tor-webwml.git

en/volunteer.wml   1) ## translation metadata

en/volunteer.wml   2) # Revision: $Revision$

volunteer.html     3) 

en/volunteer.wml   4) #include "head.wmi" TITLE="Tor: Volunteer"

volunteer.html     5) 
volunteer.html     6) <div class="main-column">
volunteer.html     7) 
volunteer.html     8) <!-- PUT CONTENT AFTER THIS TAG -->

en/volunteer.wml   9) <h2>A few things everyone can do now:</h2>

volunteer.html    10) <ol>

en/volunteer.wml  11) <li>Please consider <a href="<page docs/tor-doc-relay>">running
en/volunteer.wml  12) a relay</a> to help the Tor network grow.</li>
en/volunteer.wml  13) <li>Tell your friends! Get them to run relays. Get them to run hidden

volunteer.html    14) services. Get them to tell their friends.</li>

en/volunteer.wml  15) <li>If you like Tor's goals, please <a href="<page donate>">take a moment
en/volunteer.wml  16) to donate to support further Tor development</a>. We're also looking
en/volunteer.wml  17) for more sponsors &mdash; if you know any companies, NGOs, agencies,
en/volunteer.wml  18) or other organizations that want anonymity / privacy / communications
en/volunteer.wml  19) security, let them know about us.</li>
en/volunteer.wml  20) <li>We're looking for more <a href="<page torusers>">good examples of Tor
en/volunteer.wml  21) users and Tor use cases</a>. If you use Tor for a scenario or purpose not
en/volunteer.wml  22) yet described on that page, and you're comfortable sharing it with us,
en/volunteer.wml  23) we'd love to hear from you.</li>

volunteer.html    24) </ol>
volunteer.html    25) 

en/volunteer.wml  26) <a id="Usability"></a>

en/volunteer.wml  27) <h2><a class="anchor" href="#Usability">Supporting Applications</a></h2>

volunteer.html    28) <ol>

en/volunteer.wml  29) <li>We need more good ways to intercept DNS requests so they don't "leak" their

en/volunteer.wml  30) request to a local observer while we're trying to be anonymous. (This
en/volunteer.wml  31) happens because the application does the DNS resolve before going to
en/volunteer.wml  32) the SOCKS proxy.)</li>

en/volunteer.wml  33) <li>Tsocks/dsocks items:

en/volunteer.wml  34) <ul>
en/volunteer.wml  35) <li>We need to <a

en/volunteer.wml  36) href="https://wiki.torproject.org/noreply/TheOnionRouter/TSocksPatches">apply

en/volunteer.wml  37) all our tsocks patches</a> and maintain a new fork. We'll host it if
en/volunteer.wml  38) you want.</li>
en/volunteer.wml  39) <li>We should patch Dug Song's "dsocks" program to use Tor's
en/volunteer.wml  40) <i>mapaddress</i> commands from the controller interface, so we
en/volunteer.wml  41) don't waste a whole round-trip inside Tor doing the resolve before
en/volunteer.wml  42) connecting.</li>
en/volunteer.wml  43) <li>We need to make our <i>torify</i> script detect which of tsocks or
en/volunteer.wml  44) dsocks is installed, and call them appropriately. This probably means
en/volunteer.wml  45) unifying their interfaces, and might involve sharing code between them
en/volunteer.wml  46) or discarding one entirely.</li>
en/volunteer.wml  47) </ul>

en/volunteer.wml  48) </li>

en/volunteer.wml  49) <li>People running relays tell us they want to have one BandwidthRate

en/volunteer.wml  50) during some part of the day, and a different BandwidthRate at other
en/volunteer.wml  51) parts of the day. Rather than coding this inside Tor, we should have a
en/volunteer.wml  52) little script that speaks via the <a href="<page gui/index>">Tor
en/volunteer.wml  53) Controller Interface</a>, and does a setconf to change the bandwidth

en/volunteer.wml  54) rate. There is one for Unix and Mac already (it uses bash and cron),

en/volunteer.wml  55) but Windows users still need a solution.

en/volunteer.wml  56) </li>

en/volunteer.wml  57) <li>Tor can <a

en/volunteer.wml  58) href="https://wiki.torproject.org/noreply/TheOnionRouter/TorFAQ#ChooseEntryExit">exit

en/volunteer.wml  59) the Tor network from a particular exit node</a>, but we should be able
en/volunteer.wml  60) to specify just a country and have something automatically pick. The
en/volunteer.wml  61) best bet is to fetch Blossom's directory also, and run a local Blossom
en/volunteer.wml  62) client that fetches this directory securely (via Tor and checking its
en/volunteer.wml  63) signature), intercepts <tt>.country.blossom</tt> hostnames, and does
en/volunteer.wml  64) the right thing.</li>

volunteer.html    65) <li>Speaking of geolocation data, somebody should draw a map of the Earth

en/volunteer.wml  66) with a pin-point for each Tor relay. Bonus points if it updates as the

en/volunteer.wml  67) network grows and changes. Unfortunately, the easy ways to do this involve
en/volunteer.wml  68) sending all the data to Google and having them draw the map for you. How
en/volunteer.wml  69) much does this impact privacy, and do we have any other good options?</li>

volunteer.html    70) </ol>
volunteer.html    71) 

en/volunteer.wml  72) <a id="Documentation"></a>
en/volunteer.wml  73) <h2><a class="anchor" href="#Documentation">Documentation</a></h2>

volunteer.html    74) <ol>

en/volunteer.wml  75) <li>Please help Matt Edman with the documentation and how-tos for his

en/volunteer.wml  76) Tor controller,
en/volunteer.wml  77) <a href="http://vidalia-project.net/">Vidalia</a>.</li>

en/volunteer.wml  78) <li>Evaluate and document

en/volunteer.wml  79) <a href="https://wiki.torproject.org/wiki/TheOnionRouter/TorifyHOWTO">our

en/volunteer.wml  80) list of programs</a> that can be configured to use Tor.</li>

volunteer.html    81) <li>We need better documentation for dynamically intercepting

en/volunteer.wml  82) connections and sending them through Tor. tsocks (Linux), dsocks (BSD),

en/volunteer.wml  83) and freecap (Windows) seem to be good candidates, as would better
en/volunteer.wml  84) use of our new TransPort feature.</li>

en/volunteer.wml  85) <li>We have a huge list of <a href="https://wiki.torproject.org/noreply/TheOnionRouter/SupportPrograms">potentially useful

volunteer.html    86) programs that interface to Tor</a>. Which ones are useful in which
volunteer.html    87) situations? Please help us test them out and document your results.</li>

en/volunteer.wml  88) <li>Help translate the web page and documentation into other
en/volunteer.wml  89) languages. See the <a href="<page translation>">translation

en/volunteer.wml  90) guidelines</a> if you want to help out. We especially need Arabic or
en/volunteer.wml  91) Farsi translations, for the many Tor users in censored areas.</li>

volunteer.html    92) </ol>
volunteer.html    93) 

en/volunteer.wml  94) <a id="Coding"></a>

en/volunteer.wml  95) <a id="Summer"></a>
en/volunteer.wml  96) <a id="Projects"></a>

en/volunteer.wml  97) <h2><a class="anchor" href="#Projects">Good Coding Projects</a></h2>

en/volunteer.wml  98) 
en/volunteer.wml  99) <p>

en/volunteer.wml 100) Here is a list of ideas that were proposed for the <a href="<page gsoc>">Google Summer of Code 2008</a>
en/volunteer.wml 101) but have not been put into practice. Some of the <a href="<svnsandbox>doc/spec/proposals/">current proposals</a> 
en/volunteer.wml 102) might also be short on developers. If you think you can help out, <a href="<page contact>"> let us know!</a> 

en/volunteer.wml 103) </p>
en/volunteer.wml 104) 

en/volunteer.wml 105) <ol>
en/volunteer.wml 106) 

en/volunteer.wml 107) <li>

en/volunteer.wml 108) <b>Tor Node Scanner improvements</b>

en/volunteer.wml 109) <br />

en/volunteer.wml 110) Similar to the SoaT exit scanner (or perhaps even during exit scanning),

en/volunteer.wml 111) statistics can be gathered about the reliability of nodes. Nodes that
en/volunteer.wml 112) fail too high a percentage of their circuits should not be given
en/volunteer.wml 113) Guard status. Perhaps they should have their reported bandwidth
en/volunteer.wml 114) penalized by some ratio as well, or just get marked as Invalid. In
en/volunteer.wml 115) addition, nodes that exhibit a very low average stream capacity but
en/volunteer.wml 116) advertise a very high node bandwidth can also be marked as Invalid.
en/volunteer.wml 117) Much of this statistics gathering is already done, it just needs to be
en/volunteer.wml 118) transformed into something that can be reported to the Directory
en/volunteer.wml 119) Authorities to blacklist/penalize nodes in such a way that clients
en/volunteer.wml 120) will listen.

