se Fri Nov 1 17:13:00 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A27EC11E80F7 for ; Fri, 1 Nov 2013 17:13:00 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.649 X-Spam-Level: X-Spam-Status: No, score=-1.649 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HELO_EQ_SE=0.35, J_CHICKENPOX_33=0.6] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0RRVFzQphJ+v for ; Fri, 1 Nov 2013 17:12:56 -0700 (PDT) Received: from tiger.dc.kau.se (smtp.kau.se [193.10.220.38]) by ietfa.amsl.com (Postfix) with ESMTP id 4189411E8153 for ; Fri, 1 Nov 2013 17:12:53 -0700 (PDT) X-Spam-Processed: mail.kau.se, Sat, 02 Nov 2013 01:12:34 +0100 (not processed: spam filter heuristic analysis disabled) X-Authenticated-Sender: annabrun@kau.se X-MDRemoteIP: 217.212.115.108 X-Return-Path: anna.brunstrom@kau.se X-Envelope-From: anna.brunstrom@kau.se Message-ID: <5274437D.60801@kau.se> Date: Sat, 02 Nov 2013 01:12:45 +0100 From: Anna Brunstrom User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:15.0) Gecko/20120824 Thunderbird/15.0 MIME-Version: 1.0 To: Yuchung Cheng References: <20130917083618.1262.43665.idtracker@ietfa.amsl.com> <5238166A.7000303@kau.se> <526AFF3D.9030706@kau.se> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-rtorestart-01.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Nov 2013 00:13:00 -0000 Hi, On 2013-10-30 16:14, Yuchung Cheng wrote: > On Fri, Oct 25, 2013 at 4:31 PM, Anna Brunstrom wrote: >> Hi Yuchung, >> >> Removing the constraint has been discussed before and the very first draft >> we put together for RTO restart actually worked like that. After some >> off-list discussions with Mark Allman and others, the constraint was put in. >> >> Even if it is conceptually simpler to remove the constraint, it would be a >> disadvantage to reduce the RTO in situations where we have a continuous flow >> of data and do not want the RTO to fire even if there is a packet loss. In >> these cases it is better if a fast retransmit is triggered. > if inflight is large than spurious timeout is even less a concern b/c > we expect plenty of acks to refresh the estimator. But the negative effect of an unwanted/spurious RTO is much larger and there is no gain to be made in this case. > > Step back a bit: what's RTO? isn't it the max time we expect the ack > to come back for first unacked packet (SND.UNA)? > > The starting time should always be the last (re)transmission of that > packet (as in rto-restart). Conceptually and practically it has > nothing to do w/ number of packets I sent after (inflight). Why keep > tweaking things based on cwnd or inflight. In fact many other > transport protocols implement rto-restart natually, because that's how > timer should be re-armed. I agree that it is conceptually cleaner, but it will not work as well. The RTO value is a trade-off between timely loss detection and triggering the timer unnecessarily. This trade-off is different when you have very few segments inflight. In this case you really need the timely loss detection. When you have many segments inflight you do not want to detect loss through a RTO so having a larger value makes sense from a performance perspective. Here the RTO is only a very last resort in case you lose an entire window of packets. In this case you have big problems anyway. > The real concern is that RTO has been re-armed in the wrong way for a > quarter of century, so the formula has been tighten up for faster > timeout (e.g., Linux 200ms min-RTO). But IMO it's wrong to patch up > one thing for the other. Let's do rto-restart always, and fix the RTO > formula too. I doubt that the RTO restart issue is the main reason the RTO has been tightened. Anyway, I think it is outside the scope of the RTO restart draft to modify the RTO formula. BR, Anna > >> >> Thanks, >> Anna >> >> >> On 2013-10-23 15:57, Yuchung Cheng wrote: >>> >>> >>> >>> >>> On Tue, Sep 17, 2013 at 1:44 AM, Per Hurtig >> > wrote: >>> >>> Hi all, >>> >>> A revised draft of the RTO restart mechanism has just been submitted >>> with the details, below. >>> >>> The main changes between this and previous drafts are: >>> >>> * Improved wording throughout the document. >>> >>> * Removed the possibility for a connection limited by the receiver's >>> advertised window to use RTO restart, decreasing the risk of spurious >>> timeouts. >>> >>> * A new section that discusses the applicability of and problems >>> related >>> to the RTO restart mechanism. >>> >>> * Updates to the text that describe RTO restart's relation to TLP. >>> >>> * Acknowledgments added. >>> >>> >>> We're happy to receive any feedback! >>> >>> I suggest removing the constraint in (2) of section 3. RTO re-arm should >>> account for the time elapsed since the SND.UNA was last (re)transmitted, >>> either cwnd is 1 or 10000. There is always risk for spurious timeout but >>> that's another bug on RTT estimation. >>> >>> >>> >>> >>> Per >>> >>> >>> >>> On 09/17/2013 10:36 AM, internet-drafts@ietf.org >>> wrote: >>> > >>> > A New Internet-Draft is available from the on-line >>> Internet-Drafts directories. >>> > This draft is a work item of the TCP Maintenance and Minor >>> Extensions Working Group of the IETF. >>> > >>> > Title : TCP and SCTP RTO Restart >>> > Author(s) : Per Hurtig >>> > Anna Brunstrom >>> > Andreas Petlund >>> > Michael Welzl >>> > Filename : draft-ietf-tcpm-rtorestart-01.txt >>> > Pages : 11 >>> > Date : 2013-09-17 >>> > >>> > Abstract: >>> > This document describes a modified algorithm for managing the >>> TCP and >>> > SCTP retransmission timers that provides faster loss recovery >>> when >>> > there is a small amount of outstanding data for a connection. >>> The >>> > modification allows the transport to restart its >>> retransmission timer >>> > more aggressively in situations where fast retransmit cannot >>> be used. >>> > This enables faster loss detection and recovery for >>> connections that >>> > are short-lived or application-limited. >>> > >>> > >>> > The IETF datatracker status page for this draft is: >>> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rtorestart >>> > >>> > There's also a htmlized version available at: >>> > http://tools.ietf.org/html/draft-ietf-tcpm-rtorestart-01 >>> > >>> > A diff from the previous version is available at: >>> > http://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rtorestart-01 >>> > >>> > >>> > Please note that it may take a couple of minutes from the time of >>> submission >>> > until the htmlized version and diff are available at >>> tools.ietf.org . >>> >>> > >>> > Internet-Drafts are also available by anonymous FTP at: >>> > ftp://ftp.ietf.org/internet-drafts/ >>> > >>> > _______________________________________________ >>> > tcpm mailing list >>> > tcpm@ietf.org >>> >>> > https://www.ietf.org/mailman/listinfo/tcpm >>> > >>> >>> >>> -- >>> Per Hurtig, PhD Tel: +46 (0) 54 700 2335 >>> >>> >>> Datavetenskap Kontor: 21F-422 (Hus Vanern) >>> Karlstads universitet PGP 0x8C4FFCF6 >>> SE-651 88 Karlstad http://www.kau.se/forskare/per-hurtig >>> _______________________________________________ >>> tcpm mailing list >>> tcpm@ietf.org >>> https://www.ietf.org/mailman/listinfo/tcpm >>> >>> >>> >>> >>> >>> _______________________________________________ >>> tcpm mailing list >>> tcpm@ietf.org >>> https://www.ietf.org/mailman/listinfo/tcpm >>> >> From ycheng@google.com Fri Nov 1 17:49:35 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 99B8311E80E9 for ; Fri, 1 Nov 2013 17:49:35 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.468 X-Spam-Level: X-Spam-Status: No, score=-1.468 tagged_above=-999 required=5 tests=[AWL=-0.090, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_33=0.6, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id c6e1agQqfpz3 for ; Fri, 1 Nov 2013 17:49:34 -0700 (PDT) Received: from mail-ie0-x231.google.com (mail-ie0-x231.google.com [IPv6:2607:f8b0:4001:c03::231]) by ietfa.amsl.com (Postfix) with ESMTP id 9348E11E810C for ; Fri, 1 Nov 2013 17:49:34 -0700 (PDT) Received: by mail-ie0-f177.google.com with SMTP id e14so8436810iej.8 for ; Fri, 01 Nov 2013 17:49:34 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=dqGPu2XP8XjLFxqsz20q1AnwVt7EObHLK2eSVzKByX8=; b=h3SYDXfGe52UeGm21NwrFPU2BwMkhN8sNgaB7vcK74+c0V28M99X88sFXvV6o6KN32 rRXr1RHmsaiQpg0et+acxgXUYBv52p7gZs1WLZrtsXnqUcaD9t/VhIFD4erwgXLWcLX8 t0qRT5JYtc8Hfg4vOZACQxMaqS0LFZwqkvPQhi/L/uBSYM3j66GBHcoFqWeVneCmDX0I lvj3r26+vkCmIDzH58ZjgpoKeFbj2xbIPkYMEaw4dLRH48voDEIbe7BKafeTWR6WnpPC nGIgiX7vEbp8R6EBzb0bc2Jg1KSUtn/3SaWs3gxKHot+e3q0W0F6V+aJZ8MaqRLND9DW LnAg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=dqGPu2XP8XjLFxqsz20q1AnwVt7EObHLK2eSVzKByX8=; b=bmz13qbtYVzVJm1MAiw+ZuwcpREduKsGkgbDrvDlEWkGHu471QBv4xw16ENgbQecrQ txhROe6ggTfsDatKtXAE1n0RNjF3/hDdei/Xn6o2bYZDowiQnoCNDNv2FuOlTdtaf9Bm p5emcNGqwGP1lZ771+jCzBOJIoj+6A5Y8i85biMwbhmNcUI03UG6OHJEo2ObCIk9BCBR yWAeBK+Z4sr2oA0nN1Xg0iUjSsiPV0OPzvaKrxpbo9WEUsDLWmTjla9L62GGeYR2YYCq 3FPsMjaAhi38kSwDP9k3wDST1uIt+MM11I3pjG1HvS60TZhbQ6Od0K3zU1KVdDOwtrf6 s9fQ== X-Gm-Message-State: ALoCoQmnpG4jR5LJGUm7t0LG3P+QZDkeUKNF1MmFrwLduHpfsyZaTvZt3eSde+roQ5jyHF3XVmkQtqmjA6EPchqBn70V8YAO5E0kTavZV8JHsxXbherXLlaybZhA4RtL7WgjrpexXLm5yK4GBVGypHVrLWDt1QsUHinZ03ac5Hs2CRmdU+W88dFtD4pcoU7kbMtYwvLRRC6q X-Received: by 10.50.164.130 with SMTP id yq2mr4203168igb.53.1383353374137; Fri, 01 Nov 2013 17:49:34 -0700 (PDT) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Fri, 1 Nov 2013 17:49:14 -0700 (PDT) In-Reply-To: <5274437D.60801@kau.se> References: <20130917083618.1262.43665.idtracker@ietfa.amsl.com> <5238166A.7000303@kau.se> <526AFF3D.9030706@kau.se> <5274437D.60801@kau.se> From: Yuchung Cheng Date: Fri, 1 Nov 2013 17:49:14 -0700 Message-ID: To: Anna Brunstrom Content-Type: text/plain; charset=ISO-8859-1 Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-rtorestart-01.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Nov 2013 00:49:35 -0000 On Fri, Nov 1, 2013 at 5:12 PM, Anna Brunstrom wrote: > Hi, > > > On 2013-10-30 16:14, Yuchung Cheng wrote: >> >> On Fri, Oct 25, 2013 at 4:31 PM, Anna Brunstrom >> wrote: >>> >>> Hi Yuchung, >>> >>> Removing the constraint has been discussed before and the very first >>> draft >>> we put together for RTO restart actually worked like that. After some >>> off-list discussions with Mark Allman and others, the constraint was put >>> in. >>> >>> Even if it is conceptually simpler to remove the constraint, it would be >>> a >>> disadvantage to reduce the RTO in situations where we have a continuous >>> flow >>> of data and do not want the RTO to fire even if there is a packet loss. >>> In >>> these cases it is better if a fast retransmit is triggered. >> >> if inflight is large than spurious timeout is even less a concern b/c >> we expect plenty of acks to refresh the estimator. > > > But the negative effect of an unwanted/spurious RTO is much larger and there > is no gain to be made in this case. > > >> >> Step back a bit: what's RTO? isn't it the max time we expect the ack >> to come back for first unacked packet (SND.UNA)? >> >> The starting time should always be the last (re)transmission of that >> packet (as in rto-restart). Conceptually and practically it has >> nothing to do w/ number of packets I sent after (inflight). Why keep >> tweaking things based on cwnd or inflight. In fact many other >> transport protocols implement rto-restart natually, because that's how >> timer should be re-armed. > > > I agree that it is conceptually cleaner, but it will not work as well. > > The RTO value is a trade-off between timely loss detection and triggering > the timer unnecessarily. This trade-off is different when you have very few > segments inflight. In this case you really need the timely loss detection. This makes no sense. T0: send 6 packets P1, P2, ... P6 T100: got ACK of P1, adjust RTT, P2 should timeout at T100+rto T105: got ACK of P2, adjust RTT, P3 should timeout at T200+rto T110: got ACK of P3, adjust RTT, P4 should timeout at T0+rto because "hey we got only 3 packets inflight so I gotta be less patient"?? so maybe if I only 2 packets inflight I should do T-1 + rto. > > When you have many segments inflight you do not want to detect loss through > a RTO so having a larger value makes sense from a performance perspective. > Here the RTO is only a very last resort in case you lose an entire window of > packets. In this case you have big problems anyway. > > >> The real concern is that RTO has been re-armed in the wrong way for a >> quarter of century, so the formula has been tighten up for faster >> timeout (e.g., Linux 200ms min-RTO). But IMO it's wrong to patch up >> one thing for the other. Let's do rto-restart always, and fix the RTO >> formula too. > > > I doubt that the RTO restart issue is the main reason the RTO has been > tightened. > > Anyway, I think it is outside the scope of the RTO restart draft to modify > the RTO formula. I certainly not suggest modify the RTO formula in this draft. > > BR, > Anna > > >> >>> >>> Thanks, >>> Anna >>> >>> >>> On 2013-10-23 15:57, Yuchung Cheng wrote: >>>> >>>> >>>> >>>> >>>> >>>> On Tue, Sep 17, 2013 at 1:44 AM, Per Hurtig >>> > wrote: >>>> >>>> Hi all, >>>> >>>> A revised draft of the RTO restart mechanism has just been >>>> submitted >>>> with the details, below. >>>> >>>> The main changes between this and previous drafts are: >>>> >>>> * Improved wording throughout the document. >>>> >>>> * Removed the possibility for a connection limited by the >>>> receiver's >>>> advertised window to use RTO restart, decreasing the risk of >>>> spurious >>>> timeouts. >>>> >>>> * A new section that discusses the applicability of and problems >>>> related >>>> to the RTO restart mechanism. >>>> >>>> * Updates to the text that describe RTO restart's relation to TLP. >>>> >>>> * Acknowledgments added. >>>> >>>> >>>> We're happy to receive any feedback! >>>> >>>> I suggest removing the constraint in (2) of section 3. RTO re-arm should >>>> account for the time elapsed since the SND.UNA was last (re)transmitted, >>>> either cwnd is 1 or 10000. There is always risk for spurious timeout but >>>> that's another bug on RTT estimation. >>>> >>>> >>>> >>>> >>>> Per >>>> >>>> >>>> >>>> On 09/17/2013 10:36 AM, internet-drafts@ietf.org >>>> wrote: >>>> > >>>> > A New Internet-Draft is available from the on-line >>>> Internet-Drafts directories. >>>> > This draft is a work item of the TCP Maintenance and Minor >>>> Extensions Working Group of the IETF. >>>> > >>>> > Title : TCP and SCTP RTO Restart >>>> > Author(s) : Per Hurtig >>>> > Anna Brunstrom >>>> > Andreas Petlund >>>> > Michael Welzl >>>> > Filename : draft-ietf-tcpm-rtorestart-01.txt >>>> > Pages : 11 >>>> > Date : 2013-09-17 >>>> > >>>> > Abstract: >>>> > This document describes a modified algorithm for managing the >>>> TCP and >>>> > SCTP retransmission timers that provides faster loss recovery >>>> when >>>> > there is a small amount of outstanding data for a connection. >>>> The >>>> > modification allows the transport to restart its >>>> retransmission timer >>>> > more aggressively in situations where fast retransmit cannot >>>> be used. >>>> > This enables faster loss detection and recovery for >>>> connections that >>>> > are short-lived or application-limited. >>>> > >>>> > >>>> > The IETF datatracker status page for this draft is: >>>> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rtorestart >>>> > >>>> > There's also a htmlized version available at: >>>> > http://tools.ietf.org/html/draft-ietf-tcpm-rtorestart-01 >>>> > >>>> > A diff from the previous version is available at: >>>> > http://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rtorestart-01 >>>> > >>>> > >>>> > Please note that it may take a couple of minutes from the time >>>> of >>>> submission >>>> > until the htmlized version and diff are available at >>>> tools.ietf.org . >>>> >>>> > >>>> > Internet-Drafts are also available by anonymous FTP at: >>>> > ftp://ftp.ietf.org/internet-drafts/ >>>> > >>>> > _______________________________________________ >>>> > tcpm mailing list >>>> > tcpm@ietf.org >>>> >>>> > https://www.ietf.org/mailman/listinfo/tcpm >>>> > >>>> >>>> >>>> -- >>>> Per Hurtig, PhD Tel: +46 (0) 54 700 2335 >>>> >>>> >>>> Datavetenskap Kontor: 21F-422 (Hus Vanern) >>>> Karlstads universitet PGP 0x8C4FFCF6 >>>> SE-651 88 Karlstad http://www.kau.se/forskare/per-hurtig >>>> _______________________________________________ >>>> tcpm mailing list >>>> tcpm@ietf.org >>>> https://www.ietf.org/mailman/listinfo/tcpm >>>> >>>> >>>> >>>> >>>> >>>> _______________________________________________ >>>> tcpm mailing list >>>> tcpm@ietf.org >>>> https://www.ietf.org/mailman/listinfo/tcpm >>>> >>> > From gorry@erg.abdn.ac.uk Sat Nov 2 16:09:59 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B3EA921E811D for ; Sat, 2 Nov 2013 16:09:59 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -104.603 X-Spam-Level: X-Spam-Status: No, score=-104.603 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_46=0.6, MIME_QP_LONG_LINE=1.396, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id G185rbev1wSl for ; Sat, 2 Nov 2013 16:09:53 -0700 (PDT) Received: from SMTP.HSIA.FAIRMONT.COM (smtp.hsia.fairmont.com [142.131.15.59]) by ietfa.amsl.com (Postfix) with ESMTP id 95A0A21E80CF for ; Sat, 2 Nov 2013 16:09:53 -0700 (PDT) Received: from p1308.superclick.com (unverified [64.114.24.114]) by smtp.hsia.fairmont.com (Vircom SMTPRS 5.22.7.16456) with ESMTP id ; Sat, 2 Nov 2013 19:09:44 -0400 X-Modus-BlackList: 64.114.24.114=OK;gorry@erg.abdn.ac.uk=OK X-Modus-Trusted: 64.114.24.114=NO X-Modus-Audit: FALSE;0;0;0 Received: from [172.16.32.115] ([172.16.32.115]) (authenticated bits=0) by p1308.superclick.com (8.13.1/8.13.1) with ESMTP id rA2N9ltM017658; Sat, 2 Nov 2013 16:09:47 -0700 References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> Mime-Version: 1.0 (1.0) In-Reply-To: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Message-Id: <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> X-Mailer: iPad Mail (11B511) From: Gorry Date: Sat, 2 Nov 2013 16:09:41 -0700 To: Yuchung Cheng Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 Nov 2013 23:09:59 -0000 I have reviewed the current version of TFO and have the comments below. This= is rather long ... Nits in Section 1 - I think the document here talks of servers and clients f= rom a TCP endpoint perspective, this is normal for TCP, but the document als= o raises implications for apps designers who may have a very different view o= n what is intended by the word "server" and "client". To avoid doubt maybe y= ou should explain this in the intro? Page 3, para 2 is not clear. I think Otis just the detail of the wording, bu= t it would be good to see clear text. Last para section 2: to me, the requirements are not clear.=20 I think this piece of text means: TCP stacks should NOT enable TCP-TFO by de= fault. The stack should provide a global flag to enable this. An application= wishing to use this must enable each TCP socket on a per service/per port b= asis. I think the other comments/questions maybe more significant. section 2 also suggests the main change is to allow data in SYN, I have note= d that some people think this is the only change --- it is not. The bigger i= mplication is that the document proposes to allow new data sent during the 3= WHS. This change needs to be more clearly noted, sending an IW of data befo= re there is an ACK on the return path is a significant change to TCP semanti= cs and it's behaviour under heavy load. Page 5 fast open, bullet 3. : This says the server can send data before the 3= WHS completes. I am curious what this actually means, is this implied that t= he use is extricated to standards-track mechanisms, I.e. I am looking here f= or some explicit understanding of whether this implies IW3. Or is the experi= ment linked to both a proposal for iW10 and this new experiment? Last para 4.1.3 the text says /should include remote port numbers too/ Can w= e be clear which port numbers are used. Is this the received server port num= ber by the client? Actually, do not understand what is intended by the parag= raph, can you provide an example to be sure what is meant? Section 4.3.2. I hoped this would say something explicit about not sending S= YN segments that are larger than the default MTU. This seems like a bad thin= g to do. Why are the examples only for web? This is a TCP spec. There are two sections 6.3.2. Second Section 6.3.2 - I think this applies to SSL/TLS in general, doesn't i= t? ... I think the section should address the general case, and then provide= an example for https, rather than assuming a web-centric use. Section 7.1 first para last line.please check text, this pathology is not n= ecessarily malicious!!! The issue seems to be mainly related to NAPT where ports are used to discrim= inate senders? Also the shared NAPT can result in a range of RTT and PMTU to the same IP, b= ut this is already possible with ECMP routing. The lack of issues listed in section 7 raises serious concerns to me. I do n= ot list that is provided currently as the only, or even most significant rea= sons why this is deemed to be experimental work. Indeed, the lack of identif= ication of potential CC issues is why this document is in my opinion NOT rea= dy to complete a WGLC - if the issues are not listed how can we evaluate if t= he method is safe enough to experiment or can later be evaluated properly. Here are some additional issues I would like considered by the group and if t= he group is happy these may be safe enough to recommend testing in the gener= al internet, then I think we should document that these are things to be con= firmed. 1) I raised on the list that a negative impact of using the SYN as a probe t= o check for new option support. The examples I gave were ECN interactions an= d IPv6 probing. ECN if used will probe for ECN support in the SYN segment. I= f we include data on the SYN and then there is a lack of response, it is ass= umed by the sender that both TFO and ECN are not supported. This is an impli= cation of the model we use to negotiate for features that may be black-holed= . It is not an issue with the method, but I think we need tone clear this is= something that may happen. A similar case exists if a client attempts TFO t= o a server using IPv6 ... Is it assumed that the path does not support IPv6 o= r TFO (or actually both). It does not happen when multiple options are negot= iated. 2) the draft proposes caching RTT per path and using this for SYN timeout. T= his updates a standards track RFC, and will have implications in some networ= k paths that may seriously degrade performance, in other cases it may improv= e performance. Examples of negative performance are cases where an initial p= acket can take much longer to process than a subsequent packet. This can and= does occur in bandwidth-on-demand L2 networks, where the first packet cause= s path setup. It also occurs to some extent with policing and routing device= s that build cached state from the first packet in a flow. I think this is o= ne of the reasons why the RFC-series has to date not described a shortened R= TT method for the SYN.=20 Another issue is that previous RTT (or PMTU) is not necessarily a good indic= ation of future value in a different 5tuple, since ECMP etc can lead to very= different path characteristics for a session with a different port number. T= he current text therefore proposes an experiment that is beyond what was pre= viously considered safe. It does not specify a way to avoid the server makin= g the same mistake next connection. 3) seeding the SYN RTT from a lower value makes the sender more aggressive i= n heavily congested networks. The sender is also made more aggressive by sen= ding IW data packets before there is any indication the path can contain eve= n a single data segment. This is significant change to standard behaviour. I= f the proposal is to use IW 3 (as standard) it still needs to be called out.= If the proposal is to use a larger experimental number then I have concerns= here that this is a significant change that needs an automated recovery met= hod to prevent significant collateral damage on capacity-limited paths - the= re needs to be a way to stop a server doing this each time resulting in recu= rring loss! 4) Are there any changes to PMTUD behaviour that need experience? PMTU was n= aturally cached and I am not sure if this changes the behaviour or not. It d= oes of course allow a packet with a larger MTU to be sent before the path ha= s been initially validated via the 3WHS (I.e. After a path change while idle= , this can generate TCP segments in a SYN that are larger than the receiver= advertised MSS - I sus= pect this is not crucial, but I would also assume that the SYN itself should= not use an increased PMTU value to carry the data. Is this correct?=20 Are there other case where this change will impact the network for specific p= ath characteristics?=20 Have others thought through these issues and feel they are ok? Or at least o= k for experimental deployment? Gorry From prvs=0019bc24a5=anna.brunstrom@kau.se Sat Nov 2 21:20:07 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 652FE11E818D for ; Sat, 2 Nov 2013 21:20:07 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.149 X-Spam-Level: X-Spam-Status: No, score=-2.149 tagged_above=-999 required=5 tests=[AWL=0.500, BAYES_00=-2.599, HELO_EQ_SE=0.35, J_CHICKENPOX_33=0.6, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kNbnZ0+gglO7 for ; Sat, 2 Nov 2013 21:20:02 -0700 (PDT) Received: from nasse.dc.kau.se (smtp.kau.se [193.10.220.39]) by ietfa.amsl.com (Postfix) with ESMTP id 7690311E8171 for ; Sat, 2 Nov 2013 21:20:02 -0700 (PDT) X-Spam-Processed: mail.kau.se, Sun, 03 Nov 2013 05:19:28 +0100 (not processed: spam filter heuristic analysis disabled) X-Authenticated-Sender: annabrun@kau.se X-MDRemoteIP: 64.114.24.114 X-Return-Path: anna.brunstrom@kau.se X-Envelope-From: anna.brunstrom@kau.se Message-ID: <5275CEDD.2040205@kau.se> Date: Sun, 03 Nov 2013 05:19:41 +0100 From: Anna Brunstrom User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:15.0) Gecko/20120824 Thunderbird/15.0 MIME-Version: 1.0 To: Yuchung Cheng References: <20130917083618.1262.43665.idtracker@ietfa.amsl.com> <5238166A.7000303@kau.se> <526AFF3D.9030706@kau.se> <5274437D.60801@kau.se> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-rtorestart-01.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 03 Nov 2013 04:20:07 -0000 Hi, On 2013-11-02 01:49, Yuchung Cheng wrote: > On Fri, Nov 1, 2013 at 5:12 PM, Anna Brunstrom wrote: >> Hi, >> >> >> On 2013-10-30 16:14, Yuchung Cheng wrote: >>> >>> On Fri, Oct 25, 2013 at 4:31 PM, Anna Brunstrom >>> wrote: >>>> >>>> Hi Yuchung, >>>> >>>> Removing the constraint has been discussed before and the very first >>>> draft >>>> we put together for RTO restart actually worked like that. After some >>>> off-list discussions with Mark Allman and others, the constraint was put >>>> in. >>>> >>>> Even if it is conceptually simpler to remove the constraint, it would be >>>> a >>>> disadvantage to reduce the RTO in situations where we have a continuous >>>> flow >>>> of data and do not want the RTO to fire even if there is a packet loss. >>>> In >>>> these cases it is better if a fast retransmit is triggered. >>> >>> if inflight is large than spurious timeout is even less a concern b/c >>> we expect plenty of acks to refresh the estimator. >> >> >> But the negative effect of an unwanted/spurious RTO is much larger and there >> is no gain to be made in this case. >> >> >>> >>> Step back a bit: what's RTO? isn't it the max time we expect the ack >>> to come back for first unacked packet (SND.UNA)? >>> >>> The starting time should always be the last (re)transmission of that >>> packet (as in rto-restart). Conceptually and practically it has >>> nothing to do w/ number of packets I sent after (inflight). Why keep >>> tweaking things based on cwnd or inflight. In fact many other >>> transport protocols implement rto-restart natually, because that's how >>> timer should be re-armed. >> >> >> I agree that it is conceptually cleaner, but it will not work as well. >> >> The RTO value is a trade-off between timely loss detection and triggering >> the timer unnecessarily. This trade-off is different when you have very few >> segments inflight. In this case you really need the timely loss detection. > This makes no sense. > > T0: send 6 packets P1, P2, ... P6 > T100: got ACK of P1, adjust RTT, P2 should timeout at T100+rto > T105: got ACK of P2, adjust RTT, P3 should timeout at T200+rto > T110: got ACK of P3, adjust RTT, P4 should timeout at T0+rto because > "hey we got only 3 packets inflight so I gotta be less patient"?? The difference is that P4 can not be recovered by fast retransmit. If P4 is lost we will rely on a timeout to recover the packet. In this case there is a potential gain to be made by reducing the timeout. If we get an RTO when we have a sufficient number of ACKs on its way to trigger a fast retransmit it does not gain us anything. Consider this example. A packet is lost when you have 100 packets inflight and additional data queued for transmission. Would you prefer to recover this packet by fast retransmit or RTO? If you prefer a fast retransmit why would you want to increase the likelihood of an RTO by reducing the timeout value in this case? BR, Anna > so maybe if I only 2 packets inflight I should do T-1 + rto. > >> >> When you have many segments inflight you do not want to detect loss through >> a RTO so having a larger value makes sense from a performance perspective. >> Here the RTO is only a very last resort in case you lose an entire window of >> packets. In this case you have big problems anyway. >> >> >>> The real concern is that RTO has been re-armed in the wrong way for a >>> quarter of century, so the formula has been tighten up for faster >>> timeout (e.g., Linux 200ms min-RTO). But IMO it's wrong to patch up >>> one thing for the other. Let's do rto-restart always, and fix the RTO >>> formula too. >> >> >> I doubt that the RTO restart issue is the main reason the RTO has been >> tightened. >> >> Anyway, I think it is outside the scope of the RTO restart draft to modify >> the RTO formula. > I certainly not suggest modify the RTO formula in this draft. > >> >> BR, >> Anna >> >> >>> >>>> >>>> Thanks, >>>> Anna >>>> >>>> >>>> On 2013-10-23 15:57, Yuchung Cheng wrote: >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> On Tue, Sep 17, 2013 at 1:44 AM, Per Hurtig >>>> > wrote: >>>>> >>>>> Hi all, >>>>> >>>>> A revised draft of the RTO restart mechanism has just been >>>>> submitted >>>>> with the details, below. >>>>> >>>>> The main changes between this and previous drafts are: >>>>> >>>>> * Improved wording throughout the document. >>>>> >>>>> * Removed the possibility for a connection limited by the >>>>> receiver's >>>>> advertised window to use RTO restart, decreasing the risk of >>>>> spurious >>>>> timeouts. >>>>> >>>>> * A new section that discusses the applicability of and problems >>>>> related >>>>> to the RTO restart mechanism. >>>>> >>>>> * Updates to the text that describe RTO restart's relation to TLP. >>>>> >>>>> * Acknowledgments added. >>>>> >>>>> >>>>> We're happy to receive any feedback! >>>>> >>>>> I suggest removing the constraint in (2) of section 3. RTO re-arm should >>>>> account for the time elapsed since the SND.UNA was last (re)transmitted, >>>>> either cwnd is 1 or 10000. There is always risk for spurious timeout but >>>>> that's another bug on RTT estimation. >>>>> >>>>> >>>>> >>>>> >>>>> Per >>>>> >>>>> >>>>> >>>>> On 09/17/2013 10:36 AM, internet-drafts@ietf.org >>>>> wrote: >>>>> > >>>>> > A New Internet-Draft is available from the on-line >>>>> Internet-Drafts directories. >>>>> > This draft is a work item of the TCP Maintenance and Minor >>>>> Extensions Working Group of the IETF. >>>>> > >>>>> > Title : TCP and SCTP RTO Restart >>>>> > Author(s) : Per Hurtig >>>>> > Anna Brunstrom >>>>> > Andreas Petlund >>>>> > Michael Welzl >>>>> > Filename : draft-ietf-tcpm-rtorestart-01.txt >>>>> > Pages : 11 >>>>> > Date : 2013-09-17 >>>>> > >>>>> > Abstract: >>>>> > This document describes a modified algorithm for managing the >>>>> TCP and >>>>> > SCTP retransmission timers that provides faster loss recovery >>>>> when >>>>> > there is a small amount of outstanding data for a connection. >>>>> The >>>>> > modification allows the transport to restart its >>>>> retransmission timer >>>>> > more aggressively in situations where fast retransmit cannot >>>>> be used. >>>>> > This enables faster loss detection and recovery for >>>>> connections that >>>>> > are short-lived or application-limited. >>>>> > >>>>> > >>>>> > The IETF datatracker status page for this draft is: >>>>> > https://datatracker.ietf.org/doc/draft-ietf-tcpm-rtorestart >>>>> > >>>>> > There's also a htmlized version available at: >>>>> > http://tools.ietf.org/html/draft-ietf-tcpm-rtorestart-01 >>>>> > >>>>> > A diff from the previous version is available at: >>>>> > http://www.ietf.org/rfcdiff?url2=draft-ietf-tcpm-rtorestart-01 >>>>> > >>>>> > >>>>> > Please note that it may take a couple of minutes from the time >>>>> of >>>>> submission >>>>> > until the htmlized version and diff are available at >>>>> tools.ietf.org . >>>>> >>>>> > >>>>> > Internet-Drafts are also available by anonymous FTP at: >>>>> > ftp://ftp.ietf.org/internet-drafts/ >>>>> > >>>>> > _______________________________________________ >>>>> > tcpm mailing list >>>>> > tcpm@ietf.org >>>>> >>>>> > https://www.ietf.org/mailman/listinfo/tcpm >>>>> > >>>>> >>>>> >>>>> -- >>>>> Per Hurtig, PhD Tel: +46 (0) 54 700 2335 >>>>> >>>>> >>>>> Datavetenskap Kontor: 21F-422 (Hus Vanern) >>>>> Karlstads universitet PGP 0x8C4FFCF6 >>>>> SE-651 88 Karlstad http://www.kau.se/forskare/per-hurtig >>>>> _______________________________________________ >>>>> tcpm mailing list >>>>> tcpm@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/tcpm >>>>> >>>>> >>>>> >>>>> >>>>> >>>>> _______________________________________________ >>>>> tcpm mailing list >>>>> tcpm@ietf.org >>>>> https://www.ietf.org/mailman/listinfo/tcpm >>>>> >>>> >> From anil@csir4pi.in Sat Nov 2 23:32:44 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 89D2121E80A7 for ; Sat, 2 Nov 2013 23:32:44 -0700 (PDT) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -0.532 X-Spam-Level: X-Spam-Status: No, score=-0.532 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_NUMERIC_HELO=2.067] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3XlR6v5ab9uV for ; Sat, 2 Nov 2013 23:32:39 -0700 (PDT) Received: from mailgate.cmmacs.ernet.in (mail.cmmacs.ernet.in [202.41.64.32]) by ietfa.amsl.com (Postfix) with ESMTP id 59D6921E80AB for ; Sat, 2 Nov 2013 23:32:37 -0700 (PDT) Received: from mailgate.cmmacs.ernet.in (localhost.localdomain [127.0.0.1]) by postfix.imss70 (Postfix) with ESMTP id AA41CBC802D; Sun, 3 Nov 2013 12:02:32 +0530 (IST) Received: from email.csir4pi.in (email2.cmmacs.ernet.in [202.41.64.72]) by mailgate.cmmacs.ernet.in (Postfix) with ESMTP id 8CFF0BC802A; Sun, 3 Nov 2013 12:02:32 +0530 (IST) Received: from 202.41.64.20 (webmail.cmmacs.ernet.in [202.41.64.20]) by email.csir4pi.in (Postfix) with ESMTP id 039C328200B6; Sun, 3 Nov 2013 12:02:52 +0530 (IST) Received: from 115.242.146.148 (SquirrelMail authenticated user anil) by 202.41.64.20 with HTTP; Sun, 3 Nov 2013 12:02:32 +0530 (IST) Message-ID: <49558.115.242.146.148.1383460352.squirrel@202.41.64.20> In-Reply-To: <7A2801D5E40DD64A85E38DF22117852C7073722E@wdc1exchmbxp05.hq.corp.viasa t.com> References: <7A2801D5E40DD64A85E38DF22117852C7073722E@wdc1exchmbxp05.hq.corp.viasat.com> Date: Sun, 3 Nov 2013 12:02:32 +0530 (IST) From: anil@csir4pi.in To: "Agarwal, Anil" User-Agent: SquirrelMail/1.4.8-6.el3 MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal X-TM-AS-Product-Ver: IMSS-7.0.0.1640-7.0.0.1014-20264.003 X-TM-AS-Result: No--31.172-5.0-31-1 X-imss-scan-details: No--31.172-5.0-31-1;No--31.172-5.0-31-1 X-Mailman-Approved-At: Sun, 03 Nov 2013 08:04:22 -0800 Cc: "tcpm@ietf.org" Subject: Re: [tcpm] TCP mismatched sequence numbers issue X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: anil@csir4pi.in List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 03 Nov 2013 06:35:14 -0000 Hi: This behaviour is known and documented in "Simple Active Attack Against TCP" available at https://www.usenix.org/legacy/publications/library/proceedings/security95/full_papers/joncheray.pdf Pls. see section 2.2, 2.3 and 2.4 in this document (Desynchronised state and TCP ACK Storm) Why was the TCP connection terminated after 15-20 minutes. I guess this might be due to the timeout of application layer protocol that you used in your experiment. If so, you should be able to observer a consistency in the timeout, but you mentioned a range of 15-20 minutes. Regards, Anil On Fri, November 1, 2013 6:01 pm, Agarwal, Anil wrote: > > Hi All, > > Is the following behavior of TCP, relating to mismatched sequence numbers > between TCP peers, correct? > > Suppose host A opens a TCP connection to host B. > > After the connection is successfully opened, let's suppose > that the TCP sequence numbers are as follows - > > A: snd_una = 1, snd_nxt = 1, rcv_nxt = 1 > B: snd_una = 1, snd_nxt = 1, rcv_nxt = 1 > > Now suppose, that TCP connection A, accidentally or maliciously, changes > its > sequence numbers to the following values - > A: snd_una = -4000, snd_nxt = -4000, rcv_nxt = +4000 > > (The send sequence number is retracted, the receive sequence number is > advanced). > > The TCP states at A and B are now - > A: snd_una = -4000, snd_nxt = -4000, rcv_nxt = 4000 > B: snd_una = 1, snd_nxt = 1, rcv_nxt = 1 > > Next, A sends an ACK segment to B (or a data segment with < 4000 bytes of > data) > > A->B segment: seq = -4000, ack = 4000, datalen < 4000 > > B, on receiving the segment, rejects the segment data since it > lies outside its receive window and sends an ACK segment back to A. > > B->A segment: seq = 1, ack = 1 > > Note that B does not change rcv_nxt or snd_una. > Even if B processes the ack field (4000), it will reject it as invalid > and not update snd_una. > > A ,on receiving the segment, rejects the segment data since it > lies outside its receive window and sends an ACK segment back to B. > > A->B segment: seq = -4000, ack = 4000 > > Note that A does not change rcv_nxt or snd_una. > Even if A processes the ack field (1), it will reject it as invalid > and not update snd_una. > > And a vicious cycle is started. > Each side sends an ACK packet to each other, at the rate of 1 segment per > RTT. > > A malicious host A might send a few extra ACK segments, for good measure, > and get multiple segments into the cycle. > > The connection terminates after 15-20 minutes. > > I have tried this against Linux 2.6.18 and 2.4.20 servers and > have observed the above behavior. > Hosts A and B on a gigabit LAN merrily exchange 40000 ACK segments per > second. > > > Even if A changes either the send or the receive sequence numbers, > it results in the above cycle. > A: snd_una = -4000, snd_nxt = -4000, rcv_nxt = 1 > or > A: snd_una = 1, snd_nxt = 1, rcv_nxt = 4000 > > This behaviour can also occur due to a middle-box C that "spoofs" TCP > connections > and TCP sequences numbers. C may use a locally generated iss towards A > while B uses its > own iss value. Subsequently, if A and B communicate directly due to a > middle-box failure, > or a route change, similar conditions can arise. > > Is the above behavior expected or well-known or did I get something wrong? > If the behavior is expected, then is there a need to detect and break the > cycle? > > Thanks, > Anil Agarwal > ViaSat Inc. > > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm > -- V Anil Kumar Principal Scientist CSIR Fourth Paradigm Institute (CSIR-4PI) NAL Belur Campus Bangalore - 560037 E-mail: anil@cmmacs.ernet.in Phone: +91 80 2505 1355 From mallman@icir.org Mon Nov 4 09:32:01 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C2CD811E8370 for ; Mon, 4 Nov 2013 09:32:00 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.299 X-Spam-Level: X-Spam-Status: No, score=-106.299 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, J_CHICKENPOX_35=0.6, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3yTDspAH3EOE for ; Mon, 4 Nov 2013 09:31:53 -0800 (PST) Received: from fruitcake.ICSI.Berkeley.EDU (fruitcake.ICSI.Berkeley.EDU [192.150.186.11]) by ietfa.amsl.com (Postfix) with ESMTP id F38CE11E835D for ; Mon, 4 Nov 2013 09:29:18 -0800 (PST) Received: from lawyers.icir.org (envoy.icir.org [192.150.187.30]) by fruitcake.ICSI.Berkeley.EDU (8.12.11.20060614/8.12.11) with ESMTP id rA4HTDVC020611 for ; Mon, 4 Nov 2013 09:29:13 -0800 (PST) Received: from lawyers.icir.org (localhost [127.0.0.1]) by lawyers.icir.org (Postfix) with ESMTP id 32F932639802 for ; Mon, 4 Nov 2013 12:29:10 -0500 (EST) To: tcpm@ietf.org From: Mark Allman Organization: International Computer Science Institute (ICSI) Song-of-the-Day: Werewolves of London X-URL-0: http://www.icir.org/mallman-files/Document5949.jpg X-URL-1: http://www.icir.org/mallman-files/Document14947.docx X-URL-2: http://www.icir.org/mallman-files/Document961.xlsx MIME-Version: 1.0 Content-Type: multipart/signed; boundary="--------ma55653-1"; micalg=pgp-sha1; protocol="application/pgp-signature" Date: Mon, 04 Nov 2013 12:29:10 -0500 Sender: mallman@icir.org Message-Id: <20131104172910.32F932639802@lawyers.icir.org> Subject: [tcpm] review of 1323bis-15 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: mallman@icir.org List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 04 Nov 2013 17:32:01 -0000 ----------ma55653-1 Content-Type: text/plain Content-Disposition: inline Hi folks! Sorry to be so late with this review of 1323bis-15. I think it is in general in a lot better shape than the last time I read it. The following are a set of minor-to-modest comments. No show stoppers, IMHO. - I'd remove the mention of RTTM in the abstract. It is downplayed in the text. - Sec 1: "has resulted" --> "have resulted" - Sec 1: In the last paragraph before 1.1: - should "recommended" be "RECOMMENDED"? (I am not sure) - perhaps a mention here that the changes from 1323 are partially due to our changed understanding of how things work (i.e., this is not just a set of bug fixes) - Sec 1.1: I found this a bit strange. The two "fundamental problems" seem right to me. But, this document will go on to develop PAWS---which addresses neither problem directly. You might develop the notion that making the windows bigger causes *another* fundamental problem and that is too quickly wrapping. And, therefore, that problem must be tackled and this document does so. - Sec 1.2, point (1): This is not entirely correct. You need both a high transfer rate *and a whole bunch of data* to have a sequence wrapping issue. This point could be a lot better by noting that it takes 4GB of data to wrap. - Sec 1.2, the paragraph starting "Duplicates from earlier ....": You might also throw in there that randomizing ephemeral ports can also help probabilistically reduce the chances of duplicates from previous connections muddying the waters. - Sec 1.3: "be crashed by" --> "crash on" - Sec 1.3 (and other places): I'd like to see the TS option be discussed as the 10 byte option that it is. It is fine to note that often it consumes 12 bytes because passing is needed. However, there is a disconnect in the document between the actual specification (10 bytes) and the discussion (12 bytes) that could lead to confusion. - Sec 1.3: I'd remove the word "optimal". I don't think anyone ever claimed that any RTO scheme---TS based or non-TS based---ever produced an "optimal" RTO. The statement should be precise and say that additional RTT samples per RTT is shown to have little effect on the ultimate RTO when compared to one sample per RTT. - Sec 2.2: "to do both" --> "to both" - Sec 2.2: "is never scaled." --> "MUST NOT be scaled." - Sec 2.3: The list at the beginning of the section should be converted to standards language (MUST, SHOULD, etc.). - Sec 2.3: I find this notion in the first bullet that the window should be stored as 32-bit quantities to be weird. The maximum size for advertised windows is 30 bits. So, why specify the use of 32 bit memory locations in the control block? I mean, that is the natural thing to do, but it just adds a little confusion to things without really helping the implementer at all. - Sec 2.3: "This is safe as a sender can always choose to only partially use any signaled receive window." That is correct, but it takes a leap from the previous discussion to figure out what you mean. I'd flesh this out a bit more so the reader doesn't have to scratch their head. I.e., if the receiver is scaling by X>14 and the receiver is only scaling by 14 then the window will appear smaller than it is in reality. - Sec 3.1: The third paragraph doesn't make much sense, or at least isn't precise. You are trying to make a clear distinction with imprecise words. :-) I'd say you draw a distinction between the timestamp option that *conveys* timestamp information and the *use* of that information (e.g., by PAWS or by RTTM or by Eifel or whatnot). In particular, not all the mechanisms that use *timestamps* use "the RTT signal as input". E.g., PAWS does not. So, you are trying to push a distinction that doesn't make sense. - Sec 3.1: "option is useful" --> "option may be useful" (i.e., it depends on the balancing of costs and benefits and so I don't think an absolute statement is warranted) - Sec 3.1: "Eifel ([cites]) and others" --> I'd add cites for "and others". Or, drop it. This is a way to try to make an argument that there are scads of these things and it isn't worth listing them all, when in reality there are not. - Sec 3.2: The first paragraph feels out of place. Before you even define timestamps you are telling us how they are used. I'd move this to later in the section. Sec 3.2: This whole discussion of valid and invalid TSecr is confusing. You note that when the value is invalid it should be set to zero. But, a TSecr of zero does not mean the TS is invalid. So, how should we tell? - Sec 3.2: "A TCP MAY send the Timestamps option (TSopt) in an initial segment (i.e., segment containing a SYN bit and no ACK bit), and MAY send a TSopt in other segments only if it received a TSopt in the initial or segment for the connection." That second MAY strikes me as totally wrong. Shouldn't it be a MUST? Didn't the WG decide that once timestamps are turned on (in the 3WHS) that they must always be used? The next paragraph seems to back up this notion of using a MUST here. In fact, I am not entirely sure why there are two paragraphs here. They seem to at least largely overlap. - Likewise there is a notion that segments without timestamps "SHOULD" be dropped. But, if we're going to enforce things maybe that should also be a "MUST"? - Sec 3.3: It'd be nice to see references to this middlebox behavior you sketch. Even if just to a mailing list discussion it'd be more concrete than the nebulous language in there now. - I would suggest splitting 3.3 & 3.4 off into their own section. You say you are splitting the TS mechanism from its use, but then you don't in the document. I think a new section would make this distinction more clear. I.e., the new section could start with "One use for the timestamp option is ...". And, the PAWS section could similarly start with "Another use for the timestamp option is ...". - Sec 3.3: Since you have a section on updating the RTO you should move all the discussion of updating the RTO there. I.e., you should not have it scattered. - Sec 3.4: Swap the first two paragraphs. - Sec 3.4: You say one way to deal with the problem of the gains is addressed in the appendix. Fine. But, it might be useful to stake out a principle here. E.g., something like "in the face of multiple RTT samples per RTT an implementation SHOULD try to adhere to the spirit of the history specified in [RFC6298]". Something like that. I.e., throw out at least a high level goal. - Sec 4.2: "PAWS uses the same TCP Timestamps option as the RTTM mechanism" --> PAWS uses the TCP Timestamps option" - Sec 4.2: "In both the PAWS and the RTTM mechanism," --> "In the PAWS mechanism," - Sec 4.2: The first few paragraphs are just a mess. Get rid of the references to RTTM. PAWS does not and never did depend on RTTM. Change this to the correct and more accurate "timestamp option" sort of language. - Sec 4.2: What is 'snd.ts.ok'??? This is seemingly important, but not defined. (It is defined in an appendix, but that is not good enough for a normative discussion.) - Sec 4.2: When an segment is received, it MUST NOT be subjected to PAWS checks, and information from the Timestamps option MUST NOT be used to update connection state information. SEG.TSecr MAY be used to provide stricter acceptance checks. Don't these two sentences contradict each other? They are at least not clear. - Sec 4.4: "to measure RTTs" .... PAWS does not measure RTTs. Get rid of this. - Sec 4.4: "internet" --> "Internet" Also, one thing to think about ... Might it be useful to recommend a random offset to the TS per connection? People have shown it is possible to get signal from these things and perhaps that is not useful. I know OpenBSD uses a random offset. Perhaps just as a MAY or something. I.e., there is no harm in it and so let implementers chew on it. I hope that is useful! allman ----------ma55653-1 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (Darwin) iEYEARECAAYFAlJ32WYACgkQWyrrWs4yIs71OgCeM/ig7j+U8PcIVAfiwuWXzpfa F9EAn0Id90sx4NDV0oB2COZYRa/p404U =erbZ -----END PGP SIGNATURE----- ----------ma55653-1-- From bob.briscoe@bt.com Mon Nov 4 09:53:14 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9BE4711E81EB for ; Mon, 4 Nov 2013 09:53:09 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.917 X-Spam-Level: X-Spam-Status: No, score=-2.917 tagged_above=-999 required=5 tests=[AWL=-0.318, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gSAbPJk5Ljxc for ; Mon, 4 Nov 2013 09:52:51 -0800 (PST) Received: from hubrelay-rd.bt.com (hubrelay-rd.bt.com [62.239.224.98]) by ietfa.amsl.com (Postfix) with ESMTP id AC7E121E8096 for ; Mon, 4 Nov 2013 09:52:33 -0800 (PST) Received: from EVMHR01-UKBR.domain1.systemhost.net (193.113.108.40) by EVMHR65-UKRD.bt.com (10.187.101.20) with Microsoft SMTP Server (TLS) id 14.3.158.1; Mon, 4 Nov 2013 17:52:29 +0000 Received: from EPHR01-UKIP.domain1.systemhost.net (147.149.196.177) by EVMHR01-UKBR.domain1.systemhost.net (193.113.108.40) with Microsoft SMTP Server (TLS) id 8.3.297.1; Mon, 4 Nov 2013 17:52:29 +0000 Received: from bagheera.jungle.bt.co.uk (132.146.168.158) by EPHR01-UKIP.domain1.systemhost.net (147.149.196.177) with Microsoft SMTP Server id 14.2.347.0; Mon, 4 Nov 2013 17:52:29 +0000 Received: from BTP075694.jungle.bt.co.uk ([10.111.121.212]) by bagheera.jungle.bt.co.uk (8.13.5/8.12.8) with ESMTP id rA4HqQCl025848; Mon, 4 Nov 2013 17:52:26 GMT Message-ID: <201311041752.rA4HqQCl025848@bagheera.jungle.bt.co.uk> X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9 Date: Mon, 4 Nov 2013 17:52:25 +0000 To: From: Bob Briscoe MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed X-Scanned-By: MIMEDefang 2.56 on 132.146.168.158 Cc: tcpm IETF list Subject: [tcpm] TFO nits X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 04 Nov 2013 17:53:23 -0000 TFO authors, 5. Security Considerations Delete 'not' in the following: OLD We note that although generating bogus cookies is cost-free, the cost of validating the cookies, inherent to any authentication scheme, may not be substantial compared to processing ^^^ a regular SYN packet. 6.3.2 Speculative Connections by the Applications OLD While HTTPS adoption is still low, it will increase over time (especially given the context of recent revelations). NEW While HTTPS adoption is still low, it is likely to increase over time. RATIONALE: RFCs are meant to be timeless, and this won't be understood in a few year's time. Anyway, it's speculative. Bob ________________________________________________________________ Bob Briscoe, BT From michael.scharf@alcatel-lucent.com Wed Nov 6 03:48:20 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B9D5121E8088 for ; Wed, 6 Nov 2013 03:48:20 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -10.328 X-Spam-Level: X-Spam-Status: No, score=-10.328 tagged_above=-999 required=5 tests=[AWL=0.271, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XtRGssdZNOKh for ; Wed, 6 Nov 2013 03:48:15 -0800 (PST) Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id 4827821E80C9 for ; Wed, 6 Nov 2013 03:48:15 -0800 (PST) Received: from fr712usmtp2.zeu.alcatel-lucent.com (h135-239-2-42.lucent.com [135.239.2.42]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id rA6Bm9EA025712 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL); Wed, 6 Nov 2013 05:48:11 -0600 (CST) Received: from FR711WXCHHUB02.zeu.alcatel-lucent.com (fr711wxchhub02.zeu.alcatel-lucent.com [135.239.2.112]) by fr712usmtp2.zeu.alcatel-lucent.com (GMO) with ESMTP id rA6Bm91u017159 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 6 Nov 2013 12:48:09 +0100 Received: from FR712WXCHMBA15.zeu.alcatel-lucent.com ([169.254.7.203]) by FR711WXCHHUB02.zeu.alcatel-lucent.com ([135.239.2.112]) with mapi id 14.02.0247.003; Wed, 6 Nov 2013 12:48:09 +0100 From: "Scharf, Michael (Michael)" To: "tcpm@ietf.org" Thread-Topic: WGLC for draft-ietf-tcpm-fastopen-05 - Followup question on implementations Thread-Index: AQHO2uYPaxr3Dy5u6kmwqFzzI+avWQ== Date: Wed, 6 Nov 2013 11:48:08 +0000 Message-ID: <655C07320163294895BBADA28372AF5D0F52CC@FR712WXCHMBA15.zeu.alcatel-lucent.com> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> In-Reply-To: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> Accept-Language: de-DE, en-US Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [135.239.27.38] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39 Cc: "tcpm-chairs@tools.ietf.org" Subject: [tcpm] WGLC for draft-ietf-tcpm-fastopen-05 - Followup question on implementations X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Nov 2013 11:48:20 -0000 Hi all, Our AD asked in the meeting on Monday whether there are further planned imp= lementations of draft-ietf-tcpm-fastopen, beyond the Linux one. Therefore, the TCPM chairs would really appreciate feedback (possibly priva= tely) on potential implementations of draft-ietf-tcpm-fastopen. General int= erest in implementing the mechanism at some point in time would be useful a= s well. Thanks! Michael > -----Original Message----- > From: tcpm-bounces@ietf.org [mailto:tcpm-bounces@ietf.org] On Behalf Of > Scharf, Michael (Michael) > Sent: Friday, October 25, 2013 2:21 PM > To: tcpm@ietf.org > Cc: draft-ietf-tcpm-fastopen@tools.ietf.org > Subject: [tcpm] WGLC for draft-ietf-tcpm-fastopen-05 >=20 > Hi all, >=20 > The TCPM chairs believe that draft-ietf-tcpm-fastopen-05 addresses all > comments received so far. This email therefore starts a working group > last call running until ***November 10, 2013***. >=20 > IETF datatracker status page for this draft is: > https://datatracker.ietf.org/doc/draft-ietf-tcpm-fastopen >=20 > There's also a htmlized version available at: > http://tools.ietf.org/html/draft-ietf-tcpm-fastopen-05 >=20 > The intended status of the document is Experimental. >=20 > The document asks for IANA allocation of a new TCP option number. The > chairs' understanding is that it is TCPM consensus to ask for IESG > approval of such a codepoint allocation. >=20 > Please send any comments on the TCPM mailing list until November 10. > Notifications of approving the current version (even without other > comments) are helpful as well. >=20 > Thanks! >=20 > Michael > on behalf of the TCPM chairs >=20 > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From ycheng@google.com Wed Nov 6 13:49:21 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id EBC0421E811C for ; Wed, 6 Nov 2013 13:49:20 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 0 X-Spam-Level: X-Spam-Status: No, score=x tagged_above=-999 required=5 tests=[] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pRMYXimsXWce for ; Wed, 6 Nov 2013 13:49:20 -0800 (PST) Received: from mail-ie0-x230.google.com (mail-ie0-x230.google.com [IPv6:2607:f8b0:4001:c03::230]) by ietfa.amsl.com (Postfix) with ESMTP id AD99A21E808A for ; Wed, 6 Nov 2013 13:49:19 -0800 (PST) Received: by mail-ie0-f176.google.com with SMTP id u16so203342iet.21 for ; Wed, 06 Nov 2013 13:49:19 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:from:date:message-id:subject:to:content-type; bh=xTFMurQ0LJpcaNZH3pZdzjNFoZhdh4fTmABvtLC+CUM=; b=RFUhQUXiMbftviQtXzX7rbKVb8vcRuVxMcBQys2WCvUvhl0+Uxv/Szh08X+qMdgA98 2Va/fGP/3O2TObnuSaLBikGYxuVwjfDJmxguVILJqYx2ghSgBAhtid/Nu4XVB4gxV2FX 9AjoPn2LtWCXHDMLM2IgIhB7pcLIetiHQFui1nAQNWcVQOrjYcWS106FJNOm4Zr7mkLi RjeM+WfJw8f2h/mhSqNOlZr+RbE76DHOkFJ2TxaqnBP/pux1E1JdOreef5V46gzOYIRB T8+34og7kG54JP/+Zk8zFr5ItGQLaDGouDqFlCInEih2DJSISmh92MD3vKAf4wVcAkov JCZw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:from:date:message-id:subject:to :content-type; bh=xTFMurQ0LJpcaNZH3pZdzjNFoZhdh4fTmABvtLC+CUM=; b=BE0HyvpDXfPy/U1l83v6jI6YEr7NRyZcETUuibdBG8WuthfQdF7CYBtxd9wBVxLSqG fD+QhaoUZ7RbSYQ5zscNGNhDb7f9alp4BW7A2vT9m4zfdCwL3LS3+IjL941LdLZHt1iW UW3Ge+6OPUhjRedwzi0KWuZOI7Jmc7FJhMjjR1NdhoQenhQe7fbd11Rm5z4h9rTTEhzu 1j+aiujHcLeNp53Ra7UvyVMSjDlGJ81NjSYGbVZUhj600X6/AutbwoqbRlViPwqy4IKo F4i/LMsNpneZaSRy1UvhBL8K7NAINzsFalp+eaiOux+v9BOHmdljy/AyBh+tZn2oOLx2 EB2Q== X-Gm-Message-State: ALoCoQlpK9Ggg0UDZ4DRfbdraWJ53P20H3Xb+Jb2QBNZ9we6s6dk+IsuwQELXk4mrm24AyKcdB2umRcsDezb0fgfzPSfnZ/UceziGBQz9BpnQOEc2ywBSkRAzX/vDSnCK/YFc8bDMzcXmelLqQ7u7zfT8Unm+FP5HPHlwdn/KxG4IW+lEptb2GMipR2hkYKhr4AMlPtcEFf/ X-Received: by 10.50.127.178 with SMTP id nh18mr14705199igb.42.1383774558214; Wed, 06 Nov 2013 13:49:18 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Wed, 6 Nov 2013 13:48:57 -0800 (PST) From: Yuchung Cheng Date: Wed, 6 Nov 2013 13:48:57 -0800 Message-ID: To: "tcpm@ietf.org Extensions" Content-Type: multipart/mixed; boundary=089e013a23c0feb3ed04ea891e18 X-Mailman-Approved-At: Wed, 06 Nov 2013 19:11:54 -0800 Subject: [tcpm] TCP tail probe and the initial loss behavior X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 06 Nov 2013 21:49:21 -0000 --089e013a23c0feb3ed04ea891e18 Content-Type: text/plain; charset=ISO-8859-1 We have new data regarding the question raised in the previous two meetings: "is tail loss probe useful b/c IW10 is causing more (tail) losses". We got some data on TCP initial loss behavior with IW10. There were 15,269 outbound Google server connection traces taken from all over the world. Bottleneck bandwidths ranged from 64kbps to 40Mbps with a median of ~1Mbps. All the connections used IW10. For each of them, we recorded the ending sequence number, packet number and send time of the first lost packet (the first retransmitted packet of the connection) and the repair time (time from initial send until retransmit acked). 980 connections (6.4%) had one or more losses in the IW10 send. "lossdist1.pdf" is a histogram of % losses by position in initial window ("none" means no loss in the first 10 packets sent). Claims that IW10 "increases the loss rate" are not supported by this data: The average loss rate for the first 10 packets is 0.64%. The average loss rate for the last 5 packets of the window is only 0.44% compared to 0.84% for the first 5 packets (in fact, the loss rate for the first packet is as large as packets 6 to 10 combined). This is easier to see on a plot of the cumulative loss (lossdist2.pdf): the black line is a linear fit to the first five points and the blue to the last five points. The lines slopes show that the loss rate consistently increases faster for the first 5 packets of the window compared to the last 5. "lossdist3.pdf" shows the distribution of packet vs. cumulative % of connections with a first loss at or before that packet. The blue line is a fit of cum% to packet^0.25. The quality of the fit over the entire range (and the fact that there's no discontinuity between the portion of the data that lies within the initial window vs the portion outside it) suggests that the single connection loss probabilities have nothing to do with IW (at least for IW <= 10). We didn't graph loss recovery time since it was simple: 78% of the initial window losses (761 of 980) were repaired in 1 RTT and the remaining 22% were repaired in 2 RTT. None of them took a timeout; all were repaired by Fast Retransmit. (This is in contrast to previous IW4 measurements where 20% of the losses required a timeout to repair). --089e013a23c0feb3ed04ea891e18 Content-Type: application/pdf; name="lossdist1.pdf" Content-Disposition: attachment; filename="lossdist1.pdf" Content-Transfer-Encoding: base64 X-Attachment-Id: f_hnnfj4470 JVBERi0xLjQKJYHigeOBz4HTXHIKMSAwIG9iago8PAovQ3JlYXRpb25EYXRlIChEOjIwMTMxMTA0 MTcyMDI4KQovTW9kRGF0ZSAoRDoyMDEzMTEwNDE3MjAyOCkKL1RpdGxlIChSIEdyYXBoaWNzIE91 dHB1dCkKL1Byb2R1Y2VyIChSIDMuMC4yKQovQ3JlYXRvciAoUikKPj4KZW5kb2JqCjIgMCBvYmoK PDwgL1R5cGUgL0NhdGFsb2cgL1BhZ2VzIDMgMCBSID4+CmVuZG9iago3IDAgb2JqCjw8IC9UeXBl IC9QYWdlIC9QYXJlbnQgMyAwIFIgL0NvbnRlbnRzIDggMCBSIC9SZXNvdXJjZXMgNCAwIFIgPj4K ZW5kb2JqCjggMCBvYmoKPDwKL0xlbmd0aCAxMDEwIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlCj4+CnN0 cmVhbQp4nK1XTW8bNxC976+Yow00LD+G5PLqoClqoAVSC8ghyKFVbMBuLKWWiv79zpDc5eyGSLRt D+Jq9OZxPpbkowzcgoEn+HN4Kz7fn3798Qb2p8EorTXI8bQ/VPj1XQe+e/0L/xrg7+H9B9DwcTBw S5+nwWhygJ+H/HiVR69RPF7u4WaRBX0AjbIWXFR2BMRROXo4r3Rk93dwaKlqNaYR5MipfpX/0CrR KmECOXIlWnnXKtGrSsrcRieKBs8DjkkFPdmfhrvqYI1X3jeHajcHZ+lhmkO1mwPaoMbYHKrNDpe8 gTKJj/zjPEcxWwwTrIpWlFFsUUbI3WtlFFuUEek1oiij2OwQvLJY38JztTBa5UobjPaMTHg1pQN6 5UxzKKZ0GOmXFqGawsFaz9lODtWUDpQVthDVFA6OskI/O1RTOlBW2EJUUzpQVr6FqKZwQJtXxuRQ TelAWfkWopqyD939qvM3OfIiQZrF1XXggtIOXMoZ5Y04EhoWaFLaFszYqJxegFZpX8FAoFuBvoJW E7ic1qgUK0gZ4XpaVw8GSwmhWzFDTdZRQrieNoQKUkJer8CxTouUkF9Pm7CClJAPq4RsBW92nb7m 0/EN7SbYPYDJx5qeh2LTnNEBeuRXt3uGq+To2zXsnoYfdjzrN/i0uaIB2nw6ZTqlZPByurHERwhG GVv4mpbpBn4g/sj8MPODvpxv+Xg0zKduTvHDBj4SP+fv4sT3G+q31L+R8qesZ77bUL+j/iXD/Djn 79IGPvWP1hfxxzk+usv5SP1Luf/a/Jv+odfM5P7NfD/Hl/LLczVxxSyrbVyudL0KpUureYl7p7BU 2l8lfarjw8b4pEzpsvVbyTagimWL+M2RXdR8ujM5+nVzxKVh3ZUvxXd9aaBjhM67WYyFNGeJyGjT YinNDW9SLJW54U2JpTAzvrpe/T8lVXV3vKOfF1ovxb3CC62X2j7hUuqltE+4VHqp7BVfCL0U9gmX Oi91veILmZeyPuFS5aWqT7gUeSnqFV9ovNT0CZcSL1/Zf96OpDh05Jf1zQv7cDzcX74vjLasoo1u NnDRssg2rt3AHS1rcOP2z8r+XrY231Rnbl8n+lzadEHmvOEActQrUpfG7Z/PfS71Ksqc4wYu9SrK nPu61uUi9SrKnPua1ueSmmm5tIzuCUp/JX/9CmVX8bJtTeQ9RsHpGkfx3l99/m3/xzXdLeDq/gzH B3h4fDmd4dPxdLr+ALtbmcmG6CLkK5GJSQqpWnR8EHC5n/eH87rgLUX2L1oUMfF/GK+SL1W+mauC j4+n88s1peXh6vH3Uvpf58fjgas3/jsbEvz0zmjYHw+H+z0jX7bi7fAPQhE6WGVuZHN0cmVhbQpl bmRvYmoKMyAwIG9iago8PCAvVHlwZSAvUGFnZXMgL0tpZHMgWyA3IDAgUiBdIC9Db3VudCAxIC9N ZWRpYUJveCBbMCAwIDUwNCA1MDRdID4+CmVuZG9iago0IDAgb2JqCjw8Ci9Qcm9jU2V0IFsvUERG IC9UZXh0XQovRm9udCA8PC9GMiAxMCAwIFIgPj4KL0V4dEdTdGF0ZSA8PCA+PgovQ29sb3JTcGFj ZSA8PCAvc1JHQiA1IDAgUiA+Pgo+PgplbmRvYmoKNSAwIG9iagpbL0lDQ0Jhc2VkIDYgMCBSXQpl bmRvYmoKNiAwIG9iago8PCAvQWx0ZXJuYXRlIC9EZXZpY2VSR0IgL04gMyAvTGVuZ3RoIDI1OTYg L0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCnicnZZ3VFPZFofPvTe9UJIQipTQa2hSAkgN vUiRLioxCRBKwJAAIjZEVHBEUZGmCDIo4ICjQ5GxIoqFAVGx6wQZRNRxcBQblklkrRnfvHnvzZvf H/d+a5+9z91n733WugCQ/IMFwkxYCYAMoVgU4efFiI2LZ2AHAQzwAANsAOBws7NCFvhGApkCfNiM bJkT+Be9ug4g+fsq0z+MwQD/n5S5WSIxAFCYjOfy+NlcGRfJOD1XnCW3T8mYtjRNzjBKziJZgjJW k3PyLFt89pllDznzMoQ8GctzzuJl8OTcJ+ONORK+jJFgGRfnCPi5Mr4mY4N0SYZAxm/ksRl8TjYA KJLcLuZzU2RsLWOSKDKCLeN5AOBIyV/w0i9YzM8Tyw/FzsxaLhIkp4gZJlxTho2TE4vhz89N54vF zDAON40j4jHYmRlZHOFyAGbP/FkUeW0ZsiI72Dg5ODBtLW2+KNR/Xfybkvd2ll6Ef+4ZRB/4w/ZX fpkNALCmZbXZ+odtaRUAXesBULv9h81gLwCKsr51Dn1xHrp8XlLE4ixnK6vc3FxLAZ9rKS/o7/qf Dn9DX3zPUr7d7+VhePOTOJJ0MUNeN25meqZExMjO4nD5DOafh/gfB/51HhYR/CS+iC+URUTLpkwg TJa1W8gTiAWZQoZA+J+a+A/D/qTZuZaJ2vgR0JZYAqUhGkB+HgAoKhEgCXtkK9DvfQvGRwP5zYvR mZid+8+C/n1XuEz+yBYkf45jR0QyuBJRzuya/FoCNCAARUAD6kAb6AMTwAS2wBG4AA/gAwJBKIgE cWAx4IIUkAFEIBcUgLWgGJSCrWAnqAZ1oBE0gzZwGHSBY+A0OAcugctgBNwBUjAOnoAp8ArMQBCE hcgQFVKHdCBDyByyhViQG+QDBUMRUByUCCVDQkgCFUDroFKoHKqG6qFm6FvoKHQaugANQ7egUWgS +hV6ByMwCabBWrARbAWzYE84CI6EF8HJ8DI4Hy6Ct8CVcAN8EO6ET8OX4BFYCj+BpxGAEBE6oosw ERbCRkKReCQJESGrkBKkAmlA2pAepB+5ikiRp8hbFAZFRTFQTJQLyh8VheKilqFWoTajqlEHUJ2o PtRV1ChqCvURTUZros3RzugAdCw6GZ2LLkZXoJvQHeiz6BH0OPoVBoOhY4wxjhh/TBwmFbMCsxmz G9OOOYUZxoxhprFYrDrWHOuKDcVysGJsMbYKexB7EnsFO459gyPidHC2OF9cPE6IK8RV4FpwJ3BX cBO4GbwS3hDvjA/F8/DL8WX4RnwPfgg/jp8hKBOMCa6ESEIqYS2hktBGOEu4S3hBJBL1iE7EcKKA uIZYSTxEPE8cJb4lUUhmJDYpgSQhbSHtJ50i3SK9IJPJRmQPcjxZTN5CbiafId8nv1GgKlgqBCjw FFYr1Ch0KlxReKaIVzRU9FRcrJivWKF4RHFI8akSXslIia3EUVqlVKN0VOmG0rQyVdlGOVQ5Q3mz covyBeVHFCzFiOJD4VGKKPsoZyhjVISqT2VTudR11EbqWeo4DUMzpgXQUmmltG9og7QpFYqKnUq0 Sp5KjcpxFSkdoRvRA+jp9DL6Yfp1+jtVLVVPVb7qJtU21Suqr9XmqHmo8dVK1NrVRtTeqTPUfdTT 1Lepd6nf00BpmGmEa+Rq7NE4q/F0Dm2OyxzunJI5h+fc1oQ1zTQjNFdo7tMc0JzW0tby08rSqtI6 o/VUm67toZ2qvUP7hPakDlXHTUegs0PnpM5jhgrDk5HOqGT0MaZ0NXX9dSW69bqDujN6xnpReoV6 7Xr39An6LP0k/R36vfpTBjoGIQYFBq0Gtw3xhizDFMNdhv2Gr42MjWKMNhh1GT0yVjMOMM43bjW+ a0I2cTdZZtJgcs0UY8oyTTPdbXrZDDazN0sxqzEbMofNHcwF5rvNhy3QFk4WQosGixtMEtOTmcNs ZY5a0i2DLQstuyyfWRlYxVtts+q3+mhtb51u3Wh9x4ZiE2hTaNNj86utmS3Xtsb22lzyXN+5q+d2 z31uZ27Ht9tjd9Oeah9iv8G+1/6Dg6ODyKHNYdLRwDHRsdbxBovGCmNtZp13Qjt5Oa12Oub01tnB Wex82PkXF6ZLmkuLy6N5xvP48xrnjbnquXJc612lbgy3RLe9blJ3XXeOe4P7Aw99D55Hk8eEp6ln qudBz2de1l4irw6v12xn9kr2KW/E28+7xHvQh+IT5VPtc99XzzfZt9V3ys/eb4XfKX+0f5D/Nv8b AVoB3IDmgKlAx8CVgX1BpKAFQdVBD4LNgkXBPSFwSGDI9pC78w3nC+d3hYLQgNDtoffCjMOWhX0f jgkPC68JfxhhE1EQ0b+AumDJgpYFryK9Issi70SZREmieqMVoxOim6Nfx3jHlMdIY61iV8ZeitOI E8R1x2Pjo+Ob4qcX+izcuXA8wT6hOOH6IuNFeYsuLNZYnL74+BLFJZwlRxLRiTGJLYnvOaGcBs70 0oCltUunuGzuLu4TngdvB2+S78ov508kuSaVJz1Kdk3enjyZ4p5SkfJUwBZUC56n+qfWpb5OC03b n/YpPSa9PQOXkZhxVEgRpgn7MrUz8zKHs8yzirOky5yX7Vw2JQoSNWVD2Yuyu8U02c/UgMREsl4y muOWU5PzJjc690iecp4wb2C52fJNyyfyffO/XoFawV3RW6BbsLZgdKXnyvpV0Kqlq3pX668uWj2+ xm/NgbWEtWlrfyi0LiwvfLkuZl1PkVbRmqKx9X7rW4sVikXFNza4bKjbiNoo2Di4ae6mqk0fS3gl F0utSytK32/mbr74lc1XlV992pK0ZbDMoWzPVsxW4dbr29y3HShXLs8vH9sesr1zB2NHyY6XO5fs vFBhV1G3i7BLsktaGVzZXWVQtbXqfXVK9UiNV017rWbtptrXu3m7r+zx2NNWp1VXWvdur2DvzXq/ +s4Go4aKfZh9OfseNkY39n/N+rq5SaOptOnDfuF+6YGIA33Njs3NLZotZa1wq6R18mDCwcvfeH/T 3cZsq2+nt5ceAockhx5/m/jt9cNBh3uPsI60fWf4XW0HtaOkE+pc3jnVldIl7Y7rHj4aeLS3x6Wn 43vL7/cf0z1Wc1zleNkJwomiE59O5p+cPpV16unp5NNjvUt675yJPXOtL7xv8GzQ2fPnfM+d6ffs P3ne9fyxC84Xjl5kXey65HCpc8B+oOMH+x86Bh0GO4cch7ovO13uGZ43fOKK+5XTV72vnrsWcO3S yPyR4etR12/eSLghvcm7+ehW+q3nt3Nuz9xZcxd9t+Se0r2K+5r3G340/bFd6iA9Puo9OvBgwYM7 Y9yxJz9l//R+vOgh+WHFhM5E8yPbR8cmfScvP174ePxJ1pOZp8U/K/9c+8zk2Xe/ePwyMBU7Nf5c 9PzTr5tfqL/Y/9LuZe902PT9VxmvZl6XvFF/c+At623/u5h3EzO577HvKz+Yfuj5GPTx7qeMT59+ A/eE8/tlbmRzdHJlYW0KZW5kb2JqCjkgMCBvYmoKPDwKL1R5cGUgL0VuY29kaW5nIC9CYXNlRW5j b2RpbmcgL1dpbkFuc2lFbmNvZGluZwovRGlmZmVyZW5jZXMgWyA0NS9taW51cyA5Ni9xdW90ZWxl ZnQKMTQ0L2RvdGxlc3NpIC9ncmF2ZSAvYWN1dGUgL2NpcmN1bWZsZXggL3RpbGRlIC9tYWNyb24g L2JyZXZlIC9kb3RhY2NlbnQKL2RpZXJlc2lzIC8ubm90ZGVmIC9yaW5nIC9jZWRpbGxhIC8ubm90 ZGVmIC9odW5nYXJ1bWxhdXQgL29nb25layAvY2Fyb24gL3NwYWNlXQo+PgplbmRvYmoKMTAgMCBv YmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1R5cGUxIC9OYW1lIC9GMiAvQmFzZUZvbnQgL0hl bHZldGljYQovRW5jb2RpbmcgOSAwIFIgPj4KZW5kb2JqCnhyZWYKMCAxMQowMDAwMDAwMDAwIDY1 NTM1IGYgCjAwMDAwMDAwMjEgMDAwMDAgbiAKMDAwMDAwMDE2MyAwMDAwMCBuIAowMDAwMDAxMzc0 IDAwMDAwIG4gCjAwMDAwMDE0NTcgMDAwMDAgbiAKMDAwMDAwMTU2OSAwMDAwMCBuIAowMDAwMDAx NjAyIDAwMDAwIG4gCjAwMDAwMDAyMTIgMDAwMDAgbiAKMDAwMDAwMDI5MiAwMDAwMCBuIAowMDAw MDA0Mjk3IDAwMDAwIG4gCjAwMDAwMDQ1NTQgMDAwMDAgbiAKdHJhaWxlcgo8PCAvU2l6ZSAxMSAv SW5mbyAxIDAgUiAvUm9vdCAyIDAgUiA+PgpzdGFydHhyZWYKNDY1MQolJUVPRgo= --089e013a23c0feb3ed04ea891e18 Content-Type: application/pdf; name="lossdist2.pdf" Content-Disposition: attachment; filename="lossdist2.pdf" Content-Transfer-Encoding: base64 X-Attachment-Id: f_hnnfj7qw1 JVBERi0xLjMKJcTl8uXrp/Og0MTGCjQgMCBvYmoKPDwgL0xlbmd0aCA1IDAgUiAvRmlsdGVyIC9G bGF0ZURlY29kZSA+PgpzdHJlYW0KeAHtWkuPZDcV3t9f4Q1Sj0Q8fj+2GQEiEguUlrJALKDIkITu QKZ5/H2+Y/scu6pu36qesESjVt36yufp4/PwnZ/U79VP6v2HF6tOL8q2fy8n9f43AP76ooyKJqhn 5ZMrpT0/jecvfMwlqCcs6Y/bk/pOfWzszKAj2k/fqm/Uj8pqk1ytTv0HIr7C3w87QkmLD18ru5Ee X3+AYhacIIA+iNnpuX1/bl8MhBNIf6RG//xOfQkdyCgfdbAxOiisayjeqxCDtiVEFfCbDb5ut7Uz utRiPDTfeyJXda2bzk1r+M6R7/a0nzpZXU0uFr4NpeqkPANPWwdCytpEm2GbUC1Yt9Po6G2oycCv Bn41sLQaW9N82iYGl0I5f1O5YL2OKYlu4zt2+Ovm2etdEQ29s9qZ4oVWgH3iDVsqxM4VXaKZggXY J6Z4EGKLbQ7Vb+zQ8R3uu6l0QFj46LIozcCBxbnoWgsOhezJsxJsbuaucDLauqBLTnalF+wGPdlt s9E5er8t8hlbgmlXPtE7m3TyBmeB4+x5E+wO+S45HRHBC70SjOkP/OdN0b6s4hVDTP3K1vWQCRrS 3Crd43Q37AY9WR+MwZbXsHqPsXu8F0LSxuKILN4T7EL+Weo7yGwSySFWnUMoEsoMvOKQs2PgsZ2+ ZCexzMDRXnCmdDlq42oUYgH2T9HZ6bXF6ZRtEWIB9onP1K5GI5FZKzYzcGBzxB4kg2M7N0EJxptw YLY1yLk+n8WgYEz/ivx2hEPWxYYz+fjasRv0ZLytXqdq6xqEjN0ThM5XpGlUpmn/Jtgd8l3BvocU FnolGNMf+M9jjXd+PQRKMKY/8J/PGbXCnKUQwW7Qt0PsvEYGWt0XOnSP90JCBo8RXcFMgIJdSKdU OzoX6keMQt2nk3xY7ylEYkHDkRxCDHHhsqNeasEK+T+qiN98CNjIULWvaKskjBnZTgrprZRk11Ue 8RtcBR1zmsiQB7qJ4cDU4reFE/VBiIBF3kBAN7XiVdMa5jQRkdc7k73ux5qAE+vhDYPiAUtwBga0 BYPDAPOAeG2tRVThp+QLAD6mA4Dvkb5zoo5uQLXi1NmIajrYbAvSZZ3UAiFwPWrXwgf5p6RVVgdI VtcHGvIasULYCMKyDpyAPa7B5M2jhpTkK3lhYMpjIw3ZjE90F7UBtdQKhBNL+wkIVAOHFEwCK1re VkV4seSFkQBDGpEJRg81CyOQBZ19rtsiriNNnCg1sEVv4dStA6cuD4F74IsasW8ZZsLJ1bZ22E4M xaiixtuKyE/Jbt4g7EJAjIzMCbqBwKpKvsemCmarRX6MxF04dQSchjyik1UOZRf7sXCyOld3Jm8g RCdaySqxRjgJIvJe94YLFvUg2s3hyGaLMECbNjDl2qEtABD6AaHvIngjWShJ+YKcgKG7qhGcZFFG rvFu8vECDGFEJRh86REYgw+o4KSCXLsI6wjiYlFpYIvWzGja0cUdxoUrWUfKls4W5ABL48SC4bCb GIEg/J2BlY4SdIZPuKIJAu0K2sKMMBLMFY+6hanRWeHUEXTAQx7RySp4BdosnNA7RxPBQOQxQnSs FWOL5sLp0r6jU+K9x97VtFFD5X2loZExZaFmtIgH7x3mn4zDDINRVeysxYKcgKHxDWh8BfOualsz cp9wmsiQB7qJWW1Rs5gRyIx21uN0cjvACEJjUaqvAtKNAR0zEoClba+fEeQm7F2iDIni71BZ4YuJ YZCrCdkPkxg1dpTpkErjJm0FI6RbRu5wWC0YSNBQBGIujATp4hC2yyqrS0C1XjjRJQF5h1sbi+Pa EKJjpToEpcQWYSTIsO4oLoKL6J9Cq2Kp2ES9s0DgGCNaU4yZLlISRcktBtvMLZIAJ0B0+UCZbyza Ak5UxdUJ1d7BhwGDkoHqTFQdQlnFIBjQRjKbWnSMkWbb0aEJQFRDIcFY54XPuV2QtR8Q1GGFjPmq oJdAhkolo5tcMdiMjgl9AqZl6yoaHY+CgwqBrm+0f4xgZ6kTxJ2MXVehirnqVRROiRGWR3SM4T4G YvpQ3jihsMMRBi0ayxMEdKyVYNMa4XRlH3tjmSvp6seq37XW1NDN2WvXPK0nRdj3sQktBe4wehs2 eiyPXQguW7TtswWICNiYCJp112kk1orpYKkIo5AAk7yIymvBD7QzPaDZjBkhBEyOCW4DjEEqf+KY URiC0dqiKQbEvrOkSEEV6xOlNcPm3dtCsnUqgsQeCkoEooO5USOOiPeD217HyvStMPwM+lbIOj3G FY68++Uv9Nf6H4xorP8h/Sv2/99/b4sf3DlvOzfffLduUJ35Ivn6CQPlZfyR/798VHQ53C7B6cE5 6vNRckLCVFKQ7R6f1fvHR7So6vGjenDv1OMP6lePO9e0h+w8MlFxHkwaOyT+xs5/JjuHUocWJAx2 frALn8kOI7S2Bq1N1w4Vr2kXP5MdkliOSDaDHdqExi5NdvT6YNnK3dcE4wUG3lcc7+zOewzaCjRL FqMwZuFx20eNDN/DCXZwczboN77fW+kFu02v5IrveRP5gt1BL7d8i/6MHWQmsV8u+hZ6wbr8V07W UgDlfdId27FdHjS6S5oXIKi3mA7pHcTE+IZo3x20nfOSYtJP7Ji+3QWO8X6jDmDIlwIsN1T78te7 xNYPdfpNqvUd9FLFF/kTG/ofbKcUmpWeu4Ab8ttx4OFhoZeO4QY92c+dxOo/xpAs5VXaa++CuCue 4tEDt+vFu6i5pE/yWaZZ+vDeRW7hF6T/0zKBntbj9SLKBIYsgx5npDoMeS3V2ZnqLg/DXplAT4g5 CS8VL9hhdGrs3lh1+ryF6eKCHeauxu6NVYcucDAJXhmLcaaxe2vVwXtEXKVdGYtms/N7a9nB0J4z JtYLazFmdH5L3blnMzxekqK0Yrg631uMQJ1fftvm+hzoJTJmpAt+XLXL2/jRWBgMvdm74Mdlu97k Z2gWQMziynZtfqjnSTZdxY3lCm6NsN5uHbNW33e6Lkom6LrgzdZ04dOh6cKNCRyOxsFto+EaR+kP 6uHjO3QATj18/0590R4+vfxzQGp8PvFPf+cHLMGLeVDxkn/wL3/ih9PfvuVH8PujevyqN3bXlm30 PywkgdywjP47hWt/CGhcN7R2ElcqmKxcYevGyYZ1H1iHf/HDszyJWaIzG/79u6374t9iAxv6C+bD gPiEHck/nMYv28OPTCMPwvY0RQ73Cz9Z/CLuQ1hcu6/9B5V73WdxuTZCg54sBtxEF48o2hj42X8j lcF/v/6s6HhRIzz+wpZLdElwfRLoz7xIdubCKdvDtVPYy/LLpftRJPoeIuH1h182WzaaMjqA1NUf cKb7A3P97TD7m/EpzHBE15XbA++y+vm7vC2H5L/tXWJqCmVuZHN0cmVhbQplbmRvYmoKNSAwIG9i agoyNTIwCmVuZG9iagoyIDAgb2JqCjw8IC9UeXBlIC9QYWdlIC9QYXJlbnQgMyAwIFIgL1Jlc291 cmNlcyA2IDAgUiAvQ29udGVudHMgNCAwIFIgL01lZGlhQm94IFswIDAgNTA0IDUwNF0KPj4KZW5k b2JqCjYgMCBvYmoKPDwgL1Byb2NTZXQgWyAvUERGIC9UZXh0IF0gL0NvbG9yU3BhY2UgPDwgL0Nz MSA3IDAgUiA+PiAvRXh0R1N0YXRlIDw8IC9HczIKMjYgMCBSIC9HczMgMjcgMCBSIC9HczEgMjgg MCBSID4+IC9Gb250IDw8IC9UVDE4IDI1IDAgUiAvVFQ0IDExIDAgUiAvVFQyIDkgMCBSCi9UVDkg MTYgMCBSIC9UVDE0IDIxIDAgUiAvVFQxIDggMCBSIC9UVDEwIDE3IDAgUiAvVFQ3IDE0IDAgUiAv VFQxNSAyMiAwIFIKL1RUMTMgMjAgMCBSIC9UVDMgMTAgMCBSIC9UVDYgMTMgMCBSIC9UVDUgMTIg MCBSIC9UVDEyIDE5IDAgUiAvVFQxNiAyMyAwIFIKL1RUMTEgMTggMCBSIC9UVDE3IDI0IDAgUiAv VFQ4IDE1IDAgUiA+PiA+PgplbmRvYmoKMjYgMCBvYmoKPDwgL1R5cGUgL0V4dEdTdGF0ZSAvQ0Eg MCA+PgplbmRvYmoKMjcgMCBvYmoKPDwgL1R5cGUgL0V4dEdTdGF0ZSAvQ0EgMSA+PgplbmRvYmoK MjggMCBvYmoKPDwgL1R5cGUgL0V4dEdTdGF0ZSAvY2EgMCA+PgplbmRvYmoKMjkgMCBvYmoKPDwg L0xlbmd0aCAzMCAwIFIgL04gMyAvQWx0ZXJuYXRlIC9EZXZpY2VSR0IgL0ZpbHRlciAvRmxhdGVE ZWNvZGUgPj4Kc3RyZWFtCngBnZZ3VFPZFofPvTe90BIiICX0GnoJINI7SBUEUYlJgFAChoQmdkQF RhQRKVZkVMABR4ciY0UUC4OCYtcJ8hBQxsFRREXl3YxrCe+tNfPemv3HWd/Z57fX2Wfvfde6AFD8 ggTCdFgBgDShWBTu68FcEhPLxPcCGBABDlgBwOFmZgRH+EQC1Py9PZmZqEjGs/buLoBku9ssv1Am c9b/f5EiN0MkBgAKRdU2PH4mF+UClFOzxRky/wTK9JUpMoYxMhahCaKsIuPEr2z2p+Yru8mYlybk oRpZzhm8NJ6Mu1DemiXho4wEoVyYJeBno3wHZb1USZoA5fco09P4nEwAMBSZX8znJqFsiTJFFBnu ifICAAiUxDm8cg6L+TlongB4pmfkigSJSWKmEdeYaeXoyGb68bNT+WIxK5TDTeGIeEzP9LQMjjAX gK9vlkUBJVltmWiR7a0c7e1Z1uZo+b/Z3x5+U/09yHr7VfEm7M+eQYyeWd9s7KwvvRYA9iRamx2z vpVVALRtBkDl4axP7yAA8gUAtN6c8x6GbF6SxOIMJwuL7OxscwGfay4r6Df7n4Jvyr+GOfeZy+77 VjumFz+BI0kVM2VF5aanpktEzMwMDpfPZP33EP/jwDlpzcnDLJyfwBfxhehVUeiUCYSJaLuFPIFY kC5kCoR/1eF/GDYnBxl+nWsUaHVfAH2FOVC4SQfIbz0AQyMDJG4/egJ961sQMQrIvrxorZGvc48y ev7n+h8LXIpu4UxBIlPm9gyPZHIloiwZo9+EbMECEpAHdKAKNIEuMAIsYA0cgDNwA94gAISASBAD lgMuSAJpQASyQT7YAApBMdgBdoNqcADUgXrQBE6CNnAGXARXwA1wCwyAR0AKhsFLMAHegWkIgvAQ FaJBqpAWpA+ZQtYQG1oIeUNBUDgUA8VDiZAQkkD50CaoGCqDqqFDUD30I3Qaughdg/qgB9AgNAb9 AX2EEZgC02EN2AC2gNmwOxwIR8LL4ER4FZwHF8Db4Uq4Fj4Ot8IX4RvwACyFX8KTCEDICAPRRlgI G/FEQpBYJAERIWuRIqQCqUWakA6kG7mNSJFx5AMGh6FhmBgWxhnjh1mM4WJWYdZiSjDVmGOYVkwX 5jZmEDOB+YKlYtWxplgnrD92CTYRm40txFZgj2BbsJexA9hh7DscDsfAGeIccH64GFwybjWuBLcP 14y7gOvDDeEm8Xi8Kt4U74IPwXPwYnwhvgp/HH8e348fxr8nkAlaBGuCDyGWICRsJFQQGgjnCP2E EcI0UYGoT3QihhB5xFxiKbGO2EG8SRwmTpMUSYYkF1IkKZm0gVRJaiJdJj0mvSGTyTpkR3IYWUBe T64knyBfJQ+SP1CUKCYUT0ocRULZTjlKuUB5QHlDpVINqG7UWKqYup1aT71EfUp9L0eTM5fzl+PJ rZOrkWuV65d7JU+U15d3l18unydfIX9K/qb8uAJRwUDBU4GjsFahRuG0wj2FSUWaopViiGKaYoli g+I1xVElvJKBkrcST6lA6bDSJaUhGkLTpXnSuLRNtDraZdowHUc3pPvTk+nF9B/ovfQJZSVlW+Uo 5RzlGuWzylIGwjBg+DNSGaWMk4y7jI/zNOa5z+PP2zavaV7/vCmV+SpuKnyVIpVmlQGVj6pMVW/V FNWdqm2qT9QwaiZqYWrZavvVLquNz6fPd57PnV80/+T8h+qwuol6uPpq9cPqPeqTGpoavhoZGlUa lzTGNRmabprJmuWa5zTHtGhaC7UEWuVa57VeMJWZ7sxUZiWzizmhra7tpy3RPqTdqz2tY6izWGej TrPOE12SLls3Qbdct1N3Qk9LL1gvX69R76E+UZ+tn6S/R79bf8rA0CDaYItBm8GooYqhv2GeYaPh YyOqkavRKqNaozvGOGO2cYrxPuNbJrCJnUmSSY3JTVPY1N5UYLrPtM8Ma+ZoJjSrNbvHorDcWVms RtagOcM8yHyjeZv5Kws9i1iLnRbdFl8s7SxTLessH1kpWQVYbbTqsPrD2sSaa11jfceGauNjs86m 3ea1rakt33a/7X07ml2w3Ra7TrvP9g72Ivsm+zEHPYd4h70O99h0dii7hH3VEevo4bjO8YzjByd7 J7HTSaffnVnOKc4NzqMLDBfwF9QtGHLRceG4HHKRLmQujF94cKHUVduV41rr+sxN143ndsRtxN3Y Pdn9uPsrD0sPkUeLx5Snk+cazwteiJevV5FXr7eS92Lvau+nPjo+iT6NPhO+dr6rfS/4Yf0C/Xb6 3fPX8Of61/tPBDgErAnoCqQERgRWBz4LMgkSBXUEw8EBwbuCHy/SXyRc1BYCQvxDdoU8CTUMXRX6 cxguLDSsJux5uFV4fnh3BC1iRURDxLtIj8jSyEeLjRZLFndGyUfFRdVHTUV7RZdFS5dYLFmz5EaM Wowgpj0WHxsVeyR2cqn30t1Lh+Ps4grj7i4zXJaz7NpyteWpy8+ukF/BWXEqHhsfHd8Q/4kTwqnl TK70X7l35QTXk7uH+5LnxivnjfFd+GX8kQSXhLKE0USXxF2JY0muSRVJ4wJPQbXgdbJf8oHkqZSQ lKMpM6nRqc1phLT4tNNCJWGKsCtdMz0nvS/DNKMwQ7rKadXuVROiQNGRTChzWWa7mI7+TPVIjCSb JYNZC7Nqst5nR2WfylHMEeb05JrkbssdyfPJ+341ZjV3dWe+dv6G/ME17msOrYXWrlzbuU53XcG6 4fW+649tIG1I2fDLRsuNZRvfbore1FGgUbC+YGiz7+bGQrlCUeG9Lc5bDmzFbBVs7d1ms61q25ci XtH1YsviiuJPJdyS699ZfVf53cz2hO29pfal+3fgdgh33N3puvNYmWJZXtnQruBdreXM8qLyt7tX 7L5WYVtxYA9pj2SPtDKosr1Kr2pH1afqpOqBGo+a5r3qe7ftndrH29e/321/0wGNA8UHPh4UHLx/ yPdQa61BbcVh3OGsw8/rouq6v2d/X39E7Ujxkc9HhUelx8KPddU71Nc3qDeUNsKNksax43HHb/3g 9UN7E6vpUDOjufgEOCE58eLH+B/vngw82XmKfarpJ/2f9rbQWopaodbc1om2pDZpe0x73+mA050d zh0tP5v/fPSM9pmas8pnS8+RzhWcmzmfd37yQsaF8YuJF4c6V3Q+urTk0p2usK7ey4GXr17xuXKp 2737/FWXq2euOV07fZ19ve2G/Y3WHruell/sfmnpte9tvelws/2W462OvgV95/pd+y/e9rp95Y7/ nRsDiwb67i6+e/9e3D3pfd790QepD14/zHo4/Wj9Y+zjoicKTyqeqj+t/dX412apvfTsoNdgz7OI Z4+GuEMv/5X5r0/DBc+pzytGtEbqR61Hz4z5jN16sfTF8MuMl9Pjhb8p/rb3ldGrn353+71nYsnE 8GvR65k/St6ovjn61vZt52To5NN3ae+mp4req74/9oH9oftj9MeR6exP+E+Vn40/d3wJ/PJ4Jm1m 5t/3hPP7CmVuZHN0cmVhbQplbmRvYmoKMzAgMCBvYmoKMjYxMgplbmRvYmoKNyAwIG9iagpbIC9J Q0NCYXNlZCAyOSAwIFIgXQplbmRvYmoKMyAwIG9iago8PCAvVHlwZSAvUGFnZXMgL01lZGlhQm94 IFswIDAgNTA0IDUwNF0gL0NvdW50IDEgL0tpZHMgWyAyIDAgUiBdID4+CmVuZG9iagozMSAwIG9i ago8PCAvVHlwZSAvQ2F0YWxvZyAvUGFnZXMgMyAwIFIgL1ZlcnNpb24gLzEuNCA+PgplbmRvYmoK MTkgMCBvYmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1RydWVUeXBlIC9CYXNlRm9udCAvUUhM V0lOK0FyaWFsTVQgL0ZvbnREZXNjcmlwdG9yCjMyIDAgUiAvRW5jb2RpbmcgL01hY1JvbWFuRW5j b2RpbmcgL0ZpcnN0Q2hhciA1NSAvTGFzdENoYXIgNTUgL1dpZHRocyBbIDU1NgpdID4+CmVuZG9i agozMiAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9RSExXSU4rQXJp YWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5n bGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAv TGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRo IDIwMDAgL0ZvbnRGaWxlMiAzMyAwIFIgPj4KZW5kb2JqCjMzIDAgb2JqCjw8IC9MZW5ndGggMzQg MCBSIC9MZW5ndGgxIDY4MjQgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkJeBRV tj733uolG+kEyNpJV9OkkXRiIIIBEpPO0gEnsgSCdjPBdIBIQDDBgOAGjYpAA+LwlFFccBkV9SmV DjId0CGKy4giPHVwXEF0nssbBH2jjiip+W91g+Dzm1e3/3POPefc7dSpW7eSxVcvaaNkCpEg7+yF rZ1kXNlzwHJmX7NYjdVTsojMTVd0zl0Yqw9cQGT6au6Ca6+I1bPXgB9qb2uV7eT1E3BhOxRGjdgo 8KHtCxcvi9Wzm8GtCzpmx+3ZIdTNC1uXxcenD1BXr2pd2AaOK0f2r3Z2dC02qpQTBD+v8+q2uD/z EyX9KWY7izLInL6hCrqPLJBsVEKXEintSh6ZUJd2E9+46Zv137WkVnxrzbUajR/6ZFihFPYqnbef 3H5qro2syagmGP7SgHaWyv5JVGujk9tPXmczNNJw5uK91CTO63FnOQ4+K4bTEYCL4RFPnqNXDBN5 kXKHNypcPemDS1Ori4WKHksMqoJ2ANuBPYBCLSIfVhvoCiAEbAf2AAcBMxGotKpAB7AVOAKYRZ6w R1SHrXqYyEbbbKw3VWTScUAHBDlAS4DJQAuwEdgKmA0/qekAVgB7gBOAmbwiM7LpAsw9M7LOYD3z F5Qa1dZYtXmmUe25LBDjExtjvO7imNu4mNvIUTH1+TUxPqwoxtMLSkPovCcxpbSvOkNkYJEZmHgn KOMvUipj5KAHxGDSAC4wVUPjFek9Q92lW/cIhZjggtEccuh9gkVS0kqrE7nOj1M6OfhX/FjMwo/1 DEgr3Vr9G36UtgN7AMGPonzMP6YV/IiMOWgVsBXYAxwAjgNmfgTlMMpH/CNK5R9SCVAFtABbgT3A ccDCPwS18Q9kxhhUylUA5x+A2vj7WNb7oKn8PUjv8ff0Pv5WpGxsaa8heErigqMgLmTmxoX0jNIo fzPyw3BklBt3Ghm1WwyhSrpADIkUjHRERVakYp4jyj/pUT2OB6pH8LdJAzhm8jZGfptUYAoQBDoB M6RDkA5RCLgdeADQAGQZqA1Q+T7gdeAQjQC8wBTAyg9GMEyUH4i4axzVGfwN/gplIuL7+Z8N/jp/ 2eCv8ZcM/ip4Puz7+MuRfAdVJ8FOaGMDt4GXwG7iz/cMTXfo1Wl8DyLoAC0BqoDJQAuwETDzPXxI ZI4jHZ3spn14hh08Ql8Y/FF6yEre+Q6vuxYJqEriHncRJJCt6lY397o3342qJO7bNkGSxH3LekiS uK9bCUkS94JrIEninjMfkiTuGS2QJHFPboIEEuX3/3HoMEfZ5CuZWp3KlyJKSxGlpYjSUlL4Ulno B0XO8Z5IYSEitsXrGV7oCO1ioWdZaCoLPcRCbSy0nIVWslAFC13OQh4WsrNQPgt5WWg3G4NQhJh3 xznVsd4sFtrHQk+xUBcLuVmogIWGspDKyrxR7oxcjKcOzGewnmr50HFnz0WV2H1SuRMRdSLnndgT 9oAeAHSj5oWTOiTmnJ0v+ZCewqpY/fxxpR3VE/heNNyL27CXDgMKbtBepNFedLIX3aWCVgEtQB9w HNABM7yHYB0bDZoKWgJUAS3ACuA4YDamcxxT4dQBKqe43ZhYCWgVMFnW+F6UIShO7vTm2ew2j22C 2Ghnqflscr6ez8soIwMbc3qaNS3KUnZ+n/LP71MooTqB38Y3Uh5uxO1xvjHyQ54jyu6KuHc7qgez 31O+gqxjY8nNCsDHUJdRH012q9SPIjt/Erw0Yr8UzVIj7iLHLjZAttrp+MH+qeMLe5RD/Ny+2/GO GlVYxPEXaJ7c6XjbvtbxaknUCs2z7igD26Uarr32MY6n9hmuK2HYEnEsl2yn40b7eMeVdsPQFjNc 3oWaN9Ux1T3DMQH91dlnObxd6HOno8p+uaMi5jVattnpGIEpeGJiISY73G4M6so3OpxeFmXt3iLL ZovfMtlyoaXUUmRxWhyWPEuuZZA13WqzDrAmWxOtVqvZqli5layDovoRr0e+9QaZjZefGQnNSDFk G3YYJrcZUOLMyuk3pA0UDbxhWg1r0PpmU8MsVftumivKEhtnaCZXDdPSG6ihqUYb42mIWvSpWpmn QbNM+a2/m7HbAtBqfE2UUZM/ynSpWpWrpdf6e4mxtFUbciU/b9WGQICyMq6pyqpKr0wbW1/3KyRo KIN1np+vrJ9FT5YnT9vcMM2vPZEX0EqloOcFGrT/mKY2+3vZN+yEr66XfS1ZwN8rKtk3vqlSLyrr AoGGKLvU8COVfQ0/ZAwY/Kx4MUs/Uq35Mb8tMb8CtIffUMngl5BABYZfQUKC4acw6dfdNdRX1z0U BD6ZKnUZPl2Z6tk++wrgUwACn4wQ7TN89mWEpI9WaXRjt8MlHwQuLIfshoud5Rguxsy7DZeSuMva My5rjZFEbDaGjyToJuXIaZ+UI/A5K5D/Xmyr8XhYT3lgdrOvzeULunxtQFBbd017lhaapardswPS oGrCHZw1u13y1jYt4Gqr02a76tTucqPdL8zN0lzuquumZl+Tv7vZ21YXKfeW+1ytdYGe8VNGlZ0z 1tozY42a8itjTZGdjZJjjTfa/WKsMmkeL8cqk2OVybHGe8cbY5GR41P83VaqCdTi/knew5MSka/B XGegJsPWWWkkb7kza3nuLpxWtlGSJ6Alu2q0FEDmdXF1cbU04ZmSpgFQp8ZNWcvLnbm72La4yQZ1 mquGPIuXdC2hLN+8utivCxdUi5fIWxGjHqn71QsuPs3bWifP1g1a4bQGrapxhr/bYoE2WBeAbtxp XVKSL6r3xZTnQzlOOgpxxlHqKqQuISHu+H9zwZgT1IhOLw4au3uYN58tpq6A0PIbmji2gqYZCEPz DP8unKXkS6IrgAV2MQ/rOt2bXIchU0xDWHbXaSxeEpfisVgc54Zrl4c8XadDcro7jwyWQYxYLfZg azPtomwgx/QYZStuwveP/hnwueT98/TPpV1y/iU2umgcRNvoKTaPnqI99AI7gVbbqZd2kDwC1dG9 dAPdQavxWpsBzVqaimKC/g6Wre/Al8mDeGE+SPvhexktp12UwbL0L2gFrRJvodUqSqEhVE1TqIM2 sEv0JdRMh5WbqYwuoauok4V0v36bvkn/Az1CveLP+ilKohyajbJf/8r0V/0DKkaLO+luOsw2JTxD XowSgud9dDVtETMVps/VT2IGTlqKOSg0kfazPu5B7230GctiN4ha9PKwrukvwstOM6mdttAuNpqN 505Tsz5R308ZGGMZer2bIrQTJUrP0Xss2XRC/4N+grKpiC7GenbQG6xP9J9a2V+FuJkQpeE0FpYO +hO9QgeZiz3PO0zJplKT13Sd/jYNopE0HbN9DC3/m33Pl6OsEC8r9XoNDUBcfiejTS/RxyyHlbDJ 7FI+nHfw+8XVZMWII1Hm0DzE+y70/hHSaCdP5gfEw8qTyo/mvP4j+gDcETfdQ/fR8ywFK1VZF7uJ HWKf8Frewu/hR8UdyuPKm5ZWrPpyWkgb6En6nqWzMayR/Za1sxvYavY7djfbzw6yz3k1b+JX8uOi XSwSzyk1KNOULuVm062mdebP+/39L/b/V//3eql+KzUiH1Zi9nfS/VhZLx2gd1EO01FmYklsAIrK nGw6ux5lOdvAHmLb2ONsB0Y5yI6yL/BK+pb9yPGm5Waei8OPPAK5+NU4Yd7B7+UHUA7yv/MfRKYY IjxitKgQAdGBWa0Wt6M8Iz5WcpQDio44l5o2m7aatpmeNL1gOmFOttyEd/zrPz18qvDUR/3Uv6Z/ c3+kf4f+MQ3GPcTbA59gFZh9K8p83O/NyLjt9BZLRuxyWCGrZJcgMi1sPlvEliGSt7At7BFj7k+z ZxGld9hxzDmF2405n89H8xo+GeVy3sYX4TC2ie/gh/hJYRFJIlUMFoVivJgp2sRica3YLDTxuvhQ HBXfiZ9QdCVRcShDFLfiUcYrLcoS5X7lM+UzU7PpNdPfzInmheZbzVHz1zjVVFqmWBotMy0bLTst b1uDyM699Az9ERl45mJHxErhE8/QbfwCJRufMG8gn1tojpjIkal8G1vDb2Q7+FDTMnM5L2eT6ITi Rqxf5lv5d7xcTGQNbBrN5yNjHZoHKU9AqlD20jHlWaztDfS8zJzMlvPj5mSK4Iw0Fmekl8QIxSNe o/fEYWZRHqT3lUSWyY7xx8QUZMFzSqXJT05xLz0tFrEb6RnuI0r80boeeTyJPYF9oYmVsn8KHcfg SciiMvEJ3UxX8r/SMTzHa+j3bI4yl26jC9gN9Bk9iqdiuOkqc6F5MHuVz1PCfCDbQVx5HKsby4Yy YRpEt7CZYov5OH+XltABJZE+Ev+J2R/gT4uJygnTVNaOJ+BGupUW6SvpWpNfeZPNJcEupQLlCHa3 G0Sp4gRfgV2lGXvaTjzdu7APVIuJ0GQhcy5BXkzHDrEF5S7sEwoyaB6e8cuwi71BO8xNPEpzTQMY dh38pea1/qk0Q3+U7tbn0lX6JirGfrBavwE9bqO/0Ubaxlb1X0+d+JR8F8/2JaZ6fsBUrxfzMH+X T+Obz72/iHYBy6IvUZ7Gnak07aaw8g5Noyp9vf4XZPd52GHvplk4sH6KVX6FESaIPrqgfxLv1utF J9Z7mBr1x3QHS6R2fQFNpmfpEYuJWi0e3GONvYn1Xk9tfKq+WLT1z0McNiIKXkRrCfaftd7a6U3V 3qrKiyrKx40dUzZ61AWlI0eUnF9c5Ckcft4wd8FQ1xCn6sjPs+fmZGdlZgweNDA9zZY6ICU5KTHB ajGbFMEZFflc9UFVcwc1xe2aMKFY1l2tULSepQhqKlT15/poqmzXCtM5nl54XvELT2/M03vGk9nU CqooLlJ9LlXbX+dSo2xGox/yhjpXQNWOGfJEQ77dkFMgO51ooPqy2utUjQVVn1Z/TXvYF6wrLmLd SYm1rtq2xOIi6k5MgpgESct0dXazzEpmCDzTN66bkzUFS9RyXHU+LduFpuhGFPha52hTGv2+ulyn M1BcpLHa2a5ZGsmTksdwoVpjGM1cq1mMYdR5OONotE7tLuoLr4/aaFbQkzzHNae12a+JVvTh09I8 GLdOy7zu06yfq+gcZ7LVZ1tzRdiXNU+VzuHwalV7oNF/Vttcp+whEEAfaMsL6oPhegy9HneqQZ7F Nb4q4NfYKgyJg2WBsarY+mKn3oLgfFVLcNW42sPzg7g1OWGNpl7rjOTkeHv1I5TjU8NNfpdTq8p1 BVrr7N2DKDz12p5sr5p9rqW4qNuWFgts94DUuJCccrbQhqDHbIZkuEupYeqZyDI5R9fFOAlq6mwV M/G7sKYxkrSNofDsMbgBuAIMrbQ5uCPztITaYNg2TuqxRKaZCmwuNfwtIQNcx/5+rqY1rjEX2L4l aZR5cibVNNZ6WtY8Hq2wUKaIpRb3FHOsNOqji4uuiXKXq9OG72f50UBTENvWwLgShN/plDd4XdRL s1DRQo3+WF2lWbkR8pbgbM2D0tJ32jJ4urSETlvONA+6kMk75PcsDdas7jO/VFvGQF/7OI1l/Btz W8zeMM3VgKOx6gsH41nb0HROLWaXAUXcYItL2sBav8jl0EmJ5wrDGjshn3bBcdmfrCkF+JmNpJ4T tViRlYaGqfWaLTghRgOJTmf8mfn/GkX1E7KVwX5uFl+GNs4Tn2hs2lr5OfVzppccFg1N2HI4Tvbh cOI5NqRabJYXxxkyHh/6TrVWo+l4MgvwwyfHGIlAruZFyGBpwlNkqAO58eo5jrnxRgFcMjuLi+qx Z4bD9S61PhwMt0b10CyXanOFe/kL/IVwpw+7XSxxovqudbla/foAItbOxuHx4FTT7WJrGru9bM20 Gf5e/IlDXdPkj3DGa4M1AXlbeG2TPx4W44bI1Mc9xD80kDFKF05JBEEmD+E/LWaAcGo/rUFK0ZXQ YGuXh2ET/qCP/1WQM82ZVgCCHugnVfT95DXRj6QqfbIv9Khk4MskhTq9A15MYQp+3KokiBSK8t3e EZwpCckpXUJwr9M1ajJOsR1c8JxUa1fC/9BkHM1auKgC62ArmMKyB0TZpu51WZ5Jtn/MXFQx8R/H Jtm+W+SpmGg7RlVVFacqxqaNZWnpY0eOYItmDhztHGwmYba4LkxPL2sVz6zvP9ZwYWqvuOl/1yon n1p/Z396/4/R959iX7JX7sWKjD8EgenDcE78tYtDmWMYGKXHo2TGuZ4aJ1xy2cWTPNVXz2tdMLHp X1+5/p4KZW5kc3RyZWFtCmVuZG9iagozNCAwIG9iago0NjMwCmVuZG9iagoxMyAwIG9iago8PCAv VHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9UUFdKUU8rQXJpYWxNVCAv Rm9udERlc2NyaXB0b3IKMzUgMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RD aGFyIDQ5IC9MYXN0Q2hhciA0OSAvV2lkdGhzIFsgNTU2Cl0gPj4KZW5kb2JqCjM1IDAgb2JqCjw8 IC9UeXBlIC9Gb250RGVzY3JpcHRvciAvRm9udE5hbWUgL1RQV0pRTytBcmlhbE1UIC9GbGFncyAz MiAvRm9udEJCb3ggWy02NjUgLTMyNSAyMDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQg OTA1IC9EZXNjZW50IC0yMTIgL0NhcEhlaWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9Y SGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZp bGUyIDM2IDAgUiA+PgplbmRvYmoKMzYgMCBvYmoKPDwgL0xlbmd0aCAzNyAwIFIgL0xlbmd0aDEg Njg3MiAvRmlsdGVyIC9GbGF0ZURlY29kZSA+PgpzdHJlYW0KeAGFWQt4VNW1Xnvvc2YmLzKJkOck c4Yhg2QSAwEMkDSZPCagEUkg6AwNZgKkBASJJvhEGKoojjws16LgAx9VUaucTCKdgJYo1rYowlVr 66OCaK+Prwj1+9QqmnP/fWZA8fr1np1/rbXXWvu1zjr77DPpvWplJ6VSmAT5Fi7v6Cbzyl0Elrfw 6l4tXk/LIbK0/qJ78fJ4/ZxlROpni5dd94t4PXc9EbN0dXbIdvL6Fji/CwqzRmwS+Jiu5b3Xxuu5 beC2ZSsWJuy5YdQtyzuuTYxP76KuXdGxvBMcV14MROte0dNrVinvDvBzu6/qTPizAFHK70/F3nmK fcr+eG/cCVOCwOlzqqL7yArJTmV0CZHSpRSQirq0q3zzls83fNmeXvWFLd9mtnzog7HFUtivdN/x za7vFtvJlopqkukvDWhnrR6+mOrt9M2ub663x0eSltMXH6RWcW6/J8d5+Fkxjo4CXIyLegucg2Ks KIhWOn0x4e7PHFWeXlsqNPRYZlINdAWwC9gHKNQuCmG1g64BwsAuYB9wGLAQgUqrBqwAdgBHAYso EI6o5rTXjhW5aJuL9aaLbDoBGIAgJ2gZMAtoBzYDOwCL6Sc1K4A1wD7gJGAhn8iObpmIuWdHbzdZ /9Jl5Wa1I15tm29W+y8NxvnMljhvuCDuNi3uNmFSXH1eXZyPLYnzzKLyMDrvT04rH6rNEllYZBYm 3g3K+IuUzhg56QExinSAC0zV1PhEZv8YT/mOfUIhJrhgtIicxpBg0bSM8tpkbvATlElO/hk/Hrfw 4/0jMsp31F7Ij9EuYB8g+DGU9/n7tIYflTEHrQF2APuAQ8AJwMKPohxBeY+/R+n871QG1ADtwA5g H3ACsPK/g9r5uzJjTCrlGoDzd0Ht/B0s6x3QdP42pLf528YQfz1aMbV80BS8ZQnBWZQQsvMTQmZW eYy/Fv16HDLKgzuNjNorRlM1TRSjo0UTnDGRE61a4ozxD/o1r/OB2vH8DdIBjpm8gZHfIA1oBkJA N2CB9CakNykM3AE8AOgAsgzUDmj8APAK8CaNB3xAM2Djh6MYJsYPRT11ztos/ir/I2Uj4gf5n0z+ Cn/J5C/zP5j8z+CFsB/gL0ULnVSbAjuhjR3cDl4Gu8qf7x+T6TRqM/g+RNAJWgbUALOAdmAzYOH7 +OjoImcmOtlLB/AMO3mUPjH5o/SQjXxLnT5PPRJQk8Qz7WeQQHZoOzzc59m6DVVJPJu2QJLEc/MG SJJ4rl8LSRLPsqshSeJZtBSSJJ557ZAk8cxqhQQS4/f/bsxYZ8Wsy5lWm86vQZSuQZSuQZSuIYVf Iwt9rcg53hMtLkbEtvu844qd4T0s/CwLz2bhh1i4k4VXs/BaFq5i4ctY2MvCDhYuZGEfC+9lUxCK MPMNnFWd6sth4QMs/BQL97Cwh4WLWHgMC2uswhfjrugFeOrA/Cbrr5UPHXf1/6wau086dyGiLuS8 C3vCPtBDgGHWfHDSRsedcwslH91fXBOvnzetfEXtDL4fDffjNuynI4CCG7QfabQfnexHd+mgNUA7 MAScAAzAAu/RWMdmk6aDlgE1QDuwBjgBWMzpnMBUOK0AlVPcZU6sDLQGmCVrfD/KaBQXd/kK7A67 1z5DbHaw9EI2q9Ao5BWUlYV9OTPDlhFjabu/Svv3V2mUVJvEN/HNVIAbcUeCb45+XeCMsbujnr3O 2lHsLipUkHVsKnlYEfgU6jHrk8lhk/pJ5OBPgpdHHZegWXrUU+Lcw0bIVrudXzs+dH7iiHGIHzv2 Ov+qxRQWdf4Fmid3O99w3Ob8c1nMBs2znhgD26OZroOOKc6nDpiua2HYHnWulmy380bHdOflDtPQ GTdc1oOaL9052zPPOQP9NTgWOH096HO3s8ZxmbMq7jVZttntHI8peONiMSY7zmEO6i40O5xbEWNd vhLrVmvAOst6vrXcWmJ1WZ3WAmu+daQt02a3jbCl2pJtNpvFpti4jWwjY8ZRn1e+9UZazJefBQnN SDFlO3YYJrcZUOLMxulC0s8RTbxpTh1r0ocWUtMCTf9yjjvGklvm6aq7jumZTdTUWqdP8TbFrMZs vcLbpFubfx7oY2xTEFqdr48xag3EmCFV6/L1zPrAIDGWsW5jvuTnrtsYDFJO1tU1OTWZ1RlTGxt+ goRMZajB+/2V873ozfEW6Fub5gT0JwqCerkUjIJgk/5fc7S2wCD7nJ30Nwyyf0kWDAyKava5f7bU i+qGYLApxi4x/Uhj/4IfMgYMfja8mKUfabbCuN/2uF8R2sNvjGTwS0qiItOvKCnJ9FOY9OvrGeNv 6BsDAp9sjXpMn55s7Yc+B4rgUwQCn6wwHTB9DmSFpY9ebXbjcMClEAQuLI8cpouD5Zku5sz7TJey hMttZ1xuM0cS8dmYPpKgm7Sjp33SjsLnB4H8z2JnndfL+iuDC9v8nW5/yO3vBEL67Vd35ejhBZrW tzAoDZouPKEFC7sk7+jUg+7OBn2hu0HrqzTb/cjcJs2V7oY+avO3BvrafJ0N0Upfpd/d0RDsn948 qeKssW47M9ak5p8Yq1l2NkmONd1s96OxKqR5uhyrQo5VIcea7ptujkVmjjcH+mxUF6zH/ZO8n6ck I19D+a5gXZa9u9pM3kpXzur8PTit7KQUb1BPddfpaYDM69La0lppwjMlTSOgTk+YclZXuvL3sJ0J kx3qDHcdeXtX9qykHP+ShvhfDy6oelfKWxGnXqn7yQsuft3X0SDP1k168ZwmvaZlXqDPaoU21BCE btppXUqKP2YMxZXnQTlNOgpxxlHqqqQuKSnh+H9zwZwT1IjOIA4ae/uZr5D1Uk9Q6IVNrRxbQes8 hKFtXmAPzlLyJdETxAJ7mJf1nO5NrsOUKa4hLLvnNHpXJqRELHoT3HTt8ZK353RITnfnlcEyiRmr Xi+2NnUP5QJ56mOUq3gI3z/GR8DHkg8vMT6Wdsn5p9jo5NeJBNFOeootoadoH73ATqLVLhqkAZJH oAa6l1bRnXQrXmvzoLmNZqOo0N/Jco0BfJk8iBfmg3QQvpfSatpDWSzH+ITW0DrxOlqtozQaTbXU TCtoI7vIWEltdES5iSroIrqCulnYCBibjC3Gb+gRGhR/Mr6jFMqjhSgHjc/UvxnvUila/Jq20RG2 JekZ8mGUMDzvo6tou5ivMGOx8Q1m4KJrMAeFZtJBNsS96L2TPmI5bJWoRy8PG7rxIrwcNJ+6aDvt YZPZdO5S24yZxkHKwhjXotdtFKXdKDF6jt5mqepJ4zfGScqlEroA6xmgV9mQGP5u7XANIqYiSuNo Kiwr6Pf0RzrM3Ox5vkJNVctVn3q98QaNpAk0F7N9DC3/h33FV6OsES8pjUYdjUBcfiWjTX+g91ke K2Oz2CV8HF/B7xdXkQ0jTkBZREsQ77vR+3tIo908lR8SDytPKqcsBcNHjRG4Ix66h+6j51kaVqqx HvZL9ib7gNfzdn4PPybuVB5XXrN2YNWX0XLaSE/SVyyTTWEt7Oesi61it7JfsW3sIDvMPua1vJVf zk+ILnGleE6pQ5mj9Cg3qbeot1s+Hg4Mvzj838NfGeXGLdSCfFiL2f+a7sfKBukQvYVyhI4xlaWw ESgac7G57AaU1Wwje4jtZI+zAYxymB1jn+CV9AU7xfGm5Raej8OPPAK5+VU4Yd7J7+WHUA7zf/Kv RbYYLbxisqgSQbECs7pV3IHyjHhfyVMOKQbiXK5uVXeoO9Un1RfUk5ZU6y/xjn/l24e/K/7uvWEa Xj+8dTg6PGC8T6NwD/H2wCdYFWbfgbIU93srMm4Xvc5SEbs8Vsyq2UWITDtbyq5k1yKSN7Pt7BFz 7k+zZxGlv7ITmHMad5hzPo9P5nV8FsplvJNficPYFj7A3+TfCKtIEelilCgW08V80Sl6xXViq9DF K+Lv4pj4UnyLYijJilMZrXgUrzJdaVdWKvcrHykfqW3qy+o/LMmW5ZZbLDHLv3CqqbY2W1us862b rbutb9hCyM799Az9znxqE4QdFWuFXzxDm/hEJRefMK8in9tpkZjJkal8J1vPb2QDfIx6raWSV7KL 6aTiQaxf4jv4l7xSzGRNbA4t5RPi3VlGKk9AqlL203HlWaztVfR8rSWVreYnLKkUxRlpKs5IfxDj Fa94md4WR5hVeZDeUZJZNjvOHxPNyILnlGo1QC5xLz0trmQ30jPcT5R8yrYBeXwxewL7QisrZ/8W Bo7BFyOLKsQHdBNdzv9Gx/Ecr6e72CJlMW2iiWwVfUSP4qkYp15hKbaMYn/mS5QIP4cNEFcex+qm sjFMqCPpZjZfbLec4G/RSjqkJNN74reY/SH+tJipnFRnsy48ATfSLXSlsZauUwPKa2wxCXYJFSlH sbutEuWKC3wNdpU27Gm78XTvwT5QK2ZCk4PMuQh5MRc7xHaUu7FPKMigJXjGL8Uu9ioNWFp5jBar Ixh2HfxS8/LwbJpnPErbjMV0hbGFSrEf3GqsQo876R+0mXaydcM3UDc+Jd/Cs32R2sgPqY1GKY/w t/gcvvXs+4toF7Ec+hTladyZanUvRZS/0hyqMTYYf0F2n4sddhstwIH1Q6zyM4wwQwzRxOGLeZ/R KLqx3iPUYjxmOFkydRnLaBY9S49YVeqwenGPdfYa1nsDdfLZRq/oHF6COGxGFHyI1krsP7f56ue2 1vpqqn9WVTlt6pSKyZMmlk8YX3ZeaYm3eNy5Yz1FY9yjXZqzsMCRn5ebk501auQ5mRn29BFpqSnJ STarRVUEZ1TidzeGNN0T0hWPe8aMUll3d0DR8QNFSNegajzbR9dkuw6YzvL0wfMXP/L0xT19ZzyZ XauiqtISze/W9IMNbi3G5rUEIG9scAc1/bgpzzTlO0w5DbLLhQaaP6erQdNZSPPrjVd3RfyhhtIS 1peSXO+u70wuLaG+5BSIKZD0bHd3H8uuZqbAs/3T+jjZ0rBEPc/d4Ndz3WiKbkSRv2OR3twS8Dfk u1zB0hKd1S90L9BJnpS8pgvVm8Polnrdag6jLcEZR6fbtb6SociGmJ0WhLypi9yLOtoCuuhAH349 w4txG/Ts6z/M+b6KznEmu/WH1nwR8ecs0aRzJHKrpj/QEvhB23yX7CEYRB9oy4saQ5FGDL0Bd6pJ nsV1vi4Y0Nk6DImDZZG5qvj64qfeotBSTU9y17m7IktDuDV5EZ1mX+eK5uX5Bo2jlOfXIq0Bt0uv yXcHOxocfSMpMvu6/lyflnu2pbSkz54RD2zfiPSEkJr2Q6ETQY/bTMl0l1LT7DORZXKO7gtwEtS1 hRpmEnBjTVMk6ZxCkYVTcANwBRla6YtwR5boSfWhiH2a1GOJTFeL7G4t8gUhA9zH/3m2piOhsRTZ vyBplHlyJtV01nFa1r1evbhYpoi1HvcUc6w265NLS66Ocbe7247vZ/nRQM2IbUdwWhnC73LJG3x7 zEcLUNHDLYF4XaMF+VHyleFszUPSMnTaMmqutIRPW840D7mRyQPye5ZG6TbPmb90e9Y5/q5pOsv6 D+bOuL1pjrsJR2PNHwklsrap9axa3C4DirjBlpD0c+oDIp9DJyWeL0xr/IR82gXH5UCqrhThz2Im 9aKY1YasNDVMa9TtoRlxGkx2uRLPzP/XKGaclK1M9n2zxDL0ad7EROPT1ivPqp81vdSIaGrFlsNx so9Eks+yIdXis7wgwZDx+NB3afU6zcWTWYQ/fHJMkQjm6z6EDJZWPEWmOpifqJ7lmJ9oFMQls7O0 pBF7ZiTS6NYaI6FIR8wIL3BrdndkkL/AX4h0+7HbxRMnZuy5PV9v3BBExLrYNDwenOr63Gx9S5+P rZ8zLzCInzi09a2BKGe8PlQXlLeF17cGEmExb4hMfdxD/EMDGaP04ERHEGTyEP7TYgEIZ9LTGqQU 3tjylxIQFQWnayuRK8OVUQSCHuhbTQx961PpFGnKkOwLPfJT6hAl0y6fJnxpGZMuV9bwzXybTfmt wpLIonKRpLJUzg4kE0LnS3a5J40npqEtfroZsNv5XAif+jLS0yE5UlNBR6SlmdqTvtz0dMtcsqem SpqWBpqXqvrS0iepsq8Rsi+VaTidczU3ZQ+rYusox3ux/cP5V3q99i8TX1WoVM38ropqarKnsoyp E8az+TTf63JnWCzWyeefXzGRnxqofb31rmNlvcoN1aucT08/0I4ImD8cgRljca78qYtDOdI0MMrE euRlwXcAtTZfemHLLG/tVUs6ls1s/V+KhgoJCmVuZHN0cmVhbQplbmRvYmoKMzcgMCBvYmoKNDY2 NQplbmRvYmoKMTcgMCBvYmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1RydWVUeXBlIC9CYXNl Rm9udCAvSFNDR0VUK0FyaWFsTVQgL0ZvbnREZXNjcmlwdG9yCjM4IDAgUiAvRW5jb2RpbmcgL01h Y1JvbWFuRW5jb2RpbmcgL0ZpcnN0Q2hhciA1MyAvTGFzdENoYXIgNTMgL1dpZHRocyBbIDU1Ngpd ID4+CmVuZG9iagozOCAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9I U0NHRVQrQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQov SXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9T dGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEg L01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiAzOSAwIFIgPj4KZW5kb2JqCjM5IDAgb2JqCjw8IC9M ZW5ndGggNDAgMCBSIC9MZW5ndGgxIDcwOTIgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFt CngBhVkLfBTV1T/33tlHnrtJSLJJNtnZLFkkmxgIYAJJk81jAzYir6C7NDEbICUgmGACPqqwWBFc nvJZWqgVa1vFWmWyibgJWFKh2qIIVWvro4KPfj5+RdBfxfog8/1ndkFp/fWbu/9zzz3n3Ne5Z+7c me27cVUnJVOIBHkXrejoIf3KWYwsd9HqPjlWTrERGVu+37NkRaycsZzI8NGS5bd8P1bO2UjE3u3q 7NDqaddXwBVdEOglYpORj+1a0XdzrJzTity8vHtRXJ8TQtm4ouPmeP/0BsryDR0rOpHjyrsXRO7p 7u3Ti5SH/umynhs74/bMT5T02y+jrz/GPmTP3hczwpDAcPqEqulnZAJnpTK6hkjqkvLJgLKmN/Bt Oz7ZfK7dUv2pOc+s13zwnXHFGnNY6tn+xb7zS6xkTkYxQbfXFKhnqhm9mhqs9MW+L261xnrSNBcu PkQt4rIBt81x4qAYT6cALsZHPPmOITFO5EeqHN6ocA2kZ5Zb6kqFjBbLdCqDdgP7gEOARO2iAFor 6FogBOwDDgEnACMRqKaVgW5gD3AKMIp8YY/IDmvdOJGDujmYr0Vk0xlABQQ5QMuAWUA7sA3YAxh1 O03SDawFDgFnASN5RXZkxySMPTuySc8Gli0v14sdsWJrm14cuDYQy2fOieWNV8bMpsXMJk6OiS+v j+XjSmJ5elF5CI0PJKaUj9RliSxMMgsD7wFl/AhZGCMHPSAySQG4wFB1iVekD4x1l+85JCRiggtG i8mhjggWSUkrr0vkKj9D6eTgH/HTMQ0/PZCaVr6n7rv8bdoHHAIEfxvpLf4WreWnNJ+D1gJ7gEPA ceAMYOSnkE4ivcnfJAv/G5UBtUA7sAc4BJwBTPxvoFb+hhYxOtX4WoDzN0Ct/HVM63VQC38N3Gv8 NXWEvxSpmFo+pDOesjjjKIoz2XlxJj2rPMpfjHw+HhHlxkojog6IQqqhSaIwUjTRERW2SPVSR5S/ MyB7HA/UTeAvkwJwjORl9PwyycBsIAj0AEZwr4B7hULAduABQAEQZaBWQOZHgeeBV2gC4AVmA2Z+ IoJuovx4xF3vqMviL/BnKRseP8b/oOfP82f0/Dn+ez3/I/IC6I/yZyIFDqpLgp5Qx4rcirwMegP/ 3cDYdIdal8YPwYMO0DKgFpgFtAPbACM/xAsjix3paOQAHcU97OAR+kDPH6IHzeRd5vC6GxCAskbc 074DDmSPvMfNve6du1DUiHvrDnAacd+5GZxG3LeuA6cR9/LV4DTiXrwMnEbcC9rBacQ9qwUcSJTf /+TYcY6KWdczuc7Cb4KXboKXboKXbiKJ36Ql+lzSxvjTSHExPLbb6xlf7AgNs9BBFprLQg+yUCcL rWGhdSxUzULXsZCHhewsVMBCXhY6wCrhihDzDl5SnOq1sdBRFnqMhXpZyM1CRSw0loVkVuGNcmfk Stx1yHx6NlCn3XTcOfCdGuw+Fu6ER52IeSf2hEOgxwFVL3lhJBfGjHMKtLxwoLg2Vr58Wnl33Qx+ GBUPYxkO00lAwgIdRhgdRiOH0ZwFtBZoB0aAM4AKGGFdiHls06kFtAyoBdqBtcAZwKgP5wyGwqkb VBviPn1gZaC1wCytxA8jFSI5udObb7VbPdYZYpudWQrYrAK1gFdQVhb25fQ0c1qUpez/LOVfn6VQ Ql0C38q3UT4WYns83xb5PN8RZT+JuA846jLZj6lAQtSxqeRmRcgrqVcvTyG7WZNPJjt/FHl5xH4N qlki7hLHMEvVau13fG5/1/GBPcrBvm8/4PiLHJVYxPFnSB7d73jZfrfjj2VRMyQH3VGGbFjWTYfs lY7Hjuqm66DYHXGs0bL9jtvt0x3X23VFZ0xxXS9KXotjrnuBYwbaa7QvdHh70eZ+R639Okd1zGqK Vme/YwKG4ImxxRjseLveqatAb3B+RZR1eUtMO01+0yzTFaZyU4nJaXKY8k15pjHmdLPVnGpONiea zWajWTJzM5nHRNVTXo/21Btj1B9+RgQ0I0nnrdhhmLbNgBJnZk7fJSVDNPPmefWsWRlZRM0LZeXc PFeUJc5ZoBhc9UxJb6bmlnql0tMcNalzlQpPs2Ka/T1/P2NbA5AqfGOUUYs/ylRNtD5PSW/wDxFj aeu35Gn5Zeu3BAJky1pda6tNr0mb2tT4LSSoC4ONnq8v29esx+bJV3Y2z/Mrv84PKOUao+YHmpX/ mSe3+ofYJ+ysr3GIfaxlAf+QqGGf+OZqclHTGAg0R9k1uh3J7GPYIWKQwc6MB7NmR7K5IGa3O2ZX hPqwG6tlsEtIoCLdrighQbeTmGbX3zvW19g/FgQ22TL16ja92fI3bY4WwaYIBDZZITqq2xzNCmk2 So3ejN0OkwIQmLBcsusmdparm+gj79dNyuImd180uVvvScRGo9toBM2knLpgk3IKNt9w5H9nO+s9 HjZQFVjU6ut0+YIuXycQVDat7rIpoYWy3L8ooClkRbiDCxd1aXlHpxJwdTYqi1yNcn+VXu/f1K2a usrV2E+tvhZ/f6u3szFS5a3yuToaAwPTZ0+uuKSvuy/2NXn2t/Q1W2tsstbXdL3ev/VVoamna31V aH1VaH1N907X+yI9xmf7+81UH2jA+mn5AE9KRLwG85yB+ixrT40evFVO25q8YZxW9lKSJ6Aku+qV FECL69K60jpNhXtKU6VCbImrbGuqnHnDbG9cZYU4zVVPnr5VvavI5lvaGPv14oKob5W2FDHq0WTf esHEp3g7GrWzdbNSPK9ZqZ2zwN9vMkEabAxANu2CLCnJF1VHYsLLIZymGQpx0VCTVWuyhIS44X/G gj4miOGdIRw0DgwwbwHro96AUAqaWzi2gpYFcEPrAv8wzlLaQ6I3gAn2Mg/rvdCaNg+dp5iEMO3e C+hbFefivuiL57ppr4c8vRdccqE5j+Ysnei+6vNgazMMUw6Qa3iYciQ34f1HfQ94X8tHl6rva3ot 5x9io4vGQbSXHmNL6TE6RE+zs6i1j4ZokLQjUCPdR7fRvbQBj7UFkNxNc5EMkN/LctRBvJn8HA/M n9Mx2F5La2iYsphN/YDW0nrxEmqtpxQqpDqaTd20hV2lrqJWOin9kCroKrqBelhI9atb1R3qL+lX NCT+oJ6nJMqlRUjH1I8Mf1XfoFLU+BHtopNsR8IT5EUvIVj+jG6k3aJNYuoS9QuMwEk3YQwSzaRj bIR70Honvcds7DbRgFZ+oSrqEVjZqY26aDcNsylsOncaWtWZ6jHKQh83o9VdFKH9SFF6il5jyYaz 6i/Vs5RDJXQl5jNIL7ARMXp+3Wgt/GaAl8bTVGi66bf0LJ1gLvY73m1INpQbvIZb1ZdpDE2k+Rjt w6j5v+wzvgZprXhGalLrKRV+uUfzNv2e3mK5rIzNYtfw8byb3y9uJDN6nIi0mJbC3z9B628ijPbz ZH5c/EJ6VPrSmD96Sk3Firjpp/Qz+h1LwUxl1svuYK+wd3gDb+c/5W+Le6VHpBdNHZj1dbSCttCj 9BlLZ5VsDvse62K3sQ3sHraLHWMn2Pu8jrfw6/kZ0SVWiqekeqR5Uq/0Q8Ndhk3G90f9o0dG/zT6 mVqu3kVzEA/rMPof0f2Y2RAdp1eRTtLbzMCSWCqSzJxsPvsB0hq2hT3I9rJH2CB6OcHeZh/gkfQp +5LjScuNPA+HH+0I5OI34oR5L7+PH0c6wf/BPxfZolB4xBRRLQKiG6PaILYjPSHeknKl45IKP5cb dhr2GPYaHjU8bThrTDbdgWf881/94nzx+TdHaXTj6M7RyOig+hZlYg3x9MArWDVG34G0DOu9ExG3 j15iyfBdLitmNewqeKadLWMr2c3w5J1sN/uVPvbH2UF46S/sDMacwu36mC/nU3g9n4V0He/kK3EY 28EH+Sv8C2ESScIiMkWxmC7aRKfoE7eInUIRz4u/ibfFOfEVkiolSg6pUHJLHmm61C6tku6X3pPe M7QanjP83ZhoXGG8yxg1foxTTY1ptmmOqc20zbTf9LI5iOg8TE/Qk4jAixc7JdYJn3iCtvJJUg5e YV5APLfTYjGTI1L5XraR384G+VjDzcYqXsWuprOSG75+hu/h53iVmMma2TxaxifGGjSOkX4Nrlo6 TKelg5jbC2j5ZmMyW8PPGJMpgjPSVJyRfi8mSB7xHL0mTjKT9HN6XUpk2ew0f1jMRhQ8JdUY/OQU 99HjYiW7nZ7gPqLEL82bEcdXs19jX2hh5exfQsUx+GpEUYV4h35I1/O/0mncxxvpx2yxtIS20iR2 G71HD+GuGG+4wVhszGR/5EulMM9gg8SlRzC7qWwsE4YxdCdrE7uNZ/irtIqOS4n0pvgNRn+cPy5m SmcNc1kX7oDb6S5aqa6jWwx+6UW2hAS7hoqkU9jdbhPlkhP5WuwqrdjT9uPuHsY+UCdmQmJD5FyF uJiPHWI30k+wT0iIoKW4x6/FLvYCDRpbeJSWGFIZdh18qXludC4tUB+iXeoSukHdQaXYDzaot6HF vfR32kZ72frRH1APXiVfxb19laGJHzc0qaU8zF/l8/jOS9cX3i5iNvoQ6XGsTI3hAIWlv9A8qlU3 q39GdF+GHXYXLcSB9V3M8iP0MEOM0KTRq3m/2iR6MN+TNEd9WHWwROpSl9MsOki/Mhmow+TBGivs Rcz3B9TJ56p9onN0KfywDV7wwlursP/c7W2Y31Lnra35TnXVtKmVFVMmTyqfOKHs8tIST/H4y8a5 i8a6Cp2yoyDfnpebY8vOyhyTkZ5mtaSmJCclJphNRoMkOKMSn6spKCvuoCK5XTNmlGplVwcEHd8Q BBUZoqZLbRRZq9cB1SWWXlh+/98svTFL70VLZpWrqbq0RPa5ZOVYo0uOsgVz/OC3NLoCsnJa52fq /HadTwHvdKKC7LN1NcoKC8o+pWl1V9gXbCwtYf1JiQ2uhs7E0hLqT0wCmwROyXb19LPsGqYzPNs3 rZ+TOQVTVHJdjT4lx4WqaEYU+ToWK7Pn+H2NeU5noLREYQ2LXAsV0k5KHt2EGvRuFGODYtK7kZfi jKPQJrm/ZCS8OWqlhUFP8mLX4o5WvyI60IZPSfOg30Yl+9Z3bV8X0TjOZBu+qc0TYZ9tqawZh8Mb ZOWBOf5v1M1zai0EAmgDdXlRUzDchK43Y6WatbO4wtcH/Apbjy5xsCzSZxWbX+zUWxRcJisJrnpX V3hZEEuTG1Zo7i3OSG6ud0g9Rbk+OdzidzmV2jxXoKPR3j+GwnNvGcjxyjmXakpL+q1pMcf2p1ri THLKN5lOOD2m0zndXOOa5170LNPG6LoSJ0FFXiRjJH4X5lSpkc5KCi+qxALgCjDUUhZjRZYqCQ3B sHWaJscUmWIosrrk8KeECHCd/selko64xFhk/ZQ0pRYnF0NNYR0XeMXjUYqLtRAxNWBNMcYavTyl tGR1lLtcPVa8P2svDTQbvu0ITCuD+51ObYE3Rb20EAUlNMcfK8u0MC9C3jKcrXlQ04xc0GTO1zSh C5qL1YMuRPKg9j5LmYrZffFnsWZl+LqmKSzrv6g7Y/rmea5mHI1lXzgYj9rmlktKMb3mUPgNujin ZDT4RR6HTON4ntC1sRPyBRMcl/3JilSEn1EP6sVRkxlRqUuY3KRYgzNiNJDodMbvmf+vUlQ9q9XS s6+rxaehTPPEBxobtlJ1SfmS4SWHRXMLthyOk304nHiJDqEWG+WV8QwRjxd9p9yg0HzcmUX44ZWj UkMgT/HCZdC04C7SxYG8ePESw7x4pQAuLTpLS5qwZ4bDTS65KRwMd0TV0EKXbHWFh/jT/Olwjw+7 XSxwourwpjylaXMAHuti03B7cKrvd7GNc/q9bOO8Bf4hfOKQN7b4I5zxhmB9QFsW3tDij7tFXxAt 9LGG+EMDESP14uxKYLTgIfzTYgSIplyUIKTwD4X2pQTEgITTtYnImeZMKwJBC/SVLEa+8hroS5Kl Ea2t+ep7ktPwEBWwa/q51j2+gagj3pSUFD6f2VMTCzIz7elRfsCbZJGkAntKKiOTLap+OGix8Pk6 o1nayjyeY2XHyqj2dO3p9Kll549Yj3gmTsjzjk9PTkkxzrfotDn3lvxw/s6MhzMOJ7+S/HqeOSHD llqcKxImGCYkDWNHEvgWZM1IzEzPyDiaahmTmjEm1ZIS5b/0ZmgD8aY+kMpTUy3eTBYf1JMWib3k xSyizOaVteGltVu7rWut26ySNWTqtXmdrskTbIxsVhvHYP/5pDYY23Y5/SCbgu/9P6IUVhlJfYIN s0q8jY14k2Av4ynLaTs+w+3o32TzWM+1eU7/E8l6ztO2EmxbNWZ5vu1dfaJtaVPL2qynre9uMF/u MdxuPUJp6VMnTmCVuNjKNo8nkFGRlTWp/Iopk92uQqNpXIYz0ymumFROmWNMRlehe/5TmbuW3zH4 2OZrN1/2yFb+6vknZ915zwgz92355x/Os5A1vOnIg7sjs2qz+Me/GV3dOnruT8/eEzmFNcYS41LH 4cz8bZemt+kKRunxiDHiHYdmzGuY7mvx1N24tGP5zJb/A6GPWE4KZW5kc3RyZWFtCmVuZG9iago0 MCAwIG9iago0ODgyCmVuZG9iagoxNSAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1 ZVR5cGUgL0Jhc2VGb250IC9ORkhZVFYrQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNDEgMCBSIC9F bmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDUxIC9MYXN0Q2hhciA1MSAvV2lk dGhzIFsgNTU2Cl0gPj4KZW5kb2JqCjQxIDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRvciAv Rm9udE5hbWUgL05GSFlUVitBcmlhbE1UIC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMyNSAy MDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0NhcEhl aWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2 Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUyIDQyIDAgUiA+PgplbmRvYmoKNDIg MCBvYmoKPDwgL0xlbmd0aCA0MyAwIFIgL0xlbmd0aDEgNzEyNCAvRmlsdGVyIC9GbGF0ZURlY29k ZSA+PgpzdHJlYW0KeAGFWQt4VNW1XnvvM488hkxCSCbJkDmTIYPJJAYCmPBoMnlMwEbkFXCGgklI IgFBggFEizCovAYEyrW0UCtqtdJ6lZNJpJOgJYrVFkW4am19VPDRin6NUL+q9UHO/feZgNLr13v2 /Guvvdbar7XX2WefM6tuXt1GyRQmQf6W5c0dZFxZrciyW9asUuNlm4PI3HBDx+Ll8fLwZUSmjxcv u/WGeDlrKxFX2tuaZT15fQ1c1Q6BUSI2Hvmo9uWr1sbLWQuQW5etaBnSZ4VRNi9vXjvUP72FsnpT 8/I25LhyekDUjhWdq4wi5exAfkXHzW1D9ixIlPTbr2JvPsY+Ys/fGzciYmA4fUJT6OdkAWenEppH pLQrI8mEstSb+K49n+z4rDFlyqfWHKtR88H3RhdK5pjSsfvLQxcW28majGKCYS8VqGepGLyWauz0 5aEvb7PHe5KaixfvpQZxRbfX4Tr1pCigMwAXBVHfSFevGC1GRie7/DHh6U4bUZpSVSxUtFhiUBV0 BXAIOAoo1ChyobWDbgDCwCHgKHAKMBOBSq0KrAAOAGcAsxgpnFHVZa8aLbJQNwvzTRGZdA7QAUEu 0BJgBtAI7AIOAGbDTkpWABuAo8B5wEx+kRndMw5jz4xuN7LupctKjWJzvLhgoVHsvi4Uz6fPiue1 V8fNJsXNxo6Pi6+sjueji+J5Wn5pGI13J9pK+6syRAYmmYGBd4Ay/iylMEYuul+MIA3gAkM1JH6R 1j3KW3rgqFCICS4YtZJL7xcsakstrUrkOj9HaeTiH/OBuIYPdA9LLT1Q9X3+Lh0CjgKCv4v0Dn+H NvAz0ueglcAB4ChwEjgHmPkZpNNIb/O3KYX/hUqASqAROAAcBc4BFv4XUDt/S0aMQSVfCXD+Fqid v4lpvQmawt8A9wZ/Q+/nr0TLJpb2GoyvZIhx5Q8xmTlDTFpGaYy/HP2iABHlxUojoo6IPKqgcSIv mj/WFROO6JQlrhh/r1v1ue6vGsNfJQ3gGMmr6PlVUoGZQBPQAZjBvQbuNQoDu4H7AQ1AlIHaAZUf B14EXqMxgB+YCVj5qSi6ifGTUW+1qyqDv8Sfp0x4/AT/vZG/yJ8z8hf474z8D8hzoT/On4vmuqgq CXpCHTtyO/IS6E386e5RaS69KpUfhQddoCVAJTADaAR2AWZ+lOdFW11paOQIHcc97OJR+tDIf0kP Wsm/1OX31iAAVUm8k74HDuSAesDL/d69+1CUxLtzDzhJvHftACeJ97aN4CTxLlsDThJv61Jwknjn N4KTxDujARxIjN/3m1GjXWUzbmRqVQq/BV66BV66BV66hRR+i0z0hSLH+LNoYSE8tt/vKyh0hftY +EkWns3CD7JwGwuvZ+GNLDyFha9nYR8LO1k4l4X9LHyElcMVYebvuaw40e9g4eMs/BgLd7Kwl4Xz WXgUC6uszB/j7ujVuOuQBYysu0redNzd/b0K7D4p3A2PuhHzbuwJR0FPArpR8sNIzYsbZ+XKPK+7 sDJevnJS6YqqafwYKh7DMhyj04CCBTqGMDqGRo6huRTQSqAR6AfOATpghnUe5rHLoCmgJUAl0Ahs AM4BZmM45zAUTitA5RAPGQMrAa0EZsgSP4aUh+Tmbv9Iu9Pus08Tu5wsJZfNyNVzeRllZGBfTku1 psaY7fDntn99bqOEqgS+k++ikViI3UP5rugXI10x9tOo94iragT7CeUqiDo2kbwsH3k5dRrlCeS0 Svl4cvJHkZdGnfNQLSXqLXL1sWGy1mHXF873XR86YxzsWecR15/UmMKirj9C8uhh16vOba4/lMSs kDzpjTFkfaph2ussdz123DDdCMX+qGu9zA67bndOdd3oNBRtccX1nSj5U1yzvfNd09BerXORy9+J Ng+7Kp3Xu6bErSbIOoddYzAEX5wtxGALnEannlyjwbllMdbuL7LstQQtMyxXWUotRRa3xWUZacmx pFvTrHbrMGuyNdFqtZqtipVbyZoe08/4ffKpl242Hn5mBDQjxeDt2GGY3GZAiTMrp++TNlzU8/o5 1axe62+h+kWq9tkcT4wlzpqvmTzVTEurp/qGaq3cVx+z6LO1Ml+9Zpn5g2AXYztDkGp8a4xRQzDG dCnalKOl1QR7ibHUTXfnyPyKTXeHQuTIWFPpqEyrSJ1YV/sdpMkQNtX6vrkc37A+h2+ktrd+TlD7 9ciQVioZfWSoXvuvOeqCYC/7hJ0P1Payf8gsFOwVFeyTwGwpFxW1oVB9jM0z7Ehl/4AdIgYZ7Kx4 MEs7Uq25cbv9cbt81IfdKJnBLiGB8g27/IQEw05h0q6rc1SgtmsUCGwyVeo0bDoz1W/bHM+HTT4I bDLCdNywOZ4RljZahdGM0wmTXBCYsGxyGiZOlm2YGCPvMkxKhky2XTLZZvQk4qMxbCRBM7YzF21s Z2DzLUf+Z7at2udj3ZNDLQsCbZ5AkyfQBjRp29e0O7TwIlXtaglJhaoJb9OilnaZN7dpIU9brdbi qVW7Jhv1/k29QKone2q7aEGgIdi1wN9WG53snxzwNNeGuqfOHF92WV/bLvU1fuZ39DVTNjZe9jXV qPdvfZVJ9VTZV5nsq0z2NdU/1eiLjBifGeyyUnWoBusn826elIh4bcpxh6oz7B0VRvBOdjvW5/Th tHKQknwhLdlTrdkAGdfFVcVVUoV7SqqGQZwypHKsn+zO6WMHh1R2iFM91eRbtbpzNTkCS2rjv05c EK1aLZciTn1S9p0XTAKav7lWnq3rtcI59VrlrPnBLosF0qbaEGSTLsqSkgIxvT8uvBLCSdJQiEuG UjZFyhIShgz/bywYY4IY3unFQeNIN/PnslXUGRJabn0Dx1bQMB9uWDA/2IezlHxIdIYwwU7mY50X W5PzMHiKSwjT7ryIVauHuCFfrBrKDdNOH/k6L7rkYnM+6SyDGL5a5cPWZuqjLCDb9AhlKV7C+4/+ AXBW5oNL9LNSL3P+ETa62BCIDtJjbAk9RkfpGXYetQ5RL/WQPALV0r20ju6hLXiszYdkG81GMkF+ D8vSe/Bm8gAemA/QCdheR+upjzKYQ/+QNtAm8QpqbSIb5VEVzaQVdDe7Rl9NC+i0cieV0TV0E3Ww sB7Ud+p79IfoYeoVv9cvUBJlUwvSCf1j05/1t6gYNX5M++g025PwBPnRSxiWP6ebab9YqDB9sf4l RuCmWzAGhabTCdbPfWi9jT5gDrZO1KCVX+ia/iysnLSQ2mk/9bEJbCp3mxbo0/UTlIE+1qLVfRSl w0gxeoreYMmm8/pD+nnKoiK6GvPpoZdYvxi8sHGwEn4zwUsFNBGaFfRbep5OMQ97mq8wJZtKTX7T bfqrlE5jaS5G+whq/o19ztcjbRDPKXV6NQ2DX34kvU2/o3dYNithM9g8XsBX8PvEzWRFj2ORWmkJ /P1TtP42wugwT+YnxS+UR5WvzCMHz+jDsCJe+hn9nJ5mNsxUZZ3sDvYae4/X8Eb+M/6uuEf5lfKy pRmzvp6W0930KH3O0lg5m8V+wNrZOraF/YjtYyfYKXaWV/EGfiM/J9rFSvGUUo00R+lU7jRtNm03 nx0MDj47+D+Dn+ul+maahXjYiNH/mO7DzHrpJL2OdJreZSaWxIYhqczN5rIfIq1nd7MH2UH2K9aD Xk6xd9mHeCR9yr7ieNJyM8/B4UcegTz8Zpww7+H38pNIp/jf+RciU+QJn5ggpoiQWIFRbRG7kZ4Q 7yjZyklFh59LTXtNB0wHTY+anjGdNydb7sAz/sWvf3Gh8MLbgzS4dXDvYHSwR3+HRmAN8fTAK9gU jL4ZaSnWey8i7hC9wpLhu2xWyCrYNfBMI1vKVrK18ORdbD972Bj74+xJeOlP7BzGbONOY8xX8gm8 ms9Aup638ZU4jO3hPfw1/qWwiCSRIkaIQjFVLBRtYpW4VewVmnhR/EW8Kz4TXyPpSqLiUvIUr+JT piqNymrlPuUD5QPTAtMLpr+aE83LzZvNMfM/cKqpsMy0zLIstOyyHLa8am1CdB6jJ+g3iMBLFzsj NoqAeIJ28nFKFl5hXkI8N1KrmM4Rqfwg28pvZz18lGmteTKfzK6l84oXvn6OH+Cf8cliOqtnc2gp Hxtv0Jyu/BrcFOUYDShPYm4voeW15mS2np8zJ1MUZ6SJOCP9ToxRfOIFekOcZhblAXpTSWSZbIA/ ImYiCp5SKkxBcot76XGxkt1OT/AAUeJX1h2I42vZr7EvNLBS9i+h4xh8LaKoTLxHd9KN/M80gPt4 K/2EtSqLaSeNY+voA/ol7ooC003mQvMI9ge+RInw4awHH25+hdlNZKOYMKXTXWyh2G8+x1+n1XRS SaS3xX9j9Cf542K6ct40m7XjDridNtNKfSPdagoqL7PFJNg8ylfOYHdbJ0oVN/IN2FUWYE87jLu7 D/tAlZgOiQORcw3iYi52iP1IP8U+oSCCluAevw672EvUY27gMVpsGsaw6+BLzQuDs2m+/kvapy+m m/Q9VIz9YIu+Di0epL/SLjrINg3+kDrwKvk67u1rTHX8pKlOL+YR/jqfw/devr7wdj5z0EdIj2Nl KkxHKKL8ieZQpb5D/yOi+wrssPtoEQ6s72OWH6OHaaKfxg1ey7v0OtGB+Z6mWfojuoslUru+jGbQ k/SwxUTNFh/WWGMvY74/pDY+W18l2gaXwA+74AU/vLUa+882f83chip/ZcX3pkyeNLG8bML4caVj x5RcWVzkKyy4YrQ3f5Qnz626ckc6c7KzHJkZI9KHp6XaU4bZkpMSE6wWs0kRnFFRwFPXpGreJk3x eqZNK5ZlTzMEzd8SNGkqRHWX22iqrNcM1WWWflje8G+W/ril/5Ils6tTaEpxkRrwqNqJWo8aY/Nn BcHfXesJqdqAwU83+N0GbwPvdqOCGnC016oaa1IDWt2a9kigqba4iHUlJdZ4atoSi4uoKzEJbBI4 LdPT0cUyK5jB8MzApC5OVhumqGV7agNalgdV0YzIDzS3ajNnBQO1OW53qLhIYzUtnkUayZOSzzCh GqMbzVyjWYxu1CU442i0Xe0q6o/siNlpUZMvudXT2rwgqIlmtBHQUn3ot1bLvO19xzdFNI4z2ZZv a3NEJOBYokrjSGSLqt0/K/itujlu2UIohDZQl+fXNUXq0PUOrFS9PItrfFMoqLFN6BIHy3xjVvH5 xU+9+U1LVS3BU+1pjyxtwtJkRzSafas7mp3t79XPUHZAjTQEPW6tMscTaq51dqVTZPat3Vl+Nety TXFRlz017tiuYSlDTLLt20wbnB7XGZxhLrn62Zc8y+QYPVfjJKipLSpGEvRgTuWStJVTpKUcC4Ar xFBLa8WKLNESapoi9klSjikyzZRv96iRTwkR4Bn4++WS5iGJOd/+KUmljJNLoaax5ou85vNphYUy RCw1WFOMscIoTyguWhPjHk+HHe/P8qWBZsK3zaFJJXC/2y0XeHvMT4tQ0MKzgvGySotyouQvwdma N0lN/0XNiLlSE76ouVS9yYNI7pHvszRCs3ov/VLsGcMD7ZM0lvEf1G1xff0cTz2Oxmog0jQUtfUN l5XieulQ+A26IU4bXhMUORwyyfEcYWjjJ+SLJjguB5M1JR8/sxHUrTGLFVFpSJhap9mbpsVpKNHt Hrpn/r9KMf28rGVk31QbmoY2yTc00PiwtcmXlS8bXnJE1Ddgy+E42UciiZfpEGrxUV49lCHi8aLv Vms0mos7Mx8/vHKUS4RyND9cBk0D7iJDHMoZKl5mmDNUKYRLRmdxUR32zEikzqPWRZoizTE9vMij 2j2RXv4MfybSEcBuFw+cmN63PUer2xGCx9rZJNwenKq7PGzrrC4/2zpnfrAXnzjUrQ3BKGe8pqk6 JJeF1zQEh9xiLIgMfawh/tBAxCid8h8HMDJ4CP+0mAGicZckCCkmP5Zga5eHYRM+6OO/CnKnulPz QdACfa2K/q/9JvqKVKVftjVP/5uSYeonHwt1cdm9PynL4bfZ+FyHkxh85fclo8AKPIm2lOSU3MTE ghG5TiW3wGkqsHlsyY4sRmmqPTmZz1Ut3ph+tkeae0t8J+RPJkqbWFlpH7APpE0sGXjO/lzaRPuz vlKJsWNy/FeYbBm2gG2zTQmkXpe6JkfMzlhmX5remrHadmv6ZlskfVvOw7ZEkyrkd6KkpGTbMMXC 0C+L8Ye6/ZjAETyGC8jGJvQkJ49QHH38Icri7f7RGKUJw7SldTaqK1SuOmw281w1bOn0+t2e8WO8 jLx2L8eI//kbqfHuLnbEWHk06xXWx8rxatbvT4Kdikcup91FMbana7vDZ/9soW/gnwP2z3wLV4Lx LRygyoHKgQvvp8rJ2S8slAxLTcucOHYMKy9nC2khWxkaXpaRMa70qgnjvZ48S9kl1mwZfdVV40oz 8HQ2WyQlT553Xo/rxzduOPTg7eOuSU9L6oxtXrpkR3qP+6PH1x6/8YbWO3YPnn3taZ3d6di3Rbtj 3QPp9/G1t7fccddd6hPPL462Nt57Ze5TO/sHP/0bBm1c+micqb/rknr5KVPGQBogLzPegejaummh hnm+qpuXNC+b3vC/VIZaXAplbmRzdHJlYW0KZW5kb2JqCjQzIDAgb2JqCjQ5MDYKZW5kb2JqCjIz IDAgb2JqCjw8IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVlVHlwZSAvQmFzZUZvbnQgL1VUVlVI SitBcmlhbE1UIC9Gb250RGVzY3JpcHRvcgo0NCAwIFIgL0VuY29kaW5nIC9NYWNSb21hbkVuY29k aW5nIC9GaXJzdENoYXIgMzIgL0xhc3RDaGFyIDExNiAvV2lkdGhzIFsgMjc4CjAgMCAwIDAgMCAw IDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAg MCAwIDAKMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAw IDAgNTU2IDAgNTAwIDAgNTU2IDI3OAowIDAgMjIyIDAgNTAwIDIyMiAwIDAgNTU2IDU1NiAwIDMz MyA1MDAgMjc4IF0gPj4KZW5kb2JqCjQ0IDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRvciAv Rm9udE5hbWUgL1VUVlVISitBcmlhbE1UIC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMyNSAy MDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0NhcEhl aWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2 Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUyIDQ1IDAgUiA+PgplbmRvYmoKNDUg MCBvYmoKPDwgL0xlbmd0aCA0NiAwIFIgL0xlbmd0aDEgMTI0MTIgL0ZpbHRlciAvRmxhdGVEZWNv ZGUgPj4Kc3RyZWFtCngBxXp7fFTF2f/MnLPn7G72cnZz2d3sJrubzW4gGwwkgZAQyQlJAI3cCWYx kQSIkgBCCHipKKGKQEShttp6afFWr7VsLtKA+pKK2tdboWpttSqotEXbCPVHUVGy73dmE5T+/Ly/ P3/n7DPzzMwzt2ee55ln5uz6dRtaiYV0EYnoy1a3rCXi8fwE0fPLrl4fSKatbkKUhVesvXJ1Mp26 ihDDZ1euuu6KZDrzWkLSH1zR2rI8mSbfIJ60AhnJNC1BnLti9XrQ8cfTiGDyqjXLRsozG5DOXt1y 7Uj/5D2kA1e1rG5FjGc8pw+sXdO5XiTJeB9ife261hF6ivop//X1wF+eop/S392XJCKEArGRz0kF +TlRCSMaKSSLCJFXyFnEgDQvN7Cdd3y+4/QSe8W/jV6jqPngx3n5HDkor911Zs/ZKzVitCBpEvS8 APXUqcOzSbVGzuw58wMt2RMvGX1s+8hCaUxfxO0//Kw0lhwFMGlsbzTLv0/Kk7J6p/j1ASnU50wv sleNkwJosVCEAYRrAHsABwAyWSJlo1RDuAnQBdgDOAA4DFAIQchLA4A1gN2AowBFypJ8vQG/VpUn eVDXg/naJRc5AUgAJOJHWAiYA1gC2AnYDVAEHc9ZA9gEOAA4CVCILrl67yjG2F29t4qor31VkUi2 JJONTSLZd2ksGc+al4xrLkqSlSfJJpQksy+YlozzCpKxM1zUhcb7zNaiwaoMKQOTzMDA1yKk7AVi p5T4yf1SOokDmIShihxdcvblRop2H5BkQiUmUbKc+BODEu21OoqqzCzBThAn8bPP2FCyhA312RxF u6suZh+RPYADAIl9hPdD9iHZxI5yniOsBOwGHAAcApwAKOwo3iN4P2AfEDt7nxQCKgFLALsBBwAn ACp7H6HG3uMSI0KOVwIYew+hxv6Caf0FoZ29C+xd9m5ikL3ZW1pWtE8g0cIRxB8eQVzeEcSZUTTA 3uj9aiwkKoKVhkQ9I+WQqaRYyukNT/APSO7eijb/APu4LxD13181nr1F4gCGkbyFnt8iAcBcQDNg LUAB9jawt0kXYBfgfkAcAClDqAEC7BXAa4C3yXiADpgLMLLDvehmgB3qjUzzV2Ww37PfERc4/jr7 bxG/xl4S8avsRRG/jDgb5a+wl3qz/aQqBeUEdTTEGuJClBvYb/tynf5ElYMdAAf9CAsBlYA5gCWA nQCFHWA5vcv9TjTyDHkFOuxnveQTET9CHjQSvd2vR6ohgAEeRMovBIZgd2B3hOmRu+5GkgeR2+8A xoPIzTuA8SDyg83AeBBZdTUwHkSWtwPjQWTxEmA8iMxZCAzBAPvFb3Lz/KVzVtJAlZ1dAy5dAy5d Ay5dQ2R2DX/JVzIf4729+fng2D16dGy+v2s/7XqWds2nXQ/SrlbadSPt2ky7KmjX5bQrSrt8tCub dum06xk6Gazoonr/ecky3U27XqFdT9GuTtoVoV1h2pVLuwK0VB9gwd6LoHWIakXUV8WVjgX7LpwK 62NnQXA0CJkPwiYcQHgIkBApHUSBnCSxJ5vHOX35lcn0BeVFa6pmsoOoeBDLcJAcAchYoIMQo4No 5CCasyOsBCwBDAJOABIABdQ5mMdOEdoRFgIqAUsAmwAnAIoYzgkMhZE1CPkQ94iBFSKsBMzhKXYQ bw7eIAvqWZpPi2ozpZ0+as+mc7IT2ayUZGTALjsdRscAte79wvrlF1ZiqjKx29lOkoWF2DUS7+z9 Kss/QH/WG3nGX5VOf0qyZUgdLSMRGkY8mXSK9ETiM/L8EuJjTyIu6vUtQjV7b6TAv5/aeK29/q98 x/yf+AYY0OO+Z/x/CgzItNf/R+Q8udf/lm+7/+XCASNyno0MUET7A4J0n2+y/6lXBOlmFNzT67+R R3v9N/hm+Ff6REFrsuDyTqR0u39+ZLF/Jtqr8S31651oc6+/0ne5vyJJNZHX2esfjyFEk2g+BjvW JzoNZYsG60sH6Aq9QL1LbVDnqJPUIrVADap+NUv1qmlGp1Ez2owWo9loNCpG2ciMxJg2kDiqR/mu l6aIzU+BQFMiC1yDhaHczCAkjBoZuZjEU6U6VrdgGq2LDy4jdUsD8dMLQgPUPG9x3BCaRuPOOlK3 cFp8crRuQE3Mj5dG6+Lq3Msaeii9PYbcONs2QMnChgGa4FlbvHFndcM+Qqljy21eHo/ZclssRtwZ V1e6K51THWXTa74naBaZzTXRbx/3t2jUHc2K31W3oCH+RFYsXsSRRFasLv7jBYHGhn30c3qytmYf /RePYg37pKn089r5PF+aWhOL1Q3QRYKOBOi/QAeJQQQ6IzZmTkcCxuwk3T1JujDqgy6XR6AzmUhY 0IVNJkEnU07X05lbW9OTiwA0rgDpFDSdrsB3aV4JgyaMADQZXeQVQfNKRheniU8Vzfh8IMlGABKa SXyCxEczBYkYeY8gKRwh2X6OZLvoSUqORtDwAM1Yj47SWI+C5juM/N/R1mnRKO2bElvWWNsaqm0O 1bYCmuO3Xr3CHe9aGgj0LIvxgkBcijQvXbaCxy2t8ViotSa+LFQT6Jki6v1HcSMvnhKq6SGNtQsb ehr11preKfqU2lBLTaxvxtyS0vP62n6ur5K539PXXN5YCe9rhqj3H32V8uIZvK9S3lcp72uGPkP0 RYSMz23oMZJpsWqsH4/7WIoZ8trsDcamZWhrpwrhnRJ03+jdD2/lMZISjcUtoWlxK4DL9biqcVW8 CDrFi2zIto8UuW+cEvTup4+NFGnIdoSmkej6DZ0biLu2rSb568SDrPUb+FIkwyjP+94HJLVxvaWG +9Z18fwFdfHKeYsbelQVuc01MeSVj+alpNQOJAaTmRcgs5wTStI5Qp5XwfNMphHC/1sWxJiQDe7s g6PxTB/Vs+l60hmT4tl1CxlMwcLFYEPj4ob98KX4JtEZwwQ7aZR2jrbG5yFwkswhmHbnKKzfMIKN 8GL9SCxIO6Mk2jnKktHmopxZIhC8Wh+FaTPsJx5ApuFR4pEjxE1I4u+A4zwebksc5+U8Zp/C0A2M ACGPkadoG3mKHCDP05OotYfsI/2Eu0A15D6ykfyEbMW2thg528l8vAbk/4R6Ev04mTyADfMB8jpo LyU3kv0kg7oTn5BNZIv0JmptIVaSQ6rIXLKG3EYvSWwgjeSIfBMpJZeQq8ha2pVoSNyeuCPxMPkl 2Sf9d+IsSSGZZBne1xOfGf6ceI+MQ407yd3kCL3D9DTR0UsXKH9O1pF7pCaZJq5MnMEIguQajEEm s8jrdJBF0Xor+Tt1041SNVp5KBFPvAAqH2kiK8g9ZD+dSGewoKExMSvxOslAH9ei1btJL9mLd4A8 R96lFsPJxMOJk8RDCshFmE8/+T0dlIbPbh6uBN8M4NJYUoaSNeS/yO/IYRqiv2VrDBZDkUE3/CDx FkkjE0g9Rvsoav6NfsFuxLtJekmenpiG090W8iPObfIi+ZBm0kI6hy5iY9ka9gtpHTGixwl4l5M2 8PtnaP0DiNFeZmGHpIfkJ+WvlazhowkbViRC7iU/J7+lVsw0QDvpD+nb9GNWzZawe9lH0k/kx+U3 1BbM+nKymtxGniRfUCedTOfRy+gKupFupT+id9PX6WF6nFWxhWwlOyGtkDqk5+RpeBfInfJNhlsM tyrHhxuGXxj+w/AXiaLELWQe5GEzRn8n+QVmto8cIu/gPUI+ogaaQm14AzRI6+n1eG+kt9EH6WP0 cdqPXg7Tj+gn2JL+Tb9m2GmZwrxwfrgLFGLr4GH+hN3HDuE9zP7JvpJcUo4UlSZKFVJMWoNRbZV2 4X1a+lDOlA/JCfC5yHCXYbfhMcOThucNJxWL+kPs8a9989DZ/LMfDJPhbcN3DfcO9yc+JOlYQ+we OIJVYPQteNux3ndB4vaQN6kFvMuk+XQqvQScWULbaQe9Fpy8md5DfynG/mv6LLj0J3oCY7Yynxjz BWwim8bm4L2ctbIOOGN3sH72NjsjqVKKZJfSpXxphtQktUrrpeuku6S49Jr0vvSRdFr6Bm9CNst+ OUeOyFF5hrxE3iD/Qv67/HdDo+FVw18Vs7JauUUZUP4Fr2aqOledpzapO9W96lvGZkjnQfI0+Q0k 8NxDj0qbpVrpaXI7K5Y9OML8HvK8hCyXZjFIKnuMbmM30H6Wa7hWmcKm0NnkpBwBr19iu9lpNkWa RevoAtLOJiQbVNLkJ4BVyAfJkPws5vZ7tHytYqE3shOKhfTCRyqDj/SiNF6OSq+Sd6UjVJUfIH+R zdRFh9ij0lxIwXPyVEMDCUr3kV9LHfQG8jSrJcT8tXEH5Hg2fQJ2YSEtol9KCbjBsyFFpdLH5Cay kv2ZDEGPt5Gf0uXyleR2Ukw3kr+TR6AVYw1XKflKOn2ZtcndLJX2EyY/jtmV0VwqGdLIzbRJukc5 wd4hG8gh2Uw+kH6F0R9iv5ZmyScN8+kKaMAN5BbSkdhMrjM0yG/QK4lEF5GwfBTWbaNUJAcRb4JV aYRN2wvt3g87UCXNQo4bknMJ5KIeFuIevD+DnZAhQW3Q8UthxX5P+pWFbIBcabBRWB3c1Lw6PJ8s TjxC7k5cSa5K3EHGwR5sTWxEi4+Rv5Kd5DG6Zfh6shZHyXeg25cYprNDhumJcaybvcMWsLvOX19w O0zd5FO8v8bKTDU8Q7rlP5EFpDKxI/FHSPcYWNi7yVI4rMcwy8/Qw0xpkBQPz2Y9ienSWsz3CJmX eDThp2ayIrGKzCHPkl+qBtKiRrHGcfoG5ns9aWXzE+ul1uE28GEnuKCDWxtgf7br1fULq/TKqRdW TCkvm1w6saS4aML4wgvGFUTzx47Ji4RzQznBgD87y+fN9LhdGelpqU6HZrdZLSlmk1FVDLLEKCmo DU1vDsQjzXE5Epo5cxxPh1qQ0fKdjOZ4AFnTz6eJB3i9FhSdR6mD8or/oNSTlPo5SqoFKkjFuIJA bSgQf70mFBigi+c1AL+tJhQLxIcEPkvguwRuBR4MokKg1r2iJhCnzYHa+PSrV3TXNteMK6A9Kebq UHWreVwB6TGnAE0BFneF1vZQ11QqEOaqLe9hxGjFFOOZoZrauCeEqmhGCte2LI/PnddQW+MNBmPj CuK0elloaZxwTykqSEi16CauVMdV0U2gDT5OnNwa6CkY7N4xoJGlzVHL8tDylsaGuNSCNmrjjij6 rYm7fnDM/W0SjcMn2/rdUq/UXetuC3Di7u6tgfj98xq+U9cb5C3EYmgDdVl4enP3dHS9AytVx33x ONsSa4jTLegSjmVYzCo5v6TXG25uD8RNoWmhFd3tzViazO44mX9dsDczU9+XOEoyawPdCxtCwXil NxRrqfH1pJHu+df1efSA5/yScQU9miPJ2B6bfQSxWL+LtILpyTKBCXKO1c0/x1nKxxi6CJ5gPLAs gJE0hDCnyTxonUy6l03GAuCJUdSKL8eKtMVN1c3dWjnPxxRp3BDWQoHufxNIQGjon+fntIzkKGHt 34QXcjk5J2px2jKKx6PReH4+FxG1GmuKMU4V6YnjCq4eYKHQWg3nZ35oIHPB25ZYeSHYHwzyBb51 QCdLkYh3zWtIpgNkqbeX6IXwrVkzLxkcLUmv5yVdoyXnqjeHIMn9/DxL0uPGyLmfXctIrV1RHqcZ /0txa7K8bkGoDq5xoLa7eURq6xael0qWc4aCbygbweKp1Q2SlyGPY8wridKkhzxKAne5wRKXw/gp QqiXD6hGSKXIoYHpca15ZjKMmYPBEZ35f1UaSJzktUT0bbWRacTLoyMDTQ47PuW89HnDs3RLdQth chg8++5u83llELXkKC8aiSDxOOgHA9VxUg/NDOOHI8dkDjFvXAfLULIQWiSyY96R5HmE3pFKMTxc OscVTIfN7O6eHgpM727ubhlIdC0NBbRQ9z72PHu+e20trF1ScAYS+2/1xqfviIFjK2g51IORaT0h um1ej063LVjcsA9XHIFtCxt6GWXVzdNifFlY9cKGEbaIBeGijzXEFxNIDN/j5U5yMU5scxHXIZ1K f0e2Ib0N8U3IqwFsZU+QLcirRHkuxIwKYSP4MqNgV0ef8IC5+P3/ffiljgSQ4bXzB/fC+K6S/Ghi ImakUjBiKy/Ck07S4es0SKXwNC8zPKqEVUU9blxprjT/AaXYyrjzj4YktEGCjqAjjAC3SOSbgDT4 jW4gX5OAPAhKcnHiuOyTp2J/LmXj9AKT1ZTvsWbmj7Xm55dZJ6WXesvzL8pvsjblt1vb8pvHd1tv GXtPxr2Zj1vTxwwkjvenpCj1eUB0D8ce8TwxZq/nmTEveA6NeSP9/THGmgyaPZA4pTssFqXe6eSh wcLDifyaq55jfpffHS3ILymTywoukmcWLDLGolcY26JXW7ZaXrZ8Zf0q6igtsVFZK8wtcRUF09xL xq4Zy8b6Cm2Vtp223baEzbDbtsd2wibZLHY7q7cNJD7t1zSBnNLTNU2pt1msVoSK3Y7Q5pNcA+yJ ve4703w+lfDRZfJxkNo8c5FPShnborUQxWpl9SQczB1I/FM0xhE9hefmyhYLwoHEMUxeIKcEF5Dz np7Cu8sVHSH9TT9vOHeAXabb8nQS0SKByPjInoihDDrXb7Ox+shA4u29ApnA83RrdqhkfNlgGbu/ jJa5+NiqeIuusDunMPeAckhhfqVSYYqNz1TBtBC6+XgUCx8MD5V6xcanq2i8c2XCZO30yGG8qWPo 1FBUa+qIIqdp6NToGb3ibPSvfyWVQ5XHopVDZ485nGWFogj0HUjgV0YdTlfZhPGkiRfQDkSkI6wo oZzIxJJJk0rFO7EkLxLKUdS8qay4KCMDTlZ6WoYrFJEU1caAFheBDEel5fva9zw7o3PmxJXvXkmL a7dtui4r7r7q8PZtT8zVTK6cZ32upS+saSxa3bbiwUjWTfXTn9wye/PsNJs1MzdsvmrchbEOd8et dXrLxRdce/LrLRdOpu+P8WljZhXObL5szoXXQJPnJo5LQ5DoTLq4h3EDopfYNtmpPYXqOOGvhU7I Tl+K6vbJOAmmq0bOflWwUoVoAtc4K1XBg9ffeokzZkh7oamIw4TxXn2GyUL9vurUateC1AWu5tRm 173sXuke68Paw5kWo9VjbmdtUrthg2Wttcv6iOVp017z0xZLhuUWy8dMsuUssa+xb7JLdgoh1K8b Dxd2LmnGsHaR+8lRchKfQu32FNiB0TH6MPRcm1FIcI4X88tNifrxZY5SqvPVpjqXEDqTrzbN5GT0 Il967iGV+tVKlak2TqSaOZEqFFCd4C15wR0VYoElTopH07oRM4srX74fDK07FR1aJ+YOcXCUFWpN x/DjItBBmzpi1MVXnzhKnJOw2C41wpc+uchSRU/WiV+/O/zFuk+2P/Wef49n0+JtTzx8c/vtdIvr N4doFjX/irLNex7wrlx18M23n/8ht0J1sELZWLN0kkUX6C4/8aWzeqnJ0GSqT2mVVhrWmFpTjOlc 5/gEHUD0+RzL8vEwz/mO4Uza6Ux5grPcM8FX5ZyVWeWb52z0zPe1OFdntviuVa5NP81OuzVc/Nit LtfcjOaMtfgE6rPv0u7XmKbJXp9ZJfuxU2D2QuEpV0eNc00Dp+9MhbS4dCt0HCNg9UA+E0MB8qnQ ciCDuikvvyRupdZMP1J94UgJj/UqrtR+6s8o1nJVPTe/hC/MHFVSA0LQhP6qmAiELrlYPr6Wagaf murJLil1R2drI9raFJ119thsrSMaPQ2IRmcNYY3OQh+PVQ5hoZoqznZUQFnLypxQV9okFJZ2rBtd Lo0UFxFHmhrM4OpIgxGhstLl+ws+2/fJ8Ama9t4fcTfyzXFz75ZlO86+y+ZZJi/avvFxusj1UD/1 UwkXEWOGPxj+Sgvs2b+C3nlL9YpHIJGpWMIuw5vERa16dpqJ2j2FnvEe3bPWc6/lPuvjVmOmdYw1 7hn0yB7OjzGZ/pIso1Wy2H1mms6iaakyvjabd6fRtESqLrvCMk7cd0DEORMnTC7hsR71+Ut2EerR ueXz6FYoLUnjZp6M4Tkkh6sxKeBcBP3negpXZZLGhR9pviMI5G/9fBNAzpnfcCaTh9yeZ+l+EiSn ce7EJ4tRW8l5C65WaKcqtIohbWioiVRWVlRUnK2oHCpzgLfV1+lpmkMxqYoR9lgzOb3Eodi9uDuI 5m/eTKMdTWRdsSM0sXhiSekk6AhUhGtIenF6yNG7e3dq5k1XX9LonVw0v+bQIemeHR0rS6Zf6vy5 eXrz0h3fXAGN2AZHpQI3oXwHf12/3DSJy8oc0y7T/aa4adB0xHTSpBKT37TW1GXaPZJ11JQwmf0m 2AZVZpJJkW6kRDEosllRwwYi75bvl+PyoHxUVgblkzIjckA+jJQsDyS+FEIP5Izu4pyUxS4nm3mv suAiypJcBDIsZB/IN7qZs1GebZwxd8SkcM5FmzrWgVWcZZVDQhy5SHKBXNcRTZ1YnC45ih3b+vv7 5X8cOvR1uhz5+l3ul2HO0peYcwpr0b18d4P8K4uUxSbJbv0/htOKZLLwFVewL0LnWL15FDGNIhJ3 RTResV66xsycSiA1WGKEn93nzCsB1cl+xE6DyAiKDP1m5CiybJCVUtMM2RBWxpkbzNdIG8zvSh8r 6iMKDSkRNWwsUyabKq1zrDE5pjSoMdMN8nWGu00vKW/IbyvHlE/UL5SvjOlOs9kgSTJTFNVkMiJh MhrhmqWpqiLJcthgTjMYzGYsjGykYL9BUY3GlBRilvGJUzcZwHP4RzlGngoGhPUWm7iauQtGJSVM WJhSKEElbkcYZHhYnyCkHhOGrIMUoVgx4hRyL8w9TvlgB/FYrB8GZ1wBEZ89uvE3dVTM0iDbMCen o02z4BDAmEDctQosW4XDVbbVcEFUvkF7AbE7agOiasYKY4Ukwh5F7K3WOhP1m26WmMltdZQQrHwM 64xDk242FWSVmYxZWRVYsA96s8oQvdUbEFFPsEzISayJdDTRDnxrEN8slMRgb7AMizjYm8GjD3o1 Ts4jkbKIqCclWTkao2iEd+V8X6bGtAz0lpZWIQLUOt3r5pX/2eNNktOmWNJ/iWIDixZDAmkxxY24 CkmkT3wy3E4PfDD8wCbD/m+epfHhq88uZ/4fDF/G5fImBKVCFz/eaxCKCAka7CudXCLikonJePyE ZJwTFrEeTneV2A1+XP8eMchzEJw0SH7DWkOXIWHA125iZlI4aeTQEpZzUE+HtdhN6CDcAJwHAuQw PAIZJUnt5FZLzxIrLrRzZK2TNs44YuCSqgnShNiWgIzoKJktn6+jXElh4YSactXkKf5wztzUb9h/ ZjrfmWuwM+dhZ7YSD125N93NhSmVe/4csXN16+SYRxQ4VbPHMkOZaVykxIxXKm1GY4lW7izPmOiu 1eqcdRm17kZDo2m+1uRsypjvXm1YbVqurXauzljuvoammxSD9TJpoWGh+TLLKqnV0GpeZTG7fLLq 8KWkpOV6dS7lXvg82Cdhi3QHt/sq+kWYdNm4fgtvhyN8VKA7KQy+QMSuytmcmhsuGa9SompqANvw hCNe6uX5F/GNGrgtl1hsfFNxCsUSXiHBIJAjNmgiHGwCYwTFyhD7iI4m/aQSDJuQyTds7bRwlQVD h7BdN51u+jaD6xmcSjAdu0QHF2DTAsMC01LDUpPMpZTXStVKsWuQ9DThZ6UK93liCd9Dah7e/uJf aMb1/7j1yPDQvt6tt/T2bdnai+vmvNuvHv7w7Ov/+CHNptbXXn3tDy+++goGtHW4TQ5iBZ0kmy7V b7do47QLtTpNrgzEA8wfGGsJZRWlF2VNy1ob2BUwlrvKvRe7LvbGjJdZGl2N3nbjSkubttq10jsY eDPtfff7mW9mH0s7ln00kAhkhOSoFk2fKJdr0+WLtcXaX1P+kTWspThscK18ikqVDJ8thdg8uYfN VDPr5mZzl1kOiCUMiOWEBf+bnsIX0iwWEukzo6b9M7GWwsbzPRvIcT3EmW1eT1OLWbEzTMgg7CG9 n8bpSSr7aSW+UElw3L5JagkVdpEKu0iFhFBxPgLFaeEPCNIM3jgVh1KK9Qbu8c8oddNvfS6sBt/P ZmlnTx3TzoolFVlYRSxj5ZBDOFukiasQ6Uh1FKfzDT8Dl8wMPnIkzyF9Z/W2Plx+x4pth9s3HLl+ 8c4LHI9cfe2Tj67v7BluMzzXPW/ejsTPHhr++tZLys9+LT38+guv/vHVV/7ELdAWqOJLWEMHeVmf UphKNZmG5BK5Gh/ArpDXy4rJYTQZTdZUh8lKJCNNEcwnZtOYXUZqzAmk0lSW40gaG10rnlRyEm1+ n335Und8x74oQuT5OXbEuJzSBbeIImQ+ubHMds4YPU0I1sC2VpzFWeHUumPgD+cO/CV+gCwrI9rL W203vMAPEetoE+yMYFTSM1KVdMeWB6e2VV52+dRp06ZcnpYtRx7omFn+aN6MyuZ1Z9/iXKjEya4H XBgvufTr5Zy0nHLTxaaa3EU5rTkbTbebbs59JPXJguclq8mV6XaNryt422XwsnrGtCJqdjcaG02N 5saURkujtd3Ybmo3t6e0W9qt/ZH+PHteJDcvd+yk3MXmWMryyPIx60Prc7tyf2y+z3LHmJ8W3Dn+ YfPjlofyHsa/Tl+MZODi45QwLzmjSGgUyR1FBA3nkqDhiKDhiKDhSBZ2Nt2ZXbbYmBe2mOXMQCRd TrkgK5MfDHM8BZz5fk+lZ45niWeP55BHsXv8njWeIx7Z79npYZ7nYI7SIRfizKJjF2A4quiUafiG yQjVKONnmL60jBIe65rNUULpBY1Zq7JYli9dhffGLy7g3kEHxQJzRE/l6iD7LkjxZ9LMXI+e6i4p 4tULhd/tTobcLHpg+BAGuGh4AryWRzgrHnFu8eDSo1fNzUfVp31lh/MpsL8JzQZyHDaZ1QuE8wHI p3u55uVniq6COEU1Fw0WscqiriJWxM9fuUSYByJcOxJIcpnVC4QPgCO6hw8ikGsXqm4Xw7MH+FCx TZ3RMURgNt6hXbiW9pwjo66UZ8LIIaupY9aIdzQEYdaiOAPPHrkwiUY7cNoa9Z14CYw4iCqHOsSF Cd8/YR5EBGHHb+TSBCcFPW9cdsiQVhBxaE4tVZOUHGvAS0xjVC81jEOQnYZk0BbykpyQ1WIca/bS MXkmsxKVvcSvZXkptgTulyUD7vVE86ObN28e+Y8GkvxUvq6jKbVUHOz4VUwevv6W4MZFmKJzB3QX 7mVc2biO4dtKpLLXvv36jddODP/4pbvnVE3O/9GCG55b7IhbOts2tmdkFHpvPvDTRW0v3XDoHXqh b+W61poLQ+5w0UWbZ8+4bow/OvP6K93zG+eXhnxZqebc4qqNjYt3X/orrqe5ic9ZvuFu/Nviz/uI GTIYinBfGydhIF0enE8sVjOVSIZmitrN2CSkFLuWQ3Ko1Rm20IRqrDXVNqtr1S51lyoT7NH3q3F1 UD2MS05+7uZnACCnhBQB+VxcpQH5Elopir4UkoacM0KwgZzkuww/TvD1Rzrpv6j7WTtx00k98IvP XZNxZsLQD8Ez0o6dwiFmqBIoN/KO4mLtZe4qRaNhF+dfZCI/4TlKYclCjjTOeqZlXlKxdFXBzTf3 Pf10anRM9gO7tamtD7JlO6i6avi2HWd/PKsgE7ez2JfFk8jDN/7ve2zIlPAPlulkJv6/cQmZDZ9/ Pr6n1uNf9peKChQ7Oqw4HgX/1SD1CxfVz7w4WrWurWXVrIX/AydCdoYKZW5kc3RyZWFtCmVuZG9i ago0NiAwIG9iago4NzQ5CmVuZG9iagoxNiAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAv VHJ1ZVR5cGUgL0Jhc2VGb250IC9OS01DSkgrQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNDcgMCBS IC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDUyIC9MYXN0Q2hhciA1MiAv V2lkdGhzIFsgNTU2Cl0gPj4KZW5kb2JqCjQ3IDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRv ciAvRm9udE5hbWUgL05LTUNKSCtBcmlhbE1UIC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMy NSAyMDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0Nh cEhlaWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQg L0F2Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUyIDQ4IDAgUiA+PgplbmRvYmoK NDggMCBvYmoKPDwgL0xlbmd0aCA0OSAwIFIgL0xlbmd0aDEgNzAwNCAvRmlsdGVyIC9GbGF0ZURl Y29kZSA+PgpzdHJlYW0KeAGFWQt8VMXVPzNz95Fkl2wi5LnJ3s2SRbKJkQANIWmyeWxAAxIg6C4N ZgNEAoJEA76FpYrggmKpUsUHamtF/ZSbTaQbsCWKj4oifEq1PsFHq/ZXBP1VrQ9yv//cXVCsv353 8j/nzDlnHvfMuXPnblZctrKLbBQhQf4Fyzp7yLhyF4LlLbh8hZqo23OIzG0X9SxalqifsZTI9Omi pVddlKjnridi0e6uTtlOXt8BP+uGwqgRmwA+unvZiisT9dx2cOvS5QuS9twI6uZlnVcmx6e3UVcv 6VzWBY4rvwZE7Vneu8KoUv5I8DN7LutK+rMgUdqfvo2/9Rj7B3v+7oQTpgSB0+dUQ/eQBZKDyul8 IqVbKSAT6tJu4ps2f77xy470mi+s+Vaj5QMfjCmRwl6l59ZvdpxY5CCrDdUUw18a0M5SO3weNTro mx3fXO1IjCQtJy8+SG3izH5vjuvgk2IsHQG4GBvzFbgGxRhREKt2+ePC0585qiK9vkyo6LHcoCro cmAHsAdQqEMUwuoAXQ1EgB3AHuAgYCYClVYVWA5sA44AZlEgnDHV5agfI3LRNhf3my6y6RigA4Jc oOXADKAD2ARsA8yGn9QsB1YDe4DjgJn8Iju2eTzmnh3bYLD+JUsrjGpnoto+z6j2XxBK8OkzE7zp nITb5ITbuAkJ9VkNCT6mNMEziysi6Lw/1V4xVJ8lsnCTWZh4Dyjjz1A6Y+Si+8Qo0gAuMFVD4xeZ /aO9Fdv2CIWY4ILRQnLpQ4LF7BkV9alc58cok1z8U340YeFH+0dkVGyrP5e/TzuAPYDg76O8x9+j 1fyIjDloHbAN2AMcAI4BZn4E5TDKu/xdSufvUDlQB3QA24A9wDHAwt8BdfC3ZcYYVMp1AOdvgzr4 W7itt0DT+ZuQ3uRv6kP81VhlVcWgIfjKk4KrOClk5yeFzKyKOH8l9vVYZJQXK42M2i2KqJbGi6JY 8ThXXOTEaha74vyDftXnuq/+bH6INIBjJocw8iFSgVYgDPQAZkivQXqNIsCtwH2ABiDLQB2AyvcB LwGv0dmAH2gFrPxgDMPE+YGYt8FVn8Vf5s9TNiK+n//Z4C/x5wz+In/W4C+AF8K+jz8XK3RRfRrs hDYOcAd4Oewm/lT/6EyXXp/B9yCCLtByoA6YAXQAmwAz38OLYgtdmehkN+3DM+ziMfrE4L+nB6zk X+LyexuRgKok3sk/hwSyTd3m5X7vljtRlcR7y2ZIknhv2AhJEu/VayBJ4l16OSRJvAuXQJLEO7cD kiTeGW2QQOL83j+MHuOqnHExU+vT+RWI0hWI0hWI0hWk8Ctkoa8VOce7YiUliNhWv29siSuyi0We ZJFZLPIAi3SxyCoWWcMiNSxyIYv4WMTJIoUs4meR3WwSQhFh/oHTqlX+HBbZxyKPsUgvi3hZpJhF RrOIyir9ce6OnYOnDixgsP56+dBxd//Pa7H7pHM3IupGzruxJ+wBPQDoRs0PJ7Uo4ZxbKHlRf0ld on7W5Irl9VP5XjTci2XYS4cBBQu0F2m0F53sRXfpoHVABzAEHAN0wAzvItzHJoOmg5YDdUAHsBo4 BpiN6RzDVDgtB5VT3GFMrBy0Dpgha3wvShGKm7v9BQ6nw+eYKjY5WXohm1GoF/JKysrCvpyZYc2I M/vOr+z//spOKfUp/Ba+iQqwELcm+abY1wWuOLsj5t3tqh/FfkOFCrKOVZGXFYNPol6jPpGcVqmf QE7+KHhFzHk+mqXHvKWuXWyEbLXT9bXzQ9cnzjiH+LFzt+t1Na6wmOsv0Dy603XIeZPrhfK4FZon vXEGtks1XAedk1yP7TNc18CwNeZaJdlO13XOKa6LnYahK2G4sBc1f7prlneuayr6a3LOd/l70edO V53zQldNwmuibLPTdTam4EuIJZjsWKcxqKfQ6HBOZZx1+0stWyxBywzLzywVllKL2+KyFFjyLSOt mVaHdYTVZk21Wq1mq2LlVrKOjOtH/D751htpNl5+ZiQ0I8WQHdhhmNxmQIkzK6dzSTtDtPCW2Q2s RRtaQC3zVe3L2Z44S505VzN5GpiW2UItbQ3aJF9L3KLP0ip9LZql9RfBPsZuCUGr8fVxRm3BONOl am2+ltkYHCTGMtbenC/5mWtvDoUoJ+vyupy6zNqMquamnyBhQxlu8n1/5Xwv+nJ8BdqWltlB7ZGC kFYhBb0g1KL9erbaHhxkn7PjgaZB9plkoeCgqGWfB2ZJvahtCoVa4ux8w49U9hn8kDFg8LPixSz9 SLUWJvy2JvyK0R5+oyWDX0oKFRt+xSkphp/CpF9f7+hAU99oEPhkq9Rr+PRmqz/02VcMn2IQ+GRF aJ/hsy8rIn20WqMbpxMuhSBwYXnkNFycLM9wMWbeZ7iUJ11uOuVykzGSSMzG8JEE3diPnPSxH4HP DwL538WuBp+P9VeHFrQHujyBsCfQBYS1DZd352iR+aratyAkDaomvOH5C7ol7+zSQp6uJm2Bp0nt qzba/cjcLs3VnqY+ag+0Bfva/V1NsWp/dcDT2RTqn9I6ofK0sW46NdaE1p8Yq1V2NkGONcVo96Ox KqV5ihyrUo5VKcea4p9ijEVGjrcG+6zUEGrE+knez9NSka/hfHeoIcvRU2skb7U7Z1X+LpxWtlOa L6TZPA2aHZB5XVZfVi9NeKakaQTU6UlTzqpqd/4utj1pckCd4Wkg34qVvSspJ7C4KfHXiwuqFSvl UiSoT+p+8oJLQPN3NsmzdYtWMrtFq5s5N9hnsUAbbgpBN/mkLi0tENeHEsqzoJwsHYU45Sh1NVKX kpJ0/M9cMOYENaIziIPG7n7mL2QrqDcktMKWNo6toG0uwtA+N7gLZyn5kugN4QZ7mY/1nuxN3och U0JDuO3ek1ixMiklY7EiyQ3XXh/5ek+G5GR3PhksgxixWuHD1mbaRblAnukhylW8hO8f/SPgY8mH F+sfS7vk/B/Y6OJJEG2nx9hieoz20NPsOFrtoEEaIHkEaqK76Vq6jdbhtTYXmptoFooJ+ttYrj6A L5P78cK8n/bD9wJaRbsoi+Xon9BqWiteRau1ZKciqqdWWk43s2n6Smqnw8r1VEnT6BLqYRE9qN+i b9Z/Rw/SoPizfoLSKI8WoOzXPzX9VX+bytDidrqTDrPNKU+QH6NE4HkPXUZbxTyF6Yv0bzADN12B OSg0nfazIe5D7130Ecth14pG9PJbXdOfgZeT5lE3baVdbCKbwt2mdn26vp+yMMaV6PVOitFOlDj9 kd5kNtNx/Xf6ccqlUjoH9zNAL7MhMXxizXAd4mZClMZSFSzL6U/0PB1kHvYUX26ymSpMftPV+iEa SeNoDmb7EFr+nX3FV6GsFs8pzXoDjUBcfiWjTc/SeyyPlbMZ7Hw+li/n94rLyIoRx6EspMWI9x3o /V2k0U5u4wfEb5VHlW/NBcNH9BFYES/dRffQU8yOO1VZL/sle419wBt5B7+Lvy9uUx5WXrF04q4v pGV0Mz1KX7FMNonNZL9g3exato79it3J9rOD7GNez9v4xfyY6BaXij8qDSizlV7letONpg3mj4eD w88M/+/wV3qFfiPNRD6swexvp3txZ4N0gN5AOUzvMxNLYyNQVOZmc9g1KKvYzewBtp09zAYwykH2 PvsEr6Qv2Lccb1pu5vk4/MgjkIdfhhPmbfxufgDlIP8n/1pkiyLhExNFjQiJ5ZjVOnEryhPiPSVP OaDoiHOFaYtpm2m76VHT06bjZpvll3jHv/Tdb0+UnHh3mIbXD28Zjg0P6O/RKKwh3h74BKvB7DtR lmC9tyDjdtCrzIbY5bESVsumITIdbAm7lF2JSN7AtrIHjbk/zp5ElF5nxzBnO3cacz6LT+QNfAbK hbyLX4rD2GY+wF/j3wiLSBPpYpQoEVPEPNElVoirxBahiZfEO+J98aX4DkVXUhWXUqR4FZ8yRelQ Vir3Kh8pH5naTS+a/mZONS8z32iOmz/DqabW0mqZaZln2WTZaTlkDSM799IT9Adk4KmLHRFrREA8 Qbfw8UouPmFeRj530EIxnSNT+Xa2nl/HBvho05Xmal7NzqPjihexfo5v41/yajGdtbDZtISPS3Ro Hqk8AqlG2UtHlSdxby+j5yvNNraKHzPbKIYzUhXOSM+KsxWfeJHeFIeZRbmf3lJSWTY7yh8SrciC Pyq1piC5xd30uLiUXUdP8ABR6rfWjcjj89gj2BfaWAX7t9BxDD4PWVQpPqDr6WL+VzqK53g9/YYt VBbRLTSeXUsf0e/xVIw1XWIuMY9iL/DFSpSfwQaIKw/j7qrYaCZMI+kGNk9sNR/jb9BKOqCk0rvi fzD7A/xxMV05bprFuvEEXEc30qX6GrrKFFReYYtIsPOpWDmC3e1aUaG4wVdjV2nHnrYTT/cu7AP1 Yjo0OcicaciLOdghtqLcgX1CQQYtxjN+AXaxl2nA3MbjtMg0gmHXwS81Lw7Porn67+lOfRFdom+m MuwH6/Rr0eN2+httou1s7fA11INPyTfwbE8zNfMDpma9jEf5G3w233L6+iLaxSyH/oHyOFam1rSb osrrNJvq9I36X5DdZ2KHvZPm48D6Ie7yU4wwVQzR+OHzeJ/eLHpwv4dppv6Q7mKp1K0vpRn0JD1o MVGnxYc11tgruN9rqIvP0leIruHFiMMmRMGPaK3E/nOTv3FOW72/rvbnNdWTqyZVTpwwvmLc2eVn lZX6SsaeOcZbPNpT5FZdhQXO/LzcnOysUSPPyMxwpI+w29JSU6wWs0kRnFFpwNMcVjVvWFO8nqlT y2Td0wlF5w8UYU2Fqvl0H02V7TphOs3TD8+LfuTpT3j6T3kyh1pDNWWlasCjavubPGqczZ0ZhHxz kyekakcNeboh32rIdshuNxqogZzuJlVjYTWgNV/eHQ2Em8pKWV9aaqOnsSu1rJT6UtMgpkHSsj09 fSy7lhkCzw5M7uNkteMWtTxPU0DL9aApuhHFgc6FWuvMYKAp3+0OlZVqrHGBZ75G8qTkM1yo0RhG MzdqFmMYdTHOOBptUPtKh6Ib4w6aH/bZFnoWdrYHNdGJPgJahg/jNmnZV3+Y830VneNMtu6H1nwR DeQsVqVzNLpO1e6bGfxB23y37CEUQh9oy4ubw9FmDL0RK9Uiz+IaXxsKamwthsTBsti4q8T9JU69 xeElqpbiafB0R5eEsTR5UY1mXeWO5eX5B/UjlBdQo21Bj1ury/eEOpucfSMpOuuq/ly/mnu6pay0 z5GRCGzfiPSkYLP/UOhC0BM2QzLcpdQy61RkmZyj5xycBDV1gYqZBD24p0mSdE2i6IJJWABcIYZW 2kKsyGItpTEcdUyWetwi00zFDo8a/YKQAZ6j/zxd05nUmIsdX5A0yjw5lWoa6zwpaz6fVlIiU8TS iDXFHGuN+sSy0svj3OPpceD7WX40UCti2xmaXI7wu91ygTfE/TQfFS0yM5ioqzQ/P0b+cpyteVha hk5aRs2RlshJy6nmYQ8yeUB+z9Iozeo99ZfuyDoj0D1ZY1n/xdyVsLfM9rTgaKwGouFk1ra0nVZL 2GVAETfYkpJ2RmNQ5HPopMTzhWFNnJBPuuC4HLRpSjH+zEZSL4xbrMhKQ8PUZs0RnpqgoVS3O/nM /H+N4vpx2cpg3zdL3oY22ZecaGLaWvVp9dOmZ4uKljZsORwn+2g09TQbUi0xy3OSDBmPD3232qjR HDyZxfjDJ8ckiVC+5kfIYGnDU2SoQ/nJ6mmO+clGIVwyO8tKm7FnRqPNHrU5Go52xvXIfI/q8EQH +dP86WhPALtdInHi+q4N+VrzxhAi1s0m4/Hg1NDnYetn9vnZ+tlzg4P4iUNd3xaMccYbww0huSy8 sS2YDIuxIDL1sYb4hwYyRunFiYkgyOQh/KfFDBBNOKVBSuHsKn8pATGh4HRtIXJnuDOKQdADfaeK oe/8JvqWVGUInrJHxWHaRalkZ17/zzKDtm7bVtvDthdspmlimv02RWQy/E5jMwuLKTVNWMhms9v3 CWWkEIqwE7fZFYvYjR9LrRj2Pn8qKQpcaF+qEucX/cFkSvUXuCak4keeAYeDz4Hw9wG73RA+hWCG Br8r2i3+Is8ES8Q90XJrOscC+dPsIycQd3CVC9SPGG0gfLhTtuFPjIizjX0bcnyOf/p883yOL32+ f8kvsBrH3x0naqiuzvGvmi9rMqqqWEZmVdW6s3zKdY5n0tPTx53NjG9Fu/5uLLPKHtcP+dPGV4mi siqhFBTUyC5C82gefPwjbf60Kluktcrm91bZipzgZVXSwRdCKCey8RnjR3kyRAbjW07cwO/59XPP DQxPZB0Pip3fnfvg8P1c4befuFiuBYBLH4Oz7U9d0p5tGBhlJlfWjG8ROq9leuO5U331ly3uXDq9 7f8AoDAvcAplbmRzdHJlYW0KZW5kb2JqCjQ5IDAgb2JqCjQ3ODAKZW5kb2JqCjE4IDAgb2JqCjw8 IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVlVHlwZSAvQmFzZUZvbnQgL1FQTkdITStBcmlhbE1U IC9Gb250RGVzY3JpcHRvcgo1MCAwIFIgL0VuY29kaW5nIC9NYWNSb21hbkVuY29kaW5nIC9GaXJz dENoYXIgNTQgL0xhc3RDaGFyIDU0IC9XaWR0aHMgWyA1NTYKXSA+PgplbmRvYmoKNTAgMCBvYmoK PDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250TmFtZSAvUVBOR0hNK0FyaWFsTVQgL0ZsYWdz IDMyIC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAgMTAwNl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2Vu dCA5MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0IDcxNiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMg L1hIZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lkdGggNDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250 RmlsZTIgNTEgMCBSID4+CmVuZG9iago1MSAwIG9iago8PCAvTGVuZ3RoIDUyIDAgUiAvTGVuZ3Ro MSA3MTIwIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4AYVZC3hU1bVee+8zj7zIJEJe M2TOZMggmcRAAMOrySSZCdgIJBDsDCUmIUkJL4kkiA+EoRWFAcFyLVWqYtVW1KucTCKdAC1RfLQo wlVq6xN89Kr9GqF+VeuDnPvvMwNK69d79vxrr73W2q+119lnnzM9q9d0UCqFSZCvbWVrFxlXbjuy vLbretR4OS2HyNz4o64lK+PlS1YQmT5esuKGH8XLuZuJOHV2tMp68voauLwTAqNEbBLyMZ0re66P l3MXIbeuWNWW0OeGUTavbL0+0T+9ibJ6TevKDuS47BqI2rWqu8cokh390aVdqzsS9ixIlPK7r2Jv PM7+yp6/J25ExMBw+oRm0L1kAWejUrqKSOlURpMJZak38R07P9n2WXP6jE+tdqtR84H3xhZJ5ojS dceX+84tsZE1FcUkw14qUM9SMTyHamz05b4vb7TFe5Ka8xcfoEZxaZ8nx3nikBhHpwEuxkW9o50D YqwYHZ3u9MWEuy9zVFl6VYlQ0WKpQVXQVcA+4DCgULPIh9YGugEIA/uAw8AJwEwEKrUqsArYA5wG zGK0cERVp61qrMhF3VzMN11k0xlABwQ5QUuBuUAzsAPYA5gNOylZBWwADgNnATP5RHZ050SMPTu6 1cj6lq0oM4qt8eKiJqPY94NQPJ/dEM/9V8TNpsXNJkyKiy+rjudji+N5ZmFZGI33JaeVDVZliSxM MgsD7wJl/BlKZ4ycdL8YRRrABYZqSHwis2+Mp2zPYaEQE1wwaienPihYNC2jrCqZ6/wMZZKTf8yH 4ho+1Dcio2xP1ff5u7QPOAwI/i7SO/wd2sBPS5+DVgJ7gMPAceAMYOankU4hvc3fpnT+FpUClUAz sAc4DJwBLPwtUBt/U0aMQSVfCXD+JqiNv4FpvQGazl8H9zp/XR/kr0TLp5YNGIy3NME4CxNMtj3B ZGaVxfjL0S/GIaI8WGlE1EFRQBU0URRECyc4YyInOmOpM8bf61O9zvurxvOTpAEcIzmJnk+SCtQD LUAXYAb3KrhXKQzcAdwPaACiDNQGqPwo8CLwKo0HfEA9YOUnougmxo9HPdXOqiz+En+esuHxY/z3 Rv4if87IX+DPGvkfkOdDf5Q/F813UlUK9IQ6NuQ25KXQm/hTfWMynXpVBj8MDzpBS4FKYC7QDOwA zPwwL4i2OzPRyEE6invYyaP0kZH/mh6wkm+Z0+epQQCqknimfQ8cyB51j4f7PLvuRlESz/ad4CTx 3LINnCSeGzeCk8Sz4jpwknjal4GTxLOwGZwknrmN4EBi/L7fjBnrLJ+7nKlV6XwtvLQWXloLL60l ha+Vib5Q5Bh/ES0qgsd2+7zjipzhAyx8iIXnsfADLNzBwutZeCMLz2Dhq1nYy8IOFs5nYR8LH2RT 4Iow8/VfVJzqy2Hhoyz8OAt3s7CHhQtZeAwLq6zcF+Ou6BW465AFjKyvSt503NX3vQrsPuncBY+6 EPMu7AmHQY8DulHywUgtiBvn5su8oK+oMl6+bFrZqqpZ/AgqHsEyHKFTgIIFOoIwOoJGjqC5dNBK oBkYBM4AOmCGdQHmscOg6aClQCXQDGwAzgBmYzhnMBROq0DlEPcZAysFrQTmyhI/glSA5OIu32ib w+a1zRI7HCw9n83N1/N5OWVlYV/OzLBmxFja/s/T/vl5GiVVJfHtfAeNxkLckch3RL8Y7Yyxu6Ke g86qUeznlK8g6thU8rBC5FOo2yhPJodVyieRgz+GvCzquArV0qOeYucBNkLW2u/8wvG+8yNHjIP9 0HHQ+Sc1prCo84+QPLbfedKxxfmH0pgVkkOeGEN2QDVMBxxTnI8fNUw3QrE76lwvs/3Omx0zncsd hqIjrri6GyVfunOeZ6FzFtrzOxY7fd1oc7+z0nG1c0bcarKss985HkPwxtkiDHacw+jUnW80uKA8 xjp9xZZdlqBlruVyS5ml2OKyOC2jLXbLSGum1WYdYU21JlutVrNVsXIrWUfG9NM+r3zqjTQbDz8z ApqRYvA27DBMbjOgxJmV0/dJu0TU8br51axOG2yjusWq9tl8d4wlNyzUTO5qpmXWUV1jtTbFWxez 6PO0cm+dZqn/YbCXse0hSDW+OcaoMRhjuhRtsmuZNcEBYixj0+12mV+66fZQiHKyrqvMqcysyJha 6/8O0mIIW/zeb66cb1hvjne0tqtuflB7dHRIK5OMPjpUp/3XfHVRcIB9ws4G/APs7zILBQdEBfsk ME/KRYU/FKqLsasMO1LZ32GHiEEGOysezNKOVGt+3G533K4Q9WE3RmawS0qiQsOuMCnJsFOYtOvt HhPw944BgU22St2GTXe2+m2bo4WwKQSBTVaYjho2R7PC0karMJpxOGCSDwITlkcOw8TB8gwTY+S9 hklpwmTLBZMtRk8iPhrDRhI0k3b6vE3aadh8y5H/me2o9npZ3/RQ26JAhzvQ4g50AC3a1us6c7Tw YlXtbQtJhaoJT8vitk6Zt3ZoIXeHX2tz+9Xe6Ua9f1Evkurpbn8vLQo0BnsX+Tr80em+6QF3qz/U N7N+UvlFfW250Nek+u/oq142Nkn2NdOo9y99lUv1TNlXueyrXPY10zfT6IuMGK8P9lqpOlSD9ZN5 H09JRry22F2h6ixbV4URvNNdOevtB3Ba2Usp3pCW6q7W0gAZ1yVVJVVShXtKqkZAnJ5Q5ayf7rIf YHsTKhvEGe5q8vas6V5DOYGl/vivGxdEPWvkUsSpV8q+84JJQPO1+uXZuk4rml+nVTYsDPZaLJC2 +EOQTTsvS0kJxPTBuPAyCKdJQyEuGErZDClLSkoY/nssGGOCGN4ZwEHjYB/z5bMe6g4JLb+ukWMr aFwINyxaGDyAs5R8SHSHMMFu5mXd51uT8zB4iksI0+4+j541CS7hi55Ebph2e8nbfd4l55vzSmcZ xPBVjxdbm+kA5QJ5pocpV/EQ3n/0D4APZT68VP9Q6mXO/4qNLpYA0V56nC2lx+kwPc3OotY+GqB+ kkcgP91D6+hOug2PtYWQbKF5SCbI72S5ej/eTH6JB+Yv6Rhsf0Dr6QBlsRz9I9pAm8QrqLWJ0qiA qqieVtHt7Ep9DS2iU8pPqJyupGuoi4X1oL5d36k/RL+iAfF7/RylUB61IR3TPzb9WX+TSlDjZ3Q3 nWI7k54kH3oJw/JeWk27RZPC9CX6lxiBi9ZiDArNpmNskHvRegd9wHLYOlGDVh7UNf0ZWDmoiTpp Nx1gk9lM7jIt0mfrxygLfVyPVu+mKO1HitFv6XWWajqrP6SfpVwqpiswn356iQ2K4XMbhyvhNxO8 NI6mQrOKfkfP0wnmZk/xVaZUU5nJZ7pRP0kjaQItwGgfRs3/ZZ/z9UgbxHNKrV5NI+CXn0pv07P0 DstjpWwuu4qP46v4fWI1WdHjBKR2Wgp/34XW30YY7eep/Lh4UHlM+co8evi0PgIr4qFf0L30FEvD TFXWzX7MXmXv8RrezH/B3xV3Ko8oL1taMeuraSXdTo/R5yyTTWEN7Iesk61jt7GfsrvZMXaCfcir eCNfzs+ITnGt+K1SjTRf6VZ+YrrVtNX84XBw+Jnh/xn+XC/Tb6UGxMNGjP5ndB9mNkDH6TWkU/Qu M7EUNgJJZS62gN2EtJ7dzh5ge9kjrB+9nGDvso/wSPqUfcXxpOVmbsfhRx6B3Hw1Tph38nv4caQT /G/8C5EtCoRXTBYzREiswqhuE3cgPSneUfKU44oOP5eZdpn2mPaaHjM9bTprTrX8GM/4F79+8FzR ubeHaXjz8K7h6HC//g6Nwhri6YFXsBkYfSvSMqz3LkTcPnqFpcJ3eayIVbAr4Zlmtoxdy66HJ29h u9mvjLE/wQ7BS39iZzDmNO4wxnwZn8yr+Vykq3kHvxaHsZ28n7/KvxQWkSLSxShRJGaKJtEhesQN YpfQxIviLfGu+Ex8jaQryYpTKVA8ileZqTQra5T7lA+UD0yLTC+Y/mJONq8032qOmf+OU02Fpd7S YGmy7LDst5y0tiA6j9CT9BtE4IWLnRYbRUA8Sdv5RCUXrzAvIZ6bqV3M5ohUvpdt5jezfj7GdL15 Op/O5tBZxQNfP8f38M/4dDGb1bH5tIxPiDdoHqk8Cm6GcoSGlEOY20to+XpzKlvPz5hTKYoz0lSc kZ4V4xWveIFeF6eYRfklvaEks2w2xB8W9YiC3yoVpiC5xD30hLiW3UxP8gBR8lfWbYjjOexR7AuN rIz9U+g4Bs9BFJWL9+gntJz/mYZwH2+mn7N2ZQltp4lsHX1Av8ZdMc50jbnIPIr9gS9VIvwS1k9c eQSzm8rGMGEaSbewJrHbfIa/RmvouJJMb4v/xuiP8yfEbOWsaR7rxB1wM91K1+ob6QZTUHmZLSHB rqJC5TR2t3WiTHEh34BdZRH2tP24uw9gH6gSsyHJQeRcibhYgB1iN9Jd2CcURNBS3OM/wC72EvWb G3mMlphGMOw6+FLzwvA8Wqj/mu7Wl9A1+k4qwX5wm74OLe6lv9AO2ss2Dd9EXXiVfA339pWmWn7c VKuX8Ah/jc/nuy5eX3i7kOXQX5GewMpUmA5SRPkTzadKfZv+R0T3pdhh76bFOLC+j1l+jB5miUGa ODyH9+q1ogvzPUUN+sO6kyVTp76C5tIh+pXFRK0WL9ZYYy9jvjdRB5+n94iO4aXwww54wQdvrcH+ s8VXs6CxyldZ8b0Z06dNnVI+edLEsgnjSy8rKfYWjbt0rKdwjLvApTrzRzvsebk52VmjRl6SmWFL H5GWmpKcZLWYTYrgjIoD7toWVfO0aIrHPWtWiSy7WyFo/ZagRVMhqr3YRlNlvVaoLrL0wfJH/2Lp i1v6LlgymzqDZpQUqwG3qh3zu9UYW9gQBH+73x1StSGDn23wdxh8GniXCxXUQE6nX9VYixrQaq/r jARa/CXFrDclucZd05FcUky9ySlgU8Bp2e6uXpZdwQyGZwem9XKypmGKWp7bH9By3aiKZkRhoLVd q28IBvx2lytUUqyxmjb3Yo3kSclrmFCN0Y1mrtEsRjfqUpxxNNqq9hYPRrbFbLS4xZva7m5vXRTU RCvaCGgZXvTr17JvfD/nmyIax5nstm9r7SISyFmqSuNI5DZVu78h+K26dpdsIRRCG6jLC2tbIrXo ehtWqk6exTW+KRTU2CZ0iYNloTGr+Pzip97ClmWqluSudndGlrVgafIiGs27wRXNy/MN6KcpL6BG GoNul1Zpd4da/Y7ekRSZd0Nfrk/NvVhTUtxry4g7tndEeoJJTfs20wGnx3UGZ5hLrm7eBc8yOUb3 FTgJamqbipEE3ZjTFEk6plCkbQoWAFeIoZbWjhVZqiXVtERs06QcU2SaqdDmViOfEiLAPfS3iyWt CYm50PYpSaWMkwuhprHW87zm9WpFRTJELDVYU4yxwihPLim+Lsbd7i4b3p/lSwPVw7etoWmlcL/L JRd4a8xHi1HQwg3BeFmlxfYo+UpxtuYtUjN4XjNqgdSEz2suVG9xI5L75fssjdKsngu/dFvWJYHO aRrL+g/qjri+br67DkdjNRBpSURtXeNFpbheOhR+gy7BaZfUBIWdQyY5bheGNn5CPm+C43IwVVMK 8TMbQd0es1gRlYaEqbWarWVWnIaSXa7EPfP/VYrpZ2UtI/umWmIa2jRvYqDxYWvTLypfNLzUiKhr xJbDcbKPRJIv0iHU4qO8IpEh4vGi71JrNFqAO7MQP7xyTJEI2TUfXAZNI+4iQxyyJ4oXGdoTlUK4 ZHSWFNdiz4xEat1qbaQl0hrTw4vdqs0dGeBP86cjXQHsdvHAiekHttq12m0heKyTTcPtwam61802 N/T62Ob5C4MD+MShbm4MRjnjNS3VIbksvKYxmHCLsSAy9LGG+EMDEaN04yRNYGTwEP5pMQP4b+SC BCFlqLC1y8OwCR/08V8FuTJcGYUgaIG+VsXg1z4TfUWqMohHNc3WP1BGmQbx8aqIze3lcgA+Zzpz 4tAlmP3SfF8aS0sbacq3mwryR6Yl5zMqtMX0f/hy09L4Alt+ti09nS/ItqWmgqZKWXYp3oWOlXqP nTxme5YqhyqHMqeWNg3ZnmmaMN7uK1mey/wW3yh/rl9dmNmoLhftlnbrssx2tce6xrHJeqvjVevJ rAyLmpZmXjAW34b6JeOWH4lSJOcyFBapGKu6VZdUZMhR1qdxjNPOXmmGO2K805d0fswsxqb4MunJ wm6bz+WeNN6Gz/E2G8cszv4mNdW8wHZHcfIBNoXy2VRfVmV2c/aq7A3ZSnaW1GVnyU6zY3xMn/f5 rTle22deb9PQELKma71D/xgy5neu6f0MY44TxrMpuOS7IGuiphCzeMZ63AUWs9ly+eUTyzJHjTSb 3QWUYStHKYuNzMqaWHb55EkwMYuv+nKKr1h+VdWCxbzq0JL+c2tP3PLO8Pv3bvnw8bfOlc/dPmf1 Qw/cdOOjyvwRy8bPHl/x8ZttLcOfvxwZWo+D4jr2yFN7n/76raZHQ7H77tq3z/hghgjQx+I8/V0X hzDXUDDKTESTGe8/1FA/Z+as2d6q1UtbV8xu/D+YS1vlCmVuZHN0cmVhbQplbmRvYmoKNTIgMCBv YmoKNDg5NwplbmRvYmoKOCAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUg L0Jhc2VGb250IC9ETlJPQlErQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNTMgMCBSIC9FbmNvZGlu ZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDUwIC9MYXN0Q2hhciA1MCAvV2lkdGhzIFsg NTU2Cl0gPj4KZW5kb2JqCjUzIDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRvciAvRm9udE5h bWUgL0ROUk9CUStBcmlhbE1UIC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMyNSAyMDAwIDEw MDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0NhcEhlaWdodCA3 MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2Z1dpZHRo IDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUyIDU0IDAgUiA+PgplbmRvYmoKNTQgMCBvYmoK PDwgL0xlbmd0aCA1NSAwIFIgL0xlbmd0aDEgNzIwNCAvRmlsdGVyIC9GbGF0ZURlY29kZSA+Pgpz dHJlYW0KeAGFWQt8VMXVPzNz95H3JoE8N7t3WbJINjESHgGSJpvHBmxECAS7S4MkJJGAYCIB38LS FoEFxVJLC7Vi1VaqH3KzCXQDtEnFakURqtTWRwVfn49fEeqvYn2Q+/3n7oLS+ut3Z/9nzpxz5nXm 3Llz765aubqTkilEgnztK9p6yLhyO5Dltd+0So2VU3KIzM3X9SxZEStnLicyfbRk+a3Xxcq5G4l4 c1dnm6wnry+BKV0QGCVik5CP7Vqx6pZYObcFuXV5d3tcnxtC2byi7ZZ4//Q6yuoNbSs6keOyKyBq T3fvKqNI+VJ/Wc/Kzrg9CxAl/e6L6Gt72IfsmftjRkQMDKePqZJ+ThZwNiqla4iULqWATChLvYlv 3fbxlnOL0io/seZbjZoPvT2uSDKHlZ57P997fomNrMkoJhj2UoF6lqqRq6nORp/v/fw2W6wnqblw 8UFqFpf1e3Kcxw+J8XQK4GJ8xFvgHBTjREGkwumLCnd/xuiytJoSoaLFUoOqoN3AXmAIUGiRcEBr A10LhIC9wBBwHDATgUqtCnQDu4BTgFkUCHtEddpqxolc1M3FfNNENp0BdECQE7QUmA0sArYCuwCz YScl3cBaYAg4C5jJJ7Ij2yZi7NmRzUbWv2x5mVFsixVbFhrF/u8EY/msplhef2XMbHrMbMKkmPjy 2lg+rjiWZxSWhdB4f2JK2XBNlsjCJLMw8B5Qxp+iNMbISQ+K0aQBXGCohsQnMvrHesp2DQmFmOCC UQc59WHBIinpZTWJXOdnKIOc/CN+Oqbhp/tT08t21Xybv0V7gSFA8LeQ3uRv0lp+SvoctBrYBQwB x4AzgJmfQjqJ9AZ/g9L436gUqAYWAbuAIeAMYOF/A7Xx12XEGFTy1QDnr4Pa+GuY1mugafxVcK/y V/Vh/lKkfFrZoMF4S+OMszDOZOfHmYyssih/MfLZeESUByuNiDooxlAVTRRjIoUTnFGRE6lc6ozy t/tVr/PBmiv4CdIAjpGcQM8nSAXmAK1AD2AG9zK4lykE3As8CGgAogzUBqj8CPA88DJdAfiAOYCV H4+gmyg/FvHUOmuy+Av8GcqGx4/yPxr58/xpI3+O/8HIn0XugP4IfzricFJNEvSEOjbkNuSl0Jv4 7/vHZjj1mnQ+BA86QUuBamA2sAjYCpj5EB8T6XBmoJGDdAT3sJNH6AMj/xU9ZCXfMqfPU4cAVCXx TP8WOJBd6i4P93m270BREs8928BJ4vnBFnCSeG5bB04Sz/KbwEni6VgGThLPgkXgJPHMbgYHEuUP /GbsOGf57OuZWpPGb4aXboaXboaXbiaF3ywTfYbtzMl/Fikqgsd2+rzji5yhAyx0iIXmstBDLNTJ QmtYaB0LVbLQtSzkZSE7CzlYyMdCB9lUuCLEfAOXFKf5cljoCAvtYaFeFvKwUCELjWUhlZX7otwV uRJ3HTK/kfXXyJuOu/q/VYXdJ4274FEXYt6FPWEI9BigGyUfjNQxMeNch8zH9BdVx8qXTy/rrpnJ D6PiYSzDYToJKFigwwijw2jkMJpLA60GFgHDwBlAB8ywHoN5bDVoGmgpUA0sAtYCZwCzMZwzGAqn blA5xL3GwEpBq4HZssQPI41BcnGXr8Bmt3ltM8VWO0tzsNkO3cHLKSsL+3JGujU9ylL2f5ryr09T KKEmgd/Dt1IBFuLeeL418lmBM8p+GvEcdNaMZj8hh4KoY9PIwwqRT6VeozyZ7FYpn0R2/jjysoj9 GlRLi3iKnQdYqqy13/mZ/R3nB/YoB/u+/aDzL2pUYRHnnyF5fL/zhH2T89nSqBWSQ54oQ3ZANUwH 7VOde44Ypuug2BlxrpHZfued9hnO6+2GojOmuLYXJV+ac65ngXMm2qu3L3b6etHmfme1/VpnZcxq sqyz33kFhuCNsUUY7Hi70anbYTQ4vzzKunzFlu2WgGW2ZYqlzFJscVmclgJLvmWUNcNqs6Zak62J VqvVbFWs3ErWUVH9lM8rn3qjzMbDz4yAZqQYvA07DJPbDChxZuX0bdIyRSNvnFfLGrXhdmpcrGrn 5rmjLLFpgWZy1zIto5Eam2u1qd7GqEWfq5V7GzXLnO8G+hi7JwipxjdGGTUHokyXovX5WkZdYJAY S19/d77ML1t/dzBIOVk3VedUZ1SlT2uo/wbSaghb671fXTlfsd4cb4G2vXFeQHusIKiVSUYvCDZq P5qntgQG2cfsrL9+kP1DZsHAoKhiH/vnSrmoqg8GG6PsGsOOVPYP2CFikMHOigeztCPV6ojZ7YzZ FaI+7MbKDHYJCVRo2BUmJBh2CpN2fb1j/fV9Y0Fgk61Sr2HTm61+3eZIIWwKQWCTFaIjhs2RrJC0 0aqMZux2mDhAYMLyyG6Y2FmeYWKMvM8wKY2bbLpossnoScRGY9hIgmZSTl2wSTkFm6858r+znbVe L+uvCLa3+Dvd/la3vxNo1Tbf1JWjhRaral97UCpUTXhaF7d3ybytUwu6O+u1dne92ldh1Ps3dYtU V7jr+6jF3xzoa/F11kcqfBV+d1t9sH/GnEnll/S16WJfk+Z8Q19zZGOTZF8zjHr/1le5VM+QfZXL vsplXzN8M4y+yIjxOYE+K9UG67B+Mu/nSYmI19Z8V7A2y9ZTZQRvhStnTf4BnFZ2U5I3qCW7a7UU QMZ1SU1JjVThnpKqVIjT4qqcNRWu/ANsd1xlgzjdXUveVat7V1OOf2l97NeLC6JVq+VSxKhXyr7x golf87XVy7N1o1Y0r1GrbloQ6LNYIG2tD0I2/YIsKckf1YdjwsshnC4NhbhoKGWVUpaQEDf8z1gw xgQxvDOIg8bBfuZzsFXUGxSao7GZYytoXgA3tCwIHMBZSj4keoOYYC/zst4Lrcl5GDzFJIRp917A qtVxLu6LVfHcMO31krf3gksuNOeVzjKI4atVXmxtpgOUC+SZHqVcxUN4/9HfA96X+chS/X2plzn/ EBtdNA6i3bSHLaU9NERPsrOotZcGaYDkEaie7qc76D7agMfaAkg20VwkE+T3sVx9AG8mv8AD8xd0 FLbfoTV0gLJYjv4BraX14iXUWk8pNIZqaA51093sKn01tdBJ5ftUTlfRDdTDQnpAv0ffpj9Cv6RB 8Uf9PCVRHrUjHdU/Mv1Vf51KUOPHtINOsm0J+8iHXkKw/DmtpJ1iocL0JfrnGIGLbsYYFJpFR9kw 96L1TnqP5bA7RB1aeVjX9KdgZaeF1EU76QCbzGZwl6lFn6UfpSz0cQta3UER2o8Upd/SqyzZdFZ/ RD9LuVRMV2I+A/QCGxYj59eNVMNvJnhpPE2Dppt+R8/QceZmv+fdpmRTmclnuk0/QaNoAs3HaB9F zf9ln/I1SGvF00qDXkup8MsPpbfpD/Qmy2OlbDa7ho/n3fwBsZKs6HECUgcthb9/itbfQBjt58n8 mHhYeVz5wlwwckpPxYp46Gf0c/o9S8FMVdbLvsdeZm/zOr6I/4y/Je5Tfq28aGnDrK+lFXQ3PU6f sgw2lTWx77IudgfbwH7IdrCj7Dh7n9fwZn49PyO6xI3it0ot0jylV/m+6S7TZvP7I4GRp0b+NPKp XqbfRU2Ih3UY/Y/pAcxskI7RK0gn6S1mYkksFUllLjaf3Y60ht3NHmK72a/ZAHo5zt5iH+CR9An7 guNJy808H4cfeQRy85U4Yd7H7+fHkI7zv/PPRLYYI7xisqgUQdGNUW0Q9yLtE28qecoxRYefy0zb TbtMu02Pm540nTUnW76HZ/zzXz58vuj8GyM0snFk+0hkZEB/k0ZjDfH0wCtYJUbfhrQM670dEbeX XmLJ8F0eK2JV7Cp4ZhFbxm5kt8CTP2A72S+NsT/BDsFLf2FnMOYUbjfGfDmfzGv5bKRreSe/EYex bXyAv8w/FxaRJNLEaFEkZoiFolOsEreK7UITz4u/ibfEOfElkq4kKk5ljOJRvMoMZZGyWnlAeU95 z9Ries70rjnRvMJ8lzlq/gdONVWWOZYmy0LLVst+ywlrK6LzMO2j3yACL17slFgn/GIf3cMnKrl4 hXkB8byIOsQsjkjlu9lGficb4GNNt5greAW7ms4qHvj6ab6Ln+MVYhZrZPNoGZ8Qa9A8SnkMXKVy mE4rhzC3F9DyLeZktoafMSdTBGekaTgj/UFcoXjFc/SqOMksyi/oNSWRZbPT/FExB1HwW6XKFCCX uJ+eEDeyO2kf9xMlfmHdgji+mj2GfaGZlbF/CR3H4KsRReXibfo+Xc//SqdxH2+kn7AOZQndQxPZ HfQe/Qp3xXjTDeYi82j2LF+qhHkmGyCu/Bqzm8bGMmEaRT9gC8VO8xn+Cq2mY0oivSH+B6M/xp8Q s5SzprmsC3fAnXQX3aivo1tNAeVFtoQEu4YKlVPY3e4QZYoL+VrsKi3Y0/bj7j6AfaBGzIIkB5Fz FeJiPnaInUg/xT6hIIKW4h7/DnaxF2jA3MyjtMSUyrDr4EvNcyNzaYH+K9qhL6Eb9G1Ugv1gg34H WtxN79JW2s3Wj9xOPXiVfAX39lWmBn7M1KCX8DB/hc/j2y9dX3i7kOXQh0hPYGWqTAcprPyF5lG1 vkX/M6L7MuywO2gxDqzvYJYfoYeZYpgmjlzN+/QG0YP5nqQm/VHdyRKpS19Os+kQ/dJiojaLF2us sRcx39upk8/VV4nOkaXww1Z4wQdvrcb+s8lXN7+5xldd9a3KiunTppZPnjSxbMIVpZeXFHuLxl82 zlM41j3GpTodBfb8vNyc7KzRozIz0m1pqSnJSYkJVovZpAjOqNjvbmhVNU+rpnjcM2eWyLK7DYK2 rwlaNRWihkttNFXWa4PqEksfLK/7N0tfzNJ30ZLZ1EqqLClW/W5VO1rvVqNsQVMA/N317qCqnTb4 WQZ/r8GngHe5UEH153TVqxprVf1aw01dYX9rfUkx60tKrHPXdSaWFFNfYhLYJHBatrunj2VXMYPh 2f7pfZysKZiilueu92u5blRFM6LQ39ahzWkK+OvzXa5gSbHG6trdizWSJyWvYUJ1RjeauU6zGN2o S3HG0Wiz2lc8HN4StdHiVm9yh7ujrSWgiTa04dfSvei3Xsu+7Z2cr4poHGeyDV/X5ouwP2epKo3D 4Q2q9mBT4Gt1812yhWAQbaAuL2xoDTeg6y1YqUZ5Ftf4+mBAY+vRJQ6WhcasYvOLnXoLW5epWoK7 1t0VXtaKpckLazT3VlckL883qJ+iPL8abg64XVp1vjvYVm/vG0Xhubf25/rU3Es1JcV9tvSYY/tS 0+JMcsrXmU44PaYzOMNcco1zL3qWyTG6r8RJUFPbVYwk4MacpkrSOZXC7VOxALiCDLW0DqzIUi2h rjVsmy7lmCLTTIU2txr+hBAB7tN/v1TSFpeYC22fkFTKOLkYahpru8BrXq9WVCRDxFKHNcUYq4zy 5JLim6Lc7e6x4f1ZvjTQHPi2LTi9FO53ueQCb476aDEKWqgpECurtDg/Qr5SnK15q9QMX9CMni81 oQuai9Vb3YjkAfk+S6M1q+fiL82Wlenvmq6xrP+i7ozpG+e5G3E0Vv3h1njUNjZfUorppUPhN+ji nJZZFxD5HDLJ8XxhaGMn5AsmOC4HkjWlED+zEdQdUYsVUWlImNqg2Vpnxmgw0eWK3zP/X6WoflbW MrKvqsWnoU33xgcaG7ZWcUn5kuElh0VjM7YcjpN9OJx4iQ6hFhvllfEMEY8XfZdap9F83JmF+OGV Y6pEMF/zwWXQNOMuMsTB/HjxEsP8eKUgLhmdJcUN2DPD4Qa32hBuDbdF9dBit2pzhwf5k/zJcI8f u10scKL6gc35WsOWIDzWxabj9uBU2+dmG5v6fGzjvAWBQXziUDc2ByKc8brW2qBcFl7XHIi7xVgQ GfpYQ/yhgYhRenFWIjAyeAj/tJgBorKLEoQUk/82YGuXh2ETPujjvwpypbvSC0HQAn2piuEvfSb6 glRlWLaFFhWLaZgc7Jk+s+zel5BuS8nJzDTPT4nq/xxITzeYj3wJNhs4xyiTI6q/78uWBg6H1Drs qdA4klNSQKP8oC+ZJ2Zn49+CdM5VZ3rGtNITRyU9SqWn5WtStaRP4XGV38cvdpickcGNDn0Jaeng Yv2c8iVlZPL5jlFSJtuOoGl0P5CUxOeD+bsvNSUF3Df05vXG+pO9GZ35plSYKswHTUPmg5ZnrM/a LVcmB5ObU69P7ki9LeO2zE0ZhzLezXs3/2xe8lDSbzJ5Pj7+FdgcNvPv8LphwW5pRZ6gn/XlORJt VrP5iD1vlN2eZ7XnCcateXaR4rBF+SP9s9MZPg3m7JMzIMMdaYwnJ/ZmvwRv+1zuSewgX0cq2dhU X3L6vmq8FnTztVzhB/hYfADc2rc5x2s7t/Cfp23nvJW2f56m6srzldWnzy98Jz0jexqTZEPq5d7U O21PoTBtwhU0FReThBayhSuDwcLRLk/5lCn4TZ7kcY8xW8ZNmTKxDEcAs9mCn2L5spxnFz6888zu Hbd/7342mPmvP710buajTz7U4tizp6ayfXjNU+9ed/2P7g9nHnvlwz2Bxw49srFtgvERDlGlj8MZ /ZsuDuFoQ8EoIx6hZrxTUf3Vc2fXNnlrVi5tWz6r+f8Aa1xtyQplbmRzdHJlYW0KZW5kb2JqCjU1 IDAgb2JqCjQ5MzMKZW5kb2JqCjI0IDAgb2JqCjw8IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVl VHlwZSAvQmFzZUZvbnQgL1BNQ09aVytBcmlhbE1UIC9Gb250RGVzY3JpcHRvcgo1NiAwIFIgL0Vu Y29kaW5nIC9NYWNSb21hbkVuY29kaW5nIC9GaXJzdENoYXIgMzIgL0xhc3RDaGFyIDExOCAvV2lk dGhzIFsgMjc4CjAgMCAwIDAgODg5IDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAw IDAgMCAwIDAgMCAwIDAgMCAwIDAgMCA3MjIKMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAg MCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDU1NiAwIDUwMCAwIDU1NgoyNzggMCAwIDIyMiAw IDAgMjIyIDgzMyA1NTYgNTU2IDAgMCAwIDUwMCAyNzggNTU2IDUwMCBdID4+CmVuZG9iago1NiAw IG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9QTUNPWlcrQXJpYWxNVCAv RmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAv QXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGlu ZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAg L0ZvbnRGaWxlMiA1NyAwIFIgPj4KZW5kb2JqCjU3IDAgb2JqCjw8IC9MZW5ndGggNTggMCBSIC9M ZW5ndGgxIDEzNzY0IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4AZ17CXxURbZ3Vd17 e19ud5Lek+6k052QDgSSDiEQyQ1JAI3sBBNMJAGigCAJQcFxC6NsEQWZUUdcwGUcl1E6C9igDhnX 58LAc5vRUUFFRWciPD8GRzTp71+3G5R5873v93v35lSdqjq1nTrn1Km6ndWrrm4jJtJFBKIsWtHa TtTHvQvRK4uuWR1Qk8TsIkQz9/L2K1Yk02nLCZG+uWL5tZcn054thHjeWdLWujiZJj8iHrsEGck0 jSLOXbJi9dpk2t2EuGb5ykWpcs8SXr6idW2qf/Ih0oGrWle0JelLH+Hp9pWdq1PpFsQ17avaUvS0 gRDjH36I//Up+jV99b4kESEUiI18SyrI/URLGJFJEZlHiLhEzCQS0rxcYlu3f7vl9AJrxT90Xp1a 86HP8go48qLYvu3M7qErZKIzIalX6XkB6mknDk8n1TI5s/vML+RkT7zk7GPbR+YK+X1hl//wc8II chTAhBG9kUz/PiFPyOyd4FfiQrDPnlFsrRopBNBikRoGEK4E7AYcAIhkgZCFUhnhTYAuwG7AAcBh gIYQhLw0AFgJ2Ak4CtAImYKvN+CXq/IEN+q6MV+r4CQnAAmAQPwIiwAzAAsAWwE7ARqVjuesBNwE OAA4CdAQRXD2bi/B2J29t6pR37LlxWqyNZlsalaTfZc0JuNps5JxzYVJsvFJsjHRZPaoSck4rzAZ 20PFXWi8z2AuHqhyCA5M0oGBtyOk7CVipZT4yS4hg8QATMBQ1RxFsPflhot3HhBEQgUmULKY+BMD Au0124qrDCzBThA78bNv2GCyhA32WWzFO6suYp+S3YADAIF9ivcT9gm5iR3lPEdYCdgJOAA4BDgB 0LCjeI/g/Zh9TKzsI1IEqAQsAOwEHACcAGjZRwhl9iGXGDXkeCWAsQ8RyuyvmNZfEVrZB8A+YB8k BtjbvWXlxftUJFKUQvyhFOL0phC7ozjO3ur9fgQkKoyVhkQ9K+SQiaREyOkNjfHHBVdvxVJ/nH3W F4j4d1WNZu+QGIBhJO+g53dIADAT0AJoB2iAvQfsPdIF2AbYBYgBIGUIZUCAvQ54E/AeGQ1QADMB Ona4F93E2aHe8CR/lYP9ib1KnOD4QfYfavwme0WN32Avq/FriLNQ/jp7pTfLT6qMKCeoIyOWEReh XGJ/7Mu1+xNVNnYAHPQjLAJUAmYAFgC2AjTsAMvpXey3o5FnyevQYT/rJV+p8aPkIR1RlvmVcDUE MMCD8PgLgCHYGdgZZkr4rnuQ5EH49u3AeBC+ZQswHoR/sQ4YD8LLrwHGg/DiZcB4EJ6/ABgPwjPm AkMQZw88k5vnL5txJQ1UWdkacGkNuLQGXFpDRLaGv+R7kY/x3t6CAnBshxIZUeDv2k+7nqNds2nX Q7SrjXbdSLvW0a4K2nUZ7YrQLh/tyqJdCu16lo4DK7qo0n9eslxx0a7XaddTtKuTdoVpV4h25dKu AC1T4iy790JoHaJaNeqr4krHsvsumAjrY2XZ4Gg2ZD4bNuEAwkOAhJpSQBTISRK7s3ic01dQmUyP Gl+8smoqexEVX8QyvEiOAEQs0IsQoxfRyItozoqwErAAMAA4AUgANKDOwTy2qqEVYRGgErAAcBPg BECjDucEhsLISoR8iLvVgRUhrATM4Cn2It4cvNksW8mUfXJEnips9VFrFp2RlchiZcThgF2223S2 ODXv/c78z+/MRF+lZ7ezrSQTC7EtFW/t/T7TH6e/6Q0/66/KoHeTLBFSR8tJmIYQjyOdarqU+HQ8 P0p87EnExb2+eahm7Q0X+vdTC6+11/+975j/K1+cAT3ue9b/50BcpL3+d5Hz5F7/O77N/teK4jrk PBeOU0T7AyrpPt84/1Ovq6TrULCj138jj/b6b/BN8V/pUwvakgWXdSKlWP2zw/P9U9FejW+hX+lE m3v9lb7L/BVJqlJeZ69/NIYQSaIFGOwIn9ppMEttsL4sTpcohdq7tA3aGdqx2mJtoTZb69dmar3a dJ1dJ+ssOpPOoNPpNDpRx3RElx5PHFUifNdL16ibnwYCTYmo4jIsDOVmBiFhVMfIRSSWJtSxujmT aF1sYBGpWxiInZ4TjFPDrPkxKTiJxux1pG7upNi4SF1cm5gdK4vUxbQzL23oofT2RuTG2KY4JXMb 4jTBs9Z7Y/bqhn2EUtv627w8zl9/W2MjcTmuqXRV2ifayifX/JugRc1sqYn89Lh+QiOuSGbsrro5 DbEnMhtjxRxJZDbWxX41J9DUsI9+S0/W1uyj/8WjxoZ9wkT6be1sni9MrGlsrIvTeSodCdD/Ah0k BhHodNiYOR0J6LKSdDuSdCHUB10uj0Cn15OQShfS61U6kXK6ns7c2pqeXASgcQZIp0rT6Qz8nOb1 EGhCCEDj6CKvqzSvO7o4TWyi2ozPB5IsBCChHuJTSXzUo5KoI+9RSYpSJJvPkWxWexKSo1FpeIBm zEfP0piPguZnjPyf0bZJkQjtm9C4qKm2LVjbEqxtA7TEbr1miSvWtTAQ6FnUyAsCMSHcsnDREh63 tsUag201sUXBmkDPBLXevxQ38eIJwZoe0lQ7t6GnSWmr6Z2gTKgNttY09k2ZGS07r6/N5/qKzvw3 fc3kjUV5X1PUev/SVxkvnsL7KuN9lfG+pihT1L6IKuMzG3p0ZFJjNdaPx33MaIC8tnizGyc55PaJ qvBOyHbd6N0Pb+UxYow0xkzBSTEzgMv1yKqRVbwIOsWLLMi2popcN07I9u6nj6WKZGTbgpNIZPXV nVcTV+3SmuRfJx5krb6aL0UyjPC8f/uApDamtNZw37ouVjCnLlY5a35Dj1aL3JaaRuSNP5tnNNbG EwPJzFHIHM8JBeEcIc+r4Hl6fYrwv8uCOiZkgzv74Gg820eVLLqadDYKsay6uQymYO58sKFpfsN+ +FJ8k+hsxAQ7aYR2nm2Nz0PFSTKHYNqdZ2H11SksxYvVqVgl7YyQSOdZlpxtLsKZpQYqr1ZHYNqk /cQN8Ei/I24xTHD+SXwJOM7j4aWJ47ycx+xrGLp4Cgh5jDxFl5KnyAHyAj2JWrvJPtJPuAtUQ+4j 15Nfk43Y1uYjZzOZjVdC/q+pO9GPk8mD2DAfJAdBewm5kewnDupKfEVuIuuFt1FrPTGTHFJFZpKV 5DZ6ceJq0kSOiDeTMnIxuYq0065EQ+L2xPbEI+S3ZJ/wH4khYiQesgjvwcQ30l8SH5KRqHEnuYcc odv1e4iCXrpAeT9ZRXYIzSJNXJE4gxFkkzUYg0imkYN0gEXQehv5krro9UI1Wnk4EUu8BCofaSZL yA6yn5bSKSxbakpMSxwkDvSxFq3eQ3rJXrxx8jz5gJqkk4lHEieJmxSSCzGffvInOiAMD60brgTf JHBpBClHyUryB/IqOUyD9I9spWSSiiVF+kXiHZJOxpB6jPZ3qPkF/Y7diPcm4RVxcmISsYAvd3Bu k5fJJ9RDi+gMOo+NYCvZA8IqokOPY/AuJkvB79+g9Y8hRnuZiR0SHhafFH/QZA4fTViwImFyL7mf /JGaMdMA7aS/pO/Rz1g1W8DuZZ8KvxYfF9/StmLWl5EV5DbyJPmO2uk4OoteSpfQ6+lGege9hx6k h+lxVsXmsivZCWGJ0CE8L07CO0fsFG+WNki3ao4PNwy/NPyfw98lihMbyCzIwzqM/k7yAGa2jxwi 7+M9Qj6lEjVSC94Azab19Dq8N9Lb6EP0Mfo47Ucvh+mn9CtsSf+gPzDstEzDvHB+uAsUZKvgYf6a 3ccO4T3M/s6+F5xCjhARSoUKoVFYiVFtFLbh3SN8InrEQ2ICfC6W7pJ2So9JT0ovSCc1Ju0vsce/ +ePDQwVDHw+T4U3Ddw33DvcnPiEZWEPsHjiCVWD0rXiXYb3vgsTtJm9TE3jnoQV0Ir0YnFlAl9EO uhacvIXuoL9Vx/40fQ5c+jM9gTGbmU8d8yhWyiaxGXgvY22sA87YdtbP3mNnBK1gFKxChlAgTBGa hTZhtXCtcJcQE94UPhI+FU4LP+JNiAbRL+aIYTEiThEXiFeLD4hfil9KTdIb0ucag2aFZoMmrvkv eDUTtTO1s7TN2q3avdp3dC2QzhfJHvIMJPDcQ48K64RaYQ+5nZWIbhxh/gR5XkAWC9MYJJU9Rjex G2g/y5XWaiawCXQ6OSmGwetX2E52mk0QptE6OocsY2OSDWrSxSeAVYgvkkHxOcztT2h5rcZEb2Qn NCbSCx+pHD7Sy8JoMSK8QT4QjlCt+CD5q2igTjrIfifMhBQ8L06UGki2cB95WuigN5A9rJYQww+6 LZDj6fQJ2IW5tJj+U0jADZ4OKSoTPiM3kyvZX8gg9HgTuZsuFq8gt5MSej35kjwKrRghXaUp0GTQ 19hSsZul0X7CxMcxu3KaSwUpndxCm4UdmhPsfXI1OSQayMfC7zH6Q+xpYZp4UppNl0ADbiAbSEdi HblWahDfolcQgc4jIfEorNv1QrGYjfgmWJUm2LS90O79sANVwjTkuCA5F0Mu6mEhduD9DeyECAla Ch2/BFbsT6RfM5fFyRWShcLq4KbmjeHZZH7iUXJP4gpyVWI7GQl7sDFxPVp8jHxOtpLH6Prh60g7 jpLvQ7cvliazQ9LkxEjWzd5nc9hd568vuB2iLvI13qexMhOlZ0m3+Gcyh1QmtiTehXTnw8LeQxbC YT2GWX6DHqYKA6RkeDrrSUwW2jHfI2RW4ncJPzWQJYnlZAZ5jvxWK5FWbQRrHKNvYb7XkTY2O7Fa aBteCj5sBRcUcOtq2J/NSnX93CqlcuIFFRPGl48rK42WFI8ZXTRqZGGkYER+XjiUG8zJDvizMn1e j9vldGSkp9ltstViNhkNep1WI4kCo6SwNji5JRALt8TEcHDq1JE8HWxFRuvPMlpiAWRNPp8mFuD1 WlF0HqUCysv/hVJJUirnKKkcqCAVIwsDtcFA7GBNMBCn82c1AL+tJtgYiA2q+DQV36biZuDZ2agQ qHUtqQnEaEugNjb5miXdtS01Iwtpj9FQHaxuM4wsJD0GI1AjsJgz2N5DnROpijBn7fgeRnRmTDHm CdbUxtxBVEUzQqi2dXFs5qyG2hpvdnbjyMIYrV4UXBgj3FOKqCSkWu0mpqmOadVuAkvh48TIrYGe woHuLXGZLGyJmBYHF7c2NcSEVrRRG7NF0G9NzPmLY66fkmgcPtnGn5d6he5a19IAJ+7u3hiI7ZrV 8LO63mzeQmMj2kBdFprc0j0ZXW/BStVxXzzG1jc2xOh6dAnHMqTOKjm/pNcbalkWiOmDk4JLupe1 YGk83TEy+9rsXo9H2Zc4Sjy1ge65DcHsWKU32Nha4+tJJ92zr+1zKwH3+SUjC3tkW5KxPRZrCjGZ f460genJMhVTyTlWN/scZykfY/BCeIKxwKIARtIQxJzG8aBtHOleNA4LgKeRolZsMVZkaUxf3dIt j+f5mCKNSSE5GOj+B4EEBAf/fn5OaypHE5L/QXghl5NzohajrWfxWCQSKyjgIqKtxppijBPVdOnI wmviLBhsl3F+5ocGMhO8bW0cXwT2Z2fzBb41rpCFSMS6ZjUk0wGy0NtLlCL41qyFlwycLcmo5yVd Z0vOVW8JQpL7+XmWZMR04XN/VtmRVrtkfIw6/ofitmR53ZxgHVzjQG13S0pq6+ael0qWc4aCbyhL YbG06gbBy5DHMeYV1NKkh3yWBO5ygykmhvCnUYV6cVyrg1SqOTQwOSa3TE2GjYbs7JTO/P8qxRMn eS01+qlaahqx8ZHUQJPDjk04L33e8EzdQt1cmBwGz76723BeGUQtOcoLUxEkHgf97EB1jNRDM0P4 w5FjHIdGb0wBy1AyF1qkZjd6U8nzCL2pSo14uHSOLJwMm9ndPTkYmNzd0t0aT3QtDAbkYPc+9gJ7 obu9FtYuKTjxxP5bvbHJWxrBsSV0PNSDkUk9QbppVo9CN82Z37APVxyBTXMbehll1S2TGvmysOq5 DSm2qAvCRR9riC8mkBi+x2szyRrpVXK52EkuwsltJuI65KfRV8kmwM0a7OPI41CD8kqU5QK/GbgH IkdVwSP4SqNRTxoB7JVcFJMPv2j53z7C/7Yi6mFw8N35JLVqK8lPJ3piwLmDfy3hjxn+uRVfXZJP BnbYr+gG+MaviY9Js6U3NRdpF+pq9WHDGeMW06XmZbgtAqXEmxV4q9m2bFsIAW6WyI8BYeBHRSI/ kIA4wInWJF7X7JTeRm9O4iV58HC0imGbe5uHLdF5vF5+r2p1udNdLrfLm2F1e8ZE7M+xnUQPx8LE dipGweN2C9TrcoXyeb4f+aPYzt6Q0fcc20EiGMkYtqMv5/elGp7OQNqKJvUBsP7q6CXzXRH5dPPg qUH5NAJSOTg0KFfIFZUAGTi12cvLOWwcFblBfmnMaFf1tcpsWjIiK+InJYExfjoyDKwoF5iZWf3E KWb4qc0ALE0HrCAzH/eK2QgK80b5yeggAgs1+alDQiAb7X6SrkWQOv/iRuXsWXYdbU6Lji0phveg CeaEaY4mI91RUjy2NBoWaAml/4+yNQ/c1b3nmQ3re2h5deP8STUAIWf7j5/Qzx+4GwUbUTCeZ9Y2 zhfn3//hywf2v/YKfXn1vbd1rt5xe+eZTo3+++/o7Q/8lRe8Sl9afe+W1bwAzLo88aV0DdYpk7y9 ZxFblsloPHG832jU1BMgygKOBUixeRG8uNWZXeSWzG1kh/Sk8FvzPqHf/Kr5MDmW+X8ybRZ7pi0z UyjQ5NsKfAH/FPO89Esy5rmXSFdmXme/1b5DuMeyw/cYfYQ9ZnvXkgZ/0iOnyx6RxRMf9+aXo88B ZWR+uWwlVPSmZZkEb5aol8PWi0g4QCn1+J3hgI7qTHw0OnfWoiZXZLp8KtI8bXA61nlaco0rB23O 8jGjwevm5g7SDK6vok6NGMzJZaVRe25JsejUhsPBHA3LSLdzpov9L1ww/OLng8N/vnc3rX7hQ1o4 4UDJC796/LOmFV9sePhTxsac+OGP9Kq3Pqf1PUffGLlr+0PDJ+54dvir7ufgkJOLEsdFnzgRXmkZ G6kU6s36ArfZUzDCXFBQbh6bUeYdX3BhQbO5uWCZeWlBy+hu84YROxz3eh43Z+SfZXEeZ7GbT+pR 9xP5e93P5r/kPpT/VsZH+boaB82KJ04pNpNJU2+381Ay8bCUX+7Wc8zv9LsihQXRcrG88EJxauE8 XWPkct3SyDWmjabXTN+bv4/YyqIWKspFuVFncXa6a8GIlSPYCF+RpdKy1bLTkrBIOy27LScsgsVk tbJ6Szzxdb8sq8gpJUOWNfUWk9mMUGO1IrT4BGecPbHXdWe6z6eFeJxSPHwcpDbPUOwTjCNa5Vai MZtZPQll58YTf1cb44hi5Lm5osmEMJ44BvlSkVOqoCHnQ8XIu8tVO0L6x37ecG6cXapY8hQSlsOB 8Ojw7rBUDknpt1hYfTieeG+viozheYo5KxgdXT5QznaV03InH1sVb9EZcuUU5R7QHNIwv6ZSwzQW PlMNpoXQxcejgVAlQ029xsKnq5F555ox4+TTKbVt7oCEReTmjghyYFbOanPFUOTzz2FdKo9FYGKO waIUqUWg70CCWxiYGi6UXBwhkB2ISEdIw3W/NDp2bJn6lkbzuFhq8yYyWAYHjhYwCs5gWNBoLZBV LqplY3FBsHjfst3PTemcWnrlB1fQktpNN12bGXNddXjzpidmynpnznM+58KXVjYVr1i65KFw5s31 k59cP33d9HSL2ZMbMlw18oLGDlfHrXVK60Wj1p78Yf0F4+hH+T45f1rR1JZLZ1ywBrZgZuK4MAiJ 9tD5PYxvm0rUcpOVWo1Uwb1WO6y+aPcZtS6fiPuPDK2Os1+rslIL0QQuc1ZqVR4cfOcVzphB+aXm Yg5jRnuVKXoT9fuq06qdc9LmOFvSWpz3snuFHeZH5Ec8Jp3ZbVjGlgrLpKtN7eYu86OmPfq9hj0m k8O0wfQZEyw5C6wrrTdZBSuFECrXjsbBbSZpwbC2kV3kKDmJHwBYrUbsfGfH6MPQcy06VYJzvJhf rjHix/doGBSF6xxVuITQqXy1qYeT0Qt9GbmHtNSvrdQyrYUTaQ2cSKsqoHaMN/oS31mwlFjipHg0 r0o5F/jQwb2gwVWnIoOr1LlDHGzlRXLzMfxxEeigzR2NMElYfWKL2vk2kLJIyUUWKnoyTzz9wfB3 q77a/NSH/t3um+ZveuKRW5bdTtc7nzlEM6nh95St2/2g98rlL7793gu/5FaoDlYoC2uWQTLpHMXp J74MVi80S836emObcKW0Ut9m1GVwneMTtAFRZnMs08fDPPv70pn00x5xjH28e4yvyj7NU+WbZW9y z/a12ld4Wn1rNWszTrPTLhnXnVaz0znT0eJox4d/n3WbvEtmsix6fQYt2c+e4LNXFV415jLnmgxO 35kGaXEqZug4RsDqgXyjDgXI16qWAxlQ9HkF0ZiZmj1+pPpC4SiPlSqu1H7qd5TIuVoltyDKF2aG VtAGVEFT9VeLiUDokovl42updfCpad1Z0bLkNpHUyci0oWPT5Y5I5DQgEpmmugXQx2OVg1io5oqh jgrVL7DzPaRZVVjawXcQdblkUlJMbOnabAdXR5odVlVWuGx/4Tf7vho+QdM/fBc3gj8eN/SuX7Rl 6AM2yzRu3ubrH6fznA/3Uz8VcP2WP/zx8PdyYPf+JfTODdVLHoVEpmEJu7D7OqlZyUrXU6u7yD3a rbjb3fea7jM/btZ5zPnmmHvALbo5P/I9/mimziyYrD4DzWCR9DQRv7Ew7Eyn6Yk0RXSGRNwzbYeI cyaOGRflsRLx+aPbCHUr3PK5FTOUlqRzM0/yeQ7J4WpMCjkXQf+tYuSqTNK58CPNdwQV+aKfbwLI OfMMZzJ52OV+ju4n2eQ0blvwoe6sreS8BVcr5FNwtgblwcFmUllZUVExVFE5WG4Db+Fqpcs2jV6r 0cEey3q7l9g0Vi9uzCIF69bRSEczWVViC5aWlEbLxkJHoCIwjhkZJRlBW+/OnWmem6+5uMk7rnh2 zaFDwo4tHVdGJ19iv98wuWXhlh8vh0Zsgktegft/7qMeVC7Tj+WyMkO/Tb9LH9MP6I/oT+q1RO/X t+u79DtTWUf1Cb3Br4dt0IpM0GuEGynRSBrRoNGGJCLuFHeJMXFAPCpqBsSTIiNiQDyMlCjGE/9U hR7IGcXJOSmqu5xo4L2KKhdRluQikGFV9oH8qBg4G8XpuikzUyaFcw52ZRVYxVlW+ZObygVyVUck rbQkQ7CV2Db19/eLfzt06IcMMfzDB/wEcjOCMnXOn+2V1AlLXATKxkXVOFqajEePScY5ITVWQhnO qFXy43L5iCTOQHBSEvxSu9QlJSR8SycGJoSSwoSWVGHKwKrsJHQA5hYnmwA5DMsroiTJBS4dSqYq TyoXiMqFlCzpUoKUZAFIE6r6A0nxgkwXz+cFZwYkSWUHZwFP8acEPLi5X9p/ZjKf+0YccsKwgEH6 6j5iTpxSOayDB6Ai+njiL8o0ozkaEo+Jx/SfOD8PSO9KpwPMqQsE9S5vQC8IwSyfJsMHgddSTdDj lg2HQ3RbaFeIhZxOjyW0zUZtIlcYm4srDGwoHBo+SVs6VxWkjytOvtw2xpXGZuIThUhDXVB2Rp0l kH+q48FvHJoVkyu0zYvDDW/Oe645r9oc0t8oNt6cV+TNedUNCLnD3ItCGr4awrM+kpe3h99glQRD 9DChfC9kflKJu08BnB1OrYasajf8Gmi26u8QR0q/fzyr36eUdFXBVbEkqiNE3LmhOF3bl82XJTL9 nNujCimsKTzws8/PnCLI8ND02raaLzqwC0LxIcrTZBgCOOfcvnLzChNgMaWnhdNNNi+1mzO8FCe6 SASHI9XsYn0zuOY7nDzIsAVtUdVvT2IwBpoM28YHix9dds3d/htff+CJvmDTxPZf9zcsvnjdeDF8 5/QFCxv27947lMfuX75g/J2PDN3NeteunbnjjqH3U/IifAF5cdAblDRJ0KSxx+S4/JnwZdpJ4XSa Brp5UqmAwFwr09/Ih11HXQmXGNClW9Iddp8ECXGYDWaLyZLrgvvA6l1wJVi9MZ/jxnQuFUa+oDbO cKOqBMYclYK7pEbOdCOML8J44vvkghoNnO9In1ZUL8OolIyNJowUf8bpLm7BPdGx0ZjrpIu1u3a5 Yq4Bl+gSWEmGQ9XN0/02m2q5z6iGSFVB1RCRn6ugKkpqmU1dZabacJELFHIHFDv6PAllOqfS0504 WKkbZXKJoYUw6NwD/nkuygZh5rHOlTh9ldPU4jo0Nr1BZ9AaBI0ctmksXmo12FOLXIBV7iAQIXWV U/7uz5Z440NXf9Ty4EzZ0F9w5dTO34nhu3fXtk8rvmGok224akXV9jeH1NNXDfyePKyimbjplXsz XHwmafxcxRErV8lOjrnVArvW4DZN0UzVzdM06q7QLNXpovJ4+3hHqatWrrPXOWpdTVKTfrbcbG92 zHatkFboF8sr7Csci11raIZeI5kvFeZKcw2XmpYLbVKbYbnJ4PSJWhtMRnquV+FL7VXFQAtLr4DB 8EfQ7zmHGBKlHnRQfFIdn4rwdVAR1Wfhi5CWG4qO1lKilbUBODljjsBG8PwLuRsE3JJLTBa+ZdtV dVZ9boJBIEd1f1Jaq9of4lBXWEGT3BwwMsbD3SEs6lmVjQzCGWo+3fxTRoS7RJWDMLXYgztwaano 50hz9AulhXqRNuMzP540uQxKSXB9wd2itLOXF1DKmkc2v/xX6rjub7ceGR7c17txQ2/f+o29+ISV d/s1w58MHfzbL2kWNb/5xpv/+fIbr0PUKnHa6MEKjhacynViTnrOeP1F+prceTltOdfrb9ffkvto 2pOFLwhmvdPjco6uK3zPKXlZPWNyMTW4mnRN+iZDk7HJ1GReplumX2ZYZlxmWmbuD/fnWfPCuXm5 I8bmzjc0GheHF+evDq7O7cr9leE+0/b8uwvvHP2I4XHTw3mP4Pe/L4cdOIwnd4qcs0jwLJJ7FlFp OD9VGo6oNBxRaTiSiasMxZ5VPl+XFzIZRE8gnCEaR2V6+GElx13IV8nvrnTPcC9w73Yfcmusbr97 pfuIW/S7t7qZ+3ksYga8QdWPVuA3MLjPCmUyviYzQmXKL2YG+tId0aR/bbFFKR3VlLk8k2X6MrSw WvwwDZcjnvhCNekcUdK4kIm+UUa/h3py3UqaK1rMJaqIy6hb3X+4hgCHuCAMcNPkDvBabvUU7FZ9 aTcO4r3a3AJU3eMrP1xAgX2BfY3VAzkOxzCFcD4g5+u9XPUKPGpX2fDsW4oHilllcVcxK+Znglyi 9kmwK8H8BJJcZvUqwgfAEcXNBxHItapnSqs6PGuADxXKfUbBEIFZeIdW9Y7CmnOEUL7tMeIek3L8 mzumpXaoQUiuDFu1anrqEB+JdOAE8NP+pZ7YQFQ52KEe4rmvcUweUqPkST51kMfWpeSNzApK6YVh m2yX02RBk2MOeIk+X+ul0kgEWelIZluCXpITNJt0Iwxemp+nN2giopf45UxuByP8NjIZqFeDBZF1 69aduy2EceT+XnNamXrY4NcDefgOH8UtAN8Qfzo0OnFX4MzCFQFXxnBlr3XzddevLQ396pV7ZlSN K7hjzg3Pz7fFTJ1Lr1/mcBR5bzlw97ylr9xw6H16ge/KVW01FwRdoeIL102fcm2+PzL1uitcs5tm lwV9mWmG3JKq65vm77zk93y/zE18ywqke3CP+5d9xAAZDIaj8KpwOgPS5YbPbDIbqEAcsj5iNWgc uAmyyjkkh5rtIRNNaHGFXNuibdd2abdpRQLLtksb0w5oD2s1sH7fqJsWkFOqFAH5Vr3eAcL9JZhQ jnBJA8J3uKTNVIyq9dTw9UdB0upr97NlxEXH9lz+c4cazDx1TB7E/iUfO4WdarASqA1uiK2kRH6N u5WRSMjJ+Rcu5acOWxm8j6AtnbOeyZ6LKxYuL7zllr49e9Ii+VkP7pQntj3EFm2h2uXDt20Z+tW0 QnwCgP8NW3aU/w6JzthHPOCNHp41C6Q5opDWk0qJPT0aSaO5ujSHiaY5jDiz2cAmUuIIuZx8u/eo voRT9SKcdu5F4AordW3mVPdx9UpL9R+c6ZwLSP8z6T84VYcQ6dOKmfMj4aQDTuqc7uFrlMFdB89J D2v37PLEPAmP6DGF1NWT+Y6PE48+oD+sP6oXkZl04jmi2PgY9GrPejjxmnrknlYcnOt61XfQM668 +unu81x2+AaD/91JqBg6xvleWaFehyX9P48oW8xWM9No8bNVSQdHQTR5iVln8+LWHCfBgnWkmatG dqm6NHlYnBJbulNViLEcFyqvf/eyh2fIxn6j7apZs26f0H9f/9QVM0o72fahvtvGTJk1Z+smVo7j EeVfaYTjWB0D/Tp1r+WUdMSg01CNgUh6nUSZlMsNilQU+eig/NFBiAZOYJXcm/E+UypRkmMrN3D7 