tor-webwml.git

1) ## translation metadata

2) # Revision: $Revision$

3) # Translation-Priority: 4-optional
4) 
5) #include "head.wmi" TITLE="Tor: Volunteer" CHARSET="UTF-8"
6) <div id="content" class="clearfix">
7)   <div id="breadcrumbs">

8)     <a href="<page index>">Home &raquo; </a>

9)     <a href="<page getinvolved/volunteer>">Volunteer</a>
10)   </div>

11)   <div id="maincol">

12)     <!-- PUT CONTENT AFTER THIS TAG -->
13)     <h1>A few things everyone can do now:</h1>
14)     <ol>
15)     <li>Please consider <a href="<page docs/tor-doc-relay>">running
16)     a relay</a> to help the Tor network grow.</li>

17)     <li>Do you have an Amazon account? Are you willing to spend up to $3 a

18)     month? Then spin up your own Tor <a href="<page
19)     docs/bridges>">bridge</a> in less than 10 minutes with <a

20)     href="https://cloud.torproject.org/">tor cloud</a>!</li>

21)     <li>Tell your friends! Get them to run relays. Get them to run hidden
22)     services. Get them to tell their friends.</li>
23)     <li>If you like Tor's goals, please <a href="<page donate/donate>">take a moment
24)     to donate to support further Tor development</a>. We're also looking
25)     for more sponsors &mdash; if you know any companies, NGOs, agencies,
26)     or other organizations that want anonymity / privacy / communications
27)     security, let them know about us.</li>
28)     <li>We're looking for more <a href="<page about/torusers>">good examples of Tor
29)     users and Tor use cases</a>. If you use Tor for a scenario or purpose not
30)     yet described on that page, and you're comfortable sharing it with us,
31)     we'd love to hear from you.</li>
32)     </ol>

33) 

34)     <a id="Documentation"></a>
35)     <h2><a class="anchor" href="#Documentation">Documentation</a></h2>
36)     <ol>

37)     <li>Help translate the
38) <!-- web page and -->
39)     documentation into other

40)     languages. See the <a href="<page getinvolved/translation>">translation
41)     guidelines</a> if you want to help out. We especially need Arabic or
42)     Farsi translations, for the many Tor users in censored areas.</li>
43)     <li>Evaluate and document

44)     <a href="<wiki>doc/TorifyHOWTO">our

45)     list of programs</a> that can be configured to use Tor.</li>
46)     <li>We have a huge list of <a

47)     href="<wiki>doc/SupportPrograms">potentially useful

48)     programs that interface to Tor</a>. Which ones are useful in which
49)     situations? Please help us test them out and document your results.</li>
50)     </ol>

51) 

52)     <a id="Advocacy"></a>
53)     <h2><a class="anchor" href="#Advocacy">Advocacy</a></h2>
54)     <ol>

55)     <li>Create a presentation that can be used for various user group
56) meetings around the world.</li>

57)     <li>Create a video about the positive uses of Tor, what Tor is,
58)     or how to use it.  Some have already started on <a

59)     href="https://media.torproject.org/video/">Tor's Media server</a>,

60)     <a

61)     href="http://www.howcast.com/videos/90601-How-To-Circumvent-an-Internet-Proxy">Howcast</a>,

62)     and <a href="http://www.youtube.com/thetorproject">YouTube</a>.</li>

63)     <li>Create a poster, or a set of posters, around a theme,

64)     such as "Tor for Freedom!"</li>
65)     <li>Create a t-shirt design that incorporates "<a
66)     href="https://check.torproject.org/">Congratulations! You are using
67)     Tor!</a>" in any language.</li>

68)     </ol>

69) 

70)     <a id="Projects"></a>
71)     <h2><a class="anchor" href="#Projects">Projects</a></h2>

72) 

73)     <p>
74)     Below are a list of Tor related projects we're developing and/or
75)     maintaining. Most discussions happen on IRC so if you're interested in any
76)     of these (or you have a project idea of your own), then please <a

77)     href="<page about/contact>#irc">join us in #tor-dev</a>. Don't be shy
78)     to ask questions, and don't hesitate to ask even if the main contributors
79)     aren't active at that moment.

80)     </p>

81) 

82)     <table id="projects">
83)       <tr>
84)         <th>Name</th>
85)         <th>Category</th>
86)         <th>Language</th>
87)         <th>Activity</th>
88)         <th>Contributors</th>
89)       </tr>

90) 

91)       <tr>
92)         <td><a href="#project-tor">Tor</a></td>
93)         <td>Core</td>
94)         <td>C</td>
95)         <td>Heavy</td>

96)         <td>nickm, athena, arma</td>

97)       </tr>

98) 

99)       <tr class="alt">

100)         <td><a href="#project-torbrowser">Tor Browser</a></td>

101)         <td>Bundle</td>

102)         <td>C, Scripting</td>

103)         <td>Moderate</td>

104)         <td>mikeperry, Erinn</td>

105)       </tr>

106) 

107)       <tr>

108)         <td><a href="#project-torbutton">Torbutton</a></td>
109)         <td>Browser Add-on</td>
110)         <td>Javascript</td>
111)         <td>Moderate</td>
112)         <td>mikeperry</td>

113)       </tr>

114) 

115)       <tr class="alt">

116)         <td><a href="#project-httpseverywhere">HTTPS Everywhere</a></td>
117)         <td>Browser Add-on</td>
118)         <td>Javascript</td>
119)         <td>Moderate</td>
120)         <td>pde, mikeperry</td>

121)       </tr>

122) 

123)       <tr>

124)         <td><a href="#project-vidalia">Vidalia</a></td>
125)         <td>User Interface</td>
126)         <td>C++, Qt</td>

127)         <td>None</td>

128)         <td>chiiph</td>
129)       </tr>

130) 

131)       <tr class="alt">

132)         <td><a href="#project-arm">Arm</a></td>
133)         <td>User Interface</td>
134)         <td>Python, Curses</td>

135)         <td>Light</td>

136)         <td>atagar</td>
137)       </tr>

138) 

139)       <tr>

140)         <td><a href="#project-orbot">Orbot</a></td>
141)         <td>User Interface</td>
142)         <td>Java</td>

143)         <td>Light</td>

144)         <td>n8fr8</td>

145)       </tr>

146) 

147)       <tr class="alt">

148)         <td><a href="#project-tails">Tails</a></td>
149)         <td>OS image</td>
150)         <td>Sys Admin</td>
151)         <td>Heavy</td>
152)         <td><a href="https://tails.boum.org/">#tails</a></td>

153)       </tr>

154) 

155)       <tr>

156)         <td><a href="#project-torramdisk">tor-ramdisk</a></td>
157)         <td>OS image</td>
158)         <td>Sys Admin</td>
159)         <td>Light</td>
160)         <td>blueness</td>
161)       </tr>
162) 

163)       <tr class="alt">

164)         <td>*<a href="#project-torouter">Torouter</a></td>
165)         <td>OS image</td>
166)         <td>Sys Admin</td>

167)         <td>None</td>

168)         <td>ioerror</td>
169)       </tr>
170) 

171)       <tr>

172)         <td><a href="#project-torsocks">Torsocks</a></td>
173)         <td>Usability</td>
174)         <td>C</td>

175)         <td>Light</td>

176)         <td>ioerror, nickm</td>
177)       </tr>
178) 

179)       <tr class="alt">

180)         <td><a href="#project-torbirdy">TorBirdy</a></td>
181)         <td>Browser Add-on</td>
182)         <td>JavaScript</td>
183)         <td>Heavy</td>
184)         <td>Sukhbir (sukhe), ioerror</td>
185)       </tr>

186) 

187)       <tr>

188)         <td><a href="#project-obfsproxy">Obfsproxy</a></td>
189)         <td>Client Add-on</td>

190)         <td>C, Python</td>
191)         <td>Moderate</td>
192)         <td>asn, nickm</td>

193)       </tr>
194) 

195)       <tr class="alt">

196)         <td><a href="#project-flash-proxy">Flash Proxy</a></td>
197)         <td>Client Add-on</td>
198)         <td>Python, JavaScript, Go</td>

199)         <td>Heavy</td>

200)         <td>dcf, aallai, jct</td>
201)       </tr>

202) 

203)       <tr>

204)         <td>*<a href="#project-thandy">Thandy</a></td>
205)         <td>Updater</td>
206)         <td>Python</td>

207)         <td>None</td>

208)         <td>chiiph, Erinn, nickm</td>

209)       </tr>

210) 

211)       <tr class="alt">

212)         <td><a href="#project-shadow">Shadow</a></td>

213)         <td>Simulator</td>

214)         <td>C, Python</td>

215)         <td>Heavy</td>

216)         <td>robgjansen</td>
217)       </tr>

218) 

219)       <tr>

220)         <td><a href="#project-stem">Stem</a></td>

221)         <td>Library</td>
222)         <td>Python</td>
223)         <td>Heavy</td>

224)         <td>atagar</td>

225)       </tr>

226) 

227)       <tr class="alt">

228)         <td><a href="#project-txtorcon">Txtorcon</a></td>
229)         <td>Library</td>
230)         <td>Python, Twisted</td>

231)         <td>Heavy</td>

232)         <td>meejah</td>
233)       </tr>

234) 

235)       <tr>

236)         <td><a href="#project-tlsdate">Tlsdate</a></td>
237)         <td>Utility</td>
238)         <td>C</td>

239)         <td>Moderate</td>

240)         <td>ioerror</td>
241)       </tr>

242) 

243)       <tr class="alt">

244)         <td><a href="#project-metrics">Metrics</a></td>
245)         <td>Client Service</td>
246)         <td>Java</td>
247)         <td>Heavy</td>
248)         <td>karsten</td>
249)       </tr>

250) 

251)       <tr>

252)         <td><a href="#project-atlas">Atlas</a></td>
253)         <td>Client Service</td>
254)         <td>JavaScript</td>

255)         <td>Light</td>

256)         <td>hellais, karsten</td>

257)       </tr>

258) 

259)       <tr class="alt">

260)         <td><a href="#project-torstatus">TorStatus</a></td>
261)         <td>Client Service</td>

262)         <td>Python, Django</td>

263)         <td>None</td>
264)         <td></td>
265)       </tr>

266) 

267)       <tr>

268)         <td><a href="#project-compass">Compass</a></td>
269)         <td>Client Service</td>
270)         <td>Python</td>

271)         <td>Light</td>

272)         <td>gsathya, karsten, cwacek</td>

273)       </tr>

274) 

275)       <tr class="alt">

276)         <td><a href="#project-onionoo">Onionoo</a></td>
277)         <td>Backend Service</td>
278)         <td>Java, Python</td>

279)         <td>Light</td>

280)         <td>karsten, gsathya</td>
281)       </tr>
282) 

283)       <tr>

284)         <td><a href="#project-weather">Weather</a></td>
285)         <td>Client Service</td>
286)         <td>Python</td>

287)         <td>None</td>

288)         <td>kaner</td>
289)       </tr>

290) 

291)       <tr class="alt">

292)         <td><a href="#project-gettor">GetTor</a></td>
293)         <td>Client Service</td>
294)         <td>Python</td>

295)         <td>None</td>

296)         <td>kaner</td>

297)       </tr>

298) 

299)       <tr>

300)         <td><a href="#project-torcheck">TorCheck</a></td>
301)         <td>Client Service</td>
302)         <td>Python, Perl</td>
303)         <td>None</td>

304)         <td>ioerror</td>

305)       </tr>

306) 

307)       <tr class="alt">

308)         <td><a href="#project-bridgedb">BridgeDB</a></td>
309)         <td>Backend Service</td>
310)         <td>Python</td>

311)         <td>None</td>

312)         <td>kaner, nickm</td>

313)       </tr>

314) 

315)       <tr>

316)         <td><a href="#project-ooni-probe">Ooni Probe</a></td>

317)         <td>Scanner</td>
318)         <td>Python</td>

319)         <td>Heavy</td>

320)         <td>hellais, isis, ioerror</td>
321)       </tr>
322) 

323)       <tr class="alt">

324)         <td><a href="#project-torflow">TorFlow</a></td>
325)         <td>Backend Service</td>
326)         <td>Python</td>

327)         <td>None</td>

328)         <td>aagbsn, mikeperry</td>

329)       </tr>

330) 

331)       <tr>

332)         <td>*<a href="#project-torbel">TorBEL</a></td>
333)         <td>Backend Service</td>
334)         <td>Python</td>
335)         <td>None</td>
336)         <td>Sebastian</td>
337)       </tr>

338) 

339)       <tr class="alt">

340)         <td><a href="#project-tor2web">Tor2web</a></td>
341)         <td>Client Service</td>

342)         <td>Python</td>

343)         <td>Moderate</td>

344)         <td>hellais</td>

345)       </tr>

346) 