en/volunteer.wml 121) <br />

en/volunteer.wml 122) In addition, these same statistics can be gathered about the traffic
en/volunteer.wml 123) through a node. Events can be added to the <a

en/volunteer.wml 124) href="https://svn.torproject.org/svn/torctl/trunk/doc/howto.txt">Tor Control

en/volunteer.wml 125) Protocol</a> to
en/volunteer.wml 126) report if a circuit extend attempt through the node succeeds or fails, and
en/volunteer.wml 127) passive statistics can be gathered on both bandwidth and reliability
en/volunteer.wml 128) of other nodes via a node-based monitor using these events. Such a
en/volunteer.wml 129) scanner would also report information on oddly-behaving nodes to
en/volunteer.wml 130) the Directory Authorities, but a communication channel for this
en/volunteer.wml 131) currently does not exist and would need to be developed as well.
en/volunteer.wml 132) </li>
en/volunteer.wml 133) 
en/volunteer.wml 134) <li>
en/volunteer.wml 135) <b>Help track the overall Tor Network status</b>

en/volunteer.wml 136) <br />

en/volunteer.wml 137) It would be great to set up an automated system for tracking network
en/volunteer.wml 138) health over time, graphing it, etc. Part of this project would involve
en/volunteer.wml 139) inventing better metrics for assessing network health and growth. Is the
en/volunteer.wml 140) average uptime of the network increasing? How many relays are qualifying
en/volunteer.wml 141) for Guard status this month compared to last month? What's the turnover
en/volunteer.wml 142) in terms of new relays showing up and relays shutting off? Periodically
en/volunteer.wml 143) people collect brief snapshots, but where it gets really interesting is
en/volunteer.wml 144) when we start tracking data points over time.
en/volunteer.wml 145) <br />
en/volunteer.wml 146) Data could be collected from the "Tor Node Scanner" item above, from
en/volunteer.wml 147) the server descriptors that each relay publishes, and from other
en/volunteer.wml 148) sources. Results over time could be integrated into one of the <a
en/volunteer.wml 149) href="https://torstatus.blutmagie.de/">Tor Status</a> web pages, or be
en/volunteer.wml 150) kept separate. Speaking of the Tor Status pages, take a look at Roger's
en/volunteer.wml 151) <a href="http://archives.seul.org/or/talk/Jan-2008/msg00300.html">Tor
en/volunteer.wml 152) Status wish list</a>.
en/volunteer.wml 153) </li>
en/volunteer.wml 154) 
en/volunteer.wml 155) <li>
en/volunteer.wml 156) <b>Better Debian/Ubuntu Packaging for Tor+Vidalia</b>

en/volunteer.wml 157) <br />
en/volunteer.wml 158) Vidalia currently doesn't play nicely on Debian and Ubuntu with the
en/volunteer.wml 159) default Tor packages. The current Tor packages automatically start Tor
en/volunteer.wml 160) as a daemon running as the debian-tor user and (sensibly) do not have a

en/volunteer.wml 161) <a href="<svnsandbox>doc/spec/control-spec.txt">ControlPort</a> defined
en/volunteer.wml 162) in the default torrc. Consequently, Vidalia will try

en/volunteer.wml 163) to start its own Tor process since it could not connect to the existing

en/volunteer.wml 164) Tor, and Vidalia's Tor process will then exit with an error message

en/volunteer.wml 165) the user likely doesn't understand since Tor cannot bind its listening

en/volunteer.wml 166) ports — they're already in use by the original Tor daemon.

en/volunteer.wml 167) <br />
en/volunteer.wml 168) The current solution involves either telling the user to stop the
en/volunteer.wml 169) existing Tor daemon and let Vidalia start its own Tor process, or
en/volunteer.wml 170) explaining to the user how to set a control port and password in their
en/volunteer.wml 171) torrc. A better solution on Debian would be to use Tor's ControlSocket,
en/volunteer.wml 172) which allows Vidalia to talk to Tor via a Unix domain socket, and could
en/volunteer.wml 173) possibly be enabled by default in Tor's Debian packages. Vidalia can

en/volunteer.wml 174) then authenticate to Tor using filesystem-based (cookie) authentication
en/volunteer.wml 175) if the user running Vidalia is also in the debian-tor group.

en/volunteer.wml 176) <br />
en/volunteer.wml 177) This project will first involve adding support for Tor's ControlSocket
en/volunteer.wml 178) to Vidalia. The student will then develop and test Debian and Ubuntu
en/volunteer.wml 179) packages for Vidalia that conform to Debian's packaging standards and

en/volunteer.wml 180) make sure they work well with the existing Tor packages. We can also

en/volunteer.wml 181) set up an apt repository to host the new Vidalia packages.
en/volunteer.wml 182) <br />

en/volunteer.wml 183) The next challenge would be to find an intuitive usable way for Vidalia
en/volunteer.wml 184) to be able to change Tor's configuration (torrc) even though it is
en/volunteer.wml 185) located in <code>/etc/tor/torrc</code> and thus immutable. The best
en/volunteer.wml 186) idea we've come up with so far is to feed Tor a new configuration via
en/volunteer.wml 187) the ControlSocket when Vidalia starts, but that's bad because Tor starts

en/volunteer.wml 188) each boot with a different configuration than the user wants. The second
en/volunteer.wml 189) best idea

en/volunteer.wml 190) we've come up with is for Vidalia to write out a temporary torrc file
en/volunteer.wml 191) and ask the user to manually move it to <code>/etc/tor/torrc</code>,
en/volunteer.wml 192) but that's bad because users shouldn't have to mess with files directly.
en/volunteer.wml 193) <br />

en/volunteer.wml 194) A person undertaking this project should have prior knowledge of

en/volunteer.wml 195) Debian package management and some C++ development experience. Previous
en/volunteer.wml 196) experience with Qt is helpful, but not required.
en/volunteer.wml 197) </li>
en/volunteer.wml 198) 

