New system for modifying Tor protocol

Nick Mathewson nickm at
Tue Mar 6 22:06:50 UTC 2007

So, after a lot of discussions between me and Roger, he and I decided
that our old procedure for modifying the Tor spec was pretty stupid.
The old procedure was: we'd patch the specification (sometimes forking
first, and sometimes not), then discuss the patches, reach consensus,
and implement the changes.

This had some problems: spec and code would often be out of sync, and
it was hard for people not subscribed to or-cvs to participate in
discussions.  The canonical place to discuss protocol changes is
supposed to be or-dev, after all, but mostly that wasn't happening.

So, we came up with a new proposal system.  It's a lot like a
lighter-weight version of the Python Enhancement Process, which itself
like a lighter-weight version of the RFC process.  It's documented in
We're going to do this pretty informally, and change stuff that
doesn't work as we go along.  Please check it out?

Future big changes to Tor will follow this process, unless Roger and I
screw up (which is, alas, always a possibility).  Discussion of
proposals should take place on this list.

There are a few in-progress proposals that need attention; check out
the index and the other entries in the proposal directory.  (Note that
a few proposals are already marked "CLOSED" or "DEAD", and don't need
comments; the ones that need more work are currently 105, 104, 101,
and 103.  There's an index at )

We now return you to your regularly scheduled mailing list. :)

Nick Mathewson
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 652 bytes
Desc: not available
URL: <>

More information about the tor-dev mailing list