18:29:29 <nickm> #startmeeting
18:29:29 <MeetBot> Meeting started Wed Dec  9 18:29:29 2015 UTC.  The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:29:29 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:30:03 <nickm> hello everybody!  I see isabela and dgoulet.  I think Yawning is asleep, or should be.  I think athena has had a cold, but might be around.  Who else is here for the tor dev meeting?
18:30:15 <Yawning> ?
18:30:23 <nickm> isn't it the middle of the night for you?
18:30:36 <nickm> I thought this was the bad meeting time for you
18:30:45 <nickm> (also, I bet teor is asleep)
18:30:52 <special> hello
18:30:55 <nickm> hi special!
18:31:06 <Yawning> I woke up at like, 3am
18:31:14 <nickm> whoa
18:31:22 <special> I'll do my weekly "I wish I had more time to be a tor developer right now" meeting. soon it will be true..
18:31:57 <nickm> so my agenda is:
18:32:01 <nickm> 1. checking in
18:32:13 <nickm> 2. discussion: roles, contracts for 2016
18:32:23 <nickm> 3. discussion: code review progress, triage, priorities
18:32:43 <nickm> 4. discussion: meeting times
18:32:48 <nickm> 5. discussion: anything else?
18:32:55 <nickm> let's start with the checkins
18:33:25 <nickm> I've been working on reviewing and merging  patches.  I'm so behind that I don't know what else I might get done in December.  But I hope I can do some of my declared dec deliverables too. else I will be in a pickle.
18:34:02 <nickm> I've been looking at AEZ benchmarks with a little help from yawning; I think AEZ would be great for aesni chips but it would be slower on non-aesni chips.  How much slower, I'm not yet sure: I haven't tried the good optimizations yet
18:34:43 <nickm> This coming week I try to merge/review everything I possibly can, and fix some bugs. I have other things I'm supposed to do in december, but unless I trty to get out from under this review pile, it's never going to happen.
18:34:48 <nickm> who's next?
18:35:11 <Yawning> I can go
18:35:16 <nickm> go for it!
18:35:20 <Yawning> clang is dumb, but glad that patch is merged
18:35:31 <Yawning> I'll clean up the sha-3 branch
18:35:41 <Yawning> and deal with paperwork/my contract
18:35:54 <Yawning> and investigate multithreaded relay crypto and the pqfs link handshake
18:36:30 <Yawning> Peter Schwabe posted an updated version of his paper/code
18:37:00 <Yawning> but it was a minor correction so I'm not worried.
18:37:33 <Yawning> If there's other stuff that needs my attention let me know, though goign to that meeting has put me off balance and I need to deal with meatspace issues that I've been putting off as well...
18:37:38 <Yawning> that's it.
18:38:00 <nickm> i'll review your multithreaded client crypto branch rsn
18:38:07 <nickm> dgoulet: you ready to check in?
18:38:09 <dgoulet> I've been doing some 32c3 logistics for the .onion talk, reviewed some patches, continue work on prop#250, Metrics discussions happening also. I'm in a relocation mode tomorrow (ending on friday night).
18:38:11 <dgoulet> :)
18:38:37 <nickm> i'm a little scared about prop250.  There's so much that's been done on it that I haven't been following...
18:38:51 <dgoulet> I plan to hopefully finish with asn prop#250 by end of the month to 1) we can have it ready for 32c3 talk 2) start review rounds
18:39:12 <dgoulet> nickm: we are in deep testing mode since it's a very tricky addition to the voting system
18:39:49 <dgoulet> nickm: expect very contained code, heavily tested and spec updated by the time we ask for merge
18:40:47 <nickm> ok.  I'll pretend it doesn't exist till then? :)
18:41:16 <dgoulet> nickm: yeah, don't be too worry tbh, it's not crazy intrusive and I'm sure you will like it's simplicity with hopefully nice code :)
18:42:19 <dgoulet> hrm so to finish my status
18:42:39 <dgoulet> ah I said all the things that comes to mind :)
18:42:40 <dgoulet> --
18:42:47 <isabela> :)
18:43:40 <nickm> ok. anybody else here ? athena asn ?