en/volunteer.wml 199) <li>
en/volunteer.wml 200) <b>Improving Tor's ability to resist censorship</b>
en/volunteer.wml 201) <br />
en/volunteer.wml 202) The Tor 0.2.0.x series makes <a
en/volunteer.wml 203) href="<svnsandbox>doc/design-paper/blocking.html">significant
en/volunteer.wml 204) improvements</a> in resisting national and organizational censorship.
en/volunteer.wml 205) But Tor still needs better mechanisms for some parts of its
en/volunteer.wml 206) anti-censorship design.  For example, current Tors can only listen on a
en/volunteer.wml 207) single address/port combination at a time.  There's
en/volunteer.wml 208) <a href="<svnsandbox>doc/spec/proposals/118-multiple-orports.txt">a
en/volunteer.wml 209) proposal to address this limitation</a> and allow clients to connect
en/volunteer.wml 210) to any given Tor on multiple addresses and ports, but it needs more
en/volunteer.wml 211) work.  Another anti-censorship project (far more difficult) is to try
en/volunteer.wml 212) to make Tor more scanning-resistant.  Right now, an adversary can identify
en/volunteer.wml 213) <a href="<svnsandbox>doc/spec/proposals/125-bridges.txt">Tor bridges</a>
en/volunteer.wml 214) just by trying to connect to them, following the Tor protocol, and
en/volunteer.wml 215) seeing if they respond.  To solve this, bridges could
en/volunteer.wml 216) <a href="<svnsandbox>doc/design-paper/blocking.html#tth_sEc9.3">act like
en/volunteer.wml 217) webservers</a> (HTTP or HTTPS) when contacted by port-scanning tools,
en/volunteer.wml 218) and not act like bridges until the user provides a bridge-specific key.
en/volunteer.wml 219) <br />
en/volunteer.wml 220) This project involves a lot of research and design. One of the big
en/volunteer.wml 221) challenges will be identifying and crafting approaches that can still
en/volunteer.wml 222) resist an adversary even after the adversary knows the design, and
en/volunteer.wml 223) then trading off censorship resistance with usability and robustness.
en/volunteer.wml 224) </li>
en/volunteer.wml 225) 
en/volunteer.wml 226) <li>
en/volunteer.wml 227) <b>Tor/Polipo/Vidalia Auto-Update Framework</b>
en/volunteer.wml 228) <br />
en/volunteer.wml 229) We're in need of a good authenticated-update framework.
en/volunteer.wml 230) Vidalia already has the ability to notice when the user is running an
en/volunteer.wml 231) outdated or unrecommended version of Tor, using signed statements inside
en/volunteer.wml 232) the Tor directory information. Currently, Vidalia simply pops
en/volunteer.wml 233) up a little message box that lets the user know they should manually
en/volunteer.wml 234) upgrade. The goal of this project would be to extend Vidalia with the
en/volunteer.wml 235) ability to also fetch and install the updated Tor software for the
en/volunteer.wml 236) user. We should do the fetches via Tor when possible, but also fall back
en/volunteer.wml 237) to direct fetches in a smart way. Time permitting, we would also like
en/volunteer.wml 238) to be able to update other
en/volunteer.wml 239) applications included in the bundled installers, such as Polipo and
en/volunteer.wml 240) Vidalia itself.
en/volunteer.wml 241) <br />
en/volunteer.wml 242) To complete this project, the student will first need to first investigate
en/volunteer.wml 243) the existing auto-update frameworks (e.g., Sparkle on OS X) to evaluate
en/volunteer.wml 244) their strengths, weaknesses, security properties, and ability to be
en/volunteer.wml 245) integrated into Vidalia. If none are found to be suitable, the student
en/volunteer.wml 246) will design their own auto-update framework, document the design, and
en/volunteer.wml 247) then discuss the design with other developers to assess any security
en/volunteer.wml 248) issues. The student will then implement their framework (or integrate
en/volunteer.wml 249) an existing one) and test it.
en/volunteer.wml 250) <br />

en/volunteer.wml 251) A person undertaking this project should have good C++ development
en/volunteer.wml 252) experience. Previous experience with Qt is helpful, but not required. One
en/volunteer.wml 253) should also have a good understanding of common security

en/volunteer.wml 254) practices, such as package signature verification. Good writing ability
en/volunteer.wml 255) is also important for this project, since a vital step of the project

en/volunteer.wml 256) will be producing a design document to review and discuss
en/volunteer.wml 257) with others prior to implementation.

en/volunteer.wml 258) </li>
en/volunteer.wml 259) 
en/volunteer.wml 260) <li>
en/volunteer.wml 261) <b>An Improved and More Usable Network Map in Vidalia</b>
en/volunteer.wml 262) <br />
en/volunteer.wml 263) One of Vidalia's existing features is a network map that shows the user
en/volunteer.wml 264) the approximate geographic location of relays in the Tor network and
en/volunteer.wml 265) plots the paths the user's traffic takes as it is tunneled through the
en/volunteer.wml 266) Tor network. The map is currently not very interactive and has rather
en/volunteer.wml 267) poor graphics. Instead, we would like to leverage KDE's Marble widget
en/volunteer.wml 268) that gives us a better quality map and enables improved interactivity,
en/volunteer.wml 269) such as allowing the user to click on individual relays or circuits to
en/volunteer.wml 270) display additional information. We might also consider adding the ability
en/volunteer.wml 271) for users to click on a particular relay or a country containing one or
en/volunteer.wml 272) more Tor exit relays and say, "I want my connections to foo.com to exit
en/volunteer.wml 273) from here."
en/volunteer.wml 274) <br />

en/volunteer.wml 275) This project will first involve getting familiar with Vidalia
en/volunteer.wml 276) and the Marble widget's API. One will then integrate the widget

en/volunteer.wml 277) into Vidalia and customize Marble to be better suited for our application,
en/volunteer.wml 278) such as making circuits clickable, storing cached map data in Vidalia's
en/volunteer.wml 279) own data directory, and customizing some of the widget's dialogs.
en/volunteer.wml 280) <br />

en/volunteer.wml 281) A person undertaking this project should have good C++ development

en/volunteer.wml 282) experience. Previous experience with Qt and CMake is helpful, but not
en/volunteer.wml 283) required.
en/volunteer.wml 284) </li>
en/volunteer.wml 285) 

en/volunteer.wml 286) <li>

en/volunteer.wml 287) <b>Tor Controller Status Event Interface</b>

en/volunteer.wml 288) <br />

en/volunteer.wml 289) There are a number of status changes inside Tor of which the user may need

en/volunteer.wml 290) to be informed. For example, if the user is trying to set up his Tor as a
en/volunteer.wml 291) relay and Tor decides that its ports are not reachable from outside

en/volunteer.wml 292) the user's network, we should alert the user. Currently, all the user

en/volunteer.wml 293) gets is a couple log messages in Vidalia's 'message log' window, which they

en/volunteer.wml 294) likely never see since they don't receive a notification that something
en/volunteer.wml 295) has gone wrong. Even if the user does actually look at the message log,
en/volunteer.wml 296) most of the messages make little sense to the novice user.
en/volunteer.wml 297) <br />
en/volunteer.wml 298) Tor has the ability to inform Vidalia of many such status changes, and
en/volunteer.wml 299) we recently implemented support for a couple of these events. Still,
en/volunteer.wml 300) there are many more status events the user should be informed of and we
en/volunteer.wml 301) need a better UI for actually displaying them to the user.
en/volunteer.wml 302) <br />
en/volunteer.wml 303) The goal of this project then is to design and implement a UI for
en/volunteer.wml 304) displaying Tor status events to the user. For example, we might put a
en/volunteer.wml 305) little badge on Vidalia's tray icon that alerts the user to new status
en/volunteer.wml 306) events they should look at. Double-clicking the icon could bring up a
en/volunteer.wml 307) dialog that summarizes recent status events in simple terms and maybe

en/volunteer.wml 308) suggests a remedy for any negative events if they can be corrected by

en/volunteer.wml 309) the user. Of course, this is just an example and one is free to

en/volunteer.wml 310) suggest another approach.
en/volunteer.wml 311) <br />

en/volunteer.wml 312) A person undertaking this project should have good UI design and layout

en/volunteer.wml 313) and some C++ development experience. Previous experience with Qt and

en/volunteer.wml 314) Qt's Designer will be very helpful, but are not required. Some

en/volunteer.wml 315) English writing ability will also be useful, since this project will
en/volunteer.wml 316) likely involve writing small amounts of help documentation that should
en/volunteer.wml 317) be understandable by non-technical users. Bonus points for some graphic
en/volunteer.wml 318) design/Photoshop fu, since we might want/need some shiny new icons too.
en/volunteer.wml 319) </li>
en/volunteer.wml 320) 

en/volunteer.wml 321) <li>

en/volunteer.wml 322) <b>Improvements on our active browser configuration tester</b> -

en/volunteer.wml 323) <a href="https://check.torproject.org/">https://check.torproject.org/</a>

en/volunteer.wml 324) <br />
en/volunteer.wml 325) We currently have a functional web page to detect if Tor is working. It

en/volunteer.wml 326) has a few places where it falls short. It requires improvements with

en/volunteer.wml 327) regard to default languages and functionality. It currently only responds
en/volunteer.wml 328) in English. In addition, it is a hack of a perl script that should have
en/volunteer.wml 329) never seen the light of day. It should probably be rewritten in python

en/volunteer.wml 330) with multi-lingual support in mind. It currently uses the <a
en/volunteer.wml 331) href="http://exitlist.torproject.org/">Tor DNS exit list</a>
en/volunteer.wml 332) and should continue to do so in the future. It currently result in certain

en/volunteer.wml 333) false positives and these should be discovered, documented, and fixed
en/volunteer.wml 334) where possible. Anyone working on this project should be interested in

en/volunteer.wml 335) DNS, basic perl or preferably python programming skills, and will have

en/volunteer.wml 336) to interact minimally with Tor to test their code.