347)       <tr>

348)         <td><a href="#project-anonbib">Anonbib</a></td>
349)         <td>Website</td>

350)         <td>Python</td>

351)         <td>Light</td>
352)         <td>arma, nickm</td>
353)       </tr>
354) 

355)     </table>

356) 

357)     <sub>
358)     * Project is still in an alpha state.
359)     </sub>

360) 

361)     <br /><br />

362) 

363)     <a id="project-tor"></a>
364)     <h3>Tor (<a href="https://gitweb.torproject.org/tor.git">code</a>, <a

365)     href="https://trac.torproject.org/projects/tor/report/12">bug

366)     tracker</a>)</h3>

367) 

368)     <p>
369)     Central project, providing the core software for using and participating in
370)     the Tor network. Numerous people contribute to the project to varying
371)     extents, but the chief architects are Nick Mathewson and Roger Dingledine.
372)     </p>

373) 

374)     <p>
375)     <b>Project Ideas:</b><br />

376)     <i><a href="#limitCapabilities">Run With Limited Capabilities</a></i><br />
377)     <i><a href="#torCleanup">Tor Codebase Cleanup</a></i><br />

378)     <i><a href="#httpsImpersonation">HTTPS Server Impersonation</a></i><br />

379)     <i><a href="#chutneyExpansion">Make Chutney Do More, More Reliably</a></i>

380)     </p>

381) 

382)     <!--

383)     <a id="project-jtor"></a>
384)     <h3><a href="https://github.com/brl/JTor/wiki">JTor</a> (<a

385)     href="https://gitweb.torproject.org/jtor.git">code</a>, <a

386)     href="https://github.com/brl/JTor/issues">bug
387)     tracker</a>)</h3>

388) 

389)     <p>
390)     Java implementation of Tor and successor to <a
391)     href="http://onioncoffee.sourceforge.net/">OnionCoffee</a>. This project
392)     isn't yet complete, and has been inactive since Fall 2010.
393)     </p>

394)     -->

395) 

396)     <a id="project-torbrowser"></a>

397)     <h3><a href="<page projects/torbrowser>">Tor Browser Bundle</a> (<a
398)     href="https://gitweb.torproject.org/torbrowser.git">code</a>, <a
399)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Tor+bundles/installation&order=priority">bug

400)     tracker</a>, <a href="https://www.torproject.org/projects/torbrowser/design/">design doc</a>)</h3>
401) 

402)     <p>
403)     The Tor Browser Bundle is an easy-to-use portable package of Tor, Vidalia,

404)     Torbutton, and a Firefox fork preconfigured to work together out of

405)     the box. It contains a modified copy of Firefox that aims to resolve the
406)     privacy and security issues in mainline version.

407)     </p>

408) 

409)     <!--

410)     <p>
411)     <b>Project Ideas:</b><br />
412)     <i><a href="#auditTBB">Audit Tor Browser Bundles for data leaks</a></i><br />
413)     <i><a href="#usabilityTesting">Usability testing of Tor</a></i>
414)     </p>

415)     -->

416) 
417)     <a id="project-torbutton"></a>
418)     <h3><a href="<page torbutton/index>">Torbutton</a> (<a
419)     href="https://gitweb.torproject.org/torbutton.git">code</a>, <a
420)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Torbutton&order=priority">bug

421)     tracker</a>)</h3>

422) 

423)     <p>

424)     Firefox addon that addresses many of the client-side threats to browsing
425)     the Internet anonymously. Mike has since continued to adapt it to new
426)     threats, updated versions of Firefox, and possibly <a
427)     href="https://blog.torproject.org/blog/google-chrome-incognito-mode-tor-and-fingerprinting">Chrome
428)     as well</a>.

429)     </p>

430) 
431)     <a id="project-httpseverywhere"></a>
432)     <h3><a href="https://www.eff.org/https-everywhere">HTTPS Everywhere</a> (<a
433)     href="https://gitweb.torproject.org/https-everywhere.git">code</a>, <a
434)     href="https://trac.torproject.org/projects/tor/report/19">bug

435)     tracker</a>)</h3>

436) 

437)     <p>

438)     HTTPS Everywhere is a Firefox and Chrome extension that encrypts
439)     your communications with many major websites, making your browsing
440)     more secure.

441)     </p>

442) 

443)     <a id="project-vidalia"></a>
444)     <h3><a href="<page projects/vidalia>">Vidalia</a> (<a

445)     href="https://gitweb.torproject.org/vidalia.git">code</a>, <a

446)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Vidalia&order=priority">bug
447)     tracker</a>)</h3>

448) 

449)     <p>
450)     The most commonly used user interface for Tor. Matt Edman started the
451)     project in 2006 and brought it to its current stable state. Development

452)     slowed for several years, though Tomás Touceda has since taken the
453)     lead with pushing the project forward.

454)     </p>

455) 

456)     <!--

457)     <p>
458)     <b>Project Ideas:</b><br />

459)     <i><a href="#vidaliaStatusEventInterface">Tor Controller Status Event Interface for Vidalia</a></i><br />
460)     <i><a href="#vidalia-hidden-service-panel">Torrc plugin and improved hidden service configuration panel</a></i>

461)     </p>

462)     -->

463) 

464)     <a id="project-arm"></a>
465)     <h3><a href="http://www.atagar.com/arm/">Arm</a> (<a

466)     href="https://gitweb.torproject.org/arm.git">code</a>, <a

467)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=arm&order=priority">bug
468)     tracker</a>)</h3>

469) 

470)     <p>

471)     The anonymizing relay monitor (arm) is a terminal status monitor for Tor,
472)     intended for command-line aficionados, ssh connections, and anyone with a
473)     tty terminal. This works much like top does for system usage, providing
474)     real time statistics for bandwidth, resource usage, connections, and quite
475)     a bit more.

476)     </p>

477) 

478)     <!--

479)     <p>
480)     <b>Project Ideas:</b><br />

481)     <i><a href="#armClientMode">Client Mode Use Cases for Arm</a></i><br />
482)     <i><a href="#armGui">GUI for Arm</a></i>

483)     </p>

484)     -->

485) 

486)     <a id="project-orbot"></a>
487)     <h3><a href="https://guardianproject.info/apps/orbot/">Orbot</a> (<a

488)     href="https://gitweb.torproject.org/orbot.git">code</a>, <a

489)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Orbot&order=priority">bug
490)     tracker</a>)</h3>

491) 
492)     <p>
493)     Provides Tor on the Android platform. This was under very active
494)     development up through Fall 2010, after which things have been quiet.
495)     </p>
496) 
497)     <!--
498)     <p>
499)     <b>Project Ideas:</b><br />
500)     <i><a href="#orbot-torbutton">TorButton for Mobile Firefox 4 or Custom Browser on Android</a></i><br />
501)     <i><a href="#orbot-userInterface">Build a better user interface for Orbot</a></i><br />
502)     <i><a href="#orbot-optimisation">Core Tor mobile optimisation</a></i>
503)     </p>
504)     -->
505) 
506)     <a id="project-tails"></a>
507)     <h3><a href="https://tails.boum.org/">The Amnesic Incognito Live System</a> (<a
508)     href="http://git.immerda.ch/?p=amnesia.git;a=summary">code</a>, <a
509)     href="https://tails.boum.org/bugs/">bug
510)     tracker</a>)</h3>
511) 
512)     <p>
513)     The Amnesic Incognito Live System is a live CD/USB distribution
514)     preconfigured so that everything is safely routed through Tor and leaves no
515)     trace on the local system. This is a merger of the Amnesia and <a
516)     href="http://www.anonymityanywhere.com/incognito/">Incognito</a> projects,
517)     and still under very active development.
518)     </p>
519) 
520)     <!--
521)     <p>
522)     <b>Project Ideas:</b><br />
523)     <i><a href="#tailsHiddenServicePetnames">Petname system for Tor hidden
524)     services</a></i><br />
525)     <i><a href="#tailsServer">Tails server: Self-hosted services behind
526)     Tails-powered Tor hidden services</a></i>
527)     </p>
528)     -->
529) 

530)     <a id="project-torramdisk"></a>
531)     <h3><a href="http://opensource.dyc.edu/tor-ramdisk">Tor-ramdisk</a> (<a

532)     href="http://opensource.dyc.edu/gitweb/?p=tor-ramdisk.git;a=summary">code</a>, <a
533)     href="http://opensource.dyc.edu/tor-ramdisk-documentation">documentation</a>)</h3>

534) 
535)     <p>

536)     Tor-ramdisk is a uClibc-based micro Linux distribution whose sole
537)     purpose is to securely host a Tor server purely in RAM.

538)     </p>
539) 

540)     <a id="project-torouter"></a>
541)     <h3><a
542)     href="<wiki>doc/Torouter">Torouter</a> (<a

543)     href="https://gitweb.torproject.org/torouter.git">code</a>, <a

544)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Torouter&order=priority">bug
545)     tracker</a>)</h3>
546) 
547)     <p>
548)     Project to provide an easy-to-use, embedded Tor instance for routers. This
549)     had high activity in late 2010, but has since been rather quiet.
550)     </p>
551) 
552)     <a id="project-torsocks"></a>
553)     <h3><a href="http://code.google.com/p/torsocks/">Torsocks</a> (<a
554)     href="https://gitweb.torproject.org/torsocks.git">code</a>, <a
555)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Torify&order=priority">bug
556)     tracker</a>)</h3>
557) 

558)     <p>

559)     Utility for adapting other applications to work with Tor. Development has
560)     slowed and compatibility issues remain with some platforms, but it's
561)     otherwise feature complete.

562)     </p>

563) 

564)     <!--

565)     <p>
566)     <b>Project Ideas:</b><br />

567)     <i><a href="#torsocksForOSX">Make torsocks/dsocks work on OS X</a></i>

568)     </p>

569)     -->

570) 

571)     <a id="project-torbirdy"></a>
572)     <h3>TorBirdy (<a
573)     href="https://github.com/ioerror/torbirdy">code</a>, <a
574)     href="https://trac.torproject.org/projects/tor/wiki/torbirdy/dev">bug
575)     tracker</a>)</h3>

576) 

577)     <p>
578)     TorBirdy is Torbutton for Thunderbird and related Mozilla mail clients.
579)     </p>

580) 

581)     <a id="project-obfsproxy"></a>

582)     <h3><a href="<page projects/obfsproxy>">Obfsproxy</a> (<a

583)     href="https://gitweb.torproject.org/obfsproxy.git">C codebase</a>, <a
584)     href="https://gitweb.torproject.org/user/asn/pyobfsproxy.git">python
585)     codebase</a>, <a

586)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Pluggable+transport&order=priority">bug
587)     tracker</a>)</h3>

588) 

589)     <p>
590)     A proxy that shapes Tor traffic, making it harder for censors to detect and

591)     block Tor. This has both a C and python implementation.

592)     </p>

593) 

594)     <!--

595)     <p>
596)     <b>Project Ideas:</b><br />
597)     <i><a href="#obfsproxy-new-transports">New and innovative pluggable transports</a></i><br />
598)     <i><a href="#obfsproxy-scanning-measures">Defensive bridge active scanning measures</a></i><br />
599)     <i><a href="#obfsproxy-fuzzer">Fuzzer for the Tor protocol</a></i>
600)     </p>

601)     -->

602) 

603)     <a id="project-flash-proxy"></a>
604)     <h3><a href="https://crypto.stanford.edu/flashproxy/">Flash Proxy</a> (<a
605)     href="https://gitweb.torproject.org/flashproxy.git">code</a>, <a
606)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_information&status=needs_review&status=needs_revision&status=new&status=reopened&component=Flashproxy">bug
607)     tracker</a>)</h3>

608) 

609)     <p>
610)     Pluggable transport using proxies running in web browsers to defeat
611)     address-based blocking.
612)     </p>

613) 

614)     <a id="project-thandy"></a>
615)     <h3>Thandy (<a

616)     href="https://gitweb.torproject.org/thandy.git">code</a>)</h3>

617) 

618)     <p>

619)     Updater for Tor. The project <a
620)     href="http://google-opensource.blogspot.com/2009/03/thandy-secure-update-for-tor.html">began

621)     in the Summer of 2008</a> and produced a functional secure updater. The
622)     remaining steps are to deploy a Thandy repository, and integrate Thandy
623)     into one of our bundles &mdash; which requires UI changes in the bundles,
624)     Python packaging on various platforms, etc. Later interest in it was

625)     rekindled and many aspects of <a
626)     href="http://freehaven.net/~arma/tuf-ccs2010.pdf">its design</a> (including
627)     the language it'll be in) are currently in flux. This is related to <a
628)     href="https://www.updateframework.com/">TUF</a>.

