23:00:30 <nickm> #startmeeting network team meeting, 3 July 2019
23:00:30 <MeetBot> Meeting started Wed Jul  3 23:00:30 2019 UTC.  The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot.
23:00:30 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
23:00:32 <ahf> hello
23:01:19 <nickm> our meeting pad is https://pad.riseup.net/p/tor-netteam-2019.1-keep
23:01:32 <nickm> how are we all doing?  Who is here?
23:01:39 * catalyst is here
23:01:50 * teor4 is here
23:02:19 <nickm> hi ahf, catalyst, teor4 !
23:02:36 <nickm> Let's get started.
23:02:41 <ahf> good good
23:03:22 <nickm> First, since we've started catalyst and teor4 as long-term CI role.. how are we doing with the "how are we managing CI failures?" item?  Do we need to keep doing this on a weekly basis?
23:04:32 <nickm> Second is 041 stuff; let's all have a look at https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/CoreTorReleases/041Status if we're not looking at the first thing :)
23:06:20 <ahf> yep
23:07:02 <nickm> on 041, I'm especially wondering if there are any 041-must items that are not marked as such.  To get 041 out on time, I want to make sure we're seeing and acting on any last-minute bugs here, so we get a workable rc out after Stockholm.
23:07:57 <teor4> I think we can report on CI by exception: at the moment, we have tor stem job, chutney CI, and sbws CI failures in needs_review or later
23:08:12 <ahf> nice
23:08:26 <teor4> And we're still working on the tor stem job failure, I think I know what I need to do there
23:08:31 <nickm> ok.  So let's remove this from the "every week" section, and bring it up when there's something to work on?
23:09:43 <catalyst> we can try to update the CI wiki page on a weekly basis, and not talk about it explicitly in the meeting unless there's something exceptional?
23:09:52 <nickm> That sounds reasonable to me.
23:10:45 <catalyst> also maybe we should update the rotations page to delete the CI rotation?
23:10:46 <nickm> on 041: if there's nothing that is missing 041-must, let's move on to the roadmap and the reviewer assignments.  Anything to change there? any surprises?
23:10:55 <nickm> catalyst: yes , I think we should.
23:11:15 <nickm> since there's only a week before stockholm, maybe let's take a decision there on CI too, and postpone the wiki update till then?
23:11:50 <catalyst> what kind of decision on CI?
23:12:01 <nickm> err, not ci
23:12:02 <nickm> sorry
23:12:13 <nickm> a decision on triage, whether to put that as a long-term role too
23:12:48 <catalyst> oh, i'm not sure i remember us talking about that?
23:13:23 <nickm> right, