volunteer.html (c21530474) - tor-webwml.git

c215304748ce25a2a868da401014d04653a4b3ba

renamed contribute to volun... Andrew Lewman authored 19 years ago	`1) <!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.1//EN" "http://www.w3.org/TR/xhtml11/DTD/xhtml11.dtd"> 2) 3) <html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en"> 4) <head>`
call it Volunteer in the na... Roger Dingledine authored 19 years ago	`5) <title>Tor: Volunteer</title>`
renamed contribute to volun... Andrew Lewman authored 19 years ago	`6) <meta name="Author" content="Roger Dingledine" /> 7) <meta http-equiv="content-type" content="text/html; charset=ISO-8859-1" /> 8) <link rel="stylesheet" type="text/css" href="stylesheet.css" /> 9) <link rel="shortcut icon" type="image/x-icon" href="/favicon.ico" /> 10) </head> 11) <body> 12) 13) <!-- TITLE BAR & NAVIGATION --> 14)`
made a simple change to the... le@svn.torproject.org authored 19 years ago	`15) <div class="center"> 16)`
renamed contribute to volun... Andrew Lewman authored 19 years ago	`17) <table class="banner" border="0" cellpadding="0" cellspacing="0"> 18) <tr> 19) <td class="banner-left"></td> 20) <td class="banner-middle"> 21) <a href="index.html">Home</a>`
howitworks is now obsolete... Roger Dingledine authored 19 years ago	`22) \| <a href="overview.html">Overview</a>`
renamed contribute to volun... Andrew Lewman authored 19 years ago	`23) \| <a href="download.html">Download</a> 24) \| <a href="documentation.html">Docs</a>`
howitworks is now obsolete... Roger Dingledine authored 19 years ago	`25) \| <a href="support.html">Support</a>`
renamed contribute to volun... Andrew Lewman authored 19 years ago	`26) \| <a href="faq.html">FAQs</a>`
call it Volunteer in the na... Roger Dingledine authored 19 years ago	`27) \| <a class="current">Volunteer</a>`
renamed contribute to volun... Andrew Lewman authored 19 years ago	`28) \| <a href="developers.html">Developers</a> 29) \| <a href="research.html">Research</a> 30) \| <a href="people.html">People</a> 31) </td>`
Updated the banner to inclu... Andrew Lewman authored 19 years ago	`32) <td class="banner-right">`
Make website valid xhtml (t... Thomas Sjögren authored 19 years ago	`33) <a href="/"><img src="/images/en.png" alt="English" /></a> 34) <a href="/index.it.html"><img src="/images/it.png" alt="Italiano" /></a> 35) <a href="/index.de.html"><img src="/images/de.png" alt="Deutsch" /></a>`
Updated the banner to inclu... Andrew Lewman authored 19 years ago	`36) </td>`
renamed contribute to volun... Andrew Lewman authored 19 years ago	`37) </tr> 38) </table> 39) 40) <!-- END TITLE BAR & NAVIGATION --> 41) 42) <div class="main-column"> 43) 44) <!-- PUT CONTENT AFTER THIS TAG -->`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`45) <h2>Seven things everyone can do now:</h2> 46) <ol>`
revamp again Roger Dingledine authored 19 years ago	47) <li> We need users like you to try Tor out, and let the Tor developers 48) know about bugs you find or features you don't find.</li> 49) <li> Please consider <a href="/cvs/tor/doc/tor-doc-server.html">running 50) a server</a> to help the Tor network grow.</li> 51) <li> We especially need people with Windows programming skills to run 52) an exit server on Windows, to help us debug.</li> 53) <li> Run a <a href="/cvs/tor/doc/tor-hidden-service.html">Tor hidden 54) service</a> and put interesting content on it.</li> 55) <li> Take a look at the <a href="/gui/">Tor GUI Competition</a>, and 56) come up with ideas or designs to contribute to making Tor's interface 57) and usability better. Free T-shirt for each submission!</li> 58) <li> Tell your friends! Get them to run servers. Get them to run hidden 59) services. Get them to tell their friends.</li> 60) <li> Consider joining the <a href="http://secure.eff.org/tor">Electronic 61) Frontier Foundation</a>. More EFF donations means more freedom in the 62) world, including more Tor development.</li>
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`63) </ol> 64)`
revamp again Roger Dingledine authored 19 years ago	`65) <h2>Installers</h2>`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`66) <ol>`
