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

29) 

30)     <a id="Documentation"></a>
31)     <h2><a class="anchor" href="#Documentation">Documentation</a></h2>
32)     <ol>

33)     <li>Help translate the
34) <!-- web page and -->
35)     documentation into other

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

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

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

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

44)     programs that interface with Tor</a>. Which ones are useful in which

45)     situations? Please help us test them out and document your results.</li>
46)     </ol>

47) 

48)     <a id="Advocacy"></a>
49)     <h2><a class="anchor" href="#Advocacy">Advocacy</a></h2>
50)     <ol>

51)     <li>Monitor some of our <a 
52)     href="https://lists.torproject.org/cgi-bin/mailman/listinfo">public mailing 
53)     lists</a>, like <a 
54)     href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-talk">tor-talk</a>, <a 
55)     href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays">tor-relays</a>, <a 
56)     href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-dev">tor-dev</a>, or <a 
57)     href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tbb-dev">tbb-dev</a>, 
58)     and summarize noteworthy exchanges into articles for <a 

59)     href="https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-news">Tor 

60)     Weekly News</a>.</li>

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

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

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

66)     <a

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

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

69)     <li>Create a poster around a theme, such as "Tor for Freedom!"</li>

70)     <li>Create a t-shirt design that incorporates "<a
71)     href="https://check.torproject.org/">Congratulations! You are using
72)     Tor!</a>" in any language.</li>

73)     <li>Spread the word about Tor at a symposium or conference and use these
74)     <a href="https://media.torproject.org/misc/2015-03-tor-brochure/">Tor
75)     brochures</a> in PDF and ODG format and translated to at least ten
76)     different languages as conversation starter.</li>

77)     </ol>

78) 

79) <!--

80)     <a id="gsoc"></a>
81)     <h2><a class="anchor" href="#gsoc">Google Summer of Code</a></h2>
82) 
83)     <p>
84)     Tor is also taking part in this year's <a
85)     href="https://www.google-melange.com/gsoc/homepage/google/gsoc2013">Google
86)     Summer of Code</a>! The criteria for this is a little different - either
87)     gender can apply but you need to be either <a
88)     href="https://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2013/help_page#2._Whos_eligible_to_participate_as_a">a
89)     present student or just graduated</a>.
90)     </p>
91) 
92)     <p>
93)     As mentioned above if you're eligible for either program then please apply
94)     for both! Google Summer of Code is a far, far larger program for us than
95)     OPW so your chances of being applied that way are considerably better.
96)     </p>
97) 
98)     <p>
99)     <b>See our page for <a href="<page about/gsoc>">Google Summer of Code</a>
100)     for more information.</b>
101)     </p>

102) -->

103) 

104)     <a id="Projects"></a>
105)     <h2><a class="anchor" href="#Projects">Projects</a></h2>

106) 

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

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

114)     </p>

115) 

116)     <p>
117)     For a presentation summarizing many of these projects see...
118)     </p>
119) 
120)     <div id="ecosystem_presentation">

121)       <a href="https://www.youtube.com/watch?v=fb6iqZcQsSg">Tor Ecosystem</a> (<a href="https://media.torproject.org/video/2013-11-t3am-damian-johnson.mp4">mp4</a>, <a href="https://svn.torproject.org/svn/projects/presentations/2013-11-t3am-tor-ecosystem.pdf">slides</a>)

122)     </div>
123) 
124)     <br /></br />
125) 

126)     <table id="projects">
127)       <tr>
128)         <th>Name</th>
129)         <th>Category</th>
130)         <th>Language</th>
131)         <th>Activity</th>
132)         <th>Contributors</th>
133)       </tr>

134) 

135)       <tr>
136)         <td><a href="#project-tor">Tor</a></td>
137)         <td>Core</td>
138)         <td>C</td>
139)         <td>Heavy</td>

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

141)       </tr>

142) 

143)       <tr>

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

145)         <td>Bundle</td>

146)         <td>Javascript, XUL, Scripting</td>

147)         <td>Heavy</td>

148)         <td>mikeperry, Pearl Crescent</td>

149)       </tr>
150) 

151)       <tr>

152)         <td><a href="#project-httpseverywhere">HTTPS Everywhere</a></td>
153)         <td>Browser Add-on</td>
154)         <td>Javascript</td>

155)         <td>Heavy</td>

156)         <td>pde, mikeperry</td>

157)       </tr>

158) 

159)       <tr>

160)         <td><a href="#project-nyx">Nyx</a></td>

161)         <td>User Interface</td>
162)         <td>Python, Curses</td>

163)         <td>Heavy</td>

164)         <td>atagar</td>
165)       </tr>

166) 

167)       <tr>

168)         <td><a href="#project-orbot">Orbot</a></td>
169)         <td>User Interface</td>
170)         <td>Java</td>

171)         <td>Moderage</td>

172)         <td>n8fr8</td>

173)       </tr>

174) 

175)       <tr>

176)         <td><a href="#project-tails">Tails</a></td>
177)         <td>OS image</td>
178)         <td>Sys Admin</td>
179)         <td>Heavy</td>
180)         <td><a href="https://tails.boum.org/">#tails</a></td>

181)       </tr>

182) 

183)       <tr>

184)         <td><a href="#project-torramdisk">tor-ramdisk</a></td>
185)         <td>OS image</td>
186)         <td>Sys Admin</td>

187)         <td>Light</td>

188)         <td>blueness</td>
189)       </tr>
190) 

191)       <tr>

192)         <td><a href="#project-torsocks">Torsocks</a></td>
193)         <td>Usability</td>
194)         <td>C</td>

195)         <td>Light</td>

196)         <td>David Goulet</td>

197)       </tr>
198) 

199)       <tr>
200)         <td><a href="#project-tormessenger">Tor Messenger</a></td>
201)         <td>Bundle</td>
202)         <td>JavaScript, XUL, Scripting</td>
203)         <td>Heavy</td>
204)         <td>arlolra, boklm, sukhe</td>
205)       </tr>
206) 

207)       <tr>

208)         <td><a href="#project-torbirdy">TorBirdy</a></td>
209)         <td>Browser Add-on</td>
210)         <td>JavaScript</td>

211)         <td>Light</td>
212)         <td>sukhe</td>

213)       </tr>

214) 

215)       <tr>

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

218)         <td>Python</td>

219)         <td>None</td>

220)         <td>asn</td>

221)       </tr>
222) 

223)       <tr>

224)         <td><a href="#project-flash-proxy">Flash Proxy</a></td>
225)         <td>Client Add-on</td>
226)         <td>Python, JavaScript, Go</td>

227)         <td>Light</td>

228)         <td>dcf, infinity0, Arlo Breault</td>

229)       </tr>

230) 

231)       <tr>

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

233)         <td>Simulator</td>

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

235)         <td>Heavy</td>

236)         <td>robgjansen</td>
237)       </tr>

238) 

239)       <tr>
240)         <td><a href="#project-chutney">Chutney</a></td>
241)         <td>Simulator</td>
242)         <td>Python</td>
243)         <td>Light</td>
244)         <td>nickm</td>
245)       </tr>
246) 

247)       <tr>

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

249)         <td>Library</td>
250)         <td>Python</td>

251)         <td>Heavy</td>

252)         <td>atagar</td>

253)       </tr>

254) 

255)       <tr>

256)         <td><a href="#project-txtorcon">Txtorcon</a></td>
257)         <td>Library</td>
258)         <td>Python, Twisted</td>

259)         <td>Light</td>

260)         <td>meejah</td>
261)       </tr>

262) 

263)       <tr>

264)         <td><a href="#project-tlsdate">Tlsdate</a></td>
265)         <td>Utility</td>
266)         <td>C</td>

267)         <td>None</td>

268)         <td>ioerror</td>
269)       </tr>

270) 

271)       <tr>

272)         <td><a href="#project-metrics">Metrics</a></td>
273)         <td>Client Service</td>
274)         <td>Java</td>

275)         <td>Moderate</td>

276)         <td>karsten</td>
277)       </tr>

278) 

279)       <tr>

280)         <td><a href="#project-atlas">Atlas</a></td>
281)         <td>Client Service</td>
282)         <td>JavaScript</td>

283)         <td>None</td>

284)         <td></td>

285)       </tr>

286) 

287)       <tr>
288)         <td><a href="#project-globe">Globe</a></td>
289)         <td>Client Service</td>
290)         <td>JavaScript</td>

291)         <td>None</td>
292)         <td></td>

293)       </tr>
294) 

295)       <tr>

296)         <td><a href="#project-compass">Compass</a></td>
297)         <td>Client Service</td>
298)         <td>Python</td>

299)         <td>None</td>
300)         <td></td>

301)       </tr>

302) 

303)       <tr>

304)         <td><a href="#project-onionoo">Onionoo</a></td>
305)         <td>Backend Service</td>

306)         <td>Java</td>

307)         <td>Light</td>

308)         <td>karsten</td>

309)       </tr>
310) 

311)       <tr>
312)         <td><a href="#project-exitmap">ExitMap</a></td>
313)         <td>Backend Service</td>
314)         <td>Python</td>

