[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[freehaven-cvs] slides from paul"s talk at fc03

Update of /home/freehaven/cvsroot/doc/fc03
In directory moria.mit.edu:/home/arma/work/freehaven/doc/fc03

Added Files:
	slides-fc03.pdf slides-fc03.tex 
Log Message:
slides from paul's talk at fc03


--- NEW FILE: slides-fc03.pdf ---
%PDF-1.3
3 0 obj <<
/Length 430       
/Filter /FlateDecode
>>
stream
xڅRn0+xh<nS[==()Dr
ˇ&Chvvv=aA2nfxK|Ip9G4JN#kYi8pDqL"iwE#_VJ*\[ua59Y]BM$:I՗DC7]%vîo29˟(@0r`N~RHzԋc
-d*
AW6endstream
endobj
2 0 obj <<
/Type /Page
/Contents 3 0 R
/Resources 1 0 R
/MediaBox [0 0 792 612]
/Parent 7 0 R
>> endobj
1 0 obj <<
/Font << /F16 6 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
10 0 obj <<
/Length 475       
/Filter /FlateDecode
>>
stream
xUMo0+|4?1H*[b
	wlMDH1gLrO8^@~_?As9G(&&}MWL"Z8Kf;'d?yeL9ɤ$qc_+ʗT+0XO./JM˻4åL!|~Looɗ4qHw"yPDHs8,䚃=Nw	~y(&FJ.p+N	fU*iVL0նnVAWߵ!*k&Z1~ׅnd4c>&bIlwsdD{_B`,Ǭ
]M5k2u7S.;l
Eq]YZFؗ561NbNQy^hSql?unÊuWzRHQx&Yi%d%,Oi?{endstream
endobj
9 0 obj <<
/Type /Page
/Contents 10 0 R
/Resources 8 0 R
/MediaBox [0 0 792 612]
/Parent 7 0 R
>> endobj
8 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
16 0 obj <<
/Length 695       
/Filter /FlateDecode
>>
stream
xݖMs0Ui
&År0xHvY}8uڜ䐵J}ޕo.߮va<ϑ\F+uA2eQ$I"(*?
O))0VDB B>i2~śT!X2+'YV?BXP(צk#k;b]۸")o1ˏǼ"B(ӄEsu2	Ir'YBI&4}eD$-vLtk]5M+jj 9m=TllM:J
,ڙrh`"1b2|]DH`LpM㺶/?Gڇ{޷ôwƚaFS-R{91:9c(8Qz%6&2(VDgt+O
endobj
15 0 obj <<
/Type /Page
/Contents 16 0 R
/Resources 14 0 R
/MediaBox [0 0 792 612]
/Parent 7 0 R
>> endobj
14 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
19 0 obj <<
/Length 545       
/Filter /FlateDecode
>>
stream
xTM0+|+H[ԞZS%$qwYZE&)ٴ*03o$An_> RqR
I1Um,%Ŕp$'@}:eK^bQR$B,EmR/Po=LD6ԛo靱搑SxzrNdG,}1M<m!,?vgY.Ku1h<]3Z
i	(x]wظoV,i}7;(
;{	o;Š3ͻr"7Dk]
endobj
18 0 obj <<
/Type /Page
/Contents 19 0 R
/Resources 17 0 R
/MediaBox [0 0 792 612]
/Parent 7 0 R
>> endobj
17 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R /F35 22 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
25 0 obj <<
/Length 625       
/Filter /FlateDecode
>>
stream
xUn0+x%}=uOiM[B`iTbi||j9.gf_VشMla(ޘiEdن}]@BVPL
jݱ96_Z.D
x-/`2z\
Lwy!e]N_ϋ6S1ڵFN{›	N{YBnN]Z_ax]sc^HЙ;:ow^
uڔc|v6
y5t;bF;RgSc(M.e܆!Rrn~$Tm27a{&!ig'vIap!,/P+ɨMt%%
fjD@KKh4gc<88;TLE*	1#xiYdOfME@g3IX'|-
?4Qsy
E4/mqnOƭ w[ewgFRE{Towc={ȶwTwv
endobj
24 0 obj <<
/Type /Page
/Contents 25 0 R
/Resources 23 0 R
/MediaBox [0 0 792 612]
/Parent 7 0 R
>> endobj
23 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
28 0 obj <<
/Length 488       
/Filter /FlateDecode
>>
stream
xՔMo0
oHPZqkzͪMʿ|vpVZ=Iy'^6iqdWgsN~)2geBKɴP
}B❥z<&Es|
[fm'42[d-{)źkOyBXGBQ@q愘iVK`~\*=RE	Ka
T1GY@}vqvxtO؃m^hm%RVfZ3˧=TΝiCw]"wYHGyY`̥Kp<5Oendstream
endobj
27 0 obj <<
/Type /Page
/Contents 28 0 R
/Resources 26 0 R
/MediaBox [0 0 792 612]
/Parent 7 0 R
>> endobj
26 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R /F35 22 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
31 0 obj <<
/Length 575       
/Filter /FlateDecode
>>
stream
xՕMo@=.o=eﱑJT^E[;s׋$nIHf1hE:K2㌃$J ndKדZˌc|C(c4$Z2H_MVǭV@	&_n0DH
dZngvvg2LU˲LUllwcDչ+40a-\!X1)_pˑŤ)d׻
U%KUzՈ0)8vE1BFDMovglnytjN>iTm*.a]MX#v]'iq^2>B5}_;	FzB6 Gs4q8S}5V
ruk>	Q7un6clƃܰXwnfWkzܱƥVpexrNάSx*Bendstream
endobj
30 0 obj <<
/Type /Page
/Contents 31 0 R
/Resources 29 0 R
/MediaBox [0 0 792 612]
/Parent 32 0 R
>> endobj
29 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R /F35 22 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
35 0 obj <<
/Length 564       
/Filter /FlateDecode
>>
stream
xUr +E:3/Ŗ1$HN]$Ns/-S)l8ť,vlž`B6KPLKDžr^l}9jn$gM1Pċįq8yK]g>a