braV6+EGRnU8wOX1132IYZDVGBR/UfRZ2VGSjwCp44oeJy3iQIDUB8qN+aOiJIDAahpB8vFhpZyU GqaSKYZ5+FFUo65Bfzm9nC3VLdWvJWvoGnatbq1+jWEj3cg2CJu1m3Td+vvJb/R3GH5PHjI8T57R 9hheIy8bPiDvGv5OPjP8QE4ZCjEdg4s4DPkkbCgzzCCKQS8pdkdUgqhEezTqnZ4e8+FTJxjTKcXK l9FAVBvKecHz4F7gmh1cUXOZJJmMsIFFH0XAG8DByMEIKaqsRB6/1SszaHW6kN6QrtcbcP3A4JWm 4wpCMhgIfhehY4xqtAa9QKhUZKKmHJ2iKDh5M32cevcoOGoyCRi+3DCF5hi/fovr7qDHPdQ81Oxx DR5rRi9Q1nK4mMkvOLbyjdKoyMYbXto4ysWjRtyspW5WIS6phzQ3ZtOSNBwgytLwVeXp4eV/OBbC bfnf9w1fJYaHbrli5dxr2CZ+dsYplj+JPPxO6t89NmQK+KVaGL8EnEym4ndwF2OTmYnfV83G/ypd QhpIIygosQP4o8E3NTJzWvWM+ZdEqlYtbV0+be7/BYuwyZcKZW5kc3RyZWFtCmVuZG9iago1OCAw IG9iago5NzM4CmVuZG9iagoyMCAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5 cGUgL0Jhc2VGb250IC9SQUtGS0srQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNTkgMCBSIC9FbmNv ZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDU2IC9MYXN0Q2hhciA1NiAvV2lkdGhz IFsgNTU2Cl0gPj4KZW5kb2JqCjU5IDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRvciAvRm9u dE5hbWUgL1JBS0ZLSytBcmlhbE1UIC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMyNSAyMDAw IDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0NhcEhlaWdo dCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2Z1dp ZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUyIDYwIDAgUiA+PgplbmRvYmoKNjAgMCBv YmoKPDwgL0xlbmd0aCA2MSAwIFIgL0xlbmd0aDEgNzMxMiAvRmlsdGVyIC9GbGF0ZURlY29kZSA+ PgpzdHJlYW0KeAGFWQt8VMXVPzNz95Fkw24CeS/Zu1mymGxCIIDh1WTz2ACNyCPB7tJgEiASECQQ 8FWEReW1gCCfpUqtUNsq1U+52US6AVqiWG1RhM9aWx8VfLQ+fkWoP8WqmPv95+4C4uev3735nzlz zpmZM2fOnTt3s3LFqnayUZgE+ecvbesk48pZgCJ3/s0r1Xg9NZvI3HRD58Kl8frgJUSmjxcuue2G eD1nExH/XUd7m2wnrwvA1R0QGDViY1AO61i68tZ4PacZpXXJsvkJfU4YdfPStlsT49ObqKs3tS1t R4nLKeVq57KulUaVnFNQXtW5oj1hz4JEKb/7KvbGE+wj9vyDcSMiBobTJzSJfkYWcA4qo+uIlA5l KJlQl3oT377zk63nW+yTPrPmWY2WD787vFgyR5XOHV/u/3qhg6w2VJMMe6lAO0vlwLVU66Av9395 uyM+ktRcvHgfNYmrerzZrpOHRRGdBrgoivqGuvrEcDE0OtHljwlPT3pGub26VKjoscygKugyYD9w BFCoReRD6wBdC4SB/cAR4CRgJgKVWhVYBuwBTgNmMVQ4o6rLUT1c5KBtDuZrF1l0FtABQS7QMmA6 0AJsB/YAZsNOSpYBa4EjwDnATH6RFd05Gr5nRbcYRc/iJeVGtS1ebZ5rVHt+EIqX02bGy7qpcbMJ cbNRY+LiETXxcnhJvEwvLA+j857k1PL+6kyRiUlmwvFOUMafJTtj5KK9IoM0gAu4akj8Ir1nmLd8 zxGhEBNcMFpALr1fsGhqWnl1Mtf5WUonF/+Yn4lr+JmeQWnle6q/z9+h/cARQPB3cL/N36a1/LSM OWgVsAc4ApwAzgJmfhr3Kdxv8bfIzv9GZUAV0ALsAY4AZwEL/xuog78pM8agkq8COH8T1MHfwLTe ALXz18G9zl/X+/mfohXjy/sMxleWYFyFCSYrL8GkZ5bH+MvRL4qQUV6sNDLqkCigShotCqKFo1wx kR2dtMgV4+/2qD7X3uqR/BXSAA5PXsHIr5AKzABagU7ADO5VcK9SGNgB7AU0AFkG6gBUfgx4EXiV RgJ+YAZg5SejGCbGT0S9Na7qTP4Sf56yEPHj/A9G+SJ/zihf4L83yj+izIf+GH8umu+i6hToCW0c KB0oy6A38ad7hqW79Oo0fgQRdIGWAVXAdKAF2A6Y+RFeEF3gSkcnh+gYnmEXj9KHRvkIPWwl/2KX 31uLBFQl8U74HjiQPeoeL/d7dz2AqiTee3aCk8R791ZwknhvXwdOEu+Sm8FJ4l2wGJwk3jkt4CTx Tm8CBxLjD/1m2HBXxfQbmVpt57cgSrcgSrcgSreQwm+RN32hSB9/Gi0uRsR2+31Fxa7wQRY+zMKz WPhhFm5n4TUsvI6FJ7Hw9SzsY2EnC+ezsJ+FD7FxCEWY+XuvqI73Z7PwMRZ+goW7WNjLwoUsPIyF VVbhj3F3dCqeOhQBo+iplg8dd/d8rxK7j527EVE3ct6NPeEI6AlAN2p+GKkFceOcfFkW9BRXxesj JpQvq57Cj6LhUSzDUToFKFigo0ijo+jkKLqzg1YBLUA/cBbQATOsCzCP7Qa1g5YBVUALsBY4C5gN d87CFU7LQKWL+w3HykCrgOmyxo/iLsDt5m7/UIfT4XNMEdudzJ7Ppufr+byCMjOxL6enWdNiLPXA 56n//jyVkqqT+D18Ow3FQuxIlNujXwx1xdj9Ue8hV3UG+wnlK8g6Np68rBDlOOoy6mPJaZXyMeTk j6MsjzqvQzN71FviOsgGyVYHXF8433N96IxxsB84D7n+osYUFnX9GZLHD7hecW52/bEsZoXksDfG UBxUDdM+5zjXE8cM03VQ7I661sjigOsO52TXjU5D0R5XXN+Fmt/umuWd45qC/uqc81z+LvR5wFXl vN41KW41VrY54BoJF3xxthjOFjmNQT35RoezK2Ksw19i2WUJWqZbrraUW0osbovLMtSSZxliTbc6 rIOsNmuy1Wo1WxUrt5J1SEw/7ffJt94Qs/HyMyOhGSkG78AOw+Q2A0qcWTl9n7TBooE3NNawBq1/ PjXMU7XzjZ4YS545RzN5apiW3kANTTXaOF9DzKLP0ip8DZplxg+D3YzdE4JU45tijJqCMaZL0fo8 Lb022EeMpa3flifLq9ZvC4UoO/Pmquyq9Mq08fV130FaDWFrne/ylX2Z9WX7hmq7GhqD2mNDQ1q5 ZPShoQbtvxrV5mAf+4SdC9T1sX/JIhTsE5Xsk8AsKReVdaFQQ4xdZ9iRyv4FO2QMCthZ8WKWdqRa 8+N2u+N2hWgPu2GygF1SEhUadoVJSYadwqRdd9ewQF33MBDYZKnUZdh0ZanftDlWCJtCENhkhumY YXMsMyxttEqjG6cTJvkgMGG55DRMnCzXMDE87zZMyhImmy+ZbDZGEnFvDBtJ0E3q6Ys2qadh841A /me2vcbnYz0TQ/ObA+2eQKsn0A60altu7sjWwvNUtXt+SCpUTXhb583vkGVbuxbytNdp8z11avdE o9231M1SPdFT103NgaZgd7O/vS460T8x4GmrC/VMnjGm4oqxNl8aa8yM7xhrhuxsjBxrstHuW2NV SPVkOVaFHKtCjjXZP9kYi4wcnxHstlJNqBbrJ8senpKMfG3Nc4dqMh2dlUbyTnRnr8k7iNPKPkrx hTSbp0ZLBWRel1aXVksVnimpGgSxPaHKXjPRnXeQ7UuoHBCneWrIt3JV1yrKDiyqi/914YJo5Sq5 FHHqk7LvvGAS0PxtdfJs3aAVNzZoVTPnBLstFkhb60KQTbgoS0kJxPT+uHAEhBOkoRCXDKVskpQl JSUM/28uGD5BjOj04aBxqIf589lK6goJLb+hiWMraJqDMDTPCR7EWUq+JLpCmGAX87Gui73JeRg8 xSWEaXddxMpVCS4Ri5WJ0jDt8pGv62JILnbnk8EyiBGrlT5sbaaDlAPkmh6lHMVL+P7R3wc+kOXA Iv0DqZcl/wgbXSwBon30BFtET9AReoadQ6v91Ee9JI9AdfQgrab7aCNea3Mg2UyzcJsgv4/l6L34 Mvk5Xpg/p+Ow/QGtoYOUybL1D2ktrRd/Qqv1lEoFVE0zaBltY9foq6iZTil3UQVdQzdRJwvrQf0e faf+S/oV9Yk/6F9TCuXSfNzH9Y9Nf9XfpFK0+DE9QKfYzqSnyI9RwrD8Ga2g3WKuwvSF+pfwwE23 wAeFptFx1s996L2d3mfZbLWoRS+/0DX9WVg5aS510G46yMayydxtatan6ccpE2Pcil4foCgdwB2j 39LrzGY6p/9SP0c5VEJTMZ9eeon1i4Gv1w1UIW4mRKmIxkOzjH5Hz9NJ5mFP82Umm6nc5Dfdrr9C Q2gUzYa3j6LlP9jnfA3uteI5pV6voUGIy70y2vR7epvlsjI2nV3Hi/gy/pBYQVaMOAr3AlqEeN+P 3t9CGh3gNn5C/EJ5XPnKPHTgtD4IK+Kln9LP6GmWipmqrIvdyV5l7/Ja3sJ/yt8R9ym/Vl62tGHW 19NS2kaP0+csnY1jM9kPWQdbzTaye9kD7Dg7yT7g1byJ38jPig6xXPxWqcHdqHQpd5k2mLaYPxgI Djw78D8Dn+vl+gaaiXxYB+9/TA9hZn10gl7DfYreYSaWwgbhVpmbzWY/wr2GbWMPs33s16wXo5xk 77AP8Ur6jH3F8ablZp6Hw488Ann4Cpww7+MP8hO4T/J/8i9EligQPjFWTBIhsQxebRQ7cD8l3lZy lROKjjiXm3aZ9pj2mR43PWM6Z7ZZ7sQ7/sULv/i6+Ou3Bmhg08CugehAr/42ZWAN8fbAJ9gkeN+G ezHWexcybj/9idkQu1xWzCrZNYhMC1vMlrNbEcm72W72K8P3J9lhROkv7Cx8TuVOw+cRfCyv4dNx X8/b+XIcxnbyXv4q/1JYRIqwiwxRLCaLuaJdrBS3iV1CEy+Kv4l3xHlxAbeuJCsupUDxKj5lstKi rFIeUt5X3jc1m14w/d2cbF5q3mCOmf+FU02lZYZlpmWuZbvlgOUVayuy8yg9Rb9BBl662GmxTgTE U3QPH63k4BPmJeRzCy0Q0zgyle9jm/gdrJcPM91qnsgnsmvpnOJFrJ/je/h5PlFMYw2skRbzUfEO zUOUx8BNUo7SGeUw5vYSer7VbGNr+FmzjaI4I43HGen3YqTiEy/Q6+IUsyg/pzeUZJbFzvBHxQxk wW+VSlOQ3OJBelIsZ3fQUzxAlPyVdSvy+Fr2GPaFJlbO/i10HIOvRRZViHfpLrqR/5XO4DneRD9h C5SFdA+NZqvpfXoET0WR6SZzsTmD/ZEvUiJ8MOslrvwasxvPhjFhGkJ3s7lit/ksf41W0Qklmd4S /w3vT/AnxTTlnGkW68ATcAdtoOX6OrrNFFReZgtJsOuoUDmN3W21KFfcKNdiV2nGnnYAT/dB7APV Yhok2cica5AXs7FD7MZ9P/YJBRm0CM/4D7CLvUS95iYeo4WmQQy7Dn6peWFgFs3RH6EH9IV0k76T SrEfbNRXo8d99HfaTvvY+oEfUSc+JV/Ds32NqZ6fMNXrpTzCX+ONfNeV64toF7Js+gj3k1iZStMh iih/oUaq0rfqf0Z2X4Ud9gGahwPre5jlxxhhiuin0QPX8m69XnRivqdopv6o7mLJ1KEvoel0mH5l MVGbxYc11tjLmO+PqJ3P0leK9oFFiMN2RMGPaK3C/rPZXzu7qdpfVfm9SRMnjB9XMXbM6PJRI8tG lJb4iouuGu4tHOYpcKuu/KHOvNyc7KzMjCGD09Mc9kGptpTkJKvFbFIEZ1QS8NS3qpq3VVO8nilT SmXd0wZB2zcErZoKUf2VNpoq27VBdYWlH5Y3fMvSH7f0X7JkDnUSTSotUQMeVTte51FjbM7MIPht dZ6Qqp0x+GkGv8PgU8G73WigBrI76lSNtaoBrf7mjkigta60hHWnJNd6atuTS0uoOzkFbAo4LcvT 2c2yKpnB8KzAhG5O1lRMUcv11AW0HA+aohtRGGhboM2YGQzU5bndodISjdXO98zTSJ6UfIYJ1RrD aOZazWIMoy7CGUejLWp3SX9ka8xB81p9tgWeBW3NQU20oY+AlubDuHVa1u3vZV+uonOcyTZ+U5sn IoHsRao0jkQ2qtremcFvtM1zyx5CIfSBtrywvjVSj6G3YqUa5Flc4+tDQY2tx5A4WBYas4rPL37q LWxdrGpJnhpPR2RxK5YmN6LRrNvc0dxcf59+mnIDaqQp6HFrVXmeUFuds3sIRWbd1pPjV3Ou1JSW dDvS4oHtHmRPMLbUbzLtCHpcZ3CGueQaZl2KLJM+eqbiJKip81V4EvRgTuMkaR9HkfnjsAC4Qgyt tAVYkUVaUm1rxDFByjFFppkKHR418hkhAzxn/nmlpC0hMRc6PiOplHlyKdU01naR13w+rbhYpoil FmsKHyuN+tjSkptj3OPpdOD7WX400AzEti00oQzhd7vlAm+J+WkeKlp4ZjBeV2leXpT8ZThb81ap 6b+oyZgtNeGLmkvNWz3I5F75PUsZmtV76c/uyBwc6Jigscz/oG6P6xsaPQ04GquBSGsiaxuarqjF 9TKgiBt0CU4bXBsUeRwyyfE8YWjjJ+SLJjguB22aUog/s5HUC2IWK7LSkDC1XnO0TonTULLbnXhm /r9GMf2cbGUUl5slpqFN8CUcjbutTbyifoV7tohoaMKWw3Gyj0SSr9Ah1eJeTk0UyHh86LvVWo1m 48ksxB8+OcZJhPI0P0IGTROeIkMcyktUrzDMSzQK4ZLZWVpSjz0zEqn3qPWR1khbTA/P86gOT6SP P8OfiXQGsNvFEyemH9ySp9VvDSFiHWwCHg9ONd0etmlmt59tapwT7MNPHOqmpmCUM17bWhOSy8Jr m4KJsBgLIlMfa4h/aCBjlC68XwiMTB7Cf1rMAJH8ES8uQUqxFkiwtcvDsAk/6ON/FeROc6cVgqAH uqCK/gt+E31FqtIv9Y36+0qOqR9fCh78vkjdXLrQa6O8/BEIlD9r8GA+e8SIdHe+2XRVfnpqfpLN ZuOzk2L6pwdSU8H47FjVXrudzwbzQW9KSpwxlPZsIZUOB59tMNIKzKf+HNlUDMuwSfMMo8cMm5Rl lOFD6vgV5DhVnak6kz6+zHdm/PhRI/O6zdLD3xiOmC86AkY68h4cMUtH3sP4YOLjSxmGveAvkEI5 rM1mxrCylmHM9PL8Lg6GsRjckK5cAgb3V4zNZEWZUzOnev9h+3CkKWkkDlJ3sNXKSuvylBW2Vam3 Z22hCNuqbLCuS7nbtiF1W9aLac8NTi/Ar1xRp5orC1Utk0Wp6pU/feUXqTbKzyYbAr53BLvsSX7X kSSWFOML/Q5fl92vesaMtDOyO+zcHmP3HijP7tIEE9BHh3VlyKUa5PaMUTP8GTxjx6jnt2T7HOfn +s58OveM49O5yw0mMbe5ZXJyaekymmxc4mJzVyyn5aEQ83rHjrl6dDlODGZLJih5CgiSwUMyR5df PXaM11NgFkMyL1fY4s4l/zjS/9GNSzduGzj/2msD5++dt+HGjvWbb1i4acLUHY3r9j1x59pHRV7R /Yv3vn5q7w0/KSp5dtNhHT+09W9/mjV13H1Xy/yNd1/Qp+2Y/kj4zsf2IXONH/xQ6MPxPfBdF4cw z1AwSk/kvhnfbzSruqG+ocFXvWJR25JpTf8LFvigewplbmRzdHJlYW0KZW5kb2JqCjYxIDAgb2Jq CjQ5ODEKZW5kb2JqCjE0IDAgb2JqCjw8IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVlVHlwZSAv QmFzZUZvbnQgL1RUWkZWRCtBcmlhbE1UIC9Gb250RGVzY3JpcHRvcgo2MiAwIFIgL0VuY29kaW5n IC9NYWNSb21hbkVuY29kaW5nIC9GaXJzdENoYXIgNTAgL0xhc3RDaGFyIDUwIC9XaWR0aHMgWyA1 NTYKXSA+PgplbmRvYmoKNjIgMCBvYmoKPDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250TmFt ZSAvVFRaRlZEK0FyaWFsTVQgL0ZsYWdzIDMyIC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAgMTAw Nl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0IDcx NiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMgL1hIZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lkdGgg NDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250RmlsZTIgNjMgMCBSID4+CmVuZG9iago2MyAwIG9iago8 PCAvTGVuZ3RoIDY0IDAgUiAvTGVuZ3RoMSA3MjA0IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0 cmVhbQp4AYVZCXxU1dU/9943S/ZJAlknM28YMkgmMRKWAEmTyTIBG5Et6AwNkpBEAoKJBNyFoS0C A4qllhZqxaqtqB/yMgk4AWxSsVpRhCq1dang9rn8ilB/FetC3ve/bwaU1l+/d+d/7rnnnLude959 971ZuWJVByVTiAT52pa3dpNx5bYjy2u7caUaK6fkEJmbru1evDxWzlxGZPpk8bJbro2VczcQ8abO jlZZT15fA5M6ITBKxCYgH925fOXNsXJuM3Lrsq62uD43hLJ5eevN8f7pTZTV61uXdyDHZVdA1O6u npVGkfKl/pLuFR1xexYgSvrdV9E3drOP2XP3xYyIGBhOn1Il/Yos4GxUSlcRKZ1KAZlQlnoT37L1 081nF6ZVfmbNtxo1H3x3TJFkDind93y559xiG1mTUUww7KUC9SxVw1dSnY2+3PPlrbZYT1Jz/uID 1CQu6fPkOI8dFGPpJMDF2Ii3wDkgxoiCSIXTFxXuvoyRZWk1JUJFi6UGVUG7gD3AIKDQQuGA1ga6 BggBe4BB4BhgJgKVWhXoAnYCJwGzKBD2iOq01YwRuaibi/mmiWw6DeiAICdoKTATWAhsAXYCZsNO SrqANcAgcAYwk09kR7aOx9izI5uMrG/psjKj2BorNi8win1XB2P5jNmxvP7ymNnUmNm4CTHxpbWx fExxLM8oLAuh8b7ElLKhmiyRhUlmYeDdoIw/Q2mMkZMeECNJA7jAUA2JT2T0jfaU7RwUCjHBBaN2 cupDgkVS0stqErnOT1MGOfkn/FRMw0/1paaX7az5Pn+H9gCDgODvIL3N36Y1/KT0OWg1sBMYBI4C pwEzP4l0Aukt/hal8b9RKVANLAR2AoPAacDC/wZq42/KiDGo5KsBzt8EtfE3MK03QNP46+Be56/r Q/yVSPmUsgGD8ZbGGWdhnMnOjzMZWWVR/nLki7GIKA9WGhF1QIyiKhovRkUKxzmjIidSucQZ5e/2 qV7nAzWX8eOkARwjOY6ej5MKzAJagG7ADO5VcK9SCLgHeADQAEQZqA1Q+WHgReBVugzwAbMAKz8W QTdRfjTiqXXWZPGX+HOUDY8f4X808hf5s0b+Av+DkT+P3AH9Yf5sxOGkmiToCXVsyG3IS6E38d/3 jc5w6jXpfBAedIKWAtXATGAhsAUw80E+KtLuzEAjB+gw7mEnj9BHRv5betBKvqVOn6cOAahK4pn6 PXAgO9WdHu7zbNuOoiSeu7eCk8Tz483gJPHcuhacJJ5lN4KTxNO+FJwknvkLwUnimdkEDiTK739y 9Bhn+czrmFqTxm+Cl26Cl26Cl24ihd8kE32B7czJfxkpKoLHdvi8Y4ucof0sdJCF5rDQgyzUwUKr WWgtC1Wy0DUs5GUhOws5WMjHQgfYZLgixHz9FxWn+HJY6DAL7WahHhbysFAhC41mIZWV+6LcFbkc dx0yv5H11cibjrv6vleF3SeNu+BRF2LehT1hEPQooBslH4zUUTHjXIfMR/UVVcfKl04t66qZzg+h 4iEswyE6AShYoEMIo0No5BCaSwOtBhYCQ8BpQAfMsB6FeWwxaBpoKVANLATWAKcBszGc0xgKpy5Q OcQ9xsBKQauBmbLEDyGNQnJxl6/AZrd5bdPFFjtLc7CZDt3ByykrC/tyRro1PcpS9n2e8q/PUyih JoHfzbdQARbinni+JfJFgTPKfhHxHHDWjGQ/J4eCqGNTyMMKkU+mHqM8kexWKZ9Adv448rKI/SpU S4t4ip37Waqstc/5hf0950f2KAf7of2A8y9qVGER558heXyf87h9o/P50qgVkoOeKEO2XzVMB+yT nbsPG6ZrodgRca6W2T7nHfZpzuvshqIjprimByVfmnOOZ75zOtqrty9y+nrQ5j5ntf0aZ2XMaqKs s895GYbgjbFFGOxYu9Gp22E0OK88yjp9xZZtloBlpmWSpcxSbHFZnJYCS75lhDXDarOmWpOtiVar 1WxVrNxK1hFR/aTPK596I8zGw8+MgGakGLwNOwyT2wwocWbl9H3SMkUjb5xbyxq1oTZqXKRqZ+e6 oyxx9nzN5K5lWkYjNTbVapO9jVGLPkcr9zZqllk/CPQydncQUo1viDJqCkSZLkXr8rWMusAAMZa+ 7q58mV+y7q5gkHKybqzOqc6oSp/SUP8dpMUQttR7v7lyvmG9Od4CbVvj3ID2WEFQK5OMXhBs1H46 V20ODLBP2Rl//QD7h8yCgQFRxT71z5FyUVUfDDZG2VWGHansH7BDxCCDnRUPZmlHqtURs9sRsytE fdiNlhnsEhKo0LArTEgw7BQm7Xp7Rvvre0eDwCZbpR7Dpidb/bbN4ULYFILAJitEhw2bw1khaaNV Gc3Y7TBxgMCE5ZHdMLGzPMPEGHmvYVIaN9l4wWSj0ZOIjcawkQTNpJw8b5NyEjbfcuR/ZztqvV7W VxFsa/Z3uP0tbn8H0KJturEzRwstUtXetqBUqJrwtCxq65R5a4cWdHfUa23uerW3wqj3b+pmqa5w 1/dSs78p0Nvs66iPVPgq/O7W+mDftFkTyi/qa+OFvibM+o6+ZsnGJsi+phn1/q2vcqmeJvsql32V y76m+aYZfZER47MCvVaqDdZh/WTex5MSEa8t+a5gbZatu8oI3gpXzur8/Tit7KIkb1BLdtdqKYCM 65Kakhqpwj0lVakQp8VVOasrXPn72a64ygZxuruWvCtX9ayiHP+S+tivBxdEK1fJpYhRr5R95wUT v+ZrrZdn60ataG6jVj17fqDXYoG0pT4I2dTzsqQkf1QfigkvhXCqNBTigqGUVUpZQkLc8D9jwRgT xPDOAA4aB/qYz8FWUk9QaI7GJo6toGk+3NA8P7AfZyn5kOgJYoI9zMt6zrcm52HwFJMQpt1zHitX xbm4L1bGc8O0x0venvMuOd+cVzrLIIavVnqxtZn2Uy6QZ3qEchUP4f1H/wD4UObDS/QPpV7m/GNs dNE4iHbRbraEdtMgPc3OoNYeGqB+kkegerqPbqd7aT0ea/Mh2UhzkEyQ38ty9X68mfwaD8xf0xHY Xk2raT9lsRz9I1pD68QrqLWOUmgU1dAs6qK72BX6KmqmE8qPqJyuoOupm4X0gH63vlV/mH5DA+KP +jlKojxqQzqif2L6q/4mlaDGz2g7nWBbE/aSD72EYPkrWkE7xAKF6Yv1LzECF92EMSg0g46wIe5F 6x30Actht4s6tPKQrunPwMpOC6iTdtB+NpFN4y5Tsz5DP0JZ6ONmtLqdIrQPKUpP0ess2XRGf1g/ Q7lUTJdjPv30EhsSw+fWDlfDbyZ4aSxNgaaLfkfP0THmZr/nXaZkU5nJZ7pVP04jaBzNw2gfQc3/ ZZ/z1UhrxLNKg15LqfDLT6S36Q/0NstjpWwmu4qP5V38frGCrOhxHFI7LYG/f4HW30IY7ePJ/Kh4 SHlc+cpcMHxST8WKeOiX9Cv6PUvBTFXWw37IXmXv8jq+kP+SvyPuVR5VXra0YtbX0HK6ix6nz1kG m8xmsx+wTnY7W89+wrazI+wY+5DX8CZ+HT8tOsUN4imlFmmu0qP8yHSnaZP5w+HA8DPDfxr+XC/T 76TZiIe1GP3P6H7MbICO0mtIJ+gdZmJJLBVJZS42j92GtJrdxR5ku9ijrB+9HGPvsI/wSPqMfcXx pOVmno/DjzwCufkKnDDv5ffxo0jH+N/5FyJbjBJeMVFUiqDowqjWi3uQ9oq3lTzlqKLDz2Wmbaad pl2mx01Pm86Yky0/xDP+xa8fOld07q1hGt4wvG04Mtyvv00jsYZ4euAVrBKjb0VaivXehojbQ6+w ZPgujxWxKnYFPLOQLWU3sJvhyR+zHew3xtifYAfhpb+w0xhzCrcbY76UT+S1fCbSNbyD34DD2Fbe z1/lXwqLSBJpYqQoEtPEAtEhVopbxDahiRfF38Q74qz4GklXEhWnMkrxKF5lmrJQWaXcr3ygfGBq Nr1get+caF5uvtMcNf8Dp5oqyyzLbMsCyxbLPstxawui8xDtpScRgRcudlKsFX6xl+7m45VcvMK8 hHheSO1iBkek8l1sA7+D9fPRppvNFbyCXUlnFA98/Szfyc/yCjGDNbK5tJSPizVoHqE8Bq5SOUSn lIOY20to+WZzMlvNT5uTKYIz0hSckf4gLlO84gV6XZxgFuXX9IaSyLLZKf6ImIUoeEqpMgXIJe6j J8QN7A7ay/1EiV9ZNyOOr2SPYV9oYmXsX0LHMfhKRFG5eJd+RNfxv9Ip3Mcb6OesXVlMd9N4djt9 QL/FXTHWdL25yDySPc+XKGGeyfqJK49idlPYaCZMI+jHbIHYYT7NX6NVdFRJpLfE/2D0R/kTYoZy xjSHdeIOuIPupBv0tXSLKaC8zBaTYFdRoXISu9vtokxxIV+DXaUZe9o+3N37sQ/UiBmQ5CByrkBc zMMOsQPpF9gnFETQEtzjV2MXe4n6zU08SotNqQy7Dr7UvDA8h+brv6Xt+mK6Xt9KJdgP1uu3o8Vd 9D5toV1s3fBt1I1Xyddwb19hauBHTQ16CQ/z1/hcvu3i9YW3C1kOfYz0BFamynSAwspfaC5V65v1 PyO6L8EOu50W4cD6Hmb5CXqYLoZo/PCVvFdvEN2Y7wmarT+iO1kiderLaCYdpN9YTNRq8WKNNfYy 5nsbdfA5+krRMbwEftgCL/jgrVXYfzb66uY11fiqq75XWTF1yuTyiRPGl427rPTSkmJv0dhLxngK R7tHuVSno8Cen5ebk501ckRmRrotLTUlOSkxwWoxmxTBGRX73Q0tquZp0RSPe/r0Ell2t0LQ+i1B i6ZC1HCxjabKeq1QXWTpg+W1/2bpi1n6Llgym1pJlSXFqt+takfq3WqUzZ8dAH9XvTuoaqcMfobB 32PwKeBdLlRQ/Tmd9arGWlS/1nBjZ9jfUl9SzHqTEuvcdR2JJcXUm5gENgmclu3u7mXZVcxgeLZ/ ai8nawqmqOW56/1arhtV0Ywo9Le2a7NmB/z1+S5XsKRYY3Vt7kUayZOS1zChOqMbzVynWYxu1CU4 42i0Se0tHgpvjtpoUYs3ud3d3toc0EQr2vBr6V70W69l3/pezjdFNI4z2fpva/NF2J+zRJXG4fB6 VXtgduBbdfNdsoVgEG2gLi9saAk3oOvNWKlGeRbX+LpgQGPr0CUOloXGrGLzi516C1uWqlqCu9bd GV7agqXJC2s05xZXJC/PN6CfpDy/Gm4KuF1adb472Fpv7x1B4Tm39OX61NyLNSXFvbb0mGN7U9Pi THLKt5kOOD2mMzjDXHKNcy54lskxui/HSVBT21SMJODGnCZL0jGZwm2TsQC4ggy1tHasyBItoa4l bJsq5Zgi00yFNrca/owQAe5Tf79Y0hqXmAttn5FUyji5EGoaaz3Pa16vVlQkQ8RShzXFGKuM8sSS 4huj3O3utuH9Wb400Cz4tjU4tRTud7nkAm+K+mgRClpodiBWVmlRfoR8pThb8xapGTqvGTlPakLn NReqt7gRyf3yfZZGalbPhV+aLSvT3zlVY1n/Rd0R0zfOdTfiaKz6wy3xqG1suqgU00uHwm/QxTkt sy4g8jlkkuP5wtDGTsjnTXBcDiRrSiF+ZiOo26MWK6LSkDC1QbO1TI/RYKLLFb9n/r9KUf2MrGVk 31SLT0Ob6o0PNDZsreKi8kXDSw6LxiZsORwn+3A48SIdQi02ysvjGSIeL/outU6jebgzC/HDK8dk iWC+5oPLoGnCXWSIg/nx4kWG+fFKQVwyOkuKG7BnhsMNbrUh3BJujeqhRW7V5g4P8Kf50+FuP3a7 WOBE9f2b8rWGzUF4rJNNxe3BqbbXzTbM7vWxDXPnBwbwiUPd0BSIcMbrWmqDcll4XVMg7hZjQWTo Yw3xhwYiRunBWYnAyOAh/NNiBojKLkgQUkz+24CtXR6GTfigj/8qyJXuSi8EQQv0tSqGvvaZ6CtS lSHZFlpULKYhcrDnes2ye19Cui0lJzPTPC8lqv+zPz3dYD7xJdhs4BwjTI6o/qEvWxo4HFLrsKdC 40hOSQGN8gO+ZJ6YnY1/C9I5V53pGVNKjx+R9AiVnpKvSdWSPoPHVX4vv9BhckYGNzr0JaSlg4v1 c9KXlJHJ5zlGSJlsO4Km0X1/UhKfB+bvvtSUFHDf0ZvXG+tP9mZ05ptUYaowHzANmg9YnrM+b7dc nhxMbkq9Lrk99daMWzM3ZhzMeD/v/fwzecmDSU9m8nx8/CuwOWzm3+F1w4Ld0oo8QT/jy3Mk2qxm 82F73gi7Pc9qzxOMW/PsIsVhi/KH+2amM3wazNkrZ0CGO9IYT07syX4F3va53BPYAb6WVLKxyb7k 9L3VeC3o4mu4wvfz0fgAuKV3U47XdnbBP0/Zznorbf88RdWV5yqrT51b8F56RvYUJsn61Eu9qXfY nkFhyrjLaDIuJgktYAtWBIOFI12e8kmT8Js4weMeZbaMmTRpfBmOAGazBT/F8nU5zy58aMfpXdtv ++F9bCDzX3965ez0R55+sNmxe3dNZdvQ6mfev/a6n94Xzjz62se7A48dfHhD6zjjIxyiSh+DM/p3 XRzCkYaCUUY8Qs14p6KmpvkNV9V7a1YsaV02o+n/AGzabeUKZW5kc3RyZWFtCmVuZG9iago2NCAw IG9iago0OTMyCmVuZG9iagoyMiAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5 cGUgL0Jhc2VGb250IC9GREZNUk0rQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNjUgMCBSIC9FbmNv ZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDQ4IC9MYXN0Q2hhciA0OSAvV2lkdGhz IFsgNTU2CjU1NiBdID4+CmVuZG9iago2NSAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3Ig L0ZvbnROYW1lIC9GREZNUk0rQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUg MjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBI ZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9B dmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiA2NiAwIFIgPj4KZW5kb2JqCjY2 IDAgb2JqCjw8IC9MZW5ndGggNjcgMCBSIC9MZW5ndGgxIDczMjAgL0ZpbHRlciAvRmxhdGVEZWNv ZGUgPj4Kc3RyZWFtCngBhVkLfFTF1T8zc/eR9yaQ9yZ7N0sWySYGAhggMdk8NqAReQW7S4PZACkJ DwkmgCjC0opgQKF8FoWqqH2ItcrNJtIN2JJKqxVF+ITSilYQbX38ilC/T62iud9/7i4orb9+d/I/ c+acM68z586d2XTfuqKNEilEgrzzl7Z2kvFktyOzz1/ZrUbLSVlE5qbvdS5cGi0PW0Jk+mjhktXf i5aztxDx59vbWhdEy/Ql8mvaIYiW2TjkI9qXdt8WLWc3I09csmx+TJ99F8rWpa23xfqnN1FWb2ld 2ha1t8t6aueyru5YeTryqzpvbYvZMz9Rwm8uRt54mn3IXnwoakTEwAj6mCrpYbIQJxuV0k1ESruS RyaUpd7Et27/eMunLSmVn1hzrUbNx98ZWSSZQ0rnti/2frXQRtZEFOMMe6lAPUvV0I1UZ6Mv9n5x uy3ak9RcesQANYmr+txZjmPPiVF0BuBiVNiT5xgQI0VeuMLhjQhXX1p6WUpNiVDRYqlBVdBlwF7g IKBQi8iH1ga6DggBe4GDwDHAjAnmG1oV+TJgN3AGMIs8YQ+rDlvNSJGNutmYb4rIpPOADghygJYC 04AWYCuwGzAbdlKyDFgHHAQuAGbyiszw9rEYe2Z4s5H1LVpSZhRbo8XmuUax7zuBaD51RjSvvy5q NilqNmZcVHx1bTQfWRzN0wrLQmi8Lz6pbLAmQ2RgkhkYeCco47+jFMbIQY+KdNIALjBUQ+IVaX0j 3GW7DwqFmOCC0QJy6IOChZNSy2riuc7PUxo5+Ef8XFTDz/Ulp5btrrmen6W9wEFA8LNIb/O3aR0/ I30OWg3sBg4CR4HzgJmfQTqN9BZ/i1L4X6gUqAZagN3AQeA8YOF/AbXxN2XEGFTy1QDnb4La+BuY 1hugKfwUuFP8lD7Ij4fLJ5YNGIynNMY4CmNMZm6MScsoi/DXwp+PQkS5sdKIqAOigKporCgIF45x RERWuLLDEeHv9Kkex6M1o/kJ0gCOkZxAzydIBaYDQaATMIM7Ce4khYBtwKOABiDKQG2Ayg8DrwAn aTTgBaYDVn4sjG4i/GjYXeuoyeCv8hcpEx4/wv9g5K/wF4z8Zf57I38JeT70h/kL4XwH1SRAT6hj Q25DXgq9if+2b0SaQ69J5QfhQQdoKVANTANagK2AmR/kBeEFjjQ0coAO4x128DB9YOQ/p8et5F3k 8LrrEICqJO5J14ID2a3udnOve8dOFCVx37cdnCTuu7aAk8R9+3pwkriXrAQniXvBInCSuOe0gJPE Pa0JHEiEP/KrESMd5dMWM7Umha+Cl1bBS6vgpVWk8FUy0eeKHOOPw0VF8Ngur2dUkSO0n4WeY6GZ LPQ4C7Wx0FoWWs9ClSx0Mwt5WMjOQvks5GWhA2wCXBFi3v4rihO9WSx0mIWeZqEuFnKzUCELjWAh lZV7I9wZvg5vHTKfkfXVyJeOO/uurcLuk8Kd8KgTMe/EnnAQ9CigGyUvjNSCqHF2vswL+oqqo+Wr J5Utq5nCD6HiISzDIToNKFigQwijQ2jkEJpLAa0GWoBB4DygA2ZYF2AeWw2aAloKVAMtwDrgPGA2 hnMeQ+G0DFQOca8xsFLQamCaLPFDSAVITu705tnsNo9tithqZyn5bFq+ns/LKSMD+3JaqjU1wpL2 fZb0z8+SKK4mjt/Ht1IeFmJbLN8a/jzPEWEPht0HHDXp7AHKVxB1bCK5WSHyCdRllMeT3Srl48jO n0JeFrbfhGopYXexYz9LlrX2OT63v+v4wB7hYN+3H3D8SY0oLOz4IyRP7XOcsN/jeKk0YoXkOXeE IduvGqYD9gmOpw8bpuuh2BV2rJXZPsed9smOxXZD0RZV3NyFkjfFMdM9xzEF7dXb5zm8XWhzn6Pa frOjMmo1XtbZ5xiNIXiibBEGO8pudOrKNxqcXR5h7d5iyw6L3zLNco2lzFJscVocljxLrmW4Nc1q syZbE63xVqvVbFWs3ErW4RH9jNcjv3rDzcbHz4yAZqQYvA07DJPbDChxZuV0PWnDRCNvnFXLGrXB +dQ4T9U+neWKsPgZczSTq5ZpaY3U2FSrTfA0Riz6TK3c06hZpn/X38vYfQFINb4pwqjJH2G6FG3I 1dLq/APEWOqGe3NlftWGewMByspYWZ1VnVaVOrGh/ltI0BAG6z1fP1lfs54sT562o3GWX/tFXkAr k4yeF2jU/muW2uwfYB+zC776AfYPmQX8A6KKfeybKeWiqj4QaIywmww7Utk/YIeIQQY7az6p0o5U a37UblfUrhD1YTdCZrCLi6NCw64wLs6wU5i06+0a4avvHQECm0yVugybrkz1mzaHC2FTCAKbjBAd NmwOZ4SkjVZlNGO3wyQfBCYsh+yGiZ3lGCbGyHsNk9KYyT2XTe4xehLR0Rg2kqCZpDOXbJLOwOYb jvzPbFutx8P6KgLzm31tLl/Q5WsDgtrmle1ZWmieqvbOD0iFqgl3cN78dpm3tmkBV1u9Nt9Vr/ZW GPX+Rd0s1RWu+l5q9jX5e5u9bfXhCm+Fz9VaH+ibPH1c+RV93XO5r3HTv6Wv6bKxcbKvyUa9f+mr XKony77KZV/lsq/J3slGX2TE+HR/r5VqA3VYP5n38YR4xGsw1xmozbB1VhnBW+HMWpu7H6eVPZTg CWiJrlotCZBxXVJTUiNVeKekKhnilJgqa22FM3c/2xNT2SBOddWSp3tF1wrK8nXUR/+68EDUvUIu RZR6pOxbH5j4NG9rvTxbN2pFsxq16hlz/L0WC6TB+gBkky7JEhJ8EX0wKrwawknSUIjLhlJWKWVx cTHDf48FY0wQwzsDOGgc6GPefNZNXQGh5Tc2cWwFTXPghuY5/v04S8mPRFcAE+xiHtZ1qTU5D4On qIQw7a5L6F4R42K+6I7lhmmXhzxdl1xyqTmPdJZBDF91e7C1mfZTNpBjeoKyFTfh/qO/B7wv86EO /X2plzn/EBtdJAaiPfQ066Cn6SA9zy6g1l4aoH6SR6B6eojW0P20EZ+1OZDcQzORTJDfz7L1ftxM HsMH8zE6Atvv0FraTxksS/+A1tEGcRy1NlASFVANTadldC+7QV9BzXRa+QGV0w10C3WykO7X79O3 6z+ln9GA+IP+FSVQDs1HOqJ/ZPqz/iaVoMaPaCedZtvjniUvegnB8mG6lXaJuQrTF+pfYAROWoUx KDSVjrBB7kHrbfQey2JrRB1a+Ymu6b+DlZ3mUjvtov1sPJvMnaZmfap+hDLQx21odSeFaR9ShH5N p1ii6YL+U/0CZVMxXYf59NOrbFAMfbV+qBp+M8FLo2giNMvoN/QiHWMu9lu+zJRoKjN5TbfrJ2g4 jaHZGO0TqPk39hlfi7ROvKA06LWUDL/8UHqbfk9vsxxWyqaxm/govow/Im4lK3ocg7SAOuDvB9H6 WwijfTyRHxU/UZ5SLprzhs7oyVgRN/2YHqbfsiTMVGVd7PvsJHuH1/EW/mN+VtyvPKm8ZmnFrG+m pXQvPUWfsTQ2gc1g32XtbA3byH7IdrIj7Bh7n9fwJr6YnxftYrn4tVKLNEvpUn5gutu02fz+kH/o d0P/PfSZXqbfTTMQD+sx+h/RI5jZAB2l15FO01lmYgksGUllTjab3YG0lt3LHmd72JOsH70cY2fZ B/gkfcIucnxpuZnn4vAjj0AufitOmPfzh/hRpGP87/xzkSkKhEeMF5UiIJZhVBvFNqRnxdtKjnJU 0eHnMtMO027THtNTpudNF8yJlu/jG//Klz/5quirt4ZoaNPQjqHwUL/+NqVjDfH1wBWsEqNvRVqE 9d6BiNtLx1kifJfDilgVuwGeaWGL2HJ2Gzx5F9vFfmaM/Rn2HLz0J3YeY07idmPMV/PxvJZPQ7qZ t/HlOIxt5/38JP9CWESCSBHpokhMFnNFm+gWq8UOoYlXxF/EWfGp+BJJV+IVh1KguBWPMllpUVYo jyjvKe+Zmk0vm/5qjjcvNd9tjpj/gVNNlWW6ZYZlrmWrZZ/lhDWI6DxEz9KvEIGXH3ZGrBc+8Szd x8cq2bjCvIp4bqEFYipHpPI9bBO/k/XzEabbzBW8gt1IFxQ3fP0C380/5RViKmtks2gRHxNt0Dxc +QW4SuUQnVOew9xeRcu3mRPZWn7enEhhnJEm4oz0ezFa8YiX6ZQ4zSzKY/SGEs8y2Tn+hJiOKPi1 UmXyk1M8RM+I5exOepb7iOIvWrcgjm9kv8C+0MTK2D+FjmPwjYiicvEO/YAW8z/TObzHm+gBtkBZ SPfRWLaG3qOf460YZbrFXGROZy/xDqWHD2P9xJUnMbuJbAQTpuF0F5srdpnP89dpBR1V4ukt8UuM /ih/RkxVLphmsna8AXfS3bRcX0+rTX7lNbaQBLuJCpUz2N3WiDLFiXwddpVm7Gn78Hbvxz5QI6ZC koXIuQFxMRs7xC6kB7FPKIigDrzj38Eu9ir1m5t4hBaakhl2HfxS8/LQTJqj/5x26gvpFn07lWA/ 2KivQYt76K+0lfawDUN3UCeukq/j3b7B1MCPmhr0Et7DX+ez+I4r1xfeLmRZ9CHSM1iZKtMB6lH+ RLOoWt+i/xHRfRV22J00DwfWdzHLj9DDFDFIY4du5L16g+jEfE/TDP0J3cHiqV1fQtPoOfqZxUSt Fg/WWGOvYb53UBufqXeLtqEO+GErvOCFt1Zg/7nHWze7qcZbXXVtZcWkiRPKx48bWzZmdOnVJcWe olFXjXQXjnAVOFVHfp49Nyc7KzMjffiwtFRbSnJSYkJ8nNViNimCMyr2uRqCquYOaorbNWVKiSy7 WiFo/YYgqKkQNVxpo6myXitUV1h6Yfm9f7H0Ri29ly2ZTa2kypJi1edStSP1LjXC5szwg7+33hVQ tXMGP9Xgtxl8EninExVUX1Z7vaqxoOrTGla29/iC9SXFrDchvs5V1xZfUky98QlgE8Bpma7OXpZZ xQyGZ/om9XKyJmGKWo6r3qdlu1AVzYhCX+sCbfoMv68+1+kMlBRrrG6+a55G8qTkMUyozuhGM9dp FqMbtQNnHI02q73Fgz1bIjaaF/QkLnAtaG32a6IVbfi0VA/6rdcyb3836+siGseZbOM3tbmix5fV oUrjnp6NqvboDP836uY6ZQuBANpAXV7YEOxpQNdbsFKN8iyu8Q0Bv8Y2oEscLAuNWUXnFz31FgYX qVqcq9bV3rMoiKXJ6dFo5mpnOCfHO6CfoRyf2tPkdzm16lxXoLXe3jucemau7sv2qtlXakqKe22p Ucf2JqfEmMSkbzJtcHpUZ3CGueQaZ172LJNjdF2Hk6CmzlcxEr8Lc5ogSdsE6pk/AQuAJ8BQS1uA FenQ4uqCPbZJUo4pMs1UaHOpPZ8QIsB17u9XSlpjEnOh7ROSShknl0NNY62XeM3j0YqKZIhY6rCm GGOVUR5fUrwywl2uThvuz/LSQNPh29bApFK43+mUC7w54qV5KGihGf5oWaV5uWHyluJszYNSM3hJ kz5bakKXNJerB12I5H55n6V0zeq+/Jdiyxjma5+ksYz/oG6L6htnuRpxNFZ9PcFY1DY2XVGK6qVD 4TfoYpw2rM4vcjlkkuO5wtBGT8iXTHBc9idqSiH+zEZQL4hYrIhKQ8LUBs0WnBKlgXinM/bO/H+V IvoFWcvIvq4Wm4Y2yRMbaHTYWsUV5SuGl9gjGpuw5XCc7Ht64q/QIdSio7wuliHicdF3qnUazcab WYg/XDkmSARyNS9cBk0T3iJDHMiNFa8wzI1VCuCR0VlS3IA9s6enwaU29AR7WiN6aJ5Ltbl6Bvjz /PmeTh92u2jgRPT9m3O1hi0BeKydTcLrwam218U2zej1sk2z5vgH8BOHuqnJH+aM1wVrA3JZeF2T P+YWY0Fk6GMN8R8TRIzSRbOB0wgdZgQQ/u8BRTLKo3F0i/5QgrCi/2Gt+LUEYhMSTtgWImeqM7UQ BK3Ql6oY/NJrooukKoP4XNNs/T0l1TSI/2nksdm9XA7CG5+Tr5iG5yclZcZF9Pf7U1L4bMl4s5OS wKVSopRQRmIiaKKUUSnuQEdAjlD1uepzY0bn9pr/vaX/RUtm2dLf+pOSDOYjb3ZCArhUskkJ2RIT JZWyy01+3Wa/Wc222RFIYa4m/AabWAaQBqTgZ6R5inkj35SwKeWlZFOcJSGL+4bdkH59dl1u07Dm 9ObsmbmLLYsT5g9bkr44O5i7mq8yr0y4PWWj+UHLDttLWaf4SfPJhDdSci5PvCvO63SNGx3HKM4W x+O2OVK7CMHjTYZUxaeY07b8FzdneWyfzvWcA1nuORebOpu7HJeaCfJhQCAwzJZ2zdiyjIy0dBs3 uwpGuofZMsaWXZNqc7sKLObZi48/ujLcXbvo+GMnVv9w4Mk1a558cu2a6+fy40xh1/6ypW9IPzU0 NHTo6Qd/xR4eeuD8BdwUFn3UcbeMA0QDv4i1i6e9XlV4k1LHLVbW8a18p1X5pcLiyGziIs7EEjk7 HG+MPl7OiZiKuvjprd9mw9JF9A+9qcaC2o0FTTYWFF72ZsvlurQmxvrkJJq8SSnjTJc8MdrEVNyu uCk7YT+rZBsoy3Oj7V04A36J3YpRqJz6VSVVV2dOZKkTx4xmc2mux+lKNZst46+5pnwsv9hfc7zp gbOl3codVWscz0w+3ILohYONRx+Je8G3PQLCYTjvST+kAfIx3oeG+oapM6d6am7taF0yten/AMV6 jEUKZW5kc3RyZWFtCmVuZG9iago2NyAwIG9iago1MDA1CmVuZG9iagoyMSAwIG9iago8PCAvVHlw ZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9RQktGVkcrQXJpYWxNVCAvRm9u dERlc2NyaXB0b3IKNjggMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFy IDU3IC9MYXN0Q2hhciA1NyAvV2lkdGhzIFsgNTU2Cl0gPj4KZW5kb2JqCjY4IDAgb2JqCjw8IC9U eXBlIC9Gb250RGVzY3JpcHRvciAvRm9udE5hbWUgL1FCS0ZWRytBcmlhbE1UIC9GbGFncyAzMiAv Rm9udEJCb3ggWy02NjUgLTMyNSAyMDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1 IC9EZXNjZW50IC0yMTIgL0NhcEhlaWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVp Z2h0IDUxOSAvU3RlbUggODQgL0F2Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUy IDY5IDAgUiA+PgplbmRvYmoKNjkgMCBvYmoKPDwgL0xlbmd0aCA3MCAwIFIgL0xlbmd0aDEgNzIx NiAvRmlsdGVyIC9GbGF0ZURlY29kZSA+PgpzdHJlYW0KeAGFWQt8FNXVP/fe2UfemwCbTbLJzmbJ AtnEQADzwmTz2IANSEIC7tJgEkgkIEgwoIg8lrYoBhTKZ2lFK2prxVpksol0A7Sk4qNFET6l1jf4 aFF/jaC/ivWV+f4zuyDx8+s3k/85555z7uvcM3fubFbdtLqD4ilIgryLlrd1kX6ltYOlL7p5lRwp J9iIjE3Xdy1eHimPWkZk+Hjxsluvj5TTthDx6zo72rR62vU1cGUnFHqJ2BTwsZ3LV62JlNOawc3L ViyK2tOCKBuXt62J9k9voizf2La8AxyX3QIid63oXqUXKeMD8PFdN3VE/ZmfKO6PX4Xf2Mc+Ys/d H3EiYhA4fUrT6JdkgmShAppHJHVKmWRAWbMb+Padn2670JI07TNzhlmv+fB743I14ajUtePL/d8s tpA5HsUY3V8zoJ6pfPgaqrbQl/u/XIvBaS1dfvEBahLj+9w2x8nDYgKdAbiYEPJkOgbEOJEZKnN4 w8LVlzKmMKkyX8ioX6BTGXQFsB84AkjUIrJgtYBuBILAfuAIcBIwEoFqVhlYAewBzgBGkSnsIdlh qRwn0lA3DfNNEql0DlABQQ7QAmA20AJsB/YARt1P06wANgJHgPOAkbwiNbRzMsaeGtqqs76lywr1 Yluk2LxAL/ZdG4jwWQ0RXnN1xK004jZpSkR9RVWEj8uL8JScwiAa74tNKBystAorJmnFwLtAGX+a khgjBz0oxpACcIGh6hqvSOkb6y7cc0RIxAQXjNrJoQ4KFkpILqyM5So/Rynk4B/zoYiFD/UlJhfu qfwBf5f2A0cAwd/F/Q5/hzbyM1rMQSuAPcAR4ARwDjDyM7hP436bv01J/C0qACqAFmAPcAQ4B5j4 W6AW/qaWHzrV5AqA8zdBLfwNTOsN0CT+OqTX+evqIH85VFRSOKALnoKo4MiJCqkZUSHFWhjmL4W+ mICMcmOlkVGHRDaV02SRHcqZ5AgLW2jaEkeYv9cnexwPVk7kp0gBOEZyCj2fIhmoB1qBLsAI6RVI r1AQ2AE8CCgAsgzUAsj8GPAC8ApNBLxAPWDmJ0PoJsxPhNxVjkorf5E/R6mI+HH+Z52/wJ/V+fP8 GZ3/BTwL9mP82VCWgyrjYCfUsYBbwAtgN/A/9Y1NcaiVyfwIIugALQAqgNlAC7AdMPIjPDvU7khB I4foGJ5hBw/Rhzr/DT1sJu9Sh9ddjQSUNeIuvQoSyB55j5t73bvuRVEj7rt3QtKI+yfbIGnEvXYT JI24l90MSSPu9qWQNOKe3wJJI+7ZTZBAwvyB348d5yiafQOTK5P4LYjSLYjSLYjSLSTxW7SbvpC0 Md4Xys1FxHZ7PRNyHcGDLHiYBeew4MMs2MGCG1hwEwtOY8HrWNDDgnYWzGJBLwseYsUIRZB5+0cU S7w2FjzGgvtYsJsF3SyYw4JjWVBmRd4wd4auxlMH5tNZX6X20HFn31Xl2H2SuBMRdSLnndgTjoCe AFS95IWTnB1xTsvSeHZfbkWkfEVp4YrKGfwoKh7FMhyl04CEBTqKNDqKRo6iuSTQCqAFGATOASpg hHc25rFdp0mgBUAF0AJsBM4BRn045zAUTitAtSHu1wdWAFoBzNZK/CjubNxO7vRmWuwWj2WG2G5n SVlsdpaaxYvIasXenJJsTg6zhAOfJ/z78wSKqYzhd/PtlImF2BHl20NfZDrC7Bch9yFH5Rj2c8qS kHWshNwsB7yYuvXyVLKbNf0UsvPHwQtD9nmolhRy5zkOskSt1gHHF/b3HR/awxziB/ZDjr/JYYmF HH+F5vEDjlP2Ox1/KQiboTnsDjOwg7LuOmAvduw7prtugmF3yLFBYwcc6+3THTfYdUNHxHBdN0re JMcc93zHDLRXY1/o8HajzQOOCvt1jmkRr6lanQOOiRiCJyLmYrAT7Hqnriy9wblFYdbpzTPtMvlN s01XmgpNeSanyWHKNGWYRptTzBZzojneHGs2m41myczNZB4dVs94Pdpbb7RRf/kZkdCMJF22YIdh 2jYDSpyZOf2AlFGijtc1VrE6ZXAR1S2UlQuNrjCLbZivGFxVTEmpo7qmKqXYUxc2qXOUIk+dYqr/ ob+XsbsD0Cp8S5hRkz/MVE21OUNJqfYPEGPJm+/K0Pj4zXcFAmSz3lxhq0gpTy6prfke0qorW2s8 3162b0WPzZOp7Kpr9Cu/zQwohZqgZgbqlP9qlJv9A+xTdt5XM8A+0VjAPyDK2ae+OZpelNcEAnVh Nk/3I5l9Aj9kDBj8zHgxa34km7MifrsjfjmoD7+xGoNfTAzl6H45MTG6n8Q0v97usb6a3rEg8EmV qVv36U6VL/c5lgOfHBD4WIN0TPc5Zg1qPkq53ozdDpcsELiwdLLrLnaWrrvoI+/VXQqiLndecrlT 70lERqP7aATNJJy56JNwBj6XBfI/ix1VHg/rKwssavZ1uHytLl8H0KpsvbnTpgQXynLvooBmkBXh bl24qFPjbR1KwNVRoyxy1ci9ZXq975ibNXOZq6aXmn1N/t5mb0dNqMxb5nO11QT6ptdPKRrR152X +ppS/z191WuNTdH6mq7X+05fRZp5utZXkdZXkdbXdO90vS/Sc7ze32umqkA11k/jfTwuFvnamuEM VFktXeV68pY5bRsyDuK0spfiPAEl3lWlJABaXudX5ldqJjxTmikR6qSoybahzJlxkO2NmixQJ7uq yLNqdfdqsvmW1ET+unFBtWq1thQR6tF033vBxad422q0s3WdkttYp1Q0zPf3mkzQttYEoCu9qIuL 84XVwYjyCihLNUchLjlqummaLiYm6vi/c0EfE9SIzgAOGof6mDeLraLugFCy6po4toKm+QhD83z/ QZyltJdEdwAT7GYe1n2xNW0eukwRDWHa3RexanVUisZiVZTrrt0e8nRfDMnF5jxasHSix2qVB1ub 4SClAemGRylNchO+f9SzAL471LPDS9QPNLvG+UfY6MJREO2lfWwJ7aMj9BQ7j1r7aYD6STsC1dD9 tI7uoTvwWpsPzZ00B7cB+ntYmtqPL5OH8MJ8iI7D91raQAfJymzqh7SRNouXUWszJVA2VVI9raC7 2Ex1NTXTaenHVEQz6UbqYkHVr96t7lR/TY/QgPiz+g3FUTotwn1c/djwqvom5aPGz+heOs12xjxJ XvQShOcv6SbaLRZITF2sfokROOkWjEGiWXScDXIPWu+gs8zG1olqtPIrVVGfhpedFlAn7aaDbCqb zp2GZnWWepys6GMNWr2XQnQAd5j+QK+zeMN59dfqeUqjPLoa8+mnF9mgGP5m03AF4mZAlCZQCSwr 6I/0HJ1kLvYnvsIQbyg0eA1r1VM0mibRXIz2UdT8B/ucb8C9UTwr1apVlIi4/FSLNj1D77B0VsBm s3l8Al/BHxA3kRk9TsLdTksQ71+g9beRRgd4PD8hfiU9Ln1lzBw+oyZiRdx0H/2S/sQSMFOZdbMf sVfYe7yat/D7+LviHukx6SVTG2Z9HS2nu+hx+pylsGLWwH7IOtk6dgf7KbuXHWcn2Qe8kjfxG/g5 0SlWij9IVbgbpW7px4bbDVuNHwz7h58e/u/hz9VC9XZqQD5swuh/Rg9gZgN0gl7DfZreZQYWxxJx y8zJ5rLbcG9gd7GH2V72GOtHLyfZu+xDvJI+Y19xvGm5kWfg8KMdgVz8Jpww7+H38xO4T/J/8i9E qsgWHjFVTBMBsQKjukPswP2keEdKl05IKuJcaNhl2GPYa3jc8JThvDHe9CO841/4+lff5H7z9jAN bxneNRwa7lffoTFYQ7w98Ak2DaNvw70U670LGbefXmbxiF06y2XlbCYi08KWspVsDSL5E7abPaKP /Ql2GFH6GzuHMSdwuz7mK/hUXsVn476Od/CVOIzt5P38Ff6lMIk4kSTGiFwxXSwQHWKVuFXsEop4 Qbwl3hUXxNe4VSlWckjZklvySNOlFmm19IB0VjpraDY8b/i7Mda43Hi7MWz8BKeaclO9qcG0wLTd dMB0ytyK7DxKT9LvkYGXLnZGbBI+8STdzSdLafiEeRH53ELtYhZHpvK9bAtfz/r5WMMaYxkvY9fQ ecmNWD/L9/ALvEzMYnWskZbySZEGjaOl30KaJh2lIekw5vYiWl5jjGcb+DljPIVwRirBGekZMVHy iOfpdXGamaSH6A0plqWyIf6oqEcW/EEqN/jJKe6nJ8RKtp6e5D6i2K/M25DH17DfYl9oYoXs30LF MfgaZFGReI9+TDfwV2kIz/EW+jlrlxbT3TSZraOz9Bs8FRMMNxpzjWPYX/gSqYePYv3EpccwuxI2 lgnDaPoJWyB2G8/x12g1nZBi6W3xO4z+BH9CzJLOG+awTjwB6+l2WqluolsNfukltpgEm0c50hns butEoeQE34hdpRl72gE83QexD1SKWdDYkDkzkRdzsUPsxv0L7BMSMmgJnvFrsYu9SP3GJh6mxYZE hl0Hv9Q8PzyH5qu/oXvVxXSjupPysR/coa5Di3vp77Sd9rLNw7dRFz4lX8OzPdNQy08YatV83sNf 441818j1RbRzmI0+wv0EVqbccIh6pL9RI1Wo29S/IrvHY4e9lxbiwPo+ZvkxepghBmny8DW8V60V XZjvaWpQH1UdLJY61WU0mw7TIyYDtZk8WGOFvYT53kYdfI66SnQML0EctiMKXkRrNfafO73Vc5sq vRXlV00rKy0pLpo6ZXLhpIkFV+TneXInjB/nzhnrynbKjqxMe0Z6mi3VOmb0qJRkS1JiQnxcbIzZ ZDRIgjPK87lqW2XF3apIbteMGfla2dUGRdtlilZFhqp2pI8ia/XaYBrh6YXn9d/x9EY8vZc8mUWe RtPy82SfS1aO17jkMJvf4Id8V40rICtDujxLl3focgJkpxMVZJ+ts0ZWWKvsU2pv7uzxtdbk57He uNhqV3VHbH4e9cbGQYyDpKS6unpZajnTBZ7qK+3lZE7AFJV0V41PSXOhKpoROb62dqW+we+ryXA6 A/l5Cqte5FqokHZS8uguVK13oxirFZPejbwEZxyFtsq9eYM928IWWtjqiW93tbc1+xXRhjZ8SrIH /dYoqWvft31bROM4k91xuTVD9PhsS2TNuafnDll5sMF/Wd0Mp9ZCIIA2UJfn1Lb21KLrbVipOu0s rvDNAb/CNqNLHCxz9FlF5hc59ea0LpWVGFeVq7NnaSuWJr1HoTm3OkPp6d4B9Qyl++SeJr/LqVRk uAJtNfbe0dQz59a+NK+cNtKSn9drSY4EtjcxKSrEJ1wudCDoEZsu6e6aVDfnUmSZNkbX1TgJKvIi GSPxuzCnYo10FFPPomIsAK4AQy2lHSuyRImpbu2xlGp6TJEphhyLS+75jJABrqF/jtS0RTXGHMtn pBm1PLmUagpruygrHo+Sm6uliKkaa4oxluvlqfl5N4e5y9Vlwfez9tFA9YhtW6C0AOF3OrUF3hr2 0kIUlGCDP1KWaWFGiLwFOFvzVs0yeNEyZq5mCV60XKre6kIm92vfszRGMbsv/SVZrKN8naUKs/4H c0fEXtfoqsPRWPb1tEaztq5pRCli1wKKuMEWlZRR1X6RwaHTJJ4hdGvkhHzRBcdlf7wi5eDPqCd1 e9hkRlbqGibXKpbWGREaiHU6o8/M/1cprJ7Xauns22rRaSilnuhAI8NWykaURwwvvkfUNWHL4TjZ 9/TEjrAh1SKjvDrKkPH40HfK1QrNxZOZgz98chRrCGQoXoQMliY8Rbo6kBEtjnDMiFYK4NKyMz+v FntmT0+tS67tae1pC6vBhS7Z4uoZ4E/xp3q6fNjtIokTVg9uzVBqtwUQsU5WiseDU1Wvi21p6PWy LY3z/QP4iUPe0uQPccarW6sC2rLw6iZ/NCz6gmipjzXEPzSQMVI3zq4EIfIfgngotf8llFzSIKVY ATTY2rXDsAE/6ON/FeRMdibngKAF+loWg197DfQVydIgPGmuelZKMwzirZfLnug1agP4fZzDxign 2RZWL/QnJRnnQvhXf0KCLpz3jo+Ph5QcrymSLVoh2Zac54kbn5WU6EicnSgSE0dTPWMMgfYmWJKN c5mUZU/IzhptoIqhiqGnPQsKU0oKFgwVsuSUkpICjwd/HsvxU8ctbz0zaWJGL//OID7CILg2iI+8 uQkJkJLj4+P5/9XryL6+0xV6+rYj75TS9JlWr+uH1mtd14tl1uXpi11r09dnbUvfmrXb+lj64fSP rP+QL8ijrrI+YN1nFaUT2o183EHsnC78ZmVzykZ5fNbsxJZEnpho16bHXq7HWoR5Z782CPyAV0Jx rNibbPM6XVMmIqY2i43bduSxMCvupydzupO1ECXCKCd7k3nyDs9zW20ey4UFnqF/DYGt9AwNIWTf LHi/YkiP2KSJrBgXLVjJFgQCo4qs1smF5XzqlHFGoysbnCYXWvGe5yi52RS3K9s4ZjRcruzaZ13X 1ri+/kp25aHlB75mpme3D9229pOHf/c6f/6RVWtCj61b/xBrtKy9cebGV7vibfNuYOZXTzPL7uH3 hj8dPjvc98QRMeW+A0/fv23/fmSW/oMcmDoO5/XvuziUdt3AKCWarUZ8X1FDVV3tvOmeypuWtC2b 1fQ/QWJ5MgplbmRzdHJlYW0KZW5kb2JqCjcwIDAgb2JqCjQ5MzMKZW5kb2JqCjkgMCBvYmoKPDwg L1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1RydWVUeXBlIC9CYXNlRm9udCAvUEVRT1BHK0FyaWFsTVQg L0ZvbnREZXNjcmlwdG9yCjcxIDAgUiAvRW5jb2RpbmcgL01hY1JvbWFuRW5jb2RpbmcgL0ZpcnN0 Q2hhciA1MSAvTGFzdENoYXIgNTEgL1dpZHRocyBbIDU1NgpdID4+CmVuZG9iago3MSAwIG9iago8 PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9QRVFPUEcrQXJpYWxNVCAvRmxhZ3Mg MzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50 IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAv WEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRG aWxlMiA3MiAwIFIgPj4KZW5kb2JqCjcyIDAgb2JqCjw8IC9MZW5ndGggNzMgMCBSIC9MZW5ndGgx IDcxMjQgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLeFTVtV577zOPPIZMQkgm yZA5w5DBZBIDAUx4NJk8JmAjEiDgDAWTkEQCggkGEC3CoPIaECjX0kKtqNVK61VOJpFOQEsUqy2K cNXa+qjgoxX9GqF+VeuDnPvvMwGl16/37PnXXnuttV9rr7PPPmdW3ryqjZIpTIL8LcubO8m4slqR ZbesXqnGyzYHkbnhhs7Fy+Pl4cuITB8vXnbrDfFy1hYirrS3Nct68voauKodAqNEbALy0e3LV66J l7MWILcu62gZ0meFUTYvb14z1D+9hbJ6U/PyNuS4cnpB1M6OrpVGkXK2I7+i8+a2IXsWJEr67Vex Nx9jH7Hn740bETEwnD6hqfRzsoCzUzHNI1LalZFkQlnqTXzn7k+2f9aYMvVTa47VqPnge2MKJHNM 6dz15cELi+1kTUYxwbCXCtSzlA9eS9V2+vLgl7fZ4z1JzcWL91GDuKLH63CdelLk0xmAi/yob6Sr T4wRI6NTXP6Y8PSkjShJqSwSKlosNqgK2gEcBI4CCjWKXGjtoOuBMHAQOAqcAsxEoFKrAh3AfuAM YBYjhTOquuyVY0QW6mZhvikik84BOiDIBVoMzAQagZ3AfsBs2ElJB7AeOAqcB8zkF5nR3eMx9szo NiPrWbqsxCg2x4sLFhrFnutC8XzGrHhec3XcbHLcbNyEuPjKqng+pjCep+WVhNF4T6KtpL8yQ2Rg khkYeCco489SCmPkovvFCNIALjBUQ+IXaT2jvSX7jwqFmOCCUSu59H7BorbUkspErvNzlEYu/jEf iGv4QM+w1JL9ld/n79JB4Cgg+LtI7/B3aD0/I30OWgHsB44CJ4FzgJmfQTqN9DZ/m1L4X6gYqAAa gf3AUeAcYOF/AbXzt2TEGFTyFQDnb4Ha+ZuY1pugKfwNcG/wN/R+/kq0dFJJn8H4iocYV94Qk5kz xKRllMT4y9Ev8hFRXqw0IuqIGEXlNF6MiuaNc8WEIzp1iSvG3+tRfa77K8fyV0kDOEbyKnp+lVSg HmgCOgEzuNfAvUZhYBdwP6ABiDJQO6Dy48CLwGs0FvAD9YCVn4qimxg/GfVWuSoz+Ev8ecqEx0/w 3xv5i/w5I3+B/87I/4A8F/rj/Llorosqk6An1LEjtyMvht7En+4ZnebSK1P5UXjQBVoMVAAzgUZg J2DmR/moaKsrDY0coeO4h108Sh8a+S/pQSv5l7r83moEoCqJd/L3wIHsV/d7ud+7Zy+Kknh37AYn ifeu7eAk8d62AZwk3mWrwUnibV0KThLv/EZwknhnNoADifH7fjN6jKt05o1MrUzht8BLt8BLt8BL t5DCb5GJvlDkGH8WLSiAx/b5ffkFrvBhFn6ShWez8IMs3MbC61h4AwtPZeHrWdjHwk4WzmVhPwsf YWVwRZj5ey8rTvI7WPg4Cz/Gwl0s7GXhPBYezcIqK/XHuDt6Ne46ZAEj66mUNx1393yvHLtPCnfD o27EvBt7wlHQk4BulPwwUkfFjbNyZT6qp6AiXr5ycklH5XR+DBWPYRmO0WlAwQIdQxgdQyPH0FwK aAXQCPQD5wAdMMN6FOax06ApoMVABdAIrAfOAWZjOOcwFE4doHKIB42BFYNWADNliR9DGoXk5m7/ SLvT7rNPFzudLCWXzczVc3kpZWRgX05LtabGmO3Q57Z/fW6jhMoEvoPvpJFYiF1D+c7oFyNdMfbT qPeIq3IE+wnlKog6Nom8LA95GXUZ5YnktEr5BHLyR5GXRJ3zUC0l6i10HWbDZK1Dri+c77s+dMY4 2LPOI64/qTGFRV1/hOTRQ65XnVtdfyiOWSF50htjyA6rhmmfs8z12HHDdAMU+6KudTI75LrdOc11 o9NQtMUV13eh5E9xzfbOd01HezXORS5/F9o85KpwXu+aGreaKOscco3FEHxxtgCDzXcanXpyjQbn lsZYu7/QsscStMy0XGUpsRRa3BaXZaQlx5JuTbParcOsydZEq9VqtipWbiVrekw/4/fJp1662Xj4 mRHQjBSDt2OHYXKbASXOrJy+T9pwUcfr5lSxOq2/heoWqdpnczwxljhrvmbyVDEtrY7qGqq0Ml9d zKLP1kp9dZql/gfBbsZ2hCDV+JYYo4ZgjOlStDFHS6sO9hFjqRvvzpH5FRvvDoXIkbG6wlGRVp46 qbbmO0iTIWyq8X1zOb5hfQ7fSG1P3Zyg9uuRIa1EMvrIUJ32X3PUBcE+9gk7H6jpY/+QWSjYJ8rZ J4HZUi7Ka0KhuhibZ9iRyv4BO0QMMthZ8WCWdqRac+N2++J2eagPu9Eyg11CAuUZdnkJCYadwqRd d9foQE33aBDYZKrUZdh0ZarftjmeB5s8ENhkhOm4YXM8IyxttHKjGacTJrkgMGHZ5DRMnCzbMDFG 3m2YFA+ZbL1kstXoScRHY9hIgmZsZy7a2M7A5luO/M9sW5XPx3qmhFoWBNo8gSZPoA1o0ratbndo 4UWq2t0SkgpVE96mRS3tMm9u00KethqtxVOjdk8x6v2beoFUT/HUdNOCQEOwe4G/rSY6xT8l4Gmu CfVMq59QellfWy/1NaH+O/qql41NkH1NM+r9W1+lUj1N9lUq+yqVfU3zTzP6IiPG64PdVqoKVWP9 ZN7DkxIRr0057lBVhr2z3AjeKW7HupzDOK0coCRfSEv2VGk2QMZ1UWVRpVThnpKqYRCnDKkc66a4 cw6zA0MqO8SpniryrVzVtYocgSU18V8XLohWrpJLEac+KfvOCyYBzd9cI8/WdVrBnDqtYtb8YLfF AmlTTQiyyRdlSUmBmN4fF14J4WRpKMQlQymbKmUJCUOG/zcWjDFBDO/04aBxpIf5c9lK6goJLbeu gWMraJgPNyyYHzyMs5R8SHSFMMEu5mNdF1uT8zB4iksI0+66iJWrhrghX6wcyg3TLh/5ui665GJz Puksgxi+WunD1mY6TFlAtukRylK8hPcf/QPgrMwHl+hnpV7m/CNsdLEhEB2gx9gSeoyO0jPsPGod pD7qJXkEqqF7aS3dQ5vxWJsPyVaajWSC/B6WpffizeQBPDAfoBOwvY7W0WHKYA79Q1pPG8UrqLWR bDSKKqmeOuhudo2+ihbQaeVOKqVr6CbqZGE9qO/Qd+sP0cPUJ36vX6AkyqYWpBP6x6Y/629REWr8 mPbSabY74Qnyo5cwLH9ON9M+sVBh+mL9S4zATbdgDArNoBOsn/vQeht9wBxsrahGK7/QNf1ZWDlp IbXTPjrMJrJp3G1aoM/QT1AG+liDVvdSlA4hxegpeoMlm87rD+nnKYsK6WrMp5deYv1i8MKGwQr4 zQQv5dMkaDrot/Q8nWIe9jTvMCWbSkx+0236q5RO42guRvsIav6Nfc7XIa0Xzym1ehUNg19+JL1N v6N3WDYrZjPZPJ7PO/h94mayosdxSK20BP7+KVp/G2F0iCfzk+IXyqPKV+aRg2f0YVgRL/2Mfk5P MxtmqrIudgd7jb3Hq3kj/xl/V9yj/Ep52dKMWV9Py+luepQ+Z2msjM1iP2DtbC3bzH7E9rIT7BQ7 yyt5A7+RnxPtYoV4SqlCmqN0KXeaNpm2mc8OBgefHfyfwc/1En0TzUI8bMDof0z3YWZ9dJJeRzpN 7zITS2LDkFTmZnPZD5HWsbvZg+wA+xXrRS+n2LvsQzySPmVfcTxpuZnn4PAjj0AefjNOmPfwe/lJ pFP87/wLkSlGCZ+YKKaKkOjAqDaLXUhPiHeUbOWkosPPJaY9pv2mA6ZHTc+YzpuTLXfgGf/i17+4 UHDh7UEa3DK4ZzA62Ku/QyOwhnh64BVsKkbfjLQU670HEXeQXmHJ8F02K2Dl7Bp4ppEtZSvYGnjy LraPPWyM/XH2JLz0J3YOY7ZxpzHmK/lEXsVnIl3P2/gKHMZ2817+Gv9SWESSSBEjRIGYJhaKNrFS 3Cr2CE28KP4i3hWfia+RdCVRcSmjFK/iU6Ypjcoq5T7lA+UD0wLTC6a/mhPNy82bzDHzP3CqKbfU W2ZZFlp2Wg5ZXrU2ITqP0RP0G0TgpYudERtEQDxBO/h4JQuvMC8hnhupVczgiFR+gG3ht7NePtq0 xjyFT2HX0nnFC18/x/fzz/gUMYPVsTm0lI+LN2hOV34NbqpyjAaUJzG3l9DyGnMyW8fPmZMpijPS JJyRfifGKj7xAr0hTjOL8gC9qSSyTDbAHxH1iIKnlHJTkNziXnpcrGC30xM8QJT4lXU74vha9mvs Cw2shP1L6DgGX4soKhXv0Z10I/8zDeA+3kI/Ya3KYtpB49la+oB+ibsi33STucA8gv2BL1EifDjr xYebX2F2k9hoJkzpdBdbKPaZz/HXaRWdVBLpbfHfGP1J/riYoZw3zWbtuANup020Qt9At5qCysts MQk2j/KUM9jd1ooSxY18PXaVBdjTDuHuPox9oFLMgMSByLkGcTEXO8Q+pJ9in1AQQUtwj1+HXewl 6jU38BgtNg1j2HXwpeaFwdk0X/8l7dUX0036birCfrBZX4sWD9BfaScdYBsHf0ideJV8Hff2NaZa ftJUqxfxCH+dz+F7Ll9feDuPOegjpMexMuWmIxRR/kRzqELfrv8R0X0Fdti9tAgH1vcxy4/Rw3TR T+MHr+Xdeq3oxHxP0yz9Ed3FEqldX0Yz6Ul62GKiZosPa6yxlzHfH1Ibn62vFG2DS+CHnfCCH95a hf1nq796bkOlv6L8e1OnTJ5UVjpxwviScWOLrywq9BXkXzHGmzfaM8qtunJHOnOysxyZGSPSh6el 2lOG2ZKTEhOsFrNJEZxRYcBT26Rq3iZN8XqmTy+SZU8zBM3fEjRpKkS1l9toqqzXDNVlln5Y3vBv lv64pf+SJbOrU2lqUaEa8KjaiRqPGmPzZwXB313jCanagMHPMPhdBm8D73ajghpwtNeoGmtSA1rt 6vZIoKmmqJB1JyVWe6rbEosKqTsxCWwSOC3T09nNMsuZwfDMwORuTlYbpqhle2oCWpYHVdGMyAs0 t2r1s4KBmhy3O1RUqLHqFs8ijeRJyWeYULXRjWau1ixGN+oSnHE02qZ2F/ZHtsfstKjJl9zqaW1e ENREM9oIaKk+9FujZd72vuObIhrHmWzzt7U5IhJwLFGlcSSyWdXunxX8Vt0ct2whFEIbqMvzapsi teh6O1aqTp7FNb4xFNTYRnSJg2WeMav4/OKn3rympaqW4KnytEeWNmFpsiMazb7VHc3O9vfpZyg7 oEYagh63VpHjCTXXOLvTKTL71p4sv5p1uaaosNueGnds97CUISbZ9m2mDU6P6wzOMJdc3exLnmVy jJ6rcRLU1BYVIwl6MKcySdrKKNJShgXAFWKopbViRZZoCdVNEftkKccUmWbKs3vUyKeECPAM/P1y SfOQxJxn/5SkUsbJpVDTWPNFXvP5tIICGSKWaqwpxlhulCcWFa6OcY+n0473Z/nSQPXwbXNocjHc 73bLBd4W89MiFLTwrGC8rNKinCj5i3G25k1S039RM2Ku1IQvai5Vb/Igknvl+yyN0KzeS78Ue8bw QPtkjWX8B3VbXF83x1OHo7EaiDQNRW1dw2WluF46FH6DbojThlcHRQ6HTHI8Rxja+An5ogmOy8Fk TcnDz2wEdWvMYkVUGhKm1mr2pulxGkp0u4fumf+vUkw/L2sZ2TfVhqahTfYNDTQ+bG3KZeXLhpcc EXUN2HI4TvaRSOJlOoRafJRXD2WIeLzou9VqjebizszDD68cZRKhHM0Pl0HTgLvIEIdyhoqXGeYM VQrhktFZVFiLPTMSqfWotZGmSHNMDy/yqHZPpI8/w5+JdAaw28UDJ6Yf3paj1W4PwWPtbDJuD05V 3R62ZVa3n22ZMz/Yh08c6paGYJQzXt1UFZLLwqsbgkNuMRZEhj7WEH9oIGKULvmPAxgZPIR/WswA 0fhLEoQUkx9LsLXLw7AJH/TxXwW5U92peSBogb5WRf/XfhN9RarSL9uap/9NyTD1k4+Furns3p+U 5fDbbHyuw0kMvvL7klFg+Z5EW0pySm5iYv6IXKeSm+805ds8tmRHFqM01Z6czOeqFm9MP9srzb3F vhPyJxOlTaqosA/YB9ImFQ88Z38ubZL9WV+JxLixOf4rTLYMW8C2yaYEUq9LXZ0jZmcssy9Nb81Y Zbs1fZMtkr4152FbokkV8jtRUlKybZhiYeiXxfhDPX5M4Agew/lkYxN7k5NHKI7D/CHK4u3+MRil CcO0pXU1qh0qVx02m3muGrZ0ef1uz4SxXkZeu5djxP/8jdR4dxU5YqwsmvUKO8zK8GrW70+CnYpH LqddhTG2u3ubw2f/bKFv4J8D9s98C1eA8S0coIqBioEL76fKydkvLJQMS03LnDRuLCsrYwtpIVsR Gl6akTG+5KqJE7yeUZbSS6zZMuaqq8aXZODpbLZISp5R3nm9rh/fuP7gg7ePvyY9Lakrtmnpku3p ve6PHl9z/MYbWu/YNXj2tad1dqdj72btjrUPpN/H19zecsddd6lPPL842tp475W5T+3oH/z0bxi0 celjcKb+rkvq5adMGQNpgLzMeAei+sCsmfXTfJU3L2leNqPhfwFT5FpJCmVuZHN0cmVhbQplbmRv YmoKNzMgMCBvYmoKNDkwNwplbmRvYmoKMjUgMCBvYmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUg L1RydWVUeXBlIC9CYXNlRm9udCAvQldCRk1RK0FyaWFsTVQgL0ZvbnREZXNjcmlwdG9yCjc0IDAg UiAvRW5jb2RpbmcgL01hY1JvbWFuRW5jb2RpbmcgL0ZpcnN0Q2hhciAzMiAvTGFzdENoYXIgMTE3 IC9XaWR0aHMgWyAyNzgKMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDI3OCAwIDAgMCA1NTYgNTU2IDU1 NiAwIDAgNTU2IDU1NiAwIDAgNTU2IDAgMCAwIDAgMAowIDAgMCAwIDAgMCAwIDYxMSAwIDAgMjc4 IDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgOTQ0IDAgMCAwIDAgMCAwIDAgMCAwCjAgNTU2IDUw MCA1NTYgNTU2IDI3OCAwIDAgMjIyIDAgMCAyMjIgMCA1NTYgNTU2IDAgMCAzMzMgNTAwIDI3OCA1 NTYgXSA+PgplbmRvYmoKNzQgMCBvYmoKPDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250TmFt ZSAvQldCRk1RK0FyaWFsTVQgL0ZsYWdzIDMyIC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAgMTAw Nl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0IDcx NiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMgL1hIZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lkdGgg NDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250RmlsZTIgNzUgMCBSID4+CmVuZG9iago3NSAwIG9iago8 PCAvTGVuZ3RoIDc2IDAgUiAvTGVuZ3RoMSAxNjIzNiAvRmlsdGVyIC9GbGF0ZURlY29kZSA+Pgpz dHJlYW0KeAGFewlgVMX9/8y89/btvW93s2c2u5tssgEWDCThjuQFEsBGbtAEE0mAKKccAbxQQhXB iEJt663g0YoHsjnAgFpSRa0ohVbrr+IBVTxqjVD/lFok2f9nZhOEtv/ff1++M9853nsz33tmXlat XN1ArKSJSESft7R+ORG/QCeyA/PWrIqKIrH5CTHMvGb5tUvTZfcSQpRvr11y4zXpcrCDkFx9QUP9 /HSZnEM+bAEq0mVajDx3wdJVN6TLgRrk85csm9fbHrwd5eKl9Tf0vp98hHL0uvqlDen+pZfz8vJl javS5THfI5+5fGVDb39aRYjlNz90fLiTfk3ffCTdiRAKJEC+IyXkUaISRjRSQK4gRF4gZxEFZd6u sC33frf5zBxHyT+MmUZx5xOf5Q/gyGvy8q1nd3VfqxGjFUWT6M8bcJ86pmcyGaeRs7vO3qSl38Rb +n6BvWSm1K8t7o8ceVnqT44DmNS/NZEV2SvlS1mtoyN6hxRrc3kKHWWDpCieWCDSKNJlgF2A/QCZ zJHCaNWQrgM0AXYB9gOOAAyEIOWtUcAywDbAcYBBypJCrdGIVpYvBXBvAPN1SD5yEpACSCSCtAAw BTAHsAWwDWAQ/XjNMsA6wH7AKYCB6JKv9d4ijN3XepfI2hYtKRTF+nSxplYU266sTueTpqXz8svS 3Ualuw0pTldfMjad5w9M5668wiY8vM1sK+ws80peTNKLgS9HStkB4qCURMh2yUOSACZhqKJGl1xt ufHCbfslmVCJSZTMJ5FUp0Rbbc7CMjNLsZPERSLsW9aVbmFdbXZn4bayn7BPyS7AfoDEPsX1F/YX so4d5zRHWgrYBtgPOAw4CTCw47iO4fqEfUIc7GNSACgFzAFsA+wHnASo7GOkGvuIS4xIOV4KYOwj pBr7ENP6EKmDHQV2lB1NdbJ3W4ePLNwrkERBLxLJ60V8mb2Iy1vYwf7Y+q/+kKg4OA2JeknKIWNI kZTTmjck0iH5W0sWRjrYZ23RRGR72WD2HkkCGEbyHt78HokCpgLqAMsBBmDvA3ufNAG2ArYDkgBI GVINEGUHAe8A3ieDATpgKsDIjrTiNR3scGt8bKTMy37P3iQ+UPwQ+53I32FviPxt9rrI30IeRvtB 9kZrOELKLGgnuEdDriEvQLvCftuW64qkypxsPygYQVoAKAVMAcwBbAEY2H6W0zo/4sJDXiIHocMR 1kr+KvJfkyeMRF8U0ePjIIBRnsRHXQoMybbotjjT4/c9iCJP4vfcC4wn8ds3A+NJ/Kb1wHgSX7IG GE/i8xcB40l89hxgPIlPmQkMSQd77MXc/MjwKYtptMzBrgeVrgeVrgeVricyu55f5F8yH+PDrQMG gGIP6Yn+AyJN+2jTy7RpOm16gjY10KZbadN62lRCm66mTQnaFKJNYdqk06aX6AiQoonq7RcVR+p+ 2nSQNu2kTY20KU6b8mhTLm2K0uF6B8tuvQxah6xCZG1lXOlYdtulY2B9HCwbFM2GzGfDJuxHehiQ EiUdnaI56c6BMM9z2gaUpsuXjCpcVjaRvYYbXwMbXiPHADIY9BrE6DU85DU8zoG0FDAH0Ak4CUgB DOidg3lsEakDaQGgFDAHsA5wEmAQwzmJoTCyDCkf4i4xsAKkpYApvMRew5WDK5tl61laSEtoE6Ut IeoI0ynhVJgNJ14v7LLLaXR2UNuef9q+/6eNmMpM7B62hWSBEVt78y2t/8qKdNAHWuMvRco89H4S liF1dCSJ0zzkI0ijKA8lISOvLyYh9hzywtbQFbjN0RofGNlH7fyuPZF/hU5E/hrqYEC/Cr0U+Z9o h0xbI39CzXN7Iu+F7oy8VdBhRM3L8Q6KbF9UdN0bGhHZeVB0XY+Gh1ojt/JsT+SW0ITI4pBoaEg3 XN2Iku6ITI/PjkzE88pDcyN6I565J1IaujpSku41lN+zJzIYQ0ik0QEYbP+QeGksLB44a3gHXaAP VO9Tq9Qp6jC1UB2oZqsRNUvNVDOMLqNmtButRrPRaDQYZSMzEmNGR+q4nuBeL8MgnJ8BAk2JLHAN FoZyM4OUMGpk5Cck6ZYqWeWMsbQy2TmPVM6NJs/MiHVQ87TZSSU2liZdlaRy5tjkiERlh5qanhye qEyqU6+qaqH0nmrUJtmmDkpmVnXQFK/akJl0javaSyh1brg7k+f9NtxdXU383jWl/lLXGOfI8eX/ JakTlXXliR9//h/RhD+RlbyvckZV8tms6mQhR1JZ1ZXJn8+I1lTtpd/RUxXle+nfeVZdtVcaQ7+r mM7rpTHl1dWVHfQK0Y9E6d/RDxKDDP2McMy8H4kaw+l+D6X75eF+9MvlGfqZTCRP9MszmUQ/mfJ+ LY25FeUtuUjQxxcljaJPoy96YZ+DeeiThwR9vE3koOhz0NvE+yTHiMeEQugSRoIuNEhCokuIBkUX MfIW0aWgt8ud57vcKd4kpUcj+vAEj7Ed7+tjO44+FxDyf0cbxiYStG109byaioZYRV2sogFQl7xr zQJ/smluNNoyr5o3RJNSvG7uvAU8r29IVscaypPzYuXRltHivn9rruHNo2PlLaSmYmZVS43eUN46 Wh9dEasvr26bMLV4+EXvuvP8u4qn/pd3TeUPK+bvmiDu+7d3DefNE/i7hvN3DefvmqBPEO8iQsan VrUYydjqceAfz9uYxQx5rcvMrh7r1ZaPEcI7Ott/a+Y+RCs7iCVRnbTGxiZtAC7Xg8oGlfEm6BRv sqPa0dvkv3V0duY+uqO3SUO1MzaWJFatblxN/BULy9N/jfihatVqzop0muB1//WHLhVJvb6cx9aV yQEzKpOl02ZXtagqauvKq1E3qq/OYqnoSHWmKy9B5SjeUZLOd+R1JbzOZOrt+J+yIMaEalBnLwKN l9qoHqarSGO1lAxXzmQwBTNngww1s6v2IZbiTqKxGhNspAna2Pc0Pg+Bk3QNwbQb+2DV6l6slxar enPRtTFBEo19JOl7XIITSySCVqsSMG3KPhIABJWnSUCOEz8hqS8BX/G8Z2HqK97Oc/Y1DF1HLxCy g+ykC8lOsp+8Sk/hrl1kL2knPAQqJ4+QteQXZCPc2mzU3Emm41JQ/wsaSLVjZfI4HObj5BD6Xklu JfuIl/pTfyXryAbpXdy1gdhIDikjU8kycje9PLWa1JBj8m1kOLmcXEeW06ZUVeqe1L2pp8ivyF7p d6luYiFBMg/XodS3yp9TH5FBuOOX5EFyjN5r2k10vKUJPR8lK8lDUq1MU9emzmIE2eR6jEEmk8gh 2skSeHoD+ZL66VppHJ7yZCqZOoBeIVJLFpCHyD46lE5g2UpNalLqEPHiHTfgqQ+SVrIHVwd5hRyl VuVU6qnUKazEBpLLMJ928nvaKfV0r+8pBd0UUKk/GYmWZeQ35E1yhMbob9kyxaoUKrpyU+o9kkGG kFkY7dO48wv6T3YrrnXSG/L41FhiB11+xqlNXid/oUFaQKfQK1h/tow9Jq0kRrxxCK75ZCHo/QCe /gnEaA+zssPSk/Jz8g+GrJ7jKTs4EicPk0fJb6kNM43SRvpT+j79jI1jc9jD7FPpF/Iz8h/Vesz6 arKU3E2eI/+kLjqCTqNX0QV0Ld1If0YfpIfoEfoVK2Mz2WJ2UlogrZBekcfimiE3yrcpdyh3Gb7q qeo50POHnn+mClN3kGmQh/UY/S/JY5jZXnKYfIDrGPmUKtRC7biiNJvOojfjupXeTZ+gO+gztB1v OUI/pX+FS/oH/YHB0zIDy0Tww0OgGFuJCPMX7BF2GNcR9g37l+STcqSENFQqkaqlZRjVRmkrrt3S X+SgfFhOgc6Fyn3KNmWH8pzyqnLKYFV/Ch//zrknuwd0f9JDejb13NfT2tOe+gvxgIfwHliClWD0 9bgWgd/3QeJ2kXepFbQL0gF0DL0clJlDF9EV9AZQ8nb6EP2VGPsL9GVQ6X/oSYzZxkJizJewoWws m4LratbAViAYu5e1s/fZWUmVLJJD8kgDpAlSrdQgrZJulO6TktI70sfSp9IZ6RyulGyWI3KOHJcT 8gR5jrxafkz+Uv5SqVHeVj43mA1LDXcYOgx/R1QzRp2qTlNr1S3qHvU9Yx2k8zWym7wICTz/o8el 9VKFtJvcw4rkAJYwv4c8zyHzpUkMksp20E3sFtrOcpUbDKPZaDqZnJLjoPUbbBs7w0ZLk2glnUEW sSHpBxoy5GeBlcivkS75Zczt93jyDQYrvZWdNFhJK2KkkYiRXpcGywnpbXJUOkZV+XHyoWymPtrF npamQgpekccoVSRbeoS8IK2gt5DdrIIQ8w/GzZDjyfRZ2IWZtJB+L6UQBk+GFA2XPiO3kcXsz6QL eryJ3E/ny9eSe0gRXUu+JL+GVvRXrjMMMHjoW2yh3MzctJ0w+RnMbiTNpZKSQW6ntdJDhpPsA7Ka HJbN5BPpeYz+MHtBmiSfUqbTBdCAW8gdZEVqPblRqZL/SK8lEr2C5MnHYd3WSoVyNvJ1sCo1sGl7 oN37YAfKpEmo8UNyLodczIKFeAjXA7ATMiRoIXT8Slix35N2w0zWQa5V7BRWBzs1b/dMJ7NTvyYP pq4l16XuJYNgDzam1uKJO8jnZAvZQTf03EyWYyn5AXT7cmU8O6yMTw1izewDNoPddzF/Qe086idf 43oBnBmjvESa5f8hM0hpanPqT5DufrCwD5K5CFhPYJbf4g0TpU5S1DOZtaTGS8sx32NkWurpVISa yYLUEjKFvEx+pSqkXk2Ax0n6R8z3ZtLApqdWSQ09C0GHLaCCDmqthv25Ux83a2aZXjrm0pLRo0aO GD60uKhwyOCCSwYNTAzo3y8/npcby8mORsJZocxgwO/zejLcLqfmsNusFrPJqBoUWWKUDKyIja+L JuN1STkemzhxEC/H6lFRf0FFXTKKqvEX90lG+X31aLqop46e1/xbTz3dUz/fk2rRElIyaGC0IhZN HiqPRTvo7GlVwO8uj1VHk10CnyTwrQK3Ac/Oxg3RCv+C8miS1kUrkuPXLGiuqCsfNJC2WMzjYuMa zIMGkhazBagFWNIXW95CfWOoQJivYlQLI0YbppgMxsorkoEYbsVjpLyK+vnJqdOqKsozs7OrBw1M 0nHzYnOThEdKCdGFjBOvSRrGJVXxmuhCxDhJcle0ZWBn8+YOjcytS1jnx+bX11QlpXo8oyLpTOC9 5UnfTSf8PxbxcMRkGy9szZSaK/wLo7xzc/PGaHL7tKoL7s3M5k+orsYzcC/LG1/XPB6v3gxOVfJY PMk2VFcl6Qa8EoFlnphVen7pqDevblE0aYqNjS1oXlQH1gSbk2T6jdmtwaC+N3WcBCuizTOrYtnJ 0sxYdX15qCWDNE+/sS2gRwMXtwwa2KI504RtsTt6EavtQqQBRE+3CUx051jl9POUpXyMscsQCSaj 86IYSVUMcxrBk4YRpHneCDAAv2qKu5LzwZGFSdO4umZtFK/HFGlSydNi0eZ/EEhArOubi2vqe2sM edo/CG/kcnJe1JK0vg9PJhLJAQO4iKjjwFOMcYwoDx00cE0Hi8WWa1g/80UDmQra1lePKgD5s7M5 g+/q0MlcFJJN06rS5SiZm9lK9ALE1qyOt3T2tXhm8Zamvpbzt9fFIMntfD1LPElj/PyfQ/O6KxaM SlLv/9LckG6vnBGrRGgcrWiu65XaypkXldLtnKCgG9p6saR7XJWUyVDHMZYpidZ0hNzXBeFylTUp 5+HPIIR6fodqhFSKGhodn9TqJqbTanN2dq/O/P9u6kid4neJ7MfbeqeRHJXoHWh62MnRF5UvGp61 WaqcCZPDENk3N5svaoOopUd5WW8GicdCPzs6LklmQTPz8IclxwgO1ZlJHSRDy0xokaiuzuwtXtQx s/emavy4dA4aOB42s7l5fCw6vrmuub4j1TQ3FtVizXvZq+zV5uUVsHZpwelI7bsrMzl+czUotoCO gnowMrYlRjdNa9Hpphmzq/ZiiyO6aWZVK6NsXN3Yas4WNm5mVS9ZBEO46IOHODGBxHAfD3hWbiSz AMcAJb34JJF/Rnagfb96N9YZjeQOrOymIh8PqES9m75JNgFuQ3kjoJw9SzagTynaclG+DSJJhWAS nOIY4H8wPpxxcFH9bz++LcN/EvyuIjADTkb4L330IdB/S0yibP632guLlgsL53ErVix24sCpC/85 RepC6oa35z8P1gs+xARp3EOmIXp/lLVKW+SblF8oJw31apNxmekyc8zCLM9b49anbBvtC+x/0/Y7 p7mi2FnCfQqfhMRnkO3MduYhwS4UOReVOs/pCvmBROVOTp9nez6ht2FVYyaTd5vR/TlDB52qx6lU whg10xJixpGBVEIMI9RRUxDxLUP8sh2P3m55/AF/Qjtde/qE1lWilZBSnmpdWncXdbpGDhlcNLTI k2FQ84cNG77n0NQrC0cOkw4dWnFXfFKg/ioE5rNSX8pOpRMUyKKzWhgXFN0cDMtKRthm85k6Ul+1 OxxsFkf0gM0GzEmsvIZ4rVakVl5HCrBOPYTkEF5f2jVkcGaL4T+fdBpPMvAnfdFuswnkWz1gsQBz Eo3XEM1q5SmvO//IH5/ZbogGtBCUvZVFLb+Bo/ECXAAHtvrmyoaNbJNlk+Mtu2JSLX5W4b7c85PA uMyZ7hpPTWB65mJ1sWWee4lncaAu80Z2vWGN5SbHRsMD6n3aW/6j7H3D+5YPHcHzE2806dmx4sEm SkyaiZm2RpyNBAqu21EbRbjEyNbwm3eB8GdqE11IViS6eqdOa1dg4TmC/yigutqtuYYVFXq9Lo/G DLGc/Lhb8xYVDnNq8ViOapi1+N3ta1pXjV307uPv3fizvc+sXfvMM7eu/Ukte5fK9NLn57T1pI72 9PS8tvOBF+mjPfefPIXV3KJvF97BZeYY9PYH8M5MdulRSbc5ixfL69gW9qBRfl6mJmJQmGRSqJXR g2YxejOfE6FR3Ivt0XZNA+s6Ul/rTsHQkGCoXTAUVNYDnF19PBH8CVoV3eYoVvooMVihUayAmRKw 7KMldAPxJyZrJ0AM0KV35wKFkkndEMtS30jqhDzSWlKbyI45DQZ1KGSyiP3QXvbuzPs/LVgl3zxm beSFCQfn8LmVILJWMbcwfbNXlkxOzeZ3uw2zbB2p0+1Op0C+1U2aBiycoYS5iPp4h3CYt4ZDdrSE IaBIO9hLupWZfT6cNjoZi0agGwXvHeLpIVLQxQdbytMDCHcze9WAv9DqcjHxQt3kcAJLv+e4bnG5 2axwBq/jz27Fo7mqWCxsFpBvdEHF//Y2riP8ffxt4mX6sNHKaMNLyn7DS+qbxrdC6mXWautM+2Lr fPtNrpvcd7pedn0e/DzzVNC63/Kim2Xi8CBLC2uG32C7QoXwG5GbwK1g2KwZDYaDoWBGKBQ0hoIS ZcZgSLKFtQ72VNsUJ8XRgn83nwER5HBQZjU3+t4Ftbms05fYethljY7Qrc7dpdhWWMbWMZntY7k4 QNjSkhb20xD2RIl2GtJe0l1S2tVde8Lp4pxFstF+ScJ+i3YgbXeECnANGEFqae3K6uo8T3Z8ODg+ bNjQYoi+MEnQCxgng4o/WT03nPnynnzo5I4Hb/7pI3Sv+/s/vHtm4tOvPlET3rmzrGRe560HPr9m 8c8faXYf/uDrnVXPvvzUpvohkBRuwbKVX0NSrkgzDrvt0FQbl2MaspvDHk/IxflvcchyOGSzU6L6 IfXCrgmE9/Rz61VwqEBoMaSi+4B2IAFZ0Pu7hAQ5RFoZvDGrOes+99Pu16zvWz/MNJrcfvuAoGQa rAy27AM3JJgizW32uNzug3ZHht2dYXfYQH/dzQei27fbmd3u0D20d1AvOmT6rs71kfr1KB+ec462 TFunbdFkrUlt9Asz5KfEr/kZBnv6RS7O/q1R18t0KHHQXxIbHdFq30330RHY9+vULT8aKBz43NvL tkTXaVxgHTdUp2u5lwDnSrsw0VonAP7ixEbjJQkF7CPCbQjO0RW1iUS1e7iXG6w+rrmzPdkSLBrx ZKiwZ/FZr3geXPLT9p2br9zc75l72AfdL065/Wed1Ljq7tO/66ZNWvNdB554qHVKqZf9/fmeNTU9 Z/7w5s9aj8N0kUngnAc6nkUG0Cm9Shdx0Ag2ZiSa2S+s26jNBsXOVHLCGTZzmJI8DSRI+yEt7NO4 2fLBaSAVfsjX64cOvXdIe72Pk7Vd2oFazslBiwO0XNU95YHy6GzXzOhiab4637jINT+6yrg6tMF4 R+h943tepxrlJM7nBpIjMX6QZOFYtmhQeUN+NBbN5g1OPsqpNoZxZtJ353BGsgW6qW/MtAPq5CK7 8xo1wUgNR/awupjFqRe5r9O2DjRzzoXpSN1b6pvjW+Zb55N9cK2GWT4vf6mvg+W2JdKuJpGo7TrP xLS/EfoHBsLnCo5xQ8YtbDVV4/nCwRjUYWCWi6tZLIc4teEoeWnGBSyVfmjzD7xs8RVls+ayspev be++/sjtf+k58eidX+38uHv4lHsmr3zqiZtvelaeYV80eNLgMd9+NK+u559/bO66FZtJa+kzv93x 6rmPa5+t7njsgV27eiOKALfc4OoLadv9oiUCEc5zQoDPiCCAS7KgLpBTej8+Xb9TaJhTxABOv3Ng wtIv7LBH7FPskt2eQaZSSrnfsWmw7ZRrCoRCEUw+kKgt5JLcVchNz0gIAZcDjUvBx6+ft+cXDOJH 3dcHCOV3WrkQ/T/eevG7/u1VeNOPL9KLRwUv9+qxq7xXxq6RlniXBq+N3RS8Jbw5eFf4Ie8zwZeD X3u/iJ6Jui/1Pubd6ZVG9Z9vYPncbsQgS/7sqCHaLzzFPocbiRCfHn13alqk2vkgcHQ7klggUc6L zcLWgVzO2rmYOftcc9SpO5lza6/kIEjh6s+1n8vNed3vExtSu4LWIlQRmj6GDS3O59KCnEBYsMPD A5c4FWbbk8GNwfKd3rX1M26ZOowOe2npnnNUfWNL1803/f2J54+yt3+16obWZ9be8jidod103eXr /rzc6r9iMTX++RjVHur5rOe7ni972l7YLxU/vOfAI5shMpjhDnj7Dco+fL90t54wKGGjcYtKVZVI chirGGJUH4ki5mMsaJFNURG/RHuDFy5PIoo5A3kSyFndko5MRRjZkeppF1ElFzQz1yhSax5dwwOV 01xZ8KvlAcpkUGfSCYQpcGtclkq07hIEz04YOwE7pI/Pfc6S3VOVfTt7Ru3svgaj3o+hr8eoJfLO bj5KxgOjthGXigCprag4nQ8anM779U/nsbx0nhVO5/6gyCGJWnFU2arsUiQpipXDFsT3SSIXINqc ilDvFFGwpNhCthJJxF8wSZiuv5cM3/SR4ds+MpzRhWEkUUGGJ+T3q3m02jtjLMZbm3AoXVu9YmVJ N4x8+sfDND7pIuf+V5V9Z8djjgF8R7YGc/TTD/V4fxJ39nfF/SPJMOdI1zD/ZWSC8zLXBH8VudJZ 5brSrz1gfMDBeuO1Io0GAwlPsVJsLVfKrZWemcpM61We+cp862LPKmWV9WaPQ/FYJUJd2Mt3MCPn dCn/8WCiVqhyph6WZEVhBtVoVMzgq8lmdzis2HV0ebw+v9/TkSppw/FMlOdWl5Pn+myP0RTFlzkM VMygcJ2K0Rj2+DM8Hr/LajKFPS6gLidWMVHNmaFpTpfJavR7FAdsD2EYkiL54VpMJqMRay/md7mc TmIM+nxBrcxEpyFIsiL1AHSi0Gl7olGYpkCgg97VskPIVW0wMKk76O/uDga6/ZMrGsq/QLSERVlJ OuUYj5lgqri5EjBy5MZJlyQ23nJg4yX+/8zAnY127cABJCU8EdiFCT5acOBg04ndm1aXmZtZrDxQ mYfKAajcy2ODEfyg1IIaO2rarLqi8+UJHMfK2mxa5Pb6hg1H5kLmLsIJVzzfoFL6WM/Nbx7LDY7A jv/Xf5wSCw364rWe617qeTtf9WX0vKXsO1d6/y//lit90h3s+eb/3NUuvXB2vFy7Odow4Ycn4Qvu SH0lRyA9fHXZpD9MFasjVxmqVChKaSQZYZFITqgoNDa0PLI1YhjlLvGWwHxeHqw11tqqHLXeq4OL jEtsCxzXea8LdkY+sB71HQ186v7G903gs6zjkVQkEFUKHAUZg5VSh65c7piqXKMczfqHfFazah67 bGAkM4QpmD0hu8Wfe8RCNYtuqbM0WeQIokM2yyLWsRa/wLln4mbDAjuBMN7AkfSCFchxYUJ4jV7A TYhlFXUWEVmYGVnYlCIpj7FOSrfS7TRJT1E5QktxzifBZfWIdRaQc3qW3Y6YVIQqVGgndfEVA7Xy F6LH9+gqkHO6ly++qJ+bLprBX0ED4QnDLzRbEIpaaO8krRuG7ITW3avDqBShARbiTi5bQwbz+HvF SrICi64iJ7x/mHk0goWolOHjpjwd19FBT7evbJm7a4Xe890rLy9mxbN+tub5X61e87yyr/sfW6Zs OdjYc7Ln/Ufpfftn3XXo7SNvHIJtmJr6SuqSx5Agnd0bvRXb1zmow0K5yVoOuyi7QhbVH5JxVuhR jXz2qpi9KvytipU+arizThx67w3hyxGrFXLgAdsEk5VGQuPc43wz3DN8de4638PsYekh21PaU0Gr 0RYwL2ILpUXKautyW5Pt19bdpj3m3Var13qH9TMm2XPmOJY51jkkB+1gz+o3DhZ2tA7D2grDehz2 1EQcDmwLnR9jCEPPtRs5se05mZhfriURgf2AduuCQbrgzkTBk6DgyWUhT+5hlUbUUpWpdt5JNfNO qot3UodkFh/otbrgStr61q7s3YgTy5QR1V0rTye6VvYFqwjGtdoT+ONsg09eUU196cCtWOwi+NQ4 ojpD2glLJS1ZJ1842vPPlX+9c+dHkV2BdbM3PfvU7YvuoRt8Lx6mWdT8PGXrdz2euXjJa+++/+pP ecw9Hjw7Bo108v0e/Skzk215tmJbuU0ZmjE0dCWbaZ6eMSN0LZuvNJjmZdSFOiPvKX9yfxz43P15 xknf3wKfC83zRiKJIFfXyiDXXfUSlmu7xDuKDbVVsgrb+IzLQlear7Bda/vc8KX3LD1t16hHsls0 BzTSojoJVFKy+It4SOjI07QjTqohXKlzNjmhmlwm0grqdHHvhpDmdHqHwmngEuQUCova79AVFHfa OcVR/lZoKZDv9bGcO85Vrtz96mH1mJpSZc6iKaqkhoXIQa8geNjh4gIp2CaCTzUo2BYIF0+9QNNq V0zqOq9dXOnE9loJgoWuUiyBS0t+1LMVtVCzoTxYig/tZRh0DtH2eT2TRjQcWPen1Yveu63uvoK2 7ujzq9f8asfNNzx+x2Obf3hyG5Wap5Ux+9nxzPXOwd++cfSdA5xnlbCiYeiZBzybofsiJORhs6Ra pdY0y9IgLVaWmRosRrjCEyKkBgFO6NM5AbJCPM13faCczTgTlIe4RgWGhMpck4JloWku7ImF6l1L g/WhGww3eM6wM34Nn3M4bD7fVG+ddzk+bA45tmrbsVjR5MyQWSX7sK+LhXWfNeuENoDuGrTjl25o uE/HvsxHIvoAwsMQsUHztWAKajp1U/6A4iSWdMEID5Py4sU818vC2JOK0Ii3SMtV9dwBxX2cQuAC 7qQ5hYkATysYtqmgWmJxpIJTF9rE2sSk7hOTNew/nQEkEpPSwS7MY3qpW9K9okTEFdwsio0orJZW rOxTMQ3RLnFmqNlicUSzxQrKIF29b+C3e//ac5JmfPQnfPFw7itz64Z5m7uPsmnWEVfcufYZeoXv yXasWiV8XtCv55Oef2nRXfsW0F/eMW7Br2FF3GBhk/Iu8VGbHs4wUUegIDA4oAeWBx62PmJ7xmYM 2vrZkoHOgBzg9OgXjBRnGW2S1REyUw9LZLhlfENu3pZBM1JuXfblyThHvxdmiRNxyIhinuuJUKR4 K6EBnQfDAd0GNSGIbJD2E+FxDlccMpBTEf2hOMJ1ZXDJR/lr4XaAfCEWZ0DOitUpedIfeJnuI9nk DE6T8SFi36Yepy2CRuwDIZTp0rq6anngXCJ2hEbyTb5xN+oZWKeZVIPRwAyayZVJnAZHJr4ISAxY v54moCcri5yxoUVDi7EfVFQIs8atmqfIE3O2btvmDt625vKazBGF08sPH5Ye2rxicfH4K12PmsfX zd187hpoxCZE3iWwYnxf/ZB+tWkYl5Uppq2m7aakqdN0zHTKpBJTxLTc1GTa1lt13JQymSPY0cU3 DNgXNUi3UuyQGmSzQc1TiLxN3i4n5U75uGzolE/JjMhR+QhKspz2y2wWkLO6j1NSFq5fNvO3yoKK aEtTEUiPkH0g53Qzl1V5snEC7Elf8J1222JLFNZDiCMPAblArlyRcGO7XoLN2NTe3i7/7fDhHzxy /IejfI10G5LhYs6f7VHEhMVyY/iI9DKieGg6Hzwkneeklxl6nsdX7FAi+HjmmCJPQXJKkSLKcqVJ SSn4VpifK+SlhQlPAus7dQ+4so3QTrhIRhDtHoG3lNHCoxMhP2fTgUxvACSoQNKyZOQmAV3TJACS EuoPpJcWZLJ8MS14DANJEuTgJOAl/uOLkNvaxSIEc9+II5QvYAG99BbdrUgGN9uhdWifSV+6T0ln 3AbQ+pReYrEV36jRB7Qj/uP+lF+OGjPsGV5XSFGpwWsz2+xWe64fRgt7fcLHWPpx3JLBGYrQ7hvd KQJBwVpLDmetCPiEj7FAmURI+K/0VpDFzMMztJ/Rhae36EXDilMWij/LZITfnXqweFhx0n/Kz5b7 t/uT/k6/7JdYkccraH0G29dpSv5IUiFYF5FU7lXPs9z7gayMCxNJx5v8FS688xSIc55Fk31iiZ8m IFJQFQqKRT9f+F/4Sx8UlZSWdjn5MkToq9fgNJmNZtUsGbQ4PG0mdZhdmZRAZaGz/FQDrAJXPFBX rxeffjhjzuJ0MOLc+MTqj+sen6qZ2wcsntj4tBy/f1fF8kmFt3Q3sjuuW1p27zvdL3M/Vg4/lg8u 2kiALt7jgR83zHLzbXOOOPj2fSPHAqLBpZoD1gmGicYrDNXGaw0LjcZibZRrlHeov0KrdFV6K/w1 So1pulbrqvVO9y9Vlprma0tdS73z/ddTj8mg2K6SsOA0X2VdIjUoDeYlVrMvJKvOkMWSkZsp4v9M IQbYz+s9AFFFYNEblPYtA9B8SoxPIJwPAhE+iDPBnZtXPFjFlrKmRhFeDDmWSTN5/WXcrQG35xKr ne9QuITxFXEvwSBQI9wZwUkBuGoViiP2/YiOR0ZIKQg2JMjdG5h6nnldcG61Z2p/rPjx3InHHljX 66YZygzTXGWuScbyXnx96habftivFZGk+4JopPypO1//kHpv/ttdx3q69rZuvKO1bcPGVnxylX/P mp6/dB/6209pmNreefudP7z+9kGI2gaw8Q1w0Ene0kcXuKkm05hcLI/DB4TXyKtkg8lpNBlNNrfT ZCOSkVr4+gvOy9Rvq5Eac6Ju6mY5zrS50bV/F98LLMz3ulN4rPQSyyDIBRvS56dOp5dHxCA0QkTs ZLJrQl+sLagFkS/pRiR9eiXf1eGSDs/EDypGjiTaWxvttxzgIfZKWitk+rwPUg0e54Ynxiwsverq MWPHjr46IyzHH18xcdTT+RNK61Z2v8ctcSli6BZQYbDk02+WczJyRpl+YirPvSKnIWet6R7T7bm/ dj838FXJZvIF/b7BlQPf9ymZbBZjWiE1+2uMNaYac42lxlpjW2RcZFpkXmRZZF1ka4+35zvy47n5 uf2H5c42V1vmx+f3WxVblduU+3PzI9Z7+90/8JeDnzI/Y30y/yn8197rcW+/vk3UnD4Ee9XpbdXc PkT04VQSfTgi+nBE9OFIVkfqE90VHjnbmJ9nNcvBaNwjWy7JCvJlU05gICd+JFAamBKYE9gVOBww OAKRwLLAsYAcCWwJsMArEGUP5EJEhzr8AENQqFOm4RtQRqhGGVa2nW0Z3mKe65rdWUzpJTVZS7JY Vsijwnafbuc3AflCMJgjupurmhy6xBIJ0mBuQHf7iwv57QUiwvGnU65SAXEOHcB+ICxHlN8VEFvI AREhBjrYVa1q7gDcujs08sgACuwL+CQ2C0j6aFsgnA5Avt7DtXFAULwqG/FqXWFnISstbCpkhTzS zSXCRvQeV0fTVGazBMIHwJH0uWk01yFWtw4xPAeODhg3cWd1DBGYnb8wfabkyDlGaCk+1GMkMKQ3 nMWSo3d3kh9KagmsECcLQ85FewXi2r69S96CdQhqS7tWYN9SBBd8N4A7UhzN8R0AyLwvbeP1/EHh mJIxMO7UXJpbkww5tmgmMfVTM6kyCEk4A8VseyyT5MRsVmN/cybtl28yGxJyJoloWdwbJPjuVTqB Y4BrSKxfv773G3cU+Zp15Yra80dG+fF8fD1bzE/9hNvoW7764EF82IgQJile2uq48+a1NwzN+/kb D04pGzHgZzNueWW2M2ltXLh2kddbkHn7/vuvWPjGLYc/oJeGFq9sKL805s8rvGz95Ak39oskJt58 rX96zfThsVCW25xbVLa2Zva2K5/nepqb+o4NUB7EFxx/3kvMkMFYvBifH2DNAaQpgEjQajNTiXg1 U8JhNnixKHVoOSSH2lx5VppSjRWmijp1udqkblVlAvu+XU2qneoR1QAf8K1w3UD4LhEWJwir27nN AvI9HEYvItaZYLsQbDSdwikk9yHCcqGc9n3qPrYIO6zDWq65MEwEMcVHHd0l2onT/MiO701j33Wk s6hIe4sHS4lEni+92OSxtHM4vHPMKY52mBa8vGTukoG33962e7c70S/8+DZtTMMTbN5mqi7puXtz 988nDQyKqBK27Dj/7wE6ZS8JgjYmxIss6vYWQ1pP6UWujOKEm+Ya3V4rdXstMOZOkIkUefP8Ph70 BEVE5ROxlM/FCSDOv8XutE+Yb5RPp1fqvgyuIyh/n46ifNiAE+Uzuo3rQ8pHO33UNznIeeThAVTw VJAtD24PJoOpoBy05gnuCcfBv8yImo6YjptkVKZDU46kHYdJvNmE0FR8Z3Im7S9MIoIy4bAOtZMD FwWicBdd/xkqwYNwupeWpD2HCJSCsma3OWx8T9pgVIwIl2RrJrEZnZmEB0sDBqzHhhtUoXcfID8+ FKEsttq82AMYxnGpdO2frn5yimZptzivmzbtntHtj7RPXDplaCO7t7vt7iETps3YsomNRNCP6Jv/ Uvn4fv2//bA3jxWQQ3yt5MFXSgF8994f/08wilSQ8WQCmYj/WLgchmUa/otjFr64upJU4Q5KXAD+ M+DbJzL2yrHjJ01LlK1cWL9k0sz/C3qmffwKZW5kc3RyZWFtCmVuZG9iago3NiAwIG9iagoxMTQy MAplbmRvYmoKMTEgMCBvYmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1RydWVUeXBlIC9CYXNl Rm9udCAvUFNOTUZCK0FyaWFsTVQgL0ZvbnREZXNjcmlwdG9yCjc3IDAgUiAvRW5jb2RpbmcgL01h Y1JvbWFuRW5jb2RpbmcgL0ZpcnN0Q2hhciA1MyAvTGFzdENoYXIgNTMgL1dpZHRocyBbIDU1Ngpd ID4+CmVuZG9iago3NyAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9Q U05NRkIrQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQov SXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9T dGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEg L01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiA3OCAwIFIgPj4KZW5kb2JqCjc4IDAgb2JqCjw8IC9M ZW5ndGggNzkgMCBSIC9MZW5ndGgxIDcwOTIgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFt CngBhVkLfBTV1T/33tlHnrtJSLJJNtnZLFkkmxgIYAJJk81jAzYiAYLu0sQkQEpAMMEEfFRhURFc nvJZWqgVa31grTLZRNwELKlQbVGEqrX1UcFHPx+/IuivYn2Q+f4zu6BYf/3m7v/cc88593XumTt3 ZvuuX9lJiRQkQd6Fyzt6SL+yFiHLXriqT46Wk2xExuYf9yxeHi2nLSMyfLx42U0/jpazNhCx97o6 O7R62vU1cFkXBHqJ2GTkY7uW990YLWe1IDcv614Y02cFUTYu77gx1j+9ibJ8XcfyTuS4cu4BkXu6 e/v0IuWgf7qk5/rOmD3zEyX87qvIG4+zj9hz90aNMCQwnD6lSvolmcBZqYSuIpK6pFwyoKzpDXzr 9k83nW2zVH5mzjHrNR94d1yhxhySerZ9uffcYiuZE1GM0+01BeqZqkavpDorfbn3y5ut0Z40zfmL D1GzuGTAbXMcPyDG00mAi/FhT65jSIwTueEKhzciXAOp6aWWmmIho8USncqg3cBe4CAgUZvIg9YK ugYIAnuBg8BxwEgEqmlloBvYDZwEjCJX2MOyw1ozTmShbhbmaxGZdBpQAUEO0BJgFtAGbAV2A0bd TpN0A2uAg8AZwEhekRnePgljzwxv1LOBpctK9WJHtNjSqhcHrg5E85mzo3n95VGzaVGziZOj4ktr o/m4omieWlAaROMD8UmlIzUZIgOTzMDAe0AZP0wWxshB94t0UgAuMFRd4hWpA2PdpbsPComY4ILR InKoI4KFk1JKa+K5yk9TKjn4x/xUVMNPDSSnlO6u+SF/h/YCBwHB30F6m79Na/hJzeeg1cBu4CBw DDgNGPlJpBNIb/G3yML/TiVANdAG7AYOAqcBE/87qJW/qUWMTjW+GuD8TVArfwPTegPUwl8H9zp/ XR3hL4fLppYO6YynJMY4CmJMZk6MSc0ojfCXwl+MR0S5sdKIqP0in6poksgPF0x0RIQtXLnEEeHv Dsgex/01E/grpAAcI3kFPb9CMtAEtAM9gBHcq+BepSCwDbgfUABEGagVkPkR4AXgVZoAeIEmwMyP h9FNhB8Lu2sdNRn8Rf4cZcLjR/kf9fwF/qyeP8//oOd/Qp4H/RH+bDjPQTUJ0BPqWJFbkZdAb+C/ Hxib6lBrUvhBeNABWgJUA7OANmArYOQHeX54kSMVjeynI7iHHTxMH+r5w/SAmbxLHV53HQJQ1oh7 2g/AgeyWd7u5171jJ4oacW/ZDk4j7js2gdOI++a14DTiXrYKnEbci5aC04h7fhs4jbhnNYMDifD7 nho7zlE261om11j4DfDSDfDSDfDSDSTxG7REX0jaGH8RLiyEx3Z5PeMLHcFhFjzAgnNY8AEW7GTB 1Sy4lgUrWfAaFvSwoJ0F81jQy4L7WTlcEWTewYuKU702FjzCgo+zYC8LulmwgAXHsqDMyrwR7gxf jrsOmU/PBmq0m447B35Qhd3Hwp3wqBMx78SecBD0GKDqJS+M5PyocVaelucPFFZHy5dOK+2umcEP oeIhLMMhOgFIWKBDCKNDaOQQmrOAVgNtwAhwGlABI6zzMY+tOrWAlgDVQBuwBjgNGPXhnMZQOHWD akPcqw+sBLQamKWV+CGkfCQnd3pzrXarxzpDbLUzSx6blafm8TLKyMC+nJpiTomwpH2fJ/378ySK q4njW/hWysVCbIvlW8Nf5Doi7Odh935HTTr7GeVJiDo2ldysAHk59erlKWQ3a/LJZOePIS8N269C NUvYXeQYZslarX2OL+zvOT60RzjYD+z7HX+VIxILO/4CyWP7HK/Y73L8qSRihuSAO8KQDcu66ZC9 3PH4Ed10LRS7wo7VWrbPcat9uuNau67ojCqu6UXJa3HMcc93zEB79fYFDm8v2tznqLZf46iMWk3R 6uxzTMAQPFG2EIMdb9c7deXpDc4ri7Aub5Fph8lvmmW6zFRqKjI5TQ5TrinHNMacaraak82J5niz 2Ww0S2ZuJvOYiHrS69GeemOM+sPPiIBmJOm8FTsM07YZUOLMzOmHpKSJRt44t5Y1KiMLqXGBrJyd 64qw+NnzFYOrlimpjdTYXKuUexojJnWOUuZpVExNP/L3M7YlAKnCN0QYNfsjTNVE63KU1Dr/EDGW sm5zjpZfsm5zIEC2jFXVturUqpSpDfXfQ9p1YXu955vL9g3rsXlylR2Nc/3Kb3IDSqnGqLmBRuV/ 5sot/iH2KTvjqx9in2hZwD8kqtinvjmaXFTVBwKNEXaVbkcy+wR2iBhksDPjwazZkWzOi9rtitoV oD7sxmoZ7OLiqEC3K4iL0+0kptn194711fePBYFNpky9uk1vpvxtmyMFsCkAgU1GkI7oNkcygpqN UqU3Y7fDJA8EJiyb7LqJnWXrJvrI+3WTkpjJXRdM7tJ7EtHR6DYaQTNJJ8/bJJ2Ezbcc+d/ZzlqP hw1UBBa2+DpdvnaXrxNoVzau6rIpwQWy3L8woClkRbjbFyzs0vKOTiXg6qxXFrrq5f4Kvd531C2a usJV308tvmZ/f4u3sz5c4a3wuTrqAwPTmyaXXdTXXRf6mtz0PX01aY1N1vqartf7Tl9lmnq61leZ 1leZ1td073S9L9JjvMnfb6baQB3WT8sHeEI84rU9xxmozbD2VOnBW+G0rc4ZxmllDyV4Akqiq1ZJ ArS4Lq4prtFUuKc0VTLElpjKtrrCmTPM9sRUVohTXLXk6VvZu5JsviX10V8vLoj6VmpLEaUeTfa9 F0x8irejXjtbNyqFcxuV6tnz/f0mE6Tt9QHIpp2XJST4IupIVHgphNM0QyEuGGqySk0WFxcz/M9Y 0McEMbwzhIPG/gHmzWN91BsQSl5jM8dW0DwfbmiZ7x/GWUp7SPQGMMFe5mG951vT5qHzFJUQpt17 Hn0rY1zMF32xXDft9ZCn97xLzjfn0ZylE91XfR5sbYZhygKyDY9QluQmvP+o7wMfaPnoEvUDTa/l /CNsdJEYiPbQ42wJPU4H6Rl2BrX20hANknYEqqd76Ra6h9bjsTYfkrtoDpIB8ntYljqIN5Nf4YH5 KzoK26tpNQ1TBrOpH9IaWideRq11lET5VENN1E2b2RXqSmqhE9LtVEZX0HXUw4KqX92iblcfpIdo SPxRPUcJlE0LkY6qHxv+pr5JxajxU9pJJ9j2uCfJi16CsPwlXU+7RKvE1MXqlxiBk27AGCSaSUfZ CPeg9U56n9nYLaIOrfxaVdTDsLJTK3XRLhpmU9h07jS0qDPVo5SBPm5EqzspTPuQIvQ0vc4SDWfU B9UzlEVFdDnmM0gvshExem7taDX8ZoCXxtNUaLrpd/QcHWcu9nvebUg0lBq8hpvVV2gMTaR5GO0j qPm/7HO+GmmNeFZqUGspGX65W/M2/YHeZtmshM1iV/HxvJvfJ64nM3qciLSIlsDfP0frbyGM9vFE fkz8WnpM+sqYO3pSTcaKuOkX9Ev6PUvCTGXWy25jr7J3eR1v47/g74h7pEell0wdmPU1tJw202P0 OUtl5Ww2+xHrYrew9exutpMdZcfZB7yGN/Nr+WnRJVaIp6VapLlSr3S74U7DRuMHo/7Rw6N/Hv1c LVXvpNmIh7UY/U/pPsxsiI7Ra0gn6B1mYAksGUlmTjaP/QRpNdvMHmB72KNsEL0cZ++wD/FI+ox9 xfGk5Uaeg8OPdgRy8etxwryH38uPIR3n/+RfiEyRLzxiiqgUAdGNUa0X25CeFG9L2dIxSYWfSw07 DLsNewyPGZ4xnDEmmm7DM/6Fr399rvDcW6M0umF0x2h4dFB9m9Kxhnh64BWsEqPvQFqK9d6BiNtL L7NE+C6bFbIqdgU808aWshXsRnjyDraLPaSP/Ql2AF76KzuNMSdxuz7mS/kUXstnIV3DO/kKHMa2 80H+Kv9SmESCsIh0USimi1bRKfrETWKHUMQL4u/iHXFWfI2kSvGSQ8qX3JJHmi61SSul+6T3pfcN LYbnDf8wxhuXG+80Royf4FRTZWoyzTa1mraa9pleMbcjOg/Rk/QUIvDCxU6KtcInnqQtfJKUhVeY FxHPbbRIzOSIVL6HbeC3skE+1nCjsYJXsCvpjOSGr5/lu/lZXiFmskY2l5byidEGjWOk34CrlA7R KekA5vYiWr7RmMhW89PGRArjjDQVZ6Q/iAmSRzxPr4sTzCT9it6Q4lkmO8UfEU2IgqelKoOfnOJe ekKsYLfSk9xHFP+VeRPi+Er2G+wLzayU/VuoOAZfiSgqE+/S7XQt/xudwn28gX7GFkmLaQtNYrfQ +/Qw7orxhuuMhcZ09ie+RArxNDZIXHoUs5vKxjJhGEN3sFaxy3iav0Yr6ZgUT2+J32L0x/gTYqZ0 xjCHdeEOuJXupBXqWrrJ4JdeYotJsKuoQDqJ3e0WUSo5ka/BrtKCPW0f7u5h7AM1YiYkNkTOFYiL edghdiH9HPuEhAhagnv8auxiL9KgsZlHaLEhmWHXwZea50fn0Hz1YdqpLqbr1O1UjP1gvXoLWtxD /6CttIetG/0J9eBV8jXc21cYGvgxQ4NazEP8NT6X77h4feHtAmajj5CewMpUGfZTSPorzaVqdZP6 F0T3Jdhhd9ICHFjfwyw/Rg8zxAhNGr2S96sNogfzPUGz1UdUB4unLnUZzaID9JDJQB0mD9ZYYS9h vj+hTj5H7ROdo0vgh63wghfeWon95y5v3bzmGm911Q8qK6ZNLS+bMnlS6cQJJZcWF3kKx18yzl0w 1pXvlB15ufac7CxbZkb6mLTUFKslOSkxIT7ObDIaJMEZFflcDe2y4m5XJLdrxoxirezqgKDjW4J2 RYao4WIbRdbqdUB1kaUXlj/+jqU3aum9YMmsciVVFhfJPpesHK13yRE2f7Yf/OZ6V0BWTun8TJ3f pvNJ4J1OVJB9tq56WWHtsk9pWNUV8rXXFxex/oT4OlddZ3xxEfXHJ4BNAKdkunr6WWYV0xme6ZvW z8mchCkq2a56n5LlQlU0Iwp8HYuUptl+X32O0xkoLlJY3ULXAoW0k5JHN6E6vRvFWKeY9G7kJTjj KLRR7i8aCW2KWGlBuydxkWtRR4tfER1ow6ekeNBvvZJ583u2b4poHGey9d/W5oiQz7ZE1oxDofWy cv9s/7fq5ji1FgIBtIG6vKChPdSArjdhpRq1s7jC1wX8CluHLnGwLNBnFZ1f9NRb0L5UVuJcta6u 0NJ2LE12SKE5NznD2dneIfUkZfvkULPf5VSqc1yBjnp7/xgKzblpIMsrZ12sKS7qt6ZEHdufbIkx iUnfZjrh9KhO53RzjWucc8GzTBuj63KcBBV5oYyR+F2YU7lGOssptLAcC4ArwFBLWYQVWaLE1bWH rNM0OabIFEOB1SWHPiNEgOvUPy+WdMQkxgLrZ6QptTi5EGoK6zjPKx6PUliohYipDmuKMVbp5SnF Rasi3OXqseL9WXtpoCb4tiMwrQTudzq1Bd4Y8dICFJTgbH+0LNOCnDB5S3C25u2aZuS8Jn2epgme 11yo3u5CJA9q77OUrpjdF34Wa0aar2uawjL+i7ozqm+c62rE0Vj2hdpjUdvYfFEpqtccCr9BF+OU tDq/yOGQaRzPEbo2ekI+b4Ljsj9RkQrwM+pBvShiMiMqdQmTGxRr+4woDcQ7nbF75v+rFFHPaLX0 7JtqsWko0zyxgUaHrVRcVL5oeIkh0diMLYfjZB8KxV+kQ6hFR3l5LEPE40XfKdcpNA93ZgF+eOUo 1xDIUbxwGTTNuIt0cSAnVrzIMCdWKYBLi87iogbsmaFQg0tuCLWHOiJqcIFLtrpCQ/wZ/kyox4fd Lho4EXV4Y47SsCkAj3Wxabg9ONX2u9iG2f1etmHufP8QPnHIG5r9Yc54XXttQFsWXtfsj7lFXxAt 9LGG+EMDESP14uxKYLTgIfzTYgSIplyQIKTwD4X2pQTEgITTtYnImeJMKQBBC/S1LEa+9hroK5Kl Ea2teer7ktPwMOWxq/q51j2+gagj3qSkJD6P2ZPj89LT7akRvt+bYJGkPHtSMiOTLaJ+NGix8Hk6 o1naSjyeoyVHS6j6VPWp1Kkl5w5bD3smTsjxjk9NTEoyzrPotDH7ptxQ7o60R9IOJb6a+EaOOS7N llyYLeImGCYkDGNHEvgWZE2LT09NSzuSbBmTnDYm2ZIU4Q9607SBeJPvT+bJyRZvOosN6imLxF72 YhYRZvPK2vBS2qzd1jXWrVbJGjT12rxO1+QJNkY2q41jsP96ShuMbZuceoBNwff+n1ISKw8nP8mG WTnexka8CbCX8ZTltA2f4bb3b7R5rGdbPaf+hWQ962ldAba1ErM81/qePtHWlKklrdZT1vfWmy/1 GG61HqaU1KkTJ7ByXGxFq8cTSCvLyJhUetmUyW5XvtE0Ls2Z7hSXTSql9DEmoyvfPe/p9J3Lbht8 fNPVmy55dAt/7dxTs+64e4SZ+zb/64/nWNAa2nj4gV3hWdUZ/JPfjq5qGT375+fuDp/EGmOJcanj cGb+vkvT23QFo9RYxBjxjkNNc6+c2VDrqbl+Sceymc3/B6ItWFYKZW5kc3RyZWFtCmVuZG9iago3 OSAwIG9iago0ODgyCmVuZG9iagoxMiAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1 ZVR5cGUgL0Jhc2VGb250IC9ZQVpQVlIrQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKODAgMCBSIC9F bmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDU0IC9MYXN0Q2hhciA1NCAvV2lk dGhzIFsgNTU2Cl0gPj4KZW5kb2JqCjgwIDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRvciAv Rm9udE5hbWUgL1lBWlBWUitBcmlhbE1UIC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMyNSAy MDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAwIC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0NhcEhl aWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFkaW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2 Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAwMCAvRm9udEZpbGUyIDgxIDAgUiA+PgplbmRvYmoKODEg MCBvYmoKPDwgL0xlbmd0aCA4MiAwIFIgL0xlbmd0aDEgNzEyMCAvRmlsdGVyIC9GbGF0ZURlY29k ZSA+PgpzdHJlYW0KeAGFWQt4VNW1XnvvM4+8yCRCXjPJnMmQQTKJgQAGCE0myUzARiRAoDM0MQkk JbwkEhAfCEMrCgOC5VqqVMWqrahXOZlEOgFaovhoUYSr1NYn+OhV+zVC/arWBzn332cGlNav9+z5 1157rbVfa6+zzz5nVq9a00mpFCZBvkUr2rvJuHI7kOUtum61Gi+n5RCZm37UvXhFvHzJciLTx4uX 3/CjeDl3MxGnrs52WU9eXwOXd0FglIhNRD66a8Xq6+Pl3Gbk1uUrFyX0uWGUzSvar0/0T2+irF7T vqITOS67BqJ2r+xZbRTJjv7o0u5VnQl7FiRK+d1XsTceZ39lz98TNyJiYDh9QtPoXrKAs1EZzSdS upR8MqEs9Sa+Y+cn2z5rTZ/2qdVuNWo+8N6YYskcUbrv+HLfucU2sqaimGTYSwXqWaqGr6I6G325 78sbbfGepOb8xQeoSVza58lxnjgkxtJpgIuxUW++c0CMEfnRSqcvJtx9maPK02tKhYoWywyqgq4E 9gGHAYVaRQG0NtANQBjYBxwGTgBmIlCpVYGVwB7gNGAW+cIRVZ22mjEiF3VzMd90kU1nAB0Q5AQt A2YBrcAOYA9gNuykZCWwATgMnAXM5BPZ0Z0TMPbs6FYj61u6vNwotseLzS1Gse8HoXg+c3Y8918R N5saNxs/MS6+rDaejymJ55lF5WE03pecVj5YkyWyMMksDLwblPFnKJ0xctL9YhRpABcYqiHxicy+ 0Z7yPYeFQkxwwaiDnPqgYNG0jPKaZK7zM5RJTv4xH4pr+FDfiIzyPTXf5+/SPuAwIPi7SO/wd2gD Py19DloN7AEOA8eBM4CZn0Y6hfQ2f5vS+VtUBlQDrcAe4DBwBrDwt0Bt/E0ZMQaVfDXA+ZugNv4G pvUGaDp/Hdzr/HV9kL8SrZhSPmAw3rIE4yxKMNn2BJOZVR7jL0e/GIuI8mClEVEHRSFV0QRRGC0a 74yJnOi0Jc4Yf69P9TrvrxnHT5IGcIzkJHo+SSrQCLQB3YAZ3KvgXqUwcAdwP6ABiDJQG6Dyo8CL wKs0DvABjYCVn4iimxg/HvXUOmuy+Ev8ecqGx4/x3xv5i/w5I3+BP2vkf0BeAP1R/ly0wEk1KdAT 6tiQ25CXQW/iT/WNznTqNRn8MDzoBC0DqoFZQCuwAzDzw7ww2uHMRCMH6SjuYSeP0kdG/mt6wEq+ pU6fpw4BqErimfo9cCB71D0e7vPsuhtFSTzbd4KTxHPLNnCSeG7cCE4Sz/LrwEni6VgKThLPglZw knhmNYEDifH7fjN6jLNi1jKm1qTztfDSWnhpLby0lhS+Vib6QpFj/EW0uBge2+3zji12hg+w8CEW nsPCD7BwJwuvZ+GNLDyNha9mYS8LO1i4gIV9LHyQTYYrwszXf1Fxii+HhY+y8OMs3MPCHhYuYuHR LKyyCl+Mu6JX4K5DFjCyvhp503FX3/eqsPukcxc86kLMu7AnHAY9DuhGyQcjtTBunFsg88K+4up4 +bKp5StrZvAjqHgEy3CETgEKFugIwugIGjmC5tJBq4FWYBA4A+iAGdaFmMcOg6aDlgHVQCuwATgD mI3hnMFQOK0ElUPcZwysDLQamCVL/AhSIZKLu3z5NofNa5shdjhYegGbVaAX8ArKysK+nJlhzYix tP2fp/3z8zRKqkni2/kOysdC3JHId0S/yHfG2F1Rz0FnzSj2cypQEHVsCnlYEfLJ1GOUJ5HDKuUT ycEfQ14edcxHtfSop8R5gI2QtfY7v3C87/zIEeNgP3QcdP5JjSks6vwjJI/td550bHH+oSxmheSQ J8aQHVAN0wHHZOfjRw3TjVDsjjrXy2y/82bHdOcyh6HojCuu7kHJl+6c41ngnIH2/I6FTl8P2tzv rHZc7ZwWt5ok6+x3jsMQvHG2GIMd6zA6dRcYDc6riLEuX4lllyVomWW53FJuKbG4LE5LvsVuGWnN tNqsI6yp1mSr1Wq2KlZuJevImH7a55VPvZFm4+FnRkAzUgzehh2GyW0GlDizcvo+aZeIBt4wt5Y1 aIOLqGGhqn021x1jybMXaCZ3LdMyG6ihqVab7G2IWfQ5WoW3QbM0/jDYy9j2EKQa3xxj1BSMMV2K Ntm1zLrgADGWsel2u8wv3XR7KEQ5WddV51RnVmVMqfd/B2kzhG1+7zdXzjesN8ebr+1qmBvUHs0P aeWS0fNDDdp/zVWbgwPsE3Y24B9gf5dZKDggqtgngTlSLqr8oVBDjM037Ehlf4cdIgYZ7Kx4MEs7 Uq0Fcbvdcbsi1IfdaJnBLimJigy7oqQkw05h0q63Z3TA3zsaBDbZKvUYNj3Z6rdtjhbBpggENllh OmrYHM0KSxutymjG4YBJAQhMWB45DBMHyzNMjJH3GiZlCZMtF0y2GD2J+GgMG0nQTNrp8zZpp2Hz LUf+Z7az1utlfZWhRc2BTnegzR3oBNq0rdd15WjhharauygkFaomPG0LF3XJvL1TC7k7/doit1/t rTTq/Yu6Waor3f5eag40BXubfZ3+aKWvMuBu94f6pjdOrLiory0X+prY+B19NcrGJsq+phv1/qWv CqmeLvuqkH1VyL6m+6YbfZER443BXivVhuqwfjLv4ynJiNc2uytUm2XrrjKCt9KVs95+AKeVvZTi DWmp7lotDZBxXVpTWiNVuKekagTE6QlVzvpKl/0A25tQ2SDOcNeSd/WanjWUE1jij/96cEG0eo1c ijj1Stl3XjAJaL52vzxbN2jFcxu06tkLgr0WC6Rt/hBkU8/LUlICMX0wLrwMwqnSUIgLhlI2TcqS khKG/x4LxpgghncGcNA42Md8BWw19YSEVtDQxLEVNC2AG5oXBA/gLCUfEj0hTLCHeVnP+dbkPAye 4hLCtHvOY/WaBJfwxepEbpj2eMnbc94l55vzSmcZxPDVai+2NtMBygXyTA9TruIhvP/oHwAfynx4 if6h1Muc/xUbXSwBor30OFtCj9NhepqdRa19NED9JI9AfrqH1tGddBseawsg2UJzkEyQ38ly9X68 mfwSD8xf0jHY/oDW0wHKYjn6R7SBNolXUGsTpVEh1VAjraTb2ZX6GmqmU8pPqIKupGuom4X1oL5d 36k/RL+iAfF7/RylUB4tQjqmf2z6s/4mlaLGz+huOsV2Jj1JPvQShuW9tIp2ixaF6Yv1LzECF63F GBSaScfYIPei9U76gOWwdaIOrTyoa/ozsHJQC3XRbjrAJrHp3GVq1mfqxygLfVyPVu+mKO1HitFv 6XWWajqrP6SfpVwqoSswn356iQ2K4XMbh6vhNxO8NJamQLOSfkfP0wnmZk/xlaZUU7nJZ7pRP0kj aTzNw2gfRs3/ZZ/z9UgbxHNKvV5LI+CXn0pv07P0DstjZWwWm8/H8pX8PrGKrOhxPFIHLYG/70Lr byOM9vNUflw8qDymfGXOHz6tj8CKeOgXdC89xdIwU5X1sB+zV9l7vI638l/wd8WdyiPKy5Z2zPpq WkG302P0Octkk9ls9kPWxdax29hP2d3sGDvBPuQ1vIkv42dEl7hW/FapRZqr9Cg/Md1q2mr+cDg4 /Mzw/wx/rpfrt9JsxMNGjP5ndB9mNkDH6TWkU/QuM7EUNgJJZS42j92EtJ7dzh5ge9kjrB+9nGDv so/wSPqUfcXxpOVmbsfhRx6B3HwVTph38nv4caQT/G/8C5EtCoVXTBLTREisxKhuE3cgPSneUfKU 44oOP5ebdpn2mPaaHjM9bTprTrX8GM/4F79+8FzxubeHaXjz8K7h6HC//g6Nwhri6YFXsGkYfTvS Uqz3LkTcPnqFpcJ3eayYVbEr4ZlWtpRdy66HJ29hu9mvjLE/wQ7BS39iZzDmNO4wxnwZn8Rr+Syk q3knvxaHsZ28n7/KvxQWkSLSxShRLKaLFtEpVosbxC6hiRfFW+Jd8Zn4GklXkhWnUqh4FK8yXWlV 1ij3KR8oH5iaTS+Y/mJONq8w32qOmf+OU02VpdEy29Ji2WHZbzlpbUN0HqEn6TeIwAsXOy02ioB4 krbzCUouXmFeQjy3UoeYyRGpfC/bzG9m/Xy06XpzJa9kV9FZxQNfP8f38M94pZjJGthcWsrHxxs0 j1QeBTdNOUJDyiHM7SW0fL05la3nZ8ypFMUZaQrOSM+KcYpXvECvi1PMovyS3lCSWTYb4g+LRkTB b5UqU5Bc4h56QlzLbqYneYAo+SvrNsTxVexR7AtNrJz9U+g4Bl+FKKoQ79FPaBn/Mw3hPt5MP2cd ymLaThPYOvqAfo27YqzpGnOxeRT7A1+iRPglrJ+48ghmN4WNZsI0km5hLWK3+Qx/jdbQcSWZ3hb/ jdEf50+ImcpZ0xzWhTvgZrqVrtU30g2moPIyW0yCzaci5TR2t3WiXHEh34BdpRl72n7c3QewD9SI mZDkIHKuRFzMww6xG+ku7BMKImgJ7vEfYBd7ifrNTTxGi00jGHYdfKl5YXgOLdB/TXfri+kafSeV Yj+4TV+HFvfSX2gH7WWbhm+ibrxKvoZ7+0pTPT9uqtdLeYS/xufyXRevL7xdxHLor0hPYGWqTAcp ovyJ5lK1vk3/I6L7Uuywd9NCHFjfxyw/Rg8zxCBNGL6K9+r1ohvzPUWz9Yd1J0umLn05zaJD9CuL idotXqyxxl7GfG+iTj5HXy06h5fADzvgBR+8tQb7zxZf3bymGl911femVU6dMrli0sQJ5ePHlV1W WuItHnvpGE/RaHehS3UW5Dvsebk52VmjRl6SmWFLH5GWmpKcZLWYTYrgjEoC7vo2VfO0aYrHPWNG qSy72yFo/5agTVMhqr/YRlNlvXaoLrL0wfJH/2Lpi1v6LlgymzqNppWWqAG3qh3zu9UYWzA7CP52 vzukakMGP9Pg7zD4NPAuFyqogZwuv6qxNjWg1V/XFQm0+UtLWG9Kcp27rjO5tIR6k1PApoDTst3d vSy7ihkMzw5M7eVkTcMUtTy3P6DlulEVzYiiQHuH1jg7GPDbXa5QaYnG6ha5F2okT0pew4TqjG40 c51mMbpRl+CMo9FWtbdkMLItZqOFbd7UDndHe3NQE+1oI6BleNGvX8u+8f2cb4poHGey276ttYtI IGeJKo0jkdtU7f7ZwW/VtbtkC6EQ2kBdXlTfFqlH19uwUg3yLK7xTaGgxjahSxwsi4xZxecXP/UW tS1VtSR3rbsrsrQNS5MX0WjODa5oXp5vQD9NeQE10hR0u7RquzvU7nf0jqTInBv6cn1q7sWa0pJe W0bcsb0j0hNMatq3mU44Pa4zOMNccg1zLniWyTG6r8BJUFMXqRhJ0I05TZakczJFFk3GAuAKMdTS OrAiS7SkuraIbaqUY4pMMxXZ3GrkU0IEuIf+drGkPSExF9k+JamUcXIh1DTWfp7XvF6tuFiGiKUO a4oxVhnlSaUl18W4291tw/uzfGmgRvi2PTS1DO53ueQCb435aCEKWnh2MF5WaaE9Sr4ynK15m9QM nteMmic14fOaC9Xb3Ijkfvk+S6M0q+fCL92WdUmga6rGsv6DujOub5jrbsDRWA1E2hJR29B0USmu lw6F36BLcNoldUFh55BJjtuFoY2fkM+b4LgcTNWUIvzMRlB3xCxWRKUhYWq9ZmubEaehZJcrcc/8 f5Vi+llZy8i+qZaYhjbVmxhofNha5UXli4aXGhENTdhyOE72kUjyRTqEWnyUVyQyRDxe9F1qnUbz cGcW4YdXjskSIbvmg8ugacJdZIhD9kTxIkN7olIIl4zO0pJ67JmRSL1brY+0RdpjenihW7W5IwP8 af50pDuA3S4eODH9wFa7Vr8tBI91sam4PTjV9rrZ5tm9PrZ57oLgAD5xqJubglHOeF1bbUguC69r CibcYiyIDH2sIf7QQMQoPThJExgZPIR/WswA/hu5IEFIGSps7fIwbMIHffxXQa4MV0YRCFqgr1Ux +LXPRF+RqgziUU0z9Q+UUaZBfLwqZrN6uRyAz5nOnDh0CWa/tMCXxtLSRpoK7KbCgpFpyQWMimwx /R++3LQ0Ps9WkG1LT+fzsm2pqaCpUpZdhnehY2XeYyeP2Z6l6qHqocwpZS1Dtmdaxo+z+0qX5TK/ xTfKn+tXF2Q2qctEh6XDujSzQ11tXePYZL3V8ar1ZFaGRU1LM88bg29D/ZJxy49EKZJzGQqLVIxR 3apLKjLkKBvTOMZpZ6+0wh0x3uVLOj9mFmOTfZn0ZFGPzedyTxxnw+d4m41jFmd/k5pqnme7oyT5 AJtMBWyKL6s6uzV7ZfaGbCU7S+qys2Sn2TE+us/7/NYcr+0zr7dlaAhZy7XeoX8MGfM71/J+hjHH 8ePYZFzyXZC1UEuIWTxjPO5Ci9lsufzyCeWZo0aaze5CyrBVoJTFRmZlTSi/fNJEmJjFV305JVcs m18zbyGvObS4/9zaE7e8M/z+vVs+fPytcxWztl+16qEHbrrxUWXuiKXjZo6r+vjNRW3Dn78cGVqP g+I69shTe5/++q2WR0Ox++7at8/4YIYI0MfgPP1dF4cw11AwykxEkxnvPxSqWdA4f463ZtWS9uUz m/4PmaRcBgplbmRzdHJlYW0KZW5kb2JqCjgyIDAgb2JqCjQ4OTcKZW5kb2JqCjEwIDAgb2JqCjw8 IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVlVHlwZSAvQmFzZUZvbnQgL0hBT0NCVytBcmlhbE1U IC9Gb250RGVzY3JpcHRvcgo4MyAwIFIgL0VuY29kaW5nIC9NYWNSb21hbkVuY29kaW5nIC9GaXJz dENoYXIgNTIgL0xhc3RDaGFyIDUyIC9XaWR0aHMgWyA1NTYKXSA+PgplbmRvYmoKODMgMCBvYmoK PDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250TmFtZSAvSEFPQ0JXK0FyaWFsTVQgL0ZsYWdz IDMyIC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAgMTAwNl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2Vu dCA5MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0IDcxNiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMg L1hIZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lkdGggNDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250 RmlsZTIgODQgMCBSID4+CmVuZG9iago4NCAwIG9iago8PCAvTGVuZ3RoIDg1IDAgUiAvTGVuZ3Ro MSA3MDA0IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4AYVZC3xUxdU/M3P3kWSXbCLk ucnezZJFsomRAA0habJ5bEADEiDoLg1mA0QCgkQDvoWliuCCYqlSpSpqa8X6KTebiBvQEsVHRRE+ pVqf4KNV+yuC/qrWB7nff+4uKNZfvzv5n3PmnDOPe+bcuXM3Ky5b2UU2ipAg/4JlnT1kXLkLwfIW XL5CTdTtOUTmtot6Fi1L1M9YSmT6dNHSqy5K1HPXE7Fod1enbCev74CfdUNh1IhNAB/dvWzFlYl6 bju4denyBUl7bgR187LOK5Pj09uoq5d0LusCx5VfA6L2LO9dYVQpfyT4mT2XdSX9WZAo7U/fxt96 hP2DPX9XwglTgsDpc6qhu8kCyUHldD6R0q0UkAl1aTfxTZs/3/hlR3rNF9Z8q9Hy/g/GlEhhr9Jz 6zc7TixykNWGaorhLw1oZ6kdPo8aHfTNjm+udiRGkpaTFx+kNnFmvzfHdfAJMZaOAFyMjfkKXINi jCiIVbv8ceHpzxxVkV5fJlT0WG5QFXQ5sAPYAyjUIQphdYCuBiLADmAPcBAwE4FKqwosB7YBRwCz KBDOmOpy1I8RuWibi/tNF9l0DNABQS7QcmAG0AFsArYBZsNPapYDq4E9wHHATH6RHds8HnPPjm0w WP+SpRVGtTNRbZ9nVPsvCCX49JkJ3nROwm1ywm3chIT6rIYEH1Oa4JnFFRF03p9qrxiqzxJZuMks TLwHlPFnKJ0xctG9YhRpABeYqqHxi8z+0d6KbXuEQkxwwWghufQhwWL2jIr6VK7zY5RJLv4pP5qw 8KP9IzIqttWfy9+nHcAeQPD3Ud7j79FqfkTGHLQO2AbsAQ4AxwAzP4JyGOVd/i6l83eoHKgDOoBt wB7gGGDh74A6+NsyYwwq5TqA87dBHfwt3NZboOn8TUhv8jf1If5qrLKqYtAQfOVJwVWcFLLzk0Jm VkWcvxL7eiwyyouVRkbtFkVUS+NFUax4nCsucmI1i11x/kG/6nPdW382P0QawDGTQxj5EKlAKxAG egAzpNcgvUYR4FbgXkADkGWgDkDl+4CXgNfobMAPtAJWfjCGYeL8QMzb4KrP4i/z5ykbEd/P/2zw l/hzBn+RP2vwF8ALYd/Hn4sVuqg+DXZCGwe4A7wcdhN/qn90pkuvz+B7EEEXaDlQB8wAOoBNgJnv 4UWxha5MdLKb9uEZdvEYfWLwP9D9VvIvcfm9jUhAVRLv5J9DAtmmbvNyv3fLnahK4r1lMyRJvDds hCSJ9+o1kCTxLr0ckiTehUsgSeKd2wFJEu+MNkggcX7P46PHuCpnXMzU+nR+BaJ0BaJ0BaJ0BSn8 Clnoa0XO8bexkhJEbKvfN7bEFdnFIk+wyCwWuZ9FulhkFYusYZEaFrmQRXws4mSRQhbxs8huNgmh iDD/wGnVKn8Oi+xjkUdYpJdFvCxSzCKjWURllf44d8fOwVMHFjBYf7186Li7/+e12H3SuRsRdSPn 3dgT9oAeAHSj5oeTWpRwzi2UvKi/pC5RP2tyxfL6qXwvGu7FMuylw4CCBdqLNNqLTvaiu3TQOqAD GAKOATpghncR7mOTQdNBy4E6oANYDRwDzMZ0jmEqnJaDyinuMCZWDloHzJA1vhelCMXN3f4Ch9Ph c0wVm5wsvZDNKNQLeSVlZWFfzsywZsSZfedX9n9/ZaeU+hR+C99EBViIW5N8U+zrAlec3RHz7nbV j2K/oUIFWceqyMuKwSdRr1GfSE6r1E8gJ38YvCLmPB/N0mPeUtcuNkK22un62vmh6xNnnEP82Lnb 9boaV1jM9RdoHt7pOuS8yfVCedwKzRPeOAPbpRqug85Jrkf2Ga5rYNgac62SbKfrOucU18VOw9CV MFzYi5o/3TXLO9c1Ff01Oee7/L3oc6erznmhqybhNVG22ek6G1PwJcQSTHas0xjUU2h0OKcyzrr9 pZYtlqBlhuVnlgpLqcVtcVkKLPmWkdZMq8M6wmqzplqtVrNVsXIrWUfG9SN+n3zrjTQbLz8zEpqR YsgO7DBMbjOgxJmV07mknSFaeMvsBtaiDS2glvmq9uVsT5ylzpyrmTwNTMtsoZa2Bm2SryVu0Wdp lb4WzdL6i2AfY7eEoNX4+jijtmCc6VK1Nl/LbAwOEmMZa2/Ol/zMtTeHQpSTdXldTl1mbUZVc9NP kLChDDf5vr9yvhd9Ob4CbUvL7KD2x4KQViEFvSDUov16ttoeHGSfs+OBpkH2mWSh4KCoZZ8HZkm9 qG0KhVri7HzDj1T2GfyQMWDws+LFLP1ItRYm/LYm/IrRHn6jJYNfSgoVG37FKSmGn8KkX1/v6EBT 32gQ+GSr1Gv49GarP/TZVwyfYhD4ZEVon+GzLysifbRaoxunEy6FIHBheeQ0XJwsz3AxZt5nuJQn XW465XKTMZJIzMbwkQTd2I+c9LEfgc8PAvnfxa4Gn4/1V4cWtAe6PIGwJ9AFhLUNl3fnaJH5qtq3 ICQNqia84fkLuiXv7NJCnq4mbYGnSe2rNtr9yNwuzdWepj5qD7QF+9r9XU2xan91wNPZFOqf0jqh 8rSxbjo11oTWnxirVXY2QY41xWj3o7EqpXmKHKtSjlUpx5rin2KMRUaOtwb7rNQQasT6Sd7P01KR r+F8d6ghy9FTayRvtTtnVf4unFa2U5ovpNk8DZodkHldVl9WL014pqRpBNTpSVPOqmp3/i62PWly QJ3haSDfipW9KyknsLgp8deLC6oVK+VSJKhP6n7ygktA83c2ybN1i1Yyu0Wrmzk32GexQBtuCkE3 +aQuLS0Q14cSyrOgnCwdhTjlKHU1UpeSknT8z1ww5gQ1ojOIg8bufuYvZCuoNyS0wpY2jq2gbS7C 0D43uAtnKfmS6A3hBnuZj/We7E3ehyFTQkO47d6TWLEyKSVjsSLJDddeH/l6T4bkZHc+GSyDGLFa 4cPWZtpFuUCe6UHKVbyE7x/9I+BjyYcX6x9Lu+T8H9jo4kkQbadH2GJ6hPbQ0+w4Wu2gQRogeQRq orvoWrqN1uG1Nheam2gWign621iuPoAvk/vwwryP9sP3AlpFuyiL5eif0GpaK15Fq7VkpyKqp1Za TjezafpKaqfDyvVUSdPoEuphET2o36Jv1n9PD9Cg+LN+gtIojxag7Nc/Nf1Vf5vK0OJ2upMOs80p j5Efo0TgeTddRlvFPIXpi/RvMAM3XYE5KDSd9rMh7kPvXfQRy2HXikb08jtd05+Bl5PmUTdtpV1s IpvC3aZ2fbq+n7IwxpXo9U6K0U6UOD1JbzKb6bj+e/045VIpnYP7GaCX2ZAYPrFmuA5xMyFKY6kK luX0J3qeDjIPe4ovN9lMFSa/6Wr9EI2kcTQHs30QLf/OvuKrUFaL55RmvYFGIC6/ktGmZ+k9lsfK 2Qx2Ph/Ll/N7xGVkxYjjUBbSYsT7DvT+LtJoJ7fxA+J3ysPKt+aC4SP6CKyIl35Ld9NTzI47VVkv +yV7jX3AG3kH/y1/X9ymPKS8YunEXV9Iy+hmepi+YplsEpvJfsG62bVsHfsVu5PtZwfZx7yet/GL +THRLS4VTyoNKLOVXuV6042mDeaPh4PDzwz/7/BXeoV+I81EPqzB7G+ne3Bng3SA3kA5TO8zE0tj I1BU5mZz2DUoq9jN7H62nT3EBjDKQfY++wSvpC/YtxxvWm7m+Tj8yCOQh1+GE+Zt/C5+AOUg/yf/ WmSLIuETE0WNCInlmNU6cSvKY+I9JU85oOiIc4Vpi2mbabvpYdPTpuNmm+WXeMe/9N3vTpSceHeY htcPbxmODQ/o79EorCHeHvgEq8HsO1GWYL23ION20KvMhtjlsRJWy6YhMh1sCbuUXYlI3sC2sgeM uT/KnkCUXmfHMGc7dxpzPotP5A18BsqFvItfisPYZj7AX+PfCItIE+lilCgRU8Q80SVWiKvEFqGJ l8Q74n3xpfgORVdSFZdSpHgVnzJF6VBWKvcoHykfmdpNL5r+Zk41LzPfaI6bP8OpptbSaplpmWfZ ZNlpOWQNIzv30mP0ODLw1MWOiDUiIB6jW/h4JRefMC8jnztooZjOkal8O1vPr2MDfLTpSnM1r2bn 0XHFi1g/x7fxL3m1mM5a2GxawsclOjSPVP4IqUbZS0eVJ3BvL6PnK802toofM9sohjNSFc5Iz4qz FZ94kd4Uh5lFuY/eUlJZNjvKHxStyIInlVpTkNziLnpUXMquo8d4gCj1W+tG5PF57I/YF9pYBfu3 0HEMPg9ZVCk+oOvpYv5XOorneD39hi1UFtEtNJ5dSx/RH/BUjDVdYi4xj2Iv8MVKlJ/BBogrD+Hu qthoJkwj6QY2T2w1H+Nv0Eo6oKTSu+J/MPsD/FExXTlumsW68QRcRzfSpfoausoUVF5hi0iw86lY OYLd7VpRobjBV2NXaceethNP9y7sA/ViOjQ5yJxpyIs52CG2otyBfUJBBi3GM34BdrGXacDcxuO0 yDSCYdfBLzUvDs+iufof6E59EV2ib6Yy7Afr9GvR43b6G22i7Wzt8DXUg0/JN/BsTzM18wOmZr2M R/kbfDbfcvr6ItrFLIf+gfIoVqbWtJuiyus0m+r0jfpfkN1nYoe9k+bjwPoh7vJTjDBVDNH44fN4 n94senC/h2mm/qDuYqnUrS+lGfQEPWAxUafFhzXW2Cu432uoi8/SV4iu4cWIwyZEwY9orcT+c5O/ cU5bvb+u9uc11ZOrJlVOnDC+YtzZ5WeVlfpKxp45xls82lPkVl2FBc78vNyc7KxRI8/IzHCkj7Db 0lJTrBazSRGcUWnA0xxWNW9YU7yeqVPLZN3TCUXnDxRhTYWq+XQfTZXtOmE6zdMPz4t+5OlPePpP eTKHWkM1ZaVqwKNq+5s8apzNnRmEfHOTJ6RqRw15uiHfash2yG43GqiBnO4mVWNhNaA1X94dDYSb ykpZX1pqo6exK7WslPpS0yCmQdKyPT19LLuWGQLPDkzu42S14xa1PE9TQMv1oCm6EcWBzoVa68xg oCnf7Q6VlWqscYFnvkbypOQzXKjRGEYzN2oWYxh1Mc44Gm1Q+0qHohvjDpof9tkWehZ2tgc10Yk+ AlqGD+M2adlXf5jzfRWd40y27ofWfBEN5CxWpXM0uk7V7p0Z/EHbfLfsIRRCH2jLi5vD0WYMvREr 1SLP4hpfGwpqbC2GxMGy2LirxP0lTr3F4SWqluJp8HRHl4SxNHlRjWZd5Y7l5fkH9SOUF1CjbUGP W6vL94Q6m5x9Iyk666r+XL+ae7qlrLTPkZEIbN+I9KRgs/9Q6ELQEzZDMtyl1DLrVGSZnKPnHJwE NXWBipkEPbinSZJ0TaLogklYAFwhhlbaQqzIYi2lMRx1TJZ63CLTTMUOjxr9gpABnqP/PF3TmdSY ix1fkDTKPDmVahrrPClrPp9WUiJTxNKINcUca436xLLSy+Pc4+lx4PtZfjRQK2LbGZpcjvC73XKB N8T9NB8VLTIzmKirND8/Rv5ynK15WFqGTlpGzZGWyEnLqeZhDzJ5QH7P0ijN6j31l+7IOiPQPVlj Wf/F3JWwt8z2tOBorAai4WTWtrSdVkvYZUARN9iSknZGY1Dkc+ikxPOFYU2ckE+64LgctGlKMf7M RlIvjFusyEpDw9RmzRGemqChVLc7+cz8f43i+nHZymDfN0vehjbZl5xoYtpa9Wn106Zni4qWNmw5 HCf7aDT1NBtSLTHLc5IMGY8PfbfaqNEcPJnF+MMnxySJUL7mR8hgacNTZKhD+cnqaY75yUYhXDI7 y0qbsWdGo80etTkajnbG9ch8j+rwRAf50/zpaE8Au10iceL6rg35WvPGECLWzSbj8eDU0Odh62f2 +dn62XODg/iJQ13fFoxxxhvDDSG5LLyxLZgMi7EgMvWxhviHBjJG6cWJiSDI5CH8p8UMEE04pUFK 4ewqfykBMaHgdG0hcme4M4pB0AN9p4qh7/wm+pZUZQieskfFYdpFqWRnXv/PMoO2bttW20O2F2ym aWKa/TZFZDL8TmMzC4spNU1YyGaz2/cJZaQQirATt9kVi9iNH0utGPZefyopClxoX6oS5xc9bjKl +gtcE1LxI8+Aw8HnQPj7gN1uCJ9CMEOD3xXtFn+RZ4Il4p5ouTWdY4H8afaRE4g7uMoF6keMNhA+ 3Cnb8MdGxNnGvg05Psc/fb55PseXPt+/5BdYjePvjhM1VFfn+FfNlzUZVVUsI7Oqat1ZPuU6xzPp 6enjzmbGt6JdfzeWWWWP64f8aeOrRFFZlVAKCmpkF6F5NA8+/pE2f1qVLdJaZfN7q2xFTvCyKung CyGUE9n4jPGjPBkig/EtJ27gd//6uecGhieyjgfEzu/OfWD4Pq7w209cLNcCwKWPwdn2py5pzzYM jDKTK2vGtwhNrZ/R2HCBr/6yxZ1Lp7f9H5+pL2kKZW5kc3RyZWFtCmVuZG9iago4NSAwIG9iago0 Nzc5CmVuZG9iago4NiAwIG9iagooUXVhcnR6ICVkKQplbmRvYmoKODcgMCBvYmoKKE1hYyBPUyBY IDEwLjkgUXVhcnR6IFBERkNvbnRleHQpCmVuZG9iago4OCAwIG9iagooUXVhcnR6IFIgRGV2aWNl KQplbmRvYmoKODkgMCBvYmoKKEQ6MjAxMzExMDUxNjExMThaMDAnMDAnKQplbmRvYmoKMSAwIG9i ago8PCAvVGl0bGUgODYgMCBSIC9Qcm9kdWNlciA4NyAwIFIgL0NyZWF0b3IgODggMCBSIC9DcmVh dGlvbkRhdGUgODkgMCBSIC9Nb2REYXRlCjg5IDAgUiA+PgplbmRvYmoKeHJlZgowIDkwCjAwMDAw MDAwMDAgNjU1MzUgZiAKMDAwMDExOTYzNyAwMDAwMCBuIAowMDAwMDAyNjM2IDAwMDAwIG4gCjAw MDAwMDYwMDkgMDAwMDAgbiAKMDAwMDAwMDAyMiAwMDAwMCBuIAowMDAwMDAyNjE2IDAwMDAwIG4g CjAwMDAwMDI3NDAgMDAwMDAgbiAKMDAwMDAwNTk3MyAwMDAwMCBuIAowMDAwMDQ3NjY3IDAwMDAw IG4gCjAwMDAwODU2NjAgMDAwMDAgbiAKMDAwMDExNDE2MCAwMDAwMCBuIAowMDAwMTAzMjkzIDAw MDAwIG4gCjAwMDAxMDg3MTkgMDAwMDAgbiAKMDAwMDAxMTMzMCAwMDAwMCBuIAowMDAwMDY5MTU0 IDAwMDAwIG4gCjAwMDAwMjE5NjUgMDAwMDAgbiAKMDAwMDAzNjkwMiAwMDAwMCBuIAowMDAwMDE2 NTM5IDAwMDAwIG4gCjAwMDAwNDIyMjYgMDAwMDAgbiAKMDAwMDAwNjE1NiAwMDAwMCBuIAowMDAw MDYzNjI5IDAwMDAwIG4gCjAwMDAwODAxODMgMDAwMDAgbiAKMDAwMDA3NDYzMCAwMDAwMCBuIAow MDAwMDI3NDE1IDAwMDAwIG4gCjAwMDAwNTMxNDMgMDAwMDAgbiAKMDAwMDA5MTExMCAwMDAwMCBu IAowMDAwMDAzMTAyIDAwMDAwIG4gCjAwMDAwMDMxNDcgMDAwMDAgbiAKMDAwMDAwMzE5MiAwMDAw MCBuIAowMDAwMDAzMjM3IDAwMDAwIG4gCjAwMDAwMDU5NTIgMDAwMDAgbiAKMDAwMDAwNjA5MiAw MDAwMCBuIAowMDAwMDA2MzI5IDAwMDAwIG4gCjAwMDAwMDY1ODkgMDAwMDAgbiAKMDAwMDAxMTMw OSAwMDAwMCBuIAowMDAwMDExNTAzIDAwMDAwIG4gCjAwMDAwMTE3NjMgMDAwMDAgbiAKMDAwMDAx NjUxOCAwMDAwMCBuIAowMDAwMDE2NzEyIDAwMDAwIG4gCjAwMDAwMTY5NzIgMDAwMDAgbiAKMDAw MDAyMTk0NCAwMDAwMCBuIAowMDAwMDIyMTM4IDAwMDAwIG4gCjAwMDAwMjIzOTggMDAwMDAgbiAK MDAwMDAyNzM5NCAwMDAwMCBuIAowMDAwMDI3NzgxIDAwMDAwIG4gCjAwMDAwMjgwNDEgMDAwMDAg biAKMDAwMDAzNjg4MSAwMDAwMCBuIAowMDAwMDM3MDc1IDAwMDAwIG4gCjAwMDAwMzczMzUgMDAw MDAgbiAKMDAwMDA0MjIwNSAwMDAwMCBuIAowMDAwMDQyMzk5IDAwMDAwIG4gCjAwMDAwNDI2NTkg MDAwMDAgbiAKMDAwMDA0NzY0NiAwMDAwMCBuIAowMDAwMDQ3ODM5IDAwMDAwIG4gCjAwMDAwNDgw OTkgMDAwMDAgbiAKMDAwMDA1MzEyMiAwMDAwMCBuIAowMDAwMDUzNTE5IDAwMDAwIG4gCjAwMDAw NTM3NzkgMDAwMDAgbiAKMDAwMDA2MzYwOCAwMDAwMCBuIAowMDAwMDYzODAyIDAwMDAwIG4gCjAw MDAwNjQwNjIgMDAwMDAgbiAKMDAwMDA2OTEzMyAwMDAwMCBuIAowMDAwMDY5MzI3IDAwMDAwIG4g CjAwMDAwNjk1ODcgMDAwMDAgbiAKMDAwMDA3NDYwOSAwMDAwMCBuIAowMDAwMDc0ODA3IDAwMDAw IG4gCjAwMDAwNzUwNjcgMDAwMDAgbiAKMDAwMDA4MDE2MiAwMDAwMCBuIAowMDAwMDgwMzU2IDAw MDAwIG4gCjAwMDAwODA2MTYgMDAwMDAgbiAKMDAwMDA4NTYzOSAwMDAwMCBuIAowMDAwMDg1ODMy IDAwMDAwIG4gCjAwMDAwODYwOTIgMDAwMDAgbiAKMDAwMDA5MTA4OSAwMDAwMCBuIAowMDAwMDkx NTAwIDAwMDAwIG4gCjAwMDAwOTE3NjAgMDAwMDAgbiAKMDAwMDEwMzI3MSAwMDAwMCBuIAowMDAw MTAzNDY2IDAwMDAwIG4gCjAwMDAxMDM3MjYgMDAwMDAgbiAKMDAwMDEwODY5OCAwMDAwMCBuIAow MDAwMTA4ODkyIDAwMDAwIG4gCjAwMDAxMDkxNTIgMDAwMDAgbiAKMDAwMDExNDEzOSAwMDAwMCBu IAowMDAwMTE0MzMzIDAwMDAwIG4gCjAwMDAxMTQ1OTMgMDAwMDAgbiAKMDAwMDExOTQ2MiAwMDAw MCBuIAowMDAwMTE5NDgzIDAwMDAwIG4gCjAwMDAxMTk1MTEgMDAwMDAgbiAKMDAwMDExOTU2MSAw MDAwMCBuIAowMDAwMTE5NTk1IDAwMDAwIG4gCnRyYWlsZXIKPDwgL1NpemUgOTAgL1Jvb3QgMzEg MCBSIC9JbmZvIDEgMCBSIC9JRCBbIDw5NGEzOWE4ZjA1NDkwM2JjNjE5YzM3NmYwM2JmNjRjNT4K PDk0YTM5YThmMDU0OTAzYmM2MTljMzc2ZjAzYmY2NGM1PiBdID4+CnN0YXJ0eHJlZgoxMTk3NDIK JSVFT0YK --089e013a23c0feb3ed04ea891e18 Content-Type: application/pdf; name="lossdist3.pdf" Content-Disposition: attachment; filename="lossdist3.pdf" Content-Transfer-Encoding: base64 X-Attachment-Id: f_hnnfje7x2 JVBERi0xLjMKJcTl8uXrp/Og0MTGCjQgMCBvYmoKPDwgL0xlbmd0aCA1IDAgUiAvRmlsdGVyIC9G bGF0ZURlY29kZSA+PgpzdHJlYW0KeAGEvUuOI82SpTmPVcQGKop0OknntHJQQM0KNegFXHR3opHZ QFYOevv9nYeoGv0y6Egk7u8nxORxKKoq+jL7j9//8/d//P6v//Kf59//+M/fZ//ff/7j93/97wD/ 93/+Pv2+nj5///vvy+3j68v//W/97/9yud6/Pn//GyL5z1//9vtff/9fVnd9/OGp++XP5+fvz/Pn n/vH78v9/Of68ft//5+//4/f/+/v05/79ff/h7H/wf//P/9k/teYt1//8r+wof/7X/+Cq2f/53/R /8i1f/z77/v9zxkztz9fONo/vv58PM4fv+/XP1+P8/n+6/P8+HP9ff/8I4f0n//4ff/48zh9fV0K 4N+d/5wH85d0Ipk/rn8uyP86PPeJ4erUfyJZeyjyv8WzPDiuRee//v5vv8T8y3A+zp8Pq7hcvx4J qhD6CeP253S6neTu/c8JkV84cc0zAeTK5c/p43JXeMUI4ytC0bL+/PxjOw5UErD18efrdr8flZz/ 3OTMtlSAp7Y7I3T/U3+j5xe/ynNM/yBX/ptz5SUBl4/HF46f/1w+lH08XgR95+v5FybPH6eHGPgk THLt888tz+Rvx3++fN0Uf0XOfx4RiRL9rAKINnYcf0ROjz9fl48vxV8lpy8MHOz0b0dfVwIp1ji7 lAyw7LyN/fPzk7hQ9vg6f/1y8IHOpz+f/J7E/sHvpdihGxkFf/dDAzj6j+vlquhH6IM2+amWUDX5 e1lKi5AI0P3P5+N0U/jVcsLok6kCv0xA/VlCPP6sZwGJCluvKfiV5ny9Pm40aen7vLRVD3b/c7+4 CVyuJzLUREkIEr76GEZA5Nrlz+X2pU5IdFrq48+5UlW0gVrjsY3B7PXj9uug6PLn+s1cETMxTo3U /c/4PZo2suy95qJd2+3+eRUXNNLL6ZrGYOzX54l2/HGmld7+fN6uN1oMP9xJUpDx6HNFTMbn1+Ui Mkbq48+HpA6aimx7ZsNSYDScy4MMpGOvJpLraA9NRczGeDVS9z8TzWjaSONbmfHrZcdw/3p8io0P uiz6d7eOwS5/vj7VRG9/rl/0ikhd6R3PH7/olU99Lki6x+vjC/GR0q9+qdRo2kjsKaU2dqKPFBfR c3k86CEz5sTaQsxFfBpMvcT4HT3779j6axtpXjxOnxcxcfrzcbqnmxgIuglNRNwe108RoaGOPLki m6cC4Njlz/18OYuGynz8YVi9KCdGTZGxlY7CQr8vjy+eh+ul5/IgwY+2CqSjiD9L5o5b8merGaRx /ZUEdxRff07nx4cevzPM391RbOyhn1MkfJ1P9A/SeJUQLFz82CDpKL4+vignBtPvfKvyKtpArTkZ RogmeLq5o4iiy+P65/xkbiFOhzi1sPv2ezRtZNl701F8MVDR18MFP/IHvwhNo9ivy+PGr3QzGY+P K+OaGOO/TMZnnyvinHh8fjw+D1IfdLpoP2gqsu2ZDUuBkVWXTzeO0UQSHe2hqYjZGK9GakczmjbS +N52FFRxDGBiQxpPZWMw6oHzGTboBslysSGdJ0q7K517nguiXEfale5IKTW+KjWaNhJ76ShG6kyJ 8oCNpemL8kJebXuDYG95VewXv+N4Ppo20vgWG68Kyi9GwctDbJz+3BhJ1FtsjMHucsY7atXrFaf0 612Qkne3PlfEbJzvDAdLyuVwpUYTPcmTvfQYxYiKEpNBhAI5mr7IPHm17RXRc8urwbbnS9OOpfH9 wMbn/Yvu8fJFTQArJqMQBH1+fdB1Uq/eHtSG0EYnSpV//XPPUwVMxcfX/XaFispQcVdm1CwkttLT VIiAPh8PV9TV80VF8GSrgGmwPzhdmS9qQaL4dVlqiqy4fiDh+rhoWoa+r09Kh39Xmg0GoaqoqFu/ PulD6UdpFwjBwlcfK2IaGDuu0LCkKJolhW9VVGBb47GN0R9fr24eo4jU+GYuiPjbTo3U9ns0baTR /Y2LDiP309dNXNDoHzdPOBjWitF1fjFZxS6/1l1kkK6SYmQ5VSaIvNM8jJLqIEXT+aZpI7VnNkaK 2uN0pWbZmmji3+wVMRvj1Uhtz0fTRpa9twPJF+W82KDrpURJ+zDGD4qGO/0qbNz4icQGw5WkYOPc 54qYjdsF8CBF5Sipg6Yi257ZsBQY9eiX5qFM76qJLvBoD01FzMZ4NVJfdLSJZjRtpPGt3HhVcdJF 0AbEBgPeldmke4vBPumY6Qth4/7xuIqN659PpDQ//ehzQTKQ3D/vH2KjmGrHSo2mjcReeoyROtMr MRmAjdF0YnDCq4O9ImZjvAqmFj6ej6aNNL7FxquBhGWd6+UiNpgxX+6enW6MQfDDoyozLtYviIzJ iAquGw08jxVxatDkziIjmKZYt0pV0QZiTRwu7M4Q+nlzr1FFRHc7mvvFH0HMRZ0ajOcby9K0kEaH vbaTV5nBMHbTUHHh+cf1w5mxMDR8fmqaSs9ApyoyKB+RUmbQtem5IsmMx/3OtG1hxHmv1GhaSO0l MypFVJ83qj8yo5qwTKXnzKi9QcxGvSr2C50TzWjaSONbbLzOjI+vC5MfcX6/scxAO3nQdo19MUdh Gqyi43SnsiROokJKqXGtTBG8Y472OFMvjZRSg3U6aV+aFlJ7yY1KKSovEixN2DnJq22viJ5bXg22 PV+aViwT3w9sMKM5XWCD0eqTocVsDEYPdFGvQUnL4uSDHoz2KSnsM3HTc4OYjY/TjYnmwmCDysTa R1MRMr72eG5LabVPvQZ9dzSpN322N4jZqFeDwcZ4Ppo2suy9GVEoFc4fjwveMUY8KDXNhjEzRKLT SVAqnJjaK4FghYatiRuPbcRkXFjwuLlpR4oioFIMSVa0kZrjuY3RdZ60QEKpUE0MQN/sFTEZdWqE GgtOjaKNxBzt8i0Xt487a84XHGBaGioGojEwXRIVrFEjwrzv5B7j0WcMKGUZQC43er/KKMCPyoyS jcSSWbAQ7NGNega09Xwwf3VvMbYCmIN4UwlWOWopStR3DDKW3hJAD8Aqpgg8nVmaczIYwy/mX4xn YoChQBRo4ooQVJz6WBGTQFojvaVYQ6pUFKmKGqTmTMNgmjxdiBrpamKk+WYviHrb5dQIPRiGE8so 2sgy95oLF54MYdc7o7d+DxaxPUvdmArAdJq3T/ZeRM5dQnBx7mNB5JumarTpgxRbHZUaRRupOXMx UmzPXC4sYGxNJ62+P9krYi7q1Ahtv6NI3b+jc7Y5uh8axoPaSFywLs2qdvJiMMoihlEVMPB1R4jW JyG4+OhjRczF/et8ERcjBQOViiK1FyM4V3PmYqRYKVLRCb1RdKPDezY3iKmQT78uAxH4uF09G1jG XmeFlvbU9X/dP9VdnpkGMZfYmNm5XdJCvu4aFUgdKhtXGsxMPpNMRszE1+PG8vWWouevFPWIFW2k 5sxEpYhJRZmoqKYbndOzvUFMRZwaaGJhy7CKNtLo3naXSH+wiaWgTgyWXuE7YEy7mZCRFI+HJhla l0FGS6vXPhVEPSYFwJmNwCWlFPiqVPVsoMb+of6jQoR0e6jeXIpuzAjl0jY3iJmITwON16RJFQ2y Y3vXPpC+sFzCugtMnqkkmbBvDHZY/+dn4t9ObMoqdb4khG+3PDaIqTh/aPdwS1E1Soo8GUVFtjme 25hX7UVFFd1U1D6ZK6KuaXwaoe025LAfoIz/FtvfmFCvifQnQ7eY0LoRfd4T9uBfta5JwUbiqJ7Q mpDWIrXf4+eKyDdtwWgbcUtRMlb7aNpI7fHcxoj8emHtamm6UWY92xvEXNSrwQ7RVNMBWfbe9hbX zzuLdZQDGh7Ub6LBmHONORZpQP/F3AI2yB72SNxxfvW5ImaDhRg2FrcUA7qkDpqKbHtmw1JgxM7e p3gdTYyDR3toKmI2xqti4/nWdEAa38qNF5MypG8syIsN6js2JsLGYFSUVCpig99Ma+W0xRNS6jHY cPRzQdJjEDrDwEjpV/+o1GjaSOxlKB4pTVK0jsovW010uvLqYK+I2Rivgql9j+fVdEAa32LjxaQM 6TujpNhguGMWyKTsgNGRaZGP+vX2paVymtOHV91EQR4r4tS4fl1vIiOY6gTqLiuvog3EWurUEaLk uF474YsiuqijOTKjiLkYp4od/K6mA7LstZ28zgzmC1q3vmkJwVMyNAzmrHb/yRxCK+W0JkZHluGY pPW5IMkMzSFERjHIYA5h7aNpI7GXzBgpVspZOCYxRtGJdQLNALe5IiZjnAqmxBjHo+gANLq3ecHi GHtV4eLEDqHyYmOwydihxLizfyUuqIOQknOXPlfEicGBDGbyI+UCslKjiZLyyV4yYzBGX2p7kTGa GA/k1cFeED23vSp28LyaDgi5Ik0/sHGmkxQbbLzyk6rPoBQqRstjuiO7nCe6swMGGydJ4d1nZYqY jcdFrWpLqUSP9tFUBE21x3Nbip/2k8V02KgmzlLdnu0NYjbq1WAHz6vpgCx7b8YTFgbpLq5ujbc7 K7xmwxhRsWjPNgDpdvq4c0JDHQsN3DXYVY9tBOc0Z2OzEVUjRRVZqSgi7gFUg6HIXAymMkx9RhX9 ZmCntL5x3ElTMZsrwnPbqcGIfPyupgMSez9lBocMVFne6Im/2C0xF4PR9HQADrvnT5baCJOFfUnh Hau0fq6IyTjfvNe4pK40+UhVE2wYQVPtmY1K8QuzKOaRu5qujFrP9gYxG/VqMGKv56NpI9ve28xg vfWDzOB5xlGPrVq003+q72aN9y42Pq7MikGYE+ufYONemSJm4+POXiORjhSDyLMmWkCRscdzG6Mv YmVAuVFNxHl+tjeI2ahXgxF7PSdbrGmQHd8Pvcb166z1a7xjjqK5OxqKiU9WfV2h36mIzM9DUrDx ledgzIjZYO+MA29iMVJU1pWqJiIvMvbMxmDUtufs4VbTlWWgZ3tFyHh+I+ZK8mqktufwY00HpPH9 jQ3X5LSmx00L2LDJqkPIKAQZpxvNArPMl7TyTZh0H+KCjTQ9FUBDpKZuXu0YGW13RqZqYKJIbZmI gegq2LLMFNB6rnQxT7YKmIb6U4iYJ4qo0S9qhJxIXH8jwRN3SpUzww0ksE7H8cGwMBglOIf0XNSd PrSWDS+sfbs4P+e5QcwDlSbjzpaiiKz2aoKIQWrPTAxGaaKzMTBaTUR5frZXJClxi1cjRejjeTUd kGXvdXfRlNABE7HBUMYGZtgYjMFTp4TIifv5flIqaoFAhyQ0cctzQZIVOk4CG0uK9K5UNcHGILVn NgajHKNY9amramJ99Ju9Is6M8aoYsY/n1XRAlr3XbDg3dNjyxDay2Pi4ab/9N/s/xkguGh1zVshg l+MhMmgZEoIMTmL4sSJODc6VaLNxSbFtUCkr0mR0gFozF4OxKnbPsDqKWBX7Zi5IMmOcqtT4jQPV dECWvddcODMY7tjvZ+2KXLvqzJK5GIymx06vyHjwk4UMZpBksqZteU7HXziNZDLYGtdmI2REipq6 UtUEG4PUntkYjKknW+iZAkYTi0bf7BVxZoxXxYh9PK+mA7LsvWajmcGOulaxYeN0Zq3LbBgjM9SN s/5Ntp1YyGEhzB20VuSYt/U5ddkgRKWZm3cblxRFtaS2Jtgwsu2ZjcH4jT/Ua2xNFI3SPvbQVAQ2 tlfFiH2iqaYD0vhWH/pqmoaeL6aqeEfNyAHmsDEYfbnOFWOXTSAtnBMnu7hM4Okr+lyQsMGUXruN lXJRXalqgo3Hkz3l1Mb4jW/aQ9WsLJoYEOTVwV4RszFeBftF7ON5NR2QxrfYeDWBpzg6MdkgAgrJ B4Wvc2Mw1ibZF4AMZqxaOSdZTgjJua8+VsSpcdGu6pJSwp8rFUUHINbE4RZSaaJeQ5PAKFJpcjBH ZhQxF+NUMSIfv6vpgCx7bSevM+PGUpa4YHmEY73JjMFoe5wvExkcw9QqOK2Jg8ncFNDELc8FSWaw CHYSGcWIk1matVfTAYm99DYjRQaq+tQc0Io+MSOntrlBTEadKqbEGMej6AA0uh/y4n45Ue9R0NNL ZgKvA5nB2Ktk/ikurh9qumpLTJY8VeF0uJ8r4sTgdAnH0EZKicFJc0tV0wGJvWTGSFHXXbNtWU2E +SGv6MNrrwjPba8GI/bxvJoOSOP7gQ2mzNxWEBtsndyTGYORdzovhF2m0RzAUFviRJdrj3OfK2I2 aN868jZSxM52h7VX0yBoqj23k0oRFWWFpz3V9EmLebY3iNmoV4MRez0fTRvZ9t6OJxQHD/X4TMVS dnEeU5AJIvSzyLhfz2elBv/Gv/BLUQ1ZJICpYH1PNZcRpQV7cEctBySmzEOFiIe1T094ouaT5w+W fl0KmIR6MzLL4dECBxPCmHpNgQsMVn3ulPa/L1DIipWH1IVBC+tabh5ft5uOURAg62EuOSjR9VwR NXnN17TjOEIEzZ6FhapoI2POPFSKkFjW8kJiFH1SnD1bG8RM1KfBOIQ3sUTR70EgsNG9bRwstHA0 XlwQiSasDCELg4sLKzm/yAdOlGj9mxQjMWBHkzU/N4i5YJaaNLIQkdO0pLyKNF4YwbmaMxeVgovL h04ZaOInRUTJptzR2iDmwj6haqSW36NoR7LNvc4Ll1oMM5yzYasWaW4OuAgv5lwhKFflj6vWtZUo EoIKFii19j0IMbEYw9qNG5mFCJyrDBJSs1W1uoEaMxMjpNkKDFQPTzEQYSzTw+gJoi6T/IxLldIv 6EjMoBUdkJgjd98ywZkOrWCLSU/N6OqL8MPpGgtWP87knsM+3eQZkeoZaKA/NAkfH2VKl8Vs2Eqr 4oDYDI9Uhp+ACuymc6Oa70nJJ0OSO6Jayd+Ovn5UYnsaFb+IfXyvmbeRs12n9Wqyij2G9g2D4d71 cXXh9nHVCjm/5oekiP6rzxVx/JdPja8jRHRcwrDyKFK8RpzENmcORoqK+E53DAdVxBrxN2tFzMNF PqFqpCixa66KYGKQRvdDFtw4ia/2AJPUT+kbjBEUXHK+WXnwyR0pccEaiKTg4qHnNmIuPq86jj5C RM6lBQtV0QGpOXMxUtTDZ/IALqQI3bR2rCnvxlqQtIjxqRiRj99VdEBi7qcWwbFFNgchlIKRxHQ/ OZiXWtw5XK9XVnHEzk1S9JzUm36uiLm43rX1PEJEzmq1harogMRchpqRIqE/6FbhoorUFp+tFXFe 2CfMBVN7GL+r6IA0uh/y4uvOdpa5+GARSdMOcq4YexVc/lNe3O4nrhOqJT201AQX58oUMRcUEdpe rJTW7j+fNR0QllukyYkxUhREOnGuGV4UMUQdzfFjFDEZ49RIbcej6BDJstb+4tW0Q8couT2K/3BJ ciQvBoNN2qS4uH9ddTpDYxVSGlG5QeXngqS/5DCJdhcrpe6BE3+WqqYDEntJjJFSOZzJ+GiiV5ZX B3tFTMZ4FUyZMZ5X0wFpfG8zg3OU3G0UGxTDumNLZmyMqRN9odh4nE5aG6df8QVevCNKP1fEmUGO aX+xUsoMLl1YqpoOSOwlM0aKtqc5G6lRTRdmvvJq2yuSPqNeDXbwvJoOSOP7gQ1Oc6hEJgF1Odpk DKQhmrU8CIaKE6UCHcsVIXzj0KGfCsAPpXkSm4soCkTcXFuwTNUcENtKWowQZRzXgw56CPH8ZKsA turPb85XRoZWIFs4OGqKrLh+IIFDHFqphgQW7nwwiS64GG1O5ymxyildrY0TIaeu4EXzsTxXxDxw kET7ipVSSnBjwdqr6YDUnjuLSumnlXLtyVnRhb24o7lfOBbEVNSpwfj96/goWsBEt7h41VnohCXH RfCfBqctc2fEYHTPVPfiggNCWhlXl6Ltabi497kgSlfmOWxniYtipDlrXdZeTQck9pQVS4oQKBJN RjURprbDxX3tDWIy6lWxX5BRz5emhUx8i41Xq1fkMotyYeN0e+hOADoHoxncVF6Q4FzLFRt0okjJ O4ogP1fEbHDUWcfDK9UbC5aqJiJnc/pgL11uMaJiHcBTjWq6qFjEq22viNhfXg22PV+adiwqiNH0 AxtUdkyI1e4+76zfm41i8Plg8oBZ7hPd1FloTxghnHvksUFMxufjwwVGpQidqxZSXkUbqLVkRoWI 6cHpL+VYFV1YI382N4i5sFN4HillxsQymjbS6H7g4k5U4YK9sCRGIaXWieDggveN6F4hTU/vcmFY OeWpAmaCg4zaUixE3NxVYARwC5SaQcy79t3cRCwExHRV93AhonoIsZOf2goAfdufkWH3o1GMmo3w g9jW65mIVylUiemwnHokGpq3CzfGmMY5W1nl/J0OVhAhq7ymgVU9PVdEvsEZy/bhwVKEzeECa6+m A1J7ZmKkqNi4j+VZXjRdqPOe7Q1iLurVYKTERDOaNrLsvWbDc3NmMLxgRGyQUpp/0zyK0WeTVJ+a eVDMcbVUKwf8YpIiKXhPi54bxGywZKNNxcGInfcX6arNaBpk2zMblgKjTL+x2wKv1USc7Ddue2gq Yjbq1WDj+UHTP8W3GsirgUSnyCgf7N2Za6dhYzCdIOYfYeNBRwEbRMUSLl2Z5mp5LkgGEk6QaFex UmoSHBS39mo6ICz4S5PZGCky4ZSuczQxEMirg70iZmO8CvYLNsbz0bSR2MPPt7nB6Q0tUavl8QYN 9RdiPxiDmg5zkoFs5GnZmjg/kJJ3HDn0c0XwTvM1bytWSgMJh4wsVU0HJPbSZ4wUv6ROFhw0kZXy 6mAviNjfXhU7eD6adiyN7wc2OL3BZFAZSP3sfoO1omI6D8NEBLtn9ntZBlbvIim84xSGnytiNjhB omvgI0XsXF6wVDUN4oy3PefGSGkOJS5GD+XlN2tFzMX4NFLb7+jZfze2H5hgIfauVg2DX764DbvG nCt3jr+KCQ7w0Xm4Z5EUTGgDIK2LdUheIqYIWCznwOeWYlCvlIY9adpI7fHcxuipuQd61ET+fbNX xFyMVyO1PR9NRXZ8P7DBEQ5mAbzih9GAWalbyWBszFMJaejm/Q5ae6Wf5dgaUWm2lueKmA12wAlm S1GOVmo0baT2zMZIUeRSxYuN0cTw/c1eEI1f26uR0pnReDWaNrLsve4zPLqSPy6gxAb7Pu1BB2NF iI5EdjmFxk+rOE+Sok899bkg8k4zNm8qLilmSpUaTRupPbMxUkxedM78oInpzzd7RczGeDVSO5rR tJFl7zUbHV2ZFlFEXBgyTvT8To1CtDmt8cAFMyUdy6CXvUkILigh9FQBU3H79J7iYDQArmNIKHrU IgSQtLVlIiqjO4Qp8UcN7e7ZVgHTYIfQNEKQHVujZwHL1lsSHhSydsxHRUyCIJwnpXzqRP+rM6cK hzdNuQz1Hf8FmAReBaCtxCXkGxdHPQvQZA81JoFz0pLR7EtJZSX8LVlOs0B3LQXI6DHuFKN/q8fR sv5cdiDgLy8W5FUN5Djr1Rc6QM6JKAsWpF6So3hkAS1DS9jExgrcRW5p33QB6SxJZ20fVkg/Ouks YPQsoLZMgGTyc3IIUhSMGnpRObRtFXAWjEMjtHwmLdDzawexbL3JApbj7h7xNLOi6ZuEQpp+fVFG UboywREJmteo69e/+KkCzgLGDu0fLiGq8AhFj1gR4IhtyyRURnMvDUZyw2o+NIk72hrAJNihX1T0 FVphjJ4FNK63YwV9uNq8cpEZoFeuAhEN4wHljKtvWltIoH70VN0LrwrYAI4xAaX56kctRsxfWp3d ehZQWyahMvykOmkwan5TyzOKcmYcwmOrgLKOcikODcYP7+ViNaHoWcCy9a+//vaaSZ7m1ByOMvfO 7GsAehdNrzHIlpzWwenKeP2InOIez/wpl7TiziHOBfFrniwxKubv2OAJA/4d2cGA06WBoLxQVRv9 02HXi5Gom6SCVCj9j3H88LtzJEQlAlmj2/ZO/kK0xseV9k4/fr40zLOECJu7Y3pKmy0ADhwdOo82 GIFxZUJC0aNIBZAZteXgI0MkzBYVfLV8MIN+MjWA468/xRRwXB418/e29Lb9c+JCSw/kDEOZ5hIM b4EY6PXCUShg/4Vxyj0lQqKAaaeeoosRYAqYl9OCB9PozgAqodGzgNjSUwPhO2/bglwyJWoIjyn9 skWOFDAHdWiw5fPSs4JYtl6T4OpIRyLPWnAmiXwJwSwUw/0Te9migXM9OupA1Kw5eHTgkJOeK6Lq iPqV81jhIVK+aWCp0bSQsWcmOG1gH5h3sQTt+VM0fegu0JO9QcxFvRoMLiaa0bSRxvdDq2BdVsvO TqjMtvFAWNKXTQ+xceNWeH5eHanWWHnOczBmxGzc7uwQOgkiRVf0TVORbc9sWApMVbhm25qvWRNx 0pC2PTKjiNmoV4PBxkQzmjaidoemxcartQfWb/SOXXt3Y28+nUQxWo0OvkEGA1tz/SIhyGCFhQN3 yn4jJuP+8BbhYKQ/VySkfBQtAO5tzVxYyC3A76aDiyr6oOA5msPLIuaiTg0GF/b7oOmfoltcvFrC Zg7AIB0udLSfS9cbIrW+yBq44Nq51p2JiSmgqOCwpp4qYCY4HqgNwpEhYySDZ6OmiHmXLRNhISBV 4CyYQkT1ECKX5A62AqhBLn9GZrs8ajZCi7GtN52FzkbyM+AFR7hYcFJCbIz9Br23AapPfsuDImTq x0SYyVWeKyLfmF598NoP8xAphotqH00bqT0zMVLY4eUIW9EHFeuzuUF4bDk12MFxK1qROEUc3Q8J wW+h8gYu9C7VcDGYltOuTPOoxTisTUbw4+oNp+KC9zD4uSLmgj5OV7eXFKVbpUbTRmrPXIyUJoFZ Kh9NmtA+2wsi7rdXI8UAU3ujaSPL3tvM+GTmIjboYvTiAWfGYBQ0PltIZ+UOwakuKdhgEPVzJD9I MoOTIDqCpgZhKZYCKjWaNlJ7RLUxFS53p8ZoooP+Zq+I2RivRorusvZG00aWvddseJKtA5Lc8TMb lCyeWhRzdvk8JL8C0yu9dJQ4Oe/PMKspV54rgneadHmPcEnpngFSB01Ftj2zYSkwtH6oltmaKGKP 9tBUxGyMVyO1oxlNGyEGaVot5dUwonOS2gBXB82BMg8jG1Np406D91AxuGrk5RU8HkYeegznipgM ZqLaIhyMH53ZqpVX0QZqzVxYCFVaIuGEGlyMIirZozlUFzEX49RI1e+Dph3Jsvc2M9h61KEm9cj8 KB5GBiK1NCKSGNyc0zo0UdKTqh9n1d1PBTATnAThdPGS8Q0DnbtSC7SaIsuWiWBNw4q0Bj0zueih yni2ZSDdxfhTGZ3tfFazkcS1EuLbWOqaUwci2azDMSUUP547i8FIKR0NhQVet6r1Zw02koIGVoj8 XJB0Fvz82htcUlShlRpNG6k9MzFSNAsdFNFsK4poTd/MFXFGjFMjtYOxIg0jjsQp4uj+xoW7Cn5Q 2mm44JUKHkY2ppT68jDyxc3JcMF/wI46pjzHCC7EOcFrzLVDCBeR4veq1GjaiJpCntuY1hRVyGie F01ad3y2FyR5MV6N1PZ8NG1k2XvdPJwZOiRJdUQETLP61o6NMRlj7FRmsK6g9W0VQ5KCDbax/FyQ ZAYXoJg6bClqx0qNpo3UnjNjpJiQ6bVrsDGamLN9s1fEqTFejdT2fDRtZNl7zUZzQ5cnwsadk5dq JzpMCUZ20RoY5ZRvXAzSLUDY0FlIsXHtc0Xwjt0nHgsbkeIX+qapiNm3PbNhKTA61k9G5YMm5nRH e/hUhOe2VyNVzw+aNkLjk6bVUl4NIzo6yeEoewcHXsffmJZrdfCJXUR//8SzAaS0iqHD/XpOS6Q8 ZzbIH51LG0zldKVG00a0ooomszFSWnTVzFgzv2jSuizaD/aKmI3xKtgvvQ+59kbTRmJvs/GtD21u cHRE5x34XXjHgHdC2c0rhld6mz5scMVUr8/TcIOUvON4h58rYjY4d62LLZVyTf1NE+X6k730N4Ox wuMjPJr6WRNf1tEOsdivvSJ6bnk1GGyM56NpI41v5cZrNq68rJ1FKDKQ91J7RKFqKEbHzG1B2eVF oh8sRDHu8BI5jyhflSliNjgTol3CJcXg/01TEWe87fHcltIyl2c9VUSYnFU4mhvEZNSpwSBjHK+i DTS6H7hg+Z5uS1zwKrCOKMaIid+RNBUXLDorffjJJESnyptstIg9iKlgSeiLrePBCJIe1FKjaCM1 ZypGiiUoncqhmVQTn1XiUzxHe4OYizg1EFSM36NoIzFHq3zbg3IeQ28OhQtagqclOFCM3l+vKIAL 3oGpeppOWUJwwTsx/VgRc8GREKS3FDXkN0VFTL3NmYtK8fvSFIhcM0BrIkxuMR/tDWIu4tRAcFG/ l6KFTHQ/cMFpjK/kxYn7jhlNjBEUNdKHzp3pAFOC5L255ImmbDzlsI2YCs6D8IkLUREp3zaw1OjZ CP2/NJkKptWSIgLWszyWVBNRsoV0tDeIqbBPg8BEvV56FgITsvY+K5hcfT4oaqgztcnn6dnGGOf0 cSiYoCtTZ8GvKiGo4EyzHytiKlSghbBIMWuq1CgqgqaaMxWVIgT2+gic0jSaCJMFnqO9QUxFnBqI N+03lqVoIdvcmxaio5KfvKdcznESyJOzYvBDf6OVaR3E0rKWeg/JQAUvbda160FMBeR8JisiRXZU qno2UGNmokKExNFpD6lWRJrQMDHnfY0oKpJBJD4NpG9AVGgULWRi+8evt0yw0q/b0zDJN1LUPPht AqmBaeZB6p288UN74Ttz3oCohAE5pska7y2Cm0Aqpr9pWUhNmQYL/eLqC/nHAc6tR19t09aYpoVx MABPjTsjshyull87hGXqLQUXzhqGAhLCu0I6RwnmBNHKm4zyMqbEpyU8OXbLY0RsxCwwidOe4WCq oivFPiCKXFc/mzMPlSIkvUQeGqqIz/qxQnA0N8g/VE7Fp4GIe0Kpng3wX9Lzt87SExGtPNHJkAx0 LVy9STYUI504tk33RaelV0irnbD27kGN21N6rIhqRs3UvHe4pLTMFKlRtJAxx3MbY6zWBsTWRJSc eTvaG4TnxqmBoGJiiSLltREnm6P7GxcuNXVkUp/TEhdMwDyIbkyFHm8LxuwnbzfTmpdKOK0kaJrm 5wYxGdxa0ZbiYIRJulv7aNpI7ZmMkUIrS52itZrOlJjP9oqI/OXVktqej6aNLHuv20gz485nXsSG JvrTTwymLNOyCiM7r38TGzpQrNarCZGfK5LU4CAIuzBbiklytY+mjdSe2RgpOuIzV0pgo5qIk+Q+ 2hvEbNSrwciE8Xw0bWTZe81Gc4OXZmjlGjb4GJDPE+kYpb7io3zh5Rwig5cOaMmbPoiP0XWWlseK 4Jzmad5jXFK6ZBApK9JpjAFqzVwMRh7oqMZBEVXnN3NBkhnj1EjVb9wcTRtZ9l5z0czgaIbWrWl5 TDTTZQxEanF4RlzcyYdwoffWkRfsnvkp8pDX7ZkJToboMjVMWAYPKzNqNhJbJmKEaBq8uiBzr+ih ST3bCsBT25/K8IPUVtSotzDC7zm2XpOghOCzSRzNoG+UMDNOzcQOmGoq3i6MVfbJ6IMVIfvI8KK5 WZ4rYh44HKKb1EtKZ24iNZo2UntmYqQoqsg/1r6qh1Lsm7UiHkLGp2IHv63n8Pey9ZoJpQPSnNbQ MS+lE3kYJgbjZ/+UbwwvHNgWE/y0koKJzz4XRBmhyZne1LSlWKWo1GjaSO2ZiZHi12UlQAXLaNIE 7tleEZ7bXhU7RFNNB2TZe81G80IbBWGDnSaVV2gw5sRiZkS3ScXGkWqxQefjtT7NzfJcEbNxdom0 pXzDQDeiybpoKmL2bc9scNbEPlCI6p2BsDGamLcc7aGpCM9tr4qN52ivpgNCByxNa0h9sZaFtHb2 qRz4Ii1XaFV1HzCyi6U22dVuf9jg3eu0aM3O8hwr4HrObOjjvlTncGaMfOcGhbWPpo3EXnqZkWIK r8sDsDGaWNyW9oO9ImZjvAr26+B5NR2QxrfYeLF6gzQ3SOmfxAZnfXWX9ICRsVp2ZDxkdsGKBWHy qgo7xxVPP1bEZPB+Dv5xpNQJshRmqSraQKylZh8hVnH0cni4GEVsvB7NkRlFzMU4VezgdzUdkGWv 7eR1ZnCbRavWcKHDJqZiIEZ57leLCy7BaLmbdGe3Vb8TN2X0FooCZoKLK9pdHBnysppHzUZii6c2 pE0BHRPXVC96OKxysIXiAKZh/AlEzONy1Oj3NIKHY+ttZ3HnEK5IoCvm5HFYGIwfkItQmtpw6ErL 40SIZGdnea6IeWCnShuLS4qCvNpH00Zqz0yMlHZ6UM70q4o4wvfNXBFGke1UMSIfx6PoACxrr7no MMK7vbRUDRfc5O0wMphSSoc/KBQ4NisuGOwkRRd+7nNBMozoPV3iYqQY4ys1mjZSe+ZipJjocKtE WTGa6Mq+2SvivBivihH7eF5NB2TZe82GhxGWd/xZY3XPVJkqsvjmojGnFi2djowalre0hQ3e0gA/ mhSxJunIjTgzmIbqajVsREoXC5A6aCqCTO2ZDUuBcb6pa5mjiUH/aA9NRczGeFVsPN+aDsiy96az YKbFexYUFQNT3raKhsF04Ygze7DBHTFqLMXJtLFztTwXJMOIaqSwISn1BVySsPbRtJHYa7FaKX6H rmZWkz57bk0UMpEZxGzUq2K/Dp5X0wFpfG+HEdZo2NIOG3zeSLeM0TAYA7V2k8k3ttR1JZCOiC8c wY/manmuCN4xW2N9h42TSmkc0YZ52qE1HRCt9KHJuTFSWuvTu9eWJuK8yqttrwjPba8GO3heTQck 9mD/bUvhHIfWqzVc8UnJtJTBWMGkfWCW9X2dsSBMPt/NSKK5Wh4rYjLYRNYO45LyZQMrjyK44DLC wVoyYzBaP+fgxEUVEeXHs7lBzIWd+s1ypKWUGeN3NR2QRvcDF9o4Dhe85cZ5MQjt7EbjgAoOr3lx mmx3VrCq4Gf8t2lg85e6TDRIgnpCEpBXHYOYc2aiPFMZx8LbCzxUVwsfoteCuyZgtuO/4W17Egli ra/wYR0bUQ90OMT5rbjy+ME2CdsOip0B6cTZD0c/GGdEtNGBTc4ra21bI66k8Is7b34uiH5RTdK8 nbikKKArVU1EPEjtmYViREBPoDq2igiRwflobhATUacGI/A6HkVq3wbczzq6H/LgxpJUuODLLFrV Q8NgyqT71eUs67fhgk+besWTssvPMUYJMRdMYHWtWuOypSifK1VNcDEIedvnFkZQrGATOmV7NJ0o 75/tFUlWsHUir0Zqez6aNsJvXXuve4hmxhcH98MGm3odSweDWXpAZQZftfeyuLbLkGI84W1gfi5I MoNqUYf+aQuRom6sVDXBxiD0wdLkzChGDP6wF2xUE3FyKPJobxCnRr0ajNjr+WjaCGzU3ms2Wllw nFhbX6QW547KhjGyS3UtC3WwweKjV8hV2OpwNn1UniNyI84N3vqka9WDETvTEn1/azQNsu2ZDUuB MWLq29qwUU3EyYuftr1fygMjZqNeDUbsE001HZDGt1rKq2kIC/m8Cz9sXL9YqKClLIwYuKT1JTb4 ziKbuYqTC1hUFpql+bkiaSkcIdFxtsGIne+kSfto2gjsy55yaknBBuQTuxZ9rOnEvEdebXuDmI16 VezX9nw0HZDYw883uUHhzrfjwwYr956gLozs0lvKIYM3MunmG0FRi9PV09nlsUGcGpwgERcRUiXN uQrprp4NQIWMpZlUiB+YGTLDhlZDpIcYdY13GSMvipiJujQYcU8kVXRAlrky8TovGKu0Yg2Td73z 13lRjAh0Dc8VOef2QwU7+hwD0bTIzxG3EAWlzSvtKg5E5NxXkPJRtJExxwxrScEFxbprligiSg7p e+Gk1gaBC3I2Pg1G5PbbLcmKBtnR/ZAVnNzQpRic44Iyu2XKtMGYgzBNkF2OInuFnASRFFzwBUE9 R5hGzAVHTA4QUXJfgRuI6o6saJBtzk3EUmCkEcU7nEbPiSnNs7EiGkqWSyN1cDt6DkBj+xsTHknY 6OSlNWKCtMxrVtEwmHYy6Mswy/K+V7q1J6JxVfM0PwcTQtTqNVNruRIhCugqryJyf5Ca47GNaSOD 2ydLEUHSvo7WBjEV9Wmwg99RpH7PkThPHN3fuPA4olHgrEVrNVVuD6aFDMbYyTkHVbTMQbxMTuUl Kbh45Dm4MGIuPj8YAZUoEVKXH+VRpLiN2DmbMxcjRTHm9W4NAlJ0YobzbG0Qc2GfUFUpIh+/q+iA NLofuODgBifnPFbplLF7C2P4Qg/CC5iUF9z+IkE88kqKrpQb/1ru1lgsxFzwunVWeAcick48WqiK Dgis9LEtpVVqHb/RdE+X1IiSE3yu98ZaEKwdfCpG5ON3FR2QmCN7344hfEtA16lpqj4fYC4GI+t5 l7Dscr1UC9mEyWqd669znytiLm4+hDVCRM6FBSuvogMSc2paW4oFL15XKC6qiEW5b9aKmAv7ZMYk pV5u/K6iA9LofuCCb5zpVjVc8Jt4RsZxy2KMC/r8AlzwbXv/UtRY+qQQecF7I/xcEXNx1wcnRZiE NLfg8oWFquiAxJzSaUuxNs4ETFxUEaczjtb4LYqYi/FppLbfVQQXE8ky92ZEZYPlzvfECIAVBnpH 0mJDZC83JkQFnx/QXUIaDef+xQQFFO9rHsBE6K2Y7BZUhhBZuLXmqjkgsWUeRojNVa29w8PoYWJz sIXiAKZh/KnMdjlqVhB4OLbeNA60MAunFkCYIkoHT2BhMMZCvuCpTpNXTOt0BRGyAAsvTHP6XBHz wOsFtbc4UoTNB3KtvZoOSO2ZiZHipAoHGSBiFNH2v5kr8o+jUyO1HY+iQyTL2msuPJjqQKVW/sXF mTckhovBSCkdHSHJuPWiJXKGGX2YVlywsezngqjFa6rm3cWRInIOLViqmg5I7ZmLkdIStV5JsTUx HH6zV4S82F6N1I6mmmBjYln2XrPh4ZT1Jb59LTbY3+QFu2HDGKnFsMRn4GSX41laJCdOfWFBbPC3 nytiNrRtqMwoxk/M7QtOmC1Ng2x7PFcpME5m62U4sDGaiPhoD01FzMZ4NVI7mmqCjW/xrY7zVfmt U5WcLrN3fG7PA+rGaGiqpmCDU1ce7qmKkNKSxlefC6IeUFM17y7ChqQUJ7cvrL2aDgjHUKTJbIwU wwBr8EdNDF/SdLBXxGyMV8F+EftEU00HJPbw821ucDpCS9b8Llpq1MIeWVZMbY1gYIOTF1rGJlu0 tinvdA9AzxUxGxwj0c3vYuoHdepfUtV0QGJPLG4p7LBKKjZGEwWvvDrYC8JzB6+KHTyvpgPS+H5g g3mCLmLDBl8I8oKvDlwGU1tjTNU6OssITnj2SF1xnCqiPTMQk8HkQXezIcMYYereXMiQogMQa8mM EdJOl07+aNoXRXD+zVwRc2GnMBdMmTF+V9MBaXQ/cMFZDs0t4YKVRt2OQMNg7EnrdcOQwXV83SYk zBNSGmd5mZOfK2Iy+F6YDrZVSplBXWKpajogsZfMGClGb70uHzZGE321vDrYC5LMGK+KHTyvpgPS +H5g48HdY7Gh0qSJMRDjNS9+ERlwoheOku5s1YqLjz4VwFTwejztMlaGuLmqYM1Vc0BsK2kxQlrJ zvx89Kiie7IVwElhf7BVGdqAbOFg1cDCBJG43pKgbWe96UskcJrSs/WN8V+8i1kscPVDRyeIkI8R uvTiZpKfK2IeOMmn/cVKKSW4qGCpajogtcdzW0qJpoSIHm5HsxJwsPZrIWYiPi3ssf22nsPfy9br btMFBgd4eDdbmGAG5QJjYzQ9dWOQSzGpNW/NBiVFQrBQ5eeC8NvStek95mKiUsTILrmlqumAaMUF TWZipBiouQKuTjKaFCUOHOwtBHvj1cKIfaIZTRtZ9l6z4QJD5yt1M0ppxUEPFxjFSDWal94yil06 NK15EycLcF7NuPW5ImaDV1xrb3GkiJ0riHr3zWgaZNszG5YCo6A9ZzcqmhQndG57zgwjZiNebakd zWjaSONbreRVgUEz5l5s2OCt3C4wNiYWtAquC2pYdpycE/J6Bqv+fk72eM5s0LbpWsSGMfKddXhL VdMB0cwXTW5dI8X25CebUuRGNHEvAz612KpLc9K0ELMRrwb7RW5MNKNpI7GHn29zgxMmWrkmN5if 3zSM6ORlMPoKd5TkyJU9C8fJZ1LIFk3XIqOdQhCzoVNaYUOYS+pnTbDBFYaDPbWUwfQbs27g0jaa FCffFmI+PvYGkb3xamHL861px9L4fmCD0xy6kK1hmrejqRQn94tRtOjOOnZ5qYFeN6oyRFJ496hM EbPBiRJd5R4p4uSuhbVX0yBoqj2zESnFwOEGr49Ek+LkRZoHewsxG/FqYcvzrWnHsuy9zQ3WMvXK Uby76xCW2TBGVPQ8RAcZbA3oLiFhcpfY1Sin2XSTexCTwUqx3jg6GKFze8FSUXQAas1cREgR8KJW F7ZW5FZxwpx3PqxIcQtJZsSphcFF/N6aBpnoaJdvueAsBwO7pGlx3hVgPCjG78/30UUG37jXbULC 5EMIrkY/KlPEZHC6gu5/SRE7wtZeTYOYe9szG5FSVHwhDKphyprostRbHOwtxJkRrxYGG/F8axpk xfeWDR22pFWKTBLaaxmBiInShSvl4oIzXTovoWpEQlQeHPo8AKaCzTVdvRkhAmcZSELVM8CyZSIs I++5iEbUmvxpz1IRkieMtbU1QJIiDi3sUZ+3ngU0Lkj46yv1ePqstxLJMb80ktZRSCTyA7vE4GCK l8qZH+uFkDjGQRFVXwHkGG2RjW/lQzFi5s6ChKpngGXLJEiG34C1EBbBthre7at7BvwEtbUA50Ic Wtj4PHp+baBx/ZAJH3pvlyjk03ceTHXgygt5GtN4vTbZduJjtlr/U1mrsU1zNIsUwDEdiPLe4ggR M1NvLYVFj8YJAURcWybBMv5JtcC+1Cg+ZnPb1gKwVYfQNEL1eetZwLL1pnNQ2cx7IeWYFhzUTwZy crAuQYpS0PJltZDANRm6MhWyuoFLwAZMAjNI7o0tjJjpF5TYBC89G6gtk1AZFqb1Hh1IsBqeYl6D LWWdbQ2grBuHFrZ9Hj0riGXrbXPgwtiHM4GyxYMFqwqB1L/RAVC2cnRdHNBx6dojft0lMYD80qSM nnFjtA9ebCTNVbP+psOXJR4K5B+Uyb9HyGhRdPDmYcmmFmAK4s/CEK7HLB1qb3IDy9bbPGDr/oyj 9Ma8hjt5UEiDP280goMLNXg44AWs7rJzPQJSDJgDzow4N4oRIBexpTp6FLEAIq4tk1AZDuBwsklU Ro3iOx1tLcAk2CE0jZAPT8vW6FnAsvWWBI6Q2S/ObnRo0E0EN1t2+5UGHHphcqhgOIficVPvSVt/ mwEdck/iS0azJ06lHbTM39oeRIfjjwS7V3qbNuGPDixuO/jWvx39+DIyOgEqO6Nl/l52XsfueSeL inqfvJi7s56fBChE7cYnWBU9dzzZCFe0bFR6pDznqQCadKp0V3+2hUiECI2eBdSWCagMvZdX86vG EbOifbRFHyjAFIxDxQi9Po+eBcTW+1JJByU5ImAWuHTh6ebGWJ3ny9GigX0OuiZFyESLrltTrjxX xDxwKoR/3FK+SWHto6kImtjOlrCZ0Be75AOFDdccxOhookj7Zq+IuRivRmp7Ppo2suy9TglPvhlh rozPeMdz3N5WThTDO1bpNTLiJZ+nZylXjYQTxp6O85l3bsYsxGyw+s0O8cJoFMzPLDWaNlJ7ZmOk mBxqewk2RhOsf7NXxGyMVyO1PR9N/xTf23JBRydZdCIq+hS+VR42BmOVnoarPhKmWKdVnB+SIjfw 288VMRsPvcLpIMWoUqnRtJHaMxsjpTpATZKiu5qY8H6zF0Rtcns1Ujua0bSRZe91bri7UP+qzU+x wRKCx8yN0daYV/FPDAWsS4sMXlXLSKYpVx4LIueoHPVKgYMUpWOlqmgDtWYuRkjleja+RxE1wzdz RXhuOzVS2+/RtJFl7zUXbSefvD21LTbXbNWehaVl66W12GW/luVUtRwWBujkNeXKc0VMBt8HYkVp S1FifdNUBBl+V2niuY1pLqvteOam1aRZ6sEePhUxG+PVSNVztI+mjSx7ZePVIpXLDy5XyzvW9F1K bIwtC7a0XWQy1Q4bV6S0q8CRKj/H7r+eMxu8pcBDWjHi5FaIpUbTRmIvY9BIaQVeiadJVTSxZyHt B3tFzAYFl70K9ovfkeHX9kbTRhrfD70Gd1CYX4sNDoHoG53oHIx+joVpscHFW84J6ldnXddlxaPP FTEbXKjisNpIaWmJgdDaR9NGYi+5MVKsxPP6FLExmhhb5BVsjL0gem57Vezg+WjasTS+H9jgHA0r 6cotn1s2G4Nhj7eNyC6L9Rx20KjIPMhlF83ZzxUxG5wk8SR2pHQRIVKjqQiaao/nthQs6CgBbIwm 2Ptmr4jZGK9Gis6p9kbTRpa9t/0G5/VPH/buToWVEcUYDFHAnDlCwP/eOOut3NCYjRSZdNZzGzEb ugeoOfVIaX4aqdG0EVUIaDIbI0X7lzhsjCb6lG/2ipiN8WqkmHnV3mgqsuP7ITc4P8KCtH4reijP xzkRUoyai15Nkxpevslk0jWJpGCDezt+jjYgxGxwWoS9wy1FQf1N00Zqz2yMFFWG6nrYGE06av9s L4j6m+3VSG3PR9NGlr3XueHxlSqAJhA26JJce22Mgp/3Eckui86s6ypOFgldm6rp6LkgGWD1Koyw ESmK60qNpo2wiSNNZmOk+FfGIM15RxNTmG/2ipiN8WqktuejaSPL3ms2PMIygeEL3IqK34P6Si2l GNlFBnA7QWzwuXrPYWj/kuJfrn2uCN5pk0qbiGIoUvqF9P2PranItmc2LAVG++fFWkdNzOGP9tBU xGyMVyO1oxlNG2l8q6W8GmF1apJb7Pbu80TXYDYGYxzVTJVy7nxhgVFhniUEGby4nYXshZgMXkxN HbkwgmOIsPIq2gC9nBSZCws5zgf3TcRFFbHKw2LBNvebqV8QnltODcavaL8Pmv4pusXFtwtUzYxP rgWFizMLMzrmzXyiGFsaeps3dpmRM69OTZI1LY63+jmN4iAmg+VbftiRUr5z0cJSo2kjqhDy3MaY wbBNIjaqiTg/kdK8oPaKyN7yarCD56NpxxJ7tOe37YQ7S15vYsTWsr4zYzC8olqQXRacKbkUJ3UA /MBBnytiNpi10OFtKTyo1Ggqgqba47ktpcVPcU3lFE3EyZduj/YGMRv1ajDYGM9H00aWvbds6O24 SnhGPi57hA1jREVfptdNwAZVOeWx4qQpwI9mSzy3EbOhDxVr32CkGOUqNZo2Untmo1JEpW/6QoYV 0SxYnsVcpow2VySpUacGg4w6PooWMNG9e38aj995z7gWpRiv+R51uBiMCsg9GrVtZmIUCAhpuEXY jwVJM7nS5m+oKgaV7J1aahRtJOYyDFWKCNBAGjAbjCai5N9cCNfeIE6MOFVI7bt+L0ULmeh+aCXs x3EBRr00c45P1eTsRBTDdTa0lBd8yF5CeKk79OpHK1LEacHJEqRHSl0GS0NWPoo2EnPicGGEwMSE pIPLaCJMvUnAmxnRVCR5EacG2n4vRTuSMffmvXqamJ/1Qm9xwWVj1xkbYxxk0BMX7IupQTB4IiTf uC/mx4IkL/QGhnAhKZeSlRpFFJdP5pIXxQiKz8q7jVQTx1/Zf/NKQu0N4ryIU4V+7ViWoh1Jo3ub FxDPGk64oPryicWNUS/ydiNx4YVflyIIKS9YuvNjDN5CnBcs3bFWOFLeLqjUKEL6yVzyohghcGDY RUY1ESalXdb181yR5EWcGgguGstStBDVJnbzfV5wXEPpT17dWb9Rf8H6VTBaGR8fVqM5sQ6rGD8R UVZoprIRecby0pU3DywpZQX3EyxVNQsYWxzc3phWzciqrYgYOYzEL7HMDYK5ujTI+PxL3W30rCiW tXcjKh3TVa9YFw+s56m8WBBE8mvRdpmi6zyA+lZ+LDl2y0MFTIOKTtFQGQYNyTCgjJYiyxRPbYge TgsEGn2ihxtn3qZdtgKoQY07I7IchpVo2Uii+luz8FREByqZcHrE4lSIi4qFqdvQWzYwChWapWpe peRg/pHHisgzKnW+rhEWIkX3UalRtBC1YmniuY3RCHiJowbNaiJIvox7tDeImYhTA/HbTSxRpKQ2 4gSpuTclBUcqOVbEEgzOUdu4pFiYGhbLnBqg2AHRNgkJLiG44D6sHhvEXLAHzkLdwoiSOxWWiiLF bcTO2Zy5qJRy3FtD2pezJr304tneIOZCTqFqhJbfS9FGGt3f8sKlN/ULU2SW9NW0mJ2nizDmHGai TluggOWFI1ry8bkdz0z4CKAXv4uYjCu3KbSlMlIUjZFSM7WmjdSeyRgpll50ZInEqCbi5ID40d4g JqNeDUYejOej6Z/i+4ENDuhqFRs2ufbpxRwSuJimAVwwhw0+YO9lcwp8STGccGfQq99FzMbtI7uM I0XV+E3TRmrPbIwURS1f9BIb1UScfPfraK9IOox6taS256NpI8ve63bSPuPBgB02KOJbVgymaYA6 M7oj1rqTGx+Sgg0+m+Dlb5X8MGY2+PyAdx2DuW6s9tFERfZsz2wMxkbPjY1d2Kgmv8/5yd4gPLe8 GozcGM9H00Y8JTm8UO7VJFVe4oBz46JzcQymxdrv6CoEhRtHB8OGv3YMG7xUTc+pZ9FzZoNjI5zB WBidBLtg2rd2hylNRbY9s2EpMNo/R4aVG9VEnP566thDUxGzUa8GG88Pmv4pvtVSXi1f6Mwk7wjj eUoptoPCxmD0x+41eIMavbqGXTYAiYk81FP4VsRccHqEIxMLI3JmadYdPfvv2jITlkGTFoCyoTt6 tJR0sIbmImZifBqpRnLQ9E+xLSZe5wWHOLjEJybYI/LihQ5JBKNgI4E1QnEMS4vfRM5Emgm1pmmR KYJ3mqnp/sxIeUPlmyZWdJ7tmY3BmN/o7dkHTSxiY8+bIH0uiPJwe1WMvBjPR9NGEh+t+XWf4fGE WHhJTNhgocJ1xsbUKlnZwO6ZyQEba2Q7Wxbwg+4+V8RscJ6Ew2tbCi8rNZqKoEm9DprMxkiprxDX mvVFk/qYZ3tFzMZ4NVLb89G0kWXvLRuc7uCFN8pSFipcaegQERhRMaqyBik2WPDX4RrY4AQ8/Giq FpkiZkOLfGyoLCnKyG+aNlJ7PLcxLQu5sLUizGkFiLXEg7kgSY1xqhipMY5X0QbG2rvJiA5WnrSA DResBbjS2BhlkN7DCxecztfSN1FCDqsXtPk+F0TOabbmLb5KKUruKFj7aNpI7LV0rRQzGr6ze9TE ZEVeHewVcWaMV8F+QcZEM5o20vh+aCf6rEPY4J2PXr6gRirG78/FWLHB8Kklc+LkSikjv6ZGea6I 2eBYojYeK6UxUp96kPbRtBG4kyZnxkjRiqi/xcZoYl4rrw72giQ1xqtisDGej6aNxN5PvcYXn1IM Gx/q+j22DkYtp/dtwwYf8dTat6ozSeEdp5z9XBGzwYvUNIlZUqRrpUZTEbdL2zMbI8V4w8dkxMZo YkXsm70iPLe9GimdE41Xo2kj2sbD8x9yQ7eB1Gtoq548MBvG/IuyGE0PT2XLJR7lBvWopGBDd3/E YhGzobs/6jVGitq6UozG1rSR2jMbI6UDBKrANRWMJpZIv9krYjbGq5HSCdJ4NZqK7PjescEnz2mq WqVS7cJrQmDjgFF567vcsMExJC2aEyfnusgWzw219D2I2WBriSP6C6Ol3LjbnZYSTRupPbMxUjpF ryOumgxGE0sB3+wFUX+zvarUwfNqOiDL3usRRTW5pM96L21qGhr9E8Zv5Pc6MIayjyspVUyp0nmd LUtRg8g76mAO/oqNkWKUq9Ro2ojqJTSZjZGi/fNiB/XRo4mFg2/2ipiN8arYIZpqOiDL3ms2VG1I mje3aOSk3+EgQdkQ5rbNNyTo46nnWB8NG6zpUA/QFng2bBgxGww82niEjUjRGiR10FRk2zMblgJj BU1vS4GN0cTSwdEemorw3Paq2CGaajogjW+1lBc1uaT1/TmRcaUDDD9BKCzPWsOgmKNGDxV6iTNM UJGxGKyoBZgICimthBUi17ltQZPSyB01G4kpntqQdtJ0is4FkvWwhn2whZ4AJmH8CbRjiBb/xISA f2PpfTpwvEOea7fKxacUFvNUiYKCEo4EUYeiiaCKQU3aSM2NmAXOoWi3cUlRRFZqNG2k9szDSLFc zjiidBhNLAR8sxdEjXF7VamD59V0QJa912xMV8FClJo3K2/aM0xCFKPT1+FF7FIUauGbH5f6hxxh KOVLF/m5haSr0Kaq2BgpBrJKjaaN1J7ZGCmaPN9UhIxRRNP9Zq6IyRiniin0Om5FzgsBTgxHt5rG i0maHmf9SVMJbW4xGw8XxSjh3I1R0PJmznDhg2Z04CfewhwujOCcJm7ea4SLSLGRUqnRtJHaMxcj pe00ndJZmu4UCPJq2yuSzKhXI3WIppoOyLL3PjO+vrSC62/7Uj+FjWKabqsbo6DlgwgZeDmh6HKU PS2tf7sw8Zc0iIFtE65lD6aegDsMlhpNG1FRgCazMVL8NNw+UmpUE3Hy2YSjvUGcGvVqMMVez6vp gCx7r9mYQeShalts8OKFKTCEObtuWreCjbvfTKY4WZ91OcqpcLNYxLnBDUAO520pSmpJHTQV2fbM hqXAyHNvVGoKaE3EyYsxt71fFF9BzEa9GkyxN5pqOiCNb7WUl4OIjmJqYZehipuvOqXD6DmY5oJ0 6bDBmUcO9SlOXoPKhpYmbrwFZCEZSXirEF3gSLmIrNRooj8pEnvK+I1psZB5DLlRTcTJMpzL39ob xGzUq2K/Dp5X0wFpfIuNl/0GYzMbpmbjk0GMXWZpKMbODnfg3BfxUS5xxmTbQ50mbqyYbwTvNHXT lYAKaadNHy9OM7SiA8LhAykyGZUiKG7cuLKNojtTgKM1MqPIPw4+Dbb9HkUb4beuubaTv2TGRR8D UGZwhiTFlg5iGmNTWK/wULadL84MFqqQUmboX9T3BlFQmriJ1SIEzmfKLFM9B6TW2E9cUkTAUhtZ UD3EyCkM8nLZGiRMxKPBFLe8dkuynkF2bD9lBWOmfkoam9Y+nWiFqN3u/JuI4OSM6GIlV8uRmrDp DZQDmAam92ofFaEgkQidRbWQ+EaWKR5aECxQlrmIixoCZA3gYCkATWp7UxlFHIdHy0Ya1d8oSHHB MKz7Qeqv2OZoLxGIboZ9Dtvk/eypzvlmBIWW5kO6e6iAhaixa55Gpg+k+LiP0bHaig5IrZmFSuE/ L3wkz0bRnQLm2dog4SE+Daaw7Xb14LYBfoNl7N3gwVnIzBr4uTUxSwdRjC6G1xSQHVTmCpEOQzLw QNp4zbuIeWBHIBVKhKifKxQ1YsaIXbMx81Apfli2ZcVn9BDg9dnYIKFBDhVx+sblahEJz3H9LRsy hFKh0ic5HXR9o91kMfhkgHY+sOCtdGe2h5SGEQrxBm3EPDBBSacoyCVjhaoIHriX4cdYG5Ai81CM CHiPkqv6KCJI0mlbo9kXCRHxaTBFHr9H0UbIiJp7301yjIMfkoT6Yie5GRFM8z/14coIjhQmJ3jN Bittmppx3tnsCEk3eWc5fkNEzjEmC1XRRsYc/eSSIgTOQbmmjyKixLTL/lobJFzEp8EUufz2wGJF g+zo3ueFZnEqCnCO+QNckGqDaRjRK6zEBdefEjkrkvaOuQE1qDLFiPOCS7jqcQoRJWcWePWpxyMp GmSbc15YCowFU315kwYSRXeKimdrRdpfxqeR2n6PogPS6P7GRXpMukXdjhYXrDSlzBxMfQ1fHjMX D33IQUH5O9/khb6EKS6CpM/UNyXVs44UK5uVqibIKDL2TMZgrICwIKROs5qIkxfrQOKyN0gyI14N ptgbTTUdkMb3NzbSY1CZe8ccNtkUTZkZzN0bF4xUWDG/0cdeFCcnsVyG8+ZR9zRFiIpJGlvQYSNS FM+S2ppgw4jTwPbMxmAaOlRmLk3EyS3RbY8WUERsjFeDKfZGU00HpPEtNl6WVuS0jjkoSem0snYz GLmhJWOZ5XspqczZJ2ec1wyNsdMtwIjJoMF6o5Eix1KkuctEDWhexV7AWDMXFnLfeGZNRlxU0Z2y 62gOL4uEizg1mCJXLFvTIDu6xcXLkpvSR29AExcsnGg/VRqK0adyft1kUEinMOdakack3FinCSlX jJgMbsN4fTCY9zMipe5SmmgBXArXc2OP5zZG3nCbRGxUE3HesKf+uvaK0F/rJ4pXgx08r6YDkvho z+/qC+rhk2YNeMfamweTBTFPRp3NsuyljRFGFS05aXqmV9EMYCrYO2nNahldV7AMTFgNUQ9iW+lo BmJMoecSEdVDiJrbbVsFQoP8wenIKGZF4VZkNYOsuH4igevsyLJaqbffq8aihOBsszpT/l/FJv2B Lt0pZH1KC4TRWk8NYBLYjPccrELEzNUFjcNV5DKySK05H0aKEoy9WtFQTYTICz0P5gb4h6uq+DRY 3cZa9WxgGXudDDOEcI6DAgoi2ApOscmOZDCKSq7CmYkbRYYDZ7NC+aE5mecZ/AeIflnNlHSJU/RY Ciq4ccH0QalmTQek9kzFSNEJoCDzO2u6s1nwbG+Q5ES8Gkyh1/NqOiDL3ms2ZgihvpW/FDV8w65p IYy8IL4ri+HKC/ZEEKK9S0hkcCgpv7gRk6F7AWpllSJ0TjzpCngVKS8GqTmTMRi9wl1LekzyqomF kW/2gkD+dmoJ6ZConaoicfEc3d9aSDKDJVe9mFaJxUmHFJ0Lo83q4JHMcuNM+cPIISmRwUir54Ik MxDWFtNIETrXFnQFHDLySrCN1J7JGClWuE66FOGhz5rYzPlmr0jIqFdLanleTcwSvsX3NzaSGTpy qTkEbHA+N+ubwZwZeiOC2WBakERg/qxcUa/g5+hVhDg1WHTRCgdsGIMNhlPdvIENaxpk2zMblgKj g/CNWFpVNbHqcbSHpiJho14tqRVNNYmN5/gWGy+LC5ba9G2XlDCsmqj7XJh+SVZOlG/0Y8oE/Wp8 QUps8PokPRdEbGjC5o3FwYidOxea1NEKremAqGRCk9kYKW+vUmptTWwLI2X2a6+I2FheBcOr5Xk1 HZDYw893vQaKfDiR30VH2D1VXRjDAWdAZJaLEUog+nmE7BwvuVI/UsRkUHVqX7GYEp6NRi2GQ4YU HYBYS0aNEIcyOFYvLkYRP+zRnDPDSLioU0tq+V1N4uIpus3FXzKDTk4DKyv0eWGQNBRTmehSi4KO W8kOk/c6PZwZX3mOIlHPmQwmetpZhAxjxM6r8bUWrj5Jmg5I7KW3GSm2OHlpsNgYTYzx0q5MrL0i YaNeBXNm1PNqOsTS+H7KDD7voQjY12OQTDMpxGRT17uUGZ+86g0hEr9jqz+EOYCp+HRBNRBxc+qN M3hiwmoGGVtJCwsBsRio29EQMXow+mzLgEeS5c/I6Nyoo6gasXCMa6fEt+p7BhI2fTQSstd9ZR4R FoppjkhJJRbYe0tFxgpf6nHOouk5ugEQ/bTaTuMozsKUAEzYCU1EWNMBqT2e21JkM6cLth6a5Tdr RZIQ9WlJsf8Un6QHr/M3P8Oy9b6j0Fl0CbObleqCLtkQBQtvnjYP7OhqIVkL0V7zZJJmkQKmgcMj /HxLiADp8bRPTaZZ0QGJtbSMkaIU0DoqRIwmippncwXCg3xCeTA3jLhdPSLiKbSfmgV743KWTobz 6G0XxXQOhmMlygjaan7/k06tq9hg71zPaTYFYio4OeIOJ5h6TO5iuCSppgMSe2kbI0XNqJkFVIwm Firl1cFeEJ47eFVModfzajogje89GzpO6dki0tpAV/tYGDGcWUeQXe5vOy04OpGa3NeuTYYQk6F7 k2lEliJ0ukkVJNWzgRhLVlRG1bO265nrVQ0xfjNWJEzoRrmSQkLQtZyuogPS0H4igmUC1Qswyjtx nBYUcMG048sWsMhl9zQNnrULUaN5mp/Tfq/3gtW/cwYrFYowV45PimCCWxjbWpKiEJ2RrrAd1Ghx 9GCMH6KImFguLanl9tK0kASHufYUL4dR+OLb16birldYJyeK5ViE7VIs6Nel70PKGcuiX8pUIc5Y ZmOfrPaMlHpDlickVEUb0Pa6nhJdlaHr4zyPqagaFj4kJN5rrEioqEtLKqFA2NL0PbhFxbfhI5U3 3Zc+C6Ws4DXL1JrOtGJ0uzozp5+AbWT9mlRskpJ3bDYnA4zIO2/ZKmcjpYGAvV8JVdEGas2tqjJU jieqloMaFkeOxgiySKioS0tqhbI0LWSZe58VvLNav6QO6xBtsqKYjhC5tOEfdWlGQWoG56xgWpoM 8JwuVFw0vRwptXrmqxKqog3UGlmxZJj26ZshoqJqWCSWS+K9xoqEirq0pHRCVJdTaEbVFGQH91NW MH/UL8kPzy2VZkUxrRowGCorWJHQz6v1gDVD83NFQgVdiwaaYAoTWqS8ijZQa86KypBn+oiyqKga ZjlHY84KI6GiLi0pZh2xtjQtZJl7nxXsDuuXpBtiuaVZUUxHqcIE6yj6dXXaCiENKPrAOM4VCRPp FixC1JcojpL9dw2ZhYrQBxMXucYvjBUdsjpY2UgoqC9LSmdCHULU4F0QvFu23hVWtF5t0KmP+eJ4 VSkopjUAVgqVDRzY0s+qNQCkxAGH9/PLGwkHHB9JJWWMsPlfCUWRWooAfKs181AZLZGzzSQeqobu 95uxIqFCLjlDKrVCWZoWssy9o0KnJuMbKa0OwoC9536l5huUaplj0v2rzmYypq0OWOF/wgDnQ1I0 oIpweVmKlrwhwCoOCMObL8hsGfp9Dr2agGi5sQC2zPziirr+TOzxYwTG0aVhgBp5HXXmGDo3qcOE HmRYqUzgxRjVeKOeI+eCewLhFHyc0itPHJoQz384gfGp3THCtRThchdBy58kozUdEA1q/JNTYKT4 6flmpttCNBHh9dneIOEhXg2mn6yeL00LWfZes5EhU+O61urpXNkazqJuMDcoXqGrAZ1f06tQ9DYS UiJc81gROceP4RdZQIalCJ2rCC7L6VxRpPwYpOZMxmD0ijo6sTURJiXd0V4Rk1+nttCKZSlayDL3 movJDE5n6Nekd9XrBZIZxehfT5zTc6GqY9X6zVktyXz0lueC2DnNxLR3OFKEzjTKGVVNB6T2TMZI UX/giEfkaCJOmsvR3iAin4LaXg2mzKjnS9NClr3XbExm8Cp7ddB0tDpfFjaEpas9e+ZFRwdXjpM7 UMoV5mJ5jn8SktTgUmEKhkpRdH/TFGTb03MLU43OzEupEU03avKjPXqLImEjXg0mNhrN0rSQxreK iJdVts5NepLAD6vLcWGjmI6NuWSgXPUBCVo/O3dXfj064jwXpGxQZKcgkZTj5F6FsmVpouM/2mtv E4zfmE9IKRM4RmVNxMm3zKRp7A0SNuJVMbyaaLamQfi1Hd9i42WhrTOSnieoXmWVJGwUo9/Rq+WU k7z4XBmkirVrIHycUHlQJLnBB3RSk0jKfSFH5CW1NDH6H+2ZxcGIgTemK3Z+SGsiTr2ADk1UBH6u iJ8brwYTG/F8axrENTqafmKDJXn9nqrTeB9D2ChG7cRta7NBr8ihXmJnu17jPiupea5I2OBgQsaP Srkzk/alKYgrAdvTcyNFDJxHCBvRdGOt8NneIMmNeDWY2IjnW9MgE99PbDz4RpK941xMakytNHkG Qc/T1OAESX5iTrh3VhYRrROBhAwW99FUSFFen/UsgE5unhqMH/jOEpwTw5sExMh3w7+cFx/WU6R5 EZcGExPxeisaBCZk7oe80FlJTxZozDpRorxYGDNiqgmyk40AvoytIOk6Un9zfVIpUCRM8BumaKkU xXOkRtECYq09RoSIgInDTLmkiCgpwo/mBklWyKklBWPj99Y0CFw4uvdZod0R9FEAcKcpRAigu2Fv Th07FVRW8LUMovyli5BA/gwHnBLRUDMSdD46HbZU6O+x0S7CAEu6Z7ba1EFEA6Hy6y4b+dOEjRdL on4vFcc4sPF6AE05oQOTmh0QtN7/k6ADqZdlWRwPKFzZs3VQvLrBLuVdckQJYKc0/fJWoSFkfOdA 2bDUDFJbomtBOkvD+r/Ghui5aUvkaKuAnhp/lsyKImoQCeLyXHH9jYRUEdw/4uWVJsG1YFgIRuZw W0Crp9SsLprp8FUw6sfnY1P5rY3INf9jUqJSfl5SUYRMEPtmcyECK3JBG2RrLqfdRoJkcfxob5BQ gVOoWkIrlqVoIZrnea7yOiGGCw5kkLH8SFyqyMxDpyWd53jONUZCoEqjjnZGcCEkzvHelQRuJGTw vRhVYhyFixS58E3TIGNPzy1Md5bVJjSlY4lOYbI5krlvzBVxSxqnBlMWxHFaShQtoNG9e5mgHmfj V2ZV4jPXTF4E4+dkvUm1JDUaP7UTndf9fbjnfvg5Ul+IndMulLcIiylKTtGo8liaBqk9t6uFsame SUb0ECQbPUJYgI2XRZIW8alS7hsbS/V8j+2nFsIREWo/PDtxlrydYzD85zJ3mOAas3JAtbvXmzX9 8QyjSLKCezEaVoPhmy8bqLJcmgapPTO4MEZIHWZUWkQTUdJ5qo2OvSJNi3g1mH7XRrM0LaTx/cQG H1QjAhoruxnTawYjBr4npX5bl+M9D1HJh5S840PPidxI2GCMVuW0pHztQFJLUxDaZO2ljXA5QT7o fJZ7acpyayJOpj1He4MkM+LVYGKj0SxNC1n23vcYHG+gOMA7jl5n1qHO27MHZjK6miU2WGDVb0xC S0psMPFM5EbCBgcjVZctKWj5pmmQsRc2IgUbvF9LsWuGJ0037RQ92RskbMSrwcRGPN+aguz43ucG Duh7k2KDCU3YWBiVhV5OJjZ442QygXdJhg3eS6kZBisUQsIGL4rUuFtMLYWtQGlfmhZSe2GjUvyQ XG43G9FEnHz092iviPua8WpLLc+XpoUse69zI+WFDkxqpgAbJ8pa96ALYzzSN0TEBvc5lAn0hJJS btClKPIg9k7zpCchltqjfClaSM2FjErRPXIfFvKriCh56eHR2iDJjPg0GK9/nliiCC+DODMc3U+Z 4Zt+Got0YFaDCS1Ol//4NTnBLibUdDsvYSFGtbDWqTTQqnIGkGuaknnHsJjS4vGsZ4DY6kASGe2t OQNGDRF+PNkaQLbiEC5GCPbG56VngMb1EwkcDlHIGoZYYA8LxTTMutynetMim6LmpaOhgYvW4cFI eKDjS2UlDN8oHqN9aVpI7LVZVQozZIAZjSai5Ps20sR6fTQF8XPjVaV4TgdCHc3StJDG9xMbHMhQ 1cdj+Si2dBZjy9LrNQxuGQ1YWNG8UJMxTzG8/JJJqPbINI2AMEnhmm8fpKCMmgXUlk7fLox5kj59 p9yKops2LDoLjrlBkhVyaZDxmYaw9KwoxtpfZiEpOnUwTu+IhIezvr+qPdGFMS3kdfz4RsZplcKd n6RExSPP0R0KSVboHYRLylF6lUKdUDVRpT/Z03MLY6TiU2zmoppYMkW7qa+9IM2KelXs4PnStGJJ fLTH153msMFBEUVAXwvPaSKFGGM5m2syWErlGC99h95lrU4UrxO3gFDBMc5UbZHR3MwyS81CbMsd xYJYblP7UlJUj3a6n2wFSErIH5weGRoAtnBwqQmy4vqJBE5jyHdGHo7NNSWKMfYwqTALfM1eicMQ ICnRwJuJErOR8OBSfaScEtyWkNTSRH3+ZC8pUYyRQm+NERPVRCmHPTQte0GaEvWqmFKini9NC0l8 P6QE45leSZUJEAOjcmJhTHozylPFcrTZOcGSa9i45DnNd0DCxhcXQ81ZpXz3QNqXpiBu0Pd5bkkx FPAJbrNRTZTd3+wVSWbUqyW1PF+aFqIJnv1810Bwkxew2DvWITMvC0YMEEtBjHfUbf7mA7mBSmWL ZmpMS5QHRsIGV2lUJy4pKsdILU0LUVe85nOV0vECd0/Mw6JJpwmf7RUJG/VqSS3Pl6bv8b1vKZyd 5H4wUTFWMxw1N4rR81w4LqQMZHMivzG3glQT0i7zHFW+EHmn+ZP2f8iESrG+F6mlaSG1p+cWRrly 9dxsaWKJ7Ju9IO5rlldLakWzNC1k2XudG6k4adB6naTrfpb9y0YxSpU720ViwzcdFSc7menSeAdX Ihdi72i+LPsdhFjjjdBStBBqZykKGZXSTgvHZEVrFFFyfbNWRI8tn5bUiiWKkAlCti5zr7nIQMKG i+4kIMxu11ARiNKVF1LbKif1lf+6uICQ8sI3GQZIWnA4RCPvEqKGtuqlZ4DYMn0DMdHTyTrRUDX8 y7OtAqFBDmErGOxNGEvPAI3rp+bB6Qz5Tr5rbyNdZzHyj2/imIYL725zhOzmZvrOb5eYjYQHjoek qBKGb/xWkVqaFhJ77XIrRavgdT+moppoOdGkyZftBfFzSjN7VUxc1POlaSGN7yc2OKCRnOCLcZmQ qSwyxlCUvQiKNuod/dyMHsoSzdAsUiRksLyhsXdJsUX0TVEQ56vN6bklxW7jw+WEZnvWxBDzzV6R JEacWkLL76VoIQyzUvQTFxzY0FKn5jaUjMkMYQSlFSGvx9Pv8Q44EM1vkMqcyDJFQgaHSjT0Lil6 tG+aFlJ7IaNSOqrYzepq0urWs70iIaNeLSmt1dirpSnIju8nNji2ofqRQpJ5edkoxpSX5SYPI+yk K38o5iQlNjgDkGrVSNjgK58qr5aU70xIamlaSO2FDe4nSIp/1CeK1GVUE1XpN3tBOozUqyXF5Omb poUse6+7zgwjqgb0pjjKWb4BFTYWxkSEV5jhnSqEzlv46FTYuOQ50gakwwjH8TT0BuM5jf3WvjQt pPbCRqVIHA64ehypJiYr3+wVSW7UqyW1PF+aFrLsvWYjAwnnJLVzLza4IZvt0mBkF8OyT3yQ+bwP zmG6b8uEzY/Rj+mxpAY7p6rBirmmjnIrcp9KL3SwFi6K0XvobVPKjCpirWMW932+gNfwG2lm1Kkl tWJZmhbS6P7WTiYzmHro16TXYuHd7YTlhmDMf/0eD7pEf2KLKDkSolQh7yMSxM6RibRZc1EperJn RQuotXBRIYaak++hLUUsgH0zV0TPLaeW1PJ7aVrIsvc+L3jPu35MzX1pEepBOekA5l6Hb19qEkI5 y+vgHCZzs5DBgdIkgRF5p4mUNhmhrFIU1N80Bdn2woakwKgX9Nn2gyY2GI728KlI2KhXSyqeHzUt pPGtzHh5/Eb7a7pXBBvakAkZhVSqUFiLDN6WJXeZnnhhnEGXp9yOBISKi95PsEQoH6N4aVlITOmh BVErcLufdsQgaDUsmB8soTdASKg3I7NiiBb0BsG/MfU+IXjplFwnb3XjORwUwx++LYRj1G++7qPB QFJqHbxeOSEbCQv0kRp4lxQeRGppWkjthYhKaSEjxUsVaSf62VwRrYUtp5YUW+IxN4oWsKy95mI6 Cn/uDCqopzNJ1VqNEkQFgCtircN41qHVVF9go460SIB2Exwo0Zi7hLTG8qRngNoKDZFhEHCFS3mo 9wDr16d013R4TBVIPtSfJTQuj5r5O5bw7zUFM25QheknZDzX6fykQzE6LR3QF/M3vYmE8Fi/jGPn PFYk2cD2UQqzStEmI7UUBUFTzYWGSrESnDOr2pezJg4ffLNXJEzEqSXEEmPMLUULWebec0H3pjBZ Yr3w44cLYfDD1J8NdXPBhxiT9FSdmjQxXkSGsUVIyODcrqrNYmr93D940rSQ2gsZlWIE0It/1UdU E3sJ3+wVCRn1aknppKjtLU1BdnxvM+PMvqDO+ogNbu96broxxuELL01WZtCLKComR5ISG7RpPVck bPAZddULSwoFkVqaFlJ7YaNSOp3kxNMczJrYd/hmL4jb4/JqpLbno2kjy97r3HBXgTSnOFL3sl06 bBTjhzx50UIbAq7w+dEkJTY+81wQe6cleG0twkalfPdA2pemhWjM7KrFwhgKKOhgf2nSRvSzvSLJ jXo1Ujua0bSRZe81G+41kPYOOt+0YlSgVdBSipFdtHGd19S4xJvGFBXtWVJig9eaJQ+MyDtqSr9H cUtRNyJ11BQETbWX3JAUGHvvlCNHTRyrO9pDUxE9t7waqR3NaNrIslc2XlUVSHMzWr8nq5D6UGnY KMb4xx0Z2/VntMUGE8os+PG9sOSBEPcb2qbSEcdK8Rx9Z6SWpoXEXkegSuk0itvh0kRDlnaxX3tF wka9CvZLv2M9H00baXw/9Rus6qt8Yt7LIKHVHDQUY02E102YDfYHEjvuKls0U/NzWnL2yoByg6VR FR3ByHhq0EgtTQuJvfa+lWKQ/PACxmi6MiTIq22viJ+jmrRXgx08H007lsb3ExvcetbvSYWiryWF jWIUKSoblJO6SU6e8tNKSt5xcz15YCQtBVoOMqpRLLP0LCTWmhmVYpzUa//FqvQQI69PO9oaJHnh u+0j5byo16MHJp5j+4mJO+8pNRNnvinWvCjmybXmHdSyzMKQ0lK736ihOZGfKxImOEyiMTiY8+IS qaWJCvRor3lRDNfpicxFNBE7maZ95bFXpHkRrwZTXtTz0bQRtmyl6Sc29MIldV5c5fZ0HQWBWAFg RBUVfDZFMoyMyNg1v/ipgD3jkBj7qkuIx3QhQU+Nmvm7lkSfIfeK7GmHhWghPu67yzgzQmkZICkR fwYTCfGYLgM9zpGnqH6igPMb+hEpULRrnKYhDNcoidwrUMz6iCUDY86/a3bmp4okHZhPqeRYUpTS kbIe50eBGgsNxRjOcrRdMz0pIsTzs7kiblH1aQnBQyMZRRtZ5l4PpykudJrS0wcmupw4NBMLYxFE 7wRTPrDJnqTn5RhqLEyI8hxDLoid0zyNuaS4qBRVRqSWpoXUXsioFMPixQ1hNBEnc9ujvUGSFfFq sPN5eT6aNrLsvWYjxYXGHVf/lGp6EY3yIpjzgmoBNmCYIiOJz8nTeMfGe7LAiLzTBMqnIpl+RYoC EqmDJlqDEOeh7YWNYtR9F++wj6arVikP9n7xnbYgPLe8Gmw8R/toaixHe++KC52p1P0pWgljU8ko pD0HzRMxy0fvE3oWWJiw8RRRahMiZ5K0eUWJOJB6Ae4lSPOo2UhshYgI8fN+cSWRljR6CPHmxZzY otEECA3xZ2SIOS5XjX9RIXg4tt6nxKc/BYUwW6RetEBlMaoSXtLt3oeP3SsgFqEkpQZyz3NFFJFm azrUuKWooiM1mlRXP9nTc4MpBDZqD5qIkoOeR3tF1CD5beLVklqeL00Lmfj+1nFOd8FUXb8k/Q6v x2xOFOOUBJ+ytF2OP6g+ZkSUlNhgNq3ngri70Gyte3mVoh6P1Ggi8iK1FzaCEYPfxDF6iFIfzjxY GyRcxKfB1Dzit/U4LfQ3abFsvc+L25eOJSLNmclhohg10FUvruYX4FMgiZvzhfGNhc/kgBFFpJka P6LYqRQVdKRGE0wUqb0wEYwI2Ao8KCJI9g2P5gZhMWs5NZioiOOjaAET3U9ZwZuaVTmpImG5Nd1m MdUkWvbGLPeKEzkvdNCsiaEuzzFUgjQrbnrDEVRUiJIxQqMIKorUXKgIRgS8eVZnbzThQ9GVcvbZ 2iDJivg0mKiI31XkvBDivHB0f+NihpCHRglxwYpQZuvUP8H4jxu3h8QFSzRijBUASYmLj8gUSV5w jCRkVIrVzmdNkFGk9kJGMH5gtlyVBpp/SRNxsld/tFek/UW8WlKwEc+XpoVMfH9jI/2FTliaC45W hIsi0My1PWcFHxIMEyfJiAkvgbMaoL/DA8dMsjpWGWpP6R0tsOC/KYLmmSD4zcmLMCAdZoBXTeqk b+wQq/6WHX6V+DIYVyjs/9LSv4m9dughfv3P3//x6oopshwj0a9HvfDZlYqF0flo2JRNvmypfGEE QMp+fea5IGWAUyMaZYtpTOCagrSPpo3EnlvTSBEBRWVWDKOJGBkFNSdTNypNg4SJeFWMjB3Pl6aF THx/y4S0C5aIr5ot0K54iU3m5oMRA/epxAZTXTceHyfT9ECTMk8yiigdNE3yXmIwKKN4jtQoWkjN dcISKULgYxlKtdFEmNzS0dR87BVxWtSpgcRFYxlFG0l0vFTo3dhBp35vrc85Am8GoSGYxmHe22wu 6MP1CzMaImXnbn6uiIPSbMmHFSNFUBTPkRpNC4EM2WtmRAoyON8QMqKJOLmSHfJjbxBlhjZA5FUx Z0ajGU0rltrDz/dscHSD2glp1hQyOycYY7ChT+PZLq/aVdEN+TrKoJ+KtxumiDSS1OBlHhpvIwVn BPOsaSG1ZxZHCjZYzFKr0GRPmq7klrza9oo0NeLVYMqNRjOaNtL4fmLj5o9f4t1Fk3GNpcpzTyXo oLgIYjb43L2mEnRQM2V/RKZI2OAogvrJJaU8f9IEG0KYStSenhspfmNeQ6ZMUKJLE3Henu0NktyI V4Ntz5emFcu29z43eNMfNRnS/CzDhjCi0iCaToP3RSoRPFqp6WiGZpEiIYM3BGsYHynC5ILFQdEC xlq4iBARsPkeLqTIXDDMsjiyzBG3kGZGnBpMXMTvpWkjiY52+Z6Lr4tmD3jHxn4mp8yfjMHFlWrH lRyvgnOUvBVWiCZonnQwAxYSLpiHaNAppp/880nRAmrNfcYIEQGHj9QGNQGTIqJkODuaGyR5wUdu mA8Uc5/RWEYTXDxH9xMXvGJbvwHju76SlFYSDC44kK2JOtFxiNlhcloq3vEpRlUbjJpCQoZu9A/k LoOxUkKjCDKC1JwfWxhppN0ukWFFROnvO4l6WSN7i4SL+DSY8qKxjKKNJDrMNS9e7gFQKHGk2Fzo Ep2oGIgIWHGm/IEK9q5SMnE332lx4SknioAQwdGRlBSWIeybNS81C4EI2dJTI0RS8NmirBVGz5Vp /sEWHgYIDfFnyYzLVbOC8CJHbL1rHFTNpJ/yQe+FCwdG8J3jpKGA1+imA+DIqSngU1P+4fV3GOC4 SHLBIgR3t8goGSB22igsgt9sOaGWNIgSQtOnjZQGttO/E71cMSERsfuEOkr4BY/xrAR4WWiyuKaT LAqek1RZogjmjpOToxR+JBXv80ub55a0mocmZJ5eMMgKEQOakjG6qjVEiog5iMm6KB1BNBXZ9vRc pMAYLrmsftBE2BwL2/bQVERMjFeDKXcbzWjaSONbbPylObCxoUKTykgfwkguFGNhW5+9kN00cGJT bmgo8QyDEVZIqOCoiIaVYvpluWgh1aNmIzWm5yJFmLquoKmHencpIkZq8G1uI2EiCVER1tBjy1ro MhewTL1vD5zV4HeEBQ5SDAvC/EvS+VHywQLvMFET0J6Clk4ggmzR718kRHBYRMs5I0WI3LE4aiqy 7YUISYExLOZV2JpySRNM0Gdue84JI2EiXm0p1kWeNcHGc3w/5QQ9k34B0pQv2zQniuGUPpIjNj79 Kjji5IApm9ywwWsrkgJCmha8MC1VgzH94NynkPbRtJHYcz+xpBgqtbmivIgm4qTnVFaOvUHCRrwq 5kyo56MJNp7jW2z8pb+4+4visOFPhKWFFCNZOeZgNviYfTKBvTp5p3mQpxp0aEKSG1dWrp0bwhg9 fQtBv/Fo2kjs+bklRQx64aDYiCbi1FfCiHPsFdFz/EbxajC1i3o+mjbS+H5ig20nfk/Y4GBJFil0 hMsY3SG3QG2XWY/WjKkJfIMI79gESB4YCRu8FrSVQ+4ZUTg/a4INIWR87em5JUWhwH682YimK+cP nu0NEjbi1WBio56Ppo0se+/7jYe+JivvOAww1YQwGKLAedCg9SvwMfv0CCyNiR/mmpaBHyNhg6UX lRjFFCdn/I6aNlJ7YaNSDKBEZDaiiThvz/YGCRvxajCxUc9HU5Ed3/vcYAuD8yvODe6AhY2FwcaN j02KDQZxZZBOOekNe7DBp671XJGw4a9XDiQyuBMpoShyGgixczYXMirFzpm+caWGEkXsOH6zViRc yCdULanl9yg6L4T5njT9xMUHOytoZH1Bn1hTr6ETli7/SWDuEoJQxvl9boTJkTGmS3BxjUwReUdC +IX+Swoyvp41baT2QkalGEn1MbiDJlz5Zq+I7C2vltTyfGkKsuP7iQ3tzbnc4cWQJSMQXZs+iCWr Z97mhka6NgmJi5ufKhAqOC+SmiJCBP6w0OhZAAURajqYRIZxlP1v81A1nPR9tlUgNMghvA5G6mpv UsaXngFi68eU4KSGCk9WBeizPQvTMUtj1AgcHzMNLGhpfsLwiJRpuEemSHjgrIiKhWBeb+GWRcra aCLzi8SeB5KFMYjybUdTUU20fnkl2msviAcSysx4VUxc1POlaSGN76eU4KRGcoJtqExCdCrAGIWA FrfJCX4wR8nJKPY08Y03KOb3F2LfGGN4PeKScpvnjoWkqgcqCjDq+TETVoyjKVScpqKKKBAkdTBX JGkRn5ZQvHbnUUUrjjE3M9KXBYbOU2oSQV7wQieYcKYV0zCkl0JBBW9oy+/LNe0498hz9AZCkhec KVE5Ecxhcp9C2kcTZBSpPZ7bGM5watxkVBOG0W7uay9I86JeFVNe1POlaSGxR4t8N6RydlKfj1Ij Y8+8XUUhjXq0SJHB78sWB2FywBJAEyM/FSBUUIJonKkMMXKbQppHzUZsy13FElLT1svD6H2rRx3U k60AyQn5g+qRSRTOiar5FtdPJDw0cIsEX6VNShTTvr5nZWSgCgxF6Eu5ouEjzxUJD3yhMVWEpJwS HNGV9tEEEUVqLylRjO5P700RE9XEGow1bXtBmhL1qphSop4vTQuJvR9SghJXGxBigwvR/z9hb44m yZIz2fK1ilpCDpET2zvoNTTzmEf0/pk+RwQKM/ffI5yor25IwjCIw1Sho3XEvhhzABSdyQlmvKk5 YYONlb4yjtfy3CBlg29utYyoFLFzNEHtR9MgaBp7ZWOkKAO05sgveihnnqwNojXqy/q0Uuv36Nm/ 19ZXL4dbKN1+iGfMnHZkVgz/nWymsNIqG1/6+7KRUG8Zq/FcYgxSJlhFt0KhkYwUcXOkQqmj6ULG XpkYKdIvR0DgYjRh5sneIOVivFqp9Xw1Fbni+/otYRMlt4rgLyR6BsQaazFG0dwrmaaUj0X39+Xq /7LB8UefY8QjoneOnto2VIjQaVQVqqK0kSJxLuZKxki5F8FTAKuI5HuyNojW/PEzqlmp9XsVLTLR vePiJ2ddcY5Xi+Osw4VYQmDLpHUPL+CvjCjoDJQyM/iOmjKDlAu2jDiHdaQgg0s775ouZOyVjJGy r0o7zaivmmiYnuwNUjbGq5VywiP2VlORK753bHx4hQ9kUNZ3kErlGogu2XMJcsFx6w4f2BTU+R1u FAsVAUoFe0bsdZm/rhDF+KOeA9SWnQgvRGXw3hNIJsWoYV340dYApUGHsFWM1D0+r54DTFzvSGC3 hr57HIHbuvp6DOYGGdeqoOHn3wxK3FbjeXpS4m+fG6Q8sGGkFZVS+MZceaVW0yK1N/XISFHJeaW8 VIwmuuxqWntF8tx6NZhcjOeraZGJ7x0b7NYwAod4jMbLxmDZMlY2uJ4omc/eM7PEsVAec9QHUjJ4 m5ooI0UJWqlVVCT5GnNNi5FCKVdihYzR5NaMR3uDNDHq1AoxJq25VbTIRPeOCwvVOMeHAzs4pQSy CrYxzbfIzYxUweZ+vqsuGXytuYEHKRm/iSqUjRTzXU+aFuFn1V7JGCl+fj+1bmaMJv/DFfXL3iAl Y7xaqfV8NRW54nvHBrs3bPtplbm9YTJjMJKMjwmnG6GMNn+YYVZK77hI1ecGKRvM9Nj1rhTlY6VW 0yJjr2yMFDN2bJIMG6OJ2ZEne0XS0pD69WqlWDSqvdW0yNp7XWJ0D5IZnBEEtQv1cthYjFb4h8sV 5AY/mplA76iUbPzsc0XinQnriuIlxR60Sq2mRcZe2Rgpukd2BML+aqJHebI3SHNjvFqp9Xw1LbL2 XrPRfTgMWVjUIwK6Zb+VaatRjOyiY85bTBfIl+AT5o98UBMy2N/awIM0Nfi+aaurkXKBIFJRlDZ1 gLFWLgajJfw10yKjiBr0yVyRyYxxaqU2ltW0yNp7zcXJDL52bgCOyhmKlovBSAh+f9ghFu+j9zdn astcQXmfKxLvbNazpLdSuFmp1bTI2CsbI0VCsK8AzlYTifRkbxCfW69Wym4r0aymRdbeazZOZjCB Zm9NMzArRpQWQGl2+KhMyeCz9uYB72AWcByy5akCuuaYzZXGlaGcrgwvN2qSF4PUVokYiNkar0y+ 6aF9ebRVoDToDx4embqM8VWzyLH1NQlcnGzVSdNL9zEpIYZKGvv2crzr+TY6EXLjjrzw7lVmkPLg HbOXDO/wk55FxppPLeYkcibQeNP9WahGsAV5a6tIOp/1aDBf6vF69SxSa6Tt10z89qZEiGC5oguI jKICUXT74qecy+c6iZHq343wjtR8qkBcc6iWOa/BDPHHo54DjK3SgAzsQccMcarlg8pdf9bUAZoN bOyDq4PJQl2OGvw7f6+lrylgr4b1puftKP/bPojhGH0PJ6tDwq98U4qYmdTWM/6pzw3SZGALib3t SvG7PGkqctkrD0qBMZ5jqta3ooqIkUvx7+YGSXNEGsapS4r++1HRAhPdZ+lw2koGGfpvNtLMlIvB zMgsAzH7kY2NtE4KSQVf9GzYItNU0km0MBupHKhQahUtMuZKBZWIUrSjP7/5JjjWUxNRMrq62ztI s+JJiMntuyLcLJLXPtF9xsVpKdm+YQC0dO6qKReDsR7JVxZxjgo2t/OSKWyDqXPcTuBzgzQv2D7i G7NSLJhUajUtMvZKxkjZ52a3k4tyaiJ0TpDc7Q0yiVGvLin3i8ar1bTI2nv9ljQzKKb9WKkZyqWw HZsuRt3wkWUgKtjcHEecv3sewBFRnisyqcHw1r52pSiKK7WaFhl7ZWOkKMw9SWBqVBNxMoHme3rs HaSpUa8O9v37er6aFll7r9lobrh3km3VYYMwkhvFyC5ioPrCFzpL7jRpnErpHQf5J3IRvXOoloVF 2KgUXTtSd01FLntlQykwYnBX36XJOO/2/mO2BCkb9epgsjHRrKZFYKN+DhsvN+MwMc1pxnjHidiO URdj8MWNXnhH1cZB27DBJpBOcfH/PseP5nNlgy/IW60NZj/C6QSlVtMiJImaysZI0TW60182qumD pTa1y37tHcTnjleDkVPr+WpapPbw8+vcYJXUCGjd/ISt7YZ7KoM5uZx931RtbkE2TraFdAz/pzKD lA3WOy0LiiXjvz1popK825veuBi/MYsfZaOaiNMP4qKJMi3PDZLnjlcHuzy/NG0sE987Nth2YwT2 zTNKdVNlMFr2Fj9Ub2xUTMazscxscaQWGToLkbLBHh5LtMF8/7nfVe2rqQiaxp7PHSliYANYp0eq iTjzBd7L3kGaG/XqYLJRzy9NB7nsfZ0bzF9aeNLP+23Z5oZYYqC/cuTB0SdW0xMnd5iUDWZHG3mQ ssFsZSgbIdrxJ0WLjLmSUSlCYPTX6YrcakOUfoI3mVFrg0xm1KeDyUX9vhQdZKLjvfyaCz57RZBu h6LDLxWFKCanyWBjuj8v4w9kcM0BkQ8VmKxg8t7aaTB/758ROmrO32OpLCCSdpFjMSWhWoiPRQhL sDF1gCZE/TmYJMRjOFAPDg5wonpHAfs3dJ1OmVsYhgMxflenU/WMYpZRR+Jjn7P5wXJDH6OnEWk2 sIOkxdVIEX2logjXFhhr5aFCBEBp1QmRKiJGNnnczQ2S5vY4dUmxHBNzl6aDnOg+46IFRsy17mOb W7lYjBCoBUNG6yjGuj0j4ZCIBtVfXyTOWV9mgXGlaB4qtYoWYWStppJRKUJgRds34Wj64Hzho72D NC1akq3Q8XsU4XcRftU199Wrwa6K77R5SnNZxeFiMNp793SbGNyN08RnZby/FDfV+ByDYJEmBptI 7HYHMw84oqDUalpk7JWMShEDVxT6QjjqUxOhZ+u3iVh7g0xm1KtL6kRzaTrIie9dZrCJwwiYnOa9 tqHwJQ1EYWQL4XvH2nDzwBZC3/jEUOMGiGuO2LLA6PyzMpSPT2oWqa0SUSFS/S9TixB/9BCi3+K+ bA3gU8eflTkujxpEiiQnjOszElpx0jhbYocEfvSSMFCqw5LgtqD0Cd2V40gtT6XITD44Vsu6YiFJ 4GSCmlfNIq4G9JU6kN5zaSLeHz0GiEOSUFsDlIRsUzJjKtMoaHdXzVNc70j4cLlBEthLwLwNP4a7 KoNRnPFJ9lDPeM2A+DF7RMRBWmQG0TfW0ZzmOVJ52Xmx1L6aoOjBns8djBA4fG3gRxNR0gn4Zh57 g4R3Sup4dbDL80vTxtL4SNqvW4lcXJ55ENbXJyeKEQPdgIMCitc/GUzQ4imld7mn3MiDlA3vKQ8b I0X57EXlsjGaiqTT5Ih3ts9bZEeKGLgWIGRUEWHSodzNHaSJUacOJhkTzFG0ACVsrX3NBQee2WFA G/OH8cZwIZYI+Ca6RTA/EDfBJMoPpeSCmxyaBUHKxS/nL+VnpGg6nzQd5NhrZlQKLrghwd7K4Z6a CPPbo72DlIx6dTDJqOeXpiJXfO8y468bjjJJ4FbcthbF+D35xrVLob6UMyjJtt4OhjLcoHD0ubKR W/4gY4QsOiJURYRZJM5RRPWxI0UISJeMKPpgFkahy9pByoU+oWqlaCkayypahAq25r7OjH/8GHHu w1Mh5UKMPKAi5AIPfKFs8wfyN//esyMOhSIzSLlgD8nUZpVikeBJ00HIjNjzucWoELhRA8TBl5qI k4txy0btHaRs1KuDmRn1nFwZTUWu+L7ODDpza2WcY1qmZAwEF8ySlItMvMkFo6X6lm8cHaBUsIPE /naFLL9by46eAcbWVGaRwXvmbdpcVA0R5nIKsyK2DlAadAivK0TqnjBWzwEmrncksHGjLPiZ74zQ 3Wo57rNbzLobhqc3+IcQgMOyiDgpAlIaWPOwliqWXoRZ8/IwinJ24TKX5xYj8fLFcDKimgjS75ij 6dgbZHqROHUgmZhYVtEizvno5pen1BH67biBX4lLmFpdHYyUYBwqF7x7vzJQcb3FuwVxjo1D/v5F 4pwjpiwoDmbic1BBqdV0kLE3WVEp0oL9eSbc0UScfFDFpDz2DiL5x6vBkhcTzWo6sZz43mUGu0r4 OWm1uHAsX4qA4WLEwI5Sh0iUr38y5qBQQCre8d3FpkEQvWOUxtxZGBJLanBQQanVREl9tzepMRi9 Fbsob5qIk/XJsl97g+S549XBzI2JZjUtMvG9Y4PFDAop2ODMa/db0KoFIwZypmxwitKik0JBKX+r P5UZpGz8pNsJGyNFFf2kqQhN2djrC1YpcoPDyL6GjvjURJxc9Xi3d5DmRr06mGxMNKtpkbX3VUfi hsufzhlQofxjPcKOpFjeHi78sSzmH7/NyIQtU/LjMC2jDmekQMoG1ZIbIgdLERkp8nQ05agC+XPs lQ2KKjEKinxMHjaq6YOa89HeQcpGvTrYzfPV9Bzfu9z4wzei4x03nR82BvPShuy9dg8zJxfS0Spl q8oCl3mQ5eLDBtee2PkOJhucWFD7alqE4kpNZWOkKBbcBmluVBNxsvv5bm+QtDfHq0sKM7W3mhZZ e69zo5MX7rqkpvG34rafTmYthioW85IbfIlpYldKNvK9pkQOEu/cW5xNhLBRKU8oRPtqWmTslY2R 4vXj83e0N0eTcT7aO0hzo14dzNyYaFbTImvvNRsdq5LCLJiHDY77HDbEUqZ4lNAXhRs2rbxd58v8 giOkjEEG0TlHUm0xR4iSurpdDHQJaAHaP/WUihGiYPAKutVDjNxkUOZr7CBemXVcOphMTCSraJE1 95qJ5MUPd1S6S5H3lSv/M2y/MJqfdG3Us35kXSbYQGaiOGbL5sYikxZsIWldMVIsZ1T5UbTAWCsV I0Q1waRqsqKKiJKPnN3NHaRZUacO9v3y+2i6kLX3motkBdL5bAxUuLmK1nMRSg++q5MfgMP1rcuZ 1QsR+c4lzPh3E4INJK0oKkIlfeoXlPBMABKtdspBRVjr+Z7VdUd9Kvmg6Lrbmb8bf10p9J/L2aNk vT92vo6drSSWgPTEHIS14uT5wfj9WaTBc3lo88BKmaMEVfsY8QfRMe1nUbFY3gD2JKl8FRnWzdy8 E4O5hSfN0dFEiOzLdN7i2Bskz41TB7r5fRRdkTS6rytOpNm+Ifk2fswZlIvBGOey1TJcsD3IoGx6 +DiGznGzic8ViXMO0lxTPJgvOoculFpNi9TevEwjZRMp06sop8H8MdbcIHJPMVmniiUrxvFR9D+i +7IXRZoNHNab9Duc+5y8GMz1mByMoGrj7HmiZJel7DDu6XMu4oA0MTjqZm1QDMZyLEHtq2mR2stz i7lzx29jQMZoYo5Pr272ikxijFeDmRnj+dF0IRPfOzbYwuGvaWvbGwbRMJjtbdbuaO3wMz0JuyzL BpVPsyBI2fhgA3PYGCmau0qtpiJoGns+t1LQwlH1sDGaKNaf7A3S1BivjtTl+dF0IWvv6zbjD/sB 4h37CU6LKUZUtLh8PQHv+LHZgps4+eFsMxkVVYaBoEjZ+PWHfTnmxkjxEz1pWoT2/Ty3GC0KQ/2w MZoYwT7ZG6RsjFdHitjH86NpkCu+d7nBd9P9PWm1PAjRVmMw2i3KBPKUooBpBKRoCpWSje99bpCy wc1yrU9GKgcU1L6aFhl7PrcYHa3zgb4po4nprid7RdLarFdHitjH86PpQtbe69w4dcU/TiWFjW/M bQ4bgzH25Y3FO8pFli7CBpstzRaL2TxHqQMS7yx5XV+Es5GyeI7UalqEMZeaysZIUYEx8Qn7q4l1 tyd7gzQ3xqsjRezj+dF0IWvvNRutLBhEsXmXCKgWPQVqbhQju6gqc5cZbHCRednghZYfx2p5jv8Q aW7kcxqQMULorFAVpVkdZMyVjME8npPEc9jne8l6wZO1Qaw416cjNX7z3FF0IWvuNRfNDDdZOnqA Cw6eHC4GY/DlByZtqb/lpjfC/KGUXPzqc0WSGYiz/oOulfKEQrQfTbwTg4y9kjEYPYMbmew/RhML ME/2BuG5y6sjRezj+dF0IWvvNRsnM/JxPotYPgI3iVHIiVcGSRilJjR5aFKVkYp8GfwARmTp4Ori wWwLcoaV5sjPac/fpNtYKg0VYcorX+i4tLAX89HUACVBf9B0hAi5Hh89C6ytrylg44YzzXRDXlnf l0OMcGy/+PUhgUTtOOWvQpLwpyJOsYKUBbaO2PMO5hvPHvCbogsYaz62QnQG3n4vnSrCPissKCdB 1lwRzV1ODeZLPX4fTRdSe6Tt11ywf8OfkZrQzC4Xgznv6kIFdt1rYpRskCoXLDL31w9SLliz9f1e KRruSh1FBC4S6mOuZIwUawI/M3/mYC+a2KbyZG+QklGnjhCRj99H0YVMdO+4YFuJP6ZLEa0umM4E Slow6UpvDxUcHfX3pbL3Myf8Tv8qUqBMfPxzH2Jl+C2ppR/VXEhtlYcRYiXNCyBMitHDisLNFh4W KAvjz8gQ87h81FzIsfU6IU5ryc4Nf0X6YXYpT0IMRt7mIi9Y+JUL44iQIZsVsoOiPFdkWktufW4N MVLU0ZU6miBikLFXJgZjLZX9HGktRxOLCk/2BikX49WRIvTx/Gi6kLX3mo3TWnKFm78kwxuv8urr IZYcZhclNQ9scOGZWUKjmAu/YIOj5M2AIEbFa8F3ie5SvBaPmmBD5LJXNgZjx0Gui1hFRPdkrojk X04dKUKfYEbRBUx0n70ezQwSjEOS8e0nH10KF4vR6WUSC7NkeluBb0qZGVyc0CwQmczgkFG7mJGi 3a3U0QQXgzDRqaZyMRj9oftOHPVFD1n/ZG2QUjE+Hakf67d6IKt/59UfW19lBeufXAsbJrKF1KxY jJqEaZh0o7wr/pbUPW5ZlQk2Q7RGDWJE9LG/ZiVvpOg/K3U0wcQgVEBqKhODsVMpffQqohl8MjcI 5RU/zzh1pC7HR9EFrLXXXJysYAHGX5Klim+9yAANgzE84DMtsMPCeC6PszhQSi7Y++FzRZIVjtNK 2AhRPlfoKIKKQcZcqRjMAx1/nLU5iqgTnqwNYlasT0fq8ruK8qsaSfIi0X32hrS1cFdlxg10xux7 mTdkMEoTbhflJ6AOz9VxRI6QXQijoT43iEE5SnN98ZIijSp1NEHGIGOvZAzmj5DdJqvJgB/tFUlz sV4dKdgYz4+mC1l7X2dGvnxMuf+LVbsho5DdsWN1uOA7nM0CInV06gjN5oMu0+MkpSKl6sEQstZW 6OhZYGz51JHh1e6pEEt2PgH744Mm6tHWADx1OXSECLo+Hz0L1BYevibhpAS7N/wZGdp5c1ebisE8 IGnBg1XeWHdk0C0qJQ3/+twg5YFtJi1GRoqqsVJHE1GLoGnslYmRYjsbn95Cu8O8aHIu6dHeIOVi vDpShD6eH00Xsva+ZoPdG/4EDO74DYYNMRhiqOs9wrLh8Vh/XTaCmCWOzSIySMlgmwkH5VaK0Dlz cVd0IWOuZIwUVv5NUTWaqHKf7A1SMurUCrlDNOaOIrgYpObeZgZbOAyT3tjNp82MweiQuSgvXHDA q2MVDlV1YMLG39ITpGSwicSWkznXSBE6xyuUOpoGgbCxVzJGimEE91okM0YT458ne4OUjPFqpehH a+9ogo3H+N69J+zi8OeERCaChg2xxMDReAfplDK5NI44P5QyNQi6TUGQssHqIkcnVorYOaRw13Qh Y69sjBStklWq78loIgOf7A1SNsarlYLi2juaYOMxvq/ZIIGod5Mbv3+wPmNuLMbA/7v1he9gxiZM SSokGT/72CAl4y+3OCU1KkXonFJQeRQRZYFwH2vlYoToF35m8w+zhVH0k6bp0dxByoVO/Ycy5Uit 30fTFQnNiZreccGhW39NOmU22AwXYgnBTexywQ/WNKDDLhcfFaG3FCkXHqFNYlSK0DmscFN0AWOt XIwQXSP7VpIXOYtrlJQeFKJQX3ODaG6dOpi/4fh9NF1I7b1pM9xjmSEEnRKrHMPFYE6z+5LQUXFz XKLMvXg696uPOe/OYwbleM2FRV6SShH630qNoguotfTFK+RSfKYMjyKi/KPyy9xB5KJO/VfGlIKx jeVoupCJ7l1e8BkXSwfLE46H9B0ZjAKFQVPscmSkPzDTKtLjYC3PMe0uUjL8gKtvhFALyAodRZaU d3OTT4M5lZ+7Mxz3oYgo+XzLZS3vQ5ByUZ8uKbeGxqejCC4eosPc9KuvjhsizQY5f0xoZuagVAyU f3PmgnqO7WeJMmcjYOIvT5EozAvwVIlg/4jlSqGUodV81FiY3m351EIud/njkl3VQ4hs6gzr2goz AqWh/qzMRlE1PFUk775xLQmvbl9EmN3uuE67ON3pQahN3MotBR/c9ZaIf01nykmrhOvfZYA9I1NF KEJwuYLNgipKFoidvBRHhEaATYF5uaqE0E4nGjvzd6PXlRBSkTiLa0fJUzzvYmfriL8bPSTTUZMB YmjkTeNy94TPYUHThKJAKZsoDjT19w5SBtgqMqVDpIiYYxV3TYNc9nyuUmB0Q1y/ExqqibDZsnbZ y+sQpEzUq0vK4UW8Oprg4jG+ZeOT14GdGv6GjPdt5Po+DEYbbDNnMtiYSwYfYapzLN9ZnQ5SMtgs cpciTE5XqPwoupAxVzKUys/541tmbRzkqeknE4Z3exdSMuLUJeTu0Jij0kFRGs7H6JaLT94KJo38 NSlTuc9zuBBLZrBfsFwwk9Q8YE+hr4ojszznqgtIyWDXSF+XShE6ZxTumga57JUMpcAokGaJtooI k2/Z380N4mvFD1SnLil6pgdF5MVjdJ9x0bkKd1O6ExEu2EdQLhajXnUXjmYJsi0AaVznfvY5hEHy zjtAY/hgslSKyFkCVfvRdCFjr1xUimxnBtXJiqOJMH8/2jtIyahXB+Mk90SzmhY58X3GRoem/LBs MgobfJk0O9/RIEZmUI78czwOw9++ZSDhWot7b8kMPrnjc4PoHT8sH9G6MFsDDincNQ1y2SsbSqUV /M3Wl5sm4qQxvezRZgyivePVwWSj0cDGaDoIbCS+ZeNlm+FuSncikqPcwt+5m8XoNzz1pF2mZ/2N aa2RShX+u88VSc/GAhpX469U2kYOKaj9aIKNQWpv+pFixPDBemHYqCbi5GyW2UkbE00HKRv1ajC8 Op5fmg5y4ls2XrYadBA/M2qgRmCknuJqMQouNouFjR8/M5Kgn0Yq3v3pc4M0N1gutdcploznyIXZ cjRZjd7tTWtTjBi4/dVMcLCnJuJkdFz2a2+QsE93F68OJhuN5tJ0ENhIfO/Y4EvY/p5MhHggxP7E TZbBmPjNiXRyg+8eO1DhP3JshN+Ks1XNgyBlgw9fl41KETuXOSh1NA2SziP2eG6liIHdHmWjmn4y 8/to7yDNjXp1sMvzS9PGMvG9Y4NmyrKRMs3FirIhlhj4DIIDcwpY7wmVjSyFmLl8JaCRBykb7EQI ZRUiTO5qviu6kDFXMipFCGy+MPEcf9FmEyUnwzvLWmuDTGbUp4PJRf2+FB1kouO9/GpCizI5Gygo VPwOcqkoxI9PLRwmmNdqFrAP3gV2B0M2BrRiPlUi2DpiBzxYSugIHT0QkdtvcSy2ykMunyXT2WpT GqqGCH+rem0dQFu8h3HoYNIQn0dP2o+HuN6RwDYSfbcXmM0mTEiC8cvaW1j/YJWbLfrLsk3KFCHf +5wTZ91swu/IrhE738Hkgcr0rmmQy16ZUCpNA58kN/CjiSiZwLjs0ZEMUi7q1cHkotFAxmg6yIlv 2XjZkTAnzba8dGl//1K5mRMHs2PmVuKwQW+aMDl6bJY4OmO7u4EH0TkHaFlXHEwyWHGUnqNoETI4 5kqGUukxaVVLRjURJof5Lnt5Y4KUjDh1CR2/4UJFJMYia+6r14O5hR8OJXCOjDtciCUz7MhMDOY/ TR+H/ZnHcpiWEcgg5YKtI62uKkXknFBoZYuidCsFjrVSMRgdKPd+h9QqIkrG3KWi5gZJ7XmcuqQ2 ltW0yES3efHUpbb4ZB7Kr4vKBTu4O0g/GJn1jy1neEdR51dQ0xkoZWL8ynN2DyDxjmEbuXmTggwO KEjP0bTIsVc2KkVU3CDD9kAa5Wr6yTLco72DSP7x6mDmwUSzmhaZ+D5jo8UnM5S/HErgnZ/56Usi RGKQ9n7sWbNsy/YXdlrDVHHIlgFIAV1z3Sq7IEeGuDmgIDdV0wL7bqtEVAgi2DZsEhw9hEiBdbM1 QGnQH96rI7NRrJpFGtc7Etg9S8UACXPQEF6F4j1toiUgZSxNOQj9p4cjIOFPRQqUBDZI2dkG4md1 oPWo5iBjy6cWcsMOrYokVA8Bml/HViIWKAn1Z2WOy3A5ahZpXJ+RcN6LX5x+JhMcF3uDAWbZyRaM 35DlKYsdyleuvk6EjJ9MDodpGXI46uY4c3lgPc2uthhSlOJPmg5y7JWJShECn0pwkE37Gk1Eyarh 3d5BykW9Otjl+aXpxHLZ+7rF5KZQIkCabQQdorLqBZYXg+1Y5iblq5ePGycLOGaJA6IMOQYpG1wf ao+7Us6nPGoqctkrG0qBsZjoaRjIqCLCZNfW3dwgIf84dUnV8ZuiBSa6d5nBLSv4Dxec8TxcFCMz 2ERaLrhIo7nPPtM6x/CzcYtMZuRQMlSMEJlVoVV0kGOuVFQKKjhhyWgdLqLop9tUHqwdpHlRnw5m XkwsVQSpRdJJJ7rPuDjtJaelQgVlXcfq7rcUssHzzhWygovNzBPbMoTMCs4zJQECNCmIw952hZxt fdBzAGe6fufqSl6jyrjjIE0z84xRQ4RcmXWzdYDSoEO4WCHYOz6vngPUFg3MVy+Ha+HUkiYEOjsq XcxyIDMWWQv356crRCo0cODBmAcpD67cLJY2k3MVal9Nrnvf7E2hPpibcgjOjKgmouQzgY5KHQnm uSLTbLLWbzIP9v3Her6aFpn43rFBB2GhydIEXyJPw+n+ymAsQHgzkUnh3fyG/kchnePLuA08SMlg 00gn6EbKhYhHRUWSrzHncytFf8HNfiGjmgiTLbp3ewdpYsSpA8nF+L2KFpno3nHBPhJ7ZjpiL622 E6Hlzlcg+D0ZKFv6Ubz+noEKl0h1PudXZKAniM45OmOYJmUjRSXxpGmRsVcyRootjV5IfWkiTq7n vds7iM8drw5283w1NZYrvndssHHDUpMemUm0YWMw+mSmU/IuMmXXRPhQytRg/rFFZJCywbpNy6mR 4iWq9tW0yNgrGyPlRh7PbMFrNREn1wTf7Q2Shhqq4tUl5Q7RRLOaFll7r1uNFhikVG7Igw0+ANBB +mJZvbT0oXzl88VpD9gz5pvj9CbtqnkgEu8cpWU5sRhSNKPVvpoWGXuysRhpzU0JsH80/eQFfbR3 kOZGvTqYuTHRrKZF1t5rNtqRWLpS35obzDodNsTILqfhUghSwXnNhbHTgpcNVv1bmAZpbrBF3v5l pehenzQVueyVDaXAaPB+paBhoBdNxMkK/WUPnwYpG/XqYLIx0aymRSa+fVNeDtZpejlXHO9YjOy2 k8Wcws7iJU22u2OMk3NBvjsOiHxukLTwDtSyo7FS+dVpA9W+migr7/amNBmMioAL5HnuaCLO33p1 2TtI2ahXg+HVer6aFnHKA03LxtMQ9eQGV/FP4fedk9FtRQejZPOiHN8Uruf316PwQCrecVOQzw3S 3OAiWvu7YnhHFVqp1bQIhaaamhsjRc3ww40nsDGamKbUK9kYe0Xy3Ho1mGyM56tpkdqD/a/fFO5L MAKKFXd6lY3BXItol8KeNY4REiY36dqQOF7LY4OUjNwCcwllMU7dR88CNTaJwYpdSwa+q6I1h362 69h4MjZI80KXVip5MV7zWlcRVh5i+5oJt1L685JNLP0kLQ5EpcLXX3CNlOP+JoQY/iMUIn7mqQHK A1tMWmIplKTIl5Ks6KqH+vpma1KiEE10V+0Z+lUNr+HNVhqLAKVhHFqh4/PqOUDj+vTlaDfCzBkz ymGBW7s6f7MYCxLfM39DPctdAqGBWxVKA3cJNmaR/LKO1lxihKyRct0+UqtpEVo0NRnTYk49zS6a 0eSE1aO9QcrFeLVS6/lqWmTtffVypD0xAkuUvhuF+BGsibJuShflEgBR+kl3Xw2ahIYtoGcsqLGs fZNhBqMy1ZIUGaSmysNATs261eDSQw3zaKuAT8Ud/DsiG8NqWeSY+poCeipnWdHLCeC0D2y7BCMc 5995CTXKJ5T8+Z2jR0oWGKgoM0hpYGBAzXkwX3VOWzxoWmTs+dxiLlvMxPJoos5/sjdIqRivVmo9 X01FrvjetRHs4vCHpCfmxxk2BqMvZn4qLRELo42day3KBp++8Tl6BZGy8ZMv/l2YcXLiQqnVtMjY KxsjRQyQwzCsaviPJ2ODlApdSlqMlJVDjFXN/rmWvs4KNnD4S9IgelbKXoNkAIv3rNdYC7vRn/2r Rs0uZEeRjtMiM0h5YJ+6ddlKudftUdMiY688jBRTMjNdoyLMuWeFCaCbuSKhfZ0a7Ob4UbSRHGtf 3aRG3H/YEY1Ze2DehnIxmH1wxgXUr/lWCVEyD2+WMEDoc0UmJ35xZjSMKQVjVNCVWk2L1F6a2sXo IZHnOZa/qolWuJrW3iBNi/GqGF5tNKtpkYnv3Rvyl407YeMXE2zTjQ5GV+Tl92YGq8nNfZaxjdMR UZ5zVgCkmcGHy+1ti+EdFXSlVtMizjjMc4tZwKaNdswXTXSqenWzV2RSY7waTDbG89W0SO29qa7c T+nYgdyg9O/sxWKUJ7njDjboSfoGsA3ZN4d5hT5HRylSNv7a+8rGSFEzVmo1Fcl7GXt9T0aKbvG8 KKOJSuvJ3iDNjfFqpdbz1bTIxPd1btjruzERNri95bAhRlSWJzktQzBM64Ow6KKUbHAxhTKDlA3W CVC1Qnj5pGiRMVcyRoq9dvnKES9Kbuv7yXAEa8mMsVZkMmN8GozFvcSCT6tokZp7mxlMULTs5Vft 7IVpHozG4l/GHKY5l+AYJitnvjmmXWQGMagkp73vSpnAj5qKhPvYKxkjxethDXPTxCv0ZG8Qn1uv Vmo9X02L8GKr6V1msNqPc/QPvKhpQt1oaW9Ce+2Zca1yubuM0TwrJBdZMzpAqWCbif3cCnkm4UHP AcZWiagMdYI7AuRBNdii48FWsqK2CiQp1qHBTIr6vHoOsLb+v//8r//+7//+35cDU/oBb8uWBXqP TlosRoXlVKU0eETbAD8Qimd/+liRuEbj7UcYjxSP0cBXahUtUnOnZK+Uu2mz5WQ1UYTplKyPvUGa EXWqEE6t36tokYnuXULkmJRcsHWyg3TqA49OpRTNtXZwwbcjm/hcJNiR6t/K8E8izYmceJIMMbyj 63/StEjt5bnF+A+/Um9ajCbq92pyQBifikxejFeDycZ4vpoWmfjescH+DX9OpPl+yLwfg1mwZQsO vz/LL8l9tiaXjX99jshFyga7ju16B5MNzlaofTUVyfsYez63UsTAV7/Cxmgizid7gzQ1xquVcgtp 7K2mRSa+d2x462O8++X8hAWXc09uWTSGn6z8+J7kJkjjZAJLNhwZRWaQssE+SrveleLdfdK0CJMu aiobI8VbxFUWYWM0sVLyZG+QsjFerdR6vpqKXPG9Y4NLif09XTwmFcvGYKQnLwgZTxFHP5tfnYlc 2xHagT5Hay5SNjgi0a53pCgjK7WaFhl7ZWOk6GV6uNDhXzQx0/Fkr0iam/VqpdwwmmhW0yJr7/XA pPMXzu1mDEC75dlC2ViMlisDBbpJF9FMFr5ZUTJ+9rEicc4p2azrrRSdW6WOogXGWrkYIcuPfzbQ q8gV5kdzgzQzxqmVWr9X0yJr7zUXneh0yyW9upnB1YGdyylGbtF3psenfvqgwJQL9hT64tDh9rFB mhgcGbLvXSnKnkpFUdrUAcZauRiMutQJELkYRcyKPpkrMnkxTq3UxrKaFll7r7k4eZEVccs1buU9 XAxGfeknTW1FePVlzNmhTAM7bEt3WyTeOZAqYyPk6YUIraJFxlzJGCmqJg4EQtlRxOrBk7VBfGx9 WilWpGquipApwq+65l5zcfLip9uwUrpOmcEMXSFqV269wjeqWa+akArKYWxAQEUK6JpjNmbjLhnM 39UgUiSeWT2UhxGibGKn810P+7QebRUoDeNPIBxcl1fNIuRmbH1Ngs22HDAT34s5mcINRI567a4c sLrcrGcq3gRxpNaGI0BJYEA5VYmazPHu9l89B6itaWkrQ6dowSWXo4bV5pstmBugJIxDK3R8Xj0H WFuvSThvBcb9Aenc3Z3d1nIwmnDuvQ8NjAUQYtZXIVlgYtrHisxLwR4Si4iVoo2q1CpaZMwZ0mLM +rINQyJGEQvTT+YGKRP1aYVYh6i16EFxgWReYvu6F3U3RAp+emEv4ygTg9nvZzmTqu07848GydVN UuFQKM8N0oxgG4l97UrRdVVqNS0y9krFSLlMkvbIwV40sRD7ZK9IqF+vVmqjWU2LrL2v8yJXNjj8 YWKiU1rsvfQaB3sTrht1kETVxpJx4mQzoZniiCgydAogkxjMEfvOFEMKBU+aFhl7ZWOk6AN++llp MmM0sRD7ZG+QZsZ4tVLr+WpaZO29ZuO0l+S9vyc1CbeNTG6IkV38kN+zIYeZXaaiEyebCMsGidc8 CNLcIMtneqxSpMSTpiKXvbKhFJhrq9mpR3JVkxfZ3eylxQhSNsarlXIj1aOmRSa+fVNeLSbTHHES zN+TntirJXhTLoy+mM83Jjc4Z9bY2VPttArDjT5HS+dzZcPbb8NZMHMjY/vULNW0SO0lpxajT2SC BHuOw9TEa8/3UczOY+8gZaNeDfafm+dH0xXLxLdsvBqzI03A/p7WJbQRZWMwK5NssqCG9XMzvgNc 31vv+EhM8yBI2cB1u9xKpSdhOkKp1URberd3qvZi7L/gctewUU3GqVeyUXuD5Lnj1cFkYzw/mi5k 4nvHBmem/D2tTEiSsjEYpUAuB8XcNy6OI09dlUNK7zhH3jwIIhuO1rLGuFIuYEVqNRVB09jzuZVy G1ym0I8mDDMouds7iM8drw4mG+P50XQha++LdgNpKmsn76hQ+P2HDTEYokb5S3+pXQY8/Y3ZrV7v /lSGAkKkbOS7ESTCCFE7PilaZMz52GJs5fqWrWsO+/qesMrsxJZjMD3CEZFYOz4dTC7G76PoQmqO 9/JrLrgr1iICaa6mGS4GgwtPgMgF98e6z4AwldI7Li71uUHKBVtJ7IVXKimi1GoqUu7VVDKwoBTV E5/FQDsLoNFknI/2DuJzx6uDycZ4fjRdyMT3jo3fmXRgec+D0H1NgBIBSLngm4VmCgmtkFxkO/QB SoXHRELYCFHAP+o5wNgqEZXxIBzC8qCa5ABLaI5ZxxYhC0xS1KGDSUN9PnoWWFtfTHcizMSAPyOT hl5rUhYGY5ah12x4x0SqT1o+pHDNAVGeKzKvB6S1OFGKkGhDK7WaFqm9KU5Gil/UK1Kkopr81avp 2DtIU6JeDZaOZDw/mv5HfO9Sgo0cFlN4940bmsvGYBTz7HvAO4raTFwQOkIADtPy2CBNCnaS2PcW gzJq6kqtokVqbt6PkSIEP0ggGdVkmDp12RtkEiNOHci8GL+Pogs55r5aUUWajRz+mkySeeNIuRjM 9e5/5eJP7o4jTPY4O63reC3PFZnM4Cy9Xe9gksGxC7WvpkVqbzJjpCiqXaqUjGoizu+9PffYO0gz o14NlswYz4+m/xHf15nhrsoMAvCOU6fJjMWclc1EBSWtVwsnzp5NZcK9zxG5SFODzWztbMSSGmx4 M1lWE4Xl3d6kxmDE4GlY2agm46ymY2+QSY16dTAWLk80R9OFTHzv2PjByWD8ZZzH7sTkhjMjwZhQ zu1AMMz9cO6Udkk/u6IdJEVmENlw8JaVxpWijHzURLaIoGns+dxKUSHAxE0TcXJCxeb72DsIz61X B5ON8fxoupC191W36mggZ6uYQCVNhw2x/KK4wdwCdvnR2yDQ39U7PrfRPAhSNqgDLFcY0lSKofyj JiIfZOyVjcEw09O4DAOjCRI4hXi3d5CyUa8OJhvj+dE0SNhPfO9y48PtSP5WP7n6b9gYjMGNe0Ig g61abRDYoO3UiyO3PMY8jUjJ+OG+IskYKRYWKhVFPFYgvsWajx0hfuG/+UoaOVZFWOGTRHdzBykX OvUfpgVHihDq92pa5ET3jgtmMv01STFulRkuxAiKCWU7SbhgPrup8k8hufhbEWpskXLBshmaBjLK 7w96LmCMlYoKEQAjM5OAEZnH1rDB/YwuShxjg2hsXTqYWVGvV9EiE9ub0tO1hQwkaHo8lmxvshjD GreIaveD4WyiZEWvVLCs1qIySKlg/4k1CxO8laKertTRBBkiYT72SkaliOE3H74MGdVEnNypdrd3 kLJRrw52eb6arlgmvnd58Sdh2g4zh1cygJIVbBQpFxzONk88HICQdUdOnx2gVCDdHWUjRDX9oAcm CoytEhFI75k7Dg+qSVbwepgVY8uQNV4a6tDBpCE+r54D8P+JCxI+X2lHiAVIf0YrlUPCQG6lcpWI f8u1SrLAxT5xzLuJQgvAvBrMNLeWEPLVyEUAvmJVcyG1VRIU4m2nrPKCcV+N6iFA91FKQm0NUBLq z5HZKEYNbdtTXF9nAqWY9xJ2OMQ7YCoUI3cdxniZmAxleYjKkf0G8Sw3LEpLEANilOb1iYvJQy49 PIpgbxHHTGgqEZUyXC6Ou2kiSE7B3e0NYnF2nFqh8TsDmCpaBNkx97ofzUQ40uwt8XfkR/dUdrkY jH/M2Jl//Jd74wid0WKd++hzRVI5OnRKkb5SVI6VOpogY5CxVzKKQQbrJl20rSb4ZW/+3d5BSka9 Opi/Yj1fTYuc+N5lBrs5/Dmp37+zAFI2xJIZ3JTE9CbJxofGzR8qR6VM2l99bhCj4ndkXfwulfMJ d02wwRmGm72yUQyl33Ko7Cgy2R7NDZLMOE6tFKE3mKNogRPdZ1yczPjJAna44HRMFgbQMBitEB+Y DxecVDB/mFBWSi7oWHyuSDLDlzvLiytFFV2powkuBqFOUVO5KAaZfOvTqfCjiTA59H+3dxAzg7Yw Xh3MzKjnq2kR2Bh7r9+TTIUjzZVN/ppUaJ6VbWaIpb9jOqBsUFb7e1IqKSUbf/rcIM0MZmNajYwU 9fijJhwXuez53MGI4TcX9CbHqol/YsXqskcLO0jZqFcHk41Gs5oWOfFtbrycCmczILdopdVhiHfY KAb5nI5lpMR/cHagmcCmfZe4HKtl3EHN4HNlww8EhrNgxsmVia1mq2kRFGhvWptKEQMFp5lgv6km xBkOl43aO0jZqFeD0XlsNEfThdQefn6dGywYW9+R+iRlxqnuXMiZKVKRAyJhg20D/qKkovdV2Jh/ q8wgZcM10CMEZdSOj4oWGXMh8UjBAd2yoVNrqsjX6m6NzBikXNSng8nFxHIUXUiju7j4JDP+/qD3 yqTqL7Kg70kxG076/3BBEdTKizM/TMaadnkOcyKTGWwo6TxZMMngi57SczQtAhn0o7lQ/sJsOZMH LtipiTj/qv2yd5CyUa8GS2ZMNEcT7j3G9y4z/Oi7ZDAk6NQWLWAqTPo/N1r6lnCthq2BczA0HvrG YakLaFqwn8Qup0LJC9eI029HD1EHGFuTFYXYEtrzyTSjUUOEnOdeW0mKAKWhDl1C4zMmqgcW7nFd KfG0UtSOJAuvkk+HxtHkpMTBiJVDg9Q/BtbL3+QDKWlg6defv8h0JMz22t0cKX2q1NG0yLEnf0eK /2Anm7P/DsHURJQsst7tHaRc1KuDcV2fy+RGczRdyMT3dUowU/Nnyn52qB02xOhIDNkqmPD4EnLC ZG92naMobAIEaVKwn8TexoFips6ppitVRUmTIokXmZIxmOPSTLAzXRFNhMlNEnd7g0yJEacuoY3l KIKLx+g+4+JkBrs6/A3IS4aqkxnFfAucx7ZJ+sbMYsL8UMrM4DPozQKRZIaDp6w00v1Eih+IIwxK HU2LHHslo1L84w/mBqDsaCJONnfe7R3EzDheHczMmGiOpguZ+D5joyUGwwo3AWWagAHPsCGWzLBr 0+x3it6QwUZn6aHMz2NkQRCDYlTxnXUeM6NSRMx5jFaaKEr+F+BfyOcO6o4QXHBLG2mwipz5ezQ3 SBLjOHVJbSxHE1w8RvcZFyczqL8pHfCOre+Hi8GoIrxjUTL4KEHzgC3aJYMS2ZKDDgJkMoPv9PRl qhRK2RKq9qPpQsaeJK4UnQa3SCYzqok4mTu52ztIM6NeHczMGM+PpgtZe18VGO62dCghGz95J+xU i2UozcGessHcY5sD9mXVOxYTmwZBmhrc89DepVIohS7KWNmIpkEue2VDKTBiyOclEa8m47zbQ9Mg ZaNeHex4ftM0sdztDRsvSwyWOr2MRDY4DzhkDESNwiA9qcFV5SYCw3h3jZgQPJU3SaBU/GLYdslI bzUfNRdSWyVihBjXcylh0qJ6CJEPQK0tPCxQGurPymwUVZNf1CBg79j6OiXYoIfvCH+jDmhGDOSa rocOeT/YLukAxakAAYrRPDVASfjtR8ZWBoUf1XzUXEhs5Z1aIaZ9ekQbyqKHAH8+2BqgJOhPeKmM u0X5ZJhJVTWkwkNcn7USp8XkZlSLTZp6urT//v+UbM7VBnMZ0+IPFuhcfP/pNpSSBpr6xhykPPzJ WOUIETWnspjNJM+raJDLnI9VCowygtIyTWYUETbf4b1bG2RazPp0Sa3fR9EVyUT3GRdtMd1qWSrY 99S34kD0BtyQEya44r4J/6GQTPTWIiaBAaa5ZLOJ3W2xxJg7WyGiegi6AOWGaspDIeaAs5rsSM87 5Qnw24OpAzQf6s/BvrvokChGzf5dS/j31TvhjspU+VR7XtXhS7EYJXguLCId2EjiLDU1klKSwDni RhxE1xybZVXxSBE0By0s/I6mQYhx7JWHkaKS8qzIKmJ7DnNad3MHKRN16mA3x4+ijWStfc0FG0oo KfWNV2C4EGsEfCoM5yjXMhFBkApJBdvVLS8GKRXUAS4pHikCZ0PTXdGFjLlSMVJEkDbZMZ6bJA0S c7ysx9wgZtLx6UAyMW4fPQscY1+uqbvFMoW+O8GItlkxmDvBaJG0yivcHODIU337k+cIWyS+OTLL euJgFkgcuzCmo+lCam/eqZFiGdFbck2LaiJM1kxKfe0dpFzUq8FgbKNZTYtMfO/eERpai0yX53pK GZ2DUaaywzVssNWyPzCfL7TpYHjW55gREWliMKhuGSWWnpCDFy1hq8mX5W4vbcVibhP2lLKZVU1x BU2XvSKTGePVYLIxnq+mRSa+d2ywgcMImHJiA+DkxmC0cm5VNzf42hMvAMGwJN55Pm7PnwJSpGyw VdrO7UjxHxxXUOpoGiRvZezx3CVF0fTDeSzYGE38tE/2BmlujFcr5ebQ2FtNi0x879j4w9FHvHOG iKmqviliiYEpTmcMXBTKMMX5IKTsYL9VZpCyweDNYuZI8R+ctLhrupCxVzZGitIpXxyFjdFEA/hk b5CyMV6tlC9W7K2mIld879jgM9CWmRS8nt0vG4NRKbKwk3LjN/MI+dWZbzNbHAzlOeonkbLBd6Ct 1gbzV+fIRYvYarqQsVc2RoqiyjP+5sZoYmPsk70iKS8YHNSrlXJzaOytpkXW3uv+pOWF+ykdNqRS 47S6bCxmXehwkdzgFEzfARb/zRaGaH2OlwEk3rmyll6+WN4BjmGfanYGnoNYGaKpbAyG4ZxYh43R hOEne4M0N8arlVrPV9Mia+81Gy083WRJ4SMbubyubIjl3Wby2WkDSjqm/fKrf/OKO9n46HPOMYA0 N9y2bWpUiP9gsm6qWhSlWR1kzJWMwehMc2AdMqqIH//J2iB8IPbyaaU2llW0yJp7zcXJjJT0VGye Vi8TQajZ/FS6rTTzfv11OfNjnjDj3pYgf8uCwzOvgbcf7al3mlFljhZoyd9jpxQE4bXOqfjRwTM0 Uegg2rHTv8P2+jKYWRxvV8v8vXaI/T+fHFx3v/ov9oqYBwwFTvSDWZ1khoaXkCEVUpSQ3o2uX0z+ +lyReObYzFXFg9lGcGBBqaPpQmpvaouRorH3CnW5HE10XHp1szeIjK9XxfBqo1lNi0x8X7eYNrRs UJQNLtb7m3WQxZgu5VRU7HJTe8JkT7qvieOzPEb5IdJ0YNtIyyoxnMvpC5WvokVqbl4mzmgoRR3w 0xMIkDGaWJSuprVXZNKiTg0kFxPLKlrkmPuy6nRnJdtdHP8wadqpzcWo1Oin/QnyuQB/ci7ZNUqW F/pYkeGCnSMtqpTiOavPKD+KFhhrvumLUR9wI1WoGEUUu1W05gZpXoxTxZIXE8tq2kiOvS/HZZQj flBGLvh62cmLwajTmhUf3JOJjGtw7o+BCgaQDTtI04Lli1ZUYokyH6sxLaKHuvpua5JiMPh2vtCk GD3M5t+t0f4MUibGp5XaSFbTIo0Ne9NWvpy6cv8kp4TChN/Ja2s5GBFkKMLb922GFB/5uh6NNp8E 7Sa5IKWC8qiN6EhRNlb5UbQAvs1X+cgKhYiTOslbsORiFBHl3RxeDlIuxqmVaix3Tc/RLRdPy0Kn D+WLKAYQCuEimTaYS8DZkUWz7NkZ84LPuJkpjtTy3CAlg13eU0057WD2c1Ch/cdoWmTs+dxisMEX YMPGaEr6qWntFZnmYrwazPZiPF9Ni9QebfXrXrRs0KjwTcGwwVijmbEYk+geQrfB+OtHxmWDWyjL xs8+N0jZYFtB+9uRYrGxUqupCJp4ldVUNkaKTpIPXoaN0URD8WRvEJ9br1ZqPV9Ni6y9r9ngtKnz gkhTF/qa0AM5VrRz4SxUyWDU2TeAs3CIOE6LCFwAlIp/v93KKGHfugk2Q86bGirqPlVbJWIg3M+9 Bbwko4cQb7bwsEBpGH+OzLq8ahY5tl6T0MKKCoRBdFJitnsvQsnuHK1pxydrm/gprByc5RlK8iz2 0Yx+8wJTCIgAFXMFVsUidPZ9ZBHqZc5qYSZK8IS6xfJlrfh3eF4/gvhrje+rYpExQ+SfllVOgVr3 IMoP3F+/SI6hN/Dv/hCJizGrPuXub+PkGX8PB2KuGw7k255vCJFB7E/pQLtA7fSXz8eIaPmYlPH9 WiUEht7LTv/2GcrlunJEjvtVgm8HOHZe/+qnIWAqpcEzm9mJXMYkxVgr+Mc9YtrkPJ+5wQa53OBA /Nx612IxSBmg7m7NNFKUhk+aFkGlmnxuMRYpcj8DNIwmKuone0UysFuvVmo9X02LrL3XbJx3IJej mgrsvTtsDIYqrirOL5CcZ6igjL9RLkuVHZH45tDLtcLBEiVHEGRs9VArFhlr5WIwkp7BJtyvJkh5 sjdIsyI+rQw/XXVXD/aLpJFLbO86iHx9JUywjDlvxWBs5/jnBivyggWWZgEf2mormTu/E7dI84J9 IXaq8FOpHEAwU1bTInBxnluMGJgCkopRRJhP5gaxEl2nVspdoBnrHEULrLWvs4JdGf6W9Dx+6qwt xGC8bLk8D7N8677ZzxE333eHPHmOMQTI5AXbQCzJiiUvOICg9tWUIwk3e82LkWLI4LcK4WIU0WQ8 mRukaTFOrRQNcc1FUfJCgLyY6N7lBeMLf0sqJQrB4WIwGl3Kg/wEHPJr9lMIlAvGAM2BIM2LP86G y8VI5QiCUqtpkbFXLjiooBS1krvOJGM00XM82Ssy7cV4tVJ07k+aFll7X2VGthv4a1JE+zVsM2Mx et98Z4rMYDNDY+dQly2II7EMLNxI9z0bDImB03L2rcWQyuEJta+mRcZe2WDXiFKU7XwSIakxmigj nuwN0tQYr1ZqPV9Ni6y912y0L7GKzhiBQaw3mpQNMbKLiuwnM5e+nnztvm8An4svGx99jpGEiN5R 93vhgGyMFFuwnjQVQWbslQ2lwganvds6jya3fN3s4dMgZWO8WqmNBleqaZG1N2y8HHzxM1IK1Tv2 mQ8bg8EGSzTEyX/wnarGiRS/nuOxPEfkPlc2uG6//W8wc4Pxq9pX0yJ4p6ayMVLEwCcxw+toIk61 3+wN4nPrVTG82mhW0yK1h59f58YHu0zDBvP6ua4WnYOREu6e0C5fpOo7wBxWvfvT5yBfpGzcoST8 tyKriDf8bi6PLcYoyo0YJlmFSKS7tWRGkHIxPq3U+r2KFmHKol5+nRlMFvlr0gyw9DKZMRiNBSv7 4YLhcd8ATmjYqjrwyXO8lj5XLlhDbg8czLaQrzW2jh9Ni9Te9EMjRbOTG28gYzTRyKj9Zm+QsjFe FUtmjOeria7tIb53mdEhhb3lv17swQQkL6q1B0f57PgpE9zqYrZz3K++7WqJQKlg2rNFhULJi+zE uvS427evpLYmKwrZn/JCmRSjxm74spWkCFAaxqEVOj6vngM0Lmy9fj1aeFIZOeC1EU871KZzMKYC ubI3NHh4RxY43Na2lC2DF5JflhqH5+5S1EGVWkWLjDlDWoyCk4Om0LeaKFSf7A1SKurUCm0sVYTq IjC45l5zcboRtmn4O1LE/6KfLxeDUcZTCOEcHPCx5oRJq26SODDKc2SLSHOCfSItKEYKpyq1mhYZ eyVjpMiYfMARMkaTe8Ef7RUJ+evVSrktNF6tpkXW3ms2TmawVcOfE2mnpsrGYMTg5w1kg5HzFNS5 /AU22PzQNPgAmdRg220LCrH86JyiUGo1UWM/2CsbgxED01xJjdFEnE/2BmlqjFcrxfix2lfTIhPf Z+9Jc8NNkhlKOP6edcFiZBcJ0PkDijgPh/uicO6vufHR5wZpbvzz+4c3KcpIl1lvmoqE/dgrG0qB 0a7+y6kk0rqaHPLf7KFpkLIxXq3URrOaFpn4lo2XJQazTD8ymHAqinfe3FjMDUC0DSlIvR3JOPna vBMqjpDyXJG8KUwauWvrSCU3OHtgtqwmWvi7velIBqP+cDuA3cZocqoLr272BpEN+rB6VQyv1vPV tMjEt2y8nOF1O2WGE/RknDHNetBi9J0Mp2OXhq3vALcj1bvffc6qZ6axaLQposKGGN7R11dqNS1S e9ObjBSFk7f/ysZoYgq0muzV4meRPEd1U68Gk42JZjUtMvG9Y4Nb0vw9kfYrPc2NwYjB+/X8Fbgv b8rzfMuH3KCvaR4weul0G5PMfDSpbFTKmjhSq6lI2vjY6xs2UsTg/XqyMZqI88neIM2N8WqlWKap vdW0yMT3jg0+vucMr9PrvbwXhsWS3+wwKRu8AsbuZLqX98IGHwBUZhC9o4ry2PmB7Cf5Ht+DokXG XMkYKepIKq2jCN0uC/yyT1trRSYzxqfBzIzxexUtUnO8l697lNOG0pT5a/KTcXPXZMZgFPOpPqlo OUmYKBlQlQrattaVQUoFG0bshF2OihQeVOooWqDWph8aIYpIb0yR01HE1qwnc4M0L3QqjCmVNmP8 Xk2sZD5E946LM5fq2d5pMgrhEsNpfKOcdQXAILkZ3jzJePwCygSbReyAK4RrtFmZMV49B6DdUE1z ojJUkbm/Ax5GDQuLN1v8DgOUhnFohdwsKi+r5wBr63VCtMRgD4T5b4fGsdkuBi3GG/uLeVpp+DMB MsIpCxxH7o8vkl/W9bMsJrp8HalMuSm1ihYZcyWCGT6lqCL5uAv0rSbX3R/tDVIq6tQKrd9VhN9F YHDNveaiL4cbLDOIoI6f6a2FaMoZL+IbFRx1EN4yw57ZLaqKPlVAz2gYHbtcMvTolakaRIrEM22V hxFiTvwDui89fon8wdYAZaH+FMLBdXnVLNK4vn4r2DHJoSZ8pxjMO7EAlY5niaWAIXvz3T8doOUJ 6sgEQs3AKYWE35E242knGVcB5d7amNAHoFj8ltfs6CCs1ri14Z9JtuPD/Ps6uQoWIQ5svP7d+w64 eTIDBUYD3vtpd7mY9V0mDFjtoXJKVPwyvhVO5uc5ik+Q+OV0flYKiyHFakmlVtMiY68//khRIHKt I7/j0USE9Bl3ewfxuePVwW6er6aNZe29ZuO8BWzt8ddD2qtKyoZYfsK/fE8ndjn82xTn0yP1DqC/ ehC9c1g2o/mRgoQnTUUue2VDKTBiyOUoq8k47/b+w6efipSNenUw2ZhoVtMiE9/mxssy282TGSi4 oMB8c9kYjDxmtBw2+Np92WDC0XlYF/7ynKsMPFc2uOCtBUUwX3DOIHTGcjQtUnvTsI6UexCzscnV PTURJ51o2a+9g5SNejVY2obxfDU9x7dsvCyzNZeBggPaf7nRG4YHg41cd09OcvWwvx4vsUc1ZIOP V1hcDVI22ABu11oM7+glKrWaFqm9aV1GisFDjkzS3UQRYXK88DKX1AhSMurUJbWOH0ULrLV5Tz7J DL4q6q/pFBubqZsZg9G854NgFmy5DI4ouaCpHQi/mM8VmcyYqSll4IsyszKrZ5Gx5mLXYpTKKSoZ 20EyEVJRqIZxVtQcpDzUoYP5iug0L/dRU+CK7F1O/P2ZAQLlOUtt2VxDK12MAt3rK22n2HHeDKCY 951x2JPnqBZFmhNIt2AQS4wfT5qoqR7s+dxiFIzsoAZxoKUmwuTyRjUde4PE3vHqYJIxnq+mRRof b+TXrWcXy6nPnaNvUhRiYNBhoWcfZsxAP1/fug2XOHyqVLAxxE52MIO0GLMbHT0HGFslApn8lHyV 0RRwSKcaIqRnLe+xdQCfOg4dTBrqM1UmemDvAGvrCxL8lLGfkLSiY7gbFgZLF0Kf6OiYOjXXvxEh w926xjJRf/4g5YFdJHa7K0WdjNRdU5HLXplQCozNiQ7npaKavlMV3u3RTgxSLurVwSjFTjRH04VM fJsSrxoKpNm04S/JmrabMcmJC2NahTXksOHNIobJITbpYRTmUeBFSgZ7RuxnVopSucpX0SJjrmQo lQbx76/Zx1BNhPn7bo9MGaRkxKkDXX6TKSr6z4Wc6JaLVx0IcbOZxF8T59jDPlyIESgheCmmycZ3 zJsHXFtSMji00ywIonOOxLKYCBmVokV90lTkslcylMorQVth3h1Nxnm3l8wI4nPHq5XaaGQjmhY5 8S0bn2TGB7dexbvvTPMPG4NZrGefIOVbb2+jbkQqTdmfPlck7YXLZ6xY2Dwohb9UHpVaTYs4NECT UR2MGLjdwqb6aCJOLh8r+7V3kLJRrwZLJtTz1QQbj/EtG5/kBlMhLfo41ZArTNAwGOUQF0jmV+AW NDOBoSRSII7I8twgeuegLIuJxeCMUrRSq2kRiiE1lY1KEQOnLYz9aCJO7pHxHM6xN0ieQzxeHcz3 op6vpkVokRLfOzZY//L3RNpjbW01BiMGD82Zk3yC0llr4lRK7/gAZfMgSNmgL7RxWSmGapVaTUXQ NPbKRqWMoXfZONZTk3E+2juIzx2vDiYb9Xw1LXLZ+7pH4W5/9xaiyFFY2RAjKkoiPmMRu3x1wQyi 7FCqw6PIDFI2yHM7yZWinH7StMjYKxuVIgaOtjc3quk7pD/aO0jZqFcHk416vpoGueJ7lxt/vT5R NnI7SdkYjKVu7jBJl83EbTMht4XIxvc+58QcL1HZOIuPI+RCYZRXESQWiXMxVzIqRQicXDTxXP6i zSbK3HJyWTtIudAn2tUjReT1exUtcqJ7xwWXwbXk4efJjCYaxOCHko3awYTkDjlTxZnqTO6yvFCR QUoFg7AWFiNFQf2oaAHKQxWVigqR7Xw3t1SoKFywLNK5ypojbpFpM+rUwcyL+r2aFoEL7X1dfH7Q rCcpmD0sEQWo2dgUIAt/cxkcLwydf90yP/rnZAObRqxVBvPHb3ddFf2LXlwLDd6elAKSCWJyzoGe zxMTRyltoehErj+bAvViRer26KA2wunk2tj4qllgoo/F/MT8nYmCNAuLUZwxiy2rXKsuMdQXCunT rz42iG45Vsq64UpRL1YqihApEN9irfFXCPc5ZWKuHUXEx5mOu7mD8Nw4lRehUoTdWFbTIvz4Y+9r LtidYb1IvN5mZKNAleUxJ18EOgNyALuc5QwXuUpfLn5XhClSHysX4PYqgxk6Zw7uii5kzJWMShEC i38lo5oIPR8BuOwdpGTEqQNdfq+iK5Kae/sisEXDvEWa+9yGi8Hgwg1VcsHnUXzfCVMpneNctc8N UjLYJGKnslKsFVbqaCLtRcq9mnzuSBnDz5QuLLhFk3E+2jtIyahXB5ONer6aFjnxfd1E+uPzkUTH gNwUP2QIERQZYgsnGXzivm/+r+x8cXgWEXcFdL8MfrP5JFQA5XVnKfGuhqiLjK0SMRBNAZ9JzytS PYSYD7fDu7byNgiUhvpzZIi5LsNC1VxI4/qMhEzyIsx3OPgZcexPL+xYCBL4YRgiQgL1Q3//XqDB wDBP8eMDpGB2XMZ0hflQGerlyhw1kFBkbJWEgeh4/nyzez56CPBbLus4tgYoCfXnyKzLkKAaXG5c +XGN6zMSMreL8B8zQBLY1ZHZzMVggaa7Qzo+a++vyjyYUvYczMr3hw9iRC6NZUFupdxfEamjCSKK HHtlYjCqEW85uTR9Z4fXo71B5J3fpl6t1HoOGdV0IRPfZ2yclPjHjuGwwZ72w0Yx3iz6SAaH2OXI b9//70rJxo88Bz8ikxUcX2jDOlKePsgQ5GiCjSKwEXtlYzAPk9iDstQWRYRJn3o3d5CSUacOZh5M MFGUxBBIYoy1r3oPij1ueysXzD2mkTiYXHADcbpSJqCbB+yKqHN8CdhS1HlJEGNygNSWZIQoDStU RYgUqXPnsSNFsnMB76XHGB+NHaRM6BKajtTHcfvouQDHJfHxayaYaqMC4VdiZibzumgQI0oHR0xj mhUMD9sKcEGA00sqjwz2RcpEFnNkZ6T08kETVBQ59nxuMaicEbiK+HkgCeXkyTE3yLSZdepgfCFi HIeLKLqAie7/fHW0GOkc8dA11xssKQaCiSw4mHh8vKEZwEd16lrPkhO0TxmQy1RZMhzMENmoGUar 5wDHVmlAhl8C56eQqhbjU7MsxNQBeGj9Odi6XDVk8VNUnzUSbTIZILsEbTpQBnUus1i6+98/nDPB KFcwmSDuUUFKzzhBbYCDlAUmU+1eVipHDixBaGWiCRo4lnCzVx4GY/TQa98cyamJILle6LKX9yBI qahXK0XoE83RdCET37LxcsaKPs9tdrLB6ffDRjFi4C6EsjFRctq+vrFSY2lK/yNSLtgH0tJhpOgH o3v10JAUOdZ8rliSgoyymXAopyai5OuHl70LKRfDV2WIeyKhH0NP0uIxtmXi5WyVVZHjBTof/v8w IZZfk9sQmCEgLzix0yzgxoT6RrfSHAhSLniD7GLozCtlYfSgibhFLnvlYjBqiVOvVBNcMJq57CUv gpSLerVSsDHRHE0XMvEtG5/kBcsd/gZUcn5pry1FMX7PH2wPDBt8zH5GEUjN2Cz7E6nufK5sUJ/b yQzmb87RCXPlaFpk7E0XXCnec66PtHVwnKcm4qRAMDMZD9TPQcpGvRqpZMJEczTBxmN8y8YnufGX Gj5s/Pn2O/va0VCMGDgryiwBucFUSDOB3SL1jk0mzYMgZYP9IWVDrIVjpY4m2GA947IXFhfzKMls YKgm4mRAXPZrbxCfW68OZnM50RxNF8J4SU3v2GC/lr8CP4dXKjQ3BmNdyUsUtMvtG445qaKV8rfi 8xiNPEjZYDHIbvdIETsHDtR+NA1Cxo89n1spmg/WA5Mb1fSdUf2jvYOUjXp1MNkYz4+mC1l7X5UX zjdRS+kd1xWVjWKJgeuDGDfDBl8bbSYwiV82aOcbeZCy8Zdve4SzShEnpyda0lbThYy9sjFSdMI7 AIsm4mQT2t3eQcpGvToY066JBs+PpkGu+L7ODdYymaRJbuSSOnNjMUZqXHjH+0ht6xDU95+vM5ot DtRmN2CQssH28hZplSJ27hZoUVtNF0KnoaayMVJUXb31y5GYmoiTTzfd7Q2SAcnx6pJaz4+mK5a1 9zo3OiBxWtZBBLmRre5lYzB3J7jNknz7xvHLsMExrnrHUkgjF0lr6LysVRkVa4QIndGdyo+iCxlz JWOkKLGIP4VmFBFlttHLfa0dxMw4Ph3MX3H8rqL8riLJjET3LjNyohYquEy3yx/M19t7EAAX3JYJ zjl35wL7x8pE7+ShOhUwIkdrWUwczMYgvdDqWaC2pjOpDGM297JdaoiQHbw3WwcoDTqUxFGI1D0+ Hz0LTFzvSGDp2mIQl/92WzsKBmME4A3sks/dBf35fyMFkh/dsoOyUqQ8MPax8y3GL5ufvaVmNV1I 7eW5laLUZOE0VFQTkf/Qq8veID63Xh1MLsbzo+lCJr53bLCHxATgvebgQToSN8gE42f2TJB2Pzg7 b5gc4ykZfJ+5gQcpGWwise89UlaRj4oGIV/HnM+tFPWVh4vMi2oiTD6jcrd3kJIRpw4kF+P3UXQh a+51U9FhiVso+bakzjHw6Gx/sbwgzjxIBZ1B04A9H6aKgzXrYOIOUi7QYM87mFGy/cJak0ZERRcw 1krFCFFO9Ko13kvP6jONzSiZjwxc5gaZvKhTB5MLY+G5o+lCau9NgcGKtzfyhQs/eNZmczBCYJAA FXzXvknAW1Dfvucpw/apUsHo2m53MCPn8IG6q+f6G8/yVN6jkcF/L37o4E89xqiUTNTaQZoU9Wmw tBbj9dEEE4+xvXtD2K+hXQpXdlC4fxcNg1GaZ8AEFyyGNQmyz0Lv2HnduIOUC6Tb6YolTg4fqP1o go1Bam/ai8EorNxnIRvVRJzcFYmmtTfI5EW9OtjN86PpimXi+5oN9oWw4ydsMELsKsBilK5MPCQz oKWz46xhdUqVjRdTUoqUDTbetLCqFLFz2ECpo2mQvJOx53Mr5SkJWhzZGE2ekni0N0hzY7xaqfX8 aPpYhMK8fn7VYriXkqPh5gbl82FDrDE46+h7wpVH/saU00r5W3Ev0kQuoncMo+jC0XWkiJNTCHdN FzL2ysZIEQODtbsm4nyyN0jZGK9WqtHg52p6ju9dbvAlFntqxtUsX/uisMGsEL8Z/WFeR851NBE4 KB0ufleEHxGgVFBetayKDHH/u6tJDoiEeW2ViBGi3qYRCBGjh/1+j7YKaIv6sv4EwsF1edUs0rje kZD5UElgYi2tBZqdIjWr+daPayIUckz+JiNyEpmEyHSsP75ASeCGt0uEHOd+t5Z61TJI+MZUOKgQ EFU256/DQdUQhA3HZSlAiu715sisw6tlkUSFqddvRUtuEsoxOF2P232aBkVo+b87sQsBXCDTUcRv ZHQrVyISv383flbnnaks5EueoT0sVskCtWMKHBHqajtpXyqV4AuZ/IPKe+3k79hZVwqZAHV2lRzg 2CH2z+6+4mE2Y2iQwZs53OgHc7joQJnwOT3dH5cPCPpaeEYzz9Hh+1wZYD+Itdlgxse/qf1oupDa m2HKSFFW/2SmTBpGk2dl0H6zN4j21qtieRfG89X0HN9nmdCKin2T3kQnG0ytZm87DA1G3edXxLTL d9ATJiVLG8xcYGfgQUoGO0JaP4llEMUxBJ87iiBjkJqTxAujsHZrG1yMIpr6uzkybZByUZ9WaN0+ ehZYY1+/E+zQ6EsBpScvBmM48MvNujDB2Nrf15UET4DQd3D/YcMWyfvKoIvJppsUQXIGQamj6UJc 3EBTqRgpdul1hDF6KGeerA1SJsanlXIDaKypBx/7N/ytrddMNCeo6Vm3wVcHsbQBviGLOTDi3nGZ 4KP1TYqfSsnEzz4HNyL65pjMdUT5ihRxc3JC7UfThYy9MjFSREDVdlNEkE/mBuFwwOXUSq3jR9EC a+01F20peYqmOlxwacjhYjDXd+gkNMsWo+YAlyeUi48+Z4PAbFW5+Od1e3JRKSLnEEJry2q6EIbO aioXI4Xr6aGZEKoi6uQnc4Nobp1aqQ0mihApkLwYa6+5OHnB+6n/+OZFLc2LweCC9eG0SLy0zX2O DJYLro+YuEV0ziGZK4lyESkiZ6OpUkfThYy9cjFShJB5oVVElE/mBjEv0kjo1EqlkdDcUbTAWnvN xcmLn7/YRS0XHxyRGi4GY1xPU0WUNNPcR46Uy0wu3fGO/O5zRZIXjJI4z3STInIOIqj9aLqQsVcu RoqzRHxlhjBXEy/nk71B5H69Wik3fsbealpk7b1m42QG8z5GwGKUe+ebGWLkllNw1JPa5aRs8+Cb UrLBF+EaeZBmBjcutOcdKeruJ01FLns+txgx5FNusDGaaLTv9vBpkLIxXq0UZUW9Wk2LTHzbp75c FnK7JA26bHB6totkizHP4aqFbHBcyd/YpSbvwYINbmzwuSLznmRZdBCj5FspyqyeRcaa6b4YXeAH 4mZG9NCR69HN1iBlYjxaqXoNX6tn4zjWPqkzT15wDVyZ4PQOTOAbRXsxZu25NDNMcB2fWcBwSim9 Y3mkTARpXrBnpJWVWKJk7aRcjCYmMB7sNS8GI4ZvLp3CxWiinKqmtVckzFNQ16vBbp6vpo2FhTU1 bV68XCBzJ+WQwSr5vCTn18zdhmZFjjaGC4SIkuHPJAWfs8wqMgFQY7auCuYP7s9sVoyeA+DYeSpQ fks+DO3bt2qIUKHL1gBNinFohU4YuOodj/LyENc7EhhytPzmJqeumeaqiW55YLtvU+LPVEvc7GTb 4QgojzHvINKMYA9Ja6qRonKslDNofChYYgZxlm2eW4xhg7vHZGI0MdR4slck7TQ/e5xaofoN7ato kTX3utlsJ4KQR97tg7j9uFwsxlKmuwxNCRYQmvi/lZKMn32OMTvIdCIsubesGilmvCu1mhYZe309 Rop850pJKHOgF01w/mRvkObFeLVS6/lqWmTtvWajjQWpypdIyobXSNqJFEsnkqv1ZYOvfDX3Scyy wVGupkGQpgb3FLWwGilKcqTumopc9sqGUvlFeaGaGqOJOO/20DRI2RivVmqjgY1qWgQ21LTvyctO hPaVVeZ49+fPrJguxuSSp+t99ejqGjvjRxOetrvPMQj3Ob1znSobIAfzteD8gdpX0yJMLampbIwU PTd7+++aWAJV+83eID63XhXDq/V8NS1SexcbL5tOt1RmGEGLy39l0mYxJn387Jl2v1NPEBWD71wp CBu/+9wgZYPV33a8SqWR4ASC2bKaqLDv9sLiYtZPmSRw6BdNWR4v+2OvSJ5brwaTjYlmNS0y8W1u fMLGL+/gtt349ZPFqb4pgzH/w1XBYYPFGq4vgI0fSvlb/e1zg5SNn13pXCmKyEqtpiJoGns+t1KU 1lyoBOLwL8uA1DFP9gZpboxXK0W9XXuraZG193W7wbkkp3ip0piVGDbEYIhJl3x/gtzI5hLZ+K6U bEBP8yBI2eCiqlmtrRRF5JOmRcZe2RgplWZXCwty1eQmmkd7g5SN8WqlGJXW3moqcsX3LjdS+ctG NgY1NwZjIoQFXjIeYj26KRvZGmS7Stq0CAtSNkivuxBVZIWqKGkgEudirmSMFPnGTc1JjSpi9ONG pJu1QcqFPqFqpbBSc6tokYnuHRdc4uivydjO73CXC7FkBm2F86++chmEMNjK17rhArTlVJBywe6R 1igj5QGFR02LjL2SMVIMtnKlJO/JaGKh8sneIGVjvFopUrn2VlORK753bKRXT3PUZWR3XYYJPkle Jhh5yZf7Zl22gQkv+Dh/lwem9Ox/V4Zy+kHL/F07U5tEYhpjW4rR4XbEu535u/HrC3ZWRm/xbbWM 9xPP17HTlP3J4IGuyVMsZkIxNNLccCImmUA5ndByIMbwOX/rG8Boy8dKAHtF7HIHSx1dqSri9aKy fjDnc4vRQ+c6TWgYTbScar/ZKxLyqK7j1AptLKtokYnuMy5addLJeUjNqpVZqy4cL8Zw3+8X+Faw 96Khcxbc7HBAlOcYZIDEOQdO2QFZLGFy2ELtq4ni+sFeyRiMUQWiULaa2Kf8ZG+QZsV4tVLr+Wpa ZOL7jI1Wnaypco1w2PhrYdnMEEtmeIu+HTbHPZv0nGotGWz6bRYE0TlHallFZPRRqRxPUIr+3dvb SAKOJ9yslYvBKOt/ZRFsFVFgPJkrEu7XqZXaWFbTIgwi1PQZFycz2K/hC+BIh/W5cjEYBRsVY0o5 ukuEGFYoZGLwrfa+NyJxjjqYG3TvUhShlTqKFhhrcrGYz7MVWFJHkf/0aG4Qn0NnnVopvYtTq2mR tfdVVcFvyL0V4YLN9l04Lpa8YOuRUwn8rj/StlFHKyUZf/vcIE0MjuXb264UPxlSd01FLntlQykw Nxx6KwJsjKYM0i57aBqkbIxXK7XRrKZFXDvD882Ml6MRXm9vhjYzOAd22BiMpsIZO9ngg/VTRyPF r+doLc/5UvJc2fj4lZXFwfzVIa/17GhapPamrRkpm4q0SLzy1WQTk1Zr7Q1SNsarYni10aymRSa+ ZeNl/U3L73dRwgY7fvqaDOTgLjss6Ax++RFGfvTcWEyLyq7hZoFAqeBcTQuNytDII8OvuWqKHFt5 aiHcZ4SetBg9hJgdT2srwDQX48+R2ShWzSKwEA9fvyCnsWD3hQUjnSTTEyVhIGeEMt1N7fqHz6pK Am8/ntpb56kA01ScgqIidvgRWS2L1JTMLUQn6g42Xw3U5D1wAfZmKdNcIW69KWQijMOrZZGYwr/X FJy+gy8Z+PNRMPJ6loKBqCq9udOXgo9VWCFTreUdpjjsUwWaB+wQsbM9MhR9lVk1i8TWvBAjREXp paRyMHqoRB9tFejroD/YOjKUvdjCwVVTZOP6mgTHU7rOAILp65AQKEn8y19SDv59Y0Oa8XHYxabC sVMDDlASWLNs5TBCDooUYrSCHl7cA4ytJkJl5IcNr3IwajyF8mBrgJKgQ8mWETphrJ4DrK2vMoHB M2Ns9PELsBFsWBCL+5yEdvbEnLd5tGnI+ofDsIgMomsOxLLHcaVMzkdFi4y5EjFSNg9njJVFSYbq mIO+NVck78TxaSC62ISCl0fPAsfYl4fnMtHir8goxXsXfCkWI/VxMkwwZdPfnxXv+sbEztSNPlcq 4KxjC6V4jqaqUqtpkdqbF2OkCCGMjhrWCqsmsywaG0RjNNN1aaXqdl4M1WwUE9m714L1Tcsk2iU/ f1cexPJicI+McyYWBBQ1pgRDECNkOqzPDVIeWC9u4oyU3fijpiKXPZ9bjOLCr/SZXKOJyay7PXwa pFSMVyvFxGG9Wk2LTHzLxsv6gX1VbIKMdxy5OmwMRmfFCYuwQXhl4wdSydi/fY4Cz+fKBjcH27EO ZpyEp/bVtAhjJDWVjZEihlwfCRujiclPtcv+2BukbIxXxfBqo1lNi9Qefn7dWvzmvGTY4MRcPg2D zsEctWS2jno1d8ARJwMiM9AxWZ4bpGxwNZE9bDG8o+Ct1GpaxDESmsrGSDEk+Zk9PXT61cTEOP8W 9sdekTy3Xg0mG+P5alqk9t70oqzo8AWQsME8TNfFFqOc9vYI3xTu6XD8z7An362nU+HTss2DIGWD Tdl2bitFxVyp1VQETfyHmsrGSFEx860D7LkQF00smjzZG8Tn1quVWs9X0yJr7+vc+MdNgfGOicSu gri5MuMHnOoIkUott8CR8dy+47vjCCgyg5QNbmnv/NdIkZ5PmhbxlUFT2Rgp5yzyXjBkqiYWEZ/s DVI2xquVWs9XU5Ervq/fFHZLutnH3yo/tq3oYoxcTAl/hRzo8v1PKslGdhstUjY6b7NCtFxVXkWo LhLnYq5kjBQh8BlBpBzi2bKzddHUulkbpFzoE6pWav1eRYtMdO+4wMHWaZwe7Ky/2ykzgPDlzfCc 2i13wMEFV/P45jgqi8wg5YKtI9YrK0Wn9qRpEbpYNZWMkXINPW0SjUw1ufb+aG+QsjFerdR6vpqK 4NPa++o9Yd9QZmTs/2kgmhiF4CJfWqBlYsdAJ6f5ykG58B6qhC2gaw6VGEQtZhb8jdDqOUBtTWdS Gad8U9GsGiJ8tDWAtupQUkchUveEsXoOMHG9Swl2bFhGUbqzqyIjUbdYBmOgwGx7rPJmW30zae5n +UyJ35UZpDyws8S0LpaOhF0zal9N1NZ3e5MSg1FBMsMdRqvpG9uY9OqyN0ieo76OVweTi/F8NS0y 8b1jg0GPvyTSXqnZnBiMsvpfpo8oZ7koLmHm3k3I+NPHXFPojYgkBZtGWlEES1leqVXkwOCPI84x J4nFcIEQvAHZ9Komw3y0d5AmRpw6kFyM36tokTX39evBurelJuMZttwOF2L5PWlVmxmsqneIwmxb M4O16WZBkGYGk11OVhFLpSgjnzQtMvZKxki5Ry9H4Rn1RRNx0oTe7R2kZNSrg8nGeL6ailzxvcsM dm74czJt4Eb8ZsZgvN3sviBLKeJ6jRuTFErZsXAwuWkQpGyw/duyY6WcKojUalpk7JWNkaKoJi2S GtVEnKyl3+0NkpbmeHVJIV57q2mRtfc6NzpzI/kZTjDLxq2Jw8ZgdIFcAYR3FHH82skWvuJT77jZ xrKVNh8k3tF/fJydg5Wi1azUalpk7JWNkaKMZIcS7B9NxPnr0d5Bmhv16mDmxni+mhZZe6/Z6CQO 76Uf/zM3mJ/vpHcxsosYuEsubPDZejOBzj+XkJIbdDI+N0hzgwW9zomNlLOrfMrwpqnIZa9sKAVG GZkLRmGjmoiTu098V2vvPz8OUjbq1cFkY6JZTYtMfPumvBqo/XIXZYYTDm1oI3hTLgw2QkVK8gSu iK5ZSBMkVIiUCgrpzo0FS0Vd1VVDLftgqTwMhvekeLKiaozwbuy/4eEYY+Ng7phbqcvno+lCJrLl 4dUUL9Lf2Q2VX+kbNwQOD2L5LfOVL16RbNsy7j8KyYSbqJIUQcpEl8hWyP+IELW3Xyej1xiA1Qv1 lIrBPA2R0/iO/shAYuTToXdjB3Hn53HpYP56E8lRdCFr7vX7kdYCabZV+0MyvPM0RzNiMAZ4mYCm ks31cDDB9zDqHPML/f1F0lgwRP02c4UjRcJU6ihaYKyVihFiWMHsGlEeRUTJXvO7uYP43HHqYHIx fh9NF7L2XnORtgJpdpiUig8WBktFIQKYgo6tmc0AZvLrGhtAkhIB9MzBGj+hiTNClI8Rihok8jep hltq8aEjghO5dne1GN+DqQOUBP1Js1GhDeLoWWBtfU2BJxbrmRO/5SCnGG3kvBRL5n/2Ojfiy7c7 eDG4K7y1ZJCywD4Ra5SVokut1FG0AL6pqDyMEAFwqUzoVFFeDZpQ152OOZk4zx2nDmY6jN9H04XU Hln7NRfs0fBnZEmF9ezhYjAmXfhmS8hg0iVRctOSfYlDtDxGxyhSLtgm0v5lpCy2IrWKioT6mCsZ I0Vq/00v7XBPTYTJPVJ3ewfxOZpTnTqQXIzfR9GFTHTvuGAN0B/TkX7XS9EgBj/MLdgZaJbv2TcN fiqVIVFlnIzLOqSvB1dqWXAOZu5zQOFB0yJjr2SMlPvf0zMdTd9oFx/tHaRk1KuDycZ4fjQNcsX3 jg02bfhz4h0nQfqSDAQZ39iVJBls1m4rwNVN4YJSqEkgoGuO1LKKaEnZ8QR7tZWpGkQoqkHq2TxV CCHcZ/n8pscQH2wNUBrqT6H/yMK4fNRcSON6RwJXOVovMsRn/FASBmLGI/d/QgJ9YZtH2sSQ8KNP MaQAKAl/GDuhqJAhc8LCBFk1RY6taScUCgkc0i4J1UOA/ovJV1sF0kcdf44MIY/LlEZRcyGJC1uv 24h2n64mZLjACJ8tu5KwEGN85xjNBJYG+8N/z75e1hr6FENvtseUBE5LtGOpjFM30bxqFqktnzoQ L/o/ls/Td1YPAVJ5h4TaGqCZUH+OzLo8ajaIZJ1xfUZC+023TGaUwBKctxCXhcFYzOIOkaQb1xt2 fMGwy+aCsU+fY1wtom+MyfxKockwUq4TR/vRRNSDjL0yUYwQ+HJC28VqIkqqzbu9QeSdX7terRRk 1PPVtAiZnvg+Y+OkxE9WYZKrbMA8bAyWJU5mGbHLd+sdE7igmcUbR2R5rkiyglUr5rFuQlSJFTqK IGMQ/UZRyShGCH/4plV6zCgiSm74KBe1dpByUZ8OZi7X71GUxBBJYoy516/HyQyGN60uOWF7uBiM mQ72zSQzcpDKMFm7rnc52LSIQc3umsVsPbNPJ1M00XQhzKuoyeeOFDFwEvemiDApte7mDkKpzQ9U pw4mGXX8KFqAvBhrr7k4eeHkvs0li3tZI0NBIWZVmMRNVnDkxdxhdra3Nzv2aX8gMEnBLhI7lhWi YI7Q0UPMBcZWaQiE845hpVMt1Ag0Rg+mDtCMqD8Hk4S4fNScvycq/HtNwUkH9otYH+AXk0fTUAyG 877nMs+9j86EE59SGfb0uUGaDuwjac0wUpTdu1kgmqBBJMkae+WhUkaQsb+DOhUZ46O5g5SJOnUw majjR9ECJ7p3XLBRw8KRHogMHi7E4IeJnswvwsUvyj6pgJL2JVzs1x8/SKlgp8hsnq0UCySPiqBi kDFXKooRgdlHSqgnTLAc5AKtYyytEbWI79Nx6UASUa9Hzf59IoOH//yv//7v//7fl0NxSvGOjShH uFN4eBiMFX72yMUoX8EzcxhUIxXXfpwxlch0HvzcbS+Dpdas1NEEEZxHkNLayxt1MCLgu6dtI6qJ KCmxSnztHaRU1KvBaDVONKtpkbGHn1+9IW5wzKtMicqidaqqxZjw+uu+An4C7jJvErBp15/O0c+0 E0GaFn9/zP5CsRQHnD6wNTmaYGOQ2psGsxgx8H2EslFNxMmd6Wo69gaZxKhXB6NNnGhW0yKwkfje scG56tZ4HKbN2jEaBiMGCqCwwQvk2IRhpVJ6x6nt5kGQssFceDvckcppA6WOJtjgRIKFJ96pyeeO lDF4i4yvSTUZ56O9gzQ36tXBZKOer6ZFLntf5wY/Z9ccqeQPG2KJgWZSNmha3SAkG1w4UzZ+VWYQ o8J1P3V2k8ppg7smIudEgoiTjWgqG8WIgcGMbBxN36Dl0d5BZON4dTDZqOeraZCwP/a+ZuOnX+bU O86AdhDmnqZgJh7fdtUu/2smcMLUEZZVQGQYeIuUDT7W3G53pDx98KAJNgYZe2WjGDF84+K4sFFN xEkZerc3iO3NerVSxF7PV9MiJ77P3pQWF3QYnvqVDc6VHjYGo1HiIuLY5Wv1fQf4fEW9YyKzkYuk NXTYlAVFxqCVoiiv1NHEOzDI2JONg5HxTJzaRh9NxMkXG+/2DlI26tXBzI16vpoWOfF9xkbrDEqJ fxlE4J2rOQ5IipFdxOD6ir8C3703g4izN54zduhzgxiVY7V2PCNETV6hKiLMRcZcyaiUIfDdrkuR UT5aO4hV5/HpYMfvdMFVNJH0t1bTZ1yczPAkp4nBvtRDRaF03GWCI8FmAWNNhexhvuWpApMWbBuR iSNE3G1lj54FxlZ5iAw08B1EWyNHIO6lJEJS4mbrAE2JOnQwaYjPq+cAExceft1YsF1D3xl2+eGL JsRgFHKewZF8buHpFgxu+WvXwnU4jTlIE4IdIy3SKkXYzFQqdTQNkqYs9spEpQiBfTSlopqI8uPR 3kHKRb06mFzU89W0yInvHRts12jpyanAw4ZYYmBMxLQrbJCn+bntFJMTPyrCbyFSMtgx0oXoShF6 vgwkGVV0ISTVee5IEQLn7mwZYCWaCJPp3ru9g5SMOHUguajfq2gRuNDcm8xgiukj1T8zKJxOSGYc jBC4X7Nc/GVDM78nY1alfEG42csycpCSQbFcyipF6ByiUOpoGiRTs7Hnc0fKGDhEEDKqiTg5+3y3 d5CSUa8OxsnSE83RdCET39eZQbvH197i3V++nDlsiCWGn97JZWaQrX0DuOTON4eGrzK0siJl4x8X VIezShEnpyjumhah24u9slEp2OB4XtmoJuKkV7/bO0jZqFcHk416vpoGueJ7xwbbNmg78Y4jlFlJ R0MxflHWeNhXQL5xVr6ZwO0SZouDtgxD7NVBjMqRVLYzDuavzukEW+ajaZFjr2xUihjotciE1fTN iuTB3iC21uvVJXU8X00by2XvdRvajoTq0fZaNkiDw0YxYuCzR8y6Yfd7Jyyo75SSjd95jjdFJH0J tbS9gW9PpYg9PY9sVNMix17ZqBQxsIlgdlhFE3Gyt/du7yBlo14dzNyYaI6mC5n43uXGr2+MvGTj p5cE2qMwNgVLq/83p1Rgg2/M+BsnSXx3HLr53EGMysFUFhWPFI0hy1p3TYNc9nyuUmBJJd6L1USc TE5c9sj4QcpGvTrY8fymaWK525vceLmvwOkJ6l/ZYBL9sFGMdthhCFzwSXuLT5oPZeSCSevOJQYp F8zkzXp0pIiST0O3tM1w5vx9bJUJZYiSJsYzBTJRPcTItxQua/g4SJmoTweTCb2+aRrkiu1dXnBj N/7jHWUOi4a0EMxaBYOJ38QmFb/YdxYqONjma+P4yMcgJ0ip4J7RC0op6kVgVjHRAzXcsnkzVioG s1fxhYCK6CFGvih7NzbItBd16ZI6XsNpFW0cJ7bPmDjtBV/TLhHeVd8XJBD+exBAHrjGxZ+WqJWR h5MjAdzzYOObRcUjRNhnXBo152/dUktZiIi+czomJKgl7wF3MtxMGfA8tf4czJ+uQRw9C9TW2+KC gRCFIZ4x53Q4KAYJHMxsNvz5PaeV6EjqGq1iC8ogzQZ2jDQdKkXYtMNKHU2DEOTYKxOVggouMi8V 1USUrKvf7R2E5+CiXh1MLiaao+lC1t7rDqSjMndPckzayUQ+qFo2ijUGz5Bp1496mRJsb61zP3ns QkoGe0ZmOTpShM4xBaUgI4oWOeZKRqUg48PTaGZXNREmjt3tHaRkxKkDkaOJhV/2KLqQRvcmM1j4 48ptnmdcxNA4b8fBCIGyoVywWa/z19wlUef47l0LyiA651gty4tU3JEidD9dGzKqaZBwH3slo1LE wNdm2kRWE3Fys8fd3kFKRr06mGxMNKtpkYnvs6biZAbTvJadLlp2lcgFcXbr+nsyKc/4mWTjm3jN g5zOpq34XRHXQ7PWAxXM1jZ7gDIDwdmKuxqYKDK2SsRAVNsEEkqrhxC5FBygtvCwgDQcf1bmuAyf o2aRxvUZCW0v3Y/ODJ4keF+H7eWBICHTShpFdX//fC4ExzjN04gB0pS7cpWlRSYTJsfZEKzMUUPW FxlbkrAQRXYutIfM6vnGEqN61tYAJaH+rMxxGRJUs0HkxzWuz0g4mcC0pL0GjZtX2ZeFYrhPpe1S MsUbnyIhINY8c+E9rv3NcwfRN14jb/W9pIiR7f+WIEfTIsdemagUIbAF3BCOJqIEaD7U3iDh/Xh1 SbkZNNGspkUmvs/YOCnBHeIWmUh/MNtaNooZAxvq8Y5pb6YPGqdS/lDcZzGRg0xWfHAq4CZF7Byo UOpoWuTYKxuVkg2uEoB9B3tqMs5HewdpZtSrg5nNE81qWmTi+4yNkxt/PK4kG0zcHTbESC7ace/b lA2+X98yk/sg24pysKeRB2lucLVA661KETvflrtrGuSyVzaUAmO40cyoHqLkCObd2iCTGfXpknJP 6P8j7EyyJGly5LyvU9QRMjPmY/AMveaC99/wkwFQc28Pd77XrEj5YRjEYarQ0a569t+N7TcmJi/Y /1Qm3lmhSV4Ugwk22JsJtmolC2g44xsbp8oESPOCuQqnRYQIm0kPKR9FB8E5KQoRlYIIbdokK6xH Id4aGyRJEZcGU1LUbevB6wBuLmrsWU2hXZMaLpAT7H/J6uFi7H7gupsU4Hxq1JnOhRV6ZzQw8zCD MYiQ5AS/YOqpSBE4iwCpr6PpILUXKirFUhHTIqIiioiSN+ZqbhBN9VJf26nBuB5xghlFC6y1x1wk K9hNyVZyUcE2jqEikALR4R2M+v4KvfMcB4pr2oA+QFPCF08M5s5BF08oJaIHHgLUVmgIxEDTV9xD g9TwS+pcyNXWAMmIODSYWIjPq2eA2MLDxySkkdAuSo0WYOGPbrfXq7EYIyGOCYFQvv3LSho+Z3DK Hoj89kbkGmMy/DY1kSJqDiNI+Sgq4my1uRBRKQaavuB+NREkI/2rvUFkr04NdPF7FW0kje4VF+ze 0A/AqPibsXO4EOYQuFhDM4uUany82mEyXoxzfCskOWEkZLByrD0GtP6WInTOIlw1HaT2QkalqK+1 xiBao4k46V6v9gYJGfFqMLFRz1dTkBPfKzbYwcHPKTZYOS4bxegL/+r7FLDBJ5Abu6QyEPIWRiIX Ejb4YVNRGdM7z4ZHaV9Ni9SenluMGHT3ndiIJsV5a6+I38jx6khpD6jtraZF1t7j92QaCzaWUELB BherDBvFqJM0NSk22KScV4BJlLDBPuQU1ELsnRatvKxInRQpCs1oX02L1F7YqBTDDU4Zwv5o+sMZ pFt7gyQ34tVgyo16vpoWWXuP2ZhWg/Vffk+xoWt98qYII7tUNXp9iNqW6+n9pnCVV9jgrqLkgZHk xscfLyuqarSUlsNuNQU59sKGpMCI4VOLybARTYrzag+fioSNeDWY2Gg0q2mRxrdvysPZPOo67tOz dxy1zZrIYho/eQ6eUpZbVM0Ge5X07mhE1HNOQvqmcJVtul5jegc4UpESv5oWYTAtTWGjUgw+3rRw CBvRRJxf0n7sDRI24lUxvFrPV9MisYefz3Pjh5tezAYn37LRWZsujTG89aQKbwqfuO/QAil7x8ah ltdC5B2jNe0vUivKXlhxpt0Et5oWYTA9zy2G6zr/fTQRJ5eHhP3YK2J749VgYqOer6ZFYu9F/8r4 V6MOvSlss01uLMYwl8N9eEcRpzsm9O5wEUnYeMtzRcKGjj5dhDSvaeWraJGYa2pUilN0vsEaWqOI lLyzViSZIZ8wFwzu1+9VtEije54ZvF7aPyouKCt8PhudxRjSaOucuOCMT+ppCj6tXjFgy3NUUULC BVtJUloIc2aw5yKlbTUxkX615+cW4xVnD4ozo5povKNp7QVpZtSrYmKjnq+mRRrfKzb+Mb2o5utH 9zWrCaVvMsSELQUZAJ0cu/b9kzMbq2GDD8ktYNe0CuJNh1SbvvmZl+T7Vs8AtSUCDZFxdIPsagRY NXSeUTO2CuipdWiFxufocZbcxPWKBIam+hkp4SmiyoIwXKOK18ewZFWfoVLis2yq94VVhzxWRK6x ksb5gasUBXWkosg5UqTmQkQxikUWYK+aqDDv7AXxazVOrZC2jDqWVbTImnvccKbEYGKdCeNwwb75 clGMEPhVzIW/82cyJCUy+KpgCkvurPCCocdr9DoXKarzSK2mRfBOmkTGYsTAND+UaeRnTcR5Z69I 8qJerdRGs5oWWXuP2UiJwXCGYZnZ4HOcw4YwZwYvS9jgiF1if5OU2PjJc0xQCElq8D6pVCimOJnm utEU5NgLG5IC02/tpohFuWjS5P7FHj4VCRv1aqU2mtW0CH/Ez7LxsMSAMjZ92TvWjTPbuRiDGwaF zg2OheUV4PYRLbbzc+Q5ymw9FzY4HayCopjY4JCCtK+mRRhKSVPYqBQx+N5qcqOadNTfE8Frr0jY qFfB3EQ0mtW0scQefj7PDTbcpfRjas1XTtJ2FtO+ypDBYfW8ApQc6Vk40p40MBIyOGKetkUYzvnY gqRG0QIkcGZkL0L0VrpkX+9JFfE6Xs05M4yEizq1Uuv3alpk7T3PjB8my5wZjJCyMkIzH4yhjcfv dNjsEPRPzvEdsaMR0idDepLASLhgx0R63kr59IKUj6IFtBaGIj0WDFVUMXpdxEUV0U5fzaGoSLio UysVv6+aNpK19ywv6HepKs0F50Dhgp9zMQ3zPKtHJceaNVKU5pISGdz5peeKhIzvb+2JLObEYJVJ UquJEvvGXtgoRi3JJ4HNRjXRn2IPTWsviFlcr4pdPF9NG4sGsWh6/pZ4RUE/J9K6xVoVxmIMbLip CIRylqlVOCd233WNd+8QlMiFhA3dZnQR0h5GC62iRWKuTUalVFgz/FZqRBFR3lkrksyQT5gL5l+R 1RH5tIo2kkb3ios3LUyJC8z6mADleyDtrfNSKv+NYt9R+jPiMMFCc3JAgFzTmI1NAWIiMqwdIAN/ qybI2OorIiEgamlNAl30aBUkOVhbBsze+jMy6/KqWcRxYevx65ECQ2uvahD5yTNkH4D/fdNpGRWg vEGOTvvqIECHIhUs/0z4bBzJmrQFKBotsBoGsI1kTiTIKn18W5FLBUr5gy/krQ390zbWCSPK2bi9 GgaojWeto38U/Vy8XuzRzivAzy0I7t68Ukrb0p/9Axl59I2EXwkBiZtfST2rVl66NHCnxccVjqnE zgkGQbzwbF4gWP+S0kMeXWwtoKfGnRHZGKIFiSA4mKh++9VTSPGCci80FugEdBdzOCjGuI0FARyj r1FbqADZuRkWfvJckdDANpEkRKV8LEHaV9MitRcimIGQFN0OU2XOgmrSmeVbe0H6BtSrldpoVtMi a+/5O8A2jZR0nCHMrD/teTDaV872+SfgK/Zhg7s/9WYw8IkM5TOIvaN1ohE1Z5XSZkVrX02LaG0D TWGjUlRLfKpFWVFFbOa4M1ckeVGnVmodtyInhiJxYtTaYy4mM9jbkAaBywaGi2JUIWy7dmZwnl6/ HcWh7y/2cMzPFUlmMLWQTrVSPpQg7atpESaipSlcUJhIiuKwzUMVUVDdmSuixSAVzXZqpbSn70bR AmvtMRfTNv7o83t6S1gqHy6KEYF2CMgsX7FPM8AdXsmLf3kOLkCaF1yukHKiUj5MIe2raZHaCxcc w5AUXPBTOzGqiTDv7BVJYtSrldLu0FtNi6y9x2wkM7RT0kMEBm/coe02Ixi5BQl8D8FscFWYfjuU S0pvCReKpV40kszgigX1qivlswmqPVdTEGRqL2xwgkG6KJX5fi/aNbSzJsi72kNTkbBRr1Zqo1lN i6y9svFw8KURs8cIDFd0AkYt6GJkItOrILzy3ObmOLloW681r3yeIxX1nLzTAIlgxIYxVUTfkVpN i8Sec2oximUdp7lo4rWS9ou9IrK3XgXDq/V8NS3S+J73J9pE6VECTSLXV7uWWozhW+ZQKM901kZx cvF7vPvMc0XCBtc9rRDO6XiCla+iRWKujUalaDgZ7ZuMKKK5vVpzZhgJF/VppdbvVbTImnueGe8s ORAAgzcmmpsZxehSGQH4N+DbVxk4sJ1ObSpdVZ6j89Jz4YKNI+pziinf/0RqNS0Se21tKsW0pcpX ZUY10RVL+8VekbBRr4I5M+r5atLO0Wt8rzKDgwshg+mSJkYgIuBktLngUERaA26ZiG8cZUjYAkIF u0ZSVEgI16iFLLR6BoCIeWogvNedD+KhaojwYstJYSA01KEVosSzrdUzwNp63HSmI9FuNA8QtDGD 5E5jUYzSTp2Ua1rfB8ePzX1mocFX8ujnF+KfViMx73pcKUriaF9Ni9SeYlqMctM3h0NFNem8wa29 IuHiPV6t1EazmhZZe4/ZmI6ECYgUft8fXURm5ATm5p9DoGHD54AUuoTUz/p6UpMhJEnBhH16m0rJ JyuHDBQ5TYrg3Dy3UvzBbITzopoI885eEJMvzmVuheI3yCpaZM095mIyg1k6FQ+MqrRLOplRjEFP 5pFoHX0rHGRwQDlkMEuj54I0M9hikbemUrRgkVpNi9ReMqNSpAgLGlCmIZ41MXK7s1ckmVGvVkrT BfZqNS2y9h6zkczQ/kkPJZD+oAMVG8GcGdzOlszgaqK0BKwBhY23PAcbQpIaXLeXfqRS9G1IXTUF Cfvz3GLEwN5Yp0Y1EefVntsMI2GjXq3URrOaFml823Q+LDFYM9dHIpUbfA4vy4SLsa707tlnKqjc 7MbQFSl+PdrDPBekbPBWpLuRFFFR90RqNS0Se+1IKqWG1F02DV40aYOkNa29ImGjXgXDq/V8NS3S +JaNh3dcMDHN8MVssI6fJeTFaBsyr6aK2xW/Fh/d0VI05Lki8k4DNy8nBnMjQZ5L+2pSvX2xZxYX o8/luuyrJpZP5ZXYr70gfm69KiY2Gs1qWkRLq17Af/6mMDLX70l/RmnbN6UYMwBsPsA7SlrfEkcr wb58vTsaI/m5ImGDbfwpPyulJQpLraYgZHzt6bmVog71Le/wWk1Mgt7ZK6Ln1quV0hKM7a2mRdbe czb4vqr2nVLlqCFIuyEMhqir1FzILp+4zG/MmamwwXXcktEaKEjY4DMabiQqRHV+p2iRmtNjizFs 8eXmkIEidDMVrmZH3NdakGZGfSqmzKjfq2iRmOO9fM4FizMpP9mNlTUyttoFY1maL3yYizdmdh0m dywoVxiFRoY/hIQLNpCo/CymMDmlLe2rKYi5t72QUSkGn/o8m964amLh5M5eET1HoR2vVmo9X02L YDh+PmeDYamd40TkkKGRqiL4oN+zVW52SBZwYENcaICURDEQKthjkgKlQmTmrZ4BSAWpCRGRYdjJ V4/Ng9Q4K/hkihbvxxara/MUtWUcKqakiM+rZ4C19ex6ILZc/vMoQm8sDUNej2L8MprWEvmfvjWO H5vbPeIaX1XJzy+kKcHGERUekeI5yshIraZFYq/FSaUYc/qGfVKimlhllXZRUXtFkhL1KhiEbTSr aZHG9+oF8X2xKtnYoJeFQu3D9OBCY38v/2gqX30vnZu+aynnfDnsIkkKyhTVg5HCOU3C3ypaJOaa FpVi0MlsENo1+LMmLbbj1MVeED83ThUSF/V7FS0y5p59iuqTLVT6aqK54B4FZcbB4MK1Jvuz9PlD R4mQfXvLY8Stx8IF+0vKmDBxwVkFKR9FC+CbHzNhFWKo5csWoaKKiDKKVMdYUZHkRZ0K9p+L36Pp RDL2njWcSLOhRD+m5kDYcx0uimkWJO0mF0rq/WV4oa0T+p340nlywEi44ELhNCDCHCaLV5IaRVSA V2tNi2JEoP0VSosqYtvK1RwtWpFwUadG6sQymg6S6LDXZvNR6Yk013Pox0Saj441L4oRgjfnUc3m 0je4kJC44Di/tnsWkXMatnndbqU0SLTyUbRAremxYKhiqKXbTy+KiPJqDi+L6Ll1aqQay0XT/4pu uXhUeCLNAQj9mJoPojdkAflgmhFS40ktq11oipx9UkI0ZvNjRcIFO6TT01SK55HCuVEU4FjTY4vB xR8dP4XUKmLH1p25IG5z16mROn6PpoM0ut+48HAVaa6+VFGj2SACzztSjJFBZiEp3nyFHGRwFDBk fOe5IPaOfNDOflFWKfIhUqtpkdoLG5VioKVvLIiNamIj3529IsmMejVSJ5rRdJC196S4QJpBgH5O pPmCXdkQxg9KDB8+gEr5xkpQ4pSUUoPtuEkDZrCyhkxL2krUkH507kWTUBQR5iI1FzIqxThLn0IT GVFElHfWimipZH0aqRPLKDrImnvMxWTGl44siQu2Gg8XxXiZ/3ljAcWbzlXpN+fGCXGhoZGfC9LM 4H7MdDiV0gKFpVbTIrUXMiqlvazuqDQYsyb/VDf2iiQz6tVIEXs9H00HWXuP2fBEBtLfn3lNOL3T vAiikT/3D+gn+ORKbaigjW9J/tfP5N/yS5P/PjI0IlRYFokSJAyQbJphmG44Iuy88LX/Rwnv7o2d /Dvx1xVDbt7i7Ci5i+e3NmJiZ6OGfj2Gtl/tRYtQC/FddPymfOXWRwfv21c0NPMz1EGpMzVY8n5V I4qU0wlaf18dQWqnJYhkQBhM+V53gq8WzhHo5Vg7+nfbyHoSCSKtr6PjIIrn9Jt3fUXeAhVmLu+R 1ScD1T4uhuf6mJqi17kM/fJf/q6gxz1+LEhfAhrTVAyVomCO8lW0SM3lJagUAXCNnZvHaiLEO3tF kgRxaoSO31G0kTjdHN3zPKBg10dq1INzM8lwUUxDBF2CCRdsEtRvpuEAUvqN/HGbReScBkleVVwp qsRoX02L1F7IqJT267muX006mX1rL0izol6NFGzU89F0kLX3uEWYzGA/iX5Opjg4OtXMKMbI0dNJ 5BqX7qq8VHIjlVGZnwti7xgncVPpVYoGItpHE+9LkdoLG8UYV3KDDuyvJhW/t/aK8NzxaqSIvZ6P poOsvcdspI2ARBY9zAYl77AhzG84q4RMrmGXsXIGENTYYYN3IZEbSW4wvZLXqVLqSm400Q4IOfb0 3GL8tH4tNcKzIkZFd+aCiPzj1EgReoOpogMwhSNFv70nkxkfmlVWZnAw1OvqaCjG+J7bmmyWMytp Dt4kpczgou/ELaSZ8aY5a+VPpSiuIjWaiLtI7YWLYhwsZAu0EsOKmMZhSv9qbpFwYacWU2NXx60I vwPwq66153nBJWHyn7H9X5bX034WY3TPtEyqb05zOMq/khIXXOOVuI0oJg3MvKqoItJSVM2RGk1w UaT2wkUxuoK/dFKrSWF+3dgbpInBdibmeFaK2Ov5aDrI2nvMxmQGy+QqETSipewOG8U0ptWpIBKS 5XXlAYWipNSm8pO1dARpZnhV/CJF1Ryp0QQbRWovbBSjJKLhcmpkfZ046WUv9hS5kaSGvVpMmVDP R9NB1t5jNqbNYKijmkEjwmxWRIMw90jc1RU2OIqpTKBSzD5njYL8XJHkxhefAr0IaQLHQgxBUUSY i9RcyKgUhaLO9ys1pEhRcgG6lqJqbRGqbX4h+7SYuGgso+gga+4xF8kMBizUUSp0OKfidbLPgajW db2jrHJbaLKAgzHxTTs6FbWApgUbTdT9BiMAik4LjZ4Fais8RIahsb46Ihqk5j+KkP3Ux9YCSQk7 tBgLSg1j9CwQW3j4mIQkBEzxYQazwI/h9TEUFKNM1/yKaNC3IdytcM2E5mA1+PFz1FFCkhDsHlHH W0w8QKO0j6YiaOLXm+dWirKJC3BMhTUpSi6ku9hbJFzYq8XERT0fTQdZe8/ZYOuGJqi1qsKcuxoL 3itfTEAM3EVEAUya5aY/6gQJiQz2riZwIyKDMRklmCmrFPP/t4ogo0jN6bnFqBO4UfRoYnxPol7t LSIy4tRCF79H0Ykk5l5mBvs3VDjQ++gmxXBRTGW8JlEwywWSGVJwtY+m2zQm83OstAgJGb4vU5lR KZ8/kPbRROCcrRBhtRcyhCFFnZA7erXuJTOQwfzXxd4iIcNeLSY26vloOsjae54Z7ODQujeFCiPv siEM7yhedVWE2GCDTBLhr6TEBnv60koaCRvs/UuRVikK6ltNNAZFak/PFVO+c7JRicdQTJoUJ9+e uNhbJGzYq8XEhj0/moqc+F61GmzhUPGgGpZ9m8mNYhSvLEHQZPMfueANKcomSYmN7zxXJGywhJ6u t1IU1JEaTURepPbChjHFwK7ysGFNipNPUlzsDaLWerwaDET7GYnmaBpk4/uNjelI+AKbIuDn/8hq IRqK8YcHTdhlF5kygcJJUmpVuSBOzwVxX8LATdXRYP7VOUQhqdEEG0VqL2wYUwyc91TfW02K8+PG 3iJhw14tJjbs+dE0yMb3GxvTo3ASXRHg3T8mnZMbwsgulOvubv0KLFcpg4hdUmKD7001ciGKSj+k VxVHitg5s3DVVOTYCxuSShmpUd5qcpxXe353WE60vfHqIqUUvNEEG7fxLRsPFwE0rm2V8VffuQ0b xSjnue/MbPDd+2QCuywZO8DGvzzHdgk9Fza4/CfTg8bUInxEajQdhIZZmsKGpfSL+pCc2LAmxck3 KMK+7S3Cc7BhrwbDq0ZzNA2iX9vxLRt3UzvJDRp7XdQlaU5WZnloMb2GWkTFLrvCkwl8SE3tKt15 nqPMFhI22MvtBBLkmpqt3KltowgyisScHyumEFjPV5OpMaDefi+cH2vODJbSbW18ukjV76PocyNZ c+1RHmaGtvDIW2ordpU6MQbSwEY/ChUcqekYeQ/iGiHyHlFhCwgP7DNJjRUhwv4qD1az/66lpIRE 5Dv3wmlfhQZ/aFF8XFy4pg6QhLA/F6F4PHpw2QAOrq1nnSozrkzkmgOOV2S+ezHqrg+GSmQYyzXJ +ncJKRt8vaRYMKKANFzr4adIEbUvmFSbozbE9TXW7JuthQcLKQAKVTgfRYqRgykXc4uIiTjlFKlU /T6aoOI2ulevhnLN3lFCDBfOP4VAocDvhV1f0EzkbxISF+xaV8VRJFzghSqpkYILVpUkNYoOggJp ChmWUgh/c6mjBmLaYPGDght7i4QMObWQksB+H0WDKC9k7kXpqQ2W9JCS5tSQmsxFIJ6j42aCa5yS 9uyVFxFUQiko2WIXGtg3kg7Xm+58NkESo4Ok9+fWa0fPBCFg8im3NbMOJy2KTuP0sdN/J3x7shL1 dXTg23rveF7lAXs15LXGc+kv9LIb0oiOYZjSgFtF86tj1l79RIT3r9tmNELzzsdAio3TFtIcNY5W iHNOtkKAhew82xr8QliPwtNKtRiwrQFCgf0p5Ijt8lEDBzdxvSKBjSNq3DWOm26ikAZyussOEtjt kV/8j87I01xyyixdggCFo5GZ1w1pm32O3scQVHuPGmhhq9naKgmG9Mb/6PtWq0cB8t36Y6uAC8v6 c2TorKT5qIGEa1zYetw6pqzURkm+siUS+L/M7y5G/fRDXSEWPnOKk77Z9w1DA5eQpGAU4hJAg7Ju DowUYXMCQVKj6SC1l3SwlELgGnfNdleTouT7ZXoTa2+RJIS9WkzvgKM5mgbZ+H5jY0oH7oUIG9xv NGwI849J16e3k/oyZwyo0iQl7/7luSLJiq8vLxuOFLHz2FVTEbNve2FDUv5BuRiL5hc2rElx0pEe e24yjYQNe3WRop691QQbt/EtGw+LB8ZPzAPZO3I+bCzGgPafalbY4Kv2yQTyTg2Hlxr0HETpubDB C5W6yZhais9oH00HYcwiTWHDUvpFtclXZFiRwuT9CRmskagOLxIy7NRgOFXHV9EAygxHt1w8LCq1 eVIdPG021w7882ppMWX3O5fPmAs+a5/I+QK13hwNyDy44BUSIuc0JAthglwXsb8ts4RRBBVGxlwb jGK8VZqhqR7FyH0Cx5jTwkiYsEsXqQ1l9CyQ2E6L8UtWvP3VKwkTfJ8vqyDaUAkmJugO+JnINACF RIOFlLOCj6UnA4SUCXoY9a+R4uelMbzVNMjYa1ZYSj+mP84uLqxJYXJxdsiwvUVERr0azGnRaFZT Yxl7h41f8oJdsiaDKZhJC0HmQjdVyOo/EsFRMiMS374s4imSoYL9IameJORG0F8QFqfWQ+oLkGOy 5ZwYyHM0ehk0pNMOL3hgbmdtOScMhAY7dBHS2mw6EeZ6pGeAtfWsE2Eq1t9akGeciXYxVUw0/NWn nUWDv3XBb80xU70tGoJ5UBHE/ZsGYV5JHCmi9qcljqJBTIQ0EdLBYIKz+fyy1eQgb+wtEirk1EIK vLFEkakQ4nLN0b1qKGhz9QPQh3H0oFwYUwhssWPeBC54hZP4b5ISGbw4+f2NyDmqM1oIJ06kCJOd npIaTYOsvZBhKb0eHGpNWlgTv4SKjYu9QUx+vRrMsTea1aStodf4fmMj5QX7XzzigQ1WmYYNY46B WSSsULqyoOo4aUfTkPKVipTRQpoarFnpxdIilM90+OyBpGDDmmCD8wnUj2MvbBRjsMU12E4Na1Kc 3Ml6sbdIUsNeLabcaDSraZHG9xsbKS9Y19LVffKOjytnNSiYy2MO+4UNtucmdjJZb44GZB5k8IeQ 5AYHMdKLRIrYOY+QQjuaihx7YUNSYIy2PnSgjiyzJsXJd6iPPTK+SNiwV4uJjUazmhZpfMvGLx3J j26TERscABw2jKlT5igXbpIbzLzkHaBd16IyP39uoWGIpefCBt9dT8drTA0CMy+puKNpkLHnjqRS YoMjh2HDmhQn5VfYsL1Fwoa9GgyvtBPS0aymRRrfsvGwI9Hatwoc2OAsoc/ffhYTGyyEhA1uBkjn ws1V8Y7fTHlAAyIkbPxwL/Ri7kuo0SQ1mmDDyNhrI1qMURhbFpwb1qQ4P+SV2Le9QWyPslJeLcbq yUSzmhZpfK/Y4L48XkyxoY8KaHjOrsqc0yEGfRBBnPOh6xTU7K4PG3zmWnkAG0LEhkZs3pJYzDXk rSbY+AZxG297PFcMH4hBn1ZYTY7zxt4islevFlvPL5oay8b3io03torGO6qbsiEsv6i2n8kuX71v 7JLSb8XHAZoHQsIG58E8mqGEQogwWfS6KhrEzvWxg0EGS4YmA0X5zRFyZtia4x5r9WkxcWG/zUUU LcJvLU2vuKDjVI9NXaAyOJlhTNnNLQXh4h+NW0bgKZY1aPMAhKpDSLhgbK9euJjaQk4EqVRbTUHM ve3puUopBs57hgxrUpxMBagVr71Fkhn2ajGx0WhW0yKN7xUbHzphDxmcphwygBwB95VpKkU7zSgC FCU3GsS3b4sUCBXsHsmidIWoqG/1BFhbIcIn8+GBk5DhQWqcFRwnU6Wj4SFpBjECTHsdWkw0yGfx WT0DNC5I+P2j6jzNrg0GypQ7rOwPC8b0Q77/0XZraGjhxL6rOMY8V358IU0IpqhSUBjTT81ZCkmt niJjraWJpRTAu9ZY1FRYk2Lkk7sh3fYWSUK4lIoMZG0kq2eRxvYqHRgJi36dBPryHb3oNKYIGBmE Cb5Fm8i5zyzdCuuBKSuNJCPYOpKZPGF4JwostZqKjD1zuBjDcd1jKC6sSZGzYi5NGofZzyDNCns1 UjxXz8VqNS3S+F6xwdYNFYOMpllOa1NhTGz840ZB5wUXayXz2bkeNv7pOb0wRsIGZ1vS8VaKadZI raYgvI21lzfEUoqBq4bChjUpTj7GcrG3SDLDXi0mNhrNalpk7T0epKX41C5KPjtl79gBYjaCObuZ ow0bnIdq7JLSb8UxgeQB9296qwW/KNv20/EaU26wL1ZSsBFNRdZe2LCU2dBRC+WGNTnOG3uLhA17 tRhbjB2NmwxeOGm6j+95bijxvAdQ01CkubqRYoqBw2TaXEAJx7EwZwLtWtjgIrzkgRF5p6GblxI1 8WUpKpRIraYiYsP2woalxAbbF9VGVBNTTTpOdrE3iFubejXY8fyi6T6+39jIME27K1U/4R1HbbMg UExs8GUQ5Qanm3RSX+8F/ZhaEg3ceG4Qe0dVzoruYkj5jEW6pGoqsvbCBlNn8oGZC+41h/1qUpxM 617sLaLn6tVi+h0bzWpapPH9xsa8KSxMqbRBWjfyJjeEeZj2xwfJsMtXyhq7pMQGV8emAGNnSBb7 NXTzwiKcRYqVjTtNQY69sJEtomKDuWW0V5PjvNrjjYafsVevFhMbjWY1LdL4lo2HwzRtr2QwITbe +P/KRjHqZm7U9K/wj2+E+Vf/g5TbeL7HlDwQ4jZeh18vkBKeL79K+SpapOa0LXMxxuuc+jUZ/qAI kfPpWeWmholWVCSZYZ/ERaUSC4ytovvolouHgzTGhLTZ5sIfTtEJw8UYUfERMHPBfFEqai7eiHes jKasNCLvNJDy2iI/P1IOk2v6UtpWE0lzY0/PLcbgnG8PHU2K8yOaNCCzpiDmnjJbXo0Uz63nq2kR Jr+k6RUb/ggZifGXDrOJEYjJL+2x5CXhip4EztRLmtQ/jKXUqghoVrB1RNVHMcX4pgEXWRE18+9a Cg2I+Jf81K0q4tNaFB9fqUlfJlMLJCXsz2IiIR4TOnr8e95E9YoCdpGk9OS7jVlQJ7fAcI3JBU+n kH65NY5OwNfAe6jmgUeR5AP7SNTzrhTlR5RbkROkAPNPUhQeijFC9w3vEGFFipGzH0pIjcLsZRA3 03XqIhW/4Ws1LbL2HhcY04mwiAHvNFlMGgwXxZgRykwkB6c6EcEtNXGOnRP59YW0D2HnSKqJSrFI FOWraJGaCxmV4rXmpCmUMQyTJoXJl0cu9hZJWvjb60dI+0Q7jpEivyxC+FXX3GMu0oXQ1vpr5UhT uYSLxZgPIsVxTv1aBxcc/Yhz/oa6ksBIEuNb30JfTK+Hv9CuDKumRWovZFSK9t5rApBhTQqdz9Nc 7A3SzLBXgxH7er6aFll7j9lIZmhnpVoq2NBWHPWnCzHo1yV1IoPT58kDTY1nwMZTihugecGdJqkk IuOjCtK8ahaJrRDBeQYJqQzQLfEQYT0KkVrj2BogOWF/Brq4HDUbhHNCcT1vKlS2M4IQCbpCJyQU YqyvC47VGv3hzjiXGf6sIa8uR9nz4wuQYxo8eZGRYVhnYjijIM2rhqr6YsvULYT3fLzt6OE4PGyo sa6tAWQr/qB6ZBKF292quYvrFQnvZJ5J4IBpSRDk5pK7VDTpKA9nLKENZDjGUlZ+dn1sLyT842C9 k8Pf3+MnZb1RmpkyQI1/5CLMKvSpFSIu70SDTOtRgHzF/tgaICTIHzwcGe0Pta1Vs8jYevw6TOPA l/LUdyFMI5NMEBTv/3rjCa9abnZjQKWWCBK+JaLXQUBIYMyV3hXIIXNP+40aqsYgsaWnFsJ7VkZA NMRT8ArwYisR11b9OTJSaFurZpGx9ZiEaROYXlc5AnVspC4LxZjs4AuFeKZ23AMMBlKSEg18RzI/ vBDnNutWnuBaIZYIonwVLVJzIaJSanp0KAYipEgxUlpdrC2SdLBPi6llqN9RhJdB/BM7uldvBQs+ sut3abgopsE9Iw9xwcJbflz2y8k7tV5+jjlcIckJto/wo/IeVEqnEiy1mhapvZBRKc2x0FjCRRW5 WbgxV0Sl+Tq1Ump3bW4ULbDWnucFh9dChT4elpcjEP7r62Eyyl0nEtKGBW/Uo0PzUwGaFOwcUTmy QlTcFlo9A+CY1ISGyKhMMQlS4heB6iokxBLxzjPrTjH9/PG4WuafsYN3jwmY1oENG/oB6SKp38pA Me1V1yAZCr6Y08IxBk6SUjLkQqRBkgxsGaF6GUwvP8cQpH01BXGq2l5YqBRVovslDa+siFnUO3NF 9Ng6tVIUSzE3ihZodM+5oN6nr7Jvf/ioibkI5gjeVQzI7Lf6L4LkmstQwRb2/PRGQgXrHWrEV4rS MFLsg4miRShvpClUVIqhjK+658WoJuax7+wVCRdxaoXW71W0SMy9yAvtR9B8vsauXDtcLooxWGb2 FC6kagYfklJefOQ5RhFC5JyWqLxgV0x5wVyktK+mIBBWeyGjUgwk2INy1cQY485eET23Xq3URrOa Fll7z94S3KT0tXfc3jNsCHMMNM6aMyHJ3HnyozN1HTY+I1MkbPxljcCpUSkfQbjRtAg5Lk1hg9ZH Urx6ui1IvFYTBN/ZKxI26tVKJZqrpvv4Xr0nXJqgClADeIZXajW1u8sYgzfOAtFWq/nRRBVscJmZ skXtk2WKhA1SI69TpXzkQNpX0yK1p+cWoy2mITEb1aTpiFt7QdxOr1crtZ6vpkXW3uPcSGnBtBZH XswGkxjDRjENZXUiAza4q1y/HkMHSYkNSkA9F8Te8fp+qU05UpTfkVpNi2jgjKawUSn6aHoU2F9N bFK9s1ckuVGvVmqjWU2LrL3HbKQ/oaH/0o4fcoPbZocNYWQ8zTE7i8wGkw8dQ0hKbLCfIHnABqxE pTU0F0iMNCKl8we3moIce2FDUmCq3twqrSbtVr7Yw6ciYaNerdRGs5oWaXz7pjyczaTv95Zm2OD4 T9YKF6OfZurEbHDeOJnABRN6dyhy8xzDCz0n79gK5cQw4Co6IqtGRxS0qWSMhQphhEmt7cvuq0aT x14nXEtFwkP9Wan1OWrwcJFGtjw8nMmkMP7jkp/Blu65T4shzFnBhAMNBe8Iu6n0uyHu2/DhgeUd Gv5BwgPbRlJTVEpnEW41BTEVthcqJAVGuc0eHTNaTYhf7TkrjISNerVSG81qWqTxLRsPs4IfVeU+ vxPHsEPGQFSvmQDiV3Ujr8glpJzgXCquFQgV7CNR5bFC/GJWvXoGqK0QERm857tabiuqhggvttBb IDTUoRUan6MHDw3g4dp61lDonfbIAWld96+UWEyZoEM6pMSPLoNzhP4oADS85zlxznPhgU0lKSeM 6e0gvVPWVtMitRcmKkUI+mqoXrJqIso7e0HaidSrlVrPV9Mia+8xG+lEPKGkX1IraezLCxvMVQmD DS5owjv6AG025udmGTM5wQJaEkCInfMkk5qKlfI5hRtFi9RcyGDZS1Lac+n7VFaTdmre2iui58ap FdpYosi/qyJxZji6fT0eNhZUB24PKHTYHpHF0sU0oP7W5APlrK5wU5jMfaoZ1cDIo4Yick4jNdVm R4p+P1KraZHaCxmVUs/hbQWrST3Orb0gJn+9Wqn1fDUtsvaeZ8a7R8hiQze7JzOKEYOupxAbvCZJ BLZFhY2vPAcbIE0NDqqpownGc9SEkVpNi9Se2FiMGD70URJ4rSbivLNXJKlRr1aKOjj2VtMia+8x GykvGEFQXRIB1RNXuJUNYWQXNVZOMlHO6pI4xcnW27DxneeKJDeow9PdVIqC+k5TkGMvbEgKjBqL o5bOsmqiNrvaw6ciYaNerdRGs5oWaXz7pvzSkXDvkH5PKmsdZEtuFKPS5XyTc4Mp7zQI3GuRVvVP nqPS1XNhg+ue07gY06/OYQVpX02LxJ5zajHdg6E9DuRGNVGj53gdA7JoKhI26lUwckp7RG1vNS3S +JaNX9oN7lXR7+lxRq4Y1E5LY677aS54U1ih0a/nKj9lBxeGJw9I5WGD3QHpd4ThHQl7p2mR2GtP VCli4IoCs1FNHrGE/doL4ufWq2Jio56vpkUa3ys2fviwsNlgADG5UYyRalYiKOM4C0qe8spIysO2 PFckufHViqtCTD1FaBUtwtQLipoaleIt4oSQyYgixrd31ookM+QTngdzZtTvVcQs2010T7n40g5K DySUmrna+2CaLfC8tPpLDxs0N+BWVYM2P1ckXLBxJP2OpJwZn5FaTer4LvaaGcVoTPVNAb0n1aS5 D2tae0GaGfWq2N/j+Wg6SON7xYZO4vOWvDEyVJvB80aYCnFzRmWle5yUFGxzV4j0DmlO9W+7pb5L N19XhnCofCQzSvrPWhFzQtwictY18VcFS8hSu2b67+RCXRmZ8ZUKBi3/Wd/XzpPeg2fZoqHfju5L l9sn9mJMtegjN6pic5EAtadvwCd6vrKkx4rIMY2b5PNgCo8DCpKyIscrgIhrLRxUiHLUl9sfRcyM 3pkroufiFLpG6sQymg6y9p5zwQYN1WQUNt+UyuFCmEOgpqL5w66vuiNK2naC8xjNIkXCBRW8ftqV ooy+U7RIzYWMSjFI+XHj41pZmpgrv7NXJGTEqREi8vo9ig4SczRHz7mg4M07wX2Hw0UxykqOz5gL 7uTR70mYkhIZXGqm54qEDHaKqLddKVyI1GoKEu6lSc+tFDHwITMQDfisiTjv7BUJGfVqpIi9no+m g2Ak9p6zwQYSTXZS/2s6PZkhzJlBrRQ2uDQqLwAlT9jgCqdEbiRs8NmD1GaVIjnvNC1Se2GjUoyw /J0F2KgmlpPu7BUJG/VqpIi9no+mIie+V7nBZg3VCxoNZZ8JGoppPcqHkKnbWNDyr/4lqQyL/JwW pEDCBnOdqagqRbMV7atpkdoLG5UiBl3GpdyoJi2u3doL4g54vRqp4/loOsjae5wbHqcizWuZipPD MJMbxXCKrVl4pwP1mufjHeAOr7DB0Y9GrhsBwwafZVXdtVI+jSCp1bQI3klT2GA3hqSIgekPmmyN /KyJOO/sFdFz69VInWhG00HW3mM2PBpBmhkoRUDrRrGXF0UQyUW+6uoemWV6SwnEMN4VocZqfipA qGBKSj3uyNCRRyZqiHGR2AoRFWLo7S8fQET1sAB9aytAaJA/eFiZE8WoOcjYekYC2yXpitGnDjAZ EYhwaML88VVI4NNT+f21B0mvB9+nSsQCQgKfi4IxSAByyHwtSjKrhuIySGyFhEKa/tTn0iGherSH 9dhyxAJCQv2pzNe6PGoOMrYek5D3QtskPV5AmEVjZcJCeK9P0SoTOLOXhM8JLwajeUpbGHLsVEOl qwhFYkRWyyIxFQ4qpK0V3m1jNYnYa9hrSdsxsp603gS6ODxaTgg29aITZdekrroiDdj8FwIC0Kkz dZgy86+HJJST+n016NET+qd/FS2geeegEeU8N+3694+GBWwjkSNBnEww+AsPdMlVweab/Pax4X/y BLnNrb9KqggQY7yMBgrKAWrj8a+e919bJtX4EjNXl3j7AI8XY2qNe3RdoXL9VPKayXg1CRp1+TkK fiGKhAEq3yYwG5XSThdLjSYyu0jt6bnFmFWgnLpqYn//nb0gan3hoV6N1PF8NB1k7T1mY94Bdowo AsaC2iWVHCiGm55hxW57QC4mDxdccpW4hbhn0DjIgwlIiZTmKS01eoi7SK2Fi2KMor2OsorYj3Zn rkiosEsjQ9h12mr8KgtwE+fInlcM2h2p6Vp4YN/68FCMlGbnurOCj9ord8hYSSkr+H5NaggjikgD Ly8brhRvU6RGE0wUqb0wUYwQlO1HETsW78wVYX8NP06dGim4qONVdIC19jwn2DiiSoBZDHa0NieK MY/huVXMMpWrHGAjrKTUVP7Jc0GaFXztSR3pStEmRmo0wUWR2gsXxRhB++io+hcr4oe+M1ckWVGn RorQ67gVOS8EOC8c3au8YI+G/GcWQ9f65/0oRs/GjTP+CbimV1mgjQleptIYzM8VSV6wTpxKqVIU CpEaTXBRpPbCRTHG077Yn8SoJhquO3tB2lrUq5GCjHo+mg6y9p5lBs0v19uZjT/cqmI2FqPYoXVy K8UnjDPuYFKIjpHM4Ab7RC6kmcGJFPWa1D+RoiKM1GiCjSLMmkgmbBRjVM0eHrokjeisiV7gzl6R pEa9Gqmv9Xw0HWTtPWYjPYn7YEXAYoQOIyg3gpFdOmKgBOA9YdosbwA70sIG/ymRG0luoEI9HCFH ClqQumiCDSHHXtgoxkiSHaduNKqJWuZqD01Fwka9Gql6jvbRdJDGt2/Ko3ltpNkjoQg0y8Nmq7BR jAES5//99rG3IpnA5QuZumFHRCIX4t9Yy1NeQWRMkY3HvEaRGk2wUST2nFOLEYO/dED7XE26Y8Zb mDW+sr0iYoNCMl4F443eaEbTQRrfsvFoXhtphniqHzTTk2+lH0xzPWpgscuB8WQCE9bQwpvCcXE9 R0MgRL8x61O6u6kQCa+jBxYaRQeJub4olSIWfTzhKCLcqzVnhpFwUZ9G6vg9ig6y5vqe/JIZLP+F Ct0gnMQIxNyG5xVh4t9Hhw281GGCbx7hGkQICBEEou63mF59nyTVfFjmJwfAsXlqILznSLFL2Koh wost9BYIDXVohAg6PkcPv54BPFxbzxsL9mfIdyZ43rj/KiwUw/1/n5oYpW3ighVHyNVloYE9IInZ SHig2kzXWykqjUiNJpgpUnt6bjFaKecdqVVNtFJ39oK4I1mvRorQ6/loOsjae8xGyk76H4/7YIPt DBmFL6ZpR/WmsPGusSA/N7vraT08GNNwsYjfeY2SuumxUhSgUT6KCLxIzYWMYnQDDK+cF9XEhMmd vSJJjDg1QkTeWKLIiSHEmeHoXjUVjLRUPDCU13cvkhnFCIGJRWcG37QPGfQp6Vj+5jmqaiHJDLaJ pOuNFKHTLUj7aDpI7YWMSjGwoNR0W1FNzJvd2QvSzKhXK6UVD9sbTfBzG99vbExmsE1DEWhAm8Vi NBTTkFZ3N5EZrJwnEbglQE2phkZ+TmMr6ruwwWcj0vVGitg5kSDto+kgtRc2KkW99K7LM3hPqklH 1m/tFeG549VKreej6cSy9h6/JykxKOC9iUNs6Cuiyo1gZJcGu8p87LI3Sr8xsftbo7DBfUmNXEjY YGNeut5IETuHEjSrNJqKhP15LlJgmKHZcG5UE3Fe7aGpSNioVyu10YymxnK196wjYYzLjcuWhoZh oxjDXX3NVWywgVC/sZbjdSgHNvigvJ4LUjZ+GCQdTK0jpxIkNZoOosV/NCU3KsUgyxsAyY1qwoy8 utgrEjbqVTC82mhG00FiDz+f5wbXU+j3hFb26Oq+STQUY4DNt3hBaJTobxwnV4TGOz5HpucYgAlR VAzdeEkXI+MpqiM1mg4Se35upRhvfbeJriZSUF6JjdoLoueOV8Uuno+mE0vje8UG+zr0e/LKcYNe 35RiDHv/aGYJu+z10NI5Ay5JybvPPFckbPAqqRceKWL/idRoKoKm2tNzK0XHpZNZ4rWaaJ7u7BUJ G/VqpXjdY280wcZtfK/Y4G4jze9yXoHhbtkQRlQMs5l+NRvcaJV34F1SYoOp3+SBkbDBKRZTFiHV o7eKDlJzemylKO2++d6fyPD5GNoIvvZH93isBWlm1Kdiyoz6PYoOEnO8l8/fE3pQ/ZpIczNHuSgG F95zSGZoxJAwJSXvfvIcQzkh4YIGU71wMYXJpQ3SPpqKhPt5bqVU56qNgIxqIs47e0WSGfVqpSiv Ym80wcZtfK/YYLOqnfvg+7olA8gRcP2n04KJ7aQ/M7tpUvX1KUVtIEww+ua/FXMfqW/cigirgYf8 m1pRWpITgaipOcNqFqLlm7WGi6n/MOUfICTEn8GUEvF49Cywth4nREoMBnD+FAoJ8aEvNKbhLIb3 rq6o6rxVgJAlA+fUzB58BHFnoNGaVxBHiqBZQVKCVM8Baiw0VAj/qVrgbhQpxFtzg4QI+zSQeKjX UYSXQfL7StGrZKCd049I9rOBrEwU06ifaXreDO62S86zJK52g9TLY4w2hCgmVszy9bRiejM4TyHl VoRIAPtma6GiQrxt+sTSUUSUfAT5am6QUCGnnCeVwkrMjSa4uI3uORdKW20aExfaq66sCJYQvA8d MtikmLRng0C8e/MnqHgRjIQMdl6osCqm2DlOoRINp6zpILUXNipFDPkeitbgpEmx39obJGzEq8Eu no+mxnLie8UG9R7tFGxwM+KwUYwkY8rOqcEdRIqKlVd/M4XU4KCcnisi72i7dJv1YmKDIxaSGk0H qb2wUSl6pR9d4LGaiJOv3oX92Cuit5LfKF4dKSY3Y281LbL2nrUY7Jf844EEY/931muSG8VojPjq AHapUzgPQ1QMtSSlF4ULflNSCrF3WsvyguJIETspIanRdJDaCxuVYqj1oYUN2Igm4mRV4mpvELEx Xg2m3Kjnq2mRtfeYjQxKqIG4QBl/NROSSisQyaW5EK0HY5YDbw1dO2PggrlrPUXcAHKNcid39ATS u88hC8lEjRvGIrElIlYI9/2JmNWjEG9sFQgN8scvUmQ2ilWzyNh6TgL9uKZnKSlp5ZIRguw9xSVd ISRwMCQxf6sphITviDD+BAgJ3Hqo3taQQ/57q4aIi8RWSChE/arPhYvM6CFm38VVW45YQEiIPytD cR9bq2aRsfWYhPSkjJR8JS5DQy2Sh4RCzC7o0lCRwP+KFknzvzjGHr1UCgB9KTQ0GQkCZvFVEqPk ILEUCirEiMrrxVCgAQfBedPv2ikQAuLLymwEq2QRW8K7xwTMq/DFxi8c5YflGE4ZKMavz857M0CV oKILciWV8Y+fKyLfNE7y4uJIEfRbtI+mImiqvfBQKTKAbVMXTUT599beIOEiXg2mhqGer6ZF1t5z NvxZGJxjm+mQob/0Q1I5housYcEFy7Phgm96+sc3ECoYoGgCbYQInDsVL3oWqK0QERmaFX2SR4xK zX+VEdSb6bBsq0DfiTg0mGiIz6tngLX15KpRnmY5FOuwgPVhoRgDe3+IgpeCL/A6aLbUxDNWI/UY bbEeCw2sfR4pt30sfkpqFMFDkZjT23QwegjX7xrWSdE3XY2UKwFjbhCZq0+F6GA2lNGzQGN7/nKo +PEQQYuxXLWr5mExuiUqXVvlS9AJnOsv1WCoIPJzWg8GCRXsGUkTIczNJCcPVImNJsIuEnt67mAU Ie4iqYesiCgpKo452o0ioSJODXZxfBRtJGutr8bDNRAGCP88UkCatsANxWJEwI3rRE69SXHkJECa KOHiI8/BhZ6TcxT7NCwr5Sg5d6BEGU3EXST2mhbFCEED8lWkKKX8mBtE5sapYji1jo+iBRrdq7x4 e/dOQ4pFlhGaF8Uo3t61QIjZv76sjBeCjfZx7jPPFQkXVMXpYSXlvOCUgTJlNMFFkdhrXgTjrWcO VW+ExnfSRJjUV+E+9oqY+/FqMJERz4+mQWgBHN8rNrj0WL8mqyz6NlLekmIMN/R1JLHB+QM6N9hg P7PeGw3L/FyRsPHvs3sbK8UvFKnVFMRdiO3puZEiBi6OCBvRRJz+YtOxN0hSI14NJjbi+dE0yMT3 io2PL1+0wBhTc1NhQ5h/USbvoAo2uIxInDGa8gwWbHxHpkjY4Eo7fsgjRTt2p2mR2gsbkVIMPvRL bkQTcUJ+2I+9QcJGvBpMbMTzoymI2Xd8r9j4/PZYAe90k1LYKEYCMMnoZpork8UPsfu+Jdj4yXNF wgbpnS60UvRvkVpNi9Re2IgUMTC7ndyIJsV5a6+ISzq6bnt1pJhNtb2jaRByw/H9xkaqTW2iVIuN NDcnDBvFmK+hpXRucD07PxFscFdDepi/eS6IvdPILP1ohaj/IrSKFqm5kBEpQuDT8mokqogoKUKv 1gZJZsSnwZQZ8buK8DuIM8PR/cZFCk/tsvSoQTM/NFvJjGLqNHSRDe8JczJijK7dGxzo4zkAlSww ksxgD0nIqJROI1hqNS1SeyEjUsSQLRWQEU3E+XZrr0gzI14dqYnmaBqE39rx/cZGMkNrfkRJ3677 gcVFEUqJDy96UMLpMjgxwbKPWlRq6mSJ/x0e2FGSZdvKUEBKZrX037UTDiSB32z5Ux+lAZ4WJolN NwzzG8RO/90+JL4MNt4eLbfxOPb//J///r+HO020p9LDBoo8Dr00+mLahqIruskEvkmW350b1+MX 47P86kJaWTCMSx1hTD0Dc+aSWk2LxJ5/0cGIgCszIAQeookYOWUVvmNvELE3XhWDrYnmaBqEX9fx /ZYJeS/YKvChkhdppog/XFssRtHrWTXetT99CbiqSo0Gawx5jAGnEDmnlTOvIwbDOTyI1CpaJOb8 3GCEwFURImM0ESbz6Gqyx14RP1enBhIXieUoGqTR/fd//vNsOKZ9lR4+KG27yWIxhve6xV3tDnu2 kwfcARrn2B2ULBBSMv56fY/3x5jI4NCEpFbTIrHXzIgUMbCfXnmAuDV9s1om7SIj9gYR+eNVMWdG ojmaNpbG9yoz2LehupCmjCq7mVGMdKVTtF2+bpZRBZdX6s3RqMjP0XQKSWqwcz09rjC8o5mK1Gpa JPb83GDEkMPVsBFNxMlX1aRp7BVpasSrwZQb8fxoGmTie8UGt0CpgtK8B1meNrOYXl6PG7XZ7kNz 4gw3/S7gHfc06bkiYYM0Vzu5UnTnkVpNQdBUe3pupMh47oxIbkQTcbL5K+zH3iDJjXg1mNiI50fT IMfe8zeFL9ZoWVh9Plu3w4YwoqJJ96lRcvKD++ESp6TUfjF0TR6wJDq5wY2dqTqNKU5OK9xoWqT2 wkakFAMLCWhX5ytNivPW3iBhI14NJjbi+dEU5MT3Kjf4Zql+T6aBdMtq2ChGF89GBvKU2s3HlfjV ucgsbPCF0uSBEXlHNfXefZCVYlAUqdW0SO2FjUgRAxsT1EqMJuJkz9HVXhG3N+PVkdKeUHl1NA0y 8f3GxtQWDEhVOVAc6260sFFMBY+6e9jgS/b6jSm9fYMaucFSgZ4LYu80ZlPjskKMlyO0ihapuZAR KUJgS4Wa6CoiSvprtVtjbZBkRnwaTJkRv6uI54I4Mxzdb1ykd9VWSo8hNGzq1P9iVMJMPzozWLJS HmhQ5vl4jdf8XJFkBhtMUllUSgostZoWqb2QESli4OLNZEY0ESe7hsNG7BVpZsSrIzWeH02D8Fs7 vt/YSGawnVLrkwizMy+JMRC/4ps3nFDH8u6aCy4xi29e1YQcAU0LdpckLypEOlj16hmgtkKEZfCe gy56HzTq074KIuQM0sXWAEmKODQYX4VIGEfPBBFbePis6dTWSg0e6IhYo8tCyGA0nZzxdpZxL5P2 kPASS0j5ykA6v74RuUb38ceriSvlpldSo2iAWjN7izEEY2/0RREx/rs1N4jMUWT7prpiJO74DRXV tEije8UFWyKIDu++eNFdYpCTxghBXwmX2aTDu5sCiGDBLL+9kRDBuhqBQFelqCCRgsDREmBNtSqR kFs6rkeQLY32pIgQ/9yaK2IC7dERGY9hoWoWSVwYe5wR81qwXQPrsMBqdhZLfVYSDP89wQkLnK/M 78/yuXJEIzSPPRhpgNgzFZtq/CGiUio2LTWKBhhr4m8xBh9MbPLLjqJv2rtbc4MkI+LUYMqIxrKa Fml0v3ExDWa2i+Idh+OGC2H+LbnDRBMqlG5UgQ6Ta6pCBiVHMsBIsoJdE+liKkXxeKcpiLm3vbAh KacFVwMkLaKJOL+v9mg6ioSNeDWY2Gg0umPEmhZpfMvGw0lw7bBUxQ8b7xTV7koHoxWlYQobHKpS rvAzIuUOhVP8SQMhTnZGUNxzvFL+1TlyIanVpDHrxZ5TajFv1oN0ciOaiJNtaWJfIzE/VyRsxKtK ubVoNKtpYpn4lo2HA3cqfs5nmQ3ua1FZMQgR0CWGC8ZmyQONSvCM1yc5wL+TFdz6kRbFEjQZSLhD r44gtdOGQjIgGOT/mYNoITovJo8d/7uNRDwZifH16FiESYDfNx5NG/Hj6xxhimMKTYRABM81l0TE a8GoQD8oLbwXAFhm8FMB2kIwxat5jZGhaI7MqhmktvJKRAjnP9kt6gYiegjv/cZWgaRA/FkZ/fAa +FSNf0EhvEeJ63kC8BaxUcU/DG1hVgoHEwsMRfCMNzL3vxGhpJQEb35OMQtJGniHzEAulyMURTxG rwoS3+axYImAXXpIaYCnxhsebq0NEibkk5uLSo3foiKKFtHI314+7jTSUGqDpLYUiTimP50RwfxC UyMoI+jGcvsbkUtKXLCFr7+/EHnHqOyfD5ivFGVnpCAjmgYZe3puMWLwt09Wk2K/tTeInhuvBqON cjShNZru43uVGW86Vo5zHMAeMgyRF9zKSM5h9S9NpLsRit5w4XtGyAEDoYJtImgaTEH++MT66ilQ W20jLcNvyZR6uqOoIULW5C62BggNcsipIyFSV7PACmP1DNC4XpHAXg2sU4dw2umvW0htr/SgQQNn L4NRoeXSN8YKXtsnJb4iUyQ8sFck9Yek/MJzsgLtEFpNWju92OtrVYwSyPNDWoaVIgLnC1JhQub8 LhgJFXHqSI3jcBFFCyQ6rPX1eNhzUuNzhY+54PByGszBiIB6gOaclHhzTcWAWkJ6Oyh/UjYaCRVs F1GXslLU3VZ+FA0C9Tan54IRp2piWmVejigiSKqBY87kGAkV9ukIjdvR4yy5jW2ZeNhraiOlfglc Y+P+MCHMfR5JFyY4F61fl4GDpETFj58bJFSwYSQdTKW0TeVWU5BjL1RICox3kC2m5iKavjFztQdd RcJFvBpMb0ijWU2LNL5l45e8YBtJajuOa2VJTL+ZRw8MYv0NGvKC3zetANcR0z3RqXD5V5JASJtN Sg3VFnAmzFXznSbq6Ku9dsHFiOHNFZwGd9KkOKNp7A0SNuJVMWdCo1lNE4t+7fj5vBNhRwe/J9Is er+1xQhGcv95l1OwwceV0hbk8nnY+OfniNxIcoMKTPVlMLcYnENQtqwmKservbYYxWYugfckmohT MxBmP/aK+LnxajDlRqNZTYs0vs2Nh2+KG1n9CjQ1fIPTXepi2hOhw06wwXx3RiPczatf3W188sBI 2OAqDl4j2KiUTx1I+2oKQsbXnp5bKcZd+sJ0hnnSRJxft/YGSW7Eq8GYaFFXKHuraZG19yw3qCs4 DW/vOIo+bAhzDOz4Chvsu9ZvzEy9pMQGC0LJAyNhg5V0TWesFJP2d5oWoc+TprBRKWo6f2iR3Igm 4qQQvdobJGzEq8HERj1fTffxvcoNFn/0K+CdPkyr0YeWq4yx7YGL/Pw+cmwymcBXhOIdH3pOHhiR dxotpSmpEMvsV0WIBAn5UhQyKkUITBweRYry1togeoyyEp/crlZq/V5FizS6V1ywbVKlJzXav57M 1s5LDSvI7q/MKTEB4O+48/b/SEqZgdvIDBIuWBijGBhMof/caVqk9kJGpRjzcHLTZEQTsVOEXe0N Ejbi1WD6Fev5agpy4nvFBqe8nBi+/C2JEYgZD98Qx2/AlmOVnl7qjG8cMzMVvlYuVLBvJAVWrppT NW3Vq2cAcctwQk8NJHJ3HCY1RJgb5ODdtgYIDXJohUhdzTAojNUzQGyRgc8bC/ZryCgjPD4T445E Gy2Noenba+pk17fHHowq/DEZXGN3ZWI2Eh7YMqLON1L4xh6yO02LxJ7fj8W4A8MfiuFFiyYiZ3OF XtCxV8TPjVeDiYt6vpoWaXyv2GC/Rup4X8idnCimwb6ngmiFvWWXgYVv2qZn4Y7eBG4kZLBlJIlS KarIKF9FQchXuiZpSlpUiubYI0DIiCbC5GBJ2u7YGySJYacGEhf1exUtsuaeZwZ7Nyg15ZzK4HAh jGwhBOab3UhxF2/ygFNKcY5bFhu4kJDB92TV9ULZlNScrLjR5DMJQmovZFSKntrtM2REk+K8tTdI yIhXg4mNer6agpz4XmUGu0pUfjLzRShloxitsKmiuaBaTiK8S0pFB5ekqs9g8CUkbDDZo663mN4T DidI+2papPbCRqWo4vyZGNiIJuL8c2uviFua8epIUcnE3mpaZO09zg1Pan1TF5KTZsNfkCI3Dkb/ 7O9MwQZffUgm5FtUsPHm54hciL2jVmRJ22xUyucSpH01LULtqK9MhQ3KLElp24w2UsJGNH1rs413 NIy9QZIb8Wqwi+ej6cSy9h6z4RkdpOk4iYDcYDOBc6MY2UXt5p35sMEGg2QCU+DJDTa6JnIjyQ2+ VaoijKFupHwwQWXAagpy7IUNDjTIB2JgAcTdajQRJw4ce2gqEjbi1WBio9GMpoM0vn1THg3UkOY2 RCIQG5/0HsmNYsTAd9fwjj840pQ4keLXQ9zPKXI9J+8YTHnn+mB6U6hfpX01LRJ77VsrhQJPrK4m xSntx94gsjdeFYOqjWY0HaTxLRuPhiZIv1HvmA0OrvnW/4OREnwQxnahK5nA3U28MrDx6eeI3EjY +JfNXMHcv3L9mLSvJlL/aq9vSjFi8FdjYCOaiJOrwrRsNPaK+DleP3s1mNhoNKPpIMz2SNMrNj7+ araf3OAaSi8ko6EYMfzzcXQ6F75V7t6C3TZh49vPiQ0hYcNnsIvo7UdEulfPIrHW1qZSDNh9bwhc SI9ivLU1SPJCHq2U86Jej54TR2N7xYQ/Vy8i+Eahyi4UBNLhDJVWZOM7s5h+R9jWGx5895NoEBAa 2D6SDliYm0FupTQP1aMzCwJiy08NhPd8Dw9bGvxJjaK+2HJjYSA0xKEVWp9HzwJr63HDOd0I2zZk lHaaVZk2FcV0UsNrVJRwzNKYhnxHnKb0j5+DByH+Yan4KICuUj6cIO2raZHaU0yL0Xtw8zD8jSai zDfD194g4SJeDabfr56PpoOsvcdsTDfCpAkRwAY1/rAhjF+BPpz74ZwUn1oypP37kZCGiX/zWJEk BZ3BjZTPJkh5FDlNKEIu5kJGMQoLf0UIMqqJguTOXhCTzzjH5kaIyBvLKDpIo/vt9ZjMYIlUAah+ oozO+1FM9YwX0ilofRxSZPh+c8jgqzINHKSZwcqv8n+lqKgjtZoWqb2QUSlVWb7fReMxa1J1puL+ 2CuSzKhXI0Xs9Xw0HWTtPcsMbaZUnwUbnPPyzpPvYM6MT2/wV4nLVkLHyQ30/IF33G6fyI0kNVjP UtmEeKQoqZG6agpy7IUNSYHxH32rKalRTVTuV3tuM4yEjXo1UvX8oukgjW9z42GBwTSlL66AjT/6 hJByYzHmCz3Dwm+ts9cKk9saQsYHj+FcETmngZv2Kw7mGjLKRxEv8421cCEMVUobJcRRxJDuag7V RcJFnRqp+n3RdCJpdMvFw/KCCWzOvWKFH0N7zP4vqhbj53dJqN6iowvOvqkRYY45j9G2CgkXTKGn sqoUjTpSaKwi2koBx1q4KEZDyuWk5qKKGNffmQuit5I3oE6N1PF7NB2k0f3GRdoMba7UxKK40Gas 5EUxQvCVPNhlrJb858sEIYMRXQMHcZuh19IrfOp8ZmNXpEYT+c+Y62JPbCymbkUfHSEzqok47+wV CRv1aqSIvZ6PpoM0vt/YSG+i3ZW6q1hs6ANCYUOYf1AGF6qVNA5zjhCnPzNEanDwTM8VUVQaRmm7 4mCKk5ter5qKHHthQ1JgxODJkKOJOK/28KlI2KhXI0XsjWY0HaTxLRu/tBk0kopATTjLZWGjmCb5 NGMPG9xPkEzgfjNtlqCWzHOaVOO5sOF7byDDkMj4F6FRdJCa0+XVK6W9HV5hpUZSN0bkfDQ1mVhr RcJFfRopIpffMDaKipzolotf2gy27EWYD6C1yQhELUELZSY+PFOtICWUoVFeGwMhgr0lCmCFqKW/ 1C6OHoIGOLb01EB4nw1yGvhZDRHe2grQ5qIOjRBBx+fRswCcS81vJExjwb4O+U6Bwyx08qEQ3tNY mAWOiOoPRiSeqtaIyE8ZaEtB35deNzJU0pEZNURcJLZCQiHGKP72JG9G9TBpfmsrQHKh/lSGkOty 1OBpEH6CsfWsrtDWSo24ENYcskhYiPnSTDRT1upzO3BAQ+BE4EvYCVhAEoExeTrbyGA8MqMFDorY lJlbiDGr58/hoHpYUrq1FSAcpASrSB3Gv9FyQkhUv+VBmkl23nHHfSjgpw4FgtCI8779jXaBj9wn 0T3E9MjMT0EKT4UDdpKk1ooMaRjNUeMWoAiO9akVwntePbe11UOAyr1jK0A4kD94WBlCbhSj5iBj 63kesIlDv7HeTfqFsCAMYvhRPt0VUnH6Ehoyga3GcY3vBSVmI+JBYzOdrTpS6q5uNJEMRWpPzy2m +RxX96tJp6Zu7RURF+vVSBF6PR9NRU58r1KCkbCiwjvdVxM2isEGq06oYnmYhhAp2PCtNvxQn3mu SNjgrH4KzEr5GEJ5tSYi56jCxV7YKEYM/oASvFYTcd7ZC+Jmcr0aKWKv56PpII3vNzbSUGpPpccL SOvzSWGjGGx4wQY22E7QTPBHlmCDM9HpMLh02B/OIAYOUad/FeZs55iFpEYTbUCR2hMbixGDPs+k LKsm4ryzVyS5Ua9Gitjr+Wg6yNp7/qZQ8KcC5paXYUMYKcHU+ZcLQipwXyRMP8gdw3lTuPRPzxVJ bnB+VtXRSjH8udVE5EKQoXqQprBRjKkatPqdqya2aVzt4VORsFGvRorYG81oOsjaKxsPqyrtqfSQ AWl9TSq5UUwLCVpJJzdYRWzsSPHraXDWg0o/ei5ssPYHVWJDmH511v6kYDQdJPba/1aKGJgBMhvV RJzRtPaKhI16FQyqNprRdJDG99ubMl3Jt+/C4rfiVJsv+0BDMbp3dlqbjU+OhzlO7hjWz6+RkBtY Sm8hYcM3/EGGICjTfJqFRtFBYq6pUSmmrjRfyXsSPRzsvRpzYhgJFXVppdbt6jlxrLHnecFMrn5L hkPaY5y8KEa596b9jOTFF5/7MxNfSOGTxmZ+jnkZPRcm2EmivreYqPiI1Gg6SOw1LyrFLEynaqqI nV3e+HzMFQkXdSqY06KOVxFc3Eb3PCvYa+niFteI1hO9A+E/bZ2Z+PZ2MsUYSjQua9ACQgS7SFR3 Rcg54Z1/vB/RAw8BYqsZEQjnfSZZwztrIb6LKSeEgZBQf1ZoXK4ampVrVFh63GimC9F+SkfD/AHf nHI6LEZisw/EJGReSrsBdcEAPw93ioUEIf5VaRw+d0+kryGgcYjUKIKFIjUn8hZjFsqXUK0idj3e mSsSIuLTCq3b1oOTAaBvjT1mIk0Eevg8PI8xq6Qv6unFWIxRNvOMzjiutEngjIxCBZOv+e2NKCRV Yd0QGSmC5JiFpEbTQWovVFSKSSgfKDiaCObOXhCXFowM4tVKreej6cSy9h6zMXnBrhKVUEyL8tWp slGM1p/2nP+oszEeb7DtQlJKDD6nqOeCODEoxNmstJh/cg4oSGo0wUaR2gsbxdjuzKKxO6dqIsfv 7BVRZqxXK8W7EnujCTZu4/vtLZncYP1CJRTe8TOUDWGkBDF4nQa7fNm+sUtKbHCdsJ6DDSHJjTfW PQ/msvxWE10DZzIu9vTcYsTwz7vftRZnTcR5tecGw0jYqFcrRVcS7aMJNm7jWzYelhaq2HwwiSlW PiFVNoppatTTCpRzbHV1nAy+tOdRQzU/R4Wo58LGO7dhrJTjZIttys5oUkF5tdfGphhzDv4QF29K NdFBSvvFXpGwUa+Cmap6Pppg4za+ZePhlA3jPA4oYo5lHJZk24sUY/zwpSvYyQ2+0Jx3gIxJm/on zzF6EBI2OEidblcYGc+MZqRG00Fiz8+tFJMPXmKGjWrigK810aXXXhA9d7wqRhZNNKPpII3vFRu8 lyJfi1q5BwYNxTSa8kiEco7P0kK+Rk66lwXv/ua5ImGDy5RSdkaK2PlgjLSPpiJoqj09t1LEwGes QLTXz5qI885ekbBRr1ZqPR9NJ5a197gVnXaDLxCremJqlDs3+6YIcwxcm60XBHN8Wo0/eMUlJTYY zScPjIQNXwy9QoTJGuFV0UFqLmRUilh8Kxtk+A5ozHJtuH6EtRakmVGfiikz6vcoOkjM8V4+44LX i+uaCQ5pSklzsRgzDJztMRd4KcbgQlLyjqFKssBIuGAPSXrhSBE6x/VS1kZTkXAvTSGjUsTgb7RB RjUR5529IjwHVfVqpdbz0ZSZ4kt8r9ggRDtH6+wNfJpkd4XAvBOrAraaESo9DPdlqvHQcC1FhIFQ wRsvTSNE4D5mwTsSPQtoTho1ISIyvO70sX5DpMZGWdnXHNHY0sHFPAUNdaiYFwbs0OjZINbWk0tK EWbrhn5GTZHSaajkWkyTpFpEJ82yDkgJhZA9+8xjQZyvDNT+ek2xmFoBZgilfBQdJObaj1SKRt9f YqGXqiY6MH+JBSZqr4gyYpwKhFPr9ypapNG9Sgj2b+gHYLsgq3buRihkg7HtwcsSmOXL6QqKohkp /vAxu+SEETmnsZpblki5cOLcgqRGE4lfJPacFYsRg+9KP5qYqpJXF3tBTP56VUxs1PPVtEjje8UG Ozj04zNpqqPHyYxiTJt+eHqPIo5D5TjF4MoHlGGDIjmFpZGwwdE8UTVSxM65BUmNpiL+GW1P78hK UR3loDIjv2ii0ruzVySpUa9WipFZ7K2mRRrfKzbYwqFSE2n9sGFDWGLIT4tyjkAlzvz83HsUGSIX EjbogtXqF1OcnLi4ajpI7YWNSsmMOymNAa2JOO/sFQkb9WqlmLeIvdUU5MT3ig2+I6jfk7mgT7aS ho1izAZxXRGqmDpiR67ZYKIpzSm3ECYPjISNj69uiIwUsTPzK6nRdJDaCxuVIvUznchWuWhiLHZn L4iam+PVSkmp7a2mRdbe4241AzUmS33sUmz8686kxYjBVy9jl4MMjV1S6tIYeat/ZWwOYu80eJsN kZYidhafJDWaDlJ7YaNSxJBFxtVEnHf2ioSNerVS6/lqWmTtPWYjBRcJyyEl/CWlNBmR3BBGSjA/ xg2R/Ed6KxY5HDs/Y9j4l+e0GwxE3rFP6o9WeWAoUsTO1yGvmoogU3thQ1JgDMaZMrpq4oW82sOn ImGjXq3URrOaFll7ZePhQI2Ox+fM8Y6LgMLGYkxEkRx2kz3IYYMPIWhxTUM3D0MY2uq5sOFvV0CG ITUbLOWm/Iyig9ScVtZXig6T/akmI4rUVLsVX2tFwkV9Wqn47Re7ijaSMfe0/KQgYRsH9jU7pAVV VC2mCSJ3q5R1vkSOMFkgTGbwPeFkgRF5x1DK9ypVymEyUZXiL5oIvEjt8dzBICO7flYT3Tj2zH3t BTH361Wxi+eraWOJvRfFODW451+oUfSBNr0mAzE9pDlZePrL7Bt/aIXH+xwpfjtpI8CeadjWjYrG FKOPtaLOavbftRQaEHGojEyVEqtF2yesbkwVEOnrzwqNx1Tn/sbbhrC2njUVGsA5/zUn1I17wZwZ XqvBKPuu9Lsy4JSQB2t5rEjygU0m4mylfAgjlS2KULhAremxxQiA068gjMSiiD01d+aCuJlep1Zq Y1lNi6y9x1ykE2Eih5YYB5DmME/zoRhzXu1u3/zbEKRkMiryU0Hsm0ZrHuKvFPV0dI+eBWosVFRI Y2EdfYeKKiLIO3NFkhTxaYU2kijCyyBoXHOPmUgHou2VHkcgrQNzeTOK8R911wWaOFmaocWPT9Vp ZJTHCFtIsuLdX5tWVkTKBxaUJ1bE/waIb/PYChGBP9YHFVVElHfmioQKOWXmKrWxrKZFGt3zQosf jG9K2juuVMzGrGAJwRvXaXx9axxR0qemP/nDY4gUCRcssanqXCkfV5AUPlnRArzoUqTHFqM9eHfb o7FYp0i0KR6NGonZHDTNc+tUMVlRLKhcTYvE3ssGk+sw1ZfTFmtYmbwoxsvmS835j1yzp+xl7J3B J3VMnisSMr74RjC6VsrnLKR9NQUx97YXNjjXIClNJjpJ6Q+iibflzl4RPbderRT93Z2mRRrf88xQ 7eDSn0khzoeJjEAOgXtR09HySiQPfIiMX+qNpywrIFRwC2+KCCDC9UELydC4R80isRUiSBkJUTrQ SfGHygTroQXVgbXaQmGA0FB/RmZdXjWLjK3HTUUaTcpJX6yMLe64KgmF8D7nrSnYfHcc/DA+sGPv ecpA28wf7gLmv42MTit4VLJqFomtkFAhBt1M4jjU6qE/vbUVICTUn5HZKKLGXigI9I2txySkvdQo yUMGhLkGM5lQiJIk550Y8vsDxgpQd2WimR0p+dkFyDENiPSfRkSnFCyyWhaxKTO3EM77g4WrhvBu LQWQpXjDHyOTGAh4tdxF9eplePNuMKLSHWihQBAacd4z0Pzx16v+xJdbCRiR5akAoYDyKd1KZHQ2 wSOQqEHfIrGlpxZigO1LowmjeqguXaGvrQAhQf6gcWTYYhNbq2aRsfU8D7jaB3XIsi4cDnwfqCig bTLvXBCUH/3da8fIuuqBAf4dBtgwovfZULzrOliV4K6B2NEzg+B3Vs/5I0r4L8dOlNUOPsWVEcGe 1a6SAcbO49inIWC2LD0f08sNvpB+d+8kUXvmEYJozeiQqiC/NoDTWSOhbnic7TkUIZJZNWq7LrZC QCF+d2bFCXT1KDdubAVIAtSfkdGWT2uOGv8IQvyjKq5XbwFjXf10JNv0CEVwnu1pjoJ5O/RBgdpo DWtaUfLv/P7shEiX0X0H7HuQ0lVBF3bMtPUsgt9s3EWY7jFaCO3Gjv+d4DOBWAH6YZSauKoIgi5H 8yryDmSZz+JIfH9/YWhU+9/1MDqE/Nzs9kn/yEVFGTUYCQFsCUkzWClVMLeagtg329NzizF38NFf sZrUdVzs4VOREFGvVkoFmr1aTYs0vmXj4ZwDxW7W+DSftc1hMSoRtgfgL8WXb4TjD4a9qan/5bkg TYeu1hVTnJw1UIKspkVirylRKeqjvtZVpFk6t0EUwFFUJGTUqWBO1Do+iu6jWy4eLg1r5NxeiCna bDBaTDM4Tk+NlF3N8QdSDu4tzxVJZrDzNtRJCuc0Drb21bSI5ovQlMyoFP+xE39VxPzt1ZwTw0i4 qFMrtY6PogXWWhvJh3mh3ZIu3KnXueTSb8li1Ov+IpLSwdfCKUH81SR+aG6I0lscpHnBhFUaFEnx H/l9IrWaFom95kWlSBqW/XhO47Zu9NO3lcx97TH2EBIy6lUwmFrPV9Mije9VZvD9uLTMbJLwHWHo LEYM7DbCF8pRXwil2JGyd3zDri+QEHmnAZAPhkTKmcFH5SoeTdTHV3vNjGLE8OGKfTURp7zSa1Z7 QfzcelVMbNTz1bQIKqXpFRvvf1y983Ow46y5UYwE4KSk2eCaO40RGDFKyo1AnisSNvzt1iNEYRmh VbRIzDk1FqMl5VvQpjWKGI/fWSsia1TPuWMumDOjfq8ibQ+9RveKC76alNTk/qvJjGJMbE33wIm5 5AHXrocLtl/rOeY7hIQLNo+kfBCGd/SRkVpNizB1NM8tRm/rL/KRY9XE5E00MS8fTUGaGfWqmDKj nq+mRWLvZWZ8ei+QZnNoLFVXsm5lSC24mz9KNd8KR+RcA6sEUaerjAlQKtg5kkpCQvxHSr1bPQPU lgg0hEKqrmy1UBdoNczdRc3YKpCkqEMrND5Hj7PkJq5XKUHwKWzoPYcFYY6VLXgpmTgHmARga1ta THzMz28kKcF2kExBVYpBz52mIA7f9sKEpMAYNP14C6jGcNKkGd6rPc+GGwkX8epIsSZ9p2kR5mak adn4pSNhL4Z+SaQJvTlRjLmcjA+oZl0nUHZLSLlAIa4ar0jIoK4VdytF5Rzlq2iRmgsZknJe/E3q aYTlr4FAxtVe6BESMnK9yQppV6hjgVUUOTFuo1sufikw2JChX1MbMrrbin0gYM4MLptJZnD2IXnA XrKQwYmFZIGRkMGWkLw1lULpnaYgx17IkBQYYyl/lZDGIpr4zb+v9rJOIiRkxKsjFc+vmhZpfMvG w8zQDkm/CvxmXHHgzFiMGR1u4UE55SyXV/AHGYFUMiLPBWl7QX/ZvJEup1ukVhMZcbXXfqQYQ4uc kNBUkjQRJ1sBQzA3ROHKIGEjXhXDq/V8NS3S+JaNh7nBrsnZlsrdaNlQspimdry8RRHnK0Vggxcn 3jGAyB9G5J1Gaq3bheGdFhostZoWofmQpuRGpZio0ae5jybiZK057MdeET83Xg0mNhrNalok9l50 JNoh6dRHmvt1mhvFiGE2DzEtqmkM2JCU/uAS7uSBkbDBGmHKz0pRSERqNQUxUbYXNipFDExRmI1o Upy39gbRc5SW9mowsVHPV9Mije9VbjDjMvUAs+/qV9k1mbMftGWZS/NSr0hQX+85eirMyBQJG9wM pLmulaLmu9O0iCoLNOm5xZjg5YMXNhNNxMlW3rTLsTdI2GAqCK8Gu3i+mhKL2a+9x/M3mcPURkoP AigCWfMqG8V0BGk+Z8OOVrtJsxrvODuVMYORsMGX7NULM0CNFKOjSK2mRWovbFSKsednzUQTcb7d 2ivi9oa9CfbqSG00q2mRtfeYjcxmsZPS3wOhPecGrGGjGPXP9181nhS1fP/KcXJlbNjgc+eJXIi9 0+BN0oxQI8QgPUKraJGaCxmVYvDJJ8udW1ZElIw0rtYGSWbEp8GUGfU7ivAkCBrX3GMuJjN8kTfC 3PCT/RPaXakswO9MtFDQfvm6BYKVUIZtSRQDSQv2hKgLXiEdQLjRM0Bsmb6BGFz4O3Q8HTVEyIGP i60BQoMc8gsiIWIdn1fPAI3rVWPBXoyUfrqy2jsTOaoUjFzm6w8ERonLx8odIS9/XGPvUWI2Eh7Y NJKMEIZvJMKdpkWYaJAmPbcYIfgjdHq/8lEl9nlGk8Zisqd9nvPceDWYuKjnq2mR2HvRkWiPpEcA jHU1zaymczHaNM83Ude592CM4s/UacYzTwVxTG4vRVQxRdmWd/QsUGPsqDlCDLL8BTqep3lGEUFy 64/YHXODJC3ynbIVitdoXEUbx5h7uqOG/ZHZ2sVIl63h3lGzGP2fhxlUcD47j5Ocf4hv2m/msI0k LfjB9FqvFDUkUjg3igIgUmvJCm95UwRsV43OKCLwr1tzRdpo9kD/Sq3fq2mRtfe4oUijqT2UHkkg rc9fJCuKEYJbdOo3TjcmSn8jg9/pI48RN4ido/TwJ6TFRaQoKiI1ihaotXARIVoCFo3E3ChS5Lfm BtFz49Rgyub4fTQNIu6l6Xl7of7KAwnG/HQJ5UKYf05/6gezfI4lvxjL92olNGLzYxSbQuScBm0e vRVT9vOzqiah3PQG+gVqLVxESBFwG/hFEVFyd+PVXBFTP04dqYnlaBoELhzdb1xMXuwJSnrz4aIY IXg8QCX76XELUeZEKMsPOXgZpHnBdqyW0JHiP0ZqFC2AbxIRF4MRAbdQJS+iSFHemhskeRGnBlNe xO+jaRBxEXuP35F0puye5GAlLjELwm0MzQthzgvKJXUfFLK8ufzBDyupDNj8XJEkBp8WVOm1Urwt d5qCIFN7YUNS7hthOJkRTcTJbuZjj8FqkbARrwYTG4kGNqppkGOvbDwcrGpjpAcReMcnV8tGMaag AHGdQvbHww/iRAqiNDTyc0H6mnz4ogWxIV361bn+SvyspkVizyk1mGLIhRlMHubKBmKPprGnyIWE jXhVDK8mmqNpELGR557nBmWlAtb0El9bS/tZTONslxYkrD+3Q5ztxDU08nNFkhvZYxsI56i5IrSK Fom5viiRUgipUTT2I0WIsiVKrDkzjISL+HSkeNFs7igapNFh7nlmZBoF33S6PlQEIoA9hP/PQw+C zBF8Ddb0IhUIEewQUTaskFYW9AaungFqS08Vwns+g5oJs6hRhBdbIUZAaPiyQ0eItle2qgfFBvBw bT1JCN0e6g+UIs01oiLhQEwGcR0xmihlOJXiAHOkTaMh3bpBxAJCAvtI1OkGUnysHkgmavgjiD2T rZAQIbynmNbjWqWTHgLkPlSAsVUgJMSfQL4kN1GsmhNE4tpMeDSFg3ArAKaAuLcnJBRiDihTj1S7 ri6oFT2frZGZ90QFCAcqCfB5ZLR66lMSqyUIIjZVDlI24DyT3mJMy7DSQ3yZOx9bAdqH2p0R2RhW yyJr6nEeuANFmFVkuc5Mh+4gCQfFmJ73fVykFrf353fv/SIalfk5Rt08Z9e0cOWBezCUUjlHajUt UntJhkgRAusbaoRHE1FSgysDx94gem68GswXh9veavpf8b1KCDZtqE5muV6fqQwbwnCKGLhvx3b5 pZPy7/6YpQdnfo5xlBB5xx4ntl/wRzGxwUGJG01Bjr2wISm/2NynL9JHE3Fy6WfYYKMsmgYJG/Fq MLGRaGAjmhaZ+JaNR10o0uziUATM++jy7bBRjLkODv/yH1GeoQijaaQgSqMzPxekbHArfcbjkuI5 as9IraZFYq9VWa90Z7zgSSwNvaRINxpE0ZgbJGTEqWL5+oLNQYYV/a/olotfmooeXxUprCSEiz3S yvXf4YKdfcofZi+Q4g+8zLHXIskM9rxLPBiMsV9qDsdW0yKxZw4HI4RklnZxSRFRcqL1mCMxioSL ODWYEiOOj6IF+KUVHdaedKBI93YjEoy9L82LvfGIaWf90LTsfBrGUXKLgtZeNBQyTQy/9Vy44J4i eV5MXHB0Quywph5Ni8Re8yJSZDsFp/JJYz1pIkzGNFI59gYJGfGqmBMjnq+m/xXfsvFLZnxzogRz Wq7/O5lRTPtmfcpavb3XO4gTKXvHHTT6o0jY+Pz0GmMwZwa3iEpqNbHoerXXzAimGGarazQpTnkl NmKviJ8brwZTasTz1bTIxPeKjR/2XYaNP2yRyXtSjOkwFm34j5SA3NuCU8QpKXnHd7P1XJGwkUte VkhHDSy0ihaBfBQ1NSKlELJPV8M9tahwcWttkGSGfFopZ0b8XkVwcRvdcy4oQRybZsJYDRAVAxFA blbWqLEDCW6jV9pT//gpJiMEhAh2kqSkEua0yAWVq4da+mKrSWEI77lTJW9I1BA1U6try62FgdAQ h1ZofF49A5AQigtbzwoMjQA8IaCKMxfdoKAY82Dc2+CE4FiXfNTmAF90o0GBnwviX1ZjJReLK8WC SaRW0yK1p5gGIwQOrWYZOJqIkosyQkXsDRIu4tVg+v3i+WpaBDYc329seIyONBWOqlzNIjE/nZwQ xo+q2R8NzUkzqh6TweWPel8YmPkxDSyEyDkGS/4M0WAKk1MHqkuiyGkSZMyFjGIw7I9YribCpE64 2isi8sepFTqxjKKDNLrfuEjpiZdMu4cLfcEyXARTCJ58xSyfF0ruU33FOa7AV5iQAdLMoGBWxxuM PyghIzWaFhEZ0iQyRorfk02cyYxoUpy39gYJGfFqMGVGoxlNB1l7j9+TyQw2cqj2p23UB+DDhjBn Rj7uDhusjCb2ftydRs7PEbkRRUVS5xxAMcXJmYOrpiLHXtiQFBitsb/auJqIU0cM0B17bjOMhI14 tVLE3mhG00Ea3+bGw9KTpQKvyYkNTkCWjWDkBmu5/NiwQcgavBAnUhClwZqHIEH6orynjigmNghP 2TKaFqk959RIiQ3G/uY1mhSnvDr2Bgkb8aoYHclGM5oO0viWjYclBq28DzrjHSflMlAdjBi4aT5s cKAj7wC30adV/ePn4MdIcuODgZ45EwZnVNSRGk2L1F7flEjBBsSHjWjSJa7RRF0aP4PoOX6jeFUp kPF8NR2k8b1ig/unSEV5x+c/mhvBxIbmGGSXr6NkRpz962GDWJXCsCEkbNDiqv4oJjY4qSCp0VSE jK89PTdSioHZfBCthHlVkNhv7Sny2luvBhMbjWY0HWTtPWs3GD2xYGbvuAI0bARzDHwsJmyw1zax 86XysMG1i8kDI2Hj+69vV2eoGykdWrAUhUtmMAeBDdsLG5GCDfpnmWH4YU3EyfLl1d4gPAcb8Wow dr45GrdA0VTkxPc8N9j/wPyNc4PlXy+uoyEYPfT3F99Fkd0fZvZxE3OSUkHKl9+TB0bCRm7VWyEd WrBQFPFYEDtncyEjUrz+3z8u+TUGhAKi5JqTq7VBwoV8QtVIrd+r6CBMlEjTKy6YGGY8zi/FnR3D hTA813kEzWGQbdzwpMh1HsHNBz9sZIooKA2lvB1ypXRq4UYTZAQZeyGjGHsQeVkvmojz49beIGJj vBpMmRHPYSOaipz4XrHhi7hxzmWTulfeef8cNBWum7DKVe6aweClTG3FK2iRAqGCnSTqgleI9/tG D0zM/jPZamcSiGbCn26E0aghwvcbWwOEBjnk1JEQqVufYSF6Fmhcr0hgBwfFFCzwZXBvNUBBMFKC L3XzH6GBfHGAXG2n7pVOyo8RspHQwB4S/cdgkEanGKlRtEjNtcWNFBFwxU8yIpoInPt1VKSMvSLu R+rUQGKisYyigyS6//7Pf543nJl5lnP6HGcyorPRhJBPe2rlzcMQwvR3LnGOViS/vxAHpTFb88SY yOA4hqRG0yK117I1UiLDm65Ii2hSnNIuMmJvkORFvCrmvGg0owk2buN7lRmsAWGOJpzvrE5mBCMG vgGVzOj2JFU0XviltvVzpIGRpIZ2FTk1hDk12FakYEYTbASpvaZGMU17m05qaWsiTm4zT2rEXpGm RrwaTLnRaEbTQRrfKzbYy0FLLe+09SO5EUwx8DUbvydksJwidm8QoWPhVKxa+CJh4/OvlxhHitjZ WC+p0VSEpqz29NxIkRvM8klco0BpUpy39gZJbsSrwcRGoxlNB1l7z9+UHwpue/elbzaGDWGOwbeY 0GhwvDWhc5uCehZGbhFhaCQkZHCHgiYwiilMTjBcFC0AF7YWLiIEF9wlEi6kyFzwmUxGz2uOuIU0 M+LUYOIifq+mg8Qe7+UzLpiL8u1c8o7m0lwMxu9J2eTE+OarOolSQnLuLZd6EbeQcMFW7bSzxsQF S0wqn6pogVprmxEhRZCzXBqSSZGivDU3SPJCTq0UjI3fq+kgje4VF3y/Rb8BiclivSe4tPfSowNe BYyRFtzcoJBoSJExFdwup+KrSKjgVjiVJcHcYnxd9cBE/l1bJnAxGlt9QjHDMenR17qu1sjdImEi Pg0mJhrJaDpIYsNes+LhcJWJfC5ANRPulPSGLEYT6ctRyTRe+0T+l57SOctm1KSAEEelQVQIM6Ss YB5cykfRQWgf/ZgJqxTFRzrmKiLK9N4kYawNIi7Gp8HGbxgbRSeSMTdcPBysalOlFqn4pdiGz/q6 M60Yndm3tqFgl6nZRM6hAr00VGF+jhwworygqMm9UcEc5k+0jybIKFJ7em4xYvjjpbrRRJwcIAj3 sVdE3K9Xg4mNej6aDhJ7L9sLD39ERvYa8HwQAvABL4xyW3OSnztGnBXZoaMieHKC7SaqJgK5HlcP S0r0lBgFuoHYEQdG/DNyNbeeZdBgJQoOvWLcdvrvxB9XVmScVW2vXdrr/dh53laSnBgieDZW5Z0o QoXn6TaCf/eKppbF7RIT3DxCC+iZZXnNLomUDpIgLCadpTQqFnH3hJmEXhnGZ3wdxLFHC4H1C9Ox k38ndDtiwHHW01FB5Ndotj345R1gtwZ2iVyfU0rohRhS+NMnhM43vBOrrzWnrfzjpwhegCLRWMwL ioEUPutJ/BCEHzVFxlbjlxAQzmfCfPQQnrcFrq0A6ljWn5XRltAbNXBwjQtbj3/+zO9qV4xGCCKh OwsWwnuPOjHKN+31CxGglvFwjOIqvzeAezwNwrySWBlCZr+ENI+ag8RWkqBCDKgzdh09CvDGVoGQ EH9WZl2OGjwM4nxTXL+RkGlddoyxLGkS2DCWNfTF+FX+aTkCFvhSd4JmNSg0/PNz8GIkyfDFBdrm KlKEzSkKaR9NB9GIHU1holJaIvU6kgZYPjOlpdUbe8QtpAkRr4o59EYzmiDjNr7f2EhKaLulmyka aJ2p1ouxGG96PnhOvfrpQwh0CT55TVK8+TnYENKs4Jy+WoSRInaGamkVo+kgtRc2KkUMnJSnT6ER tSbipLoM+7E3SDIjXg128Xw0nVjW3uMXJLmhDZSu55FmEq9sCPP7zdYrNYlUrF+evyROSekV4WMQ yQMjyQ325Ks/HSliZy3nqqkI2VJ7YUNSYExD5OZjjeikSXFe7eFTkbARrwYTG41mNB1k7T2robRL 3xU9ZZ3umkpuFGMKi4t+nIEU9/zGxMk1uXp3tJfePUMQZTw9G8dUoapS/EGNHO2j6SCx55xaKUZY YmUVESZ1P6ysuUFkjkLaThXDqQ1mFC3Q6H57TyYz8skifikubsieAu2pdJVPXcc40GYZOSUPuCQ3 zrFakiwwEi7yAQjGFFKldGfxQj/wKDpIzLnRWCkGFb6rES6s6EtbuC7WyIsioSI+DSYq6vcqWmTN Pc+Lj3+u6ZHWxxuSF8Xw0p9v4CfgGGBaA67ITV5wCDA5IKR5wa6RdLnGRAbzPEiJDGs6SOy1rakU MfizWpARTYrTn5QgMWJvkLARr4o5Mer5amI/6E18rzLjs03GO9chu7LQ9kq9wYzl2b3qvNCr4eyn yNZbo1FYEsVA0oItI+l4hTkvnPOrByYCMG0gNTy1EKMpl+HwEDVEyBrV2nJSGAgNcegIjc+rZ4C1 9bjhnG6EuTXVgJrQoA9IShTTlIZ6AVLiI3thaA0kpaaUaig/vxB3choUeew4UsTISFXaR9NBai9M VIoQoIefdjQpylt7g4SLeDWYXpB6vpoW+f+EnTmaI0uSpPk6RR8hIzKWDHZu0GcYZti+PzO/LKrm QCMAor7KkKeui0DdTG31tfeYjWks2LGhao9lX7a1lw1h/Aq0aN62Chtc/eqc4M6ykPHmx2gjjcg5 Og8mCS5ShM6hBCmPIqdJkZoLGcXYxeuvxq4mwvy6tVfE5NepI6SWxOZW0SJr7jEXkxns2FBliDRn PcpFMU3waF8iXHCtTfKAe4JDxrufExkgzYyIF3P2M6kg7aMJeorUXsgoRgxsuXJmRJPivLU3SDIj Xg2mzKjnq2mRtfeYjWSGNlT61gSmivTxSb0nwZwZ+UglbORUG00guxjDRjZfDJLU8KYJNZSRInZO U2QT6n7uUojZt72wUYxpCI7ROcmqiZbvas9thpGwUa9WaqNZTYs0vm06H07SUL2wRybe8ZGislGM X5RFYbyjHPUdcYoTKX49jdP8XBB3JBqqaQWxUs4NTiVI+2iCnyJ4J01hoxgxfOg+JJrmaiJOeXWx V0RsrFfB8GqjWU2LxB5+Ps8N7k1Tt4E0E07uSVT7G6OS5rPEtsvydgYYTODHu6/IwIYQeceYjQPl YUO61E9wOEHaR9NBYq+9SaWIwSurRxNxWhPs114Q26PUjFfFxEY9X02LNL5XbHzoch6xQTkzuVFM 4wpV4PwKfM9eE+EaMahDxTu+z5Y8MBI23n0V6ZHS1jBLraYgaKo9PbdS2sunChw2qomK6c5ekeRG vVop3ctje6tpkbX3PDeYp6ThkneavU27ISy/KNfMmg0mq/Ubw4akxAadTfLASNhwOXaE6GfvFC1S cyGjUoTw2WrTdRatIxe85r2stSDNjPpUTJlRv1fRIjHHe/mci36AGWm2BpSLYoy/vRxBZrBAll0Y 3BEaLtjgl8LSSLhgb4kqCxiLlIpyS62mIOFeMiGjUsTAZh9nRjVh+M5eET23Xq0UCmJvNS3S+F6x 8e0wWevVfXBJDCDnBXuqkhfsAUkW5Do4ehhvNyVsA6GCLSOZx6oQFfWtngFqK0REhp/QJnhDpAbr HLXDIdrFsRWgSVGHiikp4vPqGWBtPfmsEk/T8ulnRPqDijssFOOH9Iwd5GfyQlEjZddoafPzC7Fv WjbTfqVK8aNRRkZqNS0Sey1OKkUIn627q4kopV1U1F6RpES9CuaOpJ6vpvv4nqcEA5l3DSS0XkCZ 745kMc3Uqf+ADT7m6zDZB6+2VAM2P6ZZOZAkBVN+6W2E4RwFaaRW0SIx1/ejUjRzzKihncFYNNFk R9PaC9LEiFOF2No7sayiRcbc0z0FbLP8q8GEuND9yMqMxQjhj6KDCz5znzz4QsqZwe3ryQIhJYMd IyooIEiYyOCkQsrPalok9qZsjRQxQLPJqCbijCam9SNTJJlRr4Lh1Xq+mhZpfK8yw/uOPA39wU1Q YaOY1gS8UY+ZFe9EIs4PpPBXAzf1MIPIO4ZunvQo5tTgpIKkVhPTw0GY9pYmPbcY1ZF7q6OJmkpe XewF8XPrVTGxUc9X0yKx96IjYUPHl0Yf5MYXm9abG8WYtuZQLP5S0nLrm0sMZiqULRoj+TnGpULC BmpSdlTKJxOkfTUFQVPthQ02LEgKM1wdZl6riYncO3tF9Nx6tVIbzWpaZO0961a1gVKNorzTkank hrDEwN5i26XKTiZw0ClsMIpIQ2IkbDCBonIFhiJFUX2naZHa03OLEYOPX5Eb1UScd/aKhI16tVKJ JrxG0318r94ULpLR78lMsr5NGDaKMZPPAQ7ylJLWN8gRJzeRhw0WipIHRsLGB68KulaK8jxSq2mR 2gsblSL1GXo5N6qJEeKdvSBub9arlUJp7K2mRdbe49zIIJ7Xid/DbLCxZ9goxlwyl5I4N9ggkExg XVRs8KrmOS32vvmGJH5Rr71DRoVYSojQKlpEvxmKQkaldELTTfQoYu7gzlqRZEZ9WqmNJYrwMogz v+Yec5EhPAuv+Rgxr5cKhWRGMR3y0woGmcGn7pMHdDF6czRE8s5BzeOCJDOYCkq/UylWEiK1mhap vZBRKTpxtmGhXYMya6JxurMXpJlRr1Zqo1lNi6y9x2xMZvg8Mq8Jl0MNGYGYV/dhfLhgUkHZw/Sd hMTFHJ0XYNcYt31q2fEI6cSCnlo9A9RWiIiM5uc8xRg1/Jaea7zaKpCkqEMrpFpCtlbPALH1oiOh Cvz0KAJp/HVKLIb7XsaBBq6b1e5faJCUaOCIjH7+IkkJdo3wHwdTo8gpC0mtpiBOWNsLE5UiBH+b kJSoJqK8s1ckXNSrlVrPV9Mije9508mwxa0jOfGma771ggRzDHwcgGBgw4eVCTPXbkMGk7ZJACMh g0ntTGhViuo8UuzMiqJFai5kVIqS3wuXkFFNFFq+e/zYKxIy4tQKrd+raJGYe5kZbN/Qr4k0/Xu5 KEYI3jhBA8R3v2k1RIaklBl8ZzxZYETOaRyld/1IUVNHajUFcWbYXsioFDG869Tl0UScd/aK6Ln1 aqWYuIm91bRI43uVGaqS7R03Pg0bqZzJbm6GUmagirrXcXKrW9jgkD2OE7mRsEHlrHKlmAvSSK0m StQghCJNem4xKPZXK2GjmhiQ3NkrEjbq1UqxFBDtqynIie8VG59vs/1P36zMe1KMGHRSADL4NkES 4VtCSg16u6SBkZDBJ3fU20BGpHzEQlJW5LjZDKqZMriQSLgQhhAR8MsipTGZFfGPO3NFwoWcQtdK ycqtpkXW3uNuJJ2q9llqCoky+OcfRwnDRW7JJARfhQgXuSSTKLkcMFxw+0/qUyPh4uNDq6pHygcV JDWKFqi1cMFpBgnRDnFizVxIEQgJgjl+DCqGyAQxh+tUMbV19Xs1LRJ7L9uMXH0PF9y26etw0FmM VQaP3inb/N1RomQrSRpU3TuPu0XCBffcp+etlE8rSGoULRBr7YsxIiE6KH8AkbyoIt/keWOuSPJC TpkxjlKoDFq/V9Mije7VO8JGPAVADcG+Qw9YtfvSmBoffifGD3z9BBmG896bSN/yFokioeLLN3VU Ct90XOGqZ/+Orb4hlaFi8dcxYaJ6aI6v1vw2GAkT9Wml1uvVtMja6xvycK1ExbEHEVSWKgfyhhSj U6K0MBfcApQkUGlBlIyN8xz71PRcuOBeZhOW+gMqPiO0ihapOe5hPFKUsv5oJWREEb2rS5RjrUi4 qE8rlVhgbBXdR/c8K7SRUuMpfGOi3ftZByIAig4zwZ6+pD4fp07G+uokEsBAiGAPCv9tMMXIba14 tnoMHFt6aiC899FfaKgaIry1FcAvFXV1HFqh8Xn1DNC4fiMhlbe2XbriZ5aQw2zKh4Uo2HgDzAK3 l4gq/pNOvPEDcQ45EQPYMQ3SujUyMj6dIJlVs0hshQSOMEiIx/laJemxepC+tRUguVB/RmZdjpoN gp9gbD3rN/RbetTAhBhTTyGhkGbEPDWv3zIdAlfOmwOunqblgBQB8oufgPd7If3GnEqQzGpZxKba TlaIaQffc3n0MA65tRVAtuwOmkckMbj9qJa7qH7Lg3Sdmof1gIFh/zAgBIWkHIembJLNWvnVPUuh EVkHGfwdAthZkhKr8xjsrkrhKh38vLRkFzt6ZhHy35dREX618IqE6drx3wlenuBbJdb71bGI43kV O1s3VFDSyul7lPn9hfmXo3FK9FwDo9+SVsZfreQt4FIWyRQJAwygVSCtlM8g3GhapPbCAicVJEUE /iIlLFQTjeidvSJhol6tlDruW01BTnyv2MiqqNh4pzsIG8U0yaFPuZJ+fO5FvyxxSipDMQ8sioSN rG8eKeobr7mSVdW0SO2FjUrRNbBjE+0a13lRkjjv7AXxq7RerdSszMJrNS2y9h43DdM+fv/T6j1s eKtY2ChGzeDNabDBynAyIZvFKCf+5Dn6Z+Ye20Sy/yhVU6WofCO1mhapvbBRKaoGpoXdSFYT1YY3 mR17RZIb9WqlKOxjbzUtsvYeszGtRD5HABt8yXtyQxjZRTH57hUrClXmDJwb35LKYMwDAuo2IckN Pl2lrrWYWgJq5htNQcy+7YUNSYFRTbpHZoIhiihC78wFaWrUqZVimHyraIFG99t7kszQBJwLe9fz 2b+4mOt50obMYD+d8sfVu9oRTYj5uSDNDD5AqA52pTS1ZqnVtEjthYtKQZRWIjWqsx6PTW6sFUle 1KeVWr+lBx/zN7/p2nqWFdprqWoJaa4XHSaKES6zzm4x+H59sv9TUmLiI8/x+ghJVvhsPURUiPc2 QlHEY0HsnM2FiEoRQOcGo4cR7J2xIiFCLqFppTaU0bNAY/stJ/J+aGdlHGMR1C2nEJjB+XdNalNc ce9hsp4bi8KC9voqYP0tt0g/f5CxGNzRKdxo6d/4NM8Uofnwx4yODkI7dhLqPLO+rEy9XS39e+08 zoJ5H9iikVKOTXgTfTGm/PTBY8X/5o22xMYR4MTPLS56LojfBw29uvWxUtRGkVpNi/D+S1PSoFJE wEZxmFtNzNjc2Sui59arldJmUXu1mhZZe4/ZmExg44iqBaS/uFxC/QYLoWD8BsTw6U1X1K0ejRG6 hPRKfOexIkkHdomkq60UVW6kosgvfpGa03MrRQ/77dEMPV40EeadvSAmf5xaofhNLKtokTX3mIvJ DPZpKACk9VHPcFGMELwbmd+AUX/zQFIig7XMBg7SzGCnSHraj0hRdEdqNS1SeyGjUsTAjVDOjGoi zjt7RZIZ9WqltDXUXq2mRdbeYzYmM9ipoQiY/vzbi9PYCgDmzKDZ1AiL+tWnrXkr2HAXNhhkJnIj SQ1WO9LTVkrzAreaghx7YUNS/kXfcmRKIzprYkv51R4+FQkb9Wql4vlV0yKNb1vMhxMS2kypG5dg g+1gkxvFmOjyB+lgg4tKkwl8xoIW28MfP6dtgDwXNhjIpy015qox2lcTdeTVXnvfYgxFGOPBvofO 0sRMuLRf7BUJG/UqGF5tNKtpkca3bDw8X8j2Sn/gCDY4hpDNm4sx3Mj0KjUbt5DiHSNrf6YSNt7z XJGwwY7r9LeSciPBiQRFtZo0Y2sk9toDF2Nw7Ys3YaOaWDGKprUXxM+tV8XERqNZTYs0vmdscEiJ flCpLzb4F+3GBcOpTLBStXHXIHlK7JLSb5W7/wYJG//oX8NZpEjPaF9NQdBUe3pupYiBjtS5UU3E eWevSHKjXhW7eF5NF2Tt/d5uSJpTpZqA4IXW8aCwIcy/KJ9nICV4U1gL1W/MjIMPEcEGX9ORTJGw 8cOFY+haKZquO02L1F7YqBSzDjQcZqOaWFa9s1ckbNSrYhMNXlXTICe+V7nxzrowz9PYU/qVjWK8 vl/awa469stjBboSSSk36Hn1XBF5p7Ga0myFqHciFEWIBLFzNhcyKsUkBLlxUcRi+521InosPqGq mCKv31V0QRrdKy4oseObvnKaxAAiJALgvIStcmguWcDp3TDh2RZFLSBEQJA0FVPc+TTQ6hkAx+ap gfDeu5PhU2qwToQIubWOrQC2pY0bdqiYgo7P1XOAtfX75k0Js0VDRil7mC4qC8Xo3biX3jRwhig/ P1+fiGtcm6fnKL30XHigRs+0nTGXkZFaTRSWV3tTtgZj+sHfHIGKamIzirQrAWuvSFKiXgX7zyWa arogje9VSrBFo2xwKQ0rHtJQjBj4uKfZ6KvAVyzi20+eggsh4YJNIup6g/FjUkVGavUsgm/z3GJE 8OHtihruWRNRyidxUXtBmhf2qcjF6+q5IGPtydZNSVNz6rdkvMDuweZFMSZhvU2Cl5LFn2QBl0zL Nw2O/BxTM3ouXLBPRB1vMXFB6Sjtq2mR2JuiNVJMxPiTr+RFNXEUQNov9ookL+pVMOdFPa+m/x3f q7xgn4Z+TbzjUv/mRTFcZ17decHJ+uQ/FyunW2FyP1lgJGzAHa6LDekSG5xPkNRqWiT2zOJixMDr ij0URBNxRpMGZNYUpJlRr4op9npeTRek8b1ig40a+j0ZPfMZneZGMWJgg6bZYAd6JsLxMmxwaDl5 YCRsfHOHgtmolJoKS62mILSvtafnVop5KZ/Yho1qYjffnb0iyY16VUyx1/NquiBr71mBwbZJprrj Hd8AMBvB8otyo7fZ4KhDY5eUMpcrRJIH3DadhQA6EA78uScxpjg5oSAp2IimRfBunluMGBiwoX01 EeedvSJho14V4wYWR4O9ahrkxPc8N1CUGSTm3bSmoRZ0MRp/TomS8RS0/zwMoaH3ygds0NolD4zI O1WmXkVcKdWcllpNi9RecqNSTNBwofZVEz/2nb0gbm/Wq0pdPK+mC7L2HueGhvCSZj5FNRKTrppC uMGYr9J0hPKNOZbEnokG2KBpSG0lxN5p5aqXH1aK0VmkqknZUqT2wkYxhuK8ru4/qomGwBMbx16R /3v1qtglmmq6IGvvMRsawkv6g/s9zAafxBk2hJFdlEP6arTYeHeNTPvIpl21JBoi+bkiyY0byNVo kCgiTOrTIExKSpEeW4yRuDcPk2QRosO7s1bk/7r/qE/FLrFU0QVZc4+5mMz45L7hcPFBexl+ihEC CxbmgmuSUn4y/xcu+NZKKjEhzQx2jKQXrhRERaqaFHgRvJOmkFGMGD71HWLIqCbivLNXJJlRr4op 9npeTRdk7T1mYzIj4w1mDdnCWzICaQjBR3CUFx+cdIMxrxGFiyxEFEhaME2aLpiFJAlprk9vUvQM 4ObMtkJEZLSBz22zphGtRifArUYDRKkpEBrkEJqKKej4XD0HaFyvmk62amgYgbT2XCUlhMV9lfmi 4dPTfQSdjVkaGFmkSGhgS0n63krRhN4oEjFFai5EFCMCXwlORlQTQd7ZKxIq4lQhBV6/q+iCxByJ +zwh2K+h35FFX/0jXBRjcoH6R1Sw3UaEMTFjafoVSy8SLthUop53paioozyKREWBWPNLtRg/qp+H iioiE+/MFQkVcsqZIikSV9ngWKrpgjS6V1xQXpLK4uKNrXDhohghUF+IC7ZoiTCiRIg8ZxEijxUJ F+zoTTUhKTeSHE6Q8igSFwVirQ1FMSJgCh7lGo5ZEZFfzfltMBIu6lSlFHn9rqYLsvaaFw/mPZme 1G4S/ZiaKSTrxcVimivMK8Itd4qJ4kBC4oKPvmiiuEi4+KFRv0pRQUZ5FImLArWWV0QYqmjz+EaL uagiRuJXc6guEi7qVLHx+2ga5ET3PC9YhmSBAmk2x3zqk47k2mJM2GqGWvxyIjTp/yUpkfGZ56ga hSgqDdu8yhjMicGpCmmvJrFRpPbCRjH69q82lNVEP452NK29INi7eFXs4nk1XZDYe9FiUK/6mCds 0LFPZhRjppA96bb7xs2jTgS+6xo22MqeNDASNtjGpm6XUUakmPCLVDWJDSFoqr2wUSnqPn8ZGV6r iWnIO3tFwka9KqbY63k1XZC196z9ZGTH1RHxjnnpvifC/ItSYocNLrRr7JLSb8U1rpIhciFhgyMN 6ueKuTC/0XRB8G6eo3yPFDG4N4WNatKk7K29ImGjXhVT7PW8mgY58b16Uz6/PZDAO2IsG8X4Rd/Y 7a2c5BVo7JISG0x/Ng+EhI1UVpARIeryCEURjwWJc/PYSmGFu1X8ykURUUbRWisSLuQTqoop8vpd RRek0b3i4ouCys7NiETzpx5ZMGWt4/Cigr4+gXs+ASb4qJ2KL2YhGMeECDaUwFIgF498QlsyVSMi ijBb3acWYjbCX8wmKaqHYZDmLmrLMQsIDfUnMoq5LlfNBRlbj1+PVN8w5UlUZH0JoR4Pgu86DycK PtlvnvCQUQ+bnSUwkIsE8ZxtJFqYDqR4s2e9Sg4QO2JtRPCbT/I5DaQEO4TGAuux478bfV0JdJyt kgOMHWL/z3//1/88WCyTLOWMfjc4ZM00PWghpoy1dqXoOTfl4LVIJKfe8xDTEDyU35/J+vSy/mgV oXEMQYqrRcEWianELwjOmBTwXgVYjJ4vcgzVx1aB/P52Z0U2hmhxx3cT1fN3gO6caZJQwI8tChbC eRas3BywYzj5rZ0f8utvnoIDgHDwjzOh4QlIEXP8QJqjZhG/b5zK91MrhPe+7RMOokcB3tgqEA7i TyAcXJer5oLAt209fgcy5qLe8hCPPNDgLyQU0hQxu3eVB4xEkwh8I8gkePionBcQEvjdUlQKEgka PBJy1QxiwrFVEjLCJBH++EPYkBA9BMjg42IrAP3/8WdlNoqqEQnXuLD1mIQ0BGxG8DkkSKCL0c48 KSjGZCglPAjzHZRM/pnfJKVcyPklaBDiEQMDMXbUHSmFzWGJVKvWdEFqL69EpTQT+ZmKKZqIku74 am8QcTFeDXbxvJouyNp7zMakxN9vV/VI68PPYUMYPyZsMJtmu5xxTdJz71C8A0gGGFFUGop5xAxD /og0bHBg4qJJbAhxWthe2ChGDLoj+mhSnFd7/6EhCRI24tVgir3RVNMFaXybGw/HGsxc/HiAQDHG B6DLRjEmX7U8o1+BC//DBrfppQHngsjkgRBnO7ugmMt0bhhzxR2parogseecoi6PFJNw/vAGvEYT cXIMJOzH3iBhI14Vc2tRz6tJbNzGt2w87DaY7/RdbbwpTEm9tb0oxt7Cf5ROYoObiDNqYNtYvPvJ c0xFCUlufPicCA2EMGoHnUTw+KOaLkjs+bmVIgZvkNW4TuuShMky/DHn1DASMuLUkdpgokhc3ER3 Wo1fMkPXdTtv33VRf96TYkTA8qm5YBChPCBKpNyl0MXruSDlotNTufIfJrhvWDLVIyaKwISs6TTQ Yoz7O9qXGkUoETWUMTVIeIhDgyns3DtO8SE1A5zIXuXEPyp7S3M5Px2p86wYlH7yGignOPmacQdX BqdX4TYJPceknJDkBOsFqR6EOcaPSFWToi7CkH+eW4zZIG95572wIqJkSHrMOSeMhIs4daR4D2Mu isTFbXTLxS85kRVyXHun60xKBMJ/LcqICS46yu/PsDmuMUh2T2kgRLBZJCWEMYXNnUbOCOs5QG3p qZHB+W7hiBbFdzHlFDEQEuLPEaKRtymrcT7wt4sXR7UUPGwi6O7pv5wOfLIgHCzGCgEHKkkQKm31 afrxvyWlXP2b54qEBXaJqMstpgg5TCHt1XRB6AmlKTxUigh0AS7tZRQRI2Pyq7lBNME/Tg32/mcd j6ILsNYed6UpLKj/uNMsXOiLMMqHxYiAPSJOCG7vSJASEhUfeYywQVpXsJlEPUww/9icR5DyKhIV RXBOmkJFMYoVf8fLIytpUpS39gZJVtipgS5+R9FG4ryoucdcpKxgOPTjkYJm/zgUFC6K0QvxgUE3 HKxhpwHwpwNFBmeCkgNG5JxGYV45ZL7SHxgkdM4jSKqaREaR2gsZxYiBL05fNBEnroT82CvigpMi 214dqY1mNS2y9h6zMZnxwTcZ8RdpfbMnbBQjBvjGO4rQucmNr1OoO2FIlueIXB8NChvvSiSlhqWU 7hyvkPZquiC1l7erUkzMMgsB+xrgSZPivLU3iOyNV4MpN+r5alpk7T1mY3KD85MqGJgl53scZUOY S056Pg0JKeV8rwpx8mGwsJG7VwYJG9nNO5hLxxtNgxx7YSO7hzVdq1OqGeppnx5xMkI79tyHGAkb 8epIaWOoo1lNizS+bUEfdiL8jL70FDa4f3XYKAb5OiIqNvIFLOLkH3/59aA6zwXh9VcMdINpXCTl plAXrzo3rElsFIm9NjfFSAlNZxxNxMm4OWzE3iBhI14Vw6uNZjUt0viWjV/6E276V+GghZN/2tYr ncUY5vFGmw2++ZCSk903aVX/5Dlm5ISEDbaypdcRRnlBGxiparogsedGdKWIgXuhzUY0fTFelSZ1 HrFXxPZome3VYBfPV9PG0vhesZF75mGDWzazkVWLmvxEetu5cl9jZmoNLoZzRcFe9bDB3FLywEjY mMVAQwr940bRBYm5pkal1Aq7v6Jz1AILUXII+2ptkGSGfFopZ0b9XkUbSaN7xQX38erXRPqL3iOZ UQwuWLM0F3znXvlDA40UiAZrfq5IuGC3SPodSTkzKFVScFmTyCgSe82MYszKcTGKMyOaFLs0HXtF mhnxajBlRj1fTYs0vudsMMI2FUy6JC0CaNmElR/lxD8uhDMPtKLxK52G/rRTDJG5PmFF3FioT60O MaA/tTbjJwrwrjNaVws0CoiKld9wnZ/cfyYN4sYRsZ/VgZ/8SeKujWfdhrZJerSANIsZ7jYWw2km opxm3F6el/5NUvpBGJPkxzYivzRU8nIh9ZmkFBsnKSRVTRek9tJtREoBcKjmokkR3tor4k6bytJe Hanx/GgaRGxI028JkJJCE5IeMiD9hw5Tr8NilLY/TFAozd4YIutt4FvEajk1MOtIQ4id01yilwsr pdApU1LJWtEFqbmQESmFwA5fyB9NCvPW3iAiv04NdPyuoo0kmSFFv3GRgkLbJD1w0KiCnXPhopjK fI75yCyftU8e/JGUyOCgdIpNI8kM9oWka6kUb2ikVtMitafnBiMGvuCazIgm4vy6tVekmRGvjtRE czQNQmY4vt/YmMxgPiDF5t8/3e+vZTxjhMzsrn+Ev7n+TanpEiPhOVmENDUwmHajUrTed5oWwTtp ChuRUgzMrTs1oklx3tobJKkRrwZTbsTzo2kQsRF7z1sNmqYUm9y2OrkhjOzS2gk7IZQb2d1Mp8qZ ueSG90CrmzWS3GBPsjqXlaJfv9MUBBkWaqQpbGTnMjEgndyIJuL8vNqj2CwSNuLVYGIj0RxNgxx7 ZeNhscleCF9siTQVwrBRjLaBu7XNhq/gVJz02VoL0pBIQ4Uijkq9NgXBYnoH2K8kqdW0SOw5pwYj BgoJtdCjiTgpNlLgxd4gYSNeFcOrieZoGmTi++1NmXaDk7YpNrm0OOsh2k5pTC0lS5HKDUadyiBN DrXIAE0eGElu8O2XFBDC8M5jRUmtpkXUTqIpucHYFCnFwCd/zUY0Kc5o0hDMMkH83HhVKZ4bz4+m QcSGND1ng71kXFiBv0hzs5Vb0cXoRDXiFht81l7FJmxISrnBzWXNAyFhI9ufVoiiMUKraJGYa2pE SiHQL5sMK1KUt9YGSWbIJxebkoL78fsoGmSie8UF5lNhcfNsMkPbKY0RAhOJ5oKiI3nAVpNwQded LDASLpiDUr/DzKt0KTMoFSS1mhahMJGmZEakiIHEDxnRROxM7kiThoXSVKSZEa8GExvx/GgaBDYc 3ys2cuIDYRUK6l7ZYtlqiU0dmkXWSfwOSzgCGd/mWLgAu6ZRk5cWoUKYqGBNJUxEzwC1FSKQyfQ+ N5vw0KhRhFGjoRdqBtBT49Bg67NYQI+z5CauVySwLUOuUvbqqqGwIMwdCfe3hYa3nbiVlH4iTtjl 5+czQMMDv33mr4yJB04k3GgKcuyFCUm5seD6sVARTUTJlO+x547ESLiIV0dKO0FvNQ0y8S0bDzsS tkqyzy/e/aWyCBvFWGrTGrd+g4lSMuLim6fceGB9uKCZzeyVMRfn0b16lKVXa4opmNOCozTpsqNJ UV7tOS+MhAv5dGQmEtJCepwWt7EtEw9nLLR50kMJDYc4WBomhDkvWIYKExxZaRZISlxww01zQIh8 I2BVNGoqWJuSFI3jnaYgYX6eG0wx8EtfNCnOq73kxTzHgMFeHamJ5mgaRHmR554VGGyfZPLU3rGJ Otu6F9OyHj+j8oK7KjqZrYst1a1w0iVZwAXF2QFE68jIxWwYExvklaRW0yJa1UNTMiNSxMBkXdiI JuJ8k/Zjb5BkRrwq5kxINEfTxhJ7/GrPik9tn/RggvadA1yes1iMGLTYLTaYKkkmsEcv3r3lOdp3 IckNvhkTNoThHUOxSK2mRWLPzw1GDNwYHTaiiTiZFg/7sVfEz41Xg6n9TDRH0yDkhuN7xcYP6ebc gNbJjWKsa+lmQLHBZ+212ZbYJaXf6j3PFQkbLBWl660URWSkVlMQNOHd2VUWKcXAhha0axAoTYrz 1t4gyY14NZjYiOdH0yDH3rPc0FbJbvPjHt5sdw/m/OZi97DBSbe0CHzFQ+8OE5w857fCSNj4x+Yf c1YpphcjRV8ZTYtQhUqTnhuMGNjtovaGQaA1fbGP5dbeIGEjXg3GNKmjcc9UTffxPc8N9kry3T3n xqf2xqkVXYwYKE3JeApaLnhL7JISGx95jhpZSNhge3uKMGOKk6VuaV9Ni8DGPDeY2GCzr9mIJsV5 a6+I25vx6kiN50fTIORG7T3OjQzhtWPSgwnGSur4wkYxVhX0VlDQfncQ4s5RXHzmKUZr6mbFhSaJ U5tXSCcTrLpq9u+aSlpEhmTnbKKMWQ19B9vY5M+xNYgeG48G0y8Yn4+iIH4nHdmrrMjhWnz7+Bwa AuG35zGoZJmicuqzSBjXcuMPPgsIC3iX3rdCrJ64Axo183cstRuxiHynXjQJ0aL4bkwNEBLkDy9C hEhaDljJ1NFTgFRwVK8oYAOHXEeaf7gL0R5LY3jvgQG1mxY41Wby/Qe1GxqmWaRIWGBpLb2upHCN pvJW0QKx1tcpQgqANtNERJFivJpLgggJE3HqSI3fR9MgjY7nnhUXmGOzsrng43fZ072Ylpf081PP 6cgDrzqf/AkVPzxFhhQJFewqUZe7UvSh0T16FmBGR4r0WDDXUCxJhYooIkgWf445J4CRUGGfjtB4 DRNS5Cy5jW2ZeFh0arulS2sm3zRtnQZCGHHSqmuyWVTQB4QLT1urAeX+v8RtJFwwYklRVSmfRLjR FMTM217I4LwCuoiBy4REPaMwayJOrglK9yV7zgsjISNeHamJ5mgahLyoved5wfaSFJ2+Di5sFKMK YILcbHDphrJHPT63gIgNNnImciH+iTVco2YcKf/obBOV1GqiTL/ac2sxGDHwg4aNaCJO7qoNG7E3 SNiIV8WcCfH8aKL3udo7b8kvucH4lAjofLiPtw1GIEKgWtK6FXWsLxqHjMz5aajmoUeAJAYnA9Tb jgwFOTLOsKoJMrb6jkjIacGdhSEieggx7cfYCmD6xp+V0T7RWzWLOK5fSUgPSjHHHiA7pq88KSMG ggQuKgsJnIXO789FZYhoPOQRB3GxzBsS2IijcsuQ04FjCGJl1UDP1ZaeWogZCi595VfVcE96CDAb w8ZWgeRC/FmZcZlUkJoNgheKiXVcft5KaOynK7kkTMMfEgLJe1UGygTdHObmkY+8AjC/7qcUMYAc Y3Tm274KKT6OToiVVTNIbPWViJC8123IR48CvLFVICTkJrOVSRRudKvmLq5XJHBCTx0yk+/6EJI2 LOKQMdynXhYLOOSPSBIh389Tc6HxjzdqFQkPnAPMxJ2k0ET/cadpkLGn5xbDcz5BYiaiiSipl6Vp 7BUx7+PVYMdzOK2miaX2YP5xZZkpb80ga0UQaS55zTbWwWCDq8HCBt+5z1Ey5tfCxrefgx8jYcOb vAdylRihVUT1cTXXN6oYg/Ef9xIM8DyJSSreWvsqIms0HviE58H8KzaWVTSRTHSvuFBdZyq45ccN JY23R+IEoBvPXLz97ciBYVeYyJ2GECEgRLCtJP2sMFzTgQR1s6ungBybpwbCe2a9nRNRowgvtsj8 AqGBKU+l8wrVZzKiegZYW48TYhpKFn3wXZ4xUGojEUzus3pnGli3SuJz2Xto+PFz+vVB/MsyGGOg 7bayUozKIrWaBhl7imkxQuCzkPA3mhTlrb1BwkW8GkyvR6NZTYs0vlcpwUYNIsA73sphQ5h7Pcrm sPEpIUJ/l5Da0D9+bJAkBXtFUmpVyocT9BxTlyhymrAsczEXMorR+MuYRnXSQ5CU31drg2gzK+Wn Fa0QhWVUV83+3ch+42Gygn0beTeomYaHYNRk3FkWHhhVJ+/5qlFc41LlBC2kWfHN5gfTVSnmcSK1 mgaBd9sLEZGi8eMAuZMiioibQejV3CBJijg1mJKiwVgRlAZwJ1prj9+QNJnMBmjeUznB8ma2sg4G F3wXUGsJVKya6VVSsHSZjpX75BK3kSTFP83PXqQ606vxXTUNMvbCheeDVT1wkbOzIoqIkvviruYG SVbEqcHe/4zjq2iBRvc8L9gbqd2Z4oJbioeLYETgOk9csPdIWaBRNVJ6QThAqbiDNC9+GCAuhpSP JkhqNQ0y9sIFBxjkgyhPYxFFRPl5a26Q5EWcGkxcNBgrwn4A54Wj+42LyQsW+FNikvTZuAktxoiA LFRe8Pr5G2BEThbqrdGoTM8NIuc0Luu2x0pRL95pGgQubC9cREpcMEd20USYvD5Xe0XM/Xh1pMbz o2mRtff4LUmLoU2TmFNmUFcOG8Fgg5kpTbbSLDMocOzfkhIbHBlJJSnE3jEu0ykOMVQpn1GQ1Goa ZOyFDU4uyAeG1xSssD+aiJMC+mpvED03Xg2mTGg0q2mRxvcqN3L/tLxjKci9Kk0bN0nTjxBDJuKp 4Xy5teKUlNjwFdiOXEhyI3dSHynK8TtNQcL+PLeY+4bkRjQpzqs9VxhGwka8OlLx3LzSy0jTIo1v 2Xi4WKgFbHouzRQwFTNsBIMNvhWS3ODr9skEndPh16Mz83NELqRseEajiLpHvmqqymT1DDLW1BAu xszEX592Zvilqkc3YaP52BokTMSjwRS3vIav1XMf2zLxcGxOEULZZCaodBmX4ttgRMDn8MIEB+6T /xw51cCUMsjPwYSR5AU7R9T7BnOUrJyEi2qisr6xp+cWY15CG4D1lkQTcVJCZmU89oqYeYprezXY 8Rw2qmligXvFx9v8uM2YFnTSgk0eGaR7v5h/TTaVaAre28UUE1EiRJQaOx6gSUGVmcpKQg7Sl/uL 0+ihomwK2laIAPJvyQ3S4SFqFGHU1NYASQraJv2EK1SfxQJ6/IvexPWKBLZrKBWp0L5Y4NMoXfsv dSoY99lgEhq41sYscFhfAWo45DEHUzpCkhHsGLmR8lEE5UgUOUfIsYu5EFEMpdx0g3YW4ayJIP/c 2ivSXsROHaH4TSyraJFG9xsX7kTetIlSAwi44D0xF4sRwps2CiglWMFI4rPMHzK4JjW/v5B2Iixt q/xiiiJSPokgqdU0yNgLGZy9kA/EwAodlHGZkjV9sdBza2+Q5EW8Goylp4lmNB2k8f3Ghl8PpKkV +TnxjsGXN+sVc2ZwT6JmoGlTWQJznHxQUc0HbZOfI3IjSQ08T8dbKZ9E0CQvbkZTkGMvbHiiW2zw 06ihhARrIk4CPvZ4kYqEjXg1mNhoNKPpII1v2XjUiSD9/q1hBGx8sHde78nBSDM+A4539Eeaz9Sv Toug3Rz0fn6uiF8Ujde8jhgpnvNZhBT51bRI7DmnFiMGrk7mudFEnN/y6tgbRGyMV8X+c/F8NJ1Y Gt+y8agjQfrvPw0kxAZfENXkxcH00nlxjMqNeiJsIGXv2IqcPODKi2xPpSFjM/tKuZEgFmlfTT6d cOyZxcXonZjdNxvRpDjlldiIvSJ+brwa7OL5aDqxNL5XbHxyob3Z4JJh78FBQzGa53daRf0KfAAq 09dcWKyWhNGUnyM3jOi30loaAyplUKUoIqN9NQUh45l8kyY9t1JULbo36mgiTi70udobJLkRrwYT G/V8NB1k7T3pVpH+4tYGe8fcybAhzL/oV2aSKeG4hIM4ta/Hmxc1bPMgpEjY0C0NbiUqpWMKt5oW 0bxsNy8uRkWrvZliI5qI8/PW3iBhI14NJjbq+Wj6X/G9yg2+o6UJM6r8P+xxyJtSjOqNXRzkKf/w F7lgg0nr9DJ8R0t5UCRs+DtaA+lH5ztaUh5FTgMhJt/m9NhKMdjha5kmw4qIkg1fV2uDhAv55HY1 UkRev0fRQRrdKy64glcljyaFWP8MF8ISglZAxQXnyRM5H4JLH8OXQFRzyQGQcMFMZmqUSumYwq2m RWovZFSKcTD9lMmIJmLnNbnaGyRsxKvBlBn1fDQVOfG9YkP1AT0bt9AOF0Y0uvGMrJYMVCISNx+h jWf8XPt3eGCeLVVFZUjw9CTV0r9jp7WJJXgl+DKHGYgOYqOTO3bm78QvX/B3ZIhW+6SovqLFf594 nseu3ZMaPCh67slzJgRDI/EzDetM4FBbwpWQig2+apJ6kuXDSQR2i6jLhaRI0d5FeRTxei1Sc0mE ShGAzk2qiYgmhXhrr4jJoxX2dNwIvW0so+gga+5xe5mqkyJAk2rigsvMhotijPs5MWUu+KB9Qucu g5DBsm5+eSF2TgO17n+sFF1htK+mRWovZFSK7odrNqFsNBE6fdHV3iDJing1GDPEE81oOsjae8xG qk6t6GkQITZYDW1mCHNmsGURNsi1P77KjR+dDXXxjm+8JA2MyDsNn7r/sVJUj7eaSA0hx17YKEYM H675R5PivNrDpyI8t14NJjYazWg6SOPb9+Rh1cnoSuNHscHSmNdQ0VCMgkKbqGWX/yV27tBSf8Jw x8/BjxC3mAwDNKrUi2JMv7E3mumNi6aDxF5zqlJMVnCFq3mNJuLkgHLWhWNvkLARr4pRda7no+kg jW/ZeFh1MgZny1LY4HScq87FNE2IK5BB6ZdWg748zvEpBxWdBC4kqcG3s1R0BnPCc0ohRacVwUUB TW72scWI4M03hml3hdcF4eJqzplhJFzEKXFhKSJvLKPpIGuv78kvmcEVZqo5KXE1YFRfyqxUMLjQ sjJc8En7RO4RaoZr+sgjcRsJF2wnS4dbKXy6UQQXBWpNjwUjTkjRAFXvWxR9UZrLpzGHl0XCRZwa bPy+aDqRrL3nbcb3lw8oaReUdho414ox+OH3ERdcGZf0p6SUcxoWeeDB+yMkXHCBaEqNSvHmIoVz VUTcAsy8rYWLYrxSbBY3F1FElP9uzRXRu7VOrRSR1+/RdJBG99s7Mr3JPyat7d0Hs7rNi2IM1lXg YJZaIdnPLuR0L9wjnBwQ4hdfg6f0LxVyVy/d1QMV9P0XY6GiGC0rS9W8WdVDjN+3xgYJE3FpMGVF vY4i3A7CT0GiSdNvTExPwv0AqliQ1gmovCHFiICTSx4J56vRvBCsDYQKvrSWFDCioGis/a3nI6XF HmsfTZBRpPZCRjFi4AzARZPivLVXpHkRr1aK2Ov5aDrI2nv8jiQv2Fqp8kFk8OkHkzEQEbBc4bT4 0SqOGgeEwwXSCzQrWDxtWkSIkZFVjx6YCFBbISIQ3jONZh6kht8SHm5sDZCkiEODcU1Zwxg9C8QW Hj4mISmhFXPPu7vVHhaKaV6P8YPSTEfTxQLHHsPCpx8bRAFpSOFFxpWifIzyUQQPQpyvdBKzLbVS zFYxWXzRRJC0K1d7g4iKOjWQmKjfo+ggje4VF+zk0A+gydLca44GYX7B+bppuGA/rfKGqRhJZXhm mSIhg/0mGgysFOXjrSbIKFJ7em4x1aTtq6uJqeM7e0VCRr0aqeP5aCpy4nvFBgugKQN1F57aCm28 MkQxxSU07lP4AKroIUzPU2uAZpEAoeKdC+svMsywXdUQdZB4hho9tUK4T26BaIOV9RDira0AoaH+ GMLBdXnUHATWbev568H6WqpNLjR1ecUmvEBMGXuVgkRkqS4k8NVJPNWAyE/xj56/1Thttj9ahvoZ GUIeNUQsBJ74R59aiOHaJzyLhOph+eTWlgE3letPZQi5Lo+ag4ytxyRMQ/mlxWI51rJKuymNMFHs 9Qk44ASnsoUESCXxEwn/nTTgJqRUDZaA/lsdEFDEdpIERRix68sKil9asMOpm0wT1Y7+VuIcT4wo A+rr6DhI7RD7bxc1I8sscopKfQs9b0EhgmeroE3ySXv9cgSr1TqqiT95yoDzmZ9ZVxdVhkhokiIz ag6CX9gKARIiS3BeFyGLgOohPE3xHlsBQkH9GZmNImr8XtzE9aopYOOCSNf8J41gWBDm/KVEVjlJ ncaeCUfI2d/0nkyX6Tkm44QoIpoOdl8dTESwR+CqqcixFyYkBcac1Keb3dXEWvLVHj4VCRf1aqW0 9/NGEz/ybXzLxsNxBlsodW+f2WB11mwsxvvpdQjY4JZ7JQ6xI+Wk8H1/g5QNz9RUyFnBJQcpKKMI MopAvsxlG1ExyNCJadEaRUQpIaVFrRUJF/VppeK38yuK3jaSMfe0fNCWSQ0OyAwun4UL4lyMBlsH opQZlJbJA27ZiHf+Moa4MJLMoCBMxyrMZTLPS/togowitZfMKMb8E6W3uKgiNlxczTkxjISMOrVS 63gVnUjWWpvJX/JCu+tFBRto0mGyj9KQ97OICfai+T444qayyoD0U19jG0ABaRRG17+YXpF/N3oW qK3QEBlmXzRbtUq8L2ctobWAOFh3Vmg8lhYE/CferZ3H/URqSW2npI2SNDskh4FiVIcM49zy8H6m AeAesDjGmfv88EbCAbtI0p9EipA5fyCp0XSQ2gsLlWJagddPPFQRu7XuzBXhhYKIOrVSG0wVQcVt dNtI3E3MpM9k/yRHB8KF7tlQk7kYEXgJBrPvvrmN2SYJ6a3lGw356YV4eKFVMs+vjhQJzlkKSY2i g0CFNImKlSICT4GtIoK8M1ckORGfVmjdtp6Nw1lRY8+zgu0Z+iUZGusz1mGiGD8qt0L5B+CL9skB FrNCBVsHkwFGkhVsEEkXEymCZDO/pEbTQWovVFSKSZcvHyag3YkmNjTe2QviSorqP16tlPaC2t5o 4le9je9VXjB1rFKBORNO/ZaNYrzmH0zIKB25CippwIep04TyqaWUGEKaGBwRThcTKWLnGEVKyGg6 SO2FjUoxEcOXEtxYVBPB3NkrksyoVysFLbE3mmDjNr7f2JgW45/WvMUG32MZNoSRXZoA1WIMbHCc Tb8xDbuk1KF4rXyR5AbNijqZkSJ2r8QfTUWOvbAhKTCG3T++55/ltmji+NzVHj4VCRv1aqV4Y241 wcZtfMvGLx3Ij65TERusdQwbxTRlpMsCYMM3tShOPpTHdhjYYDldzwXxy0/hyZ04B1NzoEtWQEbT QWLPObVSus6IQkQNaDWpHdCeyGOvSNioV8Hwii4l9kbTQRrfsnHXgiY3tFvSlT7SfIzWY6/FiIH5 LrPBBw706xE7UvaO9aHkgZHkButIKSSEkfGU0ZEaTQeJvTahlSKGL59QY0QXTcRpTbBRe0H0HL9R vSrGPYkTzWg6SON7xYZPAjk33rmISa0oVZxOB/kXZTnYdvmivYYLsCEp/VYfkSkSNrwitkKE/n2r 6CA4h6KmRqUIIR9jYYzvAStR3lkrEi7kE14GIzPW71F0kJiDw2c9ivZLdsMfF6764lE0FGM0wUZ8 IqeQ9d1whPmNlLnw1wjEjhF5p3m9TtUJc2awfV+5MpogowgT0NIkDhdjooHPJ1w1UShYE9zXXhBn xnpVTGzU89W0SOxB/nM2PrTSRaPBVU6TGIE0Pc4PJC5YwElrwLdM1N9qsJYmVYBd05pZq0xjriEt NHpUVF5tJZmAyDi8510xD1VDhFJ9bBUQ7evQCjFXYdXR4yy5iesVCezMUPnALKi+1ZHXQ5g7Er4g Exq4OTnNAHN5cY0jPCkqjSQl2DASbiKlGvJWU5FjL0xICowQ9J0Rj8GiiTL6as8diZFwUa9WCgV3 mhZpfMvGw45E75UHEEjrHG3YKKa5ca1N8RuwqQB3aQp92Jac4GJFjUWKhAw2h6iiGClC50SFnhtF B6m5kCEp5wW31apd1sqbNXlMEfJlD01FQkacWqGNhQRBkRPjNrrl4pduhF0a+jVx7p3iKlwIS2bw 4pgLvmnfPJCUmvA/eY7AhYSMzy8tHooMSxE6w7irpiLHXsiQFBgxaCAsMqqJOK/2khljj8oiXq0U NcqdpkUa37LxS2awVUM/J1PkP9nbTANUjBkpDq+YDb5pn9SgBZtBm5+jPtBzYeOLz7ibDWNqLzhQ kfIzmg4Sey1bK6VXnopNbFSTmgXt1IN9fhdpKpLUqFfBnAn1fDVtLI1v2fglN37UV5iNd76KlNwo BhuqbfWe0DmmPaBKTjfLsn/ywEjYoP9I2SEsRXWkRhNsuG8ae2LxYMzVcVbVbFQTcVoTbNReELO/ XhVTR1LPV9MisCFNz9nQQNfDCabKleZiYzHmZpV4YoMbaDJvzXdcwoa3c+utMBI2fjQ1upji9AZy NRvRVISMr72wUSn6wC8Xm9r6aE30nHf2isjeerVS6/lqWmTtPetWmQPSPbnyjhOsw4YwomLWXN8i E+e+hVdx8t2asMGmsOSBEXnHSpVO5B4pYudAwlXTQWovbFRKFVNLmGpiyvLOXhHZW69WKtHg+Wq6 j+9VbnDSX78n3jEf09woBht8RszvI9+0T4vA5GPY+Mpz8CMkbHgLkiizEKFzN3bqWCnySyEk5Pex I0UI7HkzrVFElHfWioQL+YSqldKcqc2tokUa3SsuuO5NvyaT6T/0aHlPhDkzuA8vmcFlcskDvmyi dkS3nHQQYiRcMH2rGoV5jEhp8fRW0yK1p+cWIwYuwzYZ1cTY485ekbBRr1ZKO0JtbzUFOfG9YkMf EzQZTFOVjEAsYzB1iW80SOyvSpQSygBJTxG2gFDBhpF0OsYUpDYU6W2rngG0isFOyxARGVJel83p XasaIry1VSA0yCFUByN1UWlbq2eA2MLD540FWzVSfnLRZtYJyaVgLOnqlifRwMGglJ98viQ0sNM1 ZYeR8MBukXQ3wvDNBy0ktZoW4beSpjDBTlVJMfDmI0umopo0ZWBNvAORCeLn1qti4qKer6ZFYu9l R8JWDf2SvFd8maA5UYxlBa/kwQbL2c6JPxJSG/5nZrmNhAx2iyRRKuVzFlK+ioI4X20uZHAaQ1KE oG+kKy+qidmkO3tF9Nw4tUL0FXeKFml0rzKD/RoKk/aW8qVcCHMITK0kMxhp6LdjFlNSIoMraCVT JGTw42VmrFI+mHCjaZHaCxmMRySlmsHzBBr9WZNWhm/tFQkZ9WqlqDHvNAU58T1ngzFqXle8014R NZ2LsciivSH6Edhwqx+P2CUlNjjYlKbASNjgFImKsJWijIzUalqk9sJGpWCDu1nRzlJdNBHnnb0g bmnWq5Vaz1fTImvvcauRNQImO/56OMFiCxcqlI1irLj88DqLDZqPZAKfdwkbvPSJXIi909itmx8r 5aMJklpNi9Re2KBpkBQx8PrB/mpi68SdvSLJjXq1UhvNalpk7T1mI/N92jzp4QTSSrLkhjCySzvJ eEFckP7tEIOTQmGDUwuZ1zQi7zR484KjdqlZykcTbjQFIfLaCxueNhcb3AqUFrqaiPNqD5+K6DlK y3i1UhvNalpk7ZWNhwM1ban0cALV+gZZ2CjGygYf+rBdvryZTODzHuyH4E35ynPaMpsdnB5MMUWl N8WYepSfSK2mRWKvfWulGIfqe2YZllkTCyPSfrFXJGzUq2B4tdGspkUa36t2g9leRaBVF7ZohY1i WufxFng1eR7AMKerfZryjv2VidxIcoM93ynVhDnj/e0gZVk16QDDxV7rk2LMm3CTi9moJsZn0bT2 gvi59aqY2Kjnq2mRxveKDW7VUARIs7LX3CgGse786Z3+efwCGRISGf/yWJGQ8dd72VZI5y+te/Qs EGNtbCpEAGw3MBXRQ4x3xookL+QSfgdzXtTrVaQX/RrbKyb0dQoRwfcBJy0CKcG004CKhSU/ZJjE 1DWkalD/tPU0EBrYXJL+t0KUkXxnQykRNf57LTUhpuXk4tSQUC2avbgxFcDcrT8rRG/VFrh6BkhU 2HrcbE4nwlquXGciSHc95dUoRptD7K7ofEQfEnwhlEhgqJWIhdg1rbN51XGlfDxBUqNogVoTeYtR s1C58rOuItph+XQxVyTpwNk7pfFKacuJnVpNi6y9x1xMF8KWDhUOSP/tBk5tqPzQRCQhtM3UtfzK Bz7mkQ6WxYuUG0aSEOwpSadbKZ9OkJT1uNXguMLFWKgoRgDMVhK4Bn1WRJB35oI0KeLTCsVrrK2i RRrb86xgyPThQQTSupNCWbEY/bL7NmaFvJdZVEhIPxNXUzdskGYFBXAJi5QPVUhqFC1Qa6GCkxcS IgLG686KKiLKO3NFkhV1aqXW79W0yNp7lhXaROkxhKaSOUUXLoQ5K6i30k7wTZlkATt+Qwaff0ng RpIXP9/dClkpn0640RSEyGsvbLD5RbqYraP3cGJUk+qtiz18KhI26tVKbTSraZG196ywYPpNq+ry jmtBh41i1K/am+OiPOtDEtfNxKQGu+5SaghxC6gRWzdHGnMpHanV5BMKx55TajEpaBlRTfpPfk3X XhGxgdJ4FQyvNprVtEjj++09SYuhTZj62hxsMA3iumIhhja6KlFmuQ8nifBmdiA6T0E9gFwjN3Wo Rs1KZPjBkOHXXDVBxla7EAkBMTDnDoWrHrLo1pYB07f+jMy6vGoWcVwvuhB6xj/MtYoE7mHt+1GI NiLLenSE021SB+GpxkN+irfx2x8ng4S/b15mNIQMfVxkVs0isSXqFmIyhgOO/Kqrh4bl1laA5EL9 GZmNImpQHISfYGw9byW4ViUlJrcol4RCeP/Xe5Ko0948HlELpI6VoihPBUgmcPot9WVkqHYis2oW UXvrTV6QUCG812fgREL1qHW9sRUgJMgffrmRoY7DFiGvmiD5cVHz6nX41rd3xZjuSEtLKch5zEck 5Qclmj7Hq+Y/Eyka/fipACFB29WPCCOBiGjuXVfyEHCRmNJDC9Fp6iso4iBq6Hxdca+lAKFA3uDf yDDIianVssiYep4H+kK6Gej2LG2+FMe4rswQAdy3olQhXq/mMaKzRP5O/Lw1KZwQcbgzuK0SRop+ Bp/QkfCD4Lf2wSr6KiG0Yyeh9pl1ZUTG2VUywNh5HHtqSc010RU5+OxrXoTguSLTwbM4pBQh2PSb jH0a6TSF7BNJxWUJzR5ZYnUsglN5ZhH8Zu+Tg5cW7BCZdk9rGGUt+tt8USbHEyN63ev96likdoj9 143+2i3JCQLFzk2ziX4x5j+5v8vR/6NBUvB8byNusdKZTBdixzTQ8oJhpHiMgjhSq2iRmGulVSn+ IxPcJqGaqLGlXTTUXpG8AnEqEFyt36tokUb3vBlgu/IPR9LEBR+pSLe4GBPiWZkjAZhqMhn8UnGO TiyFoRE5p9EW764oE+ZXgaMHklpN/GBXe30bipEa+vj7RRNjsWjSWMqagjQr6lUxsdFoVtMije8V G+zPSO3GNauTGcVwQedGlX0sxCTr+Yhi2OCTeYncSNh40+fDxUal6LIjtZqC8Jozdy1NaRsqRbPh mQ1+xyhi89+duSJ6bJ1aKe0QtVOjaIG19riFSL3EEi9LbfaNy1HTTQbz78lH8MIF21eSB3woT+8N r76W6BS3kXDBrs5OyUWKIi5SZHk0LaIFXjTpucXoL9xU8HZrQVBNxRtCbixqLogfozCMU8WUGAqG 50bRAmPtP8+5oFWWWWZ9+LaHe0ttnzRGc/XtnQxUYn9d+VM7IGXn+JCPngti5ygEOFB5MEXJyYp0 O9W0SOy5yViM+iHvSPUwgyWfxHytFUla1KdgbjHqd/VsHI3t1RvCDiJ5jzQL4y6jvV9IGEzoWiMl IxuPkgNULfGN7UmNWkiygoFzOldJ4RuldaRW0yKx16yolCspvX9a2LImoowmSvFoCtK0qFfFlBaN ZjUt0vhescHnhhUB/fcHFbKqKO2wNEapwFjbbPAF+2Q/Y269NRrdWIZSW0jY4B6C9LaVoja+0xTE 76Tt5R2pFMU1e07MRjUx03tnr0gyo16t1Hq+mhZpfK/YYLlPFSm13CctZdgQ5txgsVuDCapU3WCt d4JvooUN7qZOHhgJGyzdoWqFqJHvFC1ScyGjUkwpsIxgMlCENdYAsEaOaUhla0GaGfWpmH7F+r2K Fok53slnLYa2T3pogDSfBjIXi2ki3HsmKFf5qmjClJS8o8RNFkDVZAZ7Q9L7GnO5GKnVxA+kY7rm fp4LRuzEoE8n6D2pJuK8s1ckmVGvVmo9X02LNL5XbPDJCTvHWu2Qoa9QqO5gj1HygkWwRA454UJ3 RygHDCQt2C2SrrdCek9v9AzAuyI1yYrIMJcA3eZBapwVXJGXrIgtlhHnKUrKOFSMPoWZVD21egZY W//vP//nv/77v/7n4RY07ab0QAHpz27IWwwaPphtUuL94b85aIScr6xU5tdnd1T24/Gz69OEI8Vj dGORWkWLxFz7kEoRAV+P4rnVRJDSrgysvSIicJwKhFPr9ypapNG9Sgh2ZYhK2vd/nBVTU8EGy2BM 9WXVheJN3b4nVpCycxwG0HO070LknIZkXicM5m6E8wiSWk1UPFd7zYpizK38tJOqJpbW5ZXIqL0g fm69KnbxfDVtLI3vFRvsFtHPSR+vr9mEjWLMQGXNgQ00XOTEW0Q/Lyl5x621eq5I2OAaWXW9K0Xl G6nVFMTvo+3puZVifkGrKOK1mtjOe2evSFKjXq2Utobaq9W0SON7xQa7RTTNijR+lQ1h/kXJW63A UMzy5Z/EKSmxwffyJEPkQsIGX0zJ1JwxxclJihtNi9Re2KgUMWgGRGxUE3He2SsSNurVSrERNPZW U5AT3ys2uDNSvye1MHf4lI1iTHyywZY8pZzVfl7Fzp2xakcYsuW5ImGD2x9TqFWKejpSq2mR2gsb lWL4rdt/xEY1sSPnzl4QNzfr1Uppa6ijWU2LrL3H3WqmLdhMyd15YUODorwpxYhBAx6xwSUNeQc4 gZ7OhTvCUmILsXcM0Okaw1mkKCMjtZoW0UgETWGjUsTACAv2VxNx3tkrktyoVyu10aymRdbeYzYy RNNuSg8lGEeyuadsCCO7NB3s5QeKOO7ycpxfklIJ9p7ncFxIcoOrn9T9FtObwj0EN5qCIFN7YUNS YExJcJWkc6Oa2Lp2tYdPRcJGvVqpjWY1LbL2ysajfRTvbJTUJR32Tt/mJTcOBhv6KpBygzvI8w5w Kky7YDRs83NErufChm7zMBmCRIZvADGtUbQIzvkxVC1GCNoSqtSIIqKMorVWJFzUp5Gq3zA2ik4k Y+5Z+Yk0wyD9mpoc1kT/f8RFMU1F03SKC2Ys8wZwaXUyw1/gUuBGwgV7Q9ILC3OY/pqXyKgmOtob e3puMaZntFlDZFSTXLEmBop5Loi5X6+KXTwfTSeWxPe8GEf6nRVakfGt9c0kRiCaO61dqg//o9GT IuceN7Wp2kWQDkaAXdO4rdsdjSlI7QEWE9UzQG2FCGT8W2IzPVXVkLJSfWwV0FPr0AhtGMzwoMe/ 6E1cTzsSnmZLhn5G5sY4jFMWhLmx4AtsWoqgomXUIRbeJSTP+OpEfn0jck2Lwd5luFIsjkSKMgJF zpEiNRciijEvo2+tH02fNHy39oq4pa5TK3RiGUUHWXOPG053I0izU0QFItJ83KdcFCMEf4wPLvim fUOXlMj4znMEDmLnNGLzkmEwpKipI7WaFqk9kbEYsxFf7NAUGdGkOG/tDZK8iFeD6fWo56PpIGvv MRvuRpBmd0beD446eDtesWQG42X8pYjTfh+nhqTExk+eI3IhSY0Pbpy9SvEfkbpqChL257nFiOGv OymmzKxJcV7t/edtkLARrwYTG41mNB0ENmLveTfCBg21/Uyb6jr7tBbFGEf43jLYYFif1oFGWnse NEbqMERI2eDEcrpeY/rVOZEg7atpkdhraVIp5ma0Y0i5EU3E+S3tx94gYSNeFXMTUc9HE2zcxveq 1WAHnX5PJk6p6jVQQ0Mxhpi6Slu5wTHh9Jcea8g71saTB0aSG1+a8FEGCYMz2tpIraZFYs8sLoZh jW3ERjQRJ+tmYT/2ivi58Wqwi+ej6cTS+F6xQZ2l31Nzp7T7YaMYEw9qLsQG92Np5pxWMVdOaujm 54qEDeY91PmulPpcS62mIGiqPT23UsSghkdsRBNxft7aGyS5Ea8GExv1fDQdZO09azd4wajh4t3f 7Ed7D0ZUxAATZoOv2osz4pSUcsNT/ouEDSb9NX+1UlRjWhi4aFoE76QpbFRKO5691YlhoDUpzlt7 g4SNeDUYF1U5GuyNpiInvue5obUGDyeYPdXX6JQbizEVxw0OZDwlLdPDZoOPZoQNDm4nD4yEjZ9u 3KkQg7gIRZHTQIids7mQUSlC0OfvlBpWRJRUOVdrg4QL+eR2NVLH71F0kEb3igt2WunXRFonfcKF MP+eDDpSmzNUCRUSUmLo02EHkXNa1/KxKRVWlqKKvFHEG1Gg1kJFMboDrshfRf+l3oPNXipRxlwR 5RNvAB+gwKnB9BvW79F0kNh7UXxSwbNv1lSwZaZUBOLH1CABoxyeatzI2LVvP0TQeihEsG0kk1rG 9OPqWw56Qaxm/8atecgi/A74zu3PZiFaFJ+ExIJNDRAS4s9gIiEeRw/9ygBr63lDwZuoQop9UbqC IekgDNeYL9QlDGJBV02oBfiSkDz7l8eKhAX2kqTgrhTFaKSiCPIWqTk9txjFEseLzEQ0ESTnb672 iqgnHqdWiMAbyyg6yJp7zEVKT6aBGDqYC0YFw0UxzZ1q1QKzfCpNjDHklpTI+MlzQVwmaPTkYepK sVwSqdFE4EU0U4umkFEMytmjA2WjiTjZ13a1N0jIiFeDKQ3q+Wg6yNp7zEZKTyZMueMwbLDq1cwQ lszwsgVs8G37siEpNaOcEE4asA03UWm8NtsdI0VBfasJNoQce2GjGDFoA6/YiCbFebXnJtJI2IhX K0XsjWY0HQQ24mfZeDiDgRCnxOId186XjWLMJP+wMqTc4Lx8GgIm4dKpcEe1niMP9Jyi0rit2x2N 6RWgKkl7EU0HwbA0hY1KEYPu1T+aFKe0i/3YGyRsxKtibiLq+WiCjdv4XnUifMBUJaRopfNMq1EM NrQjSGywIp9M4FNc8Y4PiOo55nyEhA0+V5puV5gbCU4fSGo0wUaR2OubUowmWnuUxEY0EScn18J+ 7BXRc+vVYHpT6vloOkjje84GDS3Hk8IGjYRzYzFiYK7edrmdVHPnxC4p/VZcgp/yy0jY4IqSNLWV ooSM1GiCDSFowjtp0nMrRQy6C1hsRJPivLU3SNiIV4Mx+zjRjKaDrL1n7QYVIJtw7R3zciVDkH9Q PiAoMqjocvEb01R+r6kIIxJAIWnU5lVHQ86Lf7dqiLqIZvBnXFeIKSq1t0fPJ+30xRYtRgDRMP6M DDHX5VFzkLH1mIR0JQxZtFVNPxGf/ygJhZg/9poWRlkvye+vL/4qIbzBTSkC4Hdeg7WrCFVjFI8W OCgSU0mGQjgP4eYANYn4xpLiBQgF8aaQEqEOj5aD2BT+PaYg/QdjZlo3U8DaWCkopCl0XdcEBXzN IBPZ7wiJgu88xawdQPKAPSXpYiLDLt/IjBpyvYhKgByjWYiyUR8CVB5EDwFqG+CxVSAkyB9qzMoQ slwmm0dNkY3rFQmaZlca6KS7mklN+2gAylwxA3NT8Dd3xGmxS4t7mmJKG6m/wwD7SNKtRITK2yJR 4iIhQOzomYrwXvtLOAo/SghN14ytnf6d6OWK08QicdYVd5QM0HhexU65ojISp5iKSPCC4APH9YEw JcBHbnIj+s5XsN0uv7aAhM/OkfQjQI6W3XZXNURbJLZCQCAxwAl8Ataal/Qo4outxBtb68/IEHFc XjWLNK7fSJiGgA0bqSmp8r1tEQXFmCHnnlSzoLlmvfec8Exnwb2p/cVB2hSwj+QqRdScm5DUKDoI PEhTiIiUIuCKZvjT6EqaFOStvUGSDXZqIKVu/K4iJ7OQ/MQx97xFYNdGKjwuqx4uihGCV+z4Bfhk dxKAT2uGDC6uye9vJDnBvpG8JZEidA5LSGo0HYRxojSFjEgRA5PhyYpoIk4GYld7RUT+enWkSGHb W03wcxvf88xQ4+URgnaFZqvz+2JQ7RU77HJ7kX5ibQzV3mPqF+6ITq0opKnB0Qh1EyNF7MzsSGo0 HaT2wkakiOGbezqcGtFEnFwbd7U3SNiIV4OxgNZoVtMivCaO7zc20luwwP/XIwT3CMmNYG4r2Wyn 0TiV2lzlxkdO4h1tS2pFI4pK47FuhIwUsfPCp9WNpiLwU3thQ1Ju7rziNnoU5a21Is2M+HSkJhZx wZQoTNzG9hsTaTEo8z7d5eMZu0fdbi5Ga6RiWe8it7w3CySlvOAzQs0BEOeFCmGvIpIXliJujlFI ajQdpPbCRKQUgWfQR5GCvDU3iNJinBpMoSeYKDIXAtxiOLrfuJisYKitwpFSieMY5aIYEXzpFlHM 0j0nBxjEhIuvPMfktRDFpJGRVxGLuUKO1GiCi58be+EiGCHoBs2jiCgZj13NDcJd1uvUYOIijo+i BSa637iYvPjgA6rmgq38KSa0z9KYqgptUsQs37JPFrC3RX2LRmGWYdoXpHnBMBNVgfzKsDYo5aOI l6iIJl9RJCoGIwKuFqe1YH5GeoiRyYqrsUGSFXFpMDERt6PHWSHAWVFjz/oR9jVyvWSY0CK5KovF COAvCSYmuAC0QUpKTHBSQs8RtpBkBTtG1NsUU5CcfpbUaDoIVMxzI6UQvEFNgzkpUpS35gZJVsSp wY7jo+hEstYeczFZwbcuQgXrtqUiEP7rAxZiwl/IcIwSUjuq/zJAU4LNIulo+DItQoStT1uYB+tZ QE8jkYywjJzv9IO0uKJiiuJiSgH3ofVnMJFgl0fN/M2LIUv495iCaSTYoqGfULulSeKkQzHqdy3z iYSv3P7GlKykRAJjo/z0RpIOjCNVRIwUQXM8QVKjqQjJWnvhIVJEwPXqtAm8GtFEkAzIr/YG4bn1 ajBREc9X0yLH3nM22DiikkzraPSCYUOYY6AJChs5QM1rIyGRwQG4/PxGQgY7R7LbL1KEzgGFq6KD 1FzIiBQh8E3KkBFNhEkjfLU3SMiwUwOJi/i9ihZpdC8yQ4v8HjjgHBcom4vFaB/+UDPpN+CYY6a7 +RhEnPub57SDGiRkMEebTiZShM4JhRSc0VQk3M9zI6UYWG13ZkST4ry1N0jIiFeDsTbQaFbTImIj 9p5lhsaXf/Vz0iNom78yI5gzg3vLVKxoSOkxBY20pJQaH1qm0TthJGxwk7G68ZEidhaqUnBG00Fq T8+NFDFwyUXYiCbipP282hskbMSrwY7nq6mxnPietxrML/AlCqLCO32IJmwUoxelvKIToazNNXDE KSmxwW30yQMj8k4rU145HCni5IyCpEbTQWovbERKMaD5oklx3tororZ6vTpSE83RNMjE9xsb6UaY k9TLIDa44H7YKEZ95bV97PIxnWQCX+MNG5y7S08ixN4xb8ipOOdLpGgLOXGRriSaDlJ7YmOkiIHD p5w5gtdo+qQuu7U3SNiIV4MpN+L50TTIxPcbG+lRmKt/8wBCa2r8sMkNYWQXbHB6wb8Ci1vJBHbH hw3uUUjJYUTeabjmBUTKP0sRJ0cVrpqKhH3JhA1JeTDCznf1H6NJcV7t0dsWCRvxajCxkWjERjQN IjZir+3Gw6UP7Zz0IIIRBmVa2ShGya8xk3KS79gnE1gLVQGmwZqfYxCg58KG+kI1G4ZcQUZoFEHG +405fRhmMELgevWQYUVEyTBerfhYGyRcxKfBxIX8hrFVdB/dq8z45vygfymu/IIL4mQzZTBC8NI+ XLD3JnnAppD0MRxLTelpJFywpyT1hzCHyZ7NlJ7RROBF4F6aeG4xxaArDZQZ0aQ4rYlWPPaKmPvx arDj+dG0scTey/6VL5eYDE7vTWIE4ufwqjZc8C2T/ORMPqUw5/d3DghoWrBVJNWHMQX5YaHRswDL r30qkGtNDhSGh6ghQj60Ht5ta4AkRRwaTDTYZ1pT6fEvehPXq5TwByjVdP797Fw/O2XoYmksmL7R 6rxeD3+4UpnPVHVcY9kqP7+RpAS7RFJ+RooYOaNw1VTErNteUkJSHnz8YwepUyKaiJL9WMeeGwsj 4SJeHal4ftU0yMS3bDxsLLR10sMI2GDI7sZiMa1lak8tbOTGGsiQkKoOdhqoai0SMtgnkt4mUoTO EQWlySg6SM2FDEk5L7hvO2REk8K82iPMIiHDTg3EaYHGQmJIEYkxiLiQouXi0akypNmroV+Tfowb DcuFMGcGl+KFix/OyiNF5SUpkcGnmlNYGpFzGgl162OkCJ2zFVdNRY69kCEp/55UV+mkook4+eDA sefMMBIy4tWRmmhgo5oGmfiWjV8ygz0k+jnVq9OBqlPVBktjmvTV1g/eRW7kzgiDD/KqBNPAzTLw o+fCBttoUl4ZU3vB3iBpH00HiT2VJosphvd22dGkOKX92BtE9sarYs6EeH40bSyNb9n4JTfYz6AI WFz79+lL72GjGMtr7P2wXUbTyQTKyXjHnWqJ3EjYYKo35ZUwvKMgjdRqWiT22upGihgYFSY3ook4 P6KJ6WNrKuLnKC3t1WD6HeP50TTIxPeKDSZRFAHefebsEDqLEYP3fvArMLOiAQwjNknpt+IfjVxI 2OCuQPUvK2WlklpNQdBUe3pupBSDFOidiybFeWtvkORGvBpMbMTzo2mQY+/ZQI3NlPrGhrzjyu1s gQ/mGLjZPrnB8nYygU/zqiXR0I2kVuRGwgYfQM5exEqRnpFiJBNNi9Re2IgUMXAJivoPDQO9OMiq 0609IjcSNuLVYGJD0bgFqqb7+F7lBksTFBAUKVwzMmwE4xfl7Ci7pMgNPmufTODsYdrVP34ONoyE DfZwpcCqFA15pFbTIGMvbEQKNtjPHzaiiTh5F672ingoNF4dKe0WdTSraZHG9xsbGahxaEmfRzMb 2mCWVjSY2PCWB9hgT2sygWIn3rEVtZGD2Ds1lWoyIShC1FYRWkWDiAzJhIxIiYwvvkql1LAiRXlr bZBkRnwaTJnRWKKIHAtCT7Tmnr0n2kHZxGADjKkoJCa88QEmGDgm/7n1PUz81VOKWkDSgj0j6YIr xMDIQqungBzjKdM3EN57+wY0RI0ivLE1QGiQQ/yAEYK9CWP1DBBbePicBLZtqPSjHmAU4K1Y7Kg0 Bg3cesJWB2ign8zPz5nQ0PARGaoNIeGBfSPpfIW5I+FEgphZTUxfXO01I4oxycumQzcW0USUb9Gk 8aH9DOLnxqtK8dx4Tm5V0yKN7xUb7N8gArGhekCvh7ZZenBBDKotlGbcEuh+hNnqkMECXwMXIjIY QlGGOVEqxSzbnaIgzleb03MjpRBUpBxNCvPW3iB6rk4NdPw+iiaSie4VF+zmoEBEmkozO7HYZKl/ 6vfka4XhgjOTyX0WFNWxaMRmGe194R8hg/0l6Xor5RMKN5oGGXshg3MMSBEDF++HjGgiTo5IXO0N EjLi1WBiI54fTUFOfK/YYKMJrZ2803cskxnBxMab9p3zI3BVbxKBj+nFu+/s94YNIWHjXV+BVBtS KcrzSK2mQcaenlsMNvSNSqVGNCnOW3tF3NKMV0dKG0YdzWpapPH9xka6EV0ApR1EeMcOgGEjGGyw N4+tEbDBYkheAfYchI1/2ZzEgA3E3mnsxtSE2KgUCiK1mgYZe2EjUvyiTHWm1I+mT/a03dobJLkR rwZTbjSa1bRI4/uNjcz3afOklrrwjl592BBGI8GPxtYIs8F97+kv2ZcoNpiRzzoaP6OQ5Aa3tKfr rRTNzp2mIMde2JAUGDHoik/lRjQR57+rPTK+SNiIV4OJjUazmhZpfMvGw4EayjNporUsrkPNmxKM kvSHDTRmg93aeQfgTe+OBkmes2GGVM+FDT7Rlt7WmMvISK0m9bF6DvZlr6VJMa3Ie8pA3bU0Eee3 tB97g4SNeFXMvWujWU338S0bDwdq2l7ZKoMOzLebw3AwsfFG46k3hbsL8g7Q9ca796kzjISNb+0w 0psiDO+0X9GFxmoaRGxIk55bjBhoBUGo8K1JcUYTg7L4GcTPjVeV4rnxnJakmhaJPdh/Vm3QTHHy 0r8VM5zJjcFgQ2seIuMfWw3dv3LvbMigDkngRkKGRrTiokIanVlo9AwAFTLWxIiQAuBQp6mwHmJk 68bV2CDJC7nk7JEUzI/XR9EiMQeDz5mgaQgR79yzqZdEWy4VJP5rMRwe2Oeb5Odr1XFtaRAQGthe kv63QlTT7Pg7avz3WmpCIALEfK++eKF8iBbFfGOqgLkbf45QPT56BtBipR18TEE6ESpYXkV7xmGQ SYZgkMAXvfGMYpZJWVhhJldCem+5YCA1pRC7xrSfDrlfpCimIzWKBiAbbC0vRoQIgGv3NQ4cRcT4 eWtuED03Tg2m366xrKZF1t5jLtKFaCsm5x7ViJHS2e0azL8kAyybzcXMhCIZUcHnZRK2EfmmRbbu kqwUzEWKEYfWCWgPAoyxUFGMGU32bF0UESS3/lzNFTHz1Ne6cu0IbSSraJHG9tuLMVnBbhOVfEi/ k9l5M4KRFfr2ln4ADukmB+hc4xslVepJIc0KdpdkUq9SNJyRGkUDjDVRsRgRsAHBWRFFivLW3CDJ ijg1mLKisaymRRrdb1xMVrDpJEUnG/GGC2HOCgZZIYPdhcmCb0kpL378HC+NkeQFrqvBLKY4GW3c aApi7m0vbEgKDNf1BU1lWDQR59+rPRcWRsJGvDpSkHCnaRGWoqRp2XhYWGgXppp8fqs/OheYzAhG ZtBWhg02nSQR+CgSe1JpRznA0sj1XNhgG6E6XNgQJjYoIiW1mgapPafUYrCh04piI5oUZzSNvUHC RrwqhlcbzWpapPEtGw8LC+3EZGFGBR9TSjlDNRgxMMwKG3wKIK8AF7UoTt2ppueI3EjY+NKyz2B4 p7GbpVbTILXXzidS5AbrwGqcNfiTJuLkbErYj70ifk53r3n4ESmeG8+PpkUa3ys2+Jha2WB6JKkR SGRQ22BEpZ8PTxA6QuKCUWW5AAgVfF1Nne7IUEZHZtUMIiaYG9JTC0HEh07CQ0T0KOwbWwWSFPIn ZFlGNaZOZomFqAmSX9kePutEtL9SYwcc00fTRUIgtxYMrhQz5ets+fv0gg6jND9FxAJCAkfa0+lG RnOgHpUwu6mr6RVxkNoKCYVo8DXRIRKih5izG2VsFQgJ8sdtR2Q2ilWzSOJ6ngmMWiib0IdwKRDg F5sRVSjghsGke+pNbrTNT54lAI3LkhUZR+kOQVPY56me/YBNJPICuMwnMIi8KhQnf7GakQf0p/Zx UjVnzTj/PT7zw+3ziyiKJwGjyzsjVc2xKEXDl1+dzZL6/Ziaz9I2hZknO/mR/UkbXPqKSABFoQEY xx6V+pHRitmtmkFqK7FHCN8pgRK79OCQyyQ8HFsBnGHjT6FLFKvmLi5I+P0GTkj4YELEHOBxODAC BXyNWr86ignNbzZjH7v1rWcULs+EAfaOpCcVpATWZUbisUoKKP4+Y8SvLHPA6XyjRKFZiYZRecv1 t+yMKysy7vMfdCWC6LjG8yQB0MYSi7/bhVM6PZzgA8lxPuuNQjLLl88yd+DxpQZZHjcESPS+8kuE RMYHD1RlRYv54CgCKVJTemqE+P350nYIiJ5PBgw3tgqEAbsTxPE2htVyF9UrCmg3VT5DAV9p93CK f+a7E3jvD2jBAesbqRrp6/NiMvWUH9xIaGBZIl2mMJzzKQRJraZBas8/6mKEoANreqGiSVFGk4ZQ 9jNIkyFeVco/aKNZTRPLxPeKDX4IldBwyMc/mxHBiIHz9RpZUXd88c0w/dzMrKd/5LqqlE5GwgZ7 d/OWVMrnECS1moI4K2wvacFWYfnAsFh7S8RGNBEnnwK42hskiRGvBlN6N5rVtEjje8UG+zIoc8WG PgOX90NYYshCPqXod8cP7/5YHG30W2TgR0jYYPifWb1KUQzfaRpk7IWNSIkN9h+ZjWhSnLf2Bgkb 8WowsRHPj6YgJ75XbPzk3hjY0EXTYaMYK0Ys5Lu14Ov1eQc4Mp3fioFP80BI2GA/TvrRSlEMz600 1bRI7YWNSsGGrr9WbkST4ry1V8RVFgWyvTpSVGCxt5oWWXuPK6YMNhmF+Xw0vxV7BMLGYlo/Izy9 KdwnkNj5CFrY4M7JlAhC7J0GXZRoep8q5WMJKkpX0yK1FzaofyRFUUCeiowoIkw2T1zNDZLUiFOD 8V28CcaK8DuAM8PRPc8M5inmNA6X2w8XxVgP+9Ft2uQfewoauaRUy7Aunyzg4xnJDA2BvF4IF5HS uQQPDFbTInAxzy1GCN5DMYoU5a25QVROjVODKfQ6PooWWGvP84JDQPyW5AUFQcYT2ktpTGuDhCKz fLo+kRNuKksfBVEOCHFeaMB1I6RjCbeKFqm5pEWlKNhZKcnADT3EyKzE1dggempcGkxM1G3rcVYI cFY4tldZQZeZ6pIrDIeJYqwVcQrdPwDX+yf3WXyIc0y/pdoyIuc03PJaIexESmcSrH01LQIVktFz i2nqXRNho0hR3pobJFkRpwYTF3V8FC2w1p5nBWdanRTMYwwVgbS1UBeA8gNwx2FKb9qqMKErttVd CmhK8AumoKgQtbdVr54BWKCTmtAQGYaxXhfWYMobFdkedGOKgA0kIeLPYCIhLo+a+TuW8O8xBZmR wj+GdOaA0yDDQTH+I+lvElxUMZGNDOnBS+in4ECIo9GYqLsbjeln5viEdK+eRUhraQoLlWISQV/V UDZEEyEyCxHKY2+QENHCSzL4tJGsnkViDS+fM0FPlZKTj4H4cynoLIYqHWxTOvDB+pTQfBFTvlHT +TlxIUS+aYDkpcFgeOcDB9K+mhbBu3luMU2ouKQaTYpcXh17RWxvvBpMbNTz1bRI7L3MCzaIpOTk AxiZe6DXC8Zbpl5abLBNM4fL+KJsulY2c6aYNBI2mGdUd8vUfaQoHe80BXEjZnvJjErRI3FOxLxG E3EyLr/aGySZEa8GExv1fDUtQlcjTU9z4y+V8V/tNKRI+9EyBkVWMf+iXOweNjgDldip8uLdX55z 5EbCBkdbVXIyZR0pHzlAityopkVqL2xwVEFSzFmz78RsRBNxwu/V3iBhI14NdvF8NP2v+F6xweYI fk/YYKvbsFGM7omwyFNNpnHLo+LkM4HxjpuIkgdGwsYPG/WuUj5yIO2raZHaCxscspAUs7Ts+TQb 0UScjFuv9oq4tRmvVorY6/loOsjae9xuuOREmr3g6kWQpjZqbhQjBr5yhHdMAX+yE8psSCrDNe9Q 1NGK3O5Du/qHdeOrlM8iSPtqWqT2wgazT5IiBiY2YH80Kc5be4PoufFqMOVGPR9NB1l7j9lwf4I0 X4FTVUNtTyleNoTx9lCzZY2WQlbnTsUGiwphg09nJA+MyDsN2HjtL1JU0neaghx7YUNSZoOljvQf 0UScDFCOPXwqEjbi1WBio9Ew6WtNB2l8+6Y8WuFAmo86qezUaDRHpQ5mErRYCy18UcZxYk4LehoY 6Tn4EeI2nh/EFVkhdTIslEv5Klqk5lQ0LWY2Q4YVfZq6cBFrg4SL+DSYuFAsMDaKTiRj7ln/ijSz Eyo8te7Ej89RqYORiNoyKi6Yj0rkXNcT73xVhbgwksxguSG9sDCH6esjREY10Sje2NNzi/F6aEOs ciyaiJMbgcN97BUx9+PVYBfPR9OJJfE971+RZtdZyGDM0dckEBGwqcJcfOqrxY4cIaLUuCiJwmeE MrdNj6bLQUfIQepyFDMRPXSWV1shAsi/Jd/KCA9Rowilem0NkKSIQ4OJhvgMC+jxL3oT174ej5a8 eJpdGCoQmcjTBrp0q8JwjZZaW9qUEqxWhwY+JBjXGJDn5zeSlGAjSDrfSlEx3WkKcuyFCUmBMYPH vfbYU8UkTUT5fbWHT0XCRbwaTFw0mtF0kMa3bPzSWLCmoBJR05rUS2GjmPot6iqx8e2JCnVbCOl3 Yh9UajAjIYP9IOl7K+UTB6luq2iRmgsZnFTQUSnNa7oPZVBhTQrzai/0CAkZdmqFiLx+R5ET4za6 5eJhZjCg/PIxJZzj2j5zESyZoUV/ccG365sHkhIZVB/NAiEhgyolXS9tv6QoqpG6agoS7ue5xYiB FQCeY5RpTYrzai+ZMc+NVyv1d6MZTQdpfMvGw8zQbkkNJ8gMZmGGjWJMF2YZn5I2t8JRrSLl5uLD z9F0Cikb3GaSrteY2OAIgrSvpkViz4XTYszo6R4VsRFNxMl3J8J+7A2S1IhXxciEjWY0HaTxLRu/ 5MY7RbjZ4EMe3uGOhmKMSLh3JeV5LneDFqScG595roi80+CNPFdjKSl3JBxCkPbRBD9FWMyRJj23 GLRwRu6qiZFMNGlQZk1B9BzvBf1O2hZJgazno+kgsfeiI2Gx6007DckNlsO8wx0NxTQ+43WQXSZ0 NYAhTknpTfnOc0XCBlsuUlFUinY9UqOJX10ImmpPz60UMXDiw2xUk8Z1t/aKhI16NVLH89F0kLX3 rPxkyZvpf3vHaflhQ5h/Ub7FwoImbNBl6jfmVZWU2ODDFckDI2GD8RYb744Ulc+tJiIvQiJJU9go RgPCtmWzUU2M7+/sFQkb9WqkiL2ej6YiJ75XbwrfqNfvyWw6P0Nb0WKkJ3fUuGXmk/bKILoWSYkN vr2YPDASNj40DXaEKIkjFEU8FsTO2VzIqBTdjS7l0isXRZzeu7NWJFzIJ1SNFJHX71F0kEb3iguW g/Rr0mzBarkQRlDMIeseG2UGy3nJA774kD6GviaBGwkXbPNQjUJpEimK81tNkFGk9kJGMQqRL7eY WrOzJmbG7uwVCRv1aqSIvZ6PpiInvlds0GkQOM7p88sqNbSfUhBcvGnHC1xwnZoKfqKUkPpbXTe6 QKhgj0jntSJERX2jByYCxJY7k4HwXt96VlJUDRHe2ioQGuQQ9oO5K4nPo2eDaFyvSGDThko/+vXP v/5gCgqKUUl4YQ8a/NV0Rf2JlGngQ5h6rkh4YMeIGvZgdCQU5pEaTQeJvTYWlaJIZ0O7qagmZozl lWivvSDtSOpVMf1+9Xw0HaTxPWdDuyQ1kIANjlAmJxajyPym0lJSUCQ7J/guWchgZ0USwEjIYAuJ KgomtCKFzkiNIsgQ4ny1OT23UozTuBvbZFQTSwZ39ookMeLUCP1dv0fRQRrdKy7oS1UgIv2Z4zBo EEZQhKBdB2L8T65406qDj8Pw40emiILSmI1D2CYjUvwct5oIvEjthYxixKD9CxdNWj+5tVdEZKxX I3U8H01FTnyv2GBvh35OvOOWRLcW2jdpDDbY1puyDhfzo0tKecukedNASNjwl7bNRqR4VW81wUaR 2gsbxYjhr2cKPI2tbos47+wFUUsDGxyxk1cjRez1fDQdZO09LjEy38e+Sb6eaDY4webtOWgoJqe0 Zwa7HN9Ke8AuVO2i0sDNzzGZAOJ2EPJoU8xZpejLIjWaaBqLMC6VJrGxGLN7DFZd6lcTS4939oqE jXo1Usfz0XSQtfeYjcz3aaOkhhPkBkOxYUMY2UXqM+MjMrijPInA3Xkhg48ap+YwoqBoJ781w0Wr USmKqkhZEVEuUGvhokKa6fLRoFXk7Yo35oI0M+rUSBF5YxlNB1l7j7mYzOCaBv2aSH9m4xYaihGC a0/vxUvtyYem4txPHiNukCYGs9upOipF+RupKoKLArUWLooRgc5nitQqIso7c0WSF8x4yKmROn6P poOsvcdcTF58a2+mudAt7aowtM3yLQvVnhQnL9jM2CzwVe50L3zHuznQ+94JoaljEeL+GxGI8Ch3 ARybZxbDfU6uOLukhfjuDBUJC3KHvF2peMxzo6YxnLhetZwsB6foZM3HO9zRIAwNrO2o2oYFlprT CDCySifLGnF+fSP6aTVeY/+13o5IQQSLwhdFB6C8kqJkRIWY7fWXFarIXHwj5QFAzbGbpc8dp4rp 16vfo+kgsUfiPssI7auUtxTG1JxOiIFYWdBkLEa5tjC/PmkQ17gkkYeYE9VDIYLdGulnjLl4tFDV wIMOlayl0AAEefhOn206q4UN7tKsyiamCvDU8WeFxuPogbsBGtUrCphaTcHJ5w6GA2G4xpy/l/Ww +uMb4mg+mQqPa/72hxpUI6GB7SQpRyJF2ByduGoqYiZsL0xICoyJXq5RMhXVpK1dF3t+EYyEi3q1 UtooeqMJMm7jWzYezmppf6WHDlpcodNUE7GYllfcd+jmMqUEoUtIv9Mnj+FcEQVFevA5vYsUoXOa Qlkyig5ScyFDUs4LWEljXE2EebWHpiIiY5xaofWbmUEUOTFuo1suHs5pacelhw9MKLBsWi6EOTO+ 1BnILEmt347pCkmJDJaekwVGQgbbStSYjxShcwzhqqkIMrUXMiQFRgz6fKhorSa2QF7tOTOMhIx6 tVIbzWjid72Nb9n4JTNYDtDPiXe0cEmMQrzc+gyxyOAt0S9MmAiJC6brsy9UQKigklAHUxkaP05U SPOoOUhs6akVYmLzy0eZXZFIDyFebC0QGurPyPD/sRU1eBgE9sbW8/byk0oOE8yuZXMSzxei6f9k JCwS+NR9ZrJ94TYksK8vP76AkMB2PPUvdCHewERpGZlRQ8RFbKsVSCHm5ti14WyoHv2YMr62AoQE +WNeIhOX/YZFzQkicW0m/PJecPm0iEd4G0xBfi34RFBI4GLrvP3fbr8YAuUpIgYICVxInXbEMoTM kQpphgTUuFQqElt6aoUggYVYk1A9xHxrK0BIkD9+SSKjXaK2NWog4SauVyT8e2evpldlsttCBYrH DUymacZWicAyTX53rdvQibxFQhvoywBbBNQtCklleKuDaDmjoERhmSLPLMKU5F+N/z2msxZavGPH sfJ3oq8nlVhfRwfB1/vaefwWpKbWTkoPCpgQ5ovUbQuKUQLT6jl6vmOf9/xDUor/Pc8RAohTWiOw tJ4RIuDPCI2ig9RckqBSzEWynx7qRhGd/521IuGhPq3UxhJFeBnEv6yje54HGtB7UKDpYFZE1S4u xtCIZRk6Hl1f7VEIySspccHgMr+8EQXFzLcOaCobIkXoDC8lNZoOUnsho1LMwWny46KJAvHOXhBx f7xaqfV8NJ1Y1t6zzNCGSucqX30YLozgv7YUySbnyfPOcyo+TLRk8t/hgb0iqSYiQ9Qet4+W+Ruf pCMcWEIOaMKOBRRU4IrsUe2LcJvJ330p6kox/XB2tkrmz7XiyP/zPw8bRQjjLvDErlM3yYNi+M3N 7URP2uVSAH5hpOwWS1X9zfWcfhcNuLrrUVKEQ1cZqdF0ELyTpnBQKQKgPrsoIkIrgoWaKyJz61Qw nNpgRtECsYaXj7Mg40xeJS6mNBfcHuRdm6gspoUBDywoTLWDgGRnw7K81ZjLjxUJFWwYSS0lKTeR ZHSKyCiCiiJahUBTqChGRuvmNzitIl7Fqzne8yKhIj6t0Lo9ehZYY4+ZSEupNTSnHdJ/+UmSFcVg Qk27fgB2YScH4D5UsAe7YYP4bdVYiClHE2Ypkpy1c0mNpoPUnqhYKULQuROoqCKivDNXJFTUqZXa YKwIvwO4pXR0r7KCzRnyX8sj2e6PhmJML7KHF+cox/ihHOWbpJQWDIAStxE5x74n3e8tLiJFlEQq qdF0kNoLF5WikddM+VHErMKduSLs1zpOrdQ6PooWWGvP84K9GfIfaX0qMHlRjAj0yUGZ5aP1yQLG cOJCY5wOI4Q0L6g7089GishZJpH20XSQ2gsXlVKHJ+o1iLMiorwzVyR5UadWigFFzFkRXgZwXji6 V3nBdcjyH9/YsFQuijEJr7Ng4oILkZsFkhIXTEE0biHJCw5npqzmmyRIETnrApIaTQepvXBRKUJ4 0wcgIKOaCPPOXhBzv16tFCP42Pv/hJ3JgSNLDmTvLUWLkFm5VYoxMowEo/9lni2AB9ks8vQr7SOw GBHu8DVW0yJr73lmcMQh9R5n+4aNYsSgL1WKDfa8JP+pfsIGH/BK5EKaGWy5TF8bKWLnWIKkRtNB tEKFprBRKYpCPsfk1KgmSto7e0WSGvVqpZiSib3VtMjae8xGehKm+PjKgHOD1fqwEYzsIgZiNht8 sj6ZwIaKsEFl3MiFJDeY5k6HEyli5wu02T0ZTUWOvbAhKTBi0Cq5cqOaiPNqz32JkbBRr1Zqo1lN i8BG/CwbD4fgLI7oeJbeFD7fkVHHYqyTaF5eucE5AP3G2qfe6pNrJvRcELOh4dgR8osChxIaRZBR pOa0c3UxMpCJY8yNIibq5ZPey1orEi7q00rFbxhbRRvJmHtaY0AYM4SiglMIUEGSDsQaiVdtaYfY Zq1fjoknCck3Vs0PINc0KOt+xwgRo+Y5RUT0BDi2khSaoJT3rJGHhqph2enWVgC/kOvQCo3Pq2eA xvWvpjPlhVpzDxjII23xVDeyGDPvHNMGoXj700EEXwwNDT95jok4EPumwVl3OlaKMjRSq2mR2hN/ i9FbMVPM77CamAi5s1dEz61XK7Wer6ZF1t7zxoIdGSk7+d5FtrlTB4H5x+S782HDfZwmZpFRSnCp VepJI8kJ+t0UVpXygQRJwQV63CRyROFiLVwUIwJ23aFdwzxrIso7e0GaF+13KxOvnV7VswhMSM+r vGBThvxH+o1SInlRjIUBbYHTL8AKed4G9nmHC04aNm6Q5gU7RdTxwlikfBxBUqtpkdoLF6yQSwou qLGdF9VE5Hf2iiQv6tVKafx1q2mRtfc8L9iWoQjU90B02BDmvOBa9rDBNG7ygKuw05C+5TnmJYUk M/iclDreYsp/JqFuNAU59sKGpMwGiZHMqCbNmV7s4VORsFGvVopJkDtNizS+zY1/dCJUVIqAukQb hMNGMVZMWKJ0bvAdxWQCtUAKMCbBE7mQskFtpI6X+TphYoMpaEmtpkVizzm1GCMfVSh6T6qJqi6a NDy0piJho14Fc5tfz1eTdoFe41s2Hg7bmVbkUruwwaZm3WnxsRgxfJD6Lmh/PbggTqRANGTzc0WS G385QBk2pEtscLwixWc1LQIb0pTcqBRKuS7AbFQTcUbT2gvi5yiz41Wxi+eraWOJPdh/9qYwBU5o ZoMPX2RP82Is0TDbYjbYrKzNM0x0SkpscPlTIjcSNjiZks63Uj6XIKnVFARNhCJNYYM+Q1JMVeoE jnKjmhj93NkroudI/ni1Uuv5alpk7T1ng2VplYcaz7ANLLkhLDFQsWGXJkh375sNSYkNvm/eyIXI Ow3cvNMRziJFEXmnaZHaCxuVIoavljDVJJW39oroufVqpSA99lZTkBPfq9zgGL7qQ7zTJ7XCRjFU 0dH4feSahWaCpMQGH0lO8fkpJGxw/2Ioi5COJlgoingsSJybx1aKEPTdLdEaRUR5Z61IuJBPqFop XuyYW0WLNLpXXDCDYd/46sVQoUvcldvvfELMv8Af1mkdJN+eCBNMMRwgRLA3JMvDFaKcvtUzABMM UqOnBqKJZHOFaZAaVPOqIMTvoAGibAXwU5TVcaiYmov4vHoGWFtPboPiaXZmyCjTQLy9TYhitDt8 vME0fOgT1Xo9+JRSXPub54I0Idh/EG4k5Rg5SiHtq8lHFI69FieVYg7zb0uYatLWWWtisBVNRZIS 9SoYhG00q2mRxvcqJdgwogioqjhm3o6kGGsqOo7u4tyH75nG1Vl0/U7cI5rAjSQpmOlI3ysM56io I7WKFom5pkWlmMflqnS0ayhmTXSv0bT2gjQx9JF15YqEeIyBT8ytokXG3H+eN5wMBBQAk6Ucnm5m FKN/9qodTRRXJMmuN84pTA2P/FyQZoa/9jBSeOeDFNK+mhaJvSlbo4v6nm9um4xq0q71dmS1VySZ Ua+COTPq+Wpi4uEmvleZwSaNlJ982mAyoxgxZAsaRdx7hyEMzNPNsosmhbeRpAbf70wRJsypwRZ1 Sa0mH1A49poalSIG7j0wG9VEnNGkAZk1BWlq1Ktiyo16vpoWgX1pes6Gflp6LeUGOZp9q4tBAnP6 eKeflm3Wevvp1dWO6Kf1c0XCBlswUl5VSj+apVZTELfVtqfnVopZO2yZjWqCvDt7RZIb9Wql1vPV tEjje8UGia1SSlPq3fPvRS4xpDGfbk2Fjd9PDx6YmpGU2PjkOWSKhA1WutTNrRQvdKRW0yK1FzYq RWuhjxWp2agmRq939oqEjXq1Uuv5agpy4nvFxh/2ujk3WKuf3ChGDKxt+n3kdFhjl5TY4NLc5AGn U72nmYaVba4p1IzpV+dcgqRW0yKwMc8tRgzs8kT7aiLOO3tB3N6om7JXK6XuzfZW0yJr73Ermskd nbvzcEIzyfQRKrgWY2aEmVEQ6g0dRtevzmfz9O7Qeee5IPZOgykWSi9SFAGRWk2L1F5yo1JMWFH9 w/5qoge6s1dEz61XK7Wer6ZF1t5jNjITzDhL0/AaFvCqDhvCyC4mzTxSoHrU9/MUJsvmIeNvHisi 5xgc6ftSR4gqMkLWQ5AL1JieWox5iW9tdl89FCZ3xopo6WhdWqmNZBUtsuYeMzF5wQ8t/6nxdDY5 eVGMIs/jcCo5vojhIPksWpggixK2kKYFrWWKrEoxzRCpUbRArYWKCvE/seusqCJK0ztzRZIVdWql mEWJudW0yNp7zMVkhe4iEhUMMrJDTxN7gphA1deh+QGokRM3RX+a03zcyTW/dyLxU9LUqgwrpp/b H7iNGv/8+SZtLYWGiHgQonZotXigcjVVICTIH5J2hcbj1TPA2npOAVtLNITAOrclNB2EmQTvB4EE rjNzeHwILY75TIQCNiLPZM47EouJBI5QXPUsUGOhoUIMr/i23SrCHCNazPEj6MewoiCuLcanQmrc 6vUqWmTMPS07taPSgwikNbGZF6MYA+LsEqR35Cq9cIGUnfO3CsyFngsXfKtARRRcCBMXHEQQh6tp kdjzG7UYMWjSUqxWE3FGE69WNBVJVtSrYHi10aymRRrf886UDZO6qdh5wfE9D0gWIwY+qYl3Wgf0 +IM4kQKhDcxzRcKGv2NpNqRLbPjrl2YjmhbBO2lKalSKGD58AlyDMS9cEmc0rb0gTY16VYyZi4lm NS0Se7D/7C3RZkpvR2Sq6osmT7mxGFNHTM6YDb5zn7KTyMIGV34lD4woKnpvX11NblSK6i9SqymI 30rbCxuVoqf99rmT1cRE2J29InqOijderdR6vpoWaXyv2GACXmUnXeFwIYRsYZ7wrxfAKWQ5AuM3 oEUovV7KrlzvDw98giq9rT4BQE5wGOFGxyK2Ew4qQ53F9XFms1qYGg3jsuP2kb/1zHpSCfZOxc7q WKR2HmdCuk9tn/SgAVnKsyRCIYKnK7dJdnA2WIT0VnDKLmUWM95+1Snwc2cPpZ9lKBQjs2oWiS0F sxDOf/io9+ohvFtbAUJB/RmZjSJqUBwkvylqXiUAt/R2krWbFtl3GYgZsKxQUKCxvo5Gxh9e6tWo zE8FkGMalnltcWSoD+/ULMJk2683qkFChRiPaDPXRQ+bUG5tBQgJ8oeMHJm4jIOrZoOwrZckcIlk qgcucG8mCEKjpmCZvVPyMV8rpgiwQ9PfPBUgJOia45XAdiSihDdjEf3DpwUPhO+6UFwUWIt24ehP qbd3AUKAfMG7kdH2UMuskkXG0uNXIdWT9p77JJSK+myuCEQwGkV4ZYbilNGVw9M2CDVHb9p0qXgF hACWf1InaEOFAvbGzYuaRWJLTy2E90zfmoLq0QLRseWIBYSE+jMy6/KqWWRsPSZh2gMaPKJBlsmv NAdBoEBH65UD7BhsxJ0g84c6xQB/hwE2iWhKOJCC+7TaVTJA7CT+iDAdwTDb4UsJvhCa5/3oYKzE fzf6uhJIr32cXSUDjB1i/8//+e//e7j6pZ2Trv9plTjK2+iLMarmKkOHz2Fw/95cSabdQFQGeYxW UY+FACayjhSP0S9HahUtEnMdglSKAN5cgNB6RhPrqNJ+sVckaRCnAuHUxrKKFml0z5tEbZP0EEAt NEs1yoTFCIGJZnPBV+qT6VzwE+d8NYN+eSMhg90haRKE+YXgbIF+2tVEXXi1ZxIXY4JJx7jUKFQT cUYTI7s8F6RZUa+KXTxfTRtL43vFBntG9HNqLoh6IGwU0+yT1x2owdh+6aynSQgbzJokDYyIDYpN OgknUKWoAiO1moKgqfb03ErpeK6++nA0acPBrb0iem69WiltDrVXq2mRtfe4jUhDqQ2VPsWE9B/e yLAhzL/oJyvVtksl3NglpcT9jgxz3ULCBkdG06saU5wcnbjRtEjthY1KEQM79cxGNRHnnb0iYaNe rVSiwfPVdB/fq9ygpNXvqSlgWuWwUYyCgPuD/D7ymfq8A4y30phT5CYPjIQNtrqme60UNXGkVtMi tafnFmO6rSsrVaTPR9yaK6LZGHaWx6mV0v5QOzWKFlhrjzMjvYc2VHrrIQW/7kEJF8Wo7L48L0V9 qqtl1EZwDCRc/M1zQdwUaiSE5xcpKuxIraZFai9cVIrJNm2R5zWpIvZm3ZkrksSoUyu1wVgRLAfw W+noXuUFRyL1WzIUZBq+XBTDcQbl/gn4RH2mYfl6dLj4zXP4LSR5wUuSLrdSUBCp1bRI7YWLSjHd 5vMm8GtFn+zquzU3SPIiTg2m0Ov4KFpgrT3Pi79v3nrIhACjt3JRjOTVCUilI8Vusp/j0+lf+LBs ckBI8+Kb6khpUSFazQitokVqLlRUigj48LHzwooIkjLjam2QpEV8GkxU1O8ocl4IcV44uld58atR lfKCi5WGi2J6kdlpJi6YcGnkklLZ9SfPacIEJHnBik3JiJQ6VGtfTYvUXsioFFOPWo/WSxJNivPW XhFzP14dKe0Otb3VtMjae5IZn9ovGS50ASjtxSD4T7vmrGBqpExIRkx4Vhse9Hd4YF9Iel5jahE9 NFst/Ruf5pki+M1tvmZAOv5LJ6KhqNrs2OnftkM1aV8GW29Hy3i/doj9X/UmsuzE0K/HPOIvfWai L0an9+vN7dRt+liLEp6vbMYvviOYX11IGWDLSHpbY2KAnWOSWk2LxJ5/0cWIIDehaKQlTcTIaDR8 x94geS/iVbH/iIl6PpoO0vievhdIM/mumhBprdyFjWLEoLU7vX3c2xYyEHIyMFWkx6BHiJzzJol0 tsL4KamiI7WKFok5P7cYk9HcKnfRpDCjSQNA2Svi5+rUQOKifo+ig4y5Z3O3SLMhQ78m0px5a2YU IwRudzIXnKdtHiDljOX6wGaBngsZbA5Kb8vH7kQZVXSkVtMisdfMqBQxMCFkMqJJcUaThoKyN4js UVnbq2LOjHo+mv4nvleZwY4M/Zx490Y1lcwoRllMC2S7fJe+iYAUCAsHeY7UEBI26DXS3wqDM+rx SK2mRWLPzy1GDNzpYTaiSXFG09gr0tSIV4MpN+r5aDpI43vFBodeUm9ycfvkRjGGTL/M1OlX4Lv0 mrvFcUmJDdqT5IGRsPGpDyBepGhUI7Wagrh3sz09t1LMSOtMgV65aCJOBq5Xe4MkN+LVYGKjno+m g2jHM5pesfHz7vEDS2YfbGxJbggjKmKgqzQb3F+fMQW7YdOrMHCVDKkvJGywWJfprkpRgt1pWqT2 wkalmJFmSspsRBNxMnK92hskbMSrwcRGPR9NRcy+43vFBkcY05/6nviwUYwFLeaSyXh+Im7icuxc /xfv2Kan52gBhYQNhiJqaovpV/+M1GpaRIt1fW4xflqmTs1GNBEnI9ervSJub8arlSL2ej6aDrL2 ntcWXJirygFpyubmRjFi0P4escF1vMkE5nPjHVcM6TkiB7F3WidT4xKIx7SJxEKraJGaS2pUihCY JYD8KlKUt9YGSWbEp8GUGfU7ilwnCXFmOLrnmUGnzPdvzcUbrbG5WExrRmzaFxcsRyUP+AxYuMDL ZIGRZAZbR9TvUDxESicQLLWaFtEKFZpCRqWYsNX5f5ERTcTJBqervSLNjHi1Up/r+Wg6yNp7lhka xSgAFoz0XSO9JgMxMfntWU3qWDbhO0q+Tq5WhDUFP0XlCNC0YEdI8qJCFNIWWj0D1FaIiIy6AvdU WtP6oX4lQqa1LrYGSFLEocGOz6Nng4gtPHxMgqdxEGZ7SFnQ8eGwUAz3dYBYiffGDhPSTPWy90Np mObnisg3djYxTRiyIqXpcEuNJt4YIWjCO8mEiUoRAh9JumhSlLf2BuG59WowcVHPR9NB1t5zNtjy oCEES0j6dmrYEEZUmpvX8jV2kxGsAeRX4t6Z/PxGQgXbQdLXCPMrT2l9VaMCNUiN6bnFKP74voGp iCaCZOZdmjT+6hhFiCiMSxXhT2axonrUHGSNPdnHjDSTmfoltZSWje0HgwcvXGOUy0zSBLCLJK5x bV0yQIh/XQ2g8FK5Y0xBcqRCUqPpILHXt6pSmNFdx8qvaFKc0i4qYm+QUBGvilF6bjSj6SCN79U7 QtLrx2Q15ZODbsmKYiwqaYZTPwEvS9KAL2DFO1I8SWAkiUGKp6MR5sQgDkmNJtgoEntmcTGSj9kb sxFNxPkWTRokSlORJka8Gkxs1PPRdJDG94oNvuGk31OLCpgLG8W0rAAJYoOv06v0ZMpSUupW3vJc kbDx9c3o/SLlzUbSPppggz1KWvyuPT23Uow39W065UY0fULCrb1BkhvxajCxUc9H00HW3vMW4y/n r+wd02XDhjBi0ES69h/CBqOhZALNQthglSB5YCRs8LWHlJ6V8uGDqyYi53yCEF5xaQobwfiNmTVU 5mlIJk3E+XNrb5CwEa8GExvxfDUVOfG9yo1ffQDe3vG5kuZGMdjgRfZQkE/Up0Xgg8XaEEQzn+eY yBQSNhjFpOiolE4fWPtogo0isDHPDaYYWHgxG9GkOG/tFVF7w28Ur1aK2OP5alpE7Mfe49zwRPgn VYjmFMUG10qEjcWYSvdyFnb5+GAygYsswsZHnmPiSi9R2Phls5BelAqxjhKhUQQZRWpOjw1GCMwU q7nRCBBFRMknE6/WBgkX8WkwjkE3liqCryDODEf3PDMYKWlPlLnglXVmLEYIHBhzS80uorQHfP8q 3nHEK1lgREGxmKXdVCYjUjp9YKnRROBFIEMyISOYYvjTFjqaFOetvSLODM1lyauVIvZEs5oW0W8d e88zg30uJoPryoeMQFrE0MIeucYnipU9TNBISG8Jq2MLOC3Y46RPpl+EqCItNHpgIoAWMFATIgzh PUh6Kqlxe8G3si+2CNmAkmIcGkwpYJ9XzwCQIFt4+JiElJ5MdNA3mQUK8WGhmLbOaf8DVpmrUunJ zIykRANHD/VckaQE+0tSUlTKhw4kNZrIfI5ZpPS0PT03UoTAt0TVMOuqa2kiSgr9q71BwkW8Gkxc xPPVtAhs1N5zNtjDkdKTqx+99QYNwhyD9puIDO2KUORcIakdMXqtLVEkXLDRJNf6VcpHDi56CJsz FgJIbSkKFcEIgBuyQ4UUmYoPpGg8xhxhC9Fz49NAYiJer6JF1tzTeU/9YB5G4Bw/Q5uKYihnDkdW GTyWCoTs228egwo9FirYTpJqwph+cNYDlShVdIBa09LXCCkC2lBnRRQpSikXFTE3SKiIU8Vo0CaW 1bSIuIibz7OC90iFg4b+jExUbPFmB1PKmQp+y7QBbKFMucEFYXqKaQkhoYIKRN1MMEfJYQpJRQ9R 9+/YalIEw39O1+ht0MBPeojxz9WaGw8jYSI+HanxejWdONZemXh0Rh/pX64GMxM+uxgmitEQqQok K/i4fZoBH7pUr86lmdnWaSRU8MlH9TJMVOgYpH5wTk1IeRUdQL0zInosQh6fc54yXEQRUfrM5Zgz O0bCRZw6Uonloul/onveclIbq+pVVnAFHrtVybTFCEGLeXDBG5okYPpJTanGa36MuIWEC7ZrpZeJ FGGyxUsVbRUVONbChYSIkzkbttg7L6JIUd6aK9KuNE4dqfF7NZ1IGt2/uEiRxZ4uLeaZC31rUXmx GCGomYQLvoyWdoAPX4SLzzwGFyDtTFkWV60RzD85ByekvIrgogC+SVG4CKYI2FLDm8UYxIoU5a25 QZIXcWqw4/dqOpGsvWetBSPIfMKGWpZxe7kQ5p+Tj+8qMSjcPjyEYFJGUkoMPtORJDCiqDRcS/ZE iNC5/UJCFD8ocgNSpOZCRjBCYMtJuk8rIkq+X3y1Nggt7vo0mLhILKtoEbLe0b3KCy6P0o+J9BeV RPKiGCF4NQu7/NppCj4lJS644ShJIMSJoakXbzeAMUtBxm+kRtNBai9kREox8I01KBtNivPW3iDK jPFqMLERz1fTIhPfv9hIrYXjvnIH5zTpGTICEQFrWf4N/ui6P6X/p4TExY+fKpC0YDNJ+t0IqRrv bT7S45YAgISrLT1VGXnPlZXOrqhRhDe2BggNcsgdS4UaxuqBhZu4XpHANg4VgPzGujo0LAhzVjPm zuvB/StJAF6T0MDqYmI2Eh5YhVS5xi9jKWLkOMVV00FqL0xEihDYIZdCIpo+yalbe4OEi3g1mFIi nq+mIie+V2ywuURlgyYncDxsFCMDWfdxUnAHsXOCr6CJDMqWPEbgQkIGO0nS7UaK0DmPoOeiyCkg JM7NcyOlEJhLcl5Ek8K8tTdIyMAp8mKFKPlsbhXBxW10r7hgS0dylmNIw4WwhMAXp/xack1v80BS IuM9MpAhJGR8E2kyw1KQQWNz1XQQuJ/nRkoxMHoyGdGkOG/tDRIy4tVgyox4vpqKhPzYe96NsK1D bT/9AgtBSYxCdB6fDMfURlFHJzN8YytcUCqnxxAQKpjnSbebW10pHiMTNU4LIfZMtvTUCOE+3w0J EdFDiNrvJt5jq0BoiD+BcJBhs22tmkUa1/OUYKKFVDQJXL7lwnshygnW8ZwRbAdP4jMbYcd4ARKx gJDgCyrVUliG+HjfMqEZNUXGVkmQkAtJ+pCQED0EyLe+LrYCuLxi+sP+rMy4TLMaNZ+DQAJxYetx JqS4YjqO1bo4RqeoTFgIEvQRFGUCmwzyw/M22DH27ugp7bfP6RaNlvjlCvlHZl5SMqMGEopoIj37 ggeS939+dZ5q9CjAG1sFkgnxZ2XG5arZIJx1sfWYhPSd2lPp4QL1M4OpkFCIIlv7U5Vt3DaYcYZW XpSiXP2XiAXIMVbRGK+YBMsQMgcVJDNqDmJbrUojhPcMy9MfRQ8BMmlxsVVAtuIP+TMyicKdadXc xfWvTBgSPkggXGVARW/oIQcT+cGYRGcCz1b5fr1+VXJCUqLhOzJFwgOzb+phg6VgvtUEEdzjdbHH c4vBBDMzYSKaiJI5CzVLY6+IeScN7dVgSuJ4fjQN0vhg/nlKfOlKXLOhk9TJiWKwQedpNjgMl0uW qPTDBgcJG7mQsOGz2XBhGZWIkRk9B4F7ZKYktZQi0K2fyi7pUYy3tgZJVsijlYIvbQyVR0fPIGIi Pj5nQgvlIoJr/X1CBJWB+CmU3cpFSkX9ljRPCDkruM75AKGBfSUpnyTkd51iEaHRAw8BYkvkDYT3 zH2Hhagh6s+LLRcLBkJDHDpC9fnomSDW1mMSpplkXVC+I00KNh2K4T7Nnmlg22l+fD7fJRrUC/o5 iAFxE66xmCuuYI6RUxXSPpqIukjthYlgCkHbAMmHKFKQt+YGCRVxajBlRByPIrwM4MbS0b16NViC SlnJcGC4KKZGUbtqSYlvbRkkIzj/GyrYf5aC0Yh8ozi6lSJw6glJRZHJEWLnbC5UREoRXNQQIvM2 V2ODhAjbKuQ3I05XC3t+buJ6xQLbNBQgbT4fzG35UAzn37XzBRYYciTn6TzSl3KLcYkREhrYKJK3 RVJ+MzhNIanRBDFFYq/vRjAC0B3m4jOKFOPVnF8OIyEiTh0pbQztmlEUDdDosNa34+HMnXZNaiqB Qoy33nvdPwcjAl5FOafJ/I4YaD6IkreDrR+JW0i5YK1M9RUTq8LUAnAoIc1ENC1Se20uIwUXvngO LqKIKGlUQn3MDRIu4lQxnBrHV9ECjW65eHS6kMe5FFBlHYsYbKh1XgwGFwyKwwWHi5MFrPXHuS8/ R9xG5JxGYl44DOa84AyC2BlNcBGk9poXxTT8arcUTYSJxnAfe0XMPQW1vRpMZDSa1bRI43vFxodK YrPBjVtuN7W3Uphi+NLZb9oftm1qiECcklJm0KSmzDASNvw9kSNF7OwJvmoqQrrjnTTx3Erxg/Lq pXOKJsV5a28Q2RuvBhMbjWY1LbL2HvciKbG0Z5LNyPKOKahhQ5jZoBNhqQ67XEqRN4ApobDxExle GCFhg8gy61cpGoY7TYOMPT23GLMZzG2DaKglTZ+sI97aGyRsxKvBxEY8h9dquo/vVW58aYxlNtjU 1dwIRm68U/i5kebG0WQCn7AKG9xIn2LDSNhg/2163krRcUZqNQ0iNqQpbERKMbAdy2xEk+K8tVfE fTjFpL06UtoW6mhW0yJr73FupMLQTkqSU2yw/DJsBIMNlnuSG1w8oGxhTk9SHpv5uSL2Tktn6XQq RKEXoVU0yJgLGZESGZTvkF9FRMmI4mptkGRGfBpMmdFYoggvgzjzHd2rzNAllqaCSqeJYUhMwDMa KWJ5h9xmcOW2fNOAKK+NgaQFu0fKRISooS20egqIB1VV4cEy0MCIT22zahGpUYQ3tgYIDXJohdyT JIzVoz2il7hekcAODRnFM7UaGoGwt9KYaFAbIRoowZP4XMwRGpj/acxCwgPFc2ovSeEbQ/A7TYPU Xl+PSIkK2ghTEU2KPJo09rKfQfzceFWpi+dH08Qy8T1nQ9smWZUVG1xdno3Ng1Gk8BXBsEEd5Zzg Y9TpZz/8GG+HkZDBoCV9b6UoQyO1ioI4X20uaREp+gNuVA8Z0USYnO682hskiWGnBuIs7sSyihZp dK+4YN8Ic5lwwTai4UIYDQNlj9f2yQw+WJ88YJNSnOPMdgI3EjLYKBLKKqVJ/VtNg4y9kBEpkcFb 4syIJsIjRa72BgkZ8WowsRHPj6YgJ75XbLBRI8Unt4QOG8HEBm+l30Wq9LwBXKMT7yh1U3wakXca rnkVkX9ECgWRWk2DiA3JhI1IEQPzEGJjNCnOW3tF3NJQWtqrI6VdoI5mNS2y9p51I+yb/NRQAu+Y Txs2gsHGDxcR+W18/+zwgi+GhY1vP0fLIsTeaZHKy4jBeE7r+B6orKZBxl7YiJTYYEUQ9kcTcX7d 2htEz+m+NI9CVmqjWU2LNL5XueHvHJiND0ZIakO1v9IXuRMDf9ku52QywOAixbDB+6y+hMiFJDf4 YkEKtUrRCN9pChL257nFmNDUpTpiI5oU+9UeGV8kbMSrwcZztK+mxHK192yYhnKGJJHWKCJsBCM3 +Pp02PjgPtrEjhS/nndhpczmm9XZjq6hW3c+GlMnyTKfpFbTIMoNaVJUixEDt6+YjWhSnNJ+7A0S NuJVMbzaaFbTIrHHr/b4TUkxrmbKe/6YjuOmDvevgykG3Z2hnOSr9YmdVzzekcHJAyPJDT5hoF6Y aV7pUpzcjy+p1TQIbMiec2ox5nA5AWo2oonwmOYK+7FXxM+NV4OJjUazmhZpfK/Y+PumvYh498Y3 OpsbwYiBL1CHjW+dgVScfCwtPS6djIrxImEjN/2tEGVkhFbRIDXX1IgU6c0ebr2GGgWqz+elv7U2 SDJDPq2UM6OxrCJKk5voXnHByVuV4kxRa+kq70kwuGD9JVxwhCeR8/UrvTkauHkQwjSekHBBqaBe OJgzg0GvcmU1UV5e7TUziukqLvdXTKRZE7FraQxNY69IMyNeDabMaDSraZHG95wNxsiufVmcYMHV iVGICDiyHi4oBpP/zEjFt3xFidj0VKhg44i64GLEDE/iZvUUgHjb0lOGaBmZ0VLzoqSIGgJjz0Z4 t60BkhRxaDB2mjSM6IG9AdbWs8ZC2ymHBW6KLAvCeBlw/49XaThT316Do/Xx7NOPKWQhoYFRtqqw YgqRlbPygCLnSBARMc+NlJigBzMT0aQgb+0VaY1hp47QxrKKFllzj7nISJX5NC1K6fXgzqrhIhhc cEuWVqwoaenqE6akMmLzIITAQeycxmxeWQyGFC2otR9Ng4y9ZEWkRAatKZSNJsV5a28QPTdeDaY0 aDSraZHG9+r1YNsGPyfe/XDndjNDmDODmcewwRxocp9LN8MGV80lDYzIO4ZSOtat1KgU9cedpiDH XtiQFBiu85NcNCngqz2XGEbCRrw6UpBwp2mRxrdsPJwJ1pqMtiSKDZ1UUtM5mH5RnZ/Sr8AKY2NH yk0ZX51OCcY3TvOisOOCW0vChjC9+oxRJLWaBqk959RisMFHeMxGNCnOaGLFI34WCRvxqlJuIhrN appYJr5l4+FcMH0xHyc1G3wH+scdyWDEQPWoBS16eiWsfnXqQjWnGrp5GFIkufHJYt1K4R1lZKRW 0yB4J3ttbiLFz8Dd2GEjmjSbIK+OvSJ+brwaTL9jo1lNi8Qe7D9uN1JwUafxbaOwoTo/uRFMbGSR kyLunZOGZkNSKsHe/Jz4ERI2aE/T+RoTGzTCKT+rKQgZj3fz3EiJjU8dMOSdiybFeWtvkORGvBpM bDSa1bTI2nvOBt950lww0tpeFzaEwRAT21nspIjj6E1+dUllkNRhCJ89Hza4y0cnKGAoUtrwfatp kdpTVIsRA7d5mI1oUpy39gYJG/FqMLFRz1dTkBPfq9yg11D5yeoNizxloxhsUPmQp0yz8KE1s8EX wsIG4+fkgZHkRq5yZqgRIR1XsPIocuBC7JzNhYxKUR7lBIIW79SqMm13a22QcCGfULVSzIjE3Cpa pNE950LThborHS6Yh8x2lGDOjL98i5YQNBXJTIN+c6Y91aqyFsFzBwkXbC3JiZhK6byCpMixalqk 9kJGpTDD/eTOjGjCzPetvUHCRrwajOMCjgZ7q+k+vlds+AsxIkNf6tRrwqyy6w8iUOEnLtgskvzn AFC40MkFkyMgVLDf5IIp/X2qQkxEzwCx1c4kMnjPkq55iBpFeGNrgNAgh5w6EiJ1x+fVM0DjekUC uzoopsQCXaA7Ev3TmFYAtXLBq+fxNwmBDAAeWkIpIkSesailw2tmRprEAluWpHv0LBBjfmwx/H93 XzyKbOVizm8CN67FXH06Quv1KlpkzbXZfFhg8AbpAxxigns9Jh+KaRFOgdNd6zshqhO4zTtUsCs6 mwiMhArWhVNNVIoaMspH0QK1ljySEGHSGmldWpxGEVGyTe6Y8+9vRM+NU0cqsVw13Uf3KivY2aEG Cu90O6oOidC6ByMEb9iifmNfVbiQkNKCi8+UArAjJFxwpESVZzHXokjh3CiiOL2xFi6KwYXufhUX UaQob80V8Vs1Th2p9Xs1LdLo/sVFBiVsBOG0tLl451OhbSeKablV7QTVG4edHCWfSQsXXIKfHBBi 53Q9q1ccmemNlE8qSPkoWoC3WCLhAgMSos/mIn9+i1FElOxhv5obJHkRpwbTb1i/V9Mia+9ZaUHt rU9BKy+4q224EOafk7kJc8Fu3EYuIbWiXEWYHPgrJHnxpe86Ky+MKS9YbpZyuNC3uA5Qa+GiQkRA eeW8iCJFeWuuSPMiTh2pjWU1LbL2HnMxecFvmKLzTVu7038UY2VNX/9WYnAbeULna3DpUvhhU04K aWLww6rL1YKIpXw4QdpX0yL0cpIJG/z8kqK21SZ0ZUY0EefHrb1BkhnxajBlRj1fTYusvcdspATX nksNIgj4V0vEYUOYM4NEDBucvknsfDMubHDUJ2lgJKlBt6NygxnOSOm8wq2mIMde2JCU2WATV3Ij moiTLDz28KlI2IhXg4mNRsNYJZoWaXzbZjzsS5gY1dKV2eC6EbOxGCtx2sKq3ODe0LKBFL+ehkYe fBC5ngsb+RZGIb0W7KiS8lW0CM75Mb88lUIn1xn7RbEiRSmhY22QcBGfBuPCEMcCY6toIxlzTwdn TJmzv8NcML0GFzi3GCtP7KvDFyWZBx/aeuj9uUpEb3ksIu8YsFGlODMk5TD5uEKKzmrSCYarPT23 GPO9X+1HowmV/ErhPvaKmPvxarCL56tpY9G4BE2bGQ8H7gpYHSvC9Ip6Swbhv2wZMhXv3CKXINnr r98p3+uGCJ4JDyRwul5BCtBnV6UsSgaInXCAiH9Gd9siM0r0Dyupnf6tZ2gI7cqKjLNET9/vX/Ia z6vY2c6hepDxLAEm+EKwqKGJTBJNXn5/8p3oOTGY31xAwmc7SXrbyPhggjRHDZ4HIdzY0lMrxCQR 97yA8CZYD//Q173EdGwVCAXxJ5AjrsurhvfwGtcrElg2VplJz8AKZkgoxIuVNV4K1v68TJrZL0ZU +b0FhAOGcOllI0PZjAwRr5YgY8pPLcRcpr4QKw6iB3qY6rvYCuDuqe6siHa0OIbVsoijwtTjviI9 J/073+O2X1oFDgWFWCTiy024QcGaO+KYkUNIv82vn6JXALBf9Pi/XlAcGerlaF41i2iXf5ZbF6Is YSMWP+noIUDGpBdbBZIH8WdlNoqo4akgzrnYekxCOkyKX74KGRL0VYqwUAz3cc1vAyek88PzieZ0 GnwVLL+7keTCN3uX3GJUykcSpH01LQIR0qTnFiOEP+4eNbCTJkV5a69I8yFeHamNZjUtsvYeszEp we5q1X4U0gyHy0Yx+l5KTPylQMslcWy4k5S60D9+jqQQ0qz44cp8s1EpH0uQ9tW0SO2FDY4zSIrF Ft0zp7yIJuJkQHa1N0gyI14NpkSo56tpkbX3mA3nBsHwc+hXQJq6TmwUI7mIIQu6FGgcO02ckhIb H35ObAhJbnBXUHoTYy4Rkbpq8oGFi72wwSZpYcSga87FRjQpzqs9NBUJG/FqsPH8oul/4tvm4lEx pWC4GtZs+DPjYaMYJDAf6dxgKi+FNneFqSeBPD8nc/o8edhgBi69izGxwRSctK+mRaBTmsJGpSDB qQgrVsTz/h77MTdIyIhTxf4jMup4FR0g1v7ZdE5m8BHVFJZ8F8JnZtBQTAlBKMo/vvjCb6fIkXJm sA8lWWBEzvGq+vNiEYIwHVOw8lW0iDILRaGiUkTQAY31KMarMaeFET01Lq3Ucbt6DrDG+o78Iys+ tMlY7whfhXA1gYZiNBQsj9osH7TXr8sLiJSzgns2UzAKcUh0Y9xUvlI8p1MKllpNi8Res6JSNBTe xLqKaF+kXFlRc0XCRZ0K5qyo41V0Iml0+4Y8KiqR9veMRQW3y6umWAj/uQrRTLC5QJkDE7qCXq79 aMlrgKQEe0tUnETITaAvAlHTGz0qlfRUbPWlCoTzsqS6yEoI72LJ+WAgHNSdEVqPo2X/XDuPG0x3 HwizUyOVJUeBJhmKaYrDq9oUr74Vjug4jxMK2GKUH16If1SNwbSEeKQoEyO1mhbR+jaaFNFi9KH6 AgM8VJFma27NFQkRdWqkTjBW5N9SkcDfWnvMxTQR7NNIgclU5XBRjKkvlnFxTtMKCZIrztO3vuUx Ou5+h4wI2CqSFKmUTyVIeRQ5bE4uqOxkQnaeWymm6ijanBPVRMlxZ69IuMAp50mEoKJ+j6KDrLnn XLBZQxqRVtOWN0NYQlArKS64jyNZQEkdMt4jAxlC5JxGR1o+RLxSlI13mhapveRFpYiBc9lXTYR+ Z69IyKhXI0Xs9Xw0FTnxvWom2K+hn5NS+pcKKWwUo97++57M4Ev2aQO+JZUhmZ+DKCFhg4PkKg+K Kf/ZPi/tq2mR2gsblVLN7ZG/xlzWRF1+Zy+I30pIj1cjRez1fDQdZO09zo20Gdo9qcGD2GCIZzYW 03S590apqfBYRUnSQoPvVeu5IG0zuPU83Uyl9O5aajUtUntho1JKHBctLF5EE3He2SuS3KhXI6WF mNgbTQdZe4/ZSKuh3ZMu/HFA39BRbgQjuxj/ZYGfbiG3utF8SCrDND9XJLnBtFa6mkrRpqP9qimI 2be9sCEpMGLgCwN+U6qJcevVHpqKhI16NVL1/KLpII1v35SHxQWr3UpF5cZfDmiUjWKw4QV+8o0P niQTWODVbmRaxTwHG3pO3tEUKNGVL8ZcSEdqNKm0vtpzTi1Ga0oNcdVEnNJ+sVcEe8erYFQXG81o OkjjWzYelhfaUkn/ZTa4i8P1xWLE4AV+7LJOkxaBq1PjHdt9mwe91gM2/E1XkSFValbZmiuhUXQQ nJvHVooQckXZKCLKq7VkRh87Po0UkTeWUXSQNdf35B+Z8aUiV1xwecNkRjHmrvSlFP0GXOSZ9oAr 9pMZXBuaLBDSzGBjSfodYyKDs2mSGk0Hib22NpUiBpYCnBnVxByXtCszaq9IMqNeBXNm1PPRBBu3 8b3KjG/vQlIbzlaFNBqBiOCLBtCVqA7JKv8/EbJvfB3jAHlJGMWl05GQ88K3rokJ64GJALHVTiiQ 5vE8f6chn9Voru/YclIYCA18DVQOjRBBx+fRs8Daetx0TkfCPg39jEhryBwWiuE+Datp4BhhE0BS GRJ11IFDmdFiskRX4pmsSDHTGu2jCTNFak/8LUYIDPfhbzUR5Z29IuGiXo0Uodfz0XSQtfeYjelI 2KehX5J6iTuby4YwN//vnLA0G9+tKPgmacjgxFQSwEiSgq0iKT8rRUkdKSamUeQ0KVJzIaMYIbz5 mJ1GYdakc8i39oLoveKtjVMjROSNZRQdZM095mIyg80a+jUppH+pcZIZxVhb8CI/ZvmifZoBBk4h 40+eYzobxC+9RmtaSOQ1qhRKIzWaZObGXsgoxiz/20cyo5pYFbmzVyRk1KuRIvZ6PpoO0vieNxba PenRBNLffPxbbARzZtALsHoMG4SZROCbpGHjI88RuRBFxcoTH0ENG5HSD8LWXzWd0eQEvNoLG5Li OYaxs4uhmojzag9NRcJGvRqpen7RdJDGt2w87EjYUfnGNkHlBjcLeykIDcUI7482oMPw21vHaazm KuHpOPMc0/x6TlFpmYp6QLlhTO0BxxKkfTQdJPbakVSKyY0sIqwmlhSl/WKviNhYr4LRkazno+kg jW/ZeFhiaPekBwE0ZL98rj65UYw1LF0264L0vc0Gk+Dxjno7eWAkbLCSmA5HmBsJKnxJjSbYKBJ7 7U2KEQOTg+a1mrhyJJpc80tTED0HG0wbK6eKiY16PpoO0vhesfFJg4dGpNUzhY1ixMDap+1y4jYD Uj5LGjZ+8hyzbULCBvmRfqdSFKSRGk3QIoSMrz09t1LMAfJlR7NRTcR5Z69I2KhXI0Xs9Xw0HWTt PW5F06NokyVTd/aOgMuG59f8i+p+PP0KHGpqJkhKmQtRzQMhYQNCtQMShiKlORVLwYY1EXkRvJvn FiMGkad3rpqI885ekbBRr0aK2Ov5aCpy4nuVG+w40u+paTYqi+RGMXcSbAOBjc9vD2C0b8E71TV0 83NFwoaPzRBLhSiqIxRFTgMhds7m9NhKEQI5ZTKiSLsybq0VCRfyCVUjReT1exQdpNG94uJH388Q FUxClQog2CGAP2xFExN891FZQJASUg/zZpECIYKdI5phGyGe9z23q2eB2goPkdGCgc9GaPjHlLgj xBa/w9giZBkPDXWomJqL+Dx6FlhbT763gzA7NlJ+vvMJj7JQTMuxVCCigambjCz4X3ENcT3HUEHP hQc2jajzLabfmoMKkhpNB9HiL5rCRKUIQScr9HpUE9uWpF1U1F6RcFGvgrkjqeej6X/ie54S2lfp EQDesRjmjmQxYuAqbrPxlwbBYSJk5zjs08CFhAw2kmSikG2N2pDG85EaRQeJOT+3UiyIfLqC0WDM mgjTmiCj9oI0MeJUIcawE8soOsiYe3YZvq9Q9GCC0TNbi50Z7LX8Y4zxgT7moczg6tnkAYsghIlz 7CxO4EJKBrtHMrtnTGRwTEFSo+kgsTdla6SIQa+dMqOa2O8n7Rd7RZIZ9SoYXq3no+kgje9VZjCf pgiQ1jcO1HBqt6UxjZa0j5Jso65MInClZLzjo+ONXIhSg95Dn4dRAknKqcExBUmNJv5RJPaaGsVY 9dBU10WTdj9KE+zXXhCnxnpV7OL5aDqxNL5XbLDDJOXnGxfONDeKEQO/k9mgp9T8OXFKSt5957ki YeOdr7iGDUsROwcVUn5GUxG31ban51aKGLhmzmxUE3He2Sui3NC9wPZqpfhH7I0m2LiN7xUbfDsu PQn19rAhzL/op7fKYJeBdDNBUmLjJzJq7kDCBlum1M0VU8Zz7OKq6SD8VvPcShGDttsrN6qJYO7s FQkb9WqlaEBibzTBxm18r9j4+uvhhJb+eenzphTTlgSdJIUNjs3mHfiWlNj4m+dYJxMSNj741oVz I1LEzhkLFbej6SC1p+dWimWxn5b61cRLe2cviNqb49VK0c3G3miCjdv4/sVGBvHaXOnhBN7pAo2w UYwYdM2G7HK3Wt4BTjCJDQ2S/BxsgLg1ZPCma26VG5HiH9x1I35G00FqL2xUihi4RiTDwGgizjt7 RcJGvVqpjWY0wcZtfP9iI8U441/tC9IAmmtbfZwKDcLIeApDfbJRbLBlRL8xFZGkxMZ7niuiqDSY SjcTIaQ/IxRFhLlIzYWMSrEwpg07RxFdxJ21Inwm4Pi0UhvLKGokJ7p/cTGZwXUdFmZyaagIRABM 4ZgJrgJJ/tM/hIk/forQAJoW7C5JFxwh8Wah0bOAVsdREx4iwxoQ68umQWr4HSgcbm0VSErUoRXS VlGFMXpg4RoXHj4bnTFxqP26Sgh2qYaFxRg2aYZb5DNsUoqQ75JSQnzkuSJJCGbE0vlGCnEOLKT8 jKYiaKq9MFEppvm5PtpUVBPHi+7sFQkX9Wql1vPR5I+mXeN7xQZbOzTTiXe6+EmNBar87S1eD+Zk wgZbT6BML7/7WQ3X2KO9SMhgc0n63kghzjY4SY2ig9RcyKgUdvkyl8moJp21urVXJGTEqRVav0fR iSTmXmYGWzz0frMO3Muv0VCMELhFCHZIQ4Yozgx2FKsp1YjNzzHjLUTOaRylXY9EHCn+wYGFlJ/R VMTc217IqJR/2HRS1aSUvLVXRPbWq5Vaz1fTIo3vVWawz0Pzk0i/0WUkM4QRFdNRHqXRNOFUwpSQ yPiOCIELCRksEmnip5iaSc4rXBQdoNbCRYWIAJNmVYowR5Qop5Fcc0Fsbp0qpoyu36tpkdh7mRns 85BdFnoZRJSLYnCh5RF+Ao70J/+5ATpccHI4FYeRcMHyY96lSNGrcPZCUlV0gFhri1sh2jo+xmMu qojR+J25IskLOWXGJAVjqrFsbjUt0uhe5cU3C/3mgkO2PoCJzmJM46r2gQv2KCb9KSXDxW8ec9k4 ecHO+fS6knKVwNkLKa8ilZRXa+5HFqM//OvOmgFgFPkA7zFHyhUJF3Vqpdbv1bQIC4xyfLl4OCfO mB+pcMHeqeZFMUKgUBIXfGs8WeAbsNW58IXwVNxGkhd8ITz9bqQInT5FyqvoAPgmazxWjDiJgDGG 86KKiFJSYw5FRcJFnVqpxHLVdB/dcvFwRpwpMc6jhIs3OiCOzZAXxViJa+PJ5U7JfpYJ4hy7fBu3 kHDxzRcqUUWuWIow2VecKVUrKnCshQsJgalOcunC4C+KiPLOXBC9W/xAdWql1u/VtIgWnu3m4xoj hRbT6NoBofaCa1wmL4rBBVtnbJfT4/qBGVdKSr8UIwU9F8TeacimN2mECP0rQqPoIDUXMipFuaxN ahn7SRETunfWioSL+rRSG0sU4WUQMmXNPeYiBXiOqkiYdeVQMRBLYN4owi/A9Rv6eZmDkJCYYAeP ozYg1xit/XrZcYSI299w4gWJngViqw1nZPiROe10VcMI6NZWgdAgh0yWhGBvfF49AzSu5y8HQvqc hVnQdznVmS6G+94iAg1cTZqfn/1IoYFLBlNOGpFvGqfRsZos6VIjwDEMSY2mg2BEmnjuYITAiYqr JiK3JmivvSB+G9erYhfPV9PGEnsw/zwlGAcpAjjkA7BuNnmvgmkxUlswSTPWf50TvAAhg5c5gRsJ GWwtSVUVKcLkLIOkRlER56vNhYxKaWzdwqWaNE9wa6+I7I1TK7R+r6JFGt0rLthLkQKcjw2VCkGJ gC/s2CorPUmDH4T0erBzo1QAhAk2UqShQMbvO/sobtRQlAbBsz61EO7TI6CazVfRQ4gXW9AXICzU n5GhcojmVbPI2HqcEGkvKcJ0rYgTgoY/b0chfkYu+DUJrJI2ZoREwt88xczuu4894PyHNt/o3dB2 eDUavvnE2RA1i+AYIgpnIbxnzh/uVg8B3toKEBLqz8hsFFGD4iDJPNt6TMI0lOzm0A9I+8qek5BQ iDVpjpGjUbMg+uAcAepaTpHwm6cCJBM+dc/qkfGBC2leNYvYVvsZTmVIiNqN4QiqNcizHrqlW1sB QoL84amRYaSFGhxcNUEQSVyvXgeuDtMPSIfbuSvtvPToQcubTM2JBHYhJ9WpdQAY/eQpJoIBQgIb x9O5RsZnLVJmS41fEA4xHFt6aoU0HasrqSChejSJe2MrQEiQP35DIhOXzWXVLJK4XpHA55k1u4tw t7YynQSERkjgYheTwGfvk+q6B10kcB1JfnYBIeGbz8LzFLy8s0mN+Dh4caNmkdgKCRWCBO0zFwnV Q8wXW4m4tqic48/IrMurZpGx9fh1SJugmWZUiwQ+uJRGYTEWGjneYRbY1ZigP333CzSwxzA/vBDn Nvui9FU78VApHcCw1GpapPbCRKUY23GXtBuGaqKGv7NXJAlRr1ZqPV9Ni6y9x2ykcWD9jEmZsEGL 5tYhGD8CMWhLql4Mlsgau6SUFJ95jsiFJCt8tZ7IiJBOLVh5FDlPiuDcPLZShMDtdU6LKCLKO2tF NI25Pq3UxrKKFllzj7mYzGD1XL8mKacb39VS4nkwVHn3hwo23PRvzl0O4YI17WSBEGcGTrGod5XS oQVLraZFai+ZUSm9dfpaGw1xNelawlt7RfTcerVS6/lqWmTtPWZjMkPT6SZD+2ZCRiAi0DYZF2w0 dYlSQhnzpAVl9TBpwXv7xWSbXhJjai1m/5v08JQBt0vsapy+MzJ4/9WqqWqI8NZWgdDANZ2aXl0h 2jWHsXoGgITYek4CWzU0KYc0FzqVBWHEAw3e9QH5/EwJUEJi4ScihCxErmkM5sW6Ymr9OKpwo2iR mtNzixEBXxG5aiLIO3tFQkWcWiE66ZhbRYvE3IvqmoJJt5KJCy5fGy6K0ero6m6lBMu3YowZe0mJ DIZa+f2NhAx2jCRPKuUzFpJaTUHMve2FDAZ7kqJn1qEZWK0imuI7c0XCRZ1aqQ1mFC3Q6J73pAyI aBHjG4uizQthieCTEkFckGGNXFLiggNHGVYwcJrEYF+zZjFhLFI+p3CjaRG8m+cWIwSPNDTI8kIk USLE+7nmgjgP16liauPq+ChaYKw93V3BNFI+To30Oxd3pKUopt5IhyHhgq+SNHKk3IDzCZTmgJ5L XrD1NN3uH2FKf84pSGo1LRJ77YgrRQjsR1BeVBFRRpFGWFZUJHlRp4Lh1AYzihZodK/ygv2qKf34 yHVOoGqLpTHWMXXCWlwwiEjkHE1OB/snMtpfBxIu2HmaTlcYzvlMgrSvpkWo3+e5xVTCp72oIt1v fDFHyhUJF3VqpdbxUbTAWmvb+XASUxsnXeYjTd/pvFiMCLwLBi745H3KTW7NU5RMoec5uNBz4YLT 7WlijYkLtl4qU1bTIrHXvKgUIVDZKy+qiCil/GKuSLioU8Fwah0fRQs0uud54Z81RSdztsmLxVTD UzwrL0iQVNFsFYtz/Kx6DuqFhAt+VnW6wZwX/K6SWk3+DY89P7cYg3zsmIxq0mBEhyXgvvaC+Ln1 qpjIICGlfTUtwlqQNL1ig588RSfjgJwdovELppdDmwh5FXlfNVFDnJKSdxwc1XNFxAbvqbfcrBCv 8p2iRXAORU2MSmnGo5VLFBHlnbUishaf8CCYE6N+r6KNJObg8FmFof2VHkLQPvHlI8/gLcYQV9ea yS7f/0rkjNfCxXeeo4ESEi7YUZKCQpgz4zdSq4ly4GqvmVGMVpMvhpvVaqKxjSZKhDwXxM+tV8WU GY1mNS3S+F6xoROFKjG4+H0SIxAR+JMFcMHndJL/fMhaUVLd+ymo0FOhgo0iKSeM6c3PQYDVM0Bt 6SlDZBzec+bYPFQNEUr1sVUgSVGHVogRkB2KHmfJTVyvSKBO08+IZ7rARR0qszBguIb73usBDZ/O G4KWkDxjGru/vpDQwBaRVBKVogyNVBTh2yI1FyIqRQR/WrlUE0He2Qvi12qcWqH4TSyraJE19/j1 yJhMmys9eGCFlanzclGMmoLraAid4u2drk4Nw4+k1JS+5bkgdo7pM+2lu0hRPUZqNS1SeyGjUqxd /Lp0WU3M593ZK5K8qFcrtdGspkXW3mM2MiZjhYK9X84MbicaNoQ5MzhRHjbYiJ/cZ29l2GCglTQw ktRg/KN3DYYiRf14pynIsRc2JAVGDBwhgGtm66NJs5sXe/hUJGzUq5XaaFbTIrAhTfue/KPAYKOG fk+vJAwbxShz1SgoN7gCJbHz1YOMWnlZkgdC+qJwcUm6XmN6LTifIKnVtEjsOacWU/HtekJjP2ui Ppd25WLtFQkb9SqYm4h6vpq0LfQa37LxcJ2UNU0a77CB5+5IFiMG5lzNxt9v9mfqV/9Gyt595Lki yQ0u20rZISk3EozqpH01qW272DOLixGDrs5WblQTcUaTBoh+Loifow2LV8XYSjjRrKZFYF+aXrHB uUj9njjA4T63Gz4rmZKRXipscEpDQy8aEEnpt/rMc0XCBmcl0/lWir4+UgQTTUHQVHt6bqVoVH51 LNfjMWtibHtnr0hyo16t1Hq+mhZZe8/aDfbHEEK808W76lGC+RdlA5pqY0pa3yInNiQlNvjASPLA iLyj2ea0fziLFDV1pGAjmhbBO8mEjUoRgwq0iybivLNXRM+tVyu1nq+m+/he5QZVtn5PvGPPaNko Rgykg99HRmRpEfgYfNhgzrJ5ICRs+JMRoixC1KMRiiICCBLy57GVIoQPZhNFRhQR5Z21IuFCPqFq pbQv1D6tokUa3Ssuvujf7Ryd6nAhzJnBFzCTGdx6lTxgf3q44M6rFGFGwgULIKpRqAIipZmmW02L 0ExKRs8tRiXCbneTUU1Mft3ZKxI26tVKsVs09lZTkBPfKzb0jUklhjf35zUJpAVoH9KgttPBckXp nfV6S7LjuUCoYMtIuuAKMSdt1atngNhqgRIZZkw5pm0eqoYJUu3rP7YKhAY5hNfBSF1tFlUYq2eA xvWKBKbYVExR5TAP346kGHWVvl6j1/Ir975Rjmu2PmOjjj6MhAd2i6TzFYZvWgCz1GpaJPaaEpUi BK2n6P2oJm3XsCaNxawpiJ9br4qJi3q+mhZpfK/YoMDTL4n0F/1WcqIYMbAHxmzwCUrnxB8JZcDW 8YeRkMGWkfS9lWIVJcpXURDnq83puZVieMJVqSajmgjzzl6RJEacWiFtFbVTq2iRRveKC/aRqNTE I12yEy6EwQ/DD10QpMz40V4qvSDvkhIZVELJAiMhgxP6oaxSuHCnaZHaCxmVYiDLEXaTUU0sJN3Z KxIy6tVKsVc09lZTkBPfcza0v9LDCbz7w4qp2FgMNhh4uRuho09L8C4pscHscNLASNj45Xvp4SxS tI+RWk2L1F7YqBQx6GvZek+qiTjv7AVxS0OhHa9Waj1fTYusvcclRgZqbJ7kklG/J76lIGwUY132 r+4hV1mXO+V4ebW+IzbYM5RhiBB7x9vAJchmo1L0ZZFaTYvQfEhT2KgUMXB6DfZXE6uwd/aK6Ln1 aqU2mtW0yNp7zEYGatpj6UEArTz7KJobwsguLcyygUZ2+dEzxGDiPmzw8WvlAX2KEHnnrU9q34up QeBcwo2mIGbf9sKGpMCY/9K2nIsmjtZd7eFTkbBRr1Zqo1lNizS+fVMeDtTYZfnp4QTS2mCR3CgG G9kWQhHHvXKJU9selBt80U15QOR6LmzQwqQ4MaaMZy+/pFbTIrHnnFqMaVDKOLNRTcQp7Rd7RcJG vQqGVxvNalqk8S0bDwdq2mfpQQB1D/d1uX9dTNNsPn/JkYjcBscEue45lnfMxCUPjIQNf6MLggTh HNV5hFbRIjFnEhcjhDfPGVCtMnpVHlDlHGvODCPhoj6t1Pq9ihZZc31P/pEZ3z/MvqpHwe/JjGKE oIVjvSd8syB5AAXJjN88R+B6LlwwVal+p5jI+BOp1bRI7DUzKsUCwo8rGQ0CrYk4pV3c116RsFGv gjkz6vlqYifpTXyvMuPHYeIct9c3MQIRAW2pueBK7WYBQiAaIJUcAaGC/SXqdKBCmkSFVtr1jlTP ALHVrIiMFlJcxDBFGDVEeLGVpBhbVNhxaIW0bVS2Vs8Aa+tx0zkdCSukKT/pKnKOir2XwagCOVpp Gr5zmRzLBZISDWzYzM8vpB0J201UeKwUCu40LcKYTJrE32KEwJ2+8KdVCWtiJ9edvSJJiXq1Uuv5 alpk7T1mwx3Jt/ZReiDBupMOXNB0FnNH8smY2mxwZaXD5LvwIYNRZRLASJKCrSXpbSqlZLAUyz7Z bLJIzYWMSrHmoRMeIqOa2O53Zy+Iyae2tFMjdGIZRQdZc4+5cGYgzeYO/ZpIf1NOhItiWrnyPhGK uFwmR/ZKSpnxmeeCNDMoB9LZVIreI1KraZHaCxmVYoabGTxnRjUR5529IsmMejVSJ5rRdJC195iN yQxmmPRzMtnyxnRv2BDmzGAcrMzQddMu+ZnKkZTYYENu0sCIvKMa05VPek8qRZ11pynIsRc2JAXG nC5XDF018WNd7bnNMKLn1quRIvZGM5oO0vi26XzUkSDNQU/9nky6qc0KG8Xo7miQXJ3nUjkaeAmJ DHb4axRSJGRwm3wa1EpRU0f5KFqg1sKFhFDFHCb7gsxFFfESX83hZZFwwR2WyuqRaiwXTf8T3XLx qMBAmo+B6NdkOpZhtE7MHIwQfLSJejY3wdHzS0hc8DGSJIGRcEEepbepFBU1Ujg3igIca3psMeap 2MdrLqLog+bv1lwRv5Xj1Eodv0fTQRrdv7iYNoOLePRjMlHPPrrmRTGqZvdzVLO5Uo6ZegmJC26L SQ4IsXMat3ndcaWozCM1ihaotXBRIWapoJ4WYxQRJVuJr+YGSV7EqcH0G9bv0XSQtfe8xWANWT8m c+pM1JULYfychKCpM70kEJU0oGlN74LjSQIjSQx9+UDvTYU03WmhKHLDWKTmQkYx7/hLYlgRUdKw Xq0Nor2e49Ng4qKxjKKDrLnHXExecMJPASDNZ2rLRTHGlqzzmYvv3CdHmJLKkK3DDyFNDF6zVB6V ojWJ9tW0SO2FjEoRAyw6M6JJcd7aGySZEa8GExv1fDQdZO09ZmP6kt+dRGZKLY1nICJg37N/Azb4 JQtYWQ8XOdHP2y4gacFOlNQdFaJHserogUEDJJwmrPvUyBCGrs/SGxI1RPh9Y2uA0CCH/sPKZoUI Oj6PngXW1jMSeKuoIOwZzaA4CEKSMPWivbtKxL/aNKrWz3NPGqV5pOC/wwAbS9ScC/FbwLmFlLPV QRUShEnqTIAvQlbro6oiIFqIzYt0seNY+TvRx5NKHF9Hx0Fq53HseR1oXX88YECWCc0EX4hJ46xe 0XS7XyQyT3oyLMtDAfSiMk7iJrREbxnq5sislkViSrEspLfafVH0JN4bWwoXIAzYnSL6terwaDmI TfGuPqYg7wDdNt+UxHUmitlzEwoK4byXrEgoPtmsHKEa8sYcjX78VAD5pfEYY+ojQ7kcmVGjAvpi y43IQsxB/aUoP3oI0NfzMwyOrQLYqj//1TtgGUKWyzg4aopsXK9IoORJUalPi4YEQWhkNpZz8zb6 TjGbAF1S6oIfPwUJACGBIki9SCBlBjWQNEeNc6WIZubz1Arh/bunyBjVWY8CvLFVICTIHyeLZQi5 UYyag4yt55nAPlZVckyTZ16X5wXFe7YGmQTO6+Y91/EU9QxsNM3PLiAk8CUeVRGMJCgNRQL7Pq9q DhJbemqF6O/Z2uFMiB5i5htla8sRCwgJ8Wdkjsuj5iBj6zEJ0yKwuSNFpM4FJRMK8ROqK1D68UnF vO7+XhskcLdgfnYAJzajr082nZoEFCk+jlRIZtQcBMcQCQkVwntugYK70aMArWdsFQgJ8WdkCLku Rw0uB8mPa1uPSZg24YcJIFzFMe03DwvFcP9N23JhgQ+w9pfPrnSNb9pUcqNxtjZr3NVNj8ZEBEcT pH00HaT2wkSlMKPDmmIimhTlrb0i4n29WilCr+ej6SBr7zEbkxLsu9MvyaT4D/VR2CimJRrtf8Eu TWVSgNup0mmyAzsZIKRZ8U2bgS5mHCJFVRyp0QQbRWovbBTTIIf6S2xEE3HyueqrvUHCRrwaTIlQ z0fTQdbeYzYmN0yC2dAhsLAhzG0lLRBDTtjgA/dpAr4lpRLCbjpyIYpKQ68kUIVYFVEsSg0pctNR BOfmsZUiBPaTHEWK8tbaIHTS69Ng4qKxjKKDrLnHXCQzSB8v88MpnxozFQOxXMLHw8wEBzSSBUxs hwl98lRECGhaMA+laikYEeksQoiIngVqK0kRGSYBWP83DVLj9hFVF1uEbCApEYcGY2miYYyeBWIL Dx+TkIRQkeFhgNYFKBKVEItpZUBzEJDPDTdKEaaZJJXBjZ8roohYkeKWCvNQKSrCaB9NECEETbUX JirFfIonAkcRQXKh6dXcIKEiTg12cbyKTiRr7TkXbM3Q6XOkNSwMF8Kc04ypxQUVW9uFTwmJCr6T 1rCFhAr2kOSoTaUoDm8VQUWRmgsVxYjAe3m12qaNiwoS5bxTY66I+9H6NJCYqNvVc4Ax9ux0DNLs zlCdQMfD4c0yUYwJcdZvzQT9aPKeUXB8+8lzTNTruVDB/pB0t8bUcHDGQtpH00Fir29VpZhu6+x5 FH3Qh0q5qIi5QZQV1JJ2qhgtxwZTRQdodK/eEPpy+Y80l9JogQMNxbQ4oJIas9xakSSg54tzf/Mc XAgJF4y502YKc0fAqYRUmNEEF0Viz88tRghe+dLQSooU5dUcrUeRcBGnBrs4XkUnkrXWN+ThhKW2 SnqUgDQX9DYvisGFV2zhglntVJlcRJS8YO67ceu5cEHxny5XUvycPoMgqdF0kNhrXjBjLSlC8Pun EZ0UKcoo0phPIoOEizhVzHlRx6sILm6je5UXnMSW/5TZXKbYvCjGMonXa+GChd9k/x+kyBRGY3mO ZT0h4YJNQelwhTkv2KMp7aMJLorEXvOimJSqD9FoS4qI8u/VnPPCSLiIU0dKuz5trorg4iY6rD3P i7+/HjNoyEGBmnekGBFoT6LeEb5unyzgM0JECRfveQ4u9Fy4YBYpXa4x5QXnEFJyRtNBYq95USlC oAw3GdGkMKX92BskZMSrYk6Mej6aYOM2vmXj4VQ2AXOTpTODeZdkxmIsk7CH0GxwNCX5z7WZ8e4j zzFLLSRs/OXbWegK5syg3JL20QQbRXjx5rnB6BI4vBk2ook46VPDfuwVkT1+o3g12J/v8Xw1LULm O75XbPCJJ0WANOdjnBu84MGI4U07gLDL5+212Zc4JaXfistZG7mQsKHvvxwhQmfxW0Kj6CAx19SI lELgCiqnhhUpyltrg4QL+eSXSVJkxvi9ihaZ6F5x4UEGnYUuGdZbQmGkgEgwXfqrBvqNyWaHyJJF ePCGXraC6G+xoIEab7toiAxB+yb90eK/+f+x42ckQSRMBXLVz0UH8TJzc+z0b9G2vhyZeLtaxvu1 87i6SrmtHZPMRturT9qoRF8MzzkLa5tswcuvzlxy/GJDRaMF8Q9KQcwUaxiwFBFz7EBSo+kg6prR JOZGShF8sR9cXEaTYry1N0iYiFeD6XeL56tpEbEee4/ZSN2t8eKvIqB041bxsiGMd4COiB3+ZoM3 AyHmoSWkl4LbmPRYkaQDW0FSTkVKhXikoogwF6k5PTcYIVDlqwPWKE+aCJNx6tVekaaFnTpCE8sq govb6P71VkxmsBlDAeAc3Ue5KEYItAPmgvkK/cKELimRQbwNHKSZwXaQ1FORIkwIltRoOkjthYxI KYZv98DMU1iT4ry1N0gyI14NpsyI56tpkYnvX2xMZrAdI/UmV3cNG8KSGToQqHfzk7PIYUNSYoPR YyMXktRgYSMVFZsMkSJ2DlVcNRUJ+/NcpNxecMlzUiOaFOfVngsLI2EjXh0pzaDLK7FhTbBxG9+y 8bDi5Mdmvc7ecavLsFGMMpyboMwGe2iTCXwuhBUM+pP3PMcqiJ4LG3xAQq1tMbHBQQRpH00H0TIK mvTcSBEDn74KG9FEnAzexf7YGyRsxKti7j/i+WqCjdv4lo1/VBasdyoCLaayUpM2tBgxeAmf3ODD 98mEH6TsHXfUNHIhYYNt1amqJOVGgokISY0mIi8Se35uMMXANRhuNqJJcVoTbMReET23Xg2mNyWe r6ZFJr7nbGh3JK2URgK/ushJbCzGC8IBddvlq46aGCfxJKXfirvcErmRsMGJW/W7I0XsnEeQ1Ggq gqba03MjRQwuP9WIRhNxsqpxtTdI2IhXgx3PV9OJZe0961EonDj5EO++SPOwISwxaBCuX4Ev3ycT +PJ9vOObjY1cSNj45TKhcGYp4qTxvWo6CN7NcyOlGPTldrERTYrz1t4gYSNeDSY24vlqKnLie5Ub f7hOPbnxh4DDRjF+UZY0/D6yUa6ZICnlxleeIw+EyDuNia5ChM46WqpOKXIaCIlzfWylFILuxFhF ivLW2iCyRr2NT25XK6X9nzJ3FA2iTIy555nBcrR947IZb/mHTK1QK7epNpUXVLYaLyr/2dqgHkaj tXS3BkIEu0KkaYSIMd8XGj0LaHEfNTw1EN7zQa7QIDUuM/gKAktMYwsaBJj0cWgwpYR9PnomiLX1 5ENLPO2tNHp5+bbUsFCMGU9mX0wDO9FTcbEoE9e83UUxC2lCwFo6X2MuI7UpRoRGkwrLILE3Zasx QmCEGiqiiSjZMhnaY2+QpES8KgZhE83RNMjE9+r1YDeG/GVVhRGYOxIWNYIxBvmrQzEkhbOcqh8Z pwTnQ5ITRpIT7AdJ1yvMJTWbfsNF9KiuvFpzVgxGBBzt16unAZ80fWgfkzSRFrFXpHnhVyEyPDVe Hz2DNLb//t+nU53aL+mhhMeb3syNzmIelbKsDRN084mckat800DNz3lM6B3MRMC2kXS8knLzx/GJ lPmMJbVArrfgYm+KVmP8vgxLlQWjicgZuor7sTdI8iJeFXNexPOjaWNpfK/ygg0a+jW1mkCdmGaz GHMD2icjNthrnfxnMj/eMfOfLDCSzGBPSYpPYc4Mjk9IajTBRpHYa2YEIwZOnISNaCLOb2uCjdgr 0syIV4MpN+L50TTIxPeKDTZr6PdkFp0PUbQTKcZEOtrNBlcRaK6fvlVS+q3gSM8VCRtcOZ/iKlLE zpEKSY2mIm6pbY/nVkox+JYociOaiJOvJF/tDZLciFeDiY14fjQNMvE9Z4MdUyz4xTu9qMqNYIlB r7hyg1MIjV1SYuMjC4WwISRscIpN0z3FFCeHDlLos/dGc5mLaA0jzy1GDGwETW5Ek+K8tTdI2IhX gx3Pj6b7+F6xwfZc/Z7MYX7TWoaNYvSdFM68j5S4+mSy4mThNmwwOZA8MCLvqK71peMjRZycRpDU aDpI7fHcYsTAcnJ6j2giTqZRrvaKqL1Zr46U9oDK3tE0yMT3LzYyhGcjZWokzaXzioeNYsSgb9Qp A93UO06v9JEbfFcuRTb7h3z+ATa4LSvVlTDHyVhXUqOJyIvUXtgIphi4tNr9RzQpzlt7g4iN8Wow 5UY8P5oGERvS9C82MoTXvkkPJjSBTEMZNoSR8VSYXFNju8xvJ3auOlNLooGbn9M/QOQdQ6kr5CIy SBQRpqeplT41p8cGIwQOqqXz9mNE+XFrbRCtsnseHp8GExeJ5Sga5Jh7XHxOZmA/xeuvLngJF8Hg gvrJVOBTXgDftyIq+JRKkkCI01bjNq83kjVSpSjJWEmNogGoBmwtVESICLgEJnkRRUTJ527DfMwN kryIU4OJi8aymhZZe4+5mLzwaAPnGBgOFRmAEJvLCpbTtfavV4TbJeJaFj8KJCd4udX/rhBNpMcx VmNi+JsxRC2FBovAAh/SUMsAZ9ZCfP56u0i3qQFCgvxB0wpp36cyZPUMsLaeU8DWkmxy4DjOcCDM v+KHp6KpZT3TR3iSUYP1FokiIYG9Jel2K0UNcatnANEgRaEhQvL/U7s04UGKnPPvSLnVjjkFPc/V p4GUDPH6KFpkzD0tPLV3UoMInOPrCtm/PBj5rA+96H1kI05qab49mh7l3Y/x2wtpG8FWkvS5xvRi cJoidWcUDTDWsqUmQkTAqkioiCKi5Lt26c5jbpDkRJwqBmHj99G0SKN71WCylUM/gWZf2c6cRiIY XLC9ljULyPjROTqlvW5GdV6wOUIFZZHkBZt50+X6/lSRwRGElJ3VNEjtNTEiRQxc1xo2oumDnRry SnkYe0X83Hg1mNhoNKtpkcb3nA3tpdQwQt6xg89N5mCw8farnSGwwXHx5D/flwwb7DVJGhgJG0in 3KgU5WOkVlMQtxa2l9ckUmLjzZ0TYyFrUpy39gZJbsSrwf58j+dH0yKN7xUbVH+UnbDBBNKwIcwt Bh9xEBsUb9xE6ExgjjdsUA8mD4zIOw2ivMpIhR4pzR3cahpk7IWNSBGDT6YdTcT599beIHpuvBpM bMTzoynIie8VG+wjVaGmGbYhIxD98S9HX/068uX7hP7e94RFlGSBgFDBWku63MjwdGSiBu+D2DPZ ChER4tXnlUr3FD2E6A+F8I7EVoHQEH8C4aA2kTqKVbNI4npFApejqQ9CmJPVai3YdmlIJOgUi7jn lrzG7EUCzfr3VJU+xxUSeDK9rT+2R1/AOoLGJasmyNjyUwvhPUNVbGnAJz0K8MZWAFcv48/KjMvi MmoWcVzYetyTprDSPkoNGkQC02ohIRDec1OGCivqVu5m8GvBdBQAwzM/RYupD8qEhKm0IkJ7G5HV MkhN6aGFcF5fmBEHqCFZoODGUoHkQbxZmY1htSziqPDvMQWpp9hm+VcjBfzq6HwQMeClGhhgL2Ly XB9+ggAO/+Ynz5EMHGfmvvNVafap8CWxOuiNL3YSvhCiZR6KTsnhRwuxUUwcO/k70ceTkRhfq8Ov xTWeV7F7vlahd3pbWysVKM0537lwM8D8QEP15DaT6t058tapbSppHWNQByrIReFFyQBuA2QnsXtr vNxmb6JDjxIFdrXTvxN7XAnkUOP+Kqn3jedV7JQ7/GII8673/Q+k6Dn5jVdUY46MfsvtgYY2eoge VIC80jCr2xwjQymIjF//agkyphq/hICYm/P311YP0emajwyjLGKgr7/dWRF60TstiziqF2+/CjRe O1Og11ev/2DiQMMpccDu5cYsKXnG59ZSGP7qfQ0NbL5Q3wkzkWKcHanVNIgol0wSIVIigg6CX9Wb L1S5EPetvUGSDPFqsOP50TSxHHvPGgLqdDbyh4035u/DhjD/lkxIhQ2myxq7pMQGq2KpDLlEcJKC g1CpLYzpnSC8G01Bjr2wISkw2OBnQbuGU96GCBtXe24kjYSNeCU2IrXRrKZFYD9+lo2Ha+baR6nh ANzRkg8bwciNb921oh6TBe7ETpegsoEhjp8jD4T4N2bkxSY654YxsUHLqmxZTYPUnnNqMaZs/7Y6 jSbi5LOhYT/2BhEb41UxvNpoVtMije95Y8F8hz5BbjY4Cu3WYjBi8Olk2eWmjLwDFP3xDiB5YETe aVGq42xheKfGxFKraRCxIU3JjUiRG1xZkOo0mhRnNGmoZz+D+LnxqlJ4MJ4fTYvE3osuU3spNR7A Oz5xMLkRTGzw8Uf/Cqzsa5cIiS4p5caPnxskbDCkUI+zUvQikVpNQdx/2J6eGyliYGAXNqKJOJmc uNobRM/pRjZ5NZjYaDSraZHG9yo3uCkl8yR88WHYEOb8/vxl1lF2uT0lmfAjKbHBApVkiFxI2GB7 o5aAiylOtmHcaBpE7M9zixGD1pJJMiniJ8IuQn4xY65IUyNODSYy4rhotaIFxtrT+Qgamz8aEDAy 4aL97OEdjMxgYiFcMGpIe/CNFM4xxvFzxC2kXPQtMSIm+HZWapPqGWSsaT5iMX5ETWaz6AZ7BMjs RNrvWBokORF/BhMNioMsqZaJYS09LSmpRz5cycMCmyvbkwhzT8LnkcICm5ny+zN/Hd9wVLUWLAhJ RrCDJL1vpTRnc6spSFif5xYjAk/LaEglRQry1lwRN7lUX3bqSG0wo2iBRvevtyMjDIpy56G2O3AI MJ1qlsc1s+a9LVT7msJV3Cz2hYnZ7ivAnmnkxRzCRYh5PKkmr6qnAD+SbYm+gWjgvMEtWvz6M0Fz MUXABpIP8Wcw5UOiGDXzdyy9aC61R1JjA/mlOMXBYHJed9WpgeCiwLz6LADHM1an8tMbSTqwRyQd bqWoviO1moIQY+2Fh0iR0P4AJu9FFCnGW3ODhIk4NdhxfBVNJMfasyKLmSAOSts3dlYNF8L8Q8J0 uNCN4vqpuVU5VPCVufz4RkIF20bSv1SKUvxO0SA4Z3OhIlJEoK9IZ5Tl7YXUDZhzcxRzhC2kbaV9 GkhMxO3Vs8AYe9pUMl/6q3GDXKM9a1YEIyuYWhITlLC5kA0qkLJvHNpJBgixbxqTeYUwUjxHER2p 1TRI7fmdWgwqdBmJOqVoUpjSrqY59gYRheNVMbzaaFbTIo3vX81ERuHaHqnhg7xjfOkCazCxoVMO sss/mgYahXqA5ueUKkLknQZp3oIYDO98/CDlZjUNUnttajlyIR8wo/MSR5PilFfHXhE/N14NJjYa zWpapPG9YuOLuyHMhvcPpsUIBhvvbBs3G9xqpiFF9yLKu79+bpCwQfuSbsY7FsUGG0tUiqymIH4r bU/PjRRscG4hbETTBzPR2tV47A2S3IhXg4mNRrOaFunOx1ds8Nl1Ll7it+ILIJml0oMeUhADF2Wb Db5Yn+KbG0vi3W9k4EdI2KATzApqpXQc4VbTIGMvbERKbJBC2NOwT5oU5629QcJGvBpMbMTzoynI ie8VG2zrV0kDiXyZpe1GMH5R5qa144zyLbe8Me0kKfUv3HeZPDASNqjVU3BUSkcSLLWaBhl7YSNS xMCprLARTcTJXoWrvSJub8arI8XrkGhW0yKN719spLxQB6zhg7zj+viyEUxscPzTucHMagpOyrp4 x/nTdDFC7J0GT+luK0QRHaFVNMiYCxmREhl/+AimUsOKFOWttUGSGfFpMGVGY4ki/A7izHB0/+Ji 2lC+lKWan8ZG91Cm1SjG2jK7O5wZ9ITJAy5VCBf0g8kCI8kM/l/IqBSFdLSvpkVqL2RUiokolib9 nkQTcTLTdbVXpJkRr44U1WXsraZF1t7jSiOZwUg2VLBsYCoC4L26Sr0hnGJKBvD6xq+kiP8MCZTv OctdEcpHlK6O/Km2HQUJX/8fj/WhVCWCFIAQOyKwz7h8/2znETcqwkPxc3XcxLEJ8HDOhokdTQwo AbhnJPuVF2M9WV8BVFK9aUpB3Qa9qs45MGnk54rYL43ImA8fKZ7zKQNpX02LxJ5/yMWoZ7VXRJ1E NBGhv0x47A0i5sarYni1nq+mRRrfsvHwnAOT2tqjIjb4zlnOOSzGijLfe7dddgu7L+Qbs+pHNALy OIPVGyFyTne9eWEwGM752IGUr6JFYq7pwOEESTHC+W29Ek2EybRFyI+9In6OSlJODSQuGssqWmTM PS02WQjRbYTmQgtaahoW0+q6N2CiKte5ESZSzgxORSsLgpQM9oCkfJAUmnxeQtpX0yKolKa8TZy0 kBQxsMRkWqNJcUaTRoKWKZLMiFeVcmY0mtW0scQefj5uGtJQsgisHadmg0/ruthcjBjedA80GcmX 7/nxFDtSIBqieVtikaTGx7dXCIPhnU8dSPtqWgTvpClscDZBUsTwp8VUNClOeXXsFWlqxKvB9Ds2 mtW0SOzB/nM2GHSl2GSbxuRGMVbXsxhJI8f+Wc9f0K6HjV8/Bz9GwgYXkqtzKSY2OHcg7aspiDs1 2wsbleLt4/y9cyOaiNN3eoqN2BtEz41Xg4mNer6aFtH4DM9fsfHDsme8+2QUkTdFWGLQIofsclxf xSZxSipDIg86ioQNZixUbK6UTxvcaFoE76QpbHBIQVLEwEVYZiOaFOetvUHCRrwaTGzU89UU5MT3 ig16yxSbbJ7MgQf3fsJIM1YryFMK2Nz7xkqPpDJQ86CjSNigD0xTWykdN7D21bQI3YE0hY1K0U39 unhzb5r2kdNnV3tELsTtzXhVDK/W89W0yNp7/Ka4pPihDOS+NSJHmski5cbBiIHv9GGFos0Xt/Gr 86HVeMeRNpUaQeydhmppXCpE0Rjlq2iRmgsZlSIEeiHIryJFeWttkGRGfBrs4ncUbSTODEf3NDOI +517QeCCqTDtTg4XxWiOaeKdGXyxJu0BrWS4+PRzcGEkmcFGAxWbxdRqcJJd2lfTIrUXMipFVcUX H/yeRBNx/rm1V6SZEa9W6kQzmg6y9p5nhk92iAx9rSNkBCICffhD2feW6+CIUkIZpDVsALumUZqX F1dIpw1SdVXPALUVIiKD97SjaI4afkt4uLE1gJ4ahwZTUsTn0bNAbOHhYxLcrSLMzgy5SmnN9x3K QjHqlXfvlKSeZdrWjSKfm1VjyhKDn+PnNyLfNBvrhcViSglOHoSIagpCkLUXJirFugcZcNFElO+3 9gYJF/FqMHFRz0fTQdbeczbYqaGZTtYOvtjPk5wQ5rTmGiLt+KCkTXnFoDlcsCMq7YKRcME+EU10 soASKUrqOz2L1JqeW4wZ/+922dFElG+39gYJF/JpEDFRr0fPQWLtZV74iImZ0JniMFFMNb23/pAg utbOUUpKecE1NGVHSLjgHUrxycdYJeUzBvJ3NQUJ8/PcShGDzicrw6JJcd7aGyRcxKvBxEY9H00H 0SqR/XyeF+zh0AIkac/PWjaEOYZ3/bDKC75onzzgO6uKk07AMmSBkbDBxYi6kryY2GAv+o2mRWpP zy3Ga/bbJZho+uClurU3SNiIV4OJjXo+moqc+F61GYwSVR7inT75mdwoRlPx+aW91JRw7x2GMPUZ NphZSR4YCRtsSqBsEhuR8imDFJ/VtEjthQ1OJ0iKGLiF2rkRTYrz1l4Rt9Xj1UoRez0fTQdZe49z YwoM9s6o8GOijHtyy0YxmusfUkJssDMnxSdfFwsb7KdJ5ELak7BQlK63Uowron01LVJ7YaNS9CH6 TpnelGgiTrZOX+0NktyIV4MpN+r5aDrI2nvMRvoTbZHUYEJsfLHkodwIRnap89JZSNjg5sVkArPa KcC4d7GRC0lu/NWOLuVGpXzioMv20RTk2AsbHFSQD8Tw7QKGQaA1Kc6rPXwqEjbi1WDj+UVTY7na KxuPJneQ5uynik/YYPtv2SgGG1SlZoNtiM0EpFyCEV0j13Nhg423KcWZqBJnlNTRvpoWiT3n1GKw 8eED3hqUSZPijCYNyexnkbARryr1H7FRz0fTQRrf83ZDuyk1mIANrjH60RAeDcU0UajtQuQbX7ZP JnAcMLnBgCqRG5F3TLe2OhcEZZTUUT6KDhJzJnGltDG/9ZwVESWjrmPNmWEEa+vTSh2/R9FB1tzz zPBBWlGhK/NCRSACYCrYVoktWcCHTeNb7gmikBcQIui+UpBWyB7B8uiBCHdLY0tPDcR/uTAP+rQe 5o2L8HCx5SQxEBrikGiwEEHH5+jx7wngjsRxvUoIZshTfH7okvawUIwhGRUpaUfP8OYFDzLDV7l7 tNaxh5HwwE6TdL6Voj2P9tGkAvXGnp5bjNTw9zWgIpqI8vvWXhG9VuvVSsFFPR9NB0GlNP2LjXQk bCDhFia/HtySOWwUIwZd/SW7n+ycUpvIoCM5wb73Bg7id56s4rDTVUqFj5WPIgIvot0FaAoZxQiB 8zjOi2hSmLf2BgkZdmogvdb1O4qcGEKcGTX3vBthN4cKRP0YVBHJjGKEwDycM4N2vmRISh0LVwg2 cCEKSj1R9y2y6igpRgfRPpoOUnsho1LE8NFCP5oU5629Is2MeLVSsFHPR9NB1t5jNiYz2NGhApG3 lGs1ykYxLU7z5iozODeo/GGCRFJi4z3PBWlq8MOmvKoURXWkRhNsFKm9sFGM/oRDlE6NamIC6M5e kaRGvRopYq/no+kga+8xG1NisKlDEWhpmuIquSGM7GICiWUxkcGCWUYYzP+GDK6hSuBGkhrciKja jYIkUsysRMqKiHKBWgsXFaL64CIh51gVscJ8Zy5IM6NOjRSRN5bRdJC195iLZAY7KtkSZC7e2Gpm LhZjwlO7peCCowoZX/DJ2XDBcCRxC2lisDNHVROlaqSYf4xUFcFFAXpfiYSLYtSl3+6JNQS0IqZJ 78wVSV7UqZH6Wb9H00HW3mMukhfaT8lUo8o+7hXxESk0CHNe8JFEc8FGvmQBn1jlteEl+cpjDA2E KChN3mmdES4qRQUZKUolinXlRYFaCxfFiIAjeldFTKDfmQvSvKhTI1W/cWA0HWTtPeZi8oKJRQWA NJOPzYtihKDVflKNIyDJApa1wwV3BzRuEOeF1rW8MRIuIoXOSFURXBSotXBRjNEbW2VhbBUR5Z25 IsqLdWqkiLx+j6aDrL3HXExesClKZSfS+tae2gvttGTru9oLpiTUYFCI8jn3/OSSUmJwPFfPEbiQ JIauEV3IdWiEosjNIidoL+b0GPkbjBD02T6REUVEeWetCPtij08jVb9xYBQdpNE9rzHYWkkPKCoY EA0VgbR0Rmyyyp00SQK+NBsmfJEiRAhoVrCbRKVbMMfo4T6MRg9BB6it8BCIuQxm0E2D1PA7sNx2 a6tAUqIOjRBBx+fRs0Bs4eHzhHAiiwXa42GhGD+MdqiKBt4SpQjTOpJSQvhdWiQJQXan060UfZiy W6kVTRAhBE21FyYqxWTvD4sKyohqYh32zl6RcFGvRorQ6/loOsjae84GuzhUmNG6tzdlByaQQ6Dg DxlfBpjwU2dKl8Iey4QtIEywiyQ9roYASnrOX1y1HCSmwkOF8J7PxZuH6mFx/mILBwOEhbgzIuvw aIGDhjCmHlMwrSVR5rVQbZQGohCtFodVnA8MkZPxvoAGDthw2IDZvxUOftin4bcil9RQiEdm1MBB ERzDVjgoxIQnFyq6qaweAkRIfNdWgHBQf0ZGG0PtT9TwVJD8tLb1mIQ0k+yu1Bq68yBbW38WUtOt fSG8FHytVslCIhjQ4MxPBQgJlOfpYvPdHabRIjNqIKEIJLBBNiQUwvt8F8nFufQQ4K2tACFB/lxk 4jIOjpoThG1B+HMSfA+dSODzUc4E7ankrjq9zwwsQwKbkPLDc6eqf51P3V4nVgSEBK4hSd9qGULm dIJkosbvR5HYCgmF8J7VAr8N1cMenVtbAUKC/MHDkYnL2Bo1kHAT1ysS/nx8azkdx/ht9TqwvRIo 3nsfCLnFV++T6hRDJoHViUYMoHA0GtNKonjpaJIDFVc10FIktkJCIa2FdMGlegjwYisRx9bxZ2TW 5VFzghhbjzMhbYK2ULKpwSRQF4SEQvyE+pagXjD24feHR0jvKcOPRszesZDwzr3CIUEyhMy5BMmM moPgGCIhoUJ4z9iadGEwFz0EeGsrAE8df0Zmo4gaRILkx7WtxyRMm/Cp74eZhFJQgCC0E08mOUmr ZOE3NgE/eUJ/JnyuWUjJxP9XZZ3/PwoOQvB9dSSCRnz2xw0IvToI62pDfyZs+YCb+f9s+IyNKKCQ WiQ2nofMHq+Ub5zq9MkwHi+mn47NIYqaD9Dnd2Y7t7oHBll5Tr8TSGLv4mdkCJXr/aU7ekyHEP8a tqanVoqFDF18rZ/eeti5eGerSDiQR2haqfV69Jw4GODHx+dMcHy2vjEc6u8vzBEwNZnfn5v9kt58 wylMcPFzfnEjYYL9ICqBYMdSRMl6yFXTQbTKh6ZwUSli4Aoxc1FNxHlnr0jYqFcrxVGq2BtNsHEb 36tGcdKCGzaGjPk1OZ2ot59SjN3EjpI9zCka3v2D0xIKCBUUgymdIkTgXCYbTq1nAeapeKqVRWSY oOcuK/NQNYRxa6tAaJBDfjEk5DchPo8eWLiJ6xUJzBmpBsIzRZpWsRg9nDd0QIOKe/3WnBINC0xb pXQyEhqYRjpSLnqYRpLUKIKHIjHXjChGBFxQYyaqicDlVMZN1hREz61ThdQm1O9RdJAx92x75o+2 S7ryZyJVH/QSF4sRwrcuw8AsqxIKitl4pOwc10+lLBbSnGAZXZ1ppAiK6adIjaaDxF6zolJMyLtc 0MKaFfHyS7lHc1FUJFzUqWA4tY5X0QEa3fO80GZJ1/5I65xjuChGBLBtLvhkUSPXmcmMsfwccQtJ YrC6nUJJWKrhaB9NcMEJBCVP7DUxihGCF/00XLMioryacxtpJFzUqZXaYKoILm6iw1qbzYdrfozV 2Ywa31hwa14Uo/D9ZXFL2cd+bkVA5Eg5L77yXBDFRKfC8nHqJUk5LzhvIO2jCS6KwIU0icPFCMG9 9CoiSiuikq+5IjK3TgVzXtTxUXQf3XLxaDs3zH2w3moudHSjeVGMCPiWlM3y5flkvw9TKC+4/zU5 YCRccEAsZZMw5wVLeJIaTcRdhFUdnQIJF8WYtJ9ltSjqwZAx57zwsZBwUadWaoMZRQusted5wQc4 9VsiTVfdvChGBKyDmgu+a97IkXJe/M1z/Kh6LlzwKU11usVcHUZqNKlevNqbWjQYIWhLBglWRUQp 5eKi5oqEizoVzHlRx0eRNoFeo3uVF9/6+o644KD75EUx1mu884l0ZONusp/16jgHOckBI+GCG6XU x9EoSJe4YKQlqdF0kNhzXqwUrqeqrB4q5Ks1p4WRUFGfVgoGYq169u+19bjIyjCD94qeNExwLUez opjmCiBALyYnPRu3pNSvvOU5ogbRr8uP+eU6MZB7Ri5el/JRpEryxhyPHYwAdLph9RDinbEiIaIu rdSGYj34EwD+YEKKXuUEmz/1O7KWrNsT0ocUYy6JjeT4puLGIxI6SEmJCbYJ5fc3EibYAZJeN1IE yYYTSY2mg9ReqKgUCzS6ikFcVBPDyTt7Qcz8erVSqq9sbzUtsvae54W7cRYMude7XBjBf4574hmV 7ZfLRqokyYgJXx7Yv8MDM3CarhkZouY8kVmwlvm7dsKBJdS4uXjTSpqXFZlDwA6/Qe3kb7dH60sx /fb2drX077VD7P/5P//9fw97DYZtP94+izR3hzj6xZhM9cYN4tdcgENDyG7xRcb/T9i5o8mx9EhW /1fRSyCLLD7U3sGsYZRWWpj9K3MOzBwRmV8yS7osuwg8LBH+hHvkR+ej6zON5uejKiTd7GBmO2cN MsqMoguJuTaUlSIAbnmeRKgmQlS7NNReEQk/TgXCqfV7FS3S6N6/Fazg/JnZApullMVMD7qYC8t0 gJqd4hnJ4JOIcY5ygYy1Bkk2sKAaysSmpaTERamjCTKKxF5bymJ0/tbY+FZUEyOG0QQZtRekWVGv islGo1lNizS+r9hgUc2fE2kvhraNsC5yMPYbpoCD3ONTScl6hr5hg/vJkgaDyAYbTeiZBIoUsXNw QqmjqQjvT+3l/agUMbgNdtNEnE/2imhvvVqp9Xw1LbL2XrcRWZ1hD4hK0PGOmxcPG2KJwasgtcvH lRM79Qxhg+/+JA8GCRvsQ+UITaUYMj5pWoTNajX53GIOkttHVxO3YD3ZKxI26tVKJRo8X03P8X2V G9wS6u+Jd98YNCc3ivGL0gXxPvIa+lkF3xRGPrajztA63xgkbHD8Pv1tpeboRMaa1bRI7YUN3FeK GDi3hHaP+Y8m4nyyF2T6j/Vqpdj3fdK0yNp7nRsZV/gTzcyBxzi8UTaKuR05wz86J7++JxucZwsb bFNnzCUy3rkTNhuCKzVnDpRaTYvUXtjgZIJSpD7lj7C/miD9yV6R5Ea9WqmNZjUtsvZes3HelN/e b2VucArxsCE2oxMqAhwMM2759JOLsMEZ2/SxDECTB4MkN+huM96oFIPoJ01BLnthQykwGmtXeWWj mlgwudvDpyJho16tFI3Dk6ZFGt++KS9nqHbCM3tw+5QCgLwpxdxBpXGQDT4tkNjZisjYi2PlmXWI zPvvTC3dzEA2CBQvZkBbRYvUnDvbi9FhMsIZMqKINTR126fVWpFwUZ9WKrHA2Cp6jm65+MdYg1oR f013T/nA3v9On12M/VMn23LBJyoykOYDRulxqefIeHKQZAblHxlaiE2YFPYqtZo8iXC3l8woxjCK QcaQUU1sdEcTA8s8F2S4X6+K2RfW89W0SOLjfX73njA1ZUlNMtinz0bogdw986gcXHALdCKnKsAo naclbIGmBVuRGVgM5i/+e4RWzwFqK0QgM7/lL26mGB6qhuGFqi9bBZIUdWiFjs8My9GDhwdYW+9J oBKkLPiRQF8P1lbAprGYT/vNwG7m6Z4rQkjPPvNYEV1zltYyxkrNaQOVwwMFnJMjnD+4mQsRxYiA S4XumgjyyV6QaakZW45TKxS/MbeKFoELFf0rIdKNuAI97zeaLaAKF8XcrbJk2mGdm5x2B2zY+cI4 IZrngoxzTtRaxVipOWxglqymRWovZHAkQSkYdqtJWquJIfiTvSI+t16t1EazmhZZe+8zg+1Mh1K8 Ayz5lw2xyQyObriQwJvmOTHZYIc+bHCaLAPLQZIanO/KgKJSc0ziQVOQy57PLcZb55WLslFN/ONu D5+KhI16tVJWeY5Xq2mRxre58bIb4Uez+HHYcNE5uVEMVS47ywbXE5QNpKa54OtOGaIyHpmdYmKg LCoDisGMk8MGSq2mRfBOTWGjUsTA11eHjWoiTrWbi7VXJGzUq2DTRNTz1fQc37LxsiPhffZYs2zQ hs/d5LQaxUiz2cmGDWZeyQTONMQ77ppMHgyid7y+Hpo2g8Twbo5JKLWaFom9trrMz5RiTYbNjbsm Xplocqo4MkHmufWqmO1dPV9NizS+r9jg2FOGfnw67ORGMWLgw7F4x9htboMzTqXsXPjmQPJgkLDx 6xvzrGEjUnPqQKnVFISMxztlkhvUISlFDFyHNmxUE3E+2SuS3KhXK0Xv/aRpkbX3vt3gfgSHmvbY /Px5U8QmBogIG1y4m0zgYo10taz6JQ8GCRvc0e4eI2t6kfLcwaOmRWovbFTKkQQJYJZVk4W3j/aK hI16tVIUGsbeagpyxfc+N5gz+PnYYYMDjMPGYnRPfLGJjGcYNxfCGadS5gang5MHg4SNLO+skAcP RiiKeCxInFNRyKgUIdANDxlRRJRP1oqEC31C1Uqt36toEbiPuXeZYY3kFBki/Z3hlZkRLJkxX+CA C77R1TxQSi64zC2DMF6lBMXckaPYyYxIee5gpCAjmhapvZBRKWKA9yGjmojzyV6RsFGvVmo9X03P 8X2VGZ4jMDH4itohIxAjuNn3pmWikMDqEBZCFZKLz3mqgK6xxEJhPppWyEMHAqvnALHVziQyzDxY 0r+roUz/0VYBbcUhVAcjdVlMG1ur5wCN6ysS2IhxMIU0n0udFT/CCIb7s+mNVT7dlARgfzk0/IoM wiLhgfKSdDdi05Fw4CDjz2qaIwiXvb4flSKEuW8TRqsJ8WjCvWgKMs+tV8Xkop6vpkUa31dsUMiR nKDGLgccbPkG472YDUTS20vg+LWpcrPxsN3r9EOkXFBKkmHHYDaAHJ5Q99GzAK/XPIaqxXi9KHWc rKgiXialbuaKJC3i0wrF62k8qmjjOObezs0YTVJxN3lBzTRMTKYVo7WcFpwOca6AgwpuDoxvFC0m BQZJWlBMkp6mUgyDkMK5oyjAZc3HFmNdJtuHjJWiiEb2yVyQea/WqZXaWFbTIrTyavpXVmRawqyC 3dl4x2pjm81i9hyOBBm/5SI3m2RLrGlHuSQlKSAyzjltm03GlWIsGqmjaAF7IBSFiwoRAf/it1hF djiP5ookLerUSm0sq2mRtfe+C/nDkuZwwcrm4UJsfs6MOxm9zc1wtH9/FZILujEfK5K8oE3JmKRS jEQjxQrTT9+sBVhWVlG4qBAjCg4JopzpXxQxEnkyF6R5UadWymrQcWo1LbL2XnNx8oKjsxl2zhw+ 3WkxQphWnbHbXAxn5M685YLTfBlPijQvOOLqaGOl5tCBUkfRAvimonDByQSFiICp/+RFFRHlk7ki yYs6tVLUCD5pWmTtveYia55WWs4kgrVob4iRi2CTF3zKwZ0clvvmYjjC5J6tkMEGa5JgkCQGH73J OlalWIBD6q4pCJHXXthQatjga0TJjGpiKfhuD01Fwka9WqmNBjPRtMjaKxsvJ6ss6XJ54Hj3jeF9 2SjGPORzDgHxW8/FcLDBmNwjpk6N5rkg8xs7hcrMVCFI5LeO0CpaxAVrzw7Nq1Mp8uaXh4B4TaKI 2UsUrbUi4aI+rVRigbFV9Bzdv1rPkxl8JMZfE/uZmlmMOZMKHHBT2X6bnaxkAZ2mPHCPVzKAv/UL 2v0anC/ISNB8RSI6JuAiYyf5UIRmxOXemxbmsw925m+fiSeTHyMRXyenqmOR2nn/VvDtR5SxKMDW RN4JvwbpW01dUkKnPjNvvtcyGDsH2CZQ/07sVJhkOAk0LzmfdbgrcZNPIHZ85iCEzvxnQq8SFjYu OxOofyf2unJE2N8dtavkAMfO69jTOlpOObWIroikMG0hl2mmntfRY+cMOfdCL5inXA+Z8jJ+eqpI ysi0Ig7uRvOqWSS2QkCF6E+9xsAfv3pc2pH6tRUgFNSfI7MuRw1PBcmPipqvkp8CEn86lshd75/n C9EPcCwOhJEZwwNTm297jV+UIub3FtAvZ2CzkXhkmPNEZrUsMqbaq1QI57/NauHqYWHv0VaAcIA7 uHxEqH3BFP6tliAb1VcUzMx0KCDcUCCERpz/nMM2LMnM/W5yMJwwz3E+O6QAhINPZ88HMsepMs57 oJp5MYrAQZ9aIcaT7gnIZfUQ4KOtAOFAf/DwyFgDOrZWzSLH1uuX4TSCc8PBkMCEKCQIEQ4kMBye PGA8098dIbvL7xGBFYCQ8IsProcEP9QKCRw/eFCzCI71qYXw/seUJTqfGj0EeLOViPsUo+T4c2So /4utVbPIsfWahNMicF2uPyADtu0KCjFAtLDfl4HOLT/8XDkLCdzNm5/dzyqHBO7KTXcRGUbDkVk1 i8SWTy2E9x4PMBOqxwoT+4O1FSCZUH+OjFWf40/U8FQQqDu2XpOQTKBM3Sq7Eea7ZpMKi+E+eYdn DES9Y8tc5465DJ0INT/8IOGBT4el76jUeKDUalqEf6gpTMC/Uqw5UpxBCG6zjSaifLIXZHhfr1Zq PV9Ni6y912wkJSyLnDkA0j8YFPliLEYMXLM+BFPI3xRQamZbeQ5+QMY75lserh7OIkVvEKnVtEjt hY1KEQO3qsL+aiLOJ3tFfI4hYbxaqfV8NS2y9l6zcXLjB5srRECqut4SNsSmrZzb+bXL5+aTCcwl w8avPEf+iugdIc/CeyHT/2+EiI4L9afpKFJzIaMYKxG2Hpci9hKfrBVxtLk+rRS1nzG3ihZZc6+5 OJnBB8EzbORg9OGimD8QI33tclI0LcGHUmbG7zxH4CDNDAb/GTxWioFypFbTIninJslYjBj4JNNk RjUR55O9IsmMerVSFoRONKtpkbX3mo2TGcxsJzH4WNghI5C7Z3/9Oakx4ot3k/+M48LFnw4aBkha UBeSPqZCjKxHKHomaoBpzcZWiIgM3jMSMyeqhW31R1MFwkL9CYaHlIOOqaPm/E33pJb34whLKWcG YGfG6aK8HMVw/nPqcxmh/urM4XvuZmGWdWoBBgkLlI2kkxHDNSa5kVpNi8ReE6JSLmNmsF5FxHg3 B31FwkSdWqkN5ihaYK01HV7OL62bnBkB0l71EC6KEYF3S5gQ3qFke0hLbz44uZlSxSKhgjlFeppK MU6O8lW0SM35XDDiJIJphJyzjSKCvJvDgSKhIj6t0IYyeiZJHmP7KisoxzBItw6pBAoTYrjGdpnj BJngeFIaAM71ZZjxkecYaIqEClIg3UylGCo+aQpy2QsVSoERQnehq4h67idzQaZVWqdWypLQB0UL NLp/cTFN5h+rI9tK8Lmz4eLC3Dqc0ieGbF59YFrM4VjTgpv6kgJzFjZcsPRSLtTlD842hVKraRG8 UyZcVIodjynXcZI1iogyitZcER9bp47U5fgo+q+PAtMu1dqbBhNp0k7/8c17AciLCyMCyg2nyWQR Ia/It9wwwFwsz8GOSLhglTsdbqUYOEZqNS1Se+GiUoQwB7jdWRtFRPlkrkj60jp1pC7Hq+gC1tpr Lk5e8AvoP9IMWcpFMcfAs4DNwA1ekxdKmRd8Pqhxg0zSOi+bzUH+ESkoiNRqWqT2wkWlCIGqK/vp KiLKJ3NFfGydOlKEXsdH0fyqAskLFf3rHZmOlMe5JcoBA9NldpTKRTEioKxj2iDKi80efjGl5OJX niuic07PpvhwpWgxI7WaFqm9cFEpdmsY5N8UUa/3ZK6IecERqzh1pK5gqugC1tr7vKDKLcNNyr4O F8WI4IMff4bjfoNxuFBKLn7nOfOZ6Uq4oKDb4VowpBhZRWo1LeJSBZp8bjFCmFfSKdgoIsonc0V8 jHF1nDpShF7HR9HkhcDkRa295uLkxeffmTzg2zcmHWkvitHQ/f1IXvC5abOHKJWSC7qYxi0SLuiK 7kKMxSMURRO2SJw7j60UEXBsHSm3slREkE/WioQKfULVkYKK+n0UXUij++od8TtQviJzA2uoAMIT AnBV0h+Acidzh/adijIQp0N5ZQYIEVSBuIqyQgwcH/UcoLZ86kDsWXlxrDSoZnjw4thpt2OLH13j oaEOFbO5j89HzwJr63/+89//9Y9jQQhTfaFRpL2LLSwUYxRA5g8NfG8uP7+XsY1r3CGfX3+uZwsP FIDY5cKDuoyRgxNKraZFYq/zlkoRgmakopqIMpqceTVJBgkX9SpS/5GLen40XUjj+yoliNcIrHVg thE2ihEDB16GDb+YZJi/EQJwejaPFQkZvOj2ucGgjJFjpFbRIjHXlqJS7Flx/eGQUU026p0C1V6Q JkacKiQX9fsoupBj7t1R/D+shv2cyQPSLLtMZixGCAQ0XPC1++QBn8c587N5jsB9LmQw1C5lYpJB SaL0rKZFYq+ZUSli8OSfmVFNxBlNzgdHU5FkRr0KRmas50fThTS+95nhQvkM+5FmdDSZsRjLzX8Z /Nlc8PmXvAGU+Sc1fuY52BAJG39/z6JFsEkNDh9k0FlNZs3NXlOjGDGwZDRsVBNx6pWpWHtBmhr1 qphsNJqj6UIa31dsEKutIPUbnmfwPWFvKRhsTKUP+cY36rP+PScM9M7K+3kpBpENZ2tTUUgCqMvc 4NhEmuFoKkILW3s+t1JsZzK2u2ti8BRNa68Iz11eHanL86PpQtbeu07VOkk+H6d37MMeNsQmv9n5 JTewy/dH8g7weePkBvcWZMgxSNjgvoFsAVeKPvhRE5EXqb2wUQxnaHyHjWrC8JO9ImGjXh0pYq/n R1ORK76vcuPzG63isMHYt7lRjNUG7l6blpoxejLhUylzgxFzBh2DhI0fnviyEa0Ug9FIHU20CEVg Q00+txgxcJnUsFFNxPlkL4jtDb9RvTpSxF7Pj6YLWXuvcyOTEjoib7WSDRbYDxvFiIE9p7HLyehk AmtbYYOhWCIXmdbQSdvsItKJRopRZKSOJiIvUntho5jlzR7m552rJl/RR3tFwka9OlLEXs+PpgtZ e6/ZyPCTkdoPXtVhwyli2g0xsosYphaMX+GT1mx+dcoT0+OyMJnxxyBG5bQtTW2FXCMfoSgizEVw TkUho1KEwPfUboqI8slaEWYll09Hisgby1F0IWvuNRcnM/7MT46wFzKEikCsVf51xoRVLltpFihk D7OJAtC0YMEhXfBco2DcZ6oaPQvUVniIDN5/x4R8qobfgQgfbRVIStShI0TQ8fnoWSC28PA1CSch qMPQd8+e5UQlCopBg32gNHDxjynCmFkpaeA+lsQ8SBKCUpBkRKXmsIFSRxNEUDllkLUXJsSQolfk 0w1DRTR90J8/2jtIuIhXB7MLrOdH04WsvXdssGD8h6sqxzuOgE9OBJsYqDYOGyw5ATiBmNVZdhx4 7EJCBkUh6XsrNWcNlGK5fxRBBucRRHBOTSGjGCHwDZghI5oM89HeQULGOHUgysUnFpQfRRcSc19k BntLfH9nnON80tQUoKEYIXAIbTKDXYFMvVi6yxCMRfaMuwcxKOdss83IHnSk2D6J1NEEGSLD/dgL GZVi9JjrAY8m4uQW77u9g4SMeHUw2ajnR9OFUG+kpvfvCTUJrGTGOxa3mxliMMSW1+zrkm18by+J 8KGU7wm3AWXQMUjY+M7NYeiCjUixd/CoCTaK4J0yYaMYMXBpxfAaTcb5aO8gsnG8Ophs1POjqcgV 31dsULrhK0DFDvdTlY1i1ivxMfgZkNL8IcWKg1Kywa3AeXUGCRuUCqTrrRQD0kgdTQ5RH+z53GLM szh2PWxEE3GycHS3V8TWGjYo6sGrlSL2en40XUjj+xcb6Uj4iTz2PWx4caY9yWLEMJV92OWb9cmE uV5TNua4+PDDc9OXeM/0bDTCmbqMkzMCaj+aLgTvlAkblSIGVjFmIhJNxhlNx95Bwka8OtjN86Pp imXtvW9D50uhw4bbHGFDjOwiBnce/BX4emgzIRVCTt04JjxsiBiVk6nZaoSNSM2hg7sm2OBcQoa7 TIQPG8WIgRNiN03Gebf3n8mDPrderRSxN5qj6UJgI8+VjVc7aEj/ZpdqcoPDLoeNYlQJeFugbLAP n3eAz5BmeM51LBl4iJQNLizIIGwwc4OOVqmj6UJ4xdWU3KgUqzOfHWxGE3Fy5p4so4ePvYMkN+JV Md7ojeZoupDYw8/3ucFnu8oGs4CZxDOKD0YMNGvDBleaJRO+ITXefY8My7giyQ0+zDDpIpQx9aMi yODTIZe5eWyxWarKSH8UGeXd2mTGIOEiPl1S6/dRBBcP0V1cvMyMmednKM4PORWrfxaz1ssjUWQG V+UZAgv8SHXiNpOQIOWC0pKMs5SaVoMNArUfTZBRhHoiNSUzijHAo+pi3pNoIk42UNPDMxdA00HC Rrwqhlfr+dF0IbF3sfHqoCnSfnvRRoMDvXPOdCEi4Ns6wwWXjRgTkSM0eeGRywWSFhSWXNjkBYfQ wsTogYkAsdWsCIT3rP4ND1FjhDdbkxQDhIY4dAmdMI6eDWJtvX490pG4XzETCUYrftvEpnMx65w4 cuxLyD3NEzTnfMICVYn59UXml2VD6+fsOLJJNlJEzSkEpY6iC6m5ZESlmPkwwIK+o4kg+fjm3d5B pKJOHejmdxRtJDPEmOjeNxXs6WEOd3GOT3WUi2KEQJ3pNEiWUoUMpTJjm+cIXMSgnLPNlmMxG06O ISQpoulCai9kVIoYmKTdNBnno70ikg8Z8eqS2miOJn7Xx/j+xcbJDCo7bO3gmRNhZaMYlT2OOrVL fWpynwUnXxnyMM+xOQbS1OD4qF1vsEl2ClTVfjQ5YH+wJxuLsTzg+FVeo4k4GU/e7R0kbMSrg5kb 9fxoupDG9y82MmnFTdumKYTgLrWyITZDjJ++y7JBObFRsZmklGz8yXNFjMq522w7Hini5ADFXVOR y17YUAqMvY8/DDsuTcSJqcvetBmDhI14dUltNEcTbDzGt2z8oyNh19Df08oWCnHSahSzksYjCbDB V+z1lziRmo6EmpHkgUjfFHYA7XojNb86hyiUOppgo0jsTU4tRgyf820H2s3RZJxql/3YO0jYiFfF piOp50cTbDzGt2z8oyOBe5tHvOM+jRliOLIYjBi+cYWybHB9Q2NHCsRJ0sgQuUhyg/NE6XzFyPg5 mJDBeDRdSOzNcytFDH7P1dyIJuMcTWuviM+tVwfzTannR9OFNL6v2OA+c39PVok8M5vcKEYTNKWX 2KVxcP2cvSClZIOvcSYPBgkb3BCV7jdSxE5DoNTRVARNtedzK8XiLtW/w0Y0Eeecv73sHSRsxKuD yUY9P5ouZO297l/TbrDWQMTxjm/IDRvBEoPfivRX+IMpENhQSu+YwiUPBgkbfxinXEKEyafZFIKM UXQhOKeikFEpQvhwDENqoGiGVxw6zbQ51oy7j61PB6N2YWLB3FF0ITHHe/meC1Yd/TUZYf6h8Tcz LNEcjEEHa6vDBZcKyxh7SkrJBQuiyYJBwgWFJumFI0Xo1EQodTQVGe7HXsioFDtBf/mYoWREE3F+ e7R3EJ6DjXh1sJvnR9MVS+P7ig2qdOIcq1clAygRuMFq9nHorZErJBd+L3TIEdA151FT+ugJGIUI /M+DngVwrE8thPdu0x41kxVsAjvSqS1DPraOQweThvi8eg6wtt6VIlhqObMIpKnIKAvF+CG5tXto cO1pgkZoPPuVxwjZx0IDdSelRkwaKL3N8DOKLiTm2o9Uigi80VAmoskgRxNMxN5BtMe4W6cK4dTG sooWaXRfJQRlHg6ltMZQM69HMa15qhSzfMU+iT8rwP5Mv/Oc5hiLhQyKUdL1imVMHamjicTnuMZl b9qKxeSgg9to+pDNGQAfe0XG3vHqYLJRz1fTIo3vKzYo9PDnRJrlyGZGMWJwhw8yPAI4bcUPhSTj bx6DDJGQwafp0vNGitA5m5E3LVuFC8TaGbRGiAg+PUNDYkSRUT6aO0gSY84lct/DSE1m1O/VZI9/ j+4rLqjz8NfEO+YxzYxicNG2glNcyX/O0cuFE6TOQQYJF5xSzlRObBKD0wwqryLIKRBrzYtiRKC0 XESRUd7NMfQsEi7i1MHMi/q9mhZZe+1EXg49rcCcqQStE1/ybV4Uo3maBQTGb/hLTDRGCsnFdx6b TBkkXJBZaVciReicZcjIcxRdQK3xWDFU0fBymnG4iCKipKzhMkfrUCRcxKmDyYV+3zU9R/dVXnA0 IAPPT1arvMkCncXgggIJXhHuyUryc8dGfKOZSwoMEiqo8c9wI1JEyYGG1J5OwV7+nhwcW2FCGTCY YII7TESPMT5aK+KbtT5dUuv1aloE5tX0LyYyWbXWciYRjBWs8LftXMyCDp1jZOt9+GYFBRChgvvo kwEi4xwzNhYFb1JQwTd+lKqiC6i1cBEh+gOqMjLZiSKinLJ/kzDmDhIu4tTBLr8vTRvJ2ns3zGLh 2sO3thZTYxcuxMg0QrDaDi7gpJFbZKdz3ACYFJiyO51z/pTcGcjcZz1GodEzrUeBGgsVwQyAzurS Y4yPxg7idvtx6WAykUguRQc5sX2VFRxx86dkNO019mGiGGPuKS/jjeETEAn8m1JSwdZzUkBk0sLJ WssaIwUX1LkqdTRdSO2FjEgRA8xlsB1NxDlX4l/2DiL1x6uDyUY8vzQd5MT3LzYyFbEMM53IDy6x KhmBiIDismk9GCk1coX0jYEZQeKPgK45U6OVW8w8YaAUJqZILQDpBhHnqcro/Qer6JcaI3ywdYDQ oEPTqVSoYVx6nuL6igTqTNw8JgjHUUkJsXH/91SVQT730wCQkgrJwp+IFAkNVJU4gl8purQnRYvU nM8djAgYuIWJaCLwz0d7BwkV49SBTIj4fSk6SKMjcd83FBR2ZABIPfzhohi/O4V4uMuwjRLUhKmU ZFD56HMELhIyKC0xBYoZJocXlFpNQcL9ee5IGQM3naKdhmQ0GeejvYOEjHh1MNmI55emg8hG7L1n g9KOTMi4W7+JITQhsJwXMvimbaYUfxCy0eID3hlJCoQKvhiVQQQQLxXT5Sc1i9Cq9KkD4T57QyEi egiR5b61Ne+CQGiIPytjDan+XGoOAgux9ZqE049S0pFGwhNReTsK0SqyzIkfDF5zKRwd6kyYnFLM UwO0ueT+F/vaI8PYOTKrZhEc42nDOZDec5YT7hgTjh4DfLBVICTEn5WxTFR/qgbFQSb1Yus1Caeh /OuikW0EX2stCYXwnoHkkECldt5+rhgGYAqUp4gYIJnwm++Rh4SRYdAcmVWzyNjqACRCek8BI6qd canHAB9sFQgJ+nOTsWLUsw6XmiAb1/vGwXLKmShQr5G3IQgcspJJdzEccOtoXvrO01mhzI/O32GA FdG0pCPB5kAkomNejiJjx2eODI6zYpr4o4Vwp3+eNVXs5O9EryfzaozE8f7ScRB+1OnG3iWABZXf cu6Ej3NOAgSCO4dJFoyRUvT+ec3nA3EkwCdPza/dz9HRUnIDhd2pC8r2rwwuI7NqFnGsNWezVwjn OdudFjd6CFh9Jpu2Jl4BKTj+rMxx+VJzEMbrsfWahDQFFlLOlABh2nGbgoXwnrsjxyjTwfziFC2M Y3wfMT84wOSzU6+7CPoisloWiakkQYR0nk1INI+aRPxgyXgBQkG8KXQ5fGnZEMbUFx2kNZUzFUCY DYlQUAjnXaiXd76alfKQFH0z48pT/Og8pWPuktFiTx4A+UZ/i8yqWWRstQ2NkN4zQLjpMcDRc2wV CAn6My9HZBLFNARV8xTXVw2BuyvO8ebKqZAANA0Bi0vhgA8/5Fef261M0D/zFInqZVYhgZqQDBsq RJs3QsyL0DOtfYDa8qnKyIG3ZMll1BAyS1R5GcbWAUKCDk22VMgqUMK49BQ4cX1FArUYDoLw7JMX LiyIwQwefVp+TyrQQeZ1/1RKGviAQX74QcIDJSLpOSrF+ORJ0yK1FyYiRQg/sDxURJORP9o7SLiI VwfztYjnl6YgV3xfsUEthr8kTfVfCjXCRjFynRozfkwGoywsm/M/FcqUZh4rEjLYDH+QYsoQ5VHE c0HGuTEXMiJlCHxeZ8iIJsJkWeZu7yAhA6cmLypkDahOXYoOcqL7igsKMvw13femsQgXYpMZH1TG 4ZynfmaKQehKSQZzngwcBwkZbCP7Yq0Uix9PmhapvZARKWKYEmlfkmgy9Ed7BwkZ8epgZkY8vzQF ueJ7z4YVkjMNwDs26oeNxfhBqYybzGBzKi3BT6VkY7amJnKRsEH9bTqOSjEYjPbVtEjthY1IGcMs zzl5U5FhPpo7iMsPDBHHqYMxPWwwq+gA/tbx8l0HShXlj5kHUIDhPRJmxmL0ZtzMOJnBPTzJAzI6 XHC4IQMLkfahJGnGFpViTBip1bRI7YWLSBkC4wUTI4o+KEd5NHeQJEacOtjleBRtJJMXE91XecEM zN8S31j5LhfFiIACVJzj4HyugmM/UynzgjlgcmAQnXPK1QrESrH7GKnVtEjthYtIEQID5UuPQT5a O4hpcXw6mFTE76Pn/H1i+xcTGVZRP+kBUpngLv2cxl6M8eIfC/CxyjfqkwMfSskEw6BkgMhkhfOu 2R507jFSbH9EajUtUnthIlJG8I2CKyiNIoKksORu7iA+dpw6mFQkmChCJMBkxUT3Ly4y0bIwcmYH +MZVv82KYkTA153mF+DgX3KAa5fiHLf/Z7gxSLKC2/int6mQDIxQFOFckDinIh87UkTgwpdUjB5j fDR2kDChS3QjK3VCWT0H8HeOsddtxWHikxF5XKPUpUyITS/C0hsVgvwAVDM0SKXMCo5CZMTxIRIm uNoyCzaDGSSrrg+aFsG789zBDGFWYuiF3H+dKBHipzjmjPs8dpw6mEkQx1fRAeRiHnt7CNlayTQV P33P02wGIpl90WWCs4HJAE6s6JrTnySJQImgViTdzGASwdFAGF09B2Dh4Tw10IwQuLp9EiJaiI8V KpGaOkASIv4cTBLGZXoT1ODfU1RfvRjUgzhsxC8/MxsOxEgRnKegZ0hgSTQ/PpLxjE4uw81Bkg6U f6SDqRRzjidNQS57PncwI/CMna9GNBnk3d68B4OEinh1SZksenVpOsiJb9l4ueVn5eRMHFgj8zcJ G8Vo6T1KbErkgDXdQ344J2ZsWDumGCRkUP6R4Wal5tSESbKKFqm5kMF5C3QZAt8bHzKiiTBZrwv5 2pvXZZCQMU5dQieWKprEeIxuuXhZa8YQhiXi+aU4F5rFqWCTGSxNhAu+Xp88oJA6wwxuRk4WDBIy KBNJJ1MplqWQumsKctkLGUrN78mR+5ARTcZ5tzeZMUjIiFeX1IkGNqrpIGTGxLdsvMwM6yZn9sAy MtUJkxmLMR7yamIzgysh/YkZ/CA1jQWfmDINgrS14LJiG4divgFskSm1mhaJvQ7NImUMpwuJJuJk ASvsx95Bwka8KoZXx/NL00FgY+JbNv6RG+jw90SanaUpE2ChIBgxTJUkDRGNuplAnEjBjxO1Hq0a RO/cfpr9wUjhHaPoJ02LxF7700gRA9+hMjeOJuPUq8tekXnueHUw2Yjnl6aDnPi+YuMnH+ccNuj9 Tm4UY97gQqzNMldEW1RHMErpHfcIJA8GCRvc1ml3s1K0KZFaTUHQxP6OmnzuSBnDNy/2gI1oIk7e 0Lu9g/jc8epgshHPL00Huey9G2CwmeLca7zj1rC+KWJENbOosMH+lRlEnErJBvv4yhQJG1zI72Bl pRg/PmlaxN2u3IHhKHOkiIHP3tpkOu2b3T3YeLRn5Oc5BpXj1cFkI55fmp7j+yo35pKyYYPfsWwU Y4b9l6oMfwVu1m8mKCUbnKxrHoiEDU7EhaAIMYKMUBTxWJCQfx47UobA/vOQMYqM8tHaQZIZ+jTt aqV4fsxdig5iZsTc+8zggxQZen7jW93lQmwyg7tNkhlshCcPfitlq8oJgAwqBwkXFIQ4RqGhjBQz uSdNi7AhrIzPHYwYPr3d2syIJmJnsetu7yBhI14dzMyI55emIEN+7b1nwwmYicHrfcgIxM/54RFP 8uKXVxlPlArJheemFwgV1IOk0+E4uUIu6ii0eg4QW53RjYzeUwQ7PESNET7YOkBo0KEZeSg0Xcn4 fOmhc7rH9dXrQXGIAzZrY7k1zgGX41gx3P+ZajjHw51VsB0YGqgCUoZRg0h4oBgk3Y0Yvs1RgwdN B6m9pgQHEvSBds5vQpgS0UTkrHKpif2MyASZ545Xlbp5fmk6sZz43rNhTaTTCKWtnZONg8lG6uEY 0uLThK6QzvHpwCQAWyKHDIpBMu4YTDI4aaDUKgrCG15zknikDMFGQjKiyTAf7R0kiTFOHejy+1J0 IrnMvXs9XEbiJIfO8VHiw4XYhMA+t/Vwri3NNIShtFKS8RmZIjrnNMo9O7KlUmwaPGk6yLEXMiJF DHx1Oc1yNBEn1xbe7R0kZMSrg8lGPL80Pcf3VWZQkcHPqXfcGtzMCMYPSt/m6UkGcZzNmrfiUynZ YEUgaTBI2CAH0vVWir2DSK2mgxx7YSNSxsBnSIfXaDLOR3tFpqU5Xl1S1n1ONKtpkcb3LzaysuUu ldMJvfvWc8cHgw0+u2xuuHM1ExF+daVk4/c8Zx6AjHfO3WZfcaUYRkZqNR3k2JONxYiBvXvYP5qM 89HeQXzueHUwc6PRrKZFGt+/2MiKjvWSTifwbr77mFZDjIxnU+vvHPOQ6k5EfvrNStmYA9O+FYMk N1joTP9SqTk28aApyGUvbLCwqg+ExygM7VSMjSbi/Ha3NwOKQcJGvLqkNprVtEjjWzZeTtSs52i1 PP1ZtlIPJhue/PNXoGIo7wCFkg7BnLp1GiIyLby7bK1BHMxfnTMI6Vqr6SCwr73JqcVIfRJ02Iim D8yo/bJ3kLARr4rh1fEcXqtpkdjDz/etKOdR7K+R5sL66V+hcTDYoCw/bHCrcYaflOBn4MEJlYw1 B0luMFrNMEwM7xhUP2k6SO0Ni4vZEMz4hGngaDLOaCJJ4meQec4R2mymFJONRrOaFml8X7HxB/aH jU8uZm0rGkw2WHSa3OA7R05g2PtRyl6GSzUS+SBhw+/M34Q8dDBCq+ggkKG5pkak+EGZptlEOwu0 Zed1fLR2kGSGPq3UZEZjWUU0Fg/RveXir0WSSQw+qztn0g/EiPRvyphdZ+gk5AOhYYLGLlELhAgK TTIUU2jS4tcIrR7PIFy2mhSB8N4rp6Qhaoz6ZovmokBoiEMHow6qYRw9C8B5PHz9ekxHgjDqHfrR 4roAQ9O5mDRQhoZrDOI4yTdRzyKQCcFYLDGLtCNhzJbOt1Jz3ECp1XSQY8+YFmMsw7IQ/Dn3UxNR zoLPZe8g4SJeHezynMY0mjaWy95rNqYjQZp6Du3CBnd9lQ0xXgZiSImYa/bpIb4rpHOfuaiAZX0R nfNi6tYzVmrOG0hPFE2acCbhZi5kFHP1suOraDLMR3tFhvw6tUJXLEfRhTS6f70eJzMo63DlkZEJ 5anlIhghUAZiDRK7WtwOOZnB+dqQwdVRmYSIjHPO2aZokS2gSM2RA6VW00GOvZDBAQt9cMTPEVZp jSbiZKp6t3cQnzteHczMaDRH04U0vn+xcTKDwo4MP3+wb1g2xCYzKCkJG2yI+RMTp1KmBpcPp5Uc JKlBcVq63koxPH/SFCTsq8nnFiMGikBBmJSNJuO825s2Y5CwEa9WitgbzdF0IbARe31PXg0xkKay I8NP7kec06WL8Yv+/jF7WhLbyQqXnFiOL/lThcS+is+FDfbW7XqLGSdHEeRnNR2E3NBeW5tIkRt8 UytsRBNxMkW33Tr2DhI24lWx/5gbjeZoupDYw8/3rcbv2Q/TO1ZyHWKgIZgx8CPjCy+dp/wmTqRA aLGzj0bkImHjk6HzSk1bSNGN/KymOYrgUDP2mhuRkg2qHSY3osk49eqyV2SeO14d7PJ8NW0sJ76v 2PgzCzRI80XakxvBiIGxadigdNryG7aClNK7uRBvkbBBVZ+97UrNAQT5WU1BZpo89nzuSBHDrz8z nHMaqCbiZBx6t3eQ5Ea8OphsNJqj6UIa31ds0GlYHIc0NxC03RBLDKmnYhDnZtLEqZRs8ImhRi4S Nmj/XT2EjUh5BuFR00GOvbARKdlgLjS5EU3G+WjvIGEjXh1MNuL5aipyxfeeDQaGHnSQDRYfwsbB +EXn7kjfFD5rn3cAobDB3WqOP9hZEQkb/rIL+aNzm6VCUTRpIDLOjbmQESlCmLfeRnQUEeU0DHIf awcJF/qEqiP19/i9ii6k0X3FBZ+PjG8/GWTZaDg5awA/5h4LlqXYEE+QCunbEiEQIqgwyWb17wh5 DOFBTwFpP08dCBpYA5mcUM28Iaxrpr0YW4Z8njoOHWx9loXRc4K4bL05k44wxR0axTMP6YaFYP6O 81FBkozV5P78SE1T9jnPmRA+Jw/OoGbrsZgZwdmEDD+jaZHaa0cSKUNg/H7TZJRql/bYOwj21qti dCQbzdF0IY3vq5Sg5sTxoIum7A2HjWCw8dPfR7u5s5mdJITGOTbYHHUUCRlsjDm+Cjb9CGcUlDqK ICNIzQ2JizHA99OB0hpNhMkx/5Afe0V87jh1ILloLEfRhSQ6Pi/8rlNl5f4byzbDhWebwkUwQ6AY asxy37W/MGEiNZnBiKiB+1zIoBQlXe+cW5phZKSOJgeWd3tn2BqMGDwnJRnRZJxqNzNi7yAhI14V m8xoNEcTbDzG91VmUOjh8NOZGKfww0Yw2eBm5WGDr8M1EZAa7/7Oc/IjEjb4CmhaTjE4I5hIHU2L 1N48d6R4T35+MKaVjWgyzmjCvfgZpKkRryqFnzgTqaPpQhrfV2xQ6eHwE/K5Qba5EQw2WKkaMqg+ SX0Fl+JnzMGiZAIfJGRQ5JSet1K885GqIrgIUGvNjGIkBEPa4SKKiPL7o7mDJDN0Cs8jNZnRWI4m uHiM7isu/nA9WLhg6buZEUwu3Hnn7eSbi3kBqJEOF9/nMRJjkHDBV1nS74pNYnACQ3aqCC4CyIWK fGwxInBR27yIIqO8m5vuc5BwEadWisgby9F0IWvv3bSEcTSbc8PFby7MbV4EIwRWroaLKc6wyfip kOMNSjiywDtIuKCkIj1vpRhGj/KjiLgDwMVYCxdixMlWEWtRw0UUEeW3u7nJgUHCRZxaKSLX75um Ild07/PCYksnEnrnXXjcV/D3YIbQBoMPKiT76eHDBZ83SA4MEi44W5nut1KMppGame8ogguBy1q4 KAYXzuLNiygyykdzRXy3SNY4tVLrt6yOpgtpdP/iIksZVlvO4B/p7wwRbD0PJhd2+pjl1u5GrlBm bJ19/ASZF985WwmLEIsyo/vogYoAEq9IqChGAFxiPN3U6DHGR2MHCRNx6WCX1zKBoo2DPFlz73pV Cy4tY0R6btkJE8GIgGW7aZKQSe7P7UAywU2pyYBBjMn52mw80tN79Y8Nwd9IjaIFxrexFioiRARz 88+l6IOxRxQdcweRC8bGOIWuI8Uv2FiOpgtpdP/KiizpWHTpmU2EPeIbKrxK2ebOhWls0s817BwL dqKWHoRSixy7dKY2W47wAMZvy0D0puX8eexIQiXkgJ33ITNKjO5mKeHW0rqzQgQ8/q6eBRrTvwg4 rwUVHQ6+keZMUxkIJgVORDGab2LwX2XMhT/5nFiQvhWEnkFXpYjsfHQsE8wDHGOhIUL6zy2C81pE kSE+mjtIcmF8OpCvRSOJonktRPLrquhfTJxUoK3NwPuTXe0yEQwmWMDhQClU/KQcYwab5EW4YOkq A65BDMoZWisZK2WrNVJHE79/EMgYeyGj2CRHkiKaiJNFiLu9ItNcHq9WCjYazdF0IWvvdSNx8oJi kwy96e0PG8WYATL6m8Tgc12JnaEE3tGCcjFeskCkmcFXtvLGRIrYGbqr/Wi6EPck0RQ2KkWjQmHn pEY0ESdnPu/2DpLUiFcHMzfq+dF0IWvvNRsnN6jryNCbD9o2NYRILt4FZ9qmBtW9CX0u6oeL7/MU o4ze709icFwkfcfIEDfHL9QcNdNyFGEJAlshohDLFH9n6Hb0ECKLW8N7bBUIDfozzUdkKB+MraMG Fh7i+uoF4XB0GssftPVpK4XiPZ+2GxJ+U7A978csa0HCR+60gQSeMhynS62DBJqQOcBwVwMtRawG ONlQCO+ZhN70GLOdT20l4jzFjxJ/VsZq47F11EBCkWPrdSbkvbCm0rkCby1X3g4JC9ExTLEQRt3Q nT7kryUiOMbgwZ+dAQYluSGBTyvbwQ5ENC7ajuaj5kJiK5lQIUYYv9iRkszoIUAWX262CiQT4s/K rMtRw1NBpq00rveZQC1lDu3Q/9PkmwkL0W+zPQ3CQJNTNDM+5PjtOMZhFAdXDO4ADIcVEcYNC83o MJqPGkjg5EtGldga6hZixEnDcNNDgPZBEh5bBSQh/vDLHZm4TMhHzRVE4vqKBFbE+AHJBPduQ4IQ GvHebwpplI8dNGY3eHGMrwDkZ/ebPiHh+12EiLn6XMXRMu9HkZjyoRXCeb9MJJejxvAeLBUIBXoz 70dkrAMdU0cLFDxE9RUF9IWuX+IXF/yVA7F4bw23HHx4XHt+d6UkgSW5/uwiYYFJsSuYZMdIkdlc 5HvXdCG153MrRQi0NENENBn4o72DhIt4dTBTuJ4fTUWu+L5ig4/MTD7w8h8yArENb52pXPzwI/FG yedLwgV3sA85A4QKCkrsVY8QQXIb5XAaPQsYPfpCRGScks7lBWy9jRoiZDHjZusAoUGHUB0hRjgW h2rr6Fkgtr5sGijRcDxJgjJhnKUIqimDsdHkDFEaOhbio1BxjItHHFXxeoiEBWpE7E6DTVPHz63u owcaisTaPLcYaxJcBjAJEU3EyGviKM553mgK4nP1qTL8vV4fPRdyrL1dvLSU0jmBTHwyQEgDUYwI KLUYq3wEt5Ej1ZnXzCWI2+fCBat09qrFTHvW+9R+NF1I7DUnKkUMfOVzuIgm41S7wxY2G9VUJFzE q2KTFfX8aIKNx/i+ejmo03DEh3fsajUvihEDn7YZNvhySzL/F1LjHWfYmwUiYYPT4OlelZrM4HyC 2o8m2CgSe82MYsQwxd7O5KaCktDv5qaFHCRkxCnJiBR1oTFXRXDxEB3WOoZ4uXNu0aSzA7iYLz6b GYv51nqCjHykDiaR+w3myQxOaCQL5qvM4YJDt2k/BjMzOJ+g9qPpQmKvmVEpVjbbakYRUfo96aE+ 5g4SLuJUMZxax6voAhrdcvHqDBXSLH049GP4w6r25MViLMrNXU9wwZHIRP4TqXHu5zzHGzFIuKDI M52t2OQFpxMysIwmuCgSe82LYoQwr0jUECPfIrmMTVYMEibi0iW1bo+a58iunHjiIeNKiyWnzA+/ +HDPtBaLwcOUAZFn3OaUzGfnIa7R2uf3F/G3pXXjg5PpcCNF1CxrpQOJpgupPflbKRbu/brXKjLG R3MH0dxx6mDmRIMZRYgEgMC19nqMndkWpZLWmStNsf7hohjDnKkCwuzUntsmUvgfLhh65vcfJFzw pmaoESmiZF9N7UfThdReuKgUIczb4GRORR/UKD2aOwi3F6xTBzP0On4ULbDWXnNx8oLLXBwMznr6 4aKYQz4PqmOWCrpGzl0/IM57Or1gG/nkhaeIh7ARcvgd5UfRhdScVKwUK/ks4U1ejCKDfLR2kKRF fDqYVNTvKJq8EJm8YLfAU+SvuTh5wWKDP4GzQ4JLu1kML7m4bn4CGrSECTHhgk3P5MAgyQs2Q9Lv Roow+VKf2o+mC6m9kFEpYvg2ayDO7dREnEzO7/aKzDvJuHq8uqQoB4291bTI2nvNxsmM39xNJxm0 DZmJWkQpRIpPxQt58entef7kHLtMr+K33g/Q5oLSkHS6ESLwHw96FqitEBEZFq/9dpSNhWr4LfHi wdYBkhRx6GAmRXxePQeILTx8TcJJCUoyJJ8VJafaSYlitGfWkPl6MDXIqJuDjaGBj9nk5x8kKUFV SDrcSBE2ZxLUfjQVmYQde2GiUqxXfmM2JRXRRJRMUO/2DhIu4tXB5KKer6ZFGt9XbLAN4U/g3hmT 5rAhlhi8lV02WKaYFKCyIs5RFpUEGCRksL6QusZIETqHKe6KLqTmQkalWK9k+jFkRJNhPto7SMgY pw4kF/V7FS0Sc19kBl3WL6cPcMFGfObnizEKoN5juOAaapeIWYhUKjO0ea5IyGCJL8OISBE6NWxq P5qKDPdjL2RUihi8PsPMqCbOWz7ZKxIy6tVKreeraZHG9z4zLJ78cLWSxfG/TO/MjGATA/Mzt78Y xs6NcbDBxmXY4PbFDCYHCRtcbuKu5JEidpZ8lGLsmZ3wRWovbFSKxVs+PTVsVNOcEXiwVyRs1KuV Ws9X03N8X7HBjcgZdLIFfNgoRgzsYNORMLJ1nWriVMrc+MxzRC6id06f3DKUjZGCDahS+9F0IbDR 5y7MkrUunVYTcT7ZCzKt9Xq1UkxLYm81LbL2Xreh6UhYJqHRxl8Wx3+R5smNYi6PWwIEG1xolUzg 9lffHRbV85wvA/cTho0PqkCGjUoxGI3Ualqk9nxuMZzh83ywz75bNDl2fLRXxOfWq5XaaFbTImvv NRvpUVhG5cM4YYP6jrIhRsYTg3My7dLXpUVgHhU2WJVOHgwSNpiJZUhRKfrhJ01BLnthQykwXJ8L aGGjmojzbg+fioSNerVSG81qWgQ21LRvysupqinlVMLc+Ebvkdwoxno5n6IYNrg7INOLX0jx6zFh y3Pkhs+FDe9/90UZyB+dQm6Vr6JFas4h9WKEwHAVc+x2jSKiVPfNWpFwUZ9WKrHA2Cp6jm65eJqi ncxgO8UA2Dng6OdUU7gPNRghfHjCkMxgny95wGcm08dQQJbABwkX1Ig4sqBXRWrC5MuVSq0m5jUP 9nxuMZbr/GSwZFQT089ocj42moIM9+tVMVv/er6aFkl8X/avHNQKGZxdamIEYgWO76IPF7+8wWmi RIgombTNU55LyvFCAqBExHFFMYP8OUKr5wA4dp4aaH5LCgDCQ9UQoUKXrQJJijq0QvSnYwsW5mo2 eHmI66uUoDjDn9F9FNqDvB5i01h80F0MDb/n2jgi/JmdOOdr81yRpAT1IeluKsWI+klTkMteUkIp MFp9yowmJaqJvuJubxqLQcJFvVopRp1PmhZpfMvGy8bCicwcZUKakzbDxmL8mFa9+YIwTZqfm88S Z9RBRbujVofbICGDCpH0NpXy5MMoX0WL1FzIUGrygms7QkY1EebdHh4UCRlxaoXWb1hFEam7yJp7 140w2//eIj/uETxciE1mcNsgO5gkG5/KTjPAVnfIoBggWTCIzrlzNaWOzMsixTDySVOQyYyxFzKU mszgukHJWE2M9O728KmIz61XK7XRMM6KpkXcOcLzrzKDYg1/TjYVPOfpe8IqWTD2FdjKHLtsGKUl +EBqmgsus0oaiExqzPm6rmWJ2V5wIkGp1bRI7M3QZDHCY4LKc6uJpexoYqIYTUXCRr0KNplQz1fT xtL4lo2XHQmrxj9nOoH0D64sCBvFiIGVuGGD4q5kAv9IbnDaLYOOQZIbpEZeJzG8mzMJGX5W0yKx 1xeMUxhKEYP/kI1qIs5ocqI4MkHmOV6HeFXM37Ger6ZFGt9XbLAPkeEni1knN4qxeu9l6OYkd5in 3IJd5bDxJ8+5cAMSNn78domLpqRSc5xC7aspCBmPd+e5lSIGtkiHjWoizid7RZIb9WqlUBqvVtMi a+99u/GLWw/HO+72z946rxjYxMCt+2GDlYpkwlSsZ+rWacggYYPF7xwvqZSHEx41LcKPpj2fW4wZ 9/cZwrhfN5pICaVu9oqEjXq1Uuv5agpyxfdVbrgpQOS8YGzVtN0oxv4CG51kPEPauUmOX52F7/Qx s05vHgwSNtKQVIQRdVRHzYQtMq6NsVBRKXotriuexFANDcWTpSLhof4EmxajPq8aOpGHyL7iYW5B HB78Bk5ajGIE4Fdw5IGiqrQFnAUID9S1Zug5SHighCR9jti0GBwSV2o1eTBhEBs2NIWKYsTgpoot RjXZbI4mx9rzXJC2GPWqmC1GPV9Ni8TeF0NPS9p0F2HWeCYpDuSuDh2qXPDJjLQE3xAiSqZs8xQp 4VOhgtqSdDiDmfp8BXyYqJ4D1FaIQIY0wfvvdMvyUDVEqOrLVgGfWodW6PgcPXh4gLX1rqFwaj8z CHpiro0tC2K4Bg2WTWmVkbgBsSSslK4xoPLnLxIeqC+xq1mpOX/woCkIMrUXJhjnqYs1W74+iXbn faOJlee7PXwqEi7q1UptNKtpkbVXNl4OPS2mnEkE0l5p5AuymBst7PabeK7PTpgKScYnj+EcgYvo nNMnYlnMpOBIgmmyihapuZCh1OQF57jTQ1UTYd7toamIzx2nVmj9hgsvbLpFsubeZwYlHf6ajFjZ 2CwXYpMZLKaECz7MkzygAMVOhfXZPMdqt0jIQEUGE5WaMwkPmoJc9kIGJxf0AaV+kzyzsdFEmc3d 3mTGICGjXq0Ui71PmhZpfNt0/iMzqDnx50TaixaSGcX4QVkonx+B18SfmNi96UE2ONCRNJjLGMIG u2EOJiLFcwyoI7WaFom9GXouBhs0osNGNRGnXt3sFQkb9SrYZEI9X00WkN7jWzZeDj3dWZqpBN6x ujkdyWLEMFT8mAvkDBKRcY1LypsEIqGCm5Hsa1ZqDlNYHho1+TO8n4cWw3vOaQ8PVUOET8aCDH80 iHFppdbn1bRIIsPJ1+9I1jxpIF02MCv8mHOzohiqphV3wNZ5xW+FbE6/5TGW/f0GdJjgC2VpVSo1 hxIy5jwnVWaRYq352ArRBllI7BtSRewVPpkrkpyoUyu1saymRRrdv7jIuha9viWf4x0XVZcLsfkx WZ/CXcZt86kYfm/ORIULJg/JgEHCBWUG6W4qNacTlIJUL+8gbo4r3KyFi2K2xzAuF1VElE/mgjQv 6tRKbSyraZFpZ03e93lBN+mPScfp5dJpLYrxqs3cwAWAmbGwwK+QeUEfnBwQaV74AWlfkAqxexCh o2eBGgsVFaKfZtF50iJ6WDN6MlYkWVGXVor5TqxFEU4G4Tddc6+ZSFa02m+k/eCxTCxGBN+oryEr SN3kAIeDMuCYWsMJWyRZ4UA9VETK0wmzhDuK0Bwgvp3HVogI2LRCillWFBHlk7ki4UKn0LVS6/dq WgQuYu89F355Dt7+eP1lmGDh26RmQQbHGL9SVjcRzi2aZkQHVPN3WKCiJF2M2LwKM78/ShhOq7NW fKQII1ebP+OvCpdRO8zlyIyRzt+Jvq6sjMWj6F0t/XvtvI487SSLbdwhHq/4BFljL4bj07UzbvUE hL8551+TBbRp+cVF5n1gac2zCzepOX2g1FG0AL6pKBxwREEh/Ge1BNpWERE+mSvic+vUSm0sq2mR tfeai/NGUMJhAEjzed5yIcYvQAgpg2DcOkXLFiB2CM6R/cYtonPOk2ZrcaXmEIJSUTTZwbGEm7mQ UYwQGOfeNRHmk70gw/1xaoXccRnlq2iRRve+ndTvSXA2Dbj1vVwUYwDLKuz8Bhw9Thpw1DOJwXVv ZiM7aSBNDCpJMtKqlG6O1GpapPYkYzHY5MWczKgmxvBP9ookM+rVSm00q2mRtfc+M6hsMAJm4H5Y LC2E2GTGdz9SZkbyNfskAlUlYeNvnoN8kaQGrXwGW5WiGXvSFOSyFzaUAmNrgyvhJzWqidy628On ImGjXq2U29aPmhZpfJsbL0fcjhBmpO/2WqZiCzmro8hKMvicfRLhB0IY9FqSni4CCBUsbWa4HRm7 +dG8ahaJrRBRIYjgMu1Ji+ohxJstuAoQGurPkVmXowYPg8DesfU+Jfhunb4jzFghGVEI77/R+0mC 0wE08vsjJAmcU2zEACGBa5zTxUaGMXBkVs0iY6uDjwrhvWvqNjnVQ4CPtgKEhJmeDC8jw+AeWzi4 aoJsXJsJL2cbzKp/zyTBFbpUwAdCo+3z1M8xWGUwgEZbY1lhkpunAoQESjbTv0aGsXJkooYfeZHY 8qmF6DH+0NplkjV67GkebAUICfqDh0dmo1g1ixxb7zLBckIKMM0ET4CaCYEIGe+pSgVhlMoyc0iY ybhTns4q/PZ3SCCIdLNAEzKHJWRl1TBsDhJbIaEQ3nO30pBQPQR4s5WIa2v9OTLr8qpZ5Nh6TULG EbMer+8Id7d4IbxPOdAs0SfVP2atdVbMG3EulZ0Z2F1kFt4VWS2LxFQ4gFyFcP6jA3bW+RPxoyXi BUgesPWgqUD+auwpPGhZZEzx4r2mIMMHNu/o/WWAkqqmQRBGj7+ncJSRmacofRUo1+IXdtaV/tPr fvSKDTi+SnlEABgTjsgqOUDsJHxECJb/4+0/NyUMMKdG79jJ3z6zrhyR436U4NsBjp33sVOM0eAZ zOUdKITjmUzwuvP1xoSGkNFziWBj7WjQidZsEeLfyPCLRiZqeCrIvG9nDLlCOM8NL8NA9RDeo60A oaD+DDQR1+VV4471La6vEoD3PwM6KoZn6YWSh0B4T3M4vLfH4yK8oYBXNGMGgWQAHWd6zsjQNiND wKskSHKtTy2E7xwBGAqqh/AebQ0w3Qj9ko3SkVh3V8kiZMFYep0FpxGg+kLPLbZgVTppUMzajslF 68Rw0Jfgm1Lmwd88F2Q8w97vlhxWirFbpFbTIrUne4vRTFCewa+6mphvPtkrkmSoVytl1dtEs5oW WXuv2TjtASPnDBqZTx02xOanZC3JWicGoixewwaNo1J2FYxtfa5IcoIPRNmHrhTj+idNQZDhHVJT 2FAKjB07hqJopyWLJhYn7vbwqUjYqFcrRcHnk6ZF1l7ZeDlotGiSxZXxjolTc6MYDUHK4BiI/mEE N3Eixa/HICjPEbnPhQ02gc3bYv7qnDNQ+2paBO/UFDYqBRucNR82qok41X6zVyRs1Ktg01LU89X0 HN+7puLHN8ohOc0QNrh0mbbihhFD6uAYkFKOnDi9mlnvuLi1kYuEDS6KtR+DDXXJBtUnSq2mRWDj PLcYMXjNs7lRTcQZTWsvyNhbr4rdPK+mGxJ7b3tOpUlQI6Dz+2TMGDaK0XXRBuMdgzI2q8lTxi5K yQZZk8gHCRuc0ciYqVIs9EdqNQVBU+0lNyrFEIu1DsmoIhY0n8wVSWrUqWK3YKLoBqy1f7caSjOA yWiS6f7hQmx+T1bh5IKxGT/z/ObcphMuPiMDOyI655SrX4OqFFuRT5oWwbvz3GKEMHOLKMIcUSI0 r2XNBRlz61SxEwzPRdENONbenLBVmhUhBwwsy38ygEpeFGPs+HNOvHC2yX+YFzSpjqqcds1zQcqF X5c/QpDjaYMRWkWL1JzFqYsx15iSdmZwo4flfV26GSsi8etSsRMKKRc9B7hi+6q9oGQQYdyxejFE AKGQuo5UbDFY9YiVIf5SSNcoWryApAQFIna+K0SwIxQ9hHyA2vKpAzERpT0ysaqFmqJHUwVCgv7g YTFjThRRc/29lt6/GBRdOMr1J2GXNxyIEY0VSXMpAoNWv5I38SklCX8iY1aAhAWqPrKqN5gRMgB/ 0LRI7YWHShEBy+ISoSLMWVzFzRA3c0H6YtSpYkZex6PoBhxrX7wYlFwkH1gKOlwUw6eUvzF4nTPw LMUgNL5xXWgSQqRUsEKRHncwqSAKpVbRIjQMagoVlWKFxu8xyEU1sQuodrmovSLJijgV6D9GXr+r 6IbEHG6+zwsKQQwAaY4mtSstRgh+9sY3ko/ZJ+8/kQJxAjTPQY9I8oLKj3S4YvMycE5C7avJgfDN 3jy3GDFw3+GQUU3EGU1rL8g8t14VM/Z6Xk03pPF9xQaVF0ZAH8fB4b4lxXht2YgfNtgMT/7zffB0 rh95jj5VJGzw0YcMOivF4DFSqynIvJVjL6lRKZYxvXvC1KgmNsqf7BVJatSrYsZez6vphjS+92xY GDnXv3UMYJsRbH5RvvpreY8He2ag3/GEucGmZCN37BA2qJDLXY0z6vBF4ZxEBp3VtEjthY1KEYMF SrJRTcSp9pu9ImGjXhW7eV5NBwn78fPdm0IRJBccT25QcZHcWIx+jaIvMt4buzr9YP4Y7zgiFDYG CRscJUqnW6k5XRA2qmkRRppqChucSlCKxf2Ms5z8jSbKTJ7sBZnmZr2qlLE3mmq6IWvvNRtOVpXm qlCHUO4kswD7gDng4RIp843DX4n9Qyl/K75yk6GXyHjHlI3XFl2W9owUQ+lIVZPZUqT2wkYxdiqo N5wmupqsQHq0V+T/3r0qdoummm7I2nvNhpNVpX/8mUkE0vRkZUOM7CIGawNkY6rDJ06lZIOxRvKA xZSsaDtlm+1C2IgUY+kHTbIhEvbPc4sRA+8f2lcTcd7t4VORsFGvit2iqaYb0vi23XgxWVWaC438 PZ1F8mOHoWK6SUrIBqtziZ0CEWsfrJXp5ENkMp4pFBNLKRvI1581QJVX0Q2pOQeeK0UIXtQpGVHk yX9036wVCRf1qdiJBcaq6CBXdMvFizV+pWfVZLhwX+J/6bMXIwRuuB4uKO/JGzA7HHrH2bEG7nPm uw3s7BdChvsOhvkRqWq6IZBxnlsptjJ+dWhRTcQZTWsvCPb8hepVsZvn1XRDYu+LqSqKZn8B4Y85 dKyCQERgzZNWP71GZn5yhIiSmdE8Rdg+FSqoA8lYbDADz2dnq+cCasunRobfEu8pTB9GqwYvVH3Z KhAa6lCxy+fomV/0Ia6vUoICDAdTLDl522FeDzFco53jX0PDrykHoHFUSM+4oD1jsEFCA3sxGX5W yqH8SEXR5EiRmgsRxYjAt8ncqiaS88leEH5Zf5s4VSGZaCxVdEPW3OuG83QjtHUZivN5ysNFMULg +zVjlo/bN3SlJOMjz0EPSLuR3x5TMXHEkGJMHalquiF4p0zIqBQ7GT+ojJGMaiLOJ3tFQka9Kmbs 9byabsjae81GuhHGKn54xcwgvLARbDIDc2GDcpfMMDAXNjiRkmH2IEkNPo+eAUWlWMBB6tIkGyKX vbBRjFVtvjw+qVFNnqW82UNTkbBRr4pxBcqJpppuSOPb9+RlN8Ko/u9MJ1ict8DS92Qxxop/qCI2 JfmMWWK3DHOaC643Mg8YKfqcUbE0fi6THszYOWGgVDXdkNibnFopYrCg86aJY7ZqNxdrr0jYqFfB aCLW82q6IY1v2XjZkbAW8m2mE0gzAJiJ2mLEYL2L+cZ3jBo7UuMdRffJA6boh43vuQwRhtRl7Jw5 UKqabkjs9U2pFKv8Pzq8qibijCYnZaMpCM/dvCpm7I2mmm5I4/uKDe4uzPDzm/f3JzeKOauipku7 nJF3zdMZ1ByFZtU9zxVJbnxgfvKlUjYgo72abBFE0OReGZryhlWKldwfdNHmRjV5Hv3RXpGwUa+K GXs9r6Ybsvbetxt8VyYlfSwZHDbE5hdl+ytsTJMnG6wZnKnbyBQJGxyT9vzGStHtPmiSjSJsdKkp bBQjhm/TlToNHE38DE/2ioSNelXM2Ot5NR0En9beezaYNvt7Is1FM82NYqw10GNOlz13B0ycSvke s0WSPPglEjb4x03GGcrIRA9PBYlv56mV4of166RmxughxidbRcKEHqGpmHHX6+q5IY3tq7fkN7PE +ObBzbwlYskLFi3mLYGRxq2UbSpfxVSGHBAJE5SUOEIpZuQ/HjTdELw7z60UMVDMPlxUE3E+2SsS NurVStFwxl41ycZjfF+xYVGLaUGBxyEjEDuSforCFoOJZLKfK7XDhTs+hj1AqKBmJB3wYAb+c4Sq 5wJiq8OTyOA9H3gcHqqGufijrQKhQYewH2w6kvhcPbLwENdXJFC24VAKz7g/s91IMdy3BkYa/GzS BO1HKc0IWncfc0jVQ+hO12a3MdiMmzg+oVQVyUORmGtLUYxOwFuNfDuqiSB16mYvCM9dThUy8Ppd RTfkmHu7DmyF5EwikP7B0MrXYzFCsBZGs9z32NCRGuf4Ilx/f59LTlBEkqaCjQ2bEwaRkaqmGxJ7 Z9AaKWLgls8ho5qIU02SUXtFQka9CoZX63k13ZDG9z4zWBbzig4zw33TyYzFXMMmZBlmMSOx/3R3 Ve9Y8kiyDCIbztimEpHpiLqMnSUIparphsReU6NSxOAe7E2TIxs03ewFmdRYr4oZe6OpphvS+L5i 44cfkh02uKqnuVGMGLhof9jgwiKH1MSplN5xXVHyYJCwQTGOg5VKGTtFPUpV00GmqaZNS06tFBsf 3H48bFQTcT7ZK2JuMNSMVyvFcnjsVZNsPMb3FRv0mg6k4M6i9rwpYonBGjztfhBz4lRKNvA7g092 oROVU7efWRcbbIbUD5puSO3lDeMuZ3VphoVjc6OaiPPJXpGwUa9Wil2jB02y8RjfV2xwS4O/J95x QUXZKMbGl5tjssHlRo1dKdngHoXmgUhyg+2mdL4/xYz9e6Sq6YbUXtioFGxwHd+wUU3E+WQvCO3N zauVsgx0vKom2XiM719sZArP5vSfKQR0x57Kq+RGMWJgX2zs8k3RxP5DKdngFkYHqKzKgExr6FRq 