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)     <p>Tor has <a href="<page getinvolved/open-positions>">two open positions</a>.
31)     Please <a href="<page about/contact>">contact us</a> if you are qualified!</p>
32)     
33)     <a id="Documentation"></a>
34)     <h2><a class="anchor" href="#Documentation">Documentation</a></h2>
35)     <ol>
36)     <li>Help translate the web page and documentation into other
37)     languages. See the <a href="<page getinvolved/translation>">translation
38)     guidelines</a> if you want to help out. We especially need Arabic or
39)     Farsi translations, for the many Tor users in censored areas.</li>
40)     <li>Evaluate and document
41)     <a href="https://trac.torproject.org/projects/tor/wiki/TheOnionRouter/TorifyHOWTO">our
42)     list of programs</a> that can be configured to use Tor.</li>
43)     <li>We have a huge list of <a
44)     href="https://trac.torproject.org/projects/tor/wiki/TheOnionRouter/SupportPrograms">potentially useful
45)     programs that interface to Tor</a>. Which ones are useful in which
46)     situations? Please help us test them out and document your results.</li>
47)     </ol>
48)     
49)     <a id="Advocacy"></a>
50)     <h2><a class="anchor" href="#Advocacy">Advocacy</a></h2>
51)     <ol>

52)     <li>Create a <a
53) href="https://trac.torproject.org/projects/tor/wiki/CommunityLogos">community
54) logo</a> under a Creative Commons license that all can use and modify.</li>
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="http://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)     </ol>
66)     
67)     <a id="Coding"></a>
68)     <a id="Summer"></a>
69)     <a id="Projects"></a>
70)     <h2><a class="anchor" href="#Projects">Good Coding Projects</a></h2>
71)     
72)     <p>
73)     You may find some of these projects to be good <a href="<page
74)     about/gsoc>">Google Summer of Code 2010</a> ideas. We have labelled each idea
75)     with how useful it would be to the overall Tor project (priority), how
76)     much work we expect it would be (effort level), how much clue you should
77)     start with (skill level), and which of our <a href="<page

78)     about/corepeople>">core developers</a> would be good mentors.

79)     If one or more of these ideas looks promising to you, please <a
80)     href="<page about/contact>">contact us</a> to discuss your plans rather than
81)     sending blind applications. You may also want to propose your own project
82)     idea &mdash; which often results in the best applications.
83)     </p>
84)     
85)     <ol>
86)     
87)     <li>

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

89)     <br>

90)     Priority: <i>High</i>

91)     <br>

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

93)     <br>

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

95)     <br>

96)     Likely Mentors: <i>Steven, Erinn, Jacob, Andrew</i>

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

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

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

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

108)     and Mac OS X, and be comfortable with C/C++ and shell scripting.</p>
109)     <p>If you would like to help extend or do security auditing for
110)     TBB, please contact Erinn.</p>

111)     </li>
112)     
113)     <li>
114)     <b>Help track the overall Tor Network status</b>

115)     <br>

116)     Priority: <i>Medium to High</i>

117)     <br>

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

119)     <br>

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

121)     <br>

122)     Likely Mentors: <i>Karsten, Roger</i>

123)     <p>It would be great to set up an automated system for tracking network

124)     health over time, graphing it, etc. Part of this project would involve
125)     inventing better metrics for assessing network health and growth. Is the
126)     average uptime of the network increasing? How many relays are qualifying
127)     for Guard status this month compared to last month? What's the turnover
128)     in terms of new relays showing up and relays shutting off? Periodically
129)     people collect brief snapshots, but where it gets really interesting is

130)     when we start tracking data points over time.</p>
131)     <p>Data could be collected from the Tor Network Scanners in <a

132)     href="https://svn.torproject.org/svn/torflow/trunk/README">TorFlow</a>, from
133)     the server descriptors that each relay publishes, and from other
134)     sources. Results over time could be integrated into one of the <a
135)     href="https://torstatus.blutmagie.de/">Tor Status</a> web pages, or be
136)     kept separate. Speaking of the Tor Status pages, take a look at Roger's
137)     <a href="http://archives.seul.org/or/talk/Jan-2008/msg00300.html">Tor

138)     Status wish list</a>.</p>

139)     </li>
140)     
141)     <li>
142)     <b>Improving Tor's ability to resist censorship</b>