18:44:17 <Yawning> nickm: how much do you care about upstreaming my keccak-tiny modifications
18:44:20 <Yawning> I wasn't planning on it
18:44:31 <nickm> I'd like your changes to be in-principle upstreamable.
18:44:35 <nickm> They don't have to get upstreamed.
18:44:38 <Yawning> ok
18:44:56 <nickm> basically, we should try hard to never fork
18:44:59 <Yawning> There's no huge difference between the unrolled vs not on recent intel, with recent gcc
18:45:12 <Yawning> well, I forked the API >.>
18:45:24 <nickm> you didn't throw away the old one, so IMO that's fine
18:45:40 <Yawning> I initially did, but was like "meh, might as well keep it"
18:46:17 <nickm> so unless somebody else would like to check in, isabela, would you like to launch this roles etc discussion?
18:46:34 <nickm> I'll try to chase down athena and asn and who-else-am-i-forgetting-on-my-list?
18:46:48 <dgoulet> teor is probably sleeping
18:46:51 <nickm> yup
18:46:54 <mikeperry> I am here
18:47:00 <nickm> hi!
18:47:04 <isabela> opa
18:47:30 <mikeperry> but I don't have a whole lot to report. trying to find time to address teor's comments on #16861 and child tickets
18:48:08 <Yawning> (oh yeah, I'm poking at obfsNG still)
18:48:10 <mikeperry> not 100% sure what to do about padding one-hop tor2web and single onion circuits
18:48:28 <Yawning> "don't"?
18:48:59 <mikeperry> yeah, the current code won't, unless that orconn also happens to get used for a 3 hop circuit (which is likely, I think)
18:51:13 <mikeperry> the main thing I want to avoid is padding dircons
18:51:17 <nickm> I guess isabela is typing or thinking...
18:51:25 * isabela is waiting mikeperry's update
18:51:29 <nickm> oh!
18:51:36 <nickm> ok, I thought mikeperry was done
18:51:38 <mikeperry> but that's it for me. I worked on mobile stuff lately otherwise. also need to figure out CCC logistics and such
18:51:41 <mikeperry> all done
18:51:56 <isabela> cool i will paste 4 msgs! hang in there!
18:52:03 <isabela> OK! I am bringing this discussion here because is something the full team (contractors, employees, volunteers) should be part of.
18:52:12 <isabela> We are trying to define the different roles within the team / me and nick worked on the first approach, but dgoulet brought a second approach that i like a lot..both can be found here: https://storm.torproject.org/shared/Kzv6ny6veY24A6tjUfmONzQKbTUFrr4-wlwqoGYFip5
18:52:18 <isabela> Also, please keep in mind that we are working on organizing subsystem maintainers (remember https://storm.torproject.org/shared/nolnfrZyz5pfk-EQPpXSg4sX-gtPqUVvp3pESVcyi3w)
18:52:26 <isabela> Goal for discussion: improve roles draft :) / we can talk a little here or move on via email, this will be used for new contracts so we should do this sooner than latter
18:52:29 <isabela> [done]
18:53:15 <nickm> I'm fine with 2nd approach.
18:53:23 <nickm> Except please don't call me master of the universe.
18:53:28 <isabela> lol
18:54:27 <dgoulet> :)
18:54:43 <nickm> (I will only accept a grandiose title if it also has a connotation of arbitrariness, tyrrany, and overweening pride.)
18:55:19 <isabela> what if it comes with a cap you get to wear?
18:55:37 <nickm> (Probably best just to call it "Nick"; if we had to replace me, we wouldn't keep the role as it is; we'd split up the responsibilities.)