315)         <td>Light</td>

316)         <td>phw</td>
317)       </tr>
318) 

319)       <tr>
320)         <td><a href="#project-doctor">DocTor</a></td>
321)         <td>Backend Service</td>
322)         <td>Python</td>

323)         <td>Light</td>

324)         <td>atagar</td>
325)       </tr>
326) 

327)       <tr>

328)         <td><a href="#project-weather">Weather</a></td>
329)         <td>Client Service</td>
330)         <td>Python</td>

331)         <td>None</td>

332)         <td>kaner</td>
333)       </tr>

334) 

335)       <tr>

336)         <td><a href="#project-gettor">GetTor</a></td>
337)         <td>Client Service</td>
338)         <td>Python</td>

339)         <td>Light</td>
340)         <td>ilv</td>

341)       </tr>

342) 

343)       <tr>

344)         <td><a href="#project-torcheck">TorCheck</a></td>
345)         <td>Client Service</td>

346)         <td>Go</td>

347)         <td>None</td>

348)         <td>Arlo</td>

349)       </tr>

350) 

351)       <tr>

352)         <td><a href="#project-bridgedb">BridgeDB</a></td>
353)         <td>Backend Service</td>
354)         <td>Python</td>

355)         <td>Light</td>

356)         <td>isis</td>

357)       </tr>

358) 

359)       <tr>

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

361)         <td>Scanner</td>
362)         <td>Python</td>

363)         <td>Moderate</td>

364)         <td>hellais, aagbsn</td>

365)       </tr>
366) 

367)       <tr>
368)         <td><a href="#project-torps">TorPS</a></td>
369)         <td>Backend Service</td>
370)         <td>Python</td>

371)         <td>None</td>

372)         <td>Aaron Johnson</td>
373)       </tr>
374) 

375)       <tr>

376)         <td><a href="#project-torflow">TorFlow</a></td>
377)         <td>Backend Service</td>
378)         <td>Python</td>

379)         <td>None</td>

380)         <td>aagbsn</td>

381)       </tr>

382) 

383)       <tr>

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

386)         <td>Python</td>

387)         <td>Heavy</td>

388)         <td>evilaliv3, hellais</td>

389)       </tr>

390) 

391)       <tr>

392)         <td><a href="#project-anonbib">Anonbib</a></td>
393)         <td>Website</td>

394)         <td>Python</td>

395)         <td>Light</td>

396)         <td>arma, nickm</td>
397)       </tr>

398)     </table>

399) 

400)     <sub>
401)     * Project is still in an alpha state.
402)     </sub>

403) 

404)     <br /><br />

405) 

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

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

409)     tracker</a>)</h3>

410) 

411)     <p>
412)     Central project, providing the core software for using and participating in
413)     the Tor network. Numerous people contribute to the project to varying
414)     extents, but the chief architects are Nick Mathewson and Roger Dingledine.
415)     </p>

416) 

417)     <p>
418)     <b>Project Ideas:</b><br />

419)     <i><a href="#torCleanup">Tor Codebase Cleanup</a></i><br />

420)     <i><a href="#improveTorTestCoverage">Improve test coverage in Tor</a></i><br />
421)     <i><a href="#useMoreCores">Have the Tor daemon use more cores</a></i><br />
422)     <i><a href="#improveHiddenServices">Help improve Tor hidden services</a></i><br />

423)     <i><a href="#improvedDnsSupport">Improved DNS support for Tor</a></i>

424)     </p>

425) 

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

427)     <h3><a href="<page projects/torbrowser>">Tor Browser</a> (<a

428)     href="https://gitweb.torproject.org/tor-browser.git">code</a>, <a

429)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Torbutton&component=Tor+Launcher&component=Tor+Browser&component=Tor+bundles%2Finstallation&col=id&col=summary&col=status&col=owner&col=type&col=priority&col=milestone&order=priority">bug

430)     tracker</a>, <a href="https://www.torproject.org/projects/torbrowser/design/">design doc</a>)</h3>
431) 

432)     <p>

433)     Tor Browser is an easy-to-use, portable package of Tor, HTTPS-Everywhere, 
434)     NoScript, TorLauncher, Torbutton, and a Firefox fork, all  preconfigured 
435)     to work together out of
436)     the box. The modified copy of Firefox aims to resolve the

437)     privacy and security issues in mainline version.

438)     </p>

439) 

440)     <p>
441)     <b>Project Ideas:</b><br />
442)     <i><a href="#panopticlick">Panopticlick</a></i><br />
443)     </p>
444) 

445)     <a id="project-httpseverywhere"></a>
446)     <h3><a href="https://www.eff.org/https-everywhere">HTTPS Everywhere</a> (<a
447)     href="https://gitweb.torproject.org/https-everywhere.git">code</a>, <a
448)     href="https://trac.torproject.org/projects/tor/report/19">bug

449)     tracker</a>)</h3>

450) 

451)     <p>

452)     HTTPS Everywhere is a Firefox and Chrome extension that encrypts
453)     your communications with many major websites, making your browsing
454)     more secure.

455)     </p>

456) 

457)     <a id="project-nyx"></a>
458)     <h3><a href="https://www.atagar.com/arm/">Nyx</a> (<a
459)     href="https://gitweb.torproject.org/nyx.git">code</a>, <a

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

462) 

463)     <p>

464)     Nyx (previously <i>arm</i>) is a terminal status monitor for Tor

465)     intended for command-line aficionados, ssh connections, and anyone with a
466)     tty terminal. This works much like top does for system usage, providing
467)     real time statistics for bandwidth, resource usage, connections, and quite
468)     a bit more.

469)     </p>

470) 

471)     <p>
472)     <b>Project Ideas:</b><br />
473)     <i><a href="#expand_nyx">Expand Nyx</a></i>
474)     </p>
475) 

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

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

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

481) 
482)     <p>

483)     Provides Tor on the Android platform. The project is under active 
484)     development, updates to latest Tor releases, and working to stay up to 
485)     date with all changes in Android and mobile threats.

486)     </p>
487) 
488)     <a id="project-tails"></a>
489)     <h3><a href="https://tails.boum.org/">The Amnesic Incognito Live System</a> (<a

490)     href="https://git-tails.immerda.ch/tails/">code</a>, <a
491)     href="https://labs.riseup.net/code/projects/tails">bug

492)     tracker</a>, <a href="https://tails.boum.org/doc">documentation</a>, <a
493)     href="https://tails.boum.org/contribute/design/">design</a>, <a
494)     href="https://tails.boum.org/contribute">contribute</a>)</h3>

495) 
496)     <p>
497)     The Amnesic Incognito Live System is a live CD/USB distribution
498)     preconfigured so that everything is safely routed through Tor and leaves no
499)     trace on the local system. This is a merger of the Amnesia and <a
500)     href="http://www.anonymityanywhere.com/incognito/">Incognito</a> projects,
501)     and still under very active development.
502)     </p>
503) 

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

506)     href="https://gitweb.torproject.org/tor-ramdisk.git">code</a>, <a

507)     href="http://opensource.dyc.edu/tor-ramdisk-documentation">documentation</a>)</h3>

508) 
509)     <p>

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

512)     </p>
513) 

514)     <a id="project-torsocks"></a>

515)     <h3>Torsocks (<a

516)     href="https://gitweb.torproject.org/torsocks.git">code</a>, <a
517)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Torify&order=priority">bug
518)     tracker</a>)</h3>
519) 

520)     <p>

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

524)     </p>

525) 

526)     <a id="project-tormessenger"></a>
527)     <h3><a href="https://trac.torproject.org/projects/tor/wiki/doc/TorMessenger">Tor Messenger</a> (<a
528)     href="https://gitweb.torproject.org/tor-messenger-build.git">code</a>, <a
529)     href="https://trac.torproject.org/projects/tor/query?status=!closed&component=Tor+Messenger">bug tracker</a>)</h3>
530) 
531)     <p>
532)     Tor Messenger is a cross-platform chat program that aims to be secure by default and sends all of its traffic over Tor.
533)     </p>
534) 

535)     <a id="project-torbirdy"></a>
536)     <h3>TorBirdy (<a

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

538)     href="https://trac.torproject.org/projects/tor/wiki/torbirdy/dev">bug
539)     tracker</a>)</h3>

540) 

541)     <p>
542)     TorBirdy is Torbutton for Thunderbird and related Mozilla mail clients.
543)     </p>

544) 

545)     <p>
546)     <b>Project Ideas:</b><br />
547)     <i><a href="#makeTorbirdyBetter">Make TorBirdy Better</a></i>
548)     </p>
549) 

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

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

552)     href="https://gitweb.torproject.org/pluggable-transports/obfsproxy.git">code</a>,
553)     <a 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

554)     tracker</a>)</h3>

555) 

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

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

559)     </p>

560) 

561)     <p>
562)     <b>Project Ideas:</b><br />

563)     <i><a href="https://trac.torproject.org/projects/tor/wiki/doc/PluggableTransports#Helpneeded">Various coding tasks</a></i> <br/>

