On Mon, Sep 17, 2012 at 11:47 AM, Nick Mathewson nickm@freehaven.net wrote:
Hi, all!
From https://trac.torproject.org/projects/tor/wiki/org/roadmaps/Tor/024 :
"October 10, 2012: Big feature proposal checkpoint. Any large complicated feature which requires a design proposal must have its first design proposal draft by this date."
There are only 23 days till October 10. If there's something you want in 0.2.4.x, and it is the kind of thing that needs a design proposal, and there is no proposal yet, it could be time to start writing!
What needs a design proposal? Generally: anything that involves a change to the Tor protocol; anything whose security implications are nontrivial and need discussion; and anything that will change any Tor specification. Err on the side of "it needs a proposal."
What is a "large complicated feature"? Please assume I'm going to be extremely grumpy here, and err on the side of "it is big". If the writeup of the proposal that explains how it works, or why to do it this way is going to take more than a few paragraphs, it is probably 'big' or 'complex'.
(Note to would-be system-gamers: Please don't send a sketchy incomplete draft as a placeholder to get your foot in the door. That's not cool. If you don't have a draft ready, the feature can wait till 0.2.5.)
(Note also: a feature proposal by this deadline is a necessary condition for getting your big/tricky/complicated feature into 0.2.4, but not a sufficient condition. It also needs to have a working implementation on schedule.)
(Note finally: This is not a promise to not merge stuff that violates this deadline, but I sure will be trying not to merge such stuff.)
yrs,
Nick
On Mon, Oct 1, 2012 at 12:49 PM, Nick Mathewson nickm@freehaven.net wrote:
On Mon, Sep 17, 2012 at 11:47 AM, Nick Mathewson nickm@freehaven.net wrote:
Hi, all!
From https://trac.torproject.org/projects/tor/wiki/org/roadmaps/Tor/024 :
"October 10, 2012: Big feature proposal checkpoint. Any large complicated feature which requires a design proposal must have its first design proposal draft by this date."
Because I've had at least one question:
"Proposal" in this case means a Tor Proposal, as explained at https://gitweb.torproject.org/torspec.git/blob_plain/HEAD:/proposals/001-pro...
A proposal is not a research paper, is not a trac ticket, is not an IRC conversation, and is not a patch.
"What needs a proposal" and "what qualifies as big" are explained in my original email, upthread.
best wishes,
On 10/1/12 6:49 PM, Nick Mathewson wrote:
Hi, all!
From https://trac.torproject.org/projects/tor/wiki/org/roadmaps/Tor/024 :
"October 10, 2012: Big feature proposal checkpoint. Any large complicated feature which requires a design proposal must have its first design proposal draft by this date."
Do you think that this feature to handle TorHS Key via Tor CP would require to take a proposal to be within 0.2.4.x ?
Ref: https://trac.torproject.org/projects/tor/ticket/6411 https://trac.torproject.org/projects/tor/ticket/5976
-naif
On Sat, Oct 6, 2012 at 10:32 AM, Fabio Pietrosanti (naif) lists@infosecurity.ch wrote:
On 10/1/12 6:49 PM, Nick Mathewson wrote:
Hi, all!
From https://trac.torproject.org/projects/tor/wiki/org/roadmaps/Tor/024 :
"October 10, 2012: Big feature proposal checkpoint. Any large complicated feature which requires a design proposal must have its first design proposal draft by this date."
Do you think that this feature to handle TorHS Key via Tor CP would require to take a proposal to be within 0.2.4.x ?
Ref: https://trac.torproject.org/projects/tor/ticket/6411 https://trac.torproject.org/projects/tor/ticket/5976
I tagged it as "maybe proposal" because it involves a protocol change. I am not sure whether the change it requires would be big enough to require a proposal for the 10th of this month or not.
I don't need a huge writeup: Just enough to explain how it's supposed to work and why you'd want to do it that way. The "write the whole private key in one argument" thing seems off to me, as does putting the responsibility for private key management on the controller exclusively.
(That patch is a nonstarter for other reasons, too; see the comment I just posted to the ticket. Heap overflows are not our friends.)