629)     </p>

630) 

631)     <a id="project-shadow"></a>
632)     <h3><a href="https://shadow.cs.umn.edu/">Shadow</a> (<a
633)     href="https://github.com/shadow">code</a>, <a
634)     href="https://github.com/shadow/shadow/issues">bug
635)     tracker</a>)</h3>

636) 

637)     <p>
638)     Shadow is a discrete-event network simulator that runs the real
639)     Tor software as a plug-in. Shadow is open-source software that enables
640)     accurate, efficient, controlled, and repeatable Tor experimentation.

641)     For another simulator, see <a
642)     href="http://crysp.uwaterloo.ca/software/exptor/">ExperimenTor</a>.

643)     </p>

644) 

645)     <a id="project-stem"></a>

646)     <h3><a href="https://stem.torproject.org/">Stem</a> (<a

647)     href="https://gitweb.torproject.org/stem.git">code</a>, <a

648)     href="https://trac.torproject.org/projects/tor/wiki/doc/stem/bugs">bug

649)     tracker</a>)</h3>

650) 

651)     <p>

652)     Python controller library for scripts and controller applications using
653)     Tor.

654)     </p>

655) 

656)     <p>
657)     <b>Project Ideas:</b><br />

658)     <i><a href="#txtorcon-stemIntegration">Txtorcon/Stem Integration</a></i><br />

659)     <i><a href="#stemUsability">Stem Usability and Porting</a></i><br />
660)     <i><a href="#stemTestingForTor">Stem Tests for Tor</a></i>

661)     </p>

662) 

663)     <a id="project-txtorcon"></a>
664)     <h3>Txtorcon (<a
665)     href="https://github.com/meejah/txtorcon">code</a>, <a
666)     href="https://txtorcon.readthedocs.org">docs</a>)</h3>

667) 

668)     <p>
669)     Twisted-based asynchronous Tor control protocol implementation. Includes
670)     unit-tests, examples, state-tracking code and configuration abstraction.
671)     Used by OONI and APAF.
672)     </p>

673) 

674)     <p>
675)     <b>Project Ideas:</b><br />
676)     <i><a href="#txtorcon-stemIntegration">Txtorcon/Stem Integration</a></i>
677)     </p>
678) 

679)     <a id="project-tlsdate"></a>
680)     <h3>Tlsdate (<a href="https://github.com/ioerror/tlsdate">code</a>)</h3>

681) 

682)     <p>
683)     tlsdate: secure parasitic rdate replacement
684)     </p>

685) 

686)     <p>
687)     tlsdate sets the local clock by securely connecting with TLS to remote
688)     servers and extracting the remote time out of the secure handshake. Unlike
689)     ntpdate, tlsdate uses TCP, for instance connecting to a remote HTTPS or TLS
690)     enabled service, and provides some protection against adversaries that try
691)     to feed you malicious time information.
692)     </p>

693) 

694)     <a id="project-metrics"></a>
695)     <h3><a href="https://metrics.torproject.org/">Metrics</a> (code: <a
696)     href="https://gitweb.torproject.org/metrics-db.git">db</a>, <a
697)     href="https://gitweb.torproject.org/metrics-utils.git">utils</a>, <a
698)     href="https://gitweb.torproject.org/metrics-web.git">web</a>, <a
699)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Metrics&order=priority">bug
700)     tracker</a>)</h3>

701) 

702)     <p>
703)     Processing and analytics of consensus data, provided to users via the
704)     metrics portal. This has been under active development for several years by

705)     Karsten Loesing. See also <a
706)     href="https://gitweb.torproject.org/torperf.git">TorPerf</a>.

707)     </p>

708) 

709)     <p>
710)     <b>Project Ideas:</b><br />
711)     <i><a href="#metricsSearch">Searchable Tor descriptor and Metrics data archive</a></i> (Python/Django?)
712)     </p>

713) 

714)     <a id="project-atlas"></a>
715)     <h3><a href="https://atlas.torproject.org/">Atlas</a> (<a
716)     href="https://gitweb.torproject.org/atlas.git">code</a>)</h3>

717) 

718)     <p>
719)     Atlas is a web application to discover Tor relays and bridges. It provides
720)     useful information on how relays are configured along with graphics about
721)     their past usage. This is the third evolution of the TorStatus application.
722)     </p>

723) 

724)     <a id="project-torstatus"></a>

725)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/org/roadmaps/TorStatus">TorStatus</a> (<a

726)     href="https://gitweb.torproject.org/torstatus.git">code</a>)</h3>

727) 

728)     <p>
729)     Portal providing an overview of the Tor network, and details on any of its
730)     current relays. Though very actively used, this project has been

731)     unmaintained for a long while. The <a
732)     href="https://svn.torproject.org/svn/torstatus/trunk/">original
733)     codebase</a> was written in PHP, and students from Wesleyan wrote the new
734)     Django counterpart.

735)     </p>

736) 

737)     <a id="project-compass"></a>
738)     <h3><a href="https://compass.torproject.org/">Compass</a> (<a
739)     href="https://gitweb.torproject.org/compass.git">code</a>, <a
740)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Compass&order=priority">bug
741)     tracker</a>)</h3>

742) 

743)     <p>
744)     Compass is a web and command line application that filters and
745)     aggregates the Tor relays based on various attributes.
746)     </p>

747) 
748)     <a id="project-onionoo"></a>
749)     <h3><a href="<page projects/onionoo>">Onionoo</a> (<a
750)     href="https://gitweb.torproject.org/onionoo.git">java codebase</a>, <a
751)     href="https://gitweb.torproject.org/pyonionoo.git">python

752)     codebase</a>, <a href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Onionoo&order=priority">bug tracker</a>)</h3>

753) 
754)     <p>
755)     Onionoo is a JSON based protocol to learn information about currently
756)     running Tor relays and bridges.
757)     </p>
758) 

759)     <a id="project-weather"></a>

760)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/org/roadmaps/Weather">Weather</a> (<a

761)     href="https://gitweb.torproject.org/weather.git">code</a>, <a
762)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Tor+Weather&order=priority">bug
763)     tracker</a>)</h3>

764) 

765)     <p>
766)     Provides automatic notification to subscribed relay operators when their
767)     relay's unreachable. This underwent a rewrite by the <a
768)     href="http://hfoss.wesleyan.edu/">Wesleyan HFOSS team</a>, which went live
769)     in early 2011.
770)     </p>

771) 

772)     <a id="project-gettor"></a>

773)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/org/roadmaps/GetTor">GetTor</a> (<a

774)     href="https://gitweb.torproject.org/gettor.git">code</a>, <a

775)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=GetTor&order=priority">bug
776)     tracker</a>)</h3>

777) 

778)     <p>
779)     E-mail autoresponder providing Tor's packages over SMTP. This has been
780)     relatively unchanged for quite a while.
781)     </p>

782) 

783)     <a id="project-torcheck"></a>

784)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/org/roadmaps/TorCheck">TorCheck</a> (<a

785)     href="https://svn.torproject.org/svn/check/trunk/">code</a>, <a
786)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Tor+Check&order=priority">bug
787)     tracker</a>)</h3>

788) 

789)     <p>
790)     Provides a simple site for determining if the visitor is using Tor or not.
791)     This has been relatively unchanged for quite a while.
792)     </p>

793) 

794)     <a id="project-bridgedb"></a>

795)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/org/roadmaps/BridgeDB">BridgeDB</a> (<a

796)     href="https://gitweb.torproject.org/bridgedb.git">code</a>, <a
797)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=BridgeDB&order=priority">bug
798)     tracker</a>)</h3>

799) 

800)     <p>
801)     Backend bridge distributor, handling the various pools they're distributed
802)     in. This was actively developed until Fall of 2010.
803)     </p>

804) 
805)     <a id="project-ooni-probe"></a>

806)     <h3><a href="https://ooni.torproject.org/">Ooni Probe</a> (<a

807)     href="https://gitweb.torproject.org/ooni-probe.git">code</a>, <a
808)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Ooni&order=priority">bug
809)     tracker</a>)</h3>
810) 
811)     <p>
812)     Censorship scanner, checking your local connection for blocked or modified
813)     content.
814)     </p>
815) 

816)     <a id="project-torflow"></a>

817)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/org/roadmaps/TorFlow">TorFlow</a> (<a

818)     href="https://gitweb.torproject.org/torflow.git">code</a>, <a
819)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Torflow&order=priority">bug
820)     tracker</a>)</h3>

821) 

822)     <p>
823)     Library and collection of services for actively monitoring the Tor network.
824)     These include the Bandwidth Scanners (measuring throughput of relays) and
825)     SoaT (scans for malicious or misconfigured exit nodes). SoaT was last
826)     actively developed in the Summer of 2010, and the Bandwidth Scanners a few
827)     months later. Both have been under active use since then, but development
828)     has stopped.
829)     </p>

830) 

831)     <a id="project-torbel"></a>
832)     <h3><a

833)     href="https://blog.torproject.org/blog/torbel-tor-bulk-exit-list-tools">TorBEL</a> (<a
834)     href="https://gitweb.torproject.org/torbel.git">code</a>, <a

835)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=TorDNSEL/TorBEL&order=priority">bug
836)     tracker</a>)</h3>

837) 

838)     <p>
839)     The Tor Bulk Exitlist provides a method of identifying if IPs belong to
840)     exit nodes or not. This is a replacement for TorDNSEL which is a stable
841)     (though unmaintained) Haskell application for this purpose. The initial
842)     version of TorBEL was started in GSOC 2010 but since then the project has
843)     been inactive.
844)     </p>

845) 
846)     <a id="project-tor2web"></a>
847)     <h3><a
848)     href="http://wiki.tor2web.org/index.php/Main_Page">Tor2web</a> (<a
849)     href="https://github.com/globaleaks/tor2web-3.0/wiki">code</a>)</h3>
850) 
851)     <p>
852)     Tor2web allows Internet users to browse websites running in <a
853)     href="<page docs/hidden-services>">Tor hidden services</a>. It trades
854)     user anonymity for usability by allowing anonymous content to be
855)     distributed to non-anonymous users.
856)     </p>
857) 

858)     <a id="project-anonbib"></a>
859)     <h3><a
860)     href="http://freehaven.net/anonbib/">Anonymity Bibliography</a> (<a
861)     href="https://gitweb.torproject.org/anonbib.git">code</a>)</h3>
862) 
863)     <p>

864)     Anonbib is a list of important papers in the field of anonymity. It's
865)     also a set of scripts to generate the website from Latex (bibtex). If
866)     we're missing any important papers, please let us know!

867)     </p>
868) 

869)     <a id="Coding"></a>
870)     <a id="Summer"></a>

871)     <h2><a class="anchor" href="#Coding">Project Ideas</a></h2>

872) 

873)     <p>

874)     You may find some of these projects to be good ideas for <a href="<page
875)     about/gsoc>">Google Summer of Code</a> and the <a
876)     href="https://live.gnome.org/OutreachProgramForWomen">Outreach Program for
877)     Women</a>. We have labelled each idea with how useful it would be to the
878)     overall Tor project (priority), how much work we expect it would be (effort
879)     level), how much clue you should start with (skill level), and which of our
880)     <a href="<page about/corepeople>">core developers</a> would be good
881)     mentors. If one or more of these ideas looks promising to you, please <a

882)     href="<page about/contact>">contact us</a> to discuss your plans rather than
883)     sending blind applications. You may also want to propose your own project
884)     idea &mdash; which often results in the best applications.
885)     </p>

886) 

887)     <p>
888)     <b>If you came to this site for Google Summer of Code then please keep in
889)     mind that Google has not yet announced the participating organisations, and
890)     Tor might not be selected. See the <a
891)     href="https://www.google-melange.com/document/show/gsoc_program/google/gsoc2013/help_page#2._What_is_the_program_timeline">timeline</a>
892)     for more information.</b>
893)     </p>
894) 

895)     <ol>

896) 

897)     <!--

898)     <a id="auditTBB"></a>

899)     <li>

900)     <b>Audit Tor Browser Bundles for data leaks</b>

901)     <br>

902)     Effort Level: <i>High</i>

903)     <br>

904)     Skill Level: <i>Medium</i>

905)     <br>

906)     Likely Mentors: <i>Jacob</i>

907)     <p>The Tor Browser Bundle incorporates Tor, Firefox, Polipo, and the Vidalia

908)     user interface (and optionally the <a href="http://pidgin.im/">Pidgin</a>
909)     Instant Messaging client). Components are pre-configured to operate in a
910)     secure way, and it has very few dependencies on the installed operating
911)     system. It has therefore become one of the most easy to use, and popular,

912)     ways to use Tor on Windows.</p>
913)     <p>This project is to identify all of the traces left behind by
914)     using a Tor Browser Bundle on Windows, Mac OS X, or Linux.  Developing
915)     ways to stop, counter, or remove these traces is a final step.</p>
916)     <p>Students should be familiar with operating system analysis,

