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