revamp again Roger Dingledine authored 19 years ago	67) <li>Extend our NSIS-based Windows installer to include Privoxy. Include 68) a preconfigured config file to work well with Tor. We might also want 69) to include FreeCap -- is it stable enough and useful enough to be 70) worthwhile?</li> 71) <li>Develop a way to handle OS X uninstallation 72) that is more automated than telling people to <a 73) href="http://tor.eff.org/doc/tor-doc-osx.html#uninstall">manually remove 74) each file</a>.</li> 75) <li>Our <a href="http://tor.eff.org/cvs/tor/tor.spec.in">RPM spec file</a> 76) needs a maintainer, so we can get back to the business of writing Tor. If 77) you have RPM fu, please help out.</li> 78) </ol> 79) 80) <h2>Usability and Interface</h2> 81) <ol> 82) <li>We need a way to intercept DNS requests so they don't "leak" while 83) we're trying to be anonymous. (This happens because the application does 84) the DNS resolve before going to the SOCKS proxy.) One option is to use 85) Tor's built-in support for doing DNS resolves; but you need to ask via 86) our new socks extension for that, and no applications do this yet. A 87) nicer option is to use Tor's controller interface: you intercept the 88) DNS resolve, tell Tor about the resolve, and Tor replies with a dummy IP 89) address. Then the application makes a connection through Tor to that dummy 90) IP address, and Tor automatically maps it back to the original query.</li> 91) <li>People running servers tell us they want to have one BandwidthRate 92) during some part of the day, and a different BandwidthRate at other parts 93) of the day. Rather than coding this inside Tor, we should have a little 94) script that speaks via the <a href="/gui/">Tor Controller Interface</a>, 95) and does a setconf to change the bandwidth rate. Perhaps it would run out 96) of cron, or perhaps it would sleep until appropriate times and then do 97) its tweak (that's probably more portable). Can somebody write one for us 98) and we'll put it into <a href="/cvs/tor/contrib/">tor/contrib/</a>?</li> 99) <li>We have a variety of ways to <a 100) href="http://wiki.noreply.org/noreply/TheOnionRouter/TorFAQ#ChooseEntryExit">exit 101) the Tor network from a particular country</a>, but they all 102) require specifying the nickname of a particular Tor server. It 103) would be nice to be able to specify just a country, and 104) have something automatically pick. This requires having some 105) component that knows what country each Tor node is in. The <a 106) href="http://serifos.eecs.harvard.edu:8000/cgi-bin/exit.pl">script on 107) serifos</a> manually parses whois entries for this. Maybe geolocation 108) data will also work?</li> 109) <li>Speaking of geolocation data, somebody should draw a map of the Earth 110) with a pin-point for each Tor server. Bonus points if it updates as the 111) network grows and changes.</li> 112) <li>Tor provides anonymous connections, but we don't support 113) keeping multiple pseudonyms in practice (say, in case you 114) frequently go to two websites and if anybody knew about both of 115) them they would conclude it's you). We should find a good approach 116) and interface for handling pseudonymous profiles in Tor. See <a
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`117) href="http://archives.seul.org/or/talk/Dec-2004/msg00086.html">this`
revamp again Roger Dingledine authored 19 years ago	`118) post</a> and <a 119) href="http://archives.seul.org/or/talk/Jan-2005/msg00007.html">followup</a> 120) for details.</li>`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`121) </ol> 122)`
revamp again Roger Dingledine authored 19 years ago	`123) <h2>Documentation</h2>`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`124) <ol>`
revamp again Roger Dingledine authored 19 years ago	`125) <li>Please volunteer to help maintain this website: code, content,`
and i would like a pony. Roger Dingledine authored 19 years ago	`126) css, layout. Step one is to hang out on the IRC channel until we 127) get to know you.</li>`
revamp again Roger Dingledine authored 19 years ago	128) <li>We have too much documentation --- it's spread out too much and 129) duplicates itself in places. Please send us patches, pointers, and 130) confusions about the documentation so we can clean it up.</li> 131) <li>Help translate the web page and documentation into other 132) languages. See the <a href="translation.html">translation 133) guidelines</a> if you want to help out. We also need people to help 134) maintain the existing (Italian and German) translations.</li> 135) <li>Investigate privoxy vs. freecap vs. sockscap for win32 clients. Are 136) there usability or stability issues that we can track down and 137) resolve, or at least inform people about?</li>
and i would like a pony. Roger Dingledine authored 19 years ago	`138) <li>Can somebody help Matt Edman with the documentation and how-tos 139) for his <a href="http://freehaven.net/~edmanm/torcp/">Windows Tor`