917)     application development on one or preferably all of Windows, Linux,

918)     and Mac OS X, and be comfortable with C/C++ and shell scripting.</p>

919)     <p>Since the core of this project is researching and auditing TBB this is
920)     not likely to be a good GSoC project.</p>

921)     </li>

922)     -->

923) 

924)     <a id="txtorcon-stemIntegration"></a>
925)     <li>
926)     <b>Txtorcon/Stem Integration</b>
927)     <br>
928)     Effort Level: <i>Medium</i>
929)     <br>
930)     Skill Level: <i>Medium</i>
931)     <br>
932)     Likely Mentors: <i>meejah, Damian (atagar)</i>
933)     <p>Txtorcon is a Twisted-based Python controller library, and Stem is a
934)     synchronous (threaded) one, also in Python. There is no need to have
935)     two implementations of (at least) the protocol parsing code. This
936)     project would entail eliminating duplication by leveraging Stem's
937)     parsing in txtorcon while keeping txtorcon's API the same (or at least
938)     close).</p>
939)     <p>This would very likely involve changes to both libraries, although
940)     most would be expected to be in txtorcon. meejah is available to
941)     mentor txtorcon changes, and Damian (atagar) can help with Stem.</p>
942)     <p>It would help if you're already familiar with event-based programming,
943)     bonus points if it's Twisted.</p>
944)     </li>
945) 

946)     <!--

947)     <a id="orbot-userInterface"></a>

948)     <li>

949)     <b>Build a better user interface for Orbot</b>

950)     <br>

951)     Effort Level: <i>Medium</i>

952)     <br>

953)     Skill Level: <i>Medium</i>

954)     <br>

955)     Likely Mentors: <i>Jake</i>
956)     <p>Improved home screen to show confirmation of connection (via a TorCheck
957)     API call), better statistics about data transferred (up/down), number of
958)     circuits connected, quality of connection and so on. The &quot;Tether
959)     Wifi&quot; Android application is a good model to follow in how it shows a
960)     realtime count of bytes transferred as well as notifications when wifi
961)     clients connect. In addition, better handling of Tor system and error
962)     messages would also be very helpful, include use of standard Android
963)     operating systems notifications. The addition of a wizard or tutorial
964)     walkthrough for novice users to explain to them exactly what is and what is
965)     not anonymized or protected would greatly improve the likelihood they will
966)     use Orbot correctly. All of this should work on the range of screens and
967)     device types now offered for Android, from 2&quot; phone to 10&quot;
968)     Tablet.</p>

969)     </li>

970)     -->

971) 

972)     <!--

973)     <a id="armClientMode"></a>

974)     <li>

975)     <b>Client Mode Use Cases for Arm</b>

976)     <br>
977)     Effort Level: <i>High</i>
978)     <br>

979)     Skill Level: <i>Medium</i>

980)     <br>

981)     Likely Mentors: <i>Damian (atagar)</i>
982)     <p><a href="<page projects/arm>">Arm</a> is a Tor command line status
983)     monitor on *nix environments (Linux, Mac, and BSD). It functions much like
984)     top does, giving a CLI overlay of Tor's bandwidth usage, connections,
985)     configuration, log, etc. Thus far its design has been geared for Tor relay
986)     operators. However, this doesn't need to be the case. This project would be
987)     to expand and simplify arm to make it useful for Tor's client users
988)     too.</p>

989) 

990)     <p>This would include UI design, experimenting, and a lot of python
991)     hacking. Here's some ideas for client functionality arm could provide:</p>

992) 

993)     <ul>
994)       <li>A panel for client connections, showing each hop of the user's
995)       circuits with the ISP, country, and jurisdiction where those relays
996)       reside. Other interesting information would be the circuit's latency, how
997)       long its been around, and its possible exit ports. Some of this will be
998)       pretty tricky and require some experimentation to figure out what
999)       information can be fetched safely (for instance, scraping rdns and whois
1000)       lookups could give hints about a relay's ISP, but we'd need to do it on
1001)       all Tor relays to avoid leaking our connections to the resolver).</li>

1002) 

1003)       <li>Options to let the user request new circuits (the &quot;New
1004)       Identity&quot; feature in Vidalia), select the exit country, etc.</li>

1005) 

1006)       <li>A panel showing Internet application and if their connections are
1007)       being routed through Tor or not (giving a warning if there's leaks).</li>

1008) 

1009)       <li>The status of the bridges we're configured to use (ie, are they up?).
1010)       This would include adding control port functionality to Tor for <a
1011)       href="https://trac.torproject.org/projects/tor/ticket/2068">ticket
1012)       2068</a>.</li>

1013) 

1014)       <li>A one click option to set Tor to be a client, relay, or bridge. The
1015)       goal would be to make it trivial for users to voluntarily contribute to
1016)       the Tor network.</li>

1017) 

1018)       <li>Menus as an alternative to hotkeys to make the interface more
1019)       intuitive and usable for beginners (<a
1020)       href="http://gnosis.cx/publish/programming/charming_python_6.html">example</a>).</li>

1021) 

1022)       <li>Look at Vidalia and TorK for ideas and solicit input from the Tor community.</li>
1023)     </ul>

1024) 

1025)     <p>
1026)     More information is available in the following sections of arm's dev notes: <a href="https://trac.torproject.org/projects/tor/wiki/doc/arm#ConnectionListingExpansion">Connection Listing Expansion</a>, <a href="https://trac.torproject.org/projects/tor/wiki/doc/arm#CircuitDetails">Circuit Details</a>, and <a href="https://trac.torproject.org/projects/tor/wiki/doc/arm#ClientModeUseCases">Client Mode Use Cases</a>
1027)     </p>

1028)     </li>

1029)     -->

1030) 

1031)     <!--
1032)     <a id="orbot-optimisation"></a>
1033)     <li>
1034)     <b>Core Tor mobile optimisation</b>
1035)     <br>
1036)     Effort Level: <i>Medium</i>

1037)     <br>
1038)     Skill Level: <i>High</i>
1039)     <br>

1040)     Likely Mentors: <i>Jake</i>
1041)     <p>
1042)     The existing port of Tor to Android is basically a straight
1043)     cross-compile to Linux ARM. There has been no work done in looking at
1044)     possible optimizations of Tor within a mobile hardware environment or on
1045)     mobile networks. In addition, a number of additional Android OS APIs are
1046)     available (such as wireless network status) that could be taken
1047)     advantage of.
1048)     </p>

1049) 

1050)     <p>
1051)     It should be noted, that even without optimisation, Tor is handling the
1052)     mobile network environment very well, automatically detecting change in
1053)     IP addresses, opening circuits, etc, as the device switches from no
1054)     coverage to 2G, 3G or Wifi constantly as it changes position. However,
1055)     this observation of &quot;very well&quot;, is just based on user
1056)     experience, and not any detailed study of what exactly is happening, and
1057)     what threats might exist because of this constantly changing network state.
1058)     </p>

1059) 

1060)     <p>
1061)     Finally, the build process needs to be moved to the Android NDK from the
1062)     custom GCC toolchain we are now using, and compatibility with Android
1063)     2.3 and 3.x Honeycomb OS need to be verified.
1064)     </p>

1065) 

1066)     <p>
1067)     For more information see the <a
1068)     href="https://svn.torproject.org/svn/projects/android/trunk/Orbot/BUILD">Orbot
1069)     build documentation</a>.
1070)     </p>

1071)     </li>

1072)     -->

1073) 

1074)     <!--

1075)     <a id="obfsproxy-scanning-measures"></a>
1076)     <li>
1077)     <b>Defensive bridge active scanning measures</b>
1078)     <br>
1079)     Effort Level: <i>High</i>
1080)     <br>
1081)     Skill Level: <i>High</i>
1082)     <br>
1083)     Likely Mentors: <i>asn</i>
1084)     <p>Involves providing good answers to <a
1085)     href="https://lists.torproject.org/pipermail/tor-dev/2011-November/003073.html">this
1086)     thread</a> as well as concrete implementation plans for it.</p>

1087) 

1088)     <p>This also involves implementing proposals <a
1089)     href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/proposals/189-authorize-cell.txt">189</a>
1090)     and <a
1091)     href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/proposals/190-shared-secret-bridge-authorization.txt">190</a>.</p>
1092)     </li>

1093)     -->

1094) 

1095)     <!--
1096)     <a id="firewallProbeTool"></a>

1097)     <li>

1098)     <b>Develop a fully automatic firewall-probing system</b>

1099)     <br>
1100)     Effort Level: <i>Medium to High</i>
1101)     <br>
1102)     Skill Level: <i>High</i>
1103)     <br>

1104)     Likely Mentors: <i>Robert Ransom, Jacob</i>
1105)     <p>We would like to have a fully automatic firewall-probing system for
1106)     blocking systems with no long-term state (i.e. firewalls that can
1107)     examine each connection, but do not change their behaviour for future
1108)     connections based on the traffic they have seen).</p>
1109)     <p>Ideally, volunteers would only need to set up one or more test servers,
1110)     and run the probe client program on a publicly accessible computer
1111)     behind the firewall.</p>
1112)     <p>The test tool should:</p>
1113)     <ul>
1114)     <li>generate packet captures on both ends (and send them out to the
1115)         extent possible),</li>
1116)     <li>cycle through all the SSL configurations we might want to test
1117)         through a censorship device, and</li>
1118)     <li>also test some other protocols to see whether they are allowed
1119)         through the firewall (IMAP and other mail protocols, BitTorrent,
1120)         DTLS, etc.).</li>
1121)     </ul>
1122)     </li>
1123)     -->

1124) 

1125)     <!--
1126)     <a id="obfsproxy-fuzzer"></a>
1127)     <li>
1128)     <b>Fuzzer for the Tor protocol</b>
1129)     <br>
1130)     Effort Level: <i>Medium to High</i>
1131)     <br>
1132)     Skill Level: <i>High</i>
1133)     <br>
1134)     Likely Mentors: <i>asn</i>
1135)     <p>Involves researching good and smart ways to fuzz stateful network
1136)     protocols, and also implementing the fuzzer.</p>

1137) 

1138)     <p>We are mostly looking for a fuzzer that fuzzes the Tor protocol
1139)     itself, and not the Tor directory protocol.</p>

1140) 

1141)     <p>Bonus points if it's extremely modular. Relevant research:</p>

1142) 

1143)     <ul>
1144)       <li>PROTOS - Security Testing of Protocol Implementations</li>
1145)       <li>INTERSTATE: A Stateful Protocol Fuzzer for SIP</li>
1146)       <li>Detecting Communication Protocol Security Flaws by Formal Fuzz
1147)       Testing and Machine Learning</li>
1148)       <li>SNOOZE: Toward a Stateful NetwOrk prOtocol fuzZE</li>
1149)       <li>Michal Zalewski's &quot;bugger&quot;</li>
1150)       <li>Also look at the concepts of &quot;model checking&quot; and
1151)       &quot;symbolic execution&quot; to get inspired.</li>
1152)     </ul>
1153)     </li>
1154)     -->

1155) 
1156)     <a id="httpsImpersonation"></a>

1157)     <li>
1158)     <b>HTTPS Server Impersonation</b>
1159)     <br>
1160)     Effort Level: <i>Medium to High</i>
1161)     <br>
1162)     Skill Level: <i>Medium to High</i>
1163)     <br>
1164)     Likely Mentors: <i>Nick (nickm)</i>

1165)     <p>

1166)     We have an open proposal for a way to make Tor bridges avoid censorship by
1167)     impersonating an HTTPS server.  Specifically, we need to hack some popular
1168)     SSL "reverse proxy" (your choice) so that it relays regular web connections
1169)     to an HTTP server, but certain connections to a local Tor process.  <a
1170)     href="https://gitweb.torproject.org/torspec.git/blob_plain/HEAD:/proposals/203-https-frontend.txt">Proposal
1171)     203</a> has a general design sketch.