143)     <br>

144)     Priority: <i>Medium to High</i>

145)     <br>

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

147)     <br>

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

149)     <br>

150)     Likely Mentors: <i>Roger, Nick, Steven</i>

151)     <p>The Tor 0.2.1.x series makes <a

152)     href="<svnprojects>design-paper/blocking.html">significant
153)     improvements</a> in resisting national and organizational censorship.
154)     But Tor still needs better mechanisms for some parts of its

155)     anti-censorship design.</p>
156)     <p>One huge category of work is adding features to our <a

157)     href="http://gitweb.torproject.org/bridgedb.git?a=tree">BridgeDB</a>

158)     service (Python). Tor aims to give out <a href="<page docs/bridges>">bridge
159)     relay addresses</a> to users that can't reach the Tor network
160)     directly, but there's an arms race between algorithms for distributing
161)     addresses and algorithms for gathering and blocking them. See <a

162)     href="<blog>bridge-distribution-strategies">our

163)     blog post on the topic</a> as an overview, and then look at <a
164)     href="http://archives.seul.org/or/dev/Dec-2009/msg00000.html">Roger's
165)     or-dev post</a> from December for more recent thoughts &mdash; lots of

166)     design work remains.</p>
167)     <p>If you want to get more into the guts of Tor itself (C), a more minor problem

168)     we should address is that current Tors can only listen on a single
169)     address/port combination at a time. There's
170)     <a href="<gitblob>doc/spec/proposals/118-multiple-orports.txt">a
171)     proposal to address this limitation</a> and allow clients to connect
172)     to any given Tor on multiple addresses and ports, but it needs more

173)     work.</p>
174)     <p>This project could involve a lot of research and design. One of the big

175)     challenges will be identifying and crafting approaches that can still
176)     resist an adversary even after the adversary knows the design, and

177)     then trading off censorship resistance with usability and
178)     robustness.</p>

179)     </li>
180)     
181)     <li>
182)     <b>Tor Controller Status Event Interface for Vidalia</b>

183)     <br>

184)     Priority: <i>Medium</i>

185)     <br>

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

187)     <br>

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

189)     <br>

190)     Likely Mentors: <i>Matt</i>

191)     <p>There are a number of status changes inside Tor of which the user may need

192)     to be informed. For example, if the user is trying to set up his Tor as a
193)     relay and Tor decides that its ports are not reachable from outside
194)     the user's network, we should alert the user. Currently, all the user

195)     gets is a couple of log messages in Vidalia's 'message log' window, which they

196)     likely never see since they don't receive a notification that something
197)     has gone wrong. Even if the user does actually look at the message log,

198)     most of the messages make little sense to the novice user.</p>

199)     <p>Tor has the ability to inform Vidalia of many such status
200)     changes, and we recently implemented support for a couple of these
201)     events. Still, there are many more status events which the user should
202)     be informed of, and we need a better UI for actually displaying them
203)     to the user.</p>

204)     <p>The goal of this project then is to design and implement a UI for

205)     displaying Tor status events to the user. For example, we might put a
206)     little badge on Vidalia's tray icon that alerts the user to new status
207)     events they should look at. Double-clicking the icon could bring up a
208)     dialog that summarizes recent status events in simple terms and maybe
209)     suggests a remedy for any negative events if they can be corrected by
210)     the user. Of course, this is just an example and one is free to

211)     suggest another approach.</p>
212)     <p>A person undertaking this project should have good UI design and layout

213)     skills and some C++ development experience. Previous experience with Qt and

214)     Qt's Designer will be very helpful, but are not required. Some
215)     English writing ability will also be useful, since this project will
216)     likely involve writing small amounts of help documentation that should
217)     be understandable by non-technical users. Bonus points for some graphic

218)     design/Photoshop fu, since we might want/need some shiny new icons too.</p>

219)     </li>
220)     
221)     <li>
222)     <b>Improve our unit testing process</b>

223)     <br>

224)     Priority: <i>Medium</i>

225)     <br>

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

227)     <br>

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

229)     <br>

230)     Likely Mentors: <i>Nick, Erinn</i>

231)     <p>Tor needs to be tested far more thoroughly. This is a
232)     multi-part effort. To start with, our unit test coverage should
233)     rise substantially, especially in the areas outside the utility
234)     functions. This will require significant refactoring of some parts
235)     of Tor, in order to dissociate as much logic as possible from
236)     globals.</p>