564)     <i><a href="#betterPluggableTransports">Build Better Pluggable Transports</a></i>
565)     </p>
566) 

567)     <a id="project-flash-proxy"></a>
568)     <h3><a href="https://crypto.stanford.edu/flashproxy/">Flash Proxy</a> (<a
569)     href="https://gitweb.torproject.org/flashproxy.git">code</a>, <a
570)     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
571)     tracker</a>)</h3>

572) 

573)     <p>
574)     Pluggable transport using proxies running in web browsers to defeat
575)     address-based blocking.
576)     </p>

577) 

578)     <a id="project-chutney"></a>
579)     <h3>Chutney (<a href="https://gitweb.torproject.org/chutney.git">code</a>,
580)     <a 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=Chutney">bug
581)     tracker</a>)</h3>
582) 
583)     <p>
584)     Integration test suite that spawns a local tor network, checking the
585)     interactions of its components.
586)     </p>
587) 

588)     <a id="project-shadow"></a>
589)     <h3><a href="https://shadow.cs.umn.edu/">Shadow</a> (<a
590)     href="https://github.com/shadow">code</a>, <a
591)     href="https://github.com/shadow/shadow/issues">bug
592)     tracker</a>)</h3>

593) 

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

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

600)     </p>

601) 

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

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

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

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

606)     tracker</a>)</h3>

607) 

608)     <p>

609)     Python controller library for scripts and controller applications using
610)     Tor.

611)     </p>

612) 

613)     <p>
614)     <b>Project Ideas:</b><br />
615)     <i><a href="#descriptor_parsing_in_go">Stem Descriptor Parsing in Go</a></i>
616)     </p>
617) 

618)     <a id="project-txtorcon"></a>

619)     <h3><a href="https://txtorcon.readthedocs.org">Txtorcon</a> (<a

620)     href="https://github.com/meejah/txtorcon">code</a>, <a

621)     href="https://github.com/meejah/txtorcon/issues">bug tracker</a>)</h3>

622) 

623)     <p>
624)     Twisted-based asynchronous Tor control protocol implementation. Includes
625)     unit-tests, examples, state-tracking code and configuration abstraction.
626)     Used by OONI and APAF.
627)     </p>

628) 

629)     <p>
630)     <b>Project Ideas:</b><br />
631)     <i><a href="#txtorcon_use_txaio">Convert txtorcon to use txaio</a></i><br />
632)     <i><a href="#txtorcon_use_pytest">Convert txtorcon to py.test</a></i>
633)     </p>
634) 

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

637) 

638)     <p>
639)     tlsdate: secure parasitic rdate replacement
640)     </p>

641) 

642)     <p>
643)     tlsdate sets the local clock by securely connecting with TLS to remote
644)     servers and extracting the remote time out of the secure handshake. Unlike
645)     ntpdate, tlsdate uses TCP, for instance connecting to a remote HTTPS or TLS
646)     enabled service, and provides some protection against adversaries that try
647)     to feed you malicious time information.
648)     </p>

649) 

650)     <a id="project-metrics"></a>
651)     <h3><a href="https://metrics.torproject.org/">Metrics</a> (code: <a
652)     href="https://gitweb.torproject.org/metrics-db.git">db</a>, <a

653)     href="https://gitweb.torproject.org/metrics-web.git">web</a>)</h3>

654) 

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

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

660)     </p>

661) 

662)     <a id="project-atlas"></a>
663)     <h3><a href="https://atlas.torproject.org/">Atlas</a> (<a
664)     href="https://gitweb.torproject.org/atlas.git">code</a>)</h3>

665) 

666)     <p>
667)     Atlas is a web application to discover Tor relays and bridges. It provides
668)     useful information on how relays are configured along with graphics about

669)     their past usage.

670)     </p>

671) 

672)     <p>

673)     This is the spiritual successor to <a
674)     href="https://gitweb.torproject.org/torstatus.git">TorStatus</a>, the <a

675)     href="https://svn.torproject.org/svn/torstatus/trunk/">original

676)     codebase</a> for which was written in PHP, and rewritten by students from
677)     Wesleyan as Django.

678)     </p>

679) 

680)     <a id="project-globe"></a>
681)     <h3><a href="http://globe.rndm.de/">Globe</a> (<a
682)     href="https://github.com/makepanic/globe">code</a>, <a
683)     href="https://github.com/makepanic/globe/issues">bug tracker</a>)</h3>
684) 
685)     <p>
686)     Globe is a web application that allows you to search for Tor relays and
687)     bridges. It gives you a detailed overview of properties and configurations
688)     of a relay or bridge.
689)     </p>
690) 

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

696) 

697)     <p>
698)     Compass is a web and command line application that filters and
699)     aggregates the Tor relays based on various attributes.
700)     </p>

701) 
702)     <a id="project-onionoo"></a>

703)     <h3><a href="https://onionoo.torproject.org/">Onionoo</a> (<a
704)     href="https://gitweb.torproject.org/onionoo.git">code</a>,
705)     <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>

706) 
707)     <p>
708)     Onionoo is a JSON based protocol to learn information about currently
709)     running Tor relays and bridges.
710)     </p>
711) 

712)     <a id="project-exitmap"></a>
713)     <h3><a href="http://www.cs.kau.se/philwint/spoiled_onions/">ExitMap</a> (<a
714)     href="https://github.com/NullHypothesis/exitmap">code</a>, <a
715)     href="https://github.com/NullHypothesis/exitmap/issues">bug tracker</a>)</h3>
716) 
717)     <p>

718)     Scanner for the Tor network by Philipp Winter to detect malicious and
719)     misconfigured exits. For more information about how it works see his <a
720)     href="http://www.cs.kau.se/philwint/spoiled_onions/pets2014.pdf">Spoiled

721)     Onions</a> research paper.
722)     </p>
723) 

724)     <p>
725)     <b>Project Ideas:</b><br />
726)     <i><a href="#exitmap_improvements">Exitmap Improvements</a></i>
727)     </p>
728) 

729)     <a id="project-doctor"></a>
730)     <h3>DocTor (<a
731)     href="https://gitweb.torproject.org/doctor.git">code</a>, <a
732)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=DocTor&order=priority">bug
733)     tracker</a>)</h3>
734) 
735)     <p>
736)     DocTor is a notification service that monitors newly published descriptor
737)     information for issues. This is primarily a service to help the tor
738)     directory authority operators, but it also checks for a handful of other
739)     issues like sybil attacks.
740)     </p>
741) 

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

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

744)     href="https://gitweb.torproject.org/weather.git">code</a>, <a
745)     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
746)     tracker</a>)</h3>

747) 

748)     <p>
749)     Provides automatic notification to subscribed relay operators when their
750)     relay's unreachable. This underwent a rewrite by the <a
751)     href="http://hfoss.wesleyan.edu/">Wesleyan HFOSS team</a>, which went live
752)     in early 2011.
753)     </p>

754) 

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

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

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

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

760) 

761)     <p>
762)     E-mail autoresponder providing Tor's packages over SMTP. This has been
763)     relatively unchanged for quite a while.
764)     </p>

765) 

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

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

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

769)     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
770)     tracker</a>)</h3>

771) 

772)     <p>

773)     Site for determining if the visitor is using Tor or not.

774)     </p>

775) 

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

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

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

781) 

782)     <p>
783)     Backend bridge distributor, handling the various pools they're distributed
784)     in. This was actively developed until Fall of 2010.
785)     </p>

786) 
787)     <a id="project-ooni-probe"></a>

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

789)     href="https://gitweb.torproject.org/ooni-probe.git">code</a>, <a
790)     href="https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=needs_review&status=new&status=reopened&component=Ooni&order=priority">bug
791)     tracker</a>)</h3>
792) 
793)     <p>
794)     Censorship scanner, checking your local connection for blocked or modified
795)     content.
796)     </p>
797) 

798)     <p>
799)     <b>Project Ideas:</b><br />

800)     <i><a href="#ooniprobePcapsSupport">Add Support for Reporting Pcaps to OoniBackend and OoniProbe</a></i>

801)     </p>
802) 

803)     <a id="project-torps"></a>
804)     <h3>TorPS</a> (<a href="https://github.com/torps/torps">code</a>)</h3>
805) 
806)     <p>
807)     The Tor Path Simulator (TorPS) is a tool for efficiently simulating
808)     path selection in Tor. It chooses circuits and assigns user streams to
809)     those circuits in the same way that Tor does. TorPS is fast enough to
810)     perform thousands of simulations over periods of months.
811)     </p>
812) 

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

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

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

818) 

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

827) 
828)     <a id="project-tor2web"></a>
829)     <h3><a
830)     href="http://wiki.tor2web.org/index.php/Main_Page">Tor2web</a> (<a
831)     href="https://github.com/globaleaks/tor2web-3.0/wiki">code</a>)</h3>
832) 
833)     <p>
834)     Tor2web allows Internet users to browse websites running in <a
835)     href="<page docs/hidden-services>">Tor hidden services</a>. It trades
836)     user anonymity for usability by allowing anonymous content to be
837)     distributed to non-anonymous users.
838)     </p>
839) 

840)     <a id="project-anonbib"></a>
841)     <h3><a
842)     href="http://freehaven.net/anonbib/">Anonymity Bibliography</a> (<a
843)     href="https://gitweb.torproject.org/anonbib.git">code</a>)</h3>
844) 
845)     <p>

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

849)     </p>
850) 

851)     <a id="Coding"></a>
852)     <a id="Summer"></a>

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

854) 

855)     <p>

856)     You may find some of these projects to be good ideas for <a href="<page
857)     about/gsoc>">Google Summer of Code</a> and the <a
858)     href="https://live.gnome.org/OutreachProgramForWomen">Outreach Program for

859)     Women</a>. We have labelled each idea with which of our <a href="<page
860)     about/corepeople>">core developers</a> would be good mentors. If one or
861)     more of these ideas looks promising to you, please <a href="<page
862)     about/contact>">contact us</a> to discuss your plans rather than sending
863)     blind applications. You may also want to propose your own project idea
864)     &mdash; which often results in the best applications.