Take the pony to the vet Thomas Sjögren authored 19 years ago	`140) Controller</a>?</li>`
revamp again Roger Dingledine authored 19 years ago	`141) <li>Evaluate, create, and <a 142) href="http://wiki.noreply.org/wiki/TheOnionRouter/TorifyHOWTO">document 143) a list of programs</a> that can be routed through Tor.</li> 144) <li>We need better documentation for dynamically intercepting`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`145) connections and sending them through Tor. tsocks (Linux) and freecap`
revamp again Roger Dingledine authored 19 years ago	146) (Windows) seem to be good candidates.</li> 147) <li>We have a huge list of <a href="/support.html">potentially useful 148) programs that interface to Tor</a>. Which ones are useful in which 149) situations? Please help us test them out and document your results.</li> 150) </ol> 151) 152) <h2>Coding and Design</h2> 153) <ol> 154) <li>We recommend Privoxy as a good scrubbing web proxy, but it's 155) unmaintained and still has bugs, especially on Windows. While we're at 156) it, what sensitive information is not kept safe by Privoxy? Are there 157) other scrubbing web proxies that are more secure?</li> 158) <li>tsocks appears to be unmaintained: we have submitted several patches 159) with no response. Can somebody volunteer to start maintaining a new 160) tsocks branch? We'll help.</li>
ask people to add socks4a s... Roger Dingledine authored 19 years ago	161) <li>Some popular clients that people use with Tor 162) include <a href="http://gaim.sourceforge.net/">Gaim</a> 163) and <a href="http://www.xchat.org/">xchat</a>. These 164) programs support socks, but they don't support <a 165) href="http://wiki.noreply.org/noreply/TheOnionRouter/TorFAQ#SOCKSAndDNS">socks4a 166) or socks5-with-remote-dns</a>. Please write a patch for them and submit 167) it to the appropriate people. Let us know if you've written the patch 168) but you're having trouble getting it accepted.</li>
revamp the volunteer page.... Roger Dingledine authored 19 years ago	169) <li>Right now the hidden service descriptors are being stored on just a few 170) directory servers. This is bad for privacy and bad for robustness. To get 171) more robustness, we're going to need to make hidden service descriptors 172) even less private because we're going to have to mirror them onto many 173) places. Ideally we'd like to separate the storage/lookup system from the 174) Tor directory servers entirely. Any reliable distributed storage system 175) will do, as long as it allows authenticated updates. As far as we know, 176) no implemented DHT code supports authenticated updates. What's the right 177) next step?</li> 178) <li>Tor exit servers need to do many DNS resolves in parallel. But 179) gethostbyname() is poorly designed --- it blocks until it has finished
revamp again Roger Dingledine authored 19 years ago	`180) resolving a query --- so it requires its own thread or process. So Tor 181) is forced to spawn many separate DNS "worker" threads. There are some 182) asynchronous DNS libraries out there, but historically they are buggy and`
a gpl async dns lib is not... Roger Dingledine authored 19 years ago	`183) abandoned. Are any of them stable, fast, clean, and free software? (To 184) be clear, it needs to be more free than GPL.) If so`
revamp again Roger Dingledine authored 19 years ago	`185) (or if we can make that so), we should integrate them into Tor. See <a`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`186) href="http://archives.seul.org/or/talk/Sep-2005/msg00001.html">Agl's`
revamp again Roger Dingledine authored 19 years ago	`187) post</a> for one potential approach.</li> 188) <li>Tor 0.1.1.x includes support for hardware crypto accelerators via 189) OpenSSL. Nobody has ever tested it, though. Does somebody want to get 190) a card and let us know how it goes?</li>`
start to rewrite the volunt... Roger Dingledine authored 19 years ago	`191) <li>Because Tor servers need to store-and-forward each cell they handle, 192) high-bandwidth Tor servers end up using dozens of megabytes of memory 193) just for buffers. We need better heuristics for when to shrink/expand 194) buffers. Maybe this should be modelled after the Linux kernel buffer 195) design, where you have many smaller buffers that link to each other, 196) rather than monolithic buffers?</li>`
revamp again Roger Dingledine authored 19 years ago	`197) <li>How do ulimits work on Win32, anyway? We're having problems,`