237)     <p>Additionally, we need to automate our performance testing. We've got

238)     buildbot to automate our regular integration and compile testing already
239)     (though we need somebody to set it up on Windows),
240)     but we need to get our network simulation tests (as built in <a
241)     href="https://svn.torproject.org/svn/torflow/trunk/README">TorFlow</a>)
242)     updated for more recent versions of Tor, and designed to launch a test
243)     network either on a single machine, or across several, so we can test

244)     changes in performance on machines in different roles automatically.</p>

245)     </li>
246)     
247)     <li>
248)     <b>Help with independent Tor client implementations</b>

249)     <br>

250)     Priority: <i>Medium</i>

251)     <br>

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

253)     <br>

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

255)     <br>

256)     Likely Mentors: <i>Bruce, Nathan</i>

257)     <p>Others are currently working on Tor clients for Java, Android, and Maemo

258)     environments.  The first step is to get a handle on the current state of
259)     the project in which you are interested in helping; <a
260)     href="http://github.com/brl/JTor">Tor for Java</a>,

261)     <a href="https://svn.torproject.org/svn/projects/android/trunk/">Android/Orbot</a>,
262)      or <a href="<page docs/N900>">Tor for Maemo</a>. Check out the

263)     repository and familiarize yourself
264)     with the source code.  Further, support for requesting or even providing

265)     Tor hidden services would be neat, but not required.</p>
266)     <p>A prospective developer should be able to understand and write new Java

267)     code, including a Java cryptography API. Being able to read C code would be helpful,
268)     too. One should be willing to read the existing documentation,
269)     implement code based on it, and refine the documentation
270)     when things are underdocumented. This project is mostly about coding and

271)     to a small degree about design.</p>

272)     </li>

273) 

274)     <li>
275)     <b>More on Orbot &amp; Android OS-specific development</b>
276)     <br/>

277)     <br>

278)     Priority: <i>Medium</i>

279)     <br>

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

281)     <br>

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

283)     <br>

284)     Likely Mentors: <i>Nathan</i>

285)     <p><b>Android Java UI work:</b> Improved home screen to show better
286)     statistics about data transferred (up/down), number of circuits
287)     connected, quality of connection and so on. The "Tether Wifi"
288)     Android application is a good model to follow in how it shows
289)     a realtime count of bytes transferred as well as notifications

290)     when wifi clients connect. In addition, better display/handling

291)     of Tor system/error messages would also be very helpful. Finally,
292)     the addition of a wizard or tutorial walkthrough for novice

293)     users to explain to them exactly what is and what is not anonymized

294)     or protected would greatly improve the likelihood they will use
295)     Orbot correctly.</p>
296)     
297)     <p><b>Android Java OS/Core app work:</b> Better system-wide

298)     indication, either via the notification bar, "Toast" pop-up dialogs
299)     or some other indicator, that an application's traffic is indeed

300)     moving through Orbot/Tor. For instance, right now you need to
301)     first go to a torcheck web service to ensure your browser is
302)     routing via Tor. Orbot should be able to notify you that circuits
303)     are being opened, used, etc. The aforementioned data transfer
304)     tracker might provide this type of awareness as well.</p>
305)     
306)     <p><b>Android Java Library/Community Outreach work:</b> We need
307)     to package a simple library for use with third-party application

308)     to easily enable them to support "Torification" on non-rooted
309)     devices (i.e. w/o transparent proxying). This library should

310)     include a wrapper for the Apache HTTPClient library, a utility
311)     class for detecting the state of Orbot connectivity, and other
312)     relevant/useful things an Android app might need to anonymize
313)     itself. This work would include the creation of the library,
314)     documentation, and sample code. Outreach or effort to implement
315)     the library within other open-source apps would follow.</p>
316)     
317)     <p><b>Android OS/C/Linux work:</b> The port of Tor to Android
318)     is basically a straight cross-compile to Linux ARM. There has
319)     been no work done in looking the optimization of Tor within a
320)     mobile hardware environment, on the ARM processor or other
321)     Android hardware, or on mobile networks. It should be noted,
322)     that even without optimization, Tor is handling the mobile
323)     network environment very well, automatically detecting change