1172)     </p>
1173)     </li>

1174) 

1175)     <!--
1176)     <a id="geoIPUpgrade"></a>

1177)     <li>

1178)     <b>Improve our GeoIP file format</b>

1179)     <br>

1180)     Effort Level: <i>Medium</i>
1181)     <br>
1182)     Skill Level: <i>Medium to High</i>
1183)     <br>
1184)     Likely Mentors: <i>Robert Ransom</i>
1185)     <p>Currently, Tor bridges and relays read an entire IP-&gt;country database
1186)     into memory from a text file during startup.  We would like to
1187)     distribute this database and store it on disk in a much more compact
1188)     form, and perform IP-&gt;country lookups on it in its on-disk format if
1189)     possible.</p>
1190)     <p>We have <a href='https://trac.torproject.org/projects/tor/ticket/2506'>a
1191)     sketch of a design</a> for a moderately optimized format for IPv4 GeoIP
1192)     data; this project will involve both implementing the IPv4 format and
1193)     designing and implementing a format for IPv6 GeoIP data.</p>
1194)     <p>Since the core of this project is researching IPv6 GeoIP data and
1195)     designing the IPv6 format, this is not likely to be a good GSoC
1196)     project.</p>
1197)     </li>
1198)     -->

1199) 

1200)     <!--
1201)     <a id="unitTesting"></a>
1202)     <li>
1203)     <b>Improve our unit testing process</b>

1204)     <br>
1205)     Effort Level: <i>Medium</i>
1206)     <br>
1207)     Skill Level: <i>Medium</i>
1208)     <br>

1209)     Likely Mentors: <i>Nick, Erinn</i>
1210)     <p>Tor needs to be tested far more thoroughly. This is a
1211)     multi-part effort. To start with, our unit test coverage should
1212)     rise substantially, especially in the areas outside the utility
1213)     functions. This will require significant refactoring of some parts
1214)     of Tor, in order to dissociate as much logic as possible from
1215)     globals.</p>
1216)     <p>Additionally, we need to automate our performance testing. We've got
1217)     buildbot to automate our regular integration and compile testing already
1218)     (though we need somebody to set it up on Windows),
1219)     but we need to get our network simulation tests (as built in <a
1220)     href="https://svn.torproject.org/svn/torflow/trunk/README">TorFlow</a>)
1221)     updated for more recent versions of Tor, and designed to launch a test
1222)     network either on a single machine, or across several, so we can test
1223)     changes in performance on machines in different roles automatically.</p>
1224)     </li>
1225)     -->

1226) 

1227)     <!--
1228)     <a id="vidaliaNetworkMap"></a>
1229)     <li>
1230)     <b>Improved and More Usable Network Map in Vidalia</b>
1231)     <br>
1232)     Effort Level: <i>Medium</i>
1233)     <br>
1234)     Skill Level: <i>Medium</i>
1235)     <br>
1236)     Likely Mentors: <i>Tomás</i>

1237)     <p>

1238)     One of Vidalia's existing features is a network map that shows the user
1239)     the approximate geographic location of relays in the Tor network and
1240)     plots the paths the user's traffic takes as it is tunneled through the
1241)     Tor network. The map is currently not very interactive and has rather
1242)     poor graphics. Instead, we implemented KDE's Marble widget such
1243)     that it gives us a better quality map and enables improved interactivity,
1244)     such as allowing the user to click on individual relays or circuits to
1245)     display additional information. We want to add the ability
1246)     for users to click on a particular relay or a country containing one or
1247)     more Tor exit relays and say, "I want my connections to exit
1248)     from here."
1249)     </p>

1250) 

1251)     <p>
1252)     This project will first involve getting familiar with Vidalia
1253)     and the Marble widget's API. One will then integrate the widget
1254)     into Vidalia and customize Marble to be better suited for our application,
1255)     such as making circuits clickable, storing cached map data in Vidalia's
1256)     own data directory, and customizing some of the widget's dialogs.
1257)     </p>

1258) 

1259)     <p>
1260)     A person undertaking this project should have good C++ development
1261)     experience. Previous experience with Qt and CMake is helpful, but not
1262)     required.

1263)     </p>
1264)     </li>

1265)     -->

1266) 

1267)     <!--
1268)     <a id="resistCensorship"></a>

1269)     <li>

1270)     <b>Improving Tor's ability to resist censorship</b>

1271)     <br>

1272)     Effort Level: <i>Medium to High</i>

1273)     <br>

1274)     Skill Level: <i>High</i>
1275)     <br>
1276)     Likely Mentors: <i>Jake, Thomas</i>
1277)     <p>The Tor 0.2.1.x series makes <a
1278)     href="<svnprojects>design-paper/blocking.html">significant
1279)     improvements</a> in resisting national and organizational censorship.
1280)     But Tor still needs better mechanisms for some parts of its
1281)     anti-censorship design.</p>
1282)     <p>One huge category of work is adding features to our <a
1283)     href="https://gitweb.torproject.org/bridgedb.git?a=tree">BridgeDB</a>
1284)     service (Python). Tor aims to give out <a href="<page docs/bridges>">bridge
1285)     relay addresses</a> to users that can't reach the Tor network
1286)     directly, but there's an arms race between algorithms for distributing
1287)     addresses and algorithms for gathering and blocking them. See <a
1288)     href="<blog>bridge-distribution-strategies">our
1289)     blog post on the topic</a> as an overview, and then look at <a
1290)     href="https://lists.torproject.org/pipermail/tor-dev/2009-December/000666.html">Roger's
1291)     or-dev post</a> from December 2009 for more recent thoughts &mdash; lots of
1292)     design work remains.</p>
1293)     <p>If you want to get more into the guts of Tor itself (C), a more minor problem
1294)     we should address is that current Tors can only listen on a single
1295)     address/port combination at a time. There's
1296)     <a href="<specblob>proposals/118-multiple-orports.txt">a
1297)     proposal to address this limitation</a> and allow clients to connect
1298)     to any given Tor on multiple addresses and ports, but it needs more
1299)     work.</p>
1300)     <p>This project could involve a lot of research and design. One of the big
1301)     challenges will be identifying and crafting approaches that can still
1302)     resist an adversary even after the adversary knows the design, and
1303)     then trading off censorship resistance with usability and
1304)     robustness.</p>
1305)     </li>
1306)     -->

1307) 

1308)     <!--
1309)     <a id="user-space-transport"></a>
1310)     <li>
1311)     <b>Integrating Tor with user-space transport protocol libraries</b>
1312)     <br>
1313)     Effort Level: <i>High</i>
1314)     <br>
1315)     Skill Level: <i>High</i>
1316)     <br>
1317)     Likely Mentors: <i>Steven</i>
1318)     <p>Tor currently sends data over TCP links between nodes. <a
1319)     href="http://static.usenix.org/events/sec09/tech/full_papers/reardon.pdf">Prior
1320)     research</a> has indicated that this may not be optimal, and instead the
1321)     role that TCP plays (congestion control and reliability) should be moved
1322)     into Tor itself. This would allow a number of desirable changes, such as
1323)     preventing errors on one circuit delaying another, and giving Tor control
1324)     and visibility of congestion control.</p>
1325)     <p>There are <a
1326)     href="http://www.cl.cam.ac.uk/~sjm217/papers/tor11datagramcomparison.pdf">many
1327)     ways to do this</a>, each with their own tradeoffs and difficulty of
1328)     implementation. This project will be to select one (or more) option and
1329)     implement it in Tor. The primary goal will be to test this modified version
1330)     of Tor in simulation, but if it turns out to work well, it could be
1331)     deployed in the live Tor network.</p>
1332)     <p>Excellent C programming skills are needed, and knowledge of Tor
1333)     internals are highly desirable.</p>
1334)     </li>
1335)     -->

1336) 

1337)     <a id="chutneyExpansion"></a>
1338)     <li>
1339)     <b>Make Chutney Do More, More Reliably</b>
1340)     <br>
1341)     Effort Level: <i>Medium to High, depending on scope of project</i>
1342)     <br>
1343)     Skill Level: <i>Medium</i>
1344)     <br>

1345)     Likely Mentors: <i>Nick (nickm)</i>
1346)     <p>

1347)     We have a little Python tool called <a
1348)     href="https://gitweb.torproject.org/nickm/chutney.git">Chutney</a> for
1349)     making small local test networks.  It's small, not widely used, and not as
1350)     automated as it could be.

1351)     </p>

1352) 

1353)     <p>

1354)     It would be great to see chutney extended and a set of supporting tests
1355)     built to the point where we could use Chutney to exercise various Tor
1356)     features as an automated integration test.

1357)     </p>

1358)     </li>

1359) 

1360)     <!--
1361)     <a id="torsocksForOSX"></a>
1362)     <li>
1363)     <b>Make torsocks/dsocks work on OS X</b>
1364)     <br>
1365)     Effort Level: <i>Medium</i>
1366)     <br>
1367)     Skill Level: <i>Medium</i>
1368)     <br>
1369)     Likely Mentors: <i>Robert Hogan</i>

1370)     <p>

1371)     <a href="https://code.google.com/p/torsocks/">Torsocks</a> and <a
1372)     href="https://code.google.com/p/dsocks/">dsocks</a> are wrappers that will
1373)     run applications, intercept their outgoing network connections, and push
1374)     those connections through Tor. The goal is to handle applications that
1375)     don't support proxies (or don't supporting them well). To get it right,
1376)     they need to intercept many system calls. The syscalls you need to
1377)     intercept on Linux differ dramatically from those on BSD. So Torsocks
1378)     works fine on Linux, dsocks works ok on BSD (though it may be less
1379)     maintained and thus might miss more syscalls), and nothing works well
1380)     on both. First, we should patch dsocks to use Tor's <i>mapaddress</i>
1381)     commands from the controller interface, so we don't waste a whole
1382)     round-trip inside Tor doing the resolve before connecting. Second,
1383)     we should make our <i>torify</i> script detect which of torsocks or
1384)     dsocks is installed, and call them appropriately. This probably means
1385)     unifying their interfaces, and might involve sharing code between them
1386)     or discarding one entirely.
1387)     </p>
1388)     </li>
1389)     -->

1390) 

1391)     <!--
1392)     <a id="obfsproxy-new-transports"></a>
1393)     <li>
1394)     <b>New and innovative pluggable transports</b>
1395)     <br>
1396)     Effort Level: <i>High</i>
1397)     <br>
1398)     Skill Level: <i>High</i>
1399)     <br>
1400)     Likely Mentors: <i>asn, Steven</i>
1401)     <p>Not-very-smart transports like ROT13 and base64 are nice but not super
1402)     interesting. Other ideas like bittorrent transports might be relevant,
1403)     but you will have to provide security proofs on why they are harder to
1404)     detect and block than other less-sophisticated transports.</p>

1405) 

1406)     <p>The whole point of this project, though, is to come up with new
1407)     transports that we haven't already thought of. Be creative.</p>

1408) 

1409)     <p>Bonus points if your idea is interesting and still implementable
1410)     through the summer period.</p>

1411) 

1412)     <p>More bonus points if it's implemented on top of obfsproxy, or if your
1413)     implementation has a pluggable transport interface on top of it (as
1414)     specified <a href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/proposals/180-pluggable-transport.txt">here</a>).</p>
1415)     </li>
1416)     -->

1417) 

1418)     <!--
1419)     <a id="orbot-orlibAndOutreach"></a>
1420)     <li>
1421)     <b>Orbot integration library and community outreach</b>
1422)     <br>
1423)     Effort Level: <i>Low</i>
1424)     <br>
1425)     Skill Level: <i>Medium</i>
1426)     <br>
1427)     Likely Mentors: <i>Nathan (n8fr8)</i>
1428)     <p>
1429)     We need additional work on <a
1430)     href="https://github.com/guardianproject/orlib">ORLib</a>, our library for
1431)     use with third-party application to easily enable them to support
1432)     &quot;Torification&quot; on non-rooted devices (i.e. w/o transparent
1433)     proxying). This library includes a SOCKS client, a wrapper for the Apache
1434)     HTTPClient library, a utility class for detecting the state of Orbot
1435)     connectivity, and other relevant/useful things an Android app might need to
1436)     anonymize itself. This work would includes direct development of the
1437)     library, documentation, and sample code. Outreach or effort to implement
1438)     the library within other open-source apps is also needed.

