17:59:18 #startmeeting Network team meeting, 12 november 2019 17:59:18 Meeting started Tue Nov 12 17:59:18 2019 UTC. The chair is ahf. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:59:18 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:59:23 o/ 17:59:26 hello world 17:59:26 o/ 17:59:27 hello network team 17:59:29 hello 17:59:38 our pad is at https://pad.riseup.net/p/tor-netteam-2019.1-keep 17:59:39 hello ahf, catalyst, dgoulet! 17:59:40 o/ 18:00:00 remember we have our retrospective too in two hours from now :-) 18:00:11 asn, mikeperry: you around? 18:00:31 hello! 18:00:37 was doing pad 18:01:00 let's take a look at our 0.4.2 status: https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/CoreTorReleases/042Status 18:01:14 asn: my firefox still thinks the 2018 pad is the most important to suggest to me :-( 18:01:53 i don't see any new very high priority tickets here 18:02:14 * nickm just moved #31810 to 0.4.1, since it has been merged into 0.4.2 18:02:22 I'm here 18:02:33 late, sorry. doing pad. 18:03:00 One thing to ask ourselves is: if we fix nothing else in 0.4.2, could we call it stable tomorrow? 18:03:05 ah, thanks nickm - i see the ptach is ready for that one too for 0.4.1 18:03:08 and for 0.4.0 18:03:26 hm, good question 18:04:01 so far 042 is quite nice on my side in terms of client and relays testing I do 18:04:02 To release on time, it's pretty important to get an rc out this week or early next week. And after an rc, we should keep changes to the minimal 18:04:05 i don't know the answer to that. maybe we will get more testing for RC? 18:05:32 I don't see anything in that category. 18:05:41 but it is only bugfixes that can go in now, so wont that (hopefully) mostly be smaller changes? 18:05:48 mikeperry: are you okay with letting #31653 and #31002 wait for 0.4.3? 18:06:21 ahf: yes, but I'd like us to be even stricter about fixes during the rc period 18:06:46 makes sense 18:06:56 nickm: yes, is unfortunate, but I think the simulator is higher priority right now since tobias has just a couple weeks to work on it at the mom ent 18:07:01 okay 18:07:10 I trust your judgment here 18:08:04 are we good with 0.4.2 things? last week we discussed doing stable releases too 18:08:14 but we wanted to discuss that "later in november" or "early december" 18:08:22 yes, that's right. 18:08:23 so maybe we should wait with talking about that to the 0.4.2 rc is out? 18:08:30 good 18:08:39 okay with me. It wouldn't hurt to plan a bit, but we'll need teor aroudn for that. 18:08:57 yeah 18:09:04 ok, let's do our kanban dance: https://dip.torproject.org/torproject/core/tor/-/boards 18:09:17 please make sure everything is up-to-date there with the work you are doing 18:09:26 * asn looks 18:10:52 asn: what's up with #29294 with regards to s27 ? 18:11:16 hm? 18:11:21 #26294? 18:11:26 yes ^ 18:11:37 i think it's going into the -cans for stretch goals 18:11:44 ack 18:11:55 the -must side is the research we did (and the branch) 18:12:03 and the -can side is fixing it given the new restrictions and info 18:12:38 I've moved a few things around. 18:12:59 yes also doing a bit of that 18:14:48 catalyst: while I'm thinking about it, could you get me initial feedback on #32211 soon? It doesn't need to be deep, but even whatever you think after 30min of looking at the docs will help me improve them, and make my next docs better. 18:15:20 nickm: ok, will try to get to it today 18:15:25 thanks! 18:15:32 #30839 is in needs revision and we have it in the backlog 18:15:54 oh, is teor's ticket 18:17:33 Do we have a guess about a timeframe for a gitlab migration? Like, early-2020, mid-2020, etc? 18:17:59 we do, test begins in a week, and then move will be early 2020 most likely 18:18:05 grand 18:18:06 Januray 2020 is the safe bet 18:18:12 assuming nothing is going horribly wrong 18:18:24 okay, let's move to reviews. i think everything has been distributed already 18:18:36 i took a small patch before and reviewed that and assigned reviewer to myself 18:18:53 Can anybody take over one of my reviews? I am kind of swamped. 18:19:09 i have gotten nothing, so i can take one of yours 18:19:15 great 18:19:22 pick the one that you will like the most to get rid of and just assign it to me as reviewer 18:19:36 it will be tomorrow morning it will get review then 18:19:39 Are you comfortable with #19327? 18:20:05 it sounds fun, i know the controller part quite well, and i can probably learn more about the circuit parts of the code 18:20:12 okay, sounds good. 18:20:15 thanks! 18:21:06 how are people feeling now after a few weeks where i have the feeling that people have become better at asking our loud in #tor-dev about getting reviews during the week? 18:21:28 and secondly: have everybody remembered to add 'network-team' as a highlight in their irc client so that we have a way to broadcast questions to everybody in the team when someone wants help with something? 18:21:31 :-) 18:21:54 * gaba is getting all the network-team mentions 18:21:54 yep that email was helpful with the irssi command 18:22:04 same here. red and scary. 