8H/Dz"P.+Mw~,kU(e!nۧ=@4\J;4:-avy݁|;P4هQ6*0k"5PS-Eh%yZ	$7J7LM;-gy85솱d#(Yy\B<|fVHb&	AGbNYnU=ԢGnLuf*(q#h-n@Oq;g@Nduh`u'ٚkȑ(A-Й$ih|˙1#ȣޞZ";}M	M	GTfGSaJެ^5[-_+:t,)?Ce9c]`8zmC9mDE-uZ(M$ra<FŷH8|9_GAcdр+_sMendstream
endobj
34 0 obj <<
/Type /Page
/Contents 35 0 R
/Resources 33 0 R
/MediaBox [0 0 792 612]
/Parent 32 0 R
>> endobj
33 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
38 0 obj <<
/Length 467       
/Filter /FlateDecode
>>
stream
xՔn0~
*3':
ذÆIR8+$KS9,H# 7Yt^{fQ=[/ΞND͌ 4P`; }zځqEj֠0u J5ZPZXgׅ}lVE#Yo~æZAwdaH%Zi>C$ۛtz[q-"HIC
8RI&OL(js	|yg{*iSW7۩Ye=C5Sw]>DF *OpM~#u&w1$u3:gNcJִ
lk7ˆϻTݫRuB)E18.p3frϊ#u&*JR:|M7Ҭ]$g
mOiU/4
endobj
37 0 obj <<
/Type /Page
/Contents 38 0 R
/Resources 36 0 R
/MediaBox [0 0 792 612]
/Parent 32 0 R
>> endobj
36 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
41 0 obj <<
/Length 577       
/Filter /FlateDecode
>>
stream
xUn0+xjH<h]
 ے#X+y(v)3
5KbCү|3im7UPJ/HI~#S
TA̐,ɚo'`)`$1	}
J'#fo?  Yi!W!oÈa„`UlྫCv~E=~nvnEk?WB%*mVCK|xgPOʁ1A"1 tpɄֆ& ccz	)Ư+.|yUW[#1Ęo(2._u}tAJ?'e$
K^<_5߮ztcBپ7fdu=yNwsF4Vy׆/±0SfSq'_)aWaJa,vb36(E$?@F@EngXǛ8G:$IArBNvrQ.CUn}z8mٹv֪].pbD׈hu됢u:ly]4͚oW{&
9M0A$|TIendstream
endobj
40 0 obj <<
/Type /Page
/Contents 41 0 R
/Resources 39 0 R
/MediaBox [0 0 792 612]
/Parent 32 0 R
>> endobj
39 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
44 0 obj <<
/Length 564       
/Filter /FlateDecode
>>
stream
xUMs0Wh,oЁtJ
88l)LYIn6%9L=w(d4Zj$S-PW
 S8Ez.ْ+,Db")Òj08&?aSwśD"N0-7>eY|2&X>NH#"SMu'B7df4C>&J	_0B9ep"EEE;+|ΤB	3OqEe{6
hNbxogo=u;c^(w	i+{B\_ʮ"qS8qJxh*oOWn\.0WwJY.o{7Q.6ЍL؋nr?o*vpz]{nDtl;F\8iE_=	oIEm3Pd^	βlClcC;m^$ױ(n<1'7z+Dcp.2R)קxZ`#CV2C>`Ɏ|endstream
endobj
43 0 obj <<
/Type /Page
/Contents 44 0 R
/Resources 42 0 R
/MediaBox [0 0 792 612]
/Parent 32 0 R
>> endobj
42 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
47 0 obj <<
/Length 618       
/Filter /FlateDecode
>>
stream
xUMo0W65S6`;mڃ؉l'EGJvfIs.DHQ{<RkWhss935kV3<"~6l\[
hMfYϤZ3>|pK2_>U]@WM'
LzX!*ۢ+_`Q,U!|ՂNPs!\[6di~R9Z뿁P	!`Z}
&(}(Ox7൭6Pn]=b
:}}&}4`ѮPYMOFImSz%=AN{rCvmm!lW|;(mUNq{Wipc( xd}
o=Rp
,d`zGSx8IZIG	5ڈbOt~E;]<Uؓ>WOf4Vu c*Oд:~34{+O>G	v|FP'wHkw(.RUp<[4֭Kh.̌Du_˅ѹzHnM65|Bendstream
endobj
46 0 obj <<
/Type /Page
/Contents 47 0 R
/Resources 45 0 R
/MediaBox [0 0 792 612]
/Parent 32 0 R
>> endobj
45 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
50 0 obj <<
/Length 678       
/Filter /FlateDecode
>>
stream
xV;0+ԝ]XXO[7A*p|ĜcہɿgW󸄤>&v1"9闑t,FcfrHo<cQ3I74YLԧUbhIIK͇c{dqMO^e-yL2
ml6e]wEV'CGwCBPA2#(oB
y7yzKD2QPP8h'+#_a0`%'Y!`uMی"?r
2ؐ:d+T,'^$na74S
\~#콑kC!N:4ѮYĆuw{^e\&vjk_]Dh
ګqPt/M7%Di*ߟ"q:iO8a)ǑS_4nIVa^Av
b۴	sӶ~p^7zPLn0~T.p}b0l&]K؇=[Ԫ
ĥ9>ͧ[DcZ刘j}>`\1Ɍ(mb<n/(FO/@endstream
endobj
49 0 obj <<
/Type /Page
/Contents 50 0 R
/Resources 48 0 R
/MediaBox [0 0 792 612]
/Parent 51 0 R
>> endobj
48 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R /F35 22 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
54 0 obj <<
/Length 539       
/Filter /FlateDecode
>>
stream
xՕn0~
' CqORZ袜Tjqe9޾CRv'0@Z!@j} (cQ (7̕SHn(@"MѧcJ1P(Pd$B,%wXˠyKW{PD
X4ʼ@o{KlGnKNO9ŧ6fX`^6X%bnj1ec%j7IuWַTqf4m4zBOd_wRN_#xfh~dqM"<vy4Ot\oG]x#萿yDY2*YJQK\Ԉ` #qS>xendstream
endobj
53 0 obj <<
/Type /Page
/Contents 54 0 R
/Resources 52 0 R
/MediaBox [0 0 792 612]
/Parent 51 0 R
>> endobj
52 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
57 0 obj <<
/Length 590       
/Filter /FlateDecode
>>
stream
xTn0+xN
@u
endobj
56 0 obj <<
/Type /Page
/Contents 57 0 R
/Resources 55 0 R
/MediaBox [0 0 792 612]
/Parent 51 0 R
>> endobj
55 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
60 0 obj <<
/Length 791       
/Filter /FlateDecode
>>
stream
xݖo0W1ۉyl`*O,uh%i"xIV}@6M|㻳',?E8?ehcB(X6g_&Tbx,+Wk6>켵̸hqtX+I4yO(L^O'/
ͤ"l:'[pEЅ,4WPƆ:pu*fJ=2ITMi֕Er8YULLq$I	h0,o"IEUIF|<#!4EISfߐ<h‘

