001-process.txt 8.0 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183
  1. Filename: 001-process.txt
  2. Title: The Tor Proposal Process
  3. Author: Nick Mathewson
  4. Created: 30-Jan-2007
  5. Status: Meta
  6. Overview:
  7. This document describes how to change the Tor specifications, how Tor
  8. proposals work, and the relationship between Tor proposals and the
  9. specifications.
  10. This is an informational document.
  11. Motivation:
  12. Previously, our process for updating the Tor specifications was maximally
  13. informal: we'd patch the specification (sometimes forking first, and
  14. sometimes not), then discuss the patches, reach consensus, and implement
  15. the changes.
  16. This had a few problems.
  17. First, even at its most efficient, the old process would often have the
  18. spec out of sync with the code. The worst cases were those where
  19. implementation was deferred: the spec and code could stay out of sync for
  20. versions at a time.
  21. Second, it was hard to participate in discussion, since you had to know
  22. which portions of the spec were a proposal, and which were already
  23. implemented.
  24. Third, it littered the specifications with too many inline comments.
  25. [This was a real problem -NM]
  26. [Especially when it went to multiple levels! -NM]
  27. [XXXX especially when they weren't signed and talked about that
  28. thing that you can't remember after a year]
  29. How to change the specs now:
  30. First, somebody writes a proposal document. It should describe the change
  31. that should be made in detail, and give some idea of how to implement it.
  32. Once it's fleshed out enough, it becomes a proposal.
  33. Like an RFC, every proposal gets a number. Unlike RFCs, proposals can
  34. change over time and keep the same number, until they are finally
  35. accepted or rejected. The history for each proposal
  36. will be stored in the Tor repository.
  37. Once a proposal is in the repository, we should discuss and improve it
  38. until we've reached consensus that it's a good idea, and that it's
  39. detailed enough to implement. When this happens, we implement the
  40. proposal and incorporate it into the specifications. Thus, the specs
  41. remain the canonical documentation for the Tor protocol: no proposal is
  42. ever the canonical documentation for an implemented feature.
  43. (This process is pretty similar to the Python Enhancement Process, with
  44. the major exception that Tor proposals get re-integrated into the specs
  45. after implementation, whereas PEPs _become_ the new spec.)
  46. {It's still okay to make small changes directly to the spec if the code
  47. can be
  48. written more or less immediately, or cosmetic changes if no code change is
  49. required. This document reflects the current developers' _intent_, not
  50. a permanent promise to always use this process in the future: we reserve
  51. the right to get really excited and run off and implement something in a
  52. caffeine-or-m&m-fueled all-night hacking session.}
  53. How new proposals get added:
  54. Once an idea has been proposed on the development list, a properly formatted
  55. (see below) draft exists, and rough consensus within the active development
  56. community exists that this idea warrants consideration, the proposal editor
  57. will officially add the proposal.
  58. To get your proposal in, send it to or-dev.
  59. The current proposal editor is Nick Mathewson.
  60. What should go in a proposal:
  61. Every proposal should have a header containing these fields:
  62. Filename, Title, Author, Created, Status.
  63. These fields are optional but recommended:
  64. Target, Implemented-In.
  65. The Target field should describe which version the proposal is hoped to be
  66. implemented in (if it's Open or Accepted). The Implemented-In field
  67. should describe which version the proposal was implemented in (if it's
  68. Finished or Closed).
  69. The body of the proposal should start with an Overview section explaining
  70. what the proposal's about, what it does, and about what state it's in.
  71. After the Overview, the proposal becomes more free-form. Depending on its
  72. the length and complexity, the proposal can break into sections as
  73. appropriate, or follow a short discursive format. Every proposal should
  74. contain at least the following information before it is "ACCEPTED",
  75. though the information does not need to be in sections with these names.
  76. Motivation: What problem is the proposal trying to solve? Why does
  77. this problem matter? If several approaches are possible, why take this
  78. one?
  79. Design: A high-level view of what the new or modified features are, how
  80. the new or modified features work, how they interoperate with each
  81. other, and how they interact with the rest of Tor. This is the main
  82. body of the proposal. Some proposals will start out with only a
  83. Motivation and a Design, and wait for a specification until the
  84. Design seems approximately right.
  85. Security implications: What effects the proposed changes might have on
  86. anonymity, how well understood these effects are, and so on.
  87. Specification: A detailed description of what needs to be added to the
  88. Tor specifications in order to implement the proposal. This should
  89. be in about as much detail as the specifications will eventually
  90. contain: it should be possible for independent programmers to write
  91. mutually compatible implementations of the proposal based on its
  92. specifications.
  93. Compatibility: Will versions of Tor that follow the proposal be
  94. compatible with versions that do not? If so, how will compatibility
  95. be achieved? Generally, we try to not drop compatibility if at
  96. all possible; we haven't made a "flag day" change since May 2004,
  97. and we don't want to do another one.
  98. Implementation: If the proposal will be tricky to implement in Tor's
  99. current architecture, the document can contain some discussion of how
  100. to go about making it work.
  101. Performance and scalability notes: If the feature will have an effect
  102. on performance (in RAM, CPU, bandwidth) or scalability, there should
  103. be some analysis on how significant this effect will be, so that we
  104. can avoid really expensive performance regressions, and so we can
  105. avoid wasting time on insignificant gains.
  106. Proposal status:
  107. Open: A proposal under discussion.
  108. Accepted: The proposal is complete, and we intend to implement it.
  109. After this point, substantive changes to the proposal should be
  110. avoided, and regarded as a sign of the process having failed
  111. somewhere.
  112. Finished: The proposal has been accepted and implemented. After this
  113. point, the proposal should not be changed.
  114. Closed: The proposal has been accepted, implemented, and merged into the
  115. main specification documents. The proposal should not be changed after
  116. this point.
  117. Rejected: We're not going to implement the feature as described here,
  118. though we might do some other version. See comments in the document
  119. for details. The proposal should not be changed after this point;
  120. to bring up some other version of the idea, write a new proposal.
  121. Draft: This isn't a complete proposal yet; there are definite missing
  122. pieces. Please don't add any new proposals with this status; put them
  123. in the "ideas" sub-directory instead.
  124. Needs-Revision: The idea for the proposal is a good one, but the proposal
  125. as it stands has serious problems that keep it from being accepted.
  126. See comments in the document for details.
  127. Dead: The proposal hasn't been touched in a long time, and it doesn't look
  128. like anybody is going to complete it soon. It can become "Open" again
  129. if it gets a new proponent.
  130. Needs-Research: There are research problems that need to be solved before
  131. it's clear whether the proposal is a good idea.
  132. Meta: This is not a proposal, but a document about proposals.
  133. The editor maintains the correct status of proposals, based on rough
  134. consensus and his own discretion.
  135. Proposal numbering:
  136. Numbers 000-099 are reserved for special and meta-proposals. 100 and up
  137. are used for actual proposals. Numbers aren't recycled.