865)     </p>

866) 

867)     <ol>

868) 

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

870)     <li>

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

872)     <br>

873)     Language: <i>C</i>

874)     <br>

875)     Likely Mentors: <i>David (dgoulet)</i>

876)     <br><br>

877)     <p>

878)     The Tor code is more than 10 years old in places, and we haven't always had

879)     enough time or wisdom to write things as well as we could have.  Our unit
880)     test coverage is shamefully low, and the dependency graph of our modules is
881)     shamefully convoluted . We could use refactoring and unit tests!  Please
882)     look through the Tor source code and look for ugly or tricky code or
883)     dependencies -- the uglier and trickier the better -- and think about how
884)     you could make the code look better, read better, and (subject to testing)
885)     work better.

886)     </p>

887) 

888)     <p>

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

894)     </p>

895) 

896)     <p>

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

901)     </p>

902) 
903)     <p>
904)     <b>As part of your application for this project please identify one of the
905)     thorniest Tor functions and submit a patch refactoring it to be better. If
906)     you find this to be difficult then this likely isn't the project for
907)     you.</b>
908)     </p>

909)     </li>

910) 

911)     <a id="betterPluggableTransports"></a>
912)     <li>
913)     <b>Build Better Pluggable Transports</b>
914)     <br>

915)     Language: <i>C, Python</i>

916)     <br>

917)     Likely Mentors: <i>Ximin (infinity0)</i>

918)     <br><br>

919)     <p>

920)     For Tor users in censored countries, we have a <a

921)     href="<page docs/pluggable-transports>">

922)     pluggable transports</a> framework that uses external programs to bypass
923)     censorship in different ways. Each of these have their own strengths and
924)     weaknesses.

925)     </p>
926) 
927)     <p>

928)     We have deployed <a

929)     href="<page projects/obfsproxy>">obfsproxy</a>, 
930)     <a href="http://crypto.stanford.edu/flashproxy/">flashproxy</a>, 
931)     <a href="http://www.cs.kau.se/philwint/scramblesuit/">scramblesuit</a>, 
932)     <a href="https://trac.torproject.org/projects/tor/wiki/doc/meek">meek</a>,
933)     and <a href="https://fteproxy.org/about">FTE</a> bridges into the main 
934)     Tor Browser.</a>

935)     </p>
936) 
937)     <p>
938)     There are several possible directions for this project. Ideas include:
939)       <ol>
940)         <li>Address gaps or weaknesses in our existing pluggable transports
941)           <ul>
942)             <li>Flashproxy: Add WebRTC support to traverse NATs.</li>
943)             <li>Flashproxy: Improve the facilitator's resistance against DoS
944)             and poisoning attacks.</li>
945)           </ul>
946)         </li>

947)         <li>Finish and release our pluggable transport combiner, that chains
948)         several transports together to take advantage of orthogonal types of
949)         blocking resistance.</li>

950)         <li>Improve the UX for selecting the appropriate pluggable transport in

951)         the new Tor Browser, whilst maintaining user security.</li>

952)         <li>Implement a new pluggable transport that resists blocking in a
953)         novel way.
954)         <ul>
955)           <li>Impersonate a voice-over-IP protocol</li>
956)           <li>Impersonate HTTP <a
957)           href="http://www.cs.utexas.edu/~amir/papers/parrot.pdf">sufficiently
958)           well</a> that traffic will go through a HTTP-only proxy</li>
959)           <li>Implement <a
960)           href="http://cacr.uwaterloo.ca/techreports/2011/cacr2011-21.pdf">scanning
961)           resistance</a></li>
962)         </ul>
963)         </li>
964)       </ol>
965)     </p>
966) 
967)     <p>
968)     Applicants should be familiar with asynchronous/reactive programming, in
969)     particular the <a href="https://twistedmatrix.com/">Twisted framework</a>
970)     or something related. Most of the existing code is written in Python, with
971)     some parts in JavaScript and Go, so you should know at least one of these.
972)     You are invited to talk to us and ask questions, via our mailing lists
973)     or IRC. <b>As part of your application, please contribute a patch that
974)     implements a small feature or fixes a bug related to this area, e.g. <a
975)     href="https://trac.torproject.org/projects/tor/query?status=!closed&component=Pluggable+transport">1</a>,
976)     <a href="https://trac.torproject.org/projects/tor/query?status=!closed&component=Obfsproxy">2</a>,
977)     <a href="https://trac.torproject.org/projects/tor/query?status=!closed&component=Flashproxy">3</a>.
978)     </b>

979)     </p>
980) 

981)     <a id="makeTorbirdyBetter"></a>
982)     <li>
983)     <b>Make TorBirdy Better</b>
984)     <br>

985)     Language: <i>JavaScript, C++</i>

986)     <br>
987)     Likely Mentors: <i>Sukhbir Singh (sukhe), Jacob Appelbaum (ioerror)</i>

988)     <br><br>

989)     <p>
990) TorBirdy is an extension that configures Thunderbird to make connections over
991) the Tor anonymity network. TorBirdy has been under development for quite a
992) while but there are two known leaks that prevent it from being used by a wider

993) audience. As part of this project, you will be working on implementing a HTTP
994) proxy.

995)     </p>
996) 
997)     <p>

998) <b>HTTP Proxy</b> TorBirdy needs a HTTP proxy or a HTTP -&gt; SOCKS5 shim. Please look at

999) ticket <a href="https://trac.torproject.org/projects/tor/ticket/6958">#6958</a>

1000) for more information. Note: this has to be done using JavaScript and without using

1001) an external proxy.
1002)     </p>
1003) 
1004)     <p>
1005) If time permits and you are awesome enough to finish the above two tasks, you

1006) will be working on the <a href="https://trac.torproject.org/projects/tor/query?status=!closed&component=TorBirdy">
1007) remaining TorBirdy tickets</a>.

1008)     </p>
1009) 
1010)     <p>

1011) Applicants should be familiar with JavaScript (and preferably C++ as well). As
1012) part of your application for this project, please submit code samples for
1013) previous JavaScript projects that you have developed, or point us to projects
1014) you have been involved with (links to a public Git/GitHub repository
1015) preferred).  Prior extension development is a big plus and will be given
1016) preference during application ranking.

1017)     </p>
1018) 
1019)     <p>

1020) You may contact the mentors on IRC for more information. (sukhe on #tor-dev, #tor on irc.oftc.net)

1021)     </p>
1022)     </li>
1023) 

1024)     <a id="ooniprobePcapsSupport"></a>
1025)     <li>
1026)     <b>Add Support for Reporting Pcaps to OoniBackend and OoniProbe</b>
1027)     <br>

1028)     Language: <i>Python</i>

1029)     <br>

1030)     Likely Mentors: <i>Arturo (hellais)</i>

1031)     <br><br>

1032)     <p>
1033) The feature should also add support for including only packet capture data that
1034) is relevant to the test being run. This means that the pcap should not contain
1035) all the data sniffed on the users machine, but only that which was generated
1036) and intended to be received by ooniprobe.
1037)     </p>
1038) 
1039)     <p>
1040) This can probably be implemented by setting up a tun/tap device and routing all
1041) the ooniprobe traffic through it and only capturing data sent and received from
1042) that device. The task for the student will also be that of doing research into
1043) what are possible strategies for doing this. <b><a
1044) href="https://trac.torproject.org/projects/tor/ticket/7416">For more
1045) information see ticket 7416.</a></b>
1046)     </p>
1047)     </li>
1048) 

1049)     <a id="improveTorTestCoverage"></a>
1050)     <li>
1051)     <b>Improve test coverage in Tor</b>
1052)     <br>

1053)     Language: <i>C, Python</i>

1054)     <br>

1055)     Likely Mentors: <i>David (dgoulet)</i>

1056)     <br><br>

1057)     <p>
1058) Right now, our unit test coverage with the tests we ship is around 30%
1059) -- only 30% of the executable lines in our source are reached by the
1060) unit tests.  Improving this test coverage could make Tor development
1061) much more reliable.
1062)     </p>
1063) 
1064)     <p>
1065) So we need better unit tests, and we need better integration tests too.
1066)     </p>
1067) 
1068)     <p>

