13:29:25 #startmeeting 13:29:25 Meeting started Wed Feb 3 13:29:25 2016 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:29:25 Useful Commands: #action #agreed #help #info #idea #link #topic. 13:29:29 who is around? I am! 13:29:32 I am! 13:29:36 hi hi hi 13:29:41 strangely, i am too. 13:29:42 hello! 13:29:43 * dgoulet back from long travelling! 13:29:49 good to see you all! 13:30:47 pings out to other net team friends: teor1 Yawning athena asn_ isabela mikeperry isis . (Some of you are asleep, I know. no worries.) 13:30:59 who wants to do the first check-in? 13:31:14 * asn_ around but not first check-in plz 13:31:23 my check-in is easy: i have been distracted and commented on a few not-urgent tickets last week. 13:31:49 I can go next 13:31:54 go 13:32:14 I worked to organize a set of proposal discussion meetings: https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/MeetingSchedule . I hope that works out 13:32:25 I did a huge triage round on our 0.2.8 and TorCoreTeam20160* bugs. 13:32:34 I wrote a bunch of stuff for isa, reviewed a bunch of code, etc. 13:32:42 Right now I"m focused on putting out an alpha release 13:32:52 then I'm hoping to merge a few more things and spend the rest of the month hacking 13:32:55 i am excited for the alpha release! 13:33:12 I really want my ed fix stuff to get in. I have fewer big patches in this release than I'd like 13:33:37 armadev: you could accelerate it if you said "Hey nick, let me edit the changelog this time! I can be so fast about it!" But I bet you've got other stuff going on too 13:33:49 * asn_ can go next 13:33:54 Everybody should have a look at https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/MeetingSchedule if you haven't. 13:34:00 one more thing for me: 13:34:11 (is there something wrong with ed, that he needs a fix?) 13:34:16 ed25519 13:34:53 (he is now a robot) 13:35:11 one more thing: this term at MIT I'm mentoring a group of students in a class called 6.S194, which is an opensource large-scale software engineering lab. We start with a coding sprint this weekend. There's a set of things I need to do to prepare; I'll be asking for help with that later on. 13:35:20 Most of it will be of general use for onboarding other devs in the future. 13:35:39 I'll have between 4-6 students. We'll start with small tasks and then they'll pick a term project. 13:35:47 I hope this doesn't eat my life. 13:35:52 that sounds very interesting :) 13:35:55 indeed ^ 13:35:58 I hope we all have fun and learn things 13:36:00 also from the PoV of the students 13:36:06 asn_: ok, over to you! 13:36:15 ok hmm 13:36:16 * dgoulet goes after asn 13:36:17 so the past week 13:36:33 surprise. i worked a bit on prop250. fixed up the spec, and also worked on a tor branch of dgoulet. 13:36:49 i also worked on prop247. i wrote a simulator that aims to find ideal guard rotation lenghts and stuff. 13:37:17 it had some interesting results. I'm still not sure if they are bugs or not. i need to spend more time on this project to learn. 13:37:28 i also spent more time than expected with hackerone 13:37:39 because guido vranken found like 4 vulns in tor/libevent the past week 13:38:12 we've said "thanks", right, and not just paid him? 13:38:13 it was a bit boring work because the bugs were not very exciting and I'm not sure how much effort I should spend in reviewing/fixing them. 13:38:16 we should make sure to say thanks 13:38:19 nickm: yes for sure 13:38:20 great 13:38:25 many thanks have been said 13:38:53 ok that's for hackerone. there is still one unfixed tor bug (and one that has a patch and not merged). but they aren ot very serious 13:39:00 i also talked with olabini 13:39:13 who is starting to work on tor this month with a team of 4 people. 13:39:24 he was looking for hidden service projects. 13:39:39 ola + one of his people will be in valencia, btw 13:39:42 i suggested him to work on the prop259 simulator that isis was writing, and then maybe even write tor code that implements the best algorithm (or any algoruithm) 13:40:00 i think they started working on tor this monday. and they will go on for 3-4 weeks. 13:40:07 still haven't heard from them this week though. 13:40:19 ehm and that's that. 13:40:26 today i need to write the sponsorr rreport for january 13:40:31 so i will remember more things i did. 13:40:32 n3ext 13:40:33 asn_: didn't you get the email we sent yesterday? 