9h+DDRt8ozDD2NEkG0VqL2wUI/UpT+0kcDRh5slekbBRr1Zqo6km2XiM719sZApPU/ttJhM8ZlV2 2BDj7YENFriHDa5ub+xK2ct8z3OuUYEkN35xRGgISoE3o8gIRRFhLlJzIaNShJATWc7JVESUT9aK /N/J1/q0UhtLFcnFY3T/4uJkBh+oyYjrF1Vo5aIYm+KWVJmRrNrkDYCGcEF9a8ZcIs0MCkxKhlKG TqmsUtV0QyBDmZBRKWKgrHwyo5qI88lekWRGvVopxrSxV02y8Rjfv9hIZjiTw1srGBhbmRdF8J9y qvkFuNo/THwoY1bMCjkLuP6dnKAEJf3vYEbdaZ5ahpdOBcdOOBgJ12N22jc6aJAe7PTvxF9fguFd va2W/bvxfBX7HEWa6NnwmrEF83wLUOf3s4DKTPib++CIF6mJn8uF8qsPEgaoPMl4Q2xeAoaRN01y UgTv1BQWihGBxyQy4RtNxKgm+a69IDx386qYkdfzarohsUfGvpugsvrFvCG5wEplc6EYFQzcnI1d Nhw6D/9QSOc+8xj0iOicM7XZbixm6JxJUHkVHWQ6b/ZgDhmVIgTLZG+aCPPJXhHtHadWaGNZRYvA Rcy954I5QM7osW59uBBLCDZcmmVekDzgMuGQwbftkwWDhAzuvs5yz2CGzgmLm6YbYukImpIZlWKo b5WmZFQTA5gne0VCRr1aKVZ5Y281BRnya+89G5R3ZNTJ11sPG8X4QaltJ0sZ2+ZKOHa5lDI1OM+V kdcgYYNo0+MONmPpSFXTDYENNeX94kCGuoiBwoVho5qI88lekBlZrFcrxc7Lk6ZF1t5rNtJ/WGI5 swik/QZD2sxi1HP4ZQjZ4KrhZoJSssFFwxl5+e2G8c45VNvS+Z7DjB4jVU03pPbCBqca1IVSemLY X03E+WSvSHKjXq3URrOaFll7r9k4/cdv7p/GEwL+zVwsbIiRXcTgUQjZ4JP2yYRvStl+MX1NHgyS 3OCu4vQ0gxk7hy1umg5y2QsbSg0bfFU2uVFNTHzu9vCpSNioVytFiceTpkUa37ahLzcHCJgbHuMd Rdploxj/c4YWjANyMxxb8ApJBrdAdlNeJGQwH8lwY6QMncsjVR5FNwDf8tjBiHP2wMJFFRHl3RyK ioSLOrVSieWu6Tm65eLl1gCvM/vs4eIbMzH3mBeD1vmhmDu4/OZokp2QcMFJhyaBSLjwaoUVMkpO Gzh2jZ4DXMaSFgqBQQVtDTyT9aOHGJ+MBWmLUZdWar1eRYvAfJx8/Y6kxbDSckb9NEZeWOI7shgR MA7yFWFjzF+XrWKFzIqpOztIGwyKSzKoGCkjZ1nYPImiG1BroaJCNskcXpSKKmKh5MlckWRFnVqp 9Xs1LbL2XnOR9oI5wpyEQtgRY6gIRAA/ph/5m+vgIEIRifg5zxRIRlBqYkoUM2wP1w8NbojNnyRI 7YSESOA58yv0MoiOEqJ7tFQgFOgOmlbohLB6DrC23hNAMUcOdnGz+2FAbJxPv86olnKyhKeQHPD1 hP7wIrrm7IzbJoYEMaPmCMOl6AbgWx67YQTgie2jCHPEiBQ/wpoLMubWqWImcf1eTYvEHjn7nou5 P8ZfiRtmDxfF6IeYmttm+A6zV47EeEaDnl9epESws2b3EikemrMYGW9GywI15dxyMdbXOSQxRFQR 3anKJaLmish7PFoRNl3wiJ9h1QTBn2PsCxao5HCogDQH/KehdItvMFigF9AmB4nz+89XXPWMz+z4 GEH7WDKCchPHBcWMEb3myFEU4LLmY4sRgZ/iMyOqiChVfjMXZFqjdWql1u/VtEij+1dGnKaSLVx/ SaT5Jmnbh2KUeTGHkAt2dZsCCukcTUUTAGScc9bEjsZwEak5hqHUUbRArYULGiaFiIBiRNJkFRHl k7kiSYo6tVJsWj9pWmTtvX87OPrgj8n05jdvddpKsfk52fEYLjgxkSz4rZAdCecqkgODJC846JDO pVJzDEMpuGBpyhzg7MPNWrgoxiTPcga5qCLEn8wFaV7UqZWybHS0r6ZFGt1XecGagD8m0l6+Gy6K kQ98kmDIYD6WNPitlGSwL5AkEGlisPSWNqVS/GSRWk2L1F7YqBQxeJGvbFQTcT7ZK5LMqFcrxdsZ e6tpkbX3PjOoVvTnJO89CRs2xCYzWO4OG9wsnURgVho2PvIciSiS1GClN11NpWjenzQFueyFDaXA eD24enZyo5p4re728KlI2KhXK0UX/qRpkca3ufFy0M3QkbKWeMckcdhYjF+UUsvJDT4emEzgIF0m rD/yHGz4XNjgSC+qCvlacJJT5atoEZybx+blqRRkULc6ZEQRUSpkJtZakXBRn1YqfsPYKtpIjrm3 fQkLLuyyDBecCIULnFuM5voX1Zsz+Mw9cDTOSundZ54rondM0T5mizHYhMlAOmPNamIY+WDP5xaj E/GugpsmOhrsoWntBRnuGUjGq2I3z1fTxkLvp6bNjJdTEFat6cR9TfzQlm/JQaDCO8+kghNyaQw+ mYLpmcfjzAD+Hr+cnc3mYiAD/Dsiq+QAsRMOEJmf8QenLoeCKiE49F528rfPrCtH5DhL9CiZX/Ie z1exU7fhgIERTPc/rMIciAEMFbcoZLiWq98Yscz2h/Oxzi8EkgYUl+SdiAx7NHc1eB6EcGPLp1aI QgmK3kHYyIkehl2PtgKEgvoz0ERcl1fNBnFsvWsoqQFwS3Qc4z74ZEAhSLDcVhJ+dGTl5fP+Mn/y EKQAhAOKStKxRIahNzJEvFqCHFPz1EI471eY5KB6iO/R1gDTPR13jsjGsFoWgYJx8DUFGVFRGOOl x0PBHNH2JSiE89bZSgFH3zK6/ImQHPzNU/yfOWit71SKpDuJDP5EZtUsgmOoSR5UiP9HPQk/6eoh wEdbAZIH9efIrMtRg4dBknNj6zUJmXVSV8cnnIYEvvHclqAQ6+Y/2OmRBH5Sf1N/gYytuO8rv7pA EmG+AHXJ2DKNzKpZhDopjjaHhAqZETNIc37lxuN3z8DcbRUICfpzk0kUOLhqnuL6qkWgqtofEMfY 0MnLIDRpzPpjSKDQOm/7D4RmspOnYAUgJPDrZjQZGYYxkYEE1PAjLxJbIaFCeM/mMKonS9BjzA+2 CoQE/fnPJWNN6ESxahY5tt5lAiWTXI47acNXmXMUOxjEuJnEyNBUYCk7vzzb9Bk+ULyQ4cMg4YEP aGUHtVLOGUZqNS3CL66mMFEpGklvzZCJaCJwKtfv9g4SLuLVwRhzTjR4vpqe43ufEpZMzrzAl5Jh kjmxmG0A9+3IBl//zZv/XamZgeU5+iyRsMEmeQaSlWJVIFKraZHaCxuVcjlmhkqMdEYTcbJadbdX pO1kvLqk1vPVtMjae50baSopkMw3bJDmjH3ZKEYMFhfb8LC+nkxg9STefeY52PBsvlExjmeXwnai Qq7HzqxjFS1ScyGjUoTAvhrkV5FRPlo7iI8dnw7mr1i/owiZIJP5E91XmUGjmIEkd78dLorRQ7Jb jXNs9XzOJICxm1JmBl+2TxYMEi6oCknvUik2GyK1mhahi1RTyKgUI8oMUqOHKFmvuVs7iKs6x6eD GXn9rp79e229zwqHPbaaP7jIokkRCO/9jJw2HVDNz81njNJsOury9xdoSjASKg0RYoQ9QqvnALUl CQfCdw6ImQ9qmQaR9YmbKeMVSD7En4PJQVw+as7fsYR/rylI92m95MwEGMP9YcCQZqIYQ0aqr4YE XpDcsc+UKZ7x5XYjZr1cJMnAhMwhWTEjpDZJqdUUZFJ17IWHSrGIP5N9t6dURIwsTtzNHSRMxKmD yUQdP4oWaHRfcUEJhqMEpL01IlyIJQKraE2In2e7WyHbUO5La9gioYIqkKzNcehDKX6NJ0WL1Fyo qBQR0BWaEyoKFSiaJjvmDPuYq08Hkom6ffQscIy9LVS1eHImBvxcdO1lohjTKAtHZYLBRnKAq/bj G/2V7wb7ST4XKn4hemFSwekDpVbTIrE379RizMo7kY0iovypcpmPuYMkK+JUMQjbYI6iBRrdV1lB kYj+87rySfaZaVAoGYyf5zfHaOWCAXOSgOs94hy3qCbuQcIFyzZ2t2yIqEsuOHug1GpaJPaGw8VY rqHQffIimgiTLivcx16Ree54dTDJqOeraZHG954NL2qeqQLSHkzwHVmMGPge47CBT0n/30r5UzHR beQiYeMPm3VhI1IM8yK1moLMOzn2fG6liIG6KxBa2dFknI/2DpLUiFcHu3m+mjaWxvcVG5yKz5CT z5xlHYJySvcl5hf9xrKUuUG9QDNBKdlgIa+Ri4QNzqxklY9dO6UYRD9pWgTvznOLEYPfFpGNaDLO R3sHCRvx6mCyUc9XU5CwH3vv+pK5HzTrhnwr+LDBBaFitOnUBs7ryM16yYRfSskGt5EqwwhDRO8Y ZTMZuDDZ4NTBg6ZFGGGc5xZzmbcDiWgiTqbqd3tFprVhcD1eXVIbzWpaZO29ZiNDTk4jMcYhAqS5 E61vSjHYoDiQOBnG8iGPxKmUbPD9tkYOMt45DZ9NQxiKFAPpSK2mRWpPFhcjBn4H2D+ajPPR3kF8 7nh1MHOjnq+mRdbeazYyzrB4kp9DNiivPWyIkV0soFogp118TiawYx42GA76HJsZIsmNnumuDDsB kYkeolyk1nxqMdbqHaHKhXqIcUa6Mh9bB3HIeTw6mEw0ktWzyFp7zcTJi18cshgi+IJv0yIQ/nO6 cXhgrTk58KmQvv2dp6ABoElB0Ui63woxgByh1XMAHFNNaIgM3nOh/bCgmhl1Mnm/2TLk8xRjynHo YNIQn1fPAWILD1+TcNKBSgx9Z9roxRLpRorh9W+u3JZ8jrZk2Mnepa4xG8pzrNmLJB2YaafrrZR5 MNpXUxBSrfbCRKVYovddMiGiiShZHrrbO4jPHa8OJhf1fDUtsvbes0EtRtaU57Ro2BCDIWJgrjF2 vb7bN2EOb0qGF2pfSMigy0jPWykaiCdFi+Ccx05DRqUIgXdwyIgmw1TqsneQkDFOHUgu6vcqWiTm vswMijIMk+UVbiZoZhTjHeZTqsMFXziw1IJ3WCmd45L70DNIyKAsJB1vpRxljdRqCjKZMfZCRqVY rmexdMiIpu8UeD7aO0jIiFcHk416vpoWaXzv3xMTkfNpw8Zn73kJNpnBBQ1WuzCgzc1xsKGUbHBp fSMXCRvskmVFsFKMvyMFG9G0CN6d5xYjBu5GGjaiyTgf7R0kbMSrg/FRvIkGz1fTc3xfsTEfmB42 Phho+p7Qj/nRad8TVi6tf2JA62ewJzeUko3PyNjig+ids7bZNCw2g8gnTU7eRzdsnOcWI4YskR5N xvlor8i01serS2o9X02LrL3XrUY6Ekd3/LSy8en+Z9goxtqSn6EyA6m2SCbMJ65kg8WEDDlExjtn Ui1ArBSDyEitpkVqTxYXYzWfdQDYP5qIc3ZAL3sH8bnj1cH8Hev5alpk7b1mIz2KJZT0XsPGJ2Pr sCHG20MMqQtjCOdV5ubGh1J6xxn8DDkGSW5w9CPdb6Vc9H3UFOSy53OLEQNXWoK4C6cm47zbo7st Ejbi1cFko9GspkVgQ037przcOLeKkgUj2fhJbXLZKMaokA8Rz6/AKbZkwnek+PW4nyHPsdzvc2Hj x2fLEAczTm5iU/tqWiT2JqcORt/AV1LCRjQR52+1X/YOEjbiVTG8OtFcmg5y4ls2Xm4WW0jJPHHY +KAJTG4UIwa+cTdscHtAMoH8sJdh4pbnCEUkbOS6m0A4RyZFaBUtAhnnsYMZwo+ObEeRUd6tTWYM Ei7i0yV1YrkUHUQuYq7vyT8ywzM5UsFVmNkvs6BSyLnTlAIxoGWpF8RZWAekfJZtutsBQgRVIhl+ VoiW9FHPAWrLpwrp/TdeKN+QqCHCXzdb88oMEBri0CVUn6sHxQNMRzlxfZUQtE0Zfv74nO/Eq6AY NPzk66k2UrMWPTQoZdfyPc9BjEh4oEAko7DBjJFjCGpfTYvAxHnuYITARTZ5PaLJKB/tFZnX6nh1 SR3PL00H8VeOvddN53Qk32ljrHSenLB8ndfjwojBj5bLxhlLsUkXMgi1gYO0H2GOnr63UsxYI7WK FsE5NSUtImUI1subF9FkmI/2DpLEGKcOdPldRRtJMiPmXnMx3QhxMyHM8JPzUYeLYiy18OHQyYz5 SpkdC/cPhIz5JtoiyQwqRex6V4plv2hfTYuw+62mkBEpYmBYZzPp7E9NxMnW0d1ekWZGvFqpjWY1 LUJm1N5rNk5mULRhBEhzN2gzoxgx8JWzefX4rU0E4lQqU6R5Lsh459StVYiVmkMHal9Ni9Re2CAj jB026D5g/2gyzkd7B/E5hpbj1cHMjXi+mhY58b1tNZBmohw2vnPnUNkQI7tmUTJsfPdLu7LBomSG 57/yXBG9c+42W40rNQcPHjQFGfbHXtjg6EXY+PWrq4vRRJw/7/amIxkkbMSrldpoYCOaFoGN2nvT kSDNVb3pSbgB7bBRDDZY15pfgXdOzogTKX49diLyXJDJeCdvdIpHiucYVEdqNS1idQCajOpgxuBI Rl6jyTj16rJ3kLARr4r9x9yI56tpEdlQ01e5wYdMMhhnWjGXWaChGMv3KZRiSMv32idOpOMd55p8 jh0yEb1z8jZbjsGmMWTeq9RqYpgdhPKA89zBjIHrMG6aiJMLo8J+7BUZe8erg8lGPF9Ni8DGxPcV G+RmBuPfbLjSoxQjBsqF8I4h7R9e6IldKX+rv3mOyEXCxqdf8rtJeRJhtK+mINPGj73kRqSIgeO7 YSOajPPR3kGSG/HqYLIRz1fTIrIRP1+3oqdP+cPRc7xjw/4vL0PYEJtflOt7w8YvRkQT5y+lMkka GWbgImGDF2soqxBbCU+KFqm5kBEpQ/B4oUk2NxUSJRftmBlOyTpVEWlmxKdKTW8av1cRXDSSmOO9 fM8Fl6H6ayLtAmK4KMaWH6d1JjN+u7Xub85ANFwwtU4WDBIuKC9xnLVSDCYjtZqChHs1hYxIEQN1 FyEjmozz0d5BfI4OaLw6mJkRz1fTIie+92yQPnNgSD05i/x9oOSFXyeccSlFV4lSIX+pfOSIsAVC BcUlzYsIsY0wo3O3UUfPAWorRIyM3vuNQ5NCNTPO4OuFGemMLUM+thyi6dDBWLM2jOmho+cAkpCn /uc///1f/+e//t+r2RnCUzlnSnB6d84gXxib5B7kkgar6fytKa+LZyRLfn2R0kBSpe8dzB6CTbWM xatokZhrPxIpmeDSl2EimgiSa5bCeuwdJBkxThWiGzmxrKJFTnRfJQTVHQ6lcI4jnc5U0VAMLvxi nWk3Vy8MGX7zyJTg8oUMwbiwPWS4pTXbjlCmLsmgqVHqaLqQ2OvrESlioGROMo4m44ymY6+Iz61X B7s8X01XLI3vKzao8PDntJ6DoWbYKGZJh7VD2KUwL7nPvC1s8InGpMEg/lRM3f7OtiOr35GaowhK HU2wwXEFF89rL++I2CQ370fYiCbi5O7Lu72DhI14dTDZiOeyMZoWuey9azpdWpmPUeFdPuSpTrHE wMxg2GABr7ErZW78jgyRi4QNGEnXO5ivPgt4d00XUns+d6SIgXFW2Igm43y0d5CwEa8Odnm+mhpL 2I+f79n49X2mE3hH/WZzoxi/KM1VhnXfOhHhyHXY4CKcNCSDhI2fH7PtCEORYhgZqaOJyIvUXtgI Zgxed2WWRZNxPtorYnNDxnIy2kH8kSL2eL6aFjE31PSvNyUTNbdWZjrBWMWDBHlTilHm4lU22mXk ldg5A5hWlfFLhp8i0xi6v9VyxkrNaYUMP6MJNpAWqb2wEYwYvvtBPdmIpu/WmDzYO0jYiFcHMzfi +Wpa5LL3Pjc4FOJQioD9rHLYECO7KJ34a40ybHBhUmLn5GzY4ABHhmCDGJX3VE/VI/twkfKYwoMm 2BC57IWNYMZgwfGliThZHrrsMVErEjbi1cFkI9GspkVOfJsbr1b8kP5LFV2845uzZaOYZS7pUjjX 1dAVssNl+pwx2IdIyMj1zXARIc8rjO7qgYoCUH+eCkaYTMn56shQMXqM8W6MN6JImIhLB5OJeF1F 098+xrZMvBxp4CTvUpj4kfs9vgdLXtjZkRZ/3OWyc+UbHWGCA5EZgQ0SJqhGyeijUp5aGCk0Unhv W1kAKlTkYwczAm6kHyqiyCgfzRVpixGnVqp+T78UTYuYFbH3+h1Ji0Gj0IEGe2NJigNBMxMBmeA7 1Y1bGXMiX7+EGoEUSFCIkpHHYMY4ZxuYG42a/Ru35qGh4Rx/YBMuLKglb8KDKQPuU+vPwdh+SxCy OXoOIAU8BXWvKci8zALMmT84+WeQaTOxmLlmK8HYtvNzlkE95OLkurMOkflZGReRupMzg0kC5Uvm R/VcQI2lnjlC9ANcoi0PRxEh8oaE8pg7iG9GfTrQ8XpWfKLoiuOY+4KJOdtog0kxCEzwajFx8Lzj RGCmYnXOO/picKdJfGMFML/9IGa4E4UWNFZqzi04PoOKpjwHNXys1vI+BZOKn4y3LkVG+WiuyLwY x6mVWr9X0yKXvddZcV4Mik78LfGOIrtmRTHaYney4GIuax0uFDItuNK1KQAyPSkTtZ+z2whjkWIU Gqkqclj6YC1cBDOC711vjiKjfDR3kKRFnDqYv2H8Xk2LnOi+ekMYiDkotNaJwXbeELH5Oe3WoIJJ XwKnwJlXBir4rLtPse0qkrTgaqd0tpWaQwtKQUUGCxxruNkKE8Hw/49F92ZF9BAjPcfdWpFmRXxa KeJOJKtpkRPbv5g4WfGbbYgw4YsdJooxXqGNkAoq0hq4QjahXDWYERcXKeaQOnMnDt+EnpEiCahI U6qKLgDeFQkXETICGhtaoqPIKB/NHSRZEacOZlbE79W0iFzE3us35LSb/BgZ/HGP3uFCbLKCGhiq 6yCDH8ig2BdVSjLYL/C5IskLrpPLOlikiJ3jDHdNRXiOBXw1hQ2lZgjNh26TGdFEnHwM97I3ncog YSNeXVLsSz1qgo3H+DYzXg6vCIbTSvHOZZBkRjFI4I73YeM32xiJ3TUf2aD+zudQ4HNhg0NR9rjF ZggdqaMJNvhM42VvGpqD8Yv+ZrF63pNoMk61X/YOEjbiVbFp9eP5aoKNx/iWjZdDLFu6mTz4Y3/7 OwsYi5kSVl6SG9xnHja4oTbesWmSPBgkbOSDGf4eqJIMircUOoouJOaaGpEihF/MfIeMUUSUP+/W JjMGCRfx6ZI6fq+i7wchE+e80nLxMjMYJVHXgbtI832yyYzFaCv4TApcMFzMRXGEidRkxmeeCzJB McKhYZ3MUIqgGBhF6mi6kNhrZkTKGL53NBtNxjmayMTYO4hsHK+K4dXxfDUtcuJbNv6RGflqH00i n07zLaEoc1ZpCGBuwsEoXxlPEniNhznrYpavg3+bE84tZ68xkDTMIVJSIkoCDOnmis+MyDQVn15j eCkhuDkUeuzk7+lBjisrUmeJPkrW+4kHO69bzPQeVlzObAHhNhEL4bjLqVJOCVZe/rl/Ba84PJHf HGB+TiZlv2ZzkfB9q+kDKAzPODNqLiS2JOAI6bzLxDIQPYbX1iG2CiQF4s/KHJerZn5B46LVP7Ze k5Buw2XImSkg3N3ThVzHml/clUnHisTnggoczAzxAEkBykzsZStDeJxbyEg1Wi5kTA1zR0jn+eDh ZEH0GN+DrQLhAHdmWhqRODwD7WghDR6i+lceHArmNipnwn8p+chLIDQdJytxJj0D71wEx5jFuhA4 oF4rr4FAOKCsK53ryBAx9f3KwCRqZnQQpLZ86gjBAQf+w0H0EDLreTdbBcKB/kxjGRn2HcbWqoGE h7i+IoGaDQaQOjaVaTwuEueJfzjgs/f+pI7b+NsZz0jM32HgFyesKzHhckDhroNog8RO4i9C/E5k fQ+ixXAvO4l1qt/4NeLJkTi+GvzoWO+PndfvwGkI/lrhP7HzHicBAvHLzTUyJgDX0vX3RsjoP+Yp 2WA+kPB/W5s/BClDtBxLyBgyahbRL0RCQIR0ni/6TUMQPYb3YKsAT60/K0NVR6KIGkSC5DcdW69J yFtA8ZTfvpMEz75IwoEkgfW2MUqOpRyCr7cAOOWaGQS/CEBI4Prd9BcjQ8js3pkUR80isdWWIEKQ wMfVbUSdu6mHAD3jf9kqEBL05yYTlwn5qNkgGtf7t8DqSGcBCrOxGRKEpilg4TUksG6a19wSTB37 OU+RCQIhgW86ptsYGULmk9qmBiSgZt6NILXlU0dI77lia0iIHmN+sFUgJOjPvB6R2SiOGkh4iOsr EqivZY1Rxzg/XhbExn0WXGVhqnfzqn8qJQ0UlfSHF9E3mnpuO/OFiBAhUnJ7V7TIMRciIiURXOJ8 FOWHRjcjn2NNIo6149PBTOH4fSlaJNGRfO/fih/erCIXlGMeLoLxU7KoEi4485tl7J9K6R03P+X3 HyRcUB2SkUOkCJ0bH5Q6mooM92MvZESKGLhOLGREE3Gydnu3dxDtMS4erw4mG41mNS3S+L5ig7KI OMc5liZGPnRMBOT1WP3IbW/0mArJhddpmwMDhArKQDKiiBBvey4AOXoOIPGqkYjKyMNn5yeqmaxg uTVZMbYMuU9BQxw6mDSMz5eeA6ytdzvjVko66NczamzLQjB/SOtFZhida96IGqlJWIpr+/P7XHjg f6ZTUWpiREGGjtFE1EFqrx1NMaggFbDnupiajHI0QXvsHSQpQeG76RwpvNpoVtMije+rlKD8wp+A Tps1hplCWD450wE23jNYYjxGB2QK8IlKc8KZTGcMg4QMKkCSKGLTD3L8wAHHUQQZQWquaVGM0SRF wkNGNH1nwXc0rb0iQ36dOpBcNJZVtEiiY239fWNBDYa/AdJOGtJ9BDMEb5M2M5huJw8AzAymNfOc 9PhcyKAKJJ3MYGY/pw8k8WhapPaaGZEiuflORciIJuNU+2XvIMmMeFVsMqPRrCYLQe/xfZUZlGHY N1Pd9JtbWsJGMGJgwTFscJl8xpS8pvGOevmkwSBhgwP96WzEJjU4N6HU0QQbQWqvqVGMydasizp7 UxFhspx0maNRKxIy4tTBTI0GcxQtkOiw1sx4uehgSaQzA32jwZvMOJgRcO3nZAaXEOc1+UBqMuPH PEdmiDQz/nBF07xMg814OVJHE1xwjuKy18wo5sRTc87iVGSUKjcxYu4g4SJOFcOp4zgpFkULNLrl 4uWSgwWRzhHggvPFWXQ4GFNWbkJwsMfYNTfA8XJ4rEXn6DWTA4MkL1ixypBDLEPnSB1NcEFPetlr XhRjFYL/l9mcioiSKxwuc5MXg4SLOHVJHcdpUKNogUT3ZV7wQbWMN3/SOjYvgsnFFIySfP+fsDNH dGTZmbP/VvEv4cx92tUOtAY5crV/R18MQBb52KR1+8RFYQiickRmMcBJW/COlPPi28+RF0KcF/z4 upVVjawxj6AjNZrg4vdqr3lRjBDaX0eRopRyUR9zg4iLcaqY86LBjCIVgF6je5UXVHnpJ1CFE5WZ ai9UNSkMLvzRLJml8itZwA2wcY7q3eSAETlHN8zhWFEhCN+05HCraJCa66sVKX7NX81ERw8xsmh5 jDkrjISJuHSkxu3Vs0Bie5kVHB1XL430FwPMMBFMTLhwFibwL+0AdSvJCg6UJwOENCuo8UhvY0xU QKekVtMgtaesWCmF8N0hZjQpTGlXWsTeICEjXhVzWjSa1aTyz2t8r/LiR6VBco4duKaFIUXgax3h gn375gBC9o2PuycDBCQpKP5IV8PibrIi96ytHh1GOLaaE4G6squciBpFeLGVnBhb49ARUqWnVK+e ARLXSYm7RjNLFNRHakIoFvjqy6REMGigK1MrpqPynU/8lZSaUVaTM5wU4p9WMyfvDTJCjJTPUEhq NQ0y9pISnpQqBL4zkyFtNBEl669Xe4MkJeLVYHqtG81qWqTxvUoJCi8y6GQrMkt2KqP0lIIYdF+c 2NA3SjVCYOcz/StHpTOcNJKkYIR6I8VIOlJR5MYjiMiQppBRjBD4HjrmNOWTJoV5a6+Iya9TRyh+ OzGqaJE193jQOZlBPUgGnbR5w0Uw/Z5ajxMXfMa9eeD7cD0t8tQDekCaGVSA6K0JxnPsQlj70TSI yJCmkBEpZQYf84UyzfmkSXHe2hskmRGvBlNmNJrVtMjae8xGVrFUCukiPm1JMMBU0xmMRQhtXOic FmzgXBKBC5OTGjqK4ciNJDXovdSLsDEWKRTowMZFUxC/l7YXNiTlX/SHTWXsQac14QrLk8ee2wwj YSNeHamNRvVu1rRI49v35OGwk/JIfRVSrcYbHy8oG8Fg49vFohrP6UuRivMPUvx6mhl58hHEv7Fm bLg+Us4NzhgoW1aT2tGLvXYkxbQdSuOkbiOaiJOlvrARe4OIjfGqGF5tNKtpkca3bNy1oZMbnwyY wwZlde5JVDLpiQUxMNSwXVZRkwmudBMbP5GBDZXjyTtN2XRPjTiTLr0pnDpQtqymQcT+PLcYMahG 72hSnNE09orYHsNKezWY2Gg0q2mR2IP9528KH5rI0JNv+U5uBFMMLCyZDS4e0QIecUpKbLDH1TwQ Ejb49KuGFCulmYilVlMQMh7v5rmRot3g68FhI5oU3q29QZIb8WowsdFoVtMia+85G1yZoTVPluRU EpB2Q5h/UT5sETa4lzgtArNXtascl4sMrb+QsPHJzqk5qxQz9jtNg4w9PbcYe8DcWQGi+Zg0ER4r f1d7g4SNeDWY2Ijn8FpNQU58r3KDw4cagMIGJ5PKRjByg9VuSil5QznImEz4kZTYoFwkeWAkbPjw CqlRIcbmEYoiBy4kzklRyIiUQqC0wmTkFAxc3FpT3PNYfApjkWKUllhW0SKN7hUXXAoY37QNkcQA Sl5wEQyuMaD949kFQWavQrOjJoqAEEEc2pZdIbYCbvUUEO3z1EB4rxGuckJq+G2IGiF+h9oqYFvj 0GBKCfssOqNngLX1bM2TFYK/mkXgGRftDAvBSAhv2YqGH76l6maAY7RxjUOFGXYIKQ9UgqTzNabE 57yEpFbTILXXwUmkCIGmJVREE8Gx8JdhTuwNktcjXhVzR9JoVtN9fM9TgsVTVW2YDUY37kgGUwyu IYaNX28Ga7jd0TlvSAMXkqSgEiR9r6RwTtlhqVU0iMiY5xYjBMrMnBfRpDCjSRNEuxmkiWGnKsRj 47dYjaJFxtzTNU92Xz40mZBz/Ah+PwYTF6kAZhD33mkIX1ZIZnD3XQPXcyGDqpF0vZLCO/qTSK2m QWpvhq3xgRi4/d5kRJPijCZNyOxnkWRGvKoU5I/nYiOaFml8rzKD6gwN/djJ0ylatRbaxPYMg8ml 64kxx2gwiaBTrcho4tZZiBF5x9SNTtFvk8++ig3OHIif1TRI7ZnFxTCsE7lHkwzLq2OviJ8brwY7 niuEaJpYJr5XbPieC7PBgnxzI5hiUDBqmfjQVmstJCXvfNOGWkrW4pIb2syiIV1MbHDKIMPPagri tprmOM+NlGLgxTIb0aQ4b+0NIvbHq8HERqNZTYvwa8fesyEGLS8l0/FOH71MbgjLL6oPdsoue+V5 B7i7Imxw33MjF5LcoCQrm9KV0krXraZB9FvNc4sRAyMKsxFNivPW3iBhI14NJjbiuXiNpvv4XuUG X4TX8JPqgN9eYsHYzRi5QU0q5ceYYykw7wCf9lJXq6lbpyFGwsan6pSUG5VieH6naZCxp+cW4yXn a9JmI5povDl3d7VXxO3NeHWkxnPYqKZFGt+/2MgkntmuPgdhNvTdruRGMLHxRUqIja+/nmIQp6Qy Seo0RIi9Y/IGE2EjUgwjrf1oGkRsSCZsREoxaN9Y7UY0Kc5be4MkN+LVYMqNRrOaFll7z98UupIM P1moGjaEeerN6EH1Nwzj3joRoWUPGxCYyI0kN3L2mFFnhNhGiBALwCgizEEgw+ZCRqSIhR4sqWFF RPl7a20Qzo6sT4OJi8ayihZZc4+5mMzgZionBgfehgpD/JpvLivGKldTJf99ZZnygt14p4CApgXl JhmKVYiRZM/HVU8B8SBb4cEyooF7hfyGSA2tGDxIaG0NkJSIQ4OJhoSxegaILTx8TEJm7mw0+Yuk eEZ1V1hYTFV7LGSJfKojMu/i7Y9rn/mSKVMrIUkISkvS+VZK1Wv6lCkz02oK4qba9sJEpdhG5SNw piKaiNKLXaIi9gYJF/FqMC5/nmhW0yKN7xUbvJ86loX0d2+GYo/Wn9FSDBybMhvakFKbyH2kIYNV +wYuJGRQWpK+t1KUud0pWqTmQkalCOHHI2+mftakMG/tDRIy7NRA4qJ+r6JFYu5lZrAfnOEnZ3xz cwWVDMHgQvdeKft0vMVTL/aE1JRqgqTnRI8QOad5VIsVK6Uad2tfTUHC/Ty3UsSAlosmxXlrbxDZ G68GExv1fDUtAhux9/w9obxDBTqqfuMIn7oR/qkDXsru31+W+WSXhbwkAp/KCxs/loENI2GDPZqs /FTKZw5uNC1Se3puMc+pwkY0ESe/wtXeIGEjXg128Xw1JZYT36v3xItYZoNy4LJRDDY4q0KTzfiF rxI7E/iCTLyDpww6jIQN2pe8TpXyqQOkxGs0LYKb0qTnFiOGnw71o0lx3tor4tZ6vDpSNJT2Ss1c NC2y9h7nRjoSel7dg2o2+CZB2ShGDC6/hg2qFDP85M7vsOH7U/WmCLF3TKYILZxFyucQpH01LYJ3 kgkbnL2QFDGwegn7o0lx3tobJLkRrwZTbtTz1bTI2nvMxvQofzg+bDbYox42hJFdVAzQaftN4TNy mWL8lZT6GL5YpchZyBWS3OBLU1knrJSPJdxoCnLshQ1OWsgH+m9uWEK7poHSRJxsuB97+FQkbMSr wcRGo1lNi9CAxM+y8XCfQDN9FhyVG/w+w0Yx2FARlt4Uqpr0DngRO8Nzrt5p5HoubPA1VA3CIsVz OqFg7atpkdhzTi1GDJ8e3DINtCbFKe1iP/YGCRvxqhg5tdGspkUa36t2g7tlMvzkEzi5NYxVsmBU 1vzq0gTY4Nv2yQQ2Q+Pdh2WI3EjYyG0vgXCOMfWdokXo+aRIjy1GCLrsWKlhRUT5dbXmzDASLuLT kVq/V9Eia+5ZZrB4rrM/cEoOJDEGUpGbDnLBBNeyJwvYVQsTbOx4rG4gRFBlkjW+Cul4glSvngFq KzxERku2XHUqGqJGEV5s+ZUxEBri0BEan6MHnw24I3FczxOCvTbWtcPCB9/9Ube6GO7zyWNcY0TL pQFO/F9JKV25FbAxC5Fv2tby1iMZESmfXcjws5oWgYl5bjFC8NcSV5OivLVXpB1JvDpS6/lqWmTt PW4605GoXoHWUTnB1WTDRjEqwDS20EvIlwlNBoUzIYOXWaMO9qRB7By7WjrfMBiP+ciClK+iRWou acFOoqQIQYMU0RpN/IPhztXeIHquTg2kX7F+RxEiQZwZju5VZrDWWy6+mYQkM4oRAtuaOEeroyJd hykpZcYfPycyhCQzKDRR11tMyc6RjJIRTYtAxjy3GHs8OtsrMqJJcd7aK2Lyx6sjRcla7K2mRdbe 88yg0EODcV4F1l/LRjFUpTiX/lnXzihOf/pSbHDYImngj0+Gjc9vL5+vlM8wSPtqWqT29NxixOAb PxyeNPGPD3l17A2S1IhXgykT6vlqWmTtPWYjQwwaBo3vzIaOByY3hJFdxJDqXMb1jB7DhqTkHRf9 JQ8o2Jzc+NKYz5xFyscYbjQFOfbCBscf5ANNER+wcW5Ek+K82nNHYiRsxKsjtdGspkVgI34+70io UdKvgDSTnLJRDDa4whnvGNLqpj/HiZRb1Tc/p2zRc/KO8cBvGxdj+tU52JDBeDUtEnvO+MWIQd/3 OpoUp7SrFY+9QcJGvCqGV6optb3VtEjje9Vu/FKtZDa47D83yjFyDkYTpO8Wig0+d5/h5zdS9u7d MrBhJGxwmj3drzC8Y1B9p2kR+mpp0nOLEcNbh/rRRJwsBYf92Cvi58arwcRGPV9Ni8Qe7D95UwiG T8ZrIIy07r/kTTkYbHz71gaGtFRYeFRBwVUGHpzRTSdjJGz8cquHM6hSDCOjfTUFIeNrL2xUSou7 HnqzwGVNivPW3iB6brwa7OL5aDqxrL3nbPC6qhQBaS7TKxvCYEjdExtpssvxwcYuKWUu93s1D4SE Da7XC2UR0oGGW0WL1FzIqBQhaEdOL4rv6VOUKPJ7GWtFmhnxaTBxUb9H0UFi7mVmUE6moTijbXZ7 ykUxCsIoYMc5BnPMKx3ml6TExbefw5wRBaWpm/cgi+kN4PuZ0r6agph72wsZlSIGXRB3NBEn5U5X e4PoufFqMLFRz0fTQRrfq/eEcuA4pzW6vCZzDfeHVxcYytESOv292icqsp9E1HooTFCOkh5YmFuM VKxBRBcppoLN63yhIRAbVCSHWYgWxSfNNYV/BUJC/BlMJMTj0bMAFMTBxy+HB58IU3OiwTLSrNuU g2J47wJ2RZEX4UsycuyPn9LjIO4MtNHGyEtdTaXktnWPngVqTAEtxvNsNsHdKNL/ujU3SIiwTwOJ h3odRW7yhOT3laJXyUDNiQaH/MC66yvZUIw08FtKc0H1i4PkkHWoYPciv70R+UY68hXsqxSveZRb kcMWYN9sLVRUiAj08eWjiDfL94+J+ZgbJFTIKXSNFFzU79F0kEb3igtKPFjGFhc/TD/ChTC/3Cxt 45yGjZ2AUFUVLlivSQoYCRfUS6fJrJSGjTeKFqi1cFEhlij45qi5kCLsEyXm3GTGXBG/iuPUYMqL +j2aDhJ7L5tMKjw05EOaE8Xlohg/pwvLGMmynuEW81NC6k/49GFzQEi4+GHnMXkRKaaaUT6KFog1 v1mLEQHbqeYiihTlrblBkhe+HtiMIfUfcVG/R9NBGt2rvPhLZpoL7rnyvTdoKEb77KTVqhQNmhoC NrnTt374sUHCBe1N+ltJ8XNqVcrKR9EC9D5SlLyoEL0Pa3TmIoqInEHaMef3wUi4iFMrdfweTQdZ e08G4Fq+5aJic+E6Kr0jizGaUOWWXpJffbxBZKiSijA1XevUw7VVIYNyq7SwxkQGpxukfTUtwrhE msJGpYhBX+JWixFNxMmtQGEj9gYJG/GqGJmxno+mg8Tei9ZTu2I+moQ0HzJWoQ4aihFDqq21mcaG jtnw544Jj68cJw3ehYQNLiVMnyvMqcEnSZASG9GkXbGLveZGMQZXfGvebEST4oymsVfEz41Xg108 H00nlsb3/D1RSaZ6TNoMhki+EwkNxVg8dd02KcELo6IT/iEp5ca3nxtEbGgCFcoqxDg6ykcRyVIk 5poaxQjhreNZKyJKLg67WhsEa/UJzyPlzKjfo+hE0uhecfHJJV3hglt4mhnFCOFDh6Gw+87ZOGfG W+7qYVrk5+SSkHBBN5zuV5gzgyM30j6aIKMI3s1zi8EUZWtmNZoUZzSNvSKyt14Npsyo56PpILH3 ojdRIWYaUF2HlLckiCc+oYJheNKfaZxbjNTseFpXHigtyaADESd6zgmMEg3OL3bEnREaRRznDmxT ECUKzkr0cDpY/Z3448qIEGycjRJnxzWeV5nAcCxjTS2LJfhC2OaddxfFfZT8nryUHl/xxuehAMkC 1iQyEomMRjlWHC0OVogHMDKV+CtE06HvSOrNqh62iW5tBQgDcceI463DowVKbqJ6RQGFHBlksqTY d6EQzruSmKzj8+DiiZC9caOZj58KEA6oI9GoZGQYcyJDxKNGo9CLrXJQCO91dE8cVA8B3toyoKbk +FMZQq7Lo+Yg7LO6/OnZnIOCSg1j1Dp+5OgdzxfiLXQtMkY92nKArsdlhJSniLgH7xizUMkZEiyj 1SNrHjWQUCS2kgiF8J6bTnjBVg8B3toKEBLqT2WOy1EDTwkiSYeaV5lA5UaGlbqQJS9DIbx36TAk cE5C43CGEZ2ZM5/Mzy4gmfCHDQyTEBmWVyMzaiChiOrHMnJYiGEFV0s6E6qHdaxbWwFCgvzBVmUI WS7j4Kgp4h9XcT0ngfJJVsgiTG8uEgI5j1l9o/YYEvyFKZOAkMZUn3kKVgBCAhdpaYAQSK0dUyJp pktFDT/yIpDQpxbCex41CdVDgLe2AoQE+YOHlanL2Bo1Bxlbj18Hr2sjTEW16ljV05cDIXH+RzeF wQEfu0+me5inaY4ltExaBtgi0JBDiMPlfMJVBwQUsR2xtggjbU6gOv5q0dKuqZYdx8rfib6eRKLe O/joOEjtPI49yw9cw6CPZjp2rbMkA4rx0+miRL1YXGbf6CWlFODK+/zezALePeLRjMtVjjAQKQbD kRpNRFwE3yQTForBAkdj3OFWEzHe2SsiJtarkSLyej6aDrL2HrMxmcDoWL8hWxa6xyhsCPP7wNpS 2GCalBeene+wwXHl/PZGFJV25kNQhfAgQgyBUeRmr0jN6bGVYtKpy5WOIrZD7qwVoU4TLurTSBF5 YxlFB1lzj7mYzGCVWgMBpD/5WcNFMULQ5FF2eTcapqQy4fJzbKeANDP4GdOhCnOY79E+mgi8SO2F jGLEQJvozKgm4ryzVySZUa9Gitjr+Wg6yNp7zMZkhqoCTQY3YZaMQMwnWQ/JUCz3uxG5hMSFy5v5 r4GkBbUjGoCsEB1j2wnpMTcAJByOSU2IiAze0wkjo2mU1RDhra0CoYFPISp1R4ig4/PoWWBtPSeB JTMN6niL+aR7WRBGPNDAQN4pwS5xRpDfklLjSYVAfn4j4YGykfSskSIBOI5w1XSQ2gsTlWIrmHG0 qagmdkrv7BUJF/VqpSgDjb3RBBm38b3qOynK0C+Jd5Tplo1iWmTQh2d5QX60/KAGU0Ii4z2PFQkZ VIVkVBEpQuc0gpRHkbNCSMiXppBRKUJgfKq0qCKivDNXJFzgk9OiQir4tLXqgYnb2F4xQUmGfkuY eGPmlKZCmAPg21zJCz5u3yyQlKhg6yXziG8hoYLaZzkHPZaCCg5RXDUdpPZCRaUIwVXcDMNRhAME jiJ+ijUXROb4eepUMY1/6ngVHWCsPTs+pq8WaO9RXPz5YhAqLhajT6AixGbZtUwr8IWUnfOepeIW Ui7+8lVgdBVT78BBBGkfTQeJPTe3K8XCbhfkqojSCikXFzVXJFzUqWA4tY5X0QEa3fO8oGLy3dMD pHsbMxqKMVrgYDKuMODMRW9EqbuU5Ryn5fLSGNHvqymYawsj5d6T1W9JjSa4KBJ7fUWKEYJ3eVYR UV7N+X0wIi7WqZVax6voRLLW2nY+qkhEmsWQ9CAc+5q8KEYEroDFLD94GgLuxEle0KomB4Q4JmZi 1E47L4wpLziNIKnRdBDG2NIkDleK3RBRD6lVpGJ3lIv6misSLupUMOdFHa+i/4ruVV74igi9Iz/c EuVZONWSukjCETBH9k/A56SSBXyWIs7xBbYMNowkLygVS28jzEMEvgR91QQpRWKv71YxQshmSvUQ 5NUaaVEkVNSnlVq/owcmbmLD1uMeNYMslUV60oBnqktMa1FM9WGqoiUrWIpuDkhKPxPf7WsGgPjX ZVamM2P8qOw1IORcsNAoOkjNiUBljKUIQN970PPWQ4h3xoqEiLq0UhuK9eBjAPcKju1fTGSARa0k Ncq4z8RYR8/DRDEC+PmhuhomGAYmAz4kJSb+5jn2OIUkJygXSU8TKYL8iNRoOkjthYpKsfGp23LF RTVxLcKdvSBi/ni1UhvNaIKN2/j+xcbkxR+HiXN600NGICLQqy6rnDbPD85hKXGhCVHJAWhSsJSY jiZCBJ5bMkbPArUVIiKD99zkbB6kxu8CNq+2CFlAaKhDxZQC8Xn0LBBbePj45ZiUYJtSPyPS+mZs WCiG+3xO3jSwmq67FkgASYkGajD78wtRRNoT4xTRkSJsjh9IajQVccLaXpioFCHolN1FE1He2SsS LurVSulFsb3RBBm38T1ng46Ic6n2jg/YZK0mGDHQsuswsZKCcw8Ok2ukM9IATAIYCRkUgKTULlKE zoxUUqPoICzgSVPIqBS7nhzcNxnVxC99Z69IyIhTK7R+j6KPRWLuRWZoLMuyizLDn79TZiymXo7W UlzwOXtNV9TNdWrCR5+SBUZCBmUg6WYiRegsX0hqNBUJ99IUMipFDPp0nzKjmojzzl6RkFGvVmo9 H00nlsb3KjOoxNDPibRW1sKGsMTAIgNs6OI0j/FhQ1J6T74jU0TeMVVjZzucWYrYOTZx1XSQ2gsb lSIGiv6vmojzzl4R2VuvVoq5auytpiAnvldsfFJZ79wgDYaNYowQP78pr8YuZfLJBC7cCxuqyOc5 BohCwobfj8XUW/KiSWo0HYTl/j53MLbEqdsyG9VEP3BnL4g7kvVqpVjair3VtMjae9yGpiNhifS3 B5SoDBs2ihEDFcX+FRhbZ9jJh4HCBjdnpSERYu+YPzGNCEOWElWRGk0HwTvJKDdWihg4OAf7q4k4 7+wVSW7Uq5XaaFbTImvvMRvpUdg60FxUb8qHZlZ5U4SRXZ41hY1Pna1Vz+Bpk94UPnabIaWR5Abf YtWAZaSI8/1WU5FjL2xICowtcZxwblQTIyZ5NfbwqUjYqFcrtdGspkUa374pD6ckPKbiGbFBNf+w UYwtGW7ac26wz5+pxS9SHoRxvFrP0TToubDhz5cOpB+dk5MSGkUHqTkdL18ptkB/O7KNIsbR/toZ rXitFQkX9WmlEguMraL76JaLR1ep0eL+sroaLrgHUZ8mPBghcCGQuaBuNnlAfVn6GE6sJwuMhAuK QjKyEOYwvyI1mgi8CGRIE88djBhomZ0Z1aRZRQfAtRfE3Ov1sFfFLp6vprv4Xvav2vlxYnDHd1+T QETgmlFaUE61NQsQcl740L5yQE+FCipCNK4opiB9cZjSIrWnA0BEn4pMfkvKYM1D1RChhNS3x1aB JEUdWiF6EIcBCyqnFS83cT1PCQ1aPYtg457r0sxCMDcW3xRLOiW4SjEJQMlLXOPUS35+I+GBYhB1 N4wtLEWMHJzIBnk0FTHrtpeUkBQYb7a+MpGZnzWxoHC158bCSLioVyu10aymRRrfsvGwsWBW98cT CZVyMKBQ07mYVisZI6hb1RK2wuRLPSHji8dwrkjIoEBEvc1IETqnDzL8jKKD1FzIkJTzgs8JqZdi 8hdNLM5c7aG7SMiIUyu0ftN7oIjEWGTNPetGWAH69WQCaZVihgthzgxac3HBkJa1xYTpgk2SlhN+ yYIfIXJOczZvLELGlHVy+OCqCTKEoKn2QkYxYuC2oKsmrVShUi+J7DkzjMjeerVSG81qWmTtlY1/ ZAYVGvo5qUmhmKhsFGNRXyML2eVz9mkJuCZSQzBW2/McMzE9FzY++IadOTPmAWmkVpOrcI69Dk0o wxHG6IgrEsxGNTGmkvaLvSJho14FcybU89XEXslNfPuePOxI2HlkLS9sUJnjta3FiIEFTbPBdweS Cd+q35F3v3kONoSEDTIzr5MwvPPxgww/q2kR1UP1ucWIQUd6lWXVRJzRtPaC2B7i8aqY3otGs5oW ib0XHQkb8pxKCBsKL29KMdhwzSS5wRKCJjDELimx4YsnFgkbXhhfTGywnC7tqykImvBOmvTcShGD 6BQb1UScd/aKJDfq1UptNKtpkbX3vN3486V6Q7yjWGLYEOYYPlwzCRt8zj7DT9YxxYYmSW1WjYQN LiXptnSkfA7hRtMitRc2OK0gKWZV+sCn2KgmzNzZKxI26tVKofROU5AT36s3hSo7DSB4TCcjkxvF tCmq1UjYoNYyLQKFqmGD00sZeBgJG4xW0wtXSqcRLLWaFqm9sFEpzTG7oV1NxHlnL4jbm/VqpUiA 2FtNi6y9x7mRiRpNrT4kYzY4hVA2iql2itmq2OB73WkRoCdscAdfIwexd2xqdf5SIR1HsNAqWgTn pChkVIoQKAmH/FFElHfWiiQz6tNKbSxRhN9Bkhkx95iLTNNUKRkqfn65plaJMRBVAn8Y/ogJOpvk P2c9woR2PNU8GEhaUFmSLrhCOowgodUzAG8yatqZREZdlgebmv5ZjfYGb2wVCA1yCNXBYG98Xj0D xBacPyeBjX35jjRnFt2R+DSJMJU16SACNHB3U6NGCoTpYZ6DByHyTXMobzcGc0fCQYQbTT6acOw1 IypFCLROV01EKa8u9oL4ufWqmLhoNKtpkcb3ig0acg0f6H+4pKo5UYyaGVIQ7xjGsQWkFOCWk5Dx k8eKhAyKStL3VsonKDL8rKIgzleb03MrxRBKZR2itZroTO/sFdFz49QKsVQRp1bRIo3uFReUdWhY h7SWyvJ+CHNmeJlPZnmJ9XsSuqT0S3E9egMXEjIoLEnXWykGpHeaFqk9PbcYMbBgaDKqiTjv7BUJ GfVqpVhEjL3VFOTE94oNCjsyHOQA+rBRTNVUPnijwgbPMGBDUmLjN88VCRtM3NP1VspHKKR9NS0C G9IUNjhokeHn24fueSU1qok47+wFcUuzXq0UWx13mhZZe49bjXQjbOyxduD3hJ2rYaMYRT4uNSU3 +NRUMoGXOGz8zXNsB4LYOzbb/rQyqVI+QiHtq2kRVEpT2ODAgqSYDVKdShuxmhjz39krktyoVyu1 0aymRdbeYzamI/nD91TCxjfTkLwpwsguYmCrlt8KYnUAVG/Ku6TUqr7luSLJDaqF0vVWykcRbjQF OfbCBnVK8kFz405Eqok4r/bwqUjYqFcrRZ3BnaZF9IPi+b4pDydq1K/oUlC1G7zAqXRdjH9wENps +PpSsfGGFL+e+q9OQ4S03fjD9wjNmTG1CEwspH01LRJ7HZpUitePe/j9plQTtWnSfrFXJGzUq2B4 tZ6vpkUa37LxcKLmFzrDT1UhuX9dDGJdfElucLolUwyXBck7Po2g59gdUl1ScoOXPMMKYXin66wt tZoW0ey6zy1G06xiJb0p1aS5uzWtvSC2t14V42pnNXTyajUtEnuw/+xNUfUlO3rKDZbKkhuLEYOr DmGDq5qyVs4davmtuKqnkQsJGxo2maAI6biChVbRIjHX1KiU+inVZEBGFBHlnbUiyQz5hLlgcL9+ r6JFGt0rLjzhoN9m1b95YYTGxyWHGtmxuu0QOQ4VHnoU0X/LL0bDXy5nZG8kMgwj01pUi/4247Jj 5oqwoMtlvVcdNKA3dvK3aVtfVkYzHvSulv6deLDzOA/SfzCTZqkhXqmLUIu5GPHrXlG9eyxa51dn 3JP4KWDKeFKIPWPerGPJSoRK+bSCpFbTIngnTWJuMSJg+ZNfdDUR4529InpuvVqp9Xw1LbL2HrOR /kNnDTmqpbfi/YP6+7AhzP0HA5+wwUXcCVNCaiL+5DECF5J0oLYkAzRjym7OXkg5ZKCIMBepOT23 GCEQIggLd9FEmHf2gjQt4tQKbSyraJE195iLyQyqOxSA1o05sBIuiqmeg2TXb8DaRfKAeZr6E03U /Bz7ayDNDOpLMqqqlM9gSPtqWqT2QgbnMiTFJIePKTozqkkDrlt7RZIZ9WqlVFR6q2mRtfeYjckM 6jsUAdI65BU2hDkzfH5MbLC+nDfAB7bEBjPJRq7nkhrskWi8BEPC9KNzjuFGU5BjL2xIKmzoLJpS o5qIM5pizyMLI2GjXq0UmzF3mhZpfNtqPBxZkOd8rRJPMPfBZQxho5iqBPRWkOYkjsNkHJShBk+o 1hdrQkIG12am96gUGRzlo2iBWgsXEkIVmz+6jkNcVBHUXc3hQJFwUadWKrFcNd1Ht1z8Y1zB10wU AImvFW/tJDJUCMbr4ekBHeGHJwxknYQyXfNjRcIFVdnq2lZKfRx7sHpNoijAsRYuJAQGF6zbmIsq 4qW6MxekTUadWqn1ezUt0uj+xUXaDIaQKnYQFxSC5ZDdYiwAkPXKC1Ydk/58LkqNiKZFfowWCqRN BpUU6XIr5bMMGXBG0QK1Fi5cbiEuPikgdotRRbSsd+aKJC/q1Eqt36tpkbX3rMXQVIJiNHPxQS+q dySYf07mm6oTYVKir/zoJ2eDOGT4c0GLJDE4d5sup1KaJPBdUCVGNQU59sKGpMDoQ6grwZ4mJdZE nFd7aCoSNurVSm00q2kR2JCmzYyHLQaTbq5dt3d8W27YKEZHrgPYHtVym7/Z4Gt26iI1V/NzdOR6 Tt5ptnaEABhER2gVLcLCrR9D1WKEwFm0iyIGIBK6WCsia4yr49NKJRYYW0X30S0XD1sMLViqy8S3 Hw5BqsEYiABSP8QIVt+jVBZ8Ski+cff0AUIEpSUXTDHqpgplRfUYOLb01EAMsqjSB9B8z2qI8NZW AL+P69AKjc+rZ4DG9S8SpqmgpCMssNE9CVGMMS5XF5r8j7+dbnBUOzRwBDUxC7FvmjJ5r5GRb6Tw KVKraRG2WyQjJhZj7K6dZlFRTYzM7+wVSUrUq5VScam9Wk2LrL3njQVrvmGDT8m0rRDEb8kkTNey KhP7JvBz8acmZ80IAUkImod0uZEhQyMTJUS4iPbCd6gaIZzX1a6ioXqYt93aChASvlSFOxLazrY3 q2SRsfScAIo5utevchycUFlmDymxY65aCLpFnW/RS6G7zNV6Uhud8ZaAUEC9SQZcQA6YswsZ01cN DXoQHOtTC/Fy8MEmU1A9BHixlYj71PozMuvyqllkbD0mYd4Jqjj08yFMFWFIKMQPmNoYxk6fnkZA gksNedfzlIG+EFSfpIuNDP15ZFbNIrEl6hbCe0z4dageAry1FSB5UH9GhtKf2IoaFAeBurH1mISM tFVx6bkC7yJxhoRCrIMwxkKj5hbc76hM8N0aZMJHngqQTPjD8NrZEhlNIqx51SxiWx13VAjv+eaC M6F6WL3BIWVdbQUICfIHWyOTKHBw1dzF9a/WMSRQQpktPRzrNnogNOJ9SoMYp+pKVJPgXXRNxjqx oM/q68CqUjpLQfqROZeg1IgavyBFYiuZUEh9Ji+PWoTqIcBbWwFCgvzBw5HZKFbNImPrWSaovtLf m2Jbke04ZUIge/+HSylAtCrnqQgjKYT06zBe1M8eQI4xp6GWfiCHzPBUMuyGRo2W0YzEVkgohPeq 5L/oYQv9YssRCwgJ9Wdk1uVVs8jYekxC2gSf7tEPiDDDhbLAiR9h2mn2CTqGpjo65KAlJRo4hCMZ aABxbjPvp6hrMaQYDt5pWqT2wkSlGGPysVgIXE1EeWeviJ5br1Zqo1lNi6y9x2zMe8ESSAbVH7qI JzkhzC8G+7jKCe3weD5B7L6uBzb4rHMzQIi8Y23wCinH+T6EhBhqocjtXxGcm8dWihD4rshFEVHe WSuiA/vr00qpJNTmVtEia+4xF5MZFBJmSM1Rl+GiGCGkEohBGXUf/s2pDUtmcMF5skBIM4NJtHpW JmeRYlAYqdW0CP+QjDhcjLkYUw1nRjWxG3Bnr0gyo16tlE5Hd2JSTYusvcdsTGZklI4wBexNjEDM Iah88W/wxZdfE6WE9JbsHoaApAVLnVmmYytAQvwa3UaRngFIuNoKEZGBB92MpOyqGiK8tVUgNMgh tx8VUpmo3trVM8Daek4C++IaM5Kkmubl9RCGSmhIyQtDUy/Z8VJISG3oW0SKhAaKQ9TDrpSa7FtF i9RciKgUKxUMRUWEFGGfzTLM8TOsuSBmfXwqpJa+bo+eBcbY02PZKqH0lADpT2YQYaIYLtGD4xtD VL4F7YRgqzy+UT2YX19IM4LmQP0sVAhTRnCiImPKalok9joKqxQhuH7A7YoUEWUUaVZlRUWSFHUq GIRtMKNogUb3ajBBzUiGlfSD797h4OxSMPJUHzYTF3wDrJEjBcJsKjLELSRpQQGuhg/BcI75zZ2m RfBunluMEPxSMfyIIqK8mvPLYCRc1KmVWsdH0QJrrW/Io3WIT5VKenqghRDaRfLiYHpnXWfMWE0r BIqSW43UE2ju5eeClIu/9DMr5bzgNIV+4NXE4sjVXvOiGAubXsLWoNyKtLve8W3NFQkXdSrYfy6O V9GJpNE9zQuk2WXTb4n0Jxd2h4tiRKDz90o+DkIm+xk9zzTMz8GFEDmnmZh35YI5LzhhkYFmNTGG vNrzc4sRgiutVhFRXs05L4zI3Do1UieYKjpAosPa87zgo8f6LZHWpybDRTHWJ1MAxMhNX/ZxXiDl vPjJc8St58IF3y9Ol2tM7wjnEDLerKZFYq95USlC8DhGqzhWRJRSrjSsuSLhok4Fc17U8SqCi9vo lotH61NIc+eYOiF22nnxmxfF4CL1Pxy/5aoqc/GNlJ1jNJohp5Fw4VGrGBMGY95ukdRqWiT2zOFi 7Fl47K1Jl3fdqMe5mnNeGAkXdWqkTjBVdIC19jwvGGDrt1TVAUPt5EUxIkj1D4NY3YXjKJFyXnBj TnKAOtrJixl3GRIVXGAjoVW0SM1p5LgYEXinQ3sBGsUQo1QrK2qsSJioSyNF4HIbvqqnwIntVVb8 /HrywFSJ9kvLlmgoxlj6hwTXi0mTlrg5CBLn+DpdBl1GkhVUhaTfFeYg+bqppFYT/7ixp+cWY/dG t5BmtmdNFKhFkyaE1hTE2bReFbt4PppOLImPN/LJSAvpLM+q5IB4kxaBcFLbMeKC8VYj1+ek1MdO kQhHcycpKBxJt2tMQerKFjMRPQPUVojQLTL6KT8g0DxUjQr1fLBubBVITtShEdowose/6E1cr0hg XKSfUYsKjIzDgjBcI211TaJoYJ6EkFYnPHzWVKiTDSPJCIpB0uNWikFapKLIOVKk5kJEMW0fd0hf TdRv3tkL4rZ2nBohmGgso+gga+5xQng29ilKPW1AWpcpiIvFCMEfn4ULaoaT+FyCm2EXcqUHxM4x CeNMUSiLFCPHSK2mRWovZFSKGDikBmWrifDu7BVJXtSrkTqej6aDrL3HbHg2hjQFI/o5kX5jXBA2 hDkzqGXWWhRDNxasEqek9IJQndo0EJLUYN02ne5npFj2uNMU5NgLG5ICIwb6Ub8k1UScV3v4VCRs 1KuROtGMpoM0vn1PHg466Zj5MJq9o4McNorR0/1o5EO+8TX3xM7IK/NVvgCTPBBiNpg8UT9nzoyp vWDzTFKj6SCsb0pT2KgUW3p89xg2VhNtmbSL/dorwnPHq2A0ERvNaDpI7OHn89zgiywZdn5zKZWH FwzvgxGDa8GwyyJDY/fNoHgHkDz4EKKoNH9qEaIwNxKcoZDUaIKNIng3zy1GDHx88KqJOKNp7QWR veNVMbFRz0fTQWLvRUfCasBfj/1ZfecWib4pxRglqrJWdrmYMwUi75LSb/UnzzFIFBI2PplwOTcq xWg6UqNJA3UQMr729NxKsbfHZ03NRjVxIe6dvSJho16NFLHX89F0kLX3PDfUXMQ7CjzKhpsQLWpw 4WrY4NLhZMK7pMSGX2h1MkbCxhevPbqKKU6KMDRMG00H0d4HmsJGpYjBtxOTZdWk64Fv7RUJG/Vq pIi9no+mIie+V2+Kbwzwb8UFRWWjmAoydEaQ3PAtAo5TUmpXmVE3D4SEjb4njDYkQ28cmehZJL7N UytFBIxVkWLPSwwS452tImFCHqFppIi7Xo+eg8B8rD3Pi1+mHfaNyq5hQph/TWrMkxd8TS3DcL40 HCb4GHCGYEbCBCUi2SSvFFsGt5pgpwgbg9Kk5xZj11cXMYmLamIcfGevSNioVyNF7PV8NBU58b3K i2yK45ymlelcA/FrujSMrOCCtxSIULgdLqYo10CoYNE2HXCFdPqgnKbEbIDY6vAkMniv6al4qBoi vLVVIDTIIaeOhNyRxOfRAws3cT0ngfZW36/Qy/HJ+Rx3I4tBwxsjKL0cv7kujjYOKZzVXK1zDyPh gQKSdL3CPHAipaR9NPHzF4m9pkQxQmAhz1RUE5FH09oLoueOV8W4WGGiGU0HaXyv2KBeRL8k+/v6 xrFygtFKMBahPHPELkNisozuT0Ii4ysiRUIGNSQZT1SKtb5bRZAhxPlqc3pupXjZvz3Q11TMmmii 7uwVCRlxaoSO36PoII3uFReUZ2igqVoHxkrhQph/T59C9+A8l8YRuqREBoeBJFNEQWkW5TrDlWJI fauJwIvUXsgoRgzsYl41qb2+tVdEZDBliFcjRez1fDQVOfG9YoNLovRzqk4n11qgoZhKdXSqD7vc 1KP8YTQpKbHBNU5JAyNhg2KzrGNVikFkpEYTjUGR2tNzi9EjcpGh2agmFpvu7AVRS3O8Gqnj+Wg6 yNp73JFkmsaq8UfPKrGA4rOmaChGDHyv1na5Ly2ZwKSSrgU2uFQgw0ohbgfZWvv0NiOcRYrhaKRG E5EXwTvJhI1ixMC6Hq3NaiLOO3tFwka9Gqnj+Wg6yNp7zEamaaqg9GRCZUtUBeRNEUZ2qXCJt1e/ AtUoyQSOk4WNv3mOeZcQRaWpmzcai+lXZ0x+1VTEuWh7YUNSYGyMqlThoolly6s9fCoSNurVSBF7 oxlNB2l8+6Y8nKaxD/vtwQ3SX/zYYaMYbLiCEjb4AEMygW8zZAD2luc0C+W5sMEXtNSRFhMbVL1m 8BlNB4k959RKYYY202xUkxZx0a7eo/aKhI16FYw3eqMZTQdpfMvGw1Vg1Wx4MoE0+dr+tRhsuIQS NliVTYvA8lu882qsIxcSNmbnUEIkPCPqKB9FB4m5viiVIoQP3/KhOZl6ZaK8Wktm1NrxaaU2llEE FzfRYa7vyT8yw/cfaimBSsnJjGJs/bmOEru+E1GR42Uyg0rlZIGQcuHS6JHiF2ZIHanRdJDYa2ZU itfxl8oOvSfVxOxQ2pUZtVckmVGvgjkz6vlogo3b+JaNh5lB08JPwDO8B06LAHivaxTVXvC9raxg 8uXO+JUU8Z9JCCpK0t1EhJDVY48O/bk2mgsC8JjXwrFXAVFdbeTPdh51Y0Xi5+j4vMZxEuAu5HQb 6tldeUjU77nJg8eL4TQXOBE4g7hcEUcCSEo/yHeeC+IfUrMzby6OFMFydCKDzWg6SO2Js5UigA8N oOmUqokI7+wVUQKsVyu1no+mE8vae9ZtMCLV5z2VA7+6j0kNZTD/hHzBI2y8s86kPOdj1xmLU1Kb H92IgtKOmncXR4owOTshKaoLUOTGokjNhYxirL9wi/lVEytHd/aCOCkYE9upFdpYRlEjOdH962WY zKCEI7nKjzhcFCMEHYHRb8BXmvQLa1VbmzhkBmetM8QQ0sygiCRdS6R4xTlQIanRdBDIkCaRsVLE wGEaZ0Y1EeedvSLJjHq1UloGsb3RBBu38f2LjQwoWPLiE5jmjvuXhw1hzowP19bCxmduiWNnQ1Ji gw+GZ4hhJKlBvUjajUgRJ0cqrpqKHHthQ1JgjGn/dqBdTfScV3vuNoyEjXq1UhvNaIKN2/iWjYfd BovDDAjiHafl+54Uo6pI9a3KDRqwZAJFuhqMa5rm5+BHz4UNrgLS8KGYfnWOT0j7aDoIfb80hY1K EQNHKfyiVBNxSvvFXpGwUa+C4dVGM5oOEnv/bEMnN/jkkX5PpDn72Z6jmCreVG0LG24bHCdSIJoS ddLBMtywwbpFuhhhZLwPH0j7aDpI7PVNcWvh3NAXVtQAVRNxWtOxF0TPHa+KiY16DhvWdJDGt7lx 16MMG+Smfk9GrfoWSNrQYhR7uewWu3xUS8fc2CGWVKZEnXQYSW5wXjtNbaSInVNFGWxGUxE01V5y o1Js13CxndmIpje2Zm7tDRI24tVgYqOej6aDrL1nPQq1OxxUi3fvucbiM5h/0XdmC/4V+KB9WgQm JGGDGyaTB0bCBiVVocxChM6pFgmphs4FN4vgnGRCRqUIQZM4pYZrsxQlQrwBNIpWVKSZEZ8GY0ro WDA3ig4Sc7yXz7ngysuMsCjh9GEQNBRDOa+2ueDkZe4W+5CUvGONJVlgJFxQUpJ+J1KQwVqQpEZT EXNveyGjUrx82DEZ0UScjD6v9gbhOfI1Xg0mNur5aDqIXnU0vWKDIx9xjpvZ/Zrwhne09KY7XpR9 /sqJ3pIPCYmLOdVuQK7x5uo48hEicG7QCxPWswCOSU2IiAze66aYUfM/XMdEl8Unu9dWASfFODSY aIjPq2eAtfV///O//ud//8//e9hYqNDSdYpIs4hTForxQ7oOGRq4BrIJgJRd42vi/fn1XHjw973N gzA1nRyvKBGpaF4k9tqRVIoQuO/FVESTorQmaI+9QWRvvCqGVxvNalqk8b1KCWpMMvykAsk3NqCz GC0Ri9G2681UNsx18Fs/E0sVGX0aCRfUmGTYISxD6kiNHg2yr9acFYvRU/uLoWRXNBElH/0L97FX pHkhnwYRE/V69SzCqEB+/59nVa48T0GHhg5IkxjNi2JEoHdfvwBnh5MFbIjFN662yZBDSPOCkpIM Po15bB6p0aTR+tVeB63FiIGz1dhjecqaFKe0i/vYGyR5Ea+KOS/q+Wq6j+9VXrAnlcGnvjPrAQZL 38GIwYXZsMFn7ZsH+qqtGi1eoGaBkGQGnqer4WacZAaHLiQ1mmCjiJZT89zBiIGL+sxGNClOa1p7 RZoZ8Wow5UY9X02LxN7LboR7JzL4/GQPpLlRTLVvlOAoN1i+aVGGpMQGV18kD7ilYNjgLoqwYUy5 weEFSY2mIm6pbQ8WjxTje1YLzEY0EefPrb1BkhvxajCxUc9X0yLMJuLn806VjxpqLVJzD15CDbe0 C+wJBjHwhVqzwaftkwmuxRAbfHawkeu55AabRel6I0WcHF6Q1Gg6SO2FjUoRgyo29KZEk+KU9mNv kLARrwa7eL6aEsuJ7/mbQm0EXyF2u8EgwofsPhdjZKv9UeUGnyLVb4xyScm7zzxXJGxwUjajq0gR O2cYpH00HQQ6pSlsVAoSfjy0VTDSRJyslF7tFVF7s14dqfV8NS2y9h7nRqbwKjDzZAJpXdSn3FiM GLTHiVkqFvUTE7qERMZXHgti53hXKY68SBE6hyxETxUdoNbCRYWIgASE+1GkKG/NDRIu4tRgF79X 00ay9h5zkWmJKiw9laAZ4IqXciGM3KKx8EvsoWNmF78SEhcM5RK3ETlH795Rd4SInJGjhKzHnW2B GgsVxdga1WnN1UOMX7fGBqFIlneZcTEuDSYmGskqWmTNPWZisiLHbNUeMspMUgRSo8kCLEY58pzU p6QqPOSKHzVOAPKM+SH3/jglIgQP6YCqZv+upbAQEXznzmSTIC3/cXw3phSwTV38GUwkxOPVM0Bs vehCtFjl2QM9pM6GhINiLIUoYi1MdT7BMIlDw+QCN8slYCFuMrWcBB96e4wpwWmLJVU9+3dtibrF WEbVGRORGT1EyNcxwnisDaLMG5+KoWkjWU2LjL2nczKWarmECW+R/v7+9dBiMSLwCJOxGzc1OEZk 8EOzIT9F1ELyVlBZoqFXMEfJkQvpHj2MJa/G2ucUIwAu5TIVUaQgr+acI0ZChX06Quv1KloksWGu r8XDNS3tZ2kQy+yWzX2fyoXdYoytdKCMH4AOIFRwaUV6VgpMssJvJFQwXVFvy25ppLR5ZeWjaIFa 02PBaI4YZ1LqbSqiiCj5ANQx5+bASKiIU0cqfl81bSRr73ETMY0lAw8FoMIV3nGVxLMXHEwLpyXj hw2bhCkp9SXvkSFwISGDC67S2fpDEYqTIxcZcFbTIrUXNirF/9Qyot6RaFKcaMensVfE9jBtrwa7 eL6aNpbYe9laUMehCJBmabOtRTHYYFaBd6Q9H1eBcxyQlLxjWbiRCwkbLB+nt60U6Rqp1RQETbUX NipFDOyYmI1oUpy39gbRc+PVYGKjnq+mRdbes9yg3ohKIHv3y8zDbASDIc3IVIOL3V8uSHVufEtK bHzy3EHCBmU0GnBqPmQp2vNIraZFNNdCJmxUileG435mI5qI8/3W3iBhI14NRiWYo8Gr1XQf37Ya DxcwVHlJSZB+K16PYaMYMVA1Sp4ykNJs2nFKSmxQgZmRxrcQeUczRTt8lWLgGKnVtEjthY1KEQNV LxdNivPWXhGP6carI7XRrKZF1t7j3MjQgqpKFVOaDS0LqV9djBi47x6EgZsKI82GpMTGT54jD0Ds nTryVkH6chS1Gy7VdJZF0yJ4J01ho1LEwG0osD+aFOetvUH03Hg12MXz1bSxrL3HbKQVVV2l9+RY oeFbcmVDGBlP46Axo+zy+iYTSJaw8SfP0e8KSW7wGQFTViEGoRGKIsJcpOb02GKEwFoACINlKSJK yqWv1gbR+GR8GkxcNJZVtMiae8zFZMZcI6JFrORFEPzXV9ZllJ8r6c9FYQCapmVAob/DAv2/Fk1h xiKsRFtklQyAU31mEAYf3P1hBqQkFOhk/9hRsH1mXCmk8OPsKhlg7BD7f/61yKkCS6jWW6Fvvib4 QvRg/kAr0XPxcX5hnZaRV76zSrHqgEvCpyAjHa7PvPDz8opI86pZJLaSAxKCNJxXKZxyIHreWA5V Cd/aKqCnxp+V2SiiBg8XGVuPE2BeBqo15Lt68XJQRL24zsph89siWjXFBU19zt8hgFKRpERnUJxR yAhTKniGcTQIwdpMwq8Mg24+zYaM5nTSotCudvJ3grcjBhxoPV0VlIReonnVRVCukYElY22PsFVY aYjQXV5L6EymM7BM984ULCKQQX+f6P+o9tcZ0SEAjYGGHKuGrZKLLT+1EM7zER3HHz0Kz3rGVgC3 vuPPyqzLq2YRqLaHj3/+vP8USupjUvpdOIrj338hNhW4FhnHGKTyDWQHqKZfGcDBqEQM0G7hV9+E EgmRwXpkVs0isZUkqBCbJdwQym+qyZz0EKCv6FtbBZIG8Wdl1uWowcMgzjfF9TwTVDPJbqRJ4HIE NQQL4b224EQCNUYZR+pLtiKBeXAjBkgm/NWxO5NgGZqhyKyaRVQZ5stqeTcqhPefHq8zj7MeBXhj q0BIkD8hyjJx2e1J1WwQtvWSBA67y3clJiNnzS0oqAyGQyyVETSjMK0yK0LubE8uUF2YmI3IN6Kl 1VwpNNEx32lapPb03GIwoevtjiYCp+IsY7bYK2Lex6vBLp6vpo0l9kjax+9FmkWVRbLkYja4Nb45 UQw2dFWo2ODz9tothw1JKSl+8lyRsPFOmW/YiJSmN9a+moI4X78kEzYiRW/390/H6tGkOG/tDaLn xqvBxEY8P5oG0a8de8/Z+PrhvgazwSday4YwoiIGxtW264IPsyEpsfEnMrAhJGxwJ4NGWsX0ytNQ 3GhaBO/mucEUg0/ok2XRpDhv7Q0SNuLVYGIjnh9NQU58r3LjWyd/xQb3lQwbxajFYD3RnZPPFasT 8Bd0xAbXKuq5ImGDNF9EXPhUsViVHjgMYt9sTU+NFBGk8owxPGoIkWWrEB9Tg4SIOFTMr3ecXjUq DL1G9ooHyrXLA7XH7kE5oRQM/1V9rGzkcj79uho4McQVD9wFXR6EhAcqQ8qEdClGFqYltZoW0WCq zw1GCH9Uy01SRJGivJpjlapIuIhTgykp4vgqGkC/cqz1DXm4SqVayVLBZ4n7ggTCf2YSZoJmoHFL SH0KScRPCzUCQgSDwgydjIkISvTDQ/QMgGPzVCE5r4NOmrdJicK7WDpAOIg7R6hBWAve+c8kXuw8 byIoxNAPyLhelVLpRIupJMeDZ8ZnHV/9SkgMcHd2fncjoYBaEA3MmGhEipFhpKKI54LYOZvTcyMl /72GzQ6IFREhC3dXc4OEB1V9iqsKqSC0E5HoGWBie/5aqDjSR5VggstvzUQw/4i+NFyvBdPMBi4p UcEXF/PLcybWlzXwQ7KRrgJE/hEpGIgU/9A95gq7SO2FimBQkZcgipwATBgzqI85xT3mxqnBmPg5 GJ4bRQOICz/2tF5ABZSeGGhvhmGUsmIxfi+XS8IFe7WJnI+ixDnKrpMCQvpm0H5ovMXPKkyR08ZI ajUtEnsdmUZKIXiq7YYoQ0xWqUJ9zBG3kaRFnCrGWzWOr6IB4MLRvcqLjz+eJiCt3jRcFCMCF0vy 0lHR3cjVKyovuMhGQy7iFiLnNAXz5mAwN5ecL5DUamIkfbXn5wYjBN3XehQpyqs5vw9GZG6cOlKq BZW5VTSAuIiXz5pLFU96toA0v2PzohgRpMCR8SsRJ3KkiFLTHz+nV5vnwgVHxDRUK6a84HSBMmU1 LRJ7zYtIKQSdroXUKFKUUn7MDRIu4lQx50UcX0X30b3KC25xzMD7jdP0zYtiNOw6T6+fgCFOs19n 7uUcR071HHELCRccTU1nKwznmIJGajUtAhfz3GAKQZ9GFBnRpDCjaewVsb3xajC9JPH8aBpEmSFN r9jgK8X6NVlD5I6XZkYxYuCrGmaD+0q02E1PJymxAT3JAiNhg3O/6XArxZAxUqspCJpqT8+NFDF8 6bOWYiOaiJPVqqu9QZIa8WowsRHPj6ZBjr1nvaqKJ33vDt69MWNImyHM2f35VwetyQ2fVjIbklLv 8hYZ+BESNr751MFFiEHjnaJFai5kREohcNOlyUCRuxLWrTLUjTXFPdbGp8HERfw+igYRF9L0KjO4 d1e/pso2GBqHi2IqFKFBFBe6Is5hslYULvhwR7LASLigFiS9b6W0e2ap1RTEmWF7ISNSxMDRo5AR TcTJMtbV3iB6brwaTGzE86NpkInvFRseF8AFFTLlwpVE+J9qSR2S98CRIbFklBWUvSgB/Hd4YLks Pa8wtxdUvKvlrBaE9XfthAMh8lt3m+rdiA7Fduy4A/HfiT++HBmv8V205O9j5/FbkUUrVUN6stBd F2XCYpoRsKcjzvUlQIerzRqF/5nHNEfgftYQwCwr3a03fjx2jtQq4h26MRcSgokGjeFFQzQpxFt7 g4QIOzXQ8buKNpLknRQ9zwP2Thgk+xeiwC6ZsBghUCGJczoYzwqCyODa3ZDB95TyyxsJGRSBhLJK 0bVEajUtotqnrn8PRgzUESQnook4Wde72iti8serIzWeH02DkBm19zwzWDfTz4k01cp+L/iFghGD roPUu/f2lxVXsyEppcZPZOAHxN5pctaKQ2FIMXC807RI7SU1IqUYvviCtvqmaFKct/YG0XPj1WDK jXh+NA0y8b3KDWox8vb+YZhVNoSRXawh/GU4LbtcWpBXgHuIwgZXCDQPhMg7TdRacVgpxqB3moKE /XluMMXw3nlONCnOqz23GEbCRrw6Uup/5NXRNIjYiL1nY06GNpxntnd8gXvYKAYb9NRmgyt/Evsn Uvx6mhR13iGkbHAQP6MNY3r9OXaQMWc1LcJOuTQpqsGI4f1Ppx7R9MbLIO3H3iBhI14Vw6uJ5mga BDYc36vc+PnyDAJpmkWPOjXHMwYb/uIRucGH7DPqpBFR/6q5WmceRpIbDMsy6hSGdz5soLHpalok 9joiYdsXKcXAKiTatbslTYozmsZeET83Xg0mNuL50TSI2JCmV2zwzY2MOrm9JTd4qJDSmFYeXS/K SJbDwB5bfEhK7SrD0OSBkbDBHWgZb1RK+2a3moK4jbe95EakiIGPw4aNaCJObpi82hskuRGvBhMb 8fxoGgQ2au9xK5rFb1YVSc54pzLj9K/C/Iu+qWRZ7YbTTK2oS5bFhpN6kbDBACOUSZVSg0883iha BOckEzIipRC+OuD2SEVRIkSOjbUizYz4NJi4iN9H0SDiQppeZQbfOciok8+eDxfFWLXnWwfmgnsj MurkM+T5pbhLs1kgJFzwu6YXrpTqayy1moKE+3lupBQDGw7Y01xMmhTnrb1BkhnxajCxEc+PpkHE Ruw9yYwvNhnfsmXNj2IyApEXRJDyWUZy+iqBsoDikXDx1cGHgVDBfmjyokIMIy20egagHkpqkhWW oU/4++t5jmZ92pkkQn0zxVlhWwWaFHFoMNa9FAZPjZ4BIKG2nhyOQtiFXRpkUMU/LBTDffZG+Yn0 W6Xl5NO28YxLivLrC7FrmqdpE09cCdPrwbkDSa2iRWKu/UikFAGHa3luNCnIaNIMzPO0IiKwTlXo P2Iifq+iRSa6p68H0jQEGX567ktTcTC287gJ0WbZ/mnoNJggmhB14sF1PkMGRSPpeoW5G+HAhLSv JvYOrvaaFcEUg2b3IiOaFGc0jb0ifo5ex14NdjxfTScWyJemV2yw7ZnhJ8p8bQcaimlz04cwNC78 Su6zUpypCt/pThoY0U+lyZqrFLWZaikfMpDUagrixsL28mZx1AIpYmDtOGxEE3GyKny1N0hSI14N Jjbi+WpahNyoveeNBXNrDTVVm8TAO7khzDH88SeISUlm7hmMv0lKucF8O3lgJGww307XWykG1Xea Fqm9sBEpxaCvOIvXaFKct/YGCRvxajCxEc9XU5ET36vcYHkgw08fkA8bxSgBoJaWjNeKkycwrFjq sL3Y4IrMDCyNhA1uPtKwY6UYnkdqNS3CKy5NYSNSxPBHh/vFRjQRJ+vPV3tF3NyMVytF7PF8NS3C r117j3PDU3ik+fk1QERa36YNG8UoFEn1KEPa3OdGnJLKJMnPBbF3mrwxRDAbkfKhA2lfTYvUXtjg rIKGn7DBoXXYH02K89beIMmNeDWYciOer6ZFJr5XucElLxl+vrNZUTaEkV3EwOolsWtbq1OMX0ll 6ubntK4Pktxgu1HdbzENFig2uNEUxOzbXtiQlNngFgy10WzWWRNxfl3tMVErEjbi1WBiI9HARjQt Ahu119x4tJf6xTox17zGO8orzMZisME7YTb4kL1+Y+JUEYbY4Ghx8kBI2cjyToR4TIcPLLSKFsE5 P4aqwRQCN6OYDCtSlBI61gYJF/FpMA6+OxZPbaPov6J7nhnssunLfHpPGLJQoEefvRgh0BiaC74l lvaAZZ4MSPnAbbNASDKDehL1wpAB5jA5wySp1aTzCFd7yYxgIoOqSJMRTYozmjQl03NFbE+FGvJq sOP5ajqxJL4X/Ss73l63oahHG0NqNAZSTZlO7JBrFBIkC6gFVJSatqmDYXNQTykkzdu841jMI2oL jR4Nsa+2QgSQf0sKFTOKiRoipJgxvNvWADy1Dg22PsOC9PgXvYnrVUpQx6GfUZFSbhkWhLmxYEEj NPA1yiQAv15cY+ciP7+R8EC5iYYU8BApHzq4aoIIDiZc7IWJYAqBzTozGk2K8mrPjYWRcBGvVorf L9GspkUmvmXjH40FlRwZgPoayLBRjB/TlbX8BvpisMPU7ZHKCT5xrzEYgQsJGUxsNe4ophaQMxUZ f0bRQSB/notU8kJNktIrmhTm1Z7fBSMhw06tEJHHb3GBIifGbXTLxaNDDzxPRYeGfrx5pHozQ5gz gz126hbhgiLlkMEbEzL4JmGywEjI8MWQIqNSPnJw1QQZFH5d7Om5xWgRfjvUjybi5Fuxx54zw0jI iFcrdaIZTQdpfMvGPzKDXp2fU2xwyX/ZCEaD9+WiW9igp0/rwN3Kai40R9JzilzPhQ0+u5XOxpji 5EiFpEbTIrXXoUmk+EUZSyc1oklxSvuxN0jYiFfFnAmNZjTBxm18y8Y/coMCEoaR8o6bu7Xeh4Zg ioEu17lBUUkzASl7R5lHhp8/QsIGTUu6G2Fw5kMHkhpNi9SenxspscECiF+UaFKc0cSkLH4G0XP8 RvGqUhfPV9PGMvG9YuMv42yzwTbc5EYwYiCgsMGlSFkr/5aUfis+f5I8MBI2/ujzqMqXSvkQgqRG E2xwToHpC97Znp4bKWL45EGzEU3EyZbl1d4gYSNeDabfsdGMpoOsvceDca/3fXmBkwGivFNNsXIj WGLgAy/+FfjyQTKBmv4MO9hqTR4YCRssVaqmDTYipeMIloINayLyIGMvbBTTKM4r8V70RJPivLU3 SNiIV4Mdz8WrNf1XfM9zgyVr1pKcG/6eR9gIxi/K+0GdOjnJd+2TCbzGYYOLhZMHRsJGNtQoN4iQ DihYKIqcBkJMvs2FjEgRAjdDJTWsiCjfbq0NEi7kkxmLFJE3llF0EGYT0vSKC77lqd+A5XS+SdvM EObM4OOuFGmTbXzTNXnwKSm9J3y5MqMPIwpK+6JePGRvOVI6qXCjicwIMvZCRjEUcMf1RdMbmwO3 9gYRG+PVYMqMeA6t0VTkxPeKjS9tlMk5fuqSYYi84FrJcKF6ROc/A7FwoSpCvRAGQgWJmU6nQvyu Fho9GqFebLkzGQjvv30fknbupEYR3tgaIDS4QHKESF2S0apHzwKN6xUJFJswfBALusklr0cw0aCz WiL/MzfEEbWOgiklOIGaYYeR8EBtSbmRLo2cOIohqdG0SO21A4qUfkhOKDglokmRR9PYK+KOZLwa TFw0mtF0kMb3ig0WvfUTsESioxphIxgD2L/62qfY+PINcbyOOc9BJ+zHSAojIYPakow7KkUXGKlR pIEliPPV5vTcSNES/OXeOJMRTYTJt9FouyE/9gZJYtipgcRFYxlFB2l0r7jQ7UtxjlLaciGM35MQ mI6ICortkgZcQZt+5S0icCEkXHyrcEEvTaSIkiMMF0ULiPl5bIRExTdLiJ776Y4oZT81HJw6peGO uSLNizg1mLiI36vpILH3YqbKFJtVm+QFK5DlIpi4UCEqXLDol0kIg59wwcV/zQEh4cLf+jAXliJ0 DjBIqooWEBeItLWIkCLQjra4iCJFeWtukKSFnApj0iUuGstoOkjs4ebz4QW7NfoNWOpl764tRjBC YOXKXNALpimgMwkXH36MHDASLvh8enoaYW4wOLKgUUkVwUWAWjOFizGm+Nut3CgichYyjzmyt0i4 iFODiYvGMpoOkugOFw8nJSzmU8UcLnQlrdqLwRSCdvfIi9/cEkfkEsqczfWORcLFX87nEflIETpn GDQmqaIFxIUU6bEIEScRcFrEeRFFivJqzjlgJFzEqSMVvy+aNpJj71leqMxS0whJaxdPa1uDKQS2 6EQ4NaMZdTItCxeUtTUHhCgo5mxUHoYLSxEmpy+06FlFBY61cCEhMKSp67woUpS35oro3VqnjtT4 vZo2konuX+9I1sOpvGSTxt5RZjN5EYwQPG/G7HvuiGPAL6HMjDz3COIXXzMoETZCRE7OK1GqZwFc s7FQESEC4JL3bEVYDzFSc3s1NkiYiEuD6RdsJFGEk0HcL9Tc86ygrJCfUs5puTJvSDBF8MVCPUzw JXvlAEFKSEzwpayMSo0oJs3WvO04UjSB3B0kKSty7guIb1Kkx0ZIVLCzcFGkKG/NDRIu5BS6Vkol pI5lNMHFbXT/yopMzChU5K5hU8GAtVQAJQB1px5U6rclRImIiF8LFAgPVJZouBmMVtNDv9Uyf4py KQkJuvpYbwbNcjiIEkV3sZRw+9S6c4QcwkUPDDiAY+txMsxrQT1HBpt8d28YCMZvyMpUSPAZRsX3 KymR4BOTg/S9oKREPcxIETaHGDQWGU2L4J3thYlIQcUPw2zYo9ezJqL8urU3SNIhXg2mF6HRjKaD rL3HbEw6UM+Rweb7N4OovBrC3M6xMhU2vvnBHKaE1J2wl5WfnyKV/LyaoLH6dKQInVMMkoIMFDlJ goiMeW6kRIbOqYiMaFKYt/aKtMW0U0dINaOOZRTBxW10/3o1JjMo6iBMOffOSx4ugikEIvPr8aPb NBympETGx8y02KvJlSbaWGNieaQgg1MMInE0LTL2khmREhncxurMiCbFeWtvkGRGvBpMmdFoRtNB Gt+/2EhmqMrSrysdoT7nJDaCOTMY0IUNTosmEbgEKGzw8cukgRFFpYmTtxlZYLQUsVMbmY38aCpi 9m0vbEjKrzo3F9NhrCbiZB3t2HObYSRsxKsjtdEwbrOmxnK11/fk4SBLVZYe9MOGyvzDRjB+URbb wgaLD3lRqL7Qmo6WcvUcLYSQjixYLM9anjG1nZxjkNRoWoTckD23NiNFbrA6GDaiSXFK+7E3SNiI V8XwaqMZTQdpfK9yg58zQ04GSX89/FY1pjDFwFQaXxjMcdwrsSNl7yCqkQuRd5o8eaMRhqRLbLCg edW0iNjocwcjBn1s42hSnNYE+7FXxPbGq8HERqNZTYvEHuw/a0NVeakZBEsHf1n3T2oEggza9ZDB tSWafjJD854J//VTBUIFX4tOz2sZYuQ6Yo1JRs0isdW0iBBpwcHOEBE9hPhxY6uAbFEdhj84PTKJ gvdo1dzF9YoENvQ12sTnlhuoCNMVjHj/pTNPGP348WxC09SZonnCESAkMKLLqCvfjeU3sGaRgBo3 iUFqi6doFwppIruTL+lRgDe2CoQE+eO2IzJxmYdWzSKJ6xUJHD/V0A/hHNMluQTlJ9TylUj4VKWh M0HV62Qpt5014hzS5aXg0OiIOGKuOrtqIeAgNRUOCuE8tzqjuWoU8cVS4q2l8WZlxmFREC2LJKp/ UZBOVCWWSite4g4vi/DzsUareBmw/u0Mg6oXAE3I0kbq7yQBlSPdMpYI4U4DGSUD1E7CTwvNio2u llH0UuIMYDDpIb7tEOoMLseVQmoL4v4qGSDxOPZ/3f7Ew+r/EzzdVhqCQIreRcdE74GEfn6NxBW9 RxsFGj7te3pVjWwVvvt+NQRRs4gIQCQESMhvsMfVYiB6FJ70rK0Conr8WZmNAg4Y47u7uInrXwkw 4wZKNPBdjjGfCgmB5D31MGhkYOZzY3CAjPzi4HkDBkgKMC9LX2oZIma7UzLRYlaE+HWTqXAQIbKX u1+SBdGj+G5sFQgHdieIA24Mq0VFopeonlNADeWXpgVQ8IcSPFEwEM5/sMxq3vlGd95xbr80B1wh mIAFhAMGtOlBLQMHHErQhvmoKTK2yoGEgFjG1QdglQfRQ4D+lhN8x1YAD6d1nF/+rMy4fNQs4riw 9bhXTEOg0klNB0SCjsyGhWByn/G/WaC3yA/v47BKhS8/x/shxL5p7tUFiEgRNucRxMxoWmTsJRsi pRC4yZ1fdjQpSnl17A2SfIhXg+n3azSraZHG9y828lZQOqnP24kNSnCHDWH+MVmr1WvAKDUbO+wq SErecalj2kYj8k6TL+8TjhSxcyLhqqnIsRc2JAXGjghNykXTGwU5V3u8UkVkb7waTGw0mtW0SONb Nh6OplU6ST2g2PCmXnIjGL+ot/5klwlzst9bhmKD6XLGj952Cxt8fzTdqTE1lpxIyPgxmhapPQ+b RkpssIloNqKJOL09eOwNEjbiVTFyaqNZTYs0vmXjYR0FpZNvVFWbjW8uEQ0bwcQGn6n0r8Ch+WQC V4kkN1jPbORCwgatXoaQknLGM6GV1GpyXdyx15YmUrDxps9TKsuiSXFGk2v89FwQP8cYMX1qMbHR aFbTIrAhTa/Y+PlgyVpsfHHQvO1GMX40+nSzwV28Kt5mU09S+q04uZ7IjYQNjtRlZFUpmotoX01B 3InYnp5bKbb+OsmMIkJn/e5qbpCkRpwaTGTU8VG0QKN7xQWbfRlPcqe0by5ApTD4IYJ3ndPgPfmm CsKjRYoa0qpy1VnjFhIuuHk+hEWIbvRO0SIqpuljixGBtkU9p1L3jh/IkGCaUNmhIE2LuFQpHlu3 R88CY+zZrS88rvJIJQWtxBARiMaPeg3zwGUuyQA+Wx7XSBITI6A0UBSSbteYfuyZy1bPALWVhECG HMF5v4yavEkL8dEihXKbGiDpEH8GEwlx2WrcctxE9SoZmN3JJn79cANmmglhOAYJLpokGZygejG4 9SWeffi5QZIMVIOk260Ug+U7TUGOvfAgKTCqKjRmYc5uPQrx1loRd9/j05Gid7nRs383tn8xkaGF qiOZ45kJVV6LicVgwhWTMKFtQBMhISWq9goXsWuae3mPELoixZg5ylfRIjgnmRBRKR3lZZkOJqJI Md6aGyQpYZ8GurhtPRuHm6MaezzIyrCCMa2ujxcTfGF9mCjGGIlySXxj8Kor5dUF/JWUqGDnWmPN IvJNszCXH64UOxrRvpoWqb1QUSlC4PKYo4goWWu7mhuEi4fXqcHERR0fRQustcdcTFZQH5IB5wcj gWZFMSL4sHMM2/hiZCKXlLjgSxgZSvI9Vl0moxA4WK7kgYtI8Y9oX02L4J1kwkWl2A9nlk+Yo0lh 3tobRM+NV4OJjHq+mhZZe4/ZmMygFEODPfZlNNvOOyLMrYUvVZddSmOSB5yKCRt83zSRGwkbVKhl QFUpvy03moIce3puMWKgQsK8RhNxeg1A7MsePhUJG/FqMLHRaFbTIo1vW4yHA042mPiwYLzjsEfZ KMak6lunU2CDw5LJhF+k+PU0KdJz5IEQ/8bMnTQhXUxxcqJC2lfTIngnTWGjUuwDczur2YgmxSnt x94gYSNeFcOrjWY1LRJ7+Pk8N3741E3YYOTgAac26o0RA9+ENhsUwTQTNL6QdyydN3Ih8k7TNYYK ZkO6xAYnD6R9NS2Cd/PcYhjW8OFoUpzRNPaK2B7/014NJjbq+WpaJPZg/zkbXPOYAecb08zmRjEW HTTXUm5w96MGnGoCdS0kDdibnxskbHDXQNbpKsUwOtpXUxC38LaX3KgUruvz22IjmoiT5ayrvUGS G/FqMLFRz1fTIroNBE2v2OCbNKodRVo3DKTdEOZflNJRrdvSybGl71/d9wKIjXfLiA09FzZ+ulhR Iab0d4oWqbmQUSmyWtceiAwUYR8u0E2OjbUizYz4NJi4qN+raJGYe5EZqo70CSSkaY3NxWL8nuoi xAX35qk8k8glJe/I32aBkHBBXUj64Eoxhoz21RQk3M9zK6VXvDOPaFKct/YGSWbEq8FYNpxoVtMi je95ZjC7fddiG7Uf3E9VMoCcF5QZhgu+yJHI+RJMuNB3SUWOgVBBXUj3lyPEMPJWzwC1layIDA0/ X+pxUkiNs4JlrmSFbRGygCZFHBpMNMTn1TPA2np2UD/LPxhFWuUiej0WY5VWHxRU4rG4lKBVvqGM 4Hbn/Pp8OzCuMW/6avmhMY8hLUXLWUWMKoPEXPuRYkTw1RF+NClIaT/2BhGBdaoQTq3fq2iRRvcq IdChH4BSJ22jhItiKg9zFR6DOJUxKwO+tfkj56h1Tk4YkXPajmr9oTB3I1RGSPtqotTkas+/72KM jrRLdDQRJ4WPGf7GXhE/N14NJjbq+WpapPG9YoPaDDVSSNNhNzOKEYM2wPUj6OOBbiz4klPY4N7t DMGMhA2ueUnXWykG59G+moL4fbQ9PbdSxPDlHR923KxJcd7aGySpEa8GExv1fDUt0vhesUFxhuqM kNZlkckNYf5FuXhSawkoZyCa3JCUcuPXMsoWIWGDjf10vcYUJ2cQbjQtUntho1KY0RWWyo1oUpy3 9gYJG/FqMLFRz1dTkBPfKzbY19VgnNInYiwbxaiQopiWjOc15EZ7s8FwMmxwmWzywEjY4Nxgut5K +QyCtK+mRWovbDD9kBQpyO28ZiOaiPPt1l4RNzfj1ZGioORO0yJr7/GAKxM1/UQ+KYU0l8+UjWLa 11Q5LWxwh0Njl5Q6WghMs0pPqIsz9IvyFdIMP4WB+ChCBuPVtEjthQ0OLEgKNt49bB1NivPW3iB6 brwaTLlRz1fTImvvMRuZqNHU6g5u5Qbnh4cNYWQXbHzr8DV2+ah9Yic5wwZ7uskDI2GDsUW630rp MMKtpiDHXtiQlNnglVEHwrTMmt5YO7jaw6ciYSNeDSY2Gg0NbDQt0vj2TXk4UdNIRNMJscERNufG YrDBfr3Z4AxCY0eKd0djoU5DOHg97cZvZ3OGlBrc25VxbBUtUnNamFiMEDiBbjKsSFFK97E2SLiI T4OxXehYYGwVbSRj7unEhNkLJ0nChQqWVcy8GAUPWY9lLVGvs/pXrl7NIIxl0GaBkGQGRSLphSXl MOkvpX016YDC1V4yoxgxqK5ZmRFNijOamJLluSDtX+NVpXhuPV9Ni8CGNG1mPNwlYdWDcmolBh8E zA0OAxEBsyZsMKClUt9cfCDkvOAM/wJ2TfO2ViVKiMd0GkGqV88AtRUikPFvyafUM8SKGiKkEC28 29YAemocGkw0JAxGY+jxL3oT1ysSWFXiZxQLbww11a2yKgmGa5RnMECyVe4eSdASUr7++DFyhC2u ZARTqL+tRzQmGjiIIOVR5BwpUnMhohgRZEN9NCnIW3tF3FIzzJZTRyh+g6yiRdbc44Yz3Qj7BjRN 5oJvXA0XxQjBtaT8ApRRJvHfJSUy/vg5yBBi5zRj81ZjMKQYnUf7alqEkZA0iYzFWMBmQgxlo4k4 /9zaGyR5Ea8G069Yz1fTImvvMRvpRlRQqdkEmUHZxLAhLJmBPfzlf+pUthqLd0mJDc5u5w0xIu+o edJZ6ouUzyLcaApy7IUNTl/IB5SyoHXRpDiv9vCpSNiIV4OJjUazmhZpfPue/KMboaBDA0QFzMA7 70kxClP+uuKCNKOTSpxI8etpjtRpCB7Oi8IZ5XS9xvSrcx5B2lfTIrHXoUmlRHGH+tGkOKX92Bsk bMSrYm4i6vlqUs3BNb5l4x9NJxV1Gvoxo2E24IkargfjH5pZKDeYhmb4+aY5g7x7twzZYiS5wfGd NC7C8M7nD6R9NS0Se31TmKxKij2wbBPou0bSRJw/0TT2ivi58Wow5UY9X02LNL5XbPChWw39kNZL n9woBrHcEW82uDQrJRVscGfQwTA9eWAkbHDJVjrfSvksgrSvpiBkfO3puZUiBgYZIKSENSnOW3uD JDfi1WBio56vpkXW3rN2gz6PMZy9Yzs3bARLDIyYzAZJoqjYdJaUcoNU0q9XJGzw0+bkS6UYVEcK NqJpERpmaQoblaJp1idzxUY1sXByZ69I2KhXK7Wer6b7+J7nBiNMfZRMvxUXHQ4bxYiBgkgyniGt xraOXVJig5tJMrKkIjtRMRrS8rkIipBOI1goingsSMifx1aKEL480h9FWkS6tVYkXMgnt6uV2lhW 0SKN7hUXn3ykNs5xJ5HfE1YfwAiK5jhFlgxpVerqMCUlLqiQyMDyrxB5p6mbSxMhI1I6nnCraRG8 m+cWIwbtr180EfudvSKyt16t1Hq+moKc+F6xwTFYwvQwZsgIBBc6CyWrH7xMaGTrmLvCQTRt01MF QgUVKFnXqhADUgutngFiqwOUyGgkp8uRYLRqVKB8Y6tAaJBD2A9G6qqQVA6tngEa1ysSKOvI8BMm cuMLq9fBVOvkanNtkycjmBmHhd+IQIOQ0ED1icYKwXCNYeSdokVwbp5bjAjeu+ZcTQQpp8R67QWx vXGqkJio36tokTH3tCKF3S0uliYAxvhax1U3shhcsHvvlODmKjHGtjFSdo470vP7CykZlJckT4yJ DNYLJbWaFom9ZkWl2J3iql2nRTUxDpV2kVF7RZIX9SqY86LRrKaNpfG9ygyfazQbXHDlIQYh66yj YuDMZNjwWUezgRTeoTwyCAtJaniGN1JODY5T3GjC8SAomOcWIwbuBDQb1cT/klcXe0GaGvWqmH7H er6aFok92H/eqVLmod9T5DO4Sm4Ugw0SFu/0vvfmhR9JyTv8Th4YCRusv6TrrZReZkutpiA0PLWn 51ZK03MP5/TOW5NS6dZekeRGvVopNV62t5oWWXvP2fjLafB4xw5k2RCW3Ej1JZ2i743jTWGlImx8 RKZI2OBDXVkZqxQt/J2mRfBOmsJGpYiB+k3nRjUR5529ImGjXq0UPVjsraYgJ76nufGt6ktNJ/it OJFvNg5GK8UmJnnKkNYXx8HGt6SUG1y0mzwwEja4mSRNbaU0s7DUalqk9sJGpVg202etMg20JtrX O3tB3N6sVyN1PB9NB1l7j3PDk3ikP1h9MBuuJONNOZjWSnT6g/eRsULGUlxKo5aEqVueo/8CsXea vLkQMRhSDCMjtZoW0a44msJGpehT2ciC/dWktaJbe0X03Ho1Usfz0XSQtfeYDU/ikWalUr+nxg3s iIYNYWSXRhfsIcouq5fJBI6Chw0WS5MHRuSdJlM3QgwjIxRFblaL1FzIKKYBhhfVRpFGJbfWimjR dH0aqRPLKDrImnvMxWTGF9sT5oKPuQ8XxVg61Duh0/idhHxKRnnxJ08xugNpXlCQkj64UizxR6p6 9m/GrtIjIhZj7MqL6qyoHkald9aKJCvq00gRd70eTQdZe4+ZmKzQ5rqSwn10kiIQ/rv6kGh9YRzN BZ6HiF8/VCAZQZFJRh4VwiELWY1j5m83ZLYUGiKCqIYoyqtqIb5bUwVCgvxB0wgRcjwePQs0qleN Js2i1jiR5jsifTGEwQvee+9Qr30nHp8SUjawKtdfXkhYoLpES1kwEykS9FbRArUWHipEAB8MoUWE FGGfGFFEgqy5IDa3ThVT81a/R9NBYu/54AJpKjv0M7IXReVluSjGxMf3F9Bt6OAQQeK2fNOMqNMO IaWCCpQj5QSgvDrDzuihWb8xpnd9MV4RMSAqqoh6OSm/mCuSpIhPI9RI+ClG0X/F9ior3L+bCYoM tEGChmIE7BGxBpaes0CFhOSbO3yTIyRZQXVJephKMfRjWIBzoyhAeJ/HFiMCvgRlKqqIKO/MBXGT tE6N1PF7NB2E3zn2HjcSaS5Va+kphNKSeNVKLKbE1I4v47bc7qaVAITEBYUJSQEhbS9dUCB+KsXI MVKjaIFayxtSIToaPtLgBrOKtKZxa65I0qJOjdTxezQdZO095iINJtWXn156QZrj1+VCGD8nIXj0 rWUZzymIUkLigm9vJAeMJC9I/yz1VIr1g0hZkV+HArUWLooRgTaF9IpUEVHemQvSvKhTI0XkjWU0 HWTtPeZi8uJDxa16R76496xcFCMEVxuSa6xHNw0kJTL4LlHGD3xXPYmhyZorJaEsUiw6Rmo00TgU qb2wUYwYqP+Fs9VEnHf2ivDc8WqkiL2ej6aDrL3HbExmMBvPcNMbaXlLhDkzWJCh3BC7nBZKIrxp R0xs/OQ5xnxCFJUma9ps5DWpFMvzt5pgQ4jZt72wUYyxqHbgLppYHL3aw6ciYaNejRSxN5rRdBB6 6fhZNh7tkiDNAmOGm4Q4uVGMKgbXG8IGn7hLJlA5n8nJnzxH5HoubPCNxJAhSAn/FqFRdBCc82Nu IipFCLrORWREEVFGkSaG9rJIuKhPI9VYYGwU/Vd0z/sSTiJz1WF+KWou3Jcs5oqOZManL4vTby4p ZQZHK5oFQsIFC1EZdkjKYXIyW1KjCTKKQMY8txgxcCmwyagm4oymtRdE9viF6lUxtfyNZjQdJPZe jDFUf5lGg7MskxiBGGGzRmWrLDkpe9hwQIgoRYqeClAqqCzJgEJCzgvOgJuJ6IGJANTPSEIEGvJv yeUZGchVDRsnUTO2CoSGOjRCBB2fmQOjh9HBAGvreWNBRYd+RqR16VcaC2FuLGAhNPz4njhFLanM hfxckaQE2x9ZwaoU7dWtJqIWcuyFiWKEQCW6U6KaiPJqz42FkXBRr0aK0BvNaDpI49sX5B+NBRUd GXbqjpuQUUiFTroOiEz0OW1xsRM0lTsWCBNUlKirGRmGN1E8WiCiCI6hJjwUwntmYe5CqocIL7ZQ HCAsxJ2KEHEdjhYcDoKDY+pZQlBe+e0RPy9qKrW+F2Kfkus9TQHfSEo5Rk4OaG7mp3h1eSocUEyi DiaQB9CRGTVwwKkGJYxtdVxWiC3Cv53fVA87rt4qWlsBwoH8MS2WqcuEPGpOEIlr8+DRZjLC1HHo 90M4B9kLoRHvv6ggUh5wa2Re+18f/NE0yE8RMUBIoEVIcxEZLZFYJmrcbBaJLT21Qnivjzypw6ge Yr61FSAkyB88rMyJYtQcZGw9zwTu58hMedZpqLkEghe8d7UpuUUvkHf+162YJmUWgQQAhaNpmbcS DTlkzixc1RBxERzrUwvhvWrgL3oI8GIrEeep409lCLkuj5qDjK3HJGRoqbpK1SSSCfqykVqEhVif o6zZLxinMfO6cxcbnmrq46fUr6f6hgmZvuOq1yEyWmS3zKjRgPFqK/lTSOu9urgVMqtHdb43tgIo ExhPx5/KHJejZoOAusT16nXg6sKM8NQbhoRCeO9KW4yys6V0IUCEMhHzUwGSCXy2T0OEkWF8Gc2j RiPOiy23CQvh/Y/v/NGMznoI8NZWgJAgf7BVGUiQyzg4aorkx0XNKxI4vpaBJB/wLQmC0Ij3fLza zHN3R9733PGuGZifIuKOqZmCeagXxIPBiESLm/2Pq6kkQiGcZ23Db0PUEJ6H72spQCiQN/hXGQJu DKPlIOSB/Xv8MmRmQfHkFwc+tZaha/WTB4KgkKV73dDvYRmfJPFvrA/NqZ1kAJ0fXUDygCt/1F4z 5+BLEeKAzL6qOUhshYMKse3DAMMcVA+VWhdbjlhASKBWQP6MTKIAGDWQ0CDG1mMSpkXglCRPI5sT QjweBAr0k4gCDlvmXffZHyjQ3ScOt2d2CJw1FBzlHz5DxMMWGSUQECB2En8Q/P7xTx4lCU2Hg9YO ocbOcSWQWq84O0oWGDvE/q/Ln5ClAEO/GzNgvlfRDCjGb8i16Q6fT47nB6d754AwfnEZan5vIc0B KkXScxhTEjA7ktRoOkjstTWoFDuZfzs9qSYO2Ej7xV6R5EG9CoZXG81oOkjje94iqDDS436vB/hO k+/FlNH6YfkFfJENPzSLDXHuM48V0U/LOHHqIyTlN8LjH5NhRZBRJOZM4mKEoMsaLpp02kiaIL/2 gvSliFOFuFhxYhlFBxlzzyoFkKYKQ78m0hRIOjMWgwtX0cIFe+TNA6Ts3Heegww9JzI07XHhXzFl BjUV0j6aDhJ7zYxKEYO+UHrRRJzSLjJqr0gyo14Fw6v1fDQdpPG9ygyqQzQgYCHtlxZTLSVtcDAm /zpXqvyj2jwDSFbZ491PZOgjhIQNrtTLOFqYU4NqJGkfTbBRJPaaGsXY1c0FL6xoRRM/tjXBRu0F cWqsV8Uuno+mE0vje8WGq3XNBlUIzY1ixOAyWthgEp216C9J6bdyPa3aSCNhA/F0sJHS6FJVt2bD moq4m+L7i9PaVIoYeF6pUUWEeWeuiFJjnVopxpUxV0VwcRvdKy704bf4pq9EJjP8MThHoG9Jyix1 cxlQcsQtXPjuE3MhJFxwR4oGNPBjKSLnNIXYgVVrOgi/1Dy3UoTgo8fMqFDEY0SJkF/LmgvSxKhT xZQYdbyKDjDWXrQZfKAzXSlb4MNFMQZ9FNqbC3bgkv+csYxzlK+mOxVSLvgGcrpcY2oIOXMgqdF0 EBWJ+I6mixS7/Z5paRJnRTopgHKlYc0VSV7UqWBuMup4FcHFbXSv8oK+U/5r14ExZfKimAbAOqRD XnBYPNnPGdj0tW95TkNikOQFN1Cl/xVGi8H/jNRoOkjs+bmVIgQp5x2pIqK8miPlioSLOrVSLGXH XBXBxU10WOsI6+EqDPtFfz26Z+eJ4zd9R4oxSfI3zOGC+2OTBZTQECW9C/d8JG4h5YIvWUMr2SBI VHBvYEaaUXSQmtP230qx1d+eJHoYPlnPMVYkTNSllUoo8FU9MHEb2zLxcB1ChQZJim9KhE2EIRTC wyx18J4k8z8kJB7obhOzgKQE5R/peCNEhLrLaPQ45ACwME+NDDZd9crWWrQQ362pAiFB/jhFKjRR VM33/L2WHg+3M+cghTjBhDpKkdiCLgfCcJ4NJ1eKkgysfeanf5eUSPiKDK2dkLBA3UcG3ZGCBSqv r5oOUnt6bqXScUgJxpk1oNh5V1NBZOo4VIzTFQ6E51By/lgrz4bdNK9z6J9mZzgopk03fT2T1PId gaQ6+5/xiwFgxhdC7JcvyMuo25iC46iBpEbRQXBOmkJBpdjiZ+aCudVEhNIuymuviHgYpwJB1say ihaJOdx8ng9UXWSgybDW97qgs5jWjGgfZdY7mXrvGaOr6dScrNMNI3JO07IWGQrzS+CtTJERTZBR RBUBaAoZxYhBBd4XTbRK1nTsBTH561UxsVHPV9MisQf5z9mgHCQDTT687qsa0FmMGL6dbQxnua+L dwg2JCU2+ARE2kUjYePjt+WKkdJAPFKjqYjfRtsTGytF28bWntmoJuK8s1dE9tarlVrPV9MisCFN r9ig+EJLUEh/MJtWJ8pKR4qEiEFfgZFddinyCnAQKmz8RgZ+hIQNZpcaqBVTnJyR0EBkNB2k9sJG pYgBns1GNRHnnb0iYaNerdR6vpqCnPheseGT52KD3cZhoxhLzFxA73ELn31WJjCHkJSaLp8NXyRs sMCVwVSkiJ1TEhqKjKaD1F7YqBQxvHUgW02MZe7sBVFzw29Ur1aK3YzYW02LrL3Hb0qWbdgkzBoM 0rpvPrlRjBhcPoxdKq4zmfCF82LD99aaDZ6zd5o2sV44mN8BjhyIn9EEG0VqL2wUYzzECgnsrybi lFcXe0XCRr1aqY1mNS2y9h6zMb3q34wRkP5glBw2hJFdxODyYdhgzpDpBBcIxjtmDI1ciLxjmqZN dLNhKWL/vtVUBJnaCxuSAsMM84+rJuK82sOnImGjXq0UhRJ3mhZZe2Xj4YCTciTd3S/vWDgPG4ux SeXyYdjgFsS8A79I8etpmubnWJ/Qc2FD0iLDkJoNLnGW8lF0kJpLbWmlIEPDOdEaRWw3SPfFWpFw UZ9WKn7D2CraSMbc0x4FoaFCc0KVmw2k7TodvocJrjxMFvxKSL4tEQJCBKsOGlrBhIWI+w+fNjAR 1hMgtPephXgR/3hFXbMvq7GJq60AbS7q0AqNz6tnAP9XHj5+PdJYkDV/WFp0QugSBr0ei0HDOxsL 6lq4RjiJ72sRRAP3ATVmnrNv7JDpO2DmQbqUEZw1yKAzmg6Cd5KBv4MRAjc68DusJqK0pmOviHhf r1ZqPV9Ni6y9x2yksdBIhG0XscGOiu+rgA1h/jHdv8quL2Xm5+YCbvWzdGR5rIic0xTN24UjRZgc Ncjw0523Ai9ScyGjGCG8dY29mngJ7+wFMfnj1AptLKtokTX3mIvJDOovMvzksqLhohh7+vrUs7jw kpt+cxYT/j9h75JeOZIzUc5rFf8SMiIzXtPeQS+l9z/pc8wA8EqfQqpJpSxw8TCCdLgT7iwZ1Fb+ rshkBk0hHXpHKvsmlDpNh+CdmiTxMNZffs0MZDQR5zt7gzQzxquTYgZWr07TIWfvYzY2M+jAaPlJ 2/6yIdbMSHM9bKRhP7ErZWr86e9gQ6SpwQdO++QMZpx2579qKtJM3N8dRgysciTJRhNxvtpD0yBl Y7w6KZdCEs1pOgQ2am/Y+MswQieGETD6+EHJPjUGo/HFj1OaG7SGNBP87CT3tXOkGTPyIcqyQc9h h95gssEeCbWfpkNqLzl1GDH4rtZbbjSxlqlXL/YGKRvjVbE89cfz08Si3pv4/vYM3dzI7CNs/Pw3 J6fx1BiMGNJpDxt8yb73wE+k4t03Zy1mS5CykXnEYniX/QYtP0fTIbChJn93GPUkRUPYGE3EqVey MfaK5Hfn1WCEvZ6fpkNqD/Y/v1M4UFB/kf6X8Jobg9kUZbs9bLAJrQ28P5XSO475GzZEygZdZR1+ RyobD4aNaipCxo+9soFypVyy9qywTAOjiTjf2RvE351XJ8W7wneaDjl7n7Fh0wG1tN6xHFY2iiW/ /2U6Fru8zWvsfHG1bPBlnJZgQcoGH+HpFu6RoqiuFC121XTI2CsbIwUbtPyGjdFEzf/O3iBlY7w6 qfP8NL2P7/PcoA/FI6TDBuevJjcOI4Z8QoerwDv3iV0pc4NTYpsHnpCa3HDyNpRVyJ0UEaoiflak 5O/PTsrJ8c0CVUSU76wNUi70Kc/VkTq/T9EhcF9zn2fGv99yvALSvqzyPrFTcnZJ5RWadyPARK6U XPysDC+RRfQOR/0mkk+NYIb+552mQ8ZeM2OkiIHun1dNqHxnbxB/d16d1Hl+moqU/Pr5ORv/UXCb GOw4WTIK0cvH7slY5Y5v/nPeebmgfZBf0UEgUCpoH2kROkLum+hoMnoWqK0pUCqD9xwDHB5GDWn2 1tYApUGHkjoK8Yj1+aat07PAxPXV7UEDh+Uy0nwtLa9GqI+KQYNnakp+voJh0PmkGhnxuyKDlAbW Ozv2KoVr2XrQWnwUHVJz88Rlg4JSPI750lKYGE0EXk1nr0h+t04NJBPj9yk6ZM19+srMKVgmE0jz rZy5PQaDC3YehAs+ZK+7hI5UBjlq5Ln+/q5k8Lq0Q69S/C47D1qLj6ZDam+ygv0JShGDOxO90UYT cVaTM8TIDNK8GK+KJS/G89P0Pr6vMoNlFC8n4zXnkExmDOZzx50HZEa6xmSDi9jy/J/+bpCywTc6 e9coldRg50GL8dGUvQiPvUmNkSIGvtgQNkaT3Q7R5JQsmopMaoxXg5kb4/lpOmTi+4oNejq8nkhz DvHkxmCwka0HsEFXrq/ZiV0ph3cafZsHQcoGWzo6joxU9lEodZqK5FEWe/7upKiq+O5n2BhNxPnO 3iDNjfHqpFhkrVen6ZCJ7ys2Qn7YcJNcBxKxxMAJzs0Nvmc/sWcrHWzQOjJ5IFI2fnNpy1mlsvfg jaZD8G5/dxgxsDMxbIwm4nxnb5CyMV6dFO9m6tVpKlL2a++zgSTvfb2ezD7sypaNw6hW6MAi4yni 6IVAivYBpcwNNlQ1D4KUDd5wdPAdqew4UOo0HTL2/N1hzHc4Kz1sjCZ6pN7ZK5LnzXl1Uuf5aTrk 7H3MRifxhNf9N0izgjdsDEYMvCLFO/9jaik+3NeClM9btAQTiXdO3vLmEaWVws1KnaZDxl7ZGCkU cMZtntGjCfF39gbxd+fVSV00p+mQs/cxG52o8SBoJynSNqE1N8TILsj32zHa5bP2E7tS5gbl3OSB iN6Rnm+FyOAKVVEeq4OMuZIxGCGQoy+KiPKdtUFYNH18OqmL5RQdcuY+5mIzg4HT2oXEt1m1XAxG COnGx+73HyxOep98a48rw2N/V2Qyg9ujo/BIMfBV6jQdMvYk4zBWSZjoJTNGE7fQO3uDNDPGq5Ni +K+903TI2fuYjc2MHzk/GGHIHjIKEYHn23sN/v3Ja65woZB58Tu/ggqBpgUrMy0/gqU2j1D1JGre zuVxFlslwhd2TkHIjibFqCHCt7YGKA06hKYTsq/0jZ4FJq7PBxKWIZgW1jPeDg4LYrjGIgBt+bhP RTvxKSMJnBM1F1+kLNB60qGXoJRiQv1OzyH4tr87DCJoBOJ3zvyiiRjf2RukTMSnk2kkJbR6Dqk1 0vbzdGBabZS8o2ZRapgYDCbSmw8TfNi+sy6qHqN0XjSTjyDlgm6TDrwjRRFZqdNUJMzHnr87KTpO so8iM79o4g3/O3uDlIvx6qTsKY1Xp+mQie8rNmg58Woi7SfA+6gQa15890wV2ODL9s2MfChcNr5X hg4ef1c2+M5c131GijcJ7zQdMvbKxkgRA+2QyYzRRJxqf7E3SNkYr06Kt7+1d5qKPPF9zgZvyWEh ufHdz3vKxmEsHP7M1kRKOPd8Wij8UUrv/u3vBikb7IVtOTFSvEqo1Gk6hHpNTWVjpGg28SOg3iej iUrsnb0ieVCfVyd1np+mQ87ex3dKhxFacP7LZAJpt0SWjcHsjXSXI7mRQxPDhlKy8V9/5z9xg5cN DnJ14C2GFE5V6jQdMvbKxkj5jywqysZoIs539gbxd+fVSV00p+mQs/cxGx1G7MTMZALpf7j8ZSPL rz5BOVhaNrgd+JB941RKNn70d7gkondMpVzIDRuVImFdyH3RVASZsVc2lAIjBoagV03E+WoPTYP4 u/PqpC6a03TI2Rs2PnxLQJelh2vr3bdfTELKxmCwkTZ+i3M+upg4WW5scc6xPc0DkWGDm9bBlxQW Mzc4tkep03RI7SWnDuM2sudHXkcTM7hqcqIYTYOUjfGqGDl10ZymQya+r54b//GNmbLxL4/TsjEY LqSPHza+842yxokU/pL6/R2RizQ3OPb9hJLwnB8vcoooC1/N5WeHEQKHzoWMKiLKV2vJjCDlYnw6 KZtGY+4UHQIX9fLzzGAEbfGJy5sZg3EzpIsfLlifmDCRIkxqqP4OLvxdueC1ivXHYN79rE+o/TQd gndq8neHEcM3j28mM0YTcar9xd4gZWO8KpbMGM9PE+2kb+L7KjN+5XMNueM3MQrlwcD+Drj4wenN iZLNZ/rm9KiJEqBUsPrrEMxYjFDywi8IhQm+y8avWJh8sTVZUYhhlyYhZJxLRQ2rQC+2khQBSsM4 dEL2jmrr9CzQuLD18aNzBxL6O7yMSHuUaG+PwaDBw0SlgftkaFBKGjgzq5efQ1p6aZ2v5cUjPFSK MrJSp+mQsdeUGClC4ARg+DtNRPnO3iDlYrw6KVpNa+80HXL2PmZjBxJeZbQA/ecftgSXDbE8/qmJ y8avLFlYerlvGDLo1poEEGlSUJJ0OWukeM1ZKVtyPZ+QrBgE5/Z3hxECX3NKXowmwnxnr0juK150 x6kTsq80Tp2iQ87cx1w0M+i6/JkuRkb1/2igkIvDCCGbQsmM3z0WjoVVpSSD0xdbgonEOWds09k4 UrxOqdRpOmTsNTNGiqqck/2TGaOJheN39gZpZoxXJ3Wen6ZDzt7HbDQz7M3MJABpvjkxbIglM3g2 EToVbc+GgwyFJOO//myQZganmHTkHSk6qSplq0gTY4CxVi4GIwJ3r3mXjCKifGeuyCTGOHVSF8tp OuTsfczFZsb3P+lhpPqlBBwuBiOEOEc924PdmK4pJBe8Pm39JRLnnLlNW+NIobNSq+iAsVYuRojW Cr72RF6cIt6uvDM3iL87p06KXVo1d5oOOXsfc7F5wdfjvJhI2+Xeu0SseZFXndRvPSIOChSSC/pq mgNBdI55GzvvUXVS3KeVyiQm2T/AWPNnJ0QEHOb9qogo35krEurPqZOyozROnaZDzt7HXGxepKky XHDxh4vBCAHXqN16QhwxKiITv+3ETNQgkxVMxjrcjBQFRKWq5v7EL9WUhxHBez9IIaGqyfhJl/Ub Y8S8vzuXBvMpNz6fpiJoOnsf87A5wQEeXkjuaz+S05wYjKvMJ3BkggOMmwHUhHWOY457/YM0JzyN Q3JGiBlAhaKHIAvEtRgrFSPkws7MdKqHd7DvjA3iz+pS+BopW0vj0ik6ZGL7W3WxTPAqyCuANB2F w4QYMREBBGiWUb8ZwEquTDgnmllHkDJBb4mrPlBRKZYp3io6YKyVihEigpxwQ1aoCPtEjiIuxZkr kmw6pwYzK8bv03RI7ZG6n2fFb/ZKhQp3fDUpCsGEHyWFCRqSJgM86V/XvuVHBO2PSgQzqhYbwUwB eo9KaNTs37i1PwoEefjuKoz3xmghPoUeUwM0IcafE7KzVFPVg4MLnK3PKaCbQ+oZ63kVORyI4RpP 8BRTlK89EY5CQiE946iF1ppBygILMK0yRoryuVJRhGsHjLWmwwjRQcHmsRAximjUeGeuyDwsx6mT oom05k7TIWfvYy7ysPxpj2WqfaRplpKLByMEXmfhHeVr3gfBhUJyQSt8r77IPC5pJ2mpOVK8LajU KTpkzJWMkSIEnnhQxkSvmgjznb1BmhZ1aoUev6voIsmNn+g+vTX4PQ0dXkyKXw9oKBeD2SearQwU bD0QjnJbKcn40d8NonN5J9YqY6QoPit1mg4ZeyVjpHg74rERL5polH9nr0jIP69W6olmNT3I2fs8 M2g08XIiTdU7bAxGDN3KQMnWM+GIXSnZ+NnfFYl3TNP4dBO6Tsr2uWg/TYeMvbIxUr4r8rBJeB1N xPnO3iD+7rxaKWIfz1fTg5y9j9nIEIL0fywulg2//dzcEMszI9+W1m6WQROnUrLhomQjF2lufKd/ +sFMeDZmvNFU5LHn7w4jBtY4QKCzmojz1R4+DVI2xquVeqJZTQ8CG/Vz2PhobQ/pH6yaxDu+4rts DMaF5IbOVWCdrg8ETtfsOMIWieaByLDxrzsg5CyYcdJXpdRpOoR5u5rKxkjx6pVXJGFjNNHlp3bZ H3uDlI3xqtj/fN6N56vpQWoPPz/PDYYRryfSvC1vagxkVlNDmBosSjQRbIbXNx70zQIBXXOaNnVW ZRiRkeFqnpoiayu/Ogj3bUZ60UOIb20FmAfG+DMyxDwur5oHSVx/JWEHEncJxzHeNpWDQZx2Z/co 9WtPiePpqJ/Mg/qb/F0G2geff6Zw7j+fgkNYyGq5fQivxvlfgjfl7ITlcXEW/DtsnQ9BvPjj5Sp4 kDHCpf/bQUHI5tBdLz2nt0zYA+F1+/QpVnsuHGH+16Kb/vvW07ve70yMx6d3wS3SV+bUUDEWqa3e A0LkB87/mOp99BBeSm7nV7FVoHfA+DMyTxRVk1viTVyf5z9lGVsewwG3rZd+EXynOZ/rQH3WU+EI DxkvzL/5Tf/upadRpCVFRVzMVm2V8JsABEv86Gj8FcFvRyEzf5QQ2hs7/bvRjyuBiHWdXSUHrJ3P 7n1ma0SX4L/3ABx+PhDFTpvxKcl6jBuFTTYOO8XJrwo0fF7+W3ysDCv/yOTeHzVF1tYQoBAQL79/ e64HDIwenmpvbQWYe3/8GZnH5VXzICyUo+ZvCdB7nyUdJsh1rAsx/H4gSEgPPknmhqbEl4cUU5v+ iIAB7HNxBsSb9Ecmew9UvFrIAXZWPKaaBAPhvKetrR6oI763tgrwq3NnRB6HV8uDQEHWiT7PA7ox Wjt+bzs3vx/IotfzE7HJUmonGX/Sze1MK7+Cg2nmdqqV14OFzHr2IKh51TwIjvHVxnIwQnjPx+7C weghwLe2CpQD/UH1yIzLeZ5UzRNEbP01D1ok8U7k92wsco9cnwRCyWJOqSkJ3AteQS5ltjEx0+qv ChiOU628FVwZHhOVYUkeNSkEB6EVCzX+6oR428+HA171UL69tVWgJOhPkiUyhDxRrJoHWVufZ8KP P7Rc+5jiiZ7TflAglmvIwnNZ4Ov1vd050nXnWJEhZpHyQNns+D+YMbKH4lXTg4y9MjFShMB3JcLE aLKQfWtvkHIxXq3U4/lqGuSJ72+Phk2JX2QukfP2kY0OkxODMTWmy7wVmN1vxsm72JZLaZA7pGz8 4Cvwr1LMR6t9NcHGIGOvbAzmZZiqfDQxLXhnr4g3Fbdr+gS9PSpF7OP5anqQs/dxbuyD8rcHFoYN P1rTO2QwYkibOXbzNbzErhRPeGZZ+R38gOSWd56Vd4LFkv9sP1D7aoKNQfBOTWVjMGLgnUkeu6OJ ON/ZG6RsjFcrRezj+Wp6kLP3MRubG7zRacn4Dy0Cw4YY2UUMaTOHDQ7yN4OIUynZ4J3L5IGIUTnP 8kl2Qq5eR6iK8sQYBOf2ZydFCN8ZSB9FRPnO2iAMVI9PK0XkE8sqepAz9zEXzQynpSn9aVlgBT9c HOayGRdWu/SKe4VdN8shix4Snd8VmcygB6RkjJTbDSK1msiMQcZeM2MwnoB+/xpSRxErI+/MDdLE GKdW6uc5HkX4XYCretY+5qJ5wfKNT0ZkmYoOE0Hw/j9eMssDXwRv2y49kM2JKZ7ydzMiH3YKV5Wh xu44Wi1wkL9rJ8wNgts7Q2NK4f3Pou2Llfm7wevJiwyx5jPd1ZG/nlg+f1aymsmTv9Ke8uTToVju B5ZZjZ1ylRexSHHFlfJ++NHfDWL0TnzS9XdSlMZIvWgifpHHnr87DP/TaHCKCPqduSJ5VJ5TKzWO o3wUPQB8q+hvXOz9wCtIuafJiSP6hovBWGF36iwXnG8VKjg6z42eTrXyMyZLILmozoCmWXGkqJEr tYoIe5AxVyoGo5D7Y3vyKaKOe2duEDNifVohAh+3oyfXVCB3Q2L7GxN7N9Bq0fuBr4AsE4MRAINJ JjF8/7r3fIY2qeDzy60wgjQraPboODtSlJuVWk3c9INAhfb83WGEkHmW07coIkqFXswN4mMS6Tq1 UnAxjo+iBzhrHz8ZNitotdB/etL4KOtkxWC+grJ9GrN88r4VJifK1rk//R2v+UEmLWhxspIqlijZ GqH21UTcg4y9cjEYUxw+i2ZajCJPCntrbpCmxTi1UoQ+jkdR8kIgeZHovsoLOi30H9/4mOBwMRgR cBVzCTipoVmA/Vae//R3xC3SvGAZsWNspYicjRGtM6vpQcZeuRgpQvBUXrgYRUT5ztwgzYtx6qTQ WXOjCHLeRvc3LjYvftP2HC5++maoT87BeEvLGVDJCz5S3+ynSa1c8CGWxi0yecGHXDvGVorI2V6g 1Gp6EK6hmsrFSPkAYFFGMqLpj29kbI5y3Iq9Q5oY8eown/nj+Wp6kLP38V2yTwx2dRsB0p7PXTbE 8vSn3a1suMsUKZbQlNK7bPo8pJlBQd5aq1LEnqI9bETTII+9sqEUGDHwPiCpEU3GiTk9qL3/HVI2 4tVhsjHRrKYHmfguNz5cq6b8+p1ZA9IcFhs2DuOK2mvgM4OP1E8mIJUajwNWJ3J/VzZYZOmAw/IV V5TYOWJVqdX0ILWXnDopYuCgsLARTcZJtwX9h7ARe4eUjXi1GF6d56vpQSa+Y+OjkzaRpjXZCHgS c/JfFqsPo4eDTw+GDXbh9R6ghbuZS/NwIw8iG7xzggPTRaj1YoVWEWTA6mNubpTBKBo8onQVESUr uLwwkAutmRlFykV8OkwuJpZV9CCNDnNzn/wlM2weCxVsA57EKGQ3i93T5No3BBqkQvrGYUbcSEQt UCLo+ejgG8y0cKJpVlTPATi2v1oZvP83Z4Q7y0ONEaKGx2ltPYA0jEMvQi5Taqt6EAmQYWRsffaw YG3fL1jLAl+NWxYGsyeN/Rpa/U4JHBpYMioNnJ/QmIOUBxo+OvBWikcg2yWUWk0PMvb83UlZ12PG jIgmomRJz8cHVMTeIt5W69ViCX08X02Q8Ta+S4l3t0cHEt8FpuS3aOvpeWgYjK4bvm4YuzNY/qOM vnHyVeMWyS3vxCzTC97aRYoo2Rih1Op5kLFWLkYKpfQ/c1eNJqJkVctGrrV3SLnAp0NMgvG6esKN SPIisf2NiQ4ivKDnOK7kBYFsXgxGBLypSl4wBDQL6JWpb5xK04IjiDHx7ovjZ5M9lSJytkUotZoe BC7UVC5GihjYL/BoMk6E+tiMvUUmL+LVYol9PF9N8PM2vr+xsXnBG8UWnnxjsy856DUpRgzfGEnN R5ZTJw+UclihI3IiB5nMYBdyB14+84kUsbMtQqnV9CCwoUzZGCli8HMs8hpNxvndc9TP3iHNjHh1 mLkxnq+mBzl7nz8z6Lxo6Uk/9rIhRnahnGWhsMHqg9eYl5dK6d23/m6Q5gZvf1tcVYrYWX141TQI mngMqalsKAVGDP9Y2MJGNBEnq3yeDzH2HEaKlI14dZhsTDSr6UHO3mfDCM94Tr6rd0zEM44cBht0 DIcNvkvsNSZ2pLh3nB3ld0VmJOHwVsfUwVJEVmo1wQbbCx57yanDeEz+tLSFjWgyzh+8hwwbsXdI 2YhXi+HVeb6aHgQ2jO9vd0qfG9RBnJ1SNjg/IwXGYcSQnmHuFK6jV484kYp3LO1M5CLNDfaZtLhS KiU1hbRSqwk2BskQtbkxmIOGpS1sRJNxskTKoifsx94i2luvDpONiWY1PUjtwf5ndwrzMN8Bm7lM kzc3BuMRRNcwdinqOHWNe4dXnUrpXc54O0TvmCyxZSmcVYrY84ZZNqppkDzjY08WT4oYPE3oNBEn p8+6jLj2DtHeeHWYbIznp+mQie8rNths4OtqpJ1ROTFh3QEsVzTzIO1yFEMzIfMgvePbQRO5vysb 3/JRcigTMkxK1ldFDzLmSsZIUcb99JUQtKLIkdPpG0/VtbZIrI1Ph734fYoukpr7MjPYeN/E8EMd paIQV/ObR+jBBGvSzX/q4j4z2BjfFPBXJYIeEYdgmMgnP0h292ckK6LnABybXxXiSuA9/IeGqDFC ZER42YGaA5oScegwaajP1ZPnx5u4vkoIGkL0nQViz2EpC2K4RgXjySjS8K9fKTVCNsP19mCi0JiD lAe6RcpNpYiRN4OvmgZ57JkRlQIjBPdUmRHRRJR8UGuGce05kBQpF/HqMLmYaE7TIRPfsfHhfIR3 myxZ1zu/WVw2BoMNugzDhp/SDRkKeZ04atS6DnpESgYrHR17gxkmOwyaFVX0IDg3vxsseUHRUDKi yTA5kfSxZ2IUKRk6dZBcjN9M5VCUxHgb3XHxrhDfYYQuDQPAObaMDxdizQwOHw8XHCY9ZCjlI5wJ /QQuUjKY+HfoddN4rjlbDF41EbjIY4/fPRgxkIVo58KqyThZKtGD2ktmBCkZ8epFiluz2k/TIRPf sfGXzKBhwwi81lyEZsZgzHB+u0OL+4QX2b0D2FO707b8zsvB78oGp1B36A1mnLTdtvyspgepvS1b K+XkfKYh0UScjF72WcAGp2yHnyJlI16tVDJhPD9NVgGv8R0bH+aGfZBUmGGDLymnxDgMNogmbFAl NxN+IhXvKAYmcpGyQcddy0+lUlSz2UCp1QQbg8DG/O7BeFPOQdZop3pUk3HSemSJQS2tpkVij2JT rw7jUmw0p+mQ2vtiIKE3ksOYy8Z3FrTNjcOIgSMewwZNNtOHoZTe/dffkRsiZYOzqzr0BjM32G4w bETTINyFeDe/e6SIwbPDZSOajPONuQWaGfFpoRe3T80FcsY+q7bshMyOJ1bz7JIoFWK5nOlt8NHE Xq+mQboypIKvvDQJgugcMzfu2TxVK0WQrOK+anqQscfvHowY6N14NBGmTRlWW2PvEO2NV4fJxnh+ mt7H99Vtwvq1FxPvOMd/2BiMMTUN1djN6rijxw+l9I417SZBkLLhmja6Voo4WdRWajU9yNgrGyPF mOoZp/IaTcaZjwucvUV82KxXi4GwSl17p+mQs/dxbnQGb/dk5hJIe/5yc2MwYmDrYuzmExKJUym9 44MOrcE4FCvdVsTAgl3JqBA1ZIVO0SFjzqAOI4R/dwqIIqNEj+yMtUPKRXw6TC7G7yrid0VyUya6 rzKDAtMAKPFYcBsuBqPoZJ9BKzm+uZUwmcKWC06Ta+BB9M55VF48UglWylfAkTpNh4y9kjFSxMCB GI8m4vwHRS/2FpnMYL9t6vWV8g137J2mQ87e55lhY1rI8JzIJkYhC3CWMMzHH35CzrzIEZAOtv0a VYG4xqTNvcAvQl4fVZ+eBaytsVUiKuPbkJ37uWHZqHlOdWDX1gFNijh0mClQn0/PArX15ShCrW0l ZdMbQ3FZGIzZT3qroeFnvgDPVEchWaBRvQVYkGYEV68D70jxAKvUKSqSfI25EjFSRPCP5+aRW9H0 h6kdu2MtMMbeIaVCpw6SifH7FB0y0X11e/Css6BGmqMEhguxXErW+nSFapsHQq42f5QMRv5e/yAl 4xfNnxlGRoqi552mQ8ZeyRgpHhp+Ck8yosk435hboFTEp4WkYtw+NYfU2BdpQUsl60lNC/frmBaH cTHSdQ0V9DJ2BpJdNl4nOjQmB/xdqaDboePuSFFOV+o0FSn1+7uTIgZ3EUlFNBkmQo6oY++QkhGv Dnvx/DRdLLBRex8/KTopsfnymxeTO51td8OGWBKD94xNDD7W1thpaG9i0HLZCUgQvaPu5wTrJMZI sajyTtMhY8/fHUYMfsLnNBEn9tNnTNWnpkPKRrw6TDbG89P0Pr7PbxNaLTkSOLnBVtZlYzAejz/c gUEG8pJvYlfKa/Wzv2PxSKRsfOuBioOlgKzUaaKkfGOvbAzG4pht07IRTcaJcuuvsbdIHtXj1WJI XTSn6RDYr58f50brC1stM5FAmo9yTW4M5lJa1mCp4Hzb6xDB2aJl41d/VyTeMY3qwTwnRTVaqdN0 yNiTjcNcKnQyAxvRZJyY895hkqimQ/zdeHWYbIznp+mQs/cxG3unZMbsnfKfH0Lpc0OMjGfx10+h aNdpdNjIR070jhXR1p5Bmht8Ybpj70j57vutpiLJxdgrG0qBEcM/M+WJJuJ01dtsqT0HjyJlI14d JhsTzWk6ZOK7O+XD6bvNlmkSRNrZX9kYDDbYXhA20pdobvBNWyswXpr2d0XmTmGsWRl+RcNMZU7P IWOtHTeRolxgjXPmIOgxRg4n0PrYOqRMxKPDZEKvrU1Wz/vYjokPp+40WDLN7nViG62fXULnYBS/ ac8mL/JpzkSplIMtX+acHBBpXtBi0gGYayxf1OWVOk2HwMX+bjBj4DM/LgE7H0OTcXIAjlP3sbdI 7I1Xh53nL5oultr7cnT149feJDmrpmlRiAg80MZ7hPc5jdyDcOKb7ZOSk5NxSgVdKK08gknFfxE6 PQuMLX8VyGvpWTY+KZ35oYYIc96NNGhq/25KxJ2F5ECHV0ku55ugvsoHmjpadnLM6t4ZYnlOcNhs OfiVBk5yVCH9Yr7kzwYpB7SVOOyeFE+qStFMhqIkyCDcI2oqC8GIgK8lUrdKQzQRI0MYz6ezt8gM ITq1EEL2kOLUi6JBvMBj7uNnZkcQmywzfUCaV+B5Shxm0x2FtvnApzgndKV07r/+jsBBZgThTmpV IoYU5WOlTtMhY69kRMoY6Mdz/PZNHZqME0U+M8feIU2LeHUY+dxoXjRdLGfvYzY6gth6iePh7h+a Srw5iiUzWFOWDeo3j2vIRVdK73jDN2kgoncMmjz4X6Wcled0eFK+moo89sqGUrmiFBXaG03GybK0 43ftOYIU8Xfj1WHr+aum9/HdffLhCGI7ZloXGehsMSgbgxHDT7c7YZevdjR2DoXw6jFh6+8Y6Pxd 2WB5sutYwcwNNi+YLafpkNpLTg3GFaXYb+kSRYTJ7M6jbtbcISUjTi2GUxPMo+h9dMfFh2OI7ZiZ R2TQ/53l38OIIGngQQMcAGViUFE0MX71Z4OUirzrWqmkO/swTJVVRJ3+ai0MDmYA9Ew5gjDB96WZ QRKf18LpIdvADikVceowqUgsL5oukkaHvblL/pIXnPfqtUTaU36bF4MxrGUJi0LWZetEqZDO/eZn 3EhwIVIuOLOuixXBvPW/VfkqOmCs+bNijiIedNx6O4qM8tVayFlj49IjY1cpgayaZMnb0I6IvyQF X6+02KMG/sVTtESI5XHRq0QN64H8hs1cukTw5ZauYgYpEdweDrAnRRVdKYiw5ZugBxhrJSIYEbDj bkrtKCJIziVgkJT3mFskt9U4tRhSdovi1IumQS66v3GxA0k6YiJNy9BwMRghpLWQCpYumUapkCML ZVQT4BdInHPu1MFGKJebOkqh1cNrhAJQoUipCGYA3ITOQ6rHGJHxhhljh/izcekwmYjXqwiZIlzT M/f5IMK22lZYTLaXicHw3LP8MJvtubneCukc+2onA0SaFfSUDBWVgoFKRRE/K1Df9mcjZATMRnbC hyKjRMgbZswdUi50ysfHStkuilMvmgYJF7X3GRc2WeIbs2i333p/iORKuqdMHn734DculiI6Nr02 +bss0HHS8VYst0LmWHAQJTw21TlW/EkQvHZTrUp9/KKCyNx3a0GhGSPt340+rhw2vr5oGd/PzseR 936w8TITBaTpsZ7YB8PxBv+Hh2RuB1YTfYCzPaK/gg6QuR3oHukAO1LZp2BajJ77e2yVAYqpFIbs 8M/0gkpNPcbXL7ettUPKQnw6TBbitTyMpvex/e3J0KIKe37ZIdfHnTnNArE8JXPGKbfYP5P8dAWU iR/9lXcxiL45+aAI8KYZKWca0c1rT05i9cIPABX7s8EMwG1Hp8cY/ThLeI+1Q5y+1qWDJCKBPHre R/Y3HjYj6N3wOvLSlo9wDg+D4X8e39SrPaGN1guFTImf/VmRpIRTpLxNPCkq5kqtogPGWnMiQgTA V1U4m04mooggOWuNG+TMHeLvxqnDpCJ+v2ga5KL7GxebE7RueCXxjn1pw4VYc8JXXdrlU/e98fNl L73jxIlmQJBmBWuZHWBHioL5naYijz1/N5gx8N0yxwdneGgyzn//zP2oPZ8TRcpGvDpMNhLNi6ZB Lr5j48OCykbKTBVgg6WAYWMwhv2c3wEbfIu3zwG6Y5xZOi2bKYbI3CPs5+4QGyxlYqVOE4Xjq715 ygQzBl7td/4ZTcap8jO3QLmITwtJhW5bma2a98EdFR+WVKyG8JXwXCg+fA4VhHkYAXiAi4nBd+6b Bv8p5TjCBLVJEKSJwSHArS3E0MSKfaVO0yEscaupiREpYvjOYcU73/L9HwMAQjYWjb1F8rvx6rDz /EXTxVJ7UP/xILK3CR8r6tA2H5aCjIGogV13kgzeabd1+d/MlZ2R5Ve+oGiPhVOy6VOsDKv1lTk1 h/jqo6+cB8J9nhKpxEbPH6oAFq8cR8fYIc0KPfq/xXIJ4/WLposj5uDvcx746olXEmHaJTuCDEQA vNkJD1zBpkD6b3GN69frL9CU8JWeA0hFeH4ikmfOaCmypiYfFLII4mO2PiOdZlFMQAKNLubHWFok N5SHAzi5eKR4c/RWzwAT1l856PDhm9nMEODAGwOzB8HBP+7wIhd429W7P6eX4xnutsQAmMEjw09I yL3DaFqZU3NIbfWmiJAk0LvRsSN6DHBeQo2xQ5oL8eiw8/pF08Wx5j7LBRzItideYHJvloYizJBZ JyTbzM6ZVHS9xOlOy2j/biLk/DZJqIjzrpZUKoHKAKldtVMKFMFvOuEzn3BmxW1BbH5zvo8hDR1Q AuLMYrjXCF70vA3p81vBvrdMCYifvZWNfyAIaFrznOqZbkQ3leaP/qqAfjkZyhvDlaFGRib3wqgp knsBW8OBQiGBr+N0IhE9xvdqqn8n4cabk5gQJGB0vA3qyxuBToXWk5SMJWAQCGizOZVZ7vKMFs61 Zv7A342dPhBLDWL3tqGwrMQpOASa+5tB9JrGBAdIp1CkhnF5+2VUiKVFYktXDvDix9cXNYP4FIgp kv+vR7/ZNZm5gL0MbV0+iKUyDhnEL0ox1sgSP5V73PrdX9nJ0AZupz95J1hICth0YHVxag6pLXkr ZDnk+fotnKOHAJlu8LjzWRhjh/i78egwaTCQV00Xx5r7/CHAVKj1Ix+anyQQ4poSgH1RWuVT9X3a f0s9ydJ4f0XQAM0EJjUdNCrj289orhrYOwTH5lcDycM3PkWGLSdu6DFCFtQ7MMbYIU4pxqPD5CGB vGgaxHSouY952EGBTfuustHj48nBvRsGY3X5J+OWZjlMtZUjO8M6TnBqQFfnRHKb8qrpJ63o3hQj lc0FSp2mQ8Zek4KuPkZ7RvTfvMohzNFEmCzaUqf5RI69Q5oU8eowyYjnL5oGufi+ejbiQIvHvHMt G2JJC9ZkywYzrBaPlKplg/X8JkGQJgYvvvoQGSmbrt5qKlL21VQ2lAobzEaSGVFkmP/9yHtFp1OK DDLPyDi1GL/jDdlbRQOEi1r7PDP+0PRb377xkCsXg3E9201NOdqD37jmSnmlvvd3RSYzWKRtETFS vICo1Gk6hMxQU7mIlNfzO4duJjOiyTC/5TTZtXdIMyNeHSYZ8fxF0yBho/Y+ZiOF9C/eodOJEjbY 7Rw2BktmsAzrUGYGZxLh5NnzyvCOiaC/G6SZ4aePD/IBwVsOhdIttdsOXs2VjEgRAq/wssvB6Rvr Ma5C5+S4tXZIHxnx6TBWdRvLo2gRuUh0f7tL8sxAmlMEvJr8x3/MIMiMByOE9lIzzNlJnDCVkgtW E/0dXIAkM5gGfZ/t7GJIuccgUqfpkLFXMiJlDNTWHaOiyThR9GLvEH83Xh12nj+anljO3ueZ4eEy IUO/S0YhIvAgPK3mvJlErpBc+MbnAF1zIpS3gtBTITcYKFQ94QaAhJsrUCKU0Xt43PkUaowQLT6v auuA0qBD/ztMGvT50TPAi63PSaDnwtKPriw+CTssiMV9Vl29PajSevgbC2pK6RqfNmjVHKQ80Ovh NpeTytaCN5oOGXv+bjBC8POfM6tSE1GyjBsPxt4h5SJeHSYX8fzRNMgT399ukH1Y0AnSnGC9cNkY zJ49z52AjX9TZBGmQpLBqfpNiiAlg8RwrDmpbC5oVqgogbPg0rSIuZIh5vWkGTV7KZxbockwMWeh NfYOKRk4FcYqROTx+1G0SK61jn/FBbMhryZX6luPdUGDWDODYR1XKGF7+BthKqVzfyozSMmgPcSF ipOign6n6ZCxVzIiZQycUDHTLDUZOl958bYZe4eUjHh1mJkRzx9NgzzxfcUGz+6WnW7UmPtkMDo/ eFuTzODaNRGorVpycBhY0yBI2aAKsehg9b5S1OSVOk2H8MxXpmxEihiyTu77K9QQJGu6fqWQUSzG DikVuiRhI0XgcXvU7J9e57H0+fOCEtsribRT4T41xZIVFD7NCtocOlLyWaLywPlMyvjGAqQ8UDH3 helIUT6+03TI2CsPkTIA3tl6AzLPU5NBkmhW5cy91HRIqYhXh5kV8fzRNMgT3+dZwVvBfksZ7/iY RNg4jGvJuk2ygrXsyQGl9I7GickAkbLBftoWHL8qlX0FSp2mQ8Ze2WA/QsrOHzmnUjaiyTh5hedd M/YWyQBOUalXi1EKjOeyUU2L5GrXz49zo+WFLQRMI5jgutrVp+dgxjCzRP6VfqNkwk8XxfTuv/0o oMiUF6ywteQYqewvsCw5TYOcvbLBtoR2RvygrQr2mfqpiThdEzRbxt4hzY14dZhsNBqkq+mQje+r 3OCbvFk7pDkal7xTaEUCI7uIgQbilF/0KDQTOO+0bLAHpHkQRO+cRDWBRsgdBhGqIsIcRDJirmRE igtKF+bNxnLN2VeKT3IRa8bdn/3PuyI+HbZ+OyiNookk13rMfZ4ZTMDatpdbt1wEkwtqi3JBU3YL T87qLhecw9IsEElmMIfiAz/Jn5Fyj0GkVhNkBIEMzurmMVAyinEv/GDZEspGE3GyuouUbMTeIfwO NuLVYbLRaE7TIY+9j9lolcE1o31J53j1loVMFAiFC5YYcw1o4WsWUAn1StEw2LAFmhb0gAwVFaIq f9FDRAHSdBlb/qoyXEtatLMK7MQPNUYYIXnQ1gGlQYd8xI7Q+PzoWWDj+ur28HBkPftFi29WtVEg Fho4j4kLAvk5Ctlhg4lvaUjX7iHlgQaQvlMeKUryt5qIOsjZKxPFGCx+dvWQskJNRMkCZV7CMjNU 0yHlIl4dZkrEc8gYTYM88X3FBv0XVH2w4dpq749A5gRTaLpaIOO/eePN9+06zOYDmuFCpFzQJtI8 GSnqxx7QUEXJAJH4tj8bIQOgL0LlTkPsRoSKN9YWKBF6NAgeOt8xjNNyyAT2FQv0YeTOsCPpm0vc KAgmDSy5NidYQWjWc6JmeWAdoDEHKQ8/CTL3i1gGAjYZ9CarJnIiCKzH3gzDxXj3ybjRajua/jCQ xKvMCuPnIP6OixOvVurxHDJG08ay9nicff6YoB+DhzXesWK6xUWwsME3HWKXOnr6KRDjwjlZm5kH q8KbFSz2dNAN5lOATQbyA6/RNIhZUXuyOFLGwNTcR6QTMd/wEXu8OnuHlI14dZjXsdGcpkMee5+x wfo8+RjvONWpt0ixxpClV64CX6afTEAsbPBNsYlcpLnxm3fVDMO0TFQq2wqUgo1dxGXrAXHiXeyV jWLEwFcquroZTcYZKdmIvUPKRrw6jO8aJxpvlNE0yBPf57mRd7cUfSyd5wPs3imDGQODQB8YtA30 HuBbcmWDDrmWG0HKBrsdO+yOVLYVtPCsJvKATRMv9spGMWYd7Bux6HYqhibiZD04uwidHvq7QRzA uUbxarHHc9gYTRvLxfc3Nlp48k6fRaWwQQHREXWwsOFZr2Yg37hvJnAok3e27PE780Ak5YU+TMfh SOl7pFAQTbARRO9ir2wU4/cUI66FjybjjNTZO0Q2xqvDvI6N5jQd8tj7/E75l0onbPAB1jT2o0Es hec/uY+xyycRJxPo/wgbP/1d2BAxKh7k7c7CqUqR6W81wYbIY69sFCOG71PjR5FhxinJiLlFkhrj 1GJJjQaziiaSV2sfc7GZwUnqKTxp9GnDMhqCOanhiJpmBvvfmgchReeYUrfwFJnMyOZqbrPu0rAS iszqsaAUIS9qTSYW496AJVfCLby8I5iFpzt6bB3SrIhHh5kVjWT0hBoRnxhr7WMmWnTaYJ7BhLMK eTvRB4ZQeKCvPLn4HwVgnovfeEOTnPgTEYIWaErQP9Lhd4SoH9/oIWYBHdNWyBsI7/nW50767FKE hzgk59o6oDTo0GFUGBPG6VlgbOHh5ySk1UfP2JWY+sL1iEwnuIp0lIcFdlTOtXfvok8sm39yZ1Bo LA0sX3epRyzTCTZIKLWa4CHImJuhpxABcIWl2BmYRRREvFr73wLlIS4tJA0N5NQc0tD+ysPeFvl+ F47RDzpreyzNudNc93lLUSL8ppfPA16Rl4dsRl9k7goe8K9SRE2B8KpokTNnGo0UtSMftWZFSSKi iSBZBnYpEd5j75BSobmDpKKxjCJ+VsQbo9F9lRN0aaTo5NMWvPfonREsXDCAtbKlDigZiMW57/4u ZIgYlPOmvGB0fM8pE1TRlbIyUROBB9G72CsZxYiBLicLi9FknJGSjNhbZJ6XfJshN9FK+baqr5ZG E2y8je9vbDQzbI3kWaB33z0JVDYGCxu86QChKO/BcMSJWNigym6pITKpwTu1DriVInYK7T5yGFKN c5CzVzYiZWrwHsTeXidhaDL2eCUbsXdIUyNeHXaeP5o2lhd7nz0vaBL5h1f6SPuBqJxcggYxHpKu o7qoAhscJWom8B8cBBQ2fvi7Q4zKF1h5xbhSxM52g1dNgzz2yoZSTrJ+8a7RMms0ESd7NlPasNav pkPKRrw6TDYazWk6ZOO73PioWRVp1hNTBH7/MQdVLBY26BEndoYKNs82djayZ97KfqCWk39Acqfw Ys3+pDAkZiaw4UAp+yPVtIjsa2/GkUjJBrMU1zZHk3HGK3Mj9g6RjfFqMbzaaE7TIWPvr8/QDqi2 SPLuGO++0Yv7J4PJYMbwk8Wd2OXz9s2E37SlJzd++TsjD9LcYLOHpVexZPw/lVpN3ANB1t48b4px F/3hxNGwEU3E+V+8ko3YW8TfwUa8Okw2Gs1pOmTju9z4qHsX6R8chBY23LbZx0agkEGFErNU59OR kc2dOPfbnyVwkZLB0f+Ov5DRLaDMkioFGdEEGSI+4lmj6M9GiMzgbRkTRDMjiozyjbUFyoQu+WRR JmnRQE7N+9C+4oHNkSkBye8fXcKwpxIsRLikIv9Zp0qIiCUr/lSGoEVKxJ6RI5SkcHXpRRE8BJEH zU1SFCOCf/+9WVhGCT6EYNsLtGvN50WRMhGfDjMnGsspOuQxN0/PvzwvmIaaEnSAcJGbE0IywdmV ZeKHR23LBH0XMuGEyIfiACWCbhIH3xUibpuZHz0DnC1/VRnzgc8lS5/TPTsYWeuKQ2PrAUpDHHoR sh00z9/owcMA5l7j+ioheKeNUaT5HFDLC7orxeI+feZkHWUtE45GiFho+FYZYhYpDzSSWFkNZozs P3jVtMjZKxORgooffM4Q2qEimowyXklF7C2S8mK8WiyhN5rTBBlv4/sbGy0vbKVkh5ve0Sfe58Rg YcNPlstGqyjCtJtc59yW1sDteC8ZlBAddStF6KzqtPDMF9KlJ8iZKxnFCIE+eN8DOAFDk2HGqbN3 CL8bpw5iF8zEMooQKZLMiKK/cdEBhJZKjzhk1YOD96mKvD8Gkws26zUz+Dh9c5+WvpYbOWTRLAhS Mmgm6aBbKUJnO4X0rKZFzl7JiJSZQZ9Fh9NoIk4m7nglGbG3yGRGvFrs8fzRtLG82Pu41NrMoJmD whM2OKYnOwbRECwx0HqOFUpRDgVPIjB3rXdcxokcJN45X8srRhiKFLGzC6GFJzPKviAOcvbKRjHm Jo4cKV2iyTjjlWzE3iGmxnh1mNex0TyaFtn4vsoN5thhw8PiO12nA9MdpLLBK8OywQlVvQVo8i4b vC0tG0H0zllUXjH6kiJFNz681QQ/IrJfe2WjGDEwU2VAXU3EyWKnPQiwoT0HkiJlI14dJhuN5tG0 yGPvs4HEbkvGRL3jLKe+WB4sbGT1kavATTexI8bV4/nh77xTfoD4DCWG75xB82De/rym72DC0u0U 2K/25nETKfKbX7g2spqMM16dvUPKRrxaDK82mkfTIhvfV7lB1yAlC9JsYMieGHQGCxseymtO4uHE nn0OsMG8vmV4DuAtG0yRO/yK5WHIloqW4f9GE/wEWXszEhUjBsqnFjDRZJzxSjZib5GwP14dJhuN 5tG0yMb3FRtcjrzwx23WuPsUDSYbvNksGxw13MqTpsPeKRxt3DwIUjY4arijcKWInZOFlVpNg9i+ Xnv8bqXMDfZPqN0JGZqIkzmfK++wEXuHNDfi1WGy0WgeTYs89j5+iu6Y8vsf2wORplOg03e6MMEa A737yQ1LmdwDdFzqr9OjmYAEKRt7LGGFCJP1rVdFi5y5khEpyGC9gqUayfAEQqPUp1Q7sbbIZEZ8 Okwu4veLokUmOu7Lz7nI1wZ0jrFxa41guZ7074cLu21Sb/IxoVYfrDFMFoiUC9pKOgpXitBpkmkt znI/mgYJ97FXMiJlDASkPadjaDLOSMl97B2iPUpMvTpMNhrNo2mRje9zNmiyZCkWMn4yB+9LgUDJ C443Lxe//VyYz8ts7tS3fk9pgFJBU4kvHFeIwNsCyEATPQOcrRKhjDzQUx8a1PJ/f+jU0J8khaYW mJyIP4dR8xvFi5r5e6OCgv/9P//3//7f//fhHNUGTH+NNB8+7KA6mAlBW7pDGpvlZ8zgOtUxPuLY ay8y+cBifkfeYF59Xn21Eq+iRdbcjCKRgodvrF91jTOajDFOyXnsHdJ8yPrXCMHWxvIoWmSj+yod aOeAf1af/0C/T00aMIWkgjZCqaCedXOblxrfem9wRlXrryD65oQtbxkr1YL6rSa4IDPOnBQeRASs lfYRGUUEThfwY+1/C2hsXFrovJbRUTNxTGjY+vgh0bLTPktrf3jgC1EtLQYLEazNxyotdhO2H5Ly IvEJlCaAH41K2ckUipK5dEWKGNm80CKcQ4fRtMjZKxORIgRfUPqAHE1GGa/O3iHlIl4dJhmN5tG0 yMb3NzY6fLCXySNE9e5ffMlgWixlZ/r0vQashk7siIUN1iZbenEAZG8R52x5wUhaVIq8QupVU5HH nlEtRgw8hDpUR5Nx8h/eNE4Q0XRI2YhXh8lGo3k0LbLxHRsfrl/YbdntTUzSCaU3STGvLI1B+ELx 9s9MQJhw7ZRtph8ieWAwifJDfN5KwYyTzRhZHllNizCp115yajFi4DRSuR5NxPkrXskG2w7QtEjZ iFeL5YlRzx9N7+M7Nv7y9GQYTdlJMc4KTdkoBhs0q3tlKN7czZA4fVHTydFMPxj9lw3OI2nZKYZ3 2cWQteXVtAhsaG8eu+x1yGrdHz7H5pDldMw3hMSuFJqcjCkzSH43Xh1mbtRzXl6PpkXG3hfFhe2W TCO4U9hdz9OvbBTzRSb7csMGbRAtO9nS0SGWbpvJAxGvFZOoH9O/OFLuY1D7o6mI43ft9U6pFDH8 l6NTVpNxxivZiL1Dmhvx6jDZqOePpkUeex8/Rfe5wUfGUnayaYqMLxtixuACkRMmNs/zEbXkBg+C ssFsvnkQpGxwSHPKixFyI8NbRYuwwhdzJaNShMAZ0mHVw54JkpWu3JbOD9UzyCRGXDpMKur26Vng jP3vMyZsrrTwR5rbtcsYi8EEH6ex+qGQ5T1fmWCxI1nLtWrpxX2+eUFLSYffYN4lrN9ldXk1LTL2 5plRKXot+GqMdyVTMTUZZrwyL2LvkOZFvFoMr9bzR9MiG9/nzwwmgx312Sqck/PQGAQqeE/h44Eq 1g13PjA4db5J0e9WDtCcoHmi469CeV640cLkGj2+jUuBoal5WATB9//4JnBYiBbje7HkKJK/S0G8 WUgG4u+j5E1I2Pk4GVpc2FRpre8qvd283haLWS9yCDAINexMuPmGSQngy1AtJUVyTZmgucfDmmSk +I8ofxQtsuZ6W1SK7gI+berzcTQRIgvdOGUuxN4h/q5OHfT4PYouEh9Jje5vXPQRwasGF96U5msU nacvRgjp9qVy4zyJFAi0+JWKX/4qYYvomvOlvGAk3kqRKdG9evbvsRUCF8N/3rB3zhs9xhiPJCLW DikR+uStE6ncFI3k0bRxbGxfMUEzR0pOXsfS7N2sKOa6/HemYF4A9jFM5IiBsNLu78KFSLmgd75r V39sXPZWZ9sC2h9Ni+Cd9vK7xYiB71qonZmemoxTKTSNvUXyu/HqMPOinj+aFhl7XwykvNL3bS95 wdo/NWTZKEYMtOqbohRu7H1KZnDKvneN07TMOniFKFI26GPq2tVIUXtG+6OpiFlbe/5upfCFlka1 OwfztSB2e/ro2jukuRGvDpONev5oWuSx9/ETY+8SVvR4kY+0R1MNG2K5orzYcyDlGtG8m0xgwVp/ mRYpY24EKRs/6OQuZ5WiJH+naZG1VzYqpecz11FRyNApUmPMGXjcLBlx6jDJqOOnaIGz9sVQyrjN NUDaczeGi2KsvtA/5VDqkMbrsUTOQVBx7pu/W2Qyg+PjunTFhlIZy44LtD+aFhl789RgXwZShMDn D7wvnfjZ7Ujo8UoyYu+QkhGvFsMrh9C3mhbZ+L56avCWKAUnHfY8JHufFCMGFs98plH4cf0TOw06 zQze41h0MUCKNDMgwTG3GN75akDtj6ZF7ELFXn63GHc4+0O8L2m1UhNx8rLNEyBgI/YWye/Gq8Nk o54/mhYZe188NZztW/bbFDGvRxYihDRKSQbHi1t9u+aW1xXO1TLvGKRk9DMYJ8Rbg+h+FC1Sa5MY FSIAduVaXTgL4zkGE29sLdCs0KEXmfqcvp3RslFMYJ/nBKtp9HGEhbTZcmTar8VYrGVW4JOdkq3n whGhFYPXiC+1z/UX0TemTZ4bFbLAkhN8Fz7F5mqyN+HVnr9bDCIoULwjR5OB61VyIvYWib3x6rDH 80fTxgIZ0fQVG/nUna2hNMPkceFLhYyFPVgAKmi372OA53zT1ZUaoxYYJugt6cgbzBjdzcrNMWrm 77VUGhBJQ0U218pCtBAfB6nfZBAtC/ir8ecwSYjHVBTqyfV8E9VXFNDDQbFiWygvEoYDMV9Ks3LN vapVXuf18vNVqCYEm7B68YM0IRhthppKZdPCG01FHnv+bjFCoKWmKwDRZJTxygTUnpV3kXIRrw6T i0bzaFpk4zs2Ply/sUmB2kY2XEoaNorRG5SPXciGKwnJgAyuzs0ouhcoFTxMOt52AKaGjuZHyyJr q0wolaTgnVhLuCgyxngkE1pLVgQpE3r0IrSBWGe62Udq3oZ2RHy4dEMLjN1oEME53LzmdewoZlq4 mdEHmDk30w7GewCmZv5MJgTKBG1lHW0rQzEe1aMG1xZZY2WiUnDNopUZR7+ZigjS3ZwOXWNtkTxl x6XFkNpQHk2LPPY+LrA6JbP3kepfKnRlqChGDPneu1RwOGyTgg9jY9bp0Mw5AOKbs6bpV6xM9lig +lGzyBorFWxoQIoAOC7IHoNRZJBx6awd0qSIS4dJRd1+NC3y2PuYitaapKpbQaSC/sW+DilmVrC8 nzeD1vUz6/gxBdd3f5asAGhW/CKfDzIHGAxSskZNsqLIGisVxQiAtyGdiUaRQcYlqYi1RSYr4tJi yYqG8mh6H9zfbpBmBQOMnyuCCssab49FIIIvdfkUo1xzoc+43Xqta/nG0QIc2cOl5HDeji+VoVqM 4kfNIjVVGirDY5kd+d6JVePjscvdY2n+bjbEmZNYdx8di8QOtH2WCVSgHPia8P/lwxpDQDE8551g 6whaTNuu7A4SCfjPn4UAgGbCHxqJDjIT+JZ2CstVswiOaSxXdDFHpl9USFIQRQYYl87aIaVBl3yK RIorsKE8mhYZe7j5ORU0PVFYIs1XqttcwUQ2GI0QHGZu5Kyh9+w4cmHqblpP5sJPo4lzpN4mFXFX gkesP1qKnK0QuBj+f6M5AFvVY4g6FNpja5EQOA4thtQ6/ShaZEL7KxF7S1D3UP3xeE3/cG+KYoTg 8qU8ZPO/GcAli2+cGNTrDxDXnB7lJeLKuCFBzadlgTXVm6JCLrex34orO3r+uGBoa7QZEWOH+Lvx 6DCJqNOPpkUee59nBF8yDg9sDW8RwaK4EAF44Bm+Uaj2yDhi7ATdpYzQ0vUWZ0XTmhgBauQXDfgd wH3nfEDlnqqRwW/WQ70BR4mxRUgGtHNACdCV/x0mAfH20TPAY+vz8JlaUqohzZcuWjvwihlMAlii Z9ooAfSvz3XvPJllc2XCgYi+Of3ilfphhs2a1BtNi6w9f7cYIdDK1oI1mowyXklF7B1SLuLVYXJR zx9NRZ74vno80JeRkpKTsqmOe1cUs4+Mbo+kg8f9GSbfm2vFzfEwrSeClAyEmyQjlR0JKB9FCZwd FzmOoOZKhlgaSPj2bliNIqJkuR+7DlUxd0i5wCfTYoVoG3mrZwFqy+j5iglfxsobnZLMM8uEWAJg 95APL65iT4mDCk9h0bdvyoScnnJpXtAc0pFlpLId4Y2mRdZeqWDTAlLESwHTkjWaDDNenb1DykW8 Osy8qOePpiJPfF+xQX9GnhKcE01Cl41i9Fay2NG5HF+wnzTIF1pgg478SQIRvXNyRINhGKpUNleg /dG0CGzEXtlg5wJSXn1WkR9Nxhkp2Yi9RWbsiFeL8Tu7Qd9qWuSx9/Ezo2OHCcxDjNzglSxXXTYW QxMN/2YppSBHmyROFg6aG3yIs2OmyAwfHGXaImOksi0B7Y+mRdZe2WDzQkoDmis8ZxBeo4k4eRUy uRh7hzQ34tVhj+c8YUbTxvLY+5iN1tkME1zQsMGu05aXxVzn5RVG7mIK1p4XRyMSYvDj5GcmGEGa Gywgtc4YqWxG+O9VU5HHXtkg5qw8UVbNK8hoMs54tfZ8RBQpG/HqMNloNL4WraZFYCOa7k75cFLu zILR29xg+0SXaRaDDfsSTUASfxIBqZDBm4AuZtM72BuFksi73hslmAMFW6pQfooWWGvlQiGra3bt Zeo5iowyPsmF5hxZi/i7ceowuTCWV00byWPv88zgvX2KTXYiEK/rdzZZivn84Vis2OWo+T4PaEYX cU4204wgeudE6Y0QdSNCJsYqKiL1NefPFmOtlBWZTnVVRJR+N8Bn+FhbJHclRXXOnnuk1u9H0SKP uY+52GfGD1986RzL/a0zXIEVI4QsYXoN+D5Inwb/sPM23nGgQJNAJN45X6JMMDFGiuLxnaZF1l7J qBQx5E3C6DHy+CQXsXZI8yI+HeY1rN/Rg4/927Gksd0d8uFqDe+q7OZJTcnEqw/PQIzQdJ15TSjh OCzOyQQ9QDc741cLNCXoD0k9sUJUkKp+9AwwtuaBG5mUlhBrZlWNfdlvbC1QEnTItIlQEjk+P3re xfUVCTRmZCWTN7vU2GWhGCUbQ4K5Sh9XznajDUApECdC/m6R8kBrSIqJSGUo4IUYUo+mbEh47M1A XCnmGX6/LlRUE4VsvHrsDZLfrVeLmQD1/NG0CPOVaPqKDTozUna67XGHkGLEwCcLLAT5rx80xRF6 hHSOeVoSoEjJYD5HlfBI4WakHkVFzNea83crRQg/ds5bTfzLO3uLNDHi1EJyUb8fRYs85j5+UOxw ygpiCs9vfA++a9xmPYt5Xk8a8M0DUpwnd8KMlGRwQnKzIEjJoEUkhwpxxmb2sZu/bzUtgnf3u8WI 4TsvopIZ1USc7+wtUjLq1WKyUc8fTUVC/tj7hI3fdksyi+BS/cP52pmeHkYekra+gOEB6j4ILjrn E0xhTlGUNChSNhjNePQ/Ujz4IvVoWmTt+bvFeC3NWcAdQ6qJu+qdvUFmEKlXJ3Wen6ZDHnsfs5FB BGmPSgsbND5lEDmMGHDV6oLyzY8zG2ekzI3UMovEO8toVuMWS5yU9nl8riYL61d7ZaOYL+nRCvur yWeMXj32FvF369Vi/57np+kQn3Jq+vSpgfS/HuFgbqQnhGfoYFYFPzjMyGca5ds3i33YiJTe/czv FtE7p20eW7OYV50NCSk8V1ORx17ZUMoVurwCfBQR5jtzg4T8deqkLphVdMBG9zcuNjNYs+Rq8pr5 t9/pLRfFGEw5V9lhlfLNrgKi/BkpuWCVMlkQZDKDL6hnAWulsv6I1KNpkbUnF4sxoNI67H3pdExN DMPv7C3SzKhXi3kV6/lpOuSx9/F9kmco0hwqwtVE2pfXJUPIxKBm4MkgGXxKIY8Delq6yMMequRA gKYFY0+pqIy7EJQZNcS4yNgqERXC/R80JSe9qocQ39oaoDToj/OTylwUp+aQs/U5CZQR2QT/hx7Z zFJRIBb3ebnjo4wqljElKREpU4IXzI05SHmgIHDo/TlClOPvFC2Cb/ezxYjgO19uCREoSkIoBH1r jajvZ+vTYiZE/T5Fh4w5EvdzLlx2Jx88xCz1BQoCuVjKwblhghMScvk9Na3Div1kRB0gtY8TJ+az h3n9PeCWhFg9A6ytJgQyzqNc6++co2qo0uOQQ1hsLdCEqEOLSUN8Hj3/O+Cx9TkJtGfgOxMeVjV3 9BDztmDbKNGbED99R0KEnGPtdlpcY+aUmIs0IegPyYi7Ulyad5qKPPbKhFJ5+csrLWl3+tWU+OeN PW+EIuWiXi1m6I2GHKumQza+S4mPJuq/bZh09oA0XdtZ6jwMNpildvT43QcmnxxFSDJ8meetUKRk 0DeSAXelKHei/FG0yJorGUqZF+nclosqIso35rxhipSL+LQQ555PKNVDWizwGPssL2iWdB+gTNBI 3LGjWPKC4yo6dnAEZLMgUlKRvYJSEaRUQFNG22LeH+xEoJ6FitFU5LFXKpRKXnx3s4BcVBNhvrGX vDh7VJPx6qSIvdFARjUdsvF9lRf0aRABLXF8CCZLnWgoxp3A6aDcE+QapWTyn2ZfNh7KBqdrJfIg eVowdv/w/R1DTXTJBjsRkDpNh4y9jMArxdjCVgvvytXEe8PRtPYWgcXzarBkQj0/TRfLxndsfDRZ R9oDNMIGZZDT1IMIgY5a7gnM8lWu5P+/COUmoce/cQt4gZ2r2R8IFZVxgc25yamBCZG1lV8t5CBr A9+jJy9QX20VkL7zZ2XO5VNzCCzUw49vkBZXdk46cUCY69+RdDHc50QxRlKs8iq8CRApU+JXfkfM Irm0TNN8pSQPI0X5HKnTROoXWXvytxghsBBho/tqIsp39hYpF/VqMa9fozlNhzz2PmajxRVdkr+d ONhEwspJqqtiXkzu8d4fjCTJADYrZVbiHK3zjSJNCjZJZMhdKerwSFURUS6w1spFhVzXpEchWVFF rIe+MzfI5EWdOikibyyn6ZDH3sdcbGbwfSUCQPqHH6Dr/VGMEFgz4lFGZtBhmNznAzIzKWEhuIGL TGZwJHcG3pWi3SNSpwk2iqy9slGMGHiB7qtkpnvRRJzv7C3SzKhXi5kZ9fw0HfLY+5iNzQzeZRMB 0t9YkRk2xJIZtBjxKIMNjmXPHfA9UpYc3/I7pidBmhq+ReE2WSGbw1QOGSpKahRZcyWjGCHwaRUL XGd7KCLKd9YWobHlfFpMLhrLKTrkMfcxF5sZnuIMFR6B0FrLtkogKl668HmWYZXSM1nAl9tnTsLu EaMOMGnBoyPj7grhwBs9ZEWAtVUeAjFl/eN3jKRBNf+DB86EfbW1QFOiDi0mDfH59CwwtvDwYxKa EPRK5quhSPMOtzXWYtDASMgTHBp475SDXjkTIHsqeTjmd/AQpAlBf0lG3JXicTXfJB1NECGSjsT7 3UoRAtVW11CriSjf2VukXNSrxejgmmhO0yEb31ds0E1CuYc0J6a3srCvkjQ1KThWkDoLNv50Vkp5 PPMS1kIaeJCSQfsIRZV3TqUobt4qgowia87fLUYIfAyvhUU1EeY7e4uUjDi1kFzU71N0yJj7MjNo 4+C25IE2Z1egoBDvTdlvb11BGetnF4mSTpHWoDbme38EMCIeB3SYHmSI/ByZU3NIbeWmWiEWuP54 ozx6eNf61tYAslB/fJZUZly2Ih01F8TE9VVC0L+B7wjznUnqCp5E9lSK4f5/NFDGKut0uf5+adIE doqmDDQEKQ+8AMqAGyyPR1/Lv2iCiCJrz98t5ttjznQKE9VElGpH09obxMoMLurVYo/np+liGXtf pgQdHCn//uUs7U7SabQMBhvfeOkfu3z5NJn/T6T07ndkYCNI2fiXVwNhaKSoP99q4oko4u1Ye/5u pYiBHc6mnPM9NRHnO3uLlI16tZhs1PPTdMhj7/NHp0fB6h3zop6rhwYxr+gvuudDBs2HXGLeB0dI MpgwILJIyWCphde4ixkms4oXRQestXJRId4Rc6h1FwJUJBe8b/KbE2dukMmMOrWYXNTv03TI2Psy M/LpRrmgBaR1BS2VvA8PF7ThW2VRy3EaUsKMlI9SzodHhsCDlAxGP8bcxSSDHQivmgYJ9/e7lSIG jp20sHACpibifGdvkWZGvVpMNur5aTpk4/vqqcHJKqxkIs3Rqy0tbKtknmhm0EFvbeHJZc5ZiDNS ssHA2ciDlI2ffPcinFWK2POy69F0yNprblSKGCg+rS94NRVNxPnO3iJlo14tJhv1/DQN8sT3FRuc pZKakzkBs1ErcBoug1H68bKL3hjY4JDJZAKTkJmzfo9MpiUgZcODXqBshAids3tQPorygBDRuZor GZXyPRFTXLmIHlq63hlbpFTokgXZSq3bq+fieIx99sSwxTJpwaHiTEglIlDcZ5unxQVVLcczJcQI mRXw9QClgQaSrAly8rhTW3hwJ7U0VM8COBaJshAZnP9J23RYUI33B/eshwCsrQHmYVGHFjufT88G 8dj65AwkhGncwCjSLP+0xloM99mB4OyUqpaF4EStVFzjnD9+x7XP74yIAYUzTR9MHngUInWaDhl7 KS5WihBY9O6aajURZbRLRe0t0oyoV4Ph1Xp+mg7Z+D6/Oey4pHsCNtjj/TOrNwuR2gybDiKUcu0p +pWPk+IaN3LCDlAmaByhpiKxI0PYrLlSyp6WQdZUbqcKpWeCXixvDMrV6MGJt7YKyN66cyLrMO+k qwUOXqPC1Md3Redhkum0AQrYUtSqYjG8Z/urVQW1J71SCTBSXiC+XVYWRHJhfX3mq0RoqBR3Ohsr kDpNh6w96VspQqC5wLWb1UQo7+wtUirq1WImRKM5TYc89j5moxMynrts7g8brO137aZYriWd2CEj Z0wYZoQkg3bTXv8gzQkW8ijXToowWRBM5c2qg7u4FsC3+9liRMDbW6uKrAiiiCjfmRtk0qJOPVIb y2mCi7fRfZUZdG5wNVlj/MPZl/PALMYomOm/mcHO1YTOC4iZrnJsU7IgyGTGf1xZyRgpYmdTBVKn 6ZC118yoFA0TTNRdu/Hdl5roCX9nb5FmRr1azMyo56fpkMfe55lB8xNXwRVXv0PW4UPMzPiPMcq1 G6pa38sbZ6R8lNL9njQo0tTg2ZLRdqSInU0Kr5oGeeyVDaV8YPhVJQceX72piTjf2HPYLFI26tVi stFoTtMhG9/lxocvh1gHoosn3pGkrbEWgw0ay62xqGtzdDZsKJU6gyMCmwf5XdngUJwMsZHK84Ad C0idJtgognfRVDaKEQNbrkNGFRFmlEt+zS1SMurUYBlFGswqgou30R0XH74C4NUKb8rkgrf67Pc0 MxajKuabDVYWVLW4bmbwpcCZt7KxKllQpFz8ptPwpPKM4KR2pE4TXBQZe1OZFSMERtLOgavJIVKv JKP2Bkl5sV4t9nh+mi6Wje8rNpiIcjWR5tNkLTB4QRuMGHIim2ywbyNTLiofpPSOF2LJgiJlw2Yl KasQofMGLoXWKDpkzOVhs1I0j1DVtb6IIqJ8Z22RJoY+eS9FCsbW71N0yEb3FRe81edq8kLvD19V nMwo5tsJTswlcopQ32ERJt+n3ZlafreI3jlXa8tipJIZfB4lxfdogowiY28yoxgxsP7R5cRqwm68 knu+toimQZIZ69VislHPH02LbHxfseHZuJLBzKLDCaqE4MLqQiooQJP+DOMzMck7I1+l+qMyQRyp PIo5jk4PXNXs32vJVIpIekf46s68k4sW41SzLMTUAppafxaThHhs5OhJjryJ6isKaCahjOJlNy9v O4hwo4Gl3+83lzZWeTFrPtB5NLOSX/nZImWBBpOMuyPFWMD7jRTfLOTn3IVF1pxELEYEOUrdzKom 7sl39gZJfeHYgVOPUP1Oq8ooWuQx9/GA2sKTsoQJXrhgZ1mHkMVYZqIb3SGEcobOlpARKS8Tayu5 /Gze4IiEksGVybAbLGGyTSGF52gi8CJ4F00loxg3OkNH5znVRJzv7C2ivfVqMfOi0TyaFnnsfcxG C09mMDx9wkZWeTuIiCUz+DaDKUp5S9kVNiIlGxRnTYMgZYNiKc9NToJyCkPsFEuvmgZ57JUNpbyi LJo5DXQSpiLCfGdukMmMOvVIbTCnaAG4GC8/5mIzg/6NPCl+UPN1ps4yYTDe8/OtYgdUillPieU2 YWslUg4qHAGbLAgymUFHVobdkSJyWlfy0BlNh9BaGE3lolKE8M+/P1nkhYxqYkHmnb1Fmhn1ajEz o54/mhZ57H3MxmYGJ4JxEZD2GO0mhpCJwTqwO1gggy21DV0hueB7jckBXrHmrDAC4HiH1BuBUjxG ZtSkuGbjwmOrRBSiomH7p7fk6iHEt7YGKA368z8yZ2R8D6LmR80iE9fnD07eMvJFBjuB/svxjLJQ zLkaZ8EyNMoCH75P0JxBOvU4pWEufxF9c7LWTseRIgHYqoDUaTpk7ZWJSlnbMQd50WQV9dbeIuWi Xi32eP5oeh/fV2zwwXKuJN7xTquzVFs3xIghozz9DHy+2JVLKqapxv+LxCLlgsdeBtyRIvJfr3ru 77HlTXUYNqjH+qSoHmJ8Z22RMpFPKpMVkeKmWq8fTYtsbF8xwQVnWEKaFwztv6GxKRhM0Jzvyjc8 c7qzjwo+HO3xAtwerDRkjC2id87UmipRZZi86ULoFB0y5lJgrRQh8Kqubx2iiCjfWPNuKKK19Wkx uajfj6JFHnPzoPhwTmZfpXMHux89Eq33SDEbIOkwjt1vLIHLBV0CvA6QC9qYkwNBUmM5cbINcaRy zWmcRuo0QUaRsTeZUYwYoLx1fTVhJV499hYpG/VqsGRGPX80vY/vq8zgSREyaNnLAecQHIgIaMw3 C6iBeDwlSoVAuBodOXnUATQtWMHPmBssI0Ga/h89bsJ4bE2RGognEh/CRrGTL7UQ4Kspc6JAWag/ i5kTcfnUzN9QMf59PHJ0HGW0ZYNkOPiXL3hOQhTD+QwQksBBduEgQnLAsnSvvUiGUWdnth+SDyOV rRZIPYoWwbnIGNFiBEADksPoaiLGd/YWKRNxaiGJqN+jCNVF0jQ85j7moqOocwEnDdxSfPS309LF eOnPSrzLE9Rs9NE5IfkTKYcUvn3D7xZpRvBKPBXESmWnBVKPpkXWXslgRQ8pYvjnp0fIQEY18Th6 Z2+QkL9ePVLr+aNpkcfex2xsZng5wgYfQG/DHlPaYMTAdgNrCFfV+Q9jj5Rs8JEofjfIpAZLUWT/ I5XdCUg9mhbBu2gqG+xhQIrShVMLfBA5AVMTcb6zt0hTo14tZibU80fTIo+9j9nY3GDI5Hoy9NB7 vcOpmO+xqMGZVckGra2J3b7qzk143Zo8KNLc4PUnowgv5Ecqey3eaCry2CsbvCRFFzHkIwiyUU0s sL6x5yOjSNmoV4vJRqN5NC2y8d2D88NhhMVoD6mWDU5f7Tx9MdjgXXKHVD/FlDiV4uo5J/J3RJ7f lQ12XDLwLuYDgekcUo+mRcZecmoxFm35Fmifn9VEnNH+2FukbNSrwfBqPX80LbLxHRsfLu7xHslX M7LBPv2ceo/OYsTAcRUdVPnwfWNXimxxpubviDxI2chHOhfDO0rpSD2aFhl7M5ZUihhy8rC5UU3E Ga8ee4Pkd1TY8Wox2ajnj6ZFNr6v2PjuWYawwQjRg+TQWYwmO955eV/4r52ecPDPTE/4jkjyoIhs OFvzHaNvAyvljgWlHk1FfMbXnr9bKWLoZ0pOk61Bb+0t4u/Wq8Vko54/mhZ57H323LCTy4kE0iR1 X60XSwxsirHI4BnkVgnjjJTXiu1mlGCLlA0+D+GMboWoRiMEGaNokTXnzxZjMuxyh6z6mQkeES6b uD61xgZJYqxLi0mFofj4GT0LnLHPjurl58xDuZZI0+zf0XUx/59RNUzQSNQsUCrO0UbUHMjvygTt JhmD+U5bpnXuWFDq0bTI2JtxqFK+5PKQT7moJv+/mhgvo2mR5kW9GizPjEbzaNpYNr6v7hLe8IcM jjezCuf2FoBTZiQ+zSjj3Ezk5ab28iKl/7l/lgT6RzL0VsANFml2Wg0Bxkau6QD4zJlM3oXc9VFB XK828mcIWydWYLx8NAxA0Cj4W8gtKBjqPZiJ6T1FPtKNuhhh82QwOZnqezocYXPmZZxyFuTvBolf DOy8oTmM32WTBVKPpkXWXu8FtmIgxaOXYxIcklYTN8U7e4s0AerVYl6vev5oWuSx9/mDgaYNrh7S vAPuO3XqZDALClat3dgFG3yMIWFGyArjW34GGUGaCHSNZJwtJhlsT0D5KCLMRdZcyagUIbCf07uP WjqaCPOdvUEmKeLUI1S/w+ooWuQx9zEXmxk0bXA1kab9sMWV6yxidPrlq9VQQZtB0yBCcvE9InAh MolBt1IZG6lsT3hVtMBaKxXsYUAIKhhxLPOd5amIKN+ZW6R5UacWMy/q96Npkcfex1y00LR70okD C4BUlS2tipkXLF27sQsyWGDMvc8Meuap/+V3dFEEaWKwhTjDykpld0IKzdVU5LFXNtjDkEKTdp28 f2K+F022LL/as9AsUjbq1WJMjycaCr/RtMjGd8+MDwtN3/pY7iONw30BsJg9oExDYINyjTPjvE04 8v4maZ1wBMnTz2laexSLeVOwUo72R9MiYy8ptRgx8Aq/a4jVRJzxylRkt0aypYhsrFcjRU6t54+m RTa+Y+PDQtP+SScOSDPF7oLFYpYlfKg7drlUeRywSMdavd7Rbtk8CKJ3TNPoowpn0SUbbFPIELSa Fhl7M4ZUihiy9/QUEeYbc0mNM0cNGaceqXX8FC3wWJv75C+ZwcWgUGDK9csvXTiapMcixSGPT4+w 4Brw6ReuC/+iFNfAKVqnG0EmM+iDyJhTzLufr0S/0bTI2JuHTaV478XqQAfpakI6Xj32Fmlm1KvB khmN5tG0sWx8X2UGi5dcTaT5AlBO/oeNYnBAreMKLHnuqXGyoRQIXOd3izQz2IV9QnkUslUU5FHE VXw1FxIXIwTGcmsrrqyKiPKNNROjSLmoT4t5Fev3o2iRiQ5zn2cGn+PgaiL9z2w0ZIQIRBbSRuIt wcjokXFSoZBMsLpCU+oAJYLmkYw4I8NT/p2aRcaWv1oI9yk2O+utHkJ8tRVmBEpD/TmZdXnUoLiI 76cT15Hwl0fFL8Pjzmcu3uqCyrEzJ5bx3c4EB2w+ZG7CjRchh5XsKlugJNA1kpFmhagF3+oZYGzN nREZHjKM2Z3rVg2PlLe2FigJOvR/fwbLfRGfHz0TxMb1FQm8q+QC+vzjiIDU2SxBBfOBRBiywIuh 3Py0pM/clAbLXPcipYG3GBliguWuYHMCUqeISvLV2twUxXzasraNOedaKvLZ+mruf4eUiTr1SK3f j6ZFJrov7gr7J50tIE3zT+uKxRip2J3gXUDd5sFuJECEzAhO6UyTapFyQbtIBpiVom6M8kfRImvO 3xWz7Yw1AMZryagmwnxjL/QEKRlx6hFav0cRF2ORx9xnNRbveChH5OLHLxrtM3YUs8ZiCd9dXXDB /q6kAWXGTE55biYJipQM+koywqwUlSNSr5qKPPZKhlJWnB676czHCZ6abF54tWdmFCkZ9WoxY280 j6ZFNr67Sz4cSe2ydPqANFsIW2MtBhu8k/A24QVwDoqjYlEKzp2pddoRJPnOXI034CZQMS86exPQ /mhaZOzN86JSxMDmw45M1USc8eqxt4hsrFeDJRMazaNpY9n4jo0PH5z0z9jmLxs40E69xVgk5Vxp ayyq2BwXR5xKgThT68SjiN45WfOd4kjloUE/AtofTVTWr/bmqVGMGHhh29GpmogzXj32Bsnv1qvF zI1G82haZOP7ig2OJuF68gKIjwp20cY+SzFiYCOKzzTi47w4BxImpzNv5XiU5EGRssG6SgqLlcKr d5qKOMbVnr9bKZjiYMUupVYTc5d39hZpbtSrxWSjnj+aFnnsff7c4BUMK3BI/zdTdls1nVRABtu6 fKbx7GTHihf9m0JeKb6EmywIUCo4Hi9jjhB5wZPsnZpFxlaJqBDu82EMb0r+MXoI8dWWT4wApaH+ nMy6/KhZ5Gx9TEIn63ZOOnHQNs+peXgWM1T3WcoCk48mQKR8nP6T38GLSG57ktDd+gPxMycDKn8U LbLmSkSliOAbH1uBv1FEjO+sLVIm6tNiJkT9HkU4UCSXOJq+uj1+240JF5x1ht9OQtz9IkYI7P10 BZPCLafE8bCKlFywwyyBF9E7h/G+V1wplhDeaVpk7ZWMSvFY5Xxqa63VxLuyd/YGCffr1SO1nj+a FnnsfZ4ZzEBDBm88tsIIBBe81sQ1yrccEedzWxmpyPb9BTxIgAKXE6tNixXKKmsorZr520ZLtZSG nCnm4OLBfavFW4E3Iq+mFmhK1J/FvHzx+NEzwNiCuo8p6OoNnZO2NJAQ3+kq7li6GAtYdOR3LP1N O5wVAktMSMkCe1hSTBZpQpBAVAiPFC8hIvVoKmK61l6ZqBRVx2/OTw4V1cTK1zt7i5SLerUY24sm mkfTIo+9z9mgY4PiEWnujr78sMXS+QQxsNrQnKD4TJgRkgzO2kOEwIOUDJpGsnBVzEdFTpp9UbTI misZlSKEnOdNcqnIKot88aSPMzfIjKTxaSGpqNunZ4Ez9unLD5or7fuRCRqqu1qxGEwwL/TGpYp1 K5WBKxXfaEZvDuR3huQkzReJIwVCGR2pR9MiYy83/GKEwCEgVhWriTCjXS5qbxHtrVeD4dV6/mha ZOP76i6hcyMVJ1UN35HzsWnLpVjqKJavtMvBTUkDvnI2E1bedKSWLFI2aOBioHHFWV3e+LyLeKNp kdRjaPJ3i7nk/WOnfdHkUnk1saIfTYPkd+vVYo/nj6aNZex9+cz4wVt0c4PD++aMQfbsBIMNFx3D Bmt5mVX4Ge6dq/k7Ig9SNvjWTW6mlULrO01FvCtrr2xUihgYkb0rnfipiTjf2VvE31FNxqvFZKOe P5oWeex9/sxww2e8Yz2xQ6q9l84qYIMXnj7R7PXvrIIdLDtbi4wzWJGyYZ3mc3alyM93mhaRlf3d YsRA54t3Ju/Book439lbpGzUq8Vko54/moo88X11p3D8LFeBbKXU7ZsRui+D2RSb5jzKN/cAECad 3DdZ6zO1SMlgBzHD8yPlkKZUFRFlAX2rNX+2Qqxu0Afnfem8T0W8X3tnbpFyoVMOvyu1fj+aFnns fZ4ZOMDVRJqUbH1hY6VzCnxir4V3hbvqGO0NM1LWHN8is0jJ+MmtbWasFPXlO02LrL2yUSlioP/I MoOcjCbifGdvkbJRrxYzM+r5o6nIE99XmcEJSCk9+bgqW9v6DC0GG2y3cF5BOcsB3GEjUrLxPb8j 8iBlg/OMXoUopyM0ipIGInHufrZShMBpJ96WDEAqIsp31hYpF/qUzBgpj7M2lkfRInA/5j7JjD/2 VuIbCyh0EqXMKGResOzGY0wm2H2Z9GcpeCdq7RgoUCLoLsnZRytELa3Qo2eAteWvBuJlOss0YUEt 0sByMV/8lHTWMA6YcaT+jND/caqpUUhC1ezfj6VP9uwjTAcJNl1E8lODpMNhvsT/6fmTkMACBjZ8 a+83CvWMDcgpOYLENWdqvlwcqUTI/gSkHk2+pHmxlwJjMbsK8nkSyolqsvEgH0A8e4s0IerVYP97 PD9NF8vG9+nNgTT9GkRAyfOLJfiSUYhxjQ9jeOdSwfnooORBhr+ZEOVHA+hZ1nZTfo0M1SMyaSgd LUXWVPhbiHVxrrHJt3roSXlrq0DYG3dW5GI4LYc0Kkx9fE9kmo4w7SO4jjBH92eafpjdt20Fom5j Y1dIiJQscIZtQxaJa87RfKMIMyNF5RipR9Mia6+3U6UIgRHDefpqIsp39hbxd+vVYiZEozlNhzz2 PmYj8zKk6d8gAqT5aHrG0cFyLSm/fXx5ZFonHlSEO1PL74g8SJOCeXoelytFYxxSr5qKPPbKhlLe 4Cyp36xPTcT5xp7PxiJlo14tJhuN5jQdsvFdbny04os0DRzYZcLEZz0zkh5G3xcvKBxJqWJpsvGq /6dUHhZ/8rtB5mHBVw4z3kaK36EhUo+mRcZecmoxXpf9ZucYv3Papyb61eKVuVh7i5SNejVYHhaN 5jRdLBvfsfHRii/SshA2cCDtm4fhJY75UKeKhZewoRQI86L8Dr+DNDdoL8p4GwzOuEMi9WhaBO/u d4sRA9vwrffJuGgiznj12Bsk7K9Xi5kbjeY0HTL2YP/zO8Vj2MIGm+KygoGGYsSQTTKy4elt1k/0 BN5srTOPImXDb0uEs5FiAIn2R1MRB28aHtHk71aKGL7Tuxo2qok439lbpLlRrxaTjXp+mg557H3G hpqcQbDOwjuolhbFHAy/sUvZkYPKDbeMk1d6Hr5IbvAVn0RepGzwxaDUnCtF7RipR9Mia69sVIoX 7F31dTpjpcAph5ojydbcIJMadWoxKpAE4wOoig44a5+tYfhMoknTzKAFn/rAUXUxImBDkXcFlRu9 AeFCqTj3X34HO/ldufjTVpNCqRwj9CiilnxjzpXBxYjgh2cweZtEEUFGt1TU2iLNi/q02PqdFpUq ukg2us/vEkuVMMHZTUOEB83hPW2+JADX5h/WqEODIvrV3YN8LysfRObhRvNIht5ABucZ+KMiwfo3 /uwvKkDsHEfdZcRqIKxXG/MngY8X/zNTIkGU+n0q5u+z8fmtQKMGtTFrr7+9ML38YjrNqQgeWIEl BvZUmPSc3IRMGYb+IF4Qp2RtdFwp3p+81QQdRdaev1uMLgm+C9eirZq40d/ZW6Q81KvFTIB6fpoG eeL7KgGYWKTCxG1mw2WjGGzwWoalKNigESmhR8gs+JWfWQf5s5LBNCHjbDGTmp+jfBQlzUXi3P1u pQiBo1TMMed2aiLMd/YWKRk4ZU6MEJHX71N0iOvsdfPzzKBtg6UppFlqanHFYiFYQmAtIYlBydk0 iJBccM504w5SLugbyTLv95GidH6jCHIKrDV/thgRsGjVV5IqChUqymOo5oh7zcFBnVrMvKjfp+mQ sffF8MmLBLcXwAX9Y9zxzYtizAsY8SmtsMsLPC6nfislGdmWcEjJ4DVICs2VIm2i/TQRuYjc117Z qBQNE5z3vm/j1GRx8NbeIvzuvFpMNur5aTrksfd5ZjBAUSYgzTSzxQQvp/2QDss4bPRlbUK7tKQk ESiQkXIQ+RYZ+AlSNjxbNZyNVIqPV02wQfEBsvbKRjFi+PHHYxa5TaqJON/ZW6Rs1KvFZKOen6ZB nvi+embw2ZM80gmcgJsbxWCDnSoU2V4jOgh8aHyLlGzQr8XvFikbNHhnmF0pCsZInSbYKAIb0VQ2 ihFDaxfneyrC7DtzizDwnlOLSUYdX0UHPNY+zoxOT7m/WC7zSrF+wq4UuViMdQTo4YGGWdKV68mj LlJy4TLbIZlKsH+Kc9AOy/OAzRRInSa4KLL2ykUxritFhBtPVxMZ+c7eIs2MerXY4/lpulgeex+z 0emp9Qwtk7KRzaVlQ8wVN3riPHERNvicavKA1qiZsKbVUn6CNDPYXpFBZ6VY6nurCTZEHntloxgx 8E7d2srpl5qI8409x44iZaNeLSYbjeY0HbLx3X3y4fSUYt0vRMgGR3NNahRyiYkPVUIGxaxjjaEr ZGbwSeqU1wEMyRnaqwiFcxSfFogoMqbKQyG8x2ZHjqghwFdL4UVAEtablSHiOjxaECnisyJRHQUf zkmZ2XebCgeDMf1sPgTi5TjLp5Za1N5s17fo/BMhKeCob8fcAqWAnpEXLBWzQqcHDtg7KtuxlTtq IBol2HPU6rVqWHl4a2uBkqBD8hIhHhsTxulZYGxB3ed3Bf0a9cwvyWYZjyXiYLifrYHSwG5KsoWo 871ZaaA9vjEHKQ+U4ympIpV7nTc5JaKayPcittz65Vp/txgh8HmNndqpiSjj1WNvkEmIerWYXNTz 03TI2PuisvAVUq8TScdUqzlRjBVotldwz8KGdw4zBWWI0ulYcyJIYnKKZOvhSCVKtlmEi9FD3EVo Q4qmclHMqiAfPuMGqybqmfj02FukeaFPgyQr6vXpgYm3sX2VF7zlSu3NSag9TA8NxWCCQ1WpqGCC JcdkAV1jMzehlTE5UKR5wVI1oy1cRJdXnE0WSJ2mQ+AimspFpYghF9qHTTURZzURce0NMnlRrxYz L+r5aTpk7H2ZF0yzuZq0yXES66RFIV7QMLb2acnnX7NuzSe+W46zxJ8cCFAqaDdhsGWSHRni5tUz MqfmkNqawbdC0E3/Ut+0VA/F9VtbAzQl9MekGBn6bbDlwueogYU3cX2VEjleXRLYPcnkNGlWDPe5 9fvETGRGGCmH1O8ey35IefjFORCH5VGBU0idJogosvaaEsUIgT6sFvXVRJTaSx1Ve4NMStSrxUyJ en6aDhl7X6QEUnkPhx52tXX4WIy3mr983nuDsLshq9Y/IyUbHIKXDChSNvhHBlFWpipF7Iw/SJ2m QVxNqb2yUSkeSLQ0dQypJjh/Z28RfndeLcY0bqI5TYc89j4bRuykdOKANMfxdXpaLDEw941ZDv1p 6BGSDD4flzRgNZufGRTtT3TRO+QW82nBBgWkICOKDlhr5aJCTrQ9H2IVmf/fVZ7MqLlBJjPq1GJy YSyukY+mQ8bel5nBaR1cTaRpfuogQkESjBBYxHMQoRDnJN2EGSnJYCN9syBIyWDqkcGG5n/LdWJn EoPUaRok3N/vVooYWPlyeup0T03E+c7eImbGerWYbNTz03TIxvf5U4OpCuvLesdBRj1eEA1ixsC6 vRtE+Fe/qukdwIr8FOLsO0RmkbIRVQul+HyryHI0yJrzZ4uxHkXR5uyU2Z66WUjTWjKj1gZJZqxP i8lF/T5Fh4y5LzODF7RcA6SztdrygiWPYPaA5L0l/8p6fG6AH5EyM2j4TRYUKRc0lmToHSnC5CON SJ2mQcJ9NJWMShEDLYzhoooI8525RfjZObXY4/gqukg2uq/ywiUjqfjOAzFDKkukzjbwnwYkp6a8 mvaIfnMgQjIxrzYKlAgWQzPsjhBhp6vq9CywtkrDdF7RKrtv4lCTnNBWciK2iDjGy0IdWkwW4jMk VM8Cj63POgp4QcKXH0yIX9kx1YQoRoMUryCck1LVuqrnGxs2P+GZM6FOOILMY5PmkRQbxUx79iXw u1N0yJhLbbFSlvj2Y3pzVJO9Em62euwtUiri1EA4tbGcokM2uq8SgtVlLgDSP9wlWC6KEQLr+NZU VLF8CDpkKBXnvuV30BOkOUFHSUqOYJQXVNGROk2HjL2QuFLE8JM+4pBRTcQZrySj9gaZR0W9Wkw2 6vlpOmTj+4oN+jW4nMTL5GIH1GK00YHIBRvHYYwXm8okMThVK0kQZBKDXtyMusXkgv0UqTmj5/5e W74FWhn8/4fXLWEiekgPtvLFvlcaPYc0LeLTYsmLen2aYOJtbJ8zYdOgswe84/yAzlAX4//pRDcL GIzY7xoulEpe0JKeHOA4CpDmBQd/pN4IljjZnZCyczQReZGxN3lRjGv/07vMmySa/P94ZV7E3iKT F/HqsH+5GRvNaTpk4/uKjYzessG3eFtckNPWBtH0y+0wMOz6jMUFDaZI6R17cpsFQbxWZKvfVnik iJ31mFdNg/icrj1ZXCliYOiw6B9NYeWNvUO0N14dJhv1/NG0yGPvs7LToc53gjwhaWPuxKyYbLCe 6Q4e7H7zYHriZFzdqZr7+A8pG5xQnGF3pIiTPvjUnaPpkLVXNipFDP/YrQgZKvK2oOmC+Zbk/1TR IkmNceowyTCYdKpU0QJn7dNXxxSEltBywQdIOqIuBhc/OJM3XPAkbeRKxTmet8kCvjDD78oFz78M vMXMd55/GZxH0yFjL4PJSjlA/Z63MNFkmNEuGbF3SDMjXi2GV+v5o2mRje+r+4Rtp5R4SHPmTNrf 0VnMhkf3J5gZ3ACNXSkQJmz5HbkSpGxM9RWIhwa1dIRO0SFjLg+NlSIE+ri8LZn7MWQZ5Rtr/zuk XMSnw+Sifj+KFnnMzV3y4VovqyxeDLhg+pP9hagsxKOYvTxmLbWtq1vcEsx+drbmF6QGKBHUZRl8 K4OPJDiaT80hY4tfHYT7HunBxXXahx7+hTnbY+uA0hB/Hpl1edTgchHv/sT1VUJk3RRZDrDsm3WX qIkY5+mt9TFGLc4Ow4QcIYeVbR4JUBJoFcmoy744NBmgGwMfPQuMrbkzIkP92m9vwUHUGN8bWweU BB0KLwrlvpjl39UzQWxcX5FAWcQFRPo7n0JIjcW7rGDQQMuKV4NKbgJUCMCZUOcabKkDKQ20jmTA DZa7glfTKD9FhFJkzM1dUQwmKE//f8Le5ECSJAeyvRcVTULuCxmfhqHg83+Z90QAVfcYT4++dIUk HIsYTBW6mi+hi3FoMvA4dewtEnt16kA+/fp9FS1yzL1tL91MacHPcIujnl1FXszqjTnOcMGO5qT8 F6V4Asjnd8jmdyWDria1RjFzgkMJaD+aDjL2JisqRQxUL+08oknpaD/2DtK8iFeL4dV6fjUtsvG9 zwzecj6ZFjY4j9ENeosRAxvxW1fwZYLkPkdizxCtw44iZYNJoPS3wZIaHKdA+9GE50XwLpr43cEY 7nLEtHVFNBlnvJIN5oCSLEEmNeLVSvHU1vOraZGxB/vv6gpaf89h+KzZOdnlscWIgZ341hX2qpxu k/tI6R13wzcPgsgG7RX76mxYRwo/SS6kjqZBbMlqr2xUysF/Pv61mozz2d5BtDdeHUw2Gs3VtMi1 954NvkbHDgOaAZrtVllUln59iRjYDeNhJuxaG1t3cRvJGasps0jZ4A6nHO8eKfxkCuZR00HWXtmo FDGwISu0qshWg43C3rCIe9j/eZCkxji1UvxuHT+KFjjWPmk1fvx1BIE0ixKtshyYi8EF3/WyyqKK zR1vRK5UnPsdmUHiHJMG9IjhJ7psCmnlnzQtMvZsNY4UDSenDH0vR5NhjibGhmo6SDMjXi2WVqOe X00by8b32XvCVWCpObk00wE7k7YBoOKbH0KQCq4haRJ0zMoNz6kjmfdrQkAZJQTE+O9MPPLvruCu giJhvL9YAKcZi/gjh17oMLAHG/kzhI0P++/r5FWwCFFr4/XL0E0VWdfgoVEZ/OVQUiauFmMyiM81 mp1ZxeBB0jtHyk6VVZM87CDxy7WHR5msTqSSWD2LrLU+flY6kMJ9NqXZJUUP0bEKkMHw2DpIHz6D kiTESq3Xowevi/iiN7Z/MdENFSzn0XaHiV8WN33+xZwWY+kW3yjSciMcTERKJri/LU+8SLOAbSJ0 s1eKGjFSV9MieBdN5aJSVFOsHXegG03G/mxvkcmIeLVYYq/nV9PH+P7FxuZFVshxjl3Jna9xJ2WK Y+bufUtZT+fsYoKMjFRk+y1RB3AXkMMjV+0W85Xw1B2yo2b+XkulISL4zgy3b121OLBg/n8GwZo6 QFMi/hzMBIjHV88AYwvqXr8aTQg3Szo8cBHB/fgmxGKU1oyGragIjF3+7rlhjQwpWWAqOg+/SBOC HSJUDFcKpyJ1NRUxXWuvTFSKf3MIGiqiiSjh5tHeQcpFvDrY9fxq2liuvfdsMMhkXhrpX55JKRti Pksm7Scn/np/G2FGSDI445HnX6RksJrIdM6V4p39oGiRNVcyKkXG0FTuQAtNhvls7yAlQ6cOJBf1 +ypaZMx9mhlUDqkuOf7LcLNcFIMLFmPMAyokVksTZqQkg869WRBE58hFT/JLxkjRlUfqaioS7s/v VooYcmPDKjLMZ3MH0dw4dTDJqONH0QJwMdbe5wXHm3mYbOrjk1GtINxT6fiBh8/trDZjVGhMUhgl u0ORkgt246S+KlIu2EXD/NeVotr8oGmRtefvFiMEZvg7fxpNhMkywKO9g5SMeHUwyajnV1ORG99n bQa3UvM08Y7PgXVywj2VYrDBxIfLc9RoudCNOCMlG2xFTxYUKRvcdpEOdaWYk/ugaZG1VzYqRQfz O19+IzWiyTif7S2STmS8Wgxe1/OraZFr73VutBNxB6VjBaSZ1m6XuhitJFc7mQnUaLkbDjYiJRt/ 87tB4p0jJfcfXilm4CN1NS2y9spGpYiBjzGzvwk2osk4n+0dpLkRrw4mG43malrk2nvNxvYnbGrg ebqJ4jcjrrYaYtnhl2+bQgbnJZMIrF6eYVnrqiJNjZ8e0ISMlaJkjnLI8PPmvhMF1lq5KOamDb4n BveO8FBElCTko7lFJjPi1GLJjMZyNbk/9DG6f70nmxmMiXmaeMc22u1bixECi5m+xdSxuRqOMCNl b/s1vxtkMmPWBVeI+jlCV9Eia65kVIoQ+MjTjrgyjcc6AD4da8YdpHkRnw5mXtTvURR2RGwzGt2/ uNi8YNKSh4k0J9e7BuJuylZJzN6z30Au2MSZ1oDr3c6oTBm4CNLEyBTkwcwDVjBSb62mRdZeyagU MbCPw7fSEReTmcb5bG+RyYx4tVhir+dX08Zy7b1+S5oZ7J6UCq5aZ7zlKxLAypJLhPmTMjZXwjF/ HhGfUkYm/bMk0ORTqlwR9Nn+ro78uTYaPv/OQIyTNpYVrr6xdEBUFF1eXBgb+2fGeePGwcbPq+Mp jpMALye03TjpaAGP2MvdcmIxnGb7l8lIuZaL24hKqTjFrZwpNIPEL8dKLhyOFL+D9khdTYuMvTzI xQiAa0s6VRhNRhivJCH2DiJz49VieLWeX02LbHyHjZf7mNlM6UyEbDDE7FLHYtnL5GEd7H71GjjC VAiAfiE/W0TnHJoxflgpnKOljtRVtMiYm0F8pQiBG146UxhNhhmnjr1F8rs6dSC5aCxX0SLH3Nvp GbdTOmxgxe0XnV9fh0JE8MslYqng5v2kAV/KsamglMyvmCfwV2WCJjXdSiBTnMMUeSlWzSJjy18V cu/+Vypv3z7fIfS4cqjqY2uBpkT8WUgW6vKoSY48xfVZQrACyFPEse+/ensgKsXsN5mu97wSLPD5 gVSYfETrjMz8HTQEKQ+cOUxzsVLU3x80Fbn2yoRS2U7DsaQd5aHJKJ/s/XeQchGvDiYZjQYyRtMi G99h4x+NBTf08SSRpjzfnCjGy/+Xr1+FDaqINAPcxcr2Qx8UJ8maAfld2eAEWPqUSPE7ZowidTUt Mvamx60UMbC9ueuw0WSc0X7sHaRsxKvFkgj1/Gpyq+hjfIeNfzQW3KKeCvMnrUUn7dhfGYwYrB5M Db5mn0T4phCIoyB/xsRmkJLBREUKzGA4R9n8rGgBetDzs8WY2XTTE8od5aGIKK0Crjkzo0i5iFMH MzPq99W0yLU3Pec/MoPPLKS+5BolRl1tLYrZLfyh0JMMduInD+gnvlkPOzbrOCPItBdUg2lfi0kG eY/2q2kRvIsmo1qMGFiY3HGea37EHq+OvYOUjXi1WDKjnl9NbgJ9jO99ZrBLkrvgfE/wm1GXbCxG /85pBJ+M5+IdoZAISoE4CvJ3izQ1drNVVBkm6zcIXUWLjLmQuBghfKWsQrnDPOoKuHiylswIUi7i 04PU+n0VLXLNvcsMt1I6XLAMpr1MZixGCHxurK8JX7NPa8DOWq85hAsuzUoWBJnMYH9IeptiksEF x0hdTYuMvWkzIkXnyLRV64soIkym/b3NY80dpGTEqcVwahy/ijaSje6zvMgRD4S5aDp3NaMxEP5z GtLWzHP2Xcmg158xao6aM/wIoGcuTLl4OBiesUnScuzqGWBsTU4oo/M8CmkYNQb4aMuUKKCtcehg sqDPD3omiGvrXZHtLkn6CRKCd5K3sy9HMZym4PR5UMJmHhYXI+UT4quiefhB8mAdkvH4DGSkTASl rqZF1p4xDUYILDJwfZ9URBNRMvt9BoBoOki5iFcHk4t4/qDpY3yfpQR7M1Jx8j0vNlKXDTGn3Lgh F6ZlY6KMjFzk8zTGHaQ5wQaS9DTFfONZ+kX36CHKReDi/G4wI+CWV1PQJTc0GWWkjr1Fwn18WoRf seFTaw96BvFJj7X3ecEoK/Xmd5atOkXh1kkxB4SsfEkE52+bBBHSNU7DNgVEJi14hdLPMGzJyC01 96OiBfAtIjI4mBGwBjZDLxUZ5bO5gzQr4tTB5CJ+P2jaSK6911x0WM77x/yt7wgVJiOwZoWYWfGb Uxm+vtS2jpfxm8u4zzjN3y3SvGDERjezmHFyDuFJU5Frr2wo5fNkM1N24zkGQxNxMgf+YM/2okjZ iFcHk41E86DpY3znHXlZXNBlUdXHO3YhtrhYzOb+KzdTyAYVF484E3jfLS4cEvm7QdIGMlRzhvxg ssFJBKSupkUypYemshEpY2Cd1+JiNBlnvDr2DlI24tVieDWeP2jaWMYefr7PDb7EQQRIe4vJFBfF 6PWZAbG4oHz75qCB2JUCcajm7xbRO0drbkUcLC0EZyiQupooKR/tTUcczBj4OstVZJhP5pIaQUpG nHqQcgTjSOYoGmCju1y8zAw3STqIYCqas89dAFiMCFg4baPBF0mTByxfeY0gXHD2L3EHmczg5izq Kuqk6DIzOJCA1NW0yNibzIgUIfzi1vB2q9FEmEyCnwEimg5SMuLVYng1nj9o2lg2vveZ4USqowik PRqVzFiMGJjZNA8oZf0GrnEqlcxg/qqRB2lmsB+dwmqkkhnMi6XuXE0sozzam8wIZgxc6tnUiCbj jFeyH3uL5Hfj1cFkI9E8aNpYNr7P2PjGUnDYYM9xmlD3WAoxUqAIt8ygluUROZZnDXDHa4oMIBUM 2Ky3jwj/8UHLIjiGlsmKCOk9hcUMxFBjgE+WFtBSvXmQqcOuMa6WDaGm7uvxcoSadVoZ4GPUVMn2 IoHsRH7ydtiMUcGymmx4vyPk48ltPguUAraS0O0u5ruRrmf0JD0CrC1/VRmcZ2vFrEpGDSGzBoBD a+sAJUGHbDhWqGE86PkQ12d5wG4NlkLxDC9aYeESWBKB7ey+A1jz21ZGGCld+xOZRcoDbziV2WKp Gz9oopIMsvb83WCGgOEZ6qnJKJ/tHaRcxKuD+fzi+YOmIje+z9hgt0bqTb5BOHcpkl3BHED7OULI +MG8plFGRi648zCVRpFywX6R9LcrRd34oCdhC8S1KCoVETIANlzt0Mt9iD4CXTrmDlIq8Cl8jdB4 /aBo44D5MfeuH7Und8sh2zwtqObtEDMv2BmZjo0OjeJVLgDOAE2ZRcoFzzVTgivFxN8HTYusvZIR KWLgK/E72FMTobMg8GjvICUjXh3MvIjnD5o+xvdZXrBpg3oB7xh3d/WDOjgYbAC5sEcd/MMRBbFH yt6Eq+1SZxQpGywap79dKSbGP2haZO2VjUgZw8988i1jPvchOsDUq2NvkbS149ViSI3nD5o2lmvv dW50IciNlA4gkGaKppMWi8EGt+ybplSxvxx6wEak9I6vajVykXjHWI1FkEcp6uhIXU2LrL2yESlj 4IgA5I8iw3w2d5CmRpw6mGQkmFEUdgSS+VH0LjN+fGHih4JILv785luUvCcXYwIyV1jKBR+zTx78 jJRccMI6cRdpZrD5JnXXSlFFR+pqWmTtlYtIEQJrKjM6jybC/Ppsb5HJjHi12PX8aLqxXHv/zgyl +aYTESBtTVkyChECX0UxSSli+Uqyj5zvV7YCpRhq3ABxjcUr+okHGTZzReaqWWRslYgI6T4tgFkx egzxydYCzYn4s9CJYtWEF4MwJxLXZynBZfL4jjDLwRYWqiyG+184koNnlGtWOrIQKVPCMuYi+uYg zS2MV4rVgkhdTYusvTIRKUPgY8rtNaLJKJ/tLRLex6vFrudH043l2nufEj+s+2GDSZncraiGYu79 ZbUqLx2XwyYFmCaZASu7rhJ5kMmKb61CVojiM0JX0SJrTjIGIwRmbmbsqyKi9GqNDo9j7SD+DHF9 OtjxexWFHSMxMRrdZ5nhHSoK/3BHRhMjEAFYLmiU+2rYoMAoOzKmRa5MXKBZwW6RdLsr5KmD9Myj Zv4eSyGvkL4zwe576DAPLcb3ZOoAJUF/jtB/kqDHR88CJ6rPKPA4RThgL5SHbFVQDBJYN7XwtY/r OIOjrt6+Agt/8ztoCFIa2CeS/jZYOgHeFrRfTTmAcO1lNDWYIXzLZ96gIpqMPF4de4vkd+PVwY7n R9NBZCOaPmODTRpEAAO//+ZKRTUUY7HT75+FDb4NYkp8j5At6Jf8bJGSwT6RdLcrlfMGKL+Kipit NefvRooQvjIm64A3mrzu49neQZoYOnWg4/dRdJAHc68bCmf1lGb7SCrOX3wCY3tSsYRAwWBfQRX7 x7EJoUdKMrg/uYEHKRl8hSkV50rluMGTpkUg4/xuMGP4ZgvhSxJNxvls7yAlI14dbKPxNammRUL+ 2HvHhlU6vSXcsRjJWypDi7H4z/WODkWoYvlKhmx8i5RsMGObSqNI2WDGNh3uSuUYAlJX0yJrz98N RgyuyZqJjvrQRJx/nu0tMp1IvFrsen403ViuvddsWHEqzZ6SVJy0PZbfFyIEKqk2nWw/SiJ8U0gu +HBF4waYHuQvY8YQVhnKxshcNYukneuFgAPpPpvyWldEjyE+2VqgSRF/Fjour5oThDmRuN43Fkxp fqeyVjgXIJeFYrjvmleaZ86+NgEiJQ3cotuYg+iby1auKJI4I5VDCEitJh8/BxUe7DUlghkC97e2 eIsmo3y2twi8m/zxarGEHs+PJul5ju9fbGxKsPcppSbrMlTNZaMYO4c4pBGzX7lA3DD5SBpCksFl lwk8SJKCUdovFxWvVE4jIDWK5ILzCg/WykUwugM2Mv/wdRxFRMmg5tHcQcpFnDqYkcfvo+kgPOtG 9y8u2nSytPiL8bjSXK3TuruYswucg6IH9xlw+jV5QAGIlGTwEcMGHqSZwcUMqbtWihoaqavJZkDk 2vN3gxkDi9bOnTncQ5NxPtmz0yhSNuLVwYw90RxNBznxHTZeTHsr/dMv4iFN5UJb2MwoxmY61m6/ hw0XWH3ofLb2V+d1XO48yEy8dnojMokyX9uD1eox7iJrzaM1gxEB3dcsUaKHGHkSMwaKrYOUiXh0 sI3E6Y7qWeTGdph4McOpNOV1mfjBR4i5SO9iRJCpVvOCy/qSBSynpgg1ozvqKNK8YLdGOt5giZI1 FLSvJuMuAhfR1LwIZgzZyGpeRJNxKoWmsbcIzB+vDnY8P5oO4pOOps/Y2LT4yp69SYtARJA79rRK S9AcUAjf6MAnKfKrUkGRmV6XqY4MXzyKUCaiRyYC4Nj5VaA8S04KdMwRNUY4amrrAKUhDh3MoPVZ FtSTJ/oU12cksGsj9TCT/U7dqLAQGx+ZwfidPoT6wq6A7/+2Z/2WHw1QDtg08iiT8wcoHi15Vzhd wdQy7Y5qmgxCOs/p/l2JUw/xsTn+wdYC5UB3BknAcfhouSEcU6+LibaW9HBkvu/QN5bjvP+J9fDB 8J6PBrXrYBmDeIhZKeJhaJbfLVIaWHBPIRWpvBWcRUD7ajIXioy9ISKYIXDwqAOcaDLKeHXsLTJv Rbw62PH8aDrIie99Qrin0nED3jFX5AKAGooRA7ut/+Zt5HP22XRBrT2DEgaNTYAgZYObduluYWik cvYAqdUkG5xGSJlfe00LsaRFvjZkGxFNxvls7yBNjHh1sOO5bETTQU58n7HBFkMGDux6pLBrVcFg EswY+La7ZFCHM1NjmH6EwKqC6fqILGJQDtI8MrWYobPJ8Cq6wForFxEigl/MkrWMVZH5z4U6udZk zC2SzBinDmbk8ftoOshGR/Hz/j3hGx0pupnFYvDVzCgGF+w1+xoyGIpkuwSHGc4Qzd8ReJCSwfEx etvFjJ21eKRW0yLh/vxupIyBC/xnRU5Nxvls7yBmBgMGvTqYscfzo+kgsjH23rPB8g+P0/2w7v8p G2KJgbMZVhXUsH5j1DgjZWr8icwiZYO1HCqqxVI5Pmm6yNrzd9SXShkDpdUuzLkcCBvP9g5SNuLV wYw9nh9Ni9z4PssNPmSTepiPWjD6KhvFqL84rfEjDTWXlWRMwTTbGaP5O7aTBCkbrPHRey5mnIzJ kFpNF3G6XU1lI1LEwIr9jvec1nNu/cncQf7Pdepghh7HV9EBeNZr7XVmdCTi1hiHD0hzS0cnLhYj ArZ9/0xm8C3eRh4pa40v+R2yIhmKOFZLpxMoj/x7hFaRVBRZc6UimBFQS3h9TxUZ5LO1gzQv4tPB jDyxrCL8LmJeNLrP8oKLd3mWSOeO+eZFMUL4+oWjAb4lnhfIE4+UXLDnvjkQpHnB1pGSEczQme9M nVVNF1l7JSNSxsB3uq20HPS1/qZL56LJY8/IRTJGHa8WS+zx/GiSjef4/sXGZga1ASNcjspvWlgs MMDhnpXfSYp8yJzUZ6GoVYd3BM6fJYGZvnS8kTBiOoWjYv5eGw0fAVzm5ivePYNXg28FbHSGABvz 53Qd8WIho9TvVbF/Txwn5JeDL/dLMkzQITZBtZwYTKc539kelBvmE2eEfB7f+ZktIk+KnzV0JiTT zRYzVk4gpMasooMccyVAKSvtH78oAkJBNBngk72wMPbMSp16ENpYRhGJu8hGd7h4OfxyPoFxgs5x 7VeLiWIOppmzTgaw6TpJ7vdKOzDlBlIKSKgIUirYPlLCghk4RxJSZqKH933+PrZKRGQkgjFYW8jo McZna4vMixCfFsPLjWQ1XWRj+xcTfRHcQdlFAy4Lz3YjNQSDCSas2Xdumn3hjksj5557pEwLvl6S FAiSJtJBUnckVsrYmX9JkVlNB4GN2isbkYINSo0MO0cTcdJ3Pto7iI3keHWw4/nRdJFr73WH0QGI axhtGP7w5YUWVsXMC1by6U20y0J2Y4+UbPz2d/ITxKhcw3AZcbHUnUhdTYuYh7XXjKKBzkvODu1c rTmajPPJnsVFkbIRrw5m7I0GXqPpItfesPGPFoPTwcwpI82enxZW7qp0IAEbnM/wraB8c4OecSrF u8h6QWQGSZPhEK17EiOVV58DFg+aZCPI2ptutxgxfM9JoVFkmHHqmDtIyYhTi6WBaDCjSC6eo/vX e7KZwZRIN3xy4a5nINQQDC6Yvaf6kwsOgqbIZLvJDFZzhxvsFGlm8LmIdLLBUhNwMiHlajXJRRC4 YHM6mpoZxfi3bE00x6KJMBkN4pVkxN4i6UXGq4Mdz3lPqukiYw/u378nbPLJzVsMDxmGlo1gssGS 3a+wwT7WjLMYkGaw6lCog40iZYM92Ckwihk7s9spMqtpEfcT1l7ZiJQx/Pqxq3FqMs5newdpasSr gxl7o1lNF7n23rPB568p9ZF2+0/JEPI1YRafvDU1+IB9qkwuAGpPy0eqmwUCpcJ6wFwBSVpwH/yD FokIsqbKQyG8ZydDx7xTVrCR9VqytQhQEuLNQkYch2UzWi4yUf0rIbYb4UsGUO8iIHdDDwfB4IB5 WCZe5CCL+gTIIcEZo3J8KA8/yHQj7BVJx1spo2Yckeqymg5y7JWISBECc4D5xhsNl5r66aVHewcp F/HqYIbeaFbTRTa+f7HRxoJtk1QOOvezR5C5cyOQXHC3z19eVCo5r62WiwhZ5XgpzQGaEEzXJCMq ZODsWjSvoqdFJPqOrRKhjDzkCzE2E1FjhE+2DlAadOi/g63PR88BNq7PSGCTRj3j/93IrAKxuO/+ G3n90kH598hIAlcBlYUgesYQ4rs7Hu1p1CQLnKy4eg4gDfOzi9mt+CGmVZR8UIqnMOYMen63Ph3o eC0Ro2jiuObenEj39+zUSLX5k9tVW1UMxoNkASP1JpUc9+74/L8pFef4emuefpC05Q7HXEQcqXDB UQSkVpOBB8G72Jt+tBhzX9y8vitwaCLOP/FKMmLvIHI/Xi2GVxvN1bTI2MPPd62lOygZPOjdz9+U Q82LYMbAKRWrCpY/XasyTqVAmKj0dwdpZnAFY/rbSNFk5uRBKs5qOsjam4aW8xeZ1mS5bIad0WSc 8erYWyTsj1cHk41GAxujaZGN7zM22KzBU0Cazr3TNeyqFJMNpgysK6jIOR/tG8DmhxmfMmRsHgQp G9wSmoqzUsbOWYQHTYvYXtReW5lIkd/5kJbvSRQZ5rO5gzQ14tTBJKPBHEULXGvvM4OTKD/1jSRg CNXMEHPaghUPmgq54HRK8oDdJUj5pPi3ZEGRcsHNR+wLZMI8UkbOYseDpoMce+UiUnLx5Qs3dUtG NBEm12Q92jtIyYhXB5OMeP6gqciN77PM4Cq8VJx8fHK4CCIX3F7jO0ENm1vhXBawp3dE1DGHf5eH X25VkodMS1I+R2J1yEIQWWBNqBwUcaqfD9mGg2gxukc783fj1xNbk0pwQgs71hyro4iZbjyfxU4r kPqSA+b0ks0EMae76Mw4yWb03Krb5x4p42cmL0UWWyr5XRmgKktHW8yI2THwoGmR+Da/uxgR+PX0 sBBNxvhkzz6zSJmIVwczExrN1bQIXIy9eS9ejsPYJcmtrXrH2RRGoLIxGGyw0sFqmGzwrZi8/XxP iZuUZIOrhPPsgwwbM3NTyDC5SRihVXSQY85185HycXInna+hgzxaUA/d6NOxdpByEZ8Otn4nL0bR x+jeZ4Z7Ja3x4YKbz5i5Ic7BDIFDJqGCR5Y0+BIhneNDtUmCIk0Mdomkwy1mlByqyp7RKFrgWuvb oZCjDFpgGlepiCKjfDa3CG+VzydOLQaysVxNi2x0/+KiRTe7JZmUgQpqaEadTQshmLCklQpOWucK ONrICEkFgyM7lwBxjY7ml5sdV8jA2RNw9SxwbJUIZeTBW62xVTW+DRTiOLS2DiAN49DBpCFhXD3P ceHhu07DXZLW+LDAaYZ2oINJAyz46lJ+c2dbdspEStc4MJ8Skq0S/E7fMlJJ8VDMsNmvhNRqWsRG vPbKRKSgIpdhHUUG+WzuIKUiTh1MKhrMUbTAtfaeC3YJsPUTaeYR2oGycRIsEXD9hg+F4tVvQvGw IyQVXODdxx+kVDgxGcJGiir8g6JBjjl/txgRcGtSlyGjySif7R2kXOjUgaQifsvpKFpkovs0L5is S5nJLfUMh/p2BONpUhUx1S4XNKk5ZMc9h0hJBtdvJgeKlAz2jNDRMA89UqzJR+pqKiL3tVcyImUM 3G3g6+jkMpqIk0WAR3sHKRnx6mCy0WiupkWuvfeZkX2petfroFQpZGIwiU/EksFtUMkD2NnRmSLO vwOUCj6mkpkqIRpfN9I8qxlkbZWICEEE0w6zHBs9hvhoy8YjQGmIPwsdlx/UTBDX1msStr3koHPm MxlpsVOnKRFMFnhWVhVUabkNjqAjZcvNuK+PX2SaTNqTzFGtFHVipK6mQXQumspEpAzB3eAQ6DI7 mozy2d5BykW8OphkNJqraZFr7zUbnaRg4yQlqw0nS384UDbErLGYwafdkw125SUFGIWdkZmjDMZj QfTOsZnLhouZFuybfdJU5NorG0rZh7CacEZ5aCJO1gIe7JkYRcpGvDqYbDSaq2mRje90Iy9rLDdJ sulQNjio1dmKwWSDtWJfCC+udgxBnErx9BiZ+btF5h3hewiZrooUvyMnInU1DbL2klOLsZDIckJX IaPJOOPVsXeQshGvFsOrjeZqWmTjO2y8XB9jm6R7ifTuOxM0GaMPJhsU2a0t+HB9M0Ep4nRs5mgD foI0N7JANlBajN8RuoroXR/NTaFajBCYBtxhXkYYrAQ8WEtmHGtU1vokFyO1sVxFi0x0mJv35B+Z wSWg1HtuheTWiLwnbqd0TOEGTW+DM/9YI0we8HEfDsLKBR8kSxYESVCMwjjiKz/FzIzcif6gaZC1 N5kRKd6Tnxw5aDcVTcTJDP6OBePnInI/Xq1UMqPRXE0Ty9q7bPwjM5zOVphW6/ckhpARcHTKYosq NrfCEaVCyQsOzCTsALrmUK11aIRwjfLxWU+BtRUCB2K7ATPdR4nxPVoyJQqUhLhzMF+PBLFa5s+J 6Z8EtBOxjGHUoFffv/IfNpuDSQHzm74I1G25C84pNaVMB+b5+uhF8lgZoLmoZTqMFBVopK6mQY69 vlKRIh1YbtqhnpqM8dneQcpEvDrY8fxB08f4zsvxj3TIyiLe8Y52ZMooiGVL+xDeW4aIkvGDnpUo c8efVHD94wWaDewMSTczMiycRGa0JD2CrKnyUAge+MBOEyJ6iDDP/NhaoCzgjikyInXYTmi1LDJR fUYBezR+Z/vwL1JjEkLMd4J5e25SkQM+W5/HzwZ/pOxL+IxeHn6R0vCTzXRW3SuVAwhPmgaRh2gq EblulhDYrDErkdFklM/2DlIq4tXBTIh4/qCpyI3vMzbYqkErjXfcobkZEUw2OLbp4gfFa+6BI85I yQY7rZsAQcoG2zrTv6wU5XOkrqZBjr2yESlj4Gb6jkOjyTif7S2S13G8WoxnxARoormaFtn4/sXG NhZ/vZoM737/ZTpyciMYbDCr3+aSzi+JwBYchCSD7/6mzAgybQV7MdLBrBTkReooGuBYKxcRkosv XOgJ94z2VESUrAQ8mjtIMyNOHUwuGsvVtMhG9y8uWm+y+kifES4ohlphFbOt8KiJbwU1bG6Dy3Od oTt1X7LAwyd+xLWDNHbq854USxWN1KMm+9hHe2WjGLtsfnvz7dFknNEu+dqzfShSNuLVwWgoJpqr aRF3ddXPd1UFv2OzTNjws9HJjMFkg8UPEwEfciEccSrF02OM5u+MPL/TOxbR+CbDxWSDgwdIXU2D mBvRVDYiRW5wv0wGo6PJOKP92DuIvxuvFsOrjeZqWmTs/bNT3dwgupxl5kwnS312quSZGDEwg8/1 Ktrl+/XNBKVAHKt13GE/2EkcR2t9d6JKMv5+UDSIZJyfLYbjftJRVtVjkE/GkhjnV+PSg9S4LanV s8A19j4vuG0q1SbLSfSaZSKYTHAjke2Z3XxHHZxNpzOTCd6f5ECQvCWuVrmiSI8RXWY7lwWlbl1N g0BF7E1TEylCYGSPuFxE0x/GVKNp7B2keRGvFkteNJqryZLlMb7P2gw+kuNLwu2a7IgoGUJywfFz p3Co9N3+YZQKgThKS1dSoK8Im0VScgTDNRdEnvQUWFvTzCgjDwz/OrEcNUb4aOu/A5SGOHQwUzlh XD0DTFz/fDnaiTCknfV8ljTahwwEC+xWZzFMFjJtIgsKSQJfAuvDB8hzdXTGsPLKUINH5qoZRBZU YzgL4T1fNeiYN3oM8MnWAiUh/iwkB41i1KC4iIXF2Ho3W8HEV2ab3UdNZTHvRTBZYMLZt4DilYV+ AuQmjwxRHQB1lFGkuYB08mOlqMNnujuKErSIvtVciYgU6fCFK+Y63I0mgmQp4NHeQUoFTpkiK7Sx XEWLXHPvuWCnRipOJiMYdPa1EPO94BaTNhFMv+b5MyKZISrfbik7QcoFCZCCc6VYHnhWNIBUnJ8t RgS8zl2JVNH/jFIpngWz+CpaJLlEYa1TBzMJ4rekjqZFxh6Z+44L1mnzjiLtuedwsRiT9xzUsf2i auOorXW1t8W06GI1uUkQpGSw3J2uY6VYJIj2q6lIEuP8bqWIgROJHe9Gk3E+2ztIEyNeHYzTaRPN 1bTIxvcZG2zKZAWdbWJu+Rw2xJzL4S40zm7IBhMFSQS+woeUz+qHMvBTpGwwM05FtZgtAbM6T5oW WXv+bjGscCSgQ99oIk6m8h/tHaRsxKuDyUY9v5qK3Pg+Y4PHmYqTHiy3bKqzGAtcuGdrBtffO/Lg K3lnrNZxRxG9c7TWPYkrlQMHaL+aFoGNaCob9NiZj+LKgTPuQ5NxPttbJG31eLUYHqznsDGaFrn2 Xr8p7UbcVWndz/+7GjG5UYwYkuiywcasZIIjap8eoyJ/Rx4EiXeO1roxcaU8mKD2q2mRtVc2KsX7 yMZG16hHE3G6IfHB3kH83Xh1MNmo51fTItfeazZaccIZa4NhgyHzkCHkCg73l3PoUbOcCU/Bye03 O1TruCNAEyMfCZCeyngSQc2jhhgXwTNF/NVCuE8/uWM+d1BCxJOtBUqD/tiVjMxGcdUscmy9J4EN Ymz5Q9jWd1gQ032WJ+g6ZYFP1jcBIpVBmjLGHKQ8sDZqa1HIEPmW+JOiRdZciagUETA0l2PHX+ki WBJgyfZYM+pjbXw6mAlRv6+iRcbcpx0Jn1I1H77xLePWF1bsae64UjHbXiheufjfTpY51x+mr6MD 3pYB8mgdobk3cTGZmGZz9QywtsoDMo4t+GjMrNJGDRGyGIBDa+sATYg4dDBpiM+jBw8HuLbeJwT7 NTK56Z3LrSyY6QDztWBFyM6Cui0H7wkwQnrGjGZDDtJ8YF8JFfeVynEElFdR3gt2rjxYKw/FCICq fsd4KDLGZ3OLTHsZpxbDzfrtAG41LQIX4+ZrLra9dN9PMoKmv9srnPQVYzsMq+M+FCo3z4URZYTk ggKjcYtMc8mWkRRnK5XDB0+KFsG5aCoZTKgjRQikwA7z3HYDGc/2DtK00KkDyUX9HkVhR8RXf829 5qKtJSslbjtEmtRmvGHFuRhccLFRe1IurkgacF3VjEq4/iGBF2liUFmHspXKOQq0X02LrL2SwWkL pKgh+W4ODjjMQxFhcpDz0dxBPFTKthKdOth1/CjaSK6111w0L1xCZ/ggF4TdEfpi7CzLN8o1m9aT xKDGQ8rE4GvEjVtkEoMb8NN+rFTOIqD9alpk7ZULTiwgRQgcx7WRdLjndAVkPNs7SBMjXh1MMhrN 1bTItfeajc0M9m2k3eS2dE4ANjPEbDE450hpZf5RJyQPfkdKNthmm8iL6J2DNdcTXQ5Uly0ndcKT piKyX3tlQynZ4FjNWY1zFdDtM9VUe3adRfzdeHUw2Wg0V9Mi196w8XKt0B16TFXpHeG0Q10MNrhy xEyA2UxQEKdSPD3Y93dGnt+VDe7cyxxeMdlgy1YqztW0yNhLTi1GDD94x8NrNBlntB97Bykb8Wox vFrPr6ZFNr739TcRu9mDsZ2f0fJDU+osxs5HP7YlGd+nC/UbVh2ecEIigRcpGd/9zD0EBUvCc8Yi BecoopJ8tDZ9cTHePVZnO9KJIqL0c1zXnJlRpFzEqYNdv6+mjWSiw977zGBOmKoQaW6CasHppkox uGADjuNUqjf/0TD9j2QGEzwNPP9UMpj7zSxepPid5xGeNS0y9qa1qRQx/KRY5Xc2ne5lJPZoNzNi 7yBlI/+4WDKjnl9NG8vGd9h4uWxICcB3BcMG8zPdX7CYOcZ2urDx0yOGsqEUiKMifneQssGhyAMl M1hkSuG5ijyg8GBuUqMYIXBqqgMPFRnlk7VkRpByEZ8epNbvq2gRuJjfvc8Mzn5KxZ9ctdKXJBBM MClrjjJFxtXSBvkzQjJhfhygRFCSpPtdIY8ipL1YPQOsLX81ELOoXKALe1TZaiFA7gyZUZA5eYCy EH8O5vsRl6sGtfnbbb6N6rN0YNMGNu2r3QtaDopBAqMmN0pRybqn25gjJQnM4DXiIGWBfSPJh5XK IQS0X02LrL3ywEwfUkTAJX6+jAzQ1WSQz/YWyUs1Xi2W0Ov51cTUy1N8/2IjBcZXpvD+MA8hG0yp pUs9GDEwqdYuFWZDRoQkg+vTGrhInGP+iUmfR6mcQUD5VbTImisZzGciRQjctuNAZzQZ5rO9gzQv dOpArL5OLKPof4skM6LoX1ykvOD3LNJT9FFms6pmYhyInvkXDyiJ8ZfjhEbJ0aiOUH74qwX0zIFa tymOTI4jpOiMGkSK6FlslQdOMaTmZEbgrMihhxBR92BrgbIQfwb677rM2KBqThDH1psaC2G2bWAT Yab1ci3WwWCB1QarCuq2L20IWTvk9jczgh10qTiLlAcmj5I3wWAvZxFSca6mRcbeNJecYUjF+Y2D Q1YVDPrUZJTx6thbJL8brw4mGY0GMqrpIBvfZynBUAi7SFOAZ4yKhmKw8YVTtWGDfTS5zAqLSOkd K2WNPEjZYKtn+tuVIppov5qKJC3O71aKGDhzYRo47EOTcT7bO4j20KpXB5ONen40HWTj+4wN8jLn bHgZmKvqGyLmO069mTNMNn0de3CwCynZYHyfyIuUDa6kybmYlbLpe9a0iIs3avJ3i5ERfziKHTai 6Y+jkid7Bykb8epgslHPj6ZBZH/tvX9T+DhMKs5fPygLh41isMHlxbZl1FvsvrZ5+B4p2eB6i1Qa RcoG39JIn7tSVAbRfjUtgnfRVDYqRQwZvZsb0WScz/YWSVs9Xi2Wlq6eH02w8Rzfv3JjOxLGUak5 mRs5rWcx30tOuiUnGaU1EyJlR8vYqpGLxLuMiNL5sr8sDV9GUqleV9MitjvKlA3GW0gRA18Lde/P aDLOZ3sHaW7Eq4OZG/X8aDrItfc+N7ghIWyw9cgrN8OwmCNVPhPEgNHc4DxoMoGLn5CSDa4dDBtF mhvs3zSBVohCOspHEWEuwlJvFJWMStEZ2U/KhXoIkpMVj8YO4gzGuHQwqWgoq+cA19hrJpoXztH2 BBMTVIxIZWIxAmDfnb5RxXq1tEFGSiYox5MDQSYv2EOSvnelnKZV+9W0CN5FU6moFCEAdio1mgzz 2d5Bmhfx6mDX86PpxHLtvWajJQbrW5wmpIHh7FizYhAC8ALrPAFOSDdwZWSCS91CjX83I9hkcqFU j6odJfzGowpMuI2dcuA5Rl4NLjY9C3MoMbhHO/t3448rA1FbjLNHyQLHzvvY2bGRwtsV/VYW7rwU Y87QOybzlnxhZp5EoGudYSob8vLQi+gYQzP3+q5U3gDWxB8VWUQ+mpsGohibCf44hDiaCNG7uJwe YSY/vxvE35HyOrVCyeD6fRTBxFN0//s/707pI51CigdEI03D37eimPOn3NgUs3zNtqErFefo8/Lk uWWb35UMlo1TeRXz6TNFyaM9mg4y9qZerxTSXFLSEW80GWe0S0bsHaRkxKvFkhb1/Gj6f+L7V9+x bwXrUim8OZzC1ZZloxgx0JSHDG/tyjNXKC8GaxzNgiAlg8mNdLfsSUOVzR6HLVC+ig4AF+dnK0QE 1JkttaPIKJ/M/XeQchGnDuYrUr+PpoNce/OWvJrVQ5oNHKm/v3MbcWusxVhT4KwG1TYJScOZ0Nlo ydEXnxTX3ycLgkxmcJwmL1MxyeCsBdqPpoM4olOTJK4UrxBn0Kz3/Vc0EeeveHXsHaRsxKvFkhmN 5mj6f+L7LDNY+k39TYZxeX8zoxgx8HYwdwMbzLg1dqWSGmw3TRpwJRhIU4OD7OltgyU1OJ+A9qOJ yIsQ7/ndYsTARmnrfZo7NRlnvJKN2FtkGo14dTBzo54fTQcZe7D/vgUN+bQanAdi71HZKEYM7icK G2xrz5er2EGVq0isvzryKFI2uCIyPe5KUTkqJRvVBBsidiO119yoFDGwHz5kRJFhPps7SFMjTh1M Mur4KjrAtfaeC97U7/rG9mUau3IhZgTsJ6boNjN4ezM6YT/xjFzZKZ6p3CLlgqucWJpm6mSkqBuf NcFFkbVXLorR23K7fkfB1cSkzwd7i5SMerWYZNTzo2mQG9/7zGBnJYd5LSy+cZK10ziLwQZzk+zn hQ02GzT2SFl3Md+Vupo9bfyubLBNwGKiUArHCI0injmlJEicOz8r5qD0Nxe/drkyiojyg7VFyoU+ 2aqO1Nf1+yg6yEb3GRd865inyawlqtu3ut+SDTFkBnOcHvCCC+4PTuTU0kjJBbVhaski5YI9JFmP XilWNZ81QUaRtefvFqMa/cuZiLwm1cQU6Ad7i5SNerUYtzqO50fTIDe+z9hgysnE4AsEPHtfEzZZ Ctm325dQjZOsZj9Xj59RGgILGJDDNHqCgxmio4jVsn+OnS3XlcBzlk7bKlcJ0T1bWkAK6o6Na4RI 2/H36FlgY/qMAHaUpOxkBMgtIWWgGBTQM5YErzMzPIWSDb/ys0VKAvtHSkxUSQItRsrOUXQQnIsm f7dSREBxMeO8aCLI0cTAK5oGSQ9Cua1TC8lE/T6KDnLMvS07uZmCyWGzwVnfTlksRt3LMQW7U0pZ 2wLI4EIWepIOz/zdIGkmHKC5YXkxu0UOXaTsHE0HGXvJipViV0FGMeZWNbFyEq+uvUWaF/VqsORF ozmaYOM5vs8yg6kDHicecT6jxYXbK8Vgg2MKVJmwwTbDxq5UUoPJhKSB5zpaXDBIY29D2Igu2eDQ xaOmg4y9aWwrRQw//QaebFQTcY4mWuJoGmRSo14tZm7U86PpIBvfZ2ywjyOFJ7Oa1JZ9T4rBxpec H4INzpExqchTj5QNKZuAmwdB+qKwfTfd7kgROwcskDqaBrEZqz1/t1LEQBNlh4p8NBHnB3uLNDfq 1WKyUc+PpoNce+/KC7dXchaZ8TJTNYzByoaYMTCt76WTsMFnvhM7jcYO2SLj6yxSNvjMN13cYsbJ brNHTQdZe2WjUszuMCK0i6K9iibPbD3bW6Rs1KvFZKOeH02D3Pg+y43cIi0b7rItGYUgg+28TFxA Bnuf8grwbaUdsTloIG6BUvGbE4IHkglOK6TojJrkgEg821+tEO5zNdQO/dRDiM+2BigN9adQ2ou6 fNTAwlNc70lwG6VVPyRwoXPnLxaDBY5IW1dQ1nogy6CVAmGslt8tUh74clpqjUjRxDrnrPaj6SBj b5qLShECtwzaT1GtRxNRxqtrb5BpLurVYox8J5qj6SAb32dsMP/OE6AM+ENDnpxgz2UwGvTfnvWQ Debkk/p89eeM1fwdOxqClA2+NH8hyWCKEqGj6CBjbnqSSjEVzEyGRZbDMBSxO+ODtUWaF/p0pEiM 9fsoOshG9xkXnIPhaSL98xufIfb9cNtlzx4xo28aUNZS4ydKhZIYLDEnCfiCD4hUOFRzmyL0RJVc sLMVKbiIogOMtcmLChEB3ySxwlhFRP5kzkq7iFysU4tdv4+mE8m1927+wo2XtIgIs92yFcZABMBW TgsMCjq/pWOQETJnvcfgAGWCDSTpd0eIp+tFdBJRPQusLX81MnhPteAmRlbBooYIH23l8QcoDXXo Co3PowefA9gqNa7PEoKNGzxFhgF/2Lw5L0cxN8ey/TjtEeeDZeFHhGThT362SGlg60iKjZEiRE4q pOaMogQtonM1VyIqxarRX3a/JSOqidMFH+wtUipwyiRZIfeKdk/jKIKK5+g+44LtGzkCz4EQIunL IeajZCLfSydJCdZ1k/dslEJKMjgzkedfpGSwxyR5MlKQgYZHTQdxCKamklEpyGCjcLhQkUnA3Zws kl5zg6TVXKcWMwvq+Co6wLH2tv52e6UDCKQ5sNlJi8WIJF+3lIufdmomhlI4x5govxtkWgqu102r GamkAccUUnGOJrgoMvbSai5GCN8YvISMaiLMeHXtLdLEqFeDpdlsNEcTbDzH91lmsDCVtoK5THZ9 NzOKkeTcgGmNRV3rfXGwwfoITaPesVuI3y3SzKDeSrcbKV58qzSljqaDMFMRTc2MSrGszNjF0iJV GprY2hivrr1BJjXq1WKmRj0/mg4y9mD/XcXpVkpHEEjz6a9s50RDMWL44bemZIPGI/U3X22ekStH +5MHRcoGTNHFwVCliJ0TC0gdTYPYaNRe2agUMXCTRIfB1YQnH+wtwu+OV4vJRj0/mg5y7b1jg5aY tbp4xzcqWl4USwzcd2ImUNf6RQvjjJTPiuWP5kGQssGZ7iy8jhSxc3ABKdiopoPg3fndShED9xlY XzgSUxPRfbC3SNmoV4v9+LqeH02D3Pje5wZlm4NRnhW1Mfc4+KYsRo/H4QVWrmCYQiGZwCyDtz3A xs/8jk4wiN45XnN/4mIpISN1NMHGnyd7srgYm07YUd4WuprI/A/2BrG9OV5dqfX8alpk4/sXG11S ZpTHjo+w8c3DLmWjmP0/9+LELlewJhOYyUVKNn7ld5YEIPGOdcY2GyNEmBzjTOk5ig7iDLCKSkal CIFNazbRo4goP1hbpFzUp8V8ivV7FOFlETNjzb1/T9hey9Nkqj5nW8tFMSpHlnatNDxS7+CCojBS csHcRbKgiN6x+O6XE64U+c7MCFJH00HWnmSsFJss2Iv2oIju4IO5RdhqcJxazNDr+FG0wLX2movN C9ZnQgWf4WvnyqyyEP4z62oOUJZ7eZwxRkgmuNrmAJMULB2myxkhQnQW9+pZAMeipjREBudZJdqB 33cqHwJ8trVAU6IOLSYL8fnqGWBsfdKNsAxjL2lCfOPRzMtRDBoo/e1UqWa5Nc6C+2ukpIEPYPXh B2lCsIUktedIETZnFpA6mgYxXWuvTFSKEFhK92V0hU5NRPnB3iLlol4tJhf1/Gpa5Np7nRJdWWb1 1sP6eMee7PaphQyBKX0WhiUjR3tIY2XsVb7wI1MiQJlgB0k6GqEMKxhzInO0QESRMVUeCrHF4m8+ DAgP1UMd+WjLHAlQFuLOEVmHr5ZFjqnXFOxbwSaO1JzsSDscFHNjHhezhgNuJ8njdzGgLFAM5uEH mReDTSR5WUaKqJkZROpoOgjORVOJqBQh8KVclqshopqI8oO9RUpFvVrMhKjnV9Mi195rNpoQtLPf LPyRZuaodUUxZyKZ/GcTiRnBZUfNgEiZExwYbQIE0TuHa11c5BRlruygdkTqQRNsiFx7ZaMYMTAi 83VcTcT5ZC95ceytV1dqo7maFtn4Tk/6crOB9YhjiCxx9yLKr4s5B+N5Ddng1m2ecRa5vZQDNvI5 w0VSETtec3FxpNIvsNaWVnM0wUaRsWdOHYwYyMAd+amJ95JT+js6VNMizY16NRheredX0yIb32Hj 1UERfs8tL1QMSNOA56DIwWCDmVebMsY/bBlLnEqBOF7zd0QeRO8csbm8OFhaDHYcIXU0EXmRsTej umK84rR8nVCsJuKMV9feIGF/vVpMNhrN1bTIxvcZGwzIU3N+4/LpzmE4KSRGDKxMOYdFZeslcsYZ Kb3jFW8eBCkbTN5TRxwpYme141HTIDbVtdfcqBQxcLJux35qIs4P9hbR3nq1mGzU86tpkWvvfbvB tfxU/myN4SrGzmIwdQomG4w9aQi0y60XyQRuIzzjNWWonoOUDScJD+Tr/+WDokXWnD9bjBB2mU4G 6crVnLeSz9dha5DJi3q0mEzU61Wzfx9Tb2cwWE/xY0rywORli6zF4IFj5L4R1LC59o2olYpv+QiT Qed35YHZTfqcxQyRKbnUm6tpkbE3vVClKAV+cX829hz1qYkoo90srL1FmhX1arC0GI3matpYNr7P 3hHOSoeMb2yHsPZm04KIu4TdkiQVHJ5uAiAD4CitKeLf5YFtJPQ28FARTy10m/0oCbB28ptBcDxX U0hBlRDck53+Hd7WlSMyzl4lAxB8fXv9RrSucN+OowW2zrF1t7XVYizYcEuwrRe6vS6O2NxK297k a343SDxzZMas9YNUjiqg/WpaZO31leBAA1JY4e5ge6bVRJ3/wd4izYR6tZjPrdFcTYtce6/Z2LqC jRs8QzcSclQlZTdbAMGsBpjL99QfbOTcNKFHyGT4lp8t0nRgP0k6mJXKSYUUm1FEmIusuZLBeQak eMxUTR3wVhNhfrA3yKRFnLpC9ftR0SLX3GsumhlIcaO3Gz9YAWAtzPdiMVY+2UhoHtBL26TzzJnG REoyqBiTBUEmM9hlQs9/pXIsAamraZG1VzKoK5Ey52kbkhnVBOUf7C3SzKhXi13Pr6aN5dp7zUYz g5eJm4jCBlNL7UeLmRnMY9Nkywb3QKYN4ETkGaL5OyIP0tTg4ts0HiuVgwrzhfNqKnLtlQ1WvNDF eJpzTO1Hq4k4n+xZcRYpG/VqMdloNFfTIrAxfg4bLytOelvWC+MdI8P2o4vBBsc2zASqWC+PM3al eHr05PndIGkDHae5kLeYrwVnFVJxrqZF8C6aykaliIHtNJhbRYQZp665RfzZOjUYTq3jR9ECYw0v 32cGx+x5mrQ0UJfPfKKyGDsh+KygLzE1rKe+iJINImes5u8W0TmW1Gx2Fku+c6IAqauJmvTR3vQ9 xQiB2wFKRjV5gY5eSQa3JaFpkPxuvVpMMur51bTIxvcZGz/p7cIGlXMnLpgDC0YM1M52p1RtnOS0 eOKu4zNW67ijSNnIGfcHqZxcSL25mop4PO5PNPm7laJ+YUrV99Jxn0uVxPnB3iJNjXq1mGzU86tp kWvvfW784lLUeOex9DahQnmgjIocKLl234EHC3JW447UOuoIUCp+tP4USVpwa8WTFk8uiOCYWspD IeciOYoZHqKGAB8t2VoEKAn15sgwlx7FV8six9RrCrYb+f3DUp+ihNe8BQYLWcHchcsOmHDAAl4a AS5syXDVoVAHG0HSjThEc3mRRaGR8gSGUlfTImuvRFSKEOjX2QpHQlQTZdwHe4uUi3q1mAlRz6+m Ra6912xsN0Kahgw+V7sJEYgIOL6Ba26Q7LPm8pczOis3AfSM4VkOAsHECHliIW+GapDI32YehynR Uhoi4vCGa0UetBDfs6kFSoL+mCMrxOq5pq6eAa6t9xSwcYNlbqQ51bt1hVgeIwc3bL/U5RiCACNl 38pemT78IKWBpfJca7lSOWTxpGmRtVcmONCAFDq57NeX0BU5NRHlB3uLlIt6tZgJUc+vpiI3vvft JfvYnLqDDU4TMuqyiViM3aAc3ODAEGx4yw1ksOv5DM062ChSMtg9ki5mpSijo3wUobqIztVcyagU IbCfw7fRDxqpif//YG+RkoFT/z0Ird9X0SLX3LvM4BXgmtQ4x1XFrSqKmRmsgTAol4usT0hGpMyM n/wu9ATROZ6Q51AfpHJSIRXnaloE787vFmPuAR7MxNVE6B/sLVIy6tViPkWjMcdW08f4PssM1jl5 CBnLMu5oZhSTV7btkhnUbZzjMna+BDujkxRwBykbLJCm/l6pHFVA+9W0yNrzd4sRA9+c7bi3mjIy f7I3SBrq9epKuSfUaK6mRa6917nRboSqkis9wob3K5aMQoTATmVfYgo39guFDIXMjN/5Fex4C2Kp YCY375IQMpSNkblqFsEzRfzVQvRKDMZ2MU49hPhsawB/tf4cmY1i1CBSxJwfW69JaO+RISg2EWbK un3pYrjv7pikBF8yhGqCjpQ0sKOtMQcpD4wm06usVM4oIHU1LbL2ygSLT0jZP7MnixAy5EQTUX6w N8ikRL26Uuv51bTItfeajU2JXHUjG1wg2cE6MzZef8MAnhkoT3nxDGClqR8p2eBG8kYuMlnBZuAU 3txAnKFcjis8aVpk7ZUNdjgjRQy043bfTMtEE3F+sLdIM6NeLWYi1POraZFr7zUbmxu/fUNhg0Of vYcRnWJOd3L+zw6f6pUpWFPja4RsTfNiH6SpwRR1+t6V8sSCyquIKBdYa+WiQpSvOdEpF1VE5/rB 3CCTGXXqStVvy/XVtMi195qLzYw/njqWC3ZbbqdajBDYRWHDThnLl5hCRqQkg6s2GrjIZMZvNonL zwh5OkGhq2iRNVcyKkUIHFDa8ReKiPKDtUWaF/VpMfOifo8i/C5ik9Ho3ncjbqVoWjBh2XmLgaja mO13WMB/sSpiZ+l17TNA81cLNCvYTZL6a4U8q5CkWD0D4JhqQt9AeM+VD76PbppQDRE+21qgNOjQ ESLhxuer50Ncn5HAjg2Mulrw50c+FoHGYgTPkNyGjCf6tUMObv4+Q7MOOIqUBzaNpMcNlpeBUwpo v5qI5dHedCHFIMCNyzJRRVD0ZM6SqkipqFOLSUUdP4oWmOiwNi/Hy7kbugVvIpALCqrO/S9GBPSK Dh/oBBg1kABfIuS78ZOfmSNF9I22isrrUYr2LMqvokXWnL8rZrHNbr8d7UUTUT7ZMwWK+LtxaiGp aCyjKHnyHN3h4uVaITWkJ7nkgt6zNVYxG01u9WBbmmYZqCfz+YhHTrFTMeZ3kBGkZLBxJC3FSlEL IvWoqci1VzKUsrFjxZ6kl9ZqIs4ne0mMY2+9ulIbzdW0yMZ32PhHZjCvSASsMDJ12/LC/ZdisMFh GYssqlgnsCgTWNXzKkpSg48bIDNIsp0xAOePL2a6M7p40rTI2JvmtlLMVrHc3Bqumhiyx6trb5Gm Rr0aLJlQz6+mjWXjO2z8IzfoM1NxckKaUZglJ/NVwWDjGyPBsMGSB0+P2JUCcaw2u4yCNDdYQk9p FSm8Y3rlg6ZF8O78bjFiYIrZ95KJh2giznh17Q0S9qmt49Vi1/OraWMZe7D/ukttecHSBs11coP3 vzMXi9GR5jJdUoN9KJnGZt3Q60loUr/kZ9ATpGT85EL8UDZSVJCROooWwDdFJjMqRAQM4HwvGfdF EVF+MLdIM0OnbEgilcxoLFfTRjL2bgv6j8xghwkBUC64AXUyoxghsMXX8oLyjbGCYf7JNlXJ4DBp Ai9SMtgrQk83UskMJJG6migpH+3lCS/GLhKaKd9L1+jURDUSr669QSYz6tViZkY9v5oW2fg+ywwW rrGLNOOoFlvsEAhGDAzIfCso4BgtZE9NpPSO7w428iBlg4trqa1gY6QoIT9oKmL5U3v+bqWIgSmu rlhWE3F+sLdIc6NeLSYb9fxqWuTae/eesNuMPd7xLhtMbDWK+UR/cojD1yLxNRMiJRvcttbIg5QN LzqVjBGKUyk9V9EiOHd+thj/7/yameFVqCaGMqTYGhtkEgPKcWmx6/bRs3EcY2/3GLjZEl9db3EX VokIBA/c5dIqg/mt5ICfx+ngdWgIEM8cr3W3YjEfN2uq4XP0DLC2mhLIWGL8hD3fx1XD0CUOSUNs LdCEqEOLyUJ8Hj2Q9yGu9y+HOy0ZhsoCZ7BbYBSzLOBIl/eckHR0rXn4bFidUStTGvyO1yaIvjlU 62bFlaKjR+pRU5Frr0woZYHhl5PbRVUTUT7Zs8Ao4u/Wq8UMvdFcTYtsfIeNlwWGmyvZ0isbHjpv ShTiWTAqtx2jmu2da7RikmOn8CXr7wHKBFtH0umODKV0FF8ti4yp8lAhvGe3uHM4JHL0EOGjLV+C AGUh7hyRjWG04GARX/lEdSh42Xe405K5KiYMuLCYJC4HxcxH7ozCMepact7Hz/VQSMkC5VRCLlIa 2GCSumulnF1T6mpaZO35u8XoQ/7kM3kQUU1McXywN0h64PXqSm00V9Mi197r1rIDdfdVOrpGmi2U ncJZjBjYr2iNRd029639ipRs/M3v4Eck3jHq9LYYORspatBIXU2LrL2yUSliYEuVzeNqIs4P9hZp YtSrxUyERnM1LXLtvWajNZb7Km2okeY7kp3kLOYrzgE3OzbKtrkKhHo8w1ZHRf7MgmPnKx2tdZvi Srl8p1QVEeUCa61cVIgIuMDZt3EVEeUHc4NMZtSpK7WxXE2LXHuvudjM+MOWD7n4w56HVhXutBQj BHZZ2JCxZ4BkMHT61Rm3fo3MIJMZbOVPb7tSrG5/0LTI2isblaIaZb+G76XjPjUxMfrB3iLNjHq1 mJlRz6+mRa6912xsZnAaoWxwD8m2nGLJDA6vWFVQxc45iZ+R6lit444ieudozfVFGBop6ugPmoqE /ciUDaXsRtjEOaPgaiLOJ3t2I0XKRr1aTDYazdW0CGyMvWHjVTfyzS2UDiCQ5kblrBMdjKqSSwGt eqhiZxMfMwsMw2WDy9GaB/ld2eBytHS9keJ31NGRupoWGXvJqcWIwfMi0lpFhBnl19wiJaNODfYf G1wmmFV0gI3ubX+C9FcqFrngWfC9OvqTgxEB26d0jhp2FpmpADiGqXPcZ5ssKKJzDtZ4wPxTVEkF H8lA6CpaZMxNfVYpdtSwONnOKopYNHmyZl4UKRX1abHr91F0Irnm3ufFd6YmwwUB2rWioBARMAtr OUH5lglRolRIJn6BZBeXQImgv0n3OzK0WdF81SyCZ/urhXCf/tz6wgGfegjx0dYBtLX+rMxxedSc IHJlYW29aSr4OTfPhwMujE19sRDesEbpw6dqo6hNyBFCK8OzdCGcEORXJYGFoczzFDPF/c4U2bB6 BoADfzVtbWRwntO9nVStGuJ7trVASdCh8BJNPriEcfQsMLbw8D0JbNyAeN5S9tHmKgcUFMN9tsOE hXzIglZBmWQCnxfKYy9SFtg6krorWPoATlsgdfQ4MHwwFvIWYwKHb+O1jK0illSfzP33Z5ESEZ8W kod6fRQdZGK7TPyjraTR5ikizU6X1BRoKEYErFr5wnIi/UeHG+yA8iJKEoJppjz9IJMR1KfpbYuZ EZxOQOpqWmTsTU5Uihg4d+Q7l/oUTcQZr669RUpGvRosjWU9P5pOLBvfZ3nBrF2qby634wsRbSKK 0foxlWJVQQ3LPWA2hEzyZZzqUKgDjiLNDG5pzjsTLJnB6YRU36uJuvrR3qRGMWY/GZabiK7KqYm9 B/HK9rr2Bsnv1qvFzI16fjQdhP47mj5jg54y1Tdbjhl7lY1isGE3HzbggFls2IiU3vGFh+RBkbLB QZB0HitFFR2pq6lILwnUnr9bKWJwu7RkVBFhfjC3SFOjTi0mGXV8FR0ALqLoMy4428dkC9L035mu QIOY2c0kJv9hZnhFZDIjUnLBjopmQZBy8YeL2cLYSFFFf9C0yNorF5UiBDZt7qhPTYT5wd4iJaNe LSYZ9fxoGuTG954Nd1A6fKAqZh450xbfFnOHMDtdyXeqti8deHBZzRmpddhRpGz8pe+Vs5WiAo32 q2mRtVc2KsWaCTssfC8d9amJU88f7A2S1ma9OlLH86PpINfe694kYxGk2VCcepMxVVNjIUL4zvfb 0klz7UZeAGLbgVpHHQJxzZFadyuODEVjNB81lpGPtkpEIReP+BhuGufqIcRnWwPwKx54/VmZE8Wo QaRBmBOJ67OUYNka8hGmFW9h4V5LMdxncZcBElY5OpQSm8t9kPIF4VMBffxBjIhUeBZCwaMiRIrE t/OzlSICBqOdP4wiYvxgbZEyoU++RCNF4PX7KDrIRvcZF/SW1E8MoX+7hcCGk/2WYHJBuSARFIEJ m9bkjM460ihSIhjLp+IIxqOlrXlQc/5cS/5oRVhFpb+wzXRJTjXs7nsyZsxFykNdWuz6fDSdKK69 dwW3WyvTSnxnN0tri4FggU8G8/whgoF5BlhMPp+BWVYCCpQHRtjpcVeIikLVRw9RB8CxqCkRgfCe W4B2gKcaIny0dYHSUIeOEEEnjNGTZwmQxzu2XrcOGZvza7ZsQD6ecSQqMxUHw33OrbCDExpaTTH2 OCOyDjGKlAX2jKTWXCnGG9E9evLwReJbNJWHShEAZfSO7tSEbx/sLVImPFS/iOnQSI6eg2xsn70W DJzRyCwbh14zS4EGMR8kk/cMvmWCC65TP3NVwRmSKUP7HqRc4DlLiYuZ9pxMeNR0kLVXLirFDgyW YbtCW00+5Wd7i5SLerWYbNTzo2mQG99nbLBxI7UmGw75jzYSxWCDqQjqbNjgir3GHil7Va4ATBXJ gRH+o2yw3pxKpJhscDIBqaPpILBxfrdSxJAPs2WcpyLC/GBuEe70OE4tJhl1fBUd4Fp7/Y60B2Ws 57cUzYx8dUAuFiMCdpHQmmOWabZGHim54Etu/I4sEJlOlOOxNA2D5Z3ghAJSRxNcFFl7zYxihMCF Ee6qZmEtmgjzg71Fmhn1arHr+dF0Yrn2XrPRFsOZMOt95uH5XFsSo5BrCEzls6VGMpwSN0xuhuxw lb4+ORCgaZH1xCuTwwjIjJr0JxxYuLZKRCE2jrC1rAPf6qEPe7Y1QGnQH5uLyhBzozhqDjJxvX89 mCr+5VeoEGZzTeuJYnE/q0LmHQvsDTpSpgQHhPL42aLF74yI4RLP1MaimK8H813pP1gMyI2ei6y9 MlEpQmALTicNq4koP9hbRC7Wq8VMCaNxAmM0DXLj+4wNKomUmBTevYcTDcV4mNwbx3eQsPuVnRbJ iUjJBv07vyO6IGWDloEedzHZYO0AqaPpILBxfrdSxMDu+04bVhNxfrA3iC/k8epIHc+PpoNce69f kG0uOAyfWvML9zS2vmJdOhgDpxwQwiy35yZ0LiVASDLYCsHPeAtE0lqwA8oybCCEqMGf9ByAzWzR I4UrxJ4RLvh2rDN6qBw/GFukTNSlxcyLej2KYKuIebHmXjOxTQUzETxLpLkpuMUFrAZzaYarwpIX TEA1ykhJBVmfHOBd53fNCzIhHW+x1I3PmgicL5k+2CsZxYiBUzMd2lSTdbTar71BJi/q1ZEi9np+ NB1k4/vXW7J58duPPiDM3u7WF7bByWym7z3iRlrkW5w+8QjZkn5RZIDJCnaVpNcdIQLncIdvyOhZ YG2ViMjgPd2UTTJrcaixTeR0hVfArK0FmhR1aDFTID4fPQuMLTx8nxKU1jxGVrgZF7auoGILxmwK 09E0ZNDAmVuGKPSEkdI19svk8RdpSrBfJj3uSBE2kyVIHU2DuJOi9spEpUzmnBhylKciX0qduuYW KRV1ajGpqOOr6ADX2nsuWJ/gASDNKfQOzR3fOgggAkaq1hUZApsAfA3ey1iggp1SefxFSgX7RVJw jhSBc07/UdFB1lypqBQRcLa5I95qIsoP9hYpF4zdQ9gIrd9HEVw0kjH3SV4w1+MGCrngNW9ZsRhc 8FqboFS13A6XvIiUZHzP76AnSMlgZJlut5hNBccT0H40DRLuz+9Wihi4hK4rtNVE6B/sLVIy6tVi P76t50fTQTa+928JM8RsZdY7VwOtN4sYAZ2bKz9wQfmTNMjAFCZYVU7J5d/lgWMk6XBBUjFyoCKV 5uiAhSK1428Wofvm8F2Limqh+/bvsWOz4d+Nv56sxPp6dBzv187rt6LNJIsmfy3ykWUyM7sq+H0x oidwGi+yi0+15d3nRljvX8EvprLy1IOkpXR5qrsRR4r4qBqROpoOsvbKQqWIIJ8MsouoJmL8YG8R mVivFvO51fOj6SDX3ms22oW6A8HxAtKU2y0millus5Jh6G5c6ECDYdUMU9lm2kcfxKAYmnmenJZk pKw8I1VFqRMKrDV/tkJEwAf/nLBZRUT5wdwgcn+culIby9EEF8/R/eut2Mz46VIXXLDgwGxl34ti rNxwbMcpG7TnfjhG3JEyM7jTJn1okMkMBl0pMUaKOLlwJp3oaDrI2isblYLyv/lWImxUE2X1B3uL lI16tZiZUc+PpoNce+8zg9VBHifSXLDa4srNkw4lYINzO00Nzk8nEfiw0IxTWcJLGhQxKlbQnOWC sgoR+rcIjaKkRpE1VzKKOX7olxNHkUOMZ2uLOECnVYxPi8lFYzmKDnLNveZiM4NFHgJAmqtEW1v5 SzFCoIGwuKJkye1whBkp21A2nDVwkckMdoqk0hgpyPj+rOkga69kVIoYmECCMcd6KsLIB3OLNDHq 1GKSUcerCL8L2Bs0un+9JdtiuGcZYV7W3kCKgkAs6/JNMttwSlkODduP8kk9B5gw8dVfLdCkYCtJ OpgRImxu+ZXR0bPA2Ap9I8P4mqLGgoJBXtQwJfBsa4GyoEP/+zMYSTI+Hz0LbFyfkYAxuEfae+Kz VErXGwz3ma63nKCUzZVwRK1UaGCfRR5+kfLADpF0NcHyLnCQAqmjCSKKjL3pfosRwq+vszRbTUQZ r6S99gZJd7peLSYX9fxoOsjG954Nxq3ekyYbXoRlu7kQIfA9BOsJilkXfOACmQzLGJ/18fujMsFG kpQYyqRvYJocmaMFIoqMKX9VKNuquCirQ93qIcLogYbaGqAZEXeOyDoMB2rBwUWOqdctRN8K57Ys 8Wl1mV5YDsSy9ZUyyLeAui13uDlzrZQP6Gd+t4iuOSTrZsSRIkTq00dNg1x7JUKp7BPhksbWUtXk 8u+jvf94EYpob71azNAbDfXGaFpk4zsJ8XJfBTPi3qEhG5yLa421GNti6AV8GajjciccsSuVnPiV 3w2SdGVkxs58EydSeeCcQkidOZpgo8jYm4aiGDEwruoMajURZ7yS/dpbpGzUq8GSCI3matpYNr7D xqtdzbCZr5TLBh/CyomYg8EG53VCBtNKaQWoomaQyv6wFJxFmhpkRt6iYOkFOFKRgrOKaA8LjLV5 sYoRAQvyHYBWEVE+mUtmBCkXdepKbSxX0yLX3rwn/8gMjhen3uSyZMafbSuKsUUsU1RmBveTpyHg 4nGvpOFJ/cnvWOjK70oGq2jpVIrZF9AUov1oOgiVYDTxu4MxwmEfeYe81cR6Q7Rfe4uUjXo1WDKj nl9N7Fh4iu+zzPj12/ED3jGAyNfZyIxixMCGoqYGswdJhK9KkSyOzTruKFI2qDRTawRLalBpov1o IvIiY29yoxg6mQ32vXSkpybijFfX3iB5L9ERrxaz1ajnV9MiG99nbDBlzfNEmuzu3A072oIRAxvT bNPIM5YcrC14TzNSJRcjs0jZaPpUhMC54wnVR81BxtgW7ZEiAM/dQgRKiO+DnUWaE3rzKFWPHb9W yUawUX3GAXUi3PMGsPO2czY4DeakKGsbnnOEAzZOpRn4Eil7Vxb98/SLlAP2iKTYGCmC/vmsaZBr j98djHfub74cCRHVBCFP9mwtipSLerWYGdForqZFNr7DxsvWwi2UJYO3qpM2A8HFD6/7kws+AJBW gFMqM3b/7rdR3QkgUCqYeEuxMUIE6XBOTqtnAZue+dVCtj1+kq4jPdUQ4aOtC5SGOnSFxufRg88B LLjX1rvSguKHWYI8Ijb4tLRYDPe/ev+mNHCJYBMgUqbEj/wOHoKUB7aSpJcp5sPmWAXaj6aD4N3+ 7mCEQGnW0qKaiPKDvUF8o45XV2o9v5oWufZes9GhGMMtPhkpG3wShuduB7IYrxpfnbIJc8acnGBP TIQkg0+7pJIMEuccnbmIeKQIkyE2UkfRQdYcQR2MmVOUtXeqJueIdOraW0QyxqmFrt+j6ERiZjS6 83q8LCyYV/nlyAFpBsqdsFiMEDikYqFNHZt74QgzUjrHNQb8bhGdc4DmWuJihslBBKSOpoOsvZJR KWL47g3qVxNxfrA3SMhfr67Uen41LXLtvc8M1l55nJSVtOItLZiQDUZHRDPtg6GQzd1wbBqJlGz8 jswgkxoMAmjWr1ROIqD9alpk7cniYsTwhd3DsO8ITE10qh/sLdLUqFeLmRv1/Gpa5Np7zUYHJG6h dBCBND1UO9NidiTM4jO/LhuMvZMJTDGfYZq/I/IgzQ0qcLqbxYyTLZ/ZELqailx7ZUMpZyiYHu9o p4oI84O5QSY16tSV2mCOogU2un+9J9tmsLuGZ4B0ljbaZhSDi6/shwoXfAcmzQGfUdnhWn5H3CKT GXy3LE3LSuUUBdqvpkXWXrngrAVSDjlYeUlmVBNhfrC3SDOjXi1mZtTzq2mRa+99ZuB5ik4/1ri9 qpiZ8ZN7JexWE03ygG3fZ7jm77AWpJlhNBczM/DwSVMR2a+9sqFU2GCR1vey0fiesDDwYM8So0jZ 4Gnh1WKy0Wj4/Wha5NobNv5RYvxlD1q88/BaU6MQZDANbJNGE8gR0ISpkP0aVyNmIpTDw7k/kgB+ W2zBTkVo2KL4alkEx/ZHCzENwKexLTFGDQE+WkrE+6P15sjQZsfUaMG/IvYjY+pdQlgoWe8jzD6N FhiL4T39ggWGlZKjCQKMlCRwZrIPP0gTgk0jpWGkLIOUupoWWXtNiEoRAqekbJqsltRElB/sDTJt Rb26Uuv51bTItfeajTYWbqV0DoULAxl9mQ9FGOdyH0+LC/afpwlgM/6O0NKI9u/ywKYS1l7ZVDIy 1NDpTFdL/1475UAJ/Kbo3VEeJ5JggLWA7yYHYwgk5u8UttTT8WWx9fZqeY4nDeV//9///v+XBYWz UCYxXnHZfAuKxYifwzq24lSvuWGA2JSKX6wA5qkHiWeOyrojsViq50hdTcwrPNrLE12MOTvuOO3M YTURY7ySh9pbRPbWq8Hwaj2/mhbZ+P7VbbQLZarSFXHZYOWm52EWIwZWgWy4KF/duAwZCgE4OPNn i+icw7NuSIwUzjHxGKmraJExN+OWShECy0lOVawmwoxT194g+d04tZBcNJaraJFj7t19Ffyet5Lc cxbX+xP7XhSziHBWSi64GSB5wPzjd3YGQQYDFH5nh+3vSgYDi3S0xXzXmfJC6mpaZOxN11spYuAa Wt9EV/rUZPGh9mtvkWZGvRosmVHPr6aP8X2WGUwpYhfvuEUkN6bBUDFi+MlMZdjgMScRGCx6Cw1s cEN7Iw9SNhjApaONFN5RMkbqalpk7IXFxdg5k2+lQEYVEeaTOXvQIiWjTi1matTxo2iBa+1dB8rO IW8RlAtOuLacWIwIUgGaGbTrzQOliNLRUAcclFb8rlyQR+lni8lF7gM0M0bTImNvMqNS/BuXmu/4 S02EGe3X3iIlo14NlsxoNFfTxrLxfZYZXFGCXQYVHBHMuhA7DQMx8mAewTSgZWAexpeCY8Ytw9kh lbgDlApuuUt3OzLkKjKWaKumyNqapFDIToRPhcyiZPUwMnm2VSD0rT9HZl2+ahZpXNh634VyTBLf EeYrYFtQFMN9OLclo5Cgzw8LkfIRfcvv4EUkvjk6c+VuMH5H1xapq2mRtSd/i/GauFWBR7uaiPKD vUWaEvVqMd+Pen41LXLtvWaj3QjbJbkCMmx8Y2k8TWexPEzWsW3KqFrZ8544IyUbTAE28iDNCqbk 0ttwnUfGs9S9SD1qKnLtlQ2lzItvnCWxlyKdo4k4n+zZWhQpG/VqMQ7nTjRX0yKwMX6+ay7cQ+mI gTkI7pVsgbUYcwLshwoZHAhM8rOrGiHJoBRnUo8eJojOUc95R+9CPvPcZsQrMnoWWGOlQiHnZ2E+ y7TV85v5jidj/ztImYhLB5OJRjKKyLBFrrn3eWEnJhGMoVprsc9SyLkR78ky9766+8gYIyQRzk4e oDywwvqAScTfBz1QFcAl0toqD5HJwvyMdqPGCJ9sHcBf1aH/DmbQ8fnqGeDaek8CezXYcYc0LVbr CndZWvlDA4WwT5+6zYvvjTpS0sB0U2MOUh6Y4knNvVLUoB80LbL2/N1ihMBI05fR4R2ajPLZ3kHK Rbw6mFzU86upyI3vXw3nNhVs1kjFyV4zCiKrLG8iEaO/5MyODRmVWxoKdn4hIxfcfJU6o0i5oLtN f7tSLGw/6kncIvpWa+WiUkTgzjapiCKCZMvao7mDlApcWiRvQp0+ajaKa+t9VpCneI805yN7b4e7 J8Xgga9T2IS5naijDj6C7KlSmCB1G3WQMsGmknS2kcI5SugPmhYZe1OLVIpZYlLNcY+jLzQZZrw6 9hbJ78arg/kE6/nVtMjG91lWsI0k1SYbqgm4WVEMNrg53jaTCpbFzOzNipRdCucmG3mQssH+hHS6 3C+ILtngIAVSV1OR5MX53UoRA8W3FZZjMDQZ57O9gzQx4tXBZKOeX02LwMbYe58bbM9g6pFy9y91 c8kQcgKDbzMznSkZuTSYBuOrQnLBmYfEHaBUUDfnRRJKXnAc4UmNNkTGlr9aCPfJC4s3tk2phxA5 QXdt2XoEKA3xZyFZqMtXzSLH1msSOknBpkkuqYtj1FCtKRaDBW7L93WgdKVLs9lk0mwGq3zlvTGL pMJyiNaNiStF4xipq2kRnIumMlEpxq3UYxI4mozy2d5BykW8Ohgt7URzNS1y7b1mo82mWyu9jxlp vnjbmqKYM2Aci29KsFm5GRAhcwKXU0BwszA/0znE3dgoZSNlWaNUFSVNCqy1clGMCLiAsQPRKDLK Z3OLhHsqSZ1aDAc2lqtpkWvvNRebGVZ4csH1zJTXvh/stgzG86FT8OWlZqNWM0wuzEJKMviYYQIP Eu9oErgl+FGKqvGDpkXWXtmoFFM5v9EKZ6OJOH892zuIvxuvDnY9v5o2lmvvNRubGeypS7XJBFov VYQNMTPDo4v2cV4W3fUNygXvo4EN9vU18iB6RzniRxCgbISYOYnQKEpqFMG587OVIgTSyxa3iozy 2dpB3MQ6Ph1MLhrLVbTINfeai80MDrUmMbiJsTsL2EwpRAvGTJttGRUs3yNOkBGSid+KELXApAU7 RlJ7BeNnVNDPegbAsaiRvoHwnm9p2p062vv112by15OtAzQl4tDBpCE+Xz0DjC08fE3CJgQLT/hu j++9n309irkPtuMPSlhOCjmoojs5wzN/t0gTgk0lcLOYMXKuAqmrqYi9Ve2ViUo5A8y+5FARTUTJ SctHewcpF/HqYHJRz6+mRa6992ywYyMV4DdXfUqGUEL4ZgVpTnCMLVF2ZYi1hUgQtkCZYD4jtZkQ 2UsB/UHLIji2v1oI7/lWw47y0GOEj7bMkQBlQXcWkYM6fLUscky9pmDfCnZrpNRkqyg7aMpBMWdy v3mmDw6YeS4JkbJvzb5WH77IvBjsKUkpFYzfUTJG+9W0CM5FUxOiUoTAGqElPqM8NRnls72DlIp4 dTDJqOdX0yLX3ms29vUgYCJwJ/CfXjmKTjHbS7abM86UDZbQk/h/IiUbzEYn8iJNitwCCUMr5RbB Z01Frr2yoZRTd5yponOUjWgiTl59L7sZe+ZFkbIRrw4mG43malpk4zuNxcu1QbdSeqcb0nQLrSsW gw2Oq4QMPuyT3Od+vxmtcoooi+NFSgZb6uhwIWOkUBDlR9ECa61cKJRtVtxCM4M9FRnlk7mLlIs4 9SBVvx81bSTX3rvMYGslly6HC1YJqSvI0sV4iiz1tK74y7dvE2akfFI/8jsCD1Iy/jIiOlJpLziT gParySb00V7ZKMbzZCVgB3xoMk61o2nsLZLmieva9Opg1/OraWOBjWg6mfFyocyNlI4dmDpgGqst 52LEQGFsc0YVy+c57Ebo6HaQlt8tYlR0rby0srFSFBSRupqK2I3UXtmoFGvIfBPd93I0ESeXqD3a O4i/G68OJhuN5mpa5Np7nxscVWLN34kUdnqkDcUHsDxRmlBrC+pY/8M4I+Wz4gqP5EGRsuHtXuFs pCi8P2haZO2VjUq5oj4Fba4JM0wVJTVibpGkxjh1MMmo40fRAsfa24UyW2wHEEhDRsusxWj7SURb NC9Gc1ABF0rFOd6PZkF+Vy6++6WpEeJnVI4RuooWWXOWjosRAftdumKrIoOMbpmPtYM0L+LTwaTC WGwzVtFGcsy9Lbao7PzaA1Otf/lMFMUWzi1GCCxoWmtTv/GVLcP8FSm9oxxN4EXKBfmdHjhYwmT7 L1JXExXlkz1/txi7UFiR8a102Icm4vymvXAfe4tMYsSrg13Pr6aNZeL7pPTMSDBksNG/hcZARMAd RbZnlG/My4QLhfCNMZHLH5CTX5UKdpak+y1mkPkI7dUzAI6dXwXyWXoYwfeREZ9qjDBCa+sATYo4 dDBpYEBr7xw9eaIBrq33TQX7NXiMSDO27MQFM0hgFhh8vYXaThr8ToVBR0jPuMu9Tz9IaWDkmo53 pVhLj9QoSo4UWXPNiGJEQLfesW80GeSzvUVS3dWphfCyfoeKUbTINfeai5aeTrQ7fmCV4rczMJae ixECq7KhguOXiZwzgjNa5dL7sBMkvjFKy4fYWCYZKf4rUkfRAmutVFSI3TNcoG/hPoqIkjfp0dxB mhVx6mDX76tpI7n2XnORwvM7C0zcyRkuaH6zGjJY8oJTK7ZjFLPc6GJiMK0wA9bv+d0iTQw+kE4x sZjvBycTOD8FGaOpyLVXNpTyeX7/QaGBPRfl0GScT/YsPIuUjXh1MG7mn2iOpoPARjSd8uJV4Yk0 e0ooo5DmAGkmcQ7mqIlhSdjgzom8AH4hweKHoVp+R+T5nd6x+yn3TS8mGwz30X41LTL2ZjBTKWJg isrGaTQZZ7QfewcpG/Fqsf9ko9EcTQfZ+A4br4otpPkwNc+TrXFc9JXdFgcjBnaJ2axTzvq9EJ46 7coMXLkaKHlQpGxw60U63mBpIjijkNJzNTmke7A3I9xi7KjpVRGO/1BEmNyJ9GDO1CiiuXHqYJLR YFbRASY6rM178o/MYDGPIgrpfPKANgMNxeCCU6Y2aDRk7oCWC6WSGXzsPcVXkMTEoC3X/4wUv6OA jNTVtMjYm8amUoRAWe17OZoMsx9icIiopoOUjHi1WDKjnh9NJ5aN77Dxj8zwzs+w4aXCbrdAQzFm XhitWmBQ9rCvIGwoBcJce35H5EGaGdydmK43WNpCpnLRfjU5K/xgLywuRgzss+/gNJqMM14de4vk d+PVwa7nR9OJBTai6TM2uEuSrpkNAN7cUTIKEQIbtW3SGCkxLWF9zQeNOn7lO3WpMAKUCuZ1roTH K5S4ShappUmKCvGycwmILyQjMcZcjLif7CzQhNCXK0O8umuRMkoGOTF9RgA3fvIIcYvKhhIrKVYM 51mnNmBq2J8eM4eBSNmtMOedR1+kHLDLJP1usKQD4wikriZX5R/t+bvFCIFv9lhnsTCnJgPXHprG 3iKTDvHqYKZDPT+aDjLxwfv7LpWzjJKBecZdTYdATsWyWzpccKNeIucN8RIXfMtXPp2c9Velgs0j 6XWLGSTfxpGJ1TPA2ioRyPgoGQvTq8pD1BAhXfeOCFFzAH9FSa1DB5OG+Dx68kSf4vqMBLZtpN7k MEyXC9EoZjXATH7udqGmdXcmQUdIFjgynKdfpDSwceRJyk2NSo2i5EgRiIimElGMCJglbUcaTQb5 bG+RvFF1aqG0bI3lKJpI8ozH3OuESL35ncV4PlQeabcSJCMWIwT+136UycuGHinJ+JHfQY9InHOI 5tLiYDw5KsdIXU2LQEY0lYxKEcNXZ/9Ni2gyzmd7B2lexKuD8WHUieZoOsi195qNVpxs22GCWzYo 9nj6vh/FnMhg3t9jbjDM/Gtyny82ICUbbO9MGhTROwdpri0y8h6pHEKg4jya4IeNVw/2ykYw+oC/ HEO8igjz97O5RSR/nVrsOG53W0UTyY3uX2/JZgaNHE8TLgiwXQd7/4IRAf9BVYFZmtBGHim5oKFN FnwVSWY4VHNpES5GKocRkDqa4IKm8MFeuQhmCBw58DaA0WSYz/YOUjLi1cF8ivH8alrkxPcvNjYz mEHiacLGl14VhgKhPM4v3DUfMrhItqErJBeclm4OCBgSq2nuXpOKyuQkAjIwoZpkP7tZri1/NUK6 zzY2X8nRY4hPthYoDfrz30KykCiumkU2rs9I+MkbkNv7eE8zXYECMd1nRp+WxJRgW3qCZvi2Q7TI sOIbpDxwa172KK4U1fOzJqIuwsM7vxuMEBiGneEemoiSodqjvYOUi3h1MMmI51fTIDe+z9j49cdh g6nFHrFpLIq5Av7Ts36wwdftk/oclztDtA43ipQN3o/0uStF+RztRxNsFFl7/m4wY2CBxDLf4R6a jPPZ3iLTWsSrxdJaxPOrCTae4/sXG9tc/OErzmGDxM7EBRqKEYNntiWDq9qT/KT6jND4okoKryDT WnB5q2/IClE1Rmj1EHYBqIieUhHMACjynbdwgS41Fp8rejRm1Puzdelg5kW8XkXhRiR5Mb9734n8 9UwMecEYkFavnUgxdwf/9aQfdinfU0BzgzlSljkMFlJBFWlesIck1cRKeShBqaMJMoqsvZIRjBjY xzWrldFEnKzgPtpbZPIiXi2W2OP51QQ/z/G9zwtnyMMF56RaXhTBf15IhkYwwXUjZSIyMjGba/J3 eWCKhEVpZvlHhsoRvUcLLHQQ9u38Joh+//R+TtNBHb4ZfOnPz3HUzv5t67y+HIy9Q/p/tfRvn+3Y IRP+dXQKWfaQ8PSQ/kpOJhMWw3MOrLSc4BPXfe5KxS8W9vrU87sywEw1HS0MBPPd51gCUkfTQcZe XqaRMgIPnkhDFBniKHL418FIkRIRp1aKhB3Hj6IFTnT/yoN2oKw1c62hXLDiw55E34rF2EnGDU6W E2x96yvwR6G8FHzxh5+52CUiFQ6VXFAcLD0mxymQOoqgosiYmzFLMCJgzYdBxtFElD/ilCkRe4sk KerUgY7fV9GJ5Jh7tx6ENJs2UvYxizbVxGKEwNd2maKAC/beJQv+KJW84M7t5ECQIYNPGdLbQkZ0 pXiO1NEEGdmmK/fRJImDGQPbcjt1GE3GOZoceGHvICbGeLVYEiPRXE3/T3yfZQabN1L4feW0a/Oi CBH0M3BY5ZP2TYPcWWy75m+Imr+bFFzmQT9bxDTnhAU7FI6OQZKB85vK+Hazs2JWZ6PF6B7t9O+0 kOPJShBrfL06FpHt2HndV7TXjHcp9fhAI/1goydqMLa++vmVcM4iSJ47txLM4JSlkjz1IHnT4zNt IqvkI5WzB4+a4IDVjAd7MjcYEXAdbza9jiZiZAng0d5Bmgnx6mC+F/H8aloELhrfZ5nAZYM8Q6T5 IntrCFo3MJ9j9iKagFyGmkdPbjo6pdFmcJpHX6TpwGmxdCojRZjM9SBVRbxNC6y1chEhI/jJTXpp IqLIKJ/NLTJZEacWw6n6bW6NpkVOdP/iYjODQ/fpOb/khrJmRjFC4JiKBRU1Wq6F45lHSjL4hHmz QGQyg1owPcxIEXt2V0lGNR0ENqKpbETKGLiUzZ4pVWWqS/quGQzHnpHv79arg8lGPL+aFpGN+d3r 92R7D2525HEyTcN3vbr4wYAJzMxg2t4DftjlE/ZJBA7Oz/CUG6WTBkWaGgyjfU1GiNA5MoDQKEpq FFlzJSMYIfDCz8KsioiS6f9Hawdh4X19OphcJJaraJET3fvMYDKFr+WGi1wHYmYs5n5g9jaHi5y/ NcxImRl8BDhZEGQyg/0i6WNGCjL4IjJSR9NB3H3spSglI1LGwCUF9k3O4KHJOJ/tHYTfwUa8Otjx /Go6sVx77zKDmVfPOSHMZ4xbYg1EBB7qyzNgr0GzIEJy4X30khOgacFmkeZFhQjcb7TJhHrSUgI4 fVlbJUIZvadOb0kRNUb4ZOsApUGHHLKuUH2+emDhKa5/pURfD1ftPDHlvOg3xjxNCTHdZw7fM22k GbcFJwHYIerVo9DAJq3wUMSIXMn7lnXRkYIHziU8ajrI2isTkSIEvh+Sge5oIkpm7h/tHUQuxquD mRLx/Goa5Mb3GRvMqfMk8c7rYEtGIchgUxgtGWaTELzTHZjyQdY8/ADlgY0j6WcqY9kZmdGRvkEk bqnGX42QvvPxpM6dRo/xPdlaoBx8881RANfYsqGhq2ORjeiz8NmeQbVA+GxC/ZqKil1jwfCdhUhr Cv4Lc74CXB/otaMkw5/ILFIS2ESSOiNSec85V4H2o4mQi4y9rUqVMgQa4I5uo8ko49Wxt4i/W68O Jhnx/Gpa5MT3GRtsG+ExMt7mTFIrLF77YOQAs4i+CLzjfo4PNriteUalbF/M4y9SNthvmf5lpIid UxRIHU2DmBO15+9Gihj4wsQsxkYTcbIS8mjvIGUjXh1MNuL51bTIg733rSWj4ewd/c0m4lZY7qrs gSbWNew6KApyIRxkRMhmm88zJw2KlAxWNlk0PFKEyWGElJtVdAC4iKJyESEjYF+P89tUFygy/1kL 8b6bNbfIZEacOphcxO+raRG5iKbPMoMZS54m0uxraU3B6CEYz9MHEzIYhGWnMx8qmDEqs//NgiAl g33JaVBGitjZhYzU0TRIMuP8bqSMIcNj2Ygm43y2d5BmRrw6mGzE86tpkRPfZ2z8ZXuw3uE248+2 mmLGwBIG/2FqZPs1D/17pHxW3KWQerNI2WBqN6dpRorYuUPqUdNB1l5zI1LEwHVamUBnnKem30xa PNs7SNmIVweTjXh+NQ1y43vPBnOP7FgxN1h8Zn1DNhZzIoTzCulD+LJgMoH7ibyWBja+53fwE6Rs 8EUkG9oRInQOZaJ8FKWBEIlz52cjZQjdsKAWQ3w2dZASEYcWw8VxerTsnyeuz1jg43OpNNk2xmd7 ykIxnGdS1TkbF/c64uBzAzNe5fNwKTSK6BqDNL+iIQ3RlYoxUkcTxDCav/amJwlmAGzNbD8VTQZZ TbAee4ukvRivDiYV8fxB0yCyEU2fsfGdnhPhn1RRrbHYLiBECcOlupZYVG25CY4SRqG0Zb8UGSCu OURz/XAxqeDDJw96FlhbplFkspMoF3vJaNQQIaN2HDL7tHUAaR+HDiYN+gwLHLnyQ5sDnLg+I4HV QB4jnjEmbHHFjhkwB2LkAb2+VnlBkgDcYjcD1fS0xhykKcH2kLwZI0Xjx+GKR02DXHsyUamEwDx+ p1CjySif7NlwFCkX8epghp5oJGM0DXLiO2y83HnF/jn22cc7HnuLi8Vgg893+lhI+VZavAAzTmXp KtVGkZLBvpHUGyNFmP+XtLfZsWxJsvPm+RQ5EXgLUIMZ+ReZUzYkQByJ6AtwQHAgldigqCpK7CpJ r6/vW8vM9zmnTkYI3VW4yIgVts3NbJu7mbub++Z0Rb2ijA6C8cOpxpAqfsESWbdkw0k179qLY+xz I9QN0citLWQUx7jX7tjiad2V1ZNWHjK14jxOU4tibpywfeGLIlnIBW64ZYj02Z95bJHagjKRpBtD heacRoB5GZGGLrCt1RQhIihSE8FE1S4SRmjJNshMi9PcIlm8GaEW47nR5YbTo3a/skUXb6id9NCg tvhGAjzhoxgqsNzuQEZqy31wjgNUDTrBJLxyI1icIEik42VbM3RRoTvlolAdTgfZ9mqNUKkDX2Bw yjOc1PO+vYNo/JHqYFojkt9wetTvV9bo7NTaSScRSPfZHxpGxOIZHNtx8YbUlsL06BkqrZEf1Dw/ 1DVYE026UUxP4EDCLadBrvZqDalMMsksZps6nNTzrj3HjCK1RqQ6mNaINjecHvU71ng6ZlA9mY8w sxzDiklTi8VYHuE4jIkW6dynTj/wiJmwUmgRPwgykYS6uqSdxbQGxyugOpwOMu1l2Weo0IE18p37 wQg12RpBKI2f5g5SY0SoxXCpEfxitJqsdscWT8eMnKjnbUL9jQiXBGMxNOAqjyYYLGllOODTpsQu heO/eEGRegb73ok5wTJGsPAB1eGELYpMe0kwBlMFLrqsMcJJNSPVaW+RSTAi1cE0BvcDZCKznFaX 1e89a3D2PEknV3oTZO0nVFUGQwdurHJMI6Mjq06lWqiU7ltoSHmCaA1nbBlICukYVBPB/DA6CML5 2DhGqDTGF74dcRip5X1rB7G1yhT3Casj9w2j1WSaw4ZvTVQtmSRxwDOoIequ0EKs8rKBrY9SQ+O0 hMSDBdlOT77nqQFqCGpIsnhVGtRmPGRyctgMsm2NU0jkcPH6M1/JxKDhg9rskty0NUCG6pHnohmR b9isEtXrl0ZoGKGC0vt6oOUKvOZaA7nSzYptbMCR00wmuFYZIt3h1aewgUAEc5LmVuJgEJGOS3Tx GWDbao+SRuFZETY8lY0DJHsaN20doL4QgQ6mL6jGDZ97vZDwbU+gZiQZJ4v8nDJqryiG+Ow+mGOR 1TJfNwclxfIuGszAZd7VOUidgRqRpFRLRRIdqotTEediba+WCJUqMETYC60Dg5Na3rd3kNoiUh1M W0TyG06D3LT3tjUo0MiS4wvLrc0sqKnMVQsW7Lh9rjW4B8fXzXTQ+2gIJ5Sd3yA1BjUiKUdcKrrm A6NFMEo41RihQgUmrTPrDSfUZGvktr2D1BgKdSBtEblvGK0m09y7nkHhSPoHRfYEiHpGMWxBCad+ QObm6KYxQqUxuEMsPaRIjcFdYwyYF5XZilQXpyJ6RturMUKlDqwUdLM2nNTzvr2D1BiR6mBaI5Lf cBrkpr23PYOKDRaqkI67abqAQ00lWHRg599Ri8zN2wvUM1Rag+tSq3mQWoND9iltXCpyxwdOi2x7 tUao1IGY0ilpOKnnfXsHqTUi1cG0RiS/4VTk0u/tUcOCSacQDGs/P5FW6huLsVzA9MIFHG9Jc1rB kBkqrcEqDc8tUmvwETlyqsUcPXnZUF2cFtn2ao1QoQPlG42lYYSabJHMRDnNHcQdwxHqYEfwi9Fq crX23DMaRiiZZLcutuBL4K3kXQwNOH1slkUW+81ZBVqGSltQPVq9RSaS8CnThNylYvk0VBenRZAu nGqLUKkCIdplB6dycFLN+/YOUs+IVAfTGNHmhtPqcrX33BqdjVgy6QwCamq4xjGETAl4q971g/ux TVI3kEhb4DPVW0DRmKwxE4jBSsMGSGiGDTouMm3VECFSfDZEmkiEjyretbWAT1UeI+/QjBY3bB70 eq97cGUmG5AkQ1w21dTCy5+dTBDaqEB3KCN99UQlLkHR4kxZuUQgOhepHbg0M6F3qdggfeA0yGmv lgiVKlBk0ulvOKEl+yS37R2ktohUB9MlIvkNpyKXfu9ZgzVVsj3LzBkQxieCaQ0u23Ulj9Qtn0XH GqHSKX74nPYJUmt4EnoR3RwSaIYPVi0S2c5TQ4UGfEALx9O74KOO920dpJZQIr1iqawFVZOLzyKr 23uW4AaTiMZ11V3DoYzSqxKRn34Xp2CLoyqGRjP8lAKdA9QKDChJvoKlu3Mn4Q0bckl+1+TnoYGQ nbtSbMppHlzU77apKHyeIplUnhuiSqwNhs8AV1vPx4cdLSkiyZY+t+OeASIYRiCfy4SZrM3vLqE0 d2NlhGAO5HOLZLR0ZtZEbKks3pbq4jQIlmh7mm8xVGAW1u3qcEJLVvBu2ztIHSJSHcyuUW0uTotc 7T23xo6WRIcsFzOBYP7ZOCrmcEmNO6OBXYNsK6qHCLuTIllkgwGC1CeIE3dU+FSohlG8JIjGOM8N Fe+TVfPZrQ0n1bxvb5EYv0IthJTWhLZgZxktcjX33BbrGVRtZFObVRc0qS2CaQuqBRzEyGFzGy+q h0pjsHNRxUXGM5jEptMsFVEjVBenQTRGONUzQqUx2EfFYoRrGanmfXMHqWNEqINpjCpTRrGOgH1/ W3tui/qFpZNESqipcmAiqi0GwxZ8McKLd7AFC5VxA+7MnOkqF9tH7yJ1DHbREmyXypmMVBenQU57 tUWotAXDYKe+4YSa7E3dtrfIOEakWuyS/IbTo36/GjbrGQxQXFcXa5ApduVmMHWgdqIjJ8E2fvA1 VHoGoa+ai0Q6p2qtU1wqlmxCdXEaRGuEU60RKnT4zJKlGdZwUs/79g5S14hUB/M9VpuL0yJXe2/7 xhfS61jDMyl1DSGHDAptPfmJ/1HFWUfIwRVsQS1Wkg1S8m/JsJyoubNodYaHW8wZQzNsMmAE0RL7 1BBpCA5mm2ENH1W8a2sBzVB5DC1DQ6oVLS42i5y23jYC6R+xmaVWnLsZllWXPQ3BsVP6ha2yPRYH +BIqXeI1NMT/IMrmFK2VjktF7vnAaZDTns8thgreT3lxeqX337d3kNoiUh1Ml4jk2nQ4Fbn0+1UH 2eGCpVyCNNZgJapZt2WXYPiE5xAsfCJro0TR980WpffRYA0WgZNsFKk1WCxO6F0qN9LvOQ1y2qs1 QoUOfHjjzL7gpJ737S2SDjlSLYZUI/kNp0VWv19ZY4eLHLVGOr/Htj0kGNZg3Z98RmvgN/F+zh3v FM0j2tgnyAwXud5mIB7jpxBdjAY5zdUYoVIFNm25DQInkxFa/rhv7SD1jMh0MG1RXYZRrCOiZ1S7 X9liPYNTrToGRUAM+R0shLQEJ17sv+SwTBpdwmG3AyIDyydJFqhbUDyS/GuJcKN7PgW2rZhvIKpt WEnsZm3YqOFdWweoGRQoxpII67FnpBoXnwFWr/eMwBJcck7u9WFNrlYIphnYvrMzkMYy34jWUsUM LMtV5yC1A8tyibl8MQZeSaNDdXFiQL1tb6azxVDBqh24uyNnHSKmGE5OCJVzkDw3Uh1MW1Sbi9Mi WCOc3rYGz30iePOeOKbFsrXWGIxZO+W1WFprkPfoE5w5h0if4F7+KF6kxmBGnpC7VMxSQnUxKqK/ tjmfGyrfp7tJIM714ISaNHvb3kHqGAp1oCP3DaNH7d6zhYNjhLPCpKbIuS1Nwb5PYynfto8b8GmE naNRW6dxBGoJxrjEW6F4BQcroLnYkEqKaIh9aiFqbbh9sLE0fFTxtq3ovE8xnirPRWPRp5wvNouc tp7H0o6XhmDvDYGYRdrG0sEUn2VaY6mRu3OOl1DpEbkhVjOIpMM7TXMzcbC863ze8IYTi6N37dUj iqECQbiLh+GklvftHcTnRqqD6RLV5uK0yOr3nkswKY41uFeHUag+IebiKEd4KEi3XY5VxAW8Amkn aZ1yFFE652mtUlwqNlAeOBXR+m2v1pAqL5TDCHbH4YSeXL17056OUaTWiFQH0xrVBmsMp0Wu9sY3 nm6nM7FVAKX77rcIa41gWsMJn8ZgBy3OTxteRINrMFvLCneRGoOvpyXgLhUZtMwvRgOc1nysmJkj J/NnizaM1PKuOXtEkdoiQh1MWyj3LadH7d7zDGrOk3jzcQpuq/KuG1bXxFSBMzwO66Sxr510cAAV Ko2BzyTvKlJjcFg2mVcwOJFFP3AaRGuEU60RKq3BWRyN79YcnNRTKjhNe4tkfBqpDnYkv+E0umx7 9Ofno0YzC/enGLKgprSbqUY9IxjW8LynjkAeS6rpzIRP751pmrOORWoNPvZN6F1Ma3BZPlQXpyIO 1W2v1giVOhDD7ZdO+eCEnpxYvW3vID43Uh1Ma1Sbi9MiV3tvW4M71rPiy524zMBqDbHowHTEFINs zhtf1DNUhlmG8mTpRWoN6iVjsiEid3xgNIjGCE2NESqM0TucsYXXEaqkNDjGNLbIOEZEOpimiNgX nwVOY2+evrUEyKwfas7FdOliMN8mBf32CYR86ZyCybS3sSIcX9SuD+S5WoKCkiznhYrnPK8g1cVp kG1vsvZQqcLXfEATW4STaob7ae8g9YtItRhSrTYXp0VWv7d7CeV+5L0I98JqfG9LK4QG36mUMp3A 56gG0SvI/pgkKFu+obRATUHpCOnXYIjmYQVsc/EpsG1NWiKNdmA9ySjljA82aPj1ti3DSIGaIQId jENvVePic68XbT3vHE0wLLUcK+Sr1HaOwRSfMj3TLEKUwQAF++lqrMCmat5+kMkv2Pog7l5UOadQ OwyjQTREOKnSYvRA7oxpfhFOKnnf3kFqCoU60JEbk4bR0cSBaZt7bosOm3ainj568cOJGSgG0xac 7nMzhCzWk+uqGSpdgjGk7z+IwjlNa6XiUuWYAlQXp0GU7jy3GDpwiVRDVDip5317i8T4I9Vi0b3a XJwe9XvPM1jxT07MPSVMRusZwdAhtRwag5q7qh4ijcGHGusFIhGOyRN7HTHZUOWUQlLwYTSAtghJ HYNAldyZ20wy1xxGannf3EF8boQ6mJ5RXbDFcFrkau9tz2AzI4knFw8yI60txEw8ybRylI5k1nM9 eAYXp868lW8eZTW3SD2Dg2ikHBcVifkDpyKOTW3P5xZjwYXvCXUOHE7oyZVyN+05YhSpNSLVwbRG tbk4LXK1N9Z4mnhaQ8lEQum8GnasEUxrcJzH4cx9F3JC9ZQqYYQrcOIGQTIKsodFFdGhip4cVkjm uZxyWOFqbwabUNHX2a1Lpjuc1DNS6Ypp7yC1RqRaDKlWm4vTIqvfr/rJjhocK+zdnl7ilhUM10md XFA3Q9GtyRYJnOfftIZUIE6PpFmkvsG6T1KOUCFdzizccRpE64dTfYMFQWRAB28IcPgJI9W8ay6u cR4boW6oRvCL0QJXa297BjW8qQP96UUj4xnBtAXjpiOaNTJs5KE5H+byPlZs8eJzi4xnfPek/8G0 Bd8lSuK5nAZhjSHtjWeEChVYIpgdy3BCTUbEmTCnvYPUMyLVYvGManNxYofoTr/3PINNddJ/pPNI 7HhGMK3BZoWviso6Nn+ip1TxDGbz8QJqu0DqGbmMGZOFlcZg2EkGu4wG2eby2GKowDXfnQvLSC3v WtMxitQWkelg9pLqcjFaZLSjuTc84yt1IBQQxha4vY6xEKbgymLGTU3BpwSruET6BcugqVj9LlBD UEWS8BtIQ3C/a9LOZTOIhtinFkJ8qhjdPWQmJh9VvG3rADVD5Dk0K7K+JZujhOfZp603wgiP+6EY aH9YPzhGyLdjsAGRSZVhxE3TTjH8VGBnrflS5QI1Ahliwu4SeS4hycXyKbBtTc+QBuH9xGE3bMMG /V7u2jpAjaBA2iVEH3xxUePwWWD1Op7wrPIdYvYsMTzUbkbbKxbTFfh2q32A8cZ5OUpLFFfgfee9 F1E0RPLw4VJluOQ1JONcRjmicDU3vSJUWoJKRTvhcFLJCKXrpb1F8lyFOtCR+2K0mqx2H//41oQM aio46hEuPY9HBMMWLBjHEhzIreIpf1c0ju327eepmiIfvtQUwZKBhmr5kJHetWXB3WIYgj30LqiG jzoOn2ntILZGJqlMi8Unqsnh9De6vecVTP4YlxgNnQ+NVwTTEpzj0QnI2pik+sK5kWYmq3wPOy5Q pLbgCF5ibbDoyfmEpJvLiUzytr0ZWoqxZsZatp2R8Ckn9PwUqbR92ltk3CJSHUy/qDaH00FWv/es QbUE+aZV136OtH0kmNbgHI85hYum3E2jNUKldBwPSM5VpNbAx9OVlirnEaC6OBXJdZnh5HNDhW97 lWanvuGknvftHaS+EakOpjWqzeF0kNXvPWuwrMriINSsq2UBBw5i6sD2Rk5CssPCyBDfCJVDKTU6 9YMgtQZXS6Xg0TW6yaIfOOWkAm9926s1OJUBhg5uhugaMopryAgnm+ZU/DQ3Qh1MY0Tww+gAp7V3 Rgw2gcj0yKHyfa96RjBtQQF2xwwyq2jOjWrc06FwLz6HrwSJ3zpVIyc4mO+cnAmqi9Mg295M50KF Ci+s7RmihtMrlcqR6rR3kHpGpFoso0a1OZywxr1+73kG5cXJN7kcIgkWDAKpAl9bdGwnhWUDI2pC BMA8zafUW6Bu8eNTaxUDaQlmMbkUYNkUieX3qYVMX19nvzZ8VPGurQFivpHn0KzI2rNsDmKmHAmf ZxVZwvlKFQEluxGMlYmmFYNhBXY3cuiP1DW3xKF0qDQD2yR9/SKRzSlT+tASeS5BoovRIHpEGLV3 hEoNUMElnDJSx/vWDlKHiEwH43Do6DKMSA2K2Pm3uee2yGQMao6uJM3kUnumoXaPwVSB+5lZwsG8 3LYRB+DY7ExWc3oalyiidEzQ+ICEbrJU5IxyvzjhJkGQru3VGMV4nkMidsfhhJ7eFNXJcdpbRNuv VItdkmPW4bS63LT33BrrGawY6Ris1THFqDGE1CAn/mj0JbfFoWRodIt8DnwB0wQd0/3FxeweObG6 bOb301LNIImyMzzaD8vlg/rdNXWAGiHyHMzXFyUOnwVGK0z33ATrEJRtJP2jYJOMsjYoxgyGDJvx CyuwxZuk+3OotMKrz6lyEDVi7P/uxuJiejn133A/nAbRXduezy0VKnDsyOAxnNTyvr2D1BaR6mDa opIfTge52nvbGqS7HJmn6/PBlKzcwEEsOnBoA+fEGq6ooCafPJoJO6u8N0iNQYkJy18XFSnjPSMU L7LN1RjFGCD4XE+nu+GEmoyht+0dpMZQqANpi8p9GB1kmnvXM6gpSe9AY6af9YxivE9qnJkiYQvO jnKGCjVDpWfk7NdBaoyvjHMx2VAxrQz3wwljiMT2oakxSoUOVkvrGGGkmvfNHaS2iFAH0xgVfBkd AFtMa2/7BQlvrksnJCBIbSGmX7D1QRKsLbjzv24QKkPrJ2m0RZDagoE624rF9H8OKdxyOgjSneeW ChW4ZagT33BSzfv2DlJjRKqDaYxKfjgNEttPe29bg4uGk/9xvSuz0VqjGNbgIydUkWANLrLLaOD+ ZuesXHYcLyhSa3CiPqFmqUgcw/1wwhpF2P4PJ59bjHqjfD1c1wgn9ORq2tv2FpkgEqkW462xoZn2 DqeDXO09t0aDCKWUlFcbRThTEFssggbsfjOeYQtcO37ALSZI7zytcw5/rx24qzcXiZeCjDFcDw80 LtJ2aoMiSM4nM7plLRdHAdaJCEzTTn+3nZVkEO44GekPj4NsO+j+q2usoeUwTvJLMgH2AfWExZCc BVg6LW0yWvS9SxW5sHvfep6rBbiaI+uZVDnzg++YLVC4H04HQbZwqhVKhQbM35tOhJM6DidmguG0 SC0RqZYKj13JD6eDTHtY8LknNJZSc+gnnfAEvkhMhW+tUYyklgkJvQDv4grbzCGohpl5KiUoySGL qJWTJUyG6cJKY1AYDPPD6CDT3ObpocJ0P/yI7TJCS3ZFblozzSiiLUamg2mLyn0YHeRqbmzxbA8E auZRSTSJ0cw/a4tiqEBXMa9gXPTyFdWUKp6BmZJCBolSzs3o0EsVz2AeBffDCWMUQbpwqmcUQwcO 4JpXDCf1jFT2kLR3kFojUi0Wz6jkh9Pf6PeeZ+SLwgjH7l0Xbyy0zJjPToZX/2ALPgpQL5AIadHO zTGME6BuQdUIIWaw5J7uxGuJ8mEcDDBtxYADIT2T3S6iho0a3rYVp9i2VqCLCKUj8+GzwNXW8+6x AyXlGkk1mWoz86xLFOM1sIzpgjHpLGV7ashEd+anP31uEYdxXmSi9mIgpIzhfjgdJFN7OPncUjkh Z1cS+w0ntGRuedveQeoSkepgdpBKfjgd5GrvuTV2sKBeI0ubfAmArLnWEHM7nl0Mzz1iDa+uVc0Q OZaSI8cBitQYEMdRhgo1OaGQnDWM4iZFEC6caoxiqOBHQ7VFGKnlfXMHYbYzMh1IU1SV5XOAq7Hn lli/oGQjiSar1bzoWqIYluCbwiZXZHCeG9cUodIULD3UOCLjFxSNZNwcKkxBT4f74XQQpAunmqJU qEC1i6HTaR6cVPO+vYPULyLVwTRGJT+cDnK199wa6xdUbdChCQB+iaDGENIt2M0gpdIYfCIjXvCa zxVgi88+hXUC1CteuRfzQPo+pyoyVIRNnKLItFVDFCK/YhPGnMKZHnxQkUXNnQzKZ4CaQXkcPobG ylBpDhuscKfX2+OlhZOMkxqBCV2tUMwXyTKmu19Ygdvt4gDcJACVLsH2Tl5/kdpBRaxDGCoMwTEE Xu3hdJBtr5YoFSqwPOB2UPZvOLiDlvftHaS2iFQHY7c12sQY5TTIpd971mCZJBkmQzwzUH3CAksx dOCLzy7aWJjnXAM9Q6U1qCuoBwSpNVjlSpQZKnSnav6W00GwxnluqbAGFzYbnZznwUk979tbJPn2 SLXYjeSH09Hlau95B+lwYTEliQwLJ+67ddVmMfK8XPyPt7EiFEfgE54QaYxvPoZ5gmS0cHrGguBi KMX2RZgvowNsa/WMElFlQ3WsFfPDCC2Zst82dxA9Y4Q6mG+xuhxOB7nae26LDhbWUprvu6/xnfl4 PUPM0QJ12Pi1XfYA4gfUHM+snTrfeEERtWLPir9pixLxgklDknGGUcbEIttcHaoYKuQc8jJSy/vW DmIQGZkOpi2qy2F0kKu557ZYv8AzyQmh5qrQphcWWIphCy+gjC1Y16sbhErH4AryOoHIOAb1Igk2 VO3kVh8G73tOmKfItldjFEMHgpXphTMwOKnnfXsHqWdEqoNpjUp+OB3kau+5NdYziPwa4wcL7s0u rLHMqEcu6yW1vANKI+IF30KkLby4FR8oULdAmISaIUJx71zDpuGTkQAgtelhU0OEhjULlNcKYYJ+ 7IjMjNiWDlAjRJzFEHAkHi776+r03qCZVQAt8JW0N5MxKildGVB06t3MKjyG3+kGB2YyWXUq1MlG kdqAWpIEmmDpCZxNuOWEVYpMezNxKYYC3/IFVQwRTuoYqRy2094iPrdSHUxLVPLD6SCr33vWoKKB hBBqZuzNsSylFMMaHKdjrki7VFunYi9ECseHgPvyg9QY1JMkznwrFarnxrGL0SBxifPcUqFCPiSt V4STat63d5AaQ6EOpC0q92F0kNXuPVsQOFhoJKRxc96MmUKagq0Mty8xBbOBjADcE7rTM4sXCaoC tQT1IukuQvEKypFu2WCIItOWTy3EeifzGvMK53nwQUWC1tWWOgeoFSLPQlqhIh82BzltPR8fOlpa LenBKYi506VWWAwrWCoUKyBY3v/XUOkR5EJ5+0FmtCRmJnQMFTqyl5BUczgdZNurJUqFClxdYAUk q8FyUsv79g5SW0SqgzFZGW0Op4Nc7T23RkdLKihNarQGpf3NKYoZR9nhcC5GGkdRbTwgRBqDLznW AYIonBM01uL0k1KhOocSMmDKKG5SYFurLYqhwWc/dnoYqeV9c4skvxqhFuO51eXitMjV3nNbrGfk IiFswYy8N9TCsxjVwCzdmFOQf3LVmWp+CZXGoAI8XhAk0jlF89KbQ4UxckkJVh1OB9n2ao1SsUHC 2gfFA5g1nF4pV7pv7yAaf6Q6mNao5BenRa72nltjPQOvTLbJF9GZiJphUVUJpmf4fSI7LHmchQXq GSqtwV5N3KBIXYMS+WQaQ4W/c8TiltMgemLb0xqlcv7ApN7NaqwRTup5196Hg9QakepgK/ktp+py 295Y4+mqnuWS5vxIxxJZc6zFrBR/9SAo1mB+VE+QirfnTK2zDrIpnqs1uIOh1gimnlyrkIRzOB1k 2pvRplRYg+szYowwUs0y2uYOUmNEqMUQagU/jBZY7d4OI5RMcquVtuBYeG+ohWUx3hhfYOygwV0d 8QM+hjCzVg7KRO8itUUOyPBaw0pTUP2axHUYHWSamwBUKjZQWTOrLXJEhqMyd63pF0Vqish0ME1R uS9Gi1zNve0XLDfHFNyl3OzC+srMKvlOH6FAr/jmtYv0kc8hMsZ+8rraBWoIykoSdocIvS38o68N nwUQLGx4aiH3kX8aiOkfYaOGt219PEDNEIEOphki8/BB5gBmCtvW20MF1RrkhFAzts1IUQjpUdVx jJTWzQ2VlkgjvPjUAjUC+3mJuqVBP3Y/k22GTTQWiWSyqRFKhPSfmbWkY4SPCt61tUCNEHkGSq+o yBcbHr/V6+1eYbRiuoARvlIKnk8MfV3M0l3uFIwveCE9Or9IFCt88bGD1AyM9Im3oUp6gGIwP4yw TJFpbgxRjBqMn/mOKu4QTijJ+0UorZ72FsmARF7pqtAQIdTKfTFa5DT3Zi0aeS3fxostuK6jOdZi qPCV7x/FFlwzGwfguJ+38yAcd1lX8TxXY1AtkngbqmaN4X44YQxOVVztzWhZDB2YgTjlcY4HJ/UM 99PeQeoXkWoxpFrJL06LYI2R863u4XoA8wY9gwTpayZki6FDFpXocOR/9QOJ4hls1VTxIArHnImz L/GWsHIE4MXCfBkdYFobxygRGnymQvhipJZ3zTleFrG5Eepg2qK6XJwWudp7a7xkzYxPhWgLjpQy ETWrWIx0kfNLZhUcnP/aKQe1vMywfVNcJxjFg8RtnaN1P7GYxuAsAtwPp4NMe+MZpaLGhkvNu54a TujJ5shMj9PeQWqNSLVYPKPaXJxWl9XvvTGDapuED3Y9uIGi1iiGDlRTOZqRxbIy5Uvn5fOutQYF MckzitQ1KNdKvA2GdNCH+8VpEaQLJ59bDB345pn9kv04OalnpDrtLRLrj1QH8z1W8ovTItMe1n+7 n+QacX3jxQqxWqMYOvA9SbsFWezrT24pQO5QKR17e9U8SK3BwZHE3KVyyIVK3xhORQwlba/WKBU6 WP2mMcJINe+bO4iPjVAH0xgV/DBa4GrtbVuwnMtaJtt2XJzfuhurXbwAk30JFq4d0chhuXfDHIFj glAZZTmrnkyySG1B9VYuH1oqsvUHTotse7VFqWj3Zz6tijHKiXzxob1FaoxKtZjGqOQXpyKXfu95 Bh9RSsb5mRtim2Zh2GBYgwbMtXlFftNea4RKa/Cx4CSTRWoNzyLfEDlflWgYYcQiES6MaoxSITh3 9dotnfbByNxZma7WFqktlMlRdalW7ovRIth+mnvbM37yTTIdg88Wd4OIwZglXhVgaqoXMIZ/7nzi Z4iUjaumo3WAGoLyETIpzDVEJI/3fAbYtmqH0CA99ar2R+Z7rqupoWx4D9PWADNYVKDFdInIfPEZ 4GrrP3/4Nx//3cf/9vSoEDmaHyvFCt572wRjMZZgqIbxhZC/+QFSNOQCGMuCEI0ToMkng0wYycVz S8VzOZYA1cVpkWnPMHKoKK3hULrd0cmeBZN0y0h1tbdIXaJSDYZUK/nFaZHV7+3uwTxuPnLFBPFr y28WQwf3IzUGVyPFFtIAsO2Qp7BOkPoEhSPpHcGQjU21UB0+CyDaeWwxFOCSc1tj9yuMUPKuOXtC kZoiMi2kJarJxWiRq7m30guWfvzQg37Bnk63hRZDA44ymXOTzPKB3thCqvgF5bfVO88pnJO0KU4M pi2ozIDq4rTItDd+USp04Dxbl1TLCT2Hk1NCOS1ieyvVYPGLanNxWl1Wv/f8gtqNJJ58O5VLdAyo rB0Hy2IseyC2SwaY1XAWY7k3Qc9gt53nsjwLUmuwC5S1vFDFM6gXSObJSm84kWDftpfnFiP8EZDt lRRYhRMxM1Jd7Q2S51aqxS7JL06ri1U+6veeNagoSer5ibWcphdsxQbDGhxlsleQznKbmqOiV074 9pwd+dwitQbLOelNS5VC16Sey6mII3Xb87mlQgc+Px5jlBFqPjS3SF2jQi2mMSr4YbTA1dpbQYQa DVY4Ixv/Nr0opnczp/EOJGxB7WD9IFTago+EVO8gtQUzyyzl8cnxTHJJIEN1cVoE6c5zi6ECz3cu XE6oOZy2vUVqjEq1mMZQGxfPl9Ojfu95BofpiI6U0jEjaHpBqVcwNm+oxDe9gDs3YuoHDB6ZwDo9 8rlFag3efxKwpcoxhYTn5bTItudzi2Fz7rawX7pDJycS7Yf2Bslos1JdVCv5xWmRq73nvtFlcOst 2RfRGtxBlCx8IYzBKUI7BUMSB6yjukSGF9bqqzdARGPO5leUNU9pPKQgzcVmkWmrhigR4n8n1cDw ywcV79saoE5ReQ7NijxskLCIx9mj13su8ZPyyBiB/dymWTwYDPFZ2OPsHFbga/d1gFBpBk7LVucg dQkvfNEMQ+QBBYmGEbIViWznsaVCA/YNuyIQRrTx0NoitYQyGWeXauW+GC3Cv9Pcc4foToDDK1MR 7MdRhi7vBVKBL3xP20GdHJa7mlSSYpOZtbIbcgE1BByScS6RpxPiEMtngG3LpwZiZs2qVme/srFj vNgW74GhQDYDTPyoQIvhuKrhILF8BrjaeivjJOJ6zZNWYPG0mcVi7oD/8LgjZuB20Lx+PhPpjUWI xqWAyTWCRDYnaS1SLKaOhFCoLk6LTHvpUIuhAsGqsamc7H1KdbW3SF2iUg2GVCv5xWmR1e/t7uEk tteU/aD8rpnFYuiAMI5g5G7WK+ETEkU4BoSkGkUUjjkad7keKoTLuYQsdi6jRRAunOoWnF6AChX4 X+e+5YSaEUpjtL1BYvwRaiFtUV0uRouc5t5c7GThi9tD9AwqoEi1zbIWo2KEs0wGejI3arlVkw/a z3T1Nc8NMp5Bf8xYESqeIwsN1cVpkWlvhtpSoQPz/05Gy8kNYqXSGG1vEY24Ug0Wz6g2F6fVZfV7 zzOoK+F1Qs1SXK7txRrF0IFlPb2UzM09Ia0hVaTjgt24QZG6Bh9TSMQNhnQ5qADVxWmRaS9WXIxc zgVCfayMUPOuOUfJIjVGhVrM11jBD6MFrtbemo2wneYlEtqCG7uaZS2GBqw52ikodKIAO7aQCi2d qnXm4U1fud0EFQiHCbjFtAXhEO4Xp0WmvfGMUqECV+vYK5cTaob71d4iNUalGiyeUW0uTqvL6vee Z7xSY6A1nF1kMsK+fCAqyrjPSzcgjXWbFMfINMURlYPk8YEAdQs2gzOuDE2OXKRodtkU2bbGKaja ghHif+KOnxiifEgynbZcbRWI+VaeQ7MiX2wWqV609TyWNrmiUMcbVjQCn7BtXrEYViCpdyTjr7k6 GyuEStH4RkBef5DI5jStZYpLlWMKUF2cFtn2tN9i8OSGy05/ywktH9pbpC5RqRazf1Sbi9MiV3vP rdHMwkJLNuy0xmcvD3foLGaFMDe4eAwUa/DB++oeKq3B5VX1gCD1CtZ1E3pZgch0lLwRqltORa72 ag2p9AsuF5vpbzmh5117H14XqTUq1WIr+S2nR/2ObzytMVBjE358jkSnidZicOJYU4zBN++jOteH zYyVrxD1FtMgNYYlG/SiJfKkgrwPnwW2sZpColSc8fEpe4TKwYcp8F1jF1JLVKSLaqXGpjJi8Fjk au5tv+DbJbxLe58lSPULsVSOc6bJZIv8jXlmtAyVfsEEMD7AeWyeUzoGMBK+WyoPKtxzKnK1V2NI 5dv0IHZXVcsJPe/a0y+K+NxKtZi6V5uL0yKr33t+kZu3IeZ0XfML88QM/ZRcMQGxVS4fqeYh0hZs GqWIK0BNwZJdoi7by3ByKEi1Gl4xfAbYtmqI0CA9t0lbb28aKRs0vG3rAmqGCnQRjczDh8YDOK+o XscIT1d8mSuxku4rYnOhd/XCoBiFwRTrOZC5hB6PeAmRVuDwX95+kZqB4pEE3KUiIQ/VMEK2IgrX 5nxuqdDgUzYJnezJiBrVh+YWqSWQSR9ZohX78FngauztzkGqz5ETqNkb6mqFq3nOKCyRZpMHYclh vZuHIYCteH7QFD9Cs0hNQY1JvGSpSMcfOC2y7dUUpSJT4Yi+WYVTMDmh5kN7i9QWlWoxnaCSX5yK XPq95xeUcfAOXAskkexIUSixyAt4MAblZ3F8Ntu7fPEzT1GeIlBTUNecMXNockQBzsMmDkDhKmuD 01YNIeTU8ycHdnYCJp9E59u2BqgZKk+hDI8V+WLDGvOtXu8ZgQKOJJocEHrtzWnscwZD/G8stMYl CO7IitJSgTgX6nSjSO3ASYlE22AIl/MJcL84LYIhwqmW4BQDVGy6kGftbE9OaBmprvYGid1JqSPV YrpEJb84LTLtkf283UHY0OINQP35J7GyPlEMHSh4jTHYLIvnU7e1M7Q8hnmC1BisWSfaLlWOJ8D8 MFpgWpvMjL0wiNyAYkuI5px8yQgtH5pbpH6hUIcqjlG5L04Uetxp97YtUIasRlt8xsqtPVmMxJVT PI2kfN8MeeEtVTyDjyPx3CI1BtlicohQxTM4WAHVxQltbtvLG14MHVgJ38menLBdpNIz2t4g4xmV ajE+OjnaXJwWWf3eswYZNm8BatZPurZp7aUYOnCMx17BIMoCn5nGz1Ap3dfQLFJrkCsTcRezn7w+ cCrigNH2fG6p8Ahmhc0rygk9H9pbpL5RqRbTGpX84rTI1d5b/cRSSyMT1BwUbWJRLNbgHI+ZBbkB YkXPUGkNlmHrB0GUjliGd2mMITL9kQizDqNFtrkao1SowPDdKAWjeIaM8LFtDb1PayvTYtpCXexx y2iRae6dMcNiS2cPjIg/OUiaMWMxprl8vMdeQRbL4UHV5DrnnanluUVqC5IpQu9ivvOfobo4FdH2 ba/GKBU6sOPQ5cRyYrx9aG8Rn1upFtMa1ebitMjV3tuewfddIxz38zXDcJUqw953vl7piEYOS8FQ tAyRb4o6kQuoKSghybVCzF4zmSMdv+czAIKFwqcGYvWO8zb2SOd7blGioUTxirQ1QIaLFWgxzRCZ Lz4DXG29teLLej2HJOMSX0gejCIL8R65atGRjNRNt+FVQxO5WJruu/eh2oDKEcJuaaIeIQiaiwuJ 5G1TtYGQC1RcTXbmefJBP1lr77Y1gE+NOIdkBR4uCLgINqiAb/sBZRvYnXKErCDXBmJuHXKNgFcT 0SifiI7O3BB0ZmY+R0VKkJqB3Rxi22JJGR84uaV3214NUQwVGBDshc7y5MSq3l17Jp9FaopKtZiq V5uL0yKr34keT+fnFEp4G7HWoDiwqxWLYQ3OljuAkbh94aiCekqFyZ2b+dwg4xQcL0tnCRXPsVEQ qovTItNeMovFKKR5ZUEw1ign9IxUV3uL1BqVarA4QrW5OK0uq9+xxtMJmVVWzkGhZim3RSeLoUNq g9D4KwFfzaVBWidonbkWqWfw2azkGsGQDZlCtXz292krFlyMf1n/aKJdPuh411r8IkgtUZkuqpX6 4rTI1d70kmd+8Y1KRJbPXCn4SYdIVnEwZh6c43EMo/uzaqAt8BPusdUWX/LcIOMXXKSVaBuq+AXz uIy8ywn/vW1v/KIYOlBrYq90uicnpzFKdbW3SK1RqQb7QJ4z2hxOB1n93vQLqCla521Czc0mWec9 GNbIkqvWoCI+/s/R/ZmwUjcfLyhSz2DLL5lGsHgGhexQXZyIcbftjW8Uo7dwFV3TinJCz0ilNdre IHlupVpMa1Sbw+kgq9971sD6vAWo2aTK/BQOxZCAsx6OaGSx7GSaV/BxVqiUjivx6wdBtAZDNqci 40FDRR4cqotTEZfw2p7PLRU68M3YG0ao+dDcIj62Qi2mMSr4MjrA1dob0QRq0uosL3L8kxkp0WQw 3ydr+UyZbJZ/6weh0havPBe9g9QW7G7HYENEAhqii9EiCHceWwwNuNnPFIv00FpnlJQovbKtDRK/ WJkWu+Q+jI4m09zb+SbUXDzDu6SXUq2YFOtgqMD+uDkWWRt3xKkmBQNQaQu+OxkfKFJbcA9m4u9S se4aqotTEf2i7fncUhF1KepwdHYWJieWfB7aW6SOUakW0xrV5nA6yNXe257xnX0IheN8YDZFYAAU v8jVmtqCcvh4AR8whEhbpPgZ4wSoKSgeSb65ROzT3PMZYNuqIUKD9Nz30NmvbGIHWccr0hYqn7ZW oMU0Q2Q+fBa42noj34SYhTeMDzU160kvDkbYZGKlA3iOvhkn/3hfD0MpW5h9/XmudsAVEnhDlZfN UQWoLk4k1bftJYwshgqcRLQ7Ov2SE1qG+9XeInWJSjVYwki1OZyOLqvfewMn9Ru8Ahdw2XxwuxAO xdAhlwpgjO9dtqAcZyarnPdK8lWktqDGJIE3GBajECZUh88C09hE4hKRm1OqaW9klyqMXAe+bc78 okhNEZkWuqQ+jI4eV3NvpxdkubwBqBkXshkCh2JowIEeOwOprFfFMVTQTX66Fu58yOcGiVJO0axl XCzJdKguTqTXt+2NXxRDB/pdY0g5oWekutpbpMaoVIPFLyr54XR0Wf3e9gsrKE3+oWZJLpUn3xaz huVr7hEglXXPRj2lwj7+1OlHkXoGhyYTaoLFMzioAPeLk+3ctDeuUQwd2KOcqVg4oWekutobZMJI pVqMw2OjzeF0kNXvPWvw7fiknqR/zDHsJdRVBkMHql8Y1rEwF8TlglkSzpm2socWPyiiNdy4cmeR bZmhoqfeczKpBHGkbns+t1RsC1EafMOIvaSH5hbhsSPUYpfgy+hocrX2VhCxgpLTdcqGAk0viuV9 cp8ufkCzvKiMBswhZ/pK0hUvKFJbkExkoWepkuIkILFTIif0JsUC2fZqi2KoQKbbIFVOqPnQ3iI1 RqVaTGOojQtaw2mQS7/3PAMfTOLJKWwGhnpGMd70z29eVYs1uKaifhAq/ZZrMeoFQWoNrqlI0sEy TCa5TIxCdTgxOBbBGue5xdABo3VptZzQ86G9QRxtjlSHCt0r+eF0kKu9576RsguoWcMkeWBsQ5Ha oggacJEdIxqtsnIdP2BI5f06Yevkw99rB77BZTZWApeRJDgsMEKRNuMji7CkxVJIZ8B+yOuVLedX E4ttJb/bypGjiJ5QSQ+Lg2wzaP6ri2qhJZNLoskuEIlDdS+G4KSJdFna5EP3eet8RcWLapHrZ55D 2TxX/bkFMzGmmOp9DtXhdBBkC6caoVRoQAg2wXLyJSc33pXqam+RWqJSDYa/rjaH00GmPSz43A9S cAL1a2dj1JV9yj7ZQiymUY9DhoUt2EtOoslEJFN3Z0I1ToCaggwkASZEGIzcWaLDZwEEO08NDdJz mtuUgiXAsEHD27Y+aJh96gi0mGaIGofPAldbz42wnYGdRYxv1sORrXGJYoiPT/NCaJXBO53+JVSm GBRq5vUHSWLg3Mw9Rdeg5aUdKGKC6nA6yLZXlyhV1sFnj7acPNtaTtveInWJSrWYtqjkh9NBrvae W6MuYTmluT7UX5iExxrFXAAn66bERGtwfVTUDJHG4IxCHIDFPR6rU7AKnNhSTGOwUAzVMIqbFNnm aoxiqMAlFg2h5YSaD+0NMuNkhDpE31aXw+ggV3PPbVHPoHDSMjxtQUFEdoTgUIxiWA7wNJ3gc/f1 g1BpDCZO9QKR8QxqIhJa+PBgZrMkn6E6nDBPkW2vxiiGDi/5rmrmeXJCz4f2FqlnVKrFmMGN5IfT Qa72nltjPYPCjaSanPJgk9+h0xJLihd5oWxouMaJt3HBXhyBuiyotMa3PMfELIha8bMXENJPlgoO 95ywhojWb3u1RjE2z/wo3cWIUraH5gaJZ6xQh2oEdx5XRge4Wntui/UMvhvKO4CaD6w2uaLIMhhy cxGmyRXprNfFqWWotAXf6Y0XBIlnOFZ0R3GpHCukOpywRZFtr7Yohgo4Rqe+5YSaD+0tomesVIv5 Fiv54XSQq73n1ljP4JIu3ibUfuaojiGkYxDnXWWjWdLpjAZcnmWIY4qWp7COQN2CAJxYMzQE0tAM GwaMRaYtn1oI8dl4MakgSocPKt63NUDNoDyGltKgc7U4bA5y2nrbCJzoY3GKBRVOAK8VxHyR3uRg ZkEemyveWKwJlWZgIIFmkdqBAYfivsXUkY3RW04H2fZqiVKxqc66mOHHeZ6cOD310N4itUWlWkyX qOSH0yCXfm9nFgw73mauNTBGV28Wc5hjvzD5CxUCOI5LV1IZWHJ7+UFqDY/bH0hj0MFgPoziAiIR LoxqjFKhQkqZNEYYodtDa4vUFsoUiw3Vyn0YYYt77d6zBT2Ad4AtWLBOAQ4cxFSBAG4UIZWjWjta hkhTvIQE4wSpKagnyXU/xbQFi9g3jA6wrdUUJUIDzid1u1pGBhE2SKgKvpobJLn3CrWYflG5D6eD THsMaW/3Eq670S0omCMAdKgIhCU4xWOqRTrnZXG8cCr0qDxRtpzfH2CGCqpLiDOLaYlcGnb4LLBt 1RDQuEDFt7dntls2dP4IdNpaoC5RgRbTDJF5+CDhAFdbbxmBBQ+3YbQCJ6CbVxRzwGT6jfaagXE9 fk/+7d2zmIHTMDyHzkHqEZSMEGMW0w4cTMBHDqdBrvZqCakMgN+5u7ChtJzQ8q49u0KR2qJSLcbn 7Uebw+kgq99xiaf7hSySeuOq1uBsTvOKxdCB4tLmFWTiUTNEGiMXp6p4kBqDkhFizGIagyMVyTiH 0UG2uRpDqviF8z8GivJBybvW7C9FaoqItJCWqCZYwuqH8/vV1NteQQVJsk029hHDvsEOC5jvkmVn i2xxNL5yHx/wo82drHL7SDygiAo5RXM/UfVDhdqcTbjlNIh2b3s1hFR6BQeDMhFdTh4huW1Pryii KVaqxVbyG06jy217Y41feAWVJMn+yCdJ9WqNYliDrQl7BDktaW70lCpDBd+/iAtwByHP1Rp8biNh ppgvmJMWUB1OB8Ea4eRzS4U1KGB0wYY5WDihZ7jrhm1vkVqjUg0WT6jkhxPWuNfv9JFn1QVQs0CV fPMT2+dZ/V/ICSZTkhiD9KuOkD12ZOM4ar1AoKbgCzzpR6+7D0+JZYqbhg1ai+Q97VML2abXomuI 8kHF+7YKJOvmx8hzaFbkw+ZBLyR83kGadNOtvMaNNUCqv4j/dYlilCQypNsd6I58mVeXYKNmJqu5 /m2RJN1M0VjkvqFCbY4lwP1wOsi2V5coFa+PSgHLFpYTU6GH9hapS1SqxewglfxwOsjV3nNrNOk2 G3FEhJrRuVlFMYcLdj0cNQncFnaqeoiMKiw6xQGK1Cm+4eA3VKjOgQqoyggtF9jWaosSoQG1iGYV TvVkhJYPzQ0yflGhLqrV5XDCFvfavecZ1N7RLtKRRjatII0PhgocujGvINO3DkdjhEpjsCdQxUXG M77To2+o0J3766E6nA6y7dUapUIH9s4ts2eyF07o+dDeIvWMSrWYnlHJD6eDXO297Rl8KwoNKHRg GtAlHCb9YHoGy+5kYFoju9xY41uotAbnV6J5kbqGn9K6iFCdr0VBNIziGkW2uRqjGFVFP/JZVYwR RgzYD60twscAjkyLaYvqchgd5GruuS06ZrAm6KodxHBvpjUQCnCAx9SbnJZrIaNkiLRELkoyo3RR ooagtiRxN1iiQ25mPXwwTYBtq3YIhPTfyBvSQ2Rj8GQfA4FOWwvw1BFoMU5SVI3DZ4FpCwmfG6FD BdtpXIgaKzBLaoa1GOKTYHWsYDMsk05uNZspO2u11TmIsjlBcysR25QKtVmbhepwGsRiTa4xh1Mt USpUoDd0RbWc0PKhvUVqi0q1mLaoNhenRa723rYG9Roke1RKcWSxmQXb8mDq8JlPZ/tayOU6MlBg 6z09+ATfVo/iRWoMCktIqpiElQrVqdu5ZXSQba7GKJXpXW56x6zlZCJ1394itjdCLaQtKvfFaJFp 7l3PIFCQMkDteG9EpYQnkIXir974Q6sE1Ew6WcHtXJVcNIlGgFqCET9RtzT0dzbCoTlsDtK2plOV iBa+ccsObTnTkw8q3rc1QK2gPDc0Fdn0fdmsEqPXe92Dag3ahJi7xVm5YSSi3C8Y4n8i+sUKXHje 9x8qPYIcuToHqR2YVmfA5Fae3DCbQwhQHU4YgoMKN+3x3MG0NZdixRLlhOLldNobxG7Fu6lUi12S X5xWl9HvXZegYAMNKKvkWHmzLEssxdi+oxrM3mCkdjrBUkyotMbP0CxSa/CFw0TcoUJPDibcchok ld3hVGuUCn//kU+q4hflRCsP7S1Sa1SqxbRGJb84LbL6vecbrz/4HqHWoP5le4hYrMFxphiDbdQ4 AtfpQeQI/ikkGCNIjUF9EQuEi/nSMcsNowNsa7VFidDg2/fZmZXRx1e0lHniZpsbZDyjQi2mLSr3 xWmRae9dz/jBvZHxDG4MaF7BkBsMz+Awk8M6eS3rCFEzVBrjJTQoHqTGYJxNzC2mMTidkJRzOA0S 2+9zhwodSMQ7/S0n9Hxob5F6RqVaTGtU8ovTIlhj2nsrjFg7yVQMz6CygcmoY2cxPYNvwXvtD9Zg Pzc9gDKJmbJmxoQbFKk12M1I6B0q9PzuvOridJBtr75RKpZguEzcfumUT06MTQ/tLVJrVKrFLskv To/6vd1PWOLwEn6tQSVEU4zFsAZVFa7jufXUyQfnGmbOStFvcs4itcbPLgoPEapjzeScYZThUiTG D6Mao1SuR3mcE1uED0o+NLZITaFIMdhQrdiHzwKr23uWcN8D2hd2/McrnHVgBc5JOFyQzLEybQr1 OTT2kB4b7O+K5TRt6hOlyWzC6+8PF4zi79tODSCC2CyWdh21PFD3pp2omt9tZ2W5aFyxv+XS3692 nveJZttslHBoUKnYSunFtPSJYtQHkTWYWlnD03nGS6jUn8tckmgGMTfAAryXxNuhQmO8CKrD6SDb Xq1QKlZk2QNiW+xwYl35ob1FaolKtZh9opJfnBa52ntujabdbKtaUqU1ODXfHaFi7gX85Hyw2RXJ 6xgjRBrjNY+x5ROkxqBkJBlXseThoRpGqJmTCjfN+dxiqMDhm66jlhNqPrQ3SIw/Ql1Eq8vFaJHV 7le9Yj2DKpIkmpxOYeDvaFkMFSiudfQiefUHegZLUDNL5SrLKi4ynkFpScLtUuVYAlQXp0WQLpxq DA4vQIUO7FqzdoNnlBN6PrS3SD2jUi2mZ1Tyi9MiV3tvewa7PskA+fwYc9BaQ0zP4LLcdhM+cZ8O wDe9Z6LK3DGKF6lnsA2ShHOpciwBqjKKY3BO4aa12qIYc1+usOw8tIyY5zw0N8g4RoW6qKg4C/eL 0yK8zXB6zzOo2ki+yaVUJIu1RTFU4PU5lhGT7S54BnVAma4St/PcIOMZ3CnI6HVR5ZAF3C9OiyBd ONUaHMWACh34XghrN3hGOfHvQ3uL1DMq1WJ6RiW/OC1ytfe2ZzD7y7jOB+MRqdYQi2dwmMnIRt7G J6SiZ6iMrOyoVvMgdQ3KusmosMZQ5WDCHacijlFtr9bg+AK80IHPAppj4fnhhJ537RlLitQalWox rVFtLk6LXO2NNZ5uAhDtOC2sdJ851NQcazEqgjnMZI5F3sZtscZGPkTq9TpYgyWN5JJBkg8T4b65 qzhUPJfzCVBdnBaZ9uJTizEsU8waY5QRhYgR6mpukRqjQg2GUCv4YbTAaverftJoQoaNb8YWbIrm O4awLIYGXFPX3IJP3vPu0FIqxHWm5nOL1DN+fndrcbCMEQy8UF2cPH9x015suJghihwmxign1IxU V3uD5DnWLSPVYhqjkl+cFsEa4fSeNayyiDUY1LuIwww7GDp84rAd0pHDcmmqq57UdUKldNZBRPMg WsPJWp1liDxYIdHFaBGEk2Yco1SoQCVJl1TDCC0fWlvE1iqTs7dQxTEq98VoNZnmsOFbYwYLb6aK pFU/evck1g3CQMy9RiYZpLA/Op2g+qRTVs5saIb8XitQO3JBdo98WehiEmDbGX/IB4FI337mq6qY