#jBC/`gd7<W~]LmҶ"VvU
Xb?"`yO;G-!P8
!
!XN|Maca F'L9ؿC탥QQ~lWg<O/xe7ײHy~~P<0,endstream
endobj
59 0 obj <<
/Type /Page
/Contents 60 0 R
/Resources 58 0 R
/MediaBox [0 0 792 612]
/Parent 51 0 R
>> endobj
58 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R /F36 63 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
66 0 obj <<
/Length 893       
/Filter /FlateDecode
>>
stream
xWMo0W
ؒ_0-
`4Fqغx2u3\~uu/BEk9 FqG֙(I'X5j&%1m19,hP"qICڶ)׷vWۅbQ~\} DZ$TF)K#emd66n+KAnڊUP_n`h\NGOcpv&+HӶ7ߖZ[c8!a/=ihhN+Mߺ6e/7f՚蘒u)
endobj
65 0 obj <<
/Type /Page
/Contents 66 0 R
/Resources 64 0 R
/MediaBox [0 0 792 612]
/Parent 51 0 R
>> endobj
64 0 obj <<
/Font << /F16 6 0 R /F32 13 0 R /F36 63 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
69 0 obj <<
/Length 201       
/Filter /FlateDecode
>>
stream
xڥ?A)w&lAEuN	*a}iv
F453PdpL]I a%_[wW+nj!r–5J>"jfDI#!$VHFdޝ-*RIS}4ttNʿ:5X@1jhO_@+cYendstream
endobj
68 0 obj <<
/Type /Page
/Contents 69 0 R
/Resources 67 0 R
/MediaBox [0 0 792 612]
/Parent 51 0 R
>> endobj
67 0 obj <<
/Font << /F16 6 0 R >>
/ProcSet [ /PDF /Text ]
>> endobj
62 0 obj <<
/Length1 1475
/Length2 7608
/Length3 532
/Length 8484      
/Filter /FlateDecode
>>
stream
xeXTmװA!iP:D%[:}zwϜkk]Zfz-]k'(@MN]WWV̓,Yy88[yD<""0g
:{
%;

