TODO 8.9 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222
  1. $Id$
  2. Legend:
  3. SPEC!! - Not specified
  4. SPEC - Spec not finalized
  5. N - nick claims
  6. R - arma claims
  7. P - phobos claims
  8. - Not done
  9. * Top priority
  10. . Partially done
  11. o Done
  12. D Deferred
  13. X Abandoned
  14. Non-Coding, Soon:
  15. N - Mark up spec; note unclear points about servers
  16. N - Mention controller libs someplace.
  17. D FAQ entry: why gnutls is bad/not good for tor
  18. P - flesh out the rest of the section 6 of the faq
  19. P - gather pointers to livecd distros that include tor
  20. R . more pictures from ren. he wants to describe the tor handshake, i want to
  21. talk about hidden services.
  22. NR- write a spec appendix for 'being nice with tor'
  23. - tor-in-the-media page
  24. - Remove need for HACKING file.
  25. Website:
  26. - we need to merge documentation and support
  27. - and pare it down
  28. - and merge developers into documentation too
  29. - or at least, remove developers from the title bar
  30. - and remove home and make the "Tor" picture be the link to home.
  31. - put the logo on the website, in source form, so people can put it on
  32. stickers directly, etc.
  33. for 0.1.1.x:
  34. N - if they're trying to be a tor server and they're running
  35. win 98 or win me, give them a message talking about The Bug.
  36. R . Rename 'helper' to 'guard'.
  37. N - Display the reasons in 'destroy' and 'truncated' cells under some
  38. circumstances?
  39. N . Directory changes
  40. . Client side:
  41. - Do we want to rate-limit downloads of each identity, or do something
  42. else to download even less?
  43. - Do we want to refrain from downloading non-running or non-verified
  44. descriptors? This is potentially dangerous.
  45. - Non-directories don't need to keep descriptors in memory.
  46. R - Christian Grothoff's attack of infinite-length circuit.
  47. the solution is to have a separate 'extend-data' cell type
  48. which is used for the first N data cells, and only
  49. extend-data cells can be extend requests.
  50. - Specify, including thought about
  51. - Implement
  52. R - When we connect to a Tor server, it sends back a signed cell listing
  53. the IP it believes it is using. Use this to block dvorak's attack.
  54. Also, this is a fine time to say what time you think it is.
  55. - Verify that a new cell type is okay with deployed codebase
  56. - Specify
  57. - Implement
  58. R - clients prefer to avoid exit nodes for non-exit path positions.
  59. - find 10 dirservers.
  60. - What are criteria to be a dirserver? Write a policy.
  61. - the tor client can do the "automatic proxy config url" thing?
  62. Deferred from 0.1.1.x:
  63. - Automatically determine what ports are reachable and start using
  64. those, if circuits aren't working and it's a pattern we recognize
  65. ("port 443 worked once and port 9001 keeps not working").
  66. N - Should router info have a pointer to routerstatus?
  67. - We should at least do something about the duplicated fields.
  68. N . Additional controller features
  69. - change circuit status events to give more details, like purpose,
  70. whether they're internal, when they become dirty, when they become
  71. too dirty for further circuits, etc.
  72. R - What do we want here, exactly?
  73. N - Specify and implement it.
  74. - Change stream status events analogously.
  75. R - What do we want here, exactly?
  76. N - Specify and implement it.
  77. - Make other events "better".
  78. - Change stream status events analogously.
  79. R - What do we want here, exactly?
  80. N - Specify and implement it.
  81. - Make other events "better" analogously
  82. R - What do we want here, exactly?
  83. N - Specify and implement it.
  84. . Expose more information via getinfo:
  85. - import and export rendezvous descriptors
  86. - Review all static fields for additional candidates
  87. - Allow EXTENDCIRCUIT to unknown server.
  88. - We need some way to adjust server status, and to tell tor not to
  89. download directories/network-status, and a way to force a download.
  90. - It would be nice to request address lookups from the controller
  91. without using SOCKS.
  92. - Make everything work with hidden services
  93. X switch accountingmax to count total in+out, not either in or
  94. out. it's easy to move in this direction (not risky), but hard to
  95. back out if we decide we prefer it the way it already is. hm.
  96. - cpu fixes:
  97. - see if we should make use of truncate to retry
  98. R - kill dns workers more slowly
  99. . Directory changes
  100. . Some back-out mechanism for auto-approval
  101. - a way of rolling back approvals to before a timestamp
  102. - Consider minion-like fingerprint file/log combination.