1069) Improving unit tests would involve refactoring functions to be more

1070) testable, and writing a bunch of unit tests.
1071)     </p>
1072) 
1073)     <p>

1074) Improving integration tests would involve refactoring and improving

1075) the "chutney" program that launches a test tor network, and writing a
1076) bunch of tests to see what works and what doesn't work on such a
1077) network.  It could also involve writing tests using the library "<a
1078) href="https://stem.torproject.org/">stem</a>" to script individual clients on a
1079) Chutney network.
1080)     </p>
1081) 
1082)     <p>
1083) To get a feel for how testing works in Tor today, download Tor and
1084) Chutney, and make sure you can build Tor and use Chutney.  See how the
1085) unit tests work by skimming some of the test code in the src/test
1086) subdirectory.  Try computing test coverage (according to the
1087) instructions in the doc/HACKING file.
1088)     </p>
1089) 
1090)     <p>
1091) Also, have a look at the one current integration test that works on
1092) chutney today: it is a shell script distributed with Tor as
1093) src/test/test-tor-network.sh .  We probably don't want to have all of
1094) our integration tests be written as shell scripts, but it's interesting
1095) to see how one works.
1096)     </p>
1097) 
1098)     <p>
1099) If working on designs for an improved or refactored Chutney, watch out for <a
1100) href="http://www.joelonsoftware.com/articles/fog0000000018.html">"archicture
1101) astronautics"</a>: while it's important that we have a well-designed and
1102) maintainable Chutney architecture, it wouldn't be very useful if a good
1103) architecture were the <em>only</em> outcome here: we need tests too.
1104)     </p>
1105) 
1106)     <p>
1107) As part of the application process, please contribute a patch that makes
1108) a non-trivial improvement to chutney, and/or include a new test for some
1109) interesting Tor function. (Please pick a function that isn't completely
1110) easy to test.)
1111)     </p>
1112)     </li>
1113) 
1114)     <a id="useMoreCores"></a>
1115)     <li>
1116)     <b>Have the Tor daemon use more cores</b>
1117)     <br>

1118)     Language: <i>C</i>

1119)     <br>

1120)     Likely Mentors: <i>David (dgoulet)</i>

1121)     <br><br>

1122)     <p>
1123) Right now, if you run a busy Tor server on a multicore computer, most of
1124) the cores are mostly unused.  We have a "cpuworker" mechanism to move
1125) expensive computations into worker threads, but that mechanism is
1126) currently only used for a small fraction of our cryptography.  Moving
1127) more work into the worker threads could improve performance immensely.
1128)     </p>
1129) 
1130)     <p>
1131) So it would be great to parallelize our cryptography more in order to
1132) better handle more cores.  See
1133) <a href="https://trac.torproject.org/projects/tor/wiki/org/projects/Tor/MultithreadedCrypto">MultithreadedCrypto</a>
1134) for some background info, and
1135) <a href="https://trac.torproject.org/projects/tor/ticket/7572">ticket 7572</a> for some subtickets
1136) on our tracker.
1137)     </p>
1138) 
1139)     <p>
1140) (If you're reading through the code to see how it works today, you will
1141) also want to have a look at the new implementation for cpuworkers
1142) described in <a href="https://trac.torproject.org/projects/tor/ticket/9682">9682</a>.)
1143)     </p>
1144) 
1145)     <p>
1146) Completing the implementation of ticket #7572 --which would move our
1147) circuit crypto onto separate threads-- could be a good summer project.
1148) Alternatively, moving all of the signature generation and verification
1149) code onto the cpuworkers could be fun.  In either case, you will have
1150) some important architectural decisions to make about how to minimize
1151) shared data between the main thread and the workers, how to avoid
1152) race conditions between them, and how to test it all to make sure it has
1153) no hidden failure cases.
1154)     </p>
1155) 
1156)     <p>
1157) As part of the application process for this project, please contribute a
1158) nontrivial patch to Tor -- ideally, one that will affect some part of
1159) the codebase that you want to work on.
1160)     </p>
1161)     </li>
1162) 
1163)     <a id="improveHiddenServices"></a>
1164)     <li>
1165)     <b>Help improve Tor hidden services</b>
1166)     <br>

1167)     Language: <i>C</i>

1168)     <br>

1169)     Likely Mentors: <i>David (dgoulet), George (asn)</i>

1170)     <br><br>

1171)     <p>
1172) We're working on a revamp of the entire Tor hidden service design to
1173) improve the security and reliability of the hidden service system.
1174)     </p>
1175) 
1176)     <p>
1177) This is a big project: see

1178) <a href="https://gitweb.torproject.org/torspec.git/tree/proposals/224-rend-spec-ng.txt">proposal

1179) 224</a> for the latest design.  Are you interested in implementing some
1180) part of that?
1181)     </p>
1182) 
1183)     <p>
1184) This is a very ambitious project, so we're deliberately not suggesting
1185) particular sub-topics.  If you're interested in participating, try to
1186) read and understand the <a

1187) href="https://gitweb.torproject.org/torspec.git/tree/rend-spec.txt">existing

1188) design</a> and the design proposal for the new design, and then talk to
1189) us about what part you want to work on.
1190)     </p>
1191) 
1192)     <p>
1193) As part of the application process for this project, please contribute a
1194) nontrivial patch to Tor -- ideally, one that will affect some part of
1195) the codebase that you want to work on.
1196)     </p>
1197)     </li>
1198) 
1199)     <a id="improvedDnsSupport"></a>
1200)     <li>
1201)     <b>Improved DNS support for Tor</b>
1202)     <br>

1203)     Language: <i>C</i>

1204)     <br>

1205)     Likely Mentors: <i>David (dgoulet)</i>

1206)     <br><br>

1207)     <p>
1208) Right now, you can only use Tor's DNS support to look up IPv4 and IPv6
1209) addresses, and to fetch PTR records.  But DNS can do so much more!
1210)     </p>
1211) 
1212)     <p>
1213) <a

1214) href="https://gitweb.torproject.org/torspec.git/tree/proposals/219-expanded-dns.txt">Proposal

1215) 219</a> describes some new cell types that Tor could use to support
1216) more types of DNS over Tor.
1217)     </p>
1218) 
1219)     <p>
1220) To see how Tor implements its existing DNS lookups, start by tracing the
1221) the connection_exit_begin_resolve() function in src/or/connection_edge.c ,
1222) and see how we pass these requests downwards through src/or/dns.c to the
1223) underlying resolver.  It's not too complicated, but there are some
1224) tricky parts to understand.
1225)     </p>
1226) 
1227)     <p>
1228) As part of the application process for this project, please contribute a
1229) nontrivial patch to Tor -- ideally, one that will affect some part of
1230) the codebase that you want to work on.
1231)     </p>
1232)     </li>
1233) 

1234)     <a id="exitmap_improvements"></a>
1235)     <li>
1236)     <b>Exitmap Improvements</b>
1237)     <br>

1238)     Language: <i>Python</i>

1239)     <br>
1240)     Likely Mentors: <i>Philipp (phw)</i>

1241)     <br><br>

1242)     <p>
1243) The Tor Project makes use of the Python tool <a
1244) href="https://gitweb.torproject.org/user/phw/exitmap.git/">Exitmap</a> to
1245) systematically scan for malicious and misbehaving exit relays.  Once such a
1246) relay is found, it is assigned the BadExit flag which prevents clients from
1247) selecting the relay as last hop in their circuit.
1248)     </p>
1249) 
1250)     <p>
1251) Exitmap supports scanning modules which implement a specific scan over
1252) exit relays.  Examples are the DNS module which checks for DNS poisoning
1253) or the patching check module which looks out for tampered file
1254) downloads.
1255)     </p>
1256) 
1257)     <p>
1258) This project is meant to extend exitmap in several ways.  First, it
1259) should be made fully autonomous.  That means that exitmap should be able
1260) to run in the background, periodically fetch new relay descriptors, and
1261) have a smart algorithm which keeps scanning all exit relays
1262) periodically.  Second, exitmap should be able to emulate some user
1263) interaction and dynamically "explore" the web in order to detect
1264) tampering.  Third, unit tests should be added for existing and new code
1265) in order to make the code base more robust.
1266)     </p>
1267)     </li>

1268) 

1269)     <a id="expand_nyx"></a>
1270)     <li>
1271)     <b>Expand Nyx</b>
1272)     <br>