en/volunteer.wml 337) <br />
en/volunteer.wml 338) If you want to make the project more exciting
en/volunteer.wml 339) and involve more design and coding, take a look at <a
en/volunteer.wml 340) href="<svnsandbox>doc/spec/proposals/131-verify-tor-usage.txt">proposal
en/volunteer.wml 341) 131-verify-tor-usage.txt</a>.

en/volunteer.wml 342) </li>
en/volunteer.wml 343) 
en/volunteer.wml 344) <li>

en/volunteer.wml 345) <b>Improvements on our DNS Exit List service</b> -

en/volunteer.wml 346) <a href="http://exitlist.torproject.org/">http://exitlist.torproject.org/</a>

en/volunteer.wml 347) <br />

en/volunteer.wml 348) The <a href="http://p56soo2ibjkx23xo.onion/">exitlist software</a>
en/volunteer.wml 349) is written by our fabulous anonymous

en/volunteer.wml 350) contributer Tup. It's a DNS server written in Haskell that supports part of our <a

en/volunteer.wml 351) href="<svnsandbox>doc/contrib/torel-design.txt">exitlist

en/volunteer.wml 352) design document</a>. Currently, it is functional and it is used by

en/volunteer.wml 353) check.torproject.org and other users. The issues that are outstanding
en/volunteer.wml 354) are mostly aesthetic. This wonderful service could use a much better
en/volunteer.wml 355) website using the common Tor theme. It would be best served with better
en/volunteer.wml 356) documentation for common services that use an RBL. It could use more
en/volunteer.wml 357) publicity. A person working on this project should be interested in DNS,
en/volunteer.wml 358) basic RBL configuration for popular services, and writing documentation.
en/volunteer.wml 359) The person would require minimal Tor interaction — testing their
en/volunteer.wml 360) own documentation at the very least. Furthermore, it would be useful

en/volunteer.wml 361) if they were interested in Haskell and wanted to implement more of the

en/volunteer.wml 362) torel-design.txt suggestions.
en/volunteer.wml 363) </li>
en/volunteer.wml 364) 
en/volunteer.wml 365) <li>

en/volunteer.wml 366) <b>Testing integration of Tor with web browsers for our end users</b>

en/volunteer.wml 367) <br />

en/volunteer.wml 368) The Tor project currently lacks a solid test suite to ensure that a
en/volunteer.wml 369) user has a properly and safely configured web browser. It should test for as

en/volunteer.wml 370) many known issues as possible. It should attempt to decloak the

en/volunteer.wml 371) user in any way possible. Two current webpages that track these

en/volunteer.wml 372) kinds of issues are run by Greg Fleischer and HD Moore. Greg keeps a nice <a

en/volunteer.wml 373) href="http://pseudo-flaw.net/tor/torbutton/">list of issues along
en/volunteer.wml 374) with their proof of concept code, bug issues, etc</a>. HD Moore runs

en/volunteer.wml 375) the <a href="http://metasploit.com/research/projects/decloak/">metasploit

en/volunteer.wml 376) decloak website</a>. A person interested in defending Tor could start

en/volunteer.wml 377) by collecting as many workable and known methods for decloaking a

en/volunteer.wml 378) Tor user. (<a href="https://torcheck.xenobite.eu/">This page</a> may

en/volunteer.wml 379) be helpful as a start.) One should be familiar with the common pitfalls but

en/volunteer.wml 380) possibly have new methods in mind for implementing decloaking issues. The
en/volunteer.wml 381) website should ensure that it tells a user what their problem is. It
en/volunteer.wml 382) should help them to fix the problem or direct them to the proper support

en/volunteer.wml 383) channels. The person should also be closely familiar with using Tor and how

en/volunteer.wml 384) to prevent Tor information leakage.

en/volunteer.wml 385) </li>
en/volunteer.wml 386) 

en/volunteer.wml 387) <li>

en/volunteer.wml 388) <b>Libevent and Tor integration improvements</b>
en/volunteer.wml 389) <br />

en/volunteer.wml 390) Tor should make better use of the more recent features of Niels
en/volunteer.wml 391) Provos's <a href="http://monkey.org/~provos/libevent/">Libevent</a>

en/volunteer.wml 392) library.  Tor already uses Libevent for its low-level asynchronous IO

en/volunteer.wml 393) calls, and could also use Libevent's increasingly good implementations

en/volunteer.wml 394) of network buffers and of HTTP.  This wouldn't be simply a matter of

en/volunteer.wml 395) replacing Tor's internal calls with calls to Libevent: instead, we'll

en/volunteer.wml 396) need to refactor Tor to use Libevent calls that do not follow the

en/volunteer.wml 397) same models as Tor's existing backends. Also, we'll need to add
en/volunteer.wml 398) missing functionality to Libevent as needed — most difficult likely
en/volunteer.wml 399) will be adding OpenSSL support on top of Libevent's buffer abstraction.
en/volunteer.wml 400) Also tricky will be adding rate-limiting to Libevent.

en/volunteer.wml 401) </li>
en/volunteer.wml 402) 
en/volunteer.wml 403) <li>

en/volunteer.wml 404) <b>Tuneup Tor!</b>
en/volunteer.wml 405) <br />

en/volunteer.wml 406) Right now, Tor relays measure and report their own bandwidth, and Tor
en/volunteer.wml 407) clients choose which relays to use in part based on that bandwidth.
en/volunteer.wml 408) This approach is vulnerable to
en/volunteer.wml 409) <a href="http://freehaven.net/anonbib/#bauer:wpes2007">attacks where
en/volunteer.wml 410) relays lie about their bandwidth</a>;
en/volunteer.wml 411) to address this, Tor currently caps the maximum bandwidth
en/volunteer.wml 412) it's willing to believe any relay provides.  This is a limited fix, and
en/volunteer.wml 413) a waste of bandwidth capacity to boot.  Instead,

en/volunteer.wml 414) Tor should possibly measure bandwidth in a more distributed way, perhaps
en/volunteer.wml 415) as described in the

en/volunteer.wml 416) <a href="http://freehaven.net/anonbib/author.html#snader08">"A Tune-up for
en/volunteer.wml 417) Tor"</a> paper

en/volunteer.wml 418) by Snader and Borisov. One could use current testing code to

en/volunteer.wml 419) double-check this paper's findings and verify the extent to which they

en/volunteer.wml 420) dovetail with Tor as deployed in the wild, and determine good ways to
en/volunteer.wml 421) incorporate them into their suggestions Tor network without adding too

en/volunteer.wml 422) much communications overhead between relays and directory

en/volunteer.wml 423) authorities.