  103. - config option to publish what ports you listen on, beyond
  104. ORPort/DirPort. It should support ranges and bit prefixes (?) too.
  105. - Parse this.
  106. - Relay this in networkstatus.
  107. - packaging and ui stuff:
  108. . multiple sample torrc files
  109. - uninstallers
  110. . for os x
  111. . figure out how to make nt service stuff work?
  112. . Document it.
  113. . Add version number to directory.
  114. N - Vet all pending installer patches
  115. - Win32 installer plus privoxy, sockscap/freecap, etc.
  116. - Vet win32 systray helper code
  117. - document:
  118. - torcp needs more attention in the tor-doc-win32.
  119. - recommend gaim.
  120. - unrecommend IE because of ftp:// bug.
  121. - torrc.complete.in needs attention?
  122. - Bind to random port when making outgoing connections to Tor servers,
  123. to reduce remote sniping attacks.
  124. - Have new people be in limbo and need to demonstrate usefulness
  125. before we approve them.
  126. - Clients should estimate their skew as median of skew from servers
  127. over last N seconds.
  128. - Security
  129. - Alices avoid duplicate class C nodes.
  130. - Analyze how bad the partitioning is or isn't.
  131. . Update the hidden service stuff for the new dir approach.
  132. - switch to an ascii format, maybe sexpr?
  133. - authdirservers publish blobs of them.
  134. - other authdirservers fetch these blobs.
  135. - hidserv people have the option of not uploading their blobs.
  136. - you can insert a blob via the controller.
  137. - and there's some amount of backwards compatibility.
  138. - teach clients, intro points, and hidservs about auth mechanisms.
  139. - come up with a few more auth mechanisms.
  140. . Come up with a coherent strategy for bandwidth buckets and TLS. (The
  141. logic for reading from TLS sockets is likely to overrun the bandwidth
  142. buckets under heavy load. (Really, the logic was never right in the
  143. first place.) Also, we should audit all users of get_pending_bytes().)
  144. - Make it harder to circumvent bandwidth caps: look at number of bytes
  145. sent across sockets, not number sent inside TLS stream.
  146. - Make router_is_general_exit() a bit smarter once we're sure what it's for.
  147. - Directory "helper".
  148. - rewrite how libevent does select() on win32 so it's not so very slow.
  149. - Write limiting; separate token bucket for write
  150. - Audit everything to make sure rend and intro points are just as likely to
  151. be us as not.
  152. - Do something to prevent spurious EXTEND cells from making middleman
  153. nodes connect all over. Rate-limit failed connections, perhaps?
  154. Future version:
  155. - Limit to 2 dir, 2 OR, N SOCKS connections per IP.
  156. - Handle full buffers without totally borking
  157. - Rate-limit OR and directory connections overall and per-IP and
  158. maybe per subnet.
  159. - Hold-open-until-flushed now works by accident; it should work by
  160. design.
  161. - DoS protection: TLS puzzles, public key ops, bandwidth exhaustion.
  162. - Specify?
  163. - tor-resolve script should use socks5 to get better error messages.
  164. - make min uptime a function of the available choices (say, choose 60th
  165. percentile, not 1 day.)
  166. - Track uptime as %-of-time-up, as well as time-since-last-down.
  167. - hidserv offerers shouldn't need to define a SocksPort
  168. * figure out what breaks for this, and do it.
  169. - auth mechanisms to let hidden service midpoint and responder filter
  170. connection requests.
  171. - Relax clique assumptions.
  172. X start handling server descriptors without a socksport?
  173. - tor should be able to have a pool of outgoing IP addresses
  174. that it is able to rotate through. (maybe)
  175. Blue-sky:
  176. - Patch privoxy and socks protocol to pass strings to the browser.
  177. - Standby/hotswap/redundant hidden services.
  178. - Robust decentralized storage for hidden service descriptors.
  179. - The "China problem"
  180. - Allow small cells and large cells on the same network?
  181. - Cell buffering and resending. This will allow us to handle broken
  182. circuits as long as the endpoints don't break, plus will allow
  183. connection (tls session key) rotation.
  184. - Implement Morphmix, so we can compare its behavior, complexity, etc.
  185. - Other transport. HTTP, udp, rdp, airhook, etc. May have to do our own
  186. link crypto, unless we can bully openssl into it.
  187. - Need a relay teardown cell, separate from one-way ends.
  188. (Pending a user who needs this)
  189. - Handle half-open connections: right now we don't support all TCP
  190. streams, at least according to the protocol. But we handle all that
  191. we've seen in the wild.
  192. (Pending a user who needs this)