1273)     Language: <i>Python</i>

1274)     <br>
1275)     Likely Mentors: <i>Damian (atagar)</i>

1276)     <br><br>

1277)     <p>
1278) Nyx (previously known as <a href="https://www.atagar.com/arm/">arm</a>) is an
1279) ncurses monitor that provides Tor relay operators...
1280)     </p>
1281) 
1282)     <ul>
1283)       <li><a href="https://www.atagar.com/arm/images/screenshot_page1_full.png">bandwidth graphs and event log</a></li>
1284)       <li><a href="https://www.atagar.com/arm/images/screenshot_page2_full.png">connections</a></li>
1285)       <li><a href="https://www.atagar.com/arm/images/screenshot_page3_full.png">user's torrc</a></li>
1286)       <li><a href="https://www.atagar.com/arm/images/screenshot_configPanel_full.png">config editor</a></li>
1287)     </ul>
1288) 
1289)     <p>
1290) Nyx is presently under development in preparation for its next release. If you
1291) like python, terminals, and collaborating on an active codebase this just might
1292) be the project for you!
1293)     </p>
1294) 
1295)     <p>
1296) Proposals should involve a grab-bag of moderately small improvements you'd like
1297) to make. You're encouraged to brainstorm ideas of their own but here's a few to
1298) get ya started...
1299)     </p>
1300) 
1301)     <ul>
1302)       <li>Bring back the <a href="https://www.atagar.com/arm/images/screenshot_interpretor_full.png">interpreter panel</a>. This panel was dropped when refactoring Nyx, but it would be a neat component to bring back. You can <a href="https://gitweb.torproject.org/nyx.git/tree/src/cli/interpretorPanel.py?h=release">find old code for it in the history</a> and Stem (the library backing Nyx) has a much improved <a href="https://stem.torproject.org/tutorials/down_the_rabbit_hole.html">variant of this we can leverage</a>. This task would involve writing the curses code around Stem's interpreter functions.</li>
1303)       <li>Windows support. Like most curses applications Nyx doesn't run natively on Windows. We've had dozens of users request this and <a href="https://trac.torproject.org/projects/tor/wiki/doc/arm#Windows">it should be possible</a>. This would involve supporting PDCurses and expanding Stem to be able to query the cpu/memory usage of the tor process.</li>
1304)       <li>Unit testing! Nyx has <a href="https://gitweb.torproject.org/nyx.git/tree/test">started adding tests</a> but it's still very minimal. Achieving any substantial code coverage will require us to figure out how to unit test curses components.</li>
1305)       <li>Onionoo provides additional relay information that could enrich our connection panel such as geoip and rdns. Trick is that at present we can only query it on a per-relay basis which would leak our connections (no-go for security). However, if we could get information about all relays in bulk it would sidestep this. For some old thoughts on this see <a href="https://trac.torproject.org/projects/tor/wiki/doc/arm#CircuitDetails">here</a>.</li>
1306)       <li>... and more! Again, don't hesitate to propose ideas of your own.</li>
1307)     </ul>
1308) 
1309)     <p>
1310)     <b>As part of your application for this project please get your hands wet with the codebase by contributing patches for <a href="https://gitweb.torproject.org/nyx.git">Nyx</a> and <a href="https://stem.torproject.org/faq.html#how-do-i-get-started">Stem</a>!</b>
1311)     </p>
1312)     </li>
1313) 

1314)     <a id="descriptor_parsing_in_go"></a>
1315)     <li>
1316)     <b>Stem Descriptor Parsing in Go</b>
1317)     <br>
1318)     Language: <i>Go, Python</i>
1319)     <br>
1320)     Likely Mentors: <i>Damian (atagar), Philipp (phw)</i>
1321)     <br><br>
1322)     <p>
1323)     Tor consists of two parts: the application and a distributed network of a
1324)     few thousand volunteer relays. Information about these relays is public,
1325)     and made up of documents called <b><a href="https://stem.torproject.org/tutorials/mirror_mirror_on_the_wall.html#what-is-a-descriptor">descriptors</a></b>. We have
1326)     <a href="https://stem.torproject.org/tutorials/mirror_mirror_on_the_wall.html#are-there-any-other-parsing-libraries">three libraries capable of reading these documents</a>...
1327)     </p>
1328) 
1329)     <ul>
1330)       <li><b><a href="https://stem.torproject.org/">Stem</a></b> (Python)</li>
1331)       <li><b><a href="https://gitweb.torproject.org/metrics-lib.git/">Metrics-lib</a></b> (Java)</li>
1332)       <li><b><a href="https://gitweb.torproject.org/user/phw/zoossh.git/">Zoossh</a></b> (Go)</li>
1333)     </ul>
1334) 
1335)     <p>
1336)     Stem is the most feature rich but slowest, and conversely Zoossh is fastest
1337)     but limited. But what if Stem used CFFI bindings to do the heavy lifting in
1338)     Go? Could we unify these libraries, getting the feature set of Stem with
1339)     the performance of Zoossh?
1340)     </p>
1341) 
1342)     <p>
1343)     <b>Applicants should be familiar with both Python and Go. As part of your
1344)     application for this project please write a demo CFFI binding for Stem as a
1345)     proof of concept.</b> Bonus points if you <a
1346)     href="https://stem.torproject.org/faq.html#how-do-i-get-started">get your
1347)     hands wet by contributing patches</a>!
1348)     </p>
1349)     </li>
1350) 

1351)     <a id="txtorcon_use_txaio"></a>
1352)     <li>
1353)     <b>Convert txtorcon to use txaio</b>
1354)     <br>
1355)     Language: <i>Python, asyncio, Twisted</i>
1356)     <br>
1357)     Likely Mentors: <i>meejah</i>
1358)     <br><br>
1359)     <p>
1360)     txtorcon is currently supports only Twisted users. Re-working txtorcon
1361)     to use the txaio library would allow users to choose between Twisted and
1362)     asyncio for the client code.
1363)     </p>
1364) 
1365)     <p>
1366)     This would involve fairly extensive refactoring to txtorcon, as it
1367)     currently makes heavy use of @inlineCallbacks which doesn't work with
1368)     txaio. A prospective student should be very familiar with event-based
1369)     programming in general, and be familiar with one of Twisted or asyncio.
1370)     See also: https://github.com/meejah/txtorcon/issues/135
1371)     </p>
1372)     </li>
1373) 
1374)     <a id="txtorcon_use_pytest"></a>
1375)     <li>
1376)     <b>Convert txtorcon to py.test</b>
1377)     <br>
1378)     Language: <i>Python, Twisted</i>
1379)     <br>
1380)     Likely Mentors: <i>meejah</i>
1381)     <br><br>
1382)     <p>
1383)     Currently txtorcon uses the built-in "unittest" module, as well as
1384)     Twisted's Deferred-respecting extensions on top. However, meejah has
1385)     found py.test's "fixture" approach to be much more powerful in other
1386)     situations.
1387)     </p>
1388) 
1389)     <p>
1390)     This project would be to port at least part of txtorcon's test-suite to
1391)     use py.test style tests and fixtures and evaluate: are the tests easier
1392)     to read? are there fewer lines of code? If so, the rest of the suite
1393)     should be ported and txtorcon switched over to use py.test exclusively.
1394)     </p>
1395) 
1396)     <p>
1397)     As some of txtorcon's tests aren't very well-written, this would take a
1398)     prospective student who is very strong in unit-testing knowledge. As
1399)     txtorcon is event-based, familiarity with that style of programming
1400)     (preferrably with Twisted) is ideal.
1401)     See also: https://github.com/meejah/txtorcon/issues/136
1402)     </p>
1403)     </li>
1404) 

1405)     <a id="coniks_in_messenger"></a>
1406)     <li>
1407)     <b>Implement and Integrate CONIKS for Tor Messenger</b>
1408)     <br>

1409)     Language: <i>C, JavaScript</i>

1410)     <br>
1411)     Likely Mentors: <i>Marcela, Arlo (arlolra)</i>

1412)     <br><br>

1413)     <p>
1414) CONIKS is an end-user key management and verification system for end-to-end
1415) secure communication services, which improves upon existing key management
1416) systems by providing both strong security and better usability using a model
1417) called key transparency. CONIKS does this by requiring providers to manage
1418) tamper-evident, publicly-auditable key directories, which contain mappings from
1419) usernames to public keys, on behalf of their users. This design makes it easier
1420) for users (both "default" users and power users) to establish trust since they
1421) don't have to worry about or even see keys, but users also don't have to
1422) trust the provider to be well-behaved because the CONIKS client can run as
1423) part of the secure messaging app and automatically check that the service
1424) provider doesn’t map spurious keys to their users' usernames, and it can
1425) verify that observed name-to-key mappings are consistent with what other
1426) clients in the system are seeing. Unlike existing key transparency solutions,
1427) CONIKS also provides strong privacy guarantees by employing cryptographic
1428) primitives for robust data obfuscation.
1429)     </p>
1430) 
1431)     <p>
1432) The CONIKS system design, protocols, and proof-of-concept are described in
1433) great detail in the <a
1434) href="https://www.usenix.org/system/files/conference/usenixsecurity15/sec15-paper-melara.pdf">CONIKS
1435) research paper</a>, and basic reference implementations of a CONIKS key server
1436) and a CONIKS client are avialable on <a
1437) href="https://github.com/coniks-sys/coniks-ref-implementation">Github</a>.
1438)     </p>
1439) 
1440)     <p>
1441) This project has two main components: (1) designing and implementing a CONIKS
1442) key server tailored to Tor Messenger users, and (2) building a CONIKS client
1443) which integrates with the Tor Messenger client. One challenge the applicant
1444) will face is ensuring that the key server design is efficient and scalable for
1445) large volumes of users, concurrent traffic and guarantees this scalability even
1446) as Tor Messenger's user base grows. On the client side, the main challenges
1447) will be to focus on space efficiency as well as minimizing computational
1448) overhead when implementing the CONIKS consistency checks, and determining how
1449) to best communicate CONIKS consistency check results to users in the UI. Since
1450) Tor Messenger does not hand out online identities per se, as most online
1451) communication services do (like, say, Twitter, in which each user has a unique
1452) handle), the CONIKS key server for Tor Messenger will have to map usernames
1453) from third-party communication services to the encryption keys used in Tor
1454) Messenger. One additional important challenge that the applicant will have to
1455) help address is ensuring that each such third-party username remains unique in
1456) the Tor Messenger space and that such external, third-party identities are
1457) indeed controlled by the expected user of that third-party communication
1458) service.
1459)     </p>
1460) 
1461)     <p>
1462) Some design and implementation questions have been discussed in <a
1463) href="https://trac.torproject.org/projects/tor/ticket/17961">Ticket #17961</a>.
1464)     </p>
1465) 
1466)     <p>
1467) The applicant should have some familiarity with well-known crypto primitives
1468) and algorithms, as well as have a basic understanding of the key transparency
1469) model. Client side integration will require some basic use of JavaScript.
1470) Consider submitting a patch for <a
1471) href="https://github.com/arlolra/ctypes-otr/issues">one of the open key
1472) verification issues</a> as part of the application process.
1473)     </p>
1474)     </li>