1439)     </p>
1440)     </li>

1441)     -->

1442) 

1443)     <!--
1444)     <a id="tailsHiddenServicePetnames"></a>
1445)     <li>
1446)     <b>Petname system for Tor hidden services</b>
1447)     <br>
1448)     Effort Level: <i>High</i>
1449)     <br>
1450)     Skill Level: <i>High</i>
1451)     <br>
1452)     Likely Mentors: <i>ague</i>
1453)     <p>Tor provides hidden services. These services are only reachable through
1454)     Tor itself, and provide greater anonymity both for the providers of the
1455)     service and for its users.</p>
1456)     <p>One current downside of Tor hidden services is that they are addressed
1457)     using 80-bit base32-encoded addresses such as "v2cbb2l4lsnpio4q.onion".
1458)     These addresses are hard to remember; this makes them hard to use
1459)     within amnesic environment like Tails.</p>
1460)     <p>The project is to implement a petname system for Tor hidden services:
1461)     a way for users or providers of Tor hidden services to add a simple
1462)     'nickname' to a central database. Users could then query this central
1463)     database to retrieve a full hidden service address by giving
1464)     a nickname.</p>
1465)     <p>Adding petnames to the database could be done using a web interface or
1466)     automated fetch like those described in the <a
1467)     href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/proposals/ideas/xxx-onion-nyms.txt">&quot;.onion
1468)     nym system&quot; proposal</a>.</p>
1469)     <p>Querying the database could be done using a web interface, a REST API and
1470)     a DNS interface.</p>
1471)     <p>In order not to grow indefinitely, the software should make regular tests to
1472)     see if hidden services are still reachable and, depending on the last time
1473)     a nickname was accessed, cleanup the database as necessary.</p>
1474)     <p>The software should allow a distributed, fault-tolerant setup.
1475)     All nodes should have a synchronized copy of the database, should be
1476)     ready to answer queries and should coordinate the tests for hidden
1477)     service availability.</p>
1478)     <p>The resulting codebase must be easy to deploy: it should not be hard to
1479)     setup new databases.</p>
1480)     <p>It is expected that the volunteer will be using Behaviour Driven
1481)     Development methods. Either in Ruby using Cucumber and RSpec, or in
1482)     Python using similar tools.</p>
1483)     </li>
1484)     -->

1485) 

1486)     <a id="limitCapabilities"></a>

1487)     <li>

1488)     <b>Run With Limited Capabilities</b>

1489)     <br>
1490)     Effort Level: <i>Medium to High</i>
1491)     <br>

1492)     Skill Level: <i>High</i>

1493)     <br>
1494)     Likely Mentors: <i>Nick (nickm)</i>
1495)     <p>

1496)     Many modern operating systems give a running program the ability to drop
1497)     capabilities that it no longer needs, and other ways for a program to run
1498)     pieces of itself in a sandbox with diminished privileges.

1499)     </p>

1500) 

1501)     <p>

1502)     We'd like to do this with Tor, to improve its resistance to attacks.  The
1503)     easiest areas to address would be on systems like <a
1504)     href="https://lwn.net/Articles/475361/">recent Linux kernels</a> that make
1505)     it easy to drop or restrict the set of syscalls that a program can invoke.
1506)     That's a great project, but probably not big enough for an internship just
1507)     on its own.  For that, we'd want to make progress on at least multiple
1508)     platforms, or look into refactoring Tor into pieces that need more
1509)     privileges and pieces that don't with an eye towards sandboxing them
1510)     differently.

1511)     </p>

1512) 

1513)     <p>

1514)     See tickets <a href="https://trac.torproject.org/7005">#7005</a> and <a
1515)     href="https://trac.torproject.org/5219">#5219</a>, and their descendants,
1516)     for more information.

1517)     </p>
1518)     </li>

1519) 

1520)     <a id="metricsSearch"></a>

1521)     <li>

1522)     <b>Searchable Tor descriptor and Metrics data archive</b>

1523)     <br>
1524)     Effort Level: <i>Medium</i>
1525)     <br>
1526)     Skill Level: <i>Medium</i>
1527)     <br>

1528)     Likely Mentors: <i>Karsten</i>
1529)     <p>The <a href="https://metrics.torproject.org/data.html">Metrics data
1530)     archive</a> of Tor relay descriptors and other Tor-related network data has
1531)     grown to over 100G in size, bz2-compressed.  We have developed two search
1532)     interfaces: the <a
1533)     href="https://metrics.torproject.org/relay-search.html">relay search</a>
1534)     finds relays by nickname, fingerprint, or IP address in a given month; <a
1535)     href="https://metrics.torproject.org/exonerator.html">ExoneraTor</a> finds
1536)     whether a given IP address was a relay on a given day.</p>

1537) 

1538)     <p>We'd like to have a more general search application for Tor descriptors
1539)     and metrics data.  There are more <a
1540)     href="https://metrics.torproject.org/formats.html">descriptor types</a>
1541)     that we'd like to include in the search.  The search application should
1542)     handle most of them and understand some semantics like what's a timestamp,
1543)     what's an IP address, and what's a link to another descriptor.  Users
1544)     should then be able to search for arbitrary strings or limit their search
1545)     to given time periods or IP address ranges.  Descriptors that reference
1546)     other descriptors should contain links, and descriptors should be able to
1547)     say from where they are linked.  The goal is to make the archive easily
1548)     browsable.</p>

1549) 

1550)     <p>The search application shall be separate from the metrics website and
1551)     shouldn't rely on the metrics website codebase.  The search application
1552)     will contain hourly updated descriptor data from the metrics website via
1553)     rsync.  Programming language and database system are not specified yet,
1554)     though there's a slight preference for Python/Django and Postgres for
1555)     maintenance reasons.  If there are good reasons to pick something else,
1556)     e.g, some NoSQL variant or some search application framework, that's fine,
1557)     too.  Further requirements are that lookups should be really fast and that
1558)     changes to the search application can be implemented in reasonable
1559)     time.</p>

1560) 

1561)     <p>Applications for this project should come with a design of the proposed
1562)     search application, ideally with a proof-of-concept based on a subset of
1563)     the available data to show that it will be able to handle the 100G+ of
1564)     data.</p>

1565)     </li>

1566) 

1567)     <!--
1568)     <a id="simulateSlowConnections"></a>

1569)     <li>

1570)     <b>Simulator for slow Internet connections</b>

1571)     <br>
1572)     Effort Level: <i>Medium</i>
1573)     <br>
1574)     Skill Level: <i>Medium</i>
1575)     <br>

1576)     Likely Mentors: <i>Nick</i>

1577)     <p>

1578)     Many users of Tor have poor-quality Internet connections, giving low
1579)     bandwidth, high latency, and high packet loss/re-ordering. User
1580)     experience is that Tor reacts badly to these conditions, but it is
1581)     difficult to improve the situation without being able to repeat the
1582)     problems in the lab.
1583)     </p>

1584) 

1585)     <p>
1586)     This project would be to build a simulation environment which
1587)     replicates the poor connectivity so that the effect on Tor performance
1588)     can be measured. Other components would be a testing utility to
1589)     establish what are the properties of connections available, and to
1590)     measure the effect of performance-improving modifications to Tor.
1591)     </p>

1592) 

1593)     <p>
1594)     The tools used would be up to the student, but dummynet (for FreeBSD)
1595)     and nistnet (for Linux) are two potential components on which this
1596)     project could be built. Students should be experienced with network
1597)     programming/debugging and TCP/IP, and preferably familiar with C and a
1598)     scripting language.

1599)     </p>
1600)     </li>

1601)     -->

1602) 

1603)     <a id="stemUsability"></a>

1604)     <li>

1605)     <b>Stem Usability and Porting</b>

1606)     <br>
1607)     Effort Level: <i>Medium</i>
1608)     <br>
1609)     Skill Level: <i>Medium</i>
1610)     <br>

1611)     Likely Mentors: <i>Damian (atagar)</i>
1612)     <p>

1613)     <a href="https://stem.torproject.org/">Stem</a> is a

1614)     python controller library for tor. Like it's predecessor, <a

1615)     href="https://gitweb.torproject.org/pytorctl.git">TorCtl</a>, it uses tor's <a

1616)     href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/control-spec.txt">control
1617)     protocol</a> to help developers program against the tor process, enabling
1618)     them to build things similar to <a href="#project-vidalia">Vidalia</a> and

1619)     <a href="http://www.atagar.com/arm/">arm</a>.

1620)     </p>

1621) 

1622)     <p>
1623)     While TorCtl provided a fine first draft for this sort of functionality,
1624)     it has not proved to be extensible nor maintainable. Stem is a rewrite of
1625)     TorCtl with a heavy focus on testing, documentation, and providing a
1626)     developer friendly API.
1627)     </p>

1628) 

1629)     <p>

1630)     Stem has largely met these goals but there's still plenty of low hanging
1631)     fruit in terms of usability improvements. Also, we have clients still using
1632)     TorCtl that need to be ported over. This project includes several subtasks,
1633)     see the tickets for additional context...

1634)     </p>

1635) 

1636)     <ol>
1637)       <li>Come up with a better, more developer friendly "Module Overview" for our documentation (<a href="https://stem.torproject.org/api/control.html">example page</a>). For instance, it would be nice to provide interlinking between the overview and the classes/methods that it lists. This will probably involve asking for help from the <a href="http://sphinx-doc.org/">Sphinx user list</a>. (<a href="https://trac.torproject.org/7632">ticket</a>)</li>
1638)       <li>Navigation of <a href="https://stem.torproject.org">stem's site</a> is clunky, in large part due to the linear page flow supported by sphinx. One possible fix for this would be css-only menus, ideally as a sphinx plugin so we could contribute it back upstream.(<a href="https://trac.torproject.org/8431">ticket</a>)</li>
1639)       <li>After getting some exposure to stem it would be time to give using it a try. Tor has a couple clients (<a href="https://trac.torproject.org/8263">TorBEL</a> and <a href="https://trac.torproject.org/8264">Tor Weather</a>) that are ready to be moved to stem. Hopefully porting these will also provide us with some lessons on how we can further improve stem's API.</li>
1640)       <li><b>... other ideas?</b> The above are just some ideas I've come up with to improve usability and likely not enough to fill the summer. Feel free to suggest some of your own! For instance, one option would be to expand <a href="https://stem.torproject.org/tutorial.html">stem's tutorials</a> with more examples, maybe <a href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev/">contacting tor-dev@</a> to brainstorm ideas. This last bit is pretty open so I look forward to seeing what you come up with!</li>
1641)     </ol>

1642) 

1643)     <p>

1644)     <b>As part of your application for this project please write a script that does something interesting with stem.</b> Bonus points if this is something that we can <a href="https://stem.torproject.org/tutorial.html">make a tutorial</a> around!

1645)     </p>

1646) 

1647)     <a id="stemTestingForTor"></a>
1648)     <li>
1649)     <b>Stem Tests for Tor</b>
1650)     <br>
1651)     Effort Level: <i>Medium</i>
1652)     <br>
1653)     Skill Level: <i>Medium</i>
1654)     <br>
1655)     Likely Mentors: <i>Damian (atagar)</i>
1656)     <p>
1657)     Stem is a library for interacting with Tor (see '<a href="#stemUsability">Stem Usability and Porting</a>' above for a summary). The library has both <a href="https://gitweb.torproject.org/stem.git/tree/HEAD:/test/unit">unit</a> and <a href="https://gitweb.torproject.org/stem.git/tree/HEAD:/test/integ">integration</a> tests. The unit tests provide a quick, direct test of stem's codebase while the integration test exercises its functionality against a live instance of Tor.
1658)     </p>
1659) 
1660)     <p>
1661)     Stem's integration tests have thus far been (unsurprisingly) designed to test stem but there's no need for them to be limited to that. Stem is a complete implementation of Tor's <a href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/control-spec.txt">control-spec</a> and <a href="https://gitweb.torproject.org/torspec.git/blob/HEAD:/dir-spec.txt">dir-spec</a>. As such, stem's tests could be easily expanded to more dedicatedly test behavior involved in those portions of Tor's codebase, as well as provide a smoke test for its general functionality.
1662)     </p>
1663) 
1664)     <p>
1665)     This project would involve several components:
1666)     </p>
1667) 
1668)     <ol>
1669)       <li>Determine what kind of tests we need. <b>This should be done during the application phase</b> by <a href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev/">contacting tor-dev@</a>. Hopefully this will give us an idea of what would be the most useful kind of tests of this nature for Tor development.</li>
1670)       <li>To be useful our integration tests need to continually run against the present tip of Tor's codebase. To do this we'll want to (1) fetch and compile the latest version of Tor, (2) run our integration tests against it, (3) compose the results as an html formatted email, and send it to a list. (<a href="https://trac.torproject.org/8261">ticket</a>)</li>
1671)       <li>Implement the new suite of integration tests for Tor. This will likely include expanding Tor to support better testability. One useful candidate, for instance, would be a controller method to fetch our own descriptor. This would let us easily test various configurations to see if they provide valid descriptor content.</li>
1672)     </ol>
1673) 
1674)     <p>
1675)     <b>As part of your application for this project please write some code to expand stem's tests.</b> Bonus points if it impelments one of your suggestions for better testing Tor!
1676)     </p>
1677) 