skwoUL5rp7/HbCvKIRlhLyZ3+vxS9+YV7n6RaKo8Ib051mJIzi6hfYAclu4QdUOlH7CMUG1FIpn7 VK1OpMgM8mSOobo4kZXetdfeVAwN+OCQ85zlhI4P7S1ST6hUi9krqs3FaZHV7z1PYIu21mDK3SyL LXQwowdzRZbNtEZ7ADNFaLQFvaO2CFJvoGKEeIuDDFXOUUA1fNByEWQ7zy2GBpxwtg+6HycntHxo b5DxCmW6aCq1MWj5LHK19rxPrF+wMpJkmJDJ5Mo4armQGLkaR5h8v/zkt5fQ8jVUxpJPoRlk/IKi EaLtRZXTCHecFuH5cKpf5CZWdGBJ8ItZhfM8OUH90N4i9YtKtdgl+cVpdbnae26NjR1MgXkHUPM9 n2YVvFkw/YIy/8yXebMWxahnqLQGMSReUKSeQb1Jou1S8dYeOBW52qs1pPKNcqUlr0FrlBN63rVn VlGk1qhUi2mNanNxWmT1O73kF1kFdRvJhz+xk7BZRTGswRkmwzyDvxfeaw2peHuMxnlukIyAjNnc onCoeI5xPVQXp0WQLpxqjVKhA0cZ7ZXLCT0j1dXeIrVGpRoMqYhtae/itMi098sRtL5BdpidOUYB yq1yQ+23xRgpmHY6bSSLtXAMa3yXCoQMLs8tonTkeRxJPFQZIzhbkfx7OZEd3rYXKy7GWOEHyi9G DCd3zekaRWqMCrUYlyuMMofRAqPdZYunnmGVpfMHqNnl6+rNYtgC/sZOclg/DKEtpIpnfM1zg0Qn 1mGslV8qniOLDtXFaZFpL6PNYqjANRCOUssJNSOVtm97i9QYlWowpFrJL06LrH6nnzwtOmG9NvkM 1NSfpehkIUMIm8UIRxL708kCqrdIzZla06kAiuZUbWoUS8MqLDSOPcumSCwviU8thPhfqMm54YOK 920ViPlWnkOzIl9sFqlel0s8GKGBxPpKpw0kDJRGNcFYjOUgTn06lJGwsWSsFTgYOhNWTgfk9QeJ bEzSbiAeI/MMcjFaZJvTEIuRVVA03HXUPEYy8tDaIj62Mi2mQ1SXYQRNEXfOqt17DsFSJGEaaq53 aXrBPnAwVOAeSiuwSN/4FlJsESrdldKcpF+Uh/BcfQL6hN5iqskpIKguTotsezVGqdCBApamsuWE ng/tDTJ+UakuqpX84rTI1d7zoLqewdwixmABpWs4ll4CoYGfp9QUXDdRxUOjKWaTgy8/8JBFD0zQ LKzTd4aIUpA7NvM7YvWh+ERIkJ39E/thuTg+sug50+I0tUBdovIspgNE4ovPANMWpntugo0d1JDw EqnW+eHAbZZFyWEwcm6+Uez4hWLWLGOMEDmAfgrJIvUHKkYSbpfK9yLzi1ERvbXN1R9KhY0Y/7pd XU4soT+0t0hNEaEWuuS+GK0mV3Nv24IqktSQsizRe2mxhVhUIPyZazNMcoTJl83PM0dlfTuvv0iN QdlIjtUsFaPXA6dFkC6caoxSoUMW1HWLckLPh/YWqTEq1WJao5JfnIrE+NPe29agbIPXiXR8o7H7 QRSyB+OFklY6jpFJMe7EGqHSNSioqBsEqTUo6k28ZfjKdJZc54HTItterVEqdKBYMX2kjFDzoblF 7I8r1GIao4IfRgtcrT23RQcKqyfN+n0XXIyQXrKYs2EPldsslZXRnCNuM2slOibnCjIhhFiccLtU 5I2hujgtsu3VFqVygs5UOMNOOTGrf2hvkXpGpVqMfGe0uTgtcrX33BodM6iz/Oz8AWr6aVduipkP 4BieBMXkHASt7qHSM7gXOV5AXQnPKR0Zrpd+05uKJXGE6paTa4+37dUaxdCB8oCuqJYTet619+F1 EZ9bqRbTGtXm4rTI6ndG0KcZp3WV3W/j9kxyJEfQxRgm6fPGEfI27gNVT841eU4aa+S83SJjDT4J nMAbKp5jfyLcL06LsA4WTrVGqdCBz6foi642yIllr0h1tbdIrVGpBkOqlfzitMi098tka30DXRNT +c4Dh6lqjWLowDGeGIN94XQB7gs9U7VOO4oonJM199UwWVg5FLKDkOg8jBZAtjDyscXIuTiCpycu I7S8a07PKFJbVKjFtEXlvjgtcrU3/eQXnsEuNu8Aai4JaGy1+FIMW3COx9hKLst5kagpVTzjR2hQ PM/VGJN/FVJNziffMVpkm2tmUipU+EJTMUYYoWV46xdtbZHaojItpi2U26rHZbSanObezDOorPSD QtiCK2npzFZoLUZGxVke+zDJrEGWV075PVRKx5w2TlCktqDOJBE4GJxIy0N1cVpk2/O5xdCB7/XZ K52FyYlVcduL7dveIOmVK9Vil+QXp9Vl2nsn67LgMo7Bwn7vqcUYgbLOnuNkpLIUvccWEiEbVukK MAvvPFVTsCSY8FtMJVMVf/EZgH/OU4FMNFn5n13bssnuwEwK09YCdYoKtJhmiMzDJ2/0Tq8zcD7M y3aoIEPC+EjDV22aYFB8CWbxBaMWk1JdwupjPSJEWoGZY99+kJqBwpFE3qXCK0M1jOIjRba5ekQx NKCwwd7ovpycUPKhvUESw0eoi6hyuyy4jBa5mnseUptgWGvpDIJtL+6fb0hdjLkXGxZ2BhI49gg1 Btc3QKUxmD3EPEEiHPScGL2lIpCF6uK0yLZXY5QKHfj2kmUXy8mbE+7bW6R+UakWw89Gm4vTIld7 z61Rz6C80i8iaA1WIhJRC+kYPzjA4kBGBsfyVdSUSFuwdRgn4FB9bqpFAc5QJeoGsnewMw/NsIlb FJm2aohC7DeyYtb9yfJBxfu2BqgZlMeYMjSrxcVmkdPW20agqoQZBMS5bMRIau2lZYqIz3q/SnPA iaQqVgiVZsCT+/qDKBvRhhI7J3DciJQZHbPDB06LbHu1RKlQgc8cdBWxnNDyob1FfG6lWuyS/OJU XS793h4sKLTkALYuwcZ2r6rFGsWsgeRyyYw/3N2B7kSNUGkNAJ5bpNYgX08asVSsI4Xq4rTItldr lAodSDUbncqJPfiH9gZJhyS7jlQX1Up+cVrkau+5b+xwwSV+WeXEH1iqq28UQweU8x1wQ1FtERpt weX4PIUtRCIbCx0sch6M58jIQ7V89nckCx8tsRgrnd7sjeWXDzo+tLZI/aIyLaZfVOqL0yJXe88t sUMFV3Ek3+RaEmZhtYSYYwXr/h6BxBasu9ULQqUtmDLVB4LUL76SNMcWQ0Xm+MCpSPzwPLcYOuCQ 9kpXReWEnnftOToUqTUq1WJao9pcnBbBGtPeWONpxmmZpVk/qSDVYt0XWYxVXQ7y1DOYGsQPqJ/l QlvH0k95jlwwz9Ua7SNFfOcc+4D3xWeRbc2EczE0oPy/e5XyoWYnnK+2FqklKtFiWkJNzFGWz+px Wnsz3/TqZycPULP9SHKBbIuhAWd5HM3I2nJDnHsjUikdNb7xgSK1BOuUhN2hipZsS0N1cSKTvGvP 5xZzbyRl3075ZISad83pFkVqjAq12CX4YbSarHZn7PyFV2TrHGIW2DacBkJ+TsM6knlLWqcTXDsx M1Z31VU6QA1BQQkhdzFVtGRcOwyfAbatmiE0CM9Jss58ywYFb9uKXU5bK9BFZPEobQ0fZA5gAKle xwjPss3v1lGa6jM9JOtNhnUwhP7BuhIvlow2F8OxdRgqHYKhPi+/SO3AlhpB96LCAKG6OC2y7dUS pUKFKXDGkuHEvQAP7Q0yAaRSHaoj+eF0kKu958NmAgjUVJMk32R8W2MU8g1ylVV8gvuRoqU0moLv Pef1c9sNyzfYy+mZm4pYojRGUGkuLosgWJ6C0UJsq1J407RCPvYDJp63bQ2g9cioFWdJjg6Hy0Ha FKZ7boJEDoip3eAlQsxFtNkKORjSf85XdfAxqgFSe8HlRDNHpRagLz+Iojkzcz8RMwwVGVqoLk5F dNe2V4coFSokFFyc0PKhvUV8bqVajK+GjTaH00Gu9t62xldqS5zfs1DK/JM4Cgcx3yUL+oZRUrZc 7sY4FiJdgkPoef9Fagy+j5ISxaVyseOO0QLbmo8thgYs3HRjUka+cVaUOXdzNTdIpsAr1GKX3IfT 0WTae9czWK/hbULN9+GybgOHYtiCBKfG4Gq0VF+QGc9UlTWVeEGRGoNLL5NnLBUrpKG6OBWJ7cOp 1igVdRi5PUzPKCf0fGhvEZ8jj4xUi2mNSn44HWT1e6+fcBUVrxNqbmNIhgUHsXgGZ3hcuSFry01x VqBI5buiPj5ZRpFag/hHVL+o2C144LTItldrlMqqlHxQFGuUE3o+tLdIrVGpFtMalfxwGuTS7z1r cMs/75M16Yx87SfFyH04w6Ofcoo+V8Wx4h0qBy/WZeMHRWqNb9/dW7yoyKJDdXFaZNurNUqFDmuM MmJUf2huEcfqFWoxjVHBl9EBrtaejxkbREgoGZGhJvglsYBDMTQgkOoHZLC5KA4tQ6UtXvLcIAlx ztMSZZeIFDpEF6NFtrmaolQU03DLq8nFMELJh9YWqV9UpsU0ReUeRhn9RPSLaveeX/BJCt4l1OxS b0QthgpcX2B6QQ6bi+LQPFTa4nOeW6R+wWhXYwwVfw3VxWmRba/GKBU6cM61e5XlhJ4P7Q0S269U hwprVPLD6SBXe295BrvTMUXCSH5hjsTFDmYV5K652Q0N+I0V9XSADOVOytxM7J8Ips7h9jl+Lc+q yi9IB7kzmn0QDZanP/ZVtzl+/8AuvYKd5/orKkFZdf7dx//28RP/t9bZ//7pP3389x//60c7ysf/ l7zy3/Lff/n4r//+Ly8f/v4fQvnp4z/8PQ+95Je/8x+f++OfHRzzail9wfO1xAsZjN/RGuRPH//h o+3dP2rX+ptHL8Ad7Y/Pn7RRPoFECkJ60jbYfrlB3nv2B+tTfpP5SMwa0yI+++HX7XIDNQP/Nouy F/BGq+pKYRAplvF+7XSDvPGs2jLjsnYFT8+zaHuDvPPs49u4ez832vquP/7xL/Ou//LHv3lhCvJv fv9IFNRx/s4pMt0nn8PUMang//3PH/7177+/8KJ//8ePv336w8ff/8vH/+H3J+9+WDFVcU3m9z/e Mn35zJvNHrg8P/z+54/w/Dw8v33653Klr1HZY7+8Yftl2LKZ+M/ky2dY8P7sY9zw/bp8/9ny8m1S Jqr30n4brqz1/DOl/cIVD9yK1JzzGPf78v1nS8v5QNKN5Ec3Vngdviy8vSMvYyVjCcXXTMlIFP78 sYBB5o2++LdPscx4w+Z5L7Yv9kl2Qpif5qQ8FXw8eYDnT+q1fZLxjqUVzuUB5MkF3hix5skX5oTe 53WeHOBdPSnaY3BmyrlPHuC5tJeemvE8lV+eP7G9Mn3b/v2Jodxh2FGH4DJ98ce81nfeKUPF1b87 XtzyZPrrLVTL9Of64L+MK9V85CCsiXbYePk0bL/+C9kyJnGw8tuy3SHu+7+MLXdrsDL09XXZ7ij3 41/G9gfvm2/AfF22N6Pc20PydKiNT0zI9u1TVP6nj/N7qf40XeH+t4lt9a9f5BhJKv7/BBomjhNq /InU4IVp+svHL6w4UZhGkElo4NbNxJv/8PG3f/UHXO7zx9/+x/n3f59//+kPH/8uf/jL/vDXj/On P82//2f+/fDboTg/LNN94n+bJ5b5IfzraaZ/+vDb/zqk//f8+9d9pq19/O2/zh+W9z/uHw6r/uXD by8r+bf94b+fZz8v8H1/+Lk/LNv/aUj/ff69YYaf1TBL+ftQ/v3+4X/ev/xxkRWxsn84OvynJTiU v1QXk/3Hj7//2+e5gePWfZrx+TPTgeS/DLvUH857n3DIe79531Xnn/7y11F1xec990+IP75Qko+/ Lcn/tX/5X/aHP/4fR6m//uHDr0V+MnC+uObwkbuDmQJ+WYkn0P6HD78d++IaFefP56cj6pFjJcV9 Svz/HLk+/uFDPPu/2z+tNkfPf5xXun/44/kLb7Ds5ocPvx22f/ybJv/2qbu3aF/vUMCg4aFnw9rn j1+Yg/Ch4pld8PPD5OLjTC5e3phcwIwRiQUHgpjzJI5cMrX47AIH++N/cmDgy9zsRr0wXHHkpBDV 5yR7n1LI9KMQe6eWPpiHf3/5xoDGk1SpM5vlCnIci4oDMJbn+NYPw8xPJlv8LRA3K1GiQskCECNP MJZjuLuCgYiVGYqGg31lQgr2w89mDhn3QDho/SAx8rtcIeP2YLh8+cHMyTsPg/FVYhJ8knwyNGJt MBJLjfgD6bjyVzo2tnOwmh14FujII4JxwQylgezJsUrJA8E4uMS1BLBHWwqagnFUiap+7qjPYZnS cRkl+yG+KFSLBTwZ4GLXF/J76tgG4woHNtW4ki81GeXHhWCkvhS6oi77gWmD+iKmquyiE9r6Mth0 I7UTYwHQ6hPo2H6ieFHMow01H5eMkPiLURbqpyxCR9LOumLmO9aBBePjHqRt1EzgaryqYBzuZq+P N8GiACddgslFjMUHK+SCUXlNrsnU0QW1yMwCHt8Pafr3mau9IGMViw1OIUacl9dX2Zk6c6QcjNfm 9lYwbpNkJolhmb9j22DcaYLmvBIW+dsENWScAfgG5nocH78K3XcuEhNj3sxxgD7LKVgmP3yqznWi qMZWzY8XVphM3jnhOPJxkQH7WVzZhAuTmMCPKSy3Ehd7wfkDUUXHZjGnDlhUrXjaR0fESgjFEaLQ sSxNiTdH0Zl18KWcYN9ePXLGqi5NsEwTjDoqavjFsFTUZZmDe4/FKPNWx9BxFRZ78vQGRgEykT99 xIfZDsBhxSxLgw6MuwyYjiAR/Q87BsMdHDu/sI5H9UcxV71ewTRG5ONZFuupphLjxtt59jubsWLM JLxrPPxeWacV4+gd34Uuxm0PmEiMzzSa3MDvJyeVxLALNwNJh6TUropZXsWwFoyCLRZ4xBgz0gZD AGvAP8BaSis/1xotXRHjawnD7xvdUswjPrhm6NwuFGMm4WmUYJQ20KHEuCWi7Pg4LLV87JYw8KFP yKgnRFKx7DmBUakHCPaZlWq/BhfMl/UTzGKtn3YO6PiNe5jEXEcJxE0EGIgyHN9pJKFY/ZtfpxPj /KnOx6PfX/2MCIm2m9aOVWCcvmNhU4z7SfJ2CYHUV4m5DM7gAR224Dr1YiRzsQD7mha9s3aY066y w724RC0QFzbGxtYScXfsl89UvLGzVmZ80Y5xCoiXaAfiQW4CYGgl8LLymBdLVGCXT4S5XpTyxEhU z6rKyMU9LVqNV84FibIiTHyzzpkzjtZ6OMKDUYXNCCbGPWYRTDewCIJ4wHQSRUJH+DCIgDFGFuK4 pkGE1TFWbO2ysPtOAVoxR6BAnOc3hrw4qKJwsB/f/GyiGL1pWvjJ1QEo/4Jb+6lR6Axdft5aLJ8s C+YVr2JsYfpFm2D8xuEmMZRNu/z0hQqrYBwbj3h0BCwthqB8cKrPUvWFGcUoPIzRWbF8ZR8RDFuj OHS/WvFDxgfaPyv3D/cT8jyBJ+0Y6DhhDsZsxgUW5Ma1iI3FCFuOymB0EAMOdLyvoWP8NOC8uIee wGT4/e41jWIMikOHbxtwXpjfupsUflwQYMABM5YH4tp2Bkohv6AS6Cc3VwF98p3Te8EIc0T1Yl69 LTfDnNs+rN/SQsZ4sM/syxfjlEisz+Y7nbYYXhoLsAtNia4YB50oMSk/ju0bb8DwnHmW3sh2HJEI S9FY2v1BKlKMVZ2Rj8IU4w10JAzR1jiXvscCDiNs2iXOEXCkY0/G77bDjzGC0bsYd4OlF5GAMBKK 4R0Ir3zGOYQJRjDPsIK7000Z0D8RSPwcV+he8fxiM9ZiIG9JCRkCRF2GK+wshhM0WrtSi8zFvnq6 AnZsjfK/YlT85K0R50hqxIhBKKl4DIncjl6M1xfViDJ8EZEBAjruAE/n4qUSecRYhWMYaRvURRhv wLiyIkMLhRnc61SMC8x8lnU7Bj0GVDZHGHiiLhhTV+KNmBldyD5zrLoQg5fWgwyJMRrZLI/GemDc 3k0UFCMh802CMRrwoxihRfHA2IUi3LCD7Ckb1QX74cZBMK5/VF2wn0SPYtSKBjPMeY7nMykvm7Ca j6yWcZp4I0aNV+QzzKFmMDIN3yR0LP7hsNQYYOa8DjDvoC/G2wsV3/BgyJWKOGoEgorlLMINodOX O62SBRBuxLhzxJcLHZvi/CjGhbVp1SjnME3OxdjBTit0HHmi4qUYMSPtGuWQQDLcMsoaefFaV4rI aPVQfiLgfw3EG3boA8IpvgQihR4qPk3+ORD504hB4y9AuhiDCw8a4GSFXbioQVaW7UR8TOrFZoE+ 8+UpH+QVcWq8GHkqthej/jyqm6eTQgYzeSo7angIN9LhBRHNEEde9ZF2LCcrOzyYcCP0lehdjGuB +LMYiVfYEeVYwhejA1I4L50jO10uGOfI46/Y3juIg6Fd/JBpOZ80KMabHH7U1ODiVJkhSvIP+DED YdAQa+IMhKPBWYhN0lgFN+A2CTEGMsogIgl9jmgjRBlvWjXK+eVvMZSOxHgVF0IXo6hi6IgJRBDX ZhmnZEeQIyEpRCaWF2aQs7RUMu5aG3Z8cYQAwlVOOEmShri3N0KKcbOswycYR8BwKzHyyWhh4DJV FOPAXF6GgYsT8mAMgXRTjWzgsjpRjIlA+KmQe8hi5IlxYQOXH3IQo+g/vsIMjSRKjHFRAcKPI3uE EDGmBjEfdqT+CmcHg3NeJLNahsBiDA6Rz9Bl/TiJjF1RdkYuS32EOK2THmvM4fqLYGQ2ivcszrsc wvMPtMGYyrJqLk9EiNhGOe6fhKcxuL5LvPN4VzCOvuZt6k3cuBAsZ/0jo5/G+Q4/uoKDWjDuQyDa fGY+RIFWXjGjJJlKMfjFrEY5ukDoGHzyLBMJCqXhx7Mk/nlNRjk/4MGsFvk6IhHlGD6KcU5rnqXk EWeUjlx72uAtcb++WGbtyGeYowQKOpJsvr2nzNkupauIkWcJGeX85J4QI0BeCVGObUvZ8XK4mbx0 zC+MNsz9XT/IowywBhsgbxUqRrBBUDG/slWMjsIxYDEyuViK4YPiF+nIFjs5RBT8ymgDxiKAdLwd z6UGIqnSUEJ82+Mn7OhHzfbF6FI+Sj/yKE/pWDBQPPqRt4wWo9ZdgzLsW0ZUjOEIZwGDnUaWH7EN HwdjgVs1xDghqSyEAhdzinFOkGEDbIYNJhv6yAv87FtRV8z1EiGqZysKQY5CODDrctLdJOMLKKrh eVVWRmyCbmHKKEaE3mcpAVINzyInbZCOQxb4Mhj5oT4vRkxXDX7w9E4xDPQFfjDuuhMHkwhzNCfm 5+0lM8oZEYC4GbUWNcoJEdt9ySXj2mNfhhbLYoCPUu+oFuQUHvcoHZWavgyXmDLgSEdpo1rwklko GzoK+HgZLP94oYiPGukcl4BYfys7hyhCoxjdpu/CuZzBBIzsvRLzCqiVhZ2uT5ZRfjotCOFjqbi2 jlcBltTZNrnBERVAnPj2Ma4x5D0AkRrvg1zmhwYsTLm4FDoDneOZi1UeivNZRugMo2AsQ1QFA91X 3gMYuVSN7vqzsQ+MOX3VwlOZ+4LZ8TLvkJ8fohX6TpXxNMFNTrwI18N0+zbLrUBqQQzwnuxivBL1 J35Q5BSMSOc7YSD1MFBVM9J98UW444rb+6xnEhz5wZi6lJ+hjgAnhvn32RQRiNGhqoYbmVz+Tqkd I38GBvlZ3S9E0Fp2jC6qRqBbH2OI8EoRMZqqUxjpuJVSjHlMn0VLxhjo6KBOShTZSEeHEyMz7NvA QglWYGhRkY10XFIsXVInn2Tmp2L4H2XVw4244vshHDrJaws5DyzGvZh9Gca5zzCjy3phkmQMpZbv i+EpbdTYZXcHI0zVTgzDXvrAHxjAM/b6LGc6CBdgvNp9llRaxRgzkGXaICnz/ZhAryRYWy3oxkND p1c2l7ozkyDDp3n7OhjrujWccasQC2zV3rCFk0vmN8/DzrDFlyPE6B21iGHLvs4SKTPByuHkzL4O RvJecemBSQbAWIDTFYnyHx7rd4zyinhPW4yCMtR1KdYVNc2cfo6+YGYfJE+sVRLjHBjAuBG8dMY4 k1uWbH80/kinI4Pg8JWQ0cU8UoysxuxfqqQKYgTIaszIZNW5GGsl9TIUtZIbS2joxZjX8NIs8yVg yc+5vksyYqyB5CWBcWmesuANDhylIy9RCxPZzJ18lhmdWpBsMU8afpmwADFa5CVBlkRfMk9Glxtz PFwPiKlqXjlk3AWtFnQkO0PpfvodajE0i+1Y2PDKNDA6JnKGH+7lOrbYKxngYOSSauFCJ+cf5UeA 4eOnwdhaiG+AsZ6iFgwhTPrnWQ50BCL+bxOMdb4MRhCGuGFHIb90DCGexWkTlLWrBkMISyzB6GCW RImRduRlsI7D5FE1GJ1JX8KPCOfSO7GFtfokYQwzzmtUg9kPc9rhR79WDYYVriYbjCV11XBjquIR 4cxAQAj92yoDu1oQwrKux2IdmbzjoHcTkBOEmQHOLA+MxYr4MXRc0aESBFjjlcoa4MxqwdiX2me/ ulcgxjSn74z1SoOHGKt0fRcGOWdLuVZqPI+h01VBMeZDNSi2MGcC4i22WYOcU2owup5JOzcPE+Sc yIOx8NVXa5Bj6ihGtNlnWSOQn3PSsbFBzkQcjCGsjmyQc86TbYvGeDf7TPnEWMzeNsgUVIPUgs+P xyykgS5kin1iIyHyGeVowlvivWFFkQlyTKyDcSdLtTXIufYAHeq0CYKcA4cY6rZn4NQ6uphniMuO ZL7QpPU2wdKWzTIy8R3XofvhaqgYs9NazyBnlwdzGhQ6gpxLr2IWPdmEMc4eD0TLVZYY58IIkE4e Xe3Vwa7EhEe/fbfL+7nHrpzLjputVcwJwfQzTzXa5cH4YlkdmfHFLSQx5ng1FEGOuWcw1oEqnlHO Lp+vM8+wwiPaUYydvYhnkLPHW+TIKkk0o7VIh+guO6hsZilqhoYkLIMRotTCyUQWA6Vj7qwWDFKe i+izLNyohTlNBTbM2eGBvhOYQkaYe7XDZ5Mqi4HM0YyeKoEruGcpO96Aq9tiOHclNs7Z4cGY9y2W 7Ir9LZZo280MXfZ497xYUhl2P77b5cGwe1+jocsuD8bEquzcBLXLg7EsVcczdNnlwV5MeBUPvyQV AqOv/mhiAsZEUTVMh2YY0Kft8mBs+y0/TtOpBmME2c7wS0IIxIuqQZ2e2eXB0DovyM0JU1wxLOCr BXph0RiIwcxdjWLYSC0YzDoT+9V03uel/XBoCfRgbMGqMSkWVh2efJ1WjRn4vmflRDqmwWrsMN40 DozdRfIdMKZe8ygbGSrMWNgZuY+S7QJlxhTbZ4cmgwNDIQ6eRw1zrFFIt2MhdMzj1ZhMzDvO1Nhh MYMDf+ycDYiPyrYJ+l1CBhhr62SdbEsSbrYJ5u0qYTju64WORUVfGyuy1qy2CRYh1QKtBzLKlRtT +griLj++JBWT5SqmdZBTDPq2ihVdROMLFRjiPPvNuCiWnUhbNcklaxRjDNk2yIN4FW6kdjtPOsZY tABjH7t0xjnTYjBWUOP07KjxBtACbIdMsMwo2G9lBSn+CMQEATXAJgEG8oQn3MjXWCSKTfDgBFww BxMFNsgZvYDYiq3p6BAJBWA4Qn3UIOcEAOzT6OXUwhkAm7xuM4edQc4ZANhOH9lOZMKmDoxXsyCR SiJVzai2jxIc1IHRbyYPkLHmwqsAo/SjJmGYYJIZjLli8hfouPWf6Q77yIhZmzDEuJcuxhUEpTPG OTaAkW+Xn0HOsQGM19NnDXKEWTG6VuUzyDk2gBFHF2PSpmqMrCzxxQROCG2WiNWNK6bi9OtPTD7B 2Kys4fmR4S505EV9PwY5Bwe3vg1jviCDnIMDWPKCYkjA5BOMkFORjXJ2eLBExdKxre8bwhYuyxYj X1INJ5odWmmDD3cx+QQj5Vk6N5WEyOerLcb1u89irOoUM8qZDoCxclsLEOVYWgs7a4Rs1SAXq+A0 3phdjFmSTeBvzumKMVtTC5JJM6hiFCL4MvBVMuvBWE5RC928scXdclbEgzFlWX6muJKx5tFm48C+ DAflRlL23ikPUwsXm+Y94vzu24ux41RtDXOM92JJZhWPMMelDPAj7SShj3j2uXR5xhIHutKRt/gy SDu7kuizRCPVYBxilBw6ZvO+DDZcKIUIZpzzUYffZqKUFZDM+TIYfol9Q8b2umowmM4A4q4DdTNC bLzWAka59HmGYbYJ51GuffdlMCJvn3eK5iofGOGr75bskC1d+BkRunZh1YMBQIwFq0jMjyjky3BI jgH4iZAShO3s2BOIWm/fhZu8tZM/+v1tMfKn2JMf88FqMXLedAx+TAG3WL7Wio350c8XfyDhRMzO bcDyLV8x8rpoq0IOMWIse8R4YPmyqxiLUJXP2MU3UMWw4mIWjQnh+DEKj+arl2LujGtQsHz8kffB a+zqHZhnZWCX965qvwz0Q0s7Q9uTNfl8oBiMEOeD7eQjemLsgFcVA50zfzDCQlU20rkkCMajNauR jpFMjKzNAhIqQOjbzvzBqGCJKji2zBjmZ8UFIrYNcD773EQNMFYueG9gfhm9vPIpHqHZCZR/vkgj xnps7WeUcxIARqpebzGNchIAxoyn/AxzTgLAmJ33dTAAOD6xsYUh6sz0bOYI6kBKPStYYE5YhchW MgsAyo3+Ymz5ZcAAy832YnhIPS1raGrmGNsJKMNd7jkXY00pabtDoPdqi/lxtdiOCbV3V2NexBuv cuLt6ABGftFXZqLr6GCsYAYe8xnpHB3AWIpeulyMK8YMuyYw0vVR6i3arCm3owNkiJNE28DghaBi rODW8vzZ4i1Le9mFreca6BwejKpe6uGbNCA5PICR8CYxBsuFgmKknfUopPfGPDFWY7aNXPgmNjvn 8su1Z2LM02o+I53aYtoZCw21rsuLsQbQN+RLcHwgdWEDdrFcxSRGH0+mzbNUC6kGb56JWtQg0Hnv jhh9YJ/NLTViDO61spEu4wNeyOQgb8NI5yYAdGwilp+RzpVBMCpTawJDnZCrFZ0bmNFQahSMftl+ ZubjLoApI8vQEc9u4sogGEG6lrc3uTIIho61qJHObUIwZi6ZgtJGjj+LeZOAL81IF8dgPGBhbbCc hWWKQm+uEga6SGwaPT2NgQP2Ha/mudT7irFUXnEZcSz4FSNzquMxMlnRJ8bkMAszyGHJbzCy/aWz 5DcYC6PlR5SzIMdnqXKp+oa5dHmGcfa18iYYjKn5hY4QYMqorsa5dHnijXfCFesgRfRw1lnImt88 6kbxsLPmNxietY9a8xssC6E863SA3cNgpEGhA7Pm9yNlbDg0+wGlI0tGDTC2CSMedNb8BmO7PW8R zJrfYIi3/Kz5DUYePewyrDpfbcbKgxb8hoiZ+D5owW8wZuxtwCjnvJ9H8eYwI8hR7wtEVGK7TJsw L2M5Ux14T7PpB2a9bzDuNGoTvmXn/dDlvqM+a71vsO3vPMvSmDoQaTRs6bIVCoQHpJ9AZrlvyFhq Tp91xpktD+jYGckrc2JqnSfLfXSodgow632DYYqls943GBnh8rPeNxgjfNUwdjnvhx8esM9a7xss 1Y2K7BTN3B461snSacHYOSs2U1mm9lkQhIp6o/h2VgDs7iwfscqhts/CvAcs+KuR94a2GIW4vktX hBiJlcYw59DgkhTFToNRGqwVGOqYYucFG+YcGsDwrvR5nqWMVCswdDo6lx9JiVZgiJ3ZFgsmdDut wEjMvmD4GeccHMCS6PuscQ6EMXxWG0DYF1ULJzid44OxSqMWLidN1zDKOTqA4eeVhIhiDYMYc5nF skMlxvZMtSBCee2FGKvD1dYw5+gA1rJghCWVd3AAYjSv8YxyDg6sj+JCdQNaY5gUImWMqsY4xwZX USeNhRsldSrBNIjyp5jOGOfYAIaGfRXozddig/nt3eHnkBCMVLj9wMwgYwN2XcUMchkbsL83F2li 3hQLiTxLPtKaQzGSH7XgffqxptJlTRmoBx2kYglELfAEs4dSURWsFoQ9FqEjnSEug4Nhr8Lhwi8Z HMizZqOBBc0UBfsoG559O3QIi2zELKcIO0NcBge6SAvmGdZ5oRkc6EqTq4ClMJhISqRu+geWRXAg Fk+S+gClNFgy5irbbGqDxcg5q78hLoMDI8JkjqzzWhyMeERC5sQRzxiXwSETrQ6PDE/eQC8dkb2e 4mQugwOj0+xOw8/qYOiIj+xJDD/Lg4Oxx7bPZt3e/ZoND4yy1B6GjCynIhvlnPi718MEM+yMck78 wcg3OiYZ5Zz4gxFw+4aMcq70gU2u7zK8ZdhUujKeNpMCszg4GEsby46woBYETf+oXzhLcqUPjHcR 8dwRSAsEzVZDeToysxrJCJpR1s0Ed/7FiBlpAsx632B0pEgMZr1vMFMltQWz3jcYC155GWDW+34k daGLriTW+wYik9hmLfgNhkclprl3khkZjxJQ4vFg2RtxN9KvQdoqps0MAIx1yCWz3jfsmAml20Jn vW8wZtjVzKTHzJ5nWY1Zftb7BmNVb+ms9wXjxbOWFG3xJMa6YMzF+qxxzsweutntZeOJtL/QlB2y kkY/tMdDRnSt8XB+s18gLn/Ja4TM2z2CsWRRQ9GXTKuks4JWA9jj7PFAeGDGPDAmPipGcJ3JNhju q2J0wq5nZDPOHg/E2kRfj5HLHg9GxpV4Cx1LSipBN+/7N26ZDIAwFa3VGYXcwgciP6i4hi2TATDS mNjN7UTX+YA2o4F/LvUQS3WjajFquvnBdjadrZoatOztQKwslR1By3oOMTSucIYte7vFDK2ZlR15 kSoYVLtAD8ZumUowBLMdEvEMW211FpOfxfhszvP4HSkhHoiSYPVl8Oec1nBk+1SFCb/sjeedGeAc GMBY7al2RjgHBupDmLO1+xjh7LZgZC2V0AjnyGAdCR1o+FESrMYENiofBmMNRI0Nyl1wcpuanCeY 5oh8xrg2QWq2ojDYqwbtTy1XdsIdGcC446BqoKUr2mJuk4edIc6RgWodq/0iCvZNUTAYxQGLtSoY jP7TdnlfqR8VIykffq0LBmN8az/j/acwGIwVmprAKKdVeNGNe1YMIH0gRoF2DGNcRgb8xuisp+Gv ZJGhY4uiLejWGRlww+kY+n4GhqyZRzYjXMYFQrezw3KzMhhuOPqsG9KClcHBUsEmHb2VyuBglBWE nRGuTyJjtSLAee5HblT0tAWHCCf8YGwa1eUJcA5qYoz/y8664GCsrybbolULg8EYDVhMicTO41wP BKNubtu1MjgY8bxv1gjneiB0nbNTWWJdcBCYlZvxrUQMVBWOkdliGh9kXK4XG98y4BHNWyZLrRjh KiOIc4Ou+YDRW1XCye64J/GtLgvGEK3tKMSxLhg6xuUZ3i3OqUGN8I0+YKx7Kh/TGbdg+iwZpWq4 uFf5oFuI1d94O5BlwSEjGkRbMMuCg9XCIBYFB+HtrHAWBQcjYMX/qUyyKDhYiogVxPBmRk/RJev1 eRMWNaUG2kJMVmujrClGujtBv+XzPst7VzFnyhmQgSj0VS8kZ8UuuqJ2ggD6MVMcbtT5qheYNqwk rEypF8G/3SnVWunsmJBFspAZ3dLZnTA337bSi6qcPMpOeZUwumExFlHpCe2w0PnBu2Asl9Uohjfy AOmYeFQ84ptn1MTwxL4x41vZMfhvE7kUSTIWmvvG6AxOW8Sona3t6DRGTjG2/srO+EZvF6MoaZ9l O5pD8GKTB1ATx4pD23XHPdYjxHkil/1GgljDHnTU+aoGXYtNspiKUcOPIImxbdg2jHF9lOlO8gAK +yzzDRmJWlUzxtEdfJQEeR9Nma8YR7CWLmW+Yoway88y3zzLdkbfmoGLDi8d5zxqeUZminWKzXIZ sqTMlz6Oaq2MsrbRlyBGKt92jVxqy3A46zuWRVrmK8bCTJs1cNHlxVjAqXjEJk/4iDGULbsMkmI9 Xcbaf6t8hdCsEhu38FYxXmhiGXRW+QabrVYLPi3ylSz1Rnq3cYsOz/wUxfJ+nkX5zOT/hpYwD+ZK iI+f12uIY3AQI3D0lZNqWBEsNifMfDQVwWJM/Ot+7mQyPIhR4VUjGOLoKGIkIn3nhjiGBzHGzIyj lMymJFiM97l0bHKrskuvXXeHzpLgYHx7MoEPjEkE2wfupNR+xjh8U4w+Wel4g+6xiTFBXsyq4GDU XC2WqmDpGCeT+1A0nKpgsa3itpDY4j4xthPrGjosE/5g1v74knBsq4LFLHBKZzPM0ZwYuX61INem uwuxolgD2MX4kb5MlOsUA3apChZj+a7Go8u6giiGjcuOKGf2KsaKzGJWBRfzRiPFY0ixKlg6+mn5 Gej4sxgrdHVTAp1Xi4khU+VzJMOSYvTs5Wf6KkStRbU1zhG2xDYtptlUBSMPWWI7FoXkVgVD5xjT jNCCcyUQY4actwGWqmAxUpO8NbBUBYvNiwSyKDiPkqKpLFCqgqWaGkSgFAULeexkyFL8IsbQ3keN c6oPNEv91tZbFCxGCUoGAjCW2dSfxYepdk5ZfgYHkhXG9DRhmCOCkh/wGrtwCl2KgsVI4/MqwFIU LEb51z6bquBgI7GzuIwORKOeHGIoIznP6EDUIpTpeGAmCz7JtDgv0UMI1gWLUenVFoxz6fHQs8kT iY1zJATSsaySoMGzqQsOZnlRm0itL6uiDEpjd+NcejwZDKIMHTX9OhkqsmAS8Qx06fGYoge62fgm 0KXHs3DRCwbE4p4YsWfXRVLrKzf3wIdban3F2JLJeARdan3FcPG4NlhqfcWya6YWuiARWYzF6Rov ObQORVozi0icMEmtr9hu6YKxaaSv6Fvj2/Qa/VKMxKV+TO9KiMQrp5iARzO7lowOX5GdRJCBiu2e GnQp9hVjtrPipdg32Kxyeq2Hxb5irLNnOAdLsa8Y66iLkeP7Msg/Z10FuhT7eushSXQGOLAU+4pN JgbkvmaQqU8BSq2vGElvXwYjsbW+YhQf1M0Mc3Z4MCJptSXOWesrRh1s6YxzjAdijEEZVDiNlFpf MY9jxaUMXiYiYlMuBl1qfcVmogSU6gghP5wVX3HarSQOpd2zgSy1vmIsNNRpJ/gG65ZST1n1UTLx xB8w9tZUgkF9NvA9oWUiJMaaYc1J5LLWVywXt+p5xi7GazFebY1n7GLWH6wLPM8CfabzPH9PS6AH sy74g8+zKtJX7n5FRWTvqPYzzNGLJaOKqrY3zJFvkZrwsroIxzk164JhB198LrZXWkJUMG82VRXD HNN+MRZ9F0tdsFiWgEpnXXD4WRNVKGXBklFy02HZd4PRg/UqD9bSCSltgcsWKgkvmml1uLHO10eN cuT6PspEuZLgYHYJMbZ1li5lwSTAzvqXn3XB8DPL6myGZlOEK8YG7D6bumCxXsOjdHDxXWZFIC7E IUQLoMKO2cxi7CGrGd5q2QwWgC6xHyizzkKpC5YsM71i1gWH3e5d8WjqgqXLvKZ0qQsOxoLkNBGF xEjgM8x75pK6FvjROSkqDJ3joqODVzC4jC4/xk/rgsV2zQ0shcHBeneLdIzRYUdp0D6aayIkY/Ic N4MshcFi3D5c8QgWVr6KEZpXFAuDw49NgGIGugwPybD6rIHLqRXPzgkzDremMFiIrLnNEugoDIYd GdZU4UGXwuBg7DhEWwKdCRZkLB3soykMFmO0y+DFoxYGhx0bufNk6oIly70K2o7cwOWjYOaBxVg2 15PNuep5Hu+lLjjsOBGVjgGWumDpWFdMvwVLXbAY3hxvBEtdcDDWO6YNUsKw49z9QqkLVqafc/4J LqkLFmNmWTWMc6QpYqyTVlvjHAu1wWa/kWdT7Pv/9XU3yZbkVnaF+28UnIDMqkSxShqGBqJGzb+j b50NjyKZGTQ28nEFHA6HAzg/2I4bs/Lt0bJzHIGYdq7JGTrue0zOcu8xQ0fsG+Nhfvc9te+xJxxw j/NiFZNsX1NymvkVFeN27tGyc1IDMQnlDQsjpDzlsZdeaXQVPVq7y+1/9Z3cN0Z8sCZn6HhlsVPM 9tYM/uS+MUmN79qT+x5732Mpd3LfWD9tcG8jS7fqpGjWvBf8VOzO5eoWGbqC/1jfjI2d3jdGCfVd e3rfGH9pTbbsFI3GZPS/ctf6Zqbd4q++0/se4/LcPTJ1tCUx6tW9Iaun/68+0/UlLKyeFytA51vX PKbuPJtYopmxE/x2KanBd4sT/MYMpI+d4jcmgfBde4rfY2UaVt8pfmO2qPe4zFaS3xhHbvMl+2WB jhHXrgvYryS/rbF9eHX1Zb96WsgDPHSS39gXRbKMJ/mNSSh+tzjJb8wWzppcmEbye+wpSl17kt+Y ff/1fPbrpj3n7kl581SS/FYupca/+LDun4uy9VDi4F392ZJM3fkPlza8fJRyJw7uLiz8nphvdjOB F/iEeYolDr7qjIf1H+cvcfDZNyLuXkeWrvWByZOFvT7Iv0wbHJPsuFeUHyrNPtaeomux0wZXzjp+ cRR22uDY5xpipw2O7RSFLk0afNVxVq9LsbbvKsUZ/Go7aXDM3vCN5Rz71tmYGGeXGjfNzpi8y54i Q1enxKZ5celJg3Nw2v6+h8jOlRrELJrfLZIGH7vNoh42Q1dqUDmR3XUedtLgWAmWV18bjiFR7BFz +G/mesg+5seEKj0EZhr0KorjiopipuFakp0rM4hJj91MUy5l8DEplxvK2CmDK/cyZdAJg0PW2PVT dq5omGPJKfyqSxg89r3ZDF3PX5JtsryLn4t6MBu+e4rsXDEE9gINxRIGD71gBjthcMWkXC8MwE4Y HDsFSh1QPFfgj9EAb4wxZZ31dOwt/KUjEgbHLGF7FexcwuDYEtzlNu59WQ/beesGjFyfw1VIDeuS jFyBv7CBS/xVdrrgmET/d+3pgmNiyPU6r6JvSo5ZqN49Thcck0u+MOBSQ6UFMTm31ZeVK/LHjIqN naxckX+sTazanJUbYpYv/oROF1wxu3y7bUauyD+2zTvF7gj9ED3vd4eT+sYMyj1tNq5Un/Dv+1pf 5u6kvsfe1sVl+Ar8lVsyHzmhb4TSflO2phf2Y7beb41WLjci8gQRyCl9Y/L8a5uOTZIW4yruUb2A FF7H3hdbrj2pb+wT1pWDTeob495u6BgdSX1j1uUL8JQ7qe+x950udlLfGKduo8JITerrq9QcgD2F QX7T2Ft/g9j0aGmtlIG7GVuu+hUzT9aS7NvNdjm9Mq29VxO1Mw+7lkR34yT7drOdl/UOOyi/ntK3 ciQPX32n9I3dVtzqO6Xvse3pu/SEvqFLbKzYCX1jpsEeNvvWe4XeRzg2HU7oG5ONX99l3m62cznf 8RdtYrQrVt5F0L2hmNEqQo3JAd/TsgYJfWNWpa++E/rGbJOv91icpGTH2mWuyVmtm+7WEW157IS+ leOelApoqyifJWTG1FEhA64X1Ep1HX87Sgl9Y3qlHoid0DdmvWw5jp3Q91ix6tgJfWOW5djv4vmu r+zPr7J+TAI7UXD5PHOuromdKDgmE1U3tBF4ouBjVsShE7qEZLfrhYptqln9txMfOk1wxTg8TazY aYJj4rYGcyxNsNaxMMmkxk4TfGxflFVOzb04XuxSAbHTBMeIDFdfRk56LCYNsfsWy90b8YcZePfI yN3y4IK+W+m+muAGat1iFkgVrDJsrlDsVMExUqZ1UybuVofc2ssOVO5UwZVz7tseTG9TBV99pCpj BRe3OuiJZRZuQztVcNeam7uHkcM0j517ULFEwVcdsdJehbHZxoUrLcl7Y5m4Wx0wubF7VMO/XenK bfmptjTBVxt54C4166wJiuXoXt41YcFpgo/te7fYaYJjstFrcEbur72JN9jq4daOgv7Y2e7QSYJD TNxa15ZDjTP07ZNfgzNx5texBeRdeqLg2CVvu0NrZzE/ZoN/LbHupgo+ttx75VIFX32U8hsTpQmK +V0rw/mxUwXHqKI/dqrgYwW1u2+q4J/2Q0hov/uezYB49RuK2TizoWJW6tYpywFnuS0PTOZ7j1sg l/uPacluwcqlCj727pCRy/tXzGv/Lj1RcOym226RKPiqsyZutGfm8umVk8tfizNzVs+YZfKr70TB x97DZuX2rCLL9UlGrl0AV/pF490hI5cvELOO3MM+N6ttrXeKUGqnhL5ax/XtmKBanJFrvmMC7q16 mbl2AWILtCt3Qt+YbO9anJnzU1Mxy8LHTuh7rIWke2TmrjZbP3uKjFzTx5U23dYSnrENpqvNG95r dLOEvpXj3O/JNCorE7PIrAc8UEfRHFtU2F0T+l59S8Umuzudry1Yns9rcFFF8z02qWblTugbs4m3 W3jL1Gmh+6XInisj14SP7YSO2Cl9Y4VoK3ZC35DM+h7WsDTerjpK0XWdEZ3Qt3LvHOxkkSf0jZk9 1ynYSYmObdf55JOEvlcfL/juezLLdv1cy9e5V4FdkjjUbqLmQaf0rRh36GMpfVXH99oJtZU7pe+x Zx/Tj3It3cKA3mcFpzNNjRmTwLjxnh61zySObWxDp/QN2VW5PsZO6RvjFewpMlxtfscmp+4Wl9aG /K7ed+kpfSsmtXCdrNgpfY+J3u9ps1uchZhN692CuaL0vdZxgFrff2vliXwre9dfWVY+gXCq4Op0 ipAn/omdKjj2q1czc0X8JxvY+8jK3eJgwdrxNV16ouCKSWecc4Jd0j5hwk6mrNiJgk+ssDNOpqQu 4FdOMHIP3D6oLg8xe+c5EVyfJvhYH0Y3DDJyBfzK2eX72ImCY3I8G31ZuRKCMSYt18m1JwqOcYvX 0Zm5In6MG3lGk9L9thlCUn53W1aOx3fFHBBcrJgg/kTBipgRJ12JnSj4WPZ55U4UHLNDuXfJzF1W AWN4X3WnCa4YY/jd9kTBMX7ZniwzVwSJ2dNcOWbuwrHYfe7XBwHtllZKImg39dgXKmB0h2tcRq4Y ILat1C49TXBMmF0MEDtNcExHfHc9TfCxecCVO01w7JsGxkyS4EOaedVl5Ir4aYIKb6+fDMs0wTFn bX3sNMHHvoFnOhkMIeKI9Z1JlyY4JuJfPzVhjfJjr5uycXkDirGP6wBLQtKZWBmX17rTBB9bqNB3 MSZNHWpU5lPVKVm5NggxgtTVZwVMFHyMZbn6snK3OJROvt2Irk3UVzHJ+y0OGbnzBnLj33i3ZqcL lm7jKvgUebc9XfCxHZDd10enC455n3s/xXI34U3VwsauzUKVDFSOVn8DOTN3E9403yHnlTthcOVs 3e3RMnPnDrRCvNGYmav3OMvctWteVq4UH+bVbuhl5W7G56+fNG3fad2UtzDtTOAYG7D6dPzeRmbu prxFzcX3GJm5G94Wyn3407Wn9u2+/IItXNk5Iz0doOBgQzm/6qY8l30BXtcmnazYTtWLnNg3xAe4 DvAFHmFYL6Pk9N4txpr2IjHdVuuwE/vG2LHreOzEvjH2/lqHndj3mC6r97AT+8YIpu5psRP7HlvS tHIn9o2xECuXnevd8uLvO8m+LTyx76E0/d2hC27KK749mtiJfROMyt3d6MZO7HvM/vO1zgOl9o3x 7L+7ntw3xjdbT9WPN+exeTCqO7VvxZzOtYe9aKvVF9uB5n3RaVIccjzyWcg+8kzsVTHr9scux3CM L3WtMy5LxsXIuvYuMl3n13lRU892i5P7Vm7eL3Ji3wgPo5RRpU7se6z4K1Z4VtCvHHu5/jSpk/sm 1BUq7q6ZLsVC+v6qs74k940xAxsArUOtZel5E8zuFif3PTal9J+Z+VR4Necfy46dNLjrT1O3Ok8a fCzdUSwr12hRzmnx68DMXJ2PWTv3yJm59yxWkw2NzFwZQeVEdXuWDM2QcylKpXSLkwZX7O0rxU4a HDsXZeVOHHyMMu4178TBMSHc6svMFQXQa58ORozByhUDRDjId2VmrrUBM0nXqZm53IGU3gzOtS4z 19oQI/541542OGYafteWzz/0jVFWLtFajOuxsZyZa3HAGJKrjZXrkPdDxtNj9lCtcIrZ2vzKMb27 hU+MN9OycmXslTP59mTM3GmDMcvkV9+0wbHJCfoif9pgjBu55mXnqk78sC9pKjZtcKzzfnoVmbkW B0yguVswc0Urx8hcr/Myc+UElZPs2EgpIGhxwIT1a3JmzgJwjHfw7jFtsHL9wsSrb9pgjPOxjtes 0wbHkufXvuxc7/ECjd3Wg582OGar8qrLzJUUxMij1pQ6vNUB45rtcXtXrQ6xDtHdLaYNxsSuZ8Hd dtrgWAc3rdy0wb4BYRU+Nm0wZtNkzAA+bXDMHvu1LzvX0MsjWYjbCRunDY4lLukWmblbH3Jm3mOY rpaou7ZUyatu2mDX2gTZoLKk9AV99zAx92iZuZvz2AR23WPiYOXE5+dhdMjIqYMxG9x3LTZ1cGxx L3SCX8QwvbtCE/xizOINPWyC39hOFO4Ykwl+fXvj9veCsAl+YxzyegCb4BcTWtzTOgRkgt++2+kE gcpl6MoWxyaRik3wi9EPnxnGJviNyXTXe1jBO8Ia71kzcwX+GGO9JytxVxoq9rxdV07wi3mwr3UT /MZeMs0xIBP8Yt9wxKxtzQxr2FRPd1jOCX5j20KKTfCLvU332AS/sf++x6WSfGy10zMqNb1v31/5 IvielaHrMKx9k7V5q9z0vn269cJoJ4NM7xtL01w/ZeiK/LG2Sa4+hu70vpjM7a0Xyk3vG2u93rXT +2LUG3vfWa8i/xiz8+q75BdkT+y77fS+mFTvXqOwIZF0l375ug4pOb0vJlu+x/XPp/eNvYyLcl7f HsPP/KzJWtApwCTBvzKbyk3wG+NtXvM8eecAV05q7Lt2kt/YNzO8BMnUK9dvjlwXZL2a8+IWmxdX ndd8kl/sVvF/EdJ3oMk/lGXrseTBP9W5zwRCd2TwoZ1NGJs6WDECwfWq8Z/8o3Ks/8dSB191Mshf q6cOVo405GOTB8f2aQMTxdad/W+vYtYUs/141Xkzt6J1xsnJg11KSrpxkKkrMXjfYm64tI6Z2Yck YK7/LgMj/7HPODcps3TnERjXL5fvFlMHK2ekfbeYOjhmj/Xqy9SVF8Ts+Oy+mbr2ATCL667N1I20 wjSYM3S3PIh6vv5ky/xowNUmy/ndYepgtV1aYNdOHYzlJ736pg6O7XDGO3nOzvDVJ4Oz+2bp/Fnr iLvWOpbu1MEYM7P3k6Vrm9DXvdL0uwdLZzk+NHVwt5g6WDEzbEOAoTt1cIw86ZrHj+r4mK69HHGP wd+S6736XLBJmaG79cFSmXyqcnl5TEPXylZsDGTprkuxHcpWudTBV5/1YW3J0t36YIEuf7f6Th6s Ot+VfGjy4NgEdlWXPPiqE+x81U0erByhwjXF4SKTB8c2QiG7NU0MZsZY6a5Y6uCrzui5l4FNHexj bgvuDR9s6mDMXPluMXlwbCNKMepgtWXwtvEFTR0cq7O7qwZ0FHDl7ve/xu4o4GMvNapc8uCrj6G9 fscmD3atTdm1WDeePDi248+7x+TBmO2rr1yS36vP9uRX3x0F3H1tBt+4cO1JfkPLMDkxZIpf6Enf parYOT+N2JWcit2hgK7QP/Yqy8wV+ftMn0OwrsvMFfnHSA6uU0xWq/5P7IlHu8UdBHzsJdGxKX6V c+7JHiJDV+SPXRBSh2bobPyE3liEUvxeMTPqBhk2xa9yPOqPTfGL2e39WIrfq4/N+NidBNw9vmVV fVP8YpaBve/yXnkEsY4bWPOS/F59Yvo9BlvW10CVMxg2QjN0TXnsNg+6lqEz4A69THjHlFh1VWf+ ipXvFuxcRwHH2nrclVP8hpLAjU3xi52bNpbi96rr25gQK9dBwNVmafzYBL+xVvqVm+AX43bdGuXa BL9XHfHJWsd2vcYRqn7FJvh1qVTXV2yC39ii5pqS4PeqI8JZ33EtO9Sh5lnpN6QyXc14zKq6/mS6 TvAb80DX5GyXUxgq5xSM79oUv669uGrtOye8d4G90c12tYJ20sfkqRfT+w3pf/O//9F/CmOL6TvS 5O+LDk0bvINCNqhYuQ4NrkohxLkTLp04OLYfAbOfYda1E+AP3tYamJUrMYgJdfbArFyHBh+jos49 8c8TB/fHjmGOTRwc6wDKlZs4GOsFPnbfi4WEOq+6iYMxbstGla4UO9xt7cJuuGTmWh2UM3jXz5m5 Wx56njsPRwq/CHqPIa7Y+83MFfxfX2yTUbmJg2M9Y03OzhX8Y7yFi2mxiYNjC1Wh83gRrsRaXD7i VofitLdSZ+WK/ZXz33WeSXzaYIyXuafIyhX8x35dOmkw5MqvWNLgKyZndH1nFTtlsGKOlfuKTRkc +3XlhMEQ5+9GSmvsTUhIJqXaWp77KrJ2GM3Xc9iEwZge/i6dMDjW53q7NmGwa02lO8uh6qYLDj0h AzZdcLNwesjKTRccI0C56jJyucUYT/m6LmOcbu3YFinoZMFK2bpeg5m4kwVjHvWmqGLJgu9KmbV7 rdhkwcpxam+5wCYLjr3cDzZZMHYudA+bibOw1RKh7upj5FoTY8L53TcjV14w9vQtOXMnC8bkF+5h M3IrZVHZq+BCemV3pT3kG5yunCrYcs1926tg4zoIuPOo3pnMMhxsHNsR80nRvVkmrm9eDr11MFf7 RMGKeSerLht30734bv2ZiWsbQDFuwqoTHMinj71NXNVN6asc3/tmO3ZK39DOUuIqe59tA2ATHIYm 9IVsV90gxqb0je1XAG2hW3NvsrNqpWd6Exm4m+z9cVKt2JS+rjWc9iaycGX6Yhzod+2Uvtgk7106 pW/Ixs71na4oC8c268Pvyil9Y0+zUMydqq5yrM/6rni9TF92/S1PyrlNrwLrnI+ewjDqbNmYHYTN gIblzXasTOvKTeqrnMG7WWGCnNQ3Nh9UdVP6QpQ0X1NO6Ruyel5tZu9JfTEr5iU9ytmc1Df2duqw SX0xg+u7dlJfrB3QGmfxOqVv6L60DE3oC13maMWm9I09J0K5KX25Z6ciWLkpfWN+QOlexa3yuskI f9kCmTfyjsYTto9m7hdRTumL8Rn3ytilk/rGmJBX3aS+2D4C7NIpfUPyKvdgzOYpfTGqsU1ZFvuU vjEJ86suo3Wz3eR9jkDp11P6Kmdjoo76s7T9mfhXliPxyrLx2O3qhtiIaw77drpgjDptfZ99KyMY e5thLk0XfNVJfHzlpgtWTmJsg499K63WtXyWTdwMXBnB2EuQXMZdOsbmIX95Nvmy9cbmsTdGM3BF /IoJp9ZbGbiHnjV3ZbLgq+0ptFkmmzJF/C6VZV+Ls3CF/DHJmOuBwjihReyzoa5NF3z1cWWvp9qN 6Qjgyt22pWHVTs7pR2P7QuR+q+jEmJj9gOsB5VIGX30v9obOhWBztiADJbyufqvQtRebLlhdQtya CyULvqryr4amClaKk7k7Zt4EjLVWEvZjUwULBC0+twi0Ayi1rro0LXMksamCCxj1690jA3e+vyY9 t1m52+pXTHp+Dc6+ne/vaXaQJi9MYKX/uwVR4Doz+9YWAHZ68x7DKDpZcOy1JPNmV7BiHJebQO34 nio49nYUsKmCMWZ1T9uca3sw1omiu0Wy4KtPcLgmm+odu105WxDfPfZa26lbFN8eekePVsweyIpl 4HxMEnOo1N5P2/TtAMRe8OzayYKlDHxxtdu2wJYNjO3sYI9qVDrsI8YV3mNY2E8WjImNvntMFhyj +7hHYzw6U6VraTx2bTbO04b2ORTP1EPmCmCW6zN7mLHrZcTeIopNFox9bjmWLPjqk5bYY7BxQoO7 9n3w2D0m9XWtOOm7x6S+mBm9t5GRs4vXfTlAe4yMXBMe49KvmzNyD+38DHtvjIyOrJj0xYZyNo7h OZZD0QvPbWvCl0vaR3axaX1jL9GCTeuLfXvOWFpf9WHfm8zKtQWA0RDs0bJyaX1j+0kFLluraW0W je7L6dgqKz1Q2zJxTXhlPqUINqkvJjpdN4kBTuobe4uWckl9r23flBJ4lFivOrnn9RwTd0pfjL5h zc3GleCL7dBLLiS/vA98MQv0d21S3yv3vrmv3C1RiumQPUU2rgxAbEcPViypr+oEp8vbQXf876En pKRanNRXsSc5TsjYXK0Yp2zPf+2s66p2vqByd/rvMYmp60/dc0pf5ewq7v1ntkrvxb5F1SspiQqx fN8t7pSqY6b0q25SX+WktDbai8sS98T2o+f3448yOFefjOj6rkFetK+cgOeqMxducxS64zYaAZkt b6piRIV7CnO1WDZ2ezf/wsa7fqrgrywbj91ZwXf9/Uqsn8qznLT7pJj07Z4uC8cnP9aBPJqT3vg2 RmS392FoKFGwYtDTVWETBZcEfx4cNlVwbL/H0bVTBWPEMLdyYXdUcPW9I4UrN1Vw7AWK2FTB2H4m t2ITBYd2JAlfyFrc2zW1yib2EAzqqYJj8lD3YBm5XAHMbLlXrtx9fHXsZW+xO/43Jgtzm0/YVMGx ncvTPaYKxnTPd4+pgmNPTemDgDv+t/rMw9WXlbuVoXB3efM+JuCA2EKxwN8oQKYKjr20A5Yq+MrZ 973YCZsqWDlbKOfQYFMFY9Ije7UZuVscMA7f9QpXOAF39/WI13kZuVsboP2g5f2C7KmCFXsbsH0+ konpSsn11cbGFYlBZMLnwio2UXDsuT3YRMGYMHVdl4krHRXb11rd9Y7/jVGUnsOOTRUce0FMH+l0 /G/lfBZ6sxSbKjj2to6wO/63crlx97TZuFscMGHpdUrxZn0sAn4yT5dOFRx734Fgd/5v5eRT13n+ udPKju1nYxl4Hm+pQOVkVm/q9U3WqYJjL7uFpQq+cryMa10m7hYH998H01U3VbBLbT3saTNxCf9i Qqt7Cm+55BZEYnOxmEunCo79d7GpgjE+1S7Nwt2Mvx7bLQzpU/oqJ990rcvAcT+7BadyfdfsKt7H TkjedMzIFfDHvk7JyBXwY6KtdV5G7qYy9nVy60t9571/7yIjd8MM+15ZRu7emTHD+7zmZeSuB7Bv CLQWv/okl/dkBXJF/O4hqbj3k5HLN8doJde8rFwRf+wbomeOegzj/Ml6fV05qW9MrvNeRmZutzAz rnWs3Cl9lRI2bD3KyhXxY9Sya12WvIg/tmMPfWHDGhXxY3Lgexms3GUVYk+YoNykvphA+qtvUt/Y vuGqvkl9MdtUm7jM3IVYseLcXiR/7DqloHpeMjStb5v7z5PApvXFfAu1ecuF7JhRo54d0ourblpf bOcSdYdJfUOlnFZsUl9MPLF3kROdYx+zrlTOB+HT+mL7+dLQpL4hmqFXbFJfTNL34nblTh4XenE2 NKkvJuNyjik2qW/sfQiBTeuLEVed/cGm9Y0lAV3rpvWNPcRyndQXktVT6s8i+RPgdTLA3xdl5O8A gVsbXL0OhaYJVuGLiqFJgkMmdUMPO0kwVYkQcK3Lwp0rgLXPsGKTBKc+qbPHJgmO0ahedWrmyHoX bdsuVOwwh5MEx3asjlyazrilwc14uFdfNu6WBuwO2K3YFMEulU3bbTNxtzJgT7Gj3CmCFXMm2l5t Fu5WBmxyNKUmCFZMPnu1ZeFuYcD2Ra4MpuXqFoaC8a/Y9MAuFZrcrFVsemBM+Ruf2PTAsVIz9VNL RGl/jL7lZjc2PXDsq+3UwMRH7xSXLpwcOPbiqw42OTlwIqUyeN3Aknty4Fi/hT42OTDGMn3lJgeO bUVx6dTAkN3JW5+wqYFjTwWJTQ2M2ZNYD6fwKfESexLuzqUpswNZqPb4zFvm6ph4/lqXeSvcr1zf U9Vi9q2MTkx3fddODhyzBL1ykwNjkky3onSGScmKrpWw++qbHBjjOG8Y5z2V+I8tnu7cIKmEQ5Il Zxs7cyhFqlKeZi+WfTs5cNK1bF8tzr6V+I+VnBybHBhLEXJPm33LoYul3V+5yYExe35r3fnPniKW QVy5CYIxI3tLSgauVGCsn5OonOjgBMHYZ3w6w6T0WSgJx4pNEIyZWRt52bfC/diT8bt0Il/MIeh7 tOxbuvtYUc/qm8gX4wKu97JvDYLY670MXD0KMXR7txm41zy6hzUvA1e0X7kdgcc1MaeHhNobedm3 on3FRIxfddP4kk3acP+qm8Y3tp+kqrppfLEmXA+ReUvQE3pbRBihQA8hiN8B8pUTgj62nRRoGl/F bDxv4GXemu+xtk92i2l8MaZ5rest7672NjYvDJDT+ComP783m3nLEYg9l975IBP5YifUcwtsIt/Y 24bCJvLFvJ6rD5vINyZArnnYRL7Y96UNNpFv7H4pvGK6vXcB7TcZYhfGQl8sCXEJej2NhOXvnQ8y jS+xb1/MNy+waXxjNGHXksxWO30Yh/+hSXxDZvdjk/hitpfOD1DdJL4xubNXbhJfbD8U1F2n8A29 L14cEHJxNySkuVcBiYV7COwJgbEpfLH/tZ97EhZyy8vvxfYDVbEpfDEe8E1abArf2H5D7M8s/OXq /1CWicdSA/90nzfnoYmBoZduh6YFDi3EdpTIpMCQzwKuVzJwUgoRE7VXkfNYGhB6nydD0wGHNnIY t1MBJ+J+W+5KpQJWF/bWLGwqYEwwtBdR/HaLAvb2EJSbClg5dvHW1A40MVWvPltFe9nZt1sCShJs 47DzSwpyXCpxv6cSyNy5v5gcwW6bfSsHGNP4e9bsWznA2BM9q24yYMw3Q199kwHHvn7KvpUDxCR3 1rwMXDnA2NRmzhuZChjiBawp2bcC/dhbF5UrdY8kHr3GXVDdi8DeBrtSUwFj0mxnBbGpgGM87Hft 7WVpjdTIa0nmrUA/9tIfTg25TFJMAmEPUUOL9GM708a22C0u2oftNIzYVMDYp07ALnUUsiRfU7zR 0wFhrOUts4olA77qZtyQiYCVktNeQ4y2E3HEnvfVkSYnicCkdHeHpkFnj8UYjburKXg79pivwNYp Td/2v2NvBcy4FedDUo/fbaW6rrbvJAx3nQa4Yu8DFmwaYOztr3d6iWlztf37ExphSYCvOgmUNa5l +db7xtD2DZWbBNgXKnSCa3AWou3/mMzgPRhLUoYntrDclRMAR+rpJnFWrpQ/dtmIWLatp4deyh+i wbA5hsl37vEzbecDYIWYu3Sq3so90Ylrp+rFvqMisFS9V5/Y/rt2Z/4qJ4jaOp5xOycAIwV595iq Vzk7DT2GIxEtQVT3tU+O5qHTZiBigrqkUhP1xqbWjU3UizlXq5Utlqr3arM/30iM7cxf5WSLanFs qt7YspixqXpj+0o6lqpXfVavVrSuzbiV8Y8tDRybqhezsu7JMm5Xyo5vS2qlUvVebVLka524phxn tUndrgOybaVsMSNhPSCcasfmWAHW6kvVe/XZ6Ni1jFuihMpZDHffrNtNd2yWnFGxaCXxU85+3+6R det8j5hc2d2DdTvD2Mb+d4eJepVyCtb6hL06US9mKuz5M1id7xFrB7gWX65gTzExNUffGyjpr5jU 6W7qz9P0Yr6FWeMyWh3vEZPvveq0gKb3rhVY7GGzWzff3StNQrfVj+e6Q99zZbY636MP80TAr9iO /I21679Lp+nFSGrXPAPkjvyN7TBV1v3nv/7yR7EdSec/lf1/mpOJu8VBdLtvIWMJgD1KeY1LecYm AMbs96wHTUWHKl85BmbtzsadfVTOTvi12xJwjor3J+TlgVXdFMCqM4L2KJanUwBjouTdIhNXEjDG BbrqrIp3QDAmuVI2QeaMjSsLGHujxfJ8AmDoFK4rNgFwbKmYLt2Zvxifcc3LyBXqxzzju23qpEM7 UF86z+sv1O9zz622sRTAY7Sz17psXLG+cnzWjZdsXLF+bHnGrp0EGBP17rbZuPP9MSPs1TcJsHJy a+vkjNw5/9j6MyMnydMdZET2ejJydVNpDTuy9QkbdwLgWHca84r2EE4mWnVs3B35qxxzssZl48oB YsKREgDVNwFwrB3UsQmAMX7wd48JgGPe1Cu3M39j1o6rL5f/VgeL3TaN2N88mT0aG7j2MXPW8ZrC kdyEycp9zLx/1U0A7JtiSpzv0imAY6kBarLwq4P9HE5f/nAjLzt36wM2mZdMDye1vH/l7sKs3K22 b2GvMlbuDvxVyJbvXpgILl+2C3kqBeyVO/1vaJ9Xh6b/xYijstSx6X9jy8/EduAvpo6vuul/Y/ui zllxuucmvMxG32GqD5v+V7lzFce4/hkNbJqxyk3/i0m2XVswXtsew3/v0bBT9Sqm3u8WU/XGZEJ6 WsWm6sWs2N+lU/XGdPE1zxs9VW9sx6T/1TkiU/X26frlYkIT9YbmDsQm6r1v3l9TMnNl/WN5bTUl M9eEx2RbrnXNQe8/Mn16pabpxcT8NwUc+TFNb2xpW8kFz9h8j5HPXXUZueY7Zn248aTcNL0xqcpr SUau+Y5tBVBqit6IHcVX2xS9GM3Iui5L0XyP7XSuWnL+KyTUuvHZLzuepDfG47rqMnLN95ig7RqS +SznjwnfPmbbcg+hh7/6pulVTur7Fij3mKY3xhV89Vm/DCinLXBkPjZRb2w206UT9UJihJuznVHS FAuR415tOVqF+rF+EKWXmN0quYfZyN3YYbfS5hzjGt/TZrcK9ZWzUqwl7Nad9hvb2cvVN1Uvxi35 7jtVb6zDtLtvhitfN7bzNmP+by/DGjSdReyChFBJqV06WS/mII2vusl6Y9tv69LJejEJy679XRD/ lf3p1leWmXf9JMAxN3/3ngQYm1dcsSmAQ19PF+cVAsQsBHdpRq5EIGax3cBl5E4BHJsSQnxudFYs +cW7MAGwyhAO3mMTAGPvrLAunAA41t5wfZWFyxeIPdcCM/vFO5gdkJXLxuULxNJ7dW1GrrS/P3wd uyWpQK60f+x2byqWAviqk+tdN2XjbmnwRxZ21d3Wl/J2VvZkNSBfoDq2PV11SYCvuucnQ1MA+8Om w3ozC3dLA5Yt6Q5ZOIa36rzEhxIAX23e4XfpBMCKeYffpRMAxyhB37VTAMfM6Mfa8+wO3K6vuimA Y/S8V6wpfGuDvt7uon1HC9ytDdiCJygBsOq8pj567CEsRCcAju18u9gEwLF9XB6bABgzcb/bJgC+ +gSomy0t2KUBlfM4ezKLfXri0BTVvuLitN3awJ/a2IESAF9t76uxiknKNp5yuwQWWoxNAIzZ3fzY FMCxZ5WVSwJ89b0N666dBFg5L+q7dhJgjDn92CTAsX05KKPExDXHoE2x84CKdiBr800xbGf9xgjW rsUs3CmAMfmL6yflUgBfdRLM15/YJMDN8X2h0V0nAY4t+I5NAhzLHa9XsnHyzmSeLPaqY+LaZjr0 NSUT12yXjtn5i105VW9sco9Yqt6rzaK0PmHj+lqha/9jm5WVm6o3tr2V2FS9mG28G7RYqt6rbwkk Kks2rohfMSHjOiUbV9o/toNdKydE3FNMvRay43yVddbvPX6hqvfZlfzC3TQLl5IH4wrt9RQ2M57H XhSiukS9V9+ThXSLHfbrWvZq98jElfbHzIp1ezau/F5sXzNJYOaN7VUk5b/2ZeNK8DFYPg3ZsMjG jTxf3ZWT9SolRfWVStarddh+MrM77LjfzN+ymbEJe2PTjYrGrCQl+GJi+muJzpDuv/qeWVBsp/0q JvC9JRWbrhfz2tcBWbhc+tjzVZWj673ajIlbBKEd96vY1dHwNFJP1xvbUU0cQVbrfAEd9kKuMpvt A3YLLuhum9U6X0AgvKFo9hYXVEq2dD1s5vdx27H92Gh3SNZ7tbHmX23Jeq+cCXpdYpk7Va/qzM+9 2KxW7xX6RmKLcJ8s5/9MGfZbA79UeePxlWXgy6jfwcCx+2DiJzYFcCwleH2VhSuMxbbNEZoAOJRi bcUSAF8xi8o5dcpNAKzck0V07QTAmFzT3m7G/pYGbB9ps2pMXP0CsUl5TdAEwJgc7N5uMVzZwBhv 5JrCW7K3fozs7GM7F1g5itSvvgmAY987Kogr+Y/JAH7XTgAc28a7EIqNu7VhTu+1Lxt3awM27Unl NtGsJd8kzcbd4oDtG/+K7bBftxCeng+LTQCM+dJ6L6M4LsV/7Bl+5SYAznXfPmH17bDfmOeteW0R nwA4tnPRYhMAYzY9bvxh/IfmVUvnlkKsPrks0rmcwBTAsX0sXWVTAGNPxx2bAji2z3JiKYCvPluN X32TACsn2r4hleDgJMCx5de9a+aubCBG3H3DB+MQNFYwC8E9bEbuFgesDbM6gJFLtKyYIOqmKTQJ cIwe8hVLAnzlbKTfUFFuGmDlrNNrHht3x/1ikpa7RUaueD828bR1wyUXjGD+4ZqXlSvkV+62R2ue lp4IOCZ3fG3JyhXyF0ju47zKdfTSId7lq24i4Is3b3eiYhMBY+9LcNuVrFwbALFtusUkSXobbLvv q6++BnUhf2xBoGITAUOyX+v4zFwRf+z5jcpN2Btjq151E/Zi8jE3qZQ7YW8oZ7kOsOjc7gx2ctFY ayCfolu0tXrVZeXOrcderls5ybA9xc2uXTtlr2sZpDWZrSiKqL7LyK7clL2x6WS775S9Ma/52ped O7++vNKWLuUm7VVObLPey87tnTk5ZO8xM3dz3pg+1Sl/j5k7f0AKKZ1MLcnMnT+A2cq9p83M3ZTH XuZKuSl7S4hwet+1U/bGJI0fm7IXE7ZurGTnbmU973M9n527KY+9mZGZ003m9HvZWbmbPk3zU+9J 1u1FVcw43bTIyvGzjklSXzt46Z0KcOyFP66drte14t3NWmbudL2xfU4nL2GG3IyXVnqxd6LxE/Yq 97Jv0IS9EBt2PcdyXXyFTDBWZdP1Vmritdh0vdg7SiQ2XW8st72Xw3J1znUPIb24UZzluvne4roI q68JTtdbue2yaaPbNd8l3V7jLlS3aEU4+tdNGa5me6m5bQp25SWHQ32Z/+dhfGq7PxQdmgC4y59v rsYJgDFj/N14+t+QDeJ7XBbu9L+Y7Yw9bhaudSGWKqpuKSlU7j+2n1qPTQCMSZyt+8pOlfuPlaHf tRMAYymbhi7xHZngstqm/8UkIPdu9VpbOtVmPb5L+2zp9L+xbVuaYXz41gXMYn4txiYAjsmh9LTY BMCYkXqtwyYAjr0BhE0AHHvbEBjvZq+twzJW3aXqlSL2ur5TavrfjrR+IXBfuJ3+N7YPxxitFhzz J3Y/HRLi2vQQUldv/cQm/43NN4em/oU4Mfdmsal/Y57mGtdCX+I/9iIC5ab+xRw/eatH3zyWUAi9 lFyfS578FxMw71mzb4X7sclF7B5b3Ar3YzuYNjb5L8bQ3aKFTf4b81Kueec9NKBMkX1c07VC/+YP xn1rpGCT/8Ym/rRPqL+L9zFB1i0qfYZbZBNK8Nml7Nt9/4ddlmds8t9YMq6xyX9jfaU5NvkvZi95 I4V/W86le8iCrHkZuCKADl1fAq77cgJ7DMyCe/Vl4YoAYvKsd18W7s76xYTPG7VZuFscsGfNfJ19 cjjFhM97Gxk4q27V8bzWywKpO+s3tqM6RVgmf64AJoFy6w82AXBMwug1bwJgTJJq3cLAnaw39m6R gcsVCC252pkAd9QvJLRe52Xgyv3HOnW9Di0wX0O8gA9N16uYTvvYdL2xUyt25WS90A6usLXkBdx8 h5bdu0Mcyu4pZVXYE7BuJ+qN+e34e9IMXNm92NdLGbgy/xgrtN7MwN10L+X1JncGrglaxuuOhK4l U/XG3s4hNllv7HvX1Xzz3QX8kOuS7NtNeMwm5jUvA3fT0b0ktV+56XrV5wO6je1MnAlcW2T813eZ uJvx2p5goW7PxJX8V86SthnfcKun5LdstLxbTNkbmwbPZGfhbsZfHmzVmV2n7FVO8LXmZeNK/sd2 8EXXeqG7x062Dk3Zq5j/c9YHm7I39jIo2JS9mA/DrnUWvxP2htj/Yxmu5h2npp8i6Fkt1ifsjT1d QacInbA3pvC7dMJe7DmWik3XGypptOqm67UT7O5rCct5ut7YC3RcO11vTAPeLabrxehk7tU6+GPC 3pg5o9xvQ/k/lGXlMfmEHyuUdX9TGZoIOPb2o7GJgGMTMXbpRMCYNPc1EUsEfPXJdFynYhMBK8f1 vUfGdipwrOMBdQ02EXCs7YxYVq5kIGYZuhUTmwg41ibPyp3KFmLmdwtWrlfdpWK2e0kuTQR81QnH b+XCJgJWjhn5rp0IOJacabeYCDg2oZR5wcyVDcR0467NzpUOjL3FtgNLvvax2mtydq6QXzn5wz1u dk4+MPRcLpdOBUzFoHv2GJm5/P/YC+U7saTNtJjtza+6VMDqw17y16kgkwHHrE33aJm58oHYPXeP m50724eZiq9cOuCrz29krfskQE4H7Frld98CuVsfsJ2UUX0XFirmeTZYmDlv66oTWX1NmQ5YOV7A HpeZu8N+YzuMUZ65Bbn1ITu2RE0Hmdhkvvq0c03un0vZK8cZ+th0wLEX4HdqicX8ylnR9oY8Of/w 6mMQPnZCYJfSwn9oQuBYieB6LzNXRjC2MzBiCYGvOi/3KzchsHLyKmueYS1cdS2TP1V6104JHEt7 snukBFYfVpptbEpgzBKxLrWC3GG/MYvklcvWNUa9lbd/4gCRpMBXndTbXmS2LgFR5faFsLy66ZTC BpNEXJOzdcV8sadlUi4p8NXH6K0p2boyfcpx7jdfslOkBsfSWfQYGbu2/JTTFesqxk6zrr73yZL8 N2P3isl33qVs3Wl8XbrDnSs1jW/ohbdYKt+rzS7POi9Td6bY2H8ZGOWm8nWtAzzXukzdmQnzZj/e 0z1S+V59MnlrSrauRJ9rhXW7Nlt3sx57+0VOGbEYrdjLrECpfFXHU337YB1G0gfFx/zrPW227mZ9 ibl5gMpN5dsvYz3hBJbK9+rT23sZGbub9fJwT3TSYSSn8nUtE79uydoV9cfemlnk1MkekJ9OWFOy dlfo5T47niQdb4Ws+1+hiXxjbxXI0qUPDn0z+QLbxpMFcR3nBJBJfENvScYm8Y05NL4ewZL4XnWy OfcE2A76VU4W+dYtbBJf7Hbmdu0lkUKddDSUxPeqe8dPdYud9Kucjb57MGwi31j+UtcyXqfqwGhK bkxgiXyvPvuGN3swfkqzouyddXPXTuTbb6/xBB6byDe2AfBb+Z06/7EsW49NEdz1T6OFXY4MSh6T iwJNERx7myjYFMGY4/TXNRm6CwOw98Q6zh6XJ4FeBq+TS3pfsW83vJNLsvrHdvaGoJ9huLVB0/f1 TGyHAruWhOziAGya4Ji7Xc8Y/acJxmz47o1k6Bq10FuVnTSSJniMX3FPm51rF0C5rZdKTREckdm9 G2TlbmkQYj9tS+eW3Dm/yr1PtKT9WblbGjBrw7t2imDlpJ4+NkVw7JklB4jsnN+Y/aRrXWampcFQ 2u+Scqq0oHxg7BvMWblbGgzDN5U7t6TdOJrdTMZ52tgkwbHEor3urFzfBZ0OeEEENklw7G10Yzvo F/Nd8jovM3dLA9Z+Z/Vl5koIVq79qbHLYPEO32qu1A76VUqAeb4xNklwTJTwrpwkGLP6r/P4oCcJ ju1D/O6aJPju+r5xtdllgVNNLTHc9iKzcqUEY2sbG3eCYIR+ZH2SjbOeVOoOCuoZ2Li2yEO99qEp gjFJga01hU5tAcRM61dux/xiZv+9WBauX3Sr2KebcFrIFMGxb/CwcKcIxsgGNtozca8/rXqbFSLb UwRX7imCO7ekl0y6Luu+JDaWIjhEqbQuzsSxxRi/8b7TZOD55wxqTBdvvmfizLeYlNKuZeJSBGNU t695mTgTPiYNd0/LwiXyDYmxv0tP5Ivt0wlWk3dudY7wCPasGTgvBZM0euOEgWPCQhICa1v2zVSN SUePsW850Zjs5xtO2Te+SEzocPtHzgk5iW/sJZihU/hCxsGcX+wkvjHao3uvNUoDICHitMuKncI3 xsPYo9Zjwv6Y2f61LoUvJFN6H4qQITNxss4xmajrOczwuIcw4TfssFP4KiexMNuAncI3pnV3C+wU vn0xYVG4xR07hW9MkHXNw07he19W5AQb2tgpfDF5nUdO4RuR7+y1OhEkgW/EmYs3c6AT+GJEuHNy sRP4xtjmmybYCXwxocwtTtDpe0O3R7w7nL43JhP33eL0vZghuhnm2tP3xuzrfK07fW9MeL3bslnp ezEvc26fk0NO3xvjCdyMxVj9inkRqy2T9RDJwDqOyep834rZC70FprNJLJkxisaJqTqbJHlvTEx6 SyfmFV/nvdNZ2UHvzitRzJw/w/PbWP6V/flVln3HTgwcI9Fdc3JlbQhikqB7tuxbywJEzrahkp/9 v6utFXNdlXlrWTDtd2RJNzgpcEgK8it2UuAYUclXW1LgH4wZWTsybi0KkDTFx04JjEnPLTHQKSYp gWNvnwxKCFxtzjeao4GdEBhjUWbwsYTAIQPvu8MJgWP2vffS2LY8HYwGZO3NtLUqQNKOG+5JY+Sx YkzcrfedVyJzH+NR/GIJgauuV/uVOyEwlt/51XdC4Jimr3klTloXMEmVtSXjdl3nkIqtgW57SmDF xEevydm21gVsv8jFZ+m131OIbOZcYwmBa51/26qFcWHvKQyJN7kzbqJ95YyJNwcK4LzQ2NaEywp6 EWyXneU1t+iNyxojJN2jZtvU7z+s+C5VRwkIrF2IjcNMm4+pY+KnjSaPkxnBzJit2U4XOR1wjCnY bXsrXKmYeGUvNtPmZ+YwT7BMumtPBxyzv/Pd93TAMWmPrSlGr48uq8+asljaGSF9eheip1wxMysh cMxG5apr8ptZmBznQqaOK0kIHPOPGxQtTSwO1uOsyS111ojYbWa05GXdPFFMimo9aoVtUw+zQKyn sm5ygCF+yapjOHIXY1zzjbGs213JAD/7wYLlyiim4e/dZt2YiFgp0BrCup2dNdS4aF9tJ+1V7NMZ KHbKXt9q8tqXx8KS9v7ExKz3yhwRcmeVxi514xbYSXv7ztOCd32MnbQ3ZsW9h8W4ZlVnvZ+8AbrT fTFbg98dTtsb0p33dhQ7bS8m8FxvYml7q47l/VUube9fYqbxPaxTQ07bi/E93m2zb15ezIy9N9sh JLS91WdR/C5N21sx4py9sQ4hSa4Qsz958wLLMQuxWvU6krS3ytjXRR/Yne4b8+b2YEVkzXZrktm4 u2bdhLwxgsF1Z9bNT6zE7Fuso7JuzXbu+7c+dQhJbyomO7bWZd8suLF3IqTXYHYx2Jjlfq5RB5M0 GUtk7nCvip24N+bJ1pTsm/yTz5Ad4rDow+kfiXt/YjLAu202q16xUhN33WBU7tS9MbP8ZgV26t6Y 573ek1eyQ1p1QoTFWord6b4xy/vXlBP3xsyL9ag0mkwhxPhuCXAgSOLeqvP9tTWzFyQkS9gSMxg/ djkETAy73Q/lTt0b40TuFv6WNqw+RuV64LcW3vWnBP5VloXH7jBgzApSNPQTOyVwbPsEyOmAEd7L whLsdMAxCfcNg0ycNw7Ztz4fzBEjdxQwJsTYsxl0fRMY2inaHE3LWh44xKyuR5sQObR1j02M/EPl TgYcy4+67su+GTaxt+Wv2KmAISZqedsOLKECrphSV5mVKRFw39N/+mGlEgGH2lK++ltKTcLYt62j 2B3ti0nhT6eEnQg4dqak9mbgNDMmbb1Bxjrld2OdjLJSd7BvxD76ha5OEjkNcIzgbuWyb2JiLCu/ 1mXfzIZYbsDVl307dEO3O2TdhPhKmbHz2jqrpORgjBO94ZR1E+JjshPbr1PuDvaNedHruqwbgxj7 hPtOCDkNMNbW6erLvAmeYtJua13mzbyMeSEXRHRYSdJEzNPM+8JaURDHAG54Zdzk/2K3R9WTFRa4 GWalnN+PnQY4Rnr23fU0wJidwbk8HVaSBjjGdd9oYt1Km8fe10odVkIK85cOiXCHrR6sW4fjxN6u s2InAQ6ZI19tbe1DJcD2YrNuuf2Y2HQPxrolAY55r9dgp4GcBBgjkJitwO5o35jI8urDTgIc+z45 w04EjHFM701ApwEOyUHdQ2B3tC8msT4VFXY5lpgs1j0FdiLgmGXxmozdi7XI06PehHVCSCLgnsJA aYYhJ+uNlDcZOlUvlMJhj1ro1kwvsddUXrkL2WL+7XQ2nVRS7IzJODN8K3ey3pj1a+3NujXVMdvZ 3z1O1osJrN6zZt2uGE9+k7NLsrj+K96ZmfW3HjPEMFtyN+38fbLeWA7UPe11mcfIkDKh1zzd3TG9 MfnJNTnj5g3EWNI9btbNKMfaE7wMzA3Bpjsm3bZyDemmO0YpufqybocEr1uwXXrCQ8XoLu7wKMmJ PAePgfGi11NZt+Z7mVQOwzU568brjFnA9hjWr6QksZdIVN2peiGKifdys25N9xyT55S3LCcYxvTx GzwZgzXZ7iRz2Itkba6QTcK1LXPVbOcM2fTY4zNX9WxMZm8jMXPFfse87rMTKjtVL+ZG8+axU/XG LovRTfM5mu0lV58fhJ2qFyu5erE1dqremHzG+imbJYEQ+zeKtes7NmtNtkW3xd2lp+pVzP7dvvXE TtUb050bUdmsok0Ti7uwx2WzUvXG6Ghif2bdT3CXA/0PZVn3fO8UwE1gh2nvPixcCuCY6HKvNxOX T48xBbfEu/YUwBh148JQ7BTAsW/wiXl6IS0bqSrr1Axccb7lxfjZCxG65RjHnjSxKLCuDLGHZ+GK KgvcMHkwPVl1WTj3j/FRbqwwcLlxoc9dUOz0v5hMwRsuRenl/zCiuL22TFz5P8zCtdfLxDWCWjbf DrbqTv8bE9xupLFw7W3E5guXGEn9ixi203dgp/6NSZrsptm34nwLuFT7x079G+Of70Vk4cr/Yfuc Um2XzUNs/f1qx2l/YzYxvtpO+xvjQm1hyL79+18YJrmiOynL5nhO+/859jdj4zo4+2ZhqBydwXfp iX8xt1ow4NoT/8a8iDPc2Il/YyzXHiILZ1nD/Fr7whflTvwbk2BssjjAzowwb2K243uzsRP/YrSi l4qLMQ4RqZRaHDn5b8yuc8MkdjptzHp09id28t+Ybbrv2pP/YtajixArd/LfmA6oB2In/43xXJsT TC4LwBPALKdnp2Mn/405h685FjNbr+d9qHOvNlYqVjGZ3TsFLXQC4Nj7/ezYCYBjRCnrKCu2OKNr xbmXhqvcCYBjxn32MXYC4Jgc75qSITPdMd7ZBRciRXWb7jHr71fONy33MujJzqBV7iTA/Clb0bec xcoeh4zoNS8bZ77HKPFaAyp2ut5cMWrE9V42zoSPiUpXjpuVsDdGRbI3lI0z4TGz7b2h3EAzPmbz L4shFmLkDAisj3c2Chi5hL0xX+J97IS9Medl7O2eW10x9u82UKrulL2xE1XuFqfsje1IOrvCJp0p D9nkuXUxdsLemAzHNS4Tp2tDlDrN0Yqdrhfjt17qJNYSVjkB9OWiYqfrjVlTW0Bjp+vFMp3rqKzc 9R2Ddvmfip2wt2JihfVxRm7vTC7ndKgUFYyclUM5du/Nqozc6msbtrlcuVP2KtfnmVmg2El7Y6Lz a11GzuIUOjXeip2yV/xAtXFprC49ZW/M6ruRl+HSUZCEbp6UfzGAjfNCDyqRdV52S5QSu8m/cue4 xQR1e7PZLYMQM3pOAV19lwyLceXvIbJb/gw5leFryQl7sTQQG9wZLhFhTOJtT8tyFWBh1vJbf/lO 5uFVR4ZyiZfQCXsr9pJssRP2Yr4tuDU5dsLeGBvSmP0zK98Rd38sO9Z4/el6CZk122MlAsa09DyM rj0RcMxCvrmh1xMBx8Ri64betUQgJgdxvj3zYOlpecAkHTc2MnTMRUxuaGPNcOcqQx2W/FWXCLjq JD0v+VJ1JwKOCe6/cicCxnyYeGm6yp0IOCYgXZMtMx1hHHv6SQsrH6zlwduyyq0pmbqWB4wIdIPD Ip0MOEaWsoFVKOcrN4zA+42OMyIVA84idovbQokVNTT8GK5UGBBD9zo0Q+e9xsQ81+K+vSiHE7Pl ei3pW47yLZjNtdvWsuGuc83jGLfnegXjk/y1+qTYt8hhJwTGeGGb+NgJgWOOXfzumxA4ZG29TlEs IXDV0S1svcVOCBxLhNOj+YLohMCxfbsa8qquOru5a1x2TipQKVu6v640ze8h0s5dR7n0ZMDKcZP2 urGTAccsOzeisGTA1cf2bSRjLZUh+s97Y9DJgGOCyVsL+04tGTD2PkAWUlrjZAJDoozdQSyXhDJm X2B9wswV1skaCfm2FLj2ZMAx/fqVSwb8ExMGrynZuZaHFrbnibj2dMAxufj1QHZOp0Buv85j5rgC VUcAs9Zl5Uyu0PtNDSbh/IpjJ/ns7WTlzBDlGMjLLcfSAVedLr58Zux0wDHmYE9RKqoZbxU33r+m nA44Zrdz5dg5KZvq8+7e02bn7lLjb1cyc2WxlDKJ3qvIzIlDY4K79VP5SZYREwRdSsbGk1bl2WEi va/cSXtjt5XX02bmOPSYZfQc71jS3up7wiqTybqh5tCLJ2Kn7MVKRGxWVPUVE2Ds5WgoYW+V5Tdt AmTkWsmqddva1XbC3piuWYML5Zpk2HmuNdhb1j8/pUi/1ajzDW4EYAzu7puZqz8xa/Rad9NB8zw0 UczuYXql4o3xT9fxTWBrL8RWf+iEvZhXMrfObU/YG+PQ781awQgfq679gI28zJwXj5GAnoXsEA3L wSE5hI+drrdifIRdmplzsAfW7s5XLl1v13Js97BZOZMhJL2wBys+80IxXtvFvGwiy3V9Z+W9vF9I QtEgU0zxr7qT9WL2KV85hitZb8yqvZZkuLizsdt+7f1wjkrlY+VO9xQZrvwBTFS6e+T1STLE7PGv yVku32hgVsMFMupL1ls5Z37MOGJFqCEZhr0fhqtzojAih1t72fk/PeyuJv5jWXbeOSMCQEOymfl8 b+zOAo7RzZyfjZ0E2DYFIzLrhZ0EOKaiLZkZOS5djIO4NhZ3tTpYJaSrzpV17UmAYxq26cHMJQGO 2dO8ceUEkSTAEL9vzhh0pwHHvMarDjstbMyl95awkwBjTMveMHYS4Fgz0YuDUgB3C4nLUoyhEwBD Bv+9Nuj0vyHPeu3FTv8b4x/c8zsrJP1vtdmqPumObvVaDVLI33dPNi75b8juzJpbLNfiYG32Jm/i ujL97zGD4PxY7PS/lTvhbg0uJZdBs9Y/Ow2d/Dd0XxisWPLfK5Zhu2fNxrV0sRys47rkcpz3EDby 54M4UqQMR8UkMfcU2bjifnZIKHextmIn/435xxve2Ml/Y6S8689sXGuD+E1s9ZVL/nssZcI1b9cc e68/E2egu5IbNz++Q0uEh8ek7TZysnHW3cq1L/hqYxIQF25suvLkvzHn5n9Xnvw3Jp2xLsnC7VpB 69c0ZvIeQWywcMepIKf/dSn3eas5dvpf7G/2JPf4rRIF/ZjTBPcmrCYM77F31KbEtKFV0K+rPw8R S/8bug2WCt0BvxHm58IJ7NS/WIehbTRl3wr5sUuD8eyLwO4RSMpnBbG0v5WSojhlXOVO+xuT5tzb ysAV8GOpWhtf7Fvpd8j6OGOBnZw3ZhLeYoKV8gqxoF9tJ+fFTMLFhIqdnDfGVq/jMm8uicl6rTrm rfDIBrSc1F5h1q1wHyrLUeMYt9S8IUmgr7ZT87Z1nTLrxgjr5qOEyhkj81tcm3bsGP//VXdiXsXs 4b4GZ92uFE/rtgOs+pbvZrpiJvCth9iJebHs6KoTK5lDldsRvV15Yt6IVPyGCNtWhIvJPrznyrbl BmKXd+9ZM25N9Hxka+81uKisiV7OvA9aKpdxy+Jj0pZrScZtiGe0BmfbmuhlzV+84NIT88akz9d1 2TZzPiaZuUGXbWui59a/RI5rrXD3GERJc/CwE/NWTsp9LzaD1VS3gLb7cY+RwWqqY1aVj52YN2Yj 8mMNTsSUu2fNWhn2EL3lW9VZq473jUm07WGzVs10TCS6vmOt0vLSZIjN7/Nx+SLWSv/EbIxdOQd/ nJY31pCtwdhpedNz2HV/6MS8IW/iXq1iJ+aNiZS/cilaIC7Q+WO/jeFd/vdFf5h26HS/Xc5C3MNh p/uNCWk+drpfzLbSvEPlTvgb4zLdYHFEyAl/Y83be5LeQplA95LRX6szbvU91ik19UEDp4UBEqGe HcNO94tZ2haXYZmgkB3Ae4/Q6X5jVvHvDqf7xew6iBV3i9P9xt6ZrYS7FpMWBuydQR073S/GBL8e sCCWRIgVqlx9Gbc8D0znrS23Wt9jiEG2kndYSdJf5eS5tvpgSZJClq0bytBJfzHmdhbKcSAn/Y15 U+vQrFuZwPyetnl7NNatM5HSK72zbmKn/Y3ZcV+38CY6Ku50TfuUpXIn/sVETcssYif+jUm573Gz bwxsTDC70cyH80pDwoxb4TutpIwzZrWaX46d+Dcm27rRnH1jJ2Ly8ecaKddeVYx+4vYA7XWxb6Lj mITFxmP2TbAfa6f13kYmTjXYO5y+S0/7G+IsrKeycMxYzPbZniIbd3fgGPznWlfkxiYqZkmYKey0 EgHYMSvhnrZg04RQzsbbbZXxadg4fkKsCq91bFxnXsee/rdyJzrAyvTvBYnh2tWMad9e+CUI7sk4 3fdtixAlJ+t/Vg7aGoyd/jf2N4b3BkYpknsMLV76qANLEgArJkRYjhw7tWTMRtgeLSvHgMSM+TVF KikBcD87Zjhu9Sqp5b3EhFV7XGaOaTj24l9Hh5yoVzHr5BsXpfVM+piH2AvKzIkeYu5wVsO1nfAb emojsh2v3nqPcRS2BmMn643ZPFuHZud4AzHZ06+6O+EXk6Cbb92JJaleY5JgezJ2rhEc475u6GXo 9C3GVbjfIJM64Zdz32MG3R4jQ2eMxGxsrPeq+5piIX7rd3aOGVSMV7rtHtUl7D1GlbHHyM6Z9MqZ BO/dejEpe2MveOkgErmEQ+0Q3tAz5PpUQTFSzjfMGtay+TEe8J4sS2eRjZUau+HTRGzO6wn5xa0D Wbq7qzVv/qvbnrC3Yi+IddTH6XohVmQNxk7YGyM7vtqwO+E3JtF9Iw87YS/RK/HMOgo7YW/M6Ln+ xE7YG7OLcSYIuzN+Ywz22RbstL1YHXBPhuXChr7X7fyP0/bG+Lw3VLC0vZUTXW5+Y6ftjdE/rrrM V3O+IdjqbFFW7gRZGKXhBQRQ0t6KqXVBKHbS3pjG3YqEnbQXE3vM/cFO2huT2O9pf2vr/7nsXxh7 Z4X0cV6X+7XatZp7LfP/g9nEfC8pSyeiiZ2wqCdh6ZIBx4yXG+BYMuCYTcnt0mHJgKuPWmQhBXYy 4FjbrqvuVMAQVcV7l1k6Bi7GRt7869QSjmbVtQeznsnSGbiQ8bLqGLo8G0yU8YZLho6fELP59ZUr U1R1lqlf1Z0OGEvduV4puZA9wKz937V23a73OIuLZRwVcmf80qGbm/dg2Tk+coQH/dCpgEPPqLuu EAiRON0SD90JvzFL7Z7+kl0eAbPhtyHFyLUhiEne7j1k5FoZILXtCRi5DviNvV0h54ScBhiSgF9m CMt4dKken3eB3fm+MVN0t8jI+doj9o7P9NMcjFwrQ5Zjsp5YW2Ch04H1qrNxeQNtBthmuy7Jxjng F7O3On9IuVMBx7ile9XZuLwBrEWt6jJxOQPM2mdusUTAVWeQLOLBTgQc433uFdZSTj1mx2kxpXKJ gI9p895YfWsMd6vFO0qdBhjiI84tcXTIaYBjlsE92A22exXGyFtnDFQ+btVZrJccdO1pgGM6bE/W FGxlwMi+vvruhN+YzYB1SguMRGDfUuxARHbDAJFvCZ0+pY6y/FUbxDLtSisscWms7t+zZuH4YzH9 /pU7DXDMPsfGYhauye512l7eYM+yCYhiPLSNlCyco8JihtbGLBPXB1SYqPkNgUxcs73Qn3z13i0T 188qxayRuzYT99eq4/2+UcHEde5wbAesc89uF/EQF3BPxsIl7VXMLud2XJRL2lt1oo7dlIFrVykk SN2ryMA117G2/65x5xzfQxgmm4zO+bAW3KuQSdikxW55cC216SY3VmoC+rZ+kNP2hvZtmdFqMgjh QzMfyAl7Ed183QGdrjf0Jbix0/XGmO17Xdgd7osJJda9nUDS58Ox5xt1AEnC3pCE8y7NuHEAfYQk 7NxbVc5u7k+IX3J9BJ2yN+ZRro+wU/bGZPZuMGHXXZiMyYY1dqf7xjjP321P2Ru7LVVjuENJOv0C s7GwTUPslL0xPbdrs1iCgBgDsPZlsYxTTIsXsXYCydVm22CNy1xx1yokfPoqu8N9Mavpr3In7I3J 2d34UtkJe2MWzNWXveKLY3JsM7EdP1LuIyYteCsRdsremPzzzRzslL0xi171/ca6//De/6ks646d tXY9mcIbLqK4JjzELd0Iyrbl/GOGyzmILj0ZMEY7shyKY0RMJIYGo/PZ+2XdiDqP8QbXDf69k4CV s9U559+1t/zHbPh/154OOEYItnek2/uJe0yqefPMtZJghiQm3Fv7jIkOA469k97lYcvaHdL569UG dguDx5HBaNZ2eknn+4Z4eecgdnqJfPIxof8630xPCaychPEvdkrg2M58qroTAodsHV/YiZ0QGOMy LpbATggcE+1+TTkhcMyAWIszcJ61YJvgJmfdgSEnBI7dHtXYCYFjftBq7zE7aX3FOmNgQzcDZ4rF 2Lw1j4HLEsWefXOL0wGH6uPrqCxci4OR5fO5sSxcqwPGjl9Y7HSQ0wHH/LHOy8J5OwZ1fV2Ds2+8 xJDgZgMl++atxLife1iuaDJgzH7+koGdU5IMOGbN34Nl4ORhYqLK766nA8basz9n3eEgpwOOWag2 8LJw6+T2Ote+IqE1mYPxFiUWrhuQc9xRl02+//v/ASfn6V0KZW5kc3RyZWFtCmVuZG9iago1IDAg b2JqCjMxNTI4MQplbmRvYmoKMiAwIG9iago8PCAvVHlwZSAvUGFnZSAvUGFyZW50IDMgMCBSIC9S ZXNvdXJjZXMgNiAwIFIgL0NvbnRlbnRzIDQgMCBSIC9NZWRpYUJveCBbMCAwIDUwNCA1MDRdCj4+ CmVuZG9iago2IDAgb2JqCjw8IC9Qcm9jU2V0IFsgL1BERiAvVGV4dCBdIC9Db2xvclNwYWNlIDw8 IC9DczEgNyAwIFIgPj4gL0V4dEdTdGF0ZSA8PCAvR3MxCjI0IDAgUiA+PiAvRm9udCA8PCAvVFQx NCAyMSAwIFIgL1RUNyAxNCAwIFIgL1RUNSAxMiAwIFIgL1RUMTAgMTcgMCBSIC9UVDE2CjIzIDAg UiAvVFQ0IDExIDAgUiAvVFQ4IDE1IDAgUiAvVFQxMyAyMCAwIFIgL1RUMiA5IDAgUiAvVFQzIDEw IDAgUiAvVFQ5IDE2IDAgUgovVFQxMSAxOCAwIFIgL1RUNiAxMyAwIFIgL1RUMSA4IDAgUiAvVFQx NSAyMiAwIFIgL1RUMTIgMTkgMCBSID4+ID4+CmVuZG9iagoyNCAwIG9iago8PCAvVHlwZSAvRXh0 R1N0YXRlIC9jYSAwID4+CmVuZG9iagoyNSAwIG9iago8PCAvTGVuZ3RoIDI2IDAgUiAvTiAzIC9B bHRlcm5hdGUgL0RldmljZVJHQiAvRmlsdGVyIC9GbGF0ZURlY29kZSA+PgpzdHJlYW0KeAGdlndU U9kWh8+9N73QEiIgJfQaegkg0jtIFQRRiUmAUAKGhCZ2RAVGFBEpVmRUwAFHhyJjRRQLg4Ji1wny EFDGwVFEReXdjGsJ7601896a/cdZ39nnt9fZZ+9917oAUPyCBMJ0WAGANKFYFO7rwVwSE8vE9wIY EAEOWAHA4WZmBEf4RALU/L09mZmoSMaz9u4ugGS72yy/UCZz1v9/kSI3QyQGAApF1TY8fiYX5QKU U7PFGTL/BMr0lSkyhjEyFqEJoqwi48SvbPan5iu7yZiXJuShGlnOGbw0noy7UN6aJeGjjAShXJgl 4GejfAdlvVRJmgDl9yjT0/icTAAwFJlfzOcmoWyJMkUUGe6J8gIACJTEObxyDov5OWieAHimZ+SK BIlJYqYR15hp5ejIZvrxs1P5YjErlMNN4Yh4TM/0tAyOMBeAr2+WRQElWW2ZaJHtrRzt7VnW5mj5 v9nfHn5T/T3IevtV8Sbsz55BjJ5Z32zsrC+9FgD2JFqbHbO+lVUAtG0GQOXhrE/vIADyBQC03pzz HoZsXpLE4gwnC4vs7GxzAZ9rLivoN/ufgm/Kv4Y595nL7vtWO6YXP4EjSRUzZUXlpqemS0TMzAwO l89k/fcQ/+PAOWnNycMsnJ/AF/GF6FVR6JQJhIlou4U8gViQLmQKhH/V4X8YNicHGX6daxRodV8A fYU5ULhJB8hvPQBDIwMkbj96An3rWxAxCsi+vGitka9zjzJ6/uf6Hwtcim7hTEEiU+b2DI9kciWi LBmj34RswQISkAd0oAo0gS4wAixgDRyAM3AD3iAAhIBIEAOWAy5IAmlABLJBPtgACkEx2AF2g2pw ANSBetAEToI2cAZcBFfADXALDIBHQAqGwUswAd6BaQiC8BAVokGqkBakD5lC1hAbWgh5Q0FQOBQD xUOJkBCSQPnQJqgYKoOqoUNQPfQjdBq6CF2D+qAH0CA0Bv0BfYQRmALTYQ3YALaA2bA7HAhHwsvg RHgVnAcXwNvhSrgWPg63whfhG/AALIVfwpMIQMgIA9FGWAgb8URCkFgkAREha5EipAKpRZqQDqQb uY1IkXHkAwaHoWGYGBbGGeOHWYzhYlZh1mJKMNWYY5hWTBfmNmYQM4H5gqVi1bGmWCesP3YJNhGb jS3EVmCPYFuwl7ED2GHsOxwOx8AZ4hxwfrgYXDJuNa4Etw/XjLuA68MN4SbxeLwq3hTvgg/Bc/Bi fCG+Cn8cfx7fjx/GvyeQCVoEa4IPIZYgJGwkVBAaCOcI/YQRwjRRgahPdCKGEHnEXGIpsY7YQbxJ HCZOkxRJhiQXUiQpmbSBVElqIl0mPSa9IZPJOmRHchhZQF5PriSfIF8lD5I/UJQoJhRPShxFQtlO OUq5QHlAeUOlUg2obtRYqpi6nVpPvUR9Sn0vR5Mzl/OX48mtk6uRa5Xrl3slT5TXl3eXXy6fJ18h f0r+pvy4AlHBQMFTgaOwVqFG4bTCPYVJRZqilWKIYppiiWKD4jXFUSW8koGStxJPqUDpsNIlpSEa QtOledK4tE20Otpl2jAdRzek+9OT6cX0H+i99AllJWVb5SjlHOUa5bPKUgbCMGD4M1IZpYyTjLuM j/M05rnP48/bNq9pXv+8KZX5Km4qfJUilWaVAZWPqkxVb9UU1Z2qbapP1DBqJmphatlq+9Uuq43P p893ns+dXzT/5PyH6rC6iXq4+mr1w+o96pMamhq+GhkaVRqXNMY1GZpumsma5ZrnNMe0aFoLtQRa 5VrntV4wlZnuzFRmJbOLOaGtru2nLdE+pN2rPa1jqLNYZ6NOs84TXZIuWzdBt1y3U3dCT0svWC9f r1HvoT5Rn62fpL9Hv1t/ysDQINpgi0GbwaihiqG/YZ5ho+FjI6qRq9Eqo1qjO8Y4Y7ZxivE+41sm sImdSZJJjclNU9jU3lRgus+0zwxr5mgmNKs1u8eisNxZWaxG1qA5wzzIfKN5m/krCz2LWIudFt0W XyztLFMt6ywfWSlZBVhttOqw+sPaxJprXWN9x4Zq42Ozzqbd5rWtqS3fdr/tfTuaXbDdFrtOu8/2 DvYi+yb7MQc9h3iHvQ732HR2KLuEfdUR6+jhuM7xjOMHJ3snsdNJp9+dWc4pzg3OowsMF/AX1C0Y ctFx4bgccpEuZC6MX3hwodRV25XjWuv6zE3Xjed2xG3E3dg92f24+ysPSw+RR4vHlKeT5xrPC16I l69XkVevt5L3Yu9q76c+Oj6JPo0+E752vqt9L/hh/QL9dvrd89fw5/rX+08EOASsCegKpARGBFYH PgsyCRIFdQTDwQHBu4IfL9JfJFzUFgJC/EN2hTwJNQxdFfpzGC4sNKwm7Hm4VXh+eHcELWJFREPE u0iPyNLIR4uNFksWd0bJR8VF1UdNRXtFl0VLl1gsWbPkRoxajCCmPRYfGxV7JHZyqffS3UuH4+zi CuPuLjNclrPs2nK15anLz66QX8FZcSoeGx8d3xD/iRPCqeVMrvRfuXflBNeTu4f7kufGK+eN8V34 ZfyRBJeEsoTRRJfEXYljSa5JFUnjAk9BteB1sl/ygeSplJCUoykzqdGpzWmEtPi000IlYYqwK10z PSe9L8M0ozBDuspp1e5VE6JA0ZFMKHNZZruYjv5M9UiMJJslg1kLs2qy3mdHZZ/KUcwR5vTkmuRu yx3J88n7fjVmNXd1Z752/ob8wTXuaw6thdauXNu5Tnddwbrh9b7rj20gbUjZ8MtGy41lG99uit7U UaBRsL5gaLPv5sZCuUJR4b0tzlsObMVsFWzt3WazrWrblyJe0fViy+KK4k8l3JLr31l9V/ndzPaE 7b2l9qX7d+B2CHfc3em681iZYlle2dCu4F2t5czyovK3u1fsvlZhW3FgD2mPZI+0MqiyvUqvakfV p+qk6oEaj5rmvep7t+2d2sfb17/fbX/TAY0DxQc+HhQcvH/I91BrrUFtxWHc4azDz+ui6rq/Z39f f0TtSPGRz0eFR6XHwo911TvU1zeoN5Q2wo2SxrHjccdv/eD1Q3sTq+lQM6O5+AQ4ITnx4sf4H++e DDzZeYp9qukn/Z/2ttBailqh1tzWibakNml7THvf6YDTnR3OHS0/m/989Iz2mZqzymdLz5HOFZyb OZ93fvJCxoXxi4kXhzpXdD66tOTSna6wrt7LgZevXvG5cqnbvfv8VZerZ645XTt9nX297Yb9jdYe u56WX+x+aem172296XCz/ZbjrY6+BX3n+l37L972un3ljv+dGwOLBvruLr57/17cPel93v3RB6kP Xj/Mejj9aP1j7OOiJwpPKp6qP6391fjXZqm99Oyg12DPs4hnj4a4Qy//lfmvT8MFz6nPK0a0RupH rUfPjPmM3Xqx9MXwy4yX0+OFvyn+tveV0auffnf7vWdiycTwa9HrmT9K3qi+OfrW9m3nZOjk03dp 76anit6rvj/2gf2h+2P0x5Hp7E/4T5WfjT93fAn88ngmbWbm3/eE8/sKZW5kc3RyZWFtCmVuZG9i agoyNiAwIG9iagoyNjEyCmVuZG9iago3IDAgb2JqClsgL0lDQ0Jhc2VkIDI1IDAgUiBdCmVuZG9i agozIDAgb2JqCjw8IC9UeXBlIC9QYWdlcyAvTWVkaWFCb3ggWzAgMCA1MDQgNTA0XSAvQ291bnQg MSAvS2lkcyBbIDIgMCBSIF0gPj4KZW5kb2JqCjI3IDAgb2JqCjw8IC9UeXBlIC9DYXRhbG9nIC9Q YWdlcyAzIDAgUiAvVmVyc2lvbiAvMS40ID4+CmVuZG9iagoxOSAwIG9iago8PCAvVHlwZSAvRm9u dCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9UWURHT1krQXJpYWxNVCAvRm9udERlc2Ny aXB0b3IKMjggMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDQ4IC9M YXN0Q2hhciA1NiAvV2lkdGhzIFsgNTU2CjAgMCAwIDAgMCAwIDAgNTU2IF0gPj4KZW5kb2JqCjI4 IDAgb2JqCjw8IC9UeXBlIC9Gb250RGVzY3JpcHRvciAvRm9udE5hbWUgL1RZREdPWStBcmlhbE1U IC9GbGFncyAzMiAvRm9udEJCb3ggWy02NjUgLTMyNSAyMDAwIDEwMDZdCi9JdGFsaWNBbmdsZSAw IC9Bc2NlbnQgOTA1IC9EZXNjZW50IC0yMTIgL0NhcEhlaWdodCA3MTYgL1N0ZW1WIDk1IC9MZWFk aW5nCjMzIC9YSGVpZ2h0IDUxOSAvU3RlbUggODQgL0F2Z1dpZHRoIDQ0MSAvTWF4V2lkdGggMjAw MCAvRm9udEZpbGUyIDI5IDAgUiA+PgplbmRvYmoKMjkgMCBvYmoKPDwgL0xlbmd0aCAzMCAwIFIg L0xlbmd0aDEgNzc3NiAvRmlsdGVyIC9GbGF0ZURlY29kZSA+PgpzdHJlYW0KeAGFWQl8VNXVP/fe N0uSCTNJyB4ybzJkkExCIICsJpNlAhiRLbEzNEgCRMImgbCIRRhUtgCCfBaVqqBtFbXKyyTiBGiJ Ym1FET6ltKIVXKrYXxHqp/i55X3/+2bYWn/93p3/ueeec+527nn33fdmyeKlTWSjEAnyzVzQ2ELG ldmMrGDmsiVqtJyYQWSuva1l9oJoOWU+kenz2fNX3BYtZz5CpNzZ3NQ4K1qm75Ff3wxBtMyGIO/b vGDJHdFyZj1y2/yFM2P6zAdQti5ovCPWP72Hsnp744KmqH2uT5ZbFrYuiZUxHrquZXFTzJ4FiBJ+ p+OS+uNfVVGoAQwDBH1Bo+lRshAnBxXTLRhps9KHTChLvYlv3f7F5ovT7aO/smZbISB64qN+BTI/ rLRs+3bvD7MdZLWhGGfYSwXqWUp7bqZKB32799s7HdGepObSJbqoVlzX4clwHj8o+tMZgIv+YW8f Z5foJ/qERzl9EeHuSE4tsZcXCRUtFhtUBV0I7AUOAQpNF7nQOkBXAyFgL3AIOA6YMcFcQ6siXwjs As4AZtFH5IRVp6O8n8hE3UzM1y7S6TygA4KcoMXABGA6sBXYBZgNOylZCKwGDgEXADP5RHp4+2CM PT28ycg65s4vMYqN0WL9NKPY8ZNgNB8/KZpXjYuajYyaDRoSFQ+oiOb9CqN5cn5JCI13xCeWdJen iTRMMg0DbwFl/BWyM0ZO2i1SSQO4wFANiU8kd/T1lOw6JBRiggtGs8ipdwsWTkwqKY/nOj9PyeTk n/NzUQ0/19ErqWRX+Y38Q9oLHAIE/xDpA/4BreZnpM9By4BdwCHgGHAeMPMzSKeR3ufvk53/lYqB MmA6sAs4BJwHLPyvoA7+nowYg0q+DOD8PVAHfxfTehfUzk+BO8VP6d387fCwESVdBuMtjjHO/BiT nh1jktNKIvyt8Df9EVEerDQi6oDIo1IaLPLC+YOcEZERHj3HGeEfdahe5+7ygfwEaQDHSE6g5xOk AhOBBqAFMIM7Ce4khYBtwG5AAxBloA5A5UeAN4CTNBDwARMBKz8eRjcRfizsqXCWp/E3+R8oHR4/ yv9o5G/wV438df57I38NeS70R/ir4VwnlSdAT6jjQO5AXgy9ib/U0TfZqZcn8UPwoBO0GCgDJgDT ga2AmR/ieeFZzmQ0coCO4B528jB9ZuRP0hNW8s11+jyVCEBVEs/IG8CB7FJ3ebjPs+NhFCXx3Lcd nCSeezeDk8Rz5xpwknjmLwMniWfWXHCSeKZOByeJZ0ItOJAIf+zFvv2cwybMY2q5nS+Hl5bDS8vh peWk8OUy0TeKHOMvwgUF8NhOn7d/gTO0n4UOstBkFnqChZpYaBULrWGh0Sx0Kwt5WSiHhXJZyMdC B9hwuCLEfJ3XFEf4MljoCAs9x0KtLORhoXwW6stCKhvmi3BXeBzuOmR+I+solzcdd3XcUIrdx85d 8KgLMe/CnnAI9BigGyUfjNS8qHFmrszzOgrKouUBI0sWlo/lh1HxMJbhMJ0GFCzQYYTRYTRyGM3Z QcuA6UA3cB7QATOs8zCPrQa1gxYDZcB0YDVwHjAbwzmPoXBaCCqHuNcYWDFoGTBBlvhhpDwkF3f5 +jhyHF7HWLE1h9lz2YRcPZcPo7Q07MvJSdakCEvc93Xi/36dSHHlcfw+vpX6YCG2xfKt4W/6OCPs obDngLM8lT1IuQqijo0gD8tHPpxajfJQyrFK+RDK4c8iLwnn3IJq9rCn0Lmf9ZK19jm/yfnY+VlO hIM9m3PA+Wc1orCw80+QPLvPeSJno/O14ogVkoOeCEO2XzVMu3KGO587YpiugWJn2LlKZvucd+WM cc7LMRRNUcWtrSj57M7JnqnOsWivKmeG09eKNvc5y3JudY6OWg2VdfY5B2II3ihbgMH2zzE6deca DdYNi7BmX6FlhyVgmWC53lJiKbS4LE5LH0u2pbc12eqw9rLarPFWq9VsVazcStbeEf2Mzyufer3N xsPPjIBmpBi8AzsMk9sMKHFm5XQjaSmihtdMqWA1WvdMqpmhahenuCMsftJUzeSuYFpyDdXUVmjD vTURiz5ZG+at0SwTfxpoZ+y+IKQa3xBhVBuIMF2K1mZryZWBLmIsae2WbJlft3ZLMEgZacvKMsqS S5NGVFf9CGkwhA1V3itXxhXWm+Hto+2omRLQnukT1Eoko/cJ1mj/NUWtD3SxL9gFf1UX+6fMgoEu Ucq+8E+WclFaFQzWRNgthh2p7J+wQ8Qgg501l1RpR6o1N2q3M2qXj/qw6ysz2MXFUb5hlx8XZ9gp TNq1t/b1V7X3BYFNukqthk1runq1zZF82OSDwCYtREcMmyNpIWmjlRrN5OTAJBcEJiyLcgyTHJZl mBgjbzdMimMmGy+bbDR6EtHRGDaSoJnEM5dsEs/A5ipH/me2qcLrZR2jgjPr/U1uf4Pb3wQ0aJuW NWdooRmq2j4zKBWqJjwNM2Y2y7yxSQu6m6q0me4qtX2UUe9f1PVSPcpd1U71/tpAe72vqSo8yjfK 726sCnaMmThk2DV9bbzc15CJP9LXRNnYENnXGKPev/Q1TKrHyL6Gyb6Gyb7G+MYYfZER4xMD7Vaq CFZi/WTewRPiEa8N2a5gRZqjpdQI3lGujFXZ+3Fa2UMJ3qBmc1doiYCM66LyonKpwj0lVb0gtsdU GatGubL3sz0xlQPiJHcFeZcsbV1KGf45VdFfKy6IliyVSxGlXin70Qsmfs3XWCXP1jVawZQarWzS 1EC7xQJpQ1UQspGXZAkJ/ojeHRUOgHCkNBTisqGUjZayuLiY4b/HgjEmiOGdLhw0DnQwXy5bQq1B oeXW1HJsBbVT4Yb6qYH9OEvJh0RrEBNsZV7Weqk1OQ+Dp6iEMO3WS1iyNMbFfLEklhumrV7ytl5y yaXmvNJZBjF8tcSLrc20nzKBLNNTlKl4CO8b+qfAWZn3zNHPSr3M+d+x0UViINpDz7E59BwdopfZ BdTaS13USfIIVEWP0Ep6gNbjsTYVko00GckE+QMsU+/Em8njeGA+Tkdh+xNaRfspjWXon9FqWive Rq21lEh5VE4TaSFtYTfpS6meTiv30DC6iW6nFhbSA/p9+nb9V/Rr6hJ/1H+gBMqimUhH9c9Nf9Hf oyLU+Dk9TKfZ9rgXyIdeQrB8lBbTTjFNYfps/VuMwEXLMQaFxtNR1s29aL2JPmUZbKWoRCu/1DX9 FVjl0DRqpp20nw1lY7jLVK+P149SGvq4A60+TGHahxSh39IpZjNd0H+lX6BMKqRxmE8nvcm6Rc8P a3rK4DcTvNSfRkCzkH5Hf6DjzM1e4gtNNlOJyWe6Uz9BvWkQ1WG0T6HmJ+xrvgpptXhVqdYrqBf8 cr/0Nv2ePmBZrJhNYLfw/nwhf0wsJit6HIQ0i+bA3w+h9fcRRvu4jR8Tv1SeVb4z9+k5o/fCinjo F/QovcQSMVOVtbK72Un2Ea/k0/kv+IfiAeVp5S1LI2Z9Ky2gLfQsfc2S2XA2if2UNbOVbD27nz3M jrLj7Cwv57V8Hj8vmsUi8VulAmmK0qrcY1pn2mQ+2xPoeaXnv3u+1kv0dTQJ8bAGo/85PYaZddEx egfpNH3ITCyB9UJSmYvVsZ8hrWJb2BNsD3uadaKX4+xD9hkeSV+x7zietNzMs3H4kUcgN1+ME+YD /BF+DOk4/wf/RqSLPOEVQ8VoERQLMar1YhvSC+IDJUs5pujwc4lph2mXaY/pWdPLpgtmm+VuPOPf +P6XPxT88H4P9Wzo2dET7unUP6BUrCGeHngFG43RNyLNxXrvQMTtpbeZDb7LYgWslN0Ez0xnc9ki dgc8eS/byX5tjP15dhBe+jM7jzEn8hxjzAP4UF7BJyDdypv4IhzGtvNOfpJ/KywiQdhFqigQY8Q0 0SSWiBVih9DEG+Kv4kNxUXyPpCvxilPJUzyKVxmjTFeWKo8pnyqfmupNr5v+Zo43LzCvM0fM/8Sp ptQy0TLJMs2y1bLPcsLagOg8TC/Qi4jAyxc7I9YIv3iB7uODlUy8wryJeJ5Os8R4jkjle9gGfhfr 5H1Nd5hH8VHsZrqgeODrV/kufpGPEuNZDZtCc/mgaIPm3soz4EYrh+mcchBzexMt32G2sVX8vNlG YZyRRuCM9HsxUPGK1+mUOM0syuP0rhLP0tk5/pSYiCj4rVJqCpBLPELPi0XsLnqB+4niv7NuRhzf zJ7BvlDLStj/Ch3H4JsRRcPER3QPzeN/oXO4jzfQg2yWMpvuo8FsJX1KT+Ku6G+63VxgTmWv8TlK G09hncSVpzG7EawvE6bedC+bJnaaz/N3aCkdU+LpffEbjP4Yf16MVy6YJrNm3AF30TpapK+hFaaA 8habTYLdQvnKGexuK0WJ4kK+GrtKPfa0fbi792MfKBfjIclA5NyEuKjDDrET6SHsEwoiaA7u8Z9g F3uTOs21PEKzTb0Ydh18qXm9ZzJN1Z+kh/XZdLu+nYqwH6zXV6LFPfQ32kp72Nqen1ELXiXfwb19 k6maHzNV60W8jb/Dp/Ad164vvJ3PMujvSM9jZUpNB6hN+TNNoTJ9s/4nRPd12GEfphk4sH6MWX6O HsaKbhrcczNv16tFC+Z7mibpT+lOFk/N+nyaQAfp1xYTNVq8WGONvYX5/oya+GR9iWjqmQM/bIUX fPDWUuw/G32VdbXlvrLSG0aPGjli+LChQwaXDBpYPKCo0FvQ/7p+nvy+7jyX6sztk5OdlZmRnpba OyU5yWHvlWhLiI+zWswmRXBGhX53dYOqeRo0xeMeO7ZIlt2NEDReJWjQVIiqr7XRVFmvEaprLH2w vO1fLH1RS99lS+ZQR9PookLV71a1o1VuNcKmTgqA31LlDqraOYMfb/DbDD4RvMuFCqo/o7lK1ViD 6teqlzW3+RuqigpZe0J8pbuyKb6okNrjE8AmgNPS3S3tLL2UGQxP949s52RNxBS1LHeVX8t0oyqa Efn+xlnaxEkBf1W2yxUsKtRY5Uz3DI3kSclrmFCl0Y1mrtQsRjfqHJxxNNqkthd2t22OOGhGg9c2 yz2rsT6giUa04deSvOi3Sku/8+OMK0U0jjPZ+qu12aLNnzFHlcZtbetVbfekwFV1s12yhWAQbaAu z69uaKtG15uxUjXyLK7xtcGAxtaiSxws841ZRecXPfXmN8xVtTh3hbu5bW4DliarTaPJK1zhrCxf l36GsvxqW23A7dLKst3Bxqqc9t7UNnlFR6ZPzbxWU1TY7kiKOra9lz3G2BKvZprg9KjO4AxzydVM vuxZJsfoHoeToKbOVDGSgBtzGi5J03BqmzkcC4AryFBLm4UVmaPFVTa0OUZKOabINFO+w622fUWI APe5f1wraYxJzPmOr0gqZZxcDjWNNV7iNa9XKyiQIWKpxJpijKVGeWhR4bIId7tbHHh/li8NNBG+ bQyOLIb7XS65wJsiPpqBghaaFIiWVZqRHSZfMc7WvEFqui9pUuukJnRJc7l6gxuR3CnfZylVs3ou /+yOtBR/80iNpf0HdVNUXzPFXYOjsepva4hFbU3tNaWoXjoUfoMuxmkplQGRzSGTHM8WhjZ6Qr5k guNywKYp+fiZjaCeFbFYEZWGhKnVmqNhbJQG412u2D3z/1WK6BdkLSO7Ui02DW2kNzbQ6LC1UdeU rxmerU3U1GLL4TjZt7XFX6NDqEVHOS6WIeLxou9SKzWqw52Zjx9eOYZLBLM1H1wGTS3uIkMczI4V rzHMjlUK4pLRWVRYjT2zra3arVa3NbQ1RvTQDLfqcLd18Zf5y20tfux20cCJ6Ps3ZWvVm4PwWDMb iduDU0W7m22Y1O5jG6ZMDXThE4e6oTYQ5oxXNlQE5bLwytpAzC3GgsjQxxriHxNEjNJKdcAUhA4z Agj/e0CRg/LAq/5RQNG45GcTeaXS//Ab8PUErAkJJ24LkSvJlZQPglbpe1V0f+8z0XekKt14fFOd /qmSZOrGfxx9WF07l4PyxWflKqbeuYmJ6XER/Wyn3c7rJOPLTEwEl0Q2KaE0mw3UJmVUjHeioyBH qexc2blBA7Pbzf/e0pdoySxb+qQzMdFgPvdlJiSASyKHlJDDZpNUyi43eaXNTrOa6chBYIW5in9t zlAakAzY8VlphmJezzckbLC/1ssUZ0nI4P6Um1JvzKzMrk2pT63PnJw9zzIvYWbK/NR5mQ3ZK/hy 87KEO+3rzQ9ZdjheyzjFT5pPJrxrz7o88dY4n8s9ZGAcozhHHI/b5kxqJQSTrxekKh7NnLbl/mFT htdxcZr3HMgi77nY1Nm0RXjJGS4vBgSDKY7k6weXpKUlpzq42Z3Xz5PiSBtccn2Sw+POs5jr5r29 e1l4ScXctx8/seL+rqdXrnz66VUrb5zG32YKu+E30zt69FM9PT2Hn3voRfZoz4PnL+DNYe7nc9bJ tZ2CtcvE2qWTG9+LKbp6nTbKzh0gx5qeksLrBgxIduWaTdflJifmxtnkkmEBvtxnrKTXDmcaywvm bGdCAq+TjKG0ZwipdDh4ncHIJQfzZTQIRN9UmzRPNVpMNYIg9VIQGJFwVTgkjyj2nhsx4nJUvGgM xAgAORAwciAfG9EhGUMW619GDLr93pcnDWW3MkRkh5LKmV6Zn7EA6Ax9MRmLVwOd+4YNTWP908al jfN8YvtsoCluIA7Gd7GVyhLrooTFtqWJd6Zvoja2WVlnXZNwr21d4pb0N5JeTUnOw1fLcI6aJTNV LZZZkeqRnzJz+6s2ys0gG4axewC7MpLc1kNxLC7CZ/sc3la7T0Uk2RnZHXZuj7D795VktGqCCejD fVtTL4dVqi+Vp24bdDmsvpx2zvGljCwwsblNMyaXlCy9KaPLuNi0xYtoUTDIPJ6hQ2SopfY2W9JA yZ1HkKT0lvE2dAjizSx6p10psLkt8z851P33eQvWb+m5+M47PRfvn7FuXvPajbfN3jBy3LYpa/Y8 d/fqp0R2/4fm7j51evdtD/YvfGXDQR0fTru3vsRqm++9Z/rM9fd+r4/fNuHJ0N3P7JE7UGwn0vvh /e7HLgFhCmWDMkqO7WxmvI9TbbBqzISgt3zxnMb542v/D6wQI80KZW5kc3RyZWFtCmVuZG9iagoz MCAwIG9iago1Mjk0CmVuZG9iagoxNyAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1 ZVR5cGUgL0Jhc2VGb250IC9SRFFCTUErQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKMzEgMCBSIC9F bmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDQ4IC9MYXN0Q2hhciA1MiAvV2lk dGhzIFsgNTU2CjAgMCAwIDU1NiBdID4+CmVuZG9iagozMSAwIG9iago8PCAvVHlwZSAvRm9udERl c2NyaXB0b3IgL0ZvbnROYW1lIC9SRFFCTUErQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFst NjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAt MjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0 ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiAzMiAwIFIgPj4K ZW5kb2JqCjMyIDAgb2JqCjw8IC9MZW5ndGggMzMgMCBSIC9MZW5ndGgxIDc0NjAgL0ZpbHRlciAv RmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLfFTF1T8zc/eRZJfdBPLeZO9mySLZxEAAIRCTzWMD GpBXsLs0mASIhJcEAygWYakiGFAotVRpFR+1Qq1ys4m4AVtSsbaiCJ9SWtEKPlq1v/KoX9X6IPf7 z90FpfXX707+Z86cc+Z15szcuZsVN69sIxtFSFBg3tLWDjKe7HZkBfNWrVDjZXsWkbnxxo4FS+Pl wUuITGcXLFl9Y7ycfR+RqG9va50fL9NXyK9qhyBeZqORD21fuuLWeDm7CbltybJ5CX32FpStS1tv TfRPb6Gs3tS6tC1un6fIcseyzhXxskvqr+i4uS1hz0JEKb/W8Uj9sU/qKNIChgGCPqYKepAsxMlJ pXQ9kdKu5JEJZak38a3bP97yabOj4hNrrhUCokffG1Yk80NKx7Yv9l5Y4CSrDcUkw14qUM9SOXAd 1Trpi71f3OaM9yQ1Fx/RR43iih5flvvYc2I4nQa4GB7157n7xDCRF53gDsSEtyctvcxRXSJUtFhq UBV0GbAXOAgo1CzyoXWCrgMiwF7gIHAMMGOC+YZWRb4M2AWcBswiT7iiqttZPUxko2425usQmXQO 0AFBbtBSYCrQDGwFdgFmw05KlgHrgIPAecBMAZEZ3T4KY8+MbjaynkVLyoxia7zYNMco9nwnHM+n TI/nddfEzcbHzUaOjouvrInnw4rjeVphWQSN9yTby/qrM0QGJpmBgXeAMv4CORgjNz0s0kkDuMBQ DUlApPUM9ZXtOigUYoILRvPJrfcLFrWnllUnc52fozRy87P8TFzDz/QMSi3bVX0tf5f2AgcBwd9F eoe/Q+v4aelz0CpgF3AQOAqcA8z8NNIppLf52+Tgf6ZSoApoBnYBB4FzgIX/GdTJ35IRY1DJVwGc vwXq5G9iWm+COvhJcCf5Sb2fvx4dW17WZzD+0gTjLkwwmbkJJi2jLMZfi34+HBHlw0ojog6IAqqk UaIgWjjSHRNZ0YqF7hh/r0f1ux+uHsGPkwZwjOQ4ej5OKjANaAE6ADO4E+BOUATYBjwMaACiDNQJ qPww8ApwgkYAAWAaYOXHougmxo9GfTXu6gz+Kv8dZcLjR/jvjfwV/qKRv8x/a+QvIc+H/jB/MZrv puoU6Al1nMidyEuhN/Hf9AxNc+vVqfwgPOgGLQWqgKlAM7AVMPODvCA6352GRg7QYexhN4/SR0b+ c3rUSoFF7oCvFgGoSuIbfzU4kF3qLh8P+HY8gKIkvnu3g5PEd+cWcJL4blsPThLfklXgJPHNXwRO Et/sZnCS+KY2ggOJ8YeeHTrMPXbqYqZWO/gt8NIt8NIt8NItpPBbZKLPcZy5+U+iRUXw2M6Af3iR O7KfRZ5jkRks8iiLtLHIWhZZzyIVLHIDi/hZxMUi+SwSYJEDbBxcEWGB3suK5YEsFjnMIk+xSCeL +FikkEWGsojKxgZi3BO9BrsOWdDIeqrlpuOenqsrcfo4uAce9SDmPTgTDoIeBXSjFICRWhA3zs6X eUFPUVW8fOX4smXVk/ghVDyEZThEpwAFC3QIYXQIjRxCcw7QKqAZ6AfOATpghnUB5rHVoA7QUqAK aAbWAecAszGccxgKp2Wgcoh7jYGVglYBU2WJH0IqQPJwTyDP6XL6nZPEVhdz5LOp+Xo+H0sZGTiX 01KtqTFm3/eZ/V+f2SmpOonfy7dSHhZiWyLfGv08zx1j90d9B9zV6ezHlK8g6lg5+Vgh8nHUaZTH kMsq5aPJxZ9EXhZ1XY9qjqiv2L2fDZK19rk/d73v/sgV42A/dB1w/1GNKSzq/gMkT+5zH3fd7X6p NGaF5DlfjCHbrxqmfa5x7qcOG6brodgZda+V2T737a6J7sUuQ9EWV9zQiVLA4Z7hm+2ehPbqXHPd gU60uc9d5brBXRG3GiPr7HOPwBD8cbYIgx3uMjr15hsNzhobY+2BYssOS8gy1XKVpcxSbPFY3JY8 S65liDXN6rQOstqsyVar1WxVrNxK1iEx/XTAL996Q8zGy8+MgGakGLwTJwyTxwwocWbldC1pg0UD b5hZwxq0/nnUMFfVPp3pjbHk6bM1k7eGaWkN1NBYo43zN8Qs+gxtrL9Bs0z7bqibsXvDkGp8U4xR YyjGdCnakKul1Yb6iLHUDffkyvyKDfeEw5SVsaoqqyqtMrW8vu5bSIshbKnzf/1kfc36s/x52o6G mSHtF3lhrUwyel64QfvhTLUp1Mc+ZueDdX3sHzILh/pEJfs4OEPKRWVdONwQY9cbdqSyf8AOEYMM dtZ8UqUdqdb8uN3OuF0h6sNuqMxgl5REhYZdYVKSYacwadfdOTRY1z0UBDaZKnUaNp2Z6jdtDhfC phAENhkROmzYHM6ISBut0mjG5YJJPghMWA65DBMXyzFMjJF3GyalCZO7L5ncbfQk4qMxbCRBM/bT F23sp2HzDUf+d7atxu9nPRPC85qCbd5gizfYBrRom1e1Z2mRuaraPS8sFaomfC1z57XLvLVNC3vb 6rR53jq1e4JR79/UTVI9wVvXTU3BxlB3U6CtLjohMCHoba0L90ycNnrsZX3dfamv0dO+pa9psrHR sq+JRr1/62usVE+UfY2VfY2VfU0MTDT6IiPGp4W6rVQTrsX6ybyHpyQjXltyPeGaDGdHpRG8EzxZ a3P347aym1L8Yc3mrdHsgIzrkuqSaqnCnpKqQRA7EqqstRM8ufvZ7oTKCXGqt4b8K1Z2rqSs4MK6 +F8nHohWrJRLEad+KfvWByZBLdBaJ+/WDVrRzAatavrsULfFAmlLXRiy8RdlKSnBmN4fF14J4Xhp KMQlQymrkLKkpIThf8aCMSaI4Z0+XDQO9LBAPltBnWGh5Tc0chwFjbPhhqbZof24S8mXRGcYE+xk ftZ5sTU5D4OnuIQw7c6LWLEywSV8sSKRG6adfvJ3XnTJxeb80lkGMXy1wo+jzbSfsoEc0xOUrfgI 3z/6B8CHMh9YqH8o9TLnf8NBF0uAaDc9xRbSU3SQnmfnUWsv9VEvyStQHf2U1tB9tBGvtdmQ3E0z kEyQ38ey9V58mTyCF+YjdAS236G1tJ8yWJb+Ea2jDeJ11NpAdiqgappGy+geNllfSU10SrmDxtJk uok6WEQP6ffq2/Wf0ePUJ36vX6AUyqF5SEf0s6Y/6W9RCWr8iB6gU2x70jMUQC8RWD5IN9NOMUdh +gL9C4zAQ7dgDApNoSOsn/vReht9wLLYGlGLVh7TNf0FWLloDrXTTtrPxrCJ3GNq0qfoRygDfdyK Vh+gKO1DitGv6CSzmc7rP9PPUzYV0zWYTy+9yvrFwIX1A1XwmwleGk7l0CyjX9Pv6Bjzst/wZSab qcwUMN2mH6chNJJmYbRPoOZf2Wd8LdI68aJSr9fQIPjlB9Lb9Ft6h+WwUjaVXc+H82X8IXEzWdHj SKT5tBD+vh+tv40w2sdt/Kh4THlS+dKcN3BaH4QV8dFP6EH6DbNjpirrZN9nJ9h7vJY385/wd8V9 yh7lNUsrZn0DLaV76En6jKWxcWw6+y5rZ2vYRvYD9gA7wo6xD3k1b+SL+TnRLpaLXyk1SDOVTuUO 012mzeYPB0IDLwz8z8Bnepl+F01HPKzH6H9ED2FmfXSU3kA6Re8yE0thg5BU5mGz2PeQ1rJ72KNs N9vDetHLMfYu+wivpE/YlxxvWm7mubj8yCuQl9+MG+Z9/Kf8KNIx/nf+ucgUBcIvxogKERbLMKqN YhvSM+IdJUc5qujwc5lph2mXabfpSdPzpvNmm+X7eMe/8tVjF4ouvD1AA5sGdgxEB3r1dygda4i3 Bz7BKjD6VqRFWO8diLi99DqzwXc5rIhVssnwTDNbxJazW+HJO9lO9rgx9qfZc/DSH9k5jNnOXcaY r+RjeA2finQDb+PLcRnbznv5Cf6FsIgU4RDpokhMFHNEm1ghVosdQhOviD+Ld8Wn4iskXUlW3EqB 4lP8ykSlWVmpPKR8oHxgajK9bPqLOdm81HyXOWb+B241lZZplumWOZatln2W49YWROcheoaeRQRe ethpsV4ExTN0Lx+lZOMT5lXEczPNF1M4IpXvZpv47ayXDzXdap7AJ7Dr6Lzig69f5Lv4p3yCmMIa 2ExaxEfGGzQPUX4BrkI5RGeU5zC3V9HyrWYbW8vPmW0UxR2pHHek34oRil+8TCfFKWZRHqE3lWSW yc7wJ8Q0RMGvlEpTiDzip/S0WM5up2d4kCj5S+sWxPF17Bc4FxpZGfuX0HENvg5RNFa8R3fQYv4n OoN9vIl+zOYrC+heGsXW0Af0c+yK4aabzEXmdPYSX6h08cGsl7iyB7MrZ0OZMA2hO9kcsdN8jr9B K+mokkxvi19i9Ef502KKct40g7VjB9xOd9FyfT2tNoWU19gCEux6KlRO43RbI8oUD/J1OFWacKbt w+7ej3OgWkyBJAuRMxlxMQsnxE6k+3FOKIighdjj38Ep9ir1mht5jBaYBjGcOvil5uWBGTRb/zk9 oC+gm/TtVILzYKO+Bi3upr/QVtrNNgx8jzrwKfkG9vZkUz0/aqrXS3gXf4PP5DsuX194u5Bl0d+Q nsbKVJoOUJfyR5pJVfoW/Q+I7itwwj5Ac3FhfR+zPIseJol+GjVwHe/W60UH5nuKputP6G6WTO36 EppKz9HjFhO1WvxYY429hvl+j9r4DH2FaBtYCD9shRcC8NZKnD93B2pnNVYHqiqvrpgwvnzc2DGj R5WNHFF6ZUmxv2j4FcN8hUO9BR7VnZ/nys3JzsrMSB8yOC3V6Rhkt6UkJ1ktZpMiOKPioLe+RdV8 LZri806aVCLL3lYIWr8haNFUiOovt9FUWa8VqsssA7C88d8sA3HLwCVL5lQrqKKkWA16Ve1InVeN sdnTQ+DvqfOGVe2MwU8x+G0Gbwfv8aCCGsxqr1M11qIGtfpV7V3BlrqSYtadklzrrW1LLimm7uQU sCngtExvRzfLrGQGwzOD47s5We2YopbjrQtq2V5URTOiMNg6X5s2PRSsy/V4wiXFGqud552rkbwp +Q0TqjW60cy1msXoRl2IO45Gm9Xu4v6uLTEnzW3x2+Z757c2hTTRijaCWqof/dZpmbe9n/V1EY3j Trbxm9pc0RXMWqhK466ujar28PTQN+rmemQL4TDaQF1eWN/SVY+ut2ClGuRdXOMbwiGNbUCXuFgW GrOKzy9+6y1sWaRqSd4ab3vXohYsTU6XRjNWe6I5OYE+/TTlBNWuxpDXo1XlesOtda7uIdQ1Y3VP dkDNvlxTUtztTI07tnuQI8HY7N9k2uD0uM7gDHPJNcy45Fkmx+i9BjdBTZ2nYiQhL+Y0TpK2cdQ1 bxwWAE+YoZY2HyuyUEuqbelyjpdyTJFppkKnV+36hBAB3jN/v1zSmpCYC52fkFTKOLkUahprvchr fr9WVCRDxFKLNcUYK43ymJLiVTHu9XY48f0sPxpoGnzbGh5fCvd7PHKBN8cCNBcFLTI9FC+rNDc3 SoFS3K15i9T0X9Skz5KayEXNpeotXkRyr/yepXTN6rv053BmDA62j9dYxn9Rt8X1DTO9Dbgaq8Gu lkTUNjReVorrpUPhN+gSnDa4NiRyOWSS47nC0MZvyBdNcF0O2TSlEH9mI6jnxyxWRKUhYWq95myZ FKfhZI8nsWf+v0ox/bysZWRfV0tMQxvvTww0PmxtwmXly4Zn6xINjThyOG72XV3Jl+kQavFRXpPI EPH40PeotRrNws4sxB8+OcZJhHO1AFwGTSN2kSEO5yaKlxnmJiqF8cjoLCmux5nZ1VXvVeu7Wrpa Y3pkrld1erv6+PP8+a6OIE67eODE9P2bc7X6LWF4rJ2Nx/bgVNPtZZumdwfYppmzQ334iUPd1BiK csZrW2rCcll4bWMo4RZjQWToYw3xHxNEjNJJs4AchA4zAgj/94BiCMojcPeVISUf+XOJfNLpf9lj +NUErAkJN20LkSfVk1oIgtboK1X0fxUw0ZekKv2y3iz9AyXV1I//beSxWd1cDiaQnJOvmIbk2+2Z STH9w16Hg8+STCDbbgeXSjYpoQybDdQmZVSKb6EjIEeo6kzVmZEjcrvN/9nSP9GSWbb011673WDO BrJTUsClklNKyGmzSSpll5r8us1es5rtdCGgolzFf2tOUwaQBjjwc9JcxbyRb0rZ5HhpkCnJkpLF g4Mnp1+bXZvbOLgpvSl7Ru5iy+KUeYOXpC/ObsldzW8xr0q5zbHRfL9lh/OlrJP8hPlEypuOnEsT 70wKeLyjRyQxSnIm8aRt7tROQhAFBkGq4pXMaVv+7zZn+Z2fzvGfAVnuP5OYOpuzHB834+TDgHB4 sDPtqlFlGRlp6U5u9hYM8w12ZowquyrV6fMWWMyzFr/+8KroippFrz9yfPUP+vasWbNnz9o1187h rzOFXf3L5p4B/eTAwMChp+5/lj048ONz5/HFsOjswrvk2iEqFKdpPyWTnfkCV6WFbO22nbY9tpds pslisv0+RaQx/N5mMwuLKTlFWMiG1ToslCFCKMJO3GZXLOIAfvS2ImQeDiSTosCEDicrMX7jsyZT ciDPPToZP9b1Op18Fhi5cAZz1ljB5BgbG7BbAgXe0ZaIZ4xlm4NLH6XYh4wm7uQqFyifNuqAeX+f XGP+zKAY29ItXfd3v38OHOj3/1N+SVc4/+q8UEFVVc5/VnxakVpezlLTyss3XulXbne+4HA4Ro5g xje/XX87mlZuj+nHAymjykVBSblQ8vIqZBPhOTQHNoEhtkBKuS0yrdwW8JXbClzIS8qlgT+MbTCG jUodle5NFamM77hwJ3/why++2DswhjU/LvZ9de3jA49whf/owmJsocSu0ofhG+XbHgHhYHxvy72Z ltiLZnxT0oy66TVTqv3VNy9sXTKl8f8AB0CzWAplbmRzdHJlYW0KZW5kb2JqCjMzIDAgb2JqCjUx MzYKZW5kb2JqCjExIDAgb2JqCjw8IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVlVHlwZSAvQmFz ZUZvbnQgL1FQRFpKUitBcmlhbE1UIC9Gb250RGVzY3JpcHRvcgozNCAwIFIgL0VuY29kaW5nIC9N YWNSb21hbkVuY29kaW5nIC9GaXJzdENoYXIgNDggL0xhc3RDaGFyIDUzIC9XaWR0aHMgWyA1NTYK NTU2IDAgMCAwIDU1NiBdID4+CmVuZG9iagozNCAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0 b3IgL0ZvbnROYW1lIC9RUERaSlIrQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0z MjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9D YXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0 IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiAzNSAwIFIgPj4KZW5kb2Jq CjM1IDAgb2JqCjw8IC9MZW5ndGggMzYgMCBSIC9MZW5ndGgxIDc3ODAgL0ZpbHRlciAvRmxhdGVE ZWNvZGUgPj4Kc3RyZWFtCngBhVkLfBTV1T/3zuwjz90kJNkkm+wsSxbJJgYCGB4xmTw2qBF5Bd2l iUmAlPCSYAKKIqxWBAMK8lkUqqK2VagPJpuIG8CSitWKIlQsrWgFH63aXyPo56M+yHz/O7tBaf31 m7v/c88959zXuWfu3JntvH5FKyVSiCRS5y1taSfjylqCbMS8lZ1KtJzkIDLX/7R9wdJoOQ160ycL lqz6abSc9QCRvLqttWV+tEzfIb+kDYJomY1DPqJtaeeN0XJWA3L7kmXzYvqsLSjHL225MdY/vY2y cl3L0taofV6NKLcv6+iMlTEeKmy/vjVmzwJECb/VcQn9sS9qKNQMhgEyfUZl9CBZiJOdiulqiNrk XDKhLPQmvnnrZ5u+bLKVfWHNsUJA9Oj7IwtEfkhu3/LNnnML7GRNRDHOsBcK1LOUD15F1Xb6Zs83 N9mjPQnN0CX3Ub10UY/X4Tp2QBpFpwEujQr7cl190kgpNzzZpUYkT09qeomtskhS0GKxQRXQZcAe 4CAgU5OUB60ddC0QAvYAB4FjgJkIVGgVYBmwEzgNmKVcyRlWXPbKkVIW6mZhvjYpk84AOiCRC7QY mAY0AZuBnYDZsBOSZcBa4CBwFjCTKmWGt47F2DPDG42sZ9GSEqPYEi02NBrFnmuC0XzqjGhec3nU bFLUbMy4qPjiqmg+sjCap+aXhNB4T3xSSX9lhpSBSWZg4O2gjL9ANsbIRQ9L6aQBXMJQDYkqpfaM 8JbsPCjJxCQuMZpPLr1fYuGklJLKeK7zM5RKLv4JH4hq+EBPckrJzsor+Hu0BzgISPw9pHf5u7SW nxY+B60AdgIHgaPAGcDMTyOdQnqHv0M2/lcqBiqAJmAncBA4A1j4X0Ht/G0RMQYVfAXA+dugdv4W pvUWqI2fBHeSn9T7+fFw6cSSPoPxFccYV36MycyJMakZJRH+evjrUYgoL1YaEbVfGk7lNFYaHs4f 44pIjnDZQleEv9+j+FwPV47mb5AGcIzkDfT8BinAdKAZaAfM4E6AO0EhYAvwMKABiDJQO6Dww8Cr wAkaDajAdMDKj4XRTYQfDXurXJUZ/DX+EmXC40f4H4z8Vf6ikb/Cf2/kLyPPg/4wfzGc56LKBOgJ dezI7ciLoTfx3/WMSHXplSn8IDzoAi0GKoBpQBOwGTDzg3x4eL4rFY3sp8O4h108TB8b+WP0qJXU RS7VW40AVATxTroUHMhOZaeXq95t21EUxHv3VnCCeG/fBE4Q7023ghPEu2QlOEG88xeBE8Q7pwmc IN5p9eBAIvyhZ0eMdJVOW8yUShu/AV66AV66AV66gWR+g0j0tSzG+ItwQQE8tkP1jSpwhfax0AEW mslCj7JQKwutYaFbWaiMha5lIR8LOVkoj4VUFtrPJsAVIab2XlCcqDpY6DALPcVCHSzkZaF8FhrB QgorVSPcHb4cdx0yv5H1VIqbjrt7Li3H7mPjbnjUjZh3Y084CHoU0I2SCiNleNQ4K0/kw3sKKqLl iyeVLKu8jB9CxUNYhkN0CpCxQIcQRofQyCE0ZwOtAJqAfuAMoANmWA/HPDYb1AZaDFQATcBa4Axg NoZzBkPhtAxUDHGPMbBi0ApgmijxQ0jDkdzcrebanXaf/TJps5PZ8ti0PD2Pl1JGBvbl1BRrSoQl 7f0q6V9fJVFcZRy/m2+mXCzElli+Ofx1rivC7g9797sq09l9lCcj6thE8rJ85BOowyiPJ6dVyMeR kz+BvCTsvBrVbGFvoWsfSxa19rq+dn7g+tgZ4WA/cu53/VmJyCzs+hMkT+x1veG80/VyccQKyQFv hCHbpximfc4JrqcOG6a3QrEj7Fojsr2uW5xTXIudhqI1qri2AyXV5prpneO6DO3VOOe61A60uddV 4bzWVRa1Gi/q7HWNxhB8UbYAgx3lNDr15BkNzi6NsDa10LLNErBMs1xiKbEUWtwWlyXXkmMZZk21 2q3J1kRrvNVqNVtlK7eSdVhEP636xFNvmNl4+JkR0Ixkg7djh2FimwElzqycriAtTarjdbOqWJ3W P4/q5iral7M8ERY/Y45m8lQxLbWO6uqrtAm+uohFn6mV+uo0y/SfBLoZuzsIqcY3RBjVByJMF6J1 OVpqdaCPGEtZd1eOyC9ad1cwSI6MlRWOitTylIm1NT9Cmg1hc43v+8vxPetz+HK1bXWzAtpvcoNa iWD03GCd9j+zlIZAH/uMnfXX9LFPRRYM9Enl7DP/TCGXymuCwboIu9qwI4V9CjtEDDLYWfFgFnak WPOidjuidvmoD7sRIoNdXBzlG3b5cXGGncyEXXfHCH9N9wgQ2GQq1GHYdGQqP7Q5nA+bfBDYZITo sGFzOCMkbLRyoxmnEyZ5IDBh2eQ0TJws2zAxRt5tmBTHTO48b3Kn0ZMUHY1hIwiaSTo9ZJN0GjY/ cOR/Z1urfD7WMzk4r8Hf6vE3e/ytQLO2cWWbQwvNVZTueUGhUDTJ2zx3XpvIW1q1oKe1RpvnqVG6 Jxv1/k3dINSTPTXd1OCvD3Q3qK014cnqZL+npSbYM2X6uNIL+rrzfF/jpv9IX9NFY+NEX1OMev/W V6lQTxF9lYq+SkVfU9QpRl9kxPj0QLeVqoLVWD+R9/CEeMRrc447WJVhby83gney27EmZx9OK7so wRfUEj1VWhIg4rqosqhSqHBPCVUyxLaYyrFmsjtnH9sVU9khTvFUka9zRccKcvgX1kR/Hbgg6lwh liJKfUL2oxdM/JraUiPO1nVawaw6rWLGnEC3xQJpc00QsklDsoQEf0TvjwovhnCSMJSk84ZCViZk cXExw/+MBWNMEMM7fTho7O9hah7rpI6gpOXV1XNsBfVz4IaGOYF9OEuJh0RHEBPsYD7WMdSamIfB U1RCmHbHEDpXxLiYLzpjuWHa4SNfx5BLhprzCWcZxPBVpw9bm2kfZQHZpscpS/YS3jf0D4GPRD64 UP9I6EXO/4GNLhID0S56ii2kp+ggPc/OotYe6qNeEkegGnqAVtO9tB6PtTmQ3EkzkUyQ38uy9F68 mTyCB+YjdAS219Aa2kcZzKF/TGtpnXQctdZREg2nSppOy+gudqW+ghrolPwzKqUr6TpqZyE9oN+t b9V/Rb+mPukP+jlKoGyah3RE/8T0F/1tKkKNn9N2OsW2xj1DKnoJwfJBup52SI0y0xfo32AEbroB Y5BpKh1h/dyH1lvpQ+Zgq6VqtPJLXdNfgJWTGqmNdtA+Np5N4W5Tgz5VP0IZ6ONGtLqdwrQXKULP 0UmWaDqr/0o/S1lUSJdjPr30GuuXBs/dOlgBv5ngpVE0EZpl9Ft6iY4xD/sdX2ZKNJWYVNNN+hs0 jMbQbIz2cdT8O/uKr0FaK70o1+pVlAy/3CO8Tb+nd1k2K2bT2NV8FF/GH5KuJyt6HIM0nxbC3/ej 9XcQRnt5Ij8q/VJ+Qv7WnDt4Wk/GinjpF/Qg/Y4lYaYK62C3sRPsfV7Nm/gv+HvSvfJu+XVLC2Z9 LS2lu+gJ+oqlsglsBvsJa2Or2Xp2D9vOjrBj7CNeyev5Yn5GapOWS8/JVUiz5A75Z6Y7TBvNHw0G Bl8Y/OPgV3qJfgfNQDzcitH/nB7CzProKL2JdIreYyaWwJKRFOZms9nNSGvYXexRtovtZr3o5Rh7 j32MR9IX7FuOJy038xwcfsQRyMOvxwnzXv4AP4p0jP+Tfy1lSsMlnzReKpOC0jKMar20BekZ6V05 Wz4q6/BziWmbaadpl+kJ0/Oms+ZEy214xr/63S/PFZx7Z5AGNwxuGwwP9urvUjrWEE8PvIKVYfQt SIuw3tsQcXvoOEuE77JZAStnV8IzTWwRW85uhCdvZzvYr42xP80OwEt/Zmcw5iTuNMZ8MR/Pq/g0 pGt5K1+Ow9hW3stP8G8ki5Qg2aR0qUCaIjVKrVKntEraJmnSq9JfpfekL6XvkHQ5XnbJw2Wv7JOn yE3yCvkh+UP5Q1OD6RXT38zx5qXmO8wR86c41ZRbpltmWBotmy17LW9YmxGdh+gZehYReP5ip6Vb Jb/0DN3Nx8pZeIV5DfHcRPOlqRyRynexDfwW1stHmG40T+aT2VV0VvbC1y/ynfxLPlmayurYLFrE x0QbNA+TfwOuTD5EA/IBzO01tHyjOZGt4WfMiRTGGWkizki/l0bLPukVOimdYhb5EXpLjmeZbIA/ Lk1HFDwnl5sC5JYeoKel5ewWeob78aHkW+smxPFV7DfYF+pZCfuXpOMYfBWiqFR6n35Gi/lfaAD3 8Qa6j82XF9DdNJatpg/pMdwVo0zXmQvM6exlvlDu4mmsl7i8G7ObyEYwyTSMbmeN0g7zGf4mraCj cjy9Iz2J0R/lT0tT5bOmmawNd8AtdAct12+lVaaA/DpbQBK7mvLl09jdVkslshv5WuwqDdjT9uLu 3od9oFKaCokDkXMl4mI2dogdSPdjn5ARQQtxj1+DXew16jXX8wgtMCUz7Dr4UvPK4Eyaoz9G2/UF dJ2+lYqwH6zXV6PFXfQ32ky72LrBm6kdr5Jv4t6+0lTLj5pq9SLexd/ks/i2C9cX3s5nDvoH0tNY mXLTfuqS/0yzqELfpP8J0X0RdtjtNBcH1g8wy0/Qw2VSP40dvIp367VSO+Z7imboj+suFk9t+hKa Rgfo1xYTtVh8WGONvY753kytfKbeKbUOLoQfNsMLKry1AvvPnWr17PpKtaL80rLJkyZOKB0/bmzJ mNHFFxcV+gpGXTTSmz/CM9ytuPJynTnZWY7MjPRhaakpdltyUmJCfJzVYjbJEmdU6PfUNiuat1mT vZ7LLisSZU8LBC0/EDRrCkS1F9poiqjXAtUFliosf/pvlmrUUj1vyexKGZUVFSp+j6IdqfEoETZn RgD8XTWeoKINGPxUg99i8Eng3W5UUPyOthpFY82KX6td2dblb64pKmTdCfHVnurW+KJC6o5PAJsA Tsv0tHezzHJmMDzTP6mbkzUJU9SyPTV+LcuDqmhGyve3zNemzwj4a3Lc7mBRocaq53nmaiROSj7D hKqNbjRztWYxulEW4oyj0Ualu7C/a1PETnObfYnzPfNbGgKa1II2/FqKD/3WaJk3feD4vojGcSZb /0NtjtTldyxUhHFX13pFe3hG4Ad1c9yihWAQbaAuz69t7qpF15uwUnXiLK7xdcGAxtahSxws841Z RecXPfXmNy9StDhPlaeta1Ezlia7S6OZq9zh7Gy1Tz9N2X6lqz7gcWsVOZ5gS42zexh1zVzVk6Uq WRdqigq77SlRx3Yn22JMYtIPmVY4PaozOMNccHUzz3uWiTF6LsdJUFPmKRhJwIM5TRCkdQJ1zZuA BcAVZKilzceKLNTiqpu77JOEHFNkminf7lG6viBEgGfgnxdKWmISc779CxJKESfnQ01jLUO85vNp BQUiRCzVWFOMsdwojy8qXBnhHk+7He/P4qWBpsO3LcFJxXC/2y0WeGNEpbkoaKEZgWhZobk5YVKL cbbmzULTP6RJny00oSHN+erNHkRyr3ifpXTN6j3/s9kz0vxtkzSW8V/UrVF93SxPHY7Gir+rORa1 dfUXlKJ64VD4DboYp6VVB6QcDpngeI5kaKMn5CETHJcDiZqcj5/ZCOr5EYsVUWlImFKr2Zsvi9Jg vNsdu2f+v0oR/ayoZWTfV4tNQ5vkiw00Omxt8gXlC4aX2CXV1WPL4TjZd3XFX6BDqEVHeXksQ8Tj Rd+tVGs0G3dmPn545ZggEMzRVLgMmnrcRYY4mBMrXmCYE6sUxCWis6iwFntmV1etR6ntau5qieih uR7F7unq48/z57va/djtooET0fdtzNFqNwXhsTY2CbcHp6puD9swo1tlG2bNCfThE4eyoT4Q5oxX N1cFxbLw6vpAzC3GgojQxxriHxNEjNxBs4FTIkf4MCOICP+4mHFaJhptGImPJeKSDJpO/8taeDl4 bPfigGwSGguRO8Wdkg+CVuk7Rer/TjXRt6TI/bCk2fqHcoqpH/9x5LLZ3VwMSo3PzpNNw/KSkjLj IvpHvTYbny0YNSspCVwKJQoJZSQmgiYKGRXjnegIyBGqGKgYGDM6p9v8ny19jpbMoqW/9yYlGcwn alZCArgUsgsJ2RMTBRWy801+32avWcmyOxFYYa7gX5vTlAGkAjZ8Vporm9fzDQkbbC8nm+IsCQ7u T7sy/Yqs6pz6tIb0hqyZOYstixPmpS1JX5zVnLOK32BemXCTbb35fss2+8uOk/yE+UTCW7bs8xPv iFPdnnGj4xjF2eN43BZXSgchmNRkSBU8mjltyXtpo8Nn/7LRNwCy3DcQmzprXI6XnAniYkAwmGZP vWRsSUZGarqdmz3DR3rT7BljSy5JsXs9wy3m2YuPP7wy3Fm16Pgjb6y6p2/36tW7d69ZfUUjP85k dumTTT2D+snBwcFDT93/LHtw8L4zZ/HmsOiThXeImDiFBfwWaxdPe1RFUpNSxi2W1/LNfLtVflJm cWQ2cSnOxBI5OxxvjD5ezImYgrr4FNdrt2PpIvo/1BRjQZ3GgiYbCwovq1liuYbWxFif7ESTmmQb ZxryxGgTU/C2xU1ZCftYGVtHDt9V9g/gDPgl9paMQtnUc2VUUZE5kaVMHDOaNVKjz+1JMZst4y+5 pHQs/7a38nj9fe8Vd8o3l692PT3lcJOYm4hLt+kxymNXR6MS3+vg/yQxOuZMjs9LT3emRvh+NcEm y3nOpGRGFgfmYkSrwQhLh4jJ4iPFxtqkTiw+94L9BR/CUx2VisA1z7YZtC57VW5X7ra0x9MOJZ5I fCvHGpfmSC7IluJGm0Yn7EOASQgwe1p8empa2uFk27DktGHJtqQI/5WaJgaiJj+czJOTbWo6iw3q WZvMjqvCy8yhKmJ4KU32Zfa19s122R6ydDiM4HIwctgdHIP9/FkxGMcWJfUAG4//pn5OSWxCOPkZ to9NwJeDfjXh+7DDJ+Ot3UOR9/nA54g+nwi/zxvh5IFzjR9UDGCijSmAfcD+wXrrxT7TLfYXKCVV ON8Iy+WNPl8wrTRDhOH4cQhDs2VkmjvdLSFOKX2YBVHqnf1c+vYlt/U+temaTRftvpu/ee7Zabff 08+snXd9/odzLGTv2vjCozvC0yoy+KdPDq5sGPzyjy/dEz4tdhTj0kfi/e7HLnyCojSc2x2GklEq vCQuM97Jacb0mjlXzPRVXr+wZcnU+v8D768PPAplbmRzdHJlYW0KZW5kb2JqCjM2IDAgb2JqCjUz NzAKZW5kb2JqCjEzIDAgb2JqCjw8IC9UeXBlIC9Gb250IC9TdWJ0eXBlIC9UcnVlVHlwZSAvQmFz ZUZvbnQgL01GS1RWQytBcmlhbE1UIC9Gb250RGVzY3JpcHRvcgozNyAwIFIgL0VuY29kaW5nIC9N YWNSb21hbkVuY29kaW5nIC9GaXJzdENoYXIgNDggL0xhc3RDaGFyIDUzIC9XaWR0aHMgWyA1NTYK MCA1NTYgMCAwIDU1NiBdID4+CmVuZG9iagozNyAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0 b3IgL0ZvbnROYW1lIC9NRktUVkMrQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0z MjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9D YXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0 IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiAzOCAwIFIgPj4KZW5kb2Jq CjM4IDAgb2JqCjw8IC9MZW5ndGggMzkgMCBSIC9MZW5ndGgxIDgxMTYgL0ZpbHRlciAvRmxhdGVE ZWNvZGUgPj4Kc3RyZWFtCngBhVkJfBRF1n9V3XMkmWQ69ySZzPQwZJBMMBCOBIjJ5ASNyJVohg2S AJFwKGAARREGVwQDCrIuK6yK1yrqIp1JwAngkhVWVxRhhWXXYxWvz+O3EfRDXQ/S3796BpRdf/t1 z//Vq/deXa9eVVf3LLlxaSvZKEQSBWZd37KIjCtrAZKBs5YtUaP5RAeRuf66RXOuj+ZToTd9PmfB 8uui+awHkD/c1toyO5qnH5COaoMgmmcjkA5su37JzdF8VhNSZcHCWTF91ibk469vuTnWPr2NvHpD y/WtUXv3fSK/aGH7kli+DWnBohtbY/askSjhDzouoT/2VTWFmsEwQKYvqZQeJAtxUqiQroaoTc4l E/JCb+IbN3+54esZ9tKvrDlWCIge/WBQvkgPyos2fbfr3ByFrDZk4wx7oUA5S1n/VVSl0He7vrtF ibYkNOcvuYfqpUu6fA73sf3SYDoFcGlw2J/r7pEGSbnhse5ARPJ2paQX2SuGSCpqLDSoCroQ2AUc AGSaIbmgVUBXASFgF3AAOAaYiUCFVgUWAtuBU4BZypWcYdWtVAySslA2C+O1S5l0GtABidyghcBE YAawEdgOmA07IVkIrAIOAGcAMwWkzPDm4eh7Zni9kXTNW1BkZFui2abpRrbrmmA0nTA5mlZfHjUb EzUbNiIqvrQymg4qiKYpeUUhVN4Vn1jUW5EhZWCQGej4IlDGD5GdMXLTw1I6aQCX0FVDEpBSugb6 irYfkGRiEpcYzSa33iuxcGJyUUU81/lpSiE3/5z3RTW8ryspuWh7xRX8fdoFHAAk/j7u9/h7tIqf Ej4HLQe2AweAo8BpwMxP4X4X9zv8HbLzf1AhUA7MALYDB4DTgIX/A1Thb4uIMajgywHO3wZV+FsY 1lugdv4muDf5m3ovPx4uHl3UYzD+whjjzosxmTkxJiWjKMJfD387GBHlw0wjovZJA6iMhksDwnnD 3BHJES6d647wD7pUv/vhiqH8BGkAR09OoOUTpAKTgGZgEWAGdxLcSQoBm4CHAQ1AlIEqgMoPA68C J2koEAAmAVZ+LIxmIvxo2Ffprsjgr/GXKBMeP8L/bKSv8heN9BX+JyN9GakL+sP8xbDLTRUJ0BPK KEgVpIXQm/gfuwamuPWKZH4AHnSDFgLlwERgBrARMPMDfEB4tjsFleyjw1jDbh6mT430CXrUSoF5 7oCvCgGoCuIbcxk4kO3qdh8P+LZsRVYQ3z2bwQniu2MDOEF8t6wGJ4hvwTJwgvhmzwMniG/aDHCC +CbWgwOJ8IeeGzjIXTxxPlMr7PwmeOkmeOkmeOkmkvlN4qZvZdHH34bz8+GxbQH/4Hx3aC8L7Weh KSz0KAu1stBKFlrNQqUsdC0L+VnIyUIuFgqw0D5WAleEWKD7ouzogIOFDrPQThZqZyEfC+Wx0EAW UllxIMI94cux6pDUGElXhVh03NN1WRl2Hzv3wKMexLwHe8IB0KOAbuQCMFIHRI2zXCId0JVfHs1f OqZoYcV4fhAFD2IaDtK7gIwJOogwOohKDqI6O2g5MAPoBU4DOmCG9QCMY6NB7aCFQDkwA1gFnAbM RndOoyucFoKKLu4yOlYIWg5MFDl+EPcA3B7uCeQqTsWvjJc2OpndxSa6dBcvpowM7MspydbkCEvc 803iv75JpLiKOH4P30i5mIhNsXRj+Ntcd4TdH/btc1eks9+QS0bUsdHkY3lIS6jdyI8kp1XIR5CT P4O0KOy8GsXsYV+Bey9LEqX2uL91fuj+1BnhYD9x7nP/TY3ILOz+KyTP7HGfcN7lfrkwYoVkvy/C kOxVDdMeZ4l752HDdDUU28LulSLZ477NOc4932koWqOKa9uRC9jdU3zT3ONRX7VzpjvQjjr3uMud 17pLo1YjRZk97qHogj/K5qOzg51Go16XUWFDcYS1BQosWyyNlomWUZYiS4HFY3Fbci05ljRrilWx Jllt1nir1Wq2ylZuJWtaRD8V8IunXprZePiZEdCMZINXsMMwsc2AEmdWTleQlirV8bqplaxO651F dTNV7eup3giLnzxNM3krmZZSR3X1lVqJvy5i0adoxf46zTLpF42djN0ThFTj6yKM6hsjTBeiNTla SlVjDzGWvObuHJFesubuYJAcGcvKHeUpZcmja6t/hjQbwuZq/4+X40fW7/DnalvqpjZqT+cGtSLB 6LnBOu1XU9Wmxh72JTtTU93DvhBJsLFHKmNf1kwRcqmsOhisi7CrDTtS2RewQ8QggZ0VD2ZhR6rV FbXbFrXLQ3nYDRQJ7OLiKM+wy4uLM+xkJuw62wfWVHcOBIFNpkrthk17pvpTm8N5sMkDgU1GiA4b NoczQsJGKzOqcTph4gKBCcsmp2HiZNmGidHzTsOkMGZy1wWTu4yWpGhvDBtBUE3iqfM2iadg8xNH /ne2tdLvZ11jg7Oaalq9Nc3emlagWVu/rM2hhWaqauesoFComuRrnjmrTaQtrVrQ21qtzfJWq51j jXL/pm4S6rHe6k5qqqlv7GwKtFaHxwbG1nhbqoNd4yaNKL6orbsutDVi0s+0NUlUNkK0Nc4o929t FQv1ONFWsWirWLQ1LjDOaIuMGJ/U2GmlymAV5k+kXTwhHvHanOMJVmYoi8qM4B3rcazM2YvTyg5K 8Ac1m7dSSwREXA+pGFIhVFhTQpUEsT2mcqwc68nZy3bEVArEyd5K8i9Z2r6UHDVzq6O/dlwQLVkq piJK/UL2sxdMarRAS7U4W9dp+VPrtPLJ0xo7LRZIm6uDkI05L0tIqInovVHhpRCOEYaSdMFQyEqF LC4uZvifsWD0CWJ4pwcHjX1dLOBiS6g9KGmuunqOraB+GtzQNK1xL85S4iHRHsQA25mftZ+vTYzD 4CkqIQy7/TyWLI1xMV8siaWGabuf/O3nXXK+Or9wlkEMXy3xY2sz7aUsINv0JGXJPsL7j/4x8IlI ++fqnwi9SPln2OgiMRDtoJ1sLu2kA/QCO4NSu6iHukkcgarpAVpB99FaPNamQXIXTcFtgvw+lqV3 483kETwwH6EjsL2GVtJeymAO/VNaRWuk4yi1hhJpAFXQJFpId7Mr9aXURO/Kv6RiupJuoEUspDfq 9+ib9cfpd9Qj/Vk/RwmUTbNwH9E/N/1df5uGoMSvaSu9yzbH7aYAWgnB8kG6kbZJ02Wmz9G/Qw88 dBP6INMEOsJ6uR+1t9LHzMFWSFWo5TFd0w/ByknTqY220V42ko3jHlOTPkE/Qhlo42bUupXCtAd3 hJ6nN5nNdEZ/XD9DWVRAl2M83fQa65X6z63uL4ffTPDSYBoNzUL6A71Ex5iX/ZEvNNlMRaaA6Rb9 BKXRMGpAb59Eyf9h3/CVuFdJL8q1eiUlwS/3Cm/Tn+g9ls0K2UR2NR/MF/KHpBvJihaH4Z5Nc+Hv +1H7OwijPdzGj0qPyc/I35tz+0/pSZgRH/2WHqQ/skSMVGXt7HZ2kn3Aq/gM/lv+vnSf/JT8uqUF o76Wrqe76Rn6hqWwEjaZ/YK1sRVsLbuXbWVH2DH2Ca/g9Xw+Py21SYul5+VK3FPldvmXpjtN682f 9Df2H+r/S/83epF+J01GPKxG739ND2FkPXSU3sD9Lr3PTCyBJeFWmYc1sFtxr2R3s0fZDvYU60Yr x9j77FM8kr5i33M8abmZ5+DwI45AXn4jTpj38Qf4UdzH+D/5t1KmNEDySyOlUikoLUSv1kqbcO+W 3pOz5aOyDj8XmbaYtpt2mJ4xvWA6Y7ZZbscz/tUfHjuXf+6dfupf17+lP9zfrb9H6ZhDPD3wClaK 3rfgnof53oKI20XHmQ2+y2b5rIxdCc/MYPPYYnYzPHkH28Z+Z/T9WbYfXvobO40+J3Kn0edL+Uhe ySfivpa38sU4jG3m3fwk/06ySAmSXUqX8qVx0nSpVVoiLZe2SJr0qvQP6X3pa+kH3LocL7vlAbJP 9svj5BnyUvkh+WP5Y1OT6RXTR+Z48/XmO80R8xc41ZRZJlkmW6ZbNlr2WE5YmxGdB2k3PYcIvHCx U9JqqUbaTffw4XIWXmFeQzzPoNnSBI5I5TvYOn4b6+YDTTebx/Kx7Co6I/vg6xf5dv41HytNYHVs Ks3jw6IVmtPkp8GVygepT96Psb2Gmm8229hKftpsozDOSKNxRvqTNFT2S6/Qm9K7zCI/Qm/J8SyT 9fEnpUmIguflMlMjeaQH6FlpMbuNdvMafCj53roBcXwVexr7Qj0rYv+SdByDr0IUFUsf0C9pPv87 9WEdr6PfsNnyHLqHhrMV9DE9gVUx2HSDOd+czl7mc+UOnsq6ictPYXSj2UAmmdLoDjZd2mY+zd+g pXRUjqd3pN+j90f5s9IE+YxpCmvDCriN7qTF+mpabmqUX2dzSGJXU558CrvbCqlI9iBdhV2lCXva HqzuvdgHKqQJkDgQOVciLhqwQ2zDfT/2CRkRNBdr/BrsYq9Rt7meR2iOKYlh18GXmlf6p9A0/Qna qs+hG/TNNAT7wVp9BWrcQR/RRtrB1vTfSovwKvkG1vaVplp+1FSrD+Ed/A0+lW+5eH7h7TzmoM9w P4uZKTPtow75bzSVyvUN+l8R3Zdgh91KM3Fg/RCj/BwtjJd6aXj/VbxTr5UWYbzv0mT9Sd3N4qlN X0ATaT/9zmKiFosfc6yx1zHeW6mVT9GXSK39c+GHjfBCAN5aiv3nrkBVQ31FoLzsstKxY0aXFI8c Mbxo2NDCS4cU+PMHXzLIlzfQO8Cjul25zpzsLEdmRnpaakqyYk9KtCXEx1ktZpMscUYFNd7aZlXz NWuyzzt+/BCR97ZA0PITQbOmQlR7sY2minItUF1kGYDldf9mGYhaBi5YMkUtpdIhBWqNV9WOVHvV CJs2uRH83dXeoKr1GfwEg99k8IngPR4UUGscbdWqxprVGq12WVtHTXP1kALWmRBf5a1qjR9SQJ3x CWATwGmZ3kWdLLOMGQzPrBnTycmaiCFq2d7qGi3Li6KoRsqraZmtTZrcWFOd4/EEhxRorGqWd6ZG 4qTkN0yoymhGM1dpFqMZdS7OOBqtVzsLejs2RBSa2ey3zfbObmlq1KQW1FGjJfvRbrWWecuHjh+z qBxnsrU/1eZIHTWOuaow7uhYq2oPT278Sdkcj6ghGEQdKMvzaps7atH0BsxUnTiLa3xNsFFja9Ak DpZ5xqii44ueevOa56lanLfS29YxrxlTk92h0ZTlnnB2dqBHP0XZNWpHfaPXo5XneIMt1c7ONOqY srwrK6BmXawZUtCpJEcd25lkjzG2xJ8yrXB6VGdwhrng6qZc8CwTffRejpOgps5S0ZNGL8ZUIkhr CXXMKsEE4AoylNJmY0bmanFVzR3KGCHHEJlmylO8asdXhAjw9v3zYklLTGLOU74ioRRxciHUNNZy ntf8fi0/X4SIpQpzij6WGfmRQwqWRbjXu0jB+7N4aaBJ8G1LcEwh3O/xiAleHwnQTGS00OTGaF6l mTlhChTibM2bhab3vCa9QWhC5zUXijd7Ecnd4n2W0jWr78LPrmSk1rSN0VjGf1G3RvV1U711OBqr NR3Nsaitq78oF9ULh8Jv0MU4LbWqUcrhkAmO50iGNnpCPm+C43KjTZPz8DMbQT07YrEiKg0JU2s1 pXl8lAbjPZ7Ymvn/CkX0M6KUkfxYLDYMbYw/1tFot7WxF+Uv6p6tQ6qrx5bDcbLv6Ii/SIdQi/by 8liCiMeLvket0qgBKzMPP7xylAgEc7QAXAZNPVaRIQ7mxLIXGebECgVxiegcUlCLPbOjo9ar1nY0 d7RE9NBMr6p4O3r4C/yFjkU12O2igRPR967P0Wo3BOGxNjYGy4NTZaeXrZvcGWDrpk5r7MEnDnVd fWOYM17VXBkU08Kr6htjbjEmRIQ+5hD/mCBi5HZqAEpFivBhRhAR/nEx47RMNNQwwuMXPD7lGzSd /pdb+Ufgsd2LA7JJaCxEnmRPch4IaqUfVKn3h4CJvidV7hXlG/SP5WRTL/7jyGUNnVx0KhCf7ZJN aa7ExMy4iP5Jt93OGwQTyEpMBJdMNiGhDJsN1CZkVIh3oiMgR6i8r7xv2NCcTvN/1nQWNZlFTf/T nZhoMJ8HshISwCWTIiSk2GyCCtmFKn+ss9usZilOBFaYq/jX5hRlACmAHZ+VZsrmtXxdwjr7y0mm OEuCg9ekXpl+RVZVTn1qU3pT1pSc+Zb5CbNSF6TPz2rOWc5vMi9LuMW+1ny/ZYvysuNNftJ8MuEt e/aFgbfHBTzeEUPjGMUpcTxukzu5nRBMgSRIVTyaOW1yvbTe4Ve+nu7vA1ns74sNnU1fjJecEnEx IBhMVVJGDS/KyEhJV7jZO2CQL1XJGF40KlnxeQdYzA3zjz+8LLykct7xR04sv7fnqRUrnnpq5Yor pvPjTGaX/X5GV7/+Zn9//8Gd9z/HHuz/zekzeHOY9/ncO0VMlOKkY8HcudhLMX/HJSuJjtRUc0Ni RD/bnZxsMJ8H4hQFnCvN5BLTmCkMXC6hdTmToHFhEkEjfF/AxuMzM/HvTzLnqjs5ZXThiSOCHqHC PvHaWy7oIRw/cmKhIhq0paRwo8FAnD0ZXLSdU4GElFTe4EoTMlF3GFWLcEpI4A1g/hlIEqHzc62J OBLtidaMxgKjxprGmveZDpj3WV6yvuy0XG4L2uqT5ttmJ92SckvqXSn7Uz7K/ijnTLbtQMJzqTwH H3NzFZdi/gNeHy0IECvSOP1MINsVr1jN5sPO7DSnM9vqzJYYt2Y7pUSXEuGPd01MZvjU69gtRkCG O+yM2+LbM4/D2yIe2D6+mlRSWEnAlry7HK95C/kqLvO9fCA+6G7sjAbEWQSEv1Q5i4goPVda3ndu +ofJKZmjmSBrky71J92mHEJm9LChRpiIKCmh6Wz6jcFgXrrHVzxqFH4jRyA8zJZBo0TspKeZzRb8 ZMsPxTwz77Ftp3dsvfX2B1hP6r/+cvzr8U++8GiTa+fOitJZvSsPfXTd/F890JF69I3PdjY+vf/x dS3DEClilXtMTyBSro5OHL5+IpoTxQwwZ1K8Kz3dmSLmP8Euyy5nYhIjiyOif2asfYMRlg6xwguP FBqRjqg4d0g55EcsBAanGBFkN2hd9vLcjtwtqU+mHrSdtL2VY41LdSTlZ0txQ01DE/ZiNiQsVyU1 Pj0lNfVwkj0tKTUtyZ4I/wdSRUcCSQ8n8aQkeyCdxTr1nF1mxwMYBeYmoIruJc9QFiqrlI2KrIQs 7Q5jqToYORQHR2fPPifC2bFJTdnPRuKfvl9TIisJJ+1me1kJvsP0BhJ+XMT4AL85Nm3+vrO4MXVi MZ+dXopRYubK+zDQ6cmA0qd8uNZ6qd+E6SNj+oyZY4un+/3B1OIMsajPz1qqJ90jYeYoPc2CNe9r eD5964Lbu3duuGbDJU/dw98499zEO+7tZdYld5/98zkWUjrWH3p0W3hieQb/4vf9y5r6v/7LS/eG T2EXx0ZjXPogvC//3IVPepRK6XhHEhejFEBcZnzjoAm1dfVXV/krbpzbsmBC/f8BWmV0uAplbmRz dHJlYW0KZW5kb2JqCjM5IDAgb2JqCjU2MjIKZW5kb2JqCjkgMCBvYmoKPDwgL1R5cGUgL0ZvbnQg L1N1YnR5cGUgL1RydWVUeXBlIC9CYXNlRm9udCAvRFBNR01UK0FyaWFsTVQgL0ZvbnREZXNjcmlw dG9yCjQwIDAgUiAvRW5jb2RpbmcgL01hY1JvbWFuRW5jb2RpbmcgL0ZpcnN0Q2hhciA0OCAvTGFz dENoYXIgNTMgL1dpZHRocyBbIDU1NgowIDAgMCAwIDU1NiBdID4+CmVuZG9iago0MCAwIG9iago8 PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9EUE1HTVQrQXJpYWxNVCAvRmxhZ3Mg MzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50 IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAv WEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRG aWxlMiA0MSAwIFIgPj4KZW5kb2JqCjQxIDAgb2JqCjw8IC9MZW5ndGggNDIgMCBSIC9MZW5ndGgx IDc1NDggL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLfBTV1T/3zuwjz90kJNkk m+wsSxbJJgYCmPAw2Tw2oBF5JbpLg0mAlPCSYAKIRVisCC5P+SwVWgVtq1CrTDYRN4AlFaoVRahQ WtEKPlq1v0bQz0d9kPn+d3ZBaf31m8n/3HPPOfd1zpk7dzaddyxtpUQKkkTe2Yta2km/stpQDJm9 rFOJ1pNsRMb6H7bPXRStpy0kMnw0d+GKH0brWQ8SSbvbWlvmROv0Dcrr2iCI1tkolEPaFnXeGa1n NaJMXLh4dkyftRF186KWO2Pj05uoK7e3LGqN2ucKe6V9cUdnrC76u6b9jtaYPfMTJfxWwyX0Jz+r oWAzGAZI9AmNp4fJRJysVEy3EMltci4ZUBd6A9+y7ZONnzdZxn9mzjFDQPTYu0MLRHlEbt/61b5L c61kTkQ1TrcXCrQzlQ/cTNVW+mrfV3dZoyMJzeVL6qV66Zput81x8pA0jM4DXBoW9uQ6eqWhUm54 nMMbkVzdqekllsoiSUGPxTpVQBcD+4DDgExNUh60VtDVQBDYBxwGTgJGLDBP1yooFwO7gPOAUcqV 7GHFYa0cKmWhbRbWa5Ey6QKgARI5QIuByUATsAXYBRh1OyFZDKwGDgMXASN5pczwtpGYe2Z4g150 z19YoldbotXGmXq1+9ZAtJw0NVrW3BA1Gxs1GzEqKr62KloOLYyWqfklQXTeHZ9U0leZIWVgkRmY eDso40fJwhg5aLeUTirAJUxVl3il1O4h7pJdhyWZmMQlRnPIofVJLJyUUlIZzzV+gVLJwT/i/VEN 7+9OTinZVXkjf4f2AYcBib+D+23+Nq3m54XPQSuAXcBh4ARwATDy87jP4X6Lv0UW/lcqBiqAJmAX cBi4AJj4X0Gt/E2RMToVfAXA+ZugVv4GlvUGqIWfBXeWn9X6+Klw6ZiSXp3xFMcYR36MycyJMakZ JRH+WvjLYcgoNyKNjDooDaZyGikNDuePcEQkW3j8PEeEv9uteBy7K4fz06QCHDM5jZFPkwJMAZqB dsAI7gy4MxQEtgK7ARVAloFaAYUfA14BztBwwAtMAcz8ZBjDRPiJsLvKUZnBX+UvUiY8fpz/QS9f 4S/o5cv893r5Eso86I/xF8J5DqpMgJ7QxorSirIYegP/XfeQVIdWmcIPw4MO0GKgApgMNAFbACM/ zAeH5zhS0clBOoZn2MHD9KFePk6Pmck73+F1VyMBFUHcY68HB7JL2eXmXvf2HagK4t68DZwg7ns3 ghPEfdcacIK4Fy4DJ4h7znxwgrhnNIETxD25HhxIhD/y7JChjtLJC5hSaeHL4aXl8NJyeGk5yXy5 uOlLWczxZ+GCAnhsp9czrMARPMCCh1hwGgs+xoKtLLiKBdew4HgWvI0FPSxoZ8E8FvSy4EFWBlcE mbfnquoYr40Fj7HgUyzYwYJuFsxnwSEsqLBSb4Q7wzfgqUPh04vuSvHQcWf39eXYfSzcCY86kfNO 7AmHQU8Aml7zwkgZHDXOyhPl4O6Cimj92rEliysn8iNoeARhOELnABkBOoI0OoJOjqA7C2gF0AT0 ARcADTDCejDWsUWnFtBioAJoAlYDFwCjPp0LmAqnxaBiivv0iRWDVgCTRY0fwT0Yt5M7vblWu9Vj nShtsTNLHpucp+XxUsrIwL6cmmJOibCk/V8k/euLJIqrjOOb+RbKRSC2xsot4S9zHRH2UNh90FGZ zn5KeTKyjo0hN8tHWUYden002c1CPors/EmUJWH7LWhmCbsLHQdYsmi13/Gl/T3Hh/YIB/uB/aDj z0pEZmHHnyB5cr/jtP1+x0vFETMkh9wRhuKAopv22sscTx3TTddAsTPsWCWK/Y677RMcC+y6ojWq uK0DNa/FMc09wzER/dXYZzm8Hehzv6PCfptjfNRqtGiz3zEcU/BE2QJMdphdH9SVp3fYUBphbd5C 03aT3zTZdJ2pxFRocpocplxTjmmQOdVsNSebE83xZrPZaJbN3EzmQRHtvNcj3nqDjPrLz4iEZiTr vBU7DBPbDChxZuZ0I6lpUh2vm17F6tS+2VQ3S1E/n+6KsPipM1SDq4qpqXVUV1+llnnqIiZtmlrq qVNNU37g72JscwBSla+PMKr3R5gmRGtz1NRqfy8xlrJ2U44or1m7KRAgW8ayCltFannKmNqa7yHN urC5xvPtZfuW9dg8uer2uul+9de5AbVEMFpuoE79n+lKo7+XfcIu+mp62ceiCPh7pXL2iW+akEvl NYFAXYTdotuRwj6GHTIGBezMeaQIO1LMeVG7nVG7fLSH3RBRwC4ujvJ1u/y4ON1OZsKuq2OIr6Zr CAhsMhXq0G06MpXv2hzLh00+CGwygnRMtzmWERQ2arnejd0OkzwQmLBssusmdpatm+gz79JNimMm 918xuV8fSYrORrcRBN0knb9sk3QeNt9x5H9nW6s8HtY9LjC70dfq8jW7fK1As7phWZtNDc5SlK7Z AaFQVMndPGt2myhbWtWAq7VGne2qUbrG6e3+Td0o1ONcNV3U6Kv3dzV6W2vC47zjfK6WmkD3hCmj Sq8a6/4rY42a8j1jTRGdjRJjTdDb/dtYpUI9QYxVKsYqFWNN8E7QxyI9x6f4u8xUFahG/ETZzRPi ka/NOc5AVYa1vVxP3nFO26qcAzit7KEET0BNdFWpSYDI66LKokqhwjMlVMkQW2Iq26pxzpwDbE9M ZYU4xVVFns6lHUvJ5ptXE/3rwAVR51IRiij1CNn3XjDxqd6WGnG2rlMLptepFVNn+LtMJkibawKQ jb0sS0jwRbS+qPBaCMcKQ0m6Yihk44UsLi5m+J+5oM8JYninFweNg93Mm8c6qSMgqXl19RxbQf0M uKFxhv8AzlLiJdERwAI7mId1XO5NrEPnKSohLLvjMjqXxriYLzpjpW7a4SFPx2WXXO7OI5ylE91X nR5sbYYDlAVkG56gLNlN+P7R3gc+EOXAPO0DoRcl/wc2ukgMRHvoKTaPnqLD9Dy7iFb7qJd6SByB aujntJIepHV4rc2A5H6ahtsA+YMsS+vBl8mjeGE+SsdheyutogOUwWzah7Sa1kqn0GotJdFgqqQp tJg2sZu0pdRI5+QfUyndRLdTOwtqfm2ztk37Jf2KeqU/aJcogbJpNu7j2keGv2hvUhFa/IR20Dm2 Le4Z8mKUICwfpjtopzRTZtpc7SvMwEnLMQeZJtFx1sc96L2V3mc2tlKqRi+/0FTtKKzsNJPaaCcd YKPZBO40NGqTtOOUgTHuRK87KEz7cUfoOTrLEg0XtV9qFymLCukGrKeHXmV90sClNQMV8JsBXhpG Y6BZTL+lF+kkc7Hf8cWGREOJwWu4SztNg2gENWC2T6Dl39kXfBXu1dILcq1WRcnwywPC2/R7eptl s2I2md3Ch/HF/BHpDjJjxBG459A8+Psh9P4W0mg/T+QnpF/IT8pfG3MHzmvJiIibfkYP0+9YElaq sA52DzvD3uXVvIn/jL8jPSjvlV8ztWDVt9Ei2kRP0hcslZWxqewHrI2tZOvYA2wHO85Osg94Ja/n C/gFqU1aIj0nV+GeLnfIPzbcZ9hg/GDAP3B04I8DX2gl2n00FfmwBrP/CT2ClfXSCXod9zl6hxlY AkvGrTAna2A/wr2KbWKPsT1sL+vBKCfZO+xDvJI+Y19zvGm5kefg8COOQC5+B06YD/Kf8xO4T/J/ 8i+lTGmw5JFGS+OlgLQYs1onbcX9jPS2nC2fkDX4ucSw3bDLsMfwpOF5w0VjoukevONf+eYXlwou vTVAA+sHtg+EB3q0tykdMcTbA59g4zH7FtzzEe/tyLh9dIolwnfZrICVs5vgmSY2ny1hd8KT97Kd 7Ff63J9mh+ClP7MLmHMSt+tzvpaP5lV8Mu7beCtfgsPYNt7Dz/CvJJOUIFmkdKlAmiDNlFqlTmmF tF1SpVekv0rvSJ9L3+DW5HjZIQ+W3bJHniA3yUvlR+T35fcNjYaXDX8zxhsXGe8zRowf41RTbppi mmqaadpi2m86bW5Gdh6hZ+hZZOCVi52X1kg+6RnazEfKWfiEeRX53ERzpEkcmcr3sPX8btbDhxju NI7j49jNdFF2w9cv8F38cz5OmsTq2HSaz0dEOzQOkn8Nbrx8hPrlQ1jbq+j5TmMiW8UvGBMpjDPS GJyRfi8Nlz3Sy3RWOsdM8qP0hhzPMlk/f0Kagix4Ti43+Mkp/Zyelpawu+kZ7iOK/9q8EXl8M/s1 9oV6VsL+JWk4Bt+MLCqV3qUf0wL+F+rHc7yefsrmyHNpM41kK+l9ehxPxTDD7cYCYzp7ic+TQzyN 9RCX92J1Y9gQJhkG0b1sprTTeIG/TkvphBxPb0m/wexP8KelSfJFwzTWhifgbrqPlmhraIXBL7/G 5pLEbqF8+Tx2t5VSiexEuRq7SiP2tP14ug9gH6iUJkFiQ+bchLxowA6xE/dD2CdkZNA8POO3Yhd7 lXqM9TxCcw3JDLsOfql5eWAazdAepx3aXLpd20ZF2A/WaSvR4x76G22hPWztwI+oHZ+Sr+PZvslQ y08YarUiHuKv8+l8+9XxhbfzmY3+gftpRKbccJBC8p9pOlVoG7U/IbuvwQ67g2bhwPoeVvkRRpgo 9dHIgZt5l1YrtWO952iq9oTmYPHUpi2kyXSIfmUyUIvJgxir7DWs90fUyqdpnVLrwDz4YQu84IW3 lmL/ud9b3VBf6a0ov378uLFjykpHjxpZMmJ48bVFhZ6CYdcMdecPcQ12Ko68XHtOdpYtMyN9UFpq itWSnJSYEB9nNhkNssQZFfpctc2K6m5WZbdr4sQiUXe1QNDyHUGzqkBUe7WNqoh2LVBdZemF5Q// zdIbtfResWRWZTyNLypUfC5FPV7jUiJsxlQ/+E01roCi9uv8JJ3fqvNJ4J1ONFB8trYaRWXNik+t XdYW8jXXFBWyroT4ald1a3xRIXXFJ4BNAKdmutq7WGY50xme6RvbxcmchCWq2a4an5rlQlN0I+X7 WuaoU6b6fTU5TmegqFBl1bNds1QSJyWPbkLV+jCqsVo16cMo83DGUWmD0lXYF9oYsdKsZk/iHNec lka/KrWgD5+a4sG4NWrmXe/Zvq2ic5zJ1n1XmyOFfLZ5ijAOhdYp6u6p/u+0zXGKHgIB9IG2PL+2 OVSLoTciUnXiLK7ytQG/ytZiSBws8/VVRdcXPfXmN89X1DhXlastNL8ZockOqTRthTOcne3t1c5T tk8J1ftdTrUixxVoqbF3DaLQtBXdWV4l62pNUWGXNSXq2K5kS4xJTPou0wqnR3U6p5sLrm7aFc8y MUfXDTgJqspsBTPxu7CmMkFayyg0uwwBwBVgaKXOQUTmqXHVzSHrWCHHEplqyLe6lNBnhAxw9f/z aklLTGLMt35GQiny5EqqqazlMq96PGpBgUgRUzViijmW6/XRRYXLItzlarfi+1l8NNAU+LYlMLYY 7nc6RYA3RLw0CxU1ONUfrSs0KydM3mKcrXmz0PRd1qQ3CE3wsuZK82YXMrlHfM9Sump2X/mzWDPS fG1jVZbxX9StUX3ddFcdjsaKL9Qcy9q6+qtqUb1wKPwGXYxT06r9Ug6HTHA8R9K10RPyZRMcl/2J qpyPP6Oe1HMiJjOyUpcwpVa1Nk+M0kC80xl7Zv6/RhHtomilF982iy1DHeuJTTQ6bXXcVfWrppcY kurqseVwnOxDofirdEi16CxviBXIeHzoO5VqlRrwZObjD58cZQKBHNULl0FTj6dIFwdyYtWrDHNi jQK4RHYWFdZizwyFal1Kbag51BLRgrNcitUV6uXP8+dD7T7sdtHEiWgHNuSotRsD8FgbG4vHg1NV l4utn9rlZeunz/D34icOZX29P8wZr26uCoiw8Op6f8wtekBE6iOG+I8JMkbuoAYBpA7TEwj/94BC /CA1XDcAo1/iBxOkF/0vO4lfTcAacOOkbSJypjhT8kHQG32jSH3feA30NSlyH17b1KC9L6cY+vC/ jVzW0MXFZLzx2XmyYVBeUlJmXET7oMdi4Q2C8WYlJYFLoUQhoYzERNBEIaNifAsdBzlOFf0V/SOG 53QZ/7OnT9GTUfT0956kJJ35yJuVkAAuhaxCQtbEREGF7EqX3/bZY1SyrHYkVJgr+G/NecoAUgEL fk6aJRvX8fUJ6y0vJRviTAk27ku7Kf3GrOqc+rTG9MasaTkLTAsSZqctTF+Q1Zyzgi83Lku4y7LO +JBpu/Ul21l+xngm4Q1L9pWFd8R5na5Rw+MYxVnjeNxWR0oHIYm8yZAqeCVz2pr34gabx/r5TE8/ yBJPf2zpbOYSfNyUiYsBgUCaNfW6kSUZGanpVm50DR7qTrNmjCy5LsXqdg02GRsWnNq9LNxZNf/U o6dXPNC7d+XKvXtXrbxxJj/FZHb9b5q6B7SzAwMDR5566Fn28MBPL1zEF8P8j+bdJ/JBxM5peJzy 2C3RyOG3LMwxSYSE2ZPj89LT7akRftCbYJHlPHtSMiOTLaL9Q4+ozghLm4hb8fFiff6pY4ovHbUe 9SCE3mGpCK6xwaLTuuwVuaHc7WlPpB1JPJP4Ro45Ls2WXJAtxQ03DE84gCBICII1LT49NS3tWLJl UHLaoGRLUoT/0psmJuJN3p3Mk5Mt3nQWm9SzFpmd8mIVEWbzKmJ6KU3WxdbV1i1W2Ro0ddj0ANgY 2aw2jsl++qyYjG2rknqIjcb/bX5CSawsnPwMO8DK8FXd5034NjT4OXVb1+XofNr/KSLkESH6dOZ4 rPLSzPcq+rHQmSmAtd/63jrztR7D3dajlJI6ZsRwEbUytmSmxxNIK80QoRo9CqEymoamOdOdEmJJ 6YNMiKS74bn0HQvv6Xlq460br9m7mb9+6dnJ9z7Qx8ydmz79wyUWtIY2HH1sZ3hyRQb/+DcDyxoH Pv/jiw+Ez4unTr+0ofj2+b5LgjANp10R49TYU2/EtyrVTJk0YVK9p/KOeS0LJ9X/H4iA20oKZW5k c3RyZWFtCmVuZG9iago0MiAwIG9iago1MjAzCmVuZG9iagoxNCAwIG9iago8PCAvVHlwZSAvRm9u dCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9BUk5ZUE0rQXJpYWxNVCAvRm9udERlc2Ny aXB0b3IKNDMgMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDQ4IC9M YXN0Q2hhciA1MSAvV2lkdGhzIFsgNTU2CjAgMCA1NTYgXSA+PgplbmRvYmoKNDMgMCBvYmoKPDwg L1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250TmFtZSAvQVJOWVBNK0FyaWFsTVQgL0ZsYWdzIDMy IC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAgMTAwNl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5 MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0IDcxNiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMgL1hI ZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lkdGggNDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250Rmls ZTIgNDQgMCBSID4+CmVuZG9iago0NCAwIG9iago8PCAvTGVuZ3RoIDQ1IDAgUiAvTGVuZ3RoMSA3 NTc2IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4AYVZC3xUxdU/M/fuI8ku2YQ8Nskm ezdLFmETAwEML5PNYwMakAABd2kwCUkk4SHB8LQIi8orIFA+S4WqqNWKtcrNJuIGtESxWlGETyit aAUfVeyvCPXzUV+5/c/dBaX11+9O/mfOnHPmdebcuTObJbcubSELhUgiX9PCxnbSn4xWZErTsiVK tGy1Exlrb26fuzBaHriAyPDJ3AUrb46WM7YTSX2tLY3N0TJ9i/yaVgiiZTYS+aDWhUtWRMsZdcgt CxY1xfQZG1E2L2xcEeuf3kZZuaVxYUvUPlvUU9oXdSyJlWuQX9V+a0vMngWIEn6n4RH6459XUqgB DAMk+pTG0/1kIk42KqSZRHKrnE0GlIXewLft+HTLF/WJ4z83Z5khIHr4/cFDRX5Ybt/+9b7v5trI bEExTrcXCtQzlfTfQBU2+nrf17fZoj0JzaVH6qVa6apuj915/FlpCJ0FuDQk7M129kqDpezwOKcv Irm7k1OLEssKJAUtFupUAV0E7AMOATLVSznQ2kDXACFgH3AIOA4YMcEcXasgXwTsAc4CRilbcoQV p61ssJSBuhmYb6KUThcADZDICVoITAHqgW3AHsCo2wnJImANcAi4CBjJJ6WHd4zA2NPDm/Wse96C Ir3YGC3WzdaL3TcGo/nkqdG88rqo2dio2fCRUfHV5dF8cH40T84rCqHx7nhrUV9ZmpSGSaZh4O2g jL9IiYyRkx6UUkkFuISh6hKflNw9yFO055AkE5O4xKiZnFqfxMLWpKKyeK7xC5RMTv4JPx/V8PPd A5KK9pRdz9+jfcAhQOLvIb3L36U1/KzwOWgpsAc4BBwDLgBGfhbpDNI7/B1K5H+hQqAUqAf2AIeA C4CJ/wXUxt8WEaNTwZcCnL8NauNvYVpvgSby0+BO89NaHz8RLh5T1Ksz3sIY48yLMelZMSY5rSjC 3wh/NQQR5cFKI6IOSrlUQiOk3HDecGdEsofHtzkj/P1uxet8sGwYP0kqwDGSk+j5JClADdAAtANG cKfAnaIQsB14EFABRBmoDVD4EeA14BQNA3xADWDmx8PoJsKPhT3lzrI0/jp/mdLh8aP8D3r+Gn9J z1/lv9fzV5DnQH+EvxTOcVJZAvSEOjbkNuSF0Bv4892Dkp1aWRI/BA86QQuBUmAKUA9sA4z8EM8N NzuT0chBOoJ32MnD9LGe/5oeNpNvntPnqUAAKoJ4xl4LDmSPssfDfZ6du1AUxLN1BzhBPHdtASeI 57a14ATxLFgGThBP8zxwgnhm1YMTxDOlFhxIhD/wzKDBzuIp85lSlsiXw0vL4aXl8NJykvlykegr WYzxl+GhQ+Gx3T7vkKHO0AEWepaFprHQwyzUwkKrWWgtC41noZtYyMtCDhbKYSEfCx1ko+GKEPP1 XFEc47Oz0BEWepKFOljIw0J5LDSIhRRW7ItwV/g6vHXI/HrWXSZeOu7qvrYEu08id8GjLsS8C3vC IdBjgKaXfDBScqPGGTkiz+0eWhotXz22aFHZRH4YFQ9jGQ7TGUDGAh1GGB1GI4fRXCJoKVAP9AEX AA0wwjoX89im00TQQqAUqAfWABcAoz6cCxgKp0WgYoj79IEVgpYCU0SJH0bKRXJxly/b5rB5bROl bQ6WmMOm5Gg5vJjS0rAvJyeZkyLMuv9L6z+/tFJcWRzfyrdRNhZieyzfFv4q2xlh94Y9B51lqewX lCMj6tgY8rA85KOpQy+PIodZyEeSgz+BvCjsmIlqiWFPvvMAGyBq7Xd+5fjA+bEjwsGecxx0/kmJ yCzs/CMkT+x3nnRscr5SGDFD8qwnwpAdUHTTXsdo55NHdNO1UOwOO1eLbL/zdscE53yHrmiJKm7q QMmX6JzmmeWciPYqHXOcvg60ud9Z6rjJOT5qNUrU2e8chiF4o+xQDHaIQ+/UnaM3OKM4wlp9+aad poBpiukaU5Ep3+QyOU3ZpixTijnZbDMPMFvM8Waz2WiWzdxM5pSIdtbnFV+9FKP+8TMioBnJOm/D DsPENgNKnJk5XU/qQKmaV08vZ9VqXxNVz1HUL6a7Iyx+6izV4C5nanI1VdeWq6O91RGTNk0t9lar ppqfBLoY2xqEVOUbI4xqAxGmCdG6LDW5ItBLjCWtuztL5FetuzsYJHvaslJ7aXJJ0piqyh8hDbqw odL7/WP/nvXavdnqzurpAfU32UG1SDBadrBa/Z/pSl2gl33KLvore9k/RBYM9Eol7FP/NCGXSiqD weoIm6nbkcL+ATtEDDLYmXNIEXakmHOidrujdnmoD7tBIoNdXBzl6XZ5cXG6ncyEXVfHIH9l1yAQ 2KQr1KHbdKQrP7Q5kgebPBDYpIXoiG5zJC0kbNQSvRmHAyY5IDBhmeTQTRwsUzfRR96lmxTGTDZd Ntmk9yRFR6PbCIJmrGcv2VjPwuYHjvzvbEu518u6xwWb6vwtbn+D298CNKibl7Xa1dAcRelqCgqF okqehjlNrSJvbFGD7pZKtcldqXSN0+v9m7pOqMe5K7uozl8b6KrztVSGx/nG+d2NlcHuCTUji6/o a9PlvkbW/EhfNaKxkaKvCXq9f+urWKgniL6KRV/Foq8Jvgl6X6THeE2gy0zlwQqsn8i7eUI84rUh yxUsT7O1l+jBO85lX511AKeVvZTgDaoWd7lqBURcF5QVlAkV3imhGgBxYkxlXz3OlXWA7Y2pbBAn ucvJu2Rpx1Ky+9sqo38deCBaslQsRZR6hexHH5j4VV9jpThbV6tDp1erpVNnBbpMJkgbKoOQjb0k S0jwR7S+qPBqCMcKQ0m6bChk44UsLi5m+J+xoI8JYninFweNg93Ml8OWUEdQUnOqazm2gtpZcEPd rMABnKXER6IjiAl2MC/ruNSamIfOU1RCmHbHJSxZGuNivlgSy3XTDi95Oy655FJzXuEsnei+WuLF 1mY4QBlApuExypA9hPuP9hFwTuT9bdo5oRc5/xs2ukgMRHvpSdZGT9IheoFdRK191Es9JI5AlXQf raJ7aAM+a7Mg2UTTkAyQ38MytB7cTB7CB/MhOgrbG2k1HaA0Ztc+pjW0TjqBWuvISrlURjW0iO5m k7SlVEdn5DupmCbRLdTOQlpA26rt0B6hR6lX+oP2HSVQJjUhHdU+MfxZe5sKUOPntIvOsB1xT5MP vYRgeT/dSrul2TLT5mpfYwQuWo4xyDSZjrI+7kXrLfQRs7NVUgVa+ZWmai/CykGzqZV20wE2ik3g LkOdNlk7SmnoYwVa3UVh2o8UoefoNLMYLmqPaBcpg/LpOsynh15nfVL/d2v7S+E3A7w0hMZAs4h+ Ry/TceZmz/NFBouhyOAz3KadpBQaTjMw2sdQ80P2JV+NtEZ6Sa7SymkA/PIz4W36Pb3LMlkhm8Jm 8iF8EX9AupXM6HE4UjO1wd/3ovV3EEb7uYUfk34lPyF/Y8zuP6sNwIp46Jd0Pz3PrJipwjrYHewU e59X8Hr+S/6edI/8uPyGqRGzvokW0t30BH3JktloNpX9hLWyVWwD+xnbxY6y4+wcL+O1fD6/ILVK i6Xn5HKk6XKHfKdhvWGz8Vx/oP/F/v/t/1Ir0tbTVMTDWoz+5/QAZtZLx+hNpDP0HjOwBDYASWEu NoP9FGk1u5s9zPayx1kPejnO3mMf45P0OfuG40vLjTwLhx9xBHLzW3HCvIffx48hHed/519J6VKu 5JVGSeOloLQIo9ogbUd6WnpXzpSPyRr8XGTYadhj2Gt4wvCC4aLRYroD3/jXvv3Vd0O/e6ef+jf2 7+wP9/do71Iq1hBfD1zBxmP0jUjzsN47EXH76ASzwHeZbCgrYZPgmXo2jy1mK+DJu9hu9qg+9qfY s/DSn9gFjNnKHfqYr+ajeDmfgnQTb+GLcRjbwXv4Kf61ZJISpEQpVRoqTZBmSy3SEmmltFNSpdek v0jvSV9I3yJpcrzslHNlj+yVJ8j18lL5Afkj+SNDneFVw1+N8caFxvXGiPEfONWUmGpMU02zTdtM +00nzQ2IzsP0ND2DCLz8sLPSWskvPU1b+Qg5A1eY1xHP9dQsTeaIVL6XbeS3sx4+yLDCOI6PYzfQ RdkDX7/E9/Av+DhpMqtm02keHx5t0Jgi/wbcePkwnZefxdxeR8srjBa2ml8wWiiMM9IYnJF+Lw2T vdKrdFo6w0zyQ/SWHM/S2Xn+mFSDKHhOLjEEyCXdR09Ji9nt9DT3E8V/Y96COL6B/Qb7Qi0rYv+U NByDb0AUFUvv0500n/+ZzuM93ki/YM3yXNpKI9gq+oh+jbdiiOEW41BjKnuFt8mdfCDrIS4/jtmN YYOYZEihu9hsabfxAn+TltIxOZ7ekX6L0R/jT0mT5YuGaawVb8DttJ4Wa2tppSEgv8HmksRmUp58 FrvbKqlIdiFfg12lDnvafrzdB7APlEmTIbEjciYhLmZgh9iNdC/2CRkR1IZ3/EbsYq9Tj7GWR2iu YQDDroNfal7tn0aztF/TLm0u3aLtoALsBxu0VWhxL/2VttFetq7/p9SOq+SbeLcnGar4MUOVVsA7 +Zt8Ot955frC23nMTn9DegorU2I4SJ3yn2g6lWpbtD8iuq/CDruL5uDA+gFm+Ql6mCj10Yj+G3iX ViW1Y75naKr2mOZk8dSqLaAp9Cw9ajJQo8mLNVbZG5jvT6mFT9OWSC39bfDDNnjBB28txf6zyVcx o7bMV1py7fhxY8eMLh41ckTR8GGFVxfke4cOuWqwJ2+QO9elOHOyHVmZGfb0tNSUgclJtsQBVktC fJzZZDTIEmeU73dXNSiqp0GVPe6JEwtE2d0IQeMPBA2qAlHVlTaqIuo1QnWFpQ+WN/+bpS9q6bts yWzKeBpfkK/43Yp6tNKtRNisqQHwd1e6g4p6Xucn6/x2nbeCd7lQQfHbWysVlTUofrVqWWunv6Gy IJ91JcRXuCta4gvyqSs+AWwCODXd3d7F0kuYzvB0/9guTmYrpqhmuiv9aoYbVdGMlOdvbFZrpgb8 lVkuV7AgX2UVTe45KomTklc3oQq9G9VYoZr0bpQ2nHFU2qx05fd1bonYaE6D19Lsbm6sC6hSI9rw q0le9Fuppt/2gf37IhrHmWzDD7VZUqff3qYI487ODYr64NTAD+pmuUQLwSDaQF2eV9XQWYWut2Cl qsVZXOXrggGVrUOXOFjm6bOKzi966s1rmKeoce5yd2vnvAYsTWanStNWusKZmb5e7Sxl+pXO2oDb pZZmuYONlY6uFOqctrI7w6dkXKkpyO+yJUUd2zUgMcZYrD9kWuD0qE7ndHPBVU+77Fkmxui+DidB VWlSMJKAG3MaLUjLaOpsGo0FwBNkqKU2Y0Xa1LiKhk7bWCHHFJlqyLO5lc7PCRHgPv/3KyWNMYkx z/Y5CaWIk8uhprLGS7zq9apDh4oQMVVgTTHGEr08qiB/WYS73e023J/FpYFq4NvG4NhCuN/lEgu8 OeKjOSiooamBaFmhOVlh8hXibM0bhKbvkiZ1htCELmkuV29wI5J7xH2WUlWz5/Jfoi1toL91rMrS /ou6Jaqvnu6uxtFY8Xc2xKK2uvaKUlQvHAq/QRfj1IEVASmLQyY4niXp2ugJ+ZIJjssBiyrn4c+o B3VzxGRGVOoSplSptoaJURqMd7li78z/VymiXRS19Oz7arFpqGO9sYFGh62Ou6J8xfAsnVJ1LbYc jpN9Z2f8FTqEWnSU18UyRDwu+i6lQqUZeDPz8Icrx2iBYJbqg8ugqcVbpIuDWbHiFYZZsUpBPCI6 C/KrsGd2dla5larOhs7GiBaa41Zs7s5e/gJ/obPdj90uGjgR7cDmLLVqSxAea2Vj8XpwKu9ys41T u3xs4/RZgV78xKFsrA2EOeMVDeVBsSy8ojYQc4u+ICL0sYb4jwkiRu6gGQD+64CCCCD83wOKZOTD 8O2LSsSPJQgt+j92Fr+YgDUg4ZRtInIluZLyQNASfatIfd/6DPQNKXIfPtk0Q/tITjL04f8a2WxG FxcD8cVn5siGlByrNT0uop3rSUzkMwTjy7BawSWRRUgozWIBtQgZFeIedBTkKJWeLz0/fFhWl/E/ W/oMLRlFSx/2WK0684kvIyEBXBLZhIRsFougQna5ye/b7DEqGTYHginMFfyn5iylAclAIn5KmiMb N/CNCRsTXxlgiDMl2Ll/4KTU6zMqsmoH1qXWZUzLmm+an9A0cEHq/IyGrJV8uXFZwm2JG4z3mnba XrGf5qeMpxLeSsy8PPGOOJ/LPXJYHKM4WxyP2+5M6iAEkG8ApAo+x5y257y82e61fTHbex5ksfd8 bOps9mJcbEaLhwHB4EBb8jUjitLSklNt3OjOHewZaEsbUXRNks3jzjUZZ8w/8eCy8JLyeSceOrny Z72Pr1r1+OOrV10/m59gMrv2t/Xd/drp/v7+w0/e+wy7v/8XFy7itjDvk7b1IhZmah/KaVg7LwvG Vi4hw+4TC2J3EBPD9VpQYEPc8dZES2JOfPyQ1ByHnDPEYRhidVst9gxGyQqczmcoJo9YamHuKRRL ebRQJEoeU1pqO287nzym8PxLtpeSx9he9BYJYI19VxmsaVa/db1V9ifdmLQsS5qWtsA2L6U5bal1 Zcp6a2fKpqxHrfEGRRK/9SUkWKwDZBNDvyzCH+n2YQIHcZQaQlY2qsdiSZXtB/gjlMFbfYMxSgOG aU3uqFcWKVyxi+hQQqYOj74qHkYem4djxJ89IzSe7QX2CBsdzjjBDrDRuF73+RK+X6f8CNvRdWmp PsNiecVqfeadHV2x7z5IEpOzfTdbMCwpOX3M8GFYOTabZrPFwYHFaWK1Ro0Uq1V8mTWaBl+jr2pq itGUhnMWuXM9M3ucP5+/Zt/Dt4+YlJKc0BFZP69tS0qP629PrTgy/+bmO7b3nzv1vMbutO/aoN6x 6qGUB/iK25vuuOsu5emX54ab6++7Oue5rX39n3+I4NIfbTDuRT/2SBAOxE1ZxEBy7P034h5LZdNu CNZM9pbd2ta4YHLtvwCP3N8rCmVuZHN0cmVhbQplbmRvYmoKNDUgMCBvYmoKNTIyNwplbmRvYmoK MTUgMCBvYmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1RydWVUeXBlIC9CYXNlRm9udCAvUkhI Rk1VK0FyaWFsTVQgL0ZvbnREZXNjcmlwdG9yCjQ2IDAgUiAvRW5jb2RpbmcgL01hY1JvbWFuRW5j b2RpbmcgL0ZpcnN0Q2hhciA0OCAvTGFzdENoYXIgNDggL1dpZHRocyBbIDU1NgpdID4+CmVuZG9i ago0NiAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9SSEhGTVUrQXJp YWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5n bGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAv TGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRo IDIwMDAgL0ZvbnRGaWxlMiA0NyAwIFIgPj4KZW5kb2JqCjQ3IDAgb2JqCjw8IC9MZW5ndGggNDgg MCBSIC9MZW5ndGgxIDcwOTYgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLeFTV tV577zOPvMgkkOdMMmcyZJBMYiCA4WUyeUxAIwIh2BkazISQkvCQYAKIIgy1CA4olGup0ipqH2Kt cjKJdAK2pNLaiiJcpbSiFdS2ar8i6L1qfeXcf58ZUFq/3nPyr7X2Wmu/1l5nn30yvTev7qBUCpMg X/uKtm4yrrzFYPnta3rVeDktl8jc/K3uJSvi5ZHLiUzvLVm+7lvxct5WIva3zo42WU9enwNXdUJh lIhNBB/duaL3lng5rwXcunxle8KeF0bZvKLtlkT/9BrK6k1tKzrAcdnvA1G7V/b0GkWyS78rum/u SPizAFHKr3Rc0n7iw3oKhyAwgNMHNJ0eIAskG5XTDURKp1JAJpSl3cR37Ppg+0et6dM/tNqtUBA9 8taYEsmPKN07P93/xRIbWVNRTDL8pQH1LFXD11OdjT7d/+mttnhP0nLx4oPULK7o9+Q6TzwtxtJZ gIuxUW+Bc1CMEQXRaU5fTLj7M7Mq0mvKhIoWyw2qgq4E9gOHAYVaRSGsNtCNQBjYDxwGTgBmIlBp VYGVwF7gLGAWBcIRVZ22mjEiD3XzMN90kUPnAR0Q5AQtB2YDrcAOYC9gNvykZiWwETgMXADM5BM5 0V0TMPac6DaD9S9dXmEU2+LFloVGsf8bwTifNTfO66+Ju02Nu42fGFdfWRvnY0rjPLO4IozG+5PT KoZqskU2JpmNgXeDMv4bSmeMnPSQyCIN4AJDNTQ+kdk/2lOx97BQiAkuGC0mpz4kWDQto6Immev8 PGWSk7/Hz8Ut/Fz/iIyKvTXX8jdpP3AYEPxN3G/wN2gjPytjDloN7AUOA8eB84CZn8V9Bvfr/HVK 53+mcqAaaAX2AoeB84CF/xnUxl+TGWNQKVcDnL8GauOvYlqvgqbz05BO89P6EH85WjmlYtAQvOUJ wVmcEHLsCSEzuyLGX4p+MhYZ5cFKI6MOiSKqogmiKFo83hkTudHpXc4Yf6tf9TofqhnHT5IGcIzk JHo+SSowBwgB3YAZ0ilIpygM7AQeAjQAWQZqA1R+FHgBOEXjAB8wB7DyE1F0E+PHo55aZ002f5H/ jnIQ8WP89wZ/gT9r8Of5bw3+HHgh7Ef5s9FCJ9WkwE6oYwO3gZfDbuK/7h+d6dRrMvhhRNAJWg5U A7OBVmAHYOaHeVF0sTMTjRyio3iGnTxK7xr8p/SIlXxLnT5PHRJQlcQz9WpIIHvVvR7u8+y+H0VJ PPfsgiSJ5zvbIUniuXUTJEk8y9dAksSzeCkkSTwLWiFJ4pndDAkkxh/8xegxzsrZy5hak87XIkpr EaW1iNJaUvhaedMnihzjD6IlJYjYHp93bIkzfJCFn2bhJhZ+hIU7WHgDC29i4eksfCMLe1nYwcKF LOxj4UNsMkIRZr6By4pTfLksfJSFn2DhHhb2sHAxC49mYZVV+mLcFb0GTx2Y32D9NfKh467+q6uw +6RzFyLqQs67sCccBj0O6EbJBye1KO6cVyh5UX9Jdbx85dSKlTUz+RFUPIJlOEJnAAULdARpdASN HEFz6aDVQCswBJwHdMAM7yLMY4dB00HLgWqgFdgInAfMxnDOYyicVoLKIe43BlYOWg3MliV+BHcR bhd3+QpsDpvXNlPscLD0Qja7UC/klZSdjX05M8OaEWNpBz5O++fHaZRUk8Tv4TuoAAuxM8F3RD8p cMbYfVHPIWdNFvs+FSrIOjaFPKwYfDL1GOVJ5LBK/URy8MfBK6KOG1AtPeopdR5kI2StA85PHH9x vuuIcYjvOA45/6jGFBZ1/gGaxw84Tzrucj5XHrNC87QnxsAOqobroGOy84mjhusmGPZEnRskO+C8 3THDucxhGDrihht7UPKlO5s8C5wz0V69Y5HT14M2DzirHTc6p8e9Jsk6B5zjMARvXCzBYMc6jE7d hUaD8ytjrNNXatltCVhmW66yVFhKLS6L01JgsVtGWTOtNusIa6o12Wq1mq2KlVvJOiqmn/V55Vtv lNl4+ZmR0IwUQ7Zhh2FymwElzqycriVtpGjkjfNqWaM21E6Ni1Tto3nuGEueu0AzuWuZltlIjc21 2mRvY8yiN2mV3kbNMuebgT7G7glCq/GtMUbNgRjTpWqzXcusCwwSYxmb77ZLfsXmu4NBys1eU51b nVmVMaWh/mtIyFCG6r1fXrlfit5cb4G2u3FeQPtZQVCrkIJeEGzU/mue2hIYZB+wC/76Qfa+ZMHA oKhiH/ibpF5U1QeDjTF2g+FHKnsffsgYMPhZ8WKWfqRaC+N+e+J+xagPv9GSwS8piYoNv+KkJMNP YdKvr2e0v75vNAh8clTqMXx6ctSv+hwthk8xCHyyw3TU8DmaHZY+WpXRjMMBl0IQuLB8chguDpZv uBgj7zNcyhMud11yucvoScRHY/hIgmbSzl70STsLn68E8j+LHbVeL+ufFmxv8Xe4/SG3vwMIadvW dOZq4UWq2tcelAZVE57QovZOyds6tKC7o15rd9erfdOMev9ibpHmae76PmrxNwf6Wnwd9dFpvml+ d1t9sH/GnImVl/V116W+Js75mr7myMYmyr5mGPX+pa9KaZ4h+6qUfVXKvmb4Zhh9kZHjcwJ9VqoN 1mH9JO/nKcnI15DdFazNtnVXGck7zZW7wX4Qp5V9lOINaqnuWi0NkHldVlNWI014pqRpBNTpCVPu hmku+0G2L2GyQZ3hriVv7+qe1ZTr76qP//Xggqp3tVyKOPVK3ddecPFrvrZ6ebZu1ErmNWrVcxcE +iwWaEP1QeimXtSlpPhj+lBceSWUU6WjEJccpW661CUlJRz/PReMMUGN6AzioHGon/kKWS/1BIVW 2NjMsRU0L0AYWhYEDuIsJV8SPUFMsId5Wc/F1uQ8DJniGsK0ey6id3VCSsSiN8EN1x4veXsuhuRi c14ZLIMYser1YmszHaQ8IN/0KOUpHsL3j/428I7kw136O9IuOf87NrpYAkT76AnWRU/QYXqGXUCt /TRIAySPQPX0Q1pP99IWvNYWQHMXNeE2QX8vy9MH8GXyMF6YD9Mx+H6DNtBByma5+ru0kTaLl1Fr M6VREdXQHFpJd7Pr9NXUQmeUO6iSrqObqJuF9YB+j75L/zH9hAbF7/UvKIXyqR33Mf0905/016gM Nb5H99MZtivpKfKhlzA8H6CbaY9YqDB9if4pRuCitRiDQrPoGBviXrTeQW+zXLZe1KGVH+ma/ht4 OWghddIeOsgmsRncZWrRZ+nHKBt93IJW76coHcAdo1/SaZZquqD/WL9AeVRK12A+A/QiGxLDX2wa rkbcTIjSWJoCy0r6Ff2OTjA3+zVfaUo1VZh8plv1kzSKxtN8jPZR1Pwb+5hvwL1RPKs06LU0AnH5 row2/ZbeYPmsnM1mN/CxfCV/UNxMVvQ4Hvdi6kK870PrryONDvBUflz8SHlc+cxcMHxWH4EV8dAP 6AH6NUvDTFXWw77NTrG3eB1v5T/gb4p7lceUlyxtmPWNtILupsfpY5bJJrO57Jusk61nW9h32f3s GDvB3uE1vJkv4+dFp1glfqnU4p6n9Ch3mO40bTO/MxwY/s3wfw9/rFfod9Jc5MMmjP579CBmNkjH 6RXcZ+hNZmIpbARulbnYfHYb7g3sbvYI28ceYwPo5QR7k72LV9KH7DOONy03czsOP/II5OY344R5 L/8hP477BP8H/0TkiCLhFZPEdBEUKzGqLWIn7qfEG0q+clzREecK027TXtM+0+OmZ0wXzKmWb+Md /8LnP/qi5IvXh2l46/Du4ejwgP4GZWEN8fbAJ9h0jL4N91Ks925k3H56maUidvmshFWx6xCZVraU rWK3IJLfYXvYT4yxP8meRpT+yM5jzGncYYz5Sj6J1/LZuG/kHXwVDmO7+AA/xT8VFpEi0kWWKBEz xELRIXrFOrFbaOIF8WfxpvhIfI5bV5IVp1KkeBSvMkNpVVYrDypvK2+bWkzPm/5qTjavMN9pjpnf x6mmyjLHMtey0LLDcsBy0hpCdh6hp+gXyMBLFzsrNgm/eIru4ROUPHzCvIh8bqXFYhZHpvJ9bCu/ nQ3w0aZbzNP4NHY9XVA8iPWzfC//iE8Ts1gjm0dL+fh4g+ZRys8gTVeO0DnlacztRbR8izmVbeDn zakUxRlpCs5IvxXjFK94nk6LM8yiPEyvKsksh53jj4o5yIJfKlWmALnED+lJsYrdTk9xP1HyZ9bt yOPr2c+wLzSzCvZPoeMYfD2yqFK8RXfQMv4nOofneCt9ny1WltA9NIGtp7fpp3gqxppuMpeYs9hz vEuJ8JFsgLjyGGY3hY1mwjSKvsMWij3m8/wVWk3HlWR6Xfwcoz/OnxSzlAumJtaJJ+B2upNW6Zto nSmgvMSWkGA3ULFyFrvbelGhuMA3YldpwZ52AE/3QewDNWIWNLnInOuQF/OxQ+zBfR/2CQUZ1IVn /BvYxV6kAXMzj9ES0wiGXQf/qXl+uIkW6D+l+/UldJO+i8qwH2zR16PFffRX2kH72Obh26gbn5Kv 4Nm+ztTAj5sa9DIe4a/weXz35euLaBezXPo77iexMlWmQxRR/kjzqFrfrv8B2X0Fdtj7aREOrH/B LN9DDzPFEE0Yvp736Q2iG/M9Q3P1R3UnS6ZOfTnNpqfpJxYTtVm8WGONvYT53kYdvEnvFR3DXYjD DkTBh2itxv5zl69ufnONr7rq6unTpk6ZXDlp4oSK8ePKrywr9ZaMvWKMp3i0u8ilOgsLHPb8vNyc 7KxRIzMzbOkj0lJTkpOsFrNJEZxRqd/dEFI1T0hTPO6ZM8tk2d0GRdtXFCFNharhch9NlfXaYLrM 0wfPb/2Lpy/u6bvkyWzqdJpeVqr63ap2rN6txtiCuQHId9e7g6p2zpBnGfJOQ06D7HKhgurP7axX NRZS/VrDms6IP1RfVsr6UpLr3HUdyWWl1JecAjEFkpbj7u5jOVXMEHiOf2ofJ2sapqjlu+v9Wp4b VdGMKPa3LdbmzA346+0uV7CsVGN17e5FGsmTktdwoTqjG81cp1mMbtQunHE02qb2lQ5FtsdstCjk TV3sXtzWEtBEG9rwaxle9Fuv5dz6l9wvi2gcZ7ItX7XaRcSf26VK50hki6o9NDfwlbp2l2whGEQb qMuLG0KRBnS9HSvVKM/iGt8cDGhsM7rEwbLYmFV8fvFTb3FoqaoluWvdnZGlISxNfkSjpnWuaH6+ b1A/S/l+NdIccLu0ars72Fbv6BtFkaZ1/Xk+Ne9yS1lpny0jHti+EekJITXtq0IHgh63GZLhLqXG pkuRZXKM7mtwEtTUdhUjCbgxp8mSdEymSPtkLACuIEMtbTFWpEtLqgtFbFOlHlNkmqnY5lYjHxIy wH3uH5dr2hIac7HtQ5JGmSeXUk1jbRdlzevVSkpkiljqsKYYY5VRnlRWuibG3e5uG76f5UcDzUFs 24JTyxF+l0su8LaYjxahoIXnBuJllRbZo+Qrx9mah6Rl6KIla760hC9aLlUPuZHJA/J7lrI0q+fS X7ote6S/c6rGsv+DuSNub5znbsTRWPVHQomsbWy+rBS3y4AibrAlJG1kXUDYOXRS4nZhWOMn5Isu OC4HUjWlGH9mI6kXxyxWZKWhYWqDZgvNjNNgssuVeGb+v0ox/YKsZbAvqyWmoU31JgYaH7Y27bLy ZcNLjYjGZmw5HCf7SCT5MhtSLT7KaxIMGY8PfZdap9F8PJnF+MMnx2SJoF3zIWSwNOMpMtRBe6J4 maM9USmIS2ZnWWkD9sxIpMGtNkRCkbaYHl7kVm3uyCB/hj8T6fZjt4snTkw/uM2uNWwPImKdbCoe D061fW62dW6fj22dtyAwiH9xqFubA1HOeF2oNiiXhdc1BxJhMRZEpj7WED9oIGOUHpxdCYJMHsIv LWaAaNwlDVKK/gcabO3yMGzCP/TxWwW5MlwZxSBogT5XxdDnPhN9RqoyBE+ar7+tZJiG8HtGAZvf x+UAfMn5hYppVGFaWk5STH9nID2dz5eCLy8tDVIGpUoNZaemgqZKHZXj++cYyDGqPld9bvw4e5/5 31v6X7Rkli39bSAtzRDe8+WlpEDKIJvUkC01VVKpu9Tkl20OmNU8mwNJFOUqfqE5S9lAJpCOfyEt Usxb+NaUrenPjTAlWVJyuX/kdVnX5tXZm0e2ZLXkNdmXWZaltI9cnrUsL2Rfx9ea16Tcmr7FfJ9l t+253NP8lPlUyqvp+Zcm3pPkc7knjktilGRL4kk7nRk9hMTxjYBWxWuY087C323L9do+Wug9B7LK ey4xdbZwFT5oJsuLAcHgSFvmVRMqsrMzs2zc7C4a4xlpy55QcVWGzeMuspjnL3v5oTXR3tqlLz98 ct13Bx9bv/6xxzasv3Yhf5kp7Oqft/YP66eHh4ePPHHfL9gDw98/fwFfCUvf67oTK4xBGJc+Bufm r7ukfaRhYJSZyBozvnOoaebMhlnzvTU3d7Utn9X8f5EnWVwKZW5kc3RyZWFtCmVuZG9iago0OCAw IG9iago0ODMxCmVuZG9iagoyMSAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5 cGUgL0Jhc2VGb250IC9XS0hZUEorQXJpYWwtQm9sZE1UIC9Gb250RGVzY3JpcHRvcgo0OSAwIFIg L0VuY29kaW5nIC9NYWNSb21hbkVuY29kaW5nIC9GaXJzdENoYXIgMzIgL0xhc3RDaGFyIDExNyAv V2lkdGhzIFsgMjc4CjAgMCAwIDAgMCAwIDIzOCAwIDAgMCAwIDI3OCAwIDAgMCA1NTYgNTU2IDU1 NiAwIDAgNTU2IDU1NiAwIDAgNTU2IDAgMCAwIDAKMCAwIDAgMCAwIDcyMiAwIDAgNjExIDAgMCAy NzggMCAwIDAgMCAwIDAgNjY3IDAgMCAwIDYxMSAwIDAgOTQ0IDAgMCAwIDAgMAowIDAgMCAwIDAg NjExIDU1NiA2MTEgNTU2IDMzMyAwIDAgMjc4IDAgMCAyNzggMCA2MTEgNjExIDAgMCAzODkgNTU2 IDMzMyA2MTEKXSA+PgplbmRvYmoKNDkgMCBvYmoKPDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9G b250TmFtZSAvV0tIWVBKK0FyaWFsLUJvbGRNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94ClstNjI4IC0z NzYgMjAwMCAxMDExXSAvSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9D YXBIZWlnaHQKNzE2IC9TdGVtViAxNDUgL0xlYWRpbmcgMzMgL1hIZWlnaHQgNTE5IC9TdGVtSCAx MjEgL0F2Z1dpZHRoIDQ3OSAvTWF4V2lkdGgKMjAwMCAvRm9udEZpbGUyIDUwIDAgUiA+PgplbmRv YmoKNTAgMCBvYmoKPDwgL0xlbmd0aCA1MSAwIFIgL0xlbmd0aDEgMTYyNzYgL0ZpbHRlciAvRmxh dGVEZWNvZGUgPj4Kc3RyZWFtCngBnXsJfFTV2fc55947d/Yts2eZmUwygQwxIQshkJIbsiimQFjN gJGERYkLBNkUq0SLAkElVhEXKlErULBlMgFMWGpEW7XWQqu1aGtNWyxi5ZW+LypFMvP9z52A2Pb3 fr/vu5NznrP9z/Kc55zznOferLh95SJiJB1EIMqC21rbifoEFoO8tmDVikAqnjaKEHnOje033ZaK +zYQIt1y06133piKB1E+c9fiRa0LU3FyEXTMYiSk4rQUNGfxbSvuSMX950BX3Lp0wXB+EPWTytta 7xhun/wR8cCS1tsWpcpX3Qo6on3p8hXDcSvo9e23LxouT5sIsf02iYfnH/+ilnS0IEDhMsl/k0qy mWgII1ZSSGYRornAjhIJcZ4vsc33HQ0cmWep/ELr1SKBkOf+WpnJ6ati+w8uXLg4ZCVaJ8rq1PI8 Azh5QmIKqbGSCxcurEFneE1XPpk9MzuqTcKLZC8cGoYfgOuGA6OFF3tlU7HSB2p3qDTuihT3JweE F+PjStT0gseKOw4Le8g8UoLkPfFZPHlPr1LLi+/pLRmfooWjVRrXprJlR7G/2gdYIRwjluHQVNDN cNvhXobToEN7yEdwSThB2CU8F6/3o+IXUJGl2iG8gAEp8I/BJeEE9P4FjOUF8vlwiohePd+rM/Lm n1dR6cLzQFngW+E64PbCHYOTyFL42+GScAJCzyHvOcKE54Rn41a/tVovPEPWwjHhKWKhlPhR+xO9 VpU3T/Za0oqVaquwhTTCMRITJpMBOIZqHwHsEcJQvCFeMFplYUOv3lxsRflN6PQmdGQTmuyGT9W4 ghAvv6k3zcU7//24xabi7ooXlaYCvVZPcSO4cAehwiJhCQkRv3APaBboAtBM0PnCQmJS+6n0WqzF HWivCsWrBCcZiexqwUWKQWsFH0lXi62Mm1PtrIyPyC/GiGsEj1rEIphIKYpqBTle7A8cEhSV+Rt6 dQbevw1xq7P4iHC/IBMHSnWglNtvOSLoMcd6dSQze3Wm4q5qozATw5wJtvjRRwouc18RlsRRUbVN qBMyiAt5twiZxAlaL2SpdKfwLKlH/Ie94Qz/wCHhURX1A14pmp+QEq0JvSZz8UC1TpiA3JjwMCbg YbXxrt7w2GJSHRZGkCI4Bh6vRWgtQlahE6FOzFonZqoTM9WJTnVC+oiwETkbUaZQWEPahdWkC247 wlysnHEwlC8GZzxnRHG/4BU8YIz1EFhJkerr1Zl5zzxxe5pazNNrNBdXHRGWk6lwDENe0ev2FC89 JOSrQxnV60nngPY4xPWI4E5NDWpy8Sk5ImSAEZwxmUJW3OmPVfsR54LsJ5T9kh3nTGLvsN/x6WbH EOf0rWH69jD9dYomB9jx1KJgv+V0sDqDfYzK5rEPyXaEGDvEXiVFqOAD1sdnn73P+kkV6AnEF4L2 g5aAHowH3/D3sb5eEPT96bjJxQfLXo1HCocD/tzhgDt9OGB3FVfnsqPsFZKBKn4PmgP6Chsg2aAv g3pAB9gK8gboflZGxoPuG6avscNcxNlL7AAZC9obN/MuxOIyJ3vjGk5+GiepWGOh/zD7KdtDfCj6 k3jYh8xdveEcv+UQ6qPsBbYinum3V+vZs7SJnkOhbnKCU2Jnz8XLeSVd8cMBfz/rYl2Kp1zJVQqU HUJRblFB0Q4hkBsoCJQHdgSqrexhbCDbGdYv2wS/nAQYpAdOgetiG+Nieax6CGPi42KkA363GmqB 366GCHyrGuK5Z9VQFbufTIVjqOMeuLVwHXD3EhH+Gri74L4Hd7easgKhlXCrsZu0A9EORDsQ7Sqi HYh2INqBaFcRvOV2INpVRAsQLUC0ANGiIlqAaAGiBYgWFcH72wJEi4poBKIRiEYgGlVEIxCNQDQC 0agiGoFoBKJRRShAKEAoQCgqQgFCAUIBQlERChAKEIqKKAKiCIgiIIpURBEQRUAUAVGkIoqAKAKi SEUEgAgAEQAioCICQASACAARUBEBIAJABFSEFQgrEFYgrCrCCoQVCCsQVhVhBcIKhFVFDAIxCMQg EIMqYhCIQSAGgRhUEYNADAIxyFb3CMerfw7IcUCOA3JchRwH5DggxwE5rkKOA3IckOPDQ+eM4AIz AOwAsAPADqjYAWAHgB0AdkDFDqDkALADKjYGRAyIGBAxFREDIgZEDIiYiogBEQMipiK6gegGohuI bhXRDUQ3EN1AdKuIbiC6gehWEV1AdAHRBUSXiugCoguILiC6VEQXEF1AdKmI/+epYffSJi3OWtZB R6p0LflMpfeQEyq9m/So9Htkh0rvIvepdA0pV+lqElYpplqlK4hfS+P+cku1C1vAVLh5cEvhtsPt hXsZTlZDxxD6CC7JypRs0SJPlbfLe+WXZWmvPCgzi2aqZrtmr+ZljbRXM6hhgep0ZlL3UWwtZDNw lKyF/zkcDhH4VWqoipWi3VLss2X4lbJSxXYm8Hk+PZZPX86ne/Pp5nxarWNXU1Hd6QKknIEBtEkx hif4T8CVh/MmYGd6+MBnbn88PMbfRw+nyEglguhncD1wO+DugyuHK4YrgMuF88OVh/MBa1Kyh6s8 DJoHF4QLwJUTlwuqot2mVfqZie7o/bmJ6Hg7eSOAOxTPKwLpi+dNBXkpnjffX62jB0ge14rofiyq PaB74/6TyP5JirwY9x9CbFfcXwrSHM+7CmRuPO9tf7WJziJ+kUNnDtMZmHAenx73z0axaXH/SJBI PC/MS+ejoVzkjoRGfRIUYRWdk2opFPePR+nsuL+Cl9aSPD7xVEMK1O5JCPO40IsOfd5Pm0SqGPxn /I/6P0N//w7GQjzeD/SJIMdy++hsRe8/XPAMClf749V6Xh7nQ88wjXG6378jd6P/adRFcw/4n/Rf 5X+4oE+L5IfQ741qE3H/fYE+tkdJ83f4i/wrCk76l/uv9bf6p/ubc5Ee91/vP8y7SaK0ie054G9E hZMwity4/+pc9AVdrPff6Vf8ef6KwGHOXzKWNw1JLjjMOUCKU62PAn/zc9F63D+rvI/alHz5rNwl z5UnyuPlkJwtZ8mZskNr11q1Zq1Rq9dqtRqtqGVaonX0JQeVCL8nODTqdUEj8oiohq2Mh+HBJ4xq GbmWxNKEBtYwYyJtiA0sIA3zA7EvZ4T6qH7anJgUmkhj9gbSMHNibGykoU9OTo+VRxpicuPcph5K H44iNcY29FEys6mPJnnS/ekxew0yyf0PpfcTSr33PxSNEo9rVZWnyj7BVlFf+x+8FjWxpTbyzeO5 MpgZe7xhRlNsd2Y0VswDycxoQ+zeGYHrm/qZhZnqavuZmZNoU7/Yzix103m62F4bRbGTajFIsxnF SB4nKKadSAK8GPaTibwY5ihVLgw4ygU5QTm9iYTVcmG9SS0nUl6u50SgrrYnAA9lcgk5oZY5kUuu KAOJAba2JwwPpUIB2sRL0aZQQO3YSLUivx9FCuChCIW+p1bkp2pjscJviuQOFym7XKRMbUtI9Uet hnuoxjHiUhnHCJT5hpH/f6FFEyO0d/TKe16tWxSqawnVLYJriW1atdgT65gfCPTcs5JnBGJCuGX+ gsWcti6KrQwtqo3dE6oN9IxWcf+S/SrPHh2q7SGv1s1s6nlVWVQbH62Mrgu11kZ7qyqbqr/V1sbL bTVV/oe2KnllTbytKhX3L21V8+wq3lY1b6uat1WlVKlt1bVxuW9s6tGSidEazCunvcyghwy3pAej E13W9glcoPvHBz33pB8UCd1FDJFozBiaGDPB8ayC6oJqnoV1xrPMSLYMZ3nuGR9MP0h3DWdZkWwL TSSXJoJwfEOsbFpDLDhjThMXlZgCFvynOVvOHzXbQ+raavGH+ArVrVi+4lKNnBJe8t+fFf/pWbly 5fIV8FZGlhPSEMuf0RAbMw09kWU01VIbRdpVl9IEQU3r0enq+pIDyIygE3QFb46HIjQCDip6oiEy 69Z0y4zfIlb0+jKLlx6B3rAWDtdhtjoOUwLPWt2bnYvbEooUlqUorqs8HvcFi9FCbzmgnOamqGIr QKArt6ugq7w7t7ugu1yD3AM7kOjfwY/SeOEOgayILL/EDARXRMFsdIu392w8I1NtuJsHIpFoZDlV +XWp/DdUTUf0G8ZijOqzXK2e81vlMHweBNN5LuYj1fpKHuNPKqBiwWcVhFSUSsXUJO598yAGU9FB kqG6nSRDDOOORZInL7lEW/Ikz+OUfYqdHBYk7oafOHmR/J6OoAHSSy8QNzlPvXQ0mQTp/Ar3ib1k iGzB9X4meZzaSQ5uo7PIJCqiTIQ8SJ9OrkqeJt8hPyDPJV+i9yV3I38z+QU5jx78CSdmOZmC8rPI InJa+JhEk08RLVlPDGQ8mU5dpJW8h98X6Mej5DHyM/q95Hm06iD3ob5KUk2qk68kL5J88qDYJZ3Q 7SePkENUk1yQbIOGlE06WST5XvIjEiZR8jx5EX2K0AHxGhIkt5D7yRPUK/wCoS3kRyRBjaxZqJFe RkuTyGyyhKwmnWQ3+SW100bphHQ2eVfyFKQwjYxAn9rIaVpGJ7MXRGNyQvIDMpf0kzcwXv4bEOeK O6W5iarkD5NHcft+ierpYfqKVCw9PHRv8tnkT2GvDJPR4MgUtDOffJ+8Qt4k/yD/zdYm15JryAy0 /HOaSQM0DI6/x7zsHnaP8A65CqNtRm9Xku0kRuLkIDlEjoA3fyCD5GPqoOn0WjqfPkL/mxnZQnZM eFrYJ7wrUvHH4HeI5IJHK8gL5AD5FXmbHKMS6i+ijfRmupRupT+kgyzGPmNfiVrx++LX4pAUTgwm vk5OSX6BO7ePfJesIWvB2+dJL9lHfk1+B6vk/5AvqZWOpYvpszRGB+lnTMey2VTWzh7H7fknwhTh EeEVsUycKN4ivi1+ID0gbZJb5cTFHYlHEz9J/Cb5UvI3kB0z6g/DgNNG7oVUvEBeJu+g9vfJh+Qv XH5Q/3g6h96AVpbTDfQx+hP6c/ob+ilGCY0Dv2w2ntWi1aXsdvDpPvYoewytH+OWDhgpPmR/Z18I kpAtjBGWCc8KMaFPOC78TbSKYfEqcbQ4VZwjJjEzxdLV0gxpl7RHOiqd1VRqFmraNZ/I98nrtL8a yh/6U4IkFidiiV7IrhaStAaceIbACAheHCK/BEd/jR4PknOYBR8N0jz0u4LW0wY6mV5Hr6eL6H10 Pf0BfYI+TZ+jP8UIMAYmo+8RVs1msFa2iK1j69lDsGXsYwfZm+w9GFTOoOduISREhNHCJGGOMFdY gjGsgClvHTj7iLBbOCa8I5wSPhHOYNbcYpa4UlwjPinuFPeJv5G+K92G33PSy9KA9BvponRRwzQ+ TYamUHOzZpfmL7JGHiM3yhvld+X/0bbTDJqPngcg+5cf5sUazGK7mUNcS88gORO3DgtGHsE8zMCq +B9SJSQwL2aej745mVdM43CNIsagCK6gh0gZ/TlZq2ECFENxkMTpH9mg+Cr7DvkdbaFecaewRPol C5I92I262GF2iE4k+1glm822CYR+jFPxY8j7HeQxegtdTvbQM3QcvZuW07XkXeYSZtB1pDL5HBOp jk6iZwl6QO4VF5IbLg/hPwZoBazzpxPPiCbxe9if+sjjmNEXyUf0x+QClZKfYXcTsBu1Ypd5EPJ+ P+G7XjPW2VqsRy92kFs1x8g+qoENvVwzQVxDzpJ/ktPSQUjUROympxJt4jPiX5PlyQKsMKwysgvr bjG5GivmY0jJEcR57HqsdD32EhgfSSOZA+PZ3dj1HknGktuS30/emVxK3gL2Ah1FL9BurIg+ICph 93oDq+R9ugnr8Or/OLz/a2JiIRkgn1IPzaXFWA9npFVSl7Rb2if9THpbMxrcXkeehkT/BdKsxwgW kN+QT8lXVIu58ZJRpBT9HYu+N5FbWVQ4Qmqoj7RjzY7APj5xeCTLUct94N42rOcjWBtnsU9cT34G +xmjboxoAdrXop4G8HkeWU52YAa/T3uRshC7dj75O8ZtpmNhHhhFFNT0OHatAfTpj+Rv4HZS7dco 7Au1dDbq+opcRxaihTGkkfZgBg6QCuystcKvwO8caiUTaTb9EXAtWKFmGL8rpL9SRkYlpiTHsjbh CM6YJNK7cXqlk+/QZeiFBeMYIk46lZQlpqMP71BBjNHfqr14ki1KrhdWJ24lb5EfY04UcZVcS4hS PVOpmvCdyvHjKsaWl5WWFI8uKryqYFQkf+SIvHBuTig7GPBnZWak+7wet8vpSLPbrBazyWjQ67Sy RhIFRsmoulB9SyAWbomJ4dA11xTweKgVCa1XJLTEAkiq/3aZWIDjWpH1rZIKSt74LyWVVEnlcklq DVSSyoJRgbpQIPZ2bSjQR+dMw20i9lBtKBqInVHDk9Vwlxo2IRwMAhCo8yyuDcRoS6AuVr9qcWdd S23BKNpj0NeEahbpC0aRHr0BQQNCMXeovYe6J1A1wNx143oY0ZowxJgvVFsX84YARTVCbl3rwljj tKa62vRgMFowKkZrFoTmxwjXfiNqEVKjNhPT1MRktZlAG7TbGNkU6Bk10Plgn5XMb4kYF4YWtl7f FBNaUUddzBZBu7Ux95qTnm+iqBx68vorc9OFzjpPW4AX7uxcH4gNTGu6Apse5DVEo6gDWJZb39JZ j6YfxEw18CtVjN0fbYrR+9EkLgu56qhS40vdZHJbbg7EdKGJocWdN7dganydMTL9zmDc51P6k4PE VxfonNkUCsaq0kPR1tqMHgfpnH5nr1cJeL+dUzCqx2pLMbbHbBkOGE1XBhaB6ak8NaQW56GG6Zc5 S3kfQ5Ogj8cCCwLoSVMIYxrLvUVjSeeCsZgAPFEKVGwhZqQtpqtp6bSO4+kYIo1JudZQoPMLAgkI nfns2ymtwymaXOsXhGdyObksajHaeikci0Ri+flcROQazCn6OEGNlxWMWtXHxoTarbCNjMFFkDSC t63RcYVgfzDIJ3hTn0LmIxLrmNaUigfI/PQ4UQpxX2ItPAcTmMpxzuI5HZdyLsNbQpDkfdxuQZwx bfjyn8XqSqtbPC5GXf9L9qJUfsOMUANuN4G6zpZhqW2Y+a1YKp8zFHxD3nAollbTJKQzpPEQSxfU XAjl9XMuF0GkyRgTc/Gn4Z3G6hAglGoCDdTHrC3XpPyoPhgcXjL/jumTtVeA+pJnOUol38CGRxEb FxnuZ6rXsfHfin+rd8ZOoWEmdhzWMHNOZ6f+W3n12Ms6O+tDgfrOls7WvmTH/FDAGursZzvZzs72 OuxCqQntSx7clB6rfzCKoSym4yC2jEzsCdEN03oUugHX136YmAIbZjbFGWU1LROj0QIo4bA2VRBC T5HrQVeKy8lsuOfgSuDmwl0HVy29TlrFv5KtKPOgNJtsQXiU/BCRQdcjfxboTF6O7SYelHmUO6Q9 AjcdLxl5mSLk+RF/CAJCVTEh0N41OAnQK2jvXHC+ebgJLPUIw1QkEkrzR4bmyB+d6n/b06tRg+ob v5VlGo6ZVWr5Vt6liJXYiB33Eoea4FR9bqF1Q7Pmjxf6ezq/7eEZg9+N5BkaoK0wL58R1ov54uuS U4pKr2iMcoG8RztJ+4rudf3V+if0vzbUGA4ZC0zHzS3md2HXA1zCD9qRTCbuYzShkftYlZJGJDEh EL0sJijxajVSggmHaZjocCnwEE/E+mXlUOUU67nKyUOVpAph60V4o4uCtqAtFx6siORiQBi4qEjk axIQBzinrxd66MO4j0pkguJ6QPpKYhrpRmgqApEESqVzjAh9VKsY+B2G0T9rJtd4ImijefIQqZp8 poLa7BUVo4vo7cvShLKgU3ikjBZfhSpt588nPuf1r0z00xco13Cr9uu0Bo1e7qNZSrpmGx1r0Otv p2E5xwLDaQDvOkXiNd60arj6k0NneAPnhqitgth4G2lBp0OjkfPGjCkPPUi9+SvnlM+6hm2g3jfX PNQeWJExHx9sMDI7eUo0SwPQMwLkUaXhDv0G/U66W96t22l+SfeGTjvbFnVFfbP9N9kWuxb7bvJr K1iFZoxujGkSm6Sp09WbdureYm9qXtO9Znqf/UHzru5dk83qCXiYB8YKJdfuKvXs0Jr8lkILsyiI WXYQKfPEVNz+fNmOEwZv8J2j6ggmn5li/XLZZAziTGQZd2BSM2lupsVul80qa0LZxGYtH+PO1sga m9XlKikeUz7GZg2HWfHv7tjctfp37yUuwC9pdGWWTi1JEWngiX2JeYmWA49DDd9Bnznw+Onqmbcl 8LwCrehWzA97pRo8fw6iEwYPdGS2oruF3YXXoQIT++jI3nkSlfrYDS9pdRIlRh1u8E3gGWXNikki ol8MiDFRFL36g3QndOHUVFRO5nIFgaqqPNd8BhNBmoNBm0YuG5NTXiKEE6ee+s0SyopOiqGuumTO mw/wWS/BLcSIHmTSKmXefs8BX3/6L8XXPcc9x73Hfdqa9JqMmszZ3qfFLZ7d4o4MrcYXICM05b5r xBpPjbfGp83x5HhzfIIrLM4WN3i2pW/L2Ja5O2N3ptZOMq2ZgczRmasy12V2Zb6Xqc3k8+JyOEsz mdVoybRi3hmXJgXLh1ufMEekjz3by6jRwl8khPzGQiMz8rkz7kiTdCdcLqiflPj8lhPW1cybdWkC z6kzWFk52YpJHIosO4klFWleVgmJp7aSSDM3S5HM5EDcVsH7ELeoRDFbK0SttULS2kBtFSnLULRH w2pmNikGXbo3naWnUX5/Q0X4a45yyWiY1nSEpEM5yYDLTA6OHTs2Spc1Q15swTH2cshGWWk4BGHJ HZNTUgy9FmIjamTReDHP2v3ZzyLjFkWbFmsTn3ip9hfvn796ckniy6tdVEp8/RjV/aGn6rpZNyy6 +a6MT3756U8X9M6vPtcY5rM0F2ulTNqBWbIqI7XmgLHcXmef5H3S9Ix5q/0Ds85uS7MHbSH7/XYI BzXpjUaT3WbrY92Ky2xymM0mu97B9wWFCo20C+IH/u63QoggbMmBl8BhcUe6ydjH5igmv75Qz/Sc 6/odeMcxoBgcrtKAo8ihOARHH92jOGw2v7XQygqtVdapVsHKi1p5W2kWi1m0WAflE8fdVHFTt89v 7qNBxW5aTQ8fJ1SBlWYvJhsz10+vHhZarLxzJ7ECm3mAC69VnTskRNSp5F7zMrB/vXRVxHy39TV1 m1EnBNOxrDktV12SKtPlvDTsnvKYkmLidPCVmzOXeoyrJjetubP1zpaTXezU0H+NumH+ISq2bU68 lST0zsx5Szd3rV9/S5B9nfjnPwsTZ9/f//DRD7DSrgPH87Eu3LATHVHG32xYqV2v3erdKe3U/ti8 O63ffMB2JG3AdizN5JTG2Gqta1z72W+txx3yIXIMcJHKHrs1PQAR4izMAovSd1hM/mBhkIEhrtLg jiodVXTHdUmdgFeFU3v3UopZCSrZfrEQ08LLiDucEj1BVmedmGqkRl+u54Tdm3NJ6Id3rXMpkT/X DAYO7198p+ci25zayKgUVuURXLE7HUTdzggkkzr4VpYSVtGSOKufWRO9y9q2LfZ14vyxPyX+QvP/ a+cfhp69Z9qUxe0zp7WLM7JmNnYPfS9x7t0/J87SKN1IH6ULD108vXHLmk2b7+dfB1Un/yY6hnl2 XGkab2uwLTKs0W7U/lj6sXaHeUfaftIv7Df32fal/Zz80jaQZitNm22ImubZpqe1pGm80mrXk+4P rR85pMVpOAI5C/3phWAhZwbYJ1mDAbCPM9SqsrBIR6fqPtKdHWZhd4qF3wi2kp7ioslzYqqd2n25 KW4ar+Biau/guz8E7T9xURUyclnOyiFVrKwUDMSpAGaGqbranSozm6lVP7PuujW2m7f/5Guqe/sj mpV47/MX32U33D19yk3g4lI6I2tGY/fFu6jhvY+oLbEzsTKxJLHtJSFjw+N3Pfjw/R1Y663JU9IN 0jvQUN5Tpjyg2+jY6NpOntC8rntXeNfwhaDL1Y0wjjCNdIx0rZRW6h6QtHKa7Hanud0jWb6QK8kj pCelrbo3hZ8bpCo6FWt9upXQQVz2mbrL2jzYZbHb6k2gdI7i9hSIWrNitpeaG+ZZ6FQLtShOTyl2 4BFKtr1AL1g+N88mnxO1Kl9RBs1w5nXL1CL75SJZgLrzYG/6PTNSBxCW8BQrJBErmTP0HDbhkxFO eQDiSJop3ygljRgKcPYFA26XOyWeOFkhjWIV9U9MvP1Z4o+JDXQNLaWmXQuLE3/wvbDq+bfe6F61 m6XPPXuabob1cwndsv2GWP3t6z5NXEh8+tnjfJfcirNsHTQkHbldqdJKokbKlQPaIu3L2o+0YqG2 C69ttUQQc3H66ohWrsIXCUwzXYAex3wBQ5GBGURdgPIjCaxim3r1o4fHxfU0dWCVk881pw7YoUp7 RSHOmKFKQbJWQncrsQWd2HzgtgpnhsazhUPbpIPnEy+cH3qE9+1BePvQN4Es7ScS+F9cWirxeQjl qlSpcrhLiaRIjVKHNChJfqlFapfOSmKHxLUFgWiZ8D4lsHIPEmGAzybv53HERLJEHL19eAJuH1Yp q9Aj2rzsdhxq6JjtQTpCOnihHuPaAh6dRz8s0H9XK7kaqd/R7xGuluhN0nsSs9tyTWYzSbdyHlmI 1pW3V6Yy76bOUMrnWnH5M4syWzLbMzsypUyr5Up2ZVzJrsnWZV9GIA+qNqIyS9VwSfMy2oweYeL5 0YizMRTyMsw834jywqEt9A/UPP2e3fO3Trn5zVee27uq5oZryrqlg67gh3vX97XZnEO/F48mWq6a X9242KQHX/GVsdiP8ciY81Chrkgskhp17boOXZdO1lCJ5cKOhHuGzu32iWu5TkULFL1GxmEI4z2m GVGbYG6Ehb6DdTGRebVDL6Z4iYO+hyljo+qcD0GjwivR2pPD4+EzTpsx31Clgzb6UWKy+FBiinj0 /Pmv+eeUMi43m9ArI00q9ogQ0QQMJQaRaKhB8Y0rxdu0jl5Q4Qoa95bp+pKnFJ0vs1TvhWe8FCM8 BlEZVKLQMMUAPFmn02uMPuLUjSS5Ovm0/pTxK90/9V8ZpdelN/WvGz8g7+re179n/JR8rNPtEZ+X 9uhfMB4Se6VD+v3GN0TdVWK2VKgPGJ8WH5We1m8xalNKzz4tNZv4q75ec5B3bkDRIWBQ4CF1W6/W xFO3KU4EDAt5zKCBUVoW1fWkcgY6l3tYk1IvHOn7jhpEKdCXLOrV6HWgxcr1AjEGiMBYAF9O4NsS vUaSig16hwGGP42MBatzaLU60WA0FguiQxBENCIYsUhFoyDpDbIOX2DIsiSJIoOQGvV6HRazGfNb qKXaPlqk6AOaI4YjSqFAcSEqUoyB1I3Ia7phQWpmfd7JQ80+z9CQzzvU7JmCaf0b1zvwS/k8pPYe NyZcaOATmxv6x+SrIuvvfm39VZ5/J1hnqnaI7W1ZSpFU31Muaw7SYBpef6SB4nRalHiOFn5IjTD8 0D/T/MS2xC+w1X0oHbxoEz6/iKV8oV685us+COb65CfiY+o6zSA/7Cf25HlltKGiPP3qdGafrZmt n+2a7YlmfCVrysTxpvFpZel1YoOpIa0u/TH5SZ3eaMamQXz8Fa8kO0CUNIPBQvTuoNbXnkWzrCOZ EMYWP1Ix0nbSgfa8mVXVKneWVU4+M1T5tylYv6nVe6bqDN9NcATS5hooxjdqbtTf6LrR05YhNUdJ s7q/cN0idRzmOdMc7m8Ui/XUe1/8aCIx1D+3R7GXTrqz+fvrblr0gHRw6OxjiVOJfybOJj6YG93G 8l+Y2r59z4Fnf8j3ylkYe5U4Abf0PyvTmixRe9S12NJmb3Pd7bnTu5VtNf7C+gvP763veU5rTmtP p512ntekjU0b67zWfq2r3hM1thnlcfZyV7lHWC2ttqyXHrBs9O6y73T12w+4dGYu3570Uk732x2l 5hITT/FmlarUYis1HcRbHT14ZrcZiIKiREE5UtKFCTwIgRWRFXDLlKfSICk08YApONVMzb50Oejw +ppSrFRvljgMI+fORPhx2HwykjoNQflhuAzq67B2ltIfyiUN1At+MkI/E0cn/m5eMLXt7rW3NN7o pI7IubdPJ/5OXWeOfsw+K54x85HdR7bNXVr4s6N4BwNdiebu5LfqmeBd67DcdCkF9qgmqo/aU9Ly BETjvE7XntWRxcYJpcZxzlLvtUKt8VpnrfdJnY7LSVwycKlRzAbZbMFU6N0jzaYwFNORisVCfJu5 7AS13symSlVY+AiXfZmSGHX9Q1bOqOf8Mi4rpjZNm77NnpIWTTNs5GXDA7SXFLth57hSVMTWxNfV PXNeSnydOBq/j3qH7IW1a1o3rLtp4fptc6N4gYjdiXofY9aL7bu/u+SFH7307HaMtxrjzYOsOEgG fb6fWLFO6g0VT+qeMj1u3SXt1B/SHTL1+bRaB72GXa2p10/N2mU6oDnge13/hvE9/QnjefkrkynD kuFU0jNLnYrZVmpxvuw85hScXCosWVUqNbtB2UOK0WK2N8L2w8weO0WBA970UlpiVxWqzEBKscoe maKRghT1ZKhUsZgtpd1gKf5lhZF5djvY3Csa7B7O7hyDTIK00JkSosKseVlLs7ZniVmWoFYxWUrB 8LaUSEW+pWGd4d+BODzKCEeVR8mywEu3wsuwVfE9KFo1pN587RgIStj5gFBIpSjHafxS0XPD25YK IMjAFYLnuzmJ9er0E9RodbBK/QAjejLCLxhq82YFXDLzRs28eaiS7ip184sW4hoegQqCzbREVUew W0D504QCeeEyLuNECKqXtzR+Q5E1bnaBesac3pv4+/1t1PHOGWrXDCnCfa0T5+QJd8y+vrKS0umF Tz27/5EPIQuRxOuJI3dvuobeumZtTQ0+qqHctsf+Bs3ZRfoU6JI0XwxYA7ao2OGRtOLLHuZ02ZjD 7rKZ0yzEasYVw8ocOq3FQOcZklD++EToNdRmcdGki7p4NIt/zHcWVWvSHHpdSZV2qrZRK2hHWAtt 82zM1kdFxWROCzPHPNLtGnAxF5cJnbHU5XXf0c/aUlfcCLZUbt+72AwlwnuSeGCg4ZojXBW8imIL ntQ1ozmtpKx0DHZTt8zvbE5niTME9S3k2Vbx5Mo7lodrJnyn7Le/TZzaJoYbH1g3I+c1a8W0hg8v viRM4mv/UXT2Rax9bo1c3U90OMCrbPoqRdeoYx26mG4A183PdZJf16Jbq+tGgiRoZJgqBax0RdUl BdKMc0MjwV6hZzL2FYxH0QVzSkWvtiq15COXNUxYWTAEVf1VNWDsZrdH0tBdCvco9SZO4X32ASom Ln59rRj++gOw8ZsezlC1YGUk7x90XtYhxfBW/rj0eUr1XSt1I0FCZ2BGxVlFufipPSFe8d96ojLv dvUkQruqpkvxVp6IUewMLrJd8chp7rQ52sVaER+HlmpLrbXaWstpq6Thk5xpk6H2GA0GbHeMhl1E CeSU7sU/pqDDPtWY6MrOKe3ydHtYu+esh33uoR69IWyEbWNk3ASrCXpmAaTbSM/inu51D/cP2jc3 SVnVuwMiX6oJqoFOVTUu75Qw0pWqd3SNE5wLOjH/WcwpRhOncqZVTFoRASOlTe80PzXVz7JeXDS2 cV084RfD2/bVLF53F5/36dgDn8JITTgxtyrXfEJPab9K+8opvs4+gUrvlbw6FrXOTpvtinq2sic0 T2i3Gvt0v2N/kP6o+53xlHRK84nJulP7FvuV5lXtL4zSSu1GzTqtAPPRirje4OYscoiyo0L2taS3 4zZuDkIlveKISykKqY2fKwnLsPPr2qw3Yt9v84i0GUoCbU4rtWNYMMvgzpwTzr3C9jC9c2jbP2hp 4s3PfpD4qpMGHl+yZMuWJUseZ9kPUk1n4vXP/5F4dV1y1zO7dnVv27WLr/P1sJqWY7xWsksZsVWi OjOdAVP4SkkotDeZF5vb7aJeZzH6jWyzMWlkVcapMCP2sdXKSFnGHAtMox9BdFZdES4Kos631r7d zubZ19r32o/bRbuVhLniOFIxMHxG3w1Fymur6qcZqcWMtXx5Sr9s9k5OLefKqjO4bFXgDTc3HS/D x3xufMxXxj/w0xePBQMwxZjXywtbY6PdfFZrbqltiV539XfGTy8Uw1tvqS374qrq3Yl/YIxFmFMr xpjPjioDGpsmpM1z29yhJ+xPOLbmbcnXyY56B7MfMvWbXw9+HDpv+jJbM9I0y7TItMWw1b4zu98o V4eUnNrwTdkLw+vt6x0PZH8/R1certPUG641TbXUBydmy9k5eeFyY1mwLLssVJYja/SSTRf0mPKM 2dnZITknWxm13HiH407nqpEr8zc41+U/5dySvy97X8jUQTe7H/Q8mf/j/NgojTvoUoKhUpeSgW+p XfQjbJ0l2mBj7uZclqt4MktzfaP4CnFjpTeOokWjaOEoOiorWGSl1hIa5Ou616KrUimKpNa5DsYJ b+SOPq5kXMRRBoPCMm5a4KuI3yrPIRA5Q4ZttWUaSjXURcPZY4L1wZk06l5I29xf4psINxN9wWw2 Is1kZCN882D7rx9haPRRX32ajL0Xf/y+csk1L8OX1NnJt3pH5JfCyqTSbFy8erNyeHyw159Tqsa9 PjWupCNwi4mOya7PfsL0WPZr2e9ma4LZRpMo+vg49kM7ICVcT+h1F1SBqgepGs/OLeVUyfRBO1AN s41UbKEd9CzFjcqKj9BaoNRxZJoLJSlVJsMgNk88C9MghuBSULWrxK2gXreC09mtlJWXupXIVfBy R8JDvRa33z3PvdQtumf5FOxgFh9t9CV9bHjwyyLnmlNXFW6jaV7GrTXcXhDhZ7u74tI9hl9rljWT ZXiaU5b0nOSbis5gr7KMgBfsS352wFRhdBgreDBurACHPu0xVKgqAD7vjF62nXHLeF44D0KHMw4K bsryI2uc0AJF/gUIdN9wEfXZlyy4rTzX4ZyUeHHuPR98/MG7IxJf2eY1LS0KZITpK9Gmc5+/P0QL I9NnjcgoDDgdtoYJs5/sPPzwptETJvpdoSxnxo3XNjzwg9/GsIr8yU/YI9IPsS++rYwMkAAN6Uda xpmvNUctstdJPILLSdz2NAd125mDegSdrJeNHs5uC3F3u2NuoQVkwC24cdTHoYRjM+wlTv6WDzqy 0aAr1BcSUkjnYZfgysAIjxB222c5qxzbHXsdQoujw9HlOO4465CIw+rgRnQR14M7ulP30GW3N8TK sU+Mxz7RTxzJAW5s4JoCbA3Wc6qmgKsX3g5COzsJfctWMqwpNFOoBQ6uOpW7OdNgOSmzhcpKynJt bM2AIS8j71rP/O99d02FQXfvvdQnhgcTM++LZKR/kF8yrW70Fnps8J0fJTaCPw9hl5mBL81cZJvi vs52k+1xSdBpvJpKVmnD/yrYTjHZwodqEw0uonc6oARBEwo7nTCcjFTMLvWkTKlL/8tJqdNyUVeP SC09i1v6t4/IK8/HyWdg/+cWtitPyObUtSGMQQZTw4a5CEFhyrgjbbfs/i71+qdXXXN7PvVunzX/ ht2Ps+6EZ3DR+KkrT9IBqB04IdUnmUd+nwr9i8+/6YWaAZsYf2MMsYC8ZOALzJGwLpXgG65xpA5f Y16NL7wm4avPqWQamY6vtGbjG6smICmxw/GHfwlLrmu4Jtp4baT69rbWWwsmLr114eSZyPo/MuNf EwplbmRzdHJlYW0KZW5kb2JqCjUxIDAgb2JqCjExNjY3CmVuZG9iagoyMyAwIG9iago8PCAvVHlw ZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9LS0JWSEUrQXJpYWxNVCAvRm9u dERlc2NyaXB0b3IKNTIgMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFy IDMyIC9MYXN0Q2hhciAxMTggL1dpZHRocyBbIDI3OAowIDAgMCAwIDg4OSAwIDAgMCAwIDAgMCAw IDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgNzIyCjAgMCAwIDAg MCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCA1NTYgMCA1 MDAgMCA1NTYKMjc4IDAgMCAyMjIgMCAwIDIyMiA4MzMgNTU2IDU1NiAwIDAgMCA1MDAgMjc4IDU1 NiA1MDAgXSA+PgplbmRvYmoKNTIgMCBvYmoKPDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250 TmFtZSAvS0tCVkhFK0FyaWFsTVQgL0ZsYWdzIDMyIC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAg MTAwNl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0 IDcxNiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMgL1hIZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lk dGggNDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250RmlsZTIgNTMgMCBSID4+CmVuZG9iago1MyAwIG9i ago8PCAvTGVuZ3RoIDU0IDAgUiAvTGVuZ3RoMSAxMzc2NCAvRmlsdGVyIC9GbGF0ZURlY29kZSA+ PgpzdHJlYW0KeAGdewl8VEW296m6t/fu9O0svSfdSac7IR0IZCEEIrmBBNDITjDBRBIgCghCCAqO WxhlMaIwzKgjjoLLOC4zQ2cBAzpDxvW5MPAcdUZHBRUVnYnw/Bgc0aS/f91uUObN977f792bU3Wq 6tR26pxTp+p21q65to2s1EkSqYtXtq4m7fHsRvTS4uvWBrUk2dxE+nlXrr5qZSKdtoJI9+VVK66/ MpH2biXyvrm0rXVJIk3fIR67FBmJNCtFnLt05dr1ibSnCXHNilWLk+XepaJ8Zev6ZP/0HtLBa1pX tiXoyx4V6dWrOtYm0y2Ia1avaUvSswYiy+/jeET5kX/UUKegYAAHfUWV9AAZiJNCRTSfSF4qZ5IO aVGu49t2fLX1zEJ75T+MPiMyiB7+OK9AxM/Lq7ef3TN0lUJGK5ImjV4UoJ5h4vAMmqzQ2T1nf6Qk ehIl5x7Hfpon5fdG3IEjz0oj6BiASyN6opmB/VKelNkzIaD2S6He1Ixie/VIKYgWi7QwiHAVYA/g IECmhVIWShWEtwA6AXsABwFHAHoihKI0CFgF2AU4BtBLmZK/JxhQqvMkD+p6MF+75KKTgDhAogDC IsBMwELANsAugF6jEzmrALcADgJOAfSkSq6eHSUYu6vnDi3qXb6iWEu2JpJNzVqy97LGRDx9diKu uThBNj5BNqY0kT1qUiLOK0zEqeHiTjTea7YVD1Q7JScm6cTAVyNk/AWyM0YB2i1lUAzAJQxVy1Gl 1N7cSPGug5JMTOISoyUUiA9IrMfmKK428zg/SakU4F/ywUQJH+xNcRTvqr6Ef0R7AAcBEv8I74f8 Q7qFHxM8R1gF2AU4CDgMOAnQ82N4j+L9gH9Adv4+FQGqAAsBuwAHAScBBv4+QoW/JyRGCwVeBeD8 PYQK/yum9VeEdv4usHf5u/EB/qee8ori/RoSLUoigXAScfmSSKqzuJ+/0fPNCEhUBCsNiXpGyqGJ VCLl9ITHBPold0/lskA//7g3GA3srh7N36QYgGMkb6LnNykImAVoAawG6IG9Dext6gRsB+wGxACQ MoQKIMhfBbwOeJtGA1TALICRH+lBN/38cE9kUqDayf/IXyYXOH6I/4cWv85f0uLX+Ita/AriLJS/ yl/qyQpQtQXlhDoKYgVxEcp1/A+9uamBeLWDHwQHAwiLAFWAmYCFgG0APT/Ic3qWBFLRyDP0KnQ4 wHvocy1+jB42kro8oEYmQwCDIoiMvwgYgl3BXRGuRu65D0kRRO7aAUwEkdu2AhNB5EcbgIkgsuI6 YCKILFkOTASRBQuBiSAycx4wBP38wadz8wLlM69mwWo7XwcurQOX1oFL60jm68RL38hijPf3FBSA YzvV6IiCQOcB1vks65zDOh9mnW2s82bWuYF1VrLOK1hnlHX6WWcW61RZ5zNsHFjRydS+C5IVqpt1 vso6f8M6O1hnhHWGWWcu6wyycrWfZ/dcDK1DVKtFvdVC6Xh270UTYX3sPBsczYbMZ8MmHER4GBDX UiqIgjkJYk+WiHN6C6oS6VHji1dVT+PPo+LzWIbn6ShAxgI9DzF6Ho08j+bsCKsACwEDgJOAOEAP 6hzMY5sW2hEWAaoACwG3AE4C9NpwTmIonFYhFEPcow2sCGEVYKZI8efx5uDN5tlqpuJXoso0aZuf 2bPYzKx4Fi8npxN2OdVhdPQz276vbf/82kamahO/i2+jTCzE9mS8reebzEA/+3lP5JlAdQa7l7Jk SB2roAgLIx5HHVq6jPxGkV9Kfv4U4uIe/3xUs/dECgMHWIqotS/wjf944HN/Pwd6wv9M4M/Bfpn1 BN5CzlP7Am/6bw+8UtRvRM6zkX6G6EBQI93vHxf4zasa6QYU7OwJ3CyifYGb/FMDV/u1grZEwRUd SKn2wJzIgsA0tFfjXxRQO9DmvkCV/4pAZYKqTNTZFxiNIUQTaAEGO8KvdRrK0hqsL+9nS9VCwz2G BsNMw1hDsaHQkG0IGDINPkO6MdWoGFOMVqPZaDTqjbKRG8mY3h8/pkbFrpeu1zY/PQSakazhCiwM E2YGIXFm5HQJxdKkOl43dxKriw0sprpFwdiZuaF+Zp69IKYLTWKx1DqqmzcpNi5a12+Iz4mVR+ti hlmXN3QzdlcjcmN8Sz+jeQ39LC6yNvpiqZMb9hNjjo13+kScv/HOxkZyO6+rclelTnRUTKn5N0GL ltlSE/3+cX+PRt3RzNg9dXMbYk9mNsaKBRLPbKyL/XRusKlhP/uKnaqt2c/+S0SNDfulieyr2jki X5pY09hY18/ma3QUZP8FOkgMItAZsTELOgoasxJ0OxN0YdQHXa6IQGcyUVijC5tMGp3MBF13R25t TXcuAtC4gtSh0XS4gj+keTUMmjAC0Dg76VWN5lVnp6CJTdSa8ftBkoUAJMxLfo3Ez7waiTbybo2k KEly+3mS27WepMRoNBoRoBnbsXM0tmOg+QEj/2e0bVI0ynonNC5uqm0L1baEatsALbE7rlvqjnUu Cga7FzeKgmBMirQsWrxUxK1tscZQW01scagm2D1Bq/cvxU2ieEKoppuaauc1dDepbTU9E9QJtaHW msbeqbNKyy/o6/bzfZXO+jd9zRKNlYq+pmr1/qWvclE8VfRVLvoqF31NVadqfZEm47Mauo00qXEy 1k/Evdxihry2+LIbJzmV1RM14Z2Q7b7ZdwDeyuNkiTbGrKFJMRtAyPXI6pHVogg6JYpSkG1PFrlv npDtO8AeTxYpyHaEJlF07bUd15K7dllN4q8DD7LWXiuWIhFGRd6/fUBSG1Nba4RvXRcrmFsXq5q9 oKHbYEBuS00j8safy7NYavvjA4nMUcgcLwgl6TyhyKsUeSZTkvC/y4I2JmSDO/vhaDzTy9QstpY6 GqVYVt08DlMwbwHY0LSg4QB8KbFJdDRigh0syjrOtSbmoeGUyCFMu+McrL02iSV5sTYZa6QdUYp2 nGPJueaigllaoPFqbRSmTXeAPACv7lfkkSOE80/8M8AJEQ8vi58Q5SLmX8DQ9SeB6HH6DVtGv6GD 9Bw7hVp7aD/1kXCBaugXdCP9jDZjW1uAnNtpDl4d8n/GPPE+nEwewob5EB0C7WV0Mx0gJ3PHP6db aKP0J9TaSDbKoWqaRavoTnZp/FpqoqPyrVROl9I1tJp1xhvid8V3xB+lX9J+6T/iQ2QhLy3Geyj+ pe4v8fdoJGrcTffRUbbDtJdU9NIJygdoDe2UmmUWvyp+FiPIpnUYg0zT6RAb4FG03kafMTe7UZqM Vh6Jx+IvgMpPzbSUdtIBVsam8mxdU3x6/BA50cd6tHof9dA+vP30O3qXWXWn4o/GT5GHCulizKeP /sgGpOGhDcNV4JsOXBpBFShZRb+nl+kIC7E/8FU6q65Yp+p+FH+T0mkM1WO0v0LNT9nX/Ga8t0gv yVPikygFfPmJ4Da9SB8yLytiM9l8PoKv4g9Ka8iIHsfgXULLwO+fo/UPIEb7uJUflh6Rn5K/1WcO H4unYEUidD89QH9gNsw0yDrYj9nb7GM+mS/k9/OPpJ/JT8hvGFox6ytoJd1JT9HXLJWNY7PZ5Wwp u5FtZj9h97FD7Ag7wav5PH41Pyktldql38mT8M6VO+RbdZt0d+hPDDcMvzD8n8Nfx4vjm2g25GED Rn83PYiZ7afD9A7eo/QR0zELS8EbZNmsnt2A92Z2J3uYPc6eYH3o5Qj7iH2OLekf7FuOnZbruQ/O j3CBQnwNPMyf8V/ww3iP8L/zbySXlCNFpTKpUmqUVmFUm6XtePdKH8pe+bAcB5+Ldffoduke1z2l e053Sm81/Bh7/OvfPTJUMPTBMA1vGb5nuGe4L/4hZWANsXvgCFaJ0bfiXY71vgcSt4f+xKzgnZcV sInsUnBmIVvO2tl6cPI2tpP9Uhv7b9mz4NKf2UmM2cb92phH8TI+ic/EewVv4+1wxnbwPv42PysZ JItklzKkAmmq1Cy1SWul66V7pJj0uvS+9JF0RvoOb1w2ywE5R47IUXmqvFC+Vn5Q/kz+TNeke033 id6sX6nfpO/X/xe8momGWYbZhmbDNsM+w5vGFkjn87SXnoYEnn/YMWmDVCvtpbt4iezBEeaPkOeF tESaziGp/HG2hd/E+niubr1+Ap/AZtApOQJev8R38TN8gjSd1bG5tJyPSTSoT5efBFYpP0+D8rOY 2x/R8nq9ld3MT+qt1AMfqQI+0ovSaDkqvUbvSkeZQX6I/iqbmYsN8l9JsyAFv5Mn6hooW/oF/VZq ZzfRXl5LZP7WuBVyPIM9CbswjxWzf0pxuMEzIEXl0sd0K13N/0KD0OMtdC9bIl9Fd1EJu5E+o8eg FSN01+gL9BnsFb5M7uJprI+4/ARmV8FymaRLp9tYs7RTf5K/Q9fSYdlMH0i/xugP899K0+VTujls KTTgJtpE7fENdL2uQX6DXUUSm09h+Ris241SsZyN+BZYlSbYtH3Q7gOwA9XSdOS4ITmXQi7qYSF2 4v057IQMCVoGHb8MVuyP1Kefx/vpKl0Kg9XBTc1rw3NoQfwxui9+FV0T30EjYQ82x29Ei4/TJ7SN Hmcbh2+g1ThKvgPdvlQ3hR/WTYmP5F38HT6X33Ph+oLbYeamL/D+FiszUfcMdcl/prlUFd8afwvS nQ8Lex8tgsN6HLP8Ej1MkwaoZHgG745PkVZjvkdpdvxX8QAz09L4CppJz9IvDTpqNUSxxjH2BuZ7 A7XxOfG1UtvwMvBhG7igglvXwv7crk6un1etVk28qHLC+Ipx5WWlJcVjRheNGlkYLRiRnxcJ54Zy soOBrEy/z+txu5wZ6WmpDsWeYrNazCajQa+TJc6osDY0pSUYi7TE5Eho2rSRIh1qRUbrDzJaYkFk TbmQJhYU9VpRdAGlCsor/4VSTVCq5ymZEqykypGFwdpQMHaoJhTsZwtmNwC/sybUGIwNavh0Dd+u 4Tbg2dmoEKx1L60JxlhLsDY25bqlXbUtNSMLWbfFPDk0uc08spC6zRagFmAxV2h1N3NNZBrCXbXj uzkZbZhizBuqqY15QqiKZqRwbeuS2KzZDbU1vuzsxpGFMTZ5cWhRjISnFNVIaLLWTUw/OWbQugku g48TozuC3YUDXVv7FVrUErUuCS1pbWqISa1oozbmiKLfmpjrR8fd3yfROHyyzT8s9Uldte5lQUHc 1bU5GNs9u+EHdX3ZooXGRrSBujw8paVrCrreipWqE754jG9sbIixjegSjmVYm1VifgmvN9yyPBgz hSaFlnYtb8HSeLtiNOf67B6vV90fP0be2mDXvIZQdqzKF2psrfF3p1PXnOt7PWrQc2HJyMJuxZFg bHeKPYlYbT9E2sD0RJmGaeQCq5tznrNMjDF0MTzBWHBxECNpCGFO40TQNo66Fo/DAuBpZKgVW4IV WRYzTW7pUsaLfEyRxXRhJRTs+gdBAkKDf78wpzWZow8r/yBRKOTkvKjFWOs5PBaNxgoKhIgYJmNN McaJWrpsZOF1/TwUWq3g/CwODTQLvG1tHF8E9mdniwW+o1+lRUjEOmc3JNJBWuTrIbUIvjVvESUD 50oy6kVJ57mS89VbQpDkPnGepYyYMXL+z64402qXjo8x5/9Q3JYor5sbqoNrHKztaklKbd28C1KJ csFQ8A1lSSyWNrlB8nHkCYz7JK004SGfI4G73GCNyWH86TWhXtJvMEIqtRwWnBJTWqYlwkZzdnZS Z/5/lfrjp0QtLfq+WnIasfHR5EATw45NuCB9wfCsXVLdPJgcDs++q8t8QRlELTHKi5MRJB4H/ezg 5BjVQzPD+MORY5yARl9MBctQMg9apGU3+pLJCwh9yUqNeIR0jiycApvZ1TUlFJzS1dLV2h/vXBQK KqGu/fw5/lzX6lpYu4Tg9McP3OGLTdnaCI4tZeOhHpwmdYfYltndKtsyd0HDflxxBLfMa+jhjE9u mdQoloVPnteQZIu2IEL0sYb4YgKJEXu8IZPW6V6mK+UOugQnt1mI65Cfxl6mLYBb9djHkSegBuVV KMsFfitwL0SOaYJH+Eqj104aQeyVQhQTj7ho+d8+0v+2IuphcPDdxSQNWiuJTycmMuPcIb6WiMcG /9yOry6JJwM77OdsE3zjV+THdXN0r+svMSwy1poi5rOWrdbLbctxWwRKnWhWEq1mO7IdYQS4WaLv gtLAd6qOvqWgPCCI1sVf1e/S/Qm9uchHefBwDKp5u2e7ly81en0+ca9qd3vS3W6P25dh93jHRFOf 5bvIBMfCynepFsnr8UjM53aH80V+APmj+K6esMX/LN9JUYxkDN/Zm/PrMr1IZyBtR5OmIFh/bell C9xR5Uzz4OlB5QwCqhocGlQqlcoqgAKcOVIrKgRsHhW9SXlhzGj35OvVOaxkRFY0QCXBMQE2MgKs KBeYjdsD5JIzAsxhBpZmBFaQmY97xWwEhXmjAjQ6hCCFWQPMqUOgWFIDlG5AkDz/4kbl3Fl2A2tO Kx1bUgzvQR/KibAcfUa6s6R4bFlpRGIljP0/ytY9eE/X3qc3bexmFZMbF0yqAUg5O777kH3y4L0o 2IyC8SKztnGBvOCB9148eOCVl9iLa++/s2Ptzrs6znboTd98ze568K+i4GX2wtr7t64VBWDWlfHP dNdhnTLpT3sX8+WZnPXHT/RZLPp6AqIuFFiQim2L4cWtzeyk2zK3007dU9IvbfulPtvLtiN0PPP/ ZDpSUjMdmZlSgT7fUeAPBqba5qdfljHfs1R3deYNqXek7pTuS9npf5w9yh93vJWSBn/Sq6QrXpn3 xz/oya9AnwPqyPwKxU5M9qVlWSVflmxSIvZLKBJkjHkDrkjQyIxWMRqjJ2txkzs6QzkdbZ4+OAPr PD2xxlWDDlfFmNHgdXNzOzWD62uYSy+HcnJ5WWlqbkmx7DJEIqEcPc9ITxVMl/ueu2j4+U8Gh/98 /x42+bn3WOGEgyXP/fSJj5tWfrrpkY84H3Py2z+wa974hNV3H3tt5O4dDw+f/Mkzw593PQuHnC6J n5D98kR4peV8pFpospkKPDZvwQhbQUGFbWxGuW98wcUFzbbmguW2ZQUto7tsm0bsdN7vfcKWkX+O xXmCxR4xqcc8T+bv8zyT/4LncP4bGe/nG2ucLKs/flp1WK36+tRUEeqsIiwTl7v1Agu4Au5oYUFp hVxReLE8rXC+sTF6pXFZ9DrrZusr1m9s30Qd5aUpTFaKcktdxdnp7oUjVo3gI/xFKVUp21J2pcRT dLtS9qScTJFSrHY7r0/pj3/RpygaclrNUBR9fYrVZkOot9sRpvglVz9/cp/77nS/3wDxOK16xTio Ns9c7JcsI1qVVtLbbLyewtm5/fG/a40JRLWI3FzZakXYHz8O+dKQ05qgIec91SK6y9U6Qvq7PtFw bj+/XE3JUymiRIKR0ZE9EV0FJKUvJYXXR/rjb+/TkDEiT7VlhUpHVwxU8N0VrMIlxlYtWnSF3TlF uQf1h/U8oK/Sc32KmKke00LoFuPRQ6gSob5enyKmq1dE5/ox45QzSbVtboeERZXm9ihyYFbOaXPl UPSTT2Bdqo5HYWKOw6IUaUWgb0dCWBiYGiGUQhwhkO2IqD2sF7pfVjp2bLn2lpXmCbE05E3ksAxO HC1gFFyhiKQ3pEBWhaiWj8UFwZL9y/c8O7VjWtnV717FSmq33HJ9Zsx9zZHbtzw5SzG5cp71uxa9 sKqpeOWypQ9HMm+tn/LUxhkbZqSn2Ly5YfM1Iy9qbHe331Gntl4yav2pbzdeNI69n+9X8qcXTWu5 fOZF62ALZsVPSIOQaC9b0M3FtqmWptxiZ3YLU3GvtRpWX071Wwxuv4z7jwyDUbDfoLHSANEErghW GjQeHHrzJcGYQeWF5mIBY0b71KkmKwv4J6dNds1Nm+tqSWtx3c/vl3baHlUe9VqNNo95OV8mLddd a11t67Q9Zt1r2mfea7U6rZusH3MpJWehfZX9FrtkZxBC9frROLjNohYMazvtpmN0Cj8AsNst2PnO jdGPoeemGDUJzvFhfrmWaADfo2FQVKFzTBUSwqaJ1WZeQcYu9mfkHjawgKHKwA0pgshgFkQGTQEN Y3ylL4idBUuJJU6IR/OapHOBDx3CCxpcczo6uEabO8TBUVGkNB/HnxCBdtbc3giThNUnR2mq2AaS FimxyFJld+bJ3747/PWaz2//zXuBPZ5bFmx58tHblt/FNrqePswymfnXjG/Y85Dv6hXP/+nt534s rFAdrFAW1iyDMtlc1RUgfwavl5p1zaZ6S5t0tW6Vqc1izBA6JyboAKLOEVimX4R5qe/ozqaf8cpj Usd7xvirU6d7q/2zU5s8c/ytqSu9rf71+vUZZ/gZt4LrTrvN5ZrlbHGuxod/v327slvhiiL7/GYD HeBPitlrCq8Zc0VwTQGn706DtLhUG3QcI+D1QL7UhgLkC03LgQyopryC0piN2bwBpHrDkVIRq9VC qQMs4CxRcg1qbkGpWJiZBskQ1ARN018DJgKhSyyWX6ylwSmmZvBklZYntomETkanDx2fobRHo2cA 0eh0zS2APh6vGsRCNVcOtVdqfkGq2EOaNYVl7WIH0ZZLoZJicqQbsp1CHVl2RFNZ6YoDhV/u/3z4 JEt/7y3cCH53wtyzcfHWoXf5bOu4+bff+ASb73qkjwWYhOu3/OEPhr9RgnsOLGV3b5q89DFIZBqW sBO7r4vZ1Kx0E7N7ijyjPapnted+6y9sT9iMXlu+LeYZ8MgewY98b6A002iTrHa/mWXwaHqajN9Y mHels/R4miq7wjLumXZAxAUTx4wrFbEa9QdKtxPzqMLyeVQblJbShZmnfJFDOUKNqVBwEfRfqRah ypQuhB9psSNoyKd9YhNAztmnBZPpEbfnWXaAsukMblvwoe6crRS8BVcrldNwtgaVwcFmqqqqrKwc qqwarHCAt3C10hWH3mTQG2GPFVOqjxx6uw83ZtGCDRtYtL2Z1pQ4QmUlZaXlY6EjUBEYx4yMkoyQ o2fXrjTvrddd2uQbVzyn5vBhaefW9qtLp1yW+oB5Ssuird9dCY3YApe8Evf/wkc9pF5hGitkZaZp u2m3KWYaMB01nTIZyBQwrTZ1mnYls46Z4iZzwATbYJC5ZNJLNzPS6/SyWW8I60jeJe+WY/KAfEzW D8inZE5yUD6ClCz3x/+pCT2Qs6pLcFLWdjnZLHqVNS6iLMFFIMOa7AP5TjULNsozjFNnJU2K4Bzs yhqwSrCs6ns3VQjkmvZoWllJhuQocWzp6+uT/3b48LcZcuTbd8UJ5FYE5dqcP96n0yasEyJQPq5U i0vLEvHoMYk4J6zFajjDVWrXBXC5fFQnz0RwSicFdKt1nbq4Dt/SycylcEKY0JImTBlYlV3EBmBu cbIJ0hFYXhklCS4I6VAzNXnSuEAaF5KyZEwKUoIFII1r6g8kyQuaIV/IC8EMSJLGDsECkRJPCXhw a5/uwNkpYu6bcciJwAKG2Mv7yRY/rXHYCA9AQ0z98b+o0y220rB8XD5u+tD1SVD3lu5MkLuMwZDJ 7QuaJCmU5ddn+CHwBqYPeT2K+UiYbQ/vDvOwy+VNCW93MIcsFMbhFgoDGwqHRkzSkS5UBekTqkss t4MLpXFYxUQh0lAXlJ3VZgnkn9p48BuHZtXqDm/34XAjmvOdb86nNYf0l6pDNOeTRXM+bQNC7rDw opCGr4bwnI/kE+3hN1gloTA7QkzshTxAVbj7lMDZ4eRqKJp2w6+BZmv+DjmT+v3dOf0+raZrCq6J JWmOEHlyw/1sfW+2WJbojPNujyaksKbwwM89P3CKIMNDM2rbaj5txy4IxYcoT1dgCOCcC/sqzCtM QIo1PS2SbnX4WKotw8dwootGcTjSzC7WN0NovtMlggxHyFGq+e0JDMZAn+HY/FDxY8uvuzdw86sP Ptkbapq4+md9DUsu3TBejtw9Y+GihgN79g3l8QdWLBx/96ND9/Ke9etn7fzJ0DtJeZE+hbw42U1q mk7Sp/HHlX7lY+mztFPSmTQ9dPOUWgmBuV5hP1eOuI+54245aExPSXem+nWQEKfNbEuxpuS64T7w ejdcCV5vyRe4JV1IhUUsqEMw3KIpgSVHoxAuqUUw3QLji7A//k1iQS1mwXekz6ial2FRS8aWxi0M f5YZbmHBvaVjS2PuU26+2r3bHXMPuGW3xEsynJpunulzODTLfVYzRJoKaoaIfqiCmihpZQ5tlblm w2UhUMgdUFPR5yko03mVnuHCwUrbKBNLDC2EQRce8A9zUTYIM491rsLpq4IlF9epd5jMRrPBLOmV iEOf4mN2c2pykQuwyu0EEdJWOenv/mCJNz987fstD81SzH0FV0/r+JUcuXdP7erpxTcNdfBN16ys 3vH6kHb6qoHfk4dVtJGHXb0vwy1mkibOVQKxC5XsEJhHK0g1mD3Wqfppxvn6RuNV+mVGY6kyPnW8 s8xdq9Sl1jlr3U26JtMcpTm12TnHvVK30rREWZm60rnEvY5lmPQ62+XSPN088+XWFVKbrs28wmp2 +WWDAyYjPdeniqX2aWJggKVXwWD4I+j3vEMMidIOOig+pY1PQ8Q6aIjms4hFSMsNl442MDIohiCc nDFHYSNE/sXCDQKekkvWFLFlp2rqrPnchEEgR3N/klqr2R9yaiusoklhDjiN8Qp3CIt6TmWjg3CG ms80f58RFS5R1SBMLfbgdlxaqqa5urmmRbpFJpk14zM/njSlHEpJuL4QblHaucsLKGXNo7e/+Ffm vOFvdxwdHtzfs3lTT+/GzT34hJV313XDHw4d+tuPWRazvf7a6//54muvQtSqcNroxgqOllzqDXJO es540yWmmtz5OW05N5ruMt2W+1jaU4XPSTaTy+t2ja4rfNul8/F6zpViZnY3GZtMTeYmS5O1ybbc uNy03Lzcsty63NYX6cuz50Vy83JHjM1dYG60LIksyV8bWpvbmftT8y+sO/LvLbx79KPmJ6yP5D2K 3/++GHHiMJ7YKXLOIaFzSO45RKMR/NRoBKLRCESjEUgmrjLU1KyKBca8sNUse4ORDNkyKtMrDis5 nkKxSgFPlWemZ6Fnj+ewR2/3BDyrPEc9csCzzcM9v8MiZsAb1PxoFX4Dh/usMq7gazInpjBxMTPQ m+4sTfjXKY5SxkY1Za7I5Jn+DAOsljhMw+Xoj3+qmXSBqGlCyGT/KEvAy7y5HjXNXVosJKpIyKhH 23+EhgCHuCAMCtPkCYpaHu0U7NF8aQ8O4j2G3AJU3euvOFLAgH2KfY3XAzkBxzCJCD4g54t9QvUK vFpX2fDsW4oHinlVcWcxLxZnglzS+iTsSjA/wQSXeb2GiAEIRPWIQQRz7dqZ0q4Nzx4UQ4Vyn1Ux RGApokO7dkdhzzlKTGx7nDxjko5/c/v05A41CMlVYKvWzEge4qPRdpwAvt+/tBMbiKoG27VDvPA1 jitDWpQ4yScP8ti61LyRWSFdemHEoaQqaYqkz7EFfWTKN/iYbiSCrHQks1NCPsoJ2azGEWYfy88z mfVR2UcBJVPYwai4jUwE2tVgQXTDhg3nbwthHIW/15xWrh02xPVAHr7Dl+IWQGyI3x8aXbgrcGXh ikAoY6Sqx377DTeuLwv/9KX7ZlaPK/jJ3Jt+t8ARs3Ysu3G501nku+3gvfOXvXTT4XfYRf6r17TV XBRyh4sv3jBj6vX5gei0G65yz2maUx7yZ6aZc0uqb2xasOuyX4v9Mjf+FS/Q3Yd73L/sJzNkMBQp hVeF0xmQTg98ZqvNzCRyKqao3ax34ibIruRQDrOlhq0sbsAVcm2LYbWh07DdIBMs225DzDBgOGLQ w/p9qW1aQE5rUgTkK+16B4jwl2BCBSIkDYjY4RI2U7Vo1lMv1h8FCatvOMCXk5uN7b7yhw41mHn6 uDKI/Us5fho71WAVUAfcEEdJifKKcCuj0bBL8C9SJk4djnJ4HyFHumA9V7yXVi5aUXjbbb1796ZF 87Me2qVMbHuYL97KDCuG79w69NPphfgEAP8btuyY+B0Sm7mfvOCNCZ41D6Y5SyGtp9SS1PTSaBrL NaY5rSzNacGZzQE2UYkz7HaJ7d6r+RIuzYtwpQovAldYyWszl7aPa1damv/gShdcQPqfCf/BpTmE SJ9RbYIfcRcbcDHXDK9YowzhOnhPeflq725vzBv3yl5rWFs9Rez4OPGYgqYjpmMmGZkJJ14gqkOM waT1bIITr69H7hnVKbhu0nwHExfKa5rhucBlh28w+N+dhMqh44LvVZXadVjC//PKSorNbuN6A362 qjPCUZCtPrIZHT7cmuMkWLCBmoVqZJdpS5OHxSlxpLs0hRgrcKnqxreueGSmYumzOK6ZPfuuCX2/ 6Ju2cmZZB98x1HvnmKmz527bwitwPGLiK410AqtjZl8k77VcOiOZjXqmN5POZNQxrssVBkVXFH3/ kPL+IYgGTmBVwpvxPV2mY5TjqDAL+25zVJjgBpYaRYDL6y96EcMgazEo/qKasrJLKR8BUidUE05a 5ESA1LvqzfmjSimIwG4dQfn4sFJBZeZpNNU8Hz+KajQ2mK5kV/JlxmWm9bSOrePXG9eb1pk3s818 k3S7YYuxy/QA/dz0E/Ov6WHz7+hpQ7f5FXrR/C69Zf47fWz+lk6bCzEds5uc5nyKmMvNM0k1m3Rq qrNUB1Ep7dZrd3omzEdMnTCm06pdLKOZNBsqeCHy4F7gmh1c0XK5Tme1wAYWvR8FbwCHooeiVFRV hTxxq1duNhiNYZM53WQy4/qBwytNxxWEzmwm/C7CyDnTG8wmiZiuyMqsOUZVVXHy5qZ+5tur4qjJ dcDw5YarLMfyxRtCdwe9nqHmoWave/B4M3qBslbAxUx8wXFUbNaNim6+6YXNo9wiasTNWvJmFeKS fKi5MZuVpOEAUZ6Gryq/HV7x++Nh3Jb/ff/wNXJk6LarVs27jm8RZ2ecYsUTz8PvpP7d40CmhF+q RfBLwCk0Db+DuxSbzCz8vmoO/lfpMmqgRlAwSgWIR49valRXN2n+tNpo9ZplrSumz/u/S17KzQpl bmRzdHJlYW0KZW5kb2JqCjU0IDAgb2JqCjk3MzYKZW5kb2JqCjggMCBvYmoKPDwgL1R5cGUgL0Zv bnQgL1N1YnR5cGUgL1RydWVUeXBlIC9CYXNlRm9udCAvWUdXSEVFK0FyaWFsTVQgL0ZvbnREZXNj cmlwdG9yCjU1IDAgUiAvRW5jb2RpbmcgL01hY1JvbWFuRW5jb2RpbmcgL0ZpcnN0Q2hhciA0OCAv TGFzdENoYXIgNDggL1dpZHRocyBbIDU1NgpdID4+CmVuZG9iago1NSAwIG9iago8PCAvVHlwZSAv Rm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9ZR1dIRUUrQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRC Qm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVz Y2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1 MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiA1NiAw IFIgPj4KZW5kb2JqCjU2IDAgb2JqCjw8IC9MZW5ndGggNTcgMCBSIC9MZW5ndGgxIDcwOTYgL0Zp bHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLeFTVtV577zOPvMgkkOdMMmcyZJBMYiCA AYLJ5DEBjcgr6AwNZkJISXhIMAHEIgxtERxQKNeiYhW1D7FWOZlEOgFbUmltRRGuUlrRCmpbtV8R 6r1qfeXcf58ZUFq/3nPyr7X2Wmu/1l5nn30yvbes7qBUCpMgX/uKtm4yrrzFYPnta3rVeDktl8jc /M3uJSvi5ZHLiUzvL1m+7pvxct5WIvbXzo42WU9enwNXdUJhlIhNBB/duaL31ng5rwXcunxle8Ke F0bZvKLt1kT/9DrK6s1tKzrAcdnvA1G7V/b0GkWyS78rum/pSPizAFHKL3Vc0n7iwwYKhyAwgNMH NI0eJAskG5XTDURKp1JAJpSl3cR37Ppg+0et6dM+tNqtUBA9+vaYEsmPKN07P93/xRIbWVNRTDL8 pQH1LNXD11O9jT7d/+lttnhP0nLx4oPULK7o9+Q6TzwjxtJZgIuxUW+Bc1CMEQXRKqcvJtz9mVkV 6bVlQkWL5QZVQVcC+4HDgEKtohBWG+hGIAzsBw4DJwAzEai0qsBKYC9wFjCLAuGIqk5b7RiRh7p5 mG+6yKHzgA4IcoKWA7OAVmAHsBcwG35SsxLYCBwGLgBm8omc6K4JGHtOdJvB+pcurzCKbfFiy0Kj 2H9jMM5nzonzhmviblPjbuMnxtVX1sX5mNI4zyyuCKPx/uS0iqHabJGNSWZj4N2gjP+a0hkjJz0s skgDuMBQDY1PZPaP9lTsPSwUYoILRovJqQ8JFk3LqKhN5jo/T5nk5O/zc3ELP9c/IqNib+21/C3a DxwGBH8L95v8TdrIz8qYg9YAe4HDwHHgPGDmZ3Gfwf0Gf4PS+Z+oHKgBWoG9wGHgPGDhfwK18ddl xhhUyjUA56+D2vhrmNZroOn8NKTT/LQ+xF+JVk6pGDQEb3lCcBYnhBx7QsjMrojxl6OfjEVGebDS yKhDooiqaYIoihaPd8ZEbnRalzPG3+5Xvc6Ha8fxk6QBHCM5iZ5PkgrMBkJAN2CGdArSKQoDO4GH AQ1AloHaAJUfBV4ETtE4wAfMBqz8RBTdxPjxqKfOWZvNX+K/pRxE/Bj/ncFf5M8Z/AX+G4M/D14I +1H+XLTQSbUpsBPq2MBt4OWwm/iv+kdnOvXaDH4YEXSClgM1wCygFdgBmPlhXhRd7MxEI4foKJ5h J4/Sewb/CT1qJd9Sp89TjwRUJfFMvRoSyF51r4f7PLvvR1ESz927IEni+e52SJJ4btsESRLP8jWQ JPEsXgpJEs+CVkiSeGY1QwKJ8Yd+PnqMs3LWMqbWpvO1iNJaRGktorSWFL5W3vSJIsf4QLSkBBHb 4/OOLXGGD7LwMyw8l4UfZeEOFt7AwptYeBoL38TCXhZ2sHAhC/tY+BCbjFCEmW/gsuIUXy4LH2Xh J1m4h4U9LFzMwqNZWGWVvhh3Ra/BUwfmN1h/rXzouKv/6mrsPunchYi6kPMu7AmHQY8DulHywUkt ijvnFUpe1F9SEy9fObViZe0MfgQVj2AZjtAZQMECHUEaHUEjR9BcOmgN0AoMAecBHTDDuwjz2GHQ dNByoAZoBTYC5wGzMZzzGAqnlaByiPuNgZWD1gCzZIkfwV2E28VdvgKbw+a1zRA7HCy9kM0q1At5 JWVnY1/OzLBmxFjagY/T/vlxGiXVJvG7+Q4qwELsTPAd0U8KnDF2X9RzyFmbxe6lQgVZx6aQhxWD T6YeozyJHFapn0gO/gR4RdRxA6qlRz2lzoNshKx1wPmJ48/O9xwxDvFdxyHnH9SYwqLO30PzxAHn ScedzufLY1ZonvHEGNhB1XAddEx2PnnUcN0Ew56oc4NkB5y3O6Y7lzkMQ0fccFMPSr5051zPAucM tNfgWOT09aDNA84ax03OaXGvSbLOAec4DMEbF0sw2LEOo1N3odHg/MoY6/SVWnZbApZZlqssFZZS i8vitBRY7JZR1kyrzTrCmmpNtlqtZqti5VayjorpZ31e+dYbZTZefmYkNCPFkG3YYZjcZkCJMyun a0kbKZp407w61qQNtVPTIlX7aJ47xpLnLNBM7jqmZTZRU3OdNtnbFLPoc7VKb5Nmmf2NQB9jdweh 1fjWGKPmQIzpUrXZrmXWBwaJsYzNd9klv2LzXcEg5WavqcmtyazOmNLY8DUkZChDDd4vr9wvRW+u t0Db3TQvoP20IKhVSEEvCDZp/zVPbQkMsg/YBX/DIPuHZMHAoKhmH/jnSr2obggGm2LsBsOPVPYP +CFjwOBnxYtZ+pFqLYz77Yn7FaM+/EZLBr+kJCo2/IqTkgw/hUm/vp7R/oa+0SDwyVGpx/DpyVG/ 6nO0GD7FIPDJDtNRw+dodlj6aNVGMw4HXApB4MLyyWG4OFi+4WKMvM9wKU+43HnJ5U6jJxEfjeEj CZpJO3vRJ+0sfL4SyP8sdtR5vay/Ktje4u9w+0NufwcQ0rat6czVwotUta89KA2qJjyhRe2dkrd1 aEF3R4PW7m5Q+6qMev9ibpHmKndDH7X4mwN9Lb6OhmiVr8rvbmsI9k+fPbHysr7uvNTXxNlf09ds 2dhE2dd0o96/9FUpzdNlX5Wyr0rZ13TfdKMvMnJ8dqDPSnXBeqyf5P08JRn5GrK7gnXZtu5qI3mr XLkb7AdxWtlHKd6gluqu09IAmddltWW10oRnSppGQJ2eMOVuqHLZD7J9CZMN6gx3HXl7V/esplx/ V0P8rwcXVL2r5VLEqVfqvvaCi1/ztTXIs3WTVjKvSauZsyDQZ7FAG2oIQjf1oi4lxR/Th+LKK6Gc Kh2FuOQoddOkLikp4fjvuWCMCWpEZxAHjUP9zFfIeqknKLTCpmaOraB5AcLQsiBwEGcp+ZLoCWKC PczLei62JudhyBTXEKbdcxG9qxNSIha9CW649njJ23MxJBeb88pgGcSIVa8XW5vpIOUB+abHKE/x EL5/9HeAdyUf7tLflXbJ+d+w0cUSINpHT7IuepIO07PsAmrtp0EaIHkEaqAf0Hq6h7bgtbYAmjtp Lm4T9PewPH0AXyaP4IX5CB2D7420gQ5SNsvV36ONtFm8glqbKY2KqJZm00q6i12nr6YWOqN8hyrp OrqZullYD+h367v0H9GPaVD8Tv+CUiif2nEf0983/VF/ncpQ4/t0P51hu5KeJh96CcPzQbqF9oiF CtOX6J9iBC5aizEoNJOOsSHuResd9A7LZetFPVr5oa7pv4aXgxZSJ+2hg2wSm85dphZ9pn6MstHH rWj1forSAdwx+gWdZqmmC/qP9AuUR6V0DeYzQC+xITH8xabhGsTNhCiNpSmwrKRf0m/pBHOzX/GV plRThclnuk0/SaNoPM3HaB9Dzb+yj/kG3BvFc0qjXkcjEJfvyWjTb+hNls/K2Sx2Ax/LV/KHxC1k RY/jcS+mLsT7PrT+BtLoAE/lx8UPlSeUz8wFw2f1EVgRDz1AD9KvWBpmqrIe9m12ir3N63krf4C/ Je5RHldetrRh1jfRCrqLnqCPWSabzOawb7BOtp5tYd9j97Nj7AR7l9fyZr6MnxedYpX4hVKHe57S o3zHdIdpm/nd4cDwr4f/e/hjvUK/g+YgHzZh9N+nhzCzQTpOr+I+Q28xE0thI3CrzMXms2/h3sDu Yo+yfexxNoBeTrC32Ht4JX3IPuN403Izt+PwI49Abn4LTpj38B/w47hP8L/zT0SOKBJeMUlME0Gx EqPaInbiflq8qeQrxxUdca4w7TbtNe0zPWF61nTBnGr5Nt7xL37+wy9KvnhjmIa3Du8ejg4P6G9S FtYQbw98gk3D6NtwL8V670bG7adXWCpil89KWDW7DpFpZUvZKnYrIvldtof92Bj7U+wZROkP7DzG nMYdxpiv5JN4HZ+F+ybewVfhMLaLD/BT/FNhESkiXWSJEjFdLBQdolesE7uFJl4UfxJviY/E57h1 JVlxKkWKR/Eq05VWZbXykPKO8o6pxfSC6S/mZPMK8x3mmPkfONVUW2Zb5lgWWnZYDlhOWkPIziP0 NP0cGXjpYmfFJuEXT9PdfIKSh0+Yl5DPrbRYzOTIVL6PbeW3swE+2nSruYpXsevpguJBrJ/je/lH vErMZE1sHi3l4+MNmkcpP4U0TTlC55RnMLeX0PKt5lS2gZ83p1IUZ6QpOCP9RoxTvOIFOi3OMIvy CL2mJLMcdo4/JmYjC36hVJsC5BI/oKfEKnY7Pc39RMmfWbcjj69nP8W+0Mwq2D+FjmPw9ciiSvE2 fYeW8T/SOTzHW+letlhZQnfTBLae3qGf4KkYa7rZXGLOYs/zLiXCR7IB4srjmN0UNpoJ0yj6Llso 9pjP81dpNR1XkukN8TOM/jh/SsxULpjmsk48AbfTHbRK30TrTAHlZbaEBLuBipWz2N3WiwrFBb4R u0oL9rQDeLoPYh+oFTOhyUXmXIe8mI8dYg/u+7BPKMigLjzjN2IXe4kGzM08RktMIxh2Hfyn5oXh ubRA/wndry+hm/VdVIb9YIu+Hi3uo7/QDtrHNg9/i7rxKfkqnu3rTI38uKlRL+MR/iqfx3dfvr6I djHLpb/hfgorU206RBHlDzSPavTt+u+R3Vdgh72fFuHA+mfM8n30MEMM0YTh63mf3ii6Md8zNEd/ THeyZOrUl9MseoZ+bDFRm8WLNdbYy5jvt6iDz9V7RcdwF+KwA1HwIVqrsf/c6auf31zrq6m+elrV 1CmTKydNnFAxflz5lWWl3pKxV4zxFI92F7lUZ2GBw56fl5uTnTVqZGaGLX1EWmpKcpLVYjYpgjMq 9bsbQ6rmCWmKxz1jRpksu9ugaPuKIqSpUDVe7qOpsl4bTJd5+uD5zX/x9MU9fZc8mU2dRtPKSlW/ W9WONbjVGFswJwD5rgZ3UNXOGfJMQ95pyGmQXS5UUP25nQ2qxkKqX2tc0xnxhxrKSllfSnK9u74j uayU+pJTIKZA0nLc3X0sp5oZAs/xT+3jZE3DFLV8d4Nfy3OjKpoRxf62xdrsOQF/g93lCpaVaqy+ 3b1II3lS8houVG90o5nrNYvRjdqFM45G29S+0qHI9piNFoW8qYvdi9taAppoQxt+LcOLfhu0nNv+ nPtlEY3jTLblq1a7iPhzu1TpHIlsUbWH5wS+Utfuki0Eg2gDdXlxYyjSiK63Y6Wa5Flc45uDAY1t Rpc4WBYbs4rPL37qLQ4tVbUkd527M7I0hKXJj2g0d50rmp/vG9TPUr5fjTQH3C6txu4OtjU4+kZR ZO66/jyfmne5pay0z5YRD2zfiPSEkJr2VaEDQY/bDMlwl1LT3EuRZXKM7mtwEtTUdhUjCbgxp8mS dEymSPtkLACuIEMtbTFWpEtLqg9FbFOlHlNkmqnY5lYjHxIywH3u75dr2hIac7HtQ5JGmSeXUk1j bRdlzevVSkpkiljqsaYYY7VRnlRWuibG3e5uG76f5UcDzUZs24JTyxF+l0su8LaYjxahoIXnBOJl lRbZo+Qrx9mah6Rl6KIla760hC9aLlUPuZHJA/J7lrI0q+fSX7ote6S/c6rGsv+DuSNub5rnbsLR WPVHQomsbWq+rBS3y4AibrAlJG1kfUDYOXRS4nZhWOMn5IsuOC4HUjWlGH9mI6kXxyxWZKWhYWqj ZgvNiNNgssuVeGb+v0ox/YKsZbAvqyWmoU31JgYaH7ZWdVn5suGlRkRTM7YcjpN9JJJ8mQ2pFh/l NQmGjMeHvkut12g+nsxi/OGTY7JE0K75EDJYmvEUGeqgPVG8zNGeqBTEJbOzrLQRe2Yk0uhWGyOh SFtMDy9yqzZ3ZJA/y5+NdPux28UTJ6Yf3GbXGrcHEbFONhWPB6e6PjfbOqfPx7bOWxAYxL841K3N gShnvD5UF5TLwuubA4mwGAsiUx9riB80kDFKD86uBEEmD+GXFjNANO6SBilF/wMNtnZ5GDbhH/r4 rYJcGa6MYhC0QJ+rYuhzn4k+I1UZgifN199RMkxD+D2jgM3v43IAvuT8QsU0qjAtLScppr87kJ7O 50vBl5eWBimDUqWGslNTQVOljsrx/XMM5BjVnKs5N36cvc/87y39L1oyy5b+OpCWZgjv+/JSUiBl kE1qyJaaKqnUXWryyzYHzGqezYEkinIVv9CcpWwgE0jHv5AWKeYtfGvK1vTnR5iSLCm53D/yuqxr 8+rtzSNbslry5tqXWZaltI9cnrUsL2Rfx9ea16Tclr7FfJ9lt+353NP8lPlUymvp+Zcm3pPkc7kn jktilGRL4kk7nRk9hMTxjYBWxWuY087C327L9do+Wug9B7LKey4xdbZwFT5oJsuLAcHgSFvmVRMq srMzs2zc7C4a4xlpy55QcVWGzeMuspjnL3vl4TXR3rqlrzxyct33Bh9fv/7xxzesv3Yhf4Up7Oqf tfYP66eHh4ePPHnfz9mDw/eev4CvhKXvd92BFcYgjEsfg3Pz113SPtIwMMpMZI0Z3zkUnH7jDL/f W3tLV9vymc3/B5FmWVsKZW5kc3RyZWFtCmVuZG9iago1NyAwIG9iago0ODMxCmVuZG9iagoyMCAw IG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9RVkxDU1Er QXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNTggMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGlu ZyAvRmlyc3RDaGFyIDQ4IC9MYXN0Q2hhciA0OSAvV2lkdGhzIFsgNTU2CjU1NiBdID4+CmVuZG9i ago1OCAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9RVkxDU1ErQXJp YWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5n bGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAv TGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRo IDIwMDAgL0ZvbnRGaWxlMiA1OSAwIFIgPj4KZW5kb2JqCjU5IDAgb2JqCjw8IC9MZW5ndGggNjAg MCBSIC9MZW5ndGgxIDczMjAgL0ZpbHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLfFTF 1T8zc/eR9yaQ9yZ7lyWLZBMDAQyPmGweG9CAEAh2lwazAVISHhJMAFGEpYpgQKF8FgUfqH2ItcrN JtIN2JJKayuK8AmlFa0g2qr9FaF+n1pfud9/7m5QWn/97uR/5sw5Z15nzp07s+m6ZVUrJVKIBHkX Lm/pIOPJbkNmX7i6S42Wk7KIzI3f61i8PFoetozI9OHiZWu/Fy1nbyPiL7S1tiyKlulL5Ne0QRAt s/HIR7Yt77o1Ws5uQp64bMXCmD77LpSty1tujfVPb6Ks3tyyvDVqb5f11I4VnV2x8izkV3Xc0hqz Z36ihF/peKT+xMe1FAqCYYCgj6icHiELcbJRCd1IpLQpeWRCWepNfPvOj7Z90pxS/rE11woB0RPv jCqU+RGlY8fn+79abCNrIopxhr1UoJ6lYvAGqrHR5/s/v80W7Ulqhh7RT43iql53luPE82I0nQO4 GB325Dn6xSiRF57i8EaEqzctvTSlqlioaLHEoCroCmA/cBhQqFnkQ2sD3QCEgP3AYeAEYMYE8w2t inwFsBc4B5hFnrCHVYetapTIRt1szDdFZNJFQAcEOUBLgJlAM7Ad2AuYDTspWQFsAA4DlwAzeUVm eOc4jD0zvNXIepcsKzWKLdFi03yj2PudQDSf0RDNa6+Lmk2Omo0dHxVfXR3NRxVF87SC0hAa741P Kh2oyhAZmGQGBt4ByvhvKIUxctBjIp00gAsM1ZB4RVrvSHfp3sNCISa4YLSIHPqAYOGk1NKqeK7z i5RGDv4hvxDV8Au9yamle6uu5+dpP3AYEPw80tv8bdrAz0mfg1YCe4HDwHHgImDm55DOIr3F36IU /mcqASqBZmAvcBi4CFj4n0Ft/E0ZMQaVfCXA+ZugNv4GpvUGaAo/A+4MP6MP8JPhskml/QbjKYkx joIYk5kbY9IySiP8tfBnoxFRbqw0IuqQGEEVNE6MCBeMdUREVri83RHh7/SqHsdjVWP4KdIAjpGc Qs+nSAVmAUGgAzCDOw3uNIWAHcBjgAYgykBtgMqPAq8Ap2kM4AVmAVZ+IoxuIvx42F3tqMrgr/Lf USY8foz/3shf4S8a+cv8t0b+EvJ86I/yF8P5DqpKgJ5Qx4bchrwEehP/de/INIdelcoPw4MO0BKg EpgJNAPbATM/zEeEFznS0MghOop32MHD9IGR/5SesJJ3icPrrkEAqpK4J18LDmSvutfNve5du1GU xH3fTnCSuO/aBk4S920bwUniXrYanCTuRUvASeKe1wxOEvfMRnAgEf7oL0aOcpTNXMrUqhS+Bl5a Ay+tgZfWkMLXyESfKXKMD4ULC+GxPV7P6EJH6CALPc9Cs1noCRZqZaH1LLSRhcpZ6CYW8rCQnYXy WcjLQofYRLgixLx9VxQnebNY6CgLPcNCnSzkZqECFhrJQior80a4M3wd3jpkPiPrrZIvHXf2XluB 3SeFO+FRJ2LeiT3hMOhxQDdKXhipI6LG2fkyH9FbWBktXz25dEXVNH4EFY9gGY7QWUDBAh1BGB1B I0fQXApoJdAMDAAXAR0ww3oE5rHdoCmgJUAl0AxsAC4CZmM4FzEUTitA5RD3GwMrAa0EZsoSP4I0 AsnJnd48m93msU0T2+0sJZ/NzNfzeRllZGBfTku1pkZY0oFPk/75aRLFVcXx+/h2ysNC7Ijl28Of 5Tki7MGw+5CjKp09QPkKoo5NIjcrQD6ROo3yBLJbpXw82fnTyEvD9htRLSXsLnIcZMmy1gHHZ/Z3 HR/YIxzs+/ZDjj+qEYWFHX+A5OkDjlP2exwvlUSskDzvjjBkB1XDtN8+0fHMUcN0IxR7wo71Mjvg uMM+1bHUbihao4qbOlHypjhmu+c5pqG9WvsCh7cTbR5wVNpvcpRHrSbIOgccYzAET5QtxGBH241O XflGg3PLIqzNW2TZZfFbZlqusZRaiixOi8OSZ8m1DLemWW3WZGuiNd5qtZqtipVbyTo8op/zeuRX b7jZ+PiZEdCMFIO3YYdhcpsBJc6snK4nbZio5/Vzqlm9NrCQ6heo2idzXBEW3zBPM7mqmZZWT/WN 1dpET33Eos/Wyjz1mmXWd/09jN0XgFTjWyKMGv0RpkvRplwtrcbfT4ylbro3V+ZXbbo3EKCsjNWV WZVpFamT6mq/hQQNYbDW8/WT9TXryfLkabvq5/i1n+UFtFLJ6HmBeu2/5qhN/n72Ebvkq+1n/5BZ wN8vKthHvtlSLipqA4H6CLvRsCOV/QN2iBhksLPmkyrtSLXmR+32RO0KUB92I2UGu7g4KjDsCuLi DDuFSbuezpG+2p6RILDJVKnTsOnMVL9pc7QANgUgsMkI0VHD5mhGSNpoFUYzdjtM8kFgwnLIbpjY WY5hYoy8xzApiZncc9nkHqMnER2NYSMJmkk6N2STdA4233Dkf2Zbqz0e1jslsLDJ1+ryBV2+ViCo bV3dlqWFFqhqz8KAVKiacAcXLGyTeUurFnC11moLXbVqzxSj3r+om6R6iqu2h5p8jf6eJm9rbXiK d4rP1VIb6J06a3zZFX3dc7mv8bO+pa9ZsrHxsq+pRr1/6atMqqfKvspkX2Wyr6neqUZfZMT4LH+P laoDNVg/mffyhHjEazDXGajOsHVUGME7xZm1PvcgTiv7KMET0BJd1VoSIOO6uKq4SqrwTklVMsQp MVXW+inO3INsX0xlgzjVVU2erlWdqyjL114b/evEA1HXKrkUUeqRsm99YOLTvC218mxdrxXOqdcq G+b5eywWSIO1AcgmD8kSEnwRfSAqvBrCydJQiMuGUlYuZXFxMcN/jwVjTBDDO/04aBzqZd581kWd AaHl1zdybAWN8+CGpnn+gzhLyY9EZwAT7GQe1jnUmpyHwVNUQph25xC6VsW4mC+6Yrlh2ukhT+eQ S4aa80hnGcTwVZcHW5vpIGUDOaYnKVtxE+4/+nvA+zIfbNffl3qZ879ho4vEQLSPnmHt9AwdphfY JdTaT/3UR/IIVEsP0zq6nzbjszYPkntoNpIJ8vtZtt6Hm8nj+GA+Tsdg+x1aTwcpg2XpH9AG2iRO otYmSqIRVEWzaAXdy6brq6iJzip3UhlNp5upg4V0v36fvlP/Mf2E+sXv9a8ogXJoIdIx/UPTn/Q3 qRg1fki76SzbGfccedFLCJaP0C20R8xXmL5Y/xwjcNIajEGhGXSMDXAPWm+l91gWWydq0MqPdE3/ DazsNJ/aaA8dZBPYVO40Nekz9GOUgT5uRau7KUwHkCL0SzrDEk2X9B/rlyibiug6zKePXmUDYvCr jYOV8JsJXhpNk6BZQb+i39EJ5mK/5itMiaZSk9d0m36KhtNYmovRPomaf2Wf8vVIG8SLSp1eTcnw yw+kt+m39DbLYSVsJruRj+Yr+KPiFrKix7FIi6gd/n4Qrb+FMDrAE/lx8SPlaeULc97gOT0ZK+Km h+gR+jVLwkxV1sm+z06zd3gNb+YP8fPifuUp5TVLC2Z9Ey2ne+lp+pSlsYmsgX2XtbF1bDP7AdvN jrET7H1exRv5Un5RtImV4pdKNdIcpVO503S3aav5/UH/4G8G/3vwU71Uv5saEA8bMfof0qOYWT8d p9eRztJ5ZmIJLBlJZU42l92OtJ7dy55g+9hTrA+9nGDn2Qf4JH3MvuD40nIzz8XhRx6BXPwWnDDv 5w/z40gn+N/5ZyJTjBAeMUGUi4BYgVFtFjuQnhNvKznKcUWHn0tNu0x7TftMT5teMF0yJ1q+j2/8 K1/+6KvCr94apMEtg7sGw4N9+tuUjjXE1wNXsHKMvgVpCdZ7FyJuP51kifBdDitkFWw6PNPMlrCV 7FZ48i62h/3EGPuz7Hl46Y/sIsacxO3GmK/mE3g1n4l0E2/lK3EY28n7+Gn+ubCIBJEi0kWhmCrm i1bRJdaKXUITr4g/i/PiE/Elkq7EKw5lhOJWPMpUpVlZpTyqvKe8Z2oyvWz6iznevNx8tzli/gdO NRWWWZYGy3zLdssByylrENF5hJ6jXyACLz/snNgofOI5uo+PU7JxhXkV8dxMi8QMjkjl+9gWfgfr 4yNNt5qn8CnsBrqkuOHrF/le/gmfImawejaHlvCx0QbNw5WfgStXjtAF5XnM7VW0fKs5ka3nF82J FMYZaRLOSL8VYxSPeJnOiLPMojxObyjxLJNd4E+KWYiCXyoVJj85xcP0rFjJ7qDnuI8o/gvrNsTx Dexn2BcaWSn7p9BxDL4BUVQm3qE7aSn/E13Ae7yFHmCLlMV0H41j6+g9+ineitGmm82F5nT2Em9X uvkw1kdceQqzm8RGMmEaTnex+WKP+SJ/nVbRcSWe3hI/x+iP82fFDOWSaTZrwxtwB91NK/WNtNbk V15ji0mwG6lAOYfdbZ0oVZzIN2BXacKedgBv90HsA1ViBiRZiJzpiIu52CH2ID2IfUJBBLXjHf8O drFXqc/cyCO02JTMsOvgl5qXB2fTPP2ntFtfTDfrO6kY+8FmfR1a3Ed/oe20j20avJ06cJV8He/2 dFMdP26q04t5N3+dz+G7rlxfeLuAZdHfkJ7FylSYDlG38keaQ5X6Nv0PiO6rsMPupgU4sL6LWX6I HqaJARo3eAPv0etEB+Z7lhr0J3UHi6c2fRnNpOfpJxYTtVg8WGONvYb53k6tfLbeJVoH2+GH7fCC F95ahf3nHm/N3MYqb2XFteVTJk+aWDZh/LjSsWNKri4u8hSOvmqUu2Cka4RTdeTn2XNzsrMyM9KH D0tLtaUkJyUmxMdZLWaTIjijIp+rLqhq7qCmuF3TphXLsqsFgpZvCIKaClHdlTaaKuu1QHWFpReW 3/sXS2/U0nvZktnUciovLlJ9LlU7VutSI2xegx/8vbWugKpdMPgZBr/D4JPAO52ooPqy2mpVjQVV n1a3uq3bF6wtLmI9CfE1rprW+OIi6olPAJsATst0dfSwzApmMDzTN7mHkzUJU9RyXLU+LduFqmhG FPhaFmmzGvy+2lynM1BcpLGaha4FGsmTkscwoRqjG81co1mMbtR2nHE02qr2FA10b4vYaEHQk7jI tailya+JFrTh01I96LdWy7zt3ayvi2gcZ7LN39Tmim5fVrsqjbu7N6vaYw3+b9TNdcoWAgG0gbq8 oC7YXYeut2Gl6uVZXOObAn6NbUKXOFgWGLOKzi966i0ILlG1OFe1q617SRBLk9Ot0ey1znBOjrdf P0c5PrW70e9yapW5rkBLrb1nOHXPXtub7VWzr9QUF/XYUqOO7UlOiTGJSd9kWuH0qM7gDHPJ1c++ 7Fkmx+i6DidBTV2oYiR+F+Y0UZLWidS9cCIWAE+AoZa2CCvSrsXVBLttk6UcU2SaqcDmUrs/JkSA 68Lfr5S0xCTmAtvHJJUyTi6HmsZahnjN49EKC2WIWGqwphhjhVGeUFy0OsJdrg4b7s/y0kCz4NuW wOQSuN/plAu8NeKlBShooQZ/tKzSgtwweUtwtuZBqRkY0qTPlZrQkOZy9aALkdwn77OUrlndl/9S bBnDfG2TNZbxH9StUX39HFc9jsaqrzsYi9r6xitKUb10KPwGXYzThtX4RS6HTHI8Vxja6Al5yATH ZX+iphTgz2wE9aKIxYqoNCRMrdNswWlRGoh3OmPvzP9XKaJfkrWM7OtqsWlokz2xgUaHrU25onzF 8BK7RX0jthyOk313d/wVOoRadJTXxTJEPC76TrVGo7l4MwvwhyvHRIlAruaFy6BpxFtkiAO5seIV hrmxSgE8MjqLi+qwZ3Z317nUuu5gd0tEDy1wqTZXdz9/gb/Q3eHDbhcNnIh+cGuuVrctAI+1scl4 PThV97jYloYeL9syZ56/Hz9xqFsa/WHOeE2wOiCXhdc0+mNuMRZEhj7WEP8xQcQonTQXOIvQYUYA 4f8eUCSjPAZHt+gPJQgr+h/Wgl9LIDYh4YRtIXKmOlMLQNAKfamKgS+9JvqCVGUAn2uaq7+npJoG 8D+NPDa3h8tBeONz8hXT8PykpMy4iP5+X0oKnysZb3ZSErhUSpQSykhMBE2UMirBHegYyDGqvFB5 YeyY3B7zv7f0v2jJLFv6a19SksF86M1OSACXSjYpIVtioqRSdrnJr9vsM6vZNjsCKcxV/JfmHGUA aUAKfkZaoJg38y0JW1JeSjbFWRKyuG/Y9PTrs2tyG4c1pTdlz85dalmasHDYsvSl2cHctXyNeXXC bSmbzQ9adtleyjrDT5tPJ7yRknN54p1xXqdr/Jg4RnG2OB63w5HaSQgebzKkKj7FnHbk/25rlsf2 yXzPBZCVnguxqbP5K3GpmSgfBgQCw2xp14wrzchIS7dxs2vEKPcwW8a40mtSbW7XCIt57tKTj60O d1UvOfn4qbU/6H9q3bqnnlq/7vr5/CRT2LU/b+4d1M8MDg4eeebBX7BHBh+4eAk3hSUftt8t4wDR wL/A2sXTfq8qvEmp45cqG/h2vtuq/FxhcWQ2cRFnYomcHY03Rh8v50RMRV389NZns2HpIvrfvKnG gtqNBU02FhRe9mbL5RpaE2N9chJN3qSU8aYhT4wxMRW3K27KTjjIytkmyvLcYHsXzoBfYrdiFMpn fFVOlZWZk1jqpLFj2Hya73G6Us1my4Rrrikbx7/oqzrZ+MD5ki7l9op1jmenHm1G9MLBxqOPwr3g 2x4B4TCc96Qf0gD5GO9Dw43Ta+Y0eKpuaW9ZNqPx/wBnh42+CmVuZHN0cmVhbQplbmRvYmoKNjAg MCBvYmoKNTAwMwplbmRvYmoKMTYgMCBvYmoKPDwgL1R5cGUgL0ZvbnQgL1N1YnR5cGUgL1RydWVU eXBlIC9CYXNlRm9udCAvWFdKQ0pYK0FyaWFsTVQgL0ZvbnREZXNjcmlwdG9yCjYxIDAgUiAvRW5j b2RpbmcgL01hY1JvbWFuRW5jb2RpbmcgL0ZpcnN0Q2hhciA0OCAvTGFzdENoYXIgNTAgL1dpZHRo cyBbIDU1NgowIDU1NiBdID4+CmVuZG9iago2MSAwIG9iago8PCAvVHlwZSAvRm9udERlc2NyaXB0 b3IgL0ZvbnROYW1lIC9YV0pDSlgrQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRCQm94IFstNjY1IC0z MjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVzY2VudCAtMjEyIC9D YXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1MTkgL1N0ZW1IIDg0 IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiA2MiAwIFIgPj4KZW5kb2Jq CjYyIDAgb2JqCjw8IC9MZW5ndGggNjMgMCBSIC9MZW5ndGgxIDc2NTYgL0ZpbHRlciAvRmxhdGVE ZWNvZGUgPj4Kc3RyZWFtCngBhVkJfFTV1T/3vjdLkkkyScg6k5k3DBkkkxgIS4DEZLJMQCOyJTpD gySESNgkEBZBhKEtAgGFUkuFqri1ohZ5mQScAC2pWK0oQoXSilZwq9pfWernUre8/u+bAaX11+/d +Z977jnnbueed999b5YsXtpKFgqRRL6WBc3tpF/ZbcgcLcuWKNFyYhaRsf629tkLouW0+USGC7Pn r7gtWs7eSiRb21qbZ0XL9DXyUW0QRMtsBPJBbQuW3BEtZzcit8xf2BLTZ29A2byg+Y5Y//Qmysrt zQtao/a5fxDl9oUdS2Llh5Ff0764NWbPAkQJv9VwCf2JT2so1ASGARJ9TGX0IJmIk5WK6GaMtE3O JQPKQm/gW7Z9vPmzGclln5ptZgiIHn13cL7Ij8jtW7/c+81sK5ktKMbp9kKBeqby/puo2kpf7v1y pTXak9BcvqReqpeu6fZkOU8ckobQOYBLQ8LeXGevNFjKDZc6fRHJ3Z2aXpxcWSgpaLFIpwroQmAv cBiQaYbkgNYKugYIAXuBw8AJwIgJOnStgnwhsAs4BxilXMkeVpzWysFSNupmY77JUiZdBDRAIido ETARmAFsAXYBRt1OSBYCa4DDwCXASD4pM7xtOMaeGd6kZ91z5xfrxeZosXG6Xuy+JRjNJ0yO5jXX R83GRs2GjYiKr62K5oMLonlqXnEIjXfHJxb3VWZIGZhkBgbeDsr485TMGDnpYSmdVIBLGKou8Ump 3YM8xbsOSzIxiUuMZpFT65NYODGluDKea/wipZKTX+Dnoxp+vjsppXhX5Q38HdoLHAYk/g7S2/xt WsPPCZ+DVgC7gMPAceAiYOTnkM4ivcXfomT+VyoCKoAZwC7gMHARMPG/glr5myJidCr4CoDzN0Gt /A1M6w3QZH4G3Bl+RuvjJ8MlY4p7dcZbFGOceTEm0xZjUjOKI/y18BdDEFEerDQi6qA0kMppuDQw nDfMGZGywmVznBH+brfidT5cOZSfIhXgGMkp9HyKFGAS0AS0A0Zwp8GdphCwFXgYUAFEGagVUPhR 4BXgNA0FfMAkwMxPhNFNhB8Pe6qclRn8Vf4iZcLjx/gf9PwV/oKev8x/r+cvIXdAf5S/EHY4qTIB ekIdK3Ir8iLoDfx33YNSnVplCj8MDzpBi4AKYCIwA9gCGPlhPjA8y5mKRg7SUdzDTh6mj/T8V/So mXxznT5PNQJQEcQz9jpwILuUXR7u82zfgaIgnnu3gRPE8+PN4ATxrFwLThDP/GXgBPHMmgtOEM+0 GeAE8UysBwcS4Q89O2iws2TiPKZUJvPl8NJyeGk5vLScZL5cJPpCFmP8RTg/Hx7b6fMOyXeGDrDQ IRaawkKPslArC61mobUsVMZCt7KQl4XsLORgIR8LHWSj4YoQ8/VcVRzjy2Khoyy0h4U6WMjDQnks NIiFFFbii3BX+Hrcdcj8etZdKW467uq+rhy7TzJ3waMuxLwLe8Jh0OOAppd8MFIGRo2zHSIf2J1f ES1fO7Z4YeV4fgQVj2AZjtBZQMYCHUEYHUEjR9BcMmgFMAPoAy4CGmCE9UDMY4tOk0GLgApgBrAG uAgY9eFcxFA4LQQVQ9yrD6wItAKYKEr8CNJAJBd3+XKtdqvXOl7aYmfJDjbRoTl4CWVkYF9OTTGn RFji/s8T//V5IsVVxvF7+RbKxUJsjeVbwl/kOiPs/rDnoLMynf2cHDKijo0hD8tDPpo69PJIspuF fATZ+dPIi8P2m1EtOewpcB5gSaLWfucX9vecH9kjHOyH9oPOPysRmYWdf4Lk6f3OU/aNzpeKImZI DnkiDNkBRTfttY927jmqm66FYmfYuVpk+5132cc559l1RWtUcWsHSr5k5xTPNOd4tFdjn+n0daDN /c4K+63OsqjVSFFnv3MohuCNsvkY7BC73qnboTfYUBJhbb4C03ZTwDTRNMpUbCowuUxOU67JZhpg TjVbzUlmiznebDYbzbKZm8k8IKKd83nFU2+AUX/4GRHQDCcAQa3YYZjYZkCJMzOnG0hNk+p43dQq Vqf2tVDdTEX9bKo7wuInT1MN7iqmptZRXX2VOtpbFzFpU9QSb51qmvSDQBdj9wYhVfmGCKP6QIRp QrTOpqZWB3qJsZR199hEfs26e4JByspYVpFVkVqeMqa25ntIky5sqvF+e2V9y3qzvLnq9rqpAfWp 3KBaLBgtN1in/nSq0hjoZR+zS/6aXvZPkQUDvVI5+9g/Rcil8ppgsC7CbtbtSGH/hB0iBhnszA5S hB0pZkfUbmfULg/1YTdIZLCLi6M83S4vLk63k5mw6+oY5K/pGgQCm0yFOnSbjkzluzZH82CTBwKb jBAd1W2OZoSEjVquN2O3w8QBAhOWQ3bdxM5ydBN95F26SVHMZOMVk416T1J0NLqNIGgm8dxlm8Rz sPmOI/8321rl9bLu0mBLo7/V7W9y+1uBJnXTsrYsNTRTUbpagkKhqJKnaWZLm8ibW9Wgu7VGbXHX