1475) 
1476)     <a id="panopticlick"></a>
1477)     <li>
1478)     <b>Panopticlick</b>
1479)     <br>
1480)     Likely Mentors: <i>Georg (GeKo)</i>
1481)     <p>
1482) 
1483) The <a href="https://panopticlick.eff.org">Panopticlick project by the EFF</a>
1484) revolutionized how people think about <a
1485) href="https://panopticlick.eff.org/browser-uniqueness.pdf">browser
1486) fingerprinting</a>, both by developing tests and metrics to measure browser
1487) fingerprintability, and by crowdsourcing the evaluation and contribution of
1488) individual browser features to overall fingerprintability.
1489) 
1490)     </p>
1491)     <p>
1492) 
1493) Unfortunately, the way Panopticlick is designed <a
1494) href="https://blog.torproject.org/blog/effs-panopticlick-and-torbutton">makes
1495) it difficult</a> to evaluate defenses to browser fingerprinting, especially
1496) for browsers with a relatively small userbase such as Tor Browser. This is
1497) because any approach we take to reduce fingerprinting automatically makes our
1498) users more distinct from the previous users who submitted their fingerprint
1499) data to the EFF. Indeed, it is also impossible to ever expect that users of
1500) one browser will ever be able to blend in with users of another browser
1501) (Chrome users will always be distinguishable from Firefox users for example,
1502) based on feature set alone).
1503) 
1504)    </p>
1505)    <p>
1506) 
1507) To address this, we would like to have <a
1508) href="https://trac.torproject.org/projects/tor/ticket/6119">our own
1509) fingerprint test suite</a> to evaluate the fingerprintability of each browser
1510) feature for users running a specific Tor Browser version. There are also <a
1511) href="https://trac.torproject.org/projects/tor/query?keywords=~tbb-fingerprinting">additional
1512) fingerprinting tests</a> we can add beyond those deployed by Panopticlick.
1513)    </p>
1514)    <p>
1515) 
1516) For this project, the student would develop a website that users can
1517) voluntarily visit to test and record their Tor Browser fingerprint.  The user
1518) should get feedback on how she performed and the test results should be
1519) available in a machine readable format (e.g. JSON), broken down by Tor Browser
1520) version.  In a second step one could think about adding more sophisticated
1521) tests or supporting other browser vendors that might want to test the
1522) uniformity amongst their userbase as well. Of course, results from each
1523) browser would also need to be broken down by both browser implementation and
1524) version, so that results would only reflect the population of that specific
1525) implementation.
1526) 
1527)     </p>
1528)     </li>
1529) 

1530) <!--
1531)     <a id=""></a>
1532)     <li>
1533)     <b></b>
1534)     <br>

1535)     Language: <i>Python</i>

1536)     <br>
1537)     Likely Mentors: <i>Damian (atagar)</i>

1538)     <br><br>

1539)     <p>
1540) 
1541)     </p>
1542) 
1543)     <p>
1544) 
1545)     </p>
1546)     </li>
1547) -->
1548) 

1549)     <li>
1550)     <b>Bring up new ideas!</b>

1551)     <br>

1552)     Don't like any of these? Look at the <a

1553)     href="/press/presskit/2008-12-19-roadmap-full.pdf">Tor development

1554)     roadmap</a> for more ideas, or just try out Tor and Tor Browser,

1555)     and find out what you think needs fixing.

1556)     Some of the <a href="<spectree>proposals">current proposals</a>

1557)     might also be short on developers.
1558)     </li>

1559) 

1560)     </ol>

1561) 

1562)     <a id="OtherCoding"></a>
1563)     <h2><a class="anchor" href="#OtherCoding">Other Coding and Design related ideas</a></h2>
1564)     <ol>
1565)     <li>Tor relays don't work well on Windows XP. On
1566)     Windows, Tor uses the standard <tt>select()</tt> system
1567)     call, which uses space in the non-page pool. This means
1568)     that a medium sized Tor relay will empty the non-page pool, <a

1569)     href="<wiki>doc/WindowsBufferProblems">causing

1570)     havoc and system crashes</a>. We should probably be using overlapped IO
1571)     instead. One solution would be to teach <a
1572)     href="http://www.monkey.org/~provos/libevent/">libevent</a> how to use
1573)     overlapped IO rather than select() on Windows, and then adapt Tor to
1574)     the new libevent interface. Christian King made a
1575)     <a href="https://svn.torproject.org/svn/libevent-urz/trunk/">good
1576)     start</a> on this in the summer of 2007.</li>

1577) 

1578)     <li>We need a flexible simulator framework for studying end-to-end
1579)     traffic confirmation attacks. Many researchers have whipped up ad hoc
1580)     simulators to support their intuition either that the attacks work
1581)     really well or that some defense works great. Can we build a simulator
1582)     that's clearly documented and open enough that everybody knows it's
1583)     giving a reasonable answer? This will spur a lot of new research.
1584)     See the entry <a href="#Research">below</a> on confirmation attacks for
1585)     details on the research side of this task &mdash; who knows, when it's
1586)     done maybe you can help write a paper or three also.</li>

1587) 

1588)     <li>Tor 0.1.1.x and later include support for hardware crypto
1589)     accelerators via OpenSSL. It has been lightly tested and is
1590)     possibly very buggy.  We're looking for more rigorous testing,

1591)     performance analysis, and optimally, code fixes to OpenSSL and

1592)     Tor if needed.</li>

1593) 

1594)     <li>Write a <a
1595)     href="https://secure.wikimedia.org/wikipedia/en/wiki/Fuzz_testing">fuzzer</a>
1596)     for Tor to discover security vulnerabilities. Determine if there
1597)     are good fuzzing frameworks out there for what we want. Win fame by

1598)     getting credit when we put out a new release because of you!</li>

1599) 

1600)     <li>Tor uses TCP for transport and TLS for link
1601)     encryption. This is nice and simple, but it means all cells
1602)     on a link are delayed when a single packet gets dropped, and
1603)     it means we can only reasonably support TCP streams. We have a <a

1604)     href="<page docs/faq>#TransportIPnotTCP">list

1605)     of reasons why we haven't shifted to UDP transport</a>, but it would
1606)     be great to see that list get shorter. We also have a proposed <a

1607)     href="<specblob>proposals/100-tor-spec-udp.txt">specification

1608)     for Tor and
1609)     UDP</a> &mdash; please let us know what's wrong with it.</li>

1610) 

1611)     <li>We're not that far from having IPv6 support for destination addresses
1612)     (at exit nodes). If you care strongly about IPv6, that's probably the
1613)     first place to start.</li>

1614) 

1615)     <li>We need a way to generate the website diagrams (for example, the "How
1616)     Tor Works" pictures on the <a href="<page about/overview>">overview page</a>
1617)     from source, so we can translate them as UTF-8 text rather than edit
1618)     them by hand with Gimp. We might want to
1619)     integrate this as an wml file so translations are easy and images are
1620)     generated in multiple languages whenever we build the website.</li>

1621) 

1622)     <li>How can we make the various LiveCD/USB systems easier
1623)     to maintain, improve, and document?  One example is <a

1624)     href="https://tails.boum.org/">The Amnesic Incognito Live

1625)     System</a>.
1626)     </li>

1627) 

1628)     <li>
1629)     Another anti-censorship project is to try to make Tor
1630)     more scanning-resistant.  Right now, an adversary can identify <a