@g;

r#K

"<@oEZyH+ ?#/ryms߉n0<@WO(ER
@nА(E:Œ),:NVR@BIya:@gnX=6<ǨLzσ*׋<!1IC=fencX>ŦY*#6S>/t
/-ҀZ<oEg.l_ .0(#MKA?G˝lFэ&Ψp)(_~UUe}-Z5|j1*3m~y^ZU(Q'j}N߾ ǐnA.7Y:`uhyzAvՁxkhLK,a&)C"2Ckiҧ	(#2b<ĪwL;Tƺ#M?ޓyy,kT8*PúU7|VثD3
C -<į=kI.Zf:1嬀g~ ]YSl<4wz{:[tLs>öx59"ca9f$Mr`S3t7Ӄ욧Dzc:
+n<B7͋ }4^v>)w_<%é!qkhKZ<=l&j'A]_KYW*bΘwis*&WfR{locivo	NYBr=fGӐxМF0ݷ,_[-V
fw<‚4w33MTXݶ
vrT#g7X<cU6r(zyX]rKhJqv>gYvR|WEN21{KX'|Hr]mtQg@Qj8#9O-d$q*!4Y,%sF+L<OeS1ؾl-aYaˏDbFB+7,$#*nkoAMٖ:rZie;FF̭WYfi}>;';J	9>⎇`pϫA^?ȗ(p4j鏂8"M1yljp#h}EX<׎VL/|</"3/@/wa\47J:"ҠBWJv٦IPam8Hל53P X"GA@i)1to]m	M\Ƹv9+Ϫ'J wZkVkB3@EįlZ;5W0|QXKOݭOgC8PٻW00YXm2k;$2c+-E؃uSᲪ!CYHѣxgAC{q۹*3\l	c8s:`*qg[hL%Fr$s\#{+=~_.g@vXçs}*h':h;Iؿ͐w}eyiE"uQk;@A(O#|id$Mfx]u0 ҙbO&i`smrޯfnֈ'eS@6x;F^Wd',2=${_Q)Nޛ74ͬ/gy,WE?SI`$A
G-%Gzg.,&%th1<vԣi[kX)z(ucɁ]'c(]%F8Nr.&jV6c҂b<=$|ȷ1.K|T
S{pՋnαXzL NhϨ"@kvz6zEF6P!njTQL9,_?%O^|I嵪l66X6}E[_7u?Iޭs.cp30)sz:KekJ\@	Fu%XC୽n*8zn_ XK/j+_ybUPalX݆M|jm9)łPO?<ZmFQWnt2׋m<aZt͘I'4=Ҟ4.:X6¶LCxܧ)sI%bfbwwv.C^"Zh^B*7|q{68D
MţT+oWeK,8j.m ڮț>qu3rIwK<-</vD(ziZ#f
-DfC`#%mS@ >*Cߪ}MGt%%%2:eyV8ŠC=*o1>Ń1W
 o9ϯ+e̮7?&@ah⡂&
/8{r7{8$6^Ȁ!آz=Dh?ዃN!R0XJ	X\TS,&;{%iBfU''5wNB>elXWLh
"X=,_O}:.}J@MRn~	]k#L'U>[xxnyˠC
\Пm[M2bUxv;*0Rm9e;؟^}1f{dߗfZjWv97cPf
2w^t|:fz>A[r/
c|ڤQc[_'B4KRJIx>+s\3:DJr
ikL#=cW͊;&c'[:X篟}U
X_|.x%md#7.Q1
3y
mw>lڶu!e%-2 Et*eGR]b&yK/5])įoJ7DN̮<P8bQN"Ew$%%CŚBo2|Ahdޤ0ƗWt8U,XX^pSpA>	w
Nr%ؑpY0GÓTEuDl#K\)-fPsU?I&is=fȢxТ}}3J)׍''ί&#)P6vv8|xmP,,467VxKW>ޡbZ6L|E<|쭹 nyqx37?)>O0psY=wq\X~f$ϓe꣖ೆp^w$l'?tXYT^FM%/
,et><t?+Cφ823/'%ᣭ]8
$nfR1*'[qԊZYU9me4v~ W2ɯ<`%y"xū8=ֵvܰ-PwNW`kuYYA<mi&Gʙ%Ӣ`afRH\9d^msSW=Nf=ׁМ
37?`d7& ɂtdCVyrd	 {*`
1ABt*O,-*qrQ<3QȾE9
Q~#~͌?rxfF
[q|*]=,Z+p(!7jk6T;GʒI6;hGSO̗DYrLfD_Ff^m:޲xtu
	k4+\NnYPa:矙eZ]i7LqZS_48҇qIFWQb2z0J/*r\Y~ʭߨܸT;r6i|5DIjE	Q('ن=SH~@?tzJUF /xPyw_>K0cy}fnZ).״4bZ}-F3S՚>-=Թ,&ew= p*-Tڍjݤ	'ـ)#*=mS0,cݒˈfIcO-{ DVz"4HNLii}u
;=ϱf
endobj
63 0 obj <<
/Type /Font
/Subtype /Type1
/Encoding 70 0 R
/FirstChar 12
/LastChar 124
/Widths 71 0 R
/BaseFont /GBHUSY+LCMSSB8
/FontDescriptor 61 0 R
>> endobj
61 0 obj <<
/Ascent 694
/CapHeight 684
/Descent -139
/FontName /GBHUSY+LCMSSB8
/ItalicAngle 0
/StemV 146
/XHeight 538
/FontBBox [-99 -209 1413 840]
/Flags 4
/CharSet (/fi/ffi/parenleft/parenright/comma/hyphen/period/colon/A/G/H/I/K/L/M/P/R/S/T/U/W/a/b/c/d/e/f/g/h/i/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/emdash)
/FontFile 62 0 R
>> endobj
71 0 obj
[722 0 1104 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 535 535 0 0 382 458 382 0 0 0 0 0 0 0 0 0 0 0 382 0 0 0 0 0 0 889 0 0 0 0 0 917 899 361 0 927 778 1128 0 0 865 0 898 764 983 894 0 1271 0 0 0 0 0 0 0 0 0 646 681 611 681 627 420 681 681 299 0 642 299 1062 681 687 681 681 455 527 497 681 604 910 604 604 590 0 1375 ]
endobj
70 0 obj <<
/Type /Encoding
/Differences [ 0 /.notdef 12/fi 13/.notdef 14/ffi 15/.notdef 40/parenleft/parenright 42/.notdef 44/comma/hyphen/period 47/.notdef 58/colon 59/.notdef 65/A 66/.notdef 71/G/H/I 74/.notdef 75/K/L/M 78/.notdef 80/P 81/.notdef 82/R/S/T/U 86/.notdef 87/W 88/.notdef 97/a/b/c/d/e/f/g/h/i 106/.notdef 107/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z 123/.notdef 124/emdash 125/.notdef]
>> endobj
21 0 obj <<
/Length1 950
/Length2 3159
/Length3 532
/Length 3816      
/Filter /FlateDecode
>>
stream
xi<ƕ}do[aa2fcO(KD$L${cIM$!ٕ]*Yn[x~y^?zs}q;KFAQMO$E	XZ98!A	KH&@
E^(@‚0(w` HP^$!wR0H侧<.@̚@oңDgEH "A"w)xV 	zq P
<
ѶXAu=}z?B,-9(,!8GЅ/ߺh,PRUPD"*[)
Jt70 F(~oKNxl@hf•4j?^AD"'S1X-
2ד+ž]8Z$d'_sV2 ["nc	o|ʉOЋTLJ/2[d
R@NxjjZ4g[BGm=y~r#, \Z염[GqhǣyY5'#'(cv*&	ٰ@EDg~?p\DVL^s
ET(_juiPLe;$GK=.s@T2}=BW_o.BF_3)?tl&8eTg]_,	i%0oQw-뮠>9682߭mNTq4؃AOK~"sYh1I8JNF'm}ŗJ*yzpC=W}$Ϻ^4ӷ7"N<׺ۿĝ ħ'pɈ\th⪥K
s@mC?elL?$	I?4g3+ss~Mbwz/st
Й4ݠc"+D:#wΘJG|fvoR1Iz*"~0t{q1a]󞒌rҭj^]YedUYkPK䫘IΣph_Xާ<į>?9%1Q1\Qk
9ƈ	cmؒgK<,[a}
W(n/wϗZ$,NfcY5&O.Lebu>ЧXa^9?»@hˊXbAarv}KXЋy_ߋќY֑4o^;z# Z|ҷ
kRҴ_ Uh¾&
{f*7j
 7snIyyؓ
-7JY=Y.)+dZ-iyVpzVrE߶gpi3Ǔj|_쮄&ՇqLm~Dcrwn,P̈Y}~M^62q܆{DEf0mKdzvffRe-95b+pa[#jY\o
EPVݶ.{ؚzV]^[tQ-#E_gROB!<[7F'Į
3MN//1*IΖ.]QzseSaA;P`;Dٵ̼)k,ۮlnM&x/^4
Y-?0yRoԸ{9_{Gv!JazLعh2˨f 0+rS[+OͧNtqp2(RYY4SrُdwPKޖy<ݍ