KF2ler3/UDcKdam7posa/fWBrkZfa0241FfqdzfXBLvHTRpRclVfG6/0NWLS9/Q1STQ2QvQ1Tq/3 H32VCPU40VeJ6KtE9DXON07vi/QYnxToMlNVsBrrJ/JunhCPeG2yuYJVGdb2cj14S11Zq20HcFrZ TQneoGpxV6mJgIjrwsrCSqHCPSVUSRAnx1RZq0tdtgNsd0xlhTjFXUXeJUs7llKWf05N9NeBC6Il S8VSRKlXyL73golf9TXXiLN1nZo/tU6tmDwt0GUyQdpUE4Rs7GVZQoI/ovVFhddCOFYYStIVQyEr E7K4uJjhf8eCPiaI4Z1eHDQOdjOfgy2hjqCkOurqObaC+mlwQ+O0wAGcpcRDoiOICXYwL+u43JqY h85TVEKYdsdlLFka42K+WBLLddMOL3k7LrvkcnNe4Syd6L5a4sXWZjhA2UCO4QnKlj2E9x/tA+BD kffP0T4UepHzv2Oji8RAtJv2sDm0hw7Tc+wSau2lXuohcQSqoQdoFd1H6/FYmwbJRpqCZID8Ppat 9eDN5BE8MB+hY7C9hVbTAcpgWdpHtIbWSSdRax0l0kCqpEm0kO5hN2pLqZHOyj+iErqRbqd2FtIC 2r3aNu1x+iX1Sn/QvqEEyqEWpGPaBcNftDepEDV+RjvoLNsWt4986CUEywdpMe2UpstMm619iRG4 aDnGINMEOsb6uBett9IHLIutkqrRymOaqj0PKztNpzbaSQfYSDaOuwyN2gTtGGWgjzvQ6g4K036k CP2GzjCL4ZL2uHaJsqmArsd8euhV1if1f7O2vwJ+M8BLQ2gMNAvpt/QinWBu9ju+0GAxFBt8hpXa KRpAw6gBo30CNf/GPuerkdZIL8i1WhUlwS8/Ed6m39PbLIcVsYnsZj6EL+QPSYvJjB6HIc2iOfD3 /Wj9LYTRfm7hx6XH5Kflr4y5/ee0JKyIh35BD9LvWCJmqrAO9kN2mr3Lq/kM/gv+jnSf/KT8mqkZ s76VFtA99DR9zlLZaDaZ/YC1sVVsPfsJ28GOsRPsQ17J6/k8flFqkxZJv5GrkKbKHfKPDHcbNhk/ 7A/0P9//x/7PtWLtbpqMeFiL0f+MHsLMeuk4vY50lt5hBpbAkpAU5mIN7E6k1ewe9ijbzZ5kPejl BHuHfYRH0qfsK44nLTdyGw4/4gjk5otxwryPP8CPI53g/+BfSJnSQMkrjZTKpKC0EKNaL21F2ie9 LefIx2UNfi42bDfsMuw2PG14znDJaDH9EM/4V75+7Jv8b97qp/4N/dv7w/092tuUjjXE0wOvYGUY fTPSXKz3dkTcXjrJLPBdDstn5exGeGYGm8sWsTvgyR+zneyX+tifYYfgpT+zixhzIrfrY76Wj+RV fCLSrbyVL8JhbBvv4af5l5JJSpCSpXQpXxonTZdapSXSCmm7pEqvSH+V3pE+k75G0uR42SkPlD2y Vx4nz5CXyg/JH8gfGBoNLxveN8YbFxjvNkaM/8Spptw0yTTZNN20xbTfdMrchOg8QvvoWUTglYud k9ZKfmkf3cuHy9l4hXkV8TyDZkkTOCKV72Yb+F2shw8y3GEs5aXsJroke+DrF/gu/hkvlSawOjaV 5vJh0QaNA+SnwJXJR+i8fAhzexUt32G0sNX8otFCYZyRxuCM9HtpqOyVXqYz0llmkh+hN+R4lsnO 8yekSYiC38jlhgC5pAfoGWkRu4v2cT9R/FfmzYjjm9hT2BfqWTH7l6ThGHwToqhEepd+RPP4X+g8 7uMN9HM2S55N99Jwtoo+oF/hrhhiuN2Yb0xnL/E5cidPYz3E5ScxuzFsEJMMA+jHbLq003iRv05L 6bgcT29Jv8boj/NnpAnyJcMU1oY74C66mxZpa2mFISC/xmaTxG6mPPkcdrdVUrHsQr4Gu0oj9rT9 uLsPYB+olCZAkoXIuRFx0YAdYifS/dgnZETQHNzjt2AXe5V6jPU8QrMNSQy7Dr7UvNw/haZpv6Id 2my6XdtGhdgP1mur0OJuep+20G62rv9Oaser5Ou4t2801PLjhlqtkHfy1/lUvv3q9YW381gW/R3p GaxMueEgdcp/pqlUoW3W/oTovgY77A6aiQPre5jlBfQwXuqj4f038S6tVmrHfM/SZO0JzcniqU2b TxPpEP3SZKBmkxdrrLLXMN87qZVP0ZZIrf1z4Ict8IIP3lqK/Wejr7qhvtJXUX5dWenYMaNLRo4Y XjxsaNG1hQXe/CHXDPbkDXIPdClOR67dlpOdlZmRPiAtNcWanJRoSYiPM5uMBlnijAr87tomRfU0 qbLHPX58oSi7myFo/o6gSVUgqr3aRlVEvWaorrL0wfK2/7D0RS19VyyZVSmjssICxe9W1GM1biXC pk0OgL+nxh1U1PM6P0Hnt+p8IniXCxUUf1ZbjaKyJsWv1i5r6/Q31RQWsK6E+Gp3dWt8YQF1xSeA TQCnZrrbu1hmOdMZnukf28XJnIgpqjnuGr+a7UZVNCPl+ZtnqZMmB/w1NpcrWFigsuoW90yVxEnJ q5tQtd6NaqxWTXo3yhyccVTapHQV9HVujlhpZpPXMss9q7kxoErNaMOvpnjRb42aufK9rG+LaBxn svXf1dqkTn/WHEUYd3auV9SHJwe+U9fmEi0Eg2gDdXlebVNnLbrejJWqE2dxla8LBlS2Dl3iYJmn zyo6v+ipN69prqLGuavcbZ1zm7A0OZ0qTVnhCufk+Hq1c5TjVzrrA26XWmFzB5tr7F0DqHPKiu5s n5J9taawoMuaEnVsV1JyjLEkfpdphdOjOp3TzQVXN+WKZ5kYo/t6nARVpUXBSAJuzGm0IK2jqbNl NBYAV5ChljoLKzJHjatu6rSOFXJMkamGPKtb6fyUEAHu8/+4WtIckxjzrJ+SUIo4uRJqKmu+zKte r5qfL0LEVI01xRjL9fLIwoJlEe52t1vx/ixeGmgSfNscHFsE97tcYoE3RXw0EwU1NDkQLSs00xYm XxHO1rxJaPoua9IbhCZ0WXOlepMbkdwj3mcpXTV7rvySrRlp/raxKsv4H+rWqL5uqrsOR2PF39kU i9q6+qtKUb1wKPwGXYxT06oDko1DJjhuk3Rt9IR82QTH5YBFlfPwM+pBPStiMiMqdQlTalVr0/go Dca7XLF75v+rFNEuiVp69m212DTUsd7YQKPDVkuvKl81PEunVFePLYfjZN/ZGX+VDqEWHeX1sQwR jxd9l1KtUgPuzDz88MoxWiBoU31wGTT1uIt0cdAWK15laItVCuIS0VlYUIs9s7Oz1q3UdjZ1Nke0 0Ey3YnV39vLn+HOd7X7sdtHAiWgHNtnU2s1BeKyNjcXtwamqy802TO7ysQ1TpwV68YlD2VAfCHPG q5uqgmJZeHV9IOYWfUFE6GMN8Y8JIkbuoAagDKHD9ADC/x5QoBkainO0CCnxqURc6fR/+LiCLV4c ig3inwsTkSvFlZIHgpboa0Xq+9pnoK9IkftEvQbtAznF0IfWcllDFxcD8cXnOGTDAEdiYmZcRPuw JzmZNwjGl52YCC6FLEJCGRYLqEXIqAjvQcdAjlHF+Yrzw4bauoz/3dInaMkoWvpbT2KizlzwZSck gEshq5CQ1WIRVMiuNPltmz1GJdtqRzCFuYJ/as5RBpAKJONT0kzZuJ5vSNiQ/FKSIc6UkMX9aTem 35BdbatPa0xvzJ5im2eal9CSNj99XnaTbQVfblyWsDJ5vfF+03brS1ln+Gnj6YQ3knOuTLwjzudy jxgaxyjOGsfjtjpTOggB5EuCVMHjmNNWx4ubsrzWz6Z7z4Ms8p6PTZ1NX4QXm9HiYkAwmGZNHTW8 OCMjNd3Kje6Bgz1p1ozhxaNSrB73QJOxYd7Jh5eFl1TNPfnIqRU/6X1y1aonn1y96obp/CST2XW/ ntHdr53p7+8/suf+Z9mD/T+/eAlvC3MvzLlbxAIiQjZh7RzsxZi/41KsiVlpacaGxIj2SU9Kis5c 8MVZreAcAwwOsYyZwsDhEFqHPQkaBxYRNMIP+iw8PjMT//ikcK44U1LHFJ06JugxKjovXnUrBH0e Rw5bLFREh5bUVK536ItLTgEX7eecLyE1jTc4BgiZaDuMpkU4JSTwBjD/8CWJ0Pm+3kQcif5Eb3pn vlGlhlLjQcNh40HTi+aX7KbrLUFLfdI8y6yklakr0zamHkp9P+d926Ucy+GEZ9O4DR9wc60Oq/G3 eGU0IUDMyOO0S74cR7zVbDQetecMsNtzzPYciXFzjl1KdFgj/PHuiSkMn3ez9okZkO6OZMYt8R2Z J+FtEQ/sIF9LClnZaJ8lZV8FXu0W8jVc5gf4IHzE3dIVDYhPEBDeMusniIiyb8oqzn8z/b2U1Mwx TJD1Sdd6k+6yPo/CmGFD9TARUTKaprPpi4PBvHSXp2TUKPxGjkB4GE2DR4nYSR9gNJrwk01fl/DM vMd2Xty9484fPsB60/71x5OfjX/iuUcbHXv2VJa19K1+/v3b5v30gc6046//fU/gqUOPb2gehmDV L20w3rO+75IgTMPeIWIqNbbDGPFeTIFbbqi+IeCtXDynef6E+n8DLebxtQplbmRzdHJlYW0KZW5k b2JqCjYzIDAgb2JqCjUyNzEKZW5kb2JqCjEyIDAgb2JqCjw8IC9UeXBlIC9Gb250IC9TdWJ0eXBl IC9UcnVlVHlwZSAvQmFzZUZvbnQgL1BZU0ZMWitBcmlhbE1UIC9Gb250RGVzY3JpcHRvcgo2NCAw IFIgL0VuY29kaW5nIC9NYWNSb21hbkVuY29kaW5nIC9GaXJzdENoYXIgNDggL0xhc3RDaGFyIDUw IC9XaWR0aHMgWyA1NTYKMCA1NTYgXSA+PgplbmRvYmoKNjQgMCBvYmoKPDwgL1R5cGUgL0ZvbnRE ZXNjcmlwdG9yIC9Gb250TmFtZSAvUFlTRkxaK0FyaWFsTVQgL0ZsYWdzIDMyIC9Gb250QkJveCBb LTY2NSAtMzI1IDIwMDAgMTAwNl0KL0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5MDUgL0Rlc2NlbnQg LTIxMiAvQ2FwSGVpZ2h0IDcxNiAvU3RlbVYgOTUgL0xlYWRpbmcKMzMgL1hIZWlnaHQgNTE5IC9T dGVtSCA4NCAvQXZnV2lkdGggNDQxIC9NYXhXaWR0aCAyMDAwIC9Gb250RmlsZTIgNjUgMCBSID4+ CmVuZG9iago2NSAwIG9iago8PCAvTGVuZ3RoIDY2IDAgUiAvTGVuZ3RoMSA3NjU2IC9GaWx0ZXIg L0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4AYVZCXxU1dU/9743S5JJMknIOpOZNwwZJJMYCEuAxGSy TEADsiV2hgZJCJGwSSAsgghDWwQCCqWWClVxa0Ut8jIJOAFaUrFaUYQKpRWt4Pap/ZWlfi51y/v+ 982A0vrr9+78zz33nHO3c8+77743S5csayULhUgiX8vC5nbSr+w2ZI6W5UuVaDkxi8hYf1v7nIXR ctoCIsPFOQtW3hYtZ28jkq1trc2zo2X6GvmoNgiiZTYC+aC2hUvviJazG5FbFixqiemzN6JsXth8 R6x/ehNl5fbmha1R+9w/inL7oo6lsfLDyK9rX9Ias2cBooTfabiE/uSnNRRqAsMAiT6mMnqQTMTJ SkV0C0baJueSAWWhN/Ct2z/e8tnM5LJPzTYzBESPvjs4X+RH5fZtX+77Zo6VzBYU43R7oUA9U3n/ zVRtpS/3fbnKGu1JaK5cUi/VS9d1e7KcJw9LQ+g8wKUhYW+us1caLOWGS52+iOTuTk0vTq4slBS0 WKRTBXQRsA84Asg0U3JAawVdC4SAfcAR4CRgxAQdulZBvgjYDZwHjFKuZA8rTmvlYCkbdbMx32Qp ky4BGiCRE7QImATMBLYCuwGjbicki4C1wBHgMmAkn5QZ3j4cY88Mb9az7nkLivVic7TYOEMvdv8g GM0nTonmNTdGzcZGzYaNiIqvr4rmgwuieWpecQiNd8cnFvdVZkgZmGQGBt4OyvjzlMwYOelhKZ1U gEsYqi7xSandgzzFu49IMjGJS4xmk1Prk1g4MaW4Mp5r/BKlkpNf5BeiGn6hOymleHflTfwd2gcc AST+DtLb/G1ay88Ln4NWALuBI8AJ4BJg5OeRziG9xd+iZP43KgIqgJnAbuAIcAkw8b+BWvmbImJ0 KvgKgPM3Qa38DUzrDdBkfhbcWX5W6+OnwiVjint1xlsUY5x5MSbTFmNSM4oj/LXwF0MQUR6sNCLq kDSQymm4NDCcN8wZkbLCZXOdEf5ut+J1Plw5lJ8mFeAYyWn0fJoUYDLQBLQDRnBnwJ2hELANeBhQ AUQZqBVQ+DHgFeAMDQV8wGTAzE+G0U2Enwh7qpyVGfxV/iJlwuPH+R/1/BX+gp6/zP+g5y8hd0B/ jL8QdjipMgF6Qh0rcivyIugN/Pfdg1KdWmUKPwIPOkGLgApgEjAT2AoY+RE+MDzbmYpGDtEx3MNO HqaP9PzX9KiZfPOcPk81AlARxDP2BnAgu5XdHu7z7NiJoiCee7eDE8Tzky3gBPGsWgdOEM+C5eAE 8cyeB04Qz/SZ4ATxTKoHBxLhDz07aLCzZNJ8plQm8xXw0gp4aQW8tIJkvkIk+kIWY/xlOD8fHtvl 8w7Jd4YOstBhFprKQo+yUCsLrWGhdSxUxkK3spCXhews5GAhHwsdYqPhihDz9VxTHOPLYqFjLLSX hTpYyMNCeSw0iIUUVuKLcFf4Rtx1yPx61l0pbjru6r6hHLtPMnfBoy7EvAt7whHQE4Cml3wwUgZG jbMdIh/YnV8RLV8/tnhR5Xh+FBWPYhmO0jlAxgIdRRgdRSNH0VwyaAUwE+gDLgEaYIT1QMxjq06T QYuACmAmsBa4BBj14VzCUDgtAhVD3KcPrAi0ApgkSvwo0kAkF3f5cq12q9c6XtpqZ8kONsmhOXgJ ZWRgX05NMadEWOKBzxP/9XkixVXG8Xv5VsrFQmyL5VvDX+Q6I+z+sOeQszKd/YIcMqKOjSEPy0M+ mjr08kiym4V8BNn508iLw/ZbUC057ClwHmRJotYB5xf295wf2SMc7If2Q86/KBGZhZ1/huTpA87T 9k3Ol4oiZkgOeyIM2UFFN+21j3buPaabroNiV9i5RmQHnHfZxznn23VFa1RxawdKvmTnVM9053i0 V2Of5fR1oM0Dzgr7rc6yqNVIUeeAcyiG4I2y+RjsELveqduhN9hQEmFtvgLTDlPANMk0ylRsKjC5 TE5TrslmGmBONVvNSWaLOd5sNhvNspmbyTwgop33ecVTb4BRf/gZEdAMJwBBrdhhmNhmQIkzM6eb SE2T6njdtCpWp/a1UN0sRf1smjvC4qdMVw3uKqam1lFdfZU62lsXMWlT1RJvnWqa/MNAF2P3BiFV +cYIo/pAhGlCtN6mplYHeomxlPX32ER+3fp7gkHKylhekVWRWp4yprbme0iTLmyq8X57ZX3LerO8 ueqOumkB9ancoFosGC03WKf+bJrSGOhlH7PL/ppe9k+RBQO9Ujn72D9VyKXymmCwLsJu0e1IYf+E HSIGGezMDlKEHSlmR9RuV9QuD/VhN0hksIuLozzdLi8uTreTmbDr6hjkr+kaBAKbTIU6dJuOTOW7 NsfyYJMHApuMEB3TbY5lhISNWq43Y7fDxAECE5ZDdt3EznJ0E33kXbpJUcxk01WTTXpPUnQ0uo0g aCbx/BWbxPOw+Y4j/zvbWuX1su7SYEujv9Xtb3L7W4EmdfPytiw1NEtRulqCQqGokqdpVkubyJtb 1aC7tUZtcdcoXaV6vX9TNwp1qbumixr99YGuRl9rTbjUV+p3N9cEu8dNHlFyTV+brvY1YvL39DVZ NDZC9DVOr/dvfZUI9TjRV4noq0T0Nc43Tu+L9BifHOgyU1WwGusn8m6eEI94bbK5glUZ1vZyPXhL XVlrbAdxWtlDCd6ganFXqYmAiOvCysJKocI9JVRJECfHVFlrSl22g2xPTGWFOMVdRd6lyzqWUZZ/ bk3014ELoqXLxFJEqVfIvveCiV/1NdeIs3Wdmj+tTq2YMj3QZTJB2lQThGzsFVlCgj+i9UWF10M4 VhhK0lVDISsTsri4mOF/xoI+JojhnV4cNA51M5+DLaWOoKQ66uo5toL66XBD4/TAQZylxEOiI4gJ djAv67jSmpiHzlNUQph2xxUsXRbjYr5YGst10w4veTuuuORKc17hLJ3ovlrqxdZmOEjZQI7hCcqW PYT3H+0D4EOR98/VPhR6kfO/Y6OLxEC0h/ayubSXjtBz7DJq7aNe6iFxBKqhB2g13Ucb8FibDskm mopkgPw+lq314M3kETwwH6HjsP0BraGDlMGytI9oLa2XTqHWekqkgVRJk2kR3cMmaMuokc7JP6YS mkC3UzsLaQHtXm279jj9inqlP2rfUALlUAvSce2i4a/am1SIGj+nnXSObY/bTz70EoLlg7SEdkkz ZKbN0b7ECFy0AmOQaSIdZ33ci9Zb6QOWxVZL1WjlMU3VnoeVnWZQG+2ig2wkG8ddhkZtonacMtDH HWh1J4XpAFKEfktnmcVwWXtcu0zZVEA3Yj499Crrk/q/WddfAb8Z4KUhNAaaRfQ7epFOMjf7PV9k sBiKDT7DKu00DaBh1IDRPoGa/8M+52uQ1kovyLVaFSXBLz8V3qY/0NsshxWxSewWPoQv4g9JS8iM Hochzaa58Pf9aP0thNEBbuEnpMfkp+WvjLn957UkrIiHfkkP0u9ZImaqsA72I3aGvcur+Uz+S/6O dJ/8pPyaqRmzvpUW0j30NH3OUtloNoX9kLWx1WwD+ynbyY6zk+xDXsnr+Xx+SWqTFku/lauQpskd 8o8Ndxs2Gz/sD/Q/3/+n/s+1Yu1umoJ4WIfR/5wewsx66QS9jnSO3mEGlsCSkBTmYg3sTqQ17B72 KNvDnmQ96OUke4d9hEfSp+wrjictN3IbDj/iCOTmS3DCvI8/wE8gneT/4F9ImdJAySuNlMqkoLQI o9ogbUPaL70t58gnZA1+LjbsMOw27DE8bXjOcNloMf0Iz/hXvn7sm/xv3uqn/o39O/rD/T3a25SO NcTTA69gZRh9M9I8rPcORNw+OsUs8F0Oy2flbAI8M5PNY4vZHfDkT9gu9it97M+ww/DSX9gljDmR 2/UxX89H8io+CelW3soX4zC2nffwM/xLySQlSMlSupQvjZNmSK3SUmmltENSpVekv0nvSJ9JXyNp crzslAfKHtkrj5Nnysvkh+QP5A8MjYaXDe8b440LjXcbI8Z/4lRTbppsmmKaYdpqOmA6bW5CdB6l /fQsIvDqxc5L6yS/tJ/u5cPlbLzCvIp4nkmzpYkckcr3sI38LtbDBxnuMJbyUnYzXZY98PULfDf/ jJdKE1kdm0bz+LBog8YB8lPgyuSjdEE+jLm9ipbvMFrYGn7JaKEwzkhjcEb6gzRU9kov01npHDPJ j9AbcjzLZBf4E9JkRMFv5XJDgFzSA/SMtJjdRfu5nyj+K/MWxPHN7CnsC/WsmP1L0nAMvhlRVCK9 Sz+m+fyvdAH38Ub6BZstz6F7aThbTR/Qr3FXDDHcbsw3prOX+Fy5k6exHuLyk5jdGDaISYYB9BM2 Q9plvMRfp2V0Qo6nt6TfYPQn+DPSRPmyYSprwx1wF91Ni7V1tNIQkF9jc0hit1CefB6722qpWHYh X4tdpRF72gHc3QexD1RKEyHJQuRMQFw0YIfYhXQ/9gkZETQX9/gPsIu9Sj3Geh6hOYYkhl0HX2pe 7p9K07Vf005tDt2ubadC7AcbtNVocQ+9T1tpD1vffye141XyddzbEwy1/IShVivknfx1Po3vuHZ9 4e08lkV/R3oGK1NuOESd8l9oGlVoW7Q/I7qvww67k2bhwPoeZnkRPYyX+mh4/828S6uV2jHfczRF e0Jzsnhq0xbQJDpMvzIZqNnkxRqr7DXM905q5VO1pVJr/1z4YSu84IO3lmH/2eSrbqiv9FWU31BW OnbM6JKRI4YXDxtadH1hgTd/yHWDPXmD3ANditORa7flZGdlZqQPSEtNsSYnJVoS4uPMJqNBljij Ar+7tklRPU2q7HGPH18oyu5mCJq/I2hSFYhqr7VRFVGvGaprLH2wvO3fLH1RS99VS2ZVyqissEDx uxX1eI1bibDpUwLg76lxBxX1gs5P1PltOp8I3uVCBcWf1VajqKxJ8au1y9s6/U01hQWsKyG+2l3d Gl9YQF3xCWATwKmZ7vYullnOdIZn+sd2cTInYopqjrvGr2a7URXNSHn+5tnq5CkBf43N5QoWFqis usU9SyVxUvLqJlStd6Maq1WT3o0yF2cclTYrXQV9nVsiVprV5LXMds9ubgyoUjPa8KspXvRbo2au ei/r2yIax5lsw3e1NqnTnzVXEcadnRsU9eEpge/UtblEC8Eg2kBdnlfb1FmLrrdgperEWVzl64MB la1HlzhY5umzis4veurNa5qnqHHuKndb57wmLE1Op0pTV7rCOTm+Xu085fiVzvqA26VW2NzB5hp7 1wDqnLqyO9unZF+rKSzosqZEHduVlBxjLInfZVrh9KhO53RzwdVNvepZJsbovhEnQVVpUTCSgBtz Gi1I62jqbBmNBcAVZKilzsaKzFXjqps6rWOFHFNkqiHP6lY6PyVEgPvCP66VNMckxjzrpySUIk6u hprKmq/wqter5ueLEDFVY00xxnK9PLKwYHmEu93tVrw/i5cGmgzfNgfHFsH9LpdY4M0RH81CQQ1N CUTLCs2yhclXhLM1bxKaviua9AahCV3RXK3e5EYk94j3WUpXzZ6rv2RrRpq/bazKMv6LujWqr5vm rsPRWPF3NsWitq7+mlJULxwKv0EX49S06oBk45AJjtskXRs9IV8xwXE5YFHlPPyMelDPjpjMiEpd wpRa1do0PkqD8S5X7J75/ypFtMuilp59Wy02DXWsNzbQ6LDV0mvK1wzP0inV1WPL4TjZd3bGX6ND qEVHeWMsQ8TjRd+lVKvUgDszDz+8cowWCNpUH1wGTT3uIl0ctMWK1xjaYpWCuER0FhbUYs/s7Kx1 K7WdTZ3NES00y61Y3Z29/Dn+XGe7H7tdNHAi2sHNNrV2SxAea2NjcXtwqupys41Tunxs47TpgV58 4lA21gfCnPHqpqqgWBZeXR+IuUVfEBH6WEP8Y4KIkTuoAShD6DA9gPC/BxRohobiHC1CSnwqEVc6 /S8+rmCLF4dig/jnwkTkSnGl5IGgJfpakfq+9hnoK1LkPlGvQftATjH0obVc1tDFxUB88TkO2TDA kZiYGRfRPuxJTuYNgvFlJyaCSyGLkFCGxQJqETIqwnvQcZDjVHGh4sKwobYu43+29AlaMoqW/qcn MVFnLvqyExLApZBVSMhqsQgqZFeb/LbNHqOSbbUjmMJcwT815ykDSAWS8SlplmzcwDcmbEx+KckQ Z0rI4v60Cek3ZVfb6tMa0xuzp9rmm+YntKQtSJ+f3WRbyVcYlyesSt5gvN+0w/pS1ll+xngm4Y3k nKsT74jzudwjhsYxirPG8bhtzpQOQgD5kiBV8DjmtM3x4uYsr/WzGd4LIIu9F2JTZzMW48VmtLgY EAymWVNHDS/OyEhNt3Kje+BgT5o1Y3jxqBSrxz3QZGyYf+rh5eGlVfNOPXJ65U97n1y9+skn16y+ aQY/xWR2w29mdvdrZ/v7+4/uvf9Z9mD/Ly5dxtvCvItz7xaxgIiQTVg7B3sx5u+4FGtiVlqasSEx on3Sk5KiMxd9cVYrOMcAg0MsY6YwcDiE1mFPgsaBRQSN8EM+C4/PzMQ/PimcK86U1DFFp48LepyK LohX3QpBn8eRwxYLFdGhJTWV6x364pJTwEX7Oe9LSE3jDY4BQibaDqNpEU4JCbwBzD98SSJ0vq83 EUeiP9Gb3plvVKmh1HjIcMR4yPSi+SW76UZL0FKfNN8yO2lV6qq0TamHU9/Ped92OcdyJOHZNG7D B9xcq8Nq/B1eGU0IEDPyOO2yL8cRbzUbjcfsOQPs9hyzPUdi3JxjlxId1gh/vHtSCsPn3az9Ygak uyOZcUt8R+YpeFvEAzvE15FCVjbaZ0nZX4FXu0V8LZf5QT4IH3G3dkUD4hMEhLfM+gkiouybsooL 38x4LyU1cwwTZEPS9d6ku6zPozBm2FA9TESUjKYZbMaSYDAv3eUpGTUKv5EjEB5G0+BRInbSBxiN Jvxk09clPDPvsV2X9uy880cPsN60f/3p1Gfjn3ju0UbH3r2VZS19a55//7b5P3ugM+3E63/fG3jq 8OMbm4chWPVLG4z3rO+7JAjTsHeImEqN7TBGvBfT5OC02gnTvZVL5jYvmFj/fy5D8b8KZW5kc3Ry ZWFtCmVuZG9iago2NiAwIG9iago1MjcxCmVuZG9iagoxMCAwIG9iago8PCAvVHlwZSAvRm9udCAv U3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9XU0NRQU0rQXJpYWxNVCAvRm9udERlc2NyaXB0 b3IKNjcgMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDQ4IC9MYXN0 Q2hhciA0OSAvV2lkdGhzIFsgNTU2CjU1NiBdID4+CmVuZG9iago2NyAwIG9iago8PCAvVHlwZSAv Rm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9XU0NRQU0rQXJpYWxNVCAvRmxhZ3MgMzIgL0ZvbnRC Qm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAvRGVz Y2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdodCA1 MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiA2OCAw IFIgPj4KZW5kb2JqCjY4IDAgb2JqCjw8IC9MZW5ndGggNjkgMCBSIC9MZW5ndGgxIDczMjAgL0Zp bHRlciAvRmxhdGVEZWNvZGUgPj4Kc3RyZWFtCngBhVkLfFTF1T8zc/eR9yaQ9yZ7lyWLZBMDAQyP mGweG9CIvILu0mA2hJSEhwQTQBRhaUVwQaF8FoVWUfsQa5WbTaQbsCWV1lYU4RNKK1pBtFX7K0L9 PrW+cr//3F1QWn/97uR/5sw5Z15nzp07s+m5bWU7JVOIBHnblrV2kfHkdiCzt63qUWPllBwic9O3 uxYti5WHLSUyfbBo6Zpvx8q5W4n48x3trQtjZfoC+TUdEMTKbDzykR3Lem6PlXObkScvXd4W1+fe jbJ1Wevt8f7pDZTVW1uXtcfs7bKe2rW8uydenon8qq7b2uP2zE+U9Csdj9Qf/6ieQkEwDBD0IVXS w2QhTjYqo5uIlA6lgEwoS72Jb9vx4daPW9IqP7LmWyEgevztUcUyP6x0bf9s35eLbGRNRjHBsJcK 1LNUDd1IdTb6bN9nd9hiPUnNpUcMUJO4qs+d4zj+nBhNZwEuRkc8BY4BMUoURKY4vFHh6svILE+r KRUqWiwzqAq6HNgHHAIUahGF0NpA1wMhYB9wCDgOmDHBQkOrIl8O7AHOAmZRIOwR1WGrGSVyUTcX 800T2XQB0AFBDtAyYAbQAmwD9gBmw05KlgPrgUPARcBMXpEd2TEOY8+ObDGyvsVLy41ia6zYPN8o 9t0ciOXTZ8Xy+utiZpNjZmPHx8RX18byUSWxPKOoPITG+xJTygdrskQWJpmFgXeBMv4bSmOMHPSo yCQN4AJDNSRekdE30l2+55BQiAkuGC0khz4oWCQlvbwmkev8AmWQg3/Az8c0/Hxfanr5nprr+Tna BxwCBD+H9BZ/i9bzs9LnoNXAHuAQcAy4AJj5WaQzSG/yNymN/5nKgGqgBdgDHAIuABb+Z1Abf0NG jEElXw1w/gaojb+Oab0OmsZPgzvNT+uD/ESkYlL5gMF4yuKMoyjOZOfHmYys8ih/NfLpaESUGyuN iDooRlAVjRMjIkVjHVGRE6nsdET5232qx/FozRh+kjSAYyQn0fNJUoGZQBDoAszgToE7RSFgO/Ao oAGIMlAboPIjwMvAKRoDeIGZgJUfj6CbKD8Wcdc6arL4K/x3lA2PH+W/N/KX+QtG/hL/rZG/iLwQ +iP8hUihg2qSoCfUsSG3IS+D3sR/3Tcyw6HXpPND8KADtAyoBmYALcA2wMwP8RGRhY4MNHKQjuAd dvAIvW/kP6XHreRd7PC66xCAqiTuydeCA9mj7nFzr3vnLhQlcd+/A5wk7ru3gpPEfccGcJK4l64C J4l74WJwkrjntYCTxD2jCRxIlD/yi5GjHBUzljC1Jo2vhpdWw0ur4aXVpPDVMtGnihzjDyLFxfDY bq9ndLEjdICFnmOh2Sz0OAu1s9A6FtrAQpUsdAsLeVjIzkKFLORloYNsIlwRYt7+K4qTvDksdISF nmahbhZys1ARC41kIZVVeKPcGbkObx0yn5H11ciXjjv7rq3C7pPGnfCoEzHvxJ5wCPQYoBslL4zU ETHj3EKZj+grro6Vr55cvrxmGj+MioexDIfpDKBggQ4jjA6jkcNoLg20GmgBBoELgA6YYT0C89hm 0DTQMqAaaAHWAxcAszGcCxgKp+Wgcoj7jIGVgVYDM2SJH0YageTkTm+BzW7z2KaJbXaWVshmFOqF vIKysrAvZ6Rb06MsZf8nKf/8JIUSahL4/XwbFWAhtsfzbZFPCxxR9lDEfdBRk8kepEIFUccmkZsV IZ9I3UZ5AtmtUj6e7Pwp5OUR+02olhZxlzgOsFRZa7/jU/s7jvftUQ72PftBxx/VqMIijj9A8tR+ x0n7vY4Xy6JWSJ5zRxmyA6phOmCf6Hj6iGG6AYrdEcc6me133GWf6lhiNxTtMcUt3Sh50xyz3fMc 09BevX2Bw9uNNvc7qu23OCpjVhNknf2OMRiCJ8YWY7Cj7UanrkKjwbkVUdbhLbHstPgtMyzXWMot JRanxWEpsORbhlszrDZrqjXZmmi1Ws1WxcqtZB0e1c96PfKrN9xsfPzMCGhGisHbsMMwuc2AEmdW TteTNkw08sY5taxRG2yjxgWq9vEcV5QlzpqnmVy1TMtopMamWm2ipzFq0WdrFZ5GzTLzW/5exu4P QKrxzVFGTf4o06VoY76WUecfIMbSN96XL/OrNt4XCFBO1qrqnOqMqvRJDfXfQIKGMFjv+erJ+Yr1 5HgKtJ2Nc/zazwoCWrlk9IJAo/Zfc9Rm/wD7kF301Q+wf8gs4B8QVexD32wpF1X1gUBjlN1k2JHK /gE7RAwy2FkLSZV2pFoLY3a7Y3ZFqA+7kTKDXUICFRl2RQkJhp3CpF1v90hffe9IENhkq9Rt2HRn q1+3OVIEmyIQ2GSF6IhhcyQrJG20KqMZux0mhSAwYXlkN0zsLM8wMUbea5iUxU3uvWxyr9GTiI3G sJEEzaScvWSTchY2X3Pkf2bbaz0e1jcl0Nbsa3f5gi5fOxDUtqzqyNFCC1S1ty0gFaom3MEFbR0y b23XAq72eq3NVa/2TjHq/Yu6WaqnuOp7qdnX5O9t9rbXR6Z4p/hcrfWBvqkzx1dc0de9l/saP/Mb +popGxsv+5pq1PuXviqkeqrsq0L2VSH7muqdavRFRozP9PdaqTZQh/WTeR9PSkS8BvOdgdosW1eV EbxTnDnr8g/gtLKXkjwBLdlVq6UAMq5La0prpArvlFSlQpwWV+Wsm+LMP8D2xlU2iNNdteTpWdm9 knJ8nfWxv248EPWslEsRox4p+8YHJj7N21ovz9aNWvGcRq161jx/r8UCabA+ANnkS7KkJF9UH4wJ r4ZwsjQU4rKhlFVKWUJC3PDfY8EYE8TwzgAOGgf7mLeQ9VB3QGiFjU0cW0HTPLiheZ7/AM5S8iPR HcAEu5mHdV9qTc7D4CkmIUy7+xJ6Vsa5uC964rlh2u0hT/cll1xqziOdZRDDVz0ebG2mA5QL5Jme oFzFTbj/6O8C78l8qFN/T+plzv+GjS4aB9Feepp10tN0iJ5nF1FrHw1QP8kjUD39kNbSA7QJn7V5 kNxLs5FMkD/AcvV+3EwewwfzMToK25tpHR2gLJajv0/raaM4gVobKYVGUA3NpOV0H7tBX0nNdEb5 LlXQDXQrdbGQ7tfv13foP6af0ID4vf4lJVEetSEd1T8w/Ul/g0pR4/u0i86wHQnPkhe9hGD5MN1G u8V8hemL9M8wAietxhgUmk5H2SD3oPV2epflsLWiDq38SNf038DKTvOpg3bTATaBTeVOU7M+XT9K WejjdrS6iyK0HylKv6TTLNl0Uf+xfpFyqYSuw3z66RU2KIa+3DBUDb+Z4KXRNAma5fQr+h0dZy72 a77clGwqN3lNd+gnaTiNpbkY7ROo+Vf2CV+HtF68oDTotZQKv3xPept+S2+xPFbGZrCb+Gi+nD8i biMrehyLtJA64e+H0PqbCKP9PJkfEz9SnlI+NxcMndVTsSJu+gE9TL9mKZipyrrZd9gp9jav4y38 B/yceEB5UnnV0opZ30LL6D56ij5hGWwim8W+xTrYWraJfY/tYkfZcfYer+FNfAm/IDrECvFLpRZp jtKtfNd0j2mL+b0h/9Bvhv576BO9XL+HZiEeNmD036dHMLMBOkavIZ2hc8zEklgqksqcbC67E2kd u489zvayJ1k/ejnOzrH38Un6iH3O8aXlZp6Pw488Arn4bThhPsB/yI8hHed/55+KbDFCeMQEUSkC YjlGtUlsR3pWvKXkKccUHX4uN+007THtNT1let500Zxs+Q6+8S9/8aMvi798c4iGNg/tHIoM9etv USbWEF8PXMEqMfpWpMVY752IuH10giXDd3msmFWxG+CZFraYrWC3w5N3s93sJ8bYn2HPwUt/ZBcw 5hRuN8Z8NZ/Aa/kMpFt4O1+Bw9gO3s9P8c+ERSSJNJEpisVUMV+0ix6xRuwUmnhZ/FmcEx+LL5B0 JVFxKCMUt+JRpiotykrlEeVd5V1Ts+kl01/MieZl5nvMUfM/cKqpssy0zLLMt2yz7LectAYRnYfp WfoFIvDyw86KDcInnqX7+TglF1eYVxDPLbRQTOeIVL6XbeZ3sX4+0nS7eQqfwm6ki4obvn6B7+Ef 8yliOmtkc2gxHxtr0Dxc+Rm4SuUwnVeew9xeQcu3m5PZOn7BnEwRnJEm4Yz0WzFG8YiX6LQ4wyzK Y/S6ksiy2Xn+hJiJKPilUmXyk1P8kJ4RK9hd9Cz3ESV+bt2KOL6R/Qz7QhMrZ/8UOo7BNyKKKsTb 9F1awv9E5/Eeb6YH2UJlEd1P49haepd+irditOlWc7E5k73IO5UwH8b6iStPYnaT2EgmTMPpbjZf 7DZf4K/RSjqmJNKb4ucY/TH+jJiuXDTNZh14A+6ie2iFvoHWmPzKq2wRCXYTFSlnsbutFeWKE/l6 7CrN2NP24+0+gH2gRkyHJAeRcwPiYi52iN1ID2GfUBBBnXjHb8Yu9gr1m5t4lBaZUhl2HfxS89LQ bJqn/5R26YvoVn0HlWI/2KSvRYt76S+0jfayjUN3Uheukq/h3b7B1MCPmRr0Uh7mr/E5fOeV6wtv F7Ec+hvSM1iZKtNBCit/pDlUrW/V/4Dovgo77C5agAPrO5jlB+hhmhikcUM38l69QXRhvmdolv6E 7mCJ1KEvpRn0HP3EYqJWiwdrrLFXMd87qZ3P1ntE+1An/LANXvDCWyux/9zrrZvbVOOtrrq2csrk SRMrJowfVz52TNnVpSWe4tFXjXIXjXSNcKqOwgJ7fl5uTnZW5vBhGem2tNSU5KTEBKvFbFIEZ1Ti czUEVc0d1BS3a9q0Ull2tULQ+jVBUFMharjSRlNlvVaorrD0wvLb/2LpjVl6L1sym1pJlaUlqs+l akfrXWqUzZvlB39fvSugaucNfrrBbzf4FPBOJyqovpyOelVjQdWnNazqCPuC9aUlrDcpsc5V155Y WkK9iUlgk8Bp2a6uXpZdxQyGZ/sm93KypmCKWp6r3qflulAVzYgiX+tCbeYsv68+3+kMlJZorK7N tUAjeVLyGCZUZ3Sjmes0i9GN2okzjkZb1N6SwfDWqI0WBD3JC10LW5v9mmhFGz4t3YN+67XsO97J +aqIxnEm2/R1bb4I+3I6VWkcDm9StUdn+b9WN98pWwgE0Abq8qKGYLgBXW/FSjXKs7jGNwb8GtuI LnGwLDJmFZtf7NRbFFysagmuWldHeHEQS5MX1mj2GmckL887oJ+lPJ8abvK7nFp1vivQWm/vHU7h 2Wv6cr1q7pWa0pJeW3rMsb2paXEmOeXrTDucHtMZnGEuucbZlz3L5Bhd1+EkqKltKkbid2FOEyVp n0jhtolYADwBhlraQqxIp5ZQFwzbJks5psg0U5HNpYY/IkSA6/zfr5S0xiXmIttHJJUyTi6HmsZa L/Gax6MVF8sQsdRhTTHGKqM8obRkVZS7XF023J/lpYFmwretgcllcL/TKRd4S9RLC1DQQrP8sbJK C/Ij5C3D2ZoHpWbwkiZzrtSELmkuVw+6EMn98j5LmZrVffkvzZY1zNcxWWNZ/0HdHtM3znE14mis +sLBeNQ2Nl1RiumlQ+E36OKcNqzOL/I5ZJLj+cLQxk7Il0xwXPYna0oR/sxGUC+MWqyISkPC1AbN FpwWo4FEpzP+zvx/laL6RVnLyL6qFp+GNtkTH2hs2NqUK8pXDC85LBqbsOVwnOzD4cQrdAi12Civ i2eIeFz0nWqdRnPxZhbhD1eOiRKBfM0Ll0HThLfIEAfy48UrDPPjlQJ4ZHSWljRgzwyHG1xqQzgY bo3qoQUu1eYKD/Dn+fPhLh92u1jgRPUDW/K1hq0BeKyDTcbrwam218U2z+r1ss1z5vkH8BOHurnJ H+GM1wVrA3JZeF2TP+4WY0Fk6GMN8R8TRIzSTXOBMwgdZgQQ/u8BRSrKY3B0i/1QgrCi/2Gt+LUE YhMSTtgWIme6M70IBK3QF6oY/MJros9JVQbxuaa5+rtKumkQ/9MoYHN7uRyENzGvUDENL0xJyU6I 6u/1p6XxuZLx5qakgEunZCmhrORk0GQpozLcgY6CHKXq89Xnx47J7zX/e0v/i5bMsqW/9qekGMwH 3tykJHDpZJMSsiUnSypll5v8qs1+s5prsyOQIlzFf2nOUhaQAaThZ6QFinkT35y0Oe3FVFOCJSmH +4bdkHl9bl1+07DmzObc2flLLEuS2oYtzVySG8xfw1ebVyXdkbbJ/JBlp+3FnNP8lPlU0utpeZcn 3p3gdbrGj0lglGBL4AnbHendhODxpkKq4lPMaXvh77bkeGwfz/ecB1nhOR+fOpu/ApeaifJhQCAw zJZxzbjyrKyMTBs3u0aMcg+zZY0rvybd5naNsJjnLjnx6KpIT+3iE4+dXPO9gSfXrn3yyXVrr5/P TzCFXfvzlr4h/fTQ0NDhpx/6BXt46MELF3FTWPxB5z0yDhAN/HOsXSLt86rCm5I+fomynm/ju6zK zxWWQGYTFwkmlszZkURj9IlyTsRU1MVPb/02G5Yuqv/Nm24sqN1Y0FRjQeFlb65crktrYqxPXrLJ m5I23nTJE2NMTMXtiptykw6wSraRcjw32t6BM+CX+K0YhcrpX1ZSdXX2JJY+aewYNp/me5yudLPZ MuGaayrG8c/7a040PXiurEe5s2qt45mpR1oQvXCw8eijcC/4pkdAOAznPemHDEA+xvtw85y6WTXT PTW3dbYund70f2cKjbAKZW5kc3RyZWFtCmVuZG9iago2OSAwIG9iago1MDAyCmVuZG9iagoxOCAw IG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5cGUgL0Jhc2VGb250IC9LUVRMTkMr QXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNzAgMCBSIC9FbmNvZGluZyAvTWFjUm9tYW5FbmNvZGlu ZyAvRmlyc3RDaGFyIDQ4IC9MYXN0Q2hhciA1NCAvV2lkdGhzIFsgNTU2CjAgMCAwIDAgMCA1NTYg XSA+PgplbmRvYmoKNzAgMCBvYmoKPDwgL1R5cGUgL0ZvbnREZXNjcmlwdG9yIC9Gb250TmFtZSAv S1FUTE5DK0FyaWFsTVQgL0ZsYWdzIDMyIC9Gb250QkJveCBbLTY2NSAtMzI1IDIwMDAgMTAwNl0K L0l0YWxpY0FuZ2xlIDAgL0FzY2VudCA5MDUgL0Rlc2NlbnQgLTIxMiAvQ2FwSGVpZ2h0IDcxNiAv U3RlbVYgOTUgL0xlYWRpbmcKMzMgL1hIZWlnaHQgNTE5IC9TdGVtSCA4NCAvQXZnV2lkdGggNDQx IC9NYXhXaWR0aCAyMDAwIC9Gb250RmlsZTIgNzEgMCBSID4+CmVuZG9iago3MSAwIG9iago8PCAv TGVuZ3RoIDcyIDAgUiAvTGVuZ3RoMSA3NTgwIC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVh bQp4AYVZCXxU1dU/9743SzYyCWSdSeZNhgySSQwEMGExmSQzAQ07iZ2hiUlIUsImgQRwQRhaEQgI lM9SpFXULoJWeZlEnAAtUdRWFOETSisu4NKK/TVA/ap+LuR9//tmQGn99Xsv/3PPPefc7Zzz7rtv 0rl8RSvFU5Ak8jQvaWon/cpoQ+FqXtmpROoJ6UTGmh+0z18SqQ9dTGS4OH/xXT+I1DN2EUmH21qb WiJ1+hrlTW0QROpsLMrhbUs674zUM+pQxi9e2hzVZ2xH3byk6c7o+PQ26sodTUtaI/ZZq0W9fWlH Z7Reg/KG9uWtUXvmJ4r7nYZL6E9+6qVgIxgGSPQJTaKHyUScLFRItxHJbXIWGVAXegPftuOTLZ81 JE761Gw1Q0D0+Acj8kR5VG7f/uX+K/MtZI5HNUa3Fwq0M5UOTqdKC325/8u7LZGRhObqJfVRjXRD jyvdfvKwNJLOA1waGXJn2fukEVJWaKLdE5acPckpRYnlBZKCHgt1qoAuBfYDRwCZGqRsaC2ga4Eg sB84ApwEjFhgtq5VUC4F9gDnAaOUJdlCit1SPkLKQNsMrDdRSqNLgAZIZActBGYADcA2YA9g1O2E ZCmwFjgCXAaM5JHSQjvGYO5poc160bNwcZFebYpU6+r1as/3ApFy2qxI6b0lYjYhYjZ6bER8Y0Wk HJEfKZNzi4LovCc2oai/PFVKxSJTMfF2UMZfpETGyE6PSimkAlzCVHWJR0ruGe4q2nNEkolJXGLU QnatX2KhhKSi8liu8UuUTHZ+kQ9ENHygZ0hS0Z7yW/n7tB84Akj8fdzv8fdoLT8vfA5aBuwBjgAn gEuAkZ/HfQ73u/xdSuTvUCFQBjQAe4AjwCXAxN8BtfC3RcboVPBlAOdvg1r4W1jWW6CJ/Cy4s/ys 1s9PhYrHF/XpjLswythzo0yaNcokpxaF+RuhL0Yio1yINDLqkJRDpTRGygnljraHpfTQpAX2MP+g R3HbHy0fxU+TCnDM5DRGPk0KMBNoBNoBI7gz4M5QENgOPAqoALIM1AIo/BjwGnCGRgEeYCZg5idD GCbMT4RcFfbyVP46/z2lwePH+R/08jX+sl6+yl/Sy1dQZkN/jL8cyrZTeRz0hDYWlBaUhdAb+PM9 w5PtWnkSPwIP2kELgTJgBtAAbAOM/AjPCbXYk9HJITqGZ9jOQ/SxXv6aHjeTZ6Hd46pEAiqCuCbc DA5kj7LHxT2unQ+hKohr6w5wgrju2wJOENfd68AJ4lq8EpwgrpaF4ARxzW0AJ4hrRg04kDB/5Lnh I+zFMxYxpTyRr4KXVsFLq+ClVSTzVeKmL2Qxx5+F8vLgsd0e98g8e/AgCx5mwdks+DgLtrLgGhZc x4KTWPB2FnSzoI0Fs1nQw4KHWAlcEWSe3uuq4z3pLHiMBZ9mwQ4WdLFgLgsOZ0GFFXvC3BG6BU8d Cp9e9JSLh447em4uxe6TyB3wqAM578CecAT0BKDpNQ+MlJyIcUa2KHN68soi9RsnFC0tn8KPouFR hOEonQNkBOgo0ugoOjmK7hJBy4AGoB+4BGiAEdY5WMc2nSaCFgJlQAOwFrgEGPXpXMJUOC0FFVPc r0+sELQMmCFq/CjuHNwO7vBkWWwWt2WKtM3GErPZjGwtmxdTair25eQkc1KYJRz4POF/P0+gmPIY vpVvoywEYnu03Bb6IsseZrtCrkP28hT2U8qWkXVsPLlYLsoS6tDr48hmFvKxZONPoSwK2W5Ds8SQ K99+kA0RrQ7Yv7B9aP/YFuZgL9gO2f+khGUWsv8RkqcO2E/bNtlfKQybITnsCjMUBxXdtM9WYn/6 mG66DordIfsaURyw32ubbF9k0xWtEcXtHah5Eu2zXXPtU9Cf1zbP7ulAnwfsZbbb7ZMiVuNEmwP2 UZiCO8LmYbIjbfqgzmy9w9riMGvz5Jt2mvymGaabTEWmfJPDZDdlmaymYeZks8U8xBxvjjWbzUaz bOZmMg8La+c9bvHWG2bUX35GJDQjWect2GGY2GZAiTMzp1tJHSpV8+o5Faxa7W+m6nmK+tkcZ5jF zpqrGpwVTE2upuqaCrXEXR02abPVYne1apr5fX83Y1sDkKp8Y5hRjT/MNCFab1WTK/19xFjS+ges orxh/QOBAKWnrixLL0suTRpf5f0O0qgLG73ub670b1h3ujtL3Vk9x68+mRVQiwSjZQWq1f+ao9T5 +9gn7LLP28f+IYqAv08qZZ/4Zgu5VOoNBKrD7DbdjhT2D9ghY1DAzpxNirAjxZwdsdsdsctFe9gN FwXsYmIoV7fLjYnR7WQm7Lo7hvu83cNBYJOmUIdu05GmfNvmWC5sckFgkxqkY7rNsdSgsFFL9W5s Nphkg8CEZZJNN7GxTN1En3m3blIYNdl0zWSTPpIUmY1uIwi6STh/1SbhPGy+5cj/zLZWuN2sZ2Kg uc7X6vQ1On2tQKO6eWVbuhqcpyjdzQGhUFTJ1TivuU2UTa1qwNnqVZudXqV7ot7uX9R1Qj3R6e2m Ol+Nv7vO0+oNTfRM9DmbvIGeyTPHFl831qZrY42d+R1jzRSdjRVjTdbb/ctYxUI9WYxVLMYqFmNN 9kzWxyI9x2f6u81UEahE/ETZw+Nika+NVkegItXSXqon70RH+hrrQZxW9lKcO6DGOyvUBEDkdUF5 QblQ4ZkSqiEQJ0ZV6WsmOqwH2d6oygJxkrOC3J0rOlZQum+BN/LXgQuizhUiFBHqFrLvvGDiUz1N XnG2rlbz5lSrZbPm+rtNJkgbvQHIJlyVxcX5wlp/RHgjhBOEoSRdMxSySUIWExM1/Pdc0OcEMbzT h4PGoR7myWad1BGQ1OzqGo6toGYu3FA3138QZynxkugIYIEdzM06rvYm1qHzFJEQlt1xFZ0rolzU F53RUjftcJO746pLrnbnFs7Sie6rTje2NsNBygAyDU9QhuwifP9oHwEXRDm4QLsg9KLkf8NGF46C aC89zRbQ03SEXmCX0Wo/9VEviSOQl35Oq+lB2oDX2lxINtFs3AbIH2QZWi++TB7DC/MxOg7b79Ea OkipLF37mNbSeukUWq2nBMqhcppJS+kBNlVbQXV0Tv4RFdNUuoPaWVDza1u1Hdov6VfUJ/1Bu0Jx lEnNuI9rFw1/1t6mArT4CT1E59iOmGfJg1GCsHyYltNuqV5m2nztS8zAQaswB5mm0XHWz93ovZU+ YulstVSJXn6hqdqLsLJRPbXRbjrIxrHJ3GGo06ZpxykVY9yJXh+iEB3AHabf0lkWb7is/VK7TBmU T7dgPb30OuuXBq+sGyyD3wzw0kgaD81S+h39nk4yJ3ueLzXEG4oMHsPd2mkaRqOpFrN9Ai3/yj7n a3CvlV6Wq7QKGgK//Fh4m16i91gmK2Qz2G18JF/KH5GWkxkjjsbdQgvg713o/V2k0QEez09Iv5Cf kr8yZg2e14YgIi76GT1Mz7MErFRhHeyH7Az7gFfyBv4z/r70oLxPfsPUhFXfTkvoAXqKPmfJrITN Yt9nbWw128B+zB5ix9lJdoGX8xq+iF+S2qRl0m/lCtxz5A75R4b7DZuNFwb9gy8O/vfg51qRdj/N Qj6sw+x/Qo9gZX10gt7EfY7eZwYWx4bgVpiD1bJ7cK9hD7DH2V62j/VilJPsffYxXkmfsq843rTc yK04/IgjkJMvxwnzQf5zfgL3Sf53/oWUJuVIbmmcNEkKSEsxqw3SdtzPSu/JmfIJWYOfiww7DXsM ew1PGV4wXDbGm36Id/xrX//iSt6VdwdpcOPgzsHQYK/2HqUghnh74BNsEmbfhHsh4r0TGbefTrF4 +C6T5bFSNhWeaWAL2TJ2Jzx5H9vNfqXP/Rl2GF76E7uEOSdwmz7nG/k4XsFn4L6dt/JlOIzt4L38 DP9SMklxUqKUIuVJk6V6qVXqlO6Sdkqq9Jr0jvS+9Jn0NW5NjpXtco7skt3yZLlBXiE/In8kf2So M7xq+Isx1rjEeL8xbPwHTjWlppmmWaZ60zbTAdNpcyOy8yg9S88hA69d7Ly0TvJJz9JWPkbOwCfM 68jnBmqRpnFkKt/LNvJ7WS8fbrjTOJFPZNPpsuyCr1/me/hnfKI0jVWzObSQj450aBwmPwluknyU BuTDWNvr6PlOYzxbwy8Z4ymEM9J4nJFekkbJbulVOiudYyb5MXpLjmVpbIA/Ic1EFvxWLjX4ySH9 nJ6RlrF76VnuI4r9yrwFeTydPYl9oYYVsf+VNByDpyOLiqUP6Ee0iP+ZBvAcb6SfshZ5Pm2lMWw1 fUS/xlMx0nCHMc+Ywl7hC+QuPpT1Epf3YXXj2XAmGYbRfaxe2m28xN+kFXRCjqV3pd9g9if4M9I0 +bJhNmvDE3Av3U/LtHV0l8Evv8Hmk8Ruo1z5PHa31VKR7EC5FrtKHfa0A3i6D2IfKJemQZKOzJmK vKjFDrEb9y7sEzIyaAGe8e9hF3udeo01PEzzDUMYdh38UvPq4Gyaq/2aHtLm0x3aDirAfrBBW40e 99JfaBvtZesH76F2fEq+iWd7qqGKnzBUaQW8i7/J5/Cd18cX3s5l6fQ33M8gMqWGQ9Ql/4nmUJm2 RfsjsvsG7LAP0TwcWD/EKi9ihClSP40ZnM67tSqpHes9R7O0JzQ7i6U2bTHNoMP0K5OBmkxuxFhl b2C991Arn611Sq2DC+CHbfCCB95agf1nk6eytqbcU1Z686SJE8aXFI8bO6Zo9KjCGwvy3Xkjbxjh yh3uzHEo9uwsmzUzIz0tNWXY0OQkS+KQhPi42BizyWiQJc4o3+esalRUV6Mqu5xTphSIurMJgqZv CRpVBaKq621URbRrguo6Sw8sf/Avlp6IpeeaJbMok2hSQb7icyrqca9TCbO5s/zgH/A6A4o6oPPT dH67zieAdzjQQPGlt3kVlTUqPrVqZVuXr9FbkM+642IrnZWtsQX51B0bBzYOnJrmbO9maaVMZ3ia b0I3J3MClqhmOr0+NcOJpuhGyvU1tagzZ/l9XqvDESjIV1lls3OeSuKk5NZNqFIfRjVWqiZ9GGUB zjgqbVa68/u7toQtNK/RHd/ibGmq86tSE/rwqUlujOtV0+7+MP2bKjrHmWzDt7VWqcuXvkARxl1d GxT10Vn+b7W1OkQPgQD6QFueW9XYVYWhtyBS1eIsrvL1Ab/K1mNIHCxz9VVF1hc59eY2LlTUGGeF s61rYSNCk9ml0uy7HKHMTE+fdp4yfUpXjd/pUMuszkCT19Y9jLpm39WT4VEyrtcU5HdbkiKO7R6S GGXiE77NtMLpEZ3O6eaCq559zbNMzNF5C06CqtKsYCZ+J9ZUIkhrCXU1lyAAuAIMrdQWRGSBGlPZ 2GWZIORYIlMNuRan0vUpIQOcA3+/XtIUlRhzLZ+SUIo8uZZqKmu6yqtut5qXJ1LEVImYYo6len1c Qf7KMHc62y34fhYfDTQTvm0KTCiE+x0OEeDNYQ/NQ0UNzvJH6grNs4bIU4izNW8Umv6rmpRaoQle 1Vxr3uhEJveK71lKUc2ua3+JltShvrYJKkv9D+rWiL56jrMaR2PF19UYzdrqmutqEb1wKPwGXZRT h1b6JSuHTHDcKunayAn5qgmOy/54Vc7Fn1FP6pawyYys1CVMqVItjVMiNBDrcESfmf+vUVi7LFrp xTfNostQJ7ijE41MW514Xf266cV3SdU12HI4TvZdXbHX6ZBqkVneEi2Q8fjQdyiVKtXiyczFHz45 SgQCVtUDl0FTg6dIFwes0ep1htZoowAukZ0F+VXYM7u6qpxKVVdjV1NYC85zKhZnVx9/gb/Q1e7D bhdJnLB2cLNVrdoSgMfa2AQ8Hpwqup1s46xuD9s4Z66/Dz9xKBtr/CHOeGVjRUCEhVfW+KNu0QMi Uh8xxH9MkDFyB9UC05A6TE8g/N8DCnxP0Cj8T0Kk1NVL/GQirhT6H/YufjkBa8CN07aJyJHkSMoF QY/0tSL1f+0x0FekyP14dVOt9pGcZOjH/zeyWG03FxPyxGZmy4Zh2QkJaTFh7UJvYiKvFYwnIyEB XBLFCwmlxseDxgsZFeJ76DjIcSobKBsYPcrabfz3nv6Jnoyip7/2JiTozEVPRlwcuCSyCAlZ4uMF FbJrXX7TZ69RybDYkFQhruA/NucpFUgGEvGT0jzZuIFvjNuY+MoQQ4wpLp37hk5NuTWj0loztC6l LmO2dZFpUVzz0MUpizIarXfxVcaVcXcnbjDuMu20vJJ+lp8xnol7KzHz2sI7YjwO59hRMYxiLDE8 Zrs9qYOQSJ4hkCp4LXPanv37zeluy2f17gGQZe6B6NJZ/TJ84JSIiwGBwFBL8k1jilJTk1Ms3OjM GeEaakkdU3RTksXlzDEZaxedenRlqLNi4anHTt/14759q1fv27dm9a31/BST2c2/aegZ1M4ODg4e fXrXc+zhwZ9euoyvhoUXF9wvYjcNsUtB7LIoj82Ixs6eyOw4QEvMekO2J4ElJAwzZFsNOdnDEmKz GeVawto/I3G0ZKdZRBzT4HRQPY5p0TgeP33c8pK+nuTxhfUDlhfrEVJPwaIM5jV5UrwZXmVuco2y SGoxtZgXJrconeYVtvXm+21nzKdTk0yKiOUI/M6nh9kpfvCLEyKHrjAJxQjFqTiEIknMcmYCxzyt 7FQDkjrM2zwxV+fMwqzEk0zP5nZY9HhY8K8Vi4VjFZefE7li2Z4fe5CVUDYb70ktS2tIW5q2Nk1O Q2oaa9NSxaBpYT68xx0JldtdP4BguUW0/hmJ15X6D5P0NY4eJaJVIr7rWT3VB5jJNUIPkNF0E8KX nDLMiOBRkqVYBJMNSxUxHDcWJkbpq570/FsW3VZeO4+XH57fe2XVyfveG/zw4U0Xnn7nSvGMrdOX //Lxe+5+Up4zZOGoaaNKL77d3Dj4+RtdA2tw6F/N9j2/94Wv36l/MhB+ZNf+/Yiqfmkj8G30XZcE 4VB8S4k9ITm6BxjxLUvVs2qmTq90ly9f0LR4Ws3/AWpp4OoKZW5kc3RyZWFtCmVuZG9iago3MiAw IG9iago1MjIwCmVuZG9iagoyMiAwIG9iago8PCAvVHlwZSAvRm9udCAvU3VidHlwZSAvVHJ1ZVR5 cGUgL0Jhc2VGb250IC9OSVFOV08rQXJpYWxNVCAvRm9udERlc2NyaXB0b3IKNzMgMCBSIC9FbmNv ZGluZyAvTWFjUm9tYW5FbmNvZGluZyAvRmlyc3RDaGFyIDMyIC9MYXN0Q2hhciAxMTYgL1dpZHRo cyBbIDI3OAowIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAw IDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwCjYxMSAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAg MCAwIDAgMCAwIDAgMCAwIDAgMCAwIDAgNTU2IDAgNTAwIDAgNTU2IDAKMCAwIDIyMiAwIDUwMCAy MjIgMCAwIDU1NiA1NTYgMCAzMzMgNTAwIDI3OCBdID4+CmVuZG9iago3MyAwIG9iago8PCAvVHlw ZSAvRm9udERlc2NyaXB0b3IgL0ZvbnROYW1lIC9OSVFOV08rQXJpYWxNVCAvRmxhZ3MgMzIgL0Zv bnRCQm94IFstNjY1IC0zMjUgMjAwMCAxMDA2XQovSXRhbGljQW5nbGUgMCAvQXNjZW50IDkwNSAv RGVzY2VudCAtMjEyIC9DYXBIZWlnaHQgNzE2IC9TdGVtViA5NSAvTGVhZGluZwozMyAvWEhlaWdo dCA1MTkgL1N0ZW1IIDg0IC9BdmdXaWR0aCA0NDEgL01heFdpZHRoIDIwMDAgL0ZvbnRGaWxlMiA3 NCAwIFIgPj4KZW5kb2JqCjc0IDAgb2JqCjw8IC9MZW5ndGggNzUgMCBSIC9MZW5ndGgxIDEyMjUy IC9GaWx0ZXIgL0ZsYXRlRGVjb2RlID4+CnN0cmVhbQp4Aa16C3xUxdn3zJyz5+x9z+a2u9lN9mw2 2UA2GEgCISGSE5IAGrlfzGIiCRAlAYQQ8FJRQhWBiEJttVXb4q1ea9lcpAH1JfXW11vhVWurVUHF VttGeP0oKpLs95/ZxEq//t7v9/1+3zl5Zp6Zeeb23OaZs9m0cXMrsZMuIhFj5bqWDUQ8vh8he27l 1Zv0ZNnhJURZfMWGK9cly6lrCTF9fuXa665IljOvJSRdXt3asipZJueQT1mNimSZliLPXb1uE+j4 42tEMnXt+pWj7ZkNKGeva7l2dH7yHsr6VS3rWpHjKXoGib5hfecmUSRFW5AbGza2jtJT9Lf9RwIP bz/6j1rS1QyEApzkC1JJfkZUwohGishSQuTVchYxoczbTWzPHV/sPrPcVfkPs9+MCkIe+Di/gOfP yxv2nt0/fKVGzHYULYKeN6CfOn1kLqnRyNn9Z7+nJWfiLWOP8yBZLI3ri3iDR5+RxpPjACaN741m BQ9K+VJW77SgMSCF+1LSi13VEyQdIxaJVEe6HrAfcBggk+VSNlo1pFsBXYD9gMOAowCFEKS8VQes B+wDHAcoUpYU6NWDWnW+5ENfH/brkjzkJCABkEgQaRFgHmA5YA9gH0ARdLxmPWAr4DDgFEAhhuTp vaMEa/f03iqyvva1xaLYkiw2Noli36WxZD5nQTKvvShJVpEkm1SarL5gRjLPL0zmKXnFXRi8z+oo HqzOkDKwyQwsfANSyl4gLkpJkNwnpZM4gElYqqgxpJS+3EjxvsOSTKjEJEpWkWBiUKK9DndxtZUl 2EmSQoLsczaUbGFDfU538b7qi9lHZD/gMEBiH+H9kH1ItrLjnOdIqwD7AIcBRwAnAQo7jvcY3g/Y B8TF3idFgCrAcsA+wGHASYDK3keqsfe4xoiU41UAxt5DqrE/YVt/Qupi7wJ7l72bGGRv9paVFx8U SLRoFAnmjSIe/yiSklE8wN7o/Xo8NCoCSUOjnpZyyHRSIuX05k0KDkje3sq24AD7uE+PBu+rnsje InEAw0rewsxvER0wH9AM2ABQgL0N7G3SBdgLuA8QB0DLkGoAnb0CeA3wNpkIMADzAWZ2tBfTDLAj vZEZweoM9jv2W+IBx19n/yny19hLIn+VvSjyl5Fno/0V9lJvdpBU29BO0EdDriEvQruJ/aYvNyWY qHazw+BgEGkRoAowD7AcsAegsMMsp3dVMAWDPE1egQ0HWS/5TOQPkwfMxGgPGpEaKKDOk0jFhcCQ 7NP3RZgRuetuFHkSuf0OYDyJ3LwbGE8i39sGjCeRtVcD40lkVTswnkSWLQfGk8i8xcCQDLCf/zo3 P1g2bw3Vq13sGnDpGnDpGnDpGiKza/hLvpb5Gu/tLSgAx+4xouMLgl2HaNcztGsh7XqAdrXSrhtp 1zbaVUm7LqddUdoVoF3ZtMugXU/TqWBFFzX6zyuWG17a9QrtepJ2ddKuCO3Ko125tEunZcYAC/Ve BKtDVieyvmpudCzUd+F0eB8XC4GjIeh8CD7hMNIjgIQoGSDSc5LEvmye5/QVVCXLF1QUr6+ezZ5H x+chhufJMYAMAT0PNXoegzyP4VxIqwDLAYOAk4AEQAF1DvaxR6QupEWAKsBywFbASYAilnMSS2Fk PVK+xP1iYUVIqwDzeIk9jzcHb4iFjCwtoEW12dKeAHVl03nZiWxWRjIy4JdT3Gb3AHUc+NLx1ZcO Yqm2sNvZHpIFQewdzff0fp0VHKA/6Y08HaxOpz8m2TK0jpaTCM1DPpV0ivJkEjDz+lISYE8gL+4N LEU3V2+kMHiIOnmvA8GvAyeCnwUGGNBPA08H/6APyLQ3+HvUPHEg+FZgV/DlogEzap6JDFBkh3RB ejAwNfjkK4J0Gxru6Q3eyLMDwRsCs4JrAqKhNdlweSdKhiu4MLIsOBvj1QZWBI1OjHkgWBW4PFiZ pJrM+xwITsQSokm0AIsdHxCThrPFgEvKBuhqo1C9S21Q56lT1GK1UA2pQTVL9atp5hSzZnaa7War 2WxWzLKZmYk5bSBx3IjyUy9NEYefAoWmRBa4Bg9DuZtBShg1M3IxiadK9ax+0QxaHx9cSepX6PEz i8ID1LpgWdwUnkHjKfWkfvGM+NRo/YCaWBgvi9bH1fmXNfRQensMtXG2c4CSxQ0DNMGrtvvjKTUN Bwml7u23+Xk+bvttsRjxZlxd5a1Kme4un1n7b5JmUdlcG/3n4/0nGvVGs+J31S9qiD+eFYsXcySR FauP/3CR3thwkH5BT9XVHqT/zbNYw0FpOv2ibiGvl6bXxmL1A3SpoCM6/W/QQWOQgc6Mg5nTEd2c naS7J0mXh/6gy+UZ6CwWkifo8iwWQSdTTtfTmVtX25OLBDQenXQKmk6P/l2aV/JAk4cENBld5BVB 80pGF6eJTxfDBAIgyUYCEppJAoIkQDMFiVh5jyApGiXZ9S3JLjGTlFyNoOEJhnEcH6NxHAfNdxj5 P6OtM6JR2jcttrKxrjVc1xyuawU0x2+9erU33rVC13tWxniDHpcizStWruZ5S2s8Fm6tja8M1+o9 00S/f2lu5M3TwrU9pLFucUNPo9Fa2zvNmFYXbqmN9c2aX1p23ly7vp2rdP6/mWs+H6yUzzVL9PuX ucp48yw+Vxmfq4zPNcuYJeYiQsfnN/SYyYxYDeTH8z5ms0Jfm/2h2IwMbcN0obzTQt4b/YcQrTxK bNFY3B6eEXcAuF5PqJ5QzZtgU7zJiWrXaJP3xmkh/yH66GiThmp3eAaJbtrcuZl469pqk3+deFC1 aTMXRTKN8rp/+4CkLm601PLYuj5esKg+XrVgWUOPqqK2uTaGuoqxOputbiAxmKy8AJUVnFCSviXk dZW8zmIZJfw/dUGsCdXgzkEEGk/3USObbiKdMSmeXb+YwRUsXgY2NC5rOIRYih8SnTFssJNGaefY aHwfAifJGoJtd47Bps2j2CgvNo3mgrQzSqKdYywZGy7KmSUSwatNUbg20yHiA2SaHiE+OUK8hCT+ AviU5yNtiU95O8/ZX+HoBkaBkEfJk7SNPEkOk+foKfTaTw6SfsJDoFryU7KF/IjswLG2DDW7yEK8 JtT/iPoS/biZ3I8D837yOmgvJTeSQySDehOfka1ku/Qmem0nDpJDqsl8sp7cRi9JbCaN5Jh8Eykj l5CryAbalWhI3J64I/EQ+QU5KP1nYpjYSCZZiff1xOemPybeIxPQ405yNzlG77A8RQzM0gXKn5GN 5B6pSaaJKxNnsYIQuQZrkMkc8jodZFGM3kr+Qr10i1SDUR5MxBMvgCpAmshqcg85RCfTWSxkakzM SbxOMjDHtRj1btJLDuAdIM+Sd6nddCrxUOIU8ZFCchH2009+RwelkeFtI1XgmwlcGk/K0bKe/Af5 LTlKw/Q3bL3Jbio2GabvJd4iaWQSWYLVPoKef6ZfshvxbpVekmcmZuB2t538gHObvEg+pJm0iM6j S9l4tp79XNpIzJhxEt5VpA38/glG/wBqdIDZ2RHpQfkJ+Rsla+R4wgmJRMi95GfkN9SBneq0k36f vk0/ZjVsObuXfST9SH5MfkNtwa4vJ+vIbeQJ8iVNoVPpAnoZXU230B30B/Ru+jo9Sj9l1WwxW8NO SqulDulZeQbeRXKnfJPpFtOtyqcjDSMvjPzXyJeJ4sQtZAH0YRtWfyf5OXZ2kBwh7+A9Rj6iJmqj Trw6DdEl9Hq8N9Lb6AP0UfoY7ccsR+lH9DMcSf+g3zCctExhfgQ/PAQKs42IMH/EfsqO4D3K/s6+ ljxSjhSVJkuVUkxaj1XtkPbifUr6UM6Uj8gJ8LnYdJdpn+lR0xOm50ynFLv6fZzxr517cLhg+IMR MrJz5K6R3pH+xIckHTLE6YErWCVW34K3HfK+Cxq3n7xJ7eBdJi2g0+kl4Mxy2k476LXg5M30HvoL sfZf0WfApT/Qk1izgwXEmi9gk9kMNg/v5ayVdSAYu4P1s7fZWUmVbJJLSpcKpFlSk9QqbZKuk+6S 4tJr0vvSR9IZ6RzehGyVg3KOHJGj8ix5ubxZ/rn8F/kvpkbTq6ZPFKuyTrlFGVD+G1HNdHW+ukBt UveoB9S3zM3QzufJU+TX0MBvH3pc2ibVSU+R21mJ7MMV5nfQ5+VklTSHQVPZo3Qnu4H2s1zTtco0 No3OJafkCHj9EtvHzrBp0hxaTxeRdjYpOaCSJj8OrFJ+ngzJz2Bvv8PI1yp2eiM7qdhJL2KkcsRI L0oT5aj0KnlXOkZV+X7yJ9lKPXSIPSLNhxY8K083NZCQ9FPyK6mD3kCeYnWEWL8x74Yez6WPwy8s psX0KymBMHgutKhM+pjcRNawP5Ih2PFO8mO6Sr6S3E5K6BbyF/IwrGK86SqlQEmnL7M2uZul0n7C 5Mewu3KaSyVTGrmZNkn3KCfZO2QzOSJbyQfSL7H6I+xX0hz5lGkhXQ0LuIHcQjoS28h1pgb5DXol kehSkicfh3fbIhXLIeRb4VUa4dMOwLoPwQ9US3NQ44XmXAK9WAIPcQ/en8BPyNCgNtj4pfBivyP9 ymI2QK40OSm8Dr7UvDqykCxLPEzuTlxJrkrcQSbAH+xIbMGIj5JPyB7yKN0+cj3ZgKvkO7DtS0wz 2RHTzMQE1s3eYYvYXefLF9zOo17yV7y/gmSmm54m3fIfyCJSldid+D20exw87N1kBQLWE9jl55hh tjRISkbmsp7ETGkD9nuMLEg8kghSK1mdWEvmkWfIL1QTaVGjkHGcvoH9Xk9a2cLEJql1pA182AMu GODWZvifXUbNksXVRtX0CyunVZRPLZtcWlI8aWLRBRMKowXjx+VH8nLDOSE9mJ0V8Gf6vJ6M9LTU FLfmcjrsNqvFrComWWKUFNaFZzbr8UhzXI6EZ8+ewMvhFlS0fKeiOa6jaub5NHGd92tB03mUBiiv +BdKI0lpfEtJNb2SVE4o1OvCevz12rA+QJctaAB+W204pseHBD5H4HsF7gAeCqGDXuddXavHabNe F5959eruuubaCYW0x2atCde0WicUkh6rDagNWNwT3tBDPdOpQJinrqKHEbMDW4xnhmvr4r4wumIY Ka+uZVV8/oKGulp/KBSbUBinNSvDK+KER0pRQUJqxDRxpSauimn0NsQ4cXKr3lM42L17QCMrmqP2 VeFVLY0NcakFY9TF3VHMWxv3fO+E959FDI6YbMd3W/1Sd523TefE3d079Ph9Cxq+09cf4iPEYhgD fVnezObumZh6NyRVz2PxONsea4jT7ZgSgWWe2FVyf8moN6+5XY9bwjPCq7vbmyGazO44WXhdqDcz 0ziYOE4y6/TuxQ3hULzKH4611AZ60kj3wuv6fIbuO79lQmGP5k4ytsfpGkXsju8irWB6sk1ggpxj 9Qu/5SzlawxfhEgwrq/UsZKGMPY0lSetU0n3yqkQAJ4YRa/4KkikLW6pae7WKng9tkjjpjwtrHf/ g0ADwkN/P7+mZbRGydP+QXgj15NvVS1OW8bweDQaLyjgKqLWQKZY43RRnjyh8OoBFg5v0HB/5pcG Mh+8bYlVFIH9oRAX8K0DBlmBQrxrQUOyrJMV/l5iFCG2Zs28ZXCsJX0Jb+kaa/m2e3MYmtzP77Mk PW6OfPvn0jJS61ZXxGnG/9DcmmyvXxSuR2is13U3j2pt/eLzSsl2zlDwDW2jWDy1pkHyM9RxjPkl 0ZqMkMdIEC432ONyHv4UodSrBlQztFLUUH1mXGuenUxj1lBo1Gb+b50GEqd4L5H9s9voNuIV0dGF Jpcdn3Ze+bzl2bul+sVwOQyRfXe39bw2qFpylReNZtB4XPRDek2cLIFl5uEPV46pHGL+uAGWoWUx rEhUx/yjxfMI/aOdYni4dk4onAmf2d09M6zP7G7ubhlIdK0I61q4+yB7jj3XvaEO3i6pOAOJQ7f6 4zN3x8Cx1bQC5sHIjJ4w3bmgx6A7Fy1rOIhPHPrOxQ29jLKa5hkxLhZWs7hhlC1CIFz1IUP8YgKN 4We8/DF5VO4kF+PWNh95Pf0t2Ql8J/KbUK4F7GCPk+2oqwJ9LtSMCmUj+GVGwamOOREBc/X7//Xw zzP/74+ELlgkYvbkg+/C+F0l+aOJhVhRsmHFjtHWdJyoG+g7rE+6Q7abGkxvK79W7zYHLKes3wcF jjI+EIaSMAYJuUPuPCT4ikTO6dLgOcNEviG6PMh58Sim3Y4bl4XcZkQVU7bZvEelqkokORtyIGb1 pzrTbYxl2mSLTnX00DWNLSEDiTP9Ltco4nAI5Kxhs9uB2W02hVOM9NvtAjllWB0OYE3WaY3e6Fzt 9Oj1sKlyznDlXO1M05wTpKqqcrgypbyoqVIbrpw0scQdSseaAY9K75/7hMWH55sOPTlS8eTwFQjL L058Kgfk6YgqytgEo9DisBT4HJkF4x0FBeWOKell/oqCiwqaHE0F7Y62guaJ3Y5bxt+TcW/mY470 cQOJT/v5+vKBGD6OPex7fNwB39PjXvAdGfdG+vvjzLUZNHsgcdpw8+WnpPDUZOfpZP5xbgnHgp6g N1pYUFoulxdeJM8uXGqORa8wt0Wvtu+wv2z/2vF11F1W6qSyVpRb6ikOpXmXj18/no0PFDmrnHuc +5wJp2mfc7/zpFNy2jkbnQOJv/ZzxgI5baRrmrLEaec8cyouF1JnQPIMsMcPeO9MCwRU8Pa0kcnX QeryrcUByTa+RWshipBDXih3IPF3MRhHDBuvzZW5ZFA+gc0L5LTgAmreM2x8ulwxEcrnhNRyB9hl hjPfIBEtokcmRvZHTOXwFP1OJ1sSGUi8fUAgk3id4cgOl04sHyxn95XTcg9fWzUf0ZPnzSnKPawc UVhQqVKY4uQ7VbAtpF6+HgWakkyVJYqTb1fR+K6USVO1M2M60jF0eiiqNXVEUdM0NKY60crh6Cef kKqhqhPRqqHhE26ojujR1DHUgQL+yqk7xVM+aSJp4g20AxnpyFOUcE5kcumUKWXinVyaHwnnKGr+ dFZSnJGB0DA9LcMTjkiK6mRAS4pBhgveqoPt+5+Z1Tl78pp3r6QldTu3XpcV9151dNfOx+drFk/O MwHPihfWNxava1v9QCTrpiUzn9g+d9vcNKcjMzfPetWEC2Md3o5b642Wiy+49tQ32y+cSt8fF9DG zSma3XzZvAuvgVXNT3wqDUGjM+myHsbdnlHq3OqiLhs18F1iAyxZTgnYVG9Axv01XTVz9quClSpU E7jGWakKHrz+1kucMUPaC03FHCZN9BuzLHYaDNSk1ngWpS7yNKc2e+5l90r3OB7SHsq0mx0+aztr k9pNm+0bHF2Oh+1PWQ5Yn7LbM+y32D9mkjNnuWu9a6tLclEooXHdRMIX1Yxl7SX3kePkFNyHy2WD KxlbYwBLz3WahQbn+LG/XFs0SOF6KTW4zVGDawidzaVNMzkZvSiQnntEpUG1SmWqkxOpVk6kCgNU J/lLX/BGhVpAxEn1aNo4ejjgQzU/xYY2no4ObRR7hzq4y4u0phP44yrQQZs6YtTDpU/cpSlTIGyP GuGiTwpZquzJOvmrd0e+3PjZriffC+73bV228/GHbm6/nW73/PoIzaLWX1K2bf/9/jVrn3/z7efg ZhmphxfKhszSSRZdZHiCJJDOlkhNpibLElurtMa03tJqM6dzm+MbdAMxFnIsK8DT/JR3TGfTzmTK k1IqfJMC1SlzMqsDC1IafQsDLSnrMlsC1yrXpp9hZ7waPle5HB7P/IzmjA344Tbg2qvdpzFNk/0B q0oO4XzD7oXBU26OGueaBk7fmQpt8RgO2DhWwJYA+VwsBchfhZUDGTQs+QWlcQd1ZAZR6suLlPLc qOZGHaTBjBItVzVyC0q5YOapkqoLRRP2q2IjULqksAJclmoG35rqyy4t+46jb4rOGT4xV+uIRs8A otE5Q5DRMOzxRNWQcPnDHZUw1vLyFJgrbRIGSzs2jolLIyXFxJ2mhjK4OdJQRJisdPmhws8PfjZy kqa993t80Tn3qbV3+8rdw++yBfapS3dteYwu9TzYT4NUwueTcSMfjHyt6fsPraZ33lKz+mFIbyeO v0qcfPyMfN243DKF72ueZa/lPkvcMmg5ZjllUYklaNlg6bLsG606bklYrEEL9FiVmWRRpBspUUyK bFXUPBOR98n3yXF5UD4uK4PyKZkRWZePoiTLA4mvhICAnDU83Hhl4ZFlK59VTuPqjjZ+BghkRMgJ NecMK+erPNc8a/6o+nMORps6NuK85Kdm1ZBgHWcfZ97Gjmjq5JJ0yV3i3tnf3y//7ciRb9LlyDfv 8tMee5a+wp5trMXwc08MWSlLlWUWyeX4X6YzimSx80NIgQ+HfrAl1jHEMoZI/NjUeMcl0jVWlqLo qaFSMyLZvpT8UlCd6keeYhIVIVFh3IwaRZZNslJmmSWb8pQJ1gbrNdJm67vSx4r6sELDSkTNM5cr Uy1VjnmOmBxTGtSY5Qb5OtPdlpeUN+S3lRPKZ+qXytfm9BSr1SRJMlMU1WIxo2Axm/NUJU1VFUmW 80zWNJPJaoVgZDOiF9mkqGazzUasMn5ENCwm8BxneY6Zl0K68DTiwFEz98IAbHmE5VG6l9AqfH9g PIIxJnFZEWyYp5wlREiMpHCJEeGacI8GO4jP7vgwNOsK/O41d+yQauqonKMNaUNQ/TPRpjk4vKD4 lRpeiK3S7SnfYbogKt+gvYDcG3UCUTVzpblSEmmPIs4BR72FBi03S8zidbhLCSQfg5xxLTGslsKs cos5K6sSAvugN6sc2Vu9ush6QuVCT2JNpKOJduBrvvhVQEkM9obKIcTB3gyefdCrcXKeiZJdZD22 ZOdojGIQPlXK+zI1p2VgtrS0SpGg15leL+/89x5/kpw2xZJnbRTONloCDaQlFN+cVWgiffyzkXZ6 +IOR+7eaDp17hsZHrh5exYLfG7mM6+VNSMqELX58wCQMERo02Fc2tVTkpZOT+cRJyTwnT+RGXrqn 1GUK4gPrMZM8D8kpkxQ0bTB1mRIm/J5MrEzKoxCjGInnRnrJ5NJ9hA7iyGKE6OQoTi8ZLUnrBHLW yBISF9Y5KmthncTMTRMUSdMEkhgLb0dtlMyVz7dRbqRRBCrcTLlp8hJ/OGdu6jcdOjuTnyK1OEXy cYo4iI+uOZDu5cqUyqNUjri4uXVyzCcaUlSrzz5LmW1eqsTMVyptZnOpVpFSkTHZW6fVp9Rn1Hkb TY2WhVpTSlPGQu860zrLKm1dyrqMVd5raLpFMTkukxabFlsvs6+VWk2t1rV2qycgq+6AzZaW6ze4 lvtxPsOnwxcZbh6vqZgXaTK84PYtTmaO8FWB7hTuAaOIOAE4m1Nz80onqpSomqrjyJh0zE/9vP4i fqgAd+YSu5PfHlKEYYkIhmARqBGHCRHBIIEzAssz+LDEwJBBUgWGTcrkhwtuDUluIh3C0dJ0pumf FdzOEACB6R3Qf67AlkWmRZYVphUWmWsp75mqlSEKIOlpIiZIFaHe5FIeEdQ+tOvFP9GM6/9267GR oYO9O27p7du+oxcfdPNvv3rkw+HX//Z9mk0dr7362n+9+OorWNCOkTY5BAmmkGy6wrjdrk3QLtTq NblKj+ssqI+3h7OK04uzZmRt0Pfq5gpPhf9iz8X+mPkye6On0d9uXmNv09Z51vgH9TfT3ve+n/lm 9om0E9nH9YSeEZajWjR9slyhzZQv1pZpn9j+ljWi2dxOhAEBRaVKRsBpI05f7lEr1ayGtdnaZZV1 IUJdiBMe/M+GjQvSKgSJ8tkx1/65kKXw8VyEQD41wpzZ1k00tYSVpOQRMgh/SO+jcXqKykFahd+A JAQZ55JWQoVfpMIvUqEhVNz9QHFGnGSCNIMPTsUFikLewH3BWWVe+p2LYPI8m6MNnz6hDY+JFWEe DGcIf24RGJAmbkKkI9Vdkj6FB3D4jMsQz0Xy3dJ3pLfjoYo7Vu882r752PXL9lzgfvjqa594ZFNn z0ib6dnuBQt2J37y4Mg3t15SMfyN9NDrL7z6+1df+QP3QNthii9Bhm7ysjGtKJVqMg3LpXINfmK6 Qt4kKxa32WK2OFLdFgeRzNQmmE+slnF7zdSco6fSVJbjTjobQyuZUnoKY/47//KV4f6Of1GEyvM7 16hzOW0IbhFF6HzyYJmbMmss8hWsgW+tHEZce3ojvzVz7pTjTwQBRHt5h/OGF3jAu5E2wc8IRiHQ 5VccJd29/YHpbVWXXT59xoxpl6dly5H7O2ZXPJI/q6p54/BbnAtVuIX0gAsTJY9xvZyTllNhudhS m7s0pzVni+V2y825D6c+Ufic5LB4Mr2eifWFb3tMfraEMa2YWr2N5kZLo7XR1mhvdLSb2y3t1nZb u73d0R/pz3flR3Lzc8dPyV1mjdlWRVaN2xTelNuV+0PrT+13jPtx4Z0TH7I+Zn8w/yH8X+eLkQxc 0k8L95IzhoTHkNwxRNBwLgkajggajggajmThZDNSssuXmfPz7FY5U4+ky7YLsjL5JSbHV8iZH/RV +eb5lvv2+474FJcv6FvvO+aTg749PuZ7Fu4oHXoh4msDpwBDWG1QpuFXQkaoRhmPt/vSMkp5bmhO dymlFzRmrc1iWYF0FdEbv2QjvIMNjgV2fzZSuTnIgQtswUyameszUr2lxbx7ETdSnzeZcrfog+ND iu8t8P467+UTwYpPxNg+XNB71dwCdH0qUH60gAL7s7BsIJ+KbzMC4XwA8tcD3PIKMsVUIUT8zcWD xayquKuYFfO7Qi4R7oFHOnC2epLLbIlA+AI4Yvj4IvRclzB1l1ieS+dLxTF11sASgTn5hC4RWrpy jo2FUr5JoxeCpo45o9HREJRZg+/eOHf0ch+NduBmMBY78RY4cRBVDXWIyz0/P+EeRIZoF3+jF/ya 6wwjf0J22JRWGHFrKVqqJik5Dt1PLONUPzVNQJKdhmLIGfaTnLDDbh5v9dNx+RarEpX9JKhl+SmO BB6XJRMe9UQLotu2bRv9LwgU+Q1yY0dTapm4hPDPBvn4fbUUXweEK/r2MunBNwRPNj4d8GMlUtXr 2nX9lmsn5/3wpbvnVU8t+MGiG55d5o7bO9u2tGdkFPlvPvzjpW0v3XDkHXphYM3G1toLw9684ou2 zZ113bhgdPb1V3oXNi4sCweyUq25JdVbGpftu/SX3E5zE1+wAtPd+H+GPx4kVuhgOMJjbdzagHT5 cD+xO6xUIhmaJeqy4pCQbC4th+RQR0qenSZUc52lrlndoHape1WZ4Iy+T42rg+pRVcGB/jl0VUQA p4UWoeYL8dkHyFewStH0ldA01JwVig3kFD9l+HWCyx/lZPyiHmLtxEun9CAu/vaTDmcmHP0QIiPt xGlER0P82x938u6SEu1lHipFo3kezr/IZHd4com7DJ4s7E7jrGda5iWVK9YW3nxz31NPpUbHZd+/ T5ve+gBbuZuqa0du2z38wzmFmTyuEk8iH7+h/7vHiUoJ/x1RS2aS2fhPjLmI+Bfi98ol+C/2S0UH ivMcPhyPgv+FIHMvWjD30nnR6o1tLWvnLP7fAvlbiQplbmRzdHJlYW0KZW5kb2JqCjc1IDAgb2Jq Cjg2MDYKZW5kb2JqCjc2IDAgb2JqCihRdWFydHogJWQpCmVuZG9iago3NyAwIG9iagooTWFjIE9T IFggMTAuOSBRdWFydHogUERGQ29udGV4dCkKZW5kb2JqCjc4IDAgb2JqCihRdWFydHogUiBEZXZp Y2UpCmVuZG9iago3OSAwIG9iagooRDoyMDEzMTEwNTE1NTIzM1owMCcwMCcpCmVuZG9iagoxIDAg b2JqCjw8IC9UaXRsZSA3NiAwIFIgL1Byb2R1Y2VyIDc3IDAgUiAvQ3JlYXRvciA3OCAwIFIgL0Ny ZWF0aW9uRGF0ZSA3OSAwIFIgL01vZERhdGUKNzkgMCBSID4+CmVuZG9iagp4cmVmCjAgODAKMDAw MDAwMDAwMCA2NTUzNSBmIAowMDAwNDI1NjcyIDAwMDAwIG4gCjAwMDAzMTUzOTkgMDAwMDAgbiAK MDAwMDMxODYzMiAwMDAwMCBuIAowMDAwMDAwMDIyIDAwMDAwIG4gCjAwMDAzMTUzNzcgMDAwMDAg biAKMDAwMDMxNTUwMyAwMDAwMCBuIAowMDAwMzE4NTk2IDAwMDAwIG4gCjAwMDAzODIyNzkgMDAw MDAgbiAKMDAwMDM0MjQzMyAwMDAwMCBuIAowMDAwNDA0ODQ2IDAwMDAwIG4gCjAwMDAzMzAzMjUg MDAwMDAgbiAKMDAwMDM5OTAyNSAwMDAwMCBuIAowMDAwMzM2MjUzIDAwMDAwIG4gCjAwMDAzNDgx OTEgMDAwMDAgbiAKMDAwMDM1Mzk3MCAwMDAwMCBuIAowMDAwMzkzMjA0IDAwMDAwIG4gCjAwMDAz MjQ2MzUgMDAwMDAgbiAKMDAwMDQxMDM5NiAwMDAwMCBuIAowMDAwMzE4Nzc5IDAwMDAwIG4gCjAw MDAzODc2NTMgMDAwMDAgbiAKMDAwMDM1OTM0NSAwMDAwMCBuIAowMDAwNDE2MTc0IDAwMDAwIG4g CjAwMDAzNzE3OTUgMDAwMDAgbiAKMDAwMDMxNTgxNSAwMDAwMCBuIAowMDAwMzE1ODYwIDAwMDAw IG4gCjAwMDAzMTg1NzUgMDAwMDAgbiAKMDAwMDMxODcxNSAwMDAwMCBuIAowMDAwMzE4OTcwIDAw MDAwIG4gCjAwMDAzMTkyMzAgMDAwMDAgbiAKMDAwMDMyNDYxNCAwMDAwMCBuIAowMDAwMzI0ODE4 IDAwMDAwIG4gCjAwMDAzMjUwNzggMDAwMDAgbiAKMDAwMDMzMDMwNCAwMDAwMCBuIAowMDAwMzMw NTEyIDAwMDAwIG4gCjAwMDAzMzA3NzIgMDAwMDAgbiAKMDAwMDMzNjIzMiAwMDAwMCBuIAowMDAw MzM2NDQwIDAwMDAwIG4gCjAwMDAzMzY3MDAgMDAwMDAgbiAKMDAwMDM0MjQxMiAwMDAwMCBuIAow MDAwMzQyNjE3IDAwMDAwIG4gCjAwMDAzNDI4NzcgMDAwMDAgbiAKMDAwMDM0ODE3MCAwMDAwMCBu IAowMDAwMzQ4MzcyIDAwMDAwIG4gCjAwMDAzNDg2MzIgMDAwMDAgbiAKMDAwMDM1Mzk0OSAwMDAw MCBuIAowMDAwMzU0MTQzIDAwMDAwIG4gCjAwMDAzNTQ0MDMgMDAwMDAgbiAKMDAwMDM1OTMyNCAw MDAwMCBuIAowMDAwMzU5NzQ4IDAwMDAwIG4gCjAwMDAzNjAwMTUgMDAwMDAgbiAKMDAwMDM3MTc3 MyAwMDAwMCBuIAowMDAwMzcyMTcxIDAwMDAwIG4gCjAwMDAzNzI0MzEgMDAwMDAgbiAKMDAwMDM4 MjI1OCAwMDAwMCBuIAowMDAwMzgyNDUxIDAwMDAwIG4gCjAwMDAzODI3MTEgMDAwMDAgbiAKMDAw MDM4NzYzMiAwMDAwMCBuIAowMDAwMzg3ODMwIDAwMDAwIG4gCjAwMDAzODgwOTAgMDAwMDAgbiAK MDAwMDM5MzE4MyAwMDAwMCBuIAowMDAwMzkzMzgzIDAwMDAwIG4gCjAwMDAzOTM2NDMgMDAwMDAg biAKMDAwMDM5OTAwNCAwMDAwMCBuIAowMDAwMzk5MjA0IDAwMDAwIG4gCjAwMDAzOTk0NjQgMDAw MDAgbiAKMDAwMDQwNDgyNSAwMDAwMCBuIAowMDAwNDA1MDIzIDAwMDAwIG4gCjAwMDA0MDUyODMg MDAwMDAgbiAKMDAwMDQxMDM3NSAwMDAwMCBuIAowMDAwNDEwNTgzIDAwMDAwIG4gCjAwMDA0MTA4 NDMgMDAwMDAgbiAKMDAwMDQxNjE1MyAwMDAwMCBuIAowMDAwNDE2NTQwIDAwMDAwIG4gCjAwMDA0 MTY4MDAgMDAwMDAgbiAKMDAwMDQyNTQ5NyAwMDAwMCBuIAowMDAwNDI1NTE4IDAwMDAwIG4gCjAw MDA0MjU1NDYgMDAwMDAgbiAKMDAwMDQyNTU5NiAwMDAwMCBuIAowMDAwNDI1NjMwIDAwMDAwIG4g CnRyYWlsZXIKPDwgL1NpemUgODAgL1Jvb3QgMjcgMCBSIC9JbmZvIDEgMCBSIC9JRCBbIDxhNTAy ZmNmNmM2NDM0MTJlNmZjNDY1OTM3MTBkYjRlMj4KPGE1MDJmY2Y2YzY0MzQxMmU2ZmM0NjU5Mzcx MGRiNGUyPiBdID4+CnN0YXJ0eHJlZgo0MjU3NzcKJSVFT0YK --089e013a23c0feb3ed04ea891e18-- From prvs=00237f32c3=anna.brunstrom@kau.se Thu Nov 7 00:14:26 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 65B8B21E80C3 for ; Thu, 7 Nov 2013 00:14:26 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.649 X-Spam-Level: X-Spam-Status: No, score=-2.649 tagged_above=-999 required=5 tests=[AWL=0.600, BAYES_00=-2.599, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id p2Mo4d7gH+QL for ; Thu, 7 Nov 2013 00:14:22 -0800 (PST) Received: from nasse.dc.kau.se (smtp.kau.se [193.10.220.39]) by ietfa.amsl.com (Postfix) with ESMTP id 04CB521E80B5 for ; Thu, 7 Nov 2013 00:14:21 -0800 (PST) X-Spam-Processed: mail.kau.se, Thu, 07 Nov 2013 09:13:52 +0100 (not processed: spam filter heuristic analysis disabled) X-Authenticated-Sender: annabrun@kau.se X-MDRemoteIP: 64.114.24.114 X-Return-Path: anna.brunstrom@kau.se X-Envelope-From: anna.brunstrom@kau.se X-MDaemon-Deliver-To: tcpm@ietf.org Message-ID: <527B4BDF.60708@kau.se> Date: Thu, 07 Nov 2013 09:14:23 +0100 From: Anna Brunstrom User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:15.0) Gecko/20120824 Thunderbird/15.0 MIME-Version: 1.0 To: tcpm@ietf.org References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [tcpm] TCP tail probe and the initial loss behavior X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 08:14:26 -0000 Hi Yuchung, Thanks for posting the data, that is very interesting! If I understand it correctly all percentages are calculated in relation to the total number of connection traces? If so that makes it a bit difficult to compare the numbers. How many of the connections had 10 or more packets? Could you perhaps create an equivalent of the lossdist1 graph that shows the % of losses in position x in relation to the number of connections that had a length of x or more packets? Thanks, Anna On 2013-11-06 22:48, Yuchung Cheng wrote: > We have new data regarding the question raised in the previous two > meetings: "is tail loss probe useful b/c IW10 is causing more (tail) > losses". > > We got some data on TCP initial loss behavior with IW10. There were > 15,269 outbound Google server connection traces taken from all over > the world. Bottleneck bandwidths ranged from 64kbps to 40Mbps with a > median of ~1Mbps. All the connections used IW10. For each of them, we > recorded the ending sequence number, packet number and send time of > the first lost packet (the first retransmitted packet of the > connection) and the repair time (time from initial send until > retransmit acked). > > 980 connections (6.4%) had one or more losses in the IW10 send. > "lossdist1.pdf" is a histogram of % losses by position in initial > window ("none" means no loss in the first 10 packets sent). Claims > that IW10 "increases the loss rate" are not supported by this data: > The average loss rate for the first 10 packets is 0.64%. The average > loss rate for the last 5 packets of the window is only 0.44% compared > to 0.84% for the first 5 packets (in fact, the loss rate for the first > packet is as large as packets 6 to 10 combined). This is easier to > see on a plot of the cumulative loss (lossdist2.pdf): the black line > is a linear fit to the first five points and the blue to the last five > points. The lines slopes show that the loss rate consistently > increases faster for the first 5 packets of the window compared to the > last 5. > > "lossdist3.pdf" shows the distribution of packet vs. cumulative % of > connections with a first loss at or before that packet. The blue line > is a fit of cum% to packet^0.25. The quality of the fit over the > entire range (and the fact that there's no discontinuity between the > portion of the data that lies within the initial window vs the portion > outside it) suggests that the single connection loss probabilities > have nothing to do with IW (at least for IW <= 10). > > We didn't graph loss recovery time since it was simple: 78% of the > initial window losses (761 of 980) were repaired in 1 RTT and the > remaining 22% were repaired in 2 RTT. None of them took a timeout; all > were repaired by Fast Retransmit. (This is in contrast to previous IW4 > measurements where 20% of the losses required a timeout to repair). > > > > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm > From prvs=9023f0d9d8=david.borman@quantum.com Thu Nov 7 12:47:15 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9C66D21E818D for ; Thu, 7 Nov 2013 12:47:15 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.265 X-Spam-Level: X-Spam-Status: No, score=-3.265 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, IP_NOT_FRIENDLY=0.334, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id thl5MxrluegF for ; Thu, 7 Nov 2013 12:47:03 -0800 (PST) Received: from mx0b-000ceb01.pphosted.com (mx0b-000ceb01.pphosted.com [67.231.152.126]) by ietfa.amsl.com (Postfix) with ESMTP id 9B67121E8187 for ; Thu, 7 Nov 2013 12:46:58 -0800 (PST) Received: from pps.filterd (m0001158 [127.0.0.1]) by mx0b-000ceb01.pphosted.com (8.14.5/8.14.5) with SMTP id rA7KhfMa026619 for ; Thu, 7 Nov 2013 12:46:58 -0800 Received: from ppoxedge1.quantum.com ([146.174.252.27]) by mx0b-000ceb01.pphosted.com with ESMTP id 1g0ftj886d-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT) for ; Thu, 07 Nov 2013 12:46:57 -0800 Received: from PPOMSG2.QUANTUM.com (10.50.35.27) by PPOXEDGE1.quantum.com (146.174.252.27) with Microsoft SMTP Server (TLS) id 14.2.318.1; Thu, 7 Nov 2013 13:46:37 -0700 Received: from PPOMSG1.QUANTUM.com ([10.50.35.26]) by ppomsg2 ([10.50.35.27]) with mapi id 14.02.0318.001; Thu, 7 Nov 2013 13:46:55 -0700 From: David Borman To: "tcpm@ietf.org (tcpm@ietf.org)" Thread-Topic: Linux doesn't do ZWP correctly??? Thread-Index: AQHO2/p+1Q14vNmJ20aN/A05aCZZcw== Date: Thu, 7 Nov 2013 20:46:53 +0000 Message-ID: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.65.176.229] Content-ID: <1DDE8F5666410D4694499847C2F8191E@QUANTUM.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.10.8794, 1.0.431, 0.0.0000 definitions=2013-11-07_06:2013-11-07, 2013-11-07, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1305240000 definitions=main-1311070163 Content-Type: text/plain; charset="Windows-1252" Subject: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 20:47:15 -0000 I=92ve been doing some testing of crossing Zero Window Probes, and it appea= rs that the Linux TCP doesn=92t do ZWP properly. It is missing the 1 byte = of data. My test program creates a pair of sockets, sets SO_RCVBUF to 20K and SO_SND= BUF to 80K on both sides, then writes 40K of data in both directions. It t= hen pauses for 20 seconds to allow some ZWP packets, reads 1 byte of data f= rom both sides, pauses another 20 seconds and then reads all the data from = both sockets. I ran tcpdump to watch the packet traffic. Below are packet= traces from both a Fedora Core 19 and a FreeBSD 9.2 system. The FreeBSD t= race has what I would expect to see, the Linux trace doesn=92t send any dat= a on the probes. Any thoughts on this? -David Borman # uname -a Linux dab-fc17-vm.mdh.quantum.com 3.11.6-200.fc19.x86_64 #1 SMP Fri Oct 18 = 22:34:18 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux # tcpdump -i lo tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on lo, link-type EN10MB (Ethernet), capture size 65535 bytes 14:52:24.676876 IP localhost.56771 > localhost.57737: Flags [S], seq 109576= 2115, win 20480, options [mss 65495,sackOK,TS val 106278 ecr 0,nop,wscale 0= ], length 0 14:52:24.676905 IP localhost.57737 > localhost.56771: Flags [S.], seq 89456= 6307, ack 1095762116, win 20480, options [mss 65495,sackOK,TS val 106278 ec= r 106278,nop,wscale 0], length 0 14:52:24.676920 IP localhost.56771 > localhost.57737: Flags [.], ack 1, win= 20480, options [nop,nop,TS val 106278 ecr 106278], length 0 14:52:24.677292 IP localhost.56771 > localhost.57737: Flags [.], seq 1:1024= 1, ack 1, win 20480, options [nop,nop,TS val 106278 ecr 106278], length 102= 40 14:52:24.677303 IP localhost.57737 > localhost.56771: Flags [.], ack 10241,= win 10240, options [nop,nop,TS val 106278 ecr 106278], length 0 14:52:24.677316 IP localhost.56771 > localhost.57737: Flags [P.], seq 10241= :20481, ack 1, win 20480, options [nop,nop,TS val 106278 ecr 106278], lengt= h 10240 14:52:24.678314 IP localhost.57737 > localhost.56771: Flags [.], seq 1:1024= 1, ack 20481, win 0, options [nop,nop,TS val 106279 ecr 106278], length 102= 40 14:52:24.678340 IP localhost.56771 > localhost.57737: Flags [.], ack 10241,= win 10240, options [nop,nop,TS val 106279 ecr 106279], length 0 14:52:24.678360 IP localhost.57737 > localhost.56771: Flags [P.], seq 10241= :20481, ack 20481, win 0, options [nop,nop,TS val 106279 ecr 106278], lengt= h 10240 14:52:24.717804 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106319 ecr 106279], length 0 14:52:24.879723 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106481 ecr 106279], length 0 14:52:24.879833 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106481 ecr 106319], length 0 14:52:24.923707 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106525 ecr 106319], length 0 14:52:24.923820 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106525 ecr 106481], length 0 14:52:25.283492 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106884 ecr 106481], length 0 14:52:25.283599 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106885 ecr 106525], length 0 14:52:25.335722 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106937 ecr 106525], length 0 14:52:25.335839 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 106937 ecr 106885], length 0 14:52:26.090812 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 107692 ecr 106885], length 0 14:52:26.090919 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 107692 ecr 106937], length 0 14:52:26.159642 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 107761 ecr 106937], length 0 14:52:26.159686 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 107761 ecr 107692], length 0 14:52:27.703494 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 109304 ecr 107692], length 0 14:52:27.703541 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 109304 ecr 107761], length 0 14:52:27.803683 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 109405 ecr 107761], length 0 14:52:27.803792 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 109405 ecr 109304], length 0 14:52:30.927667 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 112529 ecr 109304], length 0 14:52:30.927712 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 112529 ecr 109405], length 0 14:52:31.087676 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 112689 ecr 109405], length 0 14:52:31.087788 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 112689 ecr 112529], length 0 14:52:37.375727 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 118977 ecr 112529], length 0 14:52:37.375832 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 118977 ecr 112689], length 0 14:52:37.664034 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 119265 ecr 112689], length 0 14:52:37.664358 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 119265 ecr 118977], length 0 14:52:50.271699 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 131873 ecr 118977], length 0 14:52:50.271807 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 131873 ecr 119265], length 0 14:52:50.815486 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 132416 ecr 119265], length 0 14:52:50.815600 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 0, options [nop,nop,TS val 132416 ecr 131873], length 0 14:53:04.679770 IP localhost.57737 > localhost.56771: Flags [.], ack 20481,= win 20480, options [nop,nop,TS val 146281 ecr 132416], length 0 14:53:04.679790 IP localhost.56771 > localhost.57737: Flags [.], seq 20481:= 30721, ack 20481, win 0, options [nop,nop,TS val 146281 ecr 146281], length= 10240 14:53:04.679795 IP localhost.56771 > localhost.57737: Flags [P.], seq 30721= :40961, ack 20481, win 0, options [nop,nop,TS val 146281 ecr 146281], lengt= h 10240 14:53:04.679908 IP localhost.56771 > localhost.57737: Flags [.], ack 20481,= win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 14:53:04.679916 IP localhost.57737 > localhost.56771: Flags [.], seq 20481:= 30721, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], length= 10240 14:53:04.679926 IP localhost.57737 > localhost.56771: Flags [P.], seq 30721= :40961, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], lengt= h 10240 14:53:04.679941 IP localhost.56771 > localhost.57737: Flags [.], ack 30721,= win 10240, options [nop,nop,TS val 146281 ecr 146281], length 0 14:53:04.679966 IP localhost.57737 > localhost.56771: Flags [.], ack 40961,= win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 14:53:04.679974 IP localhost.56771 > localhost.57737: Flags [.], seq 40961:= 51201, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], length= 10240 14:53:04.679978 IP localhost.56771 > localhost.57737: Flags [P.], seq 51201= :61441, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], lengt= h 10240 14:53:04.680002 IP localhost.56771 > localhost.57737: Flags [.], ack 40961,= win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 14:53:04.680013 IP localhost.57737 > localhost.56771: Flags [.], seq 40961:= 51201, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], length= 10240 14:53:04.680018 IP localhost.57737 > localhost.56771: Flags [P.], seq 51201= :61441, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], lengt= h 10240 14:53:04.680042 IP localhost.57737 > localhost.56771: Flags [.], ack 61441,= win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 14:53:04.680049 IP localhost.56771 > localhost.57737: Flags [.], seq 61441:= 71681, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], length= 10240 The trace from a FreeBSD 9.2 system looks like this: # uname -a FreeBSD fbsd90-64 9.2-RELEASE FreeBSD 9.2-RELEASE #0 r255898: Thu Sep 26 22= :50:31 UTC 2013 root@bake.isc.freebsd.org:/usr/obj/usr/src/sys/GENERIC = amd64 # tcpdump -i lo0 tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on lo0, link-type NULL (BSD loopback), capture size 65535 bytes 13:44:43.912939 IP localhost.22111 > localhost.49248: Flags [S], seq 158992= 5916, win 20480, options [mss 16344,nop,wscale 6,sackOK,TS val 120482 ecr 0= ], length 0 13:44:43.912987 IP localhost.49248 > localhost.22111: Flags [S.], seq 17524= 03575, ack 1589925917, win 20480, options [mss 16344,nop,wscale 6,sackOK,TS= val 2357200578 ecr 120482], length 0 13:44:43.912999 IP localhost.22111 > localhost.49248: Flags [.], ack 1, win= 510, options [nop,nop,TS val 120482 ecr 2357200578], length 0 13:44:43.913456 IP localhost.22111 > localhost.49248: Flags [.], seq 1:1633= 3, ack 1, win 510, options [nop,nop,TS val 120482 ecr 2357200578], length 1= 6332 13:44:43.914272 IP localhost.49248 > localhost.22111: Flags [.], seq 1:1633= 3, ack 16333, win 255, options [nop,nop,TS val 2357200578 ecr 120482], leng= th 16332 13:44:43.914377 IP localhost.22111 > localhost.49248: Flags [.], seq 16333:= 32653, ack 16333, win 255, options [nop,nop,TS val 120482 ecr 2357200578], = length 16320 13:44:43.914460 IP localhost.49248 > localhost.22111: Flags [.], seq 16333:= 32653, ack 32653, win 0, options [nop,nop,TS val 2357200578 ecr 120482], le= ngth 16320 13:44:44.012989 IP localhost.22111 > localhost.49248: Flags [.], ack 32653,= win 0, options [nop,nop,TS val 120583 ecr 2357200578], length 0 13:44:48.912031 IP localhost.22111 > localhost.49248: Flags [.], seq 32653:= 32654, ack 32653, win 0, options [nop,nop,TS val 125482 ecr 2357200578], le= ngth 1 13:44:48.912056 IP localhost.49248 > localhost.22111: Flags [.], ack 32654,= win 0, options [nop,nop,TS val 2357205578 ecr 125482], length 0 13:44:49.012049 IP localhost.49248 > localhost.22111: Flags [.], seq 32653:= 32654, ack 32654, win 0, options [nop,nop,TS val 2357205678 ecr 125482], le= ngth 1 13:44:49.012074 IP localhost.22111 > localhost.49248: Flags [.], ack 32654,= win 0, options [nop,nop,TS val 125582 ecr 2357205678], length 0 13:44:53.913015 IP localhost.22111 > localhost.49248: Flags [.], seq 32654:= 32655, ack 32654, win 0, options [nop,nop,TS val 130483 ecr 2357205678], le= ngth 1 13:44:53.913059 IP localhost.49248 > localhost.22111: Flags [.], ack 32655,= win 0, options [nop,nop,TS val 2357210579 ecr 130483], length 0 13:44:54.013017 IP localhost.49248 > localhost.22111: Flags [.], seq 32654:= 32655, ack 32655, win 0, options [nop,nop,TS val 2357210679 ecr 130483], le= ngth 1 13:44:54.013042 IP localhost.22111 > localhost.49248: Flags [.], ack 32655,= win 0, options [nop,nop,TS val 130583 ecr 2357210679], length 0 13:44:58.913034 IP localhost.22111 > localhost.49248: Flags [.], seq 32655:= 32656, ack 32655, win 0, options [nop,nop,TS val 135483 ecr 2357210679], le= ngth 1 13:44:58.913060 IP localhost.49248 > localhost.22111: Flags [.], ack 32656,= win 0, options [nop,nop,TS val 2357215579 ecr 135483], length 0 13:44:59.012397 IP localhost.49248 > localhost.22111: Flags [.], seq 32655:= 32656, ack 32656, win 0, options [nop,nop,TS val 2357215679 ecr 135483], le= ngth 1 13:44:59.012421 IP localhost.22111 > localhost.49248: Flags [.], ack 32656,= win 0, options [nop,nop,TS val 135583 ecr 2357215679], length 0 13:45:03.913033 IP localhost.22111 > localhost.49248: Flags [.], seq 32656:= 32657, ack 32656, win 0, options [nop,nop,TS val 140483 ecr 2357215679], le= ngth 1 13:45:03.913063 IP localhost.49248 > localhost.22111: Flags [.], ack 32657,= win 0, options [nop,nop,TS val 2357220579 ecr 140483], length 0 13:45:04.013019 IP localhost.49248 > localhost.22111: Flags [.], seq 32656:= 32657, ack 32657, win 0, options [nop,nop,TS val 2357220679 ecr 140483], le= ngth 1 13:45:04.013044 IP localhost.22111 > localhost.49248: Flags [.], ack 32657,= win 0, options [nop,nop,TS val 140583 ecr 2357220679], length 0 13:45:08.913018 IP localhost.22111 > localhost.49248: Flags [.], seq 32657:= 32658, ack 32657, win 0, options [nop,nop,TS val 145483 ecr 2357220679], le= ngth 1 13:45:08.913043 IP localhost.49248 > localhost.22111: Flags [.], ack 32658,= win 0, options [nop,nop,TS val 2357225579 ecr 145483], length 0 13:45:09.013017 IP localhost.49248 > localhost.22111: Flags [.], seq 32657:= 32658, ack 32658, win 0, options [nop,nop,TS val 2357225679 ecr 145483], le= ngth 1 13:45:09.013042 IP localhost.22111 > localhost.49248: Flags [.], ack 32658,= win 0, options [nop,nop,TS val 145583 ecr 2357225679], length 0 13:45:13.913009 IP localhost.22111 > localhost.49248: Flags [.], seq 32658:= 32659, ack 32658, win 0, options [nop,nop,TS val 150483 ecr 2357225679], le= ngth 1 13:45:13.913034 IP localhost.49248 > localhost.22111: Flags [.], ack 32659,= win 0, options [nop,nop,TS val 2357230579 ecr 150483], length 0 13:45:14.012980 IP localhost.49248 > localhost.22111: Flags [.], seq 32658:= 32659, ack 32659, win 0, options [nop,nop,TS val 2357230679 ecr 150483], le= ngth 1 13:45:14.013005 IP localhost.22111 > localhost.49248: Flags [.], ack 32659,= win 0, options [nop,nop,TS val 150583 ecr 2357230679], length 0 13:45:18.912690 IP localhost.22111 > localhost.49248: Flags [.], seq 32659:= 32660, ack 32659, win 0, options [nop,nop,TS val 155483 ecr 2357230679], le= ngth 1 13:45:18.912717 IP localhost.49248 > localhost.22111: Flags [.], ack 32660,= win 0, options [nop,nop,TS val 2357235579 ecr 155483], length 0 13:45:19.012811 IP localhost.49248 > localhost.22111: Flags [.], seq 32659:= 32660, ack 32660, win 0, options [nop,nop,TS val 2357235679 ecr 155483], le= ngth 1 13:45:19.012828 IP localhost.22111 > localhost.49248: Flags [.], ack 32660,= win 0, options [nop,nop,TS val 155583 ecr 2357235679], length 0 13:45:23.912288 IP localhost.22111 > localhost.49248: Flags [.], seq 32660:= 32661, ack 32660, win 0, options [nop,nop,TS val 160482 ecr 2357235679], le= ngth 1 13:45:23.912316 IP localhost.49248 > localhost.22111: Flags [.], ack 32661,= win 0, options [nop,nop,TS val 2357240578 ecr 160482], length 0 13:45:24.013018 IP localhost.49248 > localhost.22111: Flags [.], seq 32660:= 32661, ack 32661, win 510, options [nop,nop,TS val 2357240679 ecr 160482], = length 1 13:45:24.013046 IP localhost.22111 > localhost.49248: Flags [.], seq 32661:= 48993, ack 32661, win 510, options [nop,nop,TS val 160583 ecr 2357240679], = length 16332 13:45:24.013195 IP localhost.49248 > localhost.22111: Flags [.], seq 32661:= 48993, ack 48993, win 255, options [nop,nop,TS val 2357240679 ecr 160583], = length 16332 13:45:24.013207 IP localhost.22111 > localhost.49248: Flags [.], seq 48993:= 65313, ack 48993, win 255, options [nop,nop,TS val 160583 ecr 2357240679], = length 16320 13:45:24.013215 IP localhost.49248 > localhost.22111: Flags [.], seq 48993:= 65313, ack 65313, win 0, options [nop,nop,TS val 2357240679 ecr 160583], le= ngth 16320 13:45:24.112038 IP localhost.22111 > localhost.49248: Flags [.], ack 65313,= win 510, options [nop,nop,TS val 160682 ecr 2357240679], length 0 13:45:24.112075 IP localhost.49248 > localhost.22111: Flags [.], seq 65313:= 81645, ack 65313, win 510, options [nop,nop,TS val 2357240778 ecr 160682], = length 16332 13:45:24.112333 IP localhost.49248 > localhost.22111: Flags [P.], seq 81645= :81921, ack 65313, win 510, options [nop,nop,TS val 2357240778 ecr 160682],= length 276 13:45:24.112354 IP localhost.22111 > localhost.49248: Flags [.], seq 65313:= 81645, ack 81645, win 255, options [nop,nop,TS val 160682 ecr 2357240778], = length 16332 13:45:24.112505 IP localhost.22111 > localhost.49248: Flags [P.], seq 81645= :81921, ack 81645, win 255, options [nop,nop,TS val 160682 ecr 2357240778],= length 276 13:45:24.112520 IP localhost.49248 > localhost.22111: Flags [.], ack 81921,= win 250, options [nop,nop,TS val 2357240778 ecr 160682], length 0 13:45:24.112596 IP localhost.49248 > localhost.22111: Flags [F.], seq 81921= , ack 81921, win 510, options [nop,nop,TS val 2357240778 ecr 160682], lengt= h 0 13:45:24.112606 IP localhost.22111 > localhost.49248: Flags [.], ack 81922,= win 510, options [nop,nop,TS val 160682 ecr 2357240778], length 0 13:45:24.112624 IP localhost.22111 > localhost.49248: Flags [F.], seq 81921= , ack 81922, win 510, options [nop,nop,TS val 160682 ecr 2357240778], lengt= h 0 13:45:24.112646 IP localhost.49248 > localhost.22111: Flags [.], ack 81922,= win 510, options [nop,nop,TS val 2357240778 ecr 160682], length 0 ---------------------------------------------------------------------- The information contained in this transmission may be confidential. Any dis= closure, copying, or further distribution of confidential information is no= t permitted unless such privilege is explicitly granted in writing by Quant= um. Quantum reserves the right to have electronic communications, including= email and attachments, sent across its networks filtered through anti viru= s and spam software programs and retain such messages in order to comply wi= th applicable data security and retention requirements. Quantum is not resp= onsible for the proper and complete transmission of the substance of this c= ommunication or for any delay in its receipt. From hagen@jauu.net Thu Nov 7 14:13:36 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 59A6711E815F for ; Thu, 7 Nov 2013 14:13:36 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.6 X-Spam-Level: X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IXhvY8+0ycSP for ; Thu, 7 Nov 2013 14:13:36 -0800 (PST) Received: from Chamillionaire.breakpoint.cc (Chamillionaire.breakpoint.cc [IPv6:2001:4d88:1ffa:82:880:aa0:9009:64ae]) by ietfa.amsl.com (Postfix) with ESMTP id D211711E8127 for ; Thu, 7 Nov 2013 14:13:35 -0800 (PST) Received: from pfeifer by Chamillionaire.breakpoint.cc with local (Exim 4.80) (envelope-from ) id 1VeXpU-0000nS-71; Thu, 07 Nov 2013 23:13:32 +0100 Date: Thu, 7 Nov 2013 23:18:02 +0100 From: Hagen Paul Pfeifer To: David Borman Message-ID: <20131107221802.GA2201@localhost.localdomain> References: MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: X-Key-Id: 98350C22 X-Key-Fingerprint: 490F 557B 6C48 6D7E 5706 2EA2 4A22 8D45 9835 0C22 X-GPG-Key: gpg --recv-keys --keyserver wwwkeys.eu.pgp.net 98350C22 User-Agent: Mutt/1.5.21 (2010-09-15) Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 22:13:36 -0000 * David Borman | 2013-11-07 20:46:53 [+0000]: >I’ve been doing some testing of crossing Zero Window Probes, and it appears >that the Linux TCP doesn’t do ZWP properly. It is missing the 1 byte of >data. > >My test program creates a pair of sockets, sets SO_RCVBUF to 20K and >SO_SNDBUF to 80K on both sides, then writes 40K of data in both directions. >It then pauses for 20 seconds to allow some ZWP packets, reads 1 byte of data >from both sides, pauses another 20 seconds and then reads all the data from >both sockets. I ran tcpdump to watch the packet traffic. Below are packet >traces from both a Fedora Core 19 and a FreeBSD 9.2 system. The FreeBSD >trace has what I would expect to see, the Linux trace doesn’t send any data >on the probes. > >Any thoughts on this? Why 1 byte of data, the probe itself is not required to carry payload. The sender must accept one byte probe packets, but there is no requirement to probe with 1 byte packets. Hagen From prvs=9023f0d9d8=david.borman@quantum.com Thu Nov 7 14:30:31 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5E95711E815A for ; Thu, 7 Nov 2013 14:30:31 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.265 X-Spam-Level: X-Spam-Status: No, score=-3.265 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, IP_NOT_FRIENDLY=0.334, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Un2lN5yP-PPt for ; Thu, 7 Nov 2013 14:30:25 -0800 (PST) Received: from mx0b-000ceb01.pphosted.com (mx0b-000ceb01.pphosted.com [67.231.152.126]) by ietfa.amsl.com (Postfix) with ESMTP id 61A5311E8127 for ; Thu, 7 Nov 2013 14:30:25 -0800 (PST) Received: from pps.filterd (m0001151 [127.0.0.1]) by mx0b-000ceb01.pphosted.com (8.14.5/8.14.5) with SMTP id rA7MFtbV008176; Thu, 7 Nov 2013 14:30:24 -0800 Received: from ppoxedge1.quantum.com ([146.174.252.27]) by mx0b-000ceb01.pphosted.com with ESMTP id 1g0drvrrj7-1 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Thu, 07 Nov 2013 14:30:24 -0800 Received: from PPOMSG2.QUANTUM.com (10.50.35.27) by PPOXEDGE1.quantum.com (146.174.252.27) with Microsoft SMTP Server (TLS) id 14.2.318.1; Thu, 7 Nov 2013 15:30:04 -0700 Received: from PPOMSG1.QUANTUM.com ([10.50.35.26]) by ppomsg2 ([10.50.35.27]) with mapi id 14.02.0318.001; Thu, 7 Nov 2013 15:30:23 -0700 From: David Borman To: Hagen Paul Pfeifer Thread-Topic: [tcpm] Linux doesn't do ZWP correctly??? Thread-Index: AQHO2/p+1Q14vNmJ20aN/A05aCZZc5oay9UAgAADcYA= Date: Thu, 7 Nov 2013 22:30:21 +0000 Message-ID: References: <20131107221802.GA2201@localhost.localdomain> In-Reply-To: <20131107221802.GA2201@localhost.localdomain> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.65.176.229] Content-ID: <5AAA4C93AEC545479DC3DE905285D6A8@QUANTUM.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.10.8794, 1.0.431, 0.0.0000 definitions=2013-11-07_06:2013-11-07, 2013-11-07, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1305240000 definitions=main-1311070184 Content-Type: text/plain; charset="Windows-1252" Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 22:30:31 -0000 On Nov 7, 2013, at 4:18 PM, Hagen Paul Pfeifer wrote: > * David Borman | 2013-11-07 20:46:53 [+0000]: >=20 >> I=92ve been doing some testing of crossing Zero Window Probes, and it ap= pears >> that the Linux TCP doesn=92t do ZWP properly. It is missing the 1 byte = of >> data. >>=20 >> My test program creates a pair of sockets, sets SO_RCVBUF to 20K and >> SO_SNDBUF to 80K on both sides, then writes 40K of data in both directio= ns. >> It then pauses for 20 seconds to allow some ZWP packets, reads 1 byte of= data >> from both sides, pauses another 20 seconds and then reads all the data f= rom >> both sockets. I ran tcpdump to watch the packet traffic. Below are pac= ket >> traces from both a Fedora Core 19 and a FreeBSD 9.2 system. The FreeBSD >> trace has what I would expect to see, the Linux trace doesn=92t send any= data >> on the probes. >>=20 >> Any thoughts on this? >=20 > Why 1 byte of data, the probe itself is not required to carry payload. The > sender must accept one byte probe packets, but there is no requirement to > probe with 1 byte packets. Ugh, RFC 793 and RFC 1122 aren=92t explicit on this, but in RFC 793, page 4= 3: If a segment containing a single data octet sent to probe a zero window= ... Without at least one byte of data, it just looks like a regular ACK packet = and it won=92t be guaranteed to elicit a response, which is the whole point= of the Zero Window Probe, to elicit a response. Also, the sender isn=92t required to accept the 1 byte probe if the window = is truly zero, but it is required to send back an ACK. -David Borman > Hagen ---------------------------------------------------------------------- The information contained in this transmission may be confidential. Any dis= closure, copying, or further distribution of confidential information is no= t permitted unless such privilege is explicitly granted in writing by Quant= um. Quantum reserves the right to have electronic communications, including= email and attachments, sent across its networks filtered through anti viru= s and spam software programs and retain such messages in order to comply wi= th applicable data security and retention requirements. Quantum is not resp= onsible for the proper and complete transmission of the substance of this c= ommunication or for any delay in its receipt. From hagen@jauu.net Thu Nov 7 14:39:45 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 55EC721E80E5 for ; Thu, 7 Nov 2013 14:39:45 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.6 X-Spam-Level: X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wRsgCRN+A4-X for ; Thu, 7 Nov 2013 14:39:44 -0800 (PST) Received: from Chamillionaire.breakpoint.cc (Chamillionaire.breakpoint.cc [IPv6:2001:4d88:1ffa:82:880:aa0:9009:64ae]) by ietfa.amsl.com (Postfix) with ESMTP id C5E5B21E8144 for ; Thu, 7 Nov 2013 14:39:44 -0800 (PST) Received: from pfeifer by Chamillionaire.breakpoint.cc with local (Exim 4.80) (envelope-from ) id 1VeYEp-0001aw-7W; Thu, 07 Nov 2013 23:39:43 +0100 Date: Thu, 7 Nov 2013 23:44:13 +0100 From: Hagen Paul Pfeifer To: David Borman Message-ID: <20131107224413.GB2201@localhost.localdomain> References: <20131107221802.GA2201@localhost.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: X-Key-Id: 98350C22 X-Key-Fingerprint: 490F 557B 6C48 6D7E 5706 2EA2 4A22 8D45 9835 0C22 X-GPG-Key: gpg --recv-keys --keyserver wwwkeys.eu.pgp.net 98350C22 User-Agent: Mutt/1.5.21 (2010-09-15) Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 22:39:45 -0000 * David Borman | 2013-11-07 22:30:21 [+0000]: >Ugh, RFC 793 and RFC 1122 aren’t explicit on this, but in RFC 793, page 43: > If a segment containing a single data octet sent to probe a zero window... > >Without at least one byte of data, it just looks like a regular ACK packet >and it won’t be guaranteed to elicit a response, which is the whole point of >the Zero Window Probe, to elicit a response. If the receiver do not open the window for 0 byte zero probes this operating system may not work in the wild. The receiver must response with the actual window for every payloaded packet. "Be conservative in what you send and _liberal in what you accept_". Hagen From jakob.heitz@ericsson.com Thu Nov 7 14:49:15 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9DA5C11E815B for ; Thu, 7 Nov 2013 14:49:15 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.576 X-Spam-Level: X-Spam-Status: No, score=-2.576 tagged_above=-999 required=5 tests=[AWL=0.023, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id jvEm1DBS-8uZ for ; Thu, 7 Nov 2013 14:49:10 -0800 (PST) Received: from usevmg21.ericsson.net (usevmg21.ericsson.net [198.24.6.65]) by ietfa.amsl.com (Postfix) with ESMTP id 264A511E8140 for ; Thu, 7 Nov 2013 14:49:07 -0800 (PST) X-AuditID: c6180641-b7fbd8e0000011cc-6e-527c18e220b3 Received: from EUSAAHC002.ericsson.se (Unknown_Domain [147.117.188.78]) by usevmg21.ericsson.net (Symantec Mail Security) with SMTP id 11.03.04556.2E81C725; Thu, 7 Nov 2013 23:49:06 +0100 (CET) Received: from EUSAAMB109.ericsson.se ([147.117.188.126]) by EUSAAHC002.ericsson.se ([147.117.188.78]) with mapi id 14.02.0328.009; Thu, 7 Nov 2013 17:49:05 -0500 From: Jakob Heitz To: Hagen Paul Pfeifer , David Borman Thread-Topic: [tcpm] Linux doesn't do ZWP correctly??? Thread-Index: AQHO2/p+1Q14vNmJ20aN/A05aCZZc5oaqk4AgAADcYCAAAPggP//rOzn Date: Thu, 7 Nov 2013 22:49:05 +0000 Message-ID: <2F3EBB88EC3A454AAB08915FBF0B8C7E02EB9E48@eusaamb109.ericsson.se> References: <20131107221802.GA2201@localhost.localdomain> , <20131107224413.GB2201@localhost.localdomain> In-Reply-To: <20131107224413.GB2201@localhost.localdomain> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [147.117.188.135] Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFrrGLMWRmVeSWpSXmKPExsUyuXSPn+4jiZogg3er5Sxe/d7AarHh5VZW i20n5zM5MHssWfKTyePl7f2sHr3P17IHMEdx2aSk5mSWpRbp2yVwZSyf61CwjKviWNd7tgbG QxxdjJwcEgImEtfX/WGGsMUkLtxbz9bFyMUhJHCEUWLD7glMEM4yRokZ/xvBqtgEdCS+Xe8C sjk4RASCJLaeLAYJMwsYS6z6uoYJxBYWMJVovb0SzBYRMJO4+/0FlO0msfJgL5jNIqAisfrR CxYQm1fAV+LgkxVgcSGBi4wSE+7EgNicAlYSJ/93soHYjEDHfT8FMZ9ZQFzi1pP5TBBHC0gs 2XMe6gFRiZeP/7FC2MoS3+c8YoGoN5B4f24+M4StLbFs4WtmiL2CEidnPmGZwCg2C8nYWUha ZiFpmYWkZQEjyypGjtLi1LLcdCPDTYzAuDkmwea4g3HBJ8tDjNIcLErivF/eOgcJCaQnlqRm p6YWpBbFF5XmpBYfYmTi4JRqYNTY2rnrXbGlxYWpncwXPp5NbXRY+bD7/cLpblVRiSfN7rYa aPaoNJm9/PhoWpbA5aOnBTlXLmFzu21uqOrDsXXX84syjyZP3hwe/XxTPsOBtF4JySvX84+d 3TxXrmI9y+Ibez8FXvxeJpV5KtzvR9QHix3iPwJ22yUukdhy6b7y/F0Nci6hi4qUWIozEg21 mIuKEwE6SrGOaQIAAA== Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 22:49:15 -0000 An ACK for an ACK causes an ACK war.=0A= --=0A= =0A= Jakob Heitz.=0A= =0A= ________________________________________=0A= From: tcpm-bounces@ietf.org [tcpm-bounces@ietf.org] on behalf of Hagen Paul= Pfeifer [hagen@jauu.net]=0A= Sent: Thursday, 07 November 2013 2:44 PM=0A= To: David Borman=0A= Cc: tcpm@ietf.org (tcpm@ietf.org)=0A= Subject: Re: [tcpm] Linux doesn't do ZWP correctly???=0A= =0A= * David Borman | 2013-11-07 22:30:21 [+0000]:=0A= =0A= >Ugh, RFC 793 and RFC 1122 aren=92t explicit on this, but in RFC 793, page = 43:=0A= > If a segment containing a single data octet sent to probe a zero windo= w...=0A= >=0A= >Without at least one byte of data, it just looks like a regular ACK packet= =0A= >and it won=92t be guaranteed to elicit a response, which is the whole poin= t of=0A= >the Zero Window Probe, to elicit a response.=0A= =0A= If the receiver do not open the window for 0 byte zero probes this operatin= g=0A= system may not work in the wild. The receiver must response with the actual= =0A= window for every payloaded packet. "Be conservative in what you send and=0A= _liberal in what you accept_".=0A= =0A= Hagen=0A= _______________________________________________=0A= tcpm mailing list=0A= tcpm@ietf.org=0A= https://www.ietf.org/mailman/listinfo/tcpm=0A= From johnwheffner@gmail.com Thu Nov 7 15:11:56 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5D23811E8289 for ; Thu, 7 Nov 2013 15:11:56 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.6 X-Spam-Level: X-Spam-Status: No, score=-2.6 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id smF7JKXMEhSI for ; Thu, 7 Nov 2013 15:11:54 -0800 (PST) Received: from mail-vc0-x236.google.com (mail-vc0-x236.google.com [IPv6:2607:f8b0:400c:c03::236]) by ietfa.amsl.com (Postfix) with ESMTP id 20EB011E8197 for ; Thu, 7 Nov 2013 15:11:54 -0800 (PST) Received: by mail-vc0-f182.google.com with SMTP id if17so888692vcb.13 for ; Thu, 07 Nov 2013 15:11:53 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=vRZq6R+nO8le/4Lb62nnTa2j7SShGzHGLF5x/yhZPDo=; b=OwLQMaScKFolG4bAFDY2iKLY+zecXqENEecj55rNdLnI5Qi9WhLgIQ2EfmSMhw5mwo EiWdaqKZJaXH4QiP/0zmTm0CkWIBuirtisiU4lPo1glPAJICaS/YPJsRiHx1ZnvpryzL tNHUdZSbWkfKBQqFDDZXKlQ6PJ0u2neI+cvLnw5Kwb/ntOCL4s3WAJwH0LnkFdPBEeql 69Hd9wiDlECwnsyS/niGQTo6aarj5zEd/1z2wAXdc1d+70e/orOT9+fgBEdG1zTHbflr cEWoZefPEm+q9hJ4nb/0SMMZ5mCpItJC7034AQ0RdWgSWHNn0NAiMA2PapoiRoF6eBnf mE8g== MIME-Version: 1.0 X-Received: by 10.220.144.80 with SMTP id y16mr8845354vcu.4.1383865913481; Thu, 07 Nov 2013 15:11:53 -0800 (PST) Received: by 10.220.93.74 with HTTP; Thu, 7 Nov 2013 15:11:53 -0800 (PST) In-Reply-To: References: Date: Thu, 7 Nov 2013 18:11:53 -0500 Message-ID: From: John Heffner To: David Borman Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 23:11:56 -0000 Linux sends ZWPs by using a sequence number one before SND_UNA with zero payload, which should elicit an ack. On Thu, Nov 7, 2013 at 3:46 PM, David Borman wro= te: > I=92ve been doing some testing of crossing Zero Window Probes, and it app= ears that the Linux TCP doesn=92t do ZWP properly. It is missing the 1 byt= e of data. > > My test program creates a pair of sockets, sets SO_RCVBUF to 20K and SO_S= NDBUF to 80K on both sides, then writes 40K of data in both directions. It= then pauses for 20 seconds to allow some ZWP packets, reads 1 byte of data= from both sides, pauses another 20 seconds and then reads all the data fro= m both sockets. I ran tcpdump to watch the packet traffic. Below are pack= et traces from both a Fedora Core 19 and a FreeBSD 9.2 system. The FreeBSD= trace has what I would expect to see, the Linux trace doesn=92t send any d= ata on the probes. > > Any thoughts on this? > > -David Borman > > # uname -a > Linux dab-fc17-vm.mdh.quantum.com 3.11.6-200.fc19.x86_64 #1 SMP Fri Oct 1= 8 22:34:18 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux > # tcpdump -i lo > tcpdump: verbose output suppressed, use -v or -vv for full protocol decod= e > listening on lo, link-type EN10MB (Ethernet), capture size 65535 bytes > 14:52:24.676876 IP localhost.56771 > localhost.57737: Flags [S], seq 1095= 762115, win 20480, options [mss 65495,sackOK,TS val 106278 ecr 0,nop,wscale= 0], length 0 > 14:52:24.676905 IP localhost.57737 > localhost.56771: Flags [S.], seq 894= 566307, ack 1095762116, win 20480, options [mss 65495,sackOK,TS val 106278 = ecr 106278,nop,wscale 0], length 0 > 14:52:24.676920 IP localhost.56771 > localhost.57737: Flags [.], ack 1, w= in 20480, options [nop,nop,TS val 106278 ecr 106278], length 0 > 14:52:24.677292 IP localhost.56771 > localhost.57737: Flags [.], seq 1:10= 241, ack 1, win 20480, options [nop,nop,TS val 106278 ecr 106278], length 1= 0240 > 14:52:24.677303 IP localhost.57737 > localhost.56771: Flags [.], ack 1024= 1, win 10240, options [nop,nop,TS val 106278 ecr 106278], length 0 > 14:52:24.677316 IP localhost.56771 > localhost.57737: Flags [P.], seq 102= 41:20481, ack 1, win 20480, options [nop,nop,TS val 106278 ecr 106278], len= gth 10240 > 14:52:24.678314 IP localhost.57737 > localhost.56771: Flags [.], seq 1:10= 241, ack 20481, win 0, options [nop,nop,TS val 106279 ecr 106278], length 1= 0240 > 14:52:24.678340 IP localhost.56771 > localhost.57737: Flags [.], ack 1024= 1, win 10240, options [nop,nop,TS val 106279 ecr 106279], length 0 > 14:52:24.678360 IP localhost.57737 > localhost.56771: Flags [P.], seq 102= 41:20481, ack 20481, win 0, options [nop,nop,TS val 106279 ecr 106278], len= gth 10240 > 14:52:24.717804 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106319 ecr 106279], length 0 > 14:52:24.879723 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106481 ecr 106279], length 0 > 14:52:24.879833 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106481 ecr 106319], length 0 > 14:52:24.923707 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106525 ecr 106319], length 0 > 14:52:24.923820 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106525 ecr 106481], length 0 > 14:52:25.283492 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106884 ecr 106481], length 0 > 14:52:25.283599 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106885 ecr 106525], length 0 > 14:52:25.335722 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106937 ecr 106525], length 0 > 14:52:25.335839 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 106937 ecr 106885], length 0 > 14:52:26.090812 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 107692 ecr 106885], length 0 > 14:52:26.090919 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 107692 ecr 106937], length 0 > 14:52:26.159642 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 107761 ecr 106937], length 0 > 14:52:26.159686 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 107761 ecr 107692], length 0 > 14:52:27.703494 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 109304 ecr 107692], length 0 > 14:52:27.703541 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 109304 ecr 107761], length 0 > 14:52:27.803683 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 109405 ecr 107761], length 0 > 14:52:27.803792 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 109405 ecr 109304], length 0 > 14:52:30.927667 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 112529 ecr 109304], length 0 > 14:52:30.927712 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 112529 ecr 109405], length 0 > 14:52:31.087676 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 112689 ecr 109405], length 0 > 14:52:31.087788 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 112689 ecr 112529], length 0 > 14:52:37.375727 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 118977 ecr 112529], length 0 > 14:52:37.375832 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 118977 ecr 112689], length 0 > 14:52:37.664034 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 119265 ecr 112689], length 0 > 14:52:37.664358 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 119265 ecr 118977], length 0 > 14:52:50.271699 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 131873 ecr 118977], length 0 > 14:52:50.271807 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 131873 ecr 119265], length 0 > 14:52:50.815486 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 132416 ecr 119265], length 0 > 14:52:50.815600 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 0, options [nop,nop,TS val 132416 ecr 131873], length 0 > 14:53:04.679770 IP localhost.57737 > localhost.56771: Flags [.], ack 2048= 1, win 20480, options [nop,nop,TS val 146281 ecr 132416], length 0 > 14:53:04.679790 IP localhost.56771 > localhost.57737: Flags [.], seq 2048= 1:30721, ack 20481, win 0, options [nop,nop,TS val 146281 ecr 146281], leng= th 10240 > 14:53:04.679795 IP localhost.56771 > localhost.57737: Flags [P.], seq 307= 21:40961, ack 20481, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 > 14:53:04.679908 IP localhost.56771 > localhost.57737: Flags [.], ack 2048= 1, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 > 14:53:04.679916 IP localhost.57737 > localhost.56771: Flags [.], seq 2048= 1:30721, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], leng= th 10240 > 14:53:04.679926 IP localhost.57737 > localhost.56771: Flags [P.], seq 307= 21:40961, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 > 14:53:04.679941 IP localhost.56771 > localhost.57737: Flags [.], ack 3072= 1, win 10240, options [nop,nop,TS val 146281 ecr 146281], length 0 > 14:53:04.679966 IP localhost.57737 > localhost.56771: Flags [.], ack 4096= 1, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 > 14:53:04.679974 IP localhost.56771 > localhost.57737: Flags [.], seq 4096= 1:51201, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], leng= th 10240 > 14:53:04.679978 IP localhost.56771 > localhost.57737: Flags [P.], seq 512= 01:61441, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 > 14:53:04.680002 IP localhost.56771 > localhost.57737: Flags [.], ack 4096= 1, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 > 14:53:04.680013 IP localhost.57737 > localhost.56771: Flags [.], seq 4096= 1:51201, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], leng= th 10240 > 14:53:04.680018 IP localhost.57737 > localhost.56771: Flags [P.], seq 512= 01:61441, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 > 14:53:04.680042 IP localhost.57737 > localhost.56771: Flags [.], ack 6144= 1, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 > 14:53:04.680049 IP localhost.56771 > localhost.57737: Flags [.], seq 6144= 1:71681, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], leng= th 10240 > > The trace from a FreeBSD 9.2 system looks like this: > > # uname -a > FreeBSD fbsd90-64 9.2-RELEASE FreeBSD 9.2-RELEASE #0 r255898: Thu Sep 26 = 22:50:31 UTC 2013 root@bake.isc.freebsd.org:/usr/obj/usr/src/sys/GENERIC= amd64 > # tcpdump -i lo0 > tcpdump: verbose output suppressed, use -v or -vv for full protocol decod= e > listening on lo0, link-type NULL (BSD loopback), capture size 65535 bytes > 13:44:43.912939 IP localhost.22111 > localhost.49248: Flags [S], seq 1589= 925916, win 20480, options [mss 16344,nop,wscale 6,sackOK,TS val 120482 ecr= 0], length 0 > 13:44:43.912987 IP localhost.49248 > localhost.22111: Flags [S.], seq 175= 2403575, ack 1589925917, win 20480, options [mss 16344,nop,wscale 6,sackOK,= TS val 2357200578 ecr 120482], length 0 > 13:44:43.912999 IP localhost.22111 > localhost.49248: Flags [.], ack 1, w= in 510, options [nop,nop,TS val 120482 ecr 2357200578], length 0 > 13:44:43.913456 IP localhost.22111 > localhost.49248: Flags [.], seq 1:16= 333, ack 1, win 510, options [nop,nop,TS val 120482 ecr 2357200578], length= 16332 > 13:44:43.914272 IP localhost.49248 > localhost.22111: Flags [.], seq 1:16= 333, ack 16333, win 255, options [nop,nop,TS val 2357200578 ecr 120482], le= ngth 16332 > 13:44:43.914377 IP localhost.22111 > localhost.49248: Flags [.], seq 1633= 3:32653, ack 16333, win 255, options [nop,nop,TS val 120482 ecr 2357200578]= , length 16320 > 13:44:43.914460 IP localhost.49248 > localhost.22111: Flags [.], seq 1633= 3:32653, ack 32653, win 0, options [nop,nop,TS val 2357200578 ecr 120482], = length 16320 > 13:44:44.012989 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 3, win 0, options [nop,nop,TS val 120583 ecr 2357200578], length 0 > 13:44:48.912031 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 3:32654, ack 32653, win 0, options [nop,nop,TS val 125482 ecr 2357200578], = length 1 > 13:44:48.912056 IP localhost.49248 > localhost.22111: Flags [.], ack 3265= 4, win 0, options [nop,nop,TS val 2357205578 ecr 125482], length 0 > 13:44:49.012049 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 3:32654, ack 32654, win 0, options [nop,nop,TS val 2357205678 ecr 125482], = length 1 > 13:44:49.012074 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 4, win 0, options [nop,nop,TS val 125582 ecr 2357205678], length 0 > 13:44:53.913015 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 4:32655, ack 32654, win 0, options [nop,nop,TS val 130483 ecr 2357205678], = length 1 > 13:44:53.913059 IP localhost.49248 > localhost.22111: Flags [.], ack 3265= 5, win 0, options [nop,nop,TS val 2357210579 ecr 130483], length 0 > 13:44:54.013017 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 4:32655, ack 32655, win 0, options [nop,nop,TS val 2357210679 ecr 130483], = length 1 > 13:44:54.013042 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 5, win 0, options [nop,nop,TS val 130583 ecr 2357210679], length 0 > 13:44:58.913034 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 5:32656, ack 32655, win 0, options [nop,nop,TS val 135483 ecr 2357210679], = length 1 > 13:44:58.913060 IP localhost.49248 > localhost.22111: Flags [.], ack 3265= 6, win 0, options [nop,nop,TS val 2357215579 ecr 135483], length 0 > 13:44:59.012397 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 5:32656, ack 32656, win 0, options [nop,nop,TS val 2357215679 ecr 135483], = length 1 > 13:44:59.012421 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 6, win 0, options [nop,nop,TS val 135583 ecr 2357215679], length 0 > 13:45:03.913033 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 6:32657, ack 32656, win 0, options [nop,nop,TS val 140483 ecr 2357215679], = length 1 > 13:45:03.913063 IP localhost.49248 > localhost.22111: Flags [.], ack 3265= 7, win 0, options [nop,nop,TS val 2357220579 ecr 140483], length 0 > 13:45:04.013019 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 6:32657, ack 32657, win 0, options [nop,nop,TS val 2357220679 ecr 140483], = length 1 > 13:45:04.013044 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 7, win 0, options [nop,nop,TS val 140583 ecr 2357220679], length 0 > 13:45:08.913018 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 7:32658, ack 32657, win 0, options [nop,nop,TS val 145483 ecr 2357220679], = length 1 > 13:45:08.913043 IP localhost.49248 > localhost.22111: Flags [.], ack 3265= 8, win 0, options [nop,nop,TS val 2357225579 ecr 145483], length 0 > 13:45:09.013017 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 7:32658, ack 32658, win 0, options [nop,nop,TS val 2357225679 ecr 145483], = length 1 > 13:45:09.013042 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 8, win 0, options [nop,nop,TS val 145583 ecr 2357225679], length 0 > 13:45:13.913009 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 8:32659, ack 32658, win 0, options [nop,nop,TS val 150483 ecr 2357225679], = length 1 > 13:45:13.913034 IP localhost.49248 > localhost.22111: Flags [.], ack 3265= 9, win 0, options [nop,nop,TS val 2357230579 ecr 150483], length 0 > 13:45:14.012980 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 8:32659, ack 32659, win 0, options [nop,nop,TS val 2357230679 ecr 150483], = length 1 > 13:45:14.013005 IP localhost.22111 > localhost.49248: Flags [.], ack 3265= 9, win 0, options [nop,nop,TS val 150583 ecr 2357230679], length 0 > 13:45:18.912690 IP localhost.22111 > localhost.49248: Flags [.], seq 3265= 9:32660, ack 32659, win 0, options [nop,nop,TS val 155483 ecr 2357230679], = length 1 > 13:45:18.912717 IP localhost.49248 > localhost.22111: Flags [.], ack 3266= 0, win 0, options [nop,nop,TS val 2357235579 ecr 155483], length 0 > 13:45:19.012811 IP localhost.49248 > localhost.22111: Flags [.], seq 3265= 9:32660, ack 32660, win 0, options [nop,nop,TS val 2357235679 ecr 155483], = length 1 > 13:45:19.012828 IP localhost.22111 > localhost.49248: Flags [.], ack 3266= 0, win 0, options [nop,nop,TS val 155583 ecr 2357235679], length 0 > 13:45:23.912288 IP localhost.22111 > localhost.49248: Flags [.], seq 3266= 0:32661, ack 32660, win 0, options [nop,nop,TS val 160482 ecr 2357235679], = length 1 > 13:45:23.912316 IP localhost.49248 > localhost.22111: Flags [.], ack 3266= 1, win 0, options [nop,nop,TS val 2357240578 ecr 160482], length 0 > 13:45:24.013018 IP localhost.49248 > localhost.22111: Flags [.], seq 3266= 0:32661, ack 32661, win 510, options [nop,nop,TS val 2357240679 ecr 160482]= , length 1 > 13:45:24.013046 IP localhost.22111 > localhost.49248: Flags [.], seq 3266= 1:48993, ack 32661, win 510, options [nop,nop,TS val 160583 ecr 2357240679]= , length 16332 > 13:45:24.013195 IP localhost.49248 > localhost.22111: Flags [.], seq 3266= 1:48993, ack 48993, win 255, options [nop,nop,TS val 2357240679 ecr 160583]= , length 16332 > 13:45:24.013207 IP localhost.22111 > localhost.49248: Flags [.], seq 4899= 3:65313, ack 48993, win 255, options [nop,nop,TS val 160583 ecr 2357240679]= , length 16320 > 13:45:24.013215 IP localhost.49248 > localhost.22111: Flags [.], seq 4899= 3:65313, ack 65313, win 0, options [nop,nop,TS val 2357240679 ecr 160583], = length 16320 > 13:45:24.112038 IP localhost.22111 > localhost.49248: Flags [.], ack 6531= 3, win 510, options [nop,nop,TS val 160682 ecr 2357240679], length 0 > 13:45:24.112075 IP localhost.49248 > localhost.22111: Flags [.], seq 6531= 3:81645, ack 65313, win 510, options [nop,nop,TS val 2357240778 ecr 160682]= , length 16332 > 13:45:24.112333 IP localhost.49248 > localhost.22111: Flags [P.], seq 816= 45:81921, ack 65313, win 510, options [nop,nop,TS val 2357240778 ecr 160682= ], length 276 > 13:45:24.112354 IP localhost.22111 > localhost.49248: Flags [.], seq 6531= 3:81645, ack 81645, win 255, options [nop,nop,TS val 160682 ecr 2357240778]= , length 16332 > 13:45:24.112505 IP localhost.22111 > localhost.49248: Flags [P.], seq 816= 45:81921, ack 81645, win 255, options [nop,nop,TS val 160682 ecr 2357240778= ], length 276 > 13:45:24.112520 IP localhost.49248 > localhost.22111: Flags [.], ack 8192= 1, win 250, options [nop,nop,TS val 2357240778 ecr 160682], length 0 > 13:45:24.112596 IP localhost.49248 > localhost.22111: Flags [F.], seq 819= 21, ack 81921, win 510, options [nop,nop,TS val 2357240778 ecr 160682], len= gth 0 > 13:45:24.112606 IP localhost.22111 > localhost.49248: Flags [.], ack 8192= 2, win 510, options [nop,nop,TS val 160682 ecr 2357240778], length 0 > 13:45:24.112624 IP localhost.22111 > localhost.49248: Flags [F.], seq 819= 21, ack 81922, win 510, options [nop,nop,TS val 160682 ecr 2357240778], len= gth 0 > 13:45:24.112646 IP localhost.49248 > localhost.22111: Flags [.], ack 8192= 2, win 510, options [nop,nop,TS val 2357240778 ecr 160682], length 0 > > ---------------------------------------------------------------------- > The information contained in this transmission may be confidential. Any d= isclosure, copying, or further distribution of confidential information is = not permitted unless such privilege is explicitly granted in writing by Qua= ntum. Quantum reserves the right to have electronic communications, includi= ng email and attachments, sent across its networks filtered through anti vi= rus and spam software programs and retain such messages in order to comply = with applicable data security and retention requirements. Quantum is not re= sponsible for the proper and complete transmission of the substance of this= communication or for any delay in its receipt. > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From ncardwell@google.com Thu Nov 7 15:25:40 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1E02911E812F for ; Thu, 7 Nov 2013 15:25:40 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.378 X-Spam-Level: X-Spam-Status: No, score=-1.378 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_33=0.6, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id b1Nys0Y1KcKB for ; Thu, 7 Nov 2013 15:25:38 -0800 (PST) Received: from mail-ee0-x22e.google.com (mail-ee0-x22e.google.com [IPv6:2a00:1450:4013:c00::22e]) by ietfa.amsl.com (Postfix) with ESMTP id AC4C711E8103 for ; Thu, 7 Nov 2013 15:25:37 -0800 (PST) Received: by mail-ee0-f46.google.com with SMTP id b57so642743eek.19 for ; Thu, 07 Nov 2013 15:25:36 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type:content-transfer-encoding; bh=iX1hFLTjPP/Ns/Sh+c76+oE1SFQ+8OngdFwHZXID9T4=; b=HQE0CLpXKoMELyV+4QD3JzdztLr3x+4qWT8M57sjyT3xulFOwNv21C+vwtfBuvV9Dq XNqloU6gXgFnlrKHkX3SpUUqjmZZkPaKQnmI/AX32ClxlZN4PiWQ44pP8axFbH1I9b/z hqGH02N9ELG4oIaNIWlrc7RLdBA4OgqUEolcl9FWTOv4YyNTuiaKGCs0Q/V+mUsSyOVz 4XGxRcso39nEWzhji8Tlw5bvqwRmmGOmWSGFt4PcD51HEewDqJlL/Co0PGyXXg43J/PK Y1prDCJtlvZ9UAWDU82JdWG3TWoGszQJmT0HuXuSMsEP0Dey+6ukPhxbX5nj45h5RHbx 9+Ww== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=iX1hFLTjPP/Ns/Sh+c76+oE1SFQ+8OngdFwHZXID9T4=; b=DtcMhPJa73ZlExk9unOLNJsg9F8i7kNHYXpp6s9YB3F4DsJpJVaJRDsQjJp1+LC0tJ G/mWGyrEsWHwPACt6uFkuNvzLY1Rq4GNaWhMtvWTxGW5oJENdqDb5jJNUSq9i6/4NFvL eTiAjRAf3sdTYRpBV/A2eiE01IfIPMHMbChhdH77RE/rlNKiatlCCKJ/qHP5hpwqO3Zk 6U8G8cWgQt76EfxCg/aVggUzLHbGjD1joyv/FnNA2fHWgb4TlOeybcKdfnupGfeEDoae lmg3DDUpnzUqrkehZhp5G5qVKvTK1P0OIm1E1A8YSU/X0U5G0wdR2zuLhuCjeYvB+SMt K6xw== X-Gm-Message-State: ALoCoQkbx8WB/gj95sAMyhP6rTyrdcoV42me/Jd4twX3x6qnoWQrn91FtbnU516jFJRZJ0NuzcLojS4M4lm7BbyGqBJ+JusOHA8LEiACv6bB+8hqIXaQQBIF4RE4393sag9fD8XBgZgdv79H33QqRNOGpDakc8Adn+iv7GvmTeEK++Y8Qm3xbdy6JcRG9XtnG5k1swauMaps MIME-Version: 1.0 X-Received: by 10.14.210.8 with SMTP id t8mr12102668eeo.39.1383866736516; Thu, 07 Nov 2013 15:25:36 -0800 (PST) Received: by 10.15.33.141 with HTTP; Thu, 7 Nov 2013 15:25:36 -0800 (PST) In-Reply-To: References: Date: Thu, 7 Nov 2013 18:25:36 -0500 Message-ID: From: Neal Cardwell To: John Heffner Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: quoted-printable Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" , David Borman Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 07 Nov 2013 23:25:40 -0000 On Thu, Nov 7, 2013 at 6:11 PM, John Heffner wrote= : > Linux sends ZWPs by using a sequence number one before SND_UNA with > zero payload, which should elicit an ack. Yes, RFC 793 (page 69) requires receivers to respond with an ACK in response to an ACK when the incoming sequence number of the ACK is invalid due to being below RCV.NXT: """ There are four cases for the acceptability test for an incoming segment: Segment Receive Test Length Window ------- ------- ------------------------------------------- ... 0 >0 RCV.NXT =3D< SEG.SEQ < RCV.NXT+RCV.WND ... If an incoming segment is not acceptable, an acknowledgment should be sent in reply (unless the RST bit is set, if so drop the segment and return): After sending the acknowledgment, drop the unacceptable segment and return. """ neal > > On Thu, Nov 7, 2013 at 3:46 PM, David Borman w= rote: >> I=92ve been doing some testing of crossing Zero Window Probes, and it ap= pears that the Linux TCP doesn=92t do ZWP properly. It is missing the 1 by= te of data. >> >> My test program creates a pair of sockets, sets SO_RCVBUF to 20K and SO_= SNDBUF to 80K on both sides, then writes 40K of data in both directions. I= t then pauses for 20 seconds to allow some ZWP packets, reads 1 byte of dat= a from both sides, pauses another 20 seconds and then reads all the data fr= om both sockets. I ran tcpdump to watch the packet traffic. Below are pac= ket traces from both a Fedora Core 19 and a FreeBSD 9.2 system. The FreeBS= D trace has what I would expect to see, the Linux trace doesn=92t send any = data on the probes. >> >> Any thoughts on this? >> >> -David Borman >> >> # uname -a >> Linux dab-fc17-vm.mdh.quantum.com 3.11.6-200.fc19.x86_64 #1 SMP Fri Oct = 18 22:34:18 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux >> # tcpdump -i lo >> tcpdump: verbose output suppressed, use -v or -vv for full protocol deco= de >> listening on lo, link-type EN10MB (Ethernet), capture size 65535 bytes >> 14:52:24.676876 IP localhost.56771 > localhost.57737: Flags [S], seq 109= 5762115, win 20480, options [mss 65495,sackOK,TS val 106278 ecr 0,nop,wscal= e 0], length 0 >> 14:52:24.676905 IP localhost.57737 > localhost.56771: Flags [S.], seq 89= 4566307, ack 1095762116, win 20480, options [mss 65495,sackOK,TS val 106278= ecr 106278,nop,wscale 0], length 0 >> 14:52:24.676920 IP localhost.56771 > localhost.57737: Flags [.], ack 1, = win 20480, options [nop,nop,TS val 106278 ecr 106278], length 0 >> 14:52:24.677292 IP localhost.56771 > localhost.57737: Flags [.], seq 1:1= 0241, ack 1, win 20480, options [nop,nop,TS val 106278 ecr 106278], length = 10240 >> 14:52:24.677303 IP localhost.57737 > localhost.56771: Flags [.], ack 102= 41, win 10240, options [nop,nop,TS val 106278 ecr 106278], length 0 >> 14:52:24.677316 IP localhost.56771 > localhost.57737: Flags [P.], seq 10= 241:20481, ack 1, win 20480, options [nop,nop,TS val 106278 ecr 106278], le= ngth 10240 >> 14:52:24.678314 IP localhost.57737 > localhost.56771: Flags [.], seq 1:1= 0241, ack 20481, win 0, options [nop,nop,TS val 106279 ecr 106278], length = 10240 >> 14:52:24.678340 IP localhost.56771 > localhost.57737: Flags [.], ack 102= 41, win 10240, options [nop,nop,TS val 106279 ecr 106279], length 0 >> 14:52:24.678360 IP localhost.57737 > localhost.56771: Flags [P.], seq 10= 241:20481, ack 20481, win 0, options [nop,nop,TS val 106279 ecr 106278], le= ngth 10240 >> 14:52:24.717804 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106319 ecr 106279], length 0 >> 14:52:24.879723 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106481 ecr 106279], length 0 >> 14:52:24.879833 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106481 ecr 106319], length 0 >> 14:52:24.923707 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106525 ecr 106319], length 0 >> 14:52:24.923820 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106525 ecr 106481], length 0 >> 14:52:25.283492 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106884 ecr 106481], length 0 >> 14:52:25.283599 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106885 ecr 106525], length 0 >> 14:52:25.335722 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106937 ecr 106525], length 0 >> 14:52:25.335839 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 106937 ecr 106885], length 0 >> 14:52:26.090812 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 107692 ecr 106885], length 0 >> 14:52:26.090919 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 107692 ecr 106937], length 0 >> 14:52:26.159642 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 107761 ecr 106937], length 0 >> 14:52:26.159686 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 107761 ecr 107692], length 0 >> 14:52:27.703494 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 109304 ecr 107692], length 0 >> 14:52:27.703541 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 109304 ecr 107761], length 0 >> 14:52:27.803683 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 109405 ecr 107761], length 0 >> 14:52:27.803792 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 109405 ecr 109304], length 0 >> 14:52:30.927667 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 112529 ecr 109304], length 0 >> 14:52:30.927712 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 112529 ecr 109405], length 0 >> 14:52:31.087676 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 112689 ecr 109405], length 0 >> 14:52:31.087788 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 112689 ecr 112529], length 0 >> 14:52:37.375727 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 118977 ecr 112529], length 0 >> 14:52:37.375832 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 118977 ecr 112689], length 0 >> 14:52:37.664034 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 119265 ecr 112689], length 0 >> 14:52:37.664358 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 119265 ecr 118977], length 0 >> 14:52:50.271699 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 131873 ecr 118977], length 0 >> 14:52:50.271807 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 131873 ecr 119265], length 0 >> 14:52:50.815486 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 132416 ecr 119265], length 0 >> 14:52:50.815600 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 0, options [nop,nop,TS val 132416 ecr 131873], length 0 >> 14:53:04.679770 IP localhost.57737 > localhost.56771: Flags [.], ack 204= 81, win 20480, options [nop,nop,TS val 146281 ecr 132416], length 0 >> 14:53:04.679790 IP localhost.56771 > localhost.57737: Flags [.], seq 204= 81:30721, ack 20481, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 >> 14:53:04.679795 IP localhost.56771 > localhost.57737: Flags [P.], seq 30= 721:40961, ack 20481, win 0, options [nop,nop,TS val 146281 ecr 146281], le= ngth 10240 >> 14:53:04.679908 IP localhost.56771 > localhost.57737: Flags [.], ack 204= 81, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 >> 14:53:04.679916 IP localhost.57737 > localhost.56771: Flags [.], seq 204= 81:30721, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 >> 14:53:04.679926 IP localhost.57737 > localhost.56771: Flags [P.], seq 30= 721:40961, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], le= ngth 10240 >> 14:53:04.679941 IP localhost.56771 > localhost.57737: Flags [.], ack 307= 21, win 10240, options [nop,nop,TS val 146281 ecr 146281], length 0 >> 14:53:04.679966 IP localhost.57737 > localhost.56771: Flags [.], ack 409= 61, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 >> 14:53:04.679974 IP localhost.56771 > localhost.57737: Flags [.], seq 409= 61:51201, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 >> 14:53:04.679978 IP localhost.56771 > localhost.57737: Flags [P.], seq 51= 201:61441, ack 40961, win 0, options [nop,nop,TS val 146281 ecr 146281], le= ngth 10240 >> 14:53:04.680002 IP localhost.56771 > localhost.57737: Flags [.], ack 409= 61, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 >> 14:53:04.680013 IP localhost.57737 > localhost.56771: Flags [.], seq 409= 61:51201, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 >> 14:53:04.680018 IP localhost.57737 > localhost.56771: Flags [P.], seq 51= 201:61441, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], le= ngth 10240 >> 14:53:04.680042 IP localhost.57737 > localhost.56771: Flags [.], ack 614= 41, win 20480, options [nop,nop,TS val 146281 ecr 146281], length 0 >> 14:53:04.680049 IP localhost.56771 > localhost.57737: Flags [.], seq 614= 41:71681, ack 61441, win 0, options [nop,nop,TS val 146281 ecr 146281], len= gth 10240 >> >> The trace from a FreeBSD 9.2 system looks like this: >> >> # uname -a >> FreeBSD fbsd90-64 9.2-RELEASE FreeBSD 9.2-RELEASE #0 r255898: Thu Sep 26= 22:50:31 UTC 2013 root@bake.isc.freebsd.org:/usr/obj/usr/src/sys/GENERI= C amd64 >> # tcpdump -i lo0 >> tcpdump: verbose output suppressed, use -v or -vv for full protocol deco= de >> listening on lo0, link-type NULL (BSD loopback), capture size 65535 byte= s >> 13:44:43.912939 IP localhost.22111 > localhost.49248: Flags [S], seq 158= 9925916, win 20480, options [mss 16344,nop,wscale 6,sackOK,TS val 120482 ec= r 0], length 0 >> 13:44:43.912987 IP localhost.49248 > localhost.22111: Flags [S.], seq 17= 52403575, ack 1589925917, win 20480, options [mss 16344,nop,wscale 6,sackOK= ,TS val 2357200578 ecr 120482], length 0 >> 13:44:43.912999 IP localhost.22111 > localhost.49248: Flags [.], ack 1, = win 510, options [nop,nop,TS val 120482 ecr 2357200578], length 0 >> 13:44:43.913456 IP localhost.22111 > localhost.49248: Flags [.], seq 1:1= 6333, ack 1, win 510, options [nop,nop,TS val 120482 ecr 2357200578], lengt= h 16332 >> 13:44:43.914272 IP localhost.49248 > localhost.22111: Flags [.], seq 1:1= 6333, ack 16333, win 255, options [nop,nop,TS val 2357200578 ecr 120482], l= ength 16332 >> 13:44:43.914377 IP localhost.22111 > localhost.49248: Flags [.], seq 163= 33:32653, ack 16333, win 255, options [nop,nop,TS val 120482 ecr 2357200578= ], length 16320 >> 13:44:43.914460 IP localhost.49248 > localhost.22111: Flags [.], seq 163= 33:32653, ack 32653, win 0, options [nop,nop,TS val 2357200578 ecr 120482],= length 16320 >> 13:44:44.012989 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 53, win 0, options [nop,nop,TS val 120583 ecr 2357200578], length 0 >> 13:44:48.912031 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 53:32654, ack 32653, win 0, options [nop,nop,TS val 125482 ecr 2357200578],= length 1 >> 13:44:48.912056 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 54, win 0, options [nop,nop,TS val 2357205578 ecr 125482], length 0 >> 13:44:49.012049 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 53:32654, ack 32654, win 0, options [nop,nop,TS val 2357205678 ecr 125482],= length 1 >> 13:44:49.012074 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 54, win 0, options [nop,nop,TS val 125582 ecr 2357205678], length 0 >> 13:44:53.913015 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 54:32655, ack 32654, win 0, options [nop,nop,TS val 130483 ecr 2357205678],= length 1 >> 13:44:53.913059 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 55, win 0, options [nop,nop,TS val 2357210579 ecr 130483], length 0 >> 13:44:54.013017 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 54:32655, ack 32655, win 0, options [nop,nop,TS val 2357210679 ecr 130483],= length 1 >> 13:44:54.013042 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 55, win 0, options [nop,nop,TS val 130583 ecr 2357210679], length 0 >> 13:44:58.913034 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 55:32656, ack 32655, win 0, options [nop,nop,TS val 135483 ecr 2357210679],= length 1 >> 13:44:58.913060 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 56, win 0, options [nop,nop,TS val 2357215579 ecr 135483], length 0 >> 13:44:59.012397 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 55:32656, ack 32656, win 0, options [nop,nop,TS val 2357215679 ecr 135483],= length 1 >> 13:44:59.012421 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 56, win 0, options [nop,nop,TS val 135583 ecr 2357215679], length 0 >> 13:45:03.913033 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 56:32657, ack 32656, win 0, options [nop,nop,TS val 140483 ecr 2357215679],= length 1 >> 13:45:03.913063 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 57, win 0, options [nop,nop,TS val 2357220579 ecr 140483], length 0 >> 13:45:04.013019 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 56:32657, ack 32657, win 0, options [nop,nop,TS val 2357220679 ecr 140483],= length 1 >> 13:45:04.013044 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 57, win 0, options [nop,nop,TS val 140583 ecr 2357220679], length 0 >> 13:45:08.913018 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 57:32658, ack 32657, win 0, options [nop,nop,TS val 145483 ecr 2357220679],= length 1 >> 13:45:08.913043 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 58, win 0, options [nop,nop,TS val 2357225579 ecr 145483], length 0 >> 13:45:09.013017 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 57:32658, ack 32658, win 0, options [nop,nop,TS val 2357225679 ecr 145483],= length 1 >> 13:45:09.013042 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 58, win 0, options [nop,nop,TS val 145583 ecr 2357225679], length 0 >> 13:45:13.913009 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 58:32659, ack 32658, win 0, options [nop,nop,TS val 150483 ecr 2357225679],= length 1 >> 13:45:13.913034 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 59, win 0, options [nop,nop,TS val 2357230579 ecr 150483], length 0 >> 13:45:14.012980 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 58:32659, ack 32659, win 0, options [nop,nop,TS val 2357230679 ecr 150483],= length 1 >> 13:45:14.013005 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 59, win 0, options [nop,nop,TS val 150583 ecr 2357230679], length 0 >> 13:45:18.912690 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 59:32660, ack 32659, win 0, options [nop,nop,TS val 155483 ecr 2357230679],= length 1 >> 13:45:18.912717 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 60, win 0, options [nop,nop,TS val 2357235579 ecr 155483], length 0 >> 13:45:19.012811 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 59:32660, ack 32660, win 0, options [nop,nop,TS val 2357235679 ecr 155483],= length 1 >> 13:45:19.012828 IP localhost.22111 > localhost.49248: Flags [.], ack 326= 60, win 0, options [nop,nop,TS val 155583 ecr 2357235679], length 0 >> 13:45:23.912288 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 60:32661, ack 32660, win 0, options [nop,nop,TS val 160482 ecr 2357235679],= length 1 >> 13:45:23.912316 IP localhost.49248 > localhost.22111: Flags [.], ack 326= 61, win 0, options [nop,nop,TS val 2357240578 ecr 160482], length 0 >> 13:45:24.013018 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 60:32661, ack 32661, win 510, options [nop,nop,TS val 2357240679 ecr 160482= ], length 1 >> 13:45:24.013046 IP localhost.22111 > localhost.49248: Flags [.], seq 326= 61:48993, ack 32661, win 510, options [nop,nop,TS val 160583 ecr 2357240679= ], length 16332 >> 13:45:24.013195 IP localhost.49248 > localhost.22111: Flags [.], seq 326= 61:48993, ack 48993, win 255, options [nop,nop,TS val 2357240679 ecr 160583= ], length 16332 >> 13:45:24.013207 IP localhost.22111 > localhost.49248: Flags [.], seq 489= 93:65313, ack 48993, win 255, options [nop,nop,TS val 160583 ecr 2357240679= ], length 16320 >> 13:45:24.013215 IP localhost.49248 > localhost.22111: Flags [.], seq 489= 93:65313, ack 65313, win 0, options [nop,nop,TS val 2357240679 ecr 160583],= length 16320 >> 13:45:24.112038 IP localhost.22111 > localhost.49248: Flags [.], ack 653= 13, win 510, options [nop,nop,TS val 160682 ecr 2357240679], length 0 >> 13:45:24.112075 IP localhost.49248 > localhost.22111: Flags [.], seq 653= 13:81645, ack 65313, win 510, options [nop,nop,TS val 2357240778 ecr 160682= ], length 16332 >> 13:45:24.112333 IP localhost.49248 > localhost.22111: Flags [P.], seq 81= 645:81921, ack 65313, win 510, options [nop,nop,TS val 2357240778 ecr 16068= 2], length 276 >> 13:45:24.112354 IP localhost.22111 > localhost.49248: Flags [.], seq 653= 13:81645, ack 81645, win 255, options [nop,nop,TS val 160682 ecr 2357240778= ], length 16332 >> 13:45:24.112505 IP localhost.22111 > localhost.49248: Flags [P.], seq 81= 645:81921, ack 81645, win 255, options [nop,nop,TS val 160682 ecr 235724077= 8], length 276 >> 13:45:24.112520 IP localhost.49248 > localhost.22111: Flags [.], ack 819= 21, win 250, options [nop,nop,TS val 2357240778 ecr 160682], length 0 >> 13:45:24.112596 IP localhost.49248 > localhost.22111: Flags [F.], seq 81= 921, ack 81921, win 510, options [nop,nop,TS val 2357240778 ecr 160682], le= ngth 0 >> 13:45:24.112606 IP localhost.22111 > localhost.49248: Flags [.], ack 819= 22, win 510, options [nop,nop,TS val 160682 ecr 2357240778], length 0 >> 13:45:24.112624 IP localhost.22111 > localhost.49248: Flags [F.], seq 81= 921, ack 81922, win 510, options [nop,nop,TS val 160682 ecr 2357240778], le= ngth 0 >> 13:45:24.112646 IP localhost.49248 > localhost.22111: Flags [.], ack 819= 22, win 510, options [nop,nop,TS val 2357240778 ecr 160682], length 0 >> >> ---------------------------------------------------------------------- >> The information contained in this transmission may be confidential. Any = disclosure, copying, or further distribution of confidential information is= not permitted unless such privilege is explicitly granted in writing by Qu= antum. Quantum reserves the right to have electronic communications, includ= ing email and attachments, sent across its networks filtered through anti v= irus and spam software programs and retain such messages in order to comply= with applicable data security and retention requirements. Quantum is not r= esponsible for the proper and complete transmission of the substance of thi= s communication or for any delay in its receipt. >> _______________________________________________ >> tcpm mailing list >> tcpm@ietf.org >> https://www.ietf.org/mailman/listinfo/tcpm > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From prvs=9024a7fe46=david.borman@quantum.com Fri Nov 8 08:07:34 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1914B11E81AC for ; Fri, 8 Nov 2013 08:07:34 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.965 X-Spam-Level: X-Spam-Status: No, score=-2.965 tagged_above=-999 required=5 tests=[AWL=-0.300, BAYES_00=-2.599, IP_NOT_FRIENDLY=0.334, J_CHICKENPOX_33=0.6, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id bgfsOGKUpjI9 for ; Fri, 8 Nov 2013 08:07:26 -0800 (PST) Received: from mx0b-000ceb01.pphosted.com (mx0b-000ceb01.pphosted.com [67.231.152.126]) by ietfa.amsl.com (Postfix) with ESMTP id 7F2CD11E8189 for ; Fri, 8 Nov 2013 08:07:26 -0800 (PST) Received: from pps.filterd (m0001158 [127.0.0.1]) by mx0b-000ceb01.pphosted.com (8.14.5/8.14.5) with SMTP id rA8G4LQ0007973; Fri, 8 Nov 2013 08:07:21 -0800 Received: from ppoxedge2.quantum.com ([146.174.252.28]) by mx0b-000ceb01.pphosted.com with ESMTP id 1g0vqerp8j-6 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Fri, 08 Nov 2013 08:07:21 -0800 Received: from PPOMSG2.QUANTUM.com (10.50.35.27) by PPOXEDGE2.quantum.com (146.174.252.28) with Microsoft SMTP Server (TLS) id 14.2.318.1; Fri, 8 Nov 2013 09:04:53 -0700 Received: from PPOMSG1.QUANTUM.com ([10.50.35.26]) by ppomsg2 ([10.50.35.27]) with mapi id 14.02.0318.001; Fri, 8 Nov 2013 09:05:19 -0700 From: David Borman To: John Heffner Thread-Topic: [tcpm] Linux doesn't do ZWP correctly??? Thread-Index: AQHO2/p+1Q14vNmJ20aN/A05aCZZc5oa2uGAgAEbJIA= Date: Fri, 8 Nov 2013 16:05:17 +0000 Message-ID: References: In-Reply-To: Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [192.168.110.1] Content-ID: <9C2FE56D6C5B3B45BFD59BECF16E5BD2@QUANTUM.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.10.8794, 1.0.431, 0.0.0000 definitions=2013-11-08_07:2013-11-08, 2013-11-08, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1305240000 definitions=main-1311080090 Content-Type: text/plain; charset="Windows-1252" Cc: "tcpm@ietf.org \(tcpm@ietf.org\)" Subject: Re: [tcpm] Linux doesn't do ZWP correctly??? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 08 Nov 2013 16:07:34 -0000 On Nov 7, 2013, at 5:11 PM, John Heffner wrote: > Linux sends ZWPs by using a sequence number one before SND_UNA with > zero payload, which should elicit an ack. Ah, the sequence number of packets without data doesn=92t show up in the tc= pdump traces, unless you use -vv. So, Linux is using the same packet forma= t for ZWP as for Keep-Alive packets. Interesting. Years ago this could ha= ve been an issue, as noted in RFC 1122 page 102, when talking about Keep-Al= ives: Such a segment generally contains SEG.SEQ =3D SND.NXT-1 and may or may not contain one garbage octet of data. ... An implementation SHOULD send a keep-alive segment with no data; however, it MAY be configurable to send a keep-alive segment containing one garbage octet, for compatibility with erroneous TCP implementations. ... Unfortunately, some misbehaved TCP implementations fail to respond to a segment with SEG.SEQ =3D SND.NXT-1 unless the segment contains data. ... Hopefully those misbehaved TCP implementations are long since fixed or dead= . :-) Running the test again, using -vv on tcpdump, I see that the zero-length pr= obe packets have seq-1. -David Borman >=20 > On Thu, Nov 7, 2013 at 3:46 PM, David Borman w= rote: >> I=92ve been doing some testing of crossing Zero Window Probes, and it ap= pears that the Linux TCP doesn=92t do ZWP properly. It is missing the 1 by= te of data. >>=20 >> My test program creates a pair of sockets, sets SO_RCVBUF to 20K and SO_= SNDBUF to 80K on both sides, then writes 40K of data in both directions. I= t then pauses for 20 seconds to allow some ZWP packets, reads 1 byte of dat= a from both sides, pauses another 20 seconds and then reads all the data fr= om both sockets. I ran tcpdump to watch the packet traffic. Below are pac= ket traces from both a Fedora Core 19 and a FreeBSD 9.2 system. The FreeBS= D trace has what I would expect to see, the Linux trace doesn=92t send any = data on the probes. >>=20 >> Any thoughts on this? >>=20 >> -David Borman ... ---------------------------------------------------------------------- The information contained in this transmission may be confidential. Any dis= closure, copying, or further distribution of confidential information is no= t permitted unless such privilege is explicitly granted in writing by Quant= um. Quantum reserves the right to have electronic communications, including= email and attachments, sent across its networks filtered through anti viru= s and spam software programs and retain such messages in order to comply wi= th applicable data security and retention requirements. Quantum is not resp= onsible for the proper and complete transmission of the substance of this c= ommunication or for any delay in its receipt. From ycheng@google.com Sat Nov 9 09:13:49 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 96E7E11E8137 for ; Sat, 9 Nov 2013 09:13:49 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.76 X-Spam-Level: X-Spam-Status: No, score=-1.76 tagged_above=-999 required=5 tests=[AWL=0.218, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id G3zzFvC14S5k for ; Sat, 9 Nov 2013 09:13:48 -0800 (PST) Received: from mail-ie0-x232.google.com (mail-ie0-x232.google.com [IPv6:2607:f8b0:4001:c03::232]) by ietfa.amsl.com (Postfix) with ESMTP id C4ECE21F9FFF for ; Sat, 9 Nov 2013 09:13:48 -0800 (PST) Received: by mail-ie0-f178.google.com with SMTP id aq17so94501iec.23 for ; Sat, 09 Nov 2013 09:13:48 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=nyM1NkrVrEW0QUZKt1hCrF/Mjhju15BvCePaPWbpvdc=; b=ed/xOxIeYME+XqmmUyjqM0K09Ii9EAku2qirdKzCTx6R8rV2saHLmDdYmd+aZ9+RFi 8mNSstAOWjMR7BhnSTsc3eYLIlPndpoef/AMhB4+XutygQI9FgrlITvrqCmFAg343uqr KczwixLCl2zPiPYiHyNm0CfZEFKOp6CfOYIhvfoKV+fLCDQ5hWabfHXcGFIPQraofukj 5LROclAgLlrjtM/kEWVH7zK45e9NPpQCpACyUq0ZvYN3KqU+tjst90Y1Ht3VdmOKRYQ2 0+zcz+qJQx8HIaXMxVsyW303zBFtVROKQETHHXN9Ql6rAVe5KDzvS33zFUBgg8iIlD84 0DNg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=nyM1NkrVrEW0QUZKt1hCrF/Mjhju15BvCePaPWbpvdc=; b=auaH1OeVq6QwzAXT15JkeCF4rDKA+U+QsTTe3KbdhdlQeBm7HniYRPUx53jp/g06Z+ xQNAd9HGZlTeZ2vTSfS3b8eEQyTLKwV7E7q8DdhQOWDNkQPOLkYliL4A7T3eigksu+s3 pZaQ45uDd7bgB/UIkHDIUtMAIvM0odM+ZLCMZo1FwA/SjlfjX0WTlDl2RU71iZ6kf3BO waymfFE27eOX6E4w0cagUnms8Wza9dE0xYdCw0PcWjdhx7uiOBHJ5nGl4nYdAoomV539 pAp0fhRCnFe/m5+wTCb08N99IeGjDKsNYpAPwpCXik20QK8J2iOz06ucmenZhiFleoet BqUQ== X-Gm-Message-State: ALoCoQkC8gnkAtLZb9Zpcqi2Zlx3kJXo+FvUl/kXAtVUZdWODbyl6MVTSziACivZskVSFGckvfoqJOoisay4+UQlsin6WVEMU1euT6FPs25l+4qhSbFTf19eyGzvteUKB/tB1TpKOxQ2pJxe5MOtd75OE8e1poYVrrtPGJuzPEmimYQ14+x1NsWiXwoinJU5pc9Mumts2VZj X-Received: by 10.51.16.3 with SMTP id fs3mr6842995igd.53.1384017227906; Sat, 09 Nov 2013 09:13:47 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Sat, 9 Nov 2013 09:13:27 -0800 (PST) In-Reply-To: <527B4BDF.60708@kau.se> References: <527B4BDF.60708@kau.se> From: Yuchung Cheng Date: Sat, 9 Nov 2013 09:13:27 -0800 Message-ID: To: Anna Brunstrom Content-Type: text/plain; charset=ISO-8859-1 Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] TCP tail probe and the initial loss behavior X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 09 Nov 2013 17:13:49 -0000 On Thu, Nov 7, 2013 at 12:14 AM, Anna Brunstrom wrote: > Hi Yuchung, > > Thanks for posting the data, that is very interesting! > > If I understand it correctly all percentages are calculated in relation to > the total number of connection traces? If so that makes it a bit difficult yes > to compare the numbers. How many of the connections had 10 or more packets? why would that makes it difficult to compare the "initial" loss rate? all have 10 or more packets so they are relevant to the question we are trying to answer. > > Could you perhaps create an equivalent of the lossdist1 graph that shows the > % of losses in position x in relation to the number of connections that had > a length of x or more packets? > > Thanks, > Anna > > > > On 2013-11-06 22:48, Yuchung Cheng wrote: >> >> We have new data regarding the question raised in the previous two >> meetings: "is tail loss probe useful b/c IW10 is causing more (tail) >> losses". >> >> We got some data on TCP initial loss behavior with IW10. There were >> 15,269 outbound Google server connection traces taken from all over >> the world. Bottleneck bandwidths ranged from 64kbps to 40Mbps with a >> median of ~1Mbps. All the connections used IW10. For each of them, we >> recorded the ending sequence number, packet number and send time of >> the first lost packet (the first retransmitted packet of the >> connection) and the repair time (time from initial send until >> retransmit acked). >> >> 980 connections (6.4%) had one or more losses in the IW10 send. >> "lossdist1.pdf" is a histogram of % losses by position in initial >> window ("none" means no loss in the first 10 packets sent). Claims >> that IW10 "increases the loss rate" are not supported by this data: >> The average loss rate for the first 10 packets is 0.64%. The average >> loss rate for the last 5 packets of the window is only 0.44% compared >> to 0.84% for the first 5 packets (in fact, the loss rate for the first >> packet is as large as packets 6 to 10 combined). This is easier to >> see on a plot of the cumulative loss (lossdist2.pdf): the black line >> is a linear fit to the first five points and the blue to the last five >> points. The lines slopes show that the loss rate consistently >> increases faster for the first 5 packets of the window compared to the >> last 5. >> >> "lossdist3.pdf" shows the distribution of packet vs. cumulative % of >> connections with a first loss at or before that packet. The blue line >> is a fit of cum% to packet^0.25. The quality of the fit over the >> entire range (and the fact that there's no discontinuity between the >> portion of the data that lies within the initial window vs the portion >> outside it) suggests that the single connection loss probabilities >> have nothing to do with IW (at least for IW <= 10). >> >> We didn't graph loss recovery time since it was simple: 78% of the >> initial window losses (761 of 980) were repaired in 1 RTT and the >> remaining 22% were repaired in 2 RTT. None of them took a timeout; all >> were repaired by Fast Retransmit. (This is in contrast to previous IW4 >> measurements where 20% of the losses required a timeout to repair). >> >> >> >> _______________________________________________ >> tcpm mailing list >> tcpm@ietf.org >> https://www.ietf.org/mailman/listinfo/tcpm >> > > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From ycheng@google.com Sat Nov 9 10:06:59 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 60DEF21E8106 for ; Sat, 9 Nov 2013 10:06:59 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.478 X-Spam-Level: X-Spam-Status: No, score=-1.478 tagged_above=-999 required=5 tests=[AWL=-0.100, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_46=0.6, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VERv4brhUZhW for ; Sat, 9 Nov 2013 10:06:57 -0800 (PST) Received: from mail-ie0-x233.google.com (mail-ie0-x233.google.com [IPv6:2607:f8b0:4001:c03::233]) by ietfa.amsl.com (Postfix) with ESMTP id 70B2011E8177 for ; Sat, 9 Nov 2013 10:06:42 -0800 (PST) Received: by mail-ie0-f179.google.com with SMTP id u16so2082889iet.38 for ; Sat, 09 Nov 2013 10:06:42 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:content-transfer-encoding; bh=XYDJAyI6NY8vMuFP/FNuH44cTDaXHf6AeQlamzRHOGg=; b=NKlouZmL60jtDQaTSL0/Zfdy0PcC8iMozS9GhaVjG4AAwZ1DCaj+UaDqe/bhspaOgM XBZi/sUQk9ie3X1EkIVZS5opQ7gSLiZzD8riVt0VGMfYaaIcA/Sxunr+ZLjo8dm0gznv +D1x0hzVvvoNmz3U4b+5EwyyMI0owmpuqZUtQl51tFKu8KrT0bFj1tD2/O+nm+pvbcCD O23bLGBabD1lNjefW9OEeXwYBMxovhY6eNEvNtZNSaGe6AyLna3dZzf60SgYqg7DqRRC +/Bst7DfJ9W1uRO/SWkV+YPinveg/htPr2obdsL5fpY3aB+c8BgZGJNFQzGWxohSQB89 p2ng== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type:content-transfer-encoding; bh=XYDJAyI6NY8vMuFP/FNuH44cTDaXHf6AeQlamzRHOGg=; b=K4xcIBVhJ51HDFiH4dBR7dBQq30UA02QSy1iJsrmxvZePWFItU5kbJLDvqiEQot9AT eLXUYIPlIsF8u+Gd1NfkeZA6+HROi5pzRD3hhdkvRNWrF2rDBuEUvcC9GZYQ+7YeejWN OeirfVzDJ7QxrV/BypmjIwvmw8e1SVHUPI8nSzJNiB7+IFmOSi7N3IHYAy0/IzrnAgiY aKD8P8tG7MSeQj5ZaQaKmuqRxoXC6/IWjyMFV4yb9KJADzaduz7jCGcuhgUZTC3Q8HMp w37Cbq8BTSdeiMdFBB/lWb/6XzI//uvWlRUn84WsKQDiZXBjNUDw3OdgevJwKxE4e2Qi W4pg== X-Gm-Message-State: ALoCoQnIbS6Ey6Pzu94uaYT0KhFiphRMBac9FArIK3TmCb8JeaI8ozIg8Eqd4wx/eKb9yrrLsGTU//uPb2Qk2MtLn282ypXq0J5nsrhd9gWCCSVOqbvF5xK8Y6ewwoGoMpIcSAegCvAf9tIMcCtqaDU9ybkkKzIMy/LXarHPEuItPpipClbglgfp4acJKUjJElg4YTprG+sS X-Received: by 10.50.87.33 with SMTP id u1mr7146198igz.42.1384020401840; Sat, 09 Nov 2013 10:06:41 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Sat, 9 Nov 2013 10:06:20 -0800 (PST) In-Reply-To: <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> From: Yuchung Cheng Date: Sat, 9 Nov 2013 10:06:20 -0800 Message-ID: To: Gorry Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 09 Nov 2013 18:06:59 -0000 Hi Gorry, Sorry for the last response. Too many good sessions and dinner parties during this IETF for me to respond promptly. On Sat, Nov 2, 2013 at 4:09 PM, Gorry wrote: > > I have reviewed the current version of TFO and have the comments below. T= his is rather long ... > > Nits in Section 1 - I think the document here talks of servers and client= s from a TCP endpoint perspective, this is normal for TCP, but the document= also raises implications for apps designers who may have a very different = view on what is intended by the word "server" and "client". To avoid doubt = maybe you should explain this in the intro? They are already explained in the "Terminology" section. > > Page 3, para 2 is not clear. I think Otis just the detail of the wording,= but it would be good to see clear text. > > Last para section 2: to me, the requirements are not clear. > > I think this piece of text means: TCP stacks should NOT enable TCP-TFO by= default. The stack should provide a global flag to enable this. An applica= tion wishing to use this must enable each TCP socket on a per service/per p= ort basis. > > I think the other comments/questions maybe more significant. > > section 2 also suggests the main change is to allow data in SYN, I have n= oted that some people think this is the only change --- it is not. The bigg= er implication is that the document proposes to allow new data sent during = the 3WHS. This change needs to be more clearly noted, sending an IW of dat= a before there is an ACK on the return path is a significant change to TCP = semantics and it's behaviour under heavy load. will revise sec2 and discuss w/ you before we publish the revision. > > Page 5 fast open, bullet 3. : This says the server can send data before t= he 3WHS completes. I am curious what this actually means, is this implied t= hat the use is extricated to standards-track mechanisms, I.e. I am looking = here for some explicit understanding of whether this implies IW3. Or is the= experiment linked to both a proposal for iW10 and this new experiment? will add "initial congestion window of " before the "... the congestion control in .." to make it more clear. or > > Last para 4.1.3 the text says /should include remote port numbers too/ Ca= n we be clear which port numbers are used. Is this the received server port= number by the client? Actually, do not understand what is intended by the = paragraph, can you provide an example to be sure what is meant? a "remote" port number of the client means the "server port". but will replace remote/server to avoid confusion. The paragraph is added because of your suggestion to add negative caching. > > Section 4.3.2. I hoped this would say something explicit about not sendin= g SYN segments that are larger than the default MTU. This seems like a bad = thing to do. It's a bad thing to send any segment larger than the MTU. Why is that specific to Fast Open? > > Why are the examples only for web? This is a TCP spec. Sure there are a lot more applications can benefit from that, but Web is what motivates our work and the application is complex enough to demonstrate many aspects of Fast Open. If you have another good examples, you are welcome to suggest one. > > There are two sections 6.3.2. > > Second Section 6.3.2 - I think this applies to SSL/TLS in general, doesn'= t it? ... I think the section should address the general case, and then pro= vide an example for https, rather than assuming a web-centric use. Yes it is generic, but anyone who knows SSL/TLS good enough knows that. > > Section 7.1 first para last line.please check text, this pathology is no= t necessarily malicious!!! sure we can s/malicious/pathological. does this deserves three !!!? > The issue seems to be mainly related to NAPT where ports are used to disc= riminate senders? > Also the shared NAPT can result in a range of RTT and PMTU to the same IP= , but this is already possible with ECMP routing. > > The lack of issues listed in section 7 raises serious concerns to me. I d= o not list that is provided currently as the only, or even most significant= reasons why this is deemed to be experimental work. Indeed, the lack of id= entification of potential CC issues is why this document is in my opinion N= OT ready to complete a WGLC - if the issues are not listed how can we evalu= ate if the method is safe enough to experiment or can later be evaluated pr= operly. > > Here are some additional issues I would like considered by the group and = if the group is happy these may be safe enough to recommend testing in the = general internet, then I think we should document that these are things to = be confirmed. > > 1) I raised on the list that a negative impact of using the SYN as a prob= e to check for new option support. The examples I gave were ECN interaction= s and IPv6 probing. ECN if used will probe for ECN support in the SYN segme= nt. If we include data on the SYN and then there is a lack of response, it = is assumed by the sender that both TFO and ECN are not supported. This is a= n implication of the model we use to negotiate for features that may be bla= ck-holed. It is not an issue with the method, but I think we need tone clea= r this is something that may happen. A similar case exists if a client atte= mpts TFO to a server using IPv6 ... Is it assumed that the path does not su= pport IPv6 or TFO (or actually both). It does not happen when multiple opti= ons are negotiated. if a SYN-data is dropped, TFO will retransmit a regular SYN (with ECN if us= ed). > > 2) the draft proposes caching RTT per path and using this for SYN timeout= . This updates a standards track RFC, and will have implications in some ne= twork paths that may seriously degrade performance, in other cases it may i= mprove performance. Examples of negative performance are cases where an ini= tial packet can take much longer to process than a subsequent packet. This = can and does occur in bandwidth-on-demand L2 networks, where the first pack= et causes path setup. It also occurs to some extent with policing and routi= ng devices that build cached state from the first packet in a flow. I think= this is one of the reasons why the RFC-series has to date not described a = shortened RTT method for the SYN. > > Another issue is that previous RTT (or PMTU) is not necessarily a good in= dication of future value in a different 5tuple, since ECMP etc can lead to = very different path characteristics for a session with a different port num= ber. The current text therefore proposes an experiment that is beyond what = was previously considered safe. It does not specify a way to avoid the serv= er making the same mistake next connection. We'll just take RTT caching out of the text if that makes people happy. > > 3) seeding the SYN RTT from a lower value makes the sender more aggressiv= e in heavily congested networks. The sender is also made more aggressive by= sending IW data packets before there is any indication the path can contai= n even a single data segment. This is significant change to standard behavi= our. If the proposal is to use IW 3 (as standard) it still needs to be call= ed out. If the proposal is to use a larger experimental number then I have = concerns here that this is a significant change that needs an automated rec= overy method to prevent significant collateral damage on capacity-limited p= aths - there needs to be a way to stop a server doing this each time result= ing in recurring loss! when the IW of packets are not all acked, loss recovery is triggered and window is reduced. this is part of standard CC. also when you use "significant" four times, please back up with a good theory or data or both. and abuse that word does not it make more significant. > > 4) Are there any changes to PMTUD behaviour that need experience? PMTU wa= s naturally cached and I am not sure if this changes the behaviour or not. = It does of course allow a packet with a larger MTU to be sent before the pa= th has been initially validated via the 3WHS (I.e. After a path change whil= e idle, this can generate TCP segments in a SYN that are larger than the r= eceiver advertised MSS = - I suspect this is not crucial, but I would also assume that the SYN itse= lf should not use an increased PMTU value to carry the data. Is this correc= t? in our implementation, we use what ever MTU (possibly discovered in a prior PMTU) cached to send SYN-data. > > Are there other case where this change will impact the network for specif= ic path characteristics? > > Have others thought through these issues and feel they are ok? Or at leas= t ok for experimental deployment? > > Gorry > > > > From prvs=0025cf6a7a=anna.brunstrom@kau.se Sat Nov 9 10:19:58 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 23B1621F9E9D for ; Sat, 9 Nov 2013 10:19:58 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.269 X-Spam-Level: X-Spam-Status: No, score=-2.269 tagged_above=-999 required=5 tests=[AWL=-0.020, BAYES_00=-2.599, HELO_EQ_SE=0.35] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0yhow4-VZzzY for ; Sat, 9 Nov 2013 10:19:53 -0800 (PST) Received: from tiger.dc.kau.se (smtp.kau.se [193.10.220.38]) by ietfa.amsl.com (Postfix) with ESMTP id C094B21F9E9A for ; Sat, 9 Nov 2013 10:19:53 -0800 (PST) X-Spam-Processed: mail.kau.se, Sat, 09 Nov 2013 19:19:21 +0100 (not processed: spam filter heuristic analysis disabled) X-Authenticated-Sender: annabrun@kau.se X-MDRemoteIP: 64.114.24.114 X-Return-Path: anna.brunstrom@kau.se X-Envelope-From: anna.brunstrom@kau.se Message-ID: <527E7CD1.5060608@kau.se> Date: Sat, 09 Nov 2013 19:20:01 +0100 From: Anna Brunstrom User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64; rv:15.0) Gecko/20120824 Thunderbird/15.0 MIME-Version: 1.0 To: Yuchung Cheng References: <527B4BDF.60708@kau.se> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] TCP tail probe and the initial loss behavior X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 09 Nov 2013 18:19:58 -0000 Hi Yuchung, It was not clear to me from your mail that all 15,269 connections were 10 packets or longer. I understood it as a random selection of connections with IW10 enabled, in which case I guess many would have been shorter than 10 packets. As all 15,269 connections were 10 packets or longer the numbers are of course comparable. Thanks, Anna On 2013-11-09 18:13, Yuchung Cheng wrote: > On Thu, Nov 7, 2013 at 12:14 AM, Anna Brunstrom wrote: >> Hi Yuchung, >> >> Thanks for posting the data, that is very interesting! >> >> If I understand it correctly all percentages are calculated in relation to >> the total number of connection traces? If so that makes it a bit difficult > yes >> to compare the numbers. How many of the connections had 10 or more packets? > why would that makes it difficult to compare the "initial" loss rate? > > all have 10 or more packets so they are relevant to the question we > are trying to answer. > >> >> Could you perhaps create an equivalent of the lossdist1 graph that shows the >> % of losses in position x in relation to the number of connections that had >> a length of x or more packets? >> >> Thanks, >> Anna >> >> >> >> On 2013-11-06 22:48, Yuchung Cheng wrote: >>> >>> We have new data regarding the question raised in the previous two >>> meetings: "is tail loss probe useful b/c IW10 is causing more (tail) >>> losses". >>> >>> We got some data on TCP initial loss behavior with IW10. There were >>> 15,269 outbound Google server connection traces taken from all over >>> the world. Bottleneck bandwidths ranged from 64kbps to 40Mbps with a >>> median of ~1Mbps. All the connections used IW10. For each of them, we >>> recorded the ending sequence number, packet number and send time of >>> the first lost packet (the first retransmitted packet of the >>> connection) and the repair time (time from initial send until >>> retransmit acked). >>> >>> 980 connections (6.4%) had one or more losses in the IW10 send. >>> "lossdist1.pdf" is a histogram of % losses by position in initial >>> window ("none" means no loss in the first 10 packets sent). Claims >>> that IW10 "increases the loss rate" are not supported by this data: >>> The average loss rate for the first 10 packets is 0.64%. The average >>> loss rate for the last 5 packets of the window is only 0.44% compared >>> to 0.84% for the first 5 packets (in fact, the loss rate for the first >>> packet is as large as packets 6 to 10 combined). This is easier to >>> see on a plot of the cumulative loss (lossdist2.pdf): the black line >>> is a linear fit to the first five points and the blue to the last five >>> points. The lines slopes show that the loss rate consistently >>> increases faster for the first 5 packets of the window compared to the >>> last 5. >>> >>> "lossdist3.pdf" shows the distribution of packet vs. cumulative % of >>> connections with a first loss at or before that packet. The blue line >>> is a fit of cum% to packet^0.25. The quality of the fit over the >>> entire range (and the fact that there's no discontinuity between the >>> portion of the data that lies within the initial window vs the portion >>> outside it) suggests that the single connection loss probabilities >>> have nothing to do with IW (at least for IW <= 10). >>> >>> We didn't graph loss recovery time since it was simple: 78% of the >>> initial window losses (761 of 980) were repaired in 1 RTT and the >>> remaining 22% were repaired in 2 RTT. None of them took a timeout; all >>> were repaired by Fast Retransmit. (This is in contrast to previous IW4 >>> measurements where 20% of the losses required a timeout to repair). >>> >>> >>> >>> _______________________________________________ >>> tcpm mailing list >>> tcpm@ietf.org >>> https://www.ietf.org/mailman/listinfo/tcpm >>> >> >> _______________________________________________ >> tcpm mailing list >> tcpm@ietf.org >> https://www.ietf.org/mailman/listinfo/tcpm From ycheng@google.com Sat Nov 9 10:26:19 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 658D021F9DD6 for ; Sat, 9 Nov 2013 10:26:19 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.77 X-Spam-Level: X-Spam-Status: No, score=-1.77 tagged_above=-999 required=5 tests=[AWL=0.208, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id eMdOPkXV+lAV for ; Sat, 9 Nov 2013 10:26:18 -0800 (PST) Received: from mail-ie0-x22f.google.com (mail-ie0-x22f.google.com [IPv6:2607:f8b0:4001:c03::22f]) by ietfa.amsl.com (Postfix) with ESMTP id ABF8E21F999C for ; Sat, 9 Nov 2013 10:26:18 -0800 (PST) Received: by mail-ie0-f175.google.com with SMTP id u16so2052581iet.20 for ; Sat, 09 Nov 2013 10:26:17 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=sh1AtxImsxRllsJ0uI50tCfevbqB0zY0pkZIUQuvftM=; b=RI9rFtZTGta52LOe0XPh1b3z08VC67G7z4g7p6pkUpFWAI5YrE7B3BfgdQdesZn8ei PWSj8Q9qIVemYMUkKlIhmDi3u6ks5S/R1YVkjLU1eLkTam3WTjQASuzcPZb14c0oA9Q0 09iZB85VHd60XdFgwT72siLrLMRTHk5XZKTa1tV56CvB02q2JjmsWqXX94BwBjcorgiu cg3BNz6EAIqjCVQOHeiQst/RznEjEjqvs1TRiwQxG8VKzTOjrHxeptvMYRKiZPAUnkwg b2h6NwuXG3SFDUb5POV9mOmLop8nr6NGQEyifYSQayJhD2Gqr3cvmm0c+SvpiA4efA5Q cKcQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=sh1AtxImsxRllsJ0uI50tCfevbqB0zY0pkZIUQuvftM=; b=ZXjhXhkSLsrq4xdh3JmrkcioFsaP4nWyBWsZix7MPeG4h49+sNjlcL62V7YykucXbI MlyFrD80Y3xS5hy8Mxj9L0nkVB2LysEGamxsVK9h0u78sL+OfuZypSd0vGO4k0cZ1urr 1O/WH2QvOlGD253rorsiwEZYvv9UcSt1X/DLtakOaq5g/eY2BKIHdmGrvRLdGB3EpEwy ZSEjjiKyrwVG2jBwsOpElh6O99s/sncVaN5wNzWDSrHzdF+AlOmVMVDQ9WSbPZU/EmWc 7t6vIt1WDKnN+iQAYuEWBW6HnE2ql/O2cQBR3y7YFb4fpmnhtD1vJLLddJ0FIArhpF68 BPyA== X-Gm-Message-State: ALoCoQm1Y05o7bGd676lbt6Gapnl7XWF3hxeSOdn2wzr4aBStVZXfadAA3v92EoZ5BskyXmRxxwLx4k6mqDgZMdIzoG9zRtRHP6fSjP3v5QrCEqZRUtN84mh42IuiOIKn++7aIWYjKiSX0+yOYM8Gj6Y46c6JRhg/Cu55R8qHqrrRL2Z/o4fKVZmasnc3pWboIjcec4n05vL X-Received: by 10.50.56.44 with SMTP id x12mr7110992igp.41.1384021577796; Sat, 09 Nov 2013 10:26:17 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Sat, 9 Nov 2013 10:25:56 -0800 (PST) In-Reply-To: <527E7CD1.5060608@kau.se> References: <527B4BDF.60708@kau.se> <527E7CD1.5060608@kau.se> From: Yuchung Cheng Date: Sat, 9 Nov 2013 10:25:56 -0800 Message-ID: To: Anna Brunstrom Content-Type: text/plain; charset=ISO-8859-1 Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] TCP tail probe and the initial loss behavior X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 09 Nov 2013 18:26:19 -0000 On Sat, Nov 9, 2013 at 10:20 AM, Anna Brunstrom wrote: > Hi Yuchung, > > It was not clear to me from your mail that all 15,269 connections were 10 > packets or longer. I understood it as a random selection of connections with > IW10 enabled, in which case I guess many would have been shorter than 10 > packets. I see. Sorry for not being clear about that in my first email. > > As all 15,269 connections were 10 packets or longer the numbers are of > course comparable. > > Thanks, > Anna > > > On 2013-11-09 18:13, Yuchung Cheng wrote: >> >> On Thu, Nov 7, 2013 at 12:14 AM, Anna Brunstrom >> wrote: >>> >>> Hi Yuchung, >>> >>> Thanks for posting the data, that is very interesting! >>> >>> If I understand it correctly all percentages are calculated in relation >>> to >>> the total number of connection traces? If so that makes it a bit >>> difficult >> >> yes >>> >>> to compare the numbers. How many of the connections had 10 or more >>> packets? >> >> why would that makes it difficult to compare the "initial" loss rate? >> >> all have 10 or more packets so they are relevant to the question we >> are trying to answer. >> >>> >>> Could you perhaps create an equivalent of the lossdist1 graph that shows >>> the >>> % of losses in position x in relation to the number of connections that >>> had >>> a length of x or more packets? >>> >>> Thanks, >>> Anna >>> >>> >>> >>> On 2013-11-06 22:48, Yuchung Cheng wrote: >>>> >>>> >>>> We have new data regarding the question raised in the previous two >>>> meetings: "is tail loss probe useful b/c IW10 is causing more (tail) >>>> losses". >>>> >>>> We got some data on TCP initial loss behavior with IW10. There were >>>> 15,269 outbound Google server connection traces taken from all over >>>> the world. Bottleneck bandwidths ranged from 64kbps to 40Mbps with a >>>> median of ~1Mbps. All the connections used IW10. For each of them, we >>>> recorded the ending sequence number, packet number and send time of >>>> the first lost packet (the first retransmitted packet of the >>>> connection) and the repair time (time from initial send until >>>> retransmit acked). >>>> >>>> 980 connections (6.4%) had one or more losses in the IW10 send. >>>> "lossdist1.pdf" is a histogram of % losses by position in initial >>>> window ("none" means no loss in the first 10 packets sent). Claims >>>> that IW10 "increases the loss rate" are not supported by this data: >>>> The average loss rate for the first 10 packets is 0.64%. The average >>>> loss rate for the last 5 packets of the window is only 0.44% compared >>>> to 0.84% for the first 5 packets (in fact, the loss rate for the first >>>> packet is as large as packets 6 to 10 combined). This is easier to >>>> see on a plot of the cumulative loss (lossdist2.pdf): the black line >>>> is a linear fit to the first five points and the blue to the last five >>>> points. The lines slopes show that the loss rate consistently >>>> increases faster for the first 5 packets of the window compared to the >>>> last 5. >>>> >>>> "lossdist3.pdf" shows the distribution of packet vs. cumulative % of >>>> connections with a first loss at or before that packet. The blue line >>>> is a fit of cum% to packet^0.25. The quality of the fit over the >>>> entire range (and the fact that there's no discontinuity between the >>>> portion of the data that lies within the initial window vs the portion >>>> outside it) suggests that the single connection loss probabilities >>>> have nothing to do with IW (at least for IW <= 10). >>>> >>>> We didn't graph loss recovery time since it was simple: 78% of the >>>> initial window losses (761 of 980) were repaired in 1 RTT and the >>>> remaining 22% were repaired in 2 RTT. None of them took a timeout; all >>>> were repaired by Fast Retransmit. (This is in contrast to previous IW4 >>>> measurements where 20% of the losses required a timeout to repair). >>>> >>>> >>>> >>>> _______________________________________________ >>>> tcpm mailing list >>>> tcpm@ietf.org >>>> https://www.ietf.org/mailman/listinfo/tcpm >>>> >>> >>> _______________________________________________ >>> tcpm mailing list >>> tcpm@ietf.org >>> https://www.ietf.org/mailman/listinfo/tcpm > > From nishida@sfc.wide.ad.jp Mon Nov 11 08:52:44 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 92C0511E8128 for ; Mon, 11 Nov 2013 08:52:44 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -101.897 X-Spam-Level: X-Spam-Status: No, score=-101.897 tagged_above=-999 required=5 tests=[AWL=0.080, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id IHg0cNCxGsOm for ; Mon, 11 Nov 2013 08:52:44 -0800 (PST) Received: from mail.sfc.wide.ad.jp (ns.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by ietfa.amsl.com (Postfix) with ESMTP id 11A8311E8107 for ; Mon, 11 Nov 2013 08:52:43 -0800 (PST) Received: from mail-lb0-x230.google.com (mail-lb0-x230.google.com [IPv6:2a00:1450:4010:c04::230]) by mail.sfc.wide.ad.jp (Postfix) with ESMTPSA id 8CC4627819E for ; Tue, 12 Nov 2013 01:52:41 +0900 (JST) Received: by mail-lb0-f176.google.com with SMTP id p9so935908lbv.7 for ; Mon, 11 Nov 2013 08:52:38 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=mime-version:date:message-id:subject:from:to:content-type; bh=m6bP6XjdrIJn2eh8E3Xmw8tTtjjXn4nYCxZzJvejMnE=; b=L5eBeFdbXXMC5F3+3fHrsj9kDKiegNUGGinA2ZiSvTiEzbOg51CchZt5TS9na1g9rf o1X2MOpI2ewSILYf/Afk4dQF9SgcRXXpBHBCDeIOOrRUfb+HYlDgo7UrE3hG08Y8uwg/ 5fvM5vI1NF4UI6adE+AGx6HmwWrLEWysQ30ib5HR05g42cIPAAPYn/fYYTxR3qzjj905 jX6v1MGh5b0/1K3wYgmyxxpZPfcjtp+/mxk2wkg2/pkU8ggIk1Ec/eh3ibxI2ki3TNpM f/JxV1Gr9fWe8LEOD4uee6dNds/0Z3iAiAbVZ7+mWflq49F/YsOGp1zeLqRCnhqmDzoM gAxQ== MIME-Version: 1.0 X-Received: by 10.112.134.3 with SMTP id pg3mr22536378lbb.11.1384188758944; Mon, 11 Nov 2013 08:52:38 -0800 (PST) Received: by 10.114.99.99 with HTTP; Mon, 11 Nov 2013 08:52:38 -0800 (PST) Date: Mon, 11 Nov 2013 08:52:38 -0800 Message-ID: From: Yoshifumi Nishida To: "tcpm@ietf.org" Content-Type: multipart/alternative; boundary=047d7b343c08477c1a04eae98fd1 Subject: [tcpm] draft minutes for vancouver meeting X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 11 Nov 2013 16:52:44 -0000 --047d7b343c08477c1a04eae98fd1 Content-Type: text/plain; charset=ISO-8859-1 Hi, Thanks for very productive discussions at the vancouver meeting I've just uploaded the draft minutes for the meeting. Please let chairs know if there are suggestions or corrections. We appreciate Richard and David for note taking and Brian for Jabber scribble. http://www.ietf.org/proceedings/88/minutes/minutes-88-tcpm Regards, -- Yoshi, Pasi, Michael --047d7b343c08477c1a04eae98fd1 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Hi,