1631)     href="<specblob>proposals/125-bridges.txt">Tor bridges</a>

1632)     just by trying to connect to them, following the Tor protocol,
1633)     and seeing if they respond.  To solve this, bridges could <a
1634)     href="<svnprojects>design-paper/blocking.html#tth_sEc9.3">act like
1635)     webservers</a> (HTTP or HTTPS) when contacted by port-scanning tools,
1636)     and not act like bridges until the user provides a bridge-specific key.
1637)     To start, check out Shane Pope's <a
1638)     href="http://dl.dropbox.com/u/37735/index.html">thesis and prototype</a>.
1639)     </li>

1640) 

1641)     </ol>

1642) 

1643)     <a id="Research"></a>
1644)     <h2><a class="anchor" href="#Research">Research</a></h2>
1645)     <ol>
1646)     <li>The "end-to-end traffic confirmation attack":
1647)     by watching traffic at Alice and at Bob, we can <a
1648)     href="http://freehaven.net/anonbib/#danezis:pet2004">compare
1649)     traffic signatures and become convinced that we're watching the same
1650)     stream</a>. So far Tor accepts this as a fact of life and assumes this
1651)     attack is trivial in all cases. First of all, is that actually true? How
1652)     much traffic of what sort of distribution is needed before the adversary
1653)     is confident he has won? Are there scenarios (e.g. not transmitting much)
1654)     that slow down the attack? Do some traffic padding or traffic shaping
1655)     schemes work better than others?</li>
1656)     <li>A related question is: Does running a relay/bridge provide additional
1657)     protection against these timing attacks? Can an external adversary that can't
1658)     see inside TLS links still recognize individual streams reliably?
1659)     Does the amount of traffic carried degrade this ability any? What if the
1660)     client-relay deliberately delayed upstream relayed traffic to create a queue
1661)     that could be used to mimic timings of client downstream traffic to make it
1662)     look like it was also relayed? This same queue could also be used for masking
1663)     timings in client upstream traffic with the techniques from <a
1664)     href="http://www.freehaven.net/anonbib/#ShWa-Timing06">adaptive padding</a>,
1665)     but without the need for additional traffic. Would such an interleaving of
1666)     client upstream traffic obscure timings for external adversaries? Would the
1667)     strategies need to be adjusted for asymmetric links? For example, on
1668)     asymmetric links, is it actually possible to differentiate client traffic from
1669)     natural bursts due to their asymmetric capacity? Or is it easier than
1670)     symmetric links for some other reason?</li>
1671)     <li>Repeat Murdoch and Danezis's <a
1672)     href="http://www.cl.cam.ac.uk/~sjm217/projects/anon/#torta">attack from
1673)     Oakland 05</a> on the current Tor network. See if you can learn why it
1674)     works well on some nodes and not well on others. (My theory is that the
1675)     fast nodes with spare capacity resist the attack better.) If that's true,
1676)     then experiment with the RelayBandwidthRate and RelayBandwidthBurst
1677)     options to run a relay that is used as a client while relaying the
1678)     attacker's traffic: as we crank down the RelayBandwidthRate, does the
1679)     attack get harder? What's the right ratio of RelayBandwidthRate to
1680)     actually capacity? Or is it a ratio at all? While we're at it, does a
1681)     much larger set of candidate relays increase the false positive rate
1682)     or other complexity for the attack? (The Tor network is now almost two
1683)     orders of magnitude larger than it was when they wrote their paper.) Be
1684)     sure to read <a href="http://freehaven.net/anonbib/#clog-the-queue">Don't
1685)     Clog the Queue</a> too.</li>
1686)     <li>The "routing zones attack": most of the literature thinks of
1687)     the network path between Alice and her entry node (and between the
1688)     exit node and Bob) as a single link on some graph. In practice,
1689)     though, the path traverses many autonomous systems (ASes), and <a
1690)     href="http://freehaven.net/anonbib/#feamster:wpes2004">it's not uncommon
1691)     that the same AS appears on both the entry path and the exit path</a>.
1692)     Unfortunately, to accurately predict whether a given Alice, entry,
1693)     exit, Bob quad will be dangerous, we need to download an entire Internet
1694)     routing zone and perform expensive operations on it. Are there practical
1695)     approximations, such as avoiding IP addresses in the same /8 network?</li>
1696)     <li>Other research questions regarding geographic diversity consider
1697)     the tradeoff between choosing an efficient circuit and choosing a random
1698)     circuit. Look at Stephen Rollyson's <a
1699)     href="http://swiki.cc.gatech.edu:8080/ugResearch/uploads/7/ImprovingTor.pdf">position
1700)     paper</a> on how to discard particularly slow choices without hurting
1701)     anonymity "too much". This line of reasoning needs more work and more
1702)     thinking, but it looks very promising.</li>
1703)     <li>Tor doesn't work very well when relays have asymmetric bandwidth
1704)     (e.g. cable or DSL). Because Tor has separate TCP connections between
1705)     each hop, if the incoming bytes are arriving just fine and the outgoing
1706)     bytes are all getting dropped on the floor, the TCP push-back mechanisms
1707)     don't really transmit this information back to the incoming streams.
1708)     Perhaps Tor should detect when it's dropping a lot of outgoing packets,
1709)     and rate-limit incoming streams to regulate this itself? I can imagine
1710)     a build-up and drop-off scheme where we pick a conservative rate-limit,
1711)     slowly increase it until we get lost packets, back off, repeat. We
1712)     need somebody who's good with networks to simulate this and help design
1713)     solutions; and/or we need to understand the extent of the performance
1714)     degradation, and use this as motivation to reconsider UDP transport.</li>
1715)     <li>A related topic is congestion control. Is our
1716)     current design sufficient once we have heavy use? Maybe
1717)     we should experiment with variable-sized windows rather
1718)     than fixed-size windows? That seemed to go well in an <a

1719)     href="http://www.psc.edu/index.php/hpn-ssh/638">ssh

1720)     throughput experiment</a>. We'll need to measure and tweak, and maybe
1721)     overhaul if the results are good.</li>
1722)     <li>Our censorship-resistance goals include preventing
1723)     an attacker who's looking at Tor traffic on the wire from <a
1724)     href="<svnprojects>design-paper/blocking.html#sec:network-fingerprint">distinguishing
1725)     it from normal SSL traffic</a>. Obviously we can't achieve perfect
1726)     steganography and still remain usable, but for a first step we'd like to
1727)     block any attacks that can win by observing only a few packets. One of
1728)     the remaining attacks we haven't examined much is that Tor cells are 512
1729)     bytes, so the traffic on the wire may well be a multiple of 512 bytes.
1730)     How much does the batching and overhead in TLS records blur this on the
1731)     wire? Do different buffer flushing strategies in Tor affect this? Could
1732)     a bit of padding help a lot, or is this an attack we must accept?</li>
1733)     <li>Tor circuits are built one hop at a time, so in theory we have the
1734)     ability to make some streams exit from the second hop, some from the
1735)     third, and so on. This seems nice because it breaks up the set of exiting
1736)     streams that a given relay can see. But if we want each stream to be safe,
1737)     the "shortest" path should be at least 3 hops long by our current logic, so
1738)     the rest will be even longer. We need to examine this performance / security
1739)     tradeoff.</li>
1740)     <li>It's not that hard to DoS Tor relays or directory authorities. Are client
1741)     puzzles the right answer? What other practical approaches are there? Bonus
1742)     if they're backward-compatible with the current Tor protocol.</li>
1743)     <li>Programs like <a

1744)     href="<page docs/torbutton/index>">Torbutton</a> aim to hide

1745)     your browser's UserAgent string by replacing it with a uniform answer for
1746)     every Tor user. That way the attacker can't splinter Tor's anonymity set
1747)     by looking at that header. It tries to pick a string that is commonly used
1748)     by non-Tor users too, so it doesn't stand out. Question one: how badly
1749)     do we hurt ourselves by periodically updating the version of Firefox
1750)     that Torbutton claims to be? If we update it too often, we splinter the
1751)     anonymity sets ourselves. If we don't update it often enough, then all the
1752)     Tor users stand out because they claim to be running a quite old version
1753)     of Firefox. The answer here probably depends on the Firefox versions seen
1754)     in the wild. Question two: periodically people ask us to cycle through N
1755)     UserAgent strings rather than stick with one. Does this approach help,
1756)     hurt, or not matter? Consider: cookies and recognizing Torbutton users
1757)     by their rotating UserAgents; malicious websites who only attack certain
1758)     browsers; and whether the answers to question one impact this answer.
1759)     </li>
1760)     <li>How many bridge relays do you need to know to maintain
1761)     reachability? We should measure the churn in our bridges. If there is
1762)     lots of churn, are there ways to keep bridge users more likely to stay
1763)     connected?
1764)     </li>
1765)     </ol>

1766) 

1767)     <p>
1768)     <a href="<page about/contact>">Let us know</a> if you've made progress on any
1769)     of these!
1770)     </p>
1771)   </div>
1772)   <!-- END MAINCOL -->
1773)   <div id = "sidecol">
1774) #include "side.wmi"
1775) #include "info.wmi"
1776)   </div>
1777)   <!-- END SIDECOL -->
1778) </div>
1779) <!-- END CONTENT -->