en/volunteer.wml 424) </li>
en/volunteer.wml 425) 

en/volunteer.wml 426) <!--

en/volunteer.wml 427) <li>

en/volunteer.wml 428) <b>Improving the Tor QA process: Continuous Integration for Windows builds</b>
en/volunteer.wml 429) <br />

en/volunteer.wml 430) It would be useful to have automated build processes for Windows and
en/volunteer.wml 431) probably other platforms. The purpose of having a continuous integration
en/volunteer.wml 432) build environment is to ensure that Windows isn't left behind for any of
en/volunteer.wml 433) the software projects used in the Tor project or its accompanying.<br />
en/volunteer.wml 434) Buildbot may be a good choice for this as it appears to support all of

en/volunteer.wml 435) the platforms Tor does. See the

en/volunteer.wml 436) <a href="http://en.wikipedia.org/wiki/BuildBot">wikipedia entry for

en/volunteer.wml 437) buildbot</a>.<br />
en/volunteer.wml 438) There may be better options and the person undertaking this task should
en/volunteer.wml 439) evaluate other options. Any person working on this automatic build
en/volunteer.wml 440) process should have experience or be willing to learn how to build all
en/volunteer.wml 441) of the respective Tor related code bases from scratch. Furthermore, the
en/volunteer.wml 442) person should have some experience building software in Windows
en/volunteer.wml 443) environments as this is the target audience we want to ensure we do not
en/volunteer.wml 444) leave behind. It would require close work with the Tor source code but

en/volunteer.wml 445) probably only in the form of building, not authoring.<br />
en/volunteer.wml 446) Additionally, we need to automate our performance testing for all platforms.

en/volunteer.wml 447) We've got buildbot (except on Windows — as noted above) to automate

en/volunteer.wml 448) our regular integration and compile testing already,
en/volunteer.wml 449) but we need to get our network simulation tests (as built in torflow)
en/volunteer.wml 450) updated for more recent versions of Tor, and designed to launch a test
en/volunteer.wml 451) network either on a single machine, or across several, so we can test

en/volunteer.wml 452) changes in performance on machines in different roles automatically.

en/volunteer.wml 453) </li>

en/volunteer.wml 454) -->

en/volunteer.wml 455) 

en/volunteer.wml 456) <li>

en/volunteer.wml 457) <b>Improve our unit testing process</b>
en/volunteer.wml 458) <br />

en/volunteer.wml 459) Tor needs to be far more tested. This is a multi-part effort. To start

en/volunteer.wml 460) with, our unit test coverage should rise substantially, especially in

en/volunteer.wml 461) the areas outside the utility functions. This will require significant

en/volunteer.wml 462) refactoring of some parts of Tor, in order to dissociate as much logic

en/volunteer.wml 463) as possible from globals.
en/volunteer.wml 464) <br />

en/volunteer.wml 465) Additionally, we need to automate our performance testing. We've got

en/volunteer.wml 466) buildbot to automate our regular integration and compile testing already
en/volunteer.wml 467) (though we need somebody to set it up on Windows),
en/volunteer.wml 468) but we need to get our network simulation tests (as built in TorFlow: see
en/volunteer.wml 469) the "Tor Node Scanner improvements" item)

en/volunteer.wml 470) updated for more recent versions of Tor, and designed to launch a test
en/volunteer.wml 471) network either on a single machine, or across several, so we can test

en/volunteer.wml 472) changes in performance on machines in different roles automatically.

en/volunteer.wml 473) </li>
en/volunteer.wml 474) 

en/volunteer.wml 475) <li>

en/volunteer.wml 476) <b>Help revive an independent Tor client implementation</b>

en/volunteer.wml 477) <br />

en/volunteer.wml 478) Reanimate one of the approaches to implement a Tor client in Java,
en/volunteer.wml 479) e.g. the <a href="http://onioncoffee.sourceforge.net/">OnionCoffee
en/volunteer.wml 480) project</a>, and make it run on <a
en/volunteer.wml 481) href="http://code.google.com/android/">Android</a>. The first step
en/volunteer.wml 482) would be to port the existing code and execute it in an Android
en/volunteer.wml 483) environment. Next, the code should be updated to support the newer Tor
en/volunteer.wml 484) protocol versions like the <a href="<svnsandbox>doc/spec/dir-spec.txt">v3
en/volunteer.wml 485) directory protocol</a>. Further, support for requesting or even

en/volunteer.wml 486) providing Tor hidden services would be neat, but not required.
en/volunteer.wml 487) <br />

en/volunteer.wml 488) A prospective developer should be able to understand and write new Java
en/volunteer.wml 489) code, including

en/volunteer.wml 490) a Java cryptography API. Being able to read C code would be helpful,

en/volunteer.wml 491) too. One should be willing to read the existing documentation,

en/volunteer.wml 492) implement code based on it, and refine the documentation
en/volunteer.wml 493) when things are underdocumented. This project is mostly about coding and

en/volunteer.wml 494) to a small degree about design.
en/volunteer.wml 495) </li>
en/volunteer.wml 496) 
en/volunteer.wml 497) <li>

en/volunteer.wml 498) <b>Bring moniTor to life</b>
en/volunteer.wml 499) <br />

en/volunteer.wml 500) Implement a <a href="http://www.ss64.com/bash/top.html">top-like</a>
en/volunteer.wml 501) management tool for Tor relays. The purpose of such a tool would be
en/volunteer.wml 502) to monitor a local Tor relay via its control port and include useful
en/volunteer.wml 503) system information of the underlying machine. When running this tool, it
en/volunteer.wml 504) would dynamically update its content like top does for Linux processes.
en/volunteer.wml 505) <a href="http://archives.seul.org/or/dev/Jan-2008/msg00005.html">This

en/volunteer.wml 506) or-dev post</a> might be a good first read.
en/volunteer.wml 507) <br />

en/volunteer.wml 508) A person interested in this should be familiar

en/volunteer.wml 509) with or willing to learn about administering a Tor relay and configuring
en/volunteer.wml 510) it via its control port. As an initial prototype is written in Python,
en/volunteer.wml 511) some knowledge about writing Python code would be helpful, too. This

en/volunteer.wml 512) project is one part about identifying requirements to such a
en/volunteer.wml 513) tool and designing its interface, and one part lots of coding.

en/volunteer.wml 514) </li>
en/volunteer.wml 515) 

en/volunteer.wml 516) <li>
en/volunteer.wml 517) <b>Torbutton improvements</b>
en/volunteer.wml 518) <br />
en/volunteer.wml 519) Torbutton has a number of improvements that can be made in the post-1.2
en/volunteer.wml 520) timeframe. Most of these are documented as feature requests in the <a

en/volunteer.wml 521) href="https://bugs.torproject.org/flyspray/index.php?tasks=all&project=5">Torbutton

en/volunteer.wml 522) flyspray section</a>. Good examples include: stripping off node.exit on http
en/volunteer.wml 523) headers, more fine-grained control over formfill blocking, improved referrer

en/volunteer.wml 524) spoofing based on the domain of the site (a-la <a

en/volunteer.wml 525) href="https://addons.mozilla.org/en-US/firefox/addon/953">refcontrol extension</a>),

en/volunteer.wml 526) tighter integration with Vidalia for reporting Tor status, a New Identity
en/volunteer.wml 527) button with Tor integration and multiple identity management, and anything

en/volunteer.wml 528) else you might think of.

en/volunteer.wml 529) <br />
en/volunteer.wml 530) This work would be independent coding in Javascript and the fun world of <a
en/volunteer.wml 531) href="http://www.mozilla.org/keymaster/gatekeeper/there.is.only.xul">XUL</a>,
en/volunteer.wml 532) with not too much involvement in the Tor internals.