1678)     <!--

1679)     <a id="tailsServer"></a>
1680)     <li>
1681)     <b>Tails server: Self-hosted services behind Tails-powered Tor hidden services</b>
1682)     <br>
1683)     Effort Level: <i>High</i>
1684)     <br>
1685)     Skill Level: <i>Medium, but wide-scoped</i>
1686)     <br>
1687)     Likely Mentors: <i>intrigeri, anonym</i>
1688)     <p>Let's talk about group collaboration, communication and data sharing
1689)     infrastructure, such as chat servers, wikis, or file repositories.</p>
1690)     <p>Hosting such data and infrastructure <b>in the cloud</b> generally
1691)     implies to trust the service providers not to disclose content, usage or
1692)     users location information to third-parties. Hence, there are many threat
1693)     models in which cloud hosting is not suitable.</p>
1694)     <p>Tor partly answers the <b>users location</b> part; this is great, but
1695)     <b>content</b> is left unprotected.</p>
1696)     <p>There are two main ways to protect such content: either to encrypt it
1697)     client-side (<b>security by design</b>), or to avoid putting it into
1698)     untrusted hands in the first place.</p>
1699)     <p>Cloud solutions that offer security by design are rare and generally
1700)     not mature yet. The <b>Tails server</b> project is about exploring the
1701)     other side of the alternative: avoiding to put private data into
1702)     untrusted hands in the first place.</p>
1703)     <p>This is made possible thanks to Tor hidden services, that allow users
1704)     to offer location-hidden services, and make self-hosting possible in
1705)     many threat models. Self-hosting has its own lot of problems, however,
1706)     particularly in contexts where the physical security of the hosting
1707)     place is not assured. Combining Tor hidden services with Tails'
1708)     amnesia property and limited support for persistent encrypted data
1709)     allows to protect content, to a great degree, even in such contexts.</p>
1710)     <p>In short, setting up a new Tails server would be done by:</p>

1711) 

1712)     <ol style="list-style-type: decimal">
1713)       <li>Alice plugs a USB stick into a running desktop Tails system.</li>
1714)       <li>Alice uses a GUI to easily configure the needed services.</li>
1715)       <li>Alice unplugs the USB stick, that now contains encrypted services
1716)       configuration and data storage space.</li>
1717)       <li>Alice plugs that USB stick (and possibly a Tails Live CD) into the
1718)       old laptop that was dedicated to run Tails server.</li>
1719)       <li>Once booted, Alice enters the encryption passphrase either
1720)       directly using the keyboard or through a web interface listening on the
1721)       local network.</li>
1722)       <li>Then, Bob can use the configured services once he gets a hold on
1723)       the hidden service address. (The <b>petname system for Tor hidden
1724)       services</b> project would be very complementary to this one, by the
1725)       way.)</li>
1726)     </ol>

1727) 

1728)     <p>Tails server should content itself with hardware that is a bit old
1729)     (such as a PIII-450 laptop with 256MB of RAM) and/or half broken (e.g.
1730)     non-functional hard-disk, screen or keyboard).</p>
1731)     <p>The challenges behind this project are:</p>

1732) 

1733)     <ul>
1734)       <li>Design and write the services configuration GUI [keywords: edit
1735)       configuration files, upgrade between major Debian versions,
1736)       debconf].</li>
1737)       <li>How to create the hidden service key? [keywords: Vidalia, control
1738)       protocol].</li>
1739)       <li>Adapt the Tails boot process to allow switching to &quot;server
1740)       mode&quot; when appropriate.</li>
1741)       <li>Add support, to the Tails persistence setup process, for asking an
1742)       encryption passphrase without X, and possibly with a broken keyboard
1743)       and/or screen [keywords: local network, SSL/TLS?, certificate?].</li>
1744)     </ul>

1745) 

1746)     <p>This project can easily grow quite large, so the first task would
1747)     probably be to clarify what it would need to get an initial (minimal
1748)     but working) implementation ready to be shipped to users.</p>

1749)     <p>This project does not require to be an expert in one specific field,
1750)     but it requires to be experienced and at ease with a large scope of
1751)     software development tools, processes, and operating system knowledge.</p>
1752)     <p>Undertaking this project requires in-depth knowledge of Debian-like
1753)     systems (self-test: do the "dpkg conffile" and "debconf preseeding"
1754)     words sound new to your ear?); the Debian Live persistence system
1755)     being written in shell, being at ease with robust shell scripting is
1756)     a must; to end with, at least two pieces of software need to be
1757)     written from scratch (a GUI and a webapp): the preferred languages for
1758)     these tasks would be Python and Perl. Using Behaviour Driven
1759)     Development methods to convey expectations and acceptance criteria
1760)     would be most welcome.</p>
1761)     <p>For more information see https://tails.boum.org/todo/server_edition/</p>

1762)     </li>

1763)     -->

1764) 

1765)     <a id="torCleanup"></a>

1766)     <li>

1767)     <b>Tor Codebase Cleanup</b>

1768)     <br>

1769)     Effort Level: <i>Low to High, depending on subproject chosen</i>

1770)     <br>

1771)     Skill Level: <i>Medium to High</i>

1772)     <br>

1773)     Likely Mentors: <i>Nick (nickm)</i>

1774)     <p>

1775)     The Tor code is almost 10 years old in places, and we haven't always had
1776)     enough time or wisdom to write things as well as we could have.  Our unit
1777)     test coverage is shamefully low, and the dependency graph of our modules is
1778)     shamefully convoluted . We could use refactoring and unit tests!  Please
1779)     look through the Tor source code and look for ugly or tricky code or
1780)     dependencies -- the uglier and trickier the better -- and think about how
1781)     you could make the code look better, read better, and (subject to testing)
1782)     work better.

1783)     </p>

1784) 

1785)     <p>

1786)     If this is for a fun side-project, it would be great for you to work on
1787)     anything that can be made better and more tested.  For an internship-level
1788)     position, we'd hope that you could find a number of particularly tricky or
1789)     knotty piece of the code to clean up, and aim for resolving the ugliest
1790)     problems, not necessarily the easiest.

1791)     </p>

1792) 

1793)     <p>

1794)     For a big project here, it would be great to pick one of the major
1795)     "submodules" of Tor -- path selection, node discovery, directory authority
1796)     operations, directory service -- and refactor its interface completely, to
1797)     minify and codify its points of contact with the rest of Tor.

1798)     </p>
1799)     </li>

1800) 

1801)     <!--

1802)     <a id="vidaliaStatusEventInterface"></a>

1803)     <li>
1804)     <b>Tor Controller Status Event Interface for Vidalia</b>
1805)     <br>
1806)     Effort Level: <i>Medium</i>
1807)     <br>
1808)     Skill Level: <i>Low to Medium</i>
1809)     <br>

1810)     Likely Mentors: <i>Tomás</i>

1811)     <p>There are a number of status changes inside Tor of which the user may need
1812)     to be informed. For example, if the user is trying to set up his Tor as a
1813)     relay and Tor decides that its ports are not reachable from outside
1814)     the user's network, we should alert the user. Currently, all the user
1815)     gets is a couple of log messages in Vidalia's 'message log' window, which they
1816)     likely never see since they don't receive a notification that something
1817)     has gone wrong. Even if the user does actually look at the message log,
1818)     most of the messages make little sense to the novice user.</p>
1819)     <p>Tor has the ability to inform Vidalia of many such status
1820)     changes, and we recently implemented support for a couple of these
1821)     events. Still, there are many more status events which the user should
1822)     be informed of, and we need a better UI for actually displaying them
1823)     to the user.</p>
1824)     <p>The goal of this project then is to design and implement a UI for
1825)     displaying Tor status events to the user. For example, we might put a
1826)     little badge on Vidalia's tray icon that alerts the user to new status
1827)     events they should look at. Double-clicking the icon could bring up a
1828)     dialog that summarizes recent status events in simple terms and maybe
1829)     suggests a remedy for any negative events if they can be corrected by
1830)     the user. Of course, this is just an example and one is free to
1831)     suggest another approach.</p>
1832)     <p>A person undertaking this project should have good UI design and layout
1833)     skills and some C++ development experience. Previous experience with Qt and
1834)     Qt's Designer will be very helpful, but are not required. Some
1835)     English writing ability will also be useful, since this project will
1836)     likely involve writing small amounts of help documentation that should
1837)     be understandable by non-technical users. Bonus points for some graphic
1838)     design/Photoshop fu, since we might want/need some shiny new icons too.</p>
1839)     </li>

1840)     -->

1841) 

1842)     <!--

1843)     <a id="orbot-torbutton"></a>

1844)     <li>

1845)     <b>TorButton for Mobile Firefox 4 or Custom Browser on Android</b>

1846)     <br>

1847)     Effort Level: <i>High</i>

1848)     <br>
1849)     Skill Level: <i>High</i>
1850)     <br>

1851)     Likely Mentors: <i>Jake</i>

1852)     <p>Initial work has been done on implementing a proxy-setting add-on for
1853)     Firefox on Android (see <a
1854)     href="https://github.com/guardianproject/ProxyMob">ProxyMob</a>), but a
1855)     full port of TorButton needs to be done (dependent upon Firefox 4 port of
1856)     TorButton). The other approach is to implement a custom &quot;Tor
1857)     Browser&quot; based on Firefox or Webkit browser. See <a
1858)     href="http://code.google.com/p/torora/wiki/Android">Torora</a> for progress
1859)     on this so far.</p>

1860)     </li>

1861)     -->

1862) 

1863)     <!--

1864)     <a id="vidalia-hidden-service-panel"></a>

1865)     <li>

1866)     <b>Torrc plugin and improved hidden service configuration panel</b>

1867)     <br>
1868)     Effort Level: <i>Medium</i>
1869)     <br>
1870)     Skill Level: <i>Medium</i>
1871)     <br>

1872)     Likely Mentors: <i>Tomás</i>

1873)     <p>Vidalia's configuration handling has changed in the alpha branch. Now
1874)     every Tor option is saved in the torrc file. With that change, the
1875)     Hidden Service configuration panel was removed due to its specificity
1876)     and its multiple bugs.</p>

1877) 

1878)     <p>The idea would be to provide the new Torrc class' functionality to the
1879)     Plugin Engine and with that, create a better Hidden Service
1880)     configuration panel as a plugin.</p>

1881) 

1882)     <p>A person undertaking this project should have good UI design, layout
1883)     skills and some C++ development experience. Previous experience with Qt
1884)     and Qt's Designer will be very helpful, but are not required. Javascript
1885)     knowledge is a plus, but it shouldn't be a problem if the person
1886)     complies with the previous requirements.</p>

1887)     </li>

1888)     -->

1889) 

1890)     <!--

1891)     <a id="usabilityTesting"></a>

1892)     <li>

1893)     <b>Usability testing of Tor</b>

1894)     <br>

1895)     Effort Level: <i>Medium</i>

1896)     <br>

1897)     Skill Level: <i>Low to Medium</i>

1898)     <br>

1899)     Likely Mentors: <i>Andrew</i>

1900)     <p>

1901)     Especially the browser bundle, ideally amongst our target demographic.
1902)     That would help a lot in knowing what needs to be done in terms of bug
1903)     fixes or new features. We get this informally at the moment, but a more
1904)     structured process would be better.

1905)     </p>

1906) 

1907)     <p>

1908)     Please note that since this isn't a coding project, it isn't suitable for
1909)     Google Summer of Code.

1910)     </p>
1911)     </li>

1912)     -->

1913) 

1914)     <!--

1915)     <a id="docUpdate"></a>

1916)     <li>

1917)     <b>Website and video documentation update</b>

1918)     <br>
1919)     Effort Level: <i>Medium</i>
1920)     <br>
1921)     Skill Level: <i>Medium</i>
1922)     <br>

1923)     Likely Mentors: <i>Runa, Karen</i>

1924)     <p>

1925)     Identify outdated and/or weak pages on torproject.org and re-write the
1926)     documentation to appeal to a less technical, more goal-oriented
1927)     audience. Create a number of 3-5 minute videos with graphically
1928)     portray the written documentation. See
1929)     <a href="https://trac.torproject.org/projects/tor/query?component=Website&status=new">the
1930)     website tickets</a> for more information and a starting point.

1931)     </p>
1932)     </li>

1933)     -->

1934) 

1935)     <li>
1936)     <b>Bring up new ideas!</b>

1937)     <br>

1938)     Don't like any of these? Look at the <a

1939)     href="/press/presskit/2008-12-19-roadmap-full.pdf">Tor development

1940)     roadmap</a> for more ideas, or just try out Tor, Vidalia, and Torbutton,
1941)     and find out what you think needs fixing.