18:55:43 <nickm> isabela: I want to see the cap first
18:56:09 <isabela> fair
18:56:18 <nickm> anyways, everyone knows that this is how masters of the universe dress: http://vorpalizer.tumblr.com/image/45997649138
18:57:13 <dgoulet> clearly the Tri-Klops hat :P
18:57:58 <nickm> I have one of these: https://jet.com/product/detail/2094f7a6f1f5406b9df3a792f28c6f20?jcmp=pla:ggl:toys_games_a3:games_a3_other:na:na:na:na:na:2&code=PLA15&k_clickid=8ea282f9-0c7e-47cd-a781-4da7ce1c1ecd&gclid=COrFg-m6z8kCFYIcHwodMjIJcw
18:58:34 <isabela> should we just consider second approach moving forward and work on improving it? I think we can eventually share this with other teams as well
18:58:40 <isabela> any opinions?
18:58:51 <nickm> I think take the second approach and merge the first approach text
18:58:53 <isabela> (sorry if I am pushing !)
18:59:01 <dgoulet> nickm: +1
18:59:08 <nickm> like, with lists of what everybody does
18:59:17 <isabela> sounds good
18:59:22 <isabela> i can work on that
18:59:25 <dgoulet> and we should at least send it to other dev with contracts
18:59:25 <nickm> I'm okay with declaring that all developers are senior fwiw
18:59:29 <nickm> yes
18:59:51 <isabela> dgoulet: yes I will work on that with mikeperry ( mikeperry is that cool?)
19:00:18 <nickm> I'm revising the titles a little
19:00:24 <isabela> cool
19:00:25 <isabela> thanks
19:01:39 <mikeperry> yeah, I can help
19:02:13 <isabela> ok
19:05:14 <isabela> if nothing else, this is what I will do. I will merge both, send a note letting everyone know
19:05:19 <nickm> ok
19:05:24 <nickm> lots of drafts is a good thing
19:05:33 <nickm> (Please somebody change the title I gave myself to something real.)
19:06:55 <isabela> hehe sure
19:07:04 <isabela> I put a call on my email for that too hehe
19:07:10 <nickm> next topic ... let me jump ahead to meeting times.
19:07:19 <isabela> cool!
19:07:32 <nickm> I've noticed that we got more people on wednesday morning chats (my time) than we get at these chats.
19:07:41 <nickm> and the monday-evening ones are pretty sparse too
19:08:59 <isabela> maybe move wed time back and keep monday?
19:09:50 <nickm> that might make sense.  But if we do that we should look at our schedcule again to see who the wed time is bad for, and see if mon is really the best time for them
19:10:07 <isabela> true
19:10:41 <isabela> lets just do that then
19:10:56 <isabela> and see what is a good proposal to get it better
19:11:05 <nickm> makes sense
19:11:13 <nickm> could somebody remind me to send an email if I haven't by tomorrow?
19:11:20 <isabela> sure
19:11:25 <isabela> i do that
19:11:36 <nickm> mikeperry: this would make the wed time bad for californians.
19:11:45 <nickm> but the mon time would probably be fine. maybe.
19:12:13 <mikeperry> yeah, much earlier than this and I will be sad.
19:12:34 <nickm> does the monday one (at 8pm eastern==5pm pacific) work for you?
19:13:39 <mikeperry> yes. I need to set a calandar for it (I keep forgetting, cause it's a weird time). but it works
19:14:36 <isabela> mikeperry: wait till nick confirms the final schedule
19:15:31 <nickm> i think for next week we'll move wednesday (nearly) back and keep the current monday time... and explicitly have "discuss the schedule" be on the agenda for both.
19:16:28 <isabela> i will be -3 utc next week!
19:16:29 * nickm closes this storm pad before I add another non-proposed title for myself
19:16:30 <isabela> :P
19:17:14 <nickm> next topic?
19:17:21 <nickm> (that is, are we ready to move to the next topic?)