en/volunteer.wml 533) </li>
en/volunteer.wml 534) 

en/volunteer.wml 535) <li>

en/volunteer.wml 536) <b>Porting Polipo to Windows</b>
en/volunteer.wml 537) <br />

en/volunteer.wml 538) Help port <a
en/volunteer.wml 539) href="http://www.pps.jussieu.fr/~jch/software/polipo/">Polipo</a> to

en/volunteer.wml 540) Windows. Example topics to tackle include:
en/volunteer.wml 541) 1) handle spaces in path names and understand the filesystem
en/volunteer.wml 542) namespace — that is, where application data, personal data,

en/volunteer.wml 543) and program data typically reside in various versions of Windows. 2) the
en/volunteer.wml 544) ability to handle ipv6 communications. 3) the ability to asynchronously
en/volunteer.wml 545) query name servers, find the system nameservers, and manage netbios
en/volunteer.wml 546) and dns queries. 4) use native regex capabilities of Windows, rather
en/volunteer.wml 547) than using 3rd party GNU regex libraries. 5) manage events and buffers
en/volunteer.wml 548) natively (i.e. in Unix-like OSes, Polipo defaults to 25% of ram, in
en/volunteer.wml 549) Windows it's whatever the config specifies). 6) some sort of GUI config
en/volunteer.wml 550) and reporting tool, bonus if it has a systray icon with right clickable
en/volunteer.wml 551) menu options. Double bonus if it's cross-platform compatible.
en/volunteer.wml 552) </li>
en/volunteer.wml 553) 
en/volunteer.wml 554) <li>

en/volunteer.wml 555) <b>Make our diagrams beautiful and automated</b>

en/volunteer.wml 556) <br />

en/volunteer.wml 557) We need a way to generate the website diagrams (for example, the "How
en/volunteer.wml 558) Tor Works" pictures on the <a href="<page overview>">overview page</a>
en/volunteer.wml 559) from source, so we can translate them as UTF-8 text rather than edit
en/volunteer.wml 560) them by hand with Gimp. We might want to
en/volunteer.wml 561) integrate this as an wml file so translations are easy and images are

en/volunteer.wml 562) generated in multiple languages whenever we build the website. 

en/volunteer.wml 563) </li>
en/volunteer.wml 564) 

en/volunteer.wml 565) <li>
en/volunteer.wml 566) <b>Improve the LiveCD offerings for the Tor community</b>
en/volunteer.wml 567) <br />
en/volunteer.wml 568) How can we make the <a

en/volunteer.wml 569) href="http://anonymityanywhere.com/incognito/">Incognito LiveCD</a>

en/volunteer.wml 570) easier to maintain, improve, and document?
en/volunteer.wml 571) </li>

en/volunteer.wml 572) 

en/volunteer.wml 573) <li>
en/volunteer.wml 574) <b>Rework and extend Blossom</b>
en/volunteer.wml 575) <br />
en/volunteer.wml 576) Rework and extend Blossom (a tool for monitoring and
en/volunteer.wml 577) selecting appropriate Tor circuits based upon exit node requirements
en/volunteer.wml 578) specified by the user) to gather data in a self-contained way, with
en/volunteer.wml 579) parameters easily configurable by the user.  Blossom is presently
en/volunteer.wml 580) implemented as a single Python script that interfaces with Tor using the
en/volunteer.wml 581) Controller interface and depends upon metadata about Tor nodes obtained
en/volunteer.wml 582) via external processes, such as a webpage indicating status of the nodes
en/volunteer.wml 583) plus publically available data from DNS, whois, etc.  This project has
en/volunteer.wml 584) two parts: (1) Determine which additional metadata may be useful and
en/volunteer.wml 585) rework Blossom so that it cleanly obtains the metadata on its own rather
en/volunteer.wml 586) than depend upon external scripts (this may, for example, involve
en/volunteer.wml 587) additional threads or inter-process communication), and (2) develop a
en/volunteer.wml 588) means by which the user can easily configure Blossom, starting with a
en/volunteer.wml 589) configuration file and possibly working up to a web configuration engine.

en/volunteer.wml 590) Knowledge of Tor and Python are important; knowledge of
en/volunteer.wml 591) TCP, interprocess communication, and Perl will also be helpful.  An
en/volunteer.wml 592) interest in network neutrality is important as well, since the
en/volunteer.wml 593) principles of evaluating and understanding internet inconsistency are at
en/volunteer.wml 594) the core of the Blossom effort.

en/volunteer.wml 595) </li>
en/volunteer.wml 596) 
en/volunteer.wml 597) <li>
en/volunteer.wml 598) <b>Improve Blossom: Allow users to qualitatively describe exit nodes they desire</b>
en/volunteer.wml 599) <br />
en/volunteer.wml 600) Develop and implement a means of affording Blossom
en/volunteer.wml 601) users the ability to qualitatively describe the exit node that they
en/volunteer.wml 602) want.  The Internet is an inconsistent place: some Tor exit nodes see
en/volunteer.wml 603) the world differently than others.  As presently implemented, Blossom (a
en/volunteer.wml 604) tool for monitoring and selecting appropriate Tor circuits based upon
en/volunteer.wml 605) exit node requirements specified by the user) lacks a sufficiently rich
en/volunteer.wml 606) language to describe how the different vantage points are different.
en/volunteer.wml 607) For example, some exit nodes may have an upstream network that filters
en/volunteer.wml 608) certain kinds of traffic or certain websites.  Other exit nodes may
en/volunteer.wml 609) provide access to special content as a result of their location, perhaps
en/volunteer.wml 610) as a result of discrimination on the part of the content providers
en/volunteer.wml 611) themselves.  This project has two parts: (1) develop a language for
en/volunteer.wml 612) describing characteristics of networks in which exit nodes reside, and
en/volunteer.wml 613) (2) incorporate this language into Blossom so that users can select Tor
en/volunteer.wml 614) paths based upon the description.

en/volunteer.wml 615) Knowledge of Tor and Python are important; knowledge of
en/volunteer.wml 616) TCP, interprocess communication, and Perl will also be helpful.  An
en/volunteer.wml 617) interest in network neutrality is important as well, since the
en/volunteer.wml 618) principles of evaluating and understanding internet inconsistency are at
en/volunteer.wml 619) the core of the Blossom effort.

en/volunteer.wml 620) </li>
en/volunteer.wml 621) 

en/volunteer.wml 622) <li>
en/volunteer.wml 623) <b>Bring up new ideas!</b>
en/volunteer.wml 624) <br />
en/volunteer.wml 625) Don't like any of these? Look at the <a

en/volunteer.wml 626) href="<svnsandbox>doc/roadmaps/2008-12-19-roadmap-full.pdf">Tor development

en/volunteer.wml 627) roadmap</a> for more ideas.
en/volunteer.wml 628) </li>
en/volunteer.wml 629) 

en/volunteer.wml 630) </ol>
en/volunteer.wml 631) 

en/volunteer.wml 632) <a id="OtherCoding"></a>
en/volunteer.wml 633) <h2><a class="anchor" href="#OtherCoding">Other Coding and Design related ideas</a></h2>

volunteer.html   634) <ol>

en/volunteer.wml 635) <li>Tor relays don't work well on Windows XP. On

en/volunteer.wml 636) Windows, Tor uses the standard <tt>select()</tt> system

en/volunteer.wml 637) call, which uses space in the non-page pool. This means

en/volunteer.wml 638) that a medium sized Tor relay will empty the non-page pool, <a

en/volunteer.wml 639) href="https://wiki.torproject.org/noreply/TheOnionRouter/WindowsBufferProblems">causing

en/volunteer.wml 640) havoc and system crashes</a>. We should probably be using overlapped IO

en/volunteer.wml 641) instead. One solution would be to teach <a
en/volunteer.wml 642) href="http://www.monkey.org/~provos/libevent/">libevent</a> how to use
en/volunteer.wml 643) overlapped IO rather than select() on Windows, and then adapt Tor to

en/volunteer.wml 644) the new libevent interface. Christian King made a

en/volunteer.wml 645) <a href="https://svn.torproject.org/svn/libevent-urz/trunk/">good

en/volunteer.wml 646) start</a> on this in the summer of 2007.</li>

en/volunteer.wml 647) 