324)     in IP addresses, reconnecting circuits, etc. across switching

325)     from 2G to 3G to Wifi, and so forth.</p>

326)     </li>
327)     
328)     <li>
329)     <b>Simulator for slow Internet connections</b>

330)     <br>

331)     Priority: <i>Medium</i>

332)     <br>

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

334)     <br>

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

336)     <br>

337)     Likely Mentors: <i>Steven</i>

338)     <br>

339)     Many users of Tor have poor-quality Internet connections, giving low
340)     bandwidth, high latency, and high packet loss/re-ordering. User
341)     experience is that Tor reacts badly to these conditions, but it is
342)     difficult to improve the situation without being able to repeat the
343)     problems in the lab.

344)     <br>

345)     This project would be to build a simulation environment which
346)     replicates the poor connectivity so that the effect on Tor performance
347)     can be measured. Other components would be a testing utility to
348)     establish what are the properties of connections available, and to
349)     measure the effect of performance-improving modifications to Tor.

350)     <br>

351)     The tools used would be up to the student, but dummynet (for FreeBSD)
352)     and nistnet (for Linux) are two potential components on which this
353)     project could be built. Students should be experienced with network
354)     programming/debugging and TCP/IP, and preferably familiar with C and a
355)     scripting language.
356)     </li>
357)     
358)     <li>
359)     <b>An Improved and More Usable Network Map in Vidalia</b>

360)     <br>

361)     Priority: <i>Low to Medium</i>

362)     <br>

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

364)     <br>

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

366)     <br>

367)     Likely Mentors: <i>Matt</i>

368)     <br>

369)     One of Vidalia's existing features is a network map that shows the user
370)     the approximate geographic location of relays in the Tor network and
371)     plots the paths the user's traffic takes as it is tunneled through the
372)     Tor network. The map is currently not very interactive and has rather
373)     poor graphics. Instead, we implemented KDE's Marble widget such
374)     that it gives us a better quality map and enables improved interactivity,
375)     such as allowing the user to click on individual relays or circuits to
376)     display additional information. We want to add the ability
377)     for users to click on a particular relay or a country containing one or
378)     more Tor exit relays and say, "I want my connections to exit
379)     from here."

380)     <br>

381)     This project will first involve getting familiar with Vidalia
382)     and the Marble widget's API. One will then integrate the widget
383)     into Vidalia and customize Marble to be better suited for our application,
384)     such as making circuits clickable, storing cached map data in Vidalia's
385)     own data directory, and customizing some of the widget's dialogs.

386)     <br>

387)     A person undertaking this project should have good C++ development
388)     experience. Previous experience with Qt and CMake is helpful, but not
389)     required.
390)     </li>
391)     
392)     <li>
393)     <b>Torbutton equivalent for Thunderbird</b>

394)     <br>

395)     Priority: <i>Medium</i>

396)     <br>

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

398)     <br>

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

400)     <br>

401)     Likely Mentors: <i>Mike</i>

402)     <br>

403)     We're hearing from an increasing number of users that they want to use
404)     Thunderbird with Tor. However, there are plenty of application-level
405)     concerns, for example, by default Thunderbird will put your hostname in
406)     the outgoing mail that it sends. At some point we should start a new
407)     push to build a Thunderbird extension similar to Torbutton.
408)     </li>
409)     
410)     <li>
411)     <b>Improvements for Tor+Vidalia interaction on Linux/Unix platforms</b>

412)     <br>

413)     Priority: <i>Medium</i>

414)     <br>

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

416)     <br>

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

418)     <br>

419)     Likely Mentors: <i>Erinn, Peter</i>

420)     <br>

421)     Vidalia currently doesn't play nicely with Tor on Linux and Unix platforms.
422)     Currently, on Debian and Ubuntu, there is a configuration mechanism which
423)     allows Vidalia to override Tor's ability to start on boot (by sourcing
424)     <code>/etc/default/tor.vidalia</code> which sets <code>RUN_DAEMON=no</code> at the user's
425)     request), but full implementation of <a href="<gitblob>doc/spec/control-spec.txt">ControlPort</a> 
426)     communication is still required.