"3dxt4h^e`|Z%8atk旓*Rlju
λA$l]7xꋸ|B߮5%k@zEܘȗP..)eϥVC)EϞ˖\Y39|4ps	zqJua*0ᴃ;ƨގOV܄Ӵ?^8E$PD_endstream
endobj
22 0 obj <<
/Type /Font
/Subtype /Type1
/Encoding 72 0 R
/FirstChar 14
/LastChar 121
/Widths 73 0 R
/BaseFont /LKNSTA+LCMSSI8
/FontDescriptor 20 0 R
>> endobj
20 0 obj <<
/Ascent 694
/CapHeight 684
/Descent -139
/FontName /LKNSTA+LCMSSI8
/ItalicAngle -12
/StemV 80
/XHeight 521
/FontBBox [-110 -208 1391 857]
/Flags 4
/CharSet (/ffi/a/c/d/e/i/n/o/r/s/t/u/y)
/FontFile 21 0 R
>> endobj
73 0 obj
[1007 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 590 0 556 625 556 0 0 0 278 0 0 0 0 625 625 0 0 417 479 451 625 0 0 0 556 ]
endobj
72 0 obj <<
/Type /Encoding
/Differences [ 0 /.notdef 14/ffi 15/.notdef 97/a 98/.notdef 99/c/d/e 102/.notdef 105/i 106/.notdef 110/n/o 112/.notdef 114/r/s/t/u 118/.notdef 121/y 122/.notdef]
>> endobj
12 0 obj <<
/Length1 795
/Length2 773
/Length3 532
/Length 1344      
/Filter /FlateDecode
>>
stream
xRkPWA"d7T@@-O{d6 C>*c,*H}HK*OEEjE:
EyH7PSt~9|gw=q*P$rQ7!(XYyPJ@7SP@n|ʒ7
ՐSi_l2VHA0QrT
4' !q.EN`iqNqE{(raM"NR%abgAɿajB*]#i5!}
49Ny8MG}<x)3qL܇Hx
'x
endobj
13 0 obj <<
/Type /Font
/Subtype /Type1
/Encoding 74 0 R
/FirstChar 15
/LastChar 41
/Widths 75 0 R
/BaseFont /HRISHZ+CMSY8
/FontDescriptor 11 0 R
>> endobj
11 0 obj <<
/Ascent 750
/CapHeight 683
/Descent -194
/FontName /HRISHZ+CMSY8
/ItalicAngle -14
/StemV 89
/XHeight 431
/FontBBox [-30 -955 1185 779]
/Flags 4
/CharSet (/bullet/arrowdblright)
/FontFile 12 0 R
>> endobj
75 0 obj
[531 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1063 ]
endobj
74 0 obj <<
/Type /Encoding
/Differences [ 0 /.notdef 15/bullet 16/.notdef 41/arrowdblright 42/.notdef]
>> endobj
5 0 obj <<
/Length1 1962
/Length2 11553
/Length3 532
/Length 12631     
/Filter /FlateDecode
>>
stream
xUX̲;A
5www
@`y{ssUWW恌XQN(noBD
SQ032!8\,D\
fXڙ=
 U.!H_RT5W@<B*Ͽ/տ03H?{
{xӁΜ &fdjv$%
z\	4|\3Jlf	@amބ'*9Op\ؼ%^w1x5ܦd\Ƒz^Oe&ƇCC"g<N)@C7L6<L9Sgl
[ZEC9ç_)EU`7tB943mCYQ#X7Q6gZi֯;]'>ʂQj2}dLGWpƓ2
<[Í=SXXӭ=9$@jBt/J7;
S{L.rtsD4;lY-v=@YPM2'5(WQ
AN(ū!G	p!uYjQo=䙱U)0e11jpġ~4*.x`aZrOZ!g|~U6lgTwZ`%3_Mazǒ	sDjع6gM52oƵukHKE聵	c45L3~dYؚ*`FZʏc/"^zj8}P-hUxG^?	̩/(wr̷,"<F^^ޤ~|z'b繁`M1uM͡i{_ÊTˡ{lwU	dXDU`elnݒPU 2MDdy}$[³Hr5aŠG;cl*mIIEov9nG%e$wH'N#G%`i)	
2	Ce1GxrD32C;03G3M%,b쎠d50˯/mS%CBz:nKWxegY5SgUWIi=oܥvd΋MKi躂}Iy-{5y؊̑b\Bٻ^WsI
CTg
%<j$lȊ%li|c@
M7EW䷷/"Aɲ8#-}L\c<*(Va*HX%	fG/Sk~9A(sɤ-"ShoV%T(_ǽCҬ,d'ȅ+nfh!!#sOsD1E}vo#ҋ9R,!~`]`j3ߨ4J*yeU;U|zAEKN.n,8uyBc5Kmv8oPLq@h4hxcBw:>d7.ЄvSzXdi^ô0=l4PtOIpuިCHW>YC֋H͏:`j1zlf&55{P?Ps!dzg/9?~[&,W;׎ Ry YufM8$<~ W[-t]c1ΞP^zx͈`nu|$+PWE_W%yэF;[PR}jDTLCMدy'
;*<nz@!@Xs拳 H~OR?M3{槴ɱ)W8r
qluF
v!8>xcnTNrAJB;%EF5.1V60xқ$OVtxSrxC͑]}-=FKk%RY]'1?t!'	
xCSC'ށ0ܴ֘<+j.RF7Ő*fEu]iLME
aw0Xa@P<쏄Ɠ?3"S9!܋ae1b>鎗|PH2cfu)jIyI0cy]읜NvF*Q̛w'^x sMFplN`l&`tM$q/V5i	ٸ
s!/)k!X-Шs|V짠}\f˼4^@
fS<mzRg
U;foQKW&ZQ~C(ҽU)jPh*gK39Y9\~N=VU|1DtXRE6W
?wJ{<*g$CˣUQ8cX#I{-S>LfjUM6,i
vYgL81ξ,Sh֘ytC`5r\r~i.	P
XMJVA5v'lLmse=ciAӐ)&%bBܨ
'?THRL{4oUA{2M,ͽ3|צͥ͢l"ҰPV%Q!ѿEtVysw#i{͉B@~S!u<)V4-nVsB_Z)Nc*BuUsb/+*#6M-yvvY)>,ىU6oTO&{tRj?Gr%1’O{v56z/x	y12
=RfO]wWv:[	'̣y`<#֗;GjKh)?.eC`D.a|JI7v0[ڱugSDyӿ&Tov`nסsĠOGD1R0QQ
wu]<.
:eBxuv\K]*smp]5 e"%k!9ndu,-綒1"غMiv%q7P\(c]e}ӛsfalS˴Ɔ3FVbpW<M"z0'2;TNv>	aKA>AmZwn|C&~aS+F J[K{[͐[t6
bCؗqP,AlzoIx8RO1iҦKylvDH|.ⳤ\NmYuձJwQ|2Ua~ݶ?U=V
G(ISu|j]4)By^pIŗOxdɇqi`iaV{opdpaσ
zheaQtUx'J5Ϫf))MS62?L3^ZG,dV5g1C%6.G
f_˥K]rJ$\6Ymu.dp3SrM)kQha8
>p;Ut4X\xuOZ02İmVs	
BM*}&jQoECych>4[*eui/WyxrFCOdJC݁ę9?*%=儂@O5X>
ON4vqgqeN$
D:6bY
RjKǕk(!KQr@->bFzz1qL*MnXR-rX7d;VsuM{3IH81_zT8
Ȭ=(^g{qNPs/!6x=}Ğ<.d'uQ(VLy@||AkpׅmnCfo&~6!
9̎z1!ZM2ᨥH+ă{ٞM:c<LL[hj ̷65b4[p6{~xy_pk(7tD;pXH~ffnI(rCp(#w5UV./+,p𠑭⿔OaClD:l癩MO)C#W.b0 Nx4C\Gy*0aWbh+O<:$&-ѕ\
'I#WS{q>$o[,gXx.NJm(Bn),
AUrI<50
ם6o`f}.-8(›	)&	܈
b|^%r@[Gx	瓲?`֑>čfAt>糚iD1߼b!z&KゥTAAW0
e~=X}Hu<RK&WO!h1''8ӧscY[we<WEP?Gl>1'4#޿
Zp>SĞ%8eD!䰨=`7fPU=ccږ&$2_<5MRzwɿbM
p.>"|9vaO$Kg^Ef(SR[3_~&0٥I<CbxȽ%`HuլS0Em~ó,:~ Np/b:QxkA3mEZhV/%Sr\*!_krN
@@+@~噼?K8d&D܁X~pyf`PҠt<\[OC0ɧhaNN1</'ԇ#1
	kZ`x	3MQNWOT
'ӮƸ5O
{)si90ӍMEVC}Jdˊ]I\G??v`?IX
ŶJBс;T;Q7HiLW79^-KK9gI0̱^
Oȃo?z$J0>]LCo=$fx5M(Ŗټ&kCFbP44
%
[:IuY