13:40:40 olabini: hm... 13:40:47 olabini: where was it sent? i think i missed it 13:41:51 dgoulet: go for it 13:42:14 olabini: if you want, you can also brief us up during this meeting :) 13:42:15 hola, ok pasting 13:42:20 Little on _tor_ side, I was in Brussels (Fosdem) most of last week and changed 13:42:21 continent on Tuesday for which I arrived yesterday. I just slept 87 hours in 13:42:21 one night so I'm now semi-usable. But back in EST time!Little on _tor_ side, I was in Brussels (Fosdem) most of last week and changed 13:42:21 continent on Tuesday for which I arrived yesterday. I just slept 87 hours in 13:42:27 (double... sorry) 13:42:40 2 x 87? wow! 13:42:48 I have a _huge_ back log of tor emails so I'll prioritize this today since I think asn is waiting on me for some prop#250 stuff. 13:42:57 I still have one ticket assigned to me in TorCoreTeam201602 for which I'm 13:42:57 stalled waiting on Rob's testing/feedbacks. So my plan is to go over 028 13:42:58 tickets very very soon (today most probably). 13:43:34 I would also like to clear torsocks tickets very soon for which, after Fosdem, I realized lots of people are waiting on 13:44:05 then back to code on prop#224 13:44:09 done 13:44:26 is that everybody who is here? 13:44:33 * karsten can give an update, too. 13:44:36 wooo 13:44:45 after working on little-t-tor for a few days. 13:45:00 so, I wrote code for #8786 that asn is currently testing on his bridge. 13:45:26 the background is that dcf pointed me to a bug in the user number estimates code, 13:45:38 and having those two lines will make those estimates better. 13:45:48 I also worked on #15469, 13:46:02 which basically removes a lot of code. 13:46:19 this is motivated by robgjansen asking whether we really need those stats, and I think we don't. 13:46:45 karsten: what do you think about the resources I pointed to on that ticket, and/or #7352 ? 13:46:47 I have a branch that compiles and passes unit tests, but I didn't test this any more yet. 13:47:04 err not that one 13:47:05 #7532 13:47:40 nickm: to be honest, I didn't look closely, because if we can take out those stats entirely there's no need to do something smart anymore. 13:47:58 but ... but ... cool probailistic data structure! :( ' ' ' 13:47:58 ;) 13:48:07 I even found ways to fix the CLIENTS_SEEN event and the heartbeat log message. 13:48:09 heh, true. 13:48:18 okay, end of update, I guess. 13:48:27 thanks! 13:48:33 Any more updates? 13:48:36 Is that everyone? 13:49:12 assuming that's everyone! 13:49:16 okay, discussions! 13:49:30 #1: I would like everybody to please have a look over https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/MeetingSchedule 13:49:43 and add yourself to can-attend or conflicts for important stuff as appropriate. 13:49:50 Especially for the next 2-3 meetings 13:50:15 If you'd like to help me plan, and you don't mind publishing the info, you can add yourself to the general availability thing on the bottom. 13:50:37 armadev: I'd like you to try to come to some of these if you can. It's a high-value thing. 13:51:00 yes indeed. i've been thinking i should read some proposals too. 13:51:49 anybody have questions on the process? 13:51:51 armadev: yes you should 13:52:21 i'm not clear yet on which hats i will fit back onto my head. i'm trying to not add ten at once, since i know that will fail. :) 13:53:10 mikeperry: please add yourself to that page too when you see this? 13:53:19 armadev: this is an important hat, and not a super heavy one. 13:53:31 If we get these things right now, it should save us lots of grief in the future 13:53:39 it's only not super heavy if i don't page much back into my brain, and thus am not so useful 13:53:50 but yes, it does look useful 13:54:14 It would be better to have you kinda useful and present than to have you decide that you must not attend because you cannot spend the time to be maximally useful. :) 13:54:17 IMO 13:54:24 your call of course 13:54:40 ok 13:54:47 #item everybody should put themselves on that NetworkTeam/MeetingSchedule page! 13:55:04 I'm in the system! :) 13:55:19 next thing I'm asking for feeback on: https://storm.torproject.org/shared/MlxQEnQEThxGJJeGkcVU4OOibpaOhmL3gH2mh_R7HmG 13:56:07 it's for the 6.S194 hackathon happening this weekend. I'm mainly asking folks to go through bugs that would be good for new tor hackers to do at a hackathon, and mark them easy or onboarding or something 13:56:16 any suggestions on other activities would rock too 13:57:04 does that seem fun to anybody? 