Thanks for very producti= ve discussions at the vancouver meeting
I've just uploaded th= e draft minutes for the meeting.
Please let chairs know if there = are suggestions or corrections.
We appreciate Richard and David for note taking and Brian for Jabber s= cribble.

http://www.ietf.org/proceedings/88/minutes/minutes= -88-tcpm

Regards,
--
Yoshi, Pasi, Michael
--047d7b343c08477c1a04eae98fd1-- From nishida@sfc.wide.ad.jp Mon Nov 11 17:57:12 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 20CF221E8146 for ; Mon, 11 Nov 2013 17:57:12 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -101.902 X-Spam-Level: X-Spam-Status: No, score=-101.902 tagged_above=-999 required=5 tests=[AWL=0.075, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id m76daOz9WxwS for ; Mon, 11 Nov 2013 17:57:11 -0800 (PST) Received: from mail.sfc.wide.ad.jp (ns.sfc.wide.ad.jp [IPv6:2001:200:0:8803:203:178:142:143]) by ietfa.amsl.com (Postfix) with ESMTP id 5911021E8143 for ; Mon, 11 Nov 2013 17:57:10 -0800 (PST) Received: from mail-lb0-x232.google.com (mail-lb0-x232.google.com [IPv6:2a00:1450:4010:c04::232]) by mail.sfc.wide.ad.jp (Postfix) with ESMTPSA id 3EA2E27814A for ; Tue, 12 Nov 2013 10:57:00 +0900 (JST) Received: by mail-lb0-f178.google.com with SMTP id l4so4124737lbv.9 for ; Mon, 11 Nov 2013 17:56:58 -0800 (PST) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=8+GK7nyRP7Q/Y+QE9cj4tJB7kMvIYvrzBSdF2b04e48=; b=g8T/UqwOOvZmiKeO4KxxVpoAeuW/k5k/+lzkSusJF8qgRCMx/tDtQdml/sKi4On9Ng eZVGuxgsr9LXZU+roWknFr4AVJ6DKCegf9hLT0GyJKa+fUs/olQxFzGXs0+jAeaJZ1Lv Wckq4SF6+6eOkvKtwDWdwpPCedGGMfqx+UkgVwU2N7GjTlng5hnBPVlOASQJ0ztFWxHW 5qVKefnFCmeTSZ+s6lS6GgUS4vsWuB10LD6omoCm1J+3/qaxuhG8ISXG/R3RovdwGkjZ yus6GhQJaDOBSxSEC4g2i2AWWoYDInvJEoD9cNA6WoKFj0ArPOD5wXeFJZE5NT0XixVq ycIQ== MIME-Version: 1.0 X-Received: by 10.152.26.131 with SMTP id l3mr4454912lag.29.1384221418276; Mon, 11 Nov 2013 17:56:58 -0800 (PST) Received: by 10.114.99.99 with HTTP; Mon, 11 Nov 2013 17:56:58 -0800 (PST) In-Reply-To: <20131104172910.32F932639802@lawyers.icir.org> References: <20131104172910.32F932639802@lawyers.icir.org> Date: Mon, 11 Nov 2013 17:56:58 -0800 Message-ID: From: Yoshifumi Nishida To: Mark Allman Content-Type: multipart/alternative; boundary=089e0160ba3aed4bfe04eaf129c0 Cc: "tcpm@ietf.org" Subject: Re: [tcpm] review of 1323bis-15 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 01:57:12 -0000 --089e0160ba3aed4bfe04eaf129c0 Content-Type: text/plain; charset=ISO-8859-1 Hi Mark, On Mon, Nov 4, 2013 at 9:29 AM, Mark Allman wrote: > > > Also, one thing to think about ... Might it be useful to recommend a > random offset to the TS per connection? People have shown it is > possible to get signal from these things and perhaps that is not > useful. I know OpenBSD uses a random offset. Perhaps just as a MAY or > something. I.e., there is no harm in it and so let implementers chew on > it. > I am inclined to this as it is more secure approach, but I have one concern here. If we start using random offset per connection, I think it will be difficult to tell which one is older when we see two TCP connections with the same tuples. This won't be good thing for PAWS as it is expected to provide protections against duplicates from previous connections. I guess we might need something more than purely random value for this. Thanks, -- Yoshifumi --089e0160ba3aed4bfe04eaf129c0 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable

Hi Mark,

On Mon, Nov 4, 2013 at 9:29 AM, Mark Allman <mallman@icir.org> wrote:

Also, one thing to think about ... Might it be useful to recommend a
random offset to the TS per connection? =A0People have shown it is
possible to get signal from these things and perhaps that is not
useful. =A0I know OpenBSD uses a random offset. =A0Perhaps just as a MAY or=
something. =A0I.e., there is no harm in it and so let implementers chew on<= br> it.
=A0
I am inclined to this as it is more = secure approach, but I have one concern here.=A0
If we start usin= g random offset per connection, I think it will be difficult to tell which = one is older when we see two TCP connections with the same tuples. This won= 't be good thing for PAWS as it is expected to provide protections agai= nst duplicates from previous connections.
I guess we might need something more than purely random value for this= .