Dr?FvaK\KMDt
k:0XF!
WcPno.tH\&I)?Xǧek$a~?_B{
RjuE$t
w߿\sBA!WjqsZȺΩm/	ep+~XEcmBrٝط|d-wL.@P/,F*,v)FuC[ڠDJ	7C֐')I?23
mSEc5B5IPж}`ql^Q]!pܘe58g
Sr[etE`NBC1iU%tC/98Fns(VI+뛵ZƸ|KZ.^k$O%=RaŦ5o*M^f+5kb*UNk""Q4#_kDp:cv~..4b,LHT`sߤ~@u!uMכk9kTV P=m#?GZJu6~,ʇ6Ӯc0М
:m(}MtkezPߠؙ<Bm
NzF54H^P6)B	
s(-}>F6{Ϩr~jn~nfT"DΏBۜw[&}P+&JrS_{cZWjg6񖍹`KQϯ\fn\bhRlU<=8MxVPkl`;Ni}k>M/7I7Z:S8k;6	eoQ['RgS
m)d2ŗ=aݖ:UY$R_|
>PT%߄Z7RrmӆSsd lJT&#"Uy?;e;SJ*
!zm{2@#tc/jFxDeqn7}|'UAx\v
&^V!#B5۴.,dņ~uinO:. PܢgȬQ3ҟWi*fEfw)O/}FCPhHF.W+p8GH{bDJt&䟋ݬ+R$"mY'J*YEi]34]5k2\MZ`񡦃#ev$j"hCN1c=)#3o
pu^ͅdx)ock
.H3QAI#^XI5z}+j#,{ >Px-{hw%1? ?hbokd
endobj
6 0 obj <<
/Type /Font
/Subtype /Type1
/Encoding 76 0 R
/FirstChar 11
/LastChar 124
/Widths 77 0 R
/BaseFont /YJNJIU+LCMSS8
/FontDescriptor 4 0 R
>> endobj
4 0 obj <<
/Ascent 694
/CapHeight 684
/Descent -139
/FontName /YJNJIU+LCMSS8
/ItalicAngle 0
/StemV 80
/XHeight 521
/FontBBox [-87 -208 1285 857]
/Flags 4
/CharSet (/ff/fi/fl/ffi/quotedblright/quoteright/parenleft/parenright/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/eight/nine/colon/question/A/B/C/D/E/F/G/H/I/K/L/M/N/O/P/Q/R/S/T/U/V/W/Y/quotedblleft/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/endash/emdash)
/FontFile 5 0 R
>> endobj
77 0 obj
[729 660 660 1007 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 625 0 0 0 0 347 486 486 0 0 347 417 347 625 625 625 625 625 625 625 625 625 625 625 347 0 0 0 0 590 0 806 819 799 889 778 743 833 812 319 0 840 708 1021 812 965 785 965 816 694 896 809 806 1153 0 806 0 0 625 0 0 0 0 590 625 556 625 556 382 625 625 278 313 590 278 972 625 625 625 625 417 479 451 625 556 833 556 556 538 625 1250 ]
endobj
76 0 obj <<
/Type /Encoding
/Differences [ 0 /.notdef 11/ff/fi/fl/ffi 15/.notdef 34/quotedblright 35/.notdef 39/quoteright/parenleft/parenright 42/.notdef 44/comma/hyphen/period/slash/zero/one/two/three/four/five/six/seven/eight/nine/colon 59/.notdef 63/question 64/.notdef 65/A/B/C/D/E/F/G/H/I 74/.notdef 75/K/L/M/N/O/P/Q/R/S/T/U/V/W 88/.notdef 89/Y 90/.notdef 92/quotedblleft 93/.notdef 97/a/b/c/d/e/f/g/h/i/j/k/l/m/n/o/p/q/r/s/t/u/v/w/x/y/z/endash/emdash 125/.notdef]
>> endobj
7 0 obj <<
/Type /Pages
/Count 6
/Parent 78 0 R
/Kids [2 0 R 9 0 R 15 0 R 18 0 R 24 0 R 27 0 R]
>> endobj
32 0 obj <<
/Type /Pages
/Count 6
/Parent 78 0 R
/Kids [30 0 R 34 0 R 37 0 R 40 0 R 43 0 R 46 0 R]
>> endobj
51 0 obj <<
/Type /Pages
/Count 6
/Parent 78 0 R
/Kids [49 0 R 53 0 R 56 0 R 59 0 R 65 0 R 68 0 R]
>> endobj
78 0 obj <<
/Type /Pages
/Count 18
/Kids [7 0 R 32 0 R 51 0 R]
>> endobj
79 0 obj <<
/Type /Catalog
/Pages 78 0 R
>> endobj
80 0 obj <<
/Producer (pdfTeX-0.14h)
/Creator (TeX)
/CreationDate (D:20030311010000)
>> endobj
xref
0 81
0000000000 65535 f 
0000000621 00000 n 
0000000517 00000 n 
0000000009 00000 n 
0000044740 00000 n 
0000031833 00000 n 
0000044584 00000 n 
0000046091 00000 n 
0000001348 00000 n 
0000001243 00000 n 
0000000689 00000 n 
0000031425 00000 n 
0000029807 00000 n 
0000031269 00000 n 
0000002309 00000 n 
0000002202 00000 n 
0000001428 00000 n 
0000003121 00000 n 
0000003014 00000 n 
0000002390 00000 n 
0000029119 00000 n 
0000025025 00000 n 
0000028960 00000 n 
0000004025 00000 n 
0000003918 00000 n 
0000003214 00000 n 
0000004780 00000 n 
0000004673 00000 n 
0000004106 00000 n 
0000005635 00000 n 
0000005527 00000 n 
0000004873 00000 n 
0000046197 00000 n 
0000006479 00000 n 
0000006371 00000 n 
0000005728 00000 n 
0000007214 00000 n 
0000007106 00000 n 
0000006560 00000 n 
0000008059 00000 n 
0000007951 00000 n 
0000007295 00000 n 
0000008891 00000 n 
0000008783 00000 n 
0000008140 00000 n 
0000009777 00000 n 
0000009669 00000 n 
0000008972 00000 n 
0000010723 00000 n 
0000010615 00000 n 
0000009858 00000 n 
0000046306 00000 n 
0000011542 00000 n 
0000011434 00000 n 
0000010816 00000 n 
0000012400 00000 n 
0000012292 00000 n 
0000011623 00000 n 
0000013459 00000 n 
0000013351 00000 n 
0000012481 00000 n 
0000023945 00000 n 
0000015182 00000 n 
0000023786 00000 n 
0000014632 00000 n 
0000014524 00000 n 
0000013552 00000 n 
0000015113 00000 n 
0000015005 00000 n 
0000014725 00000 n 
0000024622 00000 n 
0000024278 00000 n 
0000029607 00000 n 
0000029345 00000 n 
0000031719 00000 n 
0000031641 00000 n 
0000045610 00000 n 
0000045207 00000 n 
0000046415 00000 n 
0000046488 00000 n 
0000046539 00000 n 
trailer
<<
/Size 81
/Root 79 0 R
/Info 80 0 R
>>
startxref
46634
%%EOF