13:57:09 * dgoulet reading 13:57:31 with luck, this gets us 1-6 long-term project contributors. 13:57:59 didn't we have a sponsorU plan to write some on-boarding docs for devs? 13:58:00 what do you have in mind for "on-boarding" here? I see that it's ~4 hours of the first day 13:58:01 with no luck, I wind up irritating a couple of MIT profs and a bunch of students and wasting a bunch of everybody's time. But I don't think that's likely :) 13:58:07 armadev: we have some! 13:58:26 then i suggest using those for some of the 'suggestions?' sections :) 13:59:03 nickm: oh nvm I see that it's basically defined in the following bullet points 13:59:08 (are those part of 5, not 4?) 14:00:03 Maybe? :) 14:00:37 are all your peoples going to be linux users? or will there osx, or heaven help us, windows? 14:00:47 I don't know; I guess we'll find out 14:00:58 nickm: should I just write suggestions or tell you and then you write them if you want to keep it? 14:01:07 dgoulet: please just write suggestions 14:04:14 ok, and I'm more or less out of big open questions. Any other discussion topics for now? 14:04:24 nickm: yes question 14:04:26 (Please note that there's a proposal discussion meeting this time tomorrow) 14:04:27 oh oh oh 14:04:30 dgoulet: go for it! 14:04:36 nickm: feature freeze is end of febuary? 14:04:39 after VLC? 14:05:36 current plan is: major features must be in by 1 march; hard freeze is 15 march. I get to decide what's major. My default answer will probably be "that can wait" :) 14:05:37 asking because prop#250 in 028 would be hrm... important 14:05:44 perfect 14:06:04 here's another question: https://lists.torproject.org/pipermail/metrics-team/2015-October/000000.html 14:06:05 hoping to get stable in april 14:06:17 hoping to put out 029 stable in november. 14:06:22 ([metrics-team] RSA-to-ed25519 migration questions) 14:06:42 karsten: can you summarize? 14:08:05 nickm: hmmmm, I'd have to read it in detail (it's been a while), but I think it was about ed25519 identities being the same length as RSA. 14:08:16 "Let's identify which measurement-related tools need work when relays switch from RSA identities to ed25519 identities" 14:09:24 this is not urgent or anything, I'm just afraid that it will keep metrics team people busy in 2016. 14:09:27 So, I'm not planning for ed25519-id-only relays to exist this year. 14:09:42 dumping RSA IDs is for 2017 at the earliest IMO, I think 14:09:50 that's good to know! 14:10:11 sounds like something for valencia? 14:10:16 this discussion? 14:10:43 it just came up when you asked for questions... :) 14:10:54 sounds like a good topic, yeah 14:11:00 great! 14:11:05 thanks for summarizing btw 14:11:10 more topics? 14:11:49 I'm good, will cleanup the email queue 14:12:20 from the #15469 ticket, it looks like nobody cares. but karsten is apparently writing the patch. do people have opinions on how to decide which stats to keep / which to remove? 14:12:33 it seems like i am frequently in the position of saying "hey, some people use that" 14:13:26 do we know somebody who uses these stats in particular? 14:14:16 me, i think. i need to read the ticket more to figure out if i can get by with fewer stats. 14:14:38 i still have hopes that my countme cell will one day be implemented. :) 14:15:20 (since all of our user stats things are wrong in ways we don't understand well, it seems wise to have more of them, which are wrong in different ways, so we can get closer to right) 14:17:05 https://lists.torproject.org/pipermail/tor-dev/2016-January/010258.html 14:17:13 you could also reply to that thread. 14:17:16 nickm: is the 0.2.8.1-alpha changelog currently in the "nick has run all his scripts" state? or is there some preprocessing remaining? 14:17:41 I'm going to re-run the formatting script after we're done editing. I have some unpushed commits. Give me a second on that. 14:18:22 armadev: pushed my edits 14:18:41 armadev: if you're going to poke it, I can step back for a while? I'm trying to make it build with openssl master at the moment. 14:18:45 any more topics? 14:20:01 then, thanks everybody! 14:20:03 #endmeeting