start to rewrite the volunt... Roger Dingledine authored 19 years ago	`198) especially on older Windowses with people running out of file 199) descriptors, connection buffer space, etc. (We should handle 200) WSAENOBUFS as needed, look at the MaxConnections registry entry, 201) look at the MaxUserPort entry, and look at the TcpTimedWaitDelay 202) entry. We may also want to provide a way to set them as needed. See <a`
Make website valid xhtml (t... Thomas Sjögren authored 19 years ago	`203) href="http://bugs.noreply.org/flyspray/index.php?do=details&id=98">bug`
start to rewrite the volunt... Roger Dingledine authored 19 years ago	`204) 98</a>.)</li> 205) <li>Encrypt identity keys on disk, and implement passphrase protection 206) for them. Right now they're just stored in plaintext.</li>`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`207) <li>Patches to Tor's autoconf scripts. First, we'd like our configure.in 208) to handle cross-compilation, e.g. so we can build Tor for obscure`
revamp again Roger Dingledine authored 19 years ago	`209) platforms like the Linksys WRTG54. Second, we'd like the with-ssl-dir 210) option to disable the search for ssl's libraries.</li>`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	211) <li>Implement reverse DNS requests inside Tor (already specified in 212) Section 5.4 of <a href="/cvs/tor/doc/tor-spec.txt">tor-spec.txt</a>).</li> 213) <li>Perform a security analysis of Tor with <a 214) href="http://en.wikipedia.org/wiki/Fuzz_testing">"fuzz"</a>. Determine 215) if there good fuzzing libraries out there for what we want. Win fame by 216) getting credit when we put out a new release because of you!</li> 217) <li>How hard is it to patch bind or a DNS proxy to redirect requests to
revamp again Roger Dingledine authored 19 years ago	`218) Tor via our <a href="http://wiki.noreply.org/noreply/TheOnionRouter/TorFAQ#CompatibleApplications">tor-resolve socks extension</a>? What about to convert UDP DNS`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`219) requests to TCP requests and send them through Tor?</li>`
and i would like a pony. Roger Dingledine authored 19 years ago	`220) <li>Tor uses TCP for transport and TLS for link 221) encryption. This is nice and simple, but it means all cells 222) on a link are delayed when a single packet gets dropped, and 223) it means we can only reasonably support TCP streams. We have a <a 224) href="http://wiki.noreply.org/noreply/TheOnionRouter/TorFAQ#TransportIPnotTCP">list 225) of reasons why we haven't shifted to UDP transport</a>, but it would be`
Take the pony to the vet Thomas Sjögren authored 19 years ago	`226) great to see that list get shorter.</li>`
revamp again Roger Dingledine authored 19 years ago	`227) <li>We're not that far from having IPv6 support for destination addresses 228) (at exit nodes). If you care strongly about IPv6, that's probably the 229) first place to start.</li>`
revamp the volunteer page.... Roger Dingledine authored 19 years ago	`230) </ol> 231) 232) <h2>Research</h2> 233) <ol>`
revamp again Roger Dingledine authored 19 years ago	234) <li>The "website fingerprinting attack": make a list of a few 235) hundred popular websites, download their pages, and make a set of 236) "signatures" for each site. Then observe a Tor client's traffic. As 237) you watch him receive data, you quickly approach a guess about which 238) (if any) of those sites he is visiting. First, how effective is 239) this attack on the deployed Tor codebase? Then start exploring 240) defenses: for example, we could change Tor's cell size from 512 241) bytes to 1024 bytes, we could employ padding techniques like <a 242) href="http://freehaven.net/anonbib/#timing-fc2004">defensive dropping</a>, 243) or we could add traffic delays. How much of an impact do these have, 244) and how much usability impact (using some suitable metric) is there from 245) a successful defense in each case?</li>
point to a paper that might... Roger Dingledine authored 19 years ago	246) <li>The "end-to-end traffic confirmation attack": 247) by watching traffic at Alice and at Bob, we can <a 248) href="http://freehaven.net/anonbib/#danezis:pet2004">compare 249) traffic signatures and become convinced that we're watching the same 250) stream</a>. So far Tor accepts this as a fact of life and assumes this 251) attack is trivial in all cases. First of all, is that actually true? How 252) much traffic of what sort of distribution is needed before the adversary 253) is confident he has won? Are there scenarios (e.g. not transmitting much) 254) that slow down the attack? Do some traffic padding or traffic shaping 255) schemes work better than others?</li>