Thanks,
--
Yoshifumi
=A0

--089e0160ba3aed4bfe04eaf129c0-- From touch@isi.edu Mon Nov 11 20:13:11 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 57FE711E81A5 for ; Mon, 11 Nov 2013 20:13:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.599 X-Spam-Level: X-Spam-Status: No, score=-102.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QaoLYBu+RGoc for ; Mon, 11 Nov 2013 20:13:00 -0800 (PST) Received: from darkstar.isi.edu (darkstar.isi.edu [128.9.128.127]) by ietfa.amsl.com (Postfix) with ESMTP id E087F11E814B for ; Mon, 11 Nov 2013 20:12:59 -0800 (PST) Received: from [75.195.232.169] (169.sub-75-195-232.myvzw.com [75.195.232.169]) (authenticated bits=0) by darkstar.isi.edu (8.13.8/8.13.8) with ESMTP id rAC4CSW7014669 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Mon, 11 Nov 2013 20:12:35 -0800 (PST) Message-ID: <5281AAAE.8000805@isi.edu> Date: Mon, 11 Nov 2013 20:12:30 -0800 From: Joe Touch User-Agent: Mozilla/5.0 (Windows NT 6.3; WOW64; rv:24.0) Gecko/20100101 Thunderbird/24.1.0 MIME-Version: 1.0 To: Yoshifumi Nishida , Mark Allman References: <20131104172910.32F932639802@lawyers.icir.org> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-ISI-4-43-8-MailScanner: Found to be clean X-MailScanner-From: touch@isi.edu Cc: "tcpm@ietf.org" Subject: Re: [tcpm] review of 1323bis-15 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 04:13:11 -0000 On 11/11/2013 5:56 PM, Yoshifumi Nishida wrote: > > Hi Mark, > > On Mon, Nov 4, 2013 at 9:29 AM, Mark Allman > wrote: > > > Also, one thing to think about ... Might it be useful to recommend a > random offset to the TS per connection? People have shown it is > possible to get signal from these things and perhaps that is not > useful. I know OpenBSD uses a random offset. Perhaps just as a MAY or > something. I.e., there is no harm in it and so let implementers chew on > it. > > I am inclined to this as it is more secure approach, IMO, it is inappropriate to suggest that such selection is related to "security". > but I have one > concern here. > If we start using random offset per connection, I think it will be > difficult to tell which one is older when we see two TCP connections > with the same tuples. This won't be good thing for PAWS as it is > expected to provide protections against duplicates from previous > connections. Indeed. Joe From martin.h.duke@gmail.com Mon Nov 11 20:35:32 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4FA9121E8087 for ; Mon, 11 Nov 2013 20:35:32 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id j23RasUAc702 for ; Mon, 11 Nov 2013 20:35:31 -0800 (PST) Received: from mail-vc0-x229.google.com (mail-vc0-x229.google.com [IPv6:2607:f8b0:400c:c03::229]) by ietfa.amsl.com (Postfix) with ESMTP id 547AC11E811D for ; Mon, 11 Nov 2013 20:35:30 -0800 (PST) Received: by mail-vc0-f169.google.com with SMTP id hu8so4067573vcb.28 for ; Mon, 11 Nov 2013 20:35:29 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=otQXHoqz3PT+IFwKOvdIYnVd6lP279YNxyEMjaS/QsE=; b=T5Zqx7n1b0QvnDxG6KgvNB7Tf4QMZD37f38yfX00449Ki0oESutPTIyqvKeOkchynf QKkFbWMyN8yxYxvlEsst650aSCfZjyrm/2CQrt952Q93jJJwVOHnZoayxXqU4TRVOMk5 ah1uRcS223WDGfVlld5OWqpHPbtKVJxq13sYMNFSAyaTuth4D+yEZyb0gUg0GLPVGcxS KlSwsxIWCO1BecxWjADAdnHu339RR/zzh3kHyih7CjgugHP7WMg6L1L8gN7y8jEmLA0q ob2zknXXOh0FriBIyAlBr8jt590p8njcj7L/BZzfIg4GAAnHk53kWKbpX5PMhMOw6uh7 3Zbw== MIME-Version: 1.0 X-Received: by 10.221.51.206 with SMTP id vj14mr27376897vcb.17.1384230929782; Mon, 11 Nov 2013 20:35:29 -0800 (PST) Received: by 10.220.121.198 with HTTP; Mon, 11 Nov 2013 20:35:29 -0800 (PST) Date: Mon, 11 Nov 2013 20:35:29 -0800 Message-ID: From: Martin Duke To: tcpm@ietf.org Content-Type: multipart/alternative; boundary=001a11332252db601d04eaf3601c Subject: [tcpm] RFC 4614 bis comments X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 04:35:32 -0000 --001a11332252db601d04eaf3601c Content-Type: text/plain; charset=ISO-8859-1 Alexander, First of all, let me emerge from deep lurking to thank you for updating this document. It is long overdue. Although I haven't been as involved in TCP work as I would have liked, perhaps I can offer some constructive comments based on the original philosophy of the RFC's organization. Martin ******** EDITORIAL ISSUES: Sec. 2: If RFC 5681 contains 3390 is 3390 a "core requirement?" It seems like a novice could skip over it entirely. 5681 even refers to 3390 for further background it they're interested. Recommendation: move RFC 3390 to Section 6 (Support Documents). Sec 3: I like how this section has been reorganized. RFC 3042, 3465: At a minimum we should indicate this has been rolled into RFC 5681. At most we could strike the reference entirely. Recommendation: Move to Section 6. Section 3.4: So I take it that, based on its inclusion in Section 3, we specifically recommend F-RTO and Eifel Response over the alternatives? If that is not the intent here than this section must be substantially revised. Sec 6.6 Generally we do not separately list obsolete RFCs in this document. >From the text I gather that everything but RFC 4022 is obsolete in this section, so I recommend we keep the introductory text and the entry for RFC 4022. **** NITS: P.3 para 6: "...important implementation practices that ARE not currently documented..." RFC 6298, last sentence is worded ambiguously. Suggest "RFC 6298 updates RFC 2988 by changing the initial RTO from 3s to 1s." (It can currently be read to say that 2988 sets it to 1s.) Sec 3.4. "implemented, and IPR status" -> "implemented, and concerns about IPR status" RFC 5682: "describes" -> "described" Sec 3.6. "Header compressions schemes... these overhead." --> "Header compression schemes for TCP/IP like "RObust Header Compression (ROHC) can significantly compress this overhead." Sec 3.7. RFC 4953. "resets (RSTs)" -> "reset (RST)" RFC 5961. "are less vulnerable" -> "less vulnerable" Sec. 4.1 RFC 3124 : "which supports sharing information among TCP connections as well" delete the words "as well". Taking it outside of connections is not formally the same as sharing it between connections. Sec. 4.4. "simultaneous" -> "simultaneously"; "additinal" -> "additional" Sec. 5 "reclassifies" -> "reclassified"; "seen widespread" -> "seen widespread deployment" RFC 721: "has not been included" -> "was not included" RFC 1078: "propose a protocol" -> "proposes a protocol" RFC 6013: para 1: "to negotiates" -> "to negotiate"; para 2: "describes method to extent" -> "describes a method to extend" para 3: "Although the RFC 6013 is publish in 2011" -> "Although RFC 6013 was published in 2011" Section 6. "implementation advices can be found in Section 6.5" -> "implementation advice can be found in Section 6.5" Section 6.1. Is RFC 675 really "infamous?" Suggest "fundamental". Section 6.2 RFC 3439 "extents" -> "extends" Sec 6.4 RFC 5166 "Among others" -> "Among other topics" Sec 6.7 RFC 5783 "The focus of the document are on" -> "The focus of the document is" Sec 6.8 RFC 1337 "Hazardsin" --> "Hazards in" --001a11332252db601d04eaf3601c Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable
Alexander,

First of all,= let me emerge from deep lurking to thank you for updating this document. I= t is long overdue. Although I haven't been as involved in TCP work as I= would have liked, perhaps I can offer some constructive comments based on = the original philosophy of the RFC's organization.

Martin

********
EDIT= ORIAL ISSUES:
Sec. 2: If RFC 5681 contains 3390 is 3390 a "c= ore requirement?" It seems like a novice could skip over it entirely. = 5681 even refers to 3390 for further background it they're interested. = Recommendation: move RFC 3390 to Section 6 (Support Documents).

Sec 3: I like how this section has been reor= ganized.

RFC 3042, 3465: At a minimum we should in= dicate this has been rolled into RFC 5681. At most we could strike the refe= rence entirely. Recommendation: Move to Section 6.

Section 3.4: So I take it that, based on its incl= usion in Section 3, we specifically recommend F-RTO and Eifel Response over= the alternatives? If that is not the intent here than this section must be= substantially revised.

Sec 6.6 Generally we do not separately list obsol= ete RFCs in this document. From the text I gather that everything but RFC 4= 022 is obsolete in this section, so I recommend we keep the introductory te= xt and the entry for RFC 4022.

****
NITS:

P.3= para 6: "...important implementation practices that ARE not currently= documented..."

RFC 6298, last sentence is wo= rded ambiguously. Suggest "RFC 6298 updates RFC 2988 by changing the i= nitial RTO from 3s to 1s." (It can currently be read to say that 2988 = sets it to 1s.)

Sec 3.4. "implemented, and IPR status" = -> "implemented, and concerns about IPR status"
=
RFC 5682: "describes" -> "described"<= /div>

Sec 3.6. "Header compressions schemes... these ove= rhead." --> "Header compression schemes for TCP/IP like "= RObust Header Compression (ROHC) can significantly compress this overhead.&= quot;

Sec 3.7.
RFC 4953. "resets (RSTs)&= quot; -> "reset (RST)"

RFC 5961. &quo= t;are less vulnerable" -> "less vulnerable"

Sec. 4.1 RFC 3124 : "which supports sharing information= among TCP connections as well" delete the words "as well". = Taking it outside of connections is not formally the same as sharing it bet= ween connections.

Sec. 4.4. "simultaneous" -> "simultan= eously"; "additinal" -> "additional"
=
Sec. 5 "reclassifies" -> "reclassified&quo= t;; "seen widespread" -> "seen widespread deployment"= ;

RFC 721: "has not been included" -> "= was not included"

RFC 1078: "propose a p= rotocol" -> "proposes a protocol"

RFC 6013: para 1: "to negotiates" -> "to negotiate&q= uot;;=A0
para 2: "describes method to extent" -> &qu= ot;describes a method to extend"
para 3: "Although the = RFC 6013 is publish in 2011" -> "Although RFC 6013 was publish= ed in 2011"

Section 6. "implementation advices can be found in= Section 6.5" -> "implementation advice can be found in Sectio= n 6.5"

Section 6.1. Is RFC 675 really "i= nfamous?" Suggest "fundamental".

Section 6.2 RFC 3439 "extents" -> "ex= tends"

Sec 6.4 RFC 5166 "Among others&qu= ot; -> "Among other topics"

Sec 6.7 R= FC 5783 "The focus of the document are on" -> "The focus = of the document is"

Sec 6.8 RFC 1337 "Hazardsin" --> &qu= ot;Hazards in"
--001a11332252db601d04eaf3601c-- From dab@weston.borman.com Tue Nov 12 05:37:32 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9619B21E8125 for ; Tue, 12 Nov 2013 05:37:25 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, J_CHICKENPOX_38=0.6] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 4dcsXYBy6TJz for ; Tue, 12 Nov 2013 05:37:21 -0800 (PST) Received: from frantic.weston.borman.com (frantic.weston.borman.com [70.57.156.33]) by ietfa.amsl.com (Postfix) with ESMTP id 5E37B21E8122 for ; Tue, 12 Nov 2013 05:37:21 -0800 (PST) Received: from [IPv6:::1] (frantic.weston.borman.com [70.57.156.33]) by frantic.weston.borman.com (8.14.7/8.14.7) with ESMTP id rACDbB1b029158; Tue, 12 Nov 2013 07:37:13 -0600 (CST) Content-Type: text/plain; charset=iso-8859-1 Mime-Version: 1.0 (Mac OS X Mail 7.0 \(1822\)) From: David Borman In-Reply-To: Date: Tue, 12 Nov 2013 07:37:11 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: References: <20131104172910.32F932639802@lawyers.icir.org> To: Yoshifumi Nishida , Richard Scheffenegger X-Mailer: Apple Mail (2.1822) Cc: "tcpm@ietf.org" , Mark Allman Subject: Re: [tcpm] review of 1323bis-15 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 13:37:36 -0000 On Nov 11, 2013, at 7:56 PM, Yoshifumi Nishida = wrote: >=20 > Hi Mark, >=20 > On Mon, Nov 4, 2013 at 9:29 AM, Mark Allman wrote: >=20 > Also, one thing to think about ... Might it be useful to recommend a > random offset to the TS per connection? People have shown it is > possible to get signal from these things and perhaps that is not > useful. I know OpenBSD uses a random offset. Perhaps just as a MAY = or > something. I.e., there is no harm in it and so let implementers chew = on > it. > =20 > I am inclined to this as it is more secure approach, but I have one = concern here.=20 > If we start using random offset per connection, I think it will be = difficult to tell which one is older when we see two TCP connections = with the same tuples. This won't be good thing for PAWS as it is = expected to provide protections against duplicates from previous = connections. > I guess we might need something more than purely random value for = this. The per connection random offset is already there on page 36, = Snd.TSoffset, but there is no text in the main body of the document. Perhaps the end of section 4.4 should have added to it something like: o A random offset may be added to the timestamp clock on a per connection basis. See RFC 6528, section 3, on randomizing the ISN. The same function with a different secret key can be use to generate the per connection timestamp offset. -David Borman >=20 > Thanks, > -- > Yoshifumi > =20 >=20 > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From mallman@icir.org Tue Nov 12 06:02:54 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F0CFE11E8140 for ; Tue, 12 Nov 2013 06:02:54 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.449 X-Spam-Level: X-Spam-Status: No, score=-106.449 tagged_above=-999 required=5 tests=[AWL=0.150, BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id QIHUM5oGOU1A for ; Tue, 12 Nov 2013 06:02:49 -0800 (PST) Received: from fruitcake.ICSI.Berkeley.EDU (fruitcake.ICSI.Berkeley.EDU [192.150.186.11]) by ietfa.amsl.com (Postfix) with ESMTP id 19DF211E813D for ; Tue, 12 Nov 2013 06:02:49 -0800 (PST) Received: from lawyers.icir.org (envoy.icir.org [192.150.187.30]) by fruitcake.ICSI.Berkeley.EDU (8.12.11.20060614/8.12.11) with ESMTP id rACE2lMO020195; Tue, 12 Nov 2013 06:02:47 -0800 (PST) Received: from lawyers.icir.org (localhost [127.0.0.1]) by lawyers.icir.org (Postfix) with ESMTP id 9E281271A366; Tue, 12 Nov 2013 09:02:43 -0500 (EST) To: Yoshifumi Nishida From: Mark Allman In-Reply-To: Organization: International Computer Science Institute (ICSI) Song-of-the-Day: Won't Get Fooled Again X-URL-0: http://www.icir.org/mallman-files/Document86130.html MIME-Version: 1.0 Content-Type: multipart/signed; boundary="--------ma13569-1"; micalg=pgp-sha1; protocol="application/pgp-signature" Date: Tue, 12 Nov 2013 09:02:43 -0500 Sender: mallman@icir.org Message-Id: <20131112140243.9E281271A366@lawyers.icir.org> Cc: "tcpm@ietf.org" Subject: Re: [tcpm] review of 1323bis-15 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: mallman@icir.org List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 14:02:55 -0000 ----------ma13569-1 Content-Type: text/plain Content-Disposition: inline > If we start using random offset per connection, I think it will be > difficult to tell which one is older when we see two TCP connections > with the same tuples. This won't be good thing for PAWS as it is > expected to provide protections against duplicates from previous > connections. So, OK, fair enough. (1) I believe---without going back to the draft right now---that 1323bis says PAWS can help protect against old segments from a previous incarnation of the connection. However, this is not part of the standard function and requires additional assumptions. (2) And, further, this is in service of dropping the MSL. I.e., if we don't re-use the tuple within the specified time frame then none of this should matter. (3) Joe, we all understand that this is not some sort of iron clad security. But, just like randomizing ephemeral port choices and ISNs and IPIDs and whatnot it seems consistent with our notion of making such things unpredictable. (4) Some implementations are in fact applying random TS offsets. If we really think this is a bad idea because of PAWS then maybe we should say that. If folks generally feel that random offsets weakens PAWS, I am not going to fight here. If Joe wants to loudly shout what we already know about security, I am not going to fight. It seems to me like something that is consistent with SOP and that we could sketch as a MAY and leave it to implementers, but I am not going to fight about it. allman ----------ma13569-1 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (Darwin) iEYEARECAAYFAlKCNQMACgkQWyrrWs4yIs5sxQCfbU5kZchU4aXQpfwkhOiiF6kw sYEAnRjNStdCDJ9SVmCtSjoztX4eVRu3 =kY1j -----END PGP SIGNATURE----- ----------ma13569-1-- From rs@netapp.com Tue Nov 12 10:28:34 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0909F11E815C for ; Tue, 12 Nov 2013 10:28:34 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.451 X-Spam-Level: X-Spam-Status: No, score=-5.451 tagged_above=-999 required=5 tests=[AWL=-3.452, BAYES_00=-2.599, J_CHICKENPOX_35=0.6] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0qb2p8BB8FNj for ; Tue, 12 Nov 2013 10:28:29 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 06C0511E810B for ; Tue, 12 Nov 2013 10:28:28 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,686,1378882800"; d="scan'208";a="74059914" Received: from vmwexceht01-prd.hq.netapp.com ([10.106.76.239]) by mx11-out.netapp.com with ESMTP; 12 Nov 2013 10:28:26 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht01-prd.hq.netapp.com ([10.106.76.239]) with mapi id 14.03.0123.003; Tue, 12 Nov 2013 10:28:21 -0800 From: "Scheffenegger, Richard" To: "mallman@icir.org" , "tcpm@ietf.org" Thread-Topic: [tcpm] review of 1323bis-15 Thread-Index: AQHO2YRNtvv/BV/OuUuBwm532/vjlpohvhLQ Date: Tue, 12 Nov 2013 18:28:13 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E8BB4B@SACEXCMBX02-PRD.hq.netapp.com> References: <20131104172910.32F932639802@lawyers.icir.org> In-Reply-To: <20131104172910.32F932639802@lawyers.icir.org> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.53] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [tcpm] review of 1323bis-15 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 18:28:34 -0000 Hi Mark, thanks for your comments. Richard Scheffenegger =20 > - I'd remove the mention of RTTM in the abstract. It is downplayed in > the text. I replace the "are used" by "can be used", and reversed the order of PAWS a= nd RTTM. > - Sec 1: "has resulted" --> "have resulted" Check > - Sec 1: In the last paragraph before 1.1: >=20 > - should "recommended" be "RECOMMENDED"? (I am not sure) This text is before the definition of RFC2119 style wording; I tend to keep= this lower case... > - perhaps a mention here that the changes from 1323 are partially > due to our changed understanding of how things work (i.e., this is > not just a set of bug fixes) Added this to the 2nd before last paragraph of section 1. > - Sec 1.1: I found this a bit strange. The two "fundamental problems" > seem right to me. But, this document will go on to develop > PAWS---which addresses neither problem directly. You might develop > the notion that making the windows bigger causes *another* > fundamental problem and that is too quickly wrapping. And, > therefore, that problem must be tackled and this document does so. Hmm... sec 1.1 is about performance; PAWS is not a performance issue but a = integrity (reliability) issue. As 1323bis also defines Window Scale, that a= ppears to address the performance issue described in 1.1 (1);=20 But I think you are correctly pointing out that the use WS will make it mor= e likely for wrap around to occur, thus PAWS is necessary. I've added a sen= tence to point this out. > - Sec 1.2, point (1): This is not entirely correct. You need both a > high transfer rate *and a whole bunch of data* to have a sequence > wrapping issue. This point could be a lot better by noting that it > takes 4GB of data to wrap. Sentence now reads: At a high enough transfer rate of large volume of data = (at least 4 GiB in the same session), the 32-bit sequence space may be wrap= ped within the time that a sement is delayed in queues. Do you think this is good enough? > - Sec 1.2, the paragraph starting "Duplicates from earlier ....": You > might also throw in there that randomizing ephemeral ports can also > help probabilistically reduce the chances of duplicates from > previous connections muddying the waters. Check. > - Sec 1.3: "be crashed by" --> "crash on" check > - Sec 1.3 (and other places): I'd like to see the TS option be > discussed as the 10 byte option that it is. It is fine to note that > often it consumes 12 bytes because passing is needed. However, > there is a disconnect in the document between the actual > specification (10 bytes) and the discussion (12 bytes) that could > lead to confusion. Fixed this instance; i couldn't find other text that incorrectly states the= option to be 12 bytes in size; only the Appendix A shows the 12-byte layou= t recommendation (but without much text). =20 > - Sec 1.3: I'd remove the word "optimal". I don't think anyone ever > claimed that any RTO scheme---TS based or non-TS based---ever > produced an "optimal" RTO. The statement should be precise and say > that additional RTT samples per RTT is shown to have little effect > on the ultimate RTO when compared to one sample per RTT. Check. > - Sec 2.2: "to do both" --> "to both" Check > - Sec 2.2: "is never scaled." --> "MUST NOT be scaled." Check > - Sec 2.3: The list at the beginning of the section should be > converted to standards language (MUST, SHOULD, etc.). Added mostly MUSTs (only one SHOULD on the receiver, which may choose to li= mit the window for memory constrains etc). > - Sec 2.3: I find this notion in the first bullet that the window > should be stored as 32-bit quantities to be weird. The maximum size > for advertised windows is 30 bits. So, why specify the use of 32 > bit memory locations in the control block? I mean, that is the > natural thing to do, but it just adds a little confusion to things > without really helping the implementer at all. Commented this point out in the xml. (I guess, current OS may want to store= this all in 64bit quanta even). > - Sec 2.3: "This is safe as a sender can always choose to only > partially use any signaled receive window." That is correct, but it > takes a leap from the previous discussion to figure out what you > mean. I'd flesh this out a bit more so the reader doesn't have to > scratch their head. I.e., if the receiver is scaling by X>14 and > the receiver is only scaling by 14 then the window will appear > smaller than it is in reality. Check. > - Sec 3.1: The third paragraph doesn't make much sense, or at least > isn't precise. You are trying to make a clear distinction with > imprecise words. :-) > > I'd say you draw a distinction between the timestamp option that > *conveys* timestamp information and the *use* of that information > (e.g., by PAWS or by RTTM or by Eifel or whatnot). In particular, > not all the mechanisms that use *timestamps* use "the RTT signal as > input". E.g., PAWS does not. So, you are trying to push a > distinction that doesn't make sense. Reworded; I'm not sure if the new wording is satisfactory though (more/spec= ific text donations gladly accepted): It is necessary to remember that there is a distinction between the= =20 Timestamps option conveying timestamp information, and the use of t= hat=20 information. In particular, the Round Trip Time Measurement (RTTM)= =20 mechanism must be viewed independently from updating the Retransmis= sion=20 Timeout (RTO) (see ). In this case, the sa= mple=20 granularity also needs to be taken into account. Other mechanisms, = such=20 as PAWS, or Eifel, are not built upon the timestamp information its= elf,=20 but are based on the intrinsic property of monotonically increasing= values. =20 > - Sec 3.1: "option is useful" --> "option may be useful" (i.e., it > depends on the balancing of costs and benefits and so I don't think > an absolute statement is warranted) Check > - Sec 3.1: "Eifel ([cites]) and others" --> I'd add cites for "and > others". Or, drop it. This is a way to try to make an argument > that there are scads of these things and it isn't worth listing them > all, when in reality there are not. Added three references that I know offhand, where the timestamps option is = used, or at least improves the mechanisms. > - Sec 3.2: The first paragraph feels out of place. Before you even > define timestamps you are telling us how they are used. I'd move > this to later in the section. I'm unsure; to me, this paragraph reads like a very short summary of the en= tire section. It has been carried forward in this form and position from 13= 23. (also, it provides the rationale why the rfc1072 options were supercede= d). > Sec 3.2: This whole discussion of valid and invalid TSecr is > confusing. You note that when the value is invalid it should be set > to zero. But, a TSecr of zero does not mean the TS is invalid. So, > how should we tell? This was my attempt at "liberal at what you accept, and conservative in wha= t you send". There are implementations that assume TSecr=3D0 =3D=3D invalid= , with not strictly defined behavior in that case (e.g. drop that segment a= nd stall until the sender timestamp increases to 1, or end up with strange = RTO values...). When TSval is more or less the uptime of the sender, this r= oll-over hardly matters [which is probably the reason this was not fixed a = long time ago in the affected stack], but when TSval is randomized, this co= rner case is more likely to occur and the implementer should properly treat= this case.=20 If anyone can think of better wording to express this, please donate text. > - Sec 3.2: >=20 > "A TCP MAY send the Timestamps option (TSopt) in an initial > segment (i.e., segment containing a SYN bit and no ACK bit), and > MAY send a TSopt in other segments only if it received a TSopt in > the initial or segment for the connection." >=20 > That second MAY strikes me as totally wrong. Shouldn't it be a > MUST? Didn't the WG decide that once timestamps are turned on (in > the 3WHS) that they must always be used? The next paragraph seems > to back up this notion of using a MUST here. In fact, I am not > entirely sure why there are two paragraphs here. They seem to at > least largely overlap. Too much editing in this space since around -06... sorry. That paragraph now describes just the handshake , case, wher= e the receiver is still free to choose not to send TSopt. The next paragrap= h handles strictly the "established" case. > - Likewise there is a notion that segments without timestamps "SHOULD" > be dropped. But, if we're going to enforce things maybe that should > also be a "MUST"? We had a discussion around this; the consensus at the time was, to allow ex= isting implementations claim compliance, to have a SHOULD here (most will a= ccept segments, which all of a sudden don't have TSopt any more; in the cas= e of FreeBSD, this was added as a feature - apparently some middleboxes ate= TSopt during path changes, leading to consternation... =20 > - Sec 3.3: It'd be nice to see references to this middlebox behavior > you sketch. Even if just to a mailing list discussion it'd be more > concrete than the nebulous language in there now. This comment seems to be around paragraph 9, sec 3.2; added a referecence t= o Andre's comment to that effect, why FreeBSD adheres to a SHOULD (again). > - I would suggest splitting 3.3 & 3.4 off into their own section. You > say you are splitting the TS mechanism from its use, but then you > don't in the document. I think a new section would make this > distinction more clear. I.e., the new section could start with "One > use for the timestamp option is ...". And, the PAWS section could > similarly start with "Another use for the timestamp option is ...". Check. > - Sec 3.3: Since you have a section on updating the RTO you should > move all the discussion of updating the RTO there. I.e., you should > not have it scattered. check >=20 > - Sec 3.4: Swap the first two paragraphs. Check.=20 > - Sec 3.4: You say one way to deal with the problem of the gains is > addressed in the appendix. Fine. But, it might be useful to stake > out a principle here. E.g., something like "in the face of multiple > RTT samples per RTT an implementation SHOULD try to adhere to the > spirit of the history specified in [RFC6298]". Something like > that. I.e., throw out at least a high level goal. Check. > - Sec 4.2: "PAWS uses the same TCP Timestamps option as the RTTM > mechanism" --> PAWS uses the TCP Timestamps option" Check > - Sec 4.2: "In both the PAWS and the RTTM mechanism," --> "In the PAWS > mechanism," Check > - Sec 4.2: The first few paragraphs are just a mess. Get rid of the > references to RTTM. PAWS does not and never did depend on RTTM. > Change this to the correct and more accurate "timestamp option" sort > of language. Tried to fix this. > - Sec 4.2: What is 'snd.ts.ok'??? This is seemingly important, but > not defined. (It is defined in an appendix, but that is not good > enough for a normative discussion.) Added this to the Timestamps option section... > - Sec 4.2: >=20 > When an segment is received, it MUST NOT be subjected to > PAWS checks, and information from the Timestamps option MUST NOT > be used to update connection state information. SEG.TSecr MAY be > used to provide stricter acceptance checks. >=20 > Don't these two sentences contradict each other? They are at least > not clear. PAWS works on Tsval; added wording to point this out. > - Sec 4.4: "to measure RTTs" .... PAWS does not measure RTTs. Get rid > of this. Updated wording >=20 > - Sec 4.4: "internet" --> "Internet" check > Also, one thing to think about ... Might it be useful to recommend a > random offset to the TS per connection? People have shown it is possible > to get signal from these things and perhaps that is not useful. I know > OpenBSD uses a random offset. Perhaps just as a MAY or something. I.e., > there is no harm in it and so let implementers chew on it. Used Davids text suggestion at the appropriate section. > I hope that is useful! Very! I'll update the formating (page flow) and post an update shortly. Best regards, Richard Scheffenegger From internet-drafts@ietf.org Tue Nov 12 11:07:30 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 187BB21E80C5; Tue, 12 Nov 2013 11:07:30 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.587 X-Spam-Level: X-Spam-Status: No, score=-102.587 tagged_above=-999 required=5 tests=[AWL=0.013, BAYES_00=-2.599, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id THTF9g9w1LFm; Tue, 12 Nov 2013 11:07:28 -0800 (PST) Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id D7B6321E8091; Tue, 12 Nov 2013 11:07:28 -0800 (PST) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: internet-drafts@ietf.org To: i-d-announce@ietf.org X-Test-IDTracker: no X-IETF-IDTracker: 4.83 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20131112190728.28759.75122.idtracker@ietfa.amsl.com> Date: Tue, 12 Nov 2013 11:07:28 -0800 Cc: tcpm@ietf.org Subject: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 19:07:30 -0000 A New Internet-Draft is available from the on-line Internet-Drafts director= ies. This draft is a work item of the TCP Maintenance and Minor Extensions Work= ing Group of the IETF. Title : TCP Extensions for High Performance Author(s) : David Borman Bob Braden Van Jacobson Richard Scheffenegger Filename : draft-ietf-tcpm-1323bis-16.txt Pages : 49 Date : 2013-11-12 Abstract: This document specifies a set of TCP extensions to improve performance over paths with a large bandwidth * delay product and to provide reliable operation over very high-speed paths. It defines TCP options for scaled windows and timestamps. The timestamps can be used for two distinct mechanisms, PAWS (Protection Against Wrapped Sequences) and RTTM (Round Trip Time Measurement). This document obsoletes RFC 1323 and describes changes from it. The IETF datatracker status page for this draft is: https://datatracker.ietf.org/doc/draft-ietf-tcpm-1323bis There's also a htmlized version available at: http://tools.ietf.org/html/draft-ietf-tcpm-1323bis-16 A diff from the previous version is available at: http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-1323bis-16 Please note that it may take a couple of minutes from the time of submission until the htmlized version and diff are available at tools.ietf.org. Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ From rs@netapp.com Tue Nov 12 11:44:04 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0EB0511E8153 for ; Tue, 12 Nov 2013 11:44:04 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -9.535 X-Spam-Level: X-Spam-Status: No, score=-9.535 tagged_above=-999 required=5 tests=[AWL=1.064, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qeaLfqAbR+Ot for ; Tue, 12 Nov 2013 11:43:59 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id D64FA11E814D for ; Tue, 12 Nov 2013 11:43:59 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,687,1378882800"; d="scan'208";a="114682875" Received: from vmwexceht04-prd.hq.netapp.com ([10.106.77.34]) by mx12-out.netapp.com with ESMTP; 12 Nov 2013 11:43:59 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht04-prd.hq.netapp.com ([10.106.77.34]) with mapi id 14.03.0123.003; Tue, 12 Nov 2013 11:43:59 -0800 From: "Scheffenegger, Richard" To: "tcpm@ietf.org" , "mallman@icir.org" , Fernando Gont Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt Thread-Index: AQHO39qr2jLQB+w1qUGbkh36wsXn+poh+oHg Date: Tue, 12 Nov 2013 19:43:58 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E8C35B@SACEXCMBX02-PRD.hq.netapp.com> References: <20131112190728.28759.75122.idtracker@ietfa.amsl.com> In-Reply-To: <20131112190728.28759.75122.idtracker@ietfa.amsl.com> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.53] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 12 Nov 2013 19:44:04 -0000 Hi group, As mentioned, I have incorporated all the changes from the latest review of= this draft, except where noted (in the earlier email). In addition, I fixed all occurances of "monotonically increasing" to "monot= onically non-decreasing" (never decreases, but may stay the same) in case s= omeone mistakenly reads the first reference identical to "strictly monotoni= cally increasing" (ie. always has to increase). And I added a privacy section as a subsection of security. Fernando, could = you please check if the paragraphs there align with your understanding of t= hese issues? As some paragraphs have shuffled to new places, you may want to review the = full document rather than the diffs, to learn how the text reads in full co= ntext. But even reviews of just the diff are fine. As a remainder, with the following feedback I was not sure how to address t= he, and WG discussion might be helpful: ------------ > - Sec 1: In the last paragraph before 1.1: >=20 > - should "recommended" be "RECOMMENDED"? (I am not sure) This text is before the definition of RFC2119 style wording; I tend to keep= this lower case... > - Sec 1.2, point (1): This is not entirely correct. You need both a > high transfer rate *and a whole bunch of data* to have a sequence > wrapping issue. This point could be a lot better by noting that it > takes 4GB of data to wrap. Sentence now reads: At a high enough transfer rate of large volume of data = (at least 4 GiB in the same session), the 32-bit sequence space may be wrap= ped within the time that a sement is delayed in queues. Do you think this is good enough? > - Sec 2.3: The list at the beginning of the section should be > converted to standards language (MUST, SHOULD, etc.). Added mostly MUSTs (only one SHOULD on the receiver, which may choose to li= mit the window for memory constrains etc). > Sec 3.2: This whole discussion of valid and invalid TSecr is > confusing. You note that when the value is invalid it should be set > to zero. But, a TSecr of zero does not mean the TS is invalid. So, > how should we tell? This was my attempt at "liberal at what you accept, and conservative in wha= t you send". There are implementations that assume TSecr=3D0 =3D=3D invalid= , with not strictly defined behavior in that case (e.g. drop that segment a= nd stall until the sender timestamp increases to 1, or end up with strange = RTO values...). When TSval is more or less the uptime of the sender, this r= oll-over hardly matters [which is probably the reason this was not fixed a = long time ago in the affected stack], but when TSval is randomized, this co= rner case is more likely to occur and the implementer should properly treat= this case.=20 If anyone can think of better wording to express this, please donate text. ------------ Richard Scheffenegger > -----Original Message----- > From: tcpm-bounces@ietf.org [mailto:tcpm-bounces@ietf.org] On Behalf Of > internet-drafts@ietf.org > Sent: Dienstag, 12. November 2013 20:07 > To: i-d-announce@ietf.org > Cc: tcpm@ietf.org > Subject: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt >=20 >=20 > A New Internet-Draft is available from the on-line Internet-Drafts > directories. > This draft is a work item of the TCP Maintenance and Minor Extensions > Working Group of the IETF. >=20 > Title : TCP Extensions for High Performance > Author(s) : David Borman > Bob Braden > Van Jacobson > Richard Scheffenegger > Filename : draft-ietf-tcpm-1323bis-16.txt > Pages : 49 > Date : 2013-11-12 >=20 > Abstract: > This document specifies a set of TCP extensions to improve > performance over paths with a large bandwidth * delay product and to > provide reliable operation over very high-speed paths. It defines > TCP options for scaled windows and timestamps. The timestamps can be > used for two distinct mechanisms, PAWS (Protection Against Wrapped > Sequences) and RTTM (Round Trip Time Measurement). >=20 > This document obsoletes RFC 1323 and describes changes from it. >=20 >=20 > The IETF datatracker status page for this draft is: > https://datatracker.ietf.org/doc/draft-ietf-tcpm-1323bis >=20 > There's also a htmlized version available at: > http://tools.ietf.org/html/draft-ietf-tcpm-1323bis-16 >=20 > A diff from the previous version is available at: > http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-1323bis-16 >=20 >=20 > Please note that it may take a couple of minutes from the time of > submission until the htmlized version and diff are available at > tools.ietf.org. >=20 > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ >=20 > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From michael.scharf@alcatel-lucent.com Wed Nov 13 08:59:43 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 06BA121E812C for ; Wed, 13 Nov 2013 08:59:43 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -10.349 X-Spam-Level: X-Spam-Status: No, score=-10.349 tagged_above=-999 required=5 tests=[AWL=0.250, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id qcVCBvg4QkBp for ; Wed, 13 Nov 2013 08:59:37 -0800 (PST) Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id 7A4EC21E8105 for ; Wed, 13 Nov 2013 08:59:37 -0800 (PST) Received: from fr711usmtp1.zeu.alcatel-lucent.com (h135-239-2-122.lucent.com [135.239.2.122]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id rADGxYSZ003412 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL) for ; Wed, 13 Nov 2013 10:59:35 -0600 (CST) Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id rADGxYTg015239 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL) for ; Wed, 13 Nov 2013 17:59:34 +0100 Received: from FR712WXCHMBA15.zeu.alcatel-lucent.com ([169.254.7.70]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.02.0247.003; Wed, 13 Nov 2013 17:59:34 +0100 From: "Scharf, Michael (Michael)" To: "tcpm@ietf.org" Thread-Topic: Completion of WGLC for draft-ietf-tcpm-fastopen-05 Thread-Index: Ac7gkbj8pVYG8aj4SEafEqzqlhf+NQ== Date: Wed, 13 Nov 2013 16:59:32 +0000 Message-ID: <655C07320163294895BBADA28372AF5D12BF20@FR712WXCHMBA15.zeu.alcatel-lucent.com> Accept-Language: de-DE, en-US Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [135.239.27.40] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39 Subject: [tcpm] Completion of WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 13 Nov 2013 16:59:43 -0000 Hi all, As an outcome of the WGLC for draft-ietf-tcpm-fastopen-05, a document revis= ion is needed. WGLC comments were submitted by: * Gorry Fairhurst: Several remarks and questions, already some list discuss= ion =3D=3D> Gorry will have to review follow-up version, possibly iteration= required * Donald Smith: Comment on Section 4.1.2, not discussed on list so far =3D= =3D> Authors, any thoughts? * Joe Touch: Objection to dedicated option codepoint, discussed on list and= in meeting, including a consensus call with strong support for a dedicated= codepoint =3D=3D> See below * Bob Briscoe: Nits =3D=3D> Trivial to address Regarding the assignment of a dedicated option codepoint, the rough but sol= id consensus in TCPM seems to be: * Fast open is a useful TCP experiment for relevant applications, and it ha= s the potential for wider experimentation and deployment * There is no significant shortage of IANA-assigned TCP option codepoints f= or users following the IETF process: Out of 256 codepoints, IANA has only 1= 4 codepoints assigned to an non-obsoleted RFC, 6 other registrations (proba= bly legacy with unknown deployment state), 9 obsoleted codepoints (availabl= e for re-use?), and of the order of 10-15 known unauthorized uses * In contrast, option space in the TCP SYN is an extremely scarce resource = and using 16bit ExID for this TCP extension could harm other, future TCP ex= tensions * There is currently one well-known implementation in a major TCP stack tha= t complies to the IETF process by using RFC 6994, but further adoption in o= ther stacks is possible * As a summary, to the TCPM community exceptional allocation of a codepoint= to this non-STD-track TCPM document seems to be a reasonable trade-off My understanding is that the TCPM chairs would have to provide this reasoni= ng to the IESG and IANA, similar to http://www.ietf.org/mail-archive/web/mu= ltipathtcp/current/msg01706.html . It is then up to the IESG to decide whet= her to approve this. Please let me know if I missed something or if you have any comments. Thanks Michael (document shepherd) From ycheng@google.com Wed Nov 13 09:11:58 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2389421E815C for ; Wed, 13 Nov 2013 09:11:58 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.784 X-Spam-Level: X-Spam-Status: No, score=-1.784 tagged_above=-999 required=5 tests=[AWL=0.193, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id cEepQplogpbc for ; Wed, 13 Nov 2013 09:11:57 -0800 (PST) Received: from mail-ie0-x22c.google.com (mail-ie0-x22c.google.com [IPv6:2607:f8b0:4001:c03::22c]) by ietfa.amsl.com (Postfix) with ESMTP id 5CF8221E8158 for ; Wed, 13 Nov 2013 09:11:54 -0800 (PST) Received: by mail-ie0-f172.google.com with SMTP id to1so986533ieb.3 for ; Wed, 13 Nov 2013 09:11:53 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=st3z1WDZgVD+NXwtrGJDC+OeZYbPDVU0rG7cCohb/wk=; b=Qb+JXF6Ql5qXwVqQhP8iyL+rKYDeU7XyoKZmv+/7Q+f9qVN7iIz3WdOtnURc81Vzil mWFhQelvHYT0XqZXOX1L3HaTTrFWNa4BTNXuqtFM5hZ9qvfH6JhT/ChASMhhh1aV7uYN usWKNchXu5+iRIeqlgVQZKkaLo27IvgZB7s/FFWaENg752+GJfvn/qHSVmSZ75Z1sIAB f6xPcQrKFTLXOH4xaL4rtZy6jyPcGYRpuK0IXpAQzTTvAtGyJxEIPll1Hk3b4Qnpv+EV kkFtpsfsVo5n7J8Xlt1V7ImAP7Spm5421isgYxDM5Yx4ec5qnqdTXC9e9jfmhI2ODvSc D9zw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=st3z1WDZgVD+NXwtrGJDC+OeZYbPDVU0rG7cCohb/wk=; b=Bo7qVEfvxTHvkV1sfsmP5MDrPEffp7Zz9DlZQUbNK1tJa3ZZ8JR386rB7wKUkL4V8V PEy7AVYO6xVn4WVxAUAt0FCgZzsgYhcws9ehbLmyv4bHr4mYK9opFZo4CQbcOHjnVETa hj3EilZmbsCT5330CwTAJDVEV7djs59xgY1uXa+FAl+cGQC4SRExHvORH6A/+HBjhqem +Qib+qy/HbUp5OseXHEZY6aDWCeVfK9ScJud6d5cZaEXBi1ADdYED5uiWTdQA7IlHS+4 wAf+mY4HAqtvdd4XKvA9f/I79DO0c9v1dWHi+hcLDssSspi90Rbr+HuVNF6L9jQhIgqi B0Zw== X-Gm-Message-State: ALoCoQnUidV80PJUdnPNBFTWIs1PykhAZVVNPrcl+ID5SVgsFdOudZO9M0zx6MdUT/xdmhqn9WKZd9arwxkS1AX2hSB5A+g7LdX6Ot+3ybwWrRAf36z+sL3JZgVY7+EtreykJUzhREXhwYJ3FA9LX2c01cSBYm5yMFaOQbbEt1AkEZpHIG+Rhq8X0A3O94Rt84/6xG97kkSH X-Received: by 10.50.87.33 with SMTP id u1mr19171095igz.42.1384362712953; Wed, 13 Nov 2013 09:11:52 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Wed, 13 Nov 2013 09:11:12 -0800 (PST) In-Reply-To: <655C07320163294895BBADA28372AF5D12BF20@FR712WXCHMBA15.zeu.alcatel-lucent.com> References: <655C07320163294895BBADA28372AF5D12BF20@FR712WXCHMBA15.zeu.alcatel-lucent.com> From: Yuchung Cheng Date: Wed, 13 Nov 2013 09:11:12 -0800 Message-ID: To: "Scharf, Michael (Michael)" Content-Type: multipart/alternative; boundary=089e0111b33abfe2ac04eb120f80 Cc: "tcpm@ietf.org" Subject: Re: [tcpm] Completion of WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 13 Nov 2013 17:11:58 -0000 --089e0111b33abfe2ac04eb120f80 Content-Type: text/plain; charset=ISO-8859-1 On Wed, Nov 13, 2013 at 8:59 AM, Scharf, Michael (Michael) < michael.scharf@alcatel-lucent.com> wrote: > Hi all, > > As an outcome of the WGLC for draft-ietf-tcpm-fastopen-05, a document > revision is needed. > > WGLC comments were submitted by: > > * Gorry Fairhurst: Several remarks and questions, already some list > discussion ==> Gorry will have to review follow-up version, possibly > iteration required > ack > > * Donald Smith: Comment on Section 4.1.2, not discussed on list so far ==> > Authors, any thoughts? > Sorry I forgot to reply his email b/c I was distracted by the other (longer) discussions of code point in the same thread. I will reply soon. > > * Joe Touch: Objection to dedicated option codepoint, discussed on list > and in meeting, including a consensus call with strong support for a > dedicated codepoint ==> See below > > * Bob Briscoe: Nits ==> Trivial to address > ack > > > Regarding the assignment of a dedicated option codepoint, the rough but > solid consensus in TCPM seems to be: > > * Fast open is a useful TCP experiment for relevant applications, and it > has the potential for wider experimentation and deployment > > * There is no significant shortage of IANA-assigned TCP option codepoints > for users following the IETF process: Out of 256 codepoints, IANA has only > 14 codepoints assigned to an non-obsoleted RFC, 6 other registrations > (probably legacy with unknown deployment state), 9 obsoleted codepoints > (available for re-use?), and of the order of 10-15 known unauthorized uses > > * In contrast, option space in the TCP SYN is an extremely scarce resource > and using 16bit ExID for this TCP extension could harm other, future TCP > extensions > > * There is currently one well-known implementation in a major TCP stack > that complies to the IETF process by using RFC 6994, but further adoption > in other stacks is possible > > * As a summary, to the TCPM community exceptional allocation of a > codepoint to this non-STD-track TCPM document seems to be a reasonable > trade-off > > My understanding is that the TCPM chairs would have to provide this > reasoning to the IESG and IANA, similar to > http://www.ietf.org/mail-archive/web/multipathtcp/current/msg01706.html . > It is then up to the IESG to decide whether to approve this. > > Please let me know if I missed something or if you have any comments. > > Thanks > > Michael (document shepherd) > > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm > --089e0111b33abfe2ac04eb120f80 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable



On Wed, Nov 13, 2013 at 8:59 AM, Scharf, Michael (Michael) <michael.scharf@alcatel-lucent.com> wrote:
Hi all,

As an outcome of the WGLC for draft-ietf-tcpm-fastopen-05, a document revis= ion is needed.

WGLC comments were submitted by:

* Gorry Fairhurst: Several remarks and questions, already some list discuss= ion =3D=3D> Gorry will have to review follow-up version, possibly iterat= ion required
ack=A0

* Joe Touch: Objection to dedicated option codepoint, discussed on list and= in meeting, including a consensus call with strong support for a dedicated= codepoint =3D=3D> See below

* Bob Briscoe: Nits =3D=3D> Trivial to address
ack= =A0


Regarding the assignment of a dedicated option codepoint, the rough but sol= id consensus in TCPM seems to be:

* Fast open is a useful TCP experiment for relevant applications, and it ha= s the potential for wider experimentation and deployment

* There is no significant shortage of IANA-assigned TCP option codepoints f= or users following the IETF process: Out of 256 codepoints, IANA has only 1= 4 codepoints assigned to an non-obsoleted RFC, 6 other registrations (proba= bly legacy with unknown deployment state), 9 obsoleted codepoints (availabl= e for re-use?), and of the order of 10-15 known unauthorized uses

* In contrast, option space in the TCP SYN is an extremely scarce resource = and using 16bit ExID for this TCP extension could harm other, future TCP ex= tensions

* There is currently one well-known implementation in a major TCP stack tha= t complies to the IETF process by using RFC 6994, but further adoption in o= ther stacks is possible

* As a summary, to the TCPM community exceptional allocation of a codepoint= to this non-STD-track TCPM document seems to be a reasonable trade-off

My understanding is that the TCPM chairs would have to provide this reasoni= ng to the IESG and IANA, similar to http://www.i= etf.org/mail-archive/web/multipathtcp/current/msg01706.html . It is the= n up to the IESG to decide whether to approve this.

Please let me know if I missed something or if you have any comments.

Thanks

Michael (document shepherd)

_______________________________________________
tcpm mailing list
tcpm@ietf.org
ht= tps://www.ietf.org/mailman/listinfo/tcpm

--089e0111b33abfe2ac04eb120f80-- From ycheng@google.com Wed Nov 13 10:56:49 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4781811E8120 for ; Wed, 13 Nov 2013 10:56:49 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.798 X-Spam-Level: X-Spam-Status: No, score=-1.798 tagged_above=-999 required=5 tests=[AWL=0.180, BAYES_00=-2.599, FM_FORGED_GMAIL=0.622, NO_RELAYS=-0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id z12Sy8J9MUf5 for ; Wed, 13 Nov 2013 10:56:48 -0800 (PST) Received: from mail-ie0-x22d.google.com (mail-ie0-x22d.google.com [IPv6:2607:f8b0:4001:c03::22d]) by ietfa.amsl.com (Postfix) with ESMTP id 9479E11E8107 for ; Wed, 13 Nov 2013 10:56:48 -0800 (PST) Received: by mail-ie0-f173.google.com with SMTP id x13so1116093ief.4 for ; Wed, 13 Nov 2013 10:56:48 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type:content-transfer-encoding; bh=0j6E3F/Mb4EZyc23QFyRPcy92kmLKybxov5B4udccDc=; b=MGeg6KaX/N0BAB+8rOyncVZc0IMWetYAZ5NEFvWB3FwMd/vziO6IGym+Al0Bnav+w5 KsWvDUhHku9KSkSf3hPmhBe4yAHampY3vVNH/UrugBLxlCcVzSeOYsvO+CO4Fyj3kRnm RZ8RM0wlZC2WuUUlN6KERN8TxOA70h6WgFUBOsv1hNe1KZN8AdsCkiLSiwQz+kJ91Prw nkWpKjt7bZUjQliUbTTZH4uX36/COrwvBxqppQrLimfxmabQtSfu3ENpXoOxTsuplWk6 6NtE1XHZHsOLIwWN791FMHPPv9zkPiGzDdCfFJ4SDOKjER+Uedq72yqQCTbx5oKm6vfQ Dtgg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type:content-transfer-encoding; bh=0j6E3F/Mb4EZyc23QFyRPcy92kmLKybxov5B4udccDc=; b=S4YMpcnYMo6ZokYZjIT/oFgziNpxcmZNHHvhmlJyyMq3PiHMn622lyTc3cKunYva/9 mHy0MqXspNJMpGhF5U37anteI9e0x+kGHbjHz039aRLfIOv8gaR+A7HxO1MNI8g9HAd9 Op2yNaloSrkTNSJykzuPgXQUttujRUrthqx/49pJpenMw12Ps9TMrKMA6npZG81YfbbJ 2qYCfJJ5MbEtRXFYqAd+Z3JKUIQehVqCBhrXSX8Z0K7+teB79X749rD6k8+Y3ODFI37S dP4R0OQMpkuQpkp0pMWyDj+k1jxR2+BaYQzvfBKqDv1f0Jq8M36y8USWzqUNZ4Be6U5B g93Q== X-Gm-Message-State: ALoCoQneB6wf2OYJcLZqi4unRcRTvDKDrx3MoYNSIUXhmZtsq7i7eT12SemeDF6yMiHPe/ntm0HCv9zffczUerhw+CHoWIBwCFkM2qIPNaiF/S8xxDBhVO/p4EOEdH6QJY46BnItyE2XueEcUjLBihmIMoElWvFT9sasdHfHWYkauIfbaxaCKZl6p+BnYENg88dWntDJasU6 X-Received: by 10.50.11.11 with SMTP id m11mr19207365igb.47.1384369008117; Wed, 13 Nov 2013 10:56:48 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Wed, 13 Nov 2013 10:56:08 -0800 (PST) In-Reply-To: <68EFACB32CF4464298EA2779B058889D0A42818F@PDDCWMBXEX503.ctl.intranet> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <68EFACB32CF4464298EA2779B058889D0A42818F@PDDCWMBXEX503.ctl.intranet> From: Yuchung Cheng Date: Wed, 13 Nov 2013 10:56:08 -0800 Message-ID: To: "Smith, Donald" Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 13 Nov 2013 18:56:49 -0000 On Mon, Oct 28, 2013 at 4:35 PM, Smith, Donald wrote: > Hi Donald, Sorry for the late reply. I dropped your email b/c it was buried in the busy discussion of code point. > This doesn't work. > Example Implementation: a simple implementation is to use AES_128 to > encrypt the IPv4 (with padding) or IPv6 address and truncate to 64 > bits. The server can periodically update the key to expire the > cookies. AES encryption on recent processors is fast and takes only > few hundred nanoseconds [RCCJR11]. > > If only one valid cookie is allowed per-IP and the server can > regenerate the cookie independently, the best validation process is > to simply regenerate a valid cookie and compare it against the > incoming cookie. In that case if the incoming cookie fails the check, > a valid cookie is readily available to be sent to the client. > > That would allow someone doing reflection attacks to spoof an syn wait a = few seconds for the server to respond (which the attacker wouldn't see) but= the attacker knows what src ip he used in the syn so he can generate the s= yn/ack with a correct TFO cookie. I am not sure I understand the issue. the attack can generate the syn/ack with a correct TFO cookie if he got the secrete key in the server. The defense is compromised if the secret key is not longer a secret just like many systems ... > > And maybe I am missing something but this seems to have an extra step in = it. > > Performing TCP Fast Open in connection 2: > > TCP A (Client) TCP B(Server) > ______________ _____________ > CLOSED LISTEN > > #1 SYN-SENT ----- ----> SYN-RCVD > > #2 ESTABLISHED <---- ---- SYN-RCVD > > #3 ESTABLISHED <---- ---- SYN-RCVD > > #4 ESTABLISHED ----- --------------------> ESTABLISHED > > #5 ESTABLISHED --- ----------> ESTABLISHED > > > > I think you could skip 4 because 5 would be a valid ack so 4 shouldn't be= required? yes skip 4 can be skipped but this is just an illustration/example. We do not imply #4 is required. > > Just one example: > The server can periodically update the key to expire the cookies. > > So that should be TFO_server_cookie_expiration (or something like that) a= nd probably should be configurable by the users. > > > Next there are several timers defined or recommended but they weren't giv= en names. > I find when working with vendors if you have a standard name for a tcp wa= it or time out value it helps a lot. > > Beside the cookie, we RECOMMEND that the > client caches the MSS and RTT to the server to enhance performance. > > Just minor but RECOMMENDED is a key word but afaik RECOMMEND is not. > It could easily be rewritten > "Besides the cookie, it is RECOMMENDED that ..." will fix that in next revision. thanks. > > This is just my first set of comments, based on a quick review. I believe= there are more areas of concern but wanted to get these posted asap. > > > (coffee !=3D sleep) & (!coffee =3D=3D sleep) > Donald.Smith@centurylink.com > From Alexander.Zimmermann@netapp.com Thu Nov 14 06:01:44 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 56B4D11E80F2 for ; Thu, 14 Nov 2013 06:01:44 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -9.266 X-Spam-Level: X-Spam-Status: No, score=-9.266 tagged_above=-999 required=5 tests=[AWL=1.333, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id pOP-a51XdKIq for ; Thu, 14 Nov 2013 06:01:39 -0800 (PST) Received: from mx2.netapp.com (mx2.netapp.com [216.240.18.37]) by ietfa.amsl.com (Postfix) with ESMTP id 4EDCD11E80E0 for ; Thu, 14 Nov 2013 06:01:39 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,699,1378882800"; d="asc'?scan'208";a="56740745" Received: from vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) by mx2-out.netapp.com with ESMTP; 14 Nov 2013 06:01:38 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.215]) by vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) with mapi id 14.03.0158.001; Thu, 14 Nov 2013 06:01:37 -0800 From: "Zimmermann, Alexander" To: "Scheffenegger, Richard" , David Borman Thread-Topic: Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAA== Date: Thu, 14 Nov 2013 14:01:35 +0000 Message-ID: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.117] Content-Type: multipart/signed; boundary="Apple-Mail=_80BEB1B3-4266-42FC-B3E0-5695D091497A"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" Subject: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 14:01:44 -0000 --Apple-Mail=_80BEB1B3-4266-42FC-B3E0-5695D091497A Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 Hi, while scanning the diff of the last version yesterday the question if the title of the draft is still appropriate came into my mind. For me high performance is somehow equal to high throughput, and for this I do not necessarily need the two options. E.g. for a 40Gb/s TCP connection in a DC I do not need Timestamps=85 So, is a title like =84The TCP Windows Scale and TCP Timestamps option=93 (or something similar) not a better title? A reader knows directly from title what is specified in the RFC. BTW: IMO the abstract is too short. Could we extent it a little bit so that the reader has more context in advanced? Alex --Apple-Mail=_80BEB1B3-4266-42FC-B3E0-5695D091497A Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKE18AACgkQdyiq39b9uS4kPQCfbo7j/MgzGo+AmslA/GJMqTbF Ek8AoI5+CE0kxA4SpDq9J5Q7KQvOOGuf =Yi8l -----END PGP SIGNATURE----- --Apple-Mail=_80BEB1B3-4266-42FC-B3E0-5695D091497A-- From rs@netapp.com Thu Nov 14 06:52:35 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7269021E80E9 for ; Thu, 14 Nov 2013 06:52:35 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.598 X-Spam-Level: X-Spam-Status: No, score=-5.598 tagged_above=-999 required=5 tests=[AWL=-2.999, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id j-hmlUCg3aUc for ; Thu, 14 Nov 2013 06:52:30 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 6388F11E80F2 for ; Thu, 14 Nov 2013 06:52:30 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,700,1378882800"; d="scan'208";a="75017743" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx11-out.netapp.com with ESMTP; 14 Nov 2013 06:52:30 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Thu, 14 Nov 2013 06:52:29 -0800 From: "Scheffenegger, Richard" To: "Zimmermann, Alexander" , David Borman , "Pasi Sarolahti (pasi.sarolahti@iki.fi)" Thread-Topic: Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnA Date: Thu, 14 Nov 2013 14:52:27 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com> References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> In-Reply-To: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.53] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 14:52:35 -0000 Hi Alex, first, thanks for the list of nits you sent me directly (making the "O" in = "Window Scale option" lowercase to mirror the style of the Timestamps optio= n; one missing "<" bracket in a xref, one missed "monotonically increasing"= ). Second, to your comments: I'd rather not change the title of the RFC, to keep this in line with other= RFCs that have been updated over time, Ie. 3517 / 6675 A Conservative Selective Acknowledgment (SACK)-based Loss Recovery Algorith= m for TCP Or 2581 / 5681 TCP Congestion Control And many (non-tcp) others... As to the slight expansion of the abstract, what about this text This document specifies a set of TCP extensions to improve=20 performance over paths with a large bandwidth * delay product and to=20 provide reliable operation over very high-speed paths. It defines=20 the TCP Window Scale (WS) option and the TCP Timestamps (TS) option=20 and their semantics. The Window Scale option is used to support=20 larger receive windows, while the Timestamps option can be used for=20 at least two distinct mechanisms, PAWS (Protection Against Wrapped=20 Sequences) and RTTM (Round Trip Time Measurement), that are also=20 described herein.=20 Is that text agreeable? I'd like to ask the Chairs to start the WGLC in the next few days, to proce= ed with this doc to the next phase. Best regards, Richard Scheffenegger > -----Original Message----- > From: Zimmermann, Alexander > Sent: Donnerstag, 14. November 2013 15:02 > To: Scheffenegger, Richard; David Borman > Cc: tcpm@ietf.org Extensions > Subject: Is RFC1323bis' title still appropriate? >=20 > Hi, >=20 > while scanning the diff of the last version yesterday the question if the > title of the draft is still appropriate came into my mind. >=20 > For me high performance is somehow equal to high throughput, and for this > I do not necessarily need the two options. E.g. > for a 40Gb/s TCP connection in a DC I do not need Timestamps... >=20 > So, is a title like "The TCP Windows Scale and TCP Timestamps option" (or > something similar) not a better title? A reader knows directly from title > what is specified in the RFC. >=20 > BTW: IMO the abstract is too short. Could we extent it a little bit so > that the reader has more context in advanced? >=20 > Alex >=20 >=20 >=20 From Alexander.Zimmermann@netapp.com Thu Nov 14 06:55:57 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BD27821E80DD for ; Thu, 14 Nov 2013 06:55:57 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -9.599 X-Spam-Level: X-Spam-Status: No, score=-9.599 tagged_above=-999 required=5 tests=[AWL=1.000, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wmFgpWFb94GC for ; Thu, 14 Nov 2013 06:55:52 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id 46CAA11E80F2 for ; Thu, 14 Nov 2013 06:55:44 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,700,1378882800"; d="asc'?scan'208";a="115593290" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx12-out.netapp.com with ESMTP; 14 Nov 2013 06:55:43 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.215]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Thu, 14 Nov 2013 06:55:43 -0800 From: "Zimmermann, Alexander" To: "Scheffenegger, Richard" Thread-Topic: Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnAgACLy4A= Date: Thu, 14 Nov 2013 14:55:41 +0000 Message-ID: References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com> In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.117] Content-Type: multipart/signed; boundary="Apple-Mail=_2F3D9244-7E0A-43F5-996E-F300F3C308C7"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" , David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 14:55:58 -0000 --Apple-Mail=_2F3D9244-7E0A-43F5-996E-F300F3C308C7 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 Inline. Am 14.11.2013 um 15:52 schrieb Scheffenegger, Richard : >=20 > Hi Alex, >=20 > first, thanks for the list of nits you sent me directly (making the = "O" in "Window Scale option" lowercase to mirror the style of the = Timestamps option; one missing "<" bracket in a xref, one missed = "monotonically increasing"). >=20 > Second, to your comments: >=20 > I'd rather not change the title of the RFC, to keep this in line with = other RFCs that have been updated over time, >=20 > Ie. 3517 / 6675 > A Conservative Selective Acknowledgment (SACK)-based Loss Recovery = Algorithm for TCP > Or > 2581 / 5681 > TCP Congestion Control Because we have already a good title here :-) >=20 > And many (non-tcp) others=85 Counter example is Gorry=92S CWV :-) >=20 >=20 >=20 > As to the slight expansion of the abstract, what about this text >=20 >=20 > This document specifies a set of TCP extensions to improve=20 > performance over paths with a large bandwidth * delay product and = to=20 > provide reliable operation over very high-speed paths. It defines=20= > the TCP Window Scale (WS) option and the TCP Timestamps (TS) option=20= > and their semantics. The Window Scale option is used to support=20 > larger receive windows, while the Timestamps option can be used for=20= > at least two distinct mechanisms, PAWS (Protection Against Wrapped=20= > Sequences) and RTTM (Round Trip Time Measurement), that are also=20 > described herein.=20 >=20 > Is that text agreeable? Yes! >=20 >=20 > I'd like to ask the Chairs to start the WGLC in the next few days, to = proceed with this doc to the next phase. >=20 >=20 > Best regards, >=20 > Richard Scheffenegger >=20 >=20 >> -----Original Message----- >> From: Zimmermann, Alexander >> Sent: Donnerstag, 14. November 2013 15:02 >> To: Scheffenegger, Richard; David Borman >> Cc: tcpm@ietf.org Extensions >> Subject: Is RFC1323bis' title still appropriate? >>=20 >> Hi, >>=20 >> while scanning the diff of the last version yesterday the question if = the >> title of the draft is still appropriate came into my mind. >>=20 >> For me high performance is somehow equal to high throughput, and for = this >> I do not necessarily need the two options. E.g. >> for a 40Gb/s TCP connection in a DC I do not need Timestamps... >>=20 >> So, is a title like "The TCP Windows Scale and TCP Timestamps option" = (or >> something similar) not a better title? A reader knows directly from = title >> what is specified in the RFC. >>=20 >> BTW: IMO the abstract is too short. Could we extent it a little bit = so >> that the reader has more context in advanced? >>=20 >> Alex >>=20 >>=20 >>=20 >=20 --Apple-Mail=_2F3D9244-7E0A-43F5-996E-F300F3C308C7 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKE5G0ACgkQdyiq39b9uS5sigCfdnBCuHrjVxr3irYj5EP7LQD3 ryoAoIVKfr31TUxXiExeXaJs2UWn1Xym =o1jV -----END PGP SIGNATURE----- --Apple-Mail=_2F3D9244-7E0A-43F5-996E-F300F3C308C7-- From pasi.sarolahti@iki.fi Thu Nov 14 07:14:34 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 31D1521E80F4 for ; Thu, 14 Nov 2013 07:14:34 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.599 X-Spam-Level: X-Spam-Status: No, score=-102.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 5Yd-4g4k7Teg for ; Thu, 14 Nov 2013 07:14:28 -0800 (PST) Received: from kirsi1.inet.fi (mta-out.inet.fi [195.156.147.13]) by ietfa.amsl.com (Postfix) with ESMTP id 017C811E80F2 for ; Thu, 14 Nov 2013 07:14:27 -0800 (PST) Received: from pc114.netlab.hut.fi (130.233.154.114) by kirsi1.inet.fi (8.5.140.03) (authenticated as saropa-1) id 526FA42301841483; Thu, 14 Nov 2013 17:14:24 +0200 Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 6.6 \(1510\)) From: Pasi Sarolahti In-Reply-To: <23362_1384285457_52828510_23362_3142_1_012C3117EDDB3C4781FD802A8C27DD4F25E8C35B@SACEXCMBX02-PRD.hq.netapp.com> Date: Thu, 14 Nov 2013 17:14:22 +0200 Content-Transfer-Encoding: quoted-printable Message-Id: <1409E0D2-B587-43BA-8173-63DEFBB3BA37@iki.fi> References: <20131112190728.28759.75122.idtracker@ietfa.amsl.com> <23362_1384285457_52828510_23362_3142_1_012C3117EDDB3C4781FD802A8C27DD4F25E8C35B@SACEXCMBX02-PRD.hq.netapp.com> To: "Scheffenegger, Richard" X-Mailer: Apple Mail (2.1510) Cc: Fernando Gont , "tcpm@ietf.org" , "mallman@icir.org" Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 15:14:34 -0000 On Nov 12, 2013, at 9:43 PM, "Scheffenegger, Richard" = wrote: > Hi group, >=20 > As mentioned, I have incorporated all the changes from the latest = review of this draft, except where noted (in the earlier email). Thanks! Let's wait for a couple of days if there are any additional = remarks, and then start a WGLC unless something major shows up. Below some thoughts on the points you raised: >> - Sec 1: In the last paragraph before 1.1: >>=20 >> - should "recommended" be "RECOMMENDED"? (I am not sure) >=20 > This text is before the definition of RFC2119 style wording; I tend to = keep this lower case... The important thing is not whether it comes before or after the = definition, but that "RECOMMENDED" has a specific meaning (synonym of = SHOULD). I agree with you to keep it lower case here (and might even = prefer less strong language about recommending an optional mechanism). = Generally, I feel that introduction is a bit too early place start = rolling out normative protocol requirements, and the typical location of = RFC 2119 reference boilerplate speaks for that, too. >> - Sec 1.2, point (1): This is not entirely correct. You need both a >> high transfer rate *and a whole bunch of data* to have a sequence >> wrapping issue. This point could be a lot better by noting that = it >> takes 4GB of data to wrap. >=20 > Sentence now reads: At a high enough transfer rate of large volume of = data (at least 4 GiB in the same session), the 32-bit sequence space may = be wrapped within the time that a sement is delayed in queues. >=20 > Do you think this is good enough? Works for me. >> - Sec 2.3: The list at the beginning of the section should be >> converted to standards language (MUST, SHOULD, etc.). >=20 > Added mostly MUSTs (only one SHOULD on the receiver, which may choose = to limit the window for memory constrains etc). Ok, but I would leave the first bullet about augmenting connection state = in its previous state. It is a local implementation detail, and doesn't = need MUST. >> Sec 3.2: This whole discussion of valid and invalid TSecr is >> confusing. You note that when the value is invalid it should be = set >> to zero. But, a TSecr of zero does not mean the TS is invalid. = So, >> how should we tell? >=20 > This was my attempt at "liberal at what you accept, and conservative = in what you send". There are implementations that assume TSecr=3D0 =3D=3D = invalid, with not strictly defined behavior in that case (e.g. drop that = segment and stall until the sender timestamp increases to 1, or end up = with strange RTO values...). When TSval is more or less the uptime of = the sender, this roll-over hardly matters [which is probably the reason = this was not fixed a long time ago in the affected stack], but when = TSval is randomized, this corner case is more likely to occur and the = implementer should properly treat this case.=20 Hmm... I can see the point of Mark's comment. How about: The Timestamp Echo Reply field (TSecr) is valid if the ACK bit is set in = the TCP header; if ACK bit is set in the TCP header, the sender of TSecr = field MUST echo a recently received timestamp sent by the remote TCP in = the TSval field of a Timestamps option (the exact rules on which = timestamp to echo will be given in Sec. 4.3). If ACK bit is not set in = the outgoing TCP header, the sender MUST (or SHOULD?) set TSecr field to = 0. When ACK bit is set, the receiver should process the incoming TSecr = value as valid (including value 0). When ACK bit is not set, the = receiver MUST ignore the value of TSecr field. Is this correct? (I don't see the need for MUST set TSecr to 0 above, because the field = should be ignored anyway in the absence of ACK. Generally I'm not fond = of excess use of MUST when not necessary) - Pasi From ilpo.jarvinen@helsinki.fi Thu Nov 14 07:27:37 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D8E8121F9C70 for ; Thu, 14 Nov 2013 07:27:37 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.186 X-Spam-Level: X-Spam-Status: No, score=-6.186 tagged_above=-999 required=5 tests=[AWL=0.114, BAYES_00=-2.599, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aNKf2fjVHcm7 for ; Thu, 14 Nov 2013 07:27:32 -0800 (PST) Received: from courier.cs.helsinki.fi (courier.cs.helsinki.fi [128.214.9.1]) by ietfa.amsl.com (Postfix) with ESMTP id 4723521F8D62 for ; Thu, 14 Nov 2013 07:27:30 -0800 (PST) Received: from melkinpaasi.cs.helsinki.fi (melkinpaasi.cs.helsinki.fi [128.214.9.14]) (TLS: TLSv1/SSLv3,256bits,AES256-SHA) by mail.cs.helsinki.fi with esmtp; Thu, 14 Nov 2013 17:27:15 +0200 id 00068149.5284EBD3.000008AA Date: Thu, 14 Nov 2013 17:27:15 +0200 (EET) From: "=?ISO-8859-15?Q?Ilpo_J=E4rvinen?=" X-X-Sender: ijjarvin@melkinpaasi.cs.helsinki.fi To: "Scheffenegger, Richard" In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com> Message-ID: References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-CS-Test-DKIM: none Cc: "tcpm@ietf.org Extensions" , David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 15:27:38 -0000 On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: > I'd rather not change the title of the RFC, to keep this in line with > other RFCs that have been updated over time, > > Ie. 3517 / 6675 > A Conservative Selective Acknowledgment (SACK)-based Loss Recovery > Algorithm for TCP ?!? "A Conservative Loss Recovery Algorithm Based on Selective Acknowledgment (SACK) for TCP" says my RFC6675 :-) ...Did you perhaps mean that its meaning wasn't significantly changed? -- i. From rs@netapp.com Thu Nov 14 07:38:40 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3387011E80F2 for ; Thu, 14 Nov 2013 07:38:40 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -9.131 X-Spam-Level: X-Spam-Status: No, score=-9.131 tagged_above=-999 required=5 tests=[AWL=0.868, BAYES_00=-2.599, J_CHICKENPOX_33=0.6, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kowjgFqBqmbg for ; Thu, 14 Nov 2013 07:38:35 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id 6294021E8117 for ; Thu, 14 Nov 2013 07:36:22 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,700,1378882800"; d="scan'208";a="115604500" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx12-out.netapp.com with ESMTP; 14 Nov 2013 07:36:21 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Thu, 14 Nov 2013 07:36:21 -0800 From: "Scheffenegger, Richard" To: Pasi Sarolahti Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt Thread-Index: AQHO39qr2jLQB+w1qUGbkh36wsXn+pok2Oa/gAADnHA= Date: Thu, 14 Nov 2013 15:36:17 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E94F1C@SACEXCMBX02-PRD.hq.netapp.com> References: <20131112190728.28759.75122.idtracker@ietfa.amsl.com> <23362_1384285457_52828510_23362_3142_1_012C3117EDDB3C4781FD802A8C27DD4F25E8C35B@SACEXCMBX02-PRD.hq.netapp.com> <1409E0D2-B587-43BA-8173-63DEFBB3BA37@iki.fi> In-Reply-To: <1409E0D2-B587-43BA-8173-63DEFBB3BA37@iki.fi> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.53] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: Fernando Gont , "tcpm@ietf.org" , "mallman@icir.org" Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 15:38:40 -0000 Hi Pasi, > >> - Sec 2.3: The list at the beginning of the section should be > >> converted to standards language (MUST, SHOULD, etc.). > > > > Added mostly MUSTs (only one SHOULD on the receiver, which may choose t= o > limit the window for memory constrains etc). >=20 > Ok, but I would leave the first bullet about augmenting connection state > in its previous state. It is a local implementation detail, and doesn't > need MUST. Will check this... > >> Sec 3.2: This whole discussion of valid and invalid TSecr is > >> confusing. You note that when the value is invalid it should be se= t > >> to zero. But, a TSecr of zero does not mean the TS is invalid. So= , > >> how should we tell? > > > > This was my attempt at "liberal at what you accept, and conservative in > what you send". There are implementations that assume TSecr=3D0 =3D=3D in= valid, > with not strictly defined behavior in that case (e.g. drop that segment > and stall until the sender timestamp increases to 1, or end up with > strange RTO values...). When TSval is more or less the uptime of the > sender, this roll-over hardly matters [which is probably the reason this > was not fixed a long time ago in the affected stack], but when TSval is > randomized, this corner case is more likely to occur and the implementer > should properly treat this case. >=20 > Hmm... I can see the point of Mark's comment. >=20 > How about: >=20 > The Timestamp Echo Reply field (TSecr) is valid if the ACK bit is set in > the TCP header; if ACK bit is set in the TCP header, the sender of TSecr > field MUST echo a recently received timestamp sent by the remote TCP in > the TSval field of a Timestamps option (the exact rules on which timestam= p > to echo will be given in Sec. 4.3). If ACK bit is not set in the outgoing > TCP header, the sender MUST (or SHOULD?) set TSecr field to 0. When ACK > bit is set, the receiver should process the incoming TSecr value as valid > (including value 0). When ACK bit is not set, the receiver MUST ignore th= e > value of TSecr field. >=20 > Is this correct? The reference here was that certain implementations also set TSecr to 0 in = the SYN,ACK or the first ACK, when they internally do something akin to SY= NCaching and don't keep the original TS in that state... I believe Windows = 95 / 98 behaved that way. I think your last sentence should read "the sender should process" :) Still a value of 0 has ambiguity currently (1323), as it may be a valid val= ue, or indicating that the receiver could not reflect a valid value for som= e reason... Your text suggestion will tie this up a bit, and is more strict to the rece= iver to not evict valid TS data :) Best regards, Richard Scheffenegger > -----Original Message----- > From: Pasi Sarolahti [mailto:pasi.sarolahti@iki.fi] > Sent: Donnerstag, 14. November 2013 16:14 > To: Scheffenegger, Richard > Cc: tcpm@ietf.org; mallman@icir.org; Fernando Gont > Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt >=20 > On Nov 12, 2013, at 9:43 PM, "Scheffenegger, Richard" > wrote: >=20 > > Hi group, > > > > As mentioned, I have incorporated all the changes from the latest revie= w > of this draft, except where noted (in the earlier email). >=20 > Thanks! Let's wait for a couple of days if there are any additional > remarks, and then start a WGLC unless something major shows up. >=20 > Below some thoughts on the points you raised: >=20 > >> - Sec 1: In the last paragraph before 1.1: > >> > >> - should "recommended" be "RECOMMENDED"? (I am not sure) > > > > This text is before the definition of RFC2119 style wording; I tend to > keep this lower case... >=20 > The important thing is not whether it comes before or after the > definition, but that "RECOMMENDED" has a specific meaning (synonym of > SHOULD). I agree with you to keep it lower case here (and might even > prefer less strong language about recommending an optional mechanism). > Generally, I feel that introduction is a bit too early place start rollin= g > out normative protocol requirements, and the typical location of RFC 2119 > reference boilerplate speaks for that, too. >=20 > >> - Sec 1.2, point (1): This is not entirely correct. You need both a > >> high transfer rate *and a whole bunch of data* to have a sequence > >> wrapping issue. This point could be a lot better by noting that it > >> takes 4GB of data to wrap. > > > > Sentence now reads: At a high enough transfer rate of large volume of > data (at least 4 GiB in the same session), the 32-bit sequence space may > be wrapped within the time that a sement is delayed in queues. > > > > Do you think this is good enough? >=20 > Works for me. >=20 > >> - Sec 2.3: The list at the beginning of the section should be > >> converted to standards language (MUST, SHOULD, etc.). > > > > Added mostly MUSTs (only one SHOULD on the receiver, which may choose t= o > limit the window for memory constrains etc). >=20 > Ok, but I would leave the first bullet about augmenting connection state > in its previous state. It is a local implementation detail, and doesn't > need MUST. >=20 > >> Sec 3.2: This whole discussion of valid and invalid TSecr is > >> confusing. You note that when the value is invalid it should be se= t > >> to zero. But, a TSecr of zero does not mean the TS is invalid. So= , > >> how should we tell? > > > > This was my attempt at "liberal at what you accept, and conservative in > what you send". There are implementations that assume TSecr=3D0 =3D=3D in= valid, > with not strictly defined behavior in that case (e.g. drop that segment > and stall until the sender timestamp increases to 1, or end up with > strange RTO values...). When TSval is more or less the uptime of the > sender, this roll-over hardly matters [which is probably the reason this > was not fixed a long time ago in the affected stack], but when TSval is > randomized, this corner case is more likely to occur and the implementer > should properly treat this case. >=20 > Hmm... I can see the point of Mark's comment. >=20 > How about: >=20 > The Timestamp Echo Reply field (TSecr) is valid if the ACK bit is set in > the TCP header; if ACK bit is set in the TCP header, the sender of TSecr > field MUST echo a recently received timestamp sent by the remote TCP in > the TSval field of a Timestamps option (the exact rules on which timestam= p > to echo will be given in Sec. 4.3). If ACK bit is not set in the outgoing > TCP header, the sender MUST (or SHOULD?) set TSecr field to 0. When ACK > bit is set, the receiver should process the incoming TSecr value as valid > (including value 0). When ACK bit is not set, the receiver MUST ignore th= e > value of TSecr field. >=20 > Is this correct? >=20 > (I don't see the need for MUST set TSecr to 0 above, because the field > should be ignored anyway in the absence of ACK. Generally I'm not fond of > excess use of MUST when not necessary) >=20 > - Pasi From rs@netapp.com Thu Nov 14 07:58:03 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 13C8D21F9C00 for ; Thu, 14 Nov 2013 07:58:03 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -9.327 X-Spam-Level: X-Spam-Status: No, score=-9.327 tagged_above=-999 required=5 tests=[AWL=0.972, BAYES_00=-2.599, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id CLXolJPXQHJB for ; Thu, 14 Nov 2013 07:57:58 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id 0BDE111E8176 for ; Thu, 14 Nov 2013 07:57:58 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,700,1378882800"; d="scan'208";a="115610916" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx12-out.netapp.com with ESMTP; 14 Nov 2013 07:57:57 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Thu, 14 Nov 2013 07:57:57 -0800 From: "Scheffenegger, Richard" To: =?iso-8859-1?Q?Ilpo_J=E4rvinen?= Thread-Topic: [tcpm] Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnAgACUnYD//4J4qA== Date: Thu, 14 Nov 2013 15:57:57 +0000 Message-ID: <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, In-Reply-To: Accept-Language: de-AT, en-US Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" , David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 15:58:03 -0000 Yes sorry! Would you like to see WS and TS in the title, or keep the old one? Best regards, Richard Am 14.11.2013 um 16:28 schrieb "Ilpo J=E4rvinen" : > On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: >=20 >> I'd rather not change the title of the RFC, to keep this in line with=20 >> other RFCs that have been updated over time,=20 >>=20 >> Ie. 3517 / 6675 >> A Conservative Selective Acknowledgment (SACK)-based Loss Recovery=20 >> Algorithm for TCP >=20 > ?!? >=20 > "A Conservative Loss Recovery Algorithm Based on Selective Acknowledgment= =20 > (SACK) for TCP" says my RFC6675 :-) >=20 > ...Did you perhaps mean that its meaning wasn't significantly changed? >=20 > --=20 > i. From ilpo.jarvinen@helsinki.fi Thu Nov 14 08:46:37 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7AB0121E8098 for ; Thu, 14 Nov 2013 08:46:35 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.242 X-Spam-Level: X-Spam-Status: No, score=-6.242 tagged_above=-999 required=5 tests=[AWL=0.057, BAYES_00=-2.599, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_MED=-4] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id R6bVff-Vx75P for ; Thu, 14 Nov 2013 08:46:29 -0800 (PST) Received: from courier.cs.helsinki.fi (courier.cs.helsinki.fi [128.214.9.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3832721E8056 for ; Thu, 14 Nov 2013 08:46:25 -0800 (PST) Received: from melkinpaasi.cs.helsinki.fi (melkinpaasi.cs.helsinki.fi [128.214.9.14]) (TLS: TLSv1/SSLv3,256bits,AES256-SHA) by mail.cs.helsinki.fi with esmtp; Thu, 14 Nov 2013 18:46:09 +0200 id 0006807D.5284FE51.00001B0E Date: Thu, 14 Nov 2013 18:46:09 +0200 (EET) From: "=?ISO-8859-15?Q?Ilpo_J=E4rvinen?=" X-X-Sender: ijjarvin@melkinpaasi.cs.helsinki.fi To: "Scheffenegger, Richard" In-Reply-To: <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> Message-ID: References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> User-Agent: Alpine 2.02 (DEB 1266 2009-07-14) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-CS-Test-DKIM: none Cc: "tcpm@ietf.org Extensions" , David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 16:46:39 -0000 On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: > Would you like to see WS and TS in the title, or keep the old one? For some reason I've never liked RFC1323 title to begin with so I'd prefer the actual option names here. It would actually be rather misleading to say everything that benefits from WS today would have particularly "high performance". BTW, I just noticed that 1323 is also a counter-example: RFC1072 TCP Extensions for Long-Delay Paths RFC1185 TCP Extension for High-Speed Paths RFC1323 TCP Extensions for High Performance (+RFC2018 TCP Selective Acknowledgment Options) -- i. From rs@netapp.com Thu Nov 14 15:38:17 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 0D65F21E80B3 for ; Thu, 14 Nov 2013 15:38:16 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -9.376 X-Spam-Level: X-Spam-Status: No, score=-9.376 tagged_above=-999 required=5 tests=[AWL=0.923, BAYES_00=-2.599, MIME_8BIT_HEADER=0.3, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id aRGMWwUzUmYm for ; Thu, 14 Nov 2013 15:38:07 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id EBC8821E808A for ; Thu, 14 Nov 2013 15:38:05 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,702,1378882800"; d="scan'208";a="115756607" Received: from vmwexceht01-prd.hq.netapp.com ([10.106.76.239]) by mx12-out.netapp.com with ESMTP; 14 Nov 2013 15:38:04 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht01-prd.hq.netapp.com ([10.106.76.239]) with mapi id 14.03.0123.003; Thu, 14 Nov 2013 15:38:04 -0800 From: "Scheffenegger, Richard" To: =?iso-8859-1?Q?Ilpo_J=E4rvinen?= Thread-Topic: [tcpm] Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnAgACUnYD//4J4qIAAk5OA///q/7A= Date: Thu, 14 Nov 2013 23:38:03 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> In-Reply-To: Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.115] Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" , David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 23:38:17 -0000 Ilpo, Very well then. But I'll need a good speaker to tell me which one is most c= orrect "TCP Window Scale and Timestamps option" (sounds like one combined option) "TCP Window Scale and Timestamps options" (sounds odd) "TCP Window Scale and TCP Timestamps options" (imho better) "TCP Window Scale option and TCP Timestamps option" (least ambiguity, but i= s this proper?) (prefixed with a "The"?) PS: I'm glad that we have come to the point to discuss the title :) =20 Richard Scheffenegger > -----Original Message----- > From: Ilpo J=E4rvinen [mailto:ilpo.jarvinen@helsinki.fi] > Sent: Donnerstag, 14. November 2013 17:46 > To: Scheffenegger, Richard > Cc: Zimmermann, Alexander; David Borman; Pasi Sarolahti > (pasi.sarolahti@iki.fi); tcpm@ietf.org Extensions > Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >=20 > On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: >=20 > > Would you like to see WS and TS in the title, or keep the old one? >=20 > For some reason I've never liked RFC1323 title to begin with so I'd prefe= r > the actual option names here. It would actually be rather misleading to > say everything that benefits from WS today would have particularly "high > performance". >=20 > BTW, I just noticed that 1323 is also a counter-example: >=20 > RFC1072 TCP Extensions for Long-Delay Paths > RFC1185 TCP Extension for High-Speed Paths > RFC1323 TCP Extensions for High Performance > (+RFC2018 TCP Selective Acknowledgment Options) >=20 >=20 > -- > i. From rs@netapp.com Thu Nov 14 15:54:29 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 99DFD11E8149 for ; Thu, 14 Nov 2013 15:54:29 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -5.57 X-Spam-Level: X-Spam-Status: No, score=-5.57 tagged_above=-999 required=5 tests=[AWL=-2.971, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id O67Z679uK5Nw for ; Thu, 14 Nov 2013 15:54:24 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 8FF7F21F9DC7 for ; Thu, 14 Nov 2013 15:54:24 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,702,1378882800"; d="scan'208";a="75180137" Received: from vmwexceht05-prd.hq.netapp.com ([10.106.77.35]) by mx11-out.netapp.com with ESMTP; 14 Nov 2013 15:54:24 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht05-prd.hq.netapp.com ([10.106.77.35]) with mapi id 14.03.0123.003; Thu, 14 Nov 2013 15:54:24 -0800 From: "Scheffenegger, Richard" To: Pasi Sarolahti Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt Thread-Index: AQHO39qr2jLQB+w1qUGbkh36wsXn+pok2Oa/gACPxKA= Date: Thu, 14 Nov 2013 23:54:22 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E97055@SACEXCMBX02-PRD.hq.netapp.com> References: <20131112190728.28759.75122.idtracker@ietfa.amsl.com> <23362_1384285457_52828510_23362_3142_1_012C3117EDDB3C4781FD802A8C27DD4F25E8C35B@SACEXCMBX02-PRD.hq.netapp.com> <1409E0D2-B587-43BA-8173-63DEFBB3BA37@iki.fi> In-Reply-To: <1409E0D2-B587-43BA-8173-63DEFBB3BA37@iki.fi> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.115] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: Fernando Gont , "tcpm@ietf.org" , "mallman@icir.org" Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-16.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 14 Nov 2013 23:54:29 -0000 Hi Pasi, > >> Sec 3.2: This whole discussion of valid and invalid TSecr is > >> confusing. You note that when the value is invalid it should be se= t > >> to zero. But, a TSecr of zero does not mean the TS is invalid. So= , > >> how should we tell? > > >=20 > Hmm... I can see the point of Mark's comment. >=20 > How about: >=20 > The Timestamp Echo Reply field (TSecr) is valid if the ACK bit is set in > the TCP header; if ACK bit is set in the TCP header, the sender of TSecr > field MUST echo a recently received timestamp sent by the remote TCP in > the TSval field of a Timestamps option (the exact rules on which timestam= p > to echo will be given in Sec. 4.3). If ACK bit is not set in the outgoing > TCP header, the sender MUST (or SHOULD?) set TSecr field to 0. When ACK > bit is set, the receiver should process the incoming TSecr value as valid > (including value 0). When ACK bit is not set, the receiver MUST ignore th= e > value of TSecr field. > I rephrased this now to=20 The Timestamps Echo Reply (TSecr) field is valid if the ACK bit=20 is set in the TCP header. If the ACK bit is not set in the=20 outgoing TCP header, the sender of that segment SHOULD set the=20 TSecr field to zero. When the ACK bit is set in an outgoing=20 segment, the sender MUST echo a recently received timestamp sent=20 by the remote TCP in the TSval field of a Timestamps option. The=20 exact rules on which TSval MUST be echoed are given in . When the ACK bit is not set, the receiver MUST=20 ignore the value of the TSecr field. (Which renders stacks that don't keep TS state during SYN-Caching not compl= iant... ). Comments? Richard Scheffenegger From dab@weston.borman.com Thu Nov 14 16:03:01 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 62E4C11E8167 for ; Thu, 14 Nov 2013 16:03:01 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.299 X-Spam-Level: X-Spam-Status: No, score=-2.299 tagged_above=-999 required=5 tests=[AWL=0.300, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6sAWOo6TA2Ub for ; Thu, 14 Nov 2013 16:02:57 -0800 (PST) Received: from frantic.weston.borman.com (frantic.weston.borman.com [70.57.156.33]) by ietfa.amsl.com (Postfix) with ESMTP id 07ACA11E8165 for ; Thu, 14 Nov 2013 16:02:55 -0800 (PST) Received: from [127.0.0.1] (frantic.weston.borman.com [70.57.156.33]) by frantic.weston.borman.com (8.14.7/8.14.7) with ESMTP id rAF00Pau009495; Thu, 14 Nov 2013 18:00:27 -0600 (CST) Content-Type: text/plain; charset=windows-1252 Mime-Version: 1.0 (Mac OS X Mail 7.0 \(1822\)) From: David Borman In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> Date: Thu, 14 Nov 2013 18:00:25 -0600 Content-Transfer-Encoding: quoted-printable Message-Id: <045012AD-1B4A-488A-B6D0-DC228CAA8380@weston.borman.com> References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> To: Richard Scheffenegger X-Mailer: Apple Mail (2.1822) Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 00:03:01 -0000 I have mixed feelings about changing the name of the document, but the = original title was from when it also included SACK, which was removed = before 1323 was published. So I guess it makes some sense to clarify = the title. On Nov 14, 2013, at 5:38 PM, Scheffenegger, Richard = wrote: > Ilpo, >=20 > Very well then. But I'll need a good speaker to tell me which one is = most correct >=20 > "TCP Window Scale and Timestamps option" (sounds like one combined = option) >=20 > "TCP Window Scale and Timestamps options" (sounds odd) This one. It sounds less odd if you switch it to =93TCP Timestamps and = Window Scale options=94, but I think we want to keep Timestamps as the = second one, given the pushback on it. But in any case, Options should be capitalized, because this is a = document title. -David Borman >=20 > "TCP Window Scale and TCP Timestamps options" (imho better) >=20 > "TCP Window Scale option and TCP Timestamps option" (least ambiguity, = but is this proper?) >=20 > (prefixed with a "The"?) >=20 >=20 > PS: I'm glad that we have come to the point to discuss the title :) >=20 >=20 > Richard Scheffenegger >=20 >=20 >> -----Original Message----- >> From: Ilpo J=E4rvinen [mailto:ilpo.jarvinen@helsinki.fi] >> Sent: Donnerstag, 14. November 2013 17:46 >> To: Scheffenegger, Richard >> Cc: Zimmermann, Alexander; David Borman; Pasi Sarolahti >> (pasi.sarolahti@iki.fi); tcpm@ietf.org Extensions >> Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >>=20 >> On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: >>=20 >>> Would you like to see WS and TS in the title, or keep the old one? >>=20 >> For some reason I've never liked RFC1323 title to begin with so I'd = prefer >> the actual option names here. It would actually be rather misleading = to >> say everything that benefits from WS today would have particularly = "high >> performance". >>=20 >> BTW, I just noticed that 1323 is also a counter-example: >>=20 >> RFC1072 TCP Extensions for Long-Delay Paths >> RFC1185 TCP Extension for High-Speed Paths >> RFC1323 TCP Extensions for High Performance >> (+RFC2018 TCP Selective Acknowledgment Options) >>=20 >>=20 >> -- >> i. > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm From l.wood@surrey.ac.uk Thu Nov 14 16:31:11 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 64EF511E8165 for ; Thu, 14 Nov 2013 16:31:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.321 X-Spam-Level: X-Spam-Status: No, score=-3.321 tagged_above=-999 required=5 tests=[AWL=-0.723, BAYES_00=-2.599, UNPARSEABLE_RELAY=0.001] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iLGc1jLmg+xw for ; Thu, 14 Nov 2013 16:31:07 -0800 (PST) Received: from mail1.bemta5.messagelabs.com (mail1.bemta5.messagelabs.com [195.245.231.152]) by ietfa.amsl.com (Postfix) with ESMTP id AA39111E8150 for ; Thu, 14 Nov 2013 16:31:06 -0800 (PST) Received: from [85.158.136.51:4924] by server-16.bemta-5.messagelabs.com id E6/DC-17144-94B65825; Fri, 15 Nov 2013 00:31:05 +0000 X-Env-Sender: l.wood@surrey.ac.uk X-Msg-Ref: server-15.tower-49.messagelabs.com!1384475465!17674198!1 X-Originating-IP: [131.227.200.39] X-StarScan-Received: X-StarScan-Version: 6.9.13; banners=-,-,- X-VirusChecked: Checked Received: (qmail 28774 invoked from network); 15 Nov 2013 00:31:05 -0000 Received: from exht012p.surrey.ac.uk (HELO EXHT012P.surrey.ac.uk) (131.227.200.39) by server-15.tower-49.messagelabs.com with AES128-SHA encrypted SMTP; 15 Nov 2013 00:31:05 -0000 Received: from EXMB01CMS.surrey.ac.uk ([169.254.1.22]) by EXHT012P.surrey.ac.uk ([131.227.200.39]) with mapi; Fri, 15 Nov 2013 00:31:03 +0000 From: To: , Date: Fri, 15 Nov 2013 00:30:07 +0000 Thread-Topic: [tcpm] Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnAgACUnYD//4J4qIAAk5OA///q/7CAABCHFg== Message-ID: <290E20B455C66743BE178C5C84F1240847E5103780@EXMB01CMS.surrey.ac.uk> References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> , <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> In-Reply-To: <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> Accept-Language: en-US, en-GB Content-Language: en-GB X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US, en-GB Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: tcpm@ietf.org, David.Borman@quantum.com Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 00:31:11 -0000 "TCP options for window scaling and for timestamps" Needs to start with "TCP options" to set context. Lloyd Wood http://sat-net.com/L.Wood/ ________________________________________ From: tcpm-bounces@ietf.org [tcpm-bounces@ietf.org] On Behalf Of Scheffeneg= ger, Richard [rs@netapp.com] Sent: 14 November 2013 23:38 To: Ilpo J=E4rvinen Cc: tcpm@ietf.org Extensions; David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? Ilpo, Very well then. But I'll need a good speaker to tell me which one is most c= orrect "TCP Window Scale and Timestamps option" (sounds like one combined option) "TCP Window Scale and Timestamps options" (sounds odd) "TCP Window Scale and TCP Timestamps options" (imho better) "TCP Window Scale option and TCP Timestamps option" (least ambiguity, but i= s this proper?) (prefixed with a "The"?) PS: I'm glad that we have come to the point to discuss the title :) Richard Scheffenegger > -----Original Message----- > From: Ilpo J=E4rvinen [mailto:ilpo.jarvinen@helsinki.fi] > Sent: Donnerstag, 14. November 2013 17:46 > To: Scheffenegger, Richard > Cc: Zimmermann, Alexander; David Borman; Pasi Sarolahti > (pasi.sarolahti@iki.fi); tcpm@ietf.org Extensions > Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? > > On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: > > > Would you like to see WS and TS in the title, or keep the old one? > > For some reason I've never liked RFC1323 title to begin with so I'd prefe= r > the actual option names here. It would actually be rather misleading to > say everything that benefits from WS today would have particularly "high > performance". > > BTW, I just noticed that 1323 is also a counter-example: > > RFC1072 TCP Extensions for Long-Delay Paths > RFC1185 TCP Extension for High-Speed Paths > RFC1323 TCP Extensions for High Performance > (+RFC2018 TCP Selective Acknowledgment Options) > > > -- > i. _______________________________________________ tcpm mailing list tcpm@ietf.org https://www.ietf.org/mailman/listinfo/tcpm From Alexander.Zimmermann@netapp.com Fri Nov 15 02:01:48 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A5A2111E814D for ; Fri, 15 Nov 2013 02:01:48 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -10.599 X-Spam-Level: X-Spam-Status: No, score=-10.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gWohGbkgBFAT for ; Fri, 15 Nov 2013 02:01:44 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id D36BD11E8150 for ; Fri, 15 Nov 2013 02:01:44 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,706,1378882800"; d="asc'?scan'208";a="115906292" Received: from vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) by mx12-out.netapp.com with ESMTP; 15 Nov 2013 02:01:44 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.215]) by vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) with mapi id 14.03.0123.003; Fri, 15 Nov 2013 02:01:43 -0800 From: "Zimmermann, Alexander" To: "L.Wood@surrey.ac.uk" Thread-Topic: [tcpm] Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnAgACUnYD//4J4qIAAk5OA///q/7CAABCHFoABJc8A Date: Fri, 15 Nov 2013 10:01:42 +0000 Message-ID: <2A8B981C-45DF-4D89-A5FA-34B0ACFACE4C@netapp.com> References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> , <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> <290E20B455C66743BE178C5C84F1240847E5103780@EXMB01CMS.surrey.ac.uk> In-Reply-To: <290E20B455C66743BE178C5C84F1240847E5103780@EXMB01CMS.surrey.ac.uk> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.117] Content-Type: multipart/signed; boundary="Apple-Mail=_E3AF39FD-E379-4462-B092-36EEBCF1BAC5"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" , David Borman Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 10:01:48 -0000 --Apple-Mail=_E3AF39FD-E379-4462-B092-36EEBCF1BAC5 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 Am 15.11.2013 um 01:30 schrieb L.Wood@surrey.ac.uk: > "TCP options for window scaling and for timestamps=93 I like this one. Maybe we can omit the 2nd for, but I=92m not a native speaker. Do we need to talk about RTTM/PAWS in the title? Eg Like this: TCP options for window scaling and timestamps and their = applications >=20 > Needs to start with "TCP options" to set context. >=20 > Lloyd Wood > http://sat-net.com/L.Wood/ >=20 >=20 > ________________________________________ > From: tcpm-bounces@ietf.org [tcpm-bounces@ietf.org] On Behalf Of = Scheffenegger, Richard [rs@netapp.com] > Sent: 14 November 2013 23:38 > To: Ilpo J=E4rvinen > Cc: tcpm@ietf.org Extensions; David Borman > Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >=20 > Ilpo, >=20 > Very well then. But I'll need a good speaker to tell me which one is = most correct >=20 > "TCP Window Scale and Timestamps option" (sounds like one combined = option) >=20 > "TCP Window Scale and Timestamps options" (sounds odd) >=20 > "TCP Window Scale and TCP Timestamps options" (imho better) >=20 > "TCP Window Scale option and TCP Timestamps option" (least ambiguity, = but is this proper?) >=20 > (prefixed with a "The"?) >=20 >=20 > PS: I'm glad that we have come to the point to discuss the title :) >=20 >=20 > Richard Scheffenegger >=20 >=20 >> -----Original Message----- >> From: Ilpo J=E4rvinen [mailto:ilpo.jarvinen@helsinki.fi] >> Sent: Donnerstag, 14. November 2013 17:46 >> To: Scheffenegger, Richard >> Cc: Zimmermann, Alexander; David Borman; Pasi Sarolahti >> (pasi.sarolahti@iki.fi); tcpm@ietf.org Extensions >> Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >>=20 >> On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: >>=20 >>> Would you like to see WS and TS in the title, or keep the old one? >>=20 >> For some reason I've never liked RFC1323 title to begin with so I'd = prefer >> the actual option names here. It would actually be rather misleading = to >> say everything that benefits from WS today would have particularly = "high >> performance". >>=20 >> BTW, I just noticed that 1323 is also a counter-example: >>=20 >> RFC1072 TCP Extensions for Long-Delay Paths >> RFC1185 TCP Extension for High-Speed Paths >> RFC1323 TCP Extensions for High Performance >> (+RFC2018 TCP Selective Acknowledgment Options) >>=20 >>=20 >> -- >> i. > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm --Apple-Mail=_E3AF39FD-E379-4462-B092-36EEBCF1BAC5 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKF8QYACgkQdyiq39b9uS7GmACfZBJJOQtquw6LydYoznm9ObLE /0MAn0kr+/iHqS9KrkGs8vz7eVEBxK4V =v7ba -----END PGP SIGNATURE----- --Apple-Mail=_E3AF39FD-E379-4462-B092-36EEBCF1BAC5-- From prvs=9031597c83=david.borman@quantum.com Fri Nov 15 07:26:32 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 7AB3911E81B5 for ; Fri, 15 Nov 2013 07:26:31 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.265 X-Spam-Level: X-Spam-Status: No, score=-3.265 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, IP_NOT_FRIENDLY=0.334, RCVD_IN_DNSWL_LOW=-1] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HtF-ZOZocSKn for ; Fri, 15 Nov 2013 07:26:23 -0800 (PST) Received: from mx0b-000ceb01.pphosted.com (mx0b-000ceb01.pphosted.com [67.231.152.126]) by ietfa.amsl.com (Postfix) with ESMTP id 9BB3911E81A7 for ; Fri, 15 Nov 2013 07:26:20 -0800 (PST) Received: from pps.filterd (m0001151 [127.0.0.1]) by mx0b-000ceb01.pphosted.com (8.14.5/8.14.5) with SMTP id rAFFMc2m020657; Fri, 15 Nov 2013 07:26:19 -0800 Received: from ppoxedge1.quantum.com ([146.174.252.27]) by mx0b-000ceb01.pphosted.com with ESMTP id 1g5fft0faw-4 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=NOT); Fri, 15 Nov 2013 07:26:19 -0800 Received: from PPOMSG2.QUANTUM.com (10.50.35.27) by PPOXEDGE1.quantum.com (146.174.252.27) with Microsoft SMTP Server (TLS) id 14.2.318.1; Fri, 15 Nov 2013 08:25:34 -0700 Received: from PPOMSG1.QUANTUM.com ([10.50.35.26]) by ppomsg2 ([10.50.35.27]) with mapi id 14.02.0318.001; Fri, 15 Nov 2013 08:25:54 -0700 From: David Borman To: "Zimmermann, Alexander" Thread-Topic: [tcpm] Is RFC1323bis' title still appropriate? Thread-Index: AQHO4UIHPG5V/lmhfEOzjestTsDhAJokyvnAgACD2oCAAAiTgIAADXiAgABzFYCAAA6MgIAAn7MAgABakgA= Date: Fri, 15 Nov 2013 15:25:53 +0000 Message-ID: References: <8D54398D-C3A0-47D7-AAB5-922A8FB7B9E4@netapp.com> <012C3117EDDB3C4781FD802A8C27DD4F25E94B57@SACEXCMBX02-PRD.hq.netapp.com>, <977B28B6-E1A6-4C89-AED3-FE14BBA057D4@netapp.com> , <012C3117EDDB3C4781FD802A8C27DD4F25E96F48@SACEXCMBX02-PRD.hq.netapp.com> <290E20B455C66743BE178C5C84F1240847E5103780@EXMB01CMS.surrey.ac.uk> <2A8B981C-45DF-4D89-A5FA-34B0ACFACE4C@netapp.com> In-Reply-To: <2A8B981C-45DF-4D89-A5FA-34B0ACFACE4C@netapp.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [192.168.110.1] Content-ID: <5068D1DA9EDFB649B5FBBB7B1BE5FF2A@QUANTUM.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.10.8794, 1.0.14, 0.0.0000 definitions=2013-11-15_02:2013-11-15, 2013-11-15, 1970-01-01 signatures=0 X-Proofpoint-Spam-Details: rule=notspam policy=default score=0 spamscore=0 suspectscore=0 phishscore=0 adultscore=0 bulkscore=0 classifier=spam adjust=0 reason=mlx scancount=1 engine=7.0.1-1305240000 definitions=main-1311150087 Content-Type: text/plain; charset="Windows-1252" Cc: "tcpm@ietf.org Extensions" , "L.Wood@surrey.ac.uk" Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 15:26:32 -0000 On Nov 15, 2013, at 4:01 AM, Zimmermann, Alexander wrote: >=20 > Am 15.11.2013 um 01:30 schrieb L.Wood@surrey.ac.uk: >=20 >> "TCP options for window scaling and for timestamps=93 >=20 > I like this one. Maybe we can omit the 2nd for, but I=92m not a > native speaker. >=20 > Do we need to talk about RTTM/PAWS in the title? >=20 > Eg Like this: TCP options for window scaling and timestamps and their app= lications Right, RFC1323 isn=92t just about defining these two options, but also PAWS= and RTTM, mechanisms that use them. It=92s first about identifying the is= sues when running TCP at high rates of speed and/or large amounts of data, = i.e., a large delay*bandwidth, and then the options and algorithms to addre= ss those issues. That=92s why I said I had mixed feelings about changing t= he title. The first sentence of the abstract is: This document specifies a set of TCP extensions to improve performance over paths with a large bandwidth * delay product and to provide reliable operation over very high-speed paths. The original title, "TCP Extensions for High Performance=94, is compact and= encompasses not only the specific defined options, but also the algorithms= such as PAWS and RTTM. So the more I think about it, the more I prefer to= just leave as it is. It just gets too unwieldy if you try and put everyth= ing explicitly into it: "TCP Options for Window Scaling and Timestamps, and the Protection Against Wrapped Sequences and Round Trip Time Measurement Algorithms=94 It=92s a document title, not an all encompassing descriptive statement of w= hat is in the document. Keep it short and simple, and the original title i= s just that. In the document, =93High Performance=94 is defined for connec= tions that suffer performance problems, specifically for large delay*bandwi= dth situations, referred to as a =93long, fat network" (LFN). The original= objection was about =93High Performance=94 in the title, so maybe change j= ust that part: =93TCP Extensions for Long Fat Networks=94 =93TCP Extensions for LFNs=94 But I don=92t like either of those more than the original title. So I=92d like to leave the title as it is, since it is an updated version o= f the original, not an entirely new document. There has to be sufficient r= eason to change the title, such as it is inaccurate or misleading, and I do= n=92t think it is. We don=92t change the title just because we think we ca= n come up with a better one. -David Borman >=20 >>=20 >> Needs to start with "TCP options" to set context. >>=20 >> Lloyd Wood >> http://sat-net.com/L.Wood/ >>=20 >>=20 >> ________________________________________ >> From: tcpm-bounces@ietf.org [tcpm-bounces@ietf.org] On Behalf Of Scheffe= negger, Richard [rs@netapp.com] >> Sent: 14 November 2013 23:38 >> To: Ilpo J=E4rvinen >> Cc: tcpm@ietf.org Extensions; David Borman >> Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >>=20 >> Ilpo, >>=20 >> Very well then. But I'll need a good speaker to tell me which one is mos= t correct >>=20 >> "TCP Window Scale and Timestamps option" (sounds like one combined optio= n) >>=20 >> "TCP Window Scale and Timestamps options" (sounds odd) >>=20 >> "TCP Window Scale and TCP Timestamps options" (imho better) >>=20 >> "TCP Window Scale option and TCP Timestamps option" (least ambiguity, bu= t is this proper?) >>=20 >> (prefixed with a "The"?) >>=20 >>=20 >> PS: I'm glad that we have come to the point to discuss the title :) >>=20 >>=20 >> Richard Scheffenegger >>=20 >>=20 >>> -----Original Message----- >>> From: Ilpo J=E4rvinen [mailto:ilpo.jarvinen@helsinki.fi] >>> Sent: Donnerstag, 14. November 2013 17:46 >>> To: Scheffenegger, Richard >>> Cc: Zimmermann, Alexander; David Borman; Pasi Sarolahti >>> (pasi.sarolahti@iki.fi); tcpm@ietf.org Extensions >>> Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >>>=20 >>> On Thu, 14 Nov 2013, Scheffenegger, Richard wrote: >>>=20 >>>> Would you like to see WS and TS in the title, or keep the old one? >>>=20 >>> For some reason I've never liked RFC1323 title to begin with so I'd pre= fer >>> the actual option names here. It would actually be rather misleading to >>> say everything that benefits from WS today would have particularly "high >>> performance". >>>=20 >>> BTW, I just noticed that 1323 is also a counter-example: >>>=20 >>> RFC1072 TCP Extensions for Long-Delay Paths >>> RFC1185 TCP Extension for High-Speed Paths >>> RFC1323 TCP Extensions for High Performance >>> (+RFC2018 TCP Selective Acknowledgment Options) >>>=20 >>>=20 >>> -- >>> i. >> _______________________________________________ >> tcpm mailing list >> tcpm@ietf.org >> https://www.ietf.org/mailman/listinfo/tcpm >> _______________________________________________ >> tcpm mailing list >> tcpm@ietf.org >> https://www.ietf.org/mailman/listinfo/tcpm >=20 > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm ---------------------------------------------------------------------- The information contained in this transmission may be confidential. Any dis= closure, copying, or further distribution of confidential information is no= t permitted unless such privilege is explicitly granted in writing by Quant= um. Quantum reserves the right to have electronic communications, including= email and attachments, sent across its networks filtered through anti viru= s and spam software programs and retain such messages in order to comply wi= th applicable data security and retention requirements. Quantum is not resp= onsible for the proper and complete transmission of the substance of this c= ommunication or for any delay in its receipt. From mallman@icir.org Fri Nov 15 07:35:10 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2346711E81D9 for ; Fri, 15 Nov 2013 07:35:03 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -106.599 X-Spam-Level: X-Spam-Status: No, score=-106.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id erbj8TMRtURg for ; Fri, 15 Nov 2013 07:34:57 -0800 (PST) Received: from fruitcake.ICSI.Berkeley.EDU (fruitcake.ICSI.Berkeley.EDU [192.150.186.11]) by ietfa.amsl.com (Postfix) with ESMTP id 5993E11E81D5 for ; Fri, 15 Nov 2013 07:34:57 -0800 (PST) Received: from lawyers.icir.org (envoy.icir.org [192.150.187.30]) by fruitcake.ICSI.Berkeley.EDU (8.12.11.20060614/8.12.11) with ESMTP id rAFFYtni024019; Fri, 15 Nov 2013 07:34:55 -0800 (PST) Received: from lawyers.icir.org (localhost [127.0.0.1]) by lawyers.icir.org (Postfix) with ESMTP id 5A8A1278617E; Fri, 15 Nov 2013 10:34:54 -0500 (EST) To: David Borman From: Mark Allman In-Reply-To: Organization: International Computer Science Institute (ICSI) Song-of-the-Day: Mrs. Robinson X-URL-0: http://www.icir.org/mallman-files/Document93276.xlsx MIME-Version: 1.0 Content-Type: multipart/signed; boundary="--------ma16156-1"; micalg=pgp-sha1; protocol="application/pgp-signature" Date: Fri, 15 Nov 2013 10:34:54 -0500 Sender: mallman@icir.org Message-Id: <20131115153454.5A8A1278617E@lawyers.icir.org> Cc: "tcpm@ietf.org Extensions" , "L.Wood@surrey.ac.uk" Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list Reply-To: mallman@icir.org List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 15:35:10 -0000 ----------ma16156-1 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable > So I=92d like to leave the title as it is,=20 Amen. Even if there is a better title to be had it isn't better enough to spend much effort on. So, my druthers is that this thread ends. :-) allman ----------ma16156-1 Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (Darwin) iEYEARECAAYFAlKGPx4ACgkQWyrrWs4yIs53kgCgigkXMthv4W5BMU/DywXhxOOI +lkAn3p+MPvjC8az5rg9Gqhi4f2eggMH =IJX/ -----END PGP SIGNATURE----- ----------ma16156-1-- From rs@netapp.com Fri Nov 15 07:48:34 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 063F811E81B3 for ; Fri, 15 Nov 2013 07:48:34 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.599 X-Spam-Level: X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id B9cflJfFlKij for ; Fri, 15 Nov 2013 07:48:28 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id E278611E821A for ; Fri, 15 Nov 2013 07:45:42 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,708,1378882800"; d="scan'208";a="75427515" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx11-out.netapp.com with ESMTP; 15 Nov 2013 07:45:17 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.86]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Fri, 15 Nov 2013 07:44:41 -0800 From: "Scheffenegger, Richard" To: "mallman@icir.org" , David Borman Thread-Topic: [tcpm] Is RFC1323bis' title still appropriate? Thread-Index: AQHO4hg7PG5V/lmhfEOzjestTsDhAJombmSA Date: Fri, 15 Nov 2013 15:44:31 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25E9A300@SACEXCMBX02-PRD.hq.netapp.com> References: <20131115153454.5A8A1278617E@lawyers.icir.org> In-Reply-To: <20131115153454.5A8A1278617E@lawyers.icir.org> Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.51] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" , "L.Wood@surrey.ac.uk" Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 15:48:34 -0000 Very well then; I'll submit the version fixing the nits and updating the abstract and one p= aragraph speaking about the validity of TSecr. Any objections against calling for WGLC then? Richard Scheffenegger > -----Original Message----- > From: tcpm-bounces@ietf.org [mailto:tcpm-bounces@ietf.org] On Behalf Of > Mark Allman > Sent: Freitag, 15. November 2013 16:35 > To: David Borman > Cc: tcpm@ietf.org Extensions; L.Wood@surrey.ac.uk > Subject: Re: [tcpm] Is RFC1323bis' title still appropriate? >=20 >=20 > > So I'd like to leave the title as it is, >=20 > Amen. >=20 > Even if there is a better title to be had it isn't better enough to spend > much effort on. So, my druthers is that this thread ends. :-) >=20 > allman >=20 >=20 From internet-drafts@ietf.org Fri Nov 15 08:06:53 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B52DC11E820C; Fri, 15 Nov 2013 08:06:53 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -101.841 X-Spam-Level: X-Spam-Status: No, score=-101.841 tagged_above=-999 required=5 tests=[AWL=0.759, BAYES_00=-2.599, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JGfNYFpcObXv; Fri, 15 Nov 2013 08:06:53 -0800 (PST) Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id AF4DC11E81C1; Fri, 15 Nov 2013 08:04:26 -0800 (PST) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: internet-drafts@ietf.org To: i-d-announce@ietf.org X-Test-IDTracker: no X-IETF-IDTracker: 4.83.p1 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20131115160426.16415.97614.idtracker@ietfa.amsl.com> Date: Fri, 15 Nov 2013 08:04:26 -0800 Cc: tcpm@ietf.org Subject: [tcpm] I-D Action: draft-ietf-tcpm-1323bis-17.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 15 Nov 2013 16:06:53 -0000 A New Internet-Draft is available from the on-line Internet-Drafts director= ies. This draft is a work item of the TCP Maintenance and Minor Extensions Work= ing Group of the IETF. Title : TCP Extensions for High Performance Author(s) : David Borman Bob Braden Van Jacobson Richard Scheffenegger Filename : draft-ietf-tcpm-1323bis-17.txt Pages : 49 Date : 2013-11-15 Abstract: This document specifies a set of TCP extensions to improve performance over paths with a large bandwidth * delay product and to provide reliable operation over very high-speed paths. It defines the TCP Window Scale (WS) option and the TCP Timestamps (TS) option and their semantics. The Window Scale option is used to support larger receive windows, while the Timestamps option can be used for at least two distinct mechanisms, PAWS (Protection Against Wrapped Sequences) and RTTM (Round Trip Time Measurement), that are also described herein. This document obsoletes RFC1323 and describes changes from it. The IETF datatracker status page for this draft is: https://datatracker.ietf.org/doc/draft-ietf-tcpm-1323bis There's also a htmlized version available at: http://tools.ietf.org/html/draft-ietf-tcpm-1323bis-17 A diff from the previous version is available at: http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-1323bis-17 Please note that it may take a couple of minutes from the time of submission until the htmlized version and diff are available at tools.ietf.org. Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ From Alexander.Zimmermann@netapp.com Mon Nov 18 04:43:25 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 18CE211E84CD for ; Mon, 18 Nov 2013 04:43:25 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.599 X-Spam-Level: X-Spam-Status: No, score=-6.599 tagged_above=-999 required=5 tests=[AWL=-4.000, BAYES_00=-2.599] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OdkWdDdIMGRk for ; Mon, 18 Nov 2013 04:43:21 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 1241A11E84B6 for ; Mon, 18 Nov 2013 04:43:19 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,723,1378882800"; d="asc'?scan'208";a="76295463" Received: from vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) by mx11-out.netapp.com with ESMTP; 18 Nov 2013 04:43:18 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.58]) by vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) with mapi id 14.03.0158.001; Mon, 18 Nov 2013 04:43:18 -0800 From: "Zimmermann, Alexander" To: "tcpm@ietf.org Extensions" Thread-Topic: New Version Notification for draft-zimmermann-tcpm-reordering-detection-00.txt Thread-Index: AQHO5FvB4MYm6JZSWUSEfb5O3dRddQ== Date: Mon, 18 Nov 2013 12:43:17 +0000 Message-ID: References: <20131118112709.18294.35017.idtracker@ietfa.amsl.com> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.115] Content-Type: multipart/signed; boundary="Apple-Mail=_ABBD418B-888D-4C44-8926-E3ADE85614E3"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Subject: [tcpm] Fwd: New Version Notification for draft-zimmermann-tcpm-reordering-detection-00.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Nov 2013 12:43:25 -0000 --Apple-Mail=_ABBD418B-888D-4C44-8926-E3ADE85614E3 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii Anfang der weitergeleiteten Nachricht: > Von: > Betreff: New Version Notification for = draft-zimmermann-tcpm-reordering-detection-00.txt > Datum: 18. November 2013 12:27:09 MEZ > An: Lennart Schulte , Alexander Zimmermann = , Arnd Hannemann = , Carsten Wolff >=20 >=20 > A new version of I-D, = draft-zimmermann-tcpm-reordering-detection-00.txt > has been successfully submitted by Alexander Zimmermann and posted to = the > IETF repository. >=20 > Filename: draft-zimmermann-tcpm-reordering-detection > Revision: 00 > Title: Detection and Quantification of Packet = Reordering with TCP > Creation date: 2013-11-18 > Group: Individual Submission > Number of pages: 22 > URL: = http://www.ietf.org/internet-drafts/draft-zimmermann-tcpm-reordering-detec= tion-00.txt > Status: = http://datatracker.ietf.org/doc/draft-zimmermann-tcpm-reordering-detection= > Htmlized: = http://tools.ietf.org/html/draft-zimmermann-tcpm-reordering-detection-00 >=20 >=20 > Abstract: > This document specifies an algorithm for the detection and > quantification of packet reordering for TCP. In the absence of > explicit congestion notification from the network, TCP uses only > packet loss as an indication of congestion. One of the signals TCP > uses to determine loss is the arrival of three duplicate > acknowledgments. However, this heuristic is not always correct, > notably in the case when paths reorder packets. This results in > degraded performance. >=20 > The algorithm for the detection and quantification of reordering in > this document uses information gathered from the TCP Timestamps > Option, the TCP SACK Option and its DSACK extension. When a > reordering event is detected, the algorithm calculates a reordering > extent by determining the number of segments the reordered segment > was late with respect to its position in the sequence number space. > Additionally, the algorithm computes a second reordering extent that > is relative to the amount of outstanding data and thus provides a > better estimation of the reordering delay when other sender state > changes. >=20 >=20 >=20 >=20 > Please note that it may take a couple of minutes from the time of = submission > until the htmlized version and diff are available at tools.ietf.org. >=20 > The IETF Secretariat >=20 --Apple-Mail=_ABBD418B-888D-4C44-8926-E3ADE85614E3 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKKC2MACgkQdyiq39b9uS4wyQCfYPkSYtSEuIGPgkeOBthavz8Q 0uwAoPE1MXsa8UCDoRCc1aDLgsWSjcyN =+xI6 -----END PGP SIGNATURE----- --Apple-Mail=_ABBD418B-888D-4C44-8926-E3ADE85614E3-- From Alexander.Zimmermann@netapp.com Mon Nov 18 04:44:00 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 2172611E849B for ; Mon, 18 Nov 2013 04:44:00 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -8.599 X-Spam-Level: X-Spam-Status: No, score=-8.599 tagged_above=-999 required=5 tests=[AWL=2.000, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id JwhSazXXh4v9 for ; Mon, 18 Nov 2013 04:43:56 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id DB04911E84F3 for ; Mon, 18 Nov 2013 04:43:45 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,723,1378882800"; d="asc'?scan'208";a="116918876" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx12-out.netapp.com with ESMTP; 18 Nov 2013 04:43:45 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.58]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Mon, 18 Nov 2013 04:43:45 -0800 From: "Zimmermann, Alexander" To: "tcpm@ietf.org Extensions" Thread-Topic: New Version Notification for draft-zimmermann-tcpm-reordering-reaction-00.txt Thread-Index: AQHO5Fuxbh8bcx5Hpk24ngRzU+PSkg== Date: Mon, 18 Nov 2013 12:43:44 +0000 Message-ID: <45657C00-3AD2-489D-952B-60066FB516F9@netapp.com> References: <20131118124143.18312.83888.idtracker@ietfa.amsl.com> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.115] Content-Type: multipart/signed; boundary="Apple-Mail=_958F5FA7-0F74-4C92-B813-F7D0AFA38B2C"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Subject: [tcpm] Fwd: New Version Notification for draft-zimmermann-tcpm-reordering-reaction-00.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Nov 2013 12:44:00 -0000 --Apple-Mail=_958F5FA7-0F74-4C92-B813-F7D0AFA38B2C Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii Anfang der weitergeleiteten Nachricht: > Von: > Betreff: New Version Notification for = draft-zimmermann-tcpm-reordering-reaction-00.txt > Datum: 18. November 2013 13:41:43 MEZ > An: Lennart Schulte , Alexander Zimmermann = , Arnd Hannemann = , Carsten Wolff >=20 >=20 > A new version of I-D, draft-zimmermann-tcpm-reordering-reaction-00.txt > has been successfully submitted by Alexander Zimmermann and posted to = the > IETF repository. >=20 > Filename: draft-zimmermann-tcpm-reordering-reaction > Revision: 00 > Title: Making TCP Adaptively Robust to Non-Congestion = Events > Creation date: 2013-11-18 > Group: Individual Submission > Number of pages: 29 > URL: = http://www.ietf.org/internet-drafts/draft-zimmermann-tcpm-reordering-react= ion-00.txt > Status: = http://datatracker.ietf.org/doc/draft-zimmermann-tcpm-reordering-reaction > Htmlized: = http://tools.ietf.org/html/draft-zimmermann-tcpm-reordering-reaction-00 >=20 >=20 > Abstract: > This document specifies an adaptive Non-Congestion Robustness (aNCR) > mechanism for TCP. In the absence of explicit congestion > notification from the network, TCP uses only packet loss as an > indication of congestion. One of the signals TCP uses to determine > loss is the arrival of three duplicate acknowledgments. However, > this heuristic is not always correct, notably in the case when paths > reorder packets. This results in degraded performance. >=20 > TCP-aNCR is designed to mitigate this performance degradation by > adaptively increasing the number of duplicate acknowledgments > required to trigger loss recovery, based on the current state of the > connection, in an effort to better disambiguate true segment loss > from segment reordering. This document specifies the changes to TCP > and TCP-NCR (on which this specification build on) and discusses the > costs and benefits of these modifications. >=20 >=20 >=20 >=20 > Please note that it may take a couple of minutes from the time of = submission > until the htmlized version and diff are available at tools.ietf.org. >=20 > The IETF Secretariat >=20 --Apple-Mail=_958F5FA7-0F74-4C92-B813-F7D0AFA38B2C Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKKC34ACgkQdyiq39b9uS4eNQCeIW/TxqtfkUFeAeiM7xZ12xUH a1gAnAwaKpXFVbDx1L7Kxvcrob9BTGlQ =gGMt -----END PGP SIGNATURE----- --Apple-Mail=_958F5FA7-0F74-4C92-B813-F7D0AFA38B2C-- From pasi.sarolahti@iki.fi Mon Nov 18 06:50:39 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 1233F11E81F9 for ; Mon, 18 Nov 2013 06:50:39 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -100.74 X-Spam-Level: X-Spam-Status: No, score=-100.74 tagged_above=-999 required=5 tests=[BAYES_20=-0.74, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id oHvU0re9oeJm for ; Mon, 18 Nov 2013 06:50:33 -0800 (PST) Received: from jenni2.inet.fi (mta-out.inet.fi [195.156.147.13]) by ietfa.amsl.com (Postfix) with ESMTP id DD33611E822A for ; Mon, 18 Nov 2013 06:44:28 -0800 (PST) Received: from pc111.netlab.hut.fi (130.233.154.111) by jenni2.inet.fi (8.5.140.03) (authenticated as saropa-1) id 52775C990137B248 for tcpm@ietf.org; Mon, 18 Nov 2013 16:44:24 +0200 From: Pasi Sarolahti Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: quoted-printable Message-Id: Date: Mon, 18 Nov 2013 16:44:23 +0200 To: "tcpm@ietf.org (tcpm@ietf.org)" Mime-Version: 1.0 (Mac OS X Mail 7.0 \(1822\)) X-Mailer: Apple Mail (2.1822) Subject: [tcpm] WGLC for draft-ietf-tcpm-1323bis X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.12 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Nov 2013 14:50:39 -0000 Hi, The WG chairs believe that the latest version of draft-ietf-tcpm-1323bis = addresses the comments made so far. Therefore, this mail starts the = working group last call for draft-ietf-tcpm-1323bis-17, to be submitted = for publication as Proposed Standard. The WG last call ends in two = weeks, on Monday, December 2. Please send any comments to the TCPM = mailing list by then. The draft can be found at = http://tools.ietf.org/html/draft-ietf-tcpm-1323bis-17 - Pasi From gorry@erg.abdn.ac.uk Tue Nov 19 09:21:45 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D2DB51AE0E7 for ; Tue, 19 Nov 2013 09:21:45 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.526 X-Spam-Level: X-Spam-Status: No, score=-3.526 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_32=0.6, J_CHICKENPOX_46=0.6, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.525, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 7D_xvq1UMGfF for ; Tue, 19 Nov 2013 09:21:42 -0800 (PST) Received: from spey.erg.abdn.ac.uk (spey.erg.abdn.ac.uk [139.133.204.173]) by ietfa.amsl.com (Postfix) with ESMTP id 46EED1AE101 for ; Tue, 19 Nov 2013 09:21:40 -0800 (PST) Received: by spey.erg.abdn.ac.uk (Postfix, from userid 5001) id 8CDFB2B451F; Tue, 19 Nov 2013 17:21:29 +0000 (GMT) Received: from gorry-mac.erg.abdn.ac.uk (gorry-mac.erg.abdn.ac.uk [139.133.207.5]) by spey.erg.abdn.ac.uk (Postfix) with ESMTPSA id B3B772B44B2; Tue, 19 Nov 2013 17:21:26 +0000 (GMT) Message-ID: <528B9E16.70602@erg.abdn.ac.uk> Date: Tue, 19 Nov 2013 17:21:26 +0000 From: Gorry Fairhurst Organization: The University of Aberdeen is a charity registered in Scotland, No SC013683. User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:24.0) Gecko/20100101 Thunderbird/24.1.0 MIME-Version: 1.0 To: Yuchung Cheng References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gorry@erg.abdn.ac.uk List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 19 Nov 2013 17:21:46 -0000 See response in-line... Many of these are small items, Somewhere down there is the significant point that I do not yet understand, i.e. what are the potential CC issues that need to be evaluated - especially how TFO respond to severe path overload? (see below) Gorry On 09/11/2013 18:06, Yuchung Cheng wrote: > Hi Gorry, > > Sorry for the last response. Too many good sessions and dinner parties > during this IETF for me to respond promptly. > > On Sat, Nov 2, 2013 at 4:09 PM, Gorry wrote: >> >> I have reviewed the current version of TFO and have the comments below. This is rather long ... >> >> Nits in Section 1 - I think the document here talks of servers and clients from a TCP endpoint perspective, this is normal for TCP, but the document also raises implications for apps designers who may have a very different view on what is intended by the word "server" and "client". To avoid doubt maybe you should explain this in the intro? > They are already explained in the "Terminology" section. > OK, but it's terse! >> >> Page 3, para 2 is not clear. I think it is just the detail of the wording, but it would be good to see clear text. >> In RFC 6936 we wrote (albeit slightly different problem, with a different protocol): "IPv6 nodes MUST provide a way for the application/protocol to indicate the set of ports that will be enabled to send datagrams with a zero UDP checksum. This may be implemented by enabling a transport mode using a socket API call when the socket is established, or by a similar mechanism. It may also be implemented by enabling the method for a pre-assigned static port" >> Last para section 2: to me, the requirements are not clear. >> >> I think this piece of text means: TCP stacks should NOT enable TCP-TFO by default. The stack should provide a global flag to enable this. An application wishing to use this must enable each TCP socket on a per service/per port basis. >> >> I think the other comments/questions maybe more significant. >> >> section 2 also suggests the main change is to allow data in SYN, I have noted that some people think this is the only change --- it is not. The bigger implication is that the document proposes to allow new data sent during the 3WHS. This change needs to be more clearly noted, sending an IW of data before there is an ACK on the return path is a significant change to TCP semantics and it's behaviour under heavy load. > > will revise sec2 and discuss w/ you before we publish the revision. > Let's do that. > >> >> Page 5 fast open, bullet 3. : This says the server can send data before the 3WHS completes. I am curious what this actually means, is this implied that the use is extricated to standards-track mechanisms, I.e. I am looking here for some explicit understanding of whether this implies IW3. Or is the experiment linked to both a proposal for iW10 and this new experiment? > will add "initial congestion window of " before the "... the > congestion control in .." to make it more clear. > or > >> >> Last para 4.1.3 the text says /should include remote port numbers too/ Can we be clear which port numbers are used. Is this the received server port number by the client? Actually, do not understand what is intended by the paragraph, can you provide an example to be sure what is meant? > a "remote" port number of the client means the "server port". but will > replace remote/server to avoid confusion. > That works. > The paragraph is added because of your suggestion to add negative caching. > OK >> >> Section 4.3.2. I hoped this would say something explicit about not sending SYN segments that are larger than the default MTU. This seems like a bad thing to do. > It's a bad thing to send any segment larger than the MTU. Why is that > specific to Fast Open? > ... Let me check: 1. TFO could allow PMTU to be cached across sessions. 2. Multi-homing/ Multipath routing could result in a different path for a new connection. 3. If TFO used a cahched PMTU with the SYN (or before the SYN-ACK), then these could be black-holded. - While local MTU is known, remote (or path) MTU can only be validated once there is a path... To me, allowing TFO to use larger MTU seems like an unnecessary extra complication to have to deal with. >> >> Why are the examples only for web? This is a TCP spec. > Sure there are a lot more applications can benefit from that, but Web > is what motivates our work and the application is complex enough to > demonstrate many aspects of Fast Open. If you have another good > examples, you are welcome to suggest one. > .. I could OR it mya be wise to just say something similar to what you have written above at the end of section 1? >> >> There are two sections 6.3.2. >> >> Second Section 6.3.2 - I think this applies to SSL/TLS in general, doesn't it? ... I think the section should address the general case, and then provide an example for https, rather than assuming a web-centric use. > > Yes it is generic, but anyone who knows SSL/TLS good enough knows that. > So could you name the section TLS over TCP ... and then give HTTP over TLS as an example? >> >> Section 7.1 first para last line.please check text, this pathology is not necessarily malicious!!! > sure we can s/malicious/pathological. does this deserves three !!!? > Fine (indeed, we want those nice people to fix all their middleboxes). >> The issue seems to be mainly related to NAPT where ports are used to discriminate senders? >> Also the shared NAPT can result in a range of RTT and PMTU to the same IP, but this is already possible with ECMP routing. >> So In this case there is need to understand the implications of using a cached RTT and cached PMTU value. What concerns me is that we need experimental data to show that this is safe - not just to realise that this may have no benefit for TFO. >> The lack of issues listed in section 7 raises serious concerns to me. I do not list that is provided currently as the only, or even most significant reasons why this is deemed to be experimental work. Indeed, the lack of identification of potential CC issues is why this document is in my opinion NOT ready to complete a WGLC - if the issues are not listed how can we evaluate if the method is safe enough to experiment or can later be evaluated properly. >> >> Here are some additional issues I would like considered by the group and if the group is happy these may be safe enough to recommend testing in the general internet, then I think we should document that these are things to be confirmed. >> >> 1) I raised on the list that a negative impact of using the SYN as a probe to check for new option support. The examples I gave were ECN interactions and IPv6 probing. ECN if used will probe for ECN support in the SYN segment. If we include data on the SYN and then there is a lack of response, it is assumed by the sender that both TFO and ECN are not supported. This is an implication of the model we use to negotiate for features that may be black-holed. It is not an issue with the method, but I think we need tone clear this is something that may happen. A similar case exists if a client attempts TFO to a server using IPv6 ... Is it assumed that the path does not support IPv6 or TFO (or actually both). It does not happen when multiple options are negotiated. > if a SYN-data is dropped, TFO will retransmit a regular SYN (with ECN if used). > May be good, I agree - but I do not see this in the text (RFC 3168 would allow this, it kind of suggests clearing ECT on retx). >> >> 2) the draft proposes caching RTT per path and using this for SYN timeout. This updates a standards track RFC, and will have implications in some network paths that may seriously degrade performance, in other cases it may improve performance. Examples of negative performance are cases where an initial packet can take much longer to process than a subsequent packet. This can and does occur in bandwidth-on-demand L2 networks, where the first packet causes path setup. It also occurs to some extent with policing and routing devices that build cached state from the first packet in a flow. I think this is one of the reasons why the RFC-series has to date not described a shortened RTT method for the SYN. > >> >> Another issue is that previous RTT (or PMTU) is not necessarily a good indication of future value in a different 5tuple, since ECMP etc can lead to very different path characteristics for a session with a different port number. The current text therefore proposes an experiment that is beyond what was previously considered safe. It does not specify a way to avoid the server making the same mistake next connection. > > We'll just take RTT caching out of the text if that makes people happy. > Maybe, although I'm not pushing for that, but it may be OK - if it is an orthogonal issue that TFO does not rely upon. If you want to mention caching these, I think the doc needs to explain more. >> >> 3) seeding the SYN RTT from a lower value makes the sender more aggressive in heavily congested networks. The sender is also made more aggressive by sending IW data packets before there is any indication the path can contain even a single data segment. This is significant change to standard behaviour. If the proposal is to use IW 3 (as standard) it still needs to be called out. If the proposal is to use a larger experimental number then I have concerns here that this is a significant change that needs an automated recovery method to prevent significant collateral damage on capacity-limited paths - there needs to be a way to stop a server doing this each time resulting in recurring loss! > > when the IW of packets are not all acked, loss recovery is triggered > and window is reduced. this is part of standard CC. > > also when you use "significant" four times, please back up with a good > theory or data or both. and abuse that word does not it make more > significant. > So let's see if we can see agree on what happens: - If the path is very lossey - severe congestion: TCP standard sends one SYN segment with some probability of loss, and if it sees loss backs off and retransmits the SYN. As (over)load increases, new sessions add 40B and the new sessions will often defer start-up, controlling their rate. - If TFO path is very lossey - severe congestion: IW full-sized segments are sent. Each new session does this, adding to the load. Each new session adds MSS*IW = 6000B (IW=3) or 15000B (IW=10). I concede this applies to severe congestion. Under this case, it seems 150x more traffic to me and this before before CC is engaged. Or is this wrong? >> >> 4) Are there any changes to PMTUD behaviour that need experience? PMTU was naturally cached and I am not sure if this changes the behaviour or not. It does of course allow a packet with a larger MTU to be sent before the path has been initially validated via the 3WHS (I.e. After a path change while idle, this can generate TCP segments in a SYN that are larger than the receiver advertised MSS - I suspect this is not crucial, but I would also assume that the SYN itself should not use an increased PMTU value to carry the data. Is this correct? > in our implementation, we use what ever MTU (possibly discovered in a > prior PMTU) cached to send SYN-data. > One option could be that the document could identify this as a topic to be explored (like RTT-caching, this does not seem to be a core technique required for FastOpen, but if you mention cachingthese, I think the WG should highlight that this can also say there are potential concerns and a need for experimentation? >> >> Are there other case where this change will impact the network for specific path characteristics? >> >> Have others thought through these issues and feel they are ok? Or at least ok for experimental deployment? >> >> Gorry >> P.S. A tiny NiT on 7.2: "Careful experimentation is necessary to evaluate if cookie-less TFO is practical." - I think the experiment that is needed by the IETF is "to evaluate if cookie-less TFO could be safe for deployment in the general Internet." From ietf-secretariat-reply@ietf.org Mon Nov 18 08:29:12 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 381B511E8135 for ; Mon, 18 Nov 2013 08:29:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -102.535 X-Spam-Level: X-Spam-Status: No, score=-102.535 tagged_above=-999 required=5 tests=[AWL=0.065, BAYES_00=-2.599, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100] Received: from mail.ietf.org ([12.22.58.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XB0CPXXlYdJv for ; Mon, 18 Nov 2013 08:29:10 -0800 (PST) Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 5599111E81C1 for ; Mon, 18 Nov 2013 08:26:59 -0800 (PST) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable To: tcpm@ietf.org X-Test-IDTracker: no X-IETF-IDTracker: 4.83.p1 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20131118162659.18268.14765.idtracker@ietfa.amsl.com> Date: Mon, 18 Nov 2013 08:26:59 -0800 From: IETF Secretariat X-Mailman-Approved-At: Tue, 19 Nov 2013 09:25:47 -0800 Subject: [tcpm] Milestones changed for tcpm WG X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 18 Nov 2013 16:29:12 -0000 Changed milestone "Submit update to RFC 1323 to the IESG for Proposed Standard RFC", added draft-ietf-tcpm-1323bis to milestone. URL: http://datatracker.ietf.org/wg/tcpm/charter/ From Alexander.Zimmermann@netapp.com Wed Nov 20 02:29:30 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 860131AD9AE for ; Wed, 20 Nov 2013 02:29:30 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -2.427 X-Spam-Level: X-Spam-Status: No, score=-2.427 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_NONE=-0.0001, RP_MATCHES_RCVD=-0.525, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id r7KH210ZUtDW for ; Wed, 20 Nov 2013 02:29:28 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id 3EE611AD8EE for ; Wed, 20 Nov 2013 02:29:28 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,736,1378882800"; d="asc'?scan'208";a="77038157" Received: from vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) by mx11-out.netapp.com with ESMTP; 20 Nov 2013 02:29:22 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.58]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.03.0158.001; Wed, 20 Nov 2013 02:29:22 -0800 From: "Zimmermann, Alexander" To: Martin Duke Thread-Topic: [tcpm] RFC 4614 bis comments Thread-Index: AQHO5dtfE62pBCLVnU67DTOrIIt3zA== Date: Wed, 20 Nov 2013 10:29:21 +0000 Message-ID: <87CC9397-9DE3-4CC4-B432-4B931355BFF7@netapp.com> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.116] Content-Type: multipart/signed; boundary="Apple-Mail=_E937E9C6-FDBF-41C9-9CBC-976B090F6E27"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" Subject: Re: [tcpm] RFC 4614 bis comments X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Nov 2013 10:29:30 -0000 --Apple-Mail=_E937E9C6-FDBF-41C9-9CBC-976B090F6E27 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 Hi Martin, thanks for reviewing the draft. Comments below. Alex Am 12.11.2013 um 05:35 schrieb Martin Duke : > Alexander, >=20 > First of all, let me emerge from deep lurking to thank you for = updating this document. It is long overdue. Although I haven't been as = involved in TCP work as I would have liked, perhaps I can offer some = constructive comments based on the original philosophy of the RFC's = organization. >=20 > Martin >=20 > ******** > EDITORIAL ISSUES: > Sec. 2: If RFC 5681 contains 3390 is 3390 a "core requirement?=93 You are right. I move rfc3390 into the (new) section 3.2. Congestion = Control Extensions > It seems like a novice could skip over it entirely. 5681 even refers = to 3390 for further background it they're interested. Recommendation: = move RFC 3390 to Section 6 (Support Documents). At the moment I suggest to not move RFC3390, RFC3042 and RFC3465 into = sec. 7. Support Documents. RFC 5861 does not obsolete any of these documents. All of them are still = =84valid=93. Maybe in further, in particular with 793bis, which obsolete a lot of RFC, we need a new subsection in = sec. 7. >=20 > Sec 3: I like how this section has been reorganized. Thanks ;-) >=20 > RFC 3042, 3465: At a minimum we should indicate this has been rolled = into RFC 5681. Good idea > At most we could strike the reference entirely. Recommendation: Move = to Section 6. See above. >=20 > Section 3.4: So I take it that, based on its inclusion in Section 3, = we specifically recommend F-RTO and Eifel Response over the = alternatives? If that is not the intent here than this section must be = substantially revised. IMO the roadmap =84recommend=93 nothing; it=92s a survey. =46rom the = introduction: =84This document is not an update of RFC 1122 and is not a rigorous standard for what needs to be implemented in TCP. This document is merely an informational roadmap that captures, organizes, and summarizes most of the RFC documents that a TCP implementer, experimenter, or student should be aware of. Particular comments or broad categorizations that this document makes about individual mechanisms and behaviors are not to be taken as definitive, nor should the content of this document alone influence implementation decisions.=93 and =84Note that the category of an RFC does not necessarily reflect its current relevance. For instance, RFC 5681 is nearly universally deployed although it is only a Draft Standard. Similarly, some Informational RFCs contain significant technical proposals for changing TCP.=93 If it is not clear that we do not make any implementation recommendation in this doc, we should instead work on introduction. Further, the introductions says: "This roadmap is divided into three main sections. Section 2 lists the RFCs that describe absolutely required TCP behaviors for proper functioning and interoperability. Further RFCs that describe strongly encouraged, but non-essential, behaviors are listed in Section 3. Experimental extensions that are not yet standard practices, but that potentially could be in the future, are described in Section 4.=93 IMO its clear, why for example Eifel response is in sec 4.4, since it is an Experimental draft. >=20 > Sec 6.6 Generally we do not separately list obsolete RFCs in this = document. =46rom the text I gather that everything but RFC 4022 is = obsolete in this section, so I recommend we keep the introductory text = and the entry for RFC 4022. I don=92t get this. I think you mean sec 7.6 and not 6.6, but still I = have no clue :-) The entire sec 7.6 is C&P from the old roadmap. Nothing that I changed. >=20 > **** > NITS: >=20 > P.3 para 6: "...important implementation practices that ARE not = currently documented=85" OK >=20 > RFC 6298, last sentence is worded ambiguously. Suggest "RFC 6298 = updates RFC 2988 by changing the initial RTO from 3s to 1s." (It can = currently be read to say that 2988 sets it to 1s.) OK >=20 > Sec 3.4. "implemented, and IPR status" -> "implemented, and concerns = about IPR status=93 OK >=20 > RFC 5682: "describes" -> =84described" OK >=20 > Sec 3.6. "Header compressions schemes... these overhead." --> "Header = compression schemes for TCP/IP like "RObust Header Compression (ROHC) = can significantly compress this overhead.=93 OK >=20 > Sec 3.7. > RFC 4953. "resets (RSTs)" -> "reset (RST)=93 OK >=20 > RFC 5961. "are less vulnerable" -> "less vulnerable=93 OK >=20 > Sec. 4.1 RFC 3124 : "which supports sharing information among TCP = connections as well" delete the words "as well". Taking it outside of = connections is not formally the same as sharing it between connections. OK >=20 > Sec. 4.4. "simultaneous" -> "simultaneously"; "additinal" -> = =84additional" OK >=20 > Sec. 5 "reclassifies" -> "reclassified"; "seen widespread" -> "seen = widespread deployment=93 OK >=20 > RFC 721: "has not been included" -> "was not included=93 OK >=20 > RFC 1078: "propose a protocol" -> "proposes a protocol=93 OK >=20 > RFC 6013: para 1: "to negotiates" -> "to negotiate";=20 > para 2: "describes method to extent" -> "describes a method to extend" > para 3: "Although the RFC 6013 is publish in 2011" -> "Although RFC = 6013 was published in 2011=93 OK >=20 > Section 6. "implementation advices can be found in Section 6.5" -> = "implementation advice can be found in Section 6.5=93 OK >=20 > Section 6.1. Is RFC 675 really "infamous?" Suggest "fundamental=93. :-) >=20 > Section 6.2 RFC 3439 "extents" -> =84extends=93 OK >=20 > Sec 6.4 RFC 5166 "Among others" -> "Among other topics=93 OK >=20 > Sec 6.7 RFC 5783 "The focus of the document are on" -> "The focus of = the document is=93 OK >=20 > Sec 6.8 RFC 1337 "Hazardsin" --> "Hazards in=93 OK > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm --Apple-Mail=_E937E9C6-FDBF-41C9-9CBC-976B090F6E27 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKMjv4ACgkQdyiq39b9uS6RCQCfU8aUC25D2Nr+/+KsZNTbuK2U SfcAn0RD1P1UG88PRXsbxHfolWeDg+mj =J2Mk -----END PGP SIGNATURE----- --Apple-Mail=_E937E9C6-FDBF-41C9-9CBC-976B090F6E27-- From lars@netapp.com Wed Nov 20 02:44:06 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5F1E51AD8F2 for ; Wed, 20 Nov 2013 02:44:06 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -7.427 X-Spam-Level: X-Spam-Status: No, score=-7.427 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.525, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id nfUUEpz-uNzP for ; Wed, 20 Nov 2013 02:44:04 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id CE2FD1AD8CD for ; Wed, 20 Nov 2013 02:44:04 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,736,1378882800"; d="asc'?scan'208";a="117806161" Received: from vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) by mx12-out.netapp.com with ESMTP; 20 Nov 2013 02:43:58 -0800 Received: from SACEXCMBX01-PRD.hq.netapp.com ([169.254.2.244]) by vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) with mapi id 14.03.0123.003; Wed, 20 Nov 2013 02:43:58 -0800 From: "Eggert, Lars" To: "tcpm@ietf.org" Thread-Topic: [tsvwg] I-D Action: draft-ietf-tsvwg-behave-requirements-update-00.txt Thread-Index: AQHO5bGG4a4BGuzkTE2ZKXecgMp20g== Date: Wed, 20 Nov 2013 10:43:57 +0000 Message-ID: <77557979-8DFA-4A0D-AE6D-7FFA99A4EE1C@netapp.com> References: <20131120052947.13956.44214.idtracker@ietfa.amsl.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.117] Content-Type: multipart/signed; boundary="Apple-Mail=_EED4B1CA-E2DF-4D88-B098-3CC9F3373AA4"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Subject: [tcpm] Fwd: [tsvwg] I-D Action: draft-ietf-tsvwg-behave-requirements-update-00.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Nov 2013 10:44:06 -0000 --Apple-Mail=_EED4B1CA-E2DF-4D88-B098-3CC9F3373AA4 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=us-ascii Quite a bit of TCP-related text in here. Would benefit from WG review. Begin forwarded message: > From: > Subject: [tsvwg] I-D Action: = draft-ietf-tsvwg-behave-requirements-update-00.txt > Date: November 20, 2013 at 0:29:47 EST > To: > Cc: >=20 >=20 > A New Internet-Draft is available from the on-line Internet-Drafts = directories. > This draft is a work item of the Transport Area Working Group Working = Group of the IETF. >=20 > Title : Network Address Translation (NAT) Behavioral = Requirements Updates > Author(s) : Reinaldo Penno > Simon Perreault > Sarat Kamiset > Mohamed Boucadair > Kengo Naito > Filename : = draft-ietf-tsvwg-behave-requirements-update-00.txt > Pages : 13 > Date : 2013-11-19 >=20 > Abstract: > This document clarifies and updates several requirements of RFC4787, > RFC5382 and RFC5508 based on operational and development experience. > The focus of this document is NAPT44. >=20 >=20 >=20 > The IETF datatracker status page for this draft is: > = https://datatracker.ietf.org/doc/draft-ietf-tsvwg-behave-requirements-upda= te >=20 > There's also a htmlized version available at: > = http://tools.ietf.org/html/draft-ietf-tsvwg-behave-requirements-update-00 >=20 >=20 > Please note that it may take a couple of minutes from the time of = submission > until the htmlized version and diff are available at tools.ietf.org. >=20 > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ >=20 --Apple-Mail=_EED4B1CA-E2DF-4D88-B098-3CC9F3373AA4 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- iQCVAwUBUoySatZcnpRveo1xAQKmiQP9F5K2GOCVjq8rlkfGKyCRhyjSWAgCJrnh h8pXrBQPcyUq3b5iIoNWLwL0oiMAxRzT/ePeygi7+0R+PHDamTj1nNHYNf83rAi0 3NoS5cEt8il2RMJBTCfkAj3krn3aeb+BMg7q2yaOaA57cS4nHUBc48QgRbwxsnUl 9fOPtGw7DWk= =Qitv -----END PGP SIGNATURE----- --Apple-Mail=_EED4B1CA-E2DF-4D88-B098-3CC9F3373AA4-- From internet-drafts@ietf.org Thu Nov 21 05:07:58 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D37541AE130; Thu, 21 Nov 2013 05:07:58 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.9 X-Spam-Level: X-Spam-Status: No, score=-1.9 tagged_above=-999 required=5 tests=[BAYES_00=-1.9] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BQ8_58VCNzlQ; Thu, 21 Nov 2013 05:07:57 -0800 (PST) Received: from ietfa.amsl.com (localhost [IPv6:::1]) by ietfa.amsl.com (Postfix) with ESMTP id 623491ADEBA; Thu, 21 Nov 2013 05:07:57 -0800 (PST) MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: quoted-printable From: internet-drafts@ietf.org To: i-d-announce@ietf.org X-Test-IDTracker: no X-IETF-IDTracker: 4.83.p1 Auto-Submitted: auto-generated Precedence: bulk Message-ID: <20131121130757.12090.36458.idtracker@ietfa.amsl.com> Date: Thu, 21 Nov 2013 05:07:57 -0800 Cc: tcpm@ietf.org Subject: [tcpm] I-D Action: draft-ietf-tcpm-tcp-rfc4614bis-01.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Nov 2013 13:07:59 -0000 A New Internet-Draft is available from the on-line Internet-Drafts director= ies. This draft is a work item of the TCP Maintenance and Minor Extensions Work= ing Group of the IETF. Title : A Roadmap for Transmission Control Protocol (TCP) Specif= ication Documents Author(s) : Martin Duke Robert Braden Wesley M. Eddy Ethan Blanton Alexander Zimmermann Filename : draft-ietf-tcpm-tcp-rfc4614bis-01.txt Pages : 50 Date : 2013-11-21 Abstract: This document contains a "roadmap" to the Requests for Comments (RFC) documents relating to the Internet's Transmission Control Protocol (TCP). This roadmap provides a brief summary of the documents defining TCP and various TCP extensions that have accumulated in the RFC series. This serves as a guide and quick reference for both TCP implementers and other parties who desire information contained in the TCP-related RFCs. The IETF datatracker status page for this draft is: https://datatracker.ietf.org/doc/draft-ietf-tcpm-tcp-rfc4614bis There's also a htmlized version available at: http://tools.ietf.org/html/draft-ietf-tcpm-tcp-rfc4614bis-01 A diff from the previous version is available at: http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-tcp-rfc4614bis-01 Please note that it may take a couple of minutes from the time of submission until the htmlized version and diff are available at tools.ietf.org. Internet-Drafts are also available by anonymous FTP at: ftp://ftp.ietf.org/internet-drafts/ From Alexander.Zimmermann@netapp.com Thu Nov 21 05:21:17 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 3FD9B1ADE86 for ; Thu, 21 Nov 2013 05:21:17 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -7.427 X-Spam-Level: X-Spam-Status: No, score=-7.427 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.525, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id i_NXNollF0qj for ; Thu, 21 Nov 2013 05:21:15 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id 954001ADAEA for ; Thu, 21 Nov 2013 05:21:15 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,744,1378882800"; d="asc'?scan'208";a="118285210" Received: from vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) by mx12-out.netapp.com with ESMTP; 21 Nov 2013 05:21:07 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.58]) by vmwexceht03-prd.hq.netapp.com ([10.106.76.241]) with mapi id 14.03.0123.003; Thu, 21 Nov 2013 05:21:05 -0800 From: "Zimmermann, Alexander" To: "tcpm@ietf.org Extensions" Thread-Topic: [tcpm] I-D Action: draft-ietf-tcpm-tcp-rfc4614bis-01.txt Thread-Index: AQHO5rrqMXOMU21OpUaUlEFSAo2M25owMbaA Date: Thu, 21 Nov 2013 13:21:05 +0000 Message-ID: References: <20131121130757.12090.36458.idtracker@ietfa.amsl.com> In-Reply-To: <20131121130757.12090.36458.idtracker@ietfa.amsl.com> Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.116] Content-Type: multipart/signed; boundary="Apple-Mail=_93790CC6-71BA-4ABF-95E1-CE2585D7343E"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Subject: Re: [tcpm] I-D Action: draft-ietf-tcpm-tcp-rfc4614bis-01.txt X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Nov 2013 13:21:17 -0000 --Apple-Mail=_93790CC6-71BA-4ABF-95E1-CE2585D7343E Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 This version addresses Yuchung=92s, Jana=92s and Martin=92s feedback. Open points are * Good location for RFC 5482 * Bob=92s feedback Alex Am 21.11.2013 um 14:07 schrieb Internet-Drafts@ietf.org: >=20 > A New Internet-Draft is available from the on-line Internet-Drafts = directories. > This draft is a work item of the TCP Maintenance and Minor Extensions = Working Group of the IETF. >=20 > Title : A Roadmap for Transmission Control Protocol = (TCP) Specification Documents > Author(s) : Martin Duke > Robert Braden > Wesley M. Eddy > Ethan Blanton > Alexander Zimmermann > Filename : draft-ietf-tcpm-tcp-rfc4614bis-01.txt > Pages : 50 > Date : 2013-11-21 >=20 > Abstract: > This document contains a "roadmap" to the Requests for Comments = (RFC) > documents relating to the Internet's Transmission Control Protocol > (TCP). This roadmap provides a brief summary of the documents > defining TCP and various TCP extensions that have accumulated in the > RFC series. This serves as a guide and quick reference for both TCP > implementers and other parties who desire information contained in > the TCP-related RFCs. >=20 >=20 > The IETF datatracker status page for this draft is: > https://datatracker.ietf.org/doc/draft-ietf-tcpm-tcp-rfc4614bis >=20 > There's also a htmlized version available at: > http://tools.ietf.org/html/draft-ietf-tcpm-tcp-rfc4614bis-01 >=20 > A diff from the previous version is available at: > http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-tcpm-tcp-rfc4614bis-01 >=20 >=20 > Please note that it may take a couple of minutes from the time of = submission > until the htmlized version and diff are available at tools.ietf.org. >=20 > Internet-Drafts are also available by anonymous FTP at: > ftp://ftp.ietf.org/internet-drafts/ >=20 > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm --Apple-Mail=_93790CC6-71BA-4ABF-95E1-CE2585D7343E Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKOCMEACgkQdyiq39b9uS4PNwCeIyctuXJJyUBLBtnQbVZo7qIc baEAoJ0rdWjH+opGWayGQoUzu5bPYORa =qQHD -----END PGP SIGNATURE----- --Apple-Mail=_93790CC6-71BA-4ABF-95E1-CE2585D7343E-- From Alexander.Zimmermann@netapp.com Thu Nov 21 05:24:57 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 343831ADF66 for ; Thu, 21 Nov 2013 05:24:57 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -0.263 X-Spam-Level: X-Spam-Status: No, score=-0.263 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, FRT_BELOW2=2.154, RCVD_IN_DNSWL_NONE=-0.0001, RP_MATCHES_RCVD=-0.525, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001, T_FRT_BELOW2=0.01] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OxFY14GdOu5P for ; Thu, 21 Nov 2013 05:24:55 -0800 (PST) Received: from mx11.netapp.com (mx11.netapp.com [216.240.18.76]) by ietfa.amsl.com (Postfix) with ESMTP id A67151ACB4E for ; Thu, 21 Nov 2013 05:24:55 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,744,1378882800"; d="asc'?scan'208";a="77468014" Received: from vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) by mx11-out.netapp.com with ESMTP; 21 Nov 2013 05:24:49 -0800 Received: from SACEXCMBX04-PRD.hq.netapp.com ([169.254.6.58]) by vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) with mapi id 14.03.0123.003; Thu, 21 Nov 2013 05:24:49 -0800 From: "Zimmermann, Alexander" To: Janardhan Iyengar Thread-Topic: [tcpm] Review of draft-ietf-tcpm-tcp-rfc4614bis-00 Thread-Index: AQHOyqa+XaYrLixeXU6FriXTRq34OJowat2A Date: Thu, 21 Nov 2013 13:24:47 +0000 Message-ID: References: In-Reply-To: Accept-Language: de-DE, en-US Content-Language: en-US X-MS-Has-Attach: yes X-MS-TNEF-Correlator: x-originating-ip: [10.104.60.116] Content-Type: multipart/signed; boundary="Apple-Mail=_34EA2695-02D3-4F6F-A7D8-31376E4B8D33"; protocol="application/pgp-signature"; micalg=pgp-sha1 MIME-Version: 1.0 Cc: "tcpm@ietf.org Extensions" , "Braden " , " Blanton" Subject: Re: [tcpm] Review of draft-ietf-tcpm-tcp-rfc4614bis-00 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 21 Nov 2013 13:24:57 -0000 --Apple-Mail=_34EA2695-02D3-4F6F-A7D8-31376E4B8D33 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=windows-1252 Hi Jana, thanks for the useful feedback. Comments bellow. Alex Am 16.10.2013 um 21:32 schrieb Janardhan Iyengar = : > Hi all, >=20 > Late, but as promised. Here's my pipe-dream: I would _LOVE_ to see a = complementary document that documents the current state of various = popular implementations, using the structure in this document as a = basis. A simple table with booleans for each of the RFCs listed here may = be a start. Such an RFC is even faster outdated as the roadmap. IMO such a list = shout be part of /net/*/tcp.c of the kernel source file. But I agree that such a list (wherever be placed) is = useful. >=20 > Thanks for all your work!! > - jana >=20 > Comments:=20 > - Page 6: "This is generally regarded as the least common = denominator among TCP flavors currently found running on Internet = hosts." I thought that the least-common denominator was NewReno. Is that = incorrect? This is C&P text from the original roadmap. Since I=92m not the author = of this section, I have problems to value your comment. If the WG has (now) feeling that NewReno and not Reno is the least = common denominator I can delete the sentence. NP. > - Page 7: What do you mean by "Fundamental Changes"? As in changes = to RFC 793? If so, can the Section be called that -- "Changes to RFC = 0793"? At least state it clearly in the intro paragraph to this section. = Without a clear explanation, it is difficult to see why 5482 is a = Fundamental Change. =46rom the introduction text of section 3 of the old roadmap: "3. Recommended Enhancements This section describes recommended TCP modifications that improve performance and security. RFCs 1323 and 3168 represent fundamental changes to the protocol. =85 " So, the authors of the RFC4614 saw this two RFCs as fundamental changes. The only thing that I did, was the creation of a new section with = exactly this wording. Nevertheless you (and also Yuchung) are right that RFC 5482 is *not* a = fundamental change. Therefore I move RFC 5482 out of this section, but leave the title as it = is. Personally, I do not like the title "Changes to RFC 0793=93 since there = are some other RFC that also make changes to 793 but do not belong to this section.=20 > - I'll echo Yuchung's earlier comment that 3.2, 3.3, 3.4, 3.5 should = be reorganized into two sections: Congestion Control, and Loss Recovery. Yes and no. Yuchung suggested to re-organize section 3.2 and 3.3, but = *not* 3.4 and 3.5. In version 01 we have now: 3.2. Congestion Control Extensions 3.3. Loss Recovery Extensions 3.4. Detection and Prevention of Spurious Retransmissions and the same structure in sec 4. (which was not suggested by Yuchung = ;-)) Strictly speaking, sec 3.4 is a subcategory of sec. 3.3. You are right. = But for a better structuring, I suggest that we put every spurious detection/reaction = algo into a separate sub-section. A reader can find the docs in question much easier. BTW, = we did the same thing with MPTCP. All MPTCP docs belongs in some way to CC and LR too, = but have their own sub-section. > You may want to note somewhere that there is some functional overlap = between these two sections -- dealing with spurious retransmissions will = fit in the Loss Recovery section, but intersects with congestion = control. > - Page 17: Consider an equivalent reorganization for section 4.2 and = 4.3. done for sec 4.2 > - Page 20: It seems to me that RFC 6182 also belongs in Section 4.4 = on MPTCP (I'm a bit ambivalent about it being in section 7.2). I'd = suggest leaving it in 7.2, and adding a reference to it in Section 4.4's = intro paragraph. The creation of the MPTCP section was with the collaboration of the = MPTCP folks. We commonly agree with the current structure. Personally, I would like to strictly follow the = current roadmap structure/philosophy to move RFCs that change the TCP protocol itself into sec 3/4 and move = other docs into sec 7. When applying this strategy, some RFCs that belong together by topic, may still belong to = multiple sections, though. To overcome this downside, we indicate in the introduction text of a section that other = RFCs to this topic are also listed in other sections: The documents listed in this section specify the Multipath TCP scheme, while the documents in Sections 7.2, 7.4, and 7.5 provide some additional background information. =20 > - Page 22: RFC 1110 -- reason for historic? RFC 6247 > - Page 30: IMO, RFC 6181 belongs in Section 4.4, not in Section 7.4.=20= see above (comment to page 20) > - Page 32: IMO, RFC 6056 belongs in Section 3.7, not in Section 7.5. RFCs in section 3.7 make changes to the protocol itself, whereas RFC = 6056 is purely an implementation advice. (This is IMO also the reason why RFC6056 is = BCP). BTW the same applies also for RFC 3360, which is (and was also in RFC = 4614) part of this section. > - Page 33: IMO, RFC 6897 belongs in Section 4.4, not in Section 7.5. see above (comment to page 20) > - Consider putting 7.5 and 7.7 next to each other -- they are = semantically related to each other, and MIB has no reason to be in = between them. yes > - WooHoo -- Self Reference!! I really don't think RFC 4614 belongs = in this list at all... the doc so far isn't listing previous versions of = documents. I don't see a reason to list a previous versions of *this* = document. ok > - Page 34: IMO, RFC 6077 belongs in Section 7.4, not in Section 7.7. Yes, why not. > - Page 34: IMO, RFC 5783 really really belongs in the Section on = Congestion Control, not in Section 7.7. I don=92t think so. All documents in sec 3 and 4 make changes to the = protocol itself. RFC 5783 is a survey similar to the TCP roadmap, which no modification at all. It=92s really = a support doc. > At least a reference to it should exist above in the CC section. Done >=20 > Editorial nits: > - Page 3: "large number of more experimental" --> "large number of = experimental=93 OK > - Page 4: "practices that is not" --> "practices that are not=93 OK > - Page 7: RFC 6298: "Abstract: "This document defines the standard = algorithm that ..." Why this sudden departure in style to quoting the = abstract? I see this departure appear later as well (Sections 3.5, 3.6, = 7.3). Because we C&P text here. Anytime we C&P an abstract from an RFC, we = quote it. The old roadmap uses the same style. > - Page 7: "Based on their investigation, .." The previous sentence = talks about RFC 6093, not about the authors of 6093 ("their"). I would = recommend removing that phrase altogether; it seems redundant.=20 OK > - Page 20: "additinal" --> =84additional" OK > - Page 24: "early precursor of the infamous RFC 793" --> "early = precursor of RFC 793". I certainly hope 793 isn't infamous! Fixed :-) > - General suggestion: use one of "predecessor of" or "precursor of" = consistently. OK > - Page 25: RFC 872 -- "Conclusion: " Style inconsistency: this = description is a departure in style from a summarizing paragraph. If C&P the abstract from a RFC we write: Abstract =84Foo Bar =85=93 = [RFC007] Here, we C&P the conclusion of the RFC. It exists a paragraph = =84Conclusion=93 in RFC 872. > - Page 35: "Hazardsin" --> "Hazards in=93 OK > - Page 36: FACK -- this loss recovery component is only a part of = the FACK document. The entire FACK algorithm includes a more accurate = estimation of outstanding bytes, and is, AFAIK, not implemented in = Linux. I would correct the reference to state that "FACK [MM96] includes = an alternate algorithm =85". OK > _______________________________________________ > tcpm mailing list > tcpm@ietf.org > https://www.ietf.org/mailman/listinfo/tcpm --Apple-Mail=_34EA2695-02D3-4F6F-A7D8-31376E4B8D33 Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="signature.asc" Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Message signed with OpenPGP using GPGMail -----BEGIN PGP SIGNATURE----- Comment: GPGTools - http://gpgtools.org iEYEARECAAYFAlKOCZcACgkQdyiq39b9uS4ehgCfUnfzWdt/xTnGyZpXyrcGF/QU oJgAoNyvM/fxWBx36zqmmQYmCsxK3lNn =+M1Z -----END PGP SIGNATURE----- --Apple-Mail=_34EA2695-02D3-4F6F-A7D8-31376E4B8D33-- From ycheng@google.com Sat Nov 23 16:42:25 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 899C11AE399 for ; Sat, 23 Nov 2013 16:42:25 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: 1.195 X-Spam-Level: * X-Spam-Status: No, score=1.195 tagged_above=-999 required=5 tests=[BAYES_20=-0.001, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, J_CHICKENPOX_32=0.6, J_CHICKENPOX_46=0.6, RP_MATCHES_RCVD=-0.525, SPF_PASS=-0.001] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3iZTEIs96J_N for ; Sat, 23 Nov 2013 16:42:22 -0800 (PST) Received: from mail-ie0-x236.google.com (mail-ie0-x236.google.com [IPv6:2607:f8b0:4001:c03::236]) by ietfa.amsl.com (Postfix) with ESMTP id B8ECA1AE395 for ; Sat, 23 Nov 2013 16:42:22 -0800 (PST) Received: by mail-ie0-f182.google.com with SMTP id as1so4677791iec.13 for ; Sat, 23 Nov 2013 16:42:15 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=j8nmrGfb73qAFlMfFNmquv4CNg33IeD7bpCcYvkkv6Y=; b=kKchkuW9ngVWwJ/roH+yoL59tdbqI0sIKh/JNyu5nZEfgHq4+ePSYeLY+xS/IW7PJt lhTjEsky/Ae6sVOEed7QvDyYf8EoLvGOHCJqgbMXqB2CRurbds97/5q+HLjMy7xlebWB SRgtPV7hMz6aekxKypOrfL1xjPwrbylUq/dbUALpeN5O8096kQ9agB2fWpvPYD+5pnlB LWyPV57m5sAyyj9bdUt/RLn3mUpHkOn5i5P1MSBFVij+CXzWmrANzwmOZ1M3DJ52GOye ggCv3dcff+jBmSjPpIeceB2FQmPR5Ydp+iYEd+mbB896FcOG/9cj+XYpdjMMrANJ3YZg q52A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=j8nmrGfb73qAFlMfFNmquv4CNg33IeD7bpCcYvkkv6Y=; b=I5lLLkk3/oDmAqTNGsNo6qvSuPKJS10yiZ/jxRM1771qSdYhiERpW2OD2I8NTP/huf ktEVylm9FgtoUHSZcvEjxDe0dYX+qDc0YplnflKXs5XSjYktb4VhIsNkc0xssSsZTcZq +ixi+yyYfX8hdUaNKif+jPxbYDd/leygCvtJEJqZyXMR1mx+GJ/vOfxdcBylKAkFF6qy Qae4LMcvFfnaOtWWwCmg5SzkCN+GfmIrpyioSWjiROH1bQv0eqxE0aC1PJW4q0Hh6S3r awPnrZ7BPWjn+O2vjGiEZO2SohPBX/i96MNm6CJjnJn2oD0ANbgbTRrO0zpzTkb2Akg0 EZSA== X-Gm-Message-State: ALoCoQkuf+L/l2ECva+Zmo9SWu/AhHUuProiN4DvNzv7V2hrEaczLDmEAJ3693WCHy1mqIZXibwoy1gVb7FkWwBiIzAzJmXfl21lR07lfiCnWMg7QaJz+09SAKWplay7WftKf+Zbu8YrW4jHnu1bIzOs3/MLVWv46Wb6hrrEEsb5Udch/Mudim9sb1KLmA1z/DAtr2geQ5h/ X-Received: by 10.51.16.3 with SMTP id fs3mr7646667igd.53.1385253734889; Sat, 23 Nov 2013 16:42:14 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Sat, 23 Nov 2013 16:41:34 -0800 (PST) In-Reply-To: <528B9E16.70602@erg.abdn.ac.uk> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> <528B9E16.70602@erg.abdn.ac.uk> From: Yuchung Cheng Date: Sat, 23 Nov 2013 16:41:34 -0800 Message-ID: To: Gorry Fairhurst Content-Type: text/plain; charset=ISO-8859-1 Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 24 Nov 2013 00:42:25 -0000 On Tue, Nov 19, 2013 at 9:21 AM, Gorry Fairhurst wrote: > > See response in-line... > > Many of these are small items, > > Somewhere down there is the significant point that I do not yet understand, > i.e. what are the potential CC issues that need to be evaluated - especially > how TFO respond to severe path overload? (see below) > > Gorry > > > On 09/11/2013 18:06, Yuchung Cheng wrote: >> >> Hi Gorry, >> >> Sorry for the last response. Too many good sessions and dinner parties >> during this IETF for me to respond promptly. >> >> On Sat, Nov 2, 2013 at 4:09 PM, Gorry wrote: >>> >>> >>> I have reviewed the current version of TFO and have the comments below. >>> This is rather long ... >>> >>> Nits in Section 1 - I think the document here talks of servers and >>> clients from a TCP endpoint perspective, this is normal for TCP, but the >>> document also raises implications for apps designers who may have a very >>> different view on what is intended by the word "server" and "client". To >>> avoid doubt maybe you should explain this in the intro? >> >> They are already explained in the "Terminology" section. >> > OK, but it's terse! > >>> >>> Page 3, para 2 is not clear. I think it is just the detail of the >>> wording, but it would be good to see clear text. >>> > In RFC 6936 we wrote (albeit slightly different problem, with a different > protocol): > "IPv6 nodes MUST provide a way for the application/protocol to > indicate the set of ports that will be enabled to send datagrams > with a zero UDP checksum. This may be implemented by enabling a > transport mode using a socket API call when the socket is > established, or by a similar mechanism. It may also be > implemented by enabling the method for a pre-assigned static > port" > > >>> Last para section 2: to me, the requirements are not clear. >>> >>> I think this piece of text means: TCP stacks should NOT enable TCP-TFO by >>> default. The stack should provide a global flag to enable this. An >>> application wishing to use this must enable each TCP socket on a per >>> service/per port basis. >>> >>> I think the other comments/questions maybe more significant. >>> >>> section 2 also suggests the main change is to allow data in SYN, I have >>> noted that some people think this is the only change --- it is not. The >>> bigger implication is that the document proposes to allow new data sent >>> during the 3WHS. This change needs to be more clearly noted, sending an IW >>> of data before there is an ACK on the return path is a significant change to >>> TCP semantics and it's behaviour under heavy load. >> >> >> will revise sec2 and discuss w/ you before we publish the revision. >> > Let's do that. > >> >>> >>> Page 5 fast open, bullet 3. : This says the server can send data before >>> the 3WHS completes. I am curious what this actually means, is this implied >>> that the use is extricated to standards-track mechanisms, I.e. I am looking >>> here for some explicit understanding of whether this implies IW3. Or is the >>> experiment linked to both a proposal for iW10 and this new experiment? >> >> will add "initial congestion window of " before the "... the >> congestion control in .." to make it more clear. >> or >> >>> >>> Last para 4.1.3 the text says /should include remote port numbers too/ >>> Can we be clear which port numbers are used. Is this the received server >>> port number by the client? Actually, do not understand what is intended by >>> the paragraph, can you provide an example to be sure what is meant? >> >> a "remote" port number of the client means the "server port". but will >> replace remote/server to avoid confusion. >> > That works. > > >> The paragraph is added because of your suggestion to add negative caching. >> > OK > > >>> >>> Section 4.3.2. I hoped this would say something explicit about not >>> sending SYN segments that are larger than the default MTU. This seems like a >>> bad thing to do. >> >> It's a bad thing to send any segment larger than the MTU. Why is that >> specific to Fast Open? >> > ... Let me check: > 1. TFO could allow PMTU to be cached across sessions. > 2. Multi-homing/ Multipath routing could result in a different path for a > new connection. > 3. If TFO used a cahched PMTU with the SYN (or before the SYN-ACK), then > these could be black-holded. > > - While local MTU is known, remote (or path) MTU can only be validated once > there is a path... To me, allowing TFO to use larger MTU seems like an > unnecessary extra complication to have to deal with. I am still not getting your concern. Path MTU is cached on a per "path" basis. If you use a different/new path you start with some default MTU. btw, there is no Section 4.3.2 in the latest draft. We only send SYN-data when we've done a previous handshake w/ the server, and have both the cookie and path information. The path (mtu) can change anyway, just like a normal data session, and the risk holds. If SYN-data is lost, we will retransmit a regular SYN (and cache the new MTU). > > >>> >>> Why are the examples only for web? This is a TCP spec. >> >> Sure there are a lot more applications can benefit from that, but Web >> is what motivates our work and the application is complex enough to >> demonstrate many aspects of Fast Open. If you have another good >> examples, you are welcome to suggest one. >> > .. I could OR it mya be wise to just say something similar to what you have > written above at the end of section 1? > > >>> >>> There are two sections 6.3.2. >>> >>> Second Section 6.3.2 - I think this applies to SSL/TLS in general, >>> doesn't it? ... I think the section should address the general case, and >>> then provide an example for https, rather than assuming a web-centric use. >> >> >> Yes it is generic, but anyone who knows SSL/TLS good enough knows that. >> > So could you name the section TLS over TCP ... and then give HTTP over TLS > as an example? Sure and that's a good idea. > > >>> >>> Section 7.1 first para last line.please check text, this pathology is >>> not necessarily malicious!!! >> >> sure we can s/malicious/pathological. does this deserves three !!!? >> > Fine (indeed, we want those nice people to fix all their middleboxes). > > >>> The issue seems to be mainly related to NAPT where ports are used to >>> discriminate senders? >>> Also the shared NAPT can result in a range of RTT and PMTU to the same >>> IP, but this is already possible with ECMP routing. >>> > So In this case there is need to understand the implications of using a > cached RTT and cached PMTU value. What concerns me is that we need > experimental data to show that this is safe - not just to realise that this > may have no benefit for TFO. > > >>> The lack of issues listed in section 7 raises serious concerns to me. I >>> do not list that is provided currently as the only, or even most significant >>> reasons why this is deemed to be experimental work. Indeed, the lack of >>> identification of potential CC issues is why this document is in my opinion >>> NOT ready to complete a WGLC - if the issues are not listed how can we >>> evaluate if the method is safe enough to experiment or can later be >>> evaluated properly. >>> >>> Here are some additional issues I would like considered by the group and >>> if the group is happy these may be safe enough to recommend testing in the >>> general internet, then I think we should document that these are things to >>> be confirmed. >>> >>> 1) I raised on the list that a negative impact of using the SYN as a >>> probe to check for new option support. The examples I gave were ECN >>> interactions and IPv6 probing. ECN if used will probe for ECN support in the >>> SYN segment. If we include data on the SYN and then there is a lack of >>> response, it is assumed by the sender that both TFO and ECN are not >>> supported. This is an implication of the model we use to negotiate for >>> features that may be black-holed. It is not an issue with the method, but I >>> think we need tone clear this is something that may happen. A similar case >>> exists if a client attempts TFO to a server using IPv6 ... Is it assumed >>> that the path does not support IPv6 or TFO (or actually both). It does not >>> happen when multiple options are negotiated. >> >> if a SYN-data is dropped, TFO will retransmit a regular SYN (with ECN if >> used). >> > May be good, I agree - but I do not see this in the text (RFC 3168 would > allow this, it kind of suggests clearing ECT on retx). > >>> >>> 2) the draft proposes caching RTT per path and using this for SYN >>> timeout. This updates a standards track RFC, and will have implications in >>> some network paths that may seriously degrade performance, in other cases it >>> may improve performance. Examples of negative performance are cases where an >>> initial packet can take much longer to process than a subsequent packet. >>> This can and does occur in bandwidth-on-demand L2 networks, where the first >>> packet causes path setup. It also occurs to some extent with policing and >>> routing devices that build cached state from the first packet in a flow. I >>> think this is one of the reasons why the RFC-series has to date not >>> described a shortened RTT method for the SYN. >> >> >>> >>> Another issue is that previous RTT (or PMTU) is not necessarily a good >>> indication of future value in a different 5tuple, since ECMP etc can lead to >>> very different path characteristics for a session with a different port >>> number. The current text therefore proposes an experiment that is beyond >>> what was previously considered safe. It does not specify a way to avoid the >>> server making the same mistake next connection. >> >> >> We'll just take RTT caching out of the text if that makes people happy. >> > Maybe, although I'm not pushing for that, but it may be OK - if it is an > orthogonal issue that TFO does not rely upon. If you want to mention caching > these, I think the doc needs to explain more. > > >>> >>> 3) seeding the SYN RTT from a lower value makes the sender more >>> aggressive in heavily congested networks. The sender is also made more >>> aggressive by sending IW data packets before there is any indication the >>> path can contain even a single data segment. This is significant change to >>> standard behaviour. If the proposal is to use IW 3 (as standard) it still >>> needs to be called out. If the proposal is to use a larger experimental >>> number then I have concerns here that this is a significant change that >>> needs an automated recovery method to prevent significant collateral damage >>> on capacity-limited paths - there needs to be a way to stop a server doing >>> this each time resulting in recurring loss! >> >> >> when the IW of packets are not all acked, loss recovery is triggered >> and window is reduced. this is part of standard CC. >> >> also when you use "significant" four times, please back up with a good >> theory or data or both. and abuse that word does not it make more >> significant. >> > So let's see if we can see agree on what happens: > > - If the path is very lossey - severe congestion: > TCP standard sends one SYN segment with some probability of loss, and if it > sees loss backs off and retransmits the SYN. As (over)load increases, new > sessions add 40B and the new sessions will often defer start-up, controlling > their rate. > > - If TFO path is very lossey - severe congestion: IW full-sized segments are > sent. Each new session does this, adding to the load. > Each new session adds MSS*IW = 6000B (IW=3) or 15000B (IW=10). > > I concede this applies to severe congestion. Under this case, it seems 150x > more traffic to me and this before before CC is engaged. Or is this wrong? I can make similar case for any recent proposal that will blow up the network, e.g., rto-restart, newcwv. The question is how practical and common those cases are. if the link is thin, not doing Fast Open with same IW will experience similar heavy losses. Let's face it: if the capacity is not enough to handle the demand, there is little can be done. The concern has been raised by Michael Scharf's, and the step 6 in Section 4.2.2 does mention that. I did ask the list for any better solution for that but I've never heard back any suggestion. > > >>> >>> 4) Are there any changes to PMTUD behaviour that need experience? PMTU >>> was naturally cached and I am not sure if this changes the behaviour or not. >>> It does of course allow a packet with a larger MTU to be sent before the >>> path has been initially validated via the 3WHS (I.e. After a path change >>> while idle, this can generate TCP segments in a SYN that are larger than the >>> receiver advertised MSS >>> - I suspect this is not crucial, but I would also assume that the SYN itself >>> should not use an increased PMTU value to carry the data. Is this correct? >> >> in our implementation, we use what ever MTU (possibly discovered in a >> prior PMTU) cached to send SYN-data. >> > One option could be that the document could identify this as a topic to be > explored (like RTT-caching, this does not seem to be a core technique > required for FastOpen, but if you mention cachingthese, I think the WG > should highlight that this > can also say there are potential concerns and a need for experimentation? Not sure RTT or PMTU caching should be part of TFO experimentation b/c the potential concerns seem orthogonal to the core protocol of Fast Open. > > >>> >>> Are there other case where this change will impact the network for >>> specific path characteristics? >>> >>> Have others thought through these issues and feel they are ok? Or at >>> least ok for experimental deployment? >>> >>> Gorry >>> > P.S. A tiny NiT on 7.2: > > "Careful experimentation is necessary to evaluate if cookie-less TFO > is practical." > - I think the experiment that is needed by the IETF is "to evaluate if > cookie-less TFO could be safe for deployment in the general Internet." Will make that change. > From michael.scharf@alcatel-lucent.com Tue Nov 26 06:06:01 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 84BE71AE21A for ; Tue, 26 Nov 2013 06:06:01 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.3 X-Spam-Level: X-Spam-Status: No, score=-6.3 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_32=0.6, RCVD_IN_DNSWL_HI=-5] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id NkOplJvJC5pw for ; Tue, 26 Nov 2013 06:05:59 -0800 (PST) Received: from ihemail4.lucent.com (ihemail4.lucent.com [135.245.0.39]) by ietfa.amsl.com (Postfix) with ESMTP id BEA5A1AE219 for ; Tue, 26 Nov 2013 06:05:59 -0800 (PST) Received: from fr711usmtp1.zeu.alcatel-lucent.com (h135-239-2-122.lucent.com [135.239.2.122]) by ihemail4.lucent.com (8.13.8/IER-o) with ESMTP id rAQE5px7000048 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=FAIL); Tue, 26 Nov 2013 08:05:54 -0600 (CST) Received: from FR712WXCHHUB03.zeu.alcatel-lucent.com (fr712wxchhub03.zeu.alcatel-lucent.com [135.239.2.74]) by fr711usmtp1.zeu.alcatel-lucent.com (GMO) with ESMTP id rAQE5psp024282 (version=TLSv1/SSLv3 cipher=AES128-SHA bits=128 verify=FAIL); Tue, 26 Nov 2013 15:05:51 +0100 Received: from FR712WXCHMBA15.zeu.alcatel-lucent.com ([169.254.7.70]) by FR712WXCHHUB03.zeu.alcatel-lucent.com ([135.239.2.74]) with mapi id 14.02.0247.003; Tue, 26 Nov 2013 15:05:51 +0100 From: "Scharf, Michael (Michael)" To: Yuchung Cheng , Gorry Fairhurst Thread-Topic: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 Thread-Index: AQHO2CCuBMxZVGNYOkqZ0Ql1yfjDJpodK8MAgA+qxQCABsRMAIAECF0Q Date: Tue, 26 Nov 2013 14:05:50 +0000 Message-ID: <655C07320163294895BBADA28372AF5D13C310@FR712WXCHMBA15.zeu.alcatel-lucent.com> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> <528B9E16.70602@erg.abdn.ac.uk> In-Reply-To: Accept-Language: de-DE, en-US Content-Language: de-DE X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [135.239.27.39] Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 X-Scanned-By: MIMEDefang 2.57 on 135.245.2.39 Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Nov 2013 14:06:01 -0000 > >>> 3) seeding the SYN RTT from a lower value makes the sender more > >>> aggressive in heavily congested networks. The sender is also made > more > >>> aggressive by sending IW data packets before there is any > indication the > >>> path can contain even a single data segment. This is significant > change to > >>> standard behaviour. If the proposal is to use IW 3 (as standard) it > still > >>> needs to be called out. If the proposal is to use a larger > experimental > >>> number then I have concerns here that this is a significant change > that > >>> needs an automated recovery method to prevent significant > collateral damage > >>> on capacity-limited paths - there needs to be a way to stop a > server doing > >>> this each time resulting in recurring loss! > >> > >> > >> when the IW of packets are not all acked, loss recovery is triggered > >> and window is reduced. this is part of standard CC. > >> > >> also when you use "significant" four times, please back up with a > good > >> theory or data or both. and abuse that word does not it make more > >> significant. > >> > > So let's see if we can see agree on what happens: > > > > - If the path is very lossey - severe congestion: > > TCP standard sends one SYN segment with some probability of loss, and > if it > > sees loss backs off and retransmits the SYN. As (over)load increases, > new > > sessions add 40B and the new sessions will often defer start-up, > controlling > > their rate. > > > > - If TFO path is very lossey - severe congestion: IW full-sized > segments are > > sent. Each new session does this, adding to the load. > > Each new session adds MSS*IW =3D 6000B (IW=3D3) or 15000B (IW=3D10). > > > > I concede this applies to severe congestion. Under this case, it > seems 150x > > more traffic to me and this before before CC is engaged. Or is this > wrong? > I can make similar case for any recent proposal that will blow up the > network, e.g., rto-restart, newcwv. The question is how practical and > common those cases are. if the link is thin, not doing Fast Open with > same IW will experience similar heavy losses. Let's face it: if the > capacity is not enough to handle the demand, there is little can be > done. >=20 > The concern has been raised by Michael Scharf's, and the step 6 in > Section 4.2.2 does mention that. I did ask the list for any better > solution for that but I've never heard back any suggestion. Indeed, I was (and I still am) worried by the combination of TFO and IW10. = Because of that, we have already had similar list discussions about highly = congested links. I am less concerned about TFO with IW3, because bursts of = that size are sent by TCP anyway pretty frequently. However, given the explicit reference to RFC 3390 in the current draft, I h= ave not further objected during the WGLC. =20 Since this question still seems to trigger discussion, what about some addi= tional explanation of "slightly more aggressive"? For instance: OLD: Note that if SYN-ACK is lost, regular TCP reduces the initial congestion window before sending any data. In this case TFO is slightly more aggressive in the first data round trip even though it does not change the congestion control. NEW: Note that in some cases TFO can be slightly more aggressive than regu= lar TCP. If SYN-ACK is lost, regular TCP reduces the initial congestion window= before sending any data. On such a congested link, TFO could first send a burst according to t= he initial window [RFC3390], and the server would then back-off after one round trip time if one or mor= e packets are lost. In this case TFO is slightly more aggressive in the first data round trip even though it = does not change the congestion control. My 2 cents, with chair hat off... Michael From ycheng@google.com Tue Nov 26 11:39:11 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 04FD11ADF34 for ; Tue, 26 Nov 2013 11:39:11 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -0.779 X-Spam-Level: X-Spam-Status: No, score=-0.779 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, J_CHICKENPOX_32=0.6, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id iRRlql12YAw8 for ; Tue, 26 Nov 2013 11:39:08 -0800 (PST) Received: from mail-ie0-x233.google.com (mail-ie0-x233.google.com [IPv6:2607:f8b0:4001:c03::233]) by ietfa.amsl.com (Postfix) with ESMTP id A83861ADF8D for ; Tue, 26 Nov 2013 11:39:08 -0800 (PST) Received: by mail-ie0-f179.google.com with SMTP id x13so10039632ief.38 for ; Tue, 26 Nov 2013 11:39:08 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=0Df0H8CBDHdvHzlUriLCDges3kkdI8VDo1sujdljoBk=; b=hZF0YKLsV5imqyU46tMVzKNVYYCQZ+BuccA6SdvrcUn4ktZaguF5tqcU/XaAk3xjar PtCQ659kTpzW9r+bKYh+Y86qm469PmRaYor6Amd4VxyFlGCjEK2t5FthpjuJJjvnVGdI gt2JtyvVkrHt8vt6QrUYeeTj8wZOn84AKBaHOXRBst+c4LViuoBGTYiBJuEoxu1HZpEH Im6eSzHbo06JUf4IDlycywZKMA21tweLKP7U3ylkGmv0Mra2/9etJ2TxgVoEDHOpioIi /CfputlEBbgNq8HxdOrgeSrV6HZ8THIfGrgsBcSs0yLxAoypqiUEtbXSD8ke3HQiXt6I u/QQ== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=0Df0H8CBDHdvHzlUriLCDges3kkdI8VDo1sujdljoBk=; b=hoy1/YfCxl6m2uTixVouQu5jXaZe874ojr5UoVn+3c5mycMPAyJdl7jzVCtgCo4V8+ 2C/f5CmPUJ9u6SuZ2ro7xe9TXXG/k9XxmE2UV6cyD7MKku8dg38bbSMoPi8to5oeuv7V in6HM1xyuKuPeabKiT3utzArGcOisZdyLP7XEl/W5b9AXKnwxHHEYp545XQwZ+miWHdy OHqID29gldaV5LdEclvSd/zMjIQuLLb8vcucLUdfPVsPPtBJDnmA54KX9qB7rmlLGb7+ FmDwiVamYNOx8ZFTig6DSJPoUx/4LwSznKI6cezzEMBR210EwucN7sg5tcgMD05jPg3e vJQw== X-Gm-Message-State: ALoCoQlL61FQBLkIKi1m9z1Ne3/DD568AdpZbhsGPdxwDGkq37LTjteK49cWMTPI1dFw+6hkH8SWouSiMd4lR0BBajzDFM6VKAGSvoOTLSbWUlJgFlopuyoWcnI2liCsrm8zLem5X0nIgrwSMUy3VxjqX12yrFVJPaq2ZMsykaXoU3YUBzO5gRBnOFFCer1ziVl+5TLL8jDC X-Received: by 10.43.145.197 with SMTP id jv5mr21440528icc.2.1385494748274; Tue, 26 Nov 2013 11:39:08 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Tue, 26 Nov 2013 11:38:28 -0800 (PST) In-Reply-To: <655C07320163294895BBADA28372AF5D13C310@FR712WXCHMBA15.zeu.alcatel-lucent.com> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> <528B9E16.70602@erg.abdn.ac.uk> <655C07320163294895BBADA28372AF5D13C310@FR712WXCHMBA15.zeu.alcatel-lucent.com> From: Yuchung Cheng Date: Tue, 26 Nov 2013 11:38:28 -0800 Message-ID: To: "Scharf, Michael (Michael)" Content-Type: multipart/alternative; boundary=001a11c1faa04f4f1f04ec19a204 Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Nov 2013 19:39:11 -0000 --001a11c1faa04f4f1f04ec19a204 Content-Type: text/plain; charset=ISO-8859-1 On Tue, Nov 26, 2013 at 6:05 AM, Scharf, Michael (Michael) < michael.scharf@alcatel-lucent.com> wrote: > > >>> 3) seeding the SYN RTT from a lower value makes the sender more > > >>> aggressive in heavily congested networks. The sender is also made > > more > > >>> aggressive by sending IW data packets before there is any > > indication the > > >>> path can contain even a single data segment. This is significant > > change to > > >>> standard behaviour. If the proposal is to use IW 3 (as standard) it > > still > > >>> needs to be called out. If the proposal is to use a larger > > experimental > > >>> number then I have concerns here that this is a significant change > > that > > >>> needs an automated recovery method to prevent significant > > collateral damage > > >>> on capacity-limited paths - there needs to be a way to stop a > > server doing > > >>> this each time resulting in recurring loss! > > >> > > >> > > >> when the IW of packets are not all acked, loss recovery is triggered > > >> and window is reduced. this is part of standard CC. > > >> > > >> also when you use "significant" four times, please back up with a > > good > > >> theory or data or both. and abuse that word does not it make more > > >> significant. > > >> > > > So let's see if we can see agree on what happens: > > > > > > - If the path is very lossey - severe congestion: > > > TCP standard sends one SYN segment with some probability of loss, and > > if it > > > sees loss backs off and retransmits the SYN. As (over)load increases, > > new > > > sessions add 40B and the new sessions will often defer start-up, > > controlling > > > their rate. > > > > > > - If TFO path is very lossey - severe congestion: IW full-sized > > segments are > > > sent. Each new session does this, adding to the load. > > > Each new session adds MSS*IW = 6000B (IW=3) or 15000B (IW=10). > > > > > > I concede this applies to severe congestion. Under this case, it > > seems 150x > > > more traffic to me and this before before CC is engaged. Or is this > > wrong? > > I can make similar case for any recent proposal that will blow up the > > network, e.g., rto-restart, newcwv. The question is how practical and > > common those cases are. if the link is thin, not doing Fast Open with > > same IW will experience similar heavy losses. Let's face it: if the > > capacity is not enough to handle the demand, there is little can be > > done. > > > > The concern has been raised by Michael Scharf's, and the step 6 in > > Section 4.2.2 does mention that. I did ask the list for any better > > solution for that but I've never heard back any suggestion. > > Indeed, I was (and I still am) worried by the combination of TFO and IW10. > Because of that, we have already had similar list discussions about highly > congested links. I am less concerned about TFO with IW3, because bursts of > that size are sent by TCP anyway pretty frequently. > > However, given the explicit reference to RFC 3390 in the current draft, I > have not further objected during the WGLC. > > Since this question still seems to trigger discussion, what about some > additional explanation of "slightly more aggressive"? For instance: > > OLD: > > Note that if SYN-ACK is lost, regular TCP reduces the initial > congestion window before sending any data. In this case TFO is > slightly more aggressive in the first data round trip even though > it does not change the congestion control. > > NEW: > > Note that in some cases TFO can be slightly more aggressive than > regular TCP. > If SYN-ACK is lost, regular TCP reduces the initial congestion > window before sending any data. > On such a congested link, TFO could first send a burst according to > the initial window [RFC3390], > and the server would then back-off after one round trip time if > one or more packets are lost. In this case TFO is > slightly more aggressive in the first data round trip even though it > does not change the congestion control. > sure > > My 2 cents, with chair hat off... > > Michael > --001a11c1faa04f4f1f04ec19a204 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable



On Tue, Nov 26, 2013 at 6:05 AM, Scharf, Michael (Michael) <michael.scharf@alcatel-lucent.com> wrote:
>= >>> 3) seeding the SYN RTT from a lower value makes the sender mo= re
> >>> aggressive in heavily congested networks. The sender is a= lso made
> more
> >>> aggressive by sending IW data packets before there is any=
> indication the
> >>> path can contain even a single data segment. This is sign= ificant
> change to
> >>> standard behaviour. If the proposal is to use IW 3 (as st= andard) it
> still
> >>> needs to be called out. If the proposal is to use a large= r
> experimental
> >>> number then I have concerns here that this is a significa= nt change
> that
> >>> needs an automated recovery method to prevent significant=
> collateral damage
> >>> on capacity-limited paths - there needs to be a way to st= op a
> server doing
> >>> this each time resulting in recurring loss!
> >>
> >>
> >> when the IW of packets are not all acked, loss recovery is tr= iggered
> >> and window is reduced. this is part of standard CC.
> >>
> >> also when you use "significant" four times, please = back up with a
> good
> >> theory or data or both. and abuse that word does not it make = more
> >> significant.
> >>
> > So let's see if we can see agree on what happens:
> >
> > - If the path is very lossey - severe congestion:
> > TCP standard sends one SYN segment with some probability of loss,= and
> if it
> > sees loss backs off and retransmits the SYN. As (over)load increa= ses,
> new
> > sessions add 40B and the new sessions will often defer start-up,<= br> > controlling
> > their rate.
> >
> > - If TFO path is very lossey - severe congestion: IW full-sized > segments are
> > sent. Each new session does this, adding to the load.
> > Each new session adds =A0MSS*IW =3D 6000B (IW=3D3) or 15000B (IW= =3D10).
> >
> > I concede this applies to severe congestion. Under this case, it<= br> > seems 150x
> > more traffic to me and this before before CC is engaged. Or is th= is
> wrong?
> I can make similar case for any recent proposal that will blow up the<= br> > network, e.g., rto-restart, newcwv. The question is how practical and<= br> > common those cases are. if the link is thin, not doing Fast Open with<= br> > same IW will experience similar heavy losses. Let's face it: if th= e
> capacity is not enough to handle the demand, there is little can be > done.
>
> The concern has been raised by Michael Scharf's, and the step 6 in=
> Section 4.2.2 does mention that. I did ask the list for any better
> solution for that but I've never heard back any suggestion.

Indeed, I was (and I still am) worried by the combination of TF= O and IW10. Because of that, we have already had similar list discussions a= bout highly congested links. I am less concerned about TFO with IW3, becaus= e bursts of that size are sent by TCP anyway pretty frequently.

However, given the explicit reference to RFC 3390 in the current draft, I h= ave not further objected during the WGLC.

Since this question still seems to trigger discussion, what about some addi= tional explanation of "slightly more aggressive"? For instance:
OLD:

=A0 =A0 =A0 Note that if SYN-ACK is lost, regular TCP reduces the initial =A0 =A0 =A0 congestion window before sending any data. In this case TFO is<= br> =A0 =A0 =A0 slightly more aggressive in the first data round trip even thou= gh
=A0 =A0 =A0 it does not change the congestion control.

NEW:

=A0 =A0 =A0 Note that in some cases TFO can be slightly more aggressive tha= n regular TCP.
=A0 =A0 =A0 If SYN-ACK is lost, regular TCP reduces the initial congestion = window before sending any data.
=A0 =A0 =A0 On such a congested link, TFO could first send a burst accordin= g to the initial window [RFC3390],
=A0 =A0 =A0 =A0 and the server would then back-off after one round trip tim= e if one or more packets are lost. In this case TFO is
=A0 =A0 =A0 slightly more aggressive in the first data round trip even thou= gh it does not change the congestion control.
sure =A0=

My 2 cents, with chair hat off...

Michael

--001a11c1faa04f4f1f04ec19a204-- From gorry@erg.abdn.ac.uk Tue Nov 26 11:51:56 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id A35FA1ADFBF for ; Tue, 26 Nov 2013 11:51:56 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.602 X-Spam-Level: X-Spam-Status: No, score=-3.602 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_32=0.6, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id SVlY9k53VJ6L for ; Tue, 26 Nov 2013 11:51:54 -0800 (PST) Received: from spey.erg.abdn.ac.uk (spey.erg.abdn.ac.uk [139.133.204.173]) by ietfa.amsl.com (Postfix) with ESMTP id D068B1ADF93 for ; Tue, 26 Nov 2013 11:51:53 -0800 (PST) Received: from www.erg.abdn.ac.uk (blake.erg.abdn.ac.uk [139.133.210.30]) by spey.erg.abdn.ac.uk (Postfix) with ESMTPSA id D0E3F2B4276; Tue, 26 Nov 2013 19:51:52 +0000 (GMT) Received: from 212.159.18.54 (SquirrelMail authenticated user gorry) by www.erg.abdn.ac.uk with HTTP; Tue, 26 Nov 2013 19:51:52 -0000 Message-ID: <763b3f8541277a7690bf859a87e69265.squirrel@www.erg.abdn.ac.uk> In-Reply-To: References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> <528B9E16.70602@erg.abdn.ac.uk> <655C07320163294895BBADA28372AF5D13C310@FR712WXCHMBA15.zeu.alcatel-lucent.com> Date: Tue, 26 Nov 2013 19:51:52 -0000 From: gorry@erg.abdn.ac.uk To: "Yuchung Cheng" User-Agent: SquirrelMail/1.4.22 MIME-Version: 1.0 Content-Type: text/plain;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 (Normal) Importance: Normal Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Nov 2013 19:51:56 -0000 See below... > On Tue, Nov 26, 2013 at 6:05 AM, Scharf, Michael (Michael) < > michael.scharf@alcatel-lucent.com> wrote: > >> > >>> 3) seeding the SYN RTT from a lower value makes the sender more >> > >>> aggressive in heavily congested networks. The sender is also made >> > more >> > >>> aggressive by sending IW data packets before there is any >> > indication the >> > >>> path can contain even a single data segment. This is significant >> > change to >> > >>> standard behaviour. If the proposal is to use IW 3 (as standard) >> it >> > still >> > >>> needs to be called out. If the proposal is to use a larger >> > experimental >> > >>> number then I have concerns here that this is a significant change >> > that >> > >>> needs an automated recovery method to prevent significant >> > collateral damage >> > >>> on capacity-limited paths - there needs to be a way to stop a >> > server doing >> > >>> this each time resulting in recurring loss! >> > >> >> > >> >> > >> when the IW of packets are not all acked, loss recovery is >> triggered >> > >> and window is reduced. this is part of standard CC. >> > >> >> > >> also when you use "significant" four times, please back up with a >> > good >> > >> theory or data or both. and abuse that word does not it make more >> > >> significant. >> > >> >> > > So let's see if we can see agree on what happens: >> > > >> > > - If the path is very lossey - severe congestion: >> > > TCP standard sends one SYN segment with some probability of loss, >> and >> > if it >> > > sees loss backs off and retransmits the SYN. As (over)load >> increases, >> > new >> > > sessions add 40B and the new sessions will often defer start-up, >> > controlling >> > > their rate. >> > > >> > > - If TFO path is very lossey - severe congestion: IW full-sized >> > segments are >> > > sent. Each new session does this, adding to the load. >> > > Each new session adds MSS*IW = 6000B (IW=3) or 15000B (IW=10). >> > > >> > > I concede this applies to severe congestion. Under this case, it >> > seems 150x >> > > more traffic to me and this before before CC is engaged. Or is this >> > wrong? >> > I can make similar case for any recent proposal that will blow up the >> > network, e.g., rto-restart, newcwv. The question is how practical and >> > common those cases are. if the link is thin, not doing Fast Open with >> > same IW will experience similar heavy losses. Let's face it: if the >> > capacity is not enough to handle the demand, there is little can be >> > done. >> > >> > The concern has been raised by Michael Scharf's, and the step 6 in >> > Section 4.2.2 does mention that. I did ask the list for any better >> > solution for that but I've never heard back any suggestion. >> >> Indeed, I was (and I still am) worried by the combination of TFO and >> IW10. >> Because of that, we have already had similar list discussions about >> highly >> congested links. I am less concerned about TFO with IW3, because bursts >> of >> that size are sent by TCP anyway pretty frequently. >> >> However, given the explicit reference to RFC 3390 in the current draft, >> I >> have not further objected during the WGLC. >> >> Since this question still seems to trigger discussion, what about some >> additional explanation of "slightly more aggressive"? For instance: >> >> OLD: >> >> Note that if SYN-ACK is lost, regular TCP reduces the initial >> congestion window before sending any data. In this case TFO is >> slightly more aggressive in the first data round trip even though >> it does not change the congestion control. >> >> NEW: >> >> Note that in some cases TFO can be slightly more aggressive than >> regular TCP. >> If SYN-ACK is lost, regular TCP reduces the initial congestion >> window before sending any data. >> On such a congested link, TFO could first send a burst according >> to >> the initial window [RFC3390], >> and the server would then back-off after one round trip time if >> one or more packets are lost. In this case TFO is >> slightly more aggressive in the first data round trip even though >> it >> does not change the congestion control. >> > sure > The description is basically OK - but I don't agree on two points: My point is that this DOES change the congestion control in the most critical starvation prevention mode (i.e. under heavy load): It allows a single TCP session to send up to IW bytes into the network without confirming the path is available. This is not a trivial change. It makes significant assumptions about the buffering on the path, and the level of current congestion - for high rate links this may be OK (likely no impact) for low rate links this is a change that I would expect could make a significant difference. I don't understand why this is described as "slightly" when a standard TCP sends 40B in 1 packet before it has validated the path, and this proposed change can send up to IW (10?) packets - that 10x the rate - in bytes this increases to x250 the sending rate (which may be what matters for slow links). Whether this matters I think needs to be determined, especially if there are robust ways to prevent this happening repeatedly over the same path. >> >> My 2 cents, with chair hat off... >> >> Michael >> > Gorry From ycheng@google.com Tue Nov 26 12:00:52 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6CFA81AE24E for ; Tue, 26 Nov 2013 12:00:52 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -0.779 X-Spam-Level: X-Spam-Status: No, score=-0.779 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FM_FORGED_GMAIL=0.622, HTML_MESSAGE=0.001, J_CHICKENPOX_32=0.6, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=no Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Txmlec7cAx4S for ; Tue, 26 Nov 2013 12:00:49 -0800 (PST) Received: from mail-ie0-x22e.google.com (mail-ie0-x22e.google.com [IPv6:2607:f8b0:4001:c03::22e]) by ietfa.amsl.com (Postfix) with ESMTP id 80BE71AE2C7 for ; Tue, 26 Nov 2013 12:00:42 -0800 (PST) Received: by mail-ie0-f174.google.com with SMTP id at1so9901299iec.19 for ; Tue, 26 Nov 2013 12:00:42 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :cc:content-type; bh=OSZqTeFi6HmmVKNsG2iO+Yk+PMWVmeyw65uFcHVF9yI=; b=K0UuavQJqLDE52EjDVA6ol5cznxaKfi0DT2Pz4626kYBHfrM099cjpx+1qvQMHe2V6 z0+dd38uT00LcZmuNUOX0NoSOHJs4mxlLFzQgsR+jA7p51vixdJd+LriSp/4Pxw6mKb/ ba4Uz5LStChyhgXgpNgYO6+1M6uoQit3KPBRkxPlVdtTXpsrO833JElwu++oqHWZAkKN FaTpqk8v19S2shqiAdMbyXjwVeirmwTJvHORJSpPbNNd4fQvskfZ2R0Zq+dXXBRcPphq qqlo2kwEropLoFFApYaLu1FtoWnmv9HB0P3dEORShlBSkyJQZxC+suJsMwVEtm/eBoVu Ziuw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:from:date :message-id:subject:to:cc:content-type; bh=OSZqTeFi6HmmVKNsG2iO+Yk+PMWVmeyw65uFcHVF9yI=; b=TyxVV45u9HtcsWyedAz40IoFoveOJtUtLoE7ti7fuO0Bd8FaL7I4XX16/B5U3Y5VvH RTffs2vsc4d8TS8N3ljkHjWoFV3fVQMh2LTOFMgxdEAyxYET7jr8mMpgap2bNWehEm0K S5s/m6Khp56qLFQ7mn8VQzdLp+Qj6s+7WOTg34rEuAlGJlgY7GBoQ2JZsQRi4/vc0Zjd SiIK5TRL3+rh/ZqqHh2uqQc+BU2SiF18Lh86yBb0tRV71jODXk5qjeZRXCW0Ky6uythN CkymdNZ8bMCS0eaRBeHCmPMhdkCmTEvAdRGrOs/uknl2WQ2v0+qD4allViyLt8ek4KgT koXw== X-Gm-Message-State: ALoCoQnTlkmQW0d0VEL1M7F0y+902pvQlXDtHv/1MoXsVkNkcL3v7+3fNiju5wYuVFpL4MujH5J0Ac/hF6P4l6k4WUT5zeolKH0XyHMw6AB/1o/+ex7sr/ql3MeyFQbf1dDrWidmgoNFdW5dzCh720g/Tis+HBCWRK0kw4FJqtiSTHKusn6/W8hZObsH8Fd9GGke2+QBoomt X-Received: by 10.51.16.3 with SMTP id fs3mr18188740igd.53.1385496042157; Tue, 26 Nov 2013 12:00:42 -0800 (PST) MIME-Version: 1.0 Received: by 10.64.142.71 with HTTP; Tue, 26 Nov 2013 12:00:01 -0800 (PST) In-Reply-To: <763b3f8541277a7690bf859a87e69265.squirrel@www.erg.abdn.ac.uk> References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526AF5B1.9070906@isi.edu> <655C07320163294895BBADA28372AF5D0EBFC6@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> <528B9E16.70602@erg.abdn.ac.uk> <655C07320163294895BBADA28372AF5D13C310@FR712WXCHMBA15.zeu.alcatel-lucent.com> <763b3f8541277a7690bf859a87e69265.squirrel@www.erg.abdn.ac.uk> From: Yuchung Cheng Date: Tue, 26 Nov 2013 12:00:01 -0800 Message-ID: To: Gorry Fairhurst Content-Type: multipart/alternative; boundary=001a1134cf1c6e52f504ec19ef95 Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: Re: [tcpm] ] WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 26 Nov 2013 20:00:52 -0000 --001a1134cf1c6e52f504ec19ef95 Content-Type: text/plain; charset=ISO-8859-1 On Tue, Nov 26, 2013 at 11:51 AM, wrote: > See below... > > > On Tue, Nov 26, 2013 at 6:05 AM, Scharf, Michael (Michael) < > > michael.scharf@alcatel-lucent.com> wrote: > > > >> > >>> 3) seeding the SYN RTT from a lower value makes the sender more > >> > >>> aggressive in heavily congested networks. The sender is also made > >> > more > >> > >>> aggressive by sending IW data packets before there is any > >> > indication the > >> > >>> path can contain even a single data segment. This is significant > >> > change to > >> > >>> standard behaviour. If the proposal is to use IW 3 (as standard) > >> it > >> > still > >> > >>> needs to be called out. If the proposal is to use a larger > >> > experimental > >> > >>> number then I have concerns here that this is a significant change > >> > that > >> > >>> needs an automated recovery method to prevent significant > >> > collateral damage > >> > >>> on capacity-limited paths - there needs to be a way to stop a > >> > server doing > >> > >>> this each time resulting in recurring loss! > >> > >> > >> > >> > >> > >> when the IW of packets are not all acked, loss recovery is > >> triggered > >> > >> and window is reduced. this is part of standard CC. > >> > >> > >> > >> also when you use "significant" four times, please back up with a > >> > good > >> > >> theory or data or both. and abuse that word does not it make more > >> > >> significant. > >> > >> > >> > > So let's see if we can see agree on what happens: > >> > > > >> > > - If the path is very lossey - severe congestion: > >> > > TCP standard sends one SYN segment with some probability of loss, > >> and > >> > if it > >> > > sees loss backs off and retransmits the SYN. As (over)load > >> increases, > >> > new > >> > > sessions add 40B and the new sessions will often defer start-up, > >> > controlling > >> > > their rate. > >> > > > >> > > - If TFO path is very lossey - severe congestion: IW full-sized > >> > segments are > >> > > sent. Each new session does this, adding to the load. > >> > > Each new session adds MSS*IW = 6000B (IW=3) or 15000B (IW=10). > >> > > > >> > > I concede this applies to severe congestion. Under this case, it > >> > seems 150x > >> > > more traffic to me and this before before CC is engaged. Or is this > >> > wrong? > >> > I can make similar case for any recent proposal that will blow up the > >> > network, e.g., rto-restart, newcwv. The question is how practical and > >> > common those cases are. if the link is thin, not doing Fast Open with > >> > same IW will experience similar heavy losses. Let's face it: if the > >> > capacity is not enough to handle the demand, there is little can be > >> > done. > >> > > >> > The concern has been raised by Michael Scharf's, and the step 6 in > >> > Section 4.2.2 does mention that. I did ask the list for any better > >> > solution for that but I've never heard back any suggestion. > >> > >> Indeed, I was (and I still am) worried by the combination of TFO and > >> IW10. > >> Because of that, we have already had similar list discussions about > >> highly > >> congested links. I am less concerned about TFO with IW3, because bursts > >> of > >> that size are sent by TCP anyway pretty frequently. > >> > >> However, given the explicit reference to RFC 3390 in the current draft, > >> I > >> have not further objected during the WGLC. > >> > >> Since this question still seems to trigger discussion, what about some > >> additional explanation of "slightly more aggressive"? For instance: > >> > >> OLD: > >> > >> Note that if SYN-ACK is lost, regular TCP reduces the initial > >> congestion window before sending any data. In this case TFO is > >> slightly more aggressive in the first data round trip even though > >> it does not change the congestion control. > >> > >> NEW: > >> > >> Note that in some cases TFO can be slightly more aggressive than > >> regular TCP. > >> If SYN-ACK is lost, regular TCP reduces the initial congestion > >> window before sending any data. > >> On such a congested link, TFO could first send a burst according > >> to > >> the initial window [RFC3390], > >> and the server would then back-off after one round trip time if > >> one or more packets are lost. In this case TFO is > >> slightly more aggressive in the first data round trip even though > >> it > >> does not change the congestion control. > >> > > sure > > > > The description is basically OK - but I don't agree on two points: > > My point is that this DOES change the congestion control in the most > critical starvation prevention mode (i.e. under heavy load): It allows a > single TCP session to send up to IW bytes into the network without > confirming the path is available. This is not a trivial change. It makes > significant assumptions about the buffering on the path, and the level of > current congestion - for high rate links this may be OK (likely no impact) > for low rate links this is a change that I would expect could make a > significant difference. > > I don't understand why this is described as "slightly" when a standard > TCP sends 40B in 1 packet before it has validated the path, and this > proposed change can send up to IW (10?) packets - that 10x the rate - in > bytes this increases to x250 the sending rate (which may be what matters > for slow links). > I will remove the word "slightly" if you can stop using "significantly", b/c both are subjective w/o good data to back-up > Whether this matters I think needs to be determined, especially if there > are robust ways to prevent this happening repeatedly over the same path. > What's your idea? > > >> > >> My 2 cents, with chair hat off... > >> > >> Michael > >> > > > > Gorry > > --001a1134cf1c6e52f504ec19ef95 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable



On Tue, Nov 26, 2013 at 11:51 AM, <gorry@erg.abdn.ac.uk>= ; wrote:
See below...

> On Tue, Nov 26, 2013 at 6:05 AM, Scharf, Michael (Michael) <
> michael.scharf@alcatel-lucent.com> wrote:
>
>> > >>> 3) seeding the SYN RTT from a lower value makes = the sender more
>> > >>> aggressive in heavily congested networks. The se= nder is also made
>> > more
>> > >>> aggressive by sending IW data packets before the= re is any
>> > indication the
>> > >>> path can contain even a single data segment. Thi= s is significant
>> > change to
>> > >>> standard behaviour. If the proposal is to use IW= 3 (as standard)
>> it
>> > still
>> > >>> needs to be called out. If the proposal is to us= e a larger
>> > experimental
>> > >>> number then I have concerns here that this is a = significant change
>> > that
>> > >>> needs an automated recovery method to prevent si= gnificant
>> > collateral damage
>> > >>> on capacity-limited paths - there needs to be a = way to stop a
>> > server doing
>> > >>> this each time resulting in recurring loss!
>> > >>
>> > >>
>> > >> when the IW of packets are not all acked, loss recov= ery is
>> triggered
>> > >> and window is reduced. this is part of standard CC.<= br> >> > >>
>> > >> also when you use "significant" four times= , please back up with a
>> > good
>> > >> theory or data or both. and abuse that word does not= it make more
>> > >> significant.
>> > >>
>> > > So let's see if we can see agree on what happens: >> > >
>> > > - If the path is very lossey - severe congestion:
>> > > TCP standard sends one SYN segment with some probability= of loss,
>> and
>> > if it
>> > > sees loss backs off and retransmits the SYN. As (over)lo= ad
>> increases,
>> > new
>> > > sessions add 40B and the new sessions will often defer s= tart-up,
>> > controlling
>> > > their rate.
>> > >
>> > > - If TFO path is very lossey - severe congestion: IW ful= l-sized
>> > segments are
>> > > sent. Each new session does this, adding to the load. >> > > Each new session adds =A0MSS*IW =3D 6000B (IW=3D3) or 15= 000B (IW=3D10).
>> > >
>> > > I concede this applies to severe congestion. Under this = case, it
>> > seems 150x
>> > > more traffic to me and this before before CC is engaged.= Or is this
>> > wrong?
>> > I can make similar case for any recent proposal that will blo= w up the
>> > network, e.g., rto-restart, newcwv. The question is how pract= ical and
>> > common those cases are. if the link is thin, not doing Fast O= pen with
>> > same IW will experience similar heavy losses. Let's face = it: if the
>> > capacity is not enough to handle the demand, there is little = can be
>> > done.
>> >
>> > The concern has been raised by Michael Scharf's, and the = step 6 in
>> > Section 4.2.2 does mention that. I did ask the list for any b= etter
>> > solution for that but I've never heard back any suggestio= n.
>>
>> Indeed, I was (and I still am) worried by the combination of TFO a= nd
>> IW10.
>> Because of that, we have already had similar list discussions abou= t
>> highly
>> congested links. I am less concerned about TFO with IW3, because b= ursts
>> of
>> that size are sent by TCP anyway pretty frequently.
>>
>> However, given the explicit reference to RFC 3390 in the current d= raft,
>> I
>> have not further objected during the WGLC.
>>
>> Since this question still seems to trigger discussion, what about = some
>> additional explanation of "slightly more aggressive"? Fo= r instance:
>>
>> OLD:
>>
>> =A0 =A0 =A0 Note that if SYN-ACK is lost, regular TCP reduces the = initial
>> =A0 =A0 =A0 congestion window before sending any data. In this cas= e TFO is
>> =A0 =A0 =A0 slightly more aggressive in the first data round trip = even though
>> =A0 =A0 =A0 it does not change the congestion control.
>>
>> NEW:
>>
>> =A0 =A0 =A0 Note that in some cases TFO can be slightly more aggre= ssive than
>> regular TCP.
>> =A0 =A0 =A0 If SYN-ACK is lost, regular TCP reduces the initial co= ngestion
>> window before sending any data.
>> =A0 =A0 =A0 On such a congested link, TFO could first send a burst= according
>> to
>> the initial window [RFC3390],
>> =A0 =A0 =A0 =A0 and the server would then back-off after one round= trip time if
>> one or more packets are lost. In this case TFO is
>> =A0 =A0 =A0 slightly more aggressive in the first data round trip = even though
>> it
>> does not change the congestion control.
>>
> sure
>

The description is basically OK - but I don't agree on two = points:

My point is that this DOES change the congestion control in the most
critical starvation prevention mode (i.e. under heavy load): It allows a single TCP session to send up to IW bytes into the network without
confirming the path is available. This is not a trivial change. It makes significant assumptions about the buffering on the path, and the level of current congestion - for high rate links this may be OK (likely no impact)<= br> for low rate links this is a change that I would expect could make a
significant difference.

I don't understand why this is described as =A0"slightly" whe= n a standard
TCP sends 40B in 1 packet before it has validated the path, and this
proposed change can send up to IW (10?) packets - that 10x the rate - in bytes this increases to x250 the sending rate (which may be what matters for slow links).
I will remove the word "slightly= " if you can stop using "significantly", b/c both are subjec= tive w/o good data to back-up


Whether this matters I think needs to be determined, especially if there are robust ways to prevent this happening repeatedly over the same path.
What's your idea?
=A0

>>
>> My 2 cents, with chair hat off...
>>
>> Michael
>>
>

Gorry


--001a1134cf1c6e52f504ec19ef95-- From erkrishna.khanal@gmail.com Wed Nov 27 21:27:44 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 333221AE06D for ; Wed, 27 Nov 2013 21:27:44 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -1.999 X-Spam-Level: X-Spam-Status: No, score=-1.999 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, DKIM_SIGNED=0.1, DKIM_VALID=-0.1, DKIM_VALID_AU=-0.1, FREEMAIL_FROM=0.001, HTML_MESSAGE=0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Co-KyL5Rs-mS for ; Wed, 27 Nov 2013 21:27:42 -0800 (PST) Received: from mail-ob0-x230.google.com (mail-ob0-x230.google.com [IPv6:2607:f8b0:4003:c01::230]) by ietfa.amsl.com (Postfix) with ESMTP id 33A481A1F62 for ; Wed, 27 Nov 2013 21:27:42 -0800 (PST) Received: by mail-ob0-f176.google.com with SMTP id va2so8279607obc.21 for ; Wed, 27 Nov 2013 21:27:41 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:date:message-id:subject:from:to:content-type; bh=cDOc2ducoBwEA8KhWQw+oCEHGEedOrTbZJ2SI91EmNo=; b=0KhHwLSUjvnt/i0q1dKrjBmCOvttgdFGDbXHwCC59/m2IBBf3mPk4MofyWHTr7S1d4 JwvPslLnlWIeavOEMNa1B5HGQzRjG5KKe7bl0uc3P+fPI6SkRwbXcf0YwpA1oXbzD3cd agQGSLnX85Z+EmL3T/F1Xy6f9VaIN7BL52FplIbL8LYtbsc6WmUNkQGimdJ0GTQTWn1S RrfNnXUfFuflAcVmuhalmhsp+Dz1qlgwmBNl1UxusKs2v4HXZj4at+ggBpLs79YeFEQi AN6saZ6S8AWBcHAGnzXsqccCUmtpglhd4FLWeZ6OSDuzYAdPkbqKvOO6R1JdhGu3Bz1q LedQ== MIME-Version: 1.0 X-Received: by 10.182.109.230 with SMTP id hv6mr234566obb.68.1385616461302; Wed, 27 Nov 2013 21:27:41 -0800 (PST) Received: by 10.182.135.1 with HTTP; Wed, 27 Nov 2013 21:27:41 -0800 (PST) Date: Thu, 28 Nov 2013 10:57:41 +0530 Message-ID: From: Krishna Khanal To: tcpm@ietf.org Content-Type: multipart/alternative; boundary=001a11c31b4ef8920e04ec35f84d Subject: [tcpm] ECN and synCookie X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Nov 2013 05:27:44 -0000 --001a11c31b4ef8920e04ec35f84d Content-Type: text/plain; charset=UTF-8 A server with synCookie support generally replies the important options received on SYN from client on its SYN+ACK packet to client after encoding it in its ISS. Since the number of options exchanged in SYN are growing, there are some alternate ways to remember this, like echoing the options in timestamp so that it can be offloaded from ISS. One of the option exchanged in handshake is ECN. When a server receives a SYN with ECE+CWR set, it replies with ECE. if server has synCookie enabled, it must remember this on ISS as well. What about echoing back the ECE option on final ack? If client receives SYN+ACK with ECE, it can set ECE on final ack and send it to server. With this, there are two advantages: 1. no need to encode the ECN capability on syncookie 2. if the path is asymmetric (though path can change anytime), client can inform server about it, if the ECE is not set on final ack, server can ignore ECN even if its enabled. Were there any discussions in the past on this? Does this make any sense? Regards, Krishna --001a11c31b4ef8920e04ec35f84d Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable

A server with synCookie support ge= nerally replies the important options received on SYN from
cl= ient on its SYN+ACK packet to client after encoding it in its ISS. Since th= e number of options
exchanged in SYN are growing, there ar= e some alternate ways to remember this, like echoing=C2=A0
th= e options in timestamp so that it can be offloaded from ISS.

=
= One of the option exchanged in handshake is ECN. When a server receives a S= YN with ECE+CWR
set, i= t replies with ECE. if server has synCookie enabled, it must remember this = on ISS as well.

What about echoing back the ECE op= tion on final ack? If client receives SYN+ACK with ECE, it can
se= t ECE on final ack and send it to server.

Wi= th this, there are two advantages:
1. no need to encode th= e ECN capability on syncookie
2.= if the path is=C2=A0asymmetric (though path can change anytime), client ca= n inform server about it,=C2=A0
if the ECE is not set on f= inal ack, server can ignore ECN even if its enabled.
Were there any discussions in the past on this? Does th= is make any sense?
Regards,
Krishna

--001a11c31b4ef8920e04ec35f84d-- From rs@netapp.com Thu Nov 28 02:41:56 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 932EE1ADFD4 for ; Thu, 28 Nov 2013 02:41:56 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -6.902 X-Spam-Level: X-Spam-Status: No, score=-6.902 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, HTML_MESSAGE=0.001, RCVD_IN_DNSWL_HI=-5, RP_MATCHES_RCVD=-0.001, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id tL-EDbLiAnsH for ; Thu, 28 Nov 2013 02:41:53 -0800 (PST) Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id 7F8A71AD939 for ; Thu, 28 Nov 2013 02:41:53 -0800 (PST) X-IronPort-AV: E=Sophos;i="4.93,790,1378882800"; d="scan'208,217";a="121255608" Received: from vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) by mx12-out.netapp.com with ESMTP; 28 Nov 2013 02:41:51 -0800 Received: from SACEXCMBX02-PRD.hq.netapp.com ([169.254.1.147]) by vmwexceht06-prd.hq.netapp.com ([10.106.77.104]) with mapi id 14.03.0123.003; Thu, 28 Nov 2013 02:41:52 -0800 From: "Scheffenegger, Richard" To: Krishna Khanal , "tcpm@ietf.org" Thread-Topic: [tcpm] ECN and synCookie Thread-Index: AQHO6/qTnHyvmDksR0KTCMF38fdvRZo6b0IQ Date: Thu, 28 Nov 2013 10:41:51 +0000 Message-ID: <012C3117EDDB3C4781FD802A8C27DD4F25EEEF6D@SACEXCMBX02-PRD.hq.netapp.com> References: In-Reply-To: Accept-Language: de-AT, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.106.53.53] Content-Type: multipart/alternative; boundary="_000_012C3117EDDB3C4781FD802A8C27DD4F25EEEF6DSACEXCMBX02PRDh_" MIME-Version: 1.0 Subject: Re: [tcpm] ECN and synCookie X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 28 Nov 2013 10:41:56 -0000 --_000_012C3117EDDB3C4781FD802A8C27DD4F25EEEF6DSACEXCMBX02PRDh_ Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: base64 SGkgS3Jpc2huYSwNCg0KYXMgdGhlIGV4YWN0IHdheSBob3cgYSBTeW4tQ29va2llIGlzIGZvcm1l ZCBpcyBhbiBhcmJpdHJhcnkgY2hvaWNlIGJ5IHRoZSBzZXJ2ZXIgZG9pbmcgdGhpcywgSSBkb27i gJl0IHNlZSB0aGF0IHRoaXMgY291bGRu4oCZdCBiZSBhY2hpZXZlZC4gQWxzbywgaXQgZG9lc27i gJl0IG5lZWQgYW55IHN0YW5kYXJkcyBhY3Rpb24uDQoNCg0KQ2hhbmdpbmcgdGhlIFNlbWFudGlj cyBvZiB0aGUgVENQIEVDTiBoYW5kc2hha2UgaG93ZXZlciwgd291bGQgY2hhbmdlIHRoZSBzZW1h bnRpY3M7IEnigJltIG9uZSBvZiB0aGUgY28tYXV0aG9ycyBvZiB0aGUgQWNjRUNOIGRyYWZ0cywg d2hlcmUgYSBkaWZmZXJlbnQgc2lnbmFsaW5nIGlzIHRvIGJlIGRlZmluZWQuIFRoZSBwb2ludCB5 b3UgbWFrZSBoYXMgbm90IHlldCBicm91Z2h0IGZvcndhcmQgYXMgYSByZXF1aXJlbWVudCB0byBi ZSBpbmNsdWRlZCBpbiBodHRwOi8vdG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFmdC1pZXRmLXRjcG0t YWNjZWNuLXJlcXMtMDQuIEJ1dCBhcyB3ZSBoYXZlIG5vdCB5ZXQgc2V0dGxlZCBvbiBhbnkgc3Bl Y2lmaWMgZW5jb2RpbmcgdG8gZ28gd2l0aCB0aGVzZSByZXF1aXJlbWVudHMsIHNvbWV0aGluZyBh bG9uZyB0aGUgbGluZXMgYmVsb3cgY291bGQgcHJvYmFibHkgYmUgZWFzaWx5IGFkZGVkLg0KDQpG b3IgZXhhbXBsZSwgd2hhdCB5b3UgYXNrIHdvdWxkIHByb2JhYmx5IGJlIGVhc3kgdG8gaW1wbGVt ZW50IGJvdGggd2l0aCBodHRwOi8vdG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFmdC1rdWVobGV3aW5k LXRjcG0tYWNjdXJhdGUtZWNuLTAyIGFuZCBodHRwOi8vdG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFm dC1rdWVobGV3aW5kLXRjcG0tYWNjdXJhdGUtZWNuLW9wdGlvbi0wMSAuDQoNCllvdSBtYXkgYWxz byBsb29rIGludG8gdGhlIHdvcmsgaGVyZSBodHRwOi8vdG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFm dC1rdWVobGV3aW5kLXRjcG0tZWNuLWZhbGxiYWNrLTAxIHRvIHNlZSBob3cgaXQgY291bGQgaW50 ZXJhY3QuDQoNCg0KQXMgYSBzaWRlIG5vdGUsIGV2ZW4gd2l0aGluIFJGQzMxNjgsIHlvdXIgKGFj dGl2ZSBvcGVuKSBjbGllbnQgd291bGQgaGF2ZSB0d28gKG5vbi1zdGFuZGFyZCkgb3B0aW9ucyB0 byBjb252ZXkgdGhlIHVzZSBvZiBFQ04gYWdhaW4gYmFjayB0byB0aGUgKHBhc3NpdmUgb3Blbikg c2VydmVyLiBPbmUgd291bGQgYmUgdG8gc2V0IHRoZSBFQ1QoMCkgb3IgRUNUKDEpIGNvZGVwb2lu dCBpbiB0aGUgQUNLIOKAkyB0aG91Z2ggUkZDMTM2OCBkb2VzbuKAmXQgYWxsb3cgdGhhdCBvbiBw dXJlIEFDS3M7IHRoZSBvdGhlciBvbmUgd291bGQgYmUgdG8gc2lnbmFsIENXUiBpbiB0aGUgcHVy ZSBBQ0suIEhvd2V2ZXIsIGFzIGJvdGggYXBwcm9hY2hlcyBhcmUgbm90IHNwZWNpZmljIChhbmQg dGhlIHVzZSBvZiBFQ1Qgb24gcHVyZSBBQ0tzIG1heSBoYXZlIG90aGVyIGltcGxpY2F0aW9ucyBm b3IgdGhlIG5ldHdvcmspLCB5b3UgY2FubuKAmXQgcmVseSBvbiB0aGUgc2VydmVyIHRvIHJlYWN0 IHByb3Blcmx5Lg0KDQpPZiBjb3Vyc2UsIGFueSBkYXRhIHNlZ21lbnQgZnJvbSB0aGUgY2xpZW50 IHRvIHRoZSBzZXJ2ZXIgd291bGQgbGVnaXRpbWF0ZWx5IGNvbnRhaW4gRUNUKDApLCBFQ1QoMSkg b3IgQ0UsIGFuZCB0aGUgc2VydmVyIGNvdWxkIOKAnGxhdGXigJ0gZW5hYmxlIEVDTiBzdXBwb3J0 IGFmdGVyIHJlY2VpdmluZyBzdWNoIGEgc2VnbWVudCwgaWYgdGhlIHNlc3Npb24gd2FzIHNldCB1 cCB1c2luZyBTeW4tQ29va2llcy4gVXAgdG8gdGhhdCBwb2ludCAodGhlIGZpcnN0IEVDTi1tYXJr ZWQgc2VnbWVudCBmcm9tIHRoZSBjbGllbnQpLCB0aGUgc2VydmVyIGNvdWxkIHNlbmQgcmVndWxh ciwgbm9uLUVDTiBzZWdtZW50cywgYW5kIEkgdGhpbmsgc3VjaCBhIHNlc3Npb24gd291bGQgYmUg ZnVsbHkgY29tcGxpYW50IOKAkyBqdXN0IG5vdCBtYWtlIHVzZSBvZiBFQ04gdW50aWwgdGhlIGZp cnN0IGluZGljYXRpb24gb2YgRUNOIGZyb20gdGhlIGNsaWVudOKApg0KDQoNCkJlc3QgcmVnYXJk cywNCg0KUmljaGFyZCBTY2hlZmZlbmVnZ2VyDQoNCg0KRnJvbTogdGNwbSBbbWFpbHRvOnRjcG0t Ym91bmNlc0BpZXRmLm9yZ10gT24gQmVoYWxmIE9mIEtyaXNobmEgS2hhbmFsDQpTZW50OiBEb25u ZXJzdGFnLCAyOC4gTm92ZW1iZXIgMjAxMyAwNjoyOA0KVG86IHRjcG1AaWV0Zi5vcmcNClN1Ympl Y3Q6IFt0Y3BtXSBFQ04gYW5kIHN5bkNvb2tpZQ0KDQoNCkEgc2VydmVyIHdpdGggc3luQ29va2ll IHN1cHBvcnQgZ2VuZXJhbGx5IHJlcGxpZXMgdGhlIGltcG9ydGFudCBvcHRpb25zIHJlY2VpdmVk IG9uIFNZTiBmcm9tDQpjbGllbnQgb24gaXRzIFNZTitBQ0sgcGFja2V0IHRvIGNsaWVudCBhZnRl ciBlbmNvZGluZyBpdCBpbiBpdHMgSVNTLiBTaW5jZSB0aGUgbnVtYmVyIG9mIG9wdGlvbnMNCmV4 Y2hhbmdlZCBpbiBTWU4gYXJlIGdyb3dpbmcsIHRoZXJlIGFyZSBzb21lIGFsdGVybmF0ZSB3YXlz IHRvIHJlbWVtYmVyIHRoaXMsIGxpa2UgZWNob2luZw0KdGhlIG9wdGlvbnMgaW4gdGltZXN0YW1w IHNvIHRoYXQgaXQgY2FuIGJlIG9mZmxvYWRlZCBmcm9tIElTUy4NCg0KT25lIG9mIHRoZSBvcHRp b24gZXhjaGFuZ2VkIGluIGhhbmRzaGFrZSBpcyBFQ04uIFdoZW4gYSBzZXJ2ZXIgcmVjZWl2ZXMg YSBTWU4gd2l0aCBFQ0UrQ1dSDQpzZXQsIGl0IHJlcGxpZXMgd2l0aCBFQ0UuIGlmIHNlcnZlciBo YXMgc3luQ29va2llIGVuYWJsZWQsIGl0IG11c3QgcmVtZW1iZXIgdGhpcyBvbiBJU1MgYXMgd2Vs bC4NCg0KV2hhdCBhYm91dCBlY2hvaW5nIGJhY2sgdGhlIEVDRSBvcHRpb24gb24gZmluYWwgYWNr PyBJZiBjbGllbnQgcmVjZWl2ZXMgU1lOK0FDSyB3aXRoIEVDRSwgaXQgY2FuDQpzZXQgRUNFIG9u IGZpbmFsIGFjayBhbmQgc2VuZCBpdCB0byBzZXJ2ZXIuDQoNCldpdGggdGhpcywgdGhlcmUgYXJl IHR3byBhZHZhbnRhZ2VzOg0KMS4gbm8gbmVlZCB0byBlbmNvZGUgdGhlIEVDTiBjYXBhYmlsaXR5 IG9uIHN5bmNvb2tpZQ0KMi4gaWYgdGhlIHBhdGggaXMgYXN5bW1ldHJpYyAodGhvdWdoIHBhdGgg Y2FuIGNoYW5nZSBhbnl0aW1lKSwgY2xpZW50IGNhbiBpbmZvcm0gc2VydmVyIGFib3V0IGl0LA0K aWYgdGhlIEVDRSBpcyBub3Qgc2V0IG9uIGZpbmFsIGFjaywgc2VydmVyIGNhbiBpZ25vcmUgRUNO IGV2ZW4gaWYgaXRzIGVuYWJsZWQuDQoNCldlcmUgdGhlcmUgYW55IGRpc2N1c3Npb25zIGluIHRo ZSBwYXN0IG9uIHRoaXM/IERvZXMgdGhpcyBtYWtlIGFueSBzZW5zZT8NCg0KUmVnYXJkcywNCkty aXNobmENCg0KDQo= --_000_012C3117EDDB3C4781FD802A8C27DD4F25EEEF6DSACEXCMBX02PRDh_ Content-Type: text/html; charset="utf-8" Content-Transfer-Encoding: base64 PGh0bWwgeG1sbnM6dj0idXJuOnNjaGVtYXMtbWljcm9zb2Z0LWNvbTp2bWwiIHhtbG5zOm89InVy bjpzY2hlbWFzLW1pY3Jvc29mdC1jb206b2ZmaWNlOm9mZmljZSIgeG1sbnM6dz0idXJuOnNjaGVt YXMtbWljcm9zb2Z0LWNvbTpvZmZpY2U6d29yZCIgeG1sbnM6eD0idXJuOnNjaGVtYXMtbWljcm9z b2Z0LWNvbTpvZmZpY2U6ZXhjZWwiIHhtbG5zOm09Imh0dHA6Ly9zY2hlbWFzLm1pY3Jvc29mdC5j b20vb2ZmaWNlLzIwMDQvMTIvb21tbCIgeG1sbnM9Imh0dHA6Ly93d3cudzMub3JnL1RSL1JFQy1o dG1sNDAiPg0KPGhlYWQ+DQo8bWV0YSBodHRwLWVxdWl2PSJDb250ZW50LVR5cGUiIGNvbnRlbnQ9 InRleHQvaHRtbDsgY2hhcnNldD11dGYtOCI+DQo8bWV0YSBuYW1lPSJHZW5lcmF0b3IiIGNvbnRl bnQ9Ik1pY3Jvc29mdCBXb3JkIDE0IChmaWx0ZXJlZCBtZWRpdW0pIj4NCjxzdHlsZT48IS0tDQov KiBGb250IERlZmluaXRpb25zICovDQpAZm9udC1mYWNlDQoJe2ZvbnQtZmFtaWx5OkNhbGlicmk7 DQoJcGFub3NlLTE6MiAxNSA1IDIgMiAyIDQgMyAyIDQ7fQ0KQGZvbnQtZmFjZQ0KCXtmb250LWZh bWlseTpUYWhvbWE7DQoJcGFub3NlLTE6MiAxMSA2IDQgMyA1IDQgNCAyIDQ7fQ0KLyogU3R5bGUg RGVmaW5pdGlvbnMgKi8NCnAuTXNvTm9ybWFsLCBsaS5Nc29Ob3JtYWwsIGRpdi5Nc29Ob3JtYWwN Cgl7bWFyZ2luOjBjbTsNCgltYXJnaW4tYm90dG9tOi4wMDAxcHQ7DQoJZm9udC1zaXplOjEyLjBw dDsNCglmb250LWZhbWlseToiVGltZXMgTmV3IFJvbWFuIiwic2VyaWYiO30NCmE6bGluaywgc3Bh bi5Nc29IeXBlcmxpbmsNCgl7bXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCWNvbG9yOmJsdWU7DQoJ dGV4dC1kZWNvcmF0aW9uOnVuZGVybGluZTt9DQphOnZpc2l0ZWQsIHNwYW4uTXNvSHlwZXJsaW5r Rm9sbG93ZWQNCgl7bXNvLXN0eWxlLXByaW9yaXR5Ojk5Ow0KCWNvbG9yOnB1cnBsZTsNCgl0ZXh0 LWRlY29yYXRpb246dW5kZXJsaW5lO30NCnNwYW4uRW1haWxTdHlsZTE3DQoJe21zby1zdHlsZS10 eXBlOnBlcnNvbmFsLXJlcGx5Ow0KCWZvbnQtZmFtaWx5OiJDYWxpYnJpIiwic2Fucy1zZXJpZiI7 DQoJY29sb3I6IzFGNDk3RDt9DQouTXNvQ2hwRGVmYXVsdA0KCXttc28tc3R5bGUtdHlwZTpleHBv cnQtb25seTsNCglmb250LWZhbWlseToiQ2FsaWJyaSIsInNhbnMtc2VyaWYiOw0KCW1zby1mYXJl YXN0LWxhbmd1YWdlOkVOLVVTO30NCkBwYWdlIFdvcmRTZWN0aW9uMQ0KCXtzaXplOjYxMi4wcHQg NzkyLjBwdDsNCgltYXJnaW46NzAuODVwdCA3MC44NXB0IDIuMGNtIDcwLjg1cHQ7fQ0KZGl2Lldv cmRTZWN0aW9uMQ0KCXtwYWdlOldvcmRTZWN0aW9uMTt9DQotLT48L3N0eWxlPjwhLS1baWYgZ3Rl IG1zbyA5XT48eG1sPg0KPG86c2hhcGVkZWZhdWx0cyB2OmV4dD0iZWRpdCIgc3BpZG1heD0iMTAy NiIgLz4NCjwveG1sPjwhW2VuZGlmXS0tPjwhLS1baWYgZ3RlIG1zbyA5XT48eG1sPg0KPG86c2hh cGVsYXlvdXQgdjpleHQ9ImVkaXQiPg0KPG86aWRtYXAgdjpleHQ9ImVkaXQiIGRhdGE9IjEiIC8+ DQo8L286c2hhcGVsYXlvdXQ+PC94bWw+PCFbZW5kaWZdLS0+DQo8L2hlYWQ+DQo8Ym9keSBsYW5n PSJERS1BVCIgbGluaz0iYmx1ZSIgdmxpbms9InB1cnBsZSI+DQo8ZGl2IGNsYXNzPSJXb3JkU2Vj dGlvbjEiPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMS4w cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7 O2NvbG9yOiMxRjQ5N0QiPkhpIEtyaXNobmEsPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xh c3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6 JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0Qi PjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFu IGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtD YWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+YXMgdGhl IGV4YWN0IHdheSBob3cgYSBTeW4tQ29va2llIGlzIGZvcm1lZCBpcyBhbiBhcmJpdHJhcnkgY2hv aWNlIGJ5IHRoZSBzZXJ2ZXIgZG9pbmcgdGhpcywgSSBkb27igJl0IHNlZSB0aGF0IHRoaXMgY291 bGRu4oCZdCBiZSBhY2hpZXZlZC4gQWxzbywNCiBpdCBkb2VzbuKAmXQgbmVlZCBhbnkgc3RhbmRh cmRzIGFjdGlvbi48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48 c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1 b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxv OnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIGxh bmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTomcXVvdDtDYWxp YnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+PG86cD4mbmJz cDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gbGFuZz0iRU4t VVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVv dDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5DaGFuZ2luZyB0aGUgU2Vt YW50aWNzIG9mIHRoZSBUQ1AgRUNOIGhhbmRzaGFrZSBob3dldmVyLCB3b3VsZCBjaGFuZ2UgdGhl IHNlbWFudGljczsgSeKAmW0gb25lIG9mIHRoZSBjby1hdXRob3JzIG9mIHRoZSBBY2NFQ04gZHJh ZnRzLCB3aGVyZSBhIGRpZmZlcmVudA0KIHNpZ25hbGluZyBpcyB0byBiZSBkZWZpbmVkLiBUaGUg cG9pbnQgeW91IG1ha2UgaGFzIG5vdCB5ZXQgYnJvdWdodCBmb3J3YXJkIGFzIGEgcmVxdWlyZW1l bnQgdG8gYmUgaW5jbHVkZWQgaW4NCjxhIGhyZWY9Imh0dHA6Ly90b29scy5pZXRmLm9yZy9odG1s L2RyYWZ0LWlldGYtdGNwbS1hY2NlY24tcmVxcy0wNCI+aHR0cDovL3Rvb2xzLmlldGYub3JnL2h0 bWwvZHJhZnQtaWV0Zi10Y3BtLWFjY2Vjbi1yZXFzLTA0PC9hPi4gQnV0IGFzIHdlIGhhdmUgbm90 IHlldCBzZXR0bGVkIG9uIGFueSBzcGVjaWZpYyBlbmNvZGluZyB0byBnbyB3aXRoIHRoZXNlIHJl cXVpcmVtZW50cywgc29tZXRoaW5nIGFsb25nIHRoZSBsaW5lcyBiZWxvdyBjb3VsZCBwcm9iYWJs eQ0KIGJlIGVhc2lseSBhZGRlZC48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNv Tm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1m YW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMx RjQ5N0QiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwi PjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWlseTom cXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3RCI+ Rm9yIGV4YW1wbGUsIHdoYXQgeW91IGFzayB3b3VsZCBwcm9iYWJseSBiZSBlYXN5IHRvIGltcGxl bWVudCBib3RoIHdpdGgNCjxhIGhyZWY9Imh0dHA6Ly90b29scy5pZXRmLm9yZy9odG1sL2RyYWZ0 LWt1ZWhsZXdpbmQtdGNwbS1hY2N1cmF0ZS1lY24tMDIiPmh0dHA6Ly90b29scy5pZXRmLm9yZy9o dG1sL2RyYWZ0LWt1ZWhsZXdpbmQtdGNwbS1hY2N1cmF0ZS1lY24tMDI8L2E+IGFuZCBodHRwOi8v dG9vbHMuaWV0Zi5vcmcvaHRtbC9kcmFmdC1rdWVobGV3aW5kLXRjcG0tYWNjdXJhdGUtZWNuLW9w dGlvbi0wMSAuPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNw YW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90 O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpw PiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5n PSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJy aSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPllvdSBtYXkgYWxz byBsb29rIGludG8gdGhlIHdvcmsgaGVyZQ0KPGEgaHJlZj0iaHR0cDovL3Rvb2xzLmlldGYub3Jn L2h0bWwvZHJhZnQta3VlaGxld2luZC10Y3BtLWVjbi1mYWxsYmFjay0wMSI+aHR0cDovL3Rvb2xz LmlldGYub3JnL2h0bWwvZHJhZnQta3VlaGxld2luZC10Y3BtLWVjbi1mYWxsYmFjay0wMTwvYT4g dG8gc2VlIGhvdyBpdCBjb3VsZCBpbnRlcmFjdC48bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBj bGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4w cHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7 O2NvbG9yOiMxRjQ5N0QiPjxvOnA+Jm5ic3A7PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJN c29Ob3JtYWwiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250 LWZhbWlseTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6 IzFGNDk3RCI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1h bCI+PHNwYW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5 OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdE Ij5BcyBhIHNpZGUgbm90ZSwgZXZlbiB3aXRoaW4gUkZDMzE2OCwgeW91ciAoYWN0aXZlIG9wZW4p IGNsaWVudCB3b3VsZCBoYXZlIHR3byAobm9uLXN0YW5kYXJkKSBvcHRpb25zIHRvIGNvbnZleSB0 aGUgdXNlIG9mIEVDTiBhZ2FpbiBiYWNrIHRvIHRoZQ0KIChwYXNzaXZlIG9wZW4pIHNlcnZlci4g T25lIHdvdWxkIGJlIHRvIHNldCB0aGUgRUNUKDApIG9yIEVDVCgxKSBjb2RlcG9pbnQgaW4gdGhl IEFDSyDigJMgdGhvdWdoIFJGQzEzNjggZG9lc27igJl0IGFsbG93IHRoYXQgb24gcHVyZSBBQ0tz OyB0aGUgb3RoZXIgb25lIHdvdWxkIGJlIHRvIHNpZ25hbCBDV1IgaW4gdGhlIHB1cmUgQUNLLiBI b3dldmVyLCBhcyBib3RoIGFwcHJvYWNoZXMgYXJlIG5vdCBzcGVjaWZpYyAoYW5kIHRoZSB1c2Ug b2YgRUNUIG9uDQogcHVyZSBBQ0tzIG1heSBoYXZlIG90aGVyIGltcGxpY2F0aW9ucyBmb3IgdGhl IG5ldHdvcmspLCB5b3UgY2FubuKAmXQgcmVseSBvbiB0aGUgc2VydmVyIHRvIHJlYWN0IHByb3Bl cmx5Lg0KPG86cD48L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4g bGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0Nh bGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpwPiZu YnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5nPSJF Ti1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZx dW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPk9mIGNvdXJzZSwgYW55 IGRhdGEgc2VnbWVudCBmcm9tIHRoZSBjbGllbnQgdG8gdGhlIHNlcnZlciB3b3VsZCBsZWdpdGlt YXRlbHkgY29udGFpbiBFQ1QoMCksIEVDVCgxKSBvciBDRSwgYW5kIHRoZSBzZXJ2ZXIgY291bGQg 4oCcbGF0ZeKAnSBlbmFibGUgRUNODQogc3VwcG9ydCBhZnRlciByZWNlaXZpbmcgc3VjaCBhIHNl Z21lbnQsIGlmIHRoZSBzZXNzaW9uIHdhcyBzZXQgdXAgdXNpbmcgU3luLUNvb2tpZXMuIFVwIHRv IHRoYXQgcG9pbnQgKHRoZSBmaXJzdCBFQ04tbWFya2VkIHNlZ21lbnQgZnJvbSB0aGUgY2xpZW50 KSwgdGhlIHNlcnZlciBjb3VsZCBzZW5kIHJlZ3VsYXIsIG5vbi1FQ04gc2VnbWVudHMsIGFuZCBJ IHRoaW5rIHN1Y2ggYSBzZXNzaW9uIHdvdWxkIGJlIGZ1bGx5IGNvbXBsaWFudCDigJMganVzdA0K IG5vdCBtYWtlIHVzZSBvZiBFQ04gdW50aWwgdGhlIGZpcnN0IGluZGljYXRpb24gb2YgRUNOIGZy b20gdGhlIGNsaWVudOKApjxvOnA+PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3Jt YWwiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1zaXplOjExLjBwdDtmb250LWZhbWls eTomcXVvdDtDYWxpYnJpJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDs7Y29sb3I6IzFGNDk3 RCI+PG86cD4mbmJzcDs8L286cD48L3NwYW4+PC9wPg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNw YW4gbGFuZz0iRU4tVVMiIHN0eWxlPSJmb250LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90 O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpw PiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5n PSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJy aSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPkJlc3QgcmVnYXJk cyw8bzpwPjwvbzpwPjwvc3Bhbj48L3A+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBsYW5n PSJFTi1VUyIgc3R5bGU9ImZvbnQtc2l6ZToxMS4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJy aSZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxvOnA+Jm5ic3A7 PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250 LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1z ZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj5SaWNoYXJkIFNjaGVmZmVuZWdnZXI8L3NwYW4+PHNw YW4gc3R5bGU9ImZvbnQtc2l6ZToxMC4wcHQ7Zm9udC1mYW1pbHk6JnF1b3Q7Q2FsaWJyaSZxdW90 OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7O2NvbG9yOiMxRjQ5N0QiPjxicj4NCjxicj4NCjxvOnA+ PC9vOnA+PC9zcGFuPjwvcD4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250 LXNpemU6MTEuMHB0O2ZvbnQtZmFtaWx5OiZxdW90O0NhbGlicmkmcXVvdDssJnF1b3Q7c2Fucy1z ZXJpZiZxdW90Oztjb2xvcjojMUY0OTdEIj48bzpwPiZuYnNwOzwvbzpwPjwvc3Bhbj48L3A+DQo8 ZGl2IHN0eWxlPSJib3JkZXI6bm9uZTtib3JkZXItbGVmdDpzb2xpZCBibHVlIDEuNXB0O3BhZGRp bmc6MGNtIDBjbSAwY20gNC4wcHQiPg0KPGRpdj4NCjxkaXYgc3R5bGU9ImJvcmRlcjpub25lO2Jv cmRlci10b3A6c29saWQgI0I1QzRERiAxLjBwdDtwYWRkaW5nOjMuMHB0IDBjbSAwY20gMGNtIj4N CjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxiPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1z aXplOjEwLjBwdDtmb250LWZhbWlseTomcXVvdDtUYWhvbWEmcXVvdDssJnF1b3Q7c2Fucy1zZXJp ZiZxdW90OyI+RnJvbTo8L3NwYW4+PC9iPjxzcGFuIGxhbmc9IkVOLVVTIiBzdHlsZT0iZm9udC1z aXplOjEwLjBwdDtmb250LWZhbWlseTomcXVvdDtUYWhvbWEmcXVvdDssJnF1b3Q7c2Fucy1zZXJp ZiZxdW90OyI+IHRjcG0gW21haWx0bzp0Y3BtLWJvdW5jZXNAaWV0Zi5vcmddDQo8Yj5PbiBCZWhh bGYgT2YgPC9iPktyaXNobmEgS2hhbmFsPGJyPg0KPGI+U2VudDo8L2I+IERvbm5lcnN0YWcsIDI4 LiBOb3ZlbWJlciAyMDEzIDA2OjI4PGJyPg0KPGI+VG86PC9iPiB0Y3BtQGlldGYub3JnPGJyPg0K PGI+U3ViamVjdDo8L2I+IFt0Y3BtXSBFQ04gYW5kIHN5bkNvb2tpZTxvOnA+PC9vOnA+PC9zcGFu PjwvcD4NCjwvZGl2Pg0KPC9kaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwv bzpwPjwvcD4NCjxkaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8 L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHls ZT0iZm9udC1mYW1pbHk6JnF1b3Q7QXJpYWwmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+ QSBzZXJ2ZXIgd2l0aCBzeW5Db29raWUgc3VwcG9ydCBnZW5lcmFsbHkgcmVwbGllcyB0aGUgaW1w b3J0YW50IG9wdGlvbnMgcmVjZWl2ZWQgb24gU1lOIGZyb208L3NwYW4+PG86cD48L286cD48L3A+ DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1m YW1pbHk6JnF1b3Q7QXJpYWwmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+Y2xpZW50IG9u IGl0cyBTWU4mIzQzO0FDSyBwYWNrZXQgdG8gY2xpZW50IGFmdGVyIGVuY29kaW5nIGl0IGluIGl0 cyBJU1MuIFNpbmNlIHRoZSBudW1iZXIgb2Ygb3B0aW9uczwvc3Bhbj48bzpwPjwvbzpwPjwvcD4N CjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZh bWlseTomcXVvdDtBcmlhbCZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5leGNoYW5nZWQg aW4gU1lOIGFyZSBncm93aW5nLCB0aGVyZSBhcmUgc29tZSBhbHRlcm5hdGUgd2F5cyB0byByZW1l bWJlciB0aGlzLCBsaWtlIGVjaG9pbmcmbmJzcDs8L3NwYW4+PG86cD48L286cD48L3A+DQo8L2Rp dj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6 JnF1b3Q7QXJpYWwmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+dGhlIG9wdGlvbnMgaW4g dGltZXN0YW1wIHNvIHRoYXQgaXQgY2FuIGJlIG9mZmxvYWRlZCBmcm9tIElTUy48L3NwYW4+PG86 cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZu YnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFu IHN0eWxlPSJmb250LWZhbWlseTomcXVvdDtBcmlhbCZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1 b3Q7Ij5PbmUgb2YgdGhlIG9wdGlvbiBleGNoYW5nZWQgaW4gaGFuZHNoYWtlIGlzIEVDTi4gV2hl biBhIHNlcnZlciByZWNlaXZlcyBhIFNZTiB3aXRoIEVDRSYjNDM7Q1dSPC9zcGFuPjxvOnA+PC9v OnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9 ImZvbnQtZmFtaWx5OiZxdW90O0FyaWFsJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPnNl dCwgaXQgcmVwbGllcyB3aXRoIEVDRS4gaWYgc2VydmVyIGhhcyBzeW5Db29raWUgZW5hYmxlZCwg aXQgbXVzdCByZW1lbWJlciB0aGlzIG9uIElTUyBhcyB3ZWxsLjwvc3Bhbj48bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+ PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZv bnQtZmFtaWx5OiZxdW90O0FyaWFsJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPldoYXQg YWJvdXQgZWNob2luZyBiYWNrIHRoZSBFQ0Ugb3B0aW9uIG9uIGZpbmFsIGFjaz8gSWYgY2xpZW50 IHJlY2VpdmVzIFNZTiYjNDM7QUNLIHdpdGggRUNFLCBpdCBjYW48L3NwYW4+PG86cD48L286cD48 L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9u dC1mYW1pbHk6JnF1b3Q7QXJpYWwmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+c2V0IEVD RSBvbiBmaW5hbCBhY2sgYW5kIHNlbmQgaXQgdG8gc2VydmVyLjwvc3Bhbj48bzpwPjwvbzpwPjwv cD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9vOnA+ PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZv bnQtZmFtaWx5OiZxdW90O0FyaWFsJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPldpdGgg dGhpcywgdGhlcmUgYXJlIHR3byBhZHZhbnRhZ2VzOjwvc3Bhbj48bzpwPjwvbzpwPjwvcD4NCjwv ZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZhbWls eTomcXVvdDtBcmlhbCZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij4xLiBubyBuZWVkIHRv IGVuY29kZSB0aGUgRUNOIGNhcGFiaWxpdHkgb24gc3luY29va2llPC9zcGFuPjxvOnA+PC9vOnA+ PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PHNwYW4gc3R5bGU9ImZv bnQtZmFtaWx5OiZxdW90O0FyaWFsJnF1b3Q7LCZxdW90O3NhbnMtc2VyaWYmcXVvdDsiPjIuIGlm IHRoZSBwYXRoIGlzJm5ic3A7YXN5bW1ldHJpYyAodGhvdWdoIHBhdGggY2FuIGNoYW5nZSBhbnl0 aW1lKSwgY2xpZW50IGNhbiBpbmZvcm0gc2VydmVyIGFib3V0IGl0LCZuYnNwOzwvc3Bhbj48bzpw PjwvbzpwPjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0 eWxlPSJmb250LWZhbWlseTomcXVvdDtBcmlhbCZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7 Ij5pZiB0aGUgRUNFIGlzIG5vdCBzZXQgb24gZmluYWwgYWNrLCBzZXJ2ZXIgY2FuIGlnbm9yZSBF Q04gZXZlbiBpZiBpdHMgZW5hYmxlZC48L3NwYW4+PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxk aXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48bzpwPiZuYnNwOzwvbzpwPjwvcD4NCjwvZGl2Pg0K PGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxzcGFuIHN0eWxlPSJmb250LWZhbWlseTomcXVv dDtBcmlhbCZxdW90OywmcXVvdDtzYW5zLXNlcmlmJnF1b3Q7Ij5XZXJlIHRoZXJlIGFueSBkaXNj dXNzaW9ucyBpbiB0aGUgcGFzdCBvbiB0aGlzPyBEb2VzIHRoaXMgbWFrZSBhbnkgc2Vuc2U/PC9z cGFuPjxvOnA+PC9vOnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+ PG86cD4mbmJzcDs8L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+DQo8cCBjbGFzcz0iTXNvTm9ybWFs Ij48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7QXJpYWwmcXVvdDssJnF1b3Q7c2Fucy1z ZXJpZiZxdW90OyI+UmVnYXJkcyw8L3NwYW4+PG86cD48L286cD48L3A+DQo8L2Rpdj4NCjxkaXY+ DQo8cCBjbGFzcz0iTXNvTm9ybWFsIj48c3BhbiBzdHlsZT0iZm9udC1mYW1pbHk6JnF1b3Q7QXJp YWwmcXVvdDssJnF1b3Q7c2Fucy1zZXJpZiZxdW90OyI+S3Jpc2huYTwvc3Bhbj48bzpwPjwvbzpw PjwvcD4NCjwvZGl2Pg0KPGRpdj4NCjxwIGNsYXNzPSJNc29Ob3JtYWwiPjxvOnA+Jm5ic3A7PC9v OnA+PC9wPg0KPC9kaXY+DQo8ZGl2Pg0KPHAgY2xhc3M9Ik1zb05vcm1hbCI+PG86cD4mbmJzcDs8 L286cD48L3A+DQo8L2Rpdj4NCjwvZGl2Pg0KPC9kaXY+DQo8L2Rpdj4NCjwvYm9keT4NCjwvaHRt bD4NCg== --_000_012C3117EDDB3C4781FD802A8C27DD4F25EEEF6DSACEXCMBX02PRDh_-- From gorry@erg.abdn.ac.uk Sat Nov 30 09:07:57 2013 Return-Path: X-Original-To: tcpm@ietfa.amsl.com Delivered-To: tcpm@ietfa.amsl.com Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 82BD91AE46D for ; Sat, 30 Nov 2013 09:07:57 -0800 (PST) X-Virus-Scanned: amavisd-new at amsl.com X-Spam-Flag: NO X-Spam-Score: -3.602 X-Spam-Level: X-Spam-Status: No, score=-3.602 tagged_above=-999 required=5 tests=[BAYES_00=-1.9, J_CHICKENPOX_36=0.6, RCVD_IN_DNSWL_MED=-2.3, RP_MATCHES_RCVD=-0.001, SPF_PASS=-0.001] autolearn=ham Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6uUGTddMXgqr for ; Sat, 30 Nov 2013 09:07:55 -0800 (PST) Received: from spey.erg.abdn.ac.uk (spey.erg.abdn.ac.uk [139.133.204.173]) by ietfa.amsl.com (Postfix) with ESMTP id 916281AE45D for ; Sat, 30 Nov 2013 09:07:55 -0800 (PST) Received: by spey.erg.abdn.ac.uk (Postfix, from userid 5001) id 8EA3B2B4525; Sat, 30 Nov 2013 17:07:53 +0000 (GMT) Received: from ra-gorry.erg.abdn.ac.uk (ra-gorry.erg.abdn.ac.uk [139.133.204.42]) by spey.erg.abdn.ac.uk (Postfix) with ESMTPSA id 22F682B43AA; Sat, 30 Nov 2013 17:07:50 +0000 (GMT) Message-ID: <529A1B65.4070008@erg.abdn.ac.uk> Date: Sat, 30 Nov 2013 17:07:49 +0000 From: Gorry Fairhurst Organization: The University of Aberdeen is a charity registered in Scotland, No SC013683. User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:24.0) Gecko/20100101 Thunderbird/24.0.1 MIME-Version: 1.0 To: Yuchung Cheng References: <655C07320163294895BBADA28372AF5D0EAFCE@FR712WXCHMBA15.zeu.alcatel-lucent.com> <526E8E6B.1060806@isi.edu> <0B96A5D7-0DAE-46FF-8D9A-311307BF7493@netapp.com> <3578D243-D0F6-41E8-B515-380C35BB27B9@isi.edu> <9267762C-FD7C-4BC6-85FB-730E774F7EEB@oracle.com> <527016BD.4090609@isi.edu> <88430820-7495-491A-AE7A-D3850973AA35@oracle.com> <527036D0.4030508@isi.edu> <2C14475E-675C-40BB-9DD6-8C2871161903@oracle.com> <2B85C60B-2301-4B1D-8176-044DAEA817A6@erg.abdn.ac.uk> <528B9E16.70602@erg.abdn.ac.uk> <655C07320163294895BBADA28372AF5D13C310@FR712WXCHMBA15.zeu.alcatel-lucent.com> <763b3f8541277a7690bf859a87e69265.squirrel@www.erg.abdn.ac.uk> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: "tcpm@ietf.org" , "draft-ietf-tcpm-fastopen@tools.ietf.org" Subject: [tcpm] Summary of mt comments/questions WGLC for draft-ietf-tcpm-fastopen-05 X-BeenThere: tcpm@ietf.org X-Mailman-Version: 2.1.15 Precedence: list Reply-To: gorry@erg.abdn.ac.uk List-Id: TCP Maintenance and Minor Extensions Working Group List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 30 Nov 2013 17:07:57 -0000 Sorry for the delay (I was unwell), so this email is to restart. I think we got a long way through discussing things, so I've summarised below all the points as I understand from my side, so we and others know where we have arrived and can fix others. I support the experimentation with TFO, although I would argue we do need to write more about how the proposed changes in this ID will modify STD behaviour and explicitly say how to "negatively" cache path info that indicates TFO should not be used. I would like to see documented the cases where this could @@potentially@@ be worse than STD mechanisms. The ID already has some good documentation, but I think it needs a little more detail especially the set of potential issues that we expect to be mitigated by a cache at both the client and server - especially when and why do we need to revoke the cookie to prevent potentially bad pathologies from being repeated in a short time between a pair of clients and servers that keep starting up new connections. --- The points below hopefully help us see what we can agree on: a) Case: cached RTT greater than minRTO The current text says "RECOMMEND that the client caches the MSS and RTT to the server to enhance performance." I think ideally we should also explicitly say if the RTT is not cached TCP SHOULD disable use of TFO (make the cookie invalid) when the RTT>MinRTO. ... & note something about "this prevents a premature RTO for flows with a long RTT and also provides an acceptable RTT for pacing" - I feel the current text focuses on performance improvement for a low RTT, but does not explain why it is important for large RTTs. b) Case: cached RTT less than minRTO and path change We could just note that this ID does not change the behaviour that a path change can result in an invalid RTT/RTO value, and that normal TCP behaviour is expected to recover. c) Case: receiver-advertised MSS less than 536 bytes. This is an unusual corner-case and is not one that I think we need to engineer for. However, I do think we need to note that an IPv4 receiver advertised MSS less than 536 would result in transmission of an unexpected large segment to the receiver. (I.e. say we noted this, but don't change the recommendation). d) Case: receiver-advertised MSS greater than 1460 B, i.e. allowing TFO with large segments. I think we should explicitly say MUST NOT use an MSS that results in a packet size greater than 1500B. - AND I think we should caution that we do not have current experience of the effect of this behaviour. We could note it as an area for further experimentation? I'm not sure what we agreed? e) Case: A sender overwhelms a path with limited capacity. This is the corner case where a SYN may fail. The issue is most severe for links that suffer major overload (think low-speed large multiplexing). First SORRY - I'd like to avoid "slightly" and "significantly" wording - we discussed on the thread - instead what matters to me really is: This draft updates the congestion-control behaviour while the sender is waiting acknowledgement for a SYN. This will result in more data (up to one IW of segments) being sent before the sender receives the ACK for the SYN. This could result in additional congestion on links that have a high loss rate where the initial ACK would have been lost, which in standard behaviour would have resulted in a lower initial congestion window. - personally I see the effect as important to document, but is OK for experimentation with an IW of 3 (see (g)). f) The cookie-less experimental method will likely also need a negative cache to disable inappropriate TFO usage. I'd like to see an explicit note for the future experimentation something a bit like this would be good: "Even if cookie-less methods are found, it is expected that a future method will still need a way to detect conditions where TFO should not be used due to path properties and some equivalent method may then be needed to disable TFO." g) I am unsure that allowing TFO to use IW=10 is safe! Best wishes, Gorry