19:17:39 <mikeperry> (I may also end up doing the wednesday meetings for a month while I travel for the holidays and CCC and post-CCC stuff. but long-term, the later meeting will be better)
19:18:04 <nickm> I figure two times roughly 10-12 hours apart should make sure we get as many folks as possible.
19:18:21 <dgoulet> I'll be for sure on Wed. in the coming month
19:19:22 <nickm> next topic is code review progress, triage, and priorities
19:19:30 <nickm> We are in the phase of the 0.2.8 cycle where:
19:19:37 <nickm> * We have some bugs in 0.2.7 that somebody needs to fix
19:19:44 <nickm> * we have a collosal amount of code to review already
19:19:55 <nickm> * The above two things are pulling me from declared 0.2.8 tasks
19:20:18 <nickm> I'm glad that I'm getting more help than previously with those things, but I'm still pretty far behind
19:20:55 <Yawning> what's busted in 0.2.7.x
19:21:25 <nickm> https://trac.torproject.org/projects/tor/query?status=needs_information&status=needs_revision&status=reopened&status=needs_review&status=assigned&status=new&status=accepted&group=status&milestone=Tor%3A+0.2.7.x-final isn't a bad introduction
19:21:43 <nickm> #17668 is pretty important
19:22:52 <Yawning> hm
19:23:10 <dgoulet> some 028 in needs_review are quite big patch set...
19:23:42 <nickm> Right now I'm focusing on review-and-merge. After a week or two I'll hit a point where the things I'm reviewing are less important than the stuff I should be writing, I guess
19:23:49 <nickm> I could use some help prioritizing all of these.
19:24:34 <nickm> Alas, the TorCoreTeam201512 ticket is already more stuff than I think we can do in the time we've got
19:25:00 <nickm> I won't be stuck figuring out what to do till next week at least, but it's something we might want to have a think about
19:25:23 <nickm> notably, there is code that we really need (ed25519 link handshake, etc) that I should write, that won't get written if I'm doing code review through January
19:25:33 <special> is there a trac query that gives a useful list of what needs review right now?
19:25:47 <special> the "Tor tickets needing review" has a lot of noise..
19:25:54 <dgoulet> https://trac.torproject.org/projects/tor/query?status=needs_review&component=Tor&milestone=Tor%3A+0.2.8.x-final&col=id&col=summary&col=status&col=owner&col=type&col=priority&col=milestone&order=priority
19:25:58 <dgoulet> 028 needs_review ^
19:26:04 <nickm> see also the 027 needs_review stuff
19:26:31 <nickm> and after looking at that, it might be a good idea to go over needs_revision from time to time to see if there's anything that's nearly-good-enough
19:26:53 <dgoulet> nickm: there are also needs_review that have ACK(s) in them
19:27:10 <nickm> I think I've merged most of those...
19:27:25 <nickm> it would be nice to have a better workflow on all of that, though
19:27:30 <dgoulet> indeed
19:27:53 <dgoulet> #4483, #7144 and #16861 are _big_ those will take some times to review
19:28:15 <isabela> ok
19:29:00 <dgoulet> nickm: #8195 seems ready to merge unless you think we need more testing?
19:29:19 <nickm> I thought you were testing it...
19:29:20 <isabela> we should clean TorCoreTeam201512
19:29:28 <nickm> agreed.
19:29:30 <isabela> and have there only what is priority for Dec
19:29:40 <dgoulet> nickm: hrm nope not right now, I was testing the ExtendPrivateAddress issue :)
19:29:41 <nickm> isabela: now, or when?  And how?
19:29:56 <isabela> keeping in mind is actually ~3 weeks left (with holidays etc could be more 2weeks depending on ppls schedule)
19:30:36 <isabela> nickm: we can close the meeting and do it now after bot is stopped
19:30:43 <nickm> sounds fine to me!
19:30:46 <nickm> #endmeeting