18:22:19 ahf: relatedly, do other teams do the group-highlight thing, and are they ok with us contacting them that way? 18:22:20 yellow and powerful here 18:22:34 yes, other teams uses network-team quite often too i think 18:22:46 and i often see it mentioned when i get in and somebody from the network team have answered 18:22:53 catalyst is asking whether there are magic words we can use to highlight them. 18:22:53 dgoulet: :P 18:22:56 I think 18:23:01 ahhh 18:23:05 i don't know what the other teams are doing there 18:23:15 I think that t b b - t e a m works, but I am not sure whether they have that highlighted or not. 18:23:19 i did not hear any team using words for them 18:23:23 Let's ask the PMs to get us a list? :) 18:23:24 the network-team thing was just pulled out of a bag i think. i have no idea if the browser dash team works or something 18:23:38 we can ask at las vegas meeting on thursday 18:23:46 great, I'll make a note 18:23:47 but it is a good idea to push this in the rest of the org a bit like our /me status have moved to other teams too! 18:24:34 i am very happy that people are using it to ask for getting things reviewed during the week and the more people that listens on this the bigger the chance is that somebody sees it and can pick it up :-) 18:24:39 ok, let's move to discussions 18:25:13 i have the first item: can anybody pick up the PR discussion that was a policy discussion a month ago but started out as an s31 recommendation? 18:25:24 i keep pushing it back on my todo list and now it's just silly by now 18:26:27 I could try to do something , but not this week. 18:26:32 anybody, ideally from s31, up for this? maybe this is something that we can wait with till s31 is actually over given that s31 is HOT right now 18:26:43 nickm: maybe we can coordinate it next week and see where we are there? 18:27:00 sure. like, at the s31 meeting? 18:27:21 either that or just the network team meeting next week 18:27:28 or we just speak on a 1:1 during the week 18:27:39 ok. please remind me? 18:27:52 will do, i keep it on my todo list but wont feel bad about not doing naything on it this week 18:27:57 thanks! 18:27:59 next item: 18:28:02 - nickm: How would we feel about moving all/most of our doc/HACKING stuff into doxygen? 18:28:11 i am personally HUGELY up for that 18:28:16 i think it would be awesome :- 18:28:17 :-) 18:28:20 (i'm mainly looking for the sense of the team on this) 18:28:43 btw, everybody should feel encouraged to look at https://src-ref.docs.torproject.org/tor/index.html and see what else they'd like there. 18:28:49 this is the best our doxygen output has ever been 18:29:27 nickm: i'm in favor of moving stuff into doxygen. among other things, i'm not sure we had a canonical dialect of markdown for doc/HACKING? 18:29:36 I don't think we do. 18:29:45 it looks great 18:30:04 It now has links to the source code, and the source-code is even cross-referenced :) 18:30:58 mikeperry, asn, dgoulet: any thoughts on this 18:31:18 0 18:31:18 ok. So, I'll open a ticket about moving more stuff into doxygen. Anybody who thinks it's a bad idea can let me know; I won't be going wild on this for at least a week 18:31:25 I'm thinking as I'm a fan of less browser :P ... and I like having basic doc of code struct or just "how to contribute" usually in the git repo 18:31:36 but overall, all our doc in doxygen is a great idea 18:31:43 (not spread around) 18:31:45 dgoulet: i don't think they are mutually exclusive? i think it is just some extra metadata in the files? 18:31:50 right 18:32:03 i'm fine with doc/HACKING as it is right now 18:32:12 i think the benefit will mainly be for other contribuotrs 18:32:32 generally a fan of doxygen.. I think I voted for markdown-only (no html) in the code style survey 18:32:51 okay 18:33:08 i think what i read from this is that we are fine keeping it as markdown, but include it in the doxygen output somehow 18:33:18 ok. I'll try to see if there's a way to do that. I believe that there is. 18:33:27 yeah markdown + 18:33:28 yeah, i think so too 18:33:29 ++* 18:33:30 cool! 18:33:36 - nickm: next steps on C style: want me to write a proposal based on our poll and catalyst's suggestions? Or wait for more feedback? 18:33:56 also, catalyst: are you getting responses on the questions you sent out? 18:34:17 does anybody have any feedback to this that they haven't shared with the team yet? if not, i think this can move forward 18:34:20 nickm: i only see yours and asn's 18:34:59 ok. if everybody answers catalyst's questions, that would rock. 18:35:08 so teor, dgoulet, mikeperry and ahf needs to prioritize this this week 18:35:12 so it can move forward 18:35:42 i see no -1 votes so far... so hopefully nobody strongly opposes it? 18:35:59 i do not strongly oppose it from what i recall reading in the airport last week :-) 18:36:22 I'm going to try to draft something under the assumption that the response is generally "BSD KNF, but..." 18:36:30 I know too little of BSD KNF... so I would need to go through taht 18:37:44 i also need to read up on this 18:38:03 okay, sounds like we have action items to this one too 18:38:12 do we have anything else we need to talk about today in the meeting? 18:38:29 * catalyst has a query about manpage splitting 18:38:47 go for it! 18:39:10 ohh, yes! 18:39:13 so someone (teor?) gave feedback to swati against splitting the manpage 18:39:36 i think it could definitely use splitting, but instead of the command line options, split torrc(5) into a separate page 18:39:44 and the non-torrc command line options stay in tor(1) 18:40:09 i would be OK with that. our man page is very long today 18:40:13 I'm very pro that 18:40:20 torrc.5 would be great imo 18:40:55 I would even be okay with splitting torrc.5 by topic 18:41:08 otherwise torrc might be the vast majority of it?) 18:41:21 anyway one of us (probably me?) would have to do the split because needing to adjust the build system 18:41:59 ok 18:42:17 (torrc, by my count, is about 80-90% of the current manpage.) 18:43:04 whoa yeah that's a lot 18:44:16 but anyway i think even splitting to tor(1) and torrc(5) is a good start. maybe we can do more splitting later? 18:45:25 ok by me 18:45:29 wfm too 18:45:45 wfm 18:45:49 i don't see any other helps with items? 18:45:56 anybody who have something else? 18:46:34 * catalyst is good for now 18:47:02 okay, gonna close the meeting now then. see you all for retrospective in 1h12min! 18:47:05 thanks all 18:47:06 #endmeeting