427)     <br>

428)     A better solution on Linux and Unix platforms would be to use Tor's
429)     ControlSocket, which allows Vidalia to talk to Tor via a Unix domain socket,
430)     and could possibly be enabled by default in Tor's distribution packages.
431)     Vidalia can then authenticate to Tor using filesystem-based (cookie)
432)     authentication if the user running Vidalia is also in the distribution-specific
433)     tor group.

434)     <br>

435)     This project will first involve adding support for Tor's ControlSocket to
436)     Vidalia. The student will then develop and test this support on various
437)     distributions to make sure it behaves in a predictable and consistent manner on
438)     all of them.

439)     <br>

440)     The next challenge would be to find an intuitive and usable way for Vidalia to be
441)     able to change Tor's configuration (torrc) even though it is located in
442)     <code>/etc/tor/torrc</code> and thus immutable. In Debian and Ubuntu we handle
443)     this with the aforementioned <code>/etc/default/tor.vidalia</code> but this
444)     functionality could (or should) be less distribution-specific. 

445)     <br>

446)     The best idea we've come up with so far is to feed Tor a new configuration via
447)     the ControlSocket when Vidalia starts, but that's bad because if the user is not
448)     using the latest Debian/Ubuntu packages, they may not have disabled Tor's
449)     ability to run on boot and will end up with a configuration that is different
450)     from what they want. The second best idea we've come up with is for Vidalia to
451)     write out a temporary torrc file and ask the user to manually move it to
452)     <code>/etc/tor/torrc</code>, but that's bad because users shouldn't have to
453)     mess with files directly.

454)     <br>

455)     A person undertaking this project should have prior knowledge of various Linux
456)     distributions and their packaging mechanisms as well as some C++ development
457)     experience. Previous experience with Qt is helpful, but not required.
458)     </li>
459)     
460)     
461)     <li>
462)     <b>Usability testing of Tor</b>

463)     <br>

464)     Priority: <i>Medium</i>

465)     <br>

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

467)     <br>

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

469)     <br>

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

471)     <br>

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

476)     </li>

477)     
478)     <li>
479)     <b>An authenticating IRC proxy</b>

480)     <br>

481)     Priority: <i>Low</i>

482)     <br>

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

484)     <br>

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

486)     <br>

487)     Likely Mentors: <i>Sebastian, Weasel, Roger</i>

488)     <br>

489)     The world needs an authenticating irc proxy. As we're periodically
490)     reminded from the Penny Arcade web comic, "Internet user + anonymity =
491)     jerk". With respect to websites we're actually doing ok, since websites
492)     can make their users log in and use other application-level authentication
493)     approaches. But IRC servers are much worse off, because most IRC server
494)     code is poorly written: hard to maintain, and harder to modify. Many
495)     IRC networks now block connections from Tor, and we're basically down to
496)     two holdouts (OFTC and Freenode). This state of affairs means that a lot
497)     of people around the world are thinking "I told you so" about anonymity
498)     online, when in fact the problem is simply lack of technology to make the
499)     problem manageable. We need some way to let the IRC networks distinguish
500)     which users have developed a reputation as not being jerks, so they can
501)     treat the two groups separately. There are some really cool research
502)     designs like <a href="http://www.cs.dartmouth.edu/~nymble/">Nymble</a>,
503)     which aim to let websites blacklist users without needing to learn who
504)     they are.  But Nymble is designed around web interactions. We need to
505)     build the glue around the IRC protocol that would let us plug in a project
506)     like Nymble (or a simpler one to start, as a proof-of-concept). One way
507)     to do that would be to build an IRC proxy that knows how to hear from
508)     IRC clients, knows how to talk to IRC servers, and has an additional
509)     layer that requires the users to authenticate.  Some work on this has
510)     begun by other volunteers, see their progress at <a
511)     href="http://github.com/anonirc/orc">http://github.com/anonirc/orc</a>.
512)     </li>
513)     
514)     <li>
515)     <b>Make torsocks/dsocks work on OS X</b>