--- NEW FILE: slides-fc03.tex ---
\documentclass[landscape]{slides}

\usepackage{color}
\pagecolor{blue}
\color{yellow}

\newif\ifpdf
\ifx\pdfoutput\undefined
   \pdffalse
\else
   \pdfoutput=1
   \pdftrue
\fi

\newenvironment{tightlist}{\begin{list}{$\bullet$}{
  \setlength{\itemsep}{0mm}
    \setlength{\parsep}{0mm}
    %  \setlength{\labelsep}{0mm}
    %  \setlength{\labelwidth}{0mm}
    %  \setlength{\topsep}{0mm}
    }}{\end{list}}

\begin{document}
\ifpdf
  \pdfcompresslevel=9
  \pdfpagewidth=\the\paperwidth
  \pdfpageheight=\the\paperheight
\fi

\newcommand\slidetitle[1]{\begin{center} \huge #1 \end{center}}

\begin{slide}
\begin{center}
\Huge
On the Economics of Anonymity\\
\vspace{1in}
\begin{tabular}{ccc}
\large Alessandro Acquisti &\hspace{.2in} & \large Roger Dingledine \\
\large SIMS, UC Berkeley &\hspace{.2in} & \large The Free Haven Project\\
\\
\multicolumn{3}{c}{\large Paul Syverson} \\  
\multicolumn{3}{c}{\large Naval Research Lab}
\end{tabular}
\end{center}
\end{slide}

\begin{slide}
\large
\slidetitle{Paper overview}
\begin{tightlist}
\item Motivation
\item Social/economic difficulties deploying anonymous (traffic
  analysis resistant) communication systems
\item An economic framework
\item Types of agents
\item Concluding remarks
\end{tightlist}
\end{slide}

\begin{slide}
\large
\slidetitle{Many people need anonymity}
\begin{tightlist}
\item Individuals are tracked and profiled daily. Imagine what they'll
have in your dossier in twenty years.
\item (If that doesn't scare you, think of your kids.)
\item Political dissidents in oppressive countries
\item Governments want to do operations secretly.
\item Corporations are vulnerable to traffic analysis,\\
corporate espionage -- VPNs, encryption don't help
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Anonymity is hard for economic/social reasons too}
\large
\begin{tightlist}
%\item Small space problem --- no place to hide\\
%      often trying to hide among only a few \emph{known} items
\item Anonymity requires \emph{inefficiencies} in computation, bandwidth,
storage
\item Unlike encryption, it's not enough for just one \\person to want
anonymity --- the infrastructure must participate
\end{tightlist}
\end{slide}

\begin{slide}
%\slidetitle{\emph{Other people} provide your anonymity (noise)}
\slidetitle{Hide users with users}
\large
\begin{tightlist}
\item Anonymity systems use messages to hide messages (the more noise,
the more anonymous something in that noise is)
\item Senders are consumers of anonymity, and providers of cover
traffic that creates anonymity for others
\item Users might be better off on crowded systems, even if those systems
have weaker anonymity designs
% \item You're always better off going where the noise is
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{More users is good}
\large
\begin{tightlist}
\item High traffic $\Rightarrow$ better performance (with same anonymity)
\item Better performance $\Rightarrow$ high traffic
\item Attracts more users: faster \emph{and} more anonymous
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{But trust bottlenecks are dangerous}
\large
\begin{tightlist}
\item Nodes with more traffic must be more trusted
\item Adversary can give good service $\Rightarrow$ see more traffic\\
  (and knock down other good providers)
\item Performance and efficiency metrics \emph{cannot} \\distinguish
bad guys from good guys
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Strong anonymity requires distributed trust}
\large
\begin{tightlist}
\item An anonymity system can't be just for one entity\\
  (even a large corporation or government)
\item You must carry traffic for others to protect yourself
\item But those others don't want to trust their traffic to just one
entity either
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Can we fund it by offering service for money?}
\large
\begin{tightlist}
\item Freedom taught us that end-users won't pay enough for strong anonymity
\item (Ok, ok, it's more complicated than that.)
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Can we get volunteers to run nodes?}
\large
\begin{tightlist}
\item Liability, especially for exit nodes
\item Having lots of nodes might work, but \ldots
\item Make an example of a few well-chosen nodes $\Rightarrow$\\ scare
  everybody
\item We can allow nodes to set individual exit policies
\item Remains an open problem
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Pseudospoofing: volunteers are a danger too}
\large
\begin{tightlist}
\item Are half your nodes run by a single bad guy?
\item Global PKI to ensure unique identities? No.
\item Decentralized trust flow algorithms? Not yet.
\item Still a major open problem for dynamic \\decentralized anonymity
systems
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Customization and preferential service are risky}
\large
\begin{tightlist}
\item We'd like to let users pay (or pay more) for\\
  preferential treatment, e.g., stronger anonymity\\ parameters
\item But the hordes in the coach seats are better off anonymity-wise
than those in first class.
\item Those who want first class anonymity have\\ incentive to
  encourage free riders
\end{tightlist}
\end{slide}


\begin{slide}
\slidetitle{It would seem we're screwed}
\large
\begin{tightlist}
\item Inefficiency costs that propagate back to the users chase users away 
\item Usability is a \emph{security} objective: anonymity \\systems are
nothing without users.
\item It's critical that we integrate privacy into the \\systems we use
to interact.
\item But it's hard enough to build a killer app. \\
It's going to be really really hard to solve all the factors at once.
\end{tightlist}
\end{slide}

\begin{slide}
\center{\slidetitle{Multiplayer Strategic agents}}
\begin{tightlist}
\large
\item ``Public good with free-riding'' (Tragedy of the commons)
%  \begin{tightlist}
    \item Collectively agents produce good (anonymity)
    \item But individuals may free ride
%  \end{tightlist}
\item Under which conditions will a system with many players
not implode?
\end{tightlist}

\end{slide}
\begin{slide}
\center\slidetitle{{Highly sensitive agents actually want some level of
    free-riding, to provide noise.}}
\large
\begin{tightlist}
\item  But, just enough free-riding for benefits to dominate costs.
\item Right distribution of valuations $\Rightarrow$ equilibria:\\
  agents with the highest valuations become nodes, others provide
  traffic.
\end{tightlist}
\end{slide}


\begin{slide}
\center{\slidetitle{Alternative node incentive mechanisms}}
\bf{
\begin{tightlist}
\item Usage fee.
\begin{tightlist}
\item Market support for low overhead services (Anonymizer)
\item Inadequate market for strong (high cost) anonymity
\item Hybrid: Larger incentive for high sensitivity volunteers to run nodes
\end{tightlist}
\item Altruistic agents. --- Motivated by public good.
\begin{tightlist}
\item Governments, Public service entities
\end{tightlist}
\item Public rankings and reputation.
\begin{tightlist}
\item High reputation attracts more cover traffic
\item Reward in itself
\end{tightlist}
\end{tightlist}}
\end{slide}





\begin{slide}
\center{\slidetitle{Conclusions}}
\bf{\begin{tightlist}
\item Systems need cover traffic (many low-sensitivity
users) to attract the high-sensitivity users

\item Most users do not want (know they want) anonymity
\begin{tightlist}
\item Weak security (small mix batch, no-delay proxy)\\
$\Rightarrow$  more users\\
 $\Rightarrow$ possibly \emph{stronger} anonymity
\end{tightlist}
%\item Reputation has a complex but critical influence on node
%participation. We must investigate its role more thoroughly.
\item High-sensitivity agents have incentive to run nodes
\begin{tightlist}
\item so they can be certain first node in their routes is trusted
\item to attract cover traffic for their messages
\end{tightlist}

\item There can be an optimal level of free-riding.
\item Key open problem: exit node liability
%\item The deployment of a completely decentralized system might
%involve coordination costs which make it unfeasible. A central
%coordination authority to redistribute payments may be more
%practical.
\end{tightlist}}
\end{slide}

\begin{slide}
\center{\slidetitle{QUESTIONS?}}
\end{slide}

\end{document}



\begin{slide}
\slidetitle{Synchronous systems}
\large
\begin{tightlist}
\item Each message has a deadline by which the node must pass it on
\item Length of paths is fixed, paths might even be public
\item Anonymity is now based on size of batch at widest point, even
for free-route systems
\item Improves flooding/trickle attacks
\item But harder to synchronize, especially for low-latency systems
\end{tightlist}
\end{slide}

\begin{slide}
\slidetitle{Privacy Enhancing Technologies workshop}
\large
\vspace{1in}
\begin{center}
March 26-28, 2003\\
Dresden, Germany\\
http://petworkshop.org/
\end{center}
\end{slide}


\begin{slide}
\begin{center}
An example: Directory servers
\end{center}
\begin{tightlist}
\item Distribute location, capabilities, key info, performance stats
\item A single directory server is a point of failure
\item Redundant directory servers: must be (provably!) \\synchronized to avoid
partitioning attacks
\item Can distinguish between clients that use static lists and clients
that update frequently
\end{tightlist}
\end{slide}

\begin{slide}
\begin{center}
Directory servers (2)
\end{center}
\begin{tightlist}
\item But even if client information is uniform, nodes can still do trickle
attack: hold message until other clients have different information.
\item Introducing reputation means adversary has new avenue to manipulate
client information
\item Tension between giving clients accurate timely information, and
preventing adversary from manipulating client behavior
\end{tightlist}
\end{slide}

\begin{slide}
\begin{center}
\slidetitle{Myopic agent: Simplest Case}
\end{center}
\begin{tightlist}
\item Does not consider affect of its choices on others