en/volunteer.wml 648) <li>We need to actually start building our <a href="<page
en/volunteer.wml 649) documentation>#DesignDoc">blocking-resistance design</a>. This involves

en/volunteer.wml 650) fleshing out the design, modifying many different pieces of Tor, adapting
en/volunteer.wml 651) <a href="http://vidalia-project.net/">Vidalia</a> so it supports the
en/volunteer.wml 652) new features, and planning for deployment.</li>

en/volunteer.wml 653) 

en/volunteer.wml 654) <li>We need a flexible simulator framework for studying end-to-end
en/volunteer.wml 655) traffic confirmation attacks. Many researchers have whipped up ad hoc
en/volunteer.wml 656) simulators to support their intuition either that the attacks work
en/volunteer.wml 657) really well or that some defense works great. Can we build a simulator
en/volunteer.wml 658) that's clearly documented and open enough that everybody knows it's
en/volunteer.wml 659) giving a reasonable answer? This will spur a lot of new research.
en/volunteer.wml 660) See the entry <a href="#Research">below</a> on confirmation attacks for
en/volunteer.wml 661) details on the research side of this task &mdash; who knows, when it's
en/volunteer.wml 662) done maybe you can help write a paper or three also.</li>

en/volunteer.wml 663) 

en/volunteer.wml 664) <li>Tor 0.1.1.x and later include support for hardware crypto accelerators

en/volunteer.wml 665) via OpenSSL. Nobody has ever tested it, though. Does somebody want to get

volunteer.html   666) a card and let us know how it goes?</li>

en/volunteer.wml 667) 

volunteer.html   668) <li>Perform a security analysis of Tor with <a
volunteer.html   669) href="http://en.wikipedia.org/wiki/Fuzz_testing">"fuzz"</a>. Determine

en/volunteer.wml 670) if there are good fuzzing libraries out there for what we want. Win fame by

volunteer.html   671) getting credit when we put out a new release because of you!</li>

en/volunteer.wml 672) 

volunteer.html   673) <li>Tor uses TCP for transport and TLS for link
volunteer.html   674) encryption. This is nice and simple, but it means all cells
volunteer.html   675) on a link are delayed when a single packet gets dropped, and
volunteer.html   676) it means we can only reasonably support TCP streams. We have a <a

en/volunteer.wml 677) href="https://wiki.torproject.org/noreply/TheOnionRouter/TorFAQ#TransportIPnotTCP">list

en/volunteer.wml 678) of reasons why we haven't shifted to UDP transport</a>, but it would
en/volunteer.wml 679) be great to see that list get shorter. We also have a proposed <a

en/volunteer.wml 680) href="<svnsandbox>doc/spec/proposals/100-tor-spec-udp.txt">specification
en/volunteer.wml 681) for Tor and

en/volunteer.wml 682) UDP</a> &mdash; please let us know what's wrong with it.</li>

en/volunteer.wml 683) 

volunteer.html   684) <li>We're not that far from having IPv6 support for destination addresses
volunteer.html   685) (at exit nodes). If you care strongly about IPv6, that's probably the
volunteer.html   686) first place to start.</li>

volunteer.html   687) </ol>
volunteer.html   688) 

en/volunteer.wml 689) <a id="Research"></a>
en/volunteer.wml 690) <h2><a class="anchor" href="#Research">Research</a></h2>

volunteer.html   691) <ol>

volunteer.html   692) <li>The "website fingerprinting attack": make a list of a few
volunteer.html   693) hundred popular websites, download their pages, and make a set of
volunteer.html   694) "signatures" for each site. Then observe a Tor client's traffic. As
volunteer.html   695) you watch him receive data, you quickly approach a guess about which
volunteer.html   696) (if any) of those sites he is visiting. First, how effective is
volunteer.html   697) this attack on the deployed Tor codebase? Then start exploring
volunteer.html   698) defenses: for example, we could change Tor's cell size from 512
volunteer.html   699) bytes to 1024 bytes, we could employ padding techniques like <a
volunteer.html   700) href="http://freehaven.net/anonbib/#timing-fc2004">defensive dropping</a>,
volunteer.html   701) or we could add traffic delays. How much of an impact do these have,
volunteer.html   702) and how much usability impact (using some suitable metric) is there from
volunteer.html   703) a successful defense in each case?</li>

volunteer.html   704) <li>The "end-to-end traffic confirmation attack":
volunteer.html   705) by watching traffic at Alice and at Bob, we can <a
volunteer.html   706) href="http://freehaven.net/anonbib/#danezis:pet2004">compare
volunteer.html   707) traffic signatures and become convinced that we're watching the same
volunteer.html   708) stream</a>. So far Tor accepts this as a fact of life and assumes this
volunteer.html   709) attack is trivial in all cases. First of all, is that actually true? How
volunteer.html   710) much traffic of what sort of distribution is needed before the adversary
volunteer.html   711) is confident he has won? Are there scenarios (e.g. not transmitting much)
volunteer.html   712) that slow down the attack? Do some traffic padding or traffic shaping
volunteer.html   713) schemes work better than others?</li>

en/volunteer.wml 714) <li>A related question is: Does running a relay/bridge provide additional
en/volunteer.wml 715) protection against these timing attacks? Can an external adversary that can't

en/volunteer.wml 716) see inside TLS links still recognize individual streams reliably?

en/volunteer.wml 717) Does the amount of traffic carried degrade this ability any? What if the
en/volunteer.wml 718) client-relay deliberately delayed upstream relayed traffic to create a queue
en/volunteer.wml 719) that could be used to mimic timings of client downstream traffic to make it
en/volunteer.wml 720) look like it was also relayed? This same queue could also be used for masking
en/volunteer.wml 721) timings in client upstream traffic with the techniques from <a
en/volunteer.wml 722) href="http://www.freehaven.net/anonbib/#ShWa-Timing06">adaptive padding</a>,
en/volunteer.wml 723) but without the need for additional traffic. Would such an interleaving of
en/volunteer.wml 724) client upstream traffic obscure timings for external adversaries? Would the
en/volunteer.wml 725) strategies need to be adjusted for asymmetric links? For example, on
en/volunteer.wml 726) asymmetric links, is it actually possible to differentiate client traffic from
en/volunteer.wml 727) natural bursts due to their asymmetric capacity? Or is it easier than
en/volunteer.wml 728) symmetric links for some other reason?</li>

en/volunteer.wml 729) <li>Repeat Murdoch and Danezis's <a
en/volunteer.wml 730) href="http://www.cl.cam.ac.uk/~sjm217/projects/anon/#torta">attack from
en/volunteer.wml 731) Oakland 05</a> on the current Tor network. See if you can learn why it
en/volunteer.wml 732) works well on some nodes and not well on others. (My theory is that the
en/volunteer.wml 733) fast nodes with spare capacity resist the attack better.) If that's true,
en/volunteer.wml 734) then experiment with the RelayBandwidthRate and RelayBandwidthBurst
en/volunteer.wml 735) options to run a relay that is used as a client while relaying the
en/volunteer.wml 736) attacker's traffic: as we crank down the RelayBandwidthRate, does the
en/volunteer.wml 737) attack get harder? What's the right ratio of RelayBandwidthRate to
en/volunteer.wml 738) actually capacity? Or is it a ratio at all? While we're at it, does a
en/volunteer.wml 739) much larger set of candidate relays increase the false positive rate
en/volunteer.wml 740) or other complexity for the attack? (The Tor network is now almost two
en/volunteer.wml 741) orders of magnitude larger than it was when they wrote their paper.) Be
en/volunteer.wml 742) sure to read <a href="http://freehaven.net/anonbib/#clog-the-queue">Don't
en/volunteer.wml 743) Clog the Queue</a> too.</li>