516)     <br>

517)     Priority: <i>Medium</i>

518)     <br>

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

520)     <br>

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

522)     <br>

523)     Likely Mentors: <i>?</i>

524)     <br>

525)     <a href="http://code.google.com/p/torsocks/">Torsocks</a> and <a
526)     href="http://code.google.com/p/dsocks/">dsocks</a> are wrappers that will
527)     run applications, intercept their outgoing network connections, and push
528)     those connections through Tor. The goal is to handle applications that
529)     don't support proxies (or don't supporting them well). To get it right,
530)     they need to intercept many system calls. The syscalls you need to
531)     intercept on Linux differ dramatically from those on BSD. So Torsocks
532)     works fine on Linux, dsocks works ok on BSD (though it may be less
533)     maintained and thus might miss more syscalls), and nothing works well
534)     on both. First, we should patch dsocks to use Tor's <i>mapaddress</i>
535)     commands from the controller interface, so we don't waste a whole
536)     round-trip inside Tor doing the resolve before connecting. Second,
537)     we should make our <i>torify</i> script detect which of torsocks or
538)     dsocks is installed, and call them appropriately. This probably means
539)     unifying their interfaces, and might involve sharing code between them
540)     or discarding one entirely.
541)     </li>
542)     
543)     <li>
544)     <b>Bring up new ideas!</b>

545)     <br>

546)     Don't like any of these? Look at the <a
547)     href="<gitblob>doc/roadmaps/2008-12-19-roadmap-full.pdf">Tor development
548)     roadmap</a> for more ideas, or just try out Tor, Vidalia, and Torbutton,
549)     and find out what you think needs fixing.
550)     Some of the <a href="<gittree>doc/spec/proposals">current proposals</a>
551)     might also be short on developers.
552)     </li>
553)     
554)     </ol>
555)     
556)     <a id="OtherCoding"></a>
557)     <h2><a class="anchor" href="#OtherCoding">Other Coding and Design related ideas</a></h2>
558)     <ol>
559)     <li>Tor relays don't work well on Windows XP. On
560)     Windows, Tor uses the standard <tt>select()</tt> system
561)     call, which uses space in the non-page pool. This means
562)     that a medium sized Tor relay will empty the non-page pool, <a
563)     href="https://trac.torproject.org/projects/tor/wiki/TheOnionRouter/WindowsBufferProblems">causing
564)     havoc and system crashes</a>. We should probably be using overlapped IO
565)     instead. One solution would be to teach <a
566)     href="http://www.monkey.org/~provos/libevent/">libevent</a> how to use
567)     overlapped IO rather than select() on Windows, and then adapt Tor to
568)     the new libevent interface. Christian King made a
569)     <a href="https://svn.torproject.org/svn/libevent-urz/trunk/">good
570)     start</a> on this in the summer of 2007.</li>
571)     
572)     <li>We need to actually start building our <a href="<page
573)     docs/documentation>#DesignDoc">blocking-resistance design</a>. This involves
574)     fleshing out the design, modifying many different pieces of Tor, adapting
575)     <a href="<page projects/vidalia>">Vidalia</a> so it supports the
576)     new features, and planning for deployment.</li>
577)     
578)     <li>We need a flexible simulator framework for studying end-to-end
579)     traffic confirmation attacks. Many researchers have whipped up ad hoc
580)     simulators to support their intuition either that the attacks work
581)     really well or that some defense works great. Can we build a simulator
582)     that's clearly documented and open enough that everybody knows it's
583)     giving a reasonable answer? This will spur a lot of new research.
584)     See the entry <a href="#Research">below</a> on confirmation attacks for
585)     details on the research side of this task &mdash; who knows, when it's
586)     done maybe you can help write a paper or three also.</li>
587)     