\item High sensitivity agents: probably
accept trade-off of becoming nodes because they risk a lot by
losing their anonymity, and because acting as nodes significantly
increases their probabilities of remaining anonymous.

\item Very Low sensitivity agents: cost/hassle of sending through the
system too high.

\end{tightlist}
\end{slide}

\begin{slide}
\begin{center}
\slidetitle{Strategic agents, one-on-one interactions}
\end{center}
\begin{tightlist}

\item Considerations:
  \begin{tightlist}
  \item Other agent's sensitivity known (high?  low?) or unknown
  \item Other agent honest or dishonest.
  \end{tightlist}
\item This case can have equilibria with free-riding even when the
  other agent's type is unknown (cf. Palfrey and Rosenthal 1989).
\item If you care enough about your anonymity, you may not care if the
  other guy freeloads.
\end{tightlist}
\end{slide}


\begin{slide}
\begin{center}
\slidetitle{The model}
\end{center}
\begin{tightlist}


\item Agent tries to minimize costs of sending messages and
risks of being tracked. Consider:

\item $$
u_{i}=-v_{a_{i}}\left( 1-p_{a}\left(
n_{s},n_{h},n_{d},a_{i}^{h}\right) \right)
-c_{s}a_{i}^{s}-c_{h}\left( n_{s},n_{h},n_{d}\right)
a_{i}^{h}-c_{n}
$$

\item Prob(anonymity loss, given number senders, honest/dishonest
nodes, and self action), weighted by disutility $v_{a_i}$ of message exposure.

\item Costs of sending a message through the mix-net, acting as a
node given $n_{s}$ agents sending messages over $n_{h}$
and $n_{d}$ nodes, and sending messages through a non-anonymous
system.

\item Each period, a rational agent can compare the utility coming
from each of these three one-period strategies: \textbf{1) act as
node, 2) act as user, 3) do not send message}.


\end{tightlist}
\end{slide}

Cut stuff about dummy traffic

Free riding is a good thing and that's economically unique

Cut types of agents discussion


\begin{slide}
\slidetitle{Need to manage incentives}
\large
\begin{tightlist}
\item Users have incentive to run a node, to get more anonymity. That's
a good start.
%\item Equilibrium where people who need anonymity run nodes, and others
%use them?
%\item Dummy traffic can help maintain anonymity -- but why should others
%send dummy traffic to help \emph{your} anonymity?
\item If anonymity for all requires each user doing similar things,
how do we deal with users who don't want as much anonymity?
\end{tightlist}
\end{slide}



\begin{slide}
\begin{center}
\slidetitle{Analytic Framework}
\end{center}
\begin{tightlist}
\item Agents are \emph{sensitive} to (value) their anonymity.
\item Sensitivity to anonymity randomly distributed in model.
\item Focus is sender (communication initiator) anonymity.

\item Agents' actions:

\begin{enumerate}
\item  Act as a sender of the anonymity system.

\item  Act as an honest node.

\item  Act as a dishonest node.

\item  Send ``around'' anonymity system or simply do not send.

\end{enumerate}
\end{tightlist}
\end{slide}

\begin{slide}
\begin{center}
\slidetitle{Benefit from Sending through Anonymity System}
\end{center}
\begin{tightlist}
\item Anonymous sending benefit, function of:

\begin{itemize}
\item  Subjective value of agent for message arriving.

\item Subjective value of keeping her identity anonymous.

\item The perceived level of anonymity in the system.

\item The perceived level of reliability in the system.
  
\item Factors affecting above probabilities: number of senders in
  system, number acting as a node, honest, dishonest, etc.


\end{itemize}
\end{tightlist}
\end{slide}

\begin{slide}
\begin{center}
\slidetitle{Benefit from Being a Node}
\end{center}
\begin{tightlist}
\item Trusted entry point to anonymity system.
\item Better cover traffic for your own messages.
\item Possible reward (money or service) for carrying traffic/making dummies.
\color{magenta}
\item  \color{yellow} Being a bad node \color{magenta}
  \begin{tightlist}
       \item \color{yellow}Disrupt service. \color{magenta}
       \item \color{yellow}Obtain traffic information. \color{magenta}
       \item \color{yellow}Possibly reward without providing service.
         \color{magenta}
  \end{tightlist}
\end{tightlist}
\end{slide}

\begin{slide}
\begin{center}
\slidetitle{Costs of Using Anonymity System}
\end{center}
\begin{tightlist}

\item  Sender cost:\\
Delay/uncertainty of delivery vs.\ sending directly to recipient


\item  Node cost:
  \begin{tightlist}
     \item Bandwidth/processing cost of providing service
     \item Possible cost of creating/sending/terminating dummies
     \item Liability for actions using your node
         (especially exiting).  \color{magenta}
     \item  \color{yellow}Dishonest node: penalty if caught. (expulsion, legal, etc.)
  \end{tightlist}
\end{tightlist}
\end{slide}


\begin{slide}
\center{\slidetitle{Types of Agents}}
\large
\begin{tightlist}
\item Myopic agent: Does not consider affect of its choices on others
\item One-on-one Strategic Agent
\item Multiplayer Strategic Agent
\end{tightlist}
\end{slide}


\begin{slide}
\slidetitle{Conclusion 2: more research remains}
\large
\begin{tightlist}
\item Our current directions aren't going to work, from an incentive and
usability perspective. We need to rethink.
\end{tightlist}
\end{slide}


%Some slides added by Alex

\begin{slide}
\center{\slidetitle{Major Open Problem}}
\bf{\begin{tightlist}

\item Cuts both ways
\begin{tightlist}
\item Liability for traffic leaving anonymity system at your node
\item Hiding/deniability of traffic entering anonymity system at your node
\end{tightlist}
\item Legal climate depends on (perception of) technology
\item YACEP:
\begin{tightlist}
\item Widespread anonymous system use\\
  $\Rightarrow$?   Deniability for traffic leaving system at your node\\
  $\Rightarrow$? Widespread anonymous system availability
\end{tightlist}
\end{tightlist}}
\end{slide}


***********************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe freehaven-cvs       in the body. http://freehaven.net/