1942)     Some of the <a href="<spectree>proposals">current proposals</a>

1943)     might also be short on developers.
1944)     </li>

1945) 

1946)     </ol>

1947) 

1948)     <a id="OtherCoding"></a>
1949)     <h2><a class="anchor" href="#OtherCoding">Other Coding and Design related ideas</a></h2>
1950)     <ol>
1951)     <li>Tor relays don't work well on Windows XP. On
1952)     Windows, Tor uses the standard <tt>select()</tt> system
1953)     call, which uses space in the non-page pool. This means
1954)     that a medium sized Tor relay will empty the non-page pool, <a

1955)     href="<wiki>doc/WindowsBufferProblems">causing

1956)     havoc and system crashes</a>. We should probably be using overlapped IO
1957)     instead. One solution would be to teach <a
1958)     href="http://www.monkey.org/~provos/libevent/">libevent</a> how to use
1959)     overlapped IO rather than select() on Windows, and then adapt Tor to
1960)     the new libevent interface. Christian King made a
1961)     <a href="https://svn.torproject.org/svn/libevent-urz/trunk/">good
1962)     start</a> on this in the summer of 2007.</li>

1963) 

1964)     <li>We need to actually start building our <a href="<page
1965)     docs/documentation>#DesignDoc">blocking-resistance design</a>. This involves
1966)     fleshing out the design, modifying many different pieces of Tor, adapting
1967)     <a href="<page projects/vidalia>">Vidalia</a> so it supports the
1968)     new features, and planning for deployment.</li>

1969) 

1970)     <li>We need a flexible simulator framework for studying end-to-end
1971)     traffic confirmation attacks. Many researchers have whipped up ad hoc
1972)     simulators to support their intuition either that the attacks work
1973)     really well or that some defense works great. Can we build a simulator
1974)     that's clearly documented and open enough that everybody knows it's
1975)     giving a reasonable answer? This will spur a lot of new research.
1976)     See the entry <a href="#Research">below</a> on confirmation attacks for
1977)     details on the research side of this task &mdash; who knows, when it's
1978)     done maybe you can help write a paper or three also.</li>

1979) 

1980)     <li>Tor 0.1.1.x and later include support for hardware crypto
1981)     accelerators via OpenSSL. It has been lightly tested and is
1982)     possibly very buggy.  We're looking for more rigorous testing,

1983)     performance analysis, and optimally, code fixes to OpenSSL and

1984)     Tor if needed.</li>

1985) 

1986)     <li>Write a <a
1987)     href="https://secure.wikimedia.org/wikipedia/en/wiki/Fuzz_testing">fuzzer</a>
1988)     for Tor to discover security vulnerabilities. Determine if there
1989)     are good fuzzing frameworks out there for what we want. Win fame by

1990)     getting credit when we put out a new release because of you!</li>

1991) 

1992)     <li>Tor uses TCP for transport and TLS for link
1993)     encryption. This is nice and simple, but it means all cells
1994)     on a link are delayed when a single packet gets dropped, and
1995)     it means we can only reasonably support TCP streams. We have a <a

1996)     href="<page docs/faq>#TransportIPnotTCP">list

1997)     of reasons why we haven't shifted to UDP transport</a>, but it would
1998)     be great to see that list get shorter. We also have a proposed <a

1999)     href="<specblob>proposals/100-tor-spec-udp.txt">specification

2000)     for Tor and
2001)     UDP</a> &mdash; please let us know what's wrong with it.</li>

2002) 

2003)     <li>We're not that far from having IPv6 support for destination addresses
2004)     (at exit nodes). If you care strongly about IPv6, that's probably the
2005)     first place to start.</li>

2006) 

2007)     <li>We need a way to generate the website diagrams (for example, the "How
2008)     Tor Works" pictures on the <a href="<page about/overview>">overview page</a>
2009)     from source, so we can translate them as UTF-8 text rather than edit
2010)     them by hand with Gimp. We might want to
2011)     integrate this as an wml file so translations are easy and images are
2012)     generated in multiple languages whenever we build the website.</li>

2013) 

2014)     <li>How can we make the various LiveCD/USB systems easier
2015)     to maintain, improve, and document?  One example is <a

2016)     href="https://tails.boum.org/">The Amnesic Incognito Live

2017)     System</a>.
2018)     </li>

2019) 

2020)     <li>
2021)     Another anti-censorship project is to try to make Tor
2022)     more scanning-resistant.  Right now, an adversary can identify <a

2023)     href="<specblob>proposals/125-bridges.txt">Tor bridges</a>

2024)     just by trying to connect to them, following the Tor protocol,
2025)     and seeing if they respond.  To solve this, bridges could <a
2026)     href="<svnprojects>design-paper/blocking.html#tth_sEc9.3">act like
2027)     webservers</a> (HTTP or HTTPS) when contacted by port-scanning tools,
2028)     and not act like bridges until the user provides a bridge-specific key.
2029)     To start, check out Shane Pope's <a
2030)     href="http://dl.dropbox.com/u/37735/index.html">thesis and prototype</a>.
2031)     </li>

2032) 

2033)     </ol>

2034) 

2035)     <a id="Research"></a>
2036)     <h2><a class="anchor" href="#Research">Research</a></h2>
2037)     <ol>
2038)     <li>The "end-to-end traffic confirmation attack":
2039)     by watching traffic at Alice and at Bob, we can <a
2040)     href="http://freehaven.net/anonbib/#danezis:pet2004">compare
2041)     traffic signatures and become convinced that we're watching the same
2042)     stream</a>. So far Tor accepts this as a fact of life and assumes this
2043)     attack is trivial in all cases. First of all, is that actually true? How
2044)     much traffic of what sort of distribution is needed before the adversary
2045)     is confident he has won? Are there scenarios (e.g. not transmitting much)
2046)     that slow down the attack? Do some traffic padding or traffic shaping
2047)     schemes work better than others?</li>
2048)     <li>A related question is: Does running a relay/bridge provide additional
2049)     protection against these timing attacks? Can an external adversary that can't
2050)     see inside TLS links still recognize individual streams reliably?
2051)     Does the amount of traffic carried degrade this ability any? What if the
2052)     client-relay deliberately delayed upstream relayed traffic to create a queue
2053)     that could be used to mimic timings of client downstream traffic to make it
2054)     look like it was also relayed? This same queue could also be used for masking
2055)     timings in client upstream traffic with the techniques from <a
2056)     href="http://www.freehaven.net/anonbib/#ShWa-Timing06">adaptive padding</a>,
2057)     but without the need for additional traffic. Would such an interleaving of
2058)     client upstream traffic obscure timings for external adversaries? Would the
2059)     strategies need to be adjusted for asymmetric links? For example, on
2060)     asymmetric links, is it actually possible to differentiate client traffic from
2061)     natural bursts due to their asymmetric capacity? Or is it easier than
2062)     symmetric links for some other reason?</li>
2063)     <li>Repeat Murdoch and Danezis's <a
2064)     href="http://www.cl.cam.ac.uk/~sjm217/projects/anon/#torta">attack from
2065)     Oakland 05</a> on the current Tor network. See if you can learn why it
2066)     works well on some nodes and not well on others. (My theory is that the
2067)     fast nodes with spare capacity resist the attack better.) If that's true,
2068)     then experiment with the RelayBandwidthRate and RelayBandwidthBurst
2069)     options to run a relay that is used as a client while relaying the
2070)     attacker's traffic: as we crank down the RelayBandwidthRate, does the
2071)     attack get harder? What's the right ratio of RelayBandwidthRate to
2072)     actually capacity? Or is it a ratio at all? While we're at it, does a
2073)     much larger set of candidate relays increase the false positive rate
2074)     or other complexity for the attack? (The Tor network is now almost two
2075)     orders of magnitude larger than it was when they wrote their paper.) Be
2076)     sure to read <a href="http://freehaven.net/anonbib/#clog-the-queue">Don't
2077)     Clog the Queue</a> too.</li>
2078)     <li>The "routing zones attack": most of the literature thinks of
2079)     the network path between Alice and her entry node (and between the
2080)     exit node and Bob) as a single link on some graph. In practice,
2081)     though, the path traverses many autonomous systems (ASes), and <a
2082)     href="http://freehaven.net/anonbib/#feamster:wpes2004">it's not uncommon
2083)     that the same AS appears on both the entry path and the exit path</a>.
2084)     Unfortunately, to accurately predict whether a given Alice, entry,
2085)     exit, Bob quad will be dangerous, we need to download an entire Internet
2086)     routing zone and perform expensive operations on it. Are there practical
2087)     approximations, such as avoiding IP addresses in the same /8 network?</li>
2088)     <li>Other research questions regarding geographic diversity consider
2089)     the tradeoff between choosing an efficient circuit and choosing a random
2090)     circuit. Look at Stephen Rollyson's <a
2091)     href="http://swiki.cc.gatech.edu:8080/ugResearch/uploads/7/ImprovingTor.pdf">position
2092)     paper</a> on how to discard particularly slow choices without hurting
2093)     anonymity "too much". This line of reasoning needs more work and more
2094)     thinking, but it looks very promising.</li>
2095)     <li>Tor doesn't work very well when relays have asymmetric bandwidth
2096)     (e.g. cable or DSL). Because Tor has separate TCP connections between
2097)     each hop, if the incoming bytes are arriving just fine and the outgoing
2098)     bytes are all getting dropped on the floor, the TCP push-back mechanisms
2099)     don't really transmit this information back to the incoming streams.
2100)     Perhaps Tor should detect when it's dropping a lot of outgoing packets,
2101)     and rate-limit incoming streams to regulate this itself? I can imagine
2102)     a build-up and drop-off scheme where we pick a conservative rate-limit,
2103)     slowly increase it until we get lost packets, back off, repeat. We
2104)     need somebody who's good with networks to simulate this and help design
2105)     solutions; and/or we need to understand the extent of the performance
2106)     degradation, and use this as motivation to reconsider UDP transport.</li>
2107)     <li>A related topic is congestion control. Is our
2108)     current design sufficient once we have heavy use? Maybe
2109)     we should experiment with variable-sized windows rather
2110)     than fixed-size windows? That seemed to go well in an <a

2111)     href="http://www.psc.edu/index.php/hpn-ssh/638">ssh

2112)     throughput experiment</a>. We'll need to measure and tweak, and maybe
2113)     overhaul if the results are good.</li>
2114)     <li>Our censorship-resistance goals include preventing
2115)     an attacker who's looking at Tor traffic on the wire from <a
2116)     href="<svnprojects>design-paper/blocking.html#sec:network-fingerprint">distinguishing
2117)     it from normal SSL traffic</a>. Obviously we can't achieve perfect
2118)     steganography and still remain usable, but for a first step we'd like to
2119)     block any attacks that can win by observing only a few packets. One of
2120)     the remaining attacks we haven't examined much is that Tor cells are 512
2121)     bytes, so the traffic on the wire may well be a multiple of 512 bytes.
2122)     How much does the batching and overhead in TLS records blur this on the
2123)     wire? Do different buffer flushing strategies in Tor affect this? Could
2124)     a bit of padding help a lot, or is this an attack we must accept?</li>
2125)     <li>Tor circuits are built one hop at a time, so in theory we have the
2126)     ability to make some streams exit from the second hop, some from the
2127)     third, and so on. This seems nice because it breaks up the set of exiting
2128)     streams that a given relay can see. But if we want each stream to be safe,
2129)     the "shortest" path should be at least 3 hops long by our current logic, so
2130)     the rest will be even longer. We need to examine this performance / security
2131)     tradeoff.</li>
2132)     <li>It's not that hard to DoS Tor relays or directory authorities. Are client
2133)     puzzles the right answer? What other practical approaches are there? Bonus
2134)     if they're backward-compatible with the current Tor protocol.</li>
2135)     <li>Programs like <a

2136)     href="<page torbutton/index>">Torbutton</a> aim to hide

2137)     your browser's UserAgent string by replacing it with a uniform answer for
2138)     every Tor user. That way the attacker can't splinter Tor's anonymity set
2139)     by looking at that header. It tries to pick a string that is commonly used
2140)     by non-Tor users too, so it doesn't stand out. Question one: how badly
2141)     do we hurt ourselves by periodically updating the version of Firefox
2142)     that Torbutton claims to be? If we update it too often, we splinter the
2143)     anonymity sets ourselves. If we don't update it often enough, then all the
2144)     Tor users stand out because they claim to be running a quite old version
2145)     of Firefox. The answer here probably depends on the Firefox versions seen
2146)     in the wild. Question two: periodically people ask us to cycle through N
2147)     UserAgent strings rather than stick with one. Does this approach help,
2148)     hurt, or not matter? Consider: cookies and recognizing Torbutton users
2149)     by their rotating UserAgents; malicious websites who only attack certain
2150)     browsers; and whether the answers to question one impact this answer.
2151)     </li>
2152)     <li>How many bridge relays do you need to know to maintain
2153)     reachability? We should measure the churn in our bridges. If there is
2154)     lots of churn, are there ways to keep bridge users more likely to stay
2155)     connected?
2156)     </li>
2157)     </ol>

2158) 

2159)     <p>
2160)     <a href="<page about/contact>">Let us know</a> if you've made progress on any
2161)     of these!
2162)     </p>
2163)   </div>
2164)   <!-- END MAINCOL -->
2165)   <div id = "sidecol">
2166) #include "side.wmi"
2167) #include "info.wmi"
2168)   </div>
2169)   <!-- END SIDECOL -->
2170) </div>
2171) <!-- END CONTENT -->