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

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

592)     Tor if needed.</li>

593)     
594)     <li>Perform a security analysis of Tor with <a
595)     href="http://en.wikipedia.org/wiki/Fuzz_testing">"fuzz"</a>. Determine
596)     if there are good fuzzing libraries out there for what we want. Win fame by
597)     getting credit when we put out a new release because of you!</li>
598)     
599)     <li>Tor uses TCP for transport and TLS for link
600)     encryption. This is nice and simple, but it means all cells
601)     on a link are delayed when a single packet gets dropped, and
602)     it means we can only reasonably support TCP streams. We have a <a
603)     href="https://trac.torproject.org/projects/tor/wiki/TheOnionRouter/TorFAQ#YoushouldtransportallIPpacketsnotjustTCPpackets.">list
604)     of reasons why we haven't shifted to UDP transport</a>, but it would
605)     be great to see that list get shorter. We also have a proposed <a
606)     href="<gitblob>doc/spec/proposals/100-tor-spec-udp.txt">specification
607)     for Tor and
608)     UDP</a> &mdash; please let us know what's wrong with it.</li>
609)     
610)     <li>We're not that far from having IPv6 support for destination addresses
611)     (at exit nodes). If you care strongly about IPv6, that's probably the
612)     first place to start.</li>
613)     
614)     <li>We need a way to generate the website diagrams (for example, the "How
615)     Tor Works" pictures on the <a href="<page about/overview>">overview page</a>
616)     from source, so we can translate them as UTF-8 text rather than edit
617)     them by hand with Gimp. We might want to
618)     integrate this as an wml file so translations are easy and images are
619)     generated in multiple languages whenever we build the website.</li>
620)     
621)     <li>How can we make the various LiveCD/USB systems easier
622)     to maintain, improve, and document?  One example is <a
623)     href="https://amnesia.boum.org/">The (Amnesic) Incognito Live
624)     System</a>.
625)     </li>
626)     
627)     <li>
628)     Another anti-censorship project is to try to make Tor
629)     more scanning-resistant.  Right now, an adversary can identify <a
630)     href="<gitblob>doc/spec/proposals/125-bridges.txt">Tor bridges</a>
631)     just by trying to connect to them, following the Tor protocol,
632)     and seeing if they respond.  To solve this, bridges could <a
633)     href="<svnprojects>design-paper/blocking.html#tth_sEc9.3">act like
634)     webservers</a> (HTTP or HTTPS) when contacted by port-scanning tools,
635)     and not act like bridges until the user provides a bridge-specific key.
636)     To start, check out Shane Pope's <a
637)     href="http://dl.dropbox.com/u/37735/index.html">thesis and prototype</a>.
638)     </li>
639)     
640)     </ol>
641)     
642)     <a id="Research"></a>
643)     <h2><a class="anchor" href="#Research">Research</a></h2>
644)     <ol>
645)     <li>The "end-to-end traffic confirmation attack":
646)     by watching traffic at Alice and at Bob, we can <a
647)     href="http://freehaven.net/anonbib/#danezis:pet2004">compare
648)     traffic signatures and become convinced that we're watching the same
649)     stream</a>. So far Tor accepts this as a fact of life and assumes this
650)     attack is trivial in all cases. First of all, is that actually true? How
651)     much traffic of what sort of distribution is needed before the adversary
652)     is confident he has won? Are there scenarios (e.g. not transmitting much)
653)     that slow down the attack? Do some traffic padding or traffic shaping
654)     schemes work better than others?</li>
655)     <li>A related question is: Does running a relay/bridge provide additional
656)     protection against these timing attacks? Can an external adversary that can't
657)     see inside TLS links still recognize individual streams reliably?
658)     Does the amount of traffic carried degrade this ability any? What if the
659)     client-relay deliberately delayed upstream relayed traffic to create a queue
660)     that could be used to mimic timings of client downstream traffic to make it
661)     look like it was also relayed? This same queue could also be used for masking
662)     timings in client upstream traffic with the techniques from <a
663)     href="http://www.freehaven.net/anonbib/#ShWa-Timing06">adaptive padding</a>,
664)     but without the need for additional traffic. Would such an interleaving of
665)     client upstream traffic obscure timings for external adversaries? Would the
666)     strategies need to be adjusted for asymmetric links? For example, on
667)     asymmetric links, is it actually possible to differentiate client traffic from
668)     natural bursts due to their asymmetric capacity? Or is it easier than
669)     symmetric links for some other reason?</li>
670)     <li>Repeat Murdoch and Danezis's <a
671)     href="http://www.cl.cam.ac.uk/~sjm217/projects/anon/#torta">attack from
672)     Oakland 05</a> on the current Tor network. See if you can learn why it
673)     works well on some nodes and not well on others. (My theory is that the