volunteer.html   744) <li>The "routing zones attack": most of the literature thinks of
volunteer.html   745) the network path between Alice and her entry node (and between the
volunteer.html   746) exit node and Bob) as a single link on some graph. In practice,
volunteer.html   747) though, the path traverses many autonomous systems (ASes), and <a
volunteer.html   748) href="http://freehaven.net/anonbib/#feamster:wpes2004">it's not uncommon
volunteer.html   749) that the same AS appears on both the entry path and the exit path</a>.
volunteer.html   750) Unfortunately, to accurately predict whether a given Alice, entry,
volunteer.html   751) exit, Bob quad will be dangerous, we need to download an entire Internet
volunteer.html   752) routing zone and perform expensive operations on it. Are there practical
volunteer.html   753) approximations, such as avoiding IP addresses in the same /8 network?</li>

en/volunteer.wml 754) <li>Other research questions regarding geographic diversity consider
en/volunteer.wml 755) the tradeoff between choosing an efficient circuit and choosing a random

en/volunteer.wml 756) circuit. Look at Stephen Rollyson's <a

en/volunteer.wml 757) href="http://swiki.cc.gatech.edu:8080/ugResearch/uploads/7/ImprovingTor.pdf">position
en/volunteer.wml 758) paper</a> on how to discard particularly slow choices without hurting

en/volunteer.wml 759) anonymity "too much". This line of reasoning needs more work and more
en/volunteer.wml 760) thinking, but it looks very promising.</li>

en/volunteer.wml 761) <li>Tor doesn't work very well when relays have asymmetric bandwidth

volunteer.html   762) (e.g. cable or DSL). Because Tor has separate TCP connections between
volunteer.html   763) each hop, if the incoming bytes are arriving just fine and the outgoing
volunteer.html   764) bytes are all getting dropped on the floor, the TCP push-back mechanisms
volunteer.html   765) don't really transmit this information back to the incoming streams.
volunteer.html   766) Perhaps Tor should detect when it's dropping a lot of outgoing packets,
volunteer.html   767) and rate-limit incoming streams to regulate this itself? I can imagine
volunteer.html   768) a build-up and drop-off scheme where we pick a conservative rate-limit,
volunteer.html   769) slowly increase it until we get lost packets, back off, repeat. We
volunteer.html   770) need somebody who's good with networks to simulate this and help design
volunteer.html   771) solutions; and/or we need to understand the extent of the performance
volunteer.html   772) degradation, and use this as motivation to reconsider UDP transport.</li>
volunteer.html   773) <li>A related topic is congestion control. Is our
volunteer.html   774) current design sufficient once we have heavy use? Maybe
volunteer.html   775) we should experiment with variable-sized windows rather
volunteer.html   776) than fixed-size windows? That seemed to go well in an <a
volunteer.html   777) href="http://www.psc.edu/networking/projects/hpn-ssh/theory.php">ssh
volunteer.html   778) throughput experiment</a>. We'll need to measure and tweak, and maybe
volunteer.html   779) overhaul if the results are good.</li>

en/volunteer.wml 780) <li>Our censorship-resistance goals include preventing
en/volunteer.wml 781) an attacker who's looking at Tor traffic on the wire from <a

en/volunteer.wml 782) href="<svnsandbox>doc/design-paper/blocking.html#sec:network-fingerprint">distinguishing

en/volunteer.wml 783) it from normal SSL traffic</a>. Obviously we can't achieve perfect
en/volunteer.wml 784) steganography and still remain usable, but for a first step we'd like to
en/volunteer.wml 785) block any attacks that can win by observing only a few packets. One of
en/volunteer.wml 786) the remaining attacks we haven't examined much is that Tor cells are 512
en/volunteer.wml 787) bytes, so the traffic on the wire may well be a multiple of 512 bytes.
en/volunteer.wml 788) How much does the batching and overhead in TLS records blur this on the
en/volunteer.wml 789) wire? Do different buffer flushing strategies in Tor affect this? Could
en/volunteer.wml 790) a bit of padding help a lot, or is this an attack we must accept?</li>

volunteer.html   791) <li>Tor circuits are built one hop at a time, so in theory we have the
volunteer.html   792) ability to make some streams exit from the second hop, some from the
volunteer.html   793) third, and so on. This seems nice because it breaks up the set of exiting

en/volunteer.wml 794) streams that a given relay can see. But if we want each stream to be safe,

volunteer.html   795) the "shortest" path should be at least 3 hops long by our current logic, so
volunteer.html   796) the rest will be even longer. We need to examine this performance / security
volunteer.html   797) tradeoff.</li>

en/volunteer.wml 798) <li>It's not that hard to DoS Tor relays or directory authorities. Are client

volunteer.html   799) puzzles the right answer? What other practical approaches are there? Bonus
volunteer.html   800) if they're backward-compatible with the current Tor protocol.</li>

en/volunteer.wml 801) <li>Programs like <a

en/volunteer.wml 802) href="<page torbutton/index>">Torbutton</a> aim to hide

en/volunteer.wml 803) your browser's UserAgent string by replacing it with a uniform answer for
en/volunteer.wml 804) every Tor user. That way the attacker can't splinter Tor's anonymity set
en/volunteer.wml 805) by looking at that header. It tries to pick a string that is commonly used
en/volunteer.wml 806) by non-Tor users too, so it doesn't stand out. Question one: how badly
en/volunteer.wml 807) do we hurt ourselves by periodically updating the version of Firefox
en/volunteer.wml 808) that Torbutton claims to be? If we update it too often, we splinter the
en/volunteer.wml 809) anonymity sets ourselves. If we don't update it often enough, then all the
en/volunteer.wml 810) Tor users stand out because they claim to be running a quite old version
en/volunteer.wml 811) of Firefox. The answer here probably depends on the Firefox versions seen
en/volunteer.wml 812) in the wild. Question two: periodically people ask us to cycle through N
en/volunteer.wml 813) UserAgent strings rather than stick with one. Does this approach help,
en/volunteer.wml 814) hurt, or not matter? Consider: cookies and recognizing Torbutton users
en/volunteer.wml 815) by their rotating UserAgents; malicious websites who only attack certain
en/volunteer.wml 816) browsers; and whether the answers to question one impact this answer.
en/volunteer.wml 817) </li>

en/volunteer.wml 818) <li>Right now Tor clients are willing to reuse a given circuit for ten
en/volunteer.wml 819) minutes after it's first used. The goal is to avoid loading down the
en/volunteer.wml 820) network with too many circuit extend operations, yet to also avoid having
en/volunteer.wml 821) clients use the same circuit for so long that the exit node can build a
en/volunteer.wml 822) useful pseudonymous profile of them. Alas, ten minutes is probably way
en/volunteer.wml 823) too long, especially if connections from multiple protocols (e.g. IM and
en/volunteer.wml 824) web browsing) are put on the same circuit. If we keep fixed the overall
en/volunteer.wml 825) number of circuit extends that the network needs to do, are there more
en/volunteer.wml 826) efficient and/or safer ways for clients to allocate streams to circuits,
en/volunteer.wml 827) or for clients to build preemptive circuits? Perhaps this research item
en/volunteer.wml 828) needs to start with gathering some traces of what connections typical
en/volunteer.wml 829) clients try to launch, so you have something realistic to try to optimize.
en/volunteer.wml 830) </li>
en/volunteer.wml 831) <li>How many bridge relays do you need to know to maintain
en/volunteer.wml 832) reachability? We should measure the churn in our bridges. If there is
en/volunteer.wml 833) lots of churn, are there ways to keep bridge users more likely to stay
en/volunteer.wml 834) connected?
en/volunteer.wml 835) </li>

volunteer.html   836) </ol>
volunteer.html   837) 

en/volunteer.wml 838) <p>

en/volunteer.wml 839) <a href="<page contact>">Let us know</a> if you've made progress on any
en/volunteer.wml 840) of these!

en/volunteer.wml 841) </p>

volunteer.html   842) 

en/volunteer.wml 843)   </div><!-- #main -->

volunteer.html   844) 

en/volunteer.wml 845) #include <foot.wmi>