one pony is not enough. Roger Dingledine authored 19 years ago	256) <li>The "run two servers and wait attack": Tor clients pick a new path 257) periodically. If the adversary runs an entry and an exit, eventually some 258) Alice will build a circuit that begins and ends with his nodes. The 259) current Tor threat model assumes the end-to-end traffic confirmation attack 260) is trivial, and instead aims to limit the chance that the adversary will 261) be able to see both sides of a circuit. One way to help this is 262) <a href="http://freehaven.net/anonbib/#wright03">helper 263) nodes</a> -- Alice picks a small set of entry nodes and uses them always. 264) But in reality, Tor nodes disappear sometimes. So it would seem that the 265) attack continues, albeit slower than before. How much slower?</li>
revamp again Roger Dingledine authored 19 years ago	266) <li>The "routing zones attack": most of the literature thinks of 267) the network path between Alice and her entry node (and between the 268) exit node and Bob) as a single link on some graph. In practice, 269) though, the path traverses many autonomous systems (ASes), and <a 270) href="http://freehaven.net/anonbib/#feamster:wpes2004">it's not uncommon 271) that the same AS appears on both the entry path and the exit path</a>. 272) Unfortunately, to accurately predict whether a given Alice, entry, 273) exit, Bob quad will be dangerous, we need to download an entire Internet 274) routing zone and perform expensive operations on it. Are there practical 275) approximations, such as avoiding IP addresses in the same /8 network?</li>
revamp the volunteer page.... Roger Dingledine authored 19 years ago	276) <li>Tor doesn't work very well when servers have asymmetric bandwidth 277) (e.g. cable or DSL). Because Tor has separate TCP connections between 278) each hop, if the incoming bytes are arriving just fine and the outgoing 279) bytes are all getting dropped on the floor, the TCP push-back mechanisms 280) don't really transmit this information back to the incoming streams. 281) Perhaps Tor should detect when it's dropping a lot of outgoing packets, 282) and rate-limit incoming streams to regulate this itself? I can imagine 283) a build-up and drop-off scheme where we pick a conservative rate-limit, 284) slowly increase it until we get lost packets, back off, repeat. We 285) need somebody who's good with networks to simulate this and help design 286) solutions; and/or we need to understand the extent of the performance 287) degradation, and use this as motivation to reconsider UDP transport.</li> 288) <li>A related topic is congestion control. Is our 289) current design sufficient once we have heavy use? Maybe 290) we should experiment with variable-sized windows rather 291) than fixed-size windows? That seemed to go well in an <a 292) href="http://www.psc.edu/networking/projects/hpn-ssh/theory.php">ssh 293) throughput experiment</a>. We'll need to measure and tweak, and maybe 294) overhaul if the results are good.</li>
revamp again Roger Dingledine authored 19 years ago	295) <li>To let dissidents in remote countries use Tor without being blocked 296) at their country's firewall, we need a way to get tens of thousands of 297) relays, not just a few hundred. We can imagine a Tor client GUI that 298) has a "help China" button at the top that opens a port and relays a 299) few KB/s of traffic into the Tor network. (A few KB/s shouldn't be too 300) much hassle, and there are few abuse issues since they're not being exit 301) nodes.) But how do we distribute a list of these volunteer clients to the 302) good dissidents in an automated way that doesn't let the country-level 303) firewalls intercept and enumerate them? Probably needs to work on a 304) human-trust level.</li>
one pony is not enough. Roger Dingledine authored 19 years ago	305) <li>Tor circuits are built one hop at a time, so in theory we have the 306) ability to make some streams exit from the second hop, some from the 307) third, and so on. This seems nice because it breaks up the set of exiting 308) streams that a given server can see. But if we want each stream to be safe, 309) the "shortest" path should be at least 3 hops long by our current logic, so 310) the rest will be even longer. We need to examine this performance / security 311) tradeoff.</li> 312) <li>It's not that hard to DoS Tor servers or dirservers. Are client 313) puzzles the right answer? What other practical approaches are there? Bonus 314) if they're backward-compatible with the current Tor protocol.</li>