674)     fast nodes with spare capacity resist the attack better.) If that's true,
675)     then experiment with the RelayBandwidthRate and RelayBandwidthBurst
676)     options to run a relay that is used as a client while relaying the
677)     attacker's traffic: as we crank down the RelayBandwidthRate, does the
678)     attack get harder? What's the right ratio of RelayBandwidthRate to
679)     actually capacity? Or is it a ratio at all? While we're at it, does a
680)     much larger set of candidate relays increase the false positive rate
681)     or other complexity for the attack? (The Tor network is now almost two
682)     orders of magnitude larger than it was when they wrote their paper.) Be
683)     sure to read <a href="http://freehaven.net/anonbib/#clog-the-queue">Don't
684)     Clog the Queue</a> too.</li>
685)     <li>The "routing zones attack": most of the literature thinks of
686)     the network path between Alice and her entry node (and between the
687)     exit node and Bob) as a single link on some graph. In practice,
688)     though, the path traverses many autonomous systems (ASes), and <a
689)     href="http://freehaven.net/anonbib/#feamster:wpes2004">it's not uncommon
690)     that the same AS appears on both the entry path and the exit path</a>.
691)     Unfortunately, to accurately predict whether a given Alice, entry,
692)     exit, Bob quad will be dangerous, we need to download an entire Internet
693)     routing zone and perform expensive operations on it. Are there practical
694)     approximations, such as avoiding IP addresses in the same /8 network?</li>
695)     <li>Other research questions regarding geographic diversity consider
696)     the tradeoff between choosing an efficient circuit and choosing a random
697)     circuit. Look at Stephen Rollyson's <a
698)     href="http://swiki.cc.gatech.edu:8080/ugResearch/uploads/7/ImprovingTor.pdf">position
699)     paper</a> on how to discard particularly slow choices without hurting
700)     anonymity "too much". This line of reasoning needs more work and more
701)     thinking, but it looks very promising.</li>
702)     <li>Tor doesn't work very well when relays have asymmetric bandwidth
703)     (e.g. cable or DSL). Because Tor has separate TCP connections between
704)     each hop, if the incoming bytes are arriving just fine and the outgoing
705)     bytes are all getting dropped on the floor, the TCP push-back mechanisms
706)     don't really transmit this information back to the incoming streams.
707)     Perhaps Tor should detect when it's dropping a lot of outgoing packets,
708)     and rate-limit incoming streams to regulate this itself? I can imagine
709)     a build-up and drop-off scheme where we pick a conservative rate-limit,
710)     slowly increase it until we get lost packets, back off, repeat. We
711)     need somebody who's good with networks to simulate this and help design
712)     solutions; and/or we need to understand the extent of the performance
713)     degradation, and use this as motivation to reconsider UDP transport.</li>
714)     <li>A related topic is congestion control. Is our
715)     current design sufficient once we have heavy use? Maybe
716)     we should experiment with variable-sized windows rather
717)     than fixed-size windows? That seemed to go well in an <a
718)     href="http://www.psc.edu/networking/projects/hpn-ssh/theory.php">ssh
719)     throughput experiment</a>. We'll need to measure and tweak, and maybe
720)     overhaul if the results are good.</li>
721)     <li>Our censorship-resistance goals include preventing
722)     an attacker who's looking at Tor traffic on the wire from <a
723)     href="<svnprojects>design-paper/blocking.html#sec:network-fingerprint">distinguishing
724)     it from normal SSL traffic</a>. Obviously we can't achieve perfect
725)     steganography and still remain usable, but for a first step we'd like to
726)     block any attacks that can win by observing only a few packets. One of
727)     the remaining attacks we haven't examined much is that Tor cells are 512
728)     bytes, so the traffic on the wire may well be a multiple of 512 bytes.
729)     How much does the batching and overhead in TLS records blur this on the
730)     wire? Do different buffer flushing strategies in Tor affect this? Could
731)     a bit of padding help a lot, or is this an attack we must accept?</li>
732)     <li>Tor circuits are built one hop at a time, so in theory we have the
733)     ability to make some streams exit from the second hop, some from the
734)     third, and so on. This seems nice because it breaks up the set of exiting
735)     streams that a given relay can see. But if we want each stream to be safe,
736)     the "shortest" path should be at least 3 hops long by our current logic, so
737)     the rest will be even longer. We need to examine this performance / security
738)     tradeoff.</li>
739)     <li>It's not that hard to DoS Tor relays or directory authorities. Are client
740)     puzzles the right answer? What other practical approaches are there? Bonus
741)     if they're backward-compatible with the current Tor protocol.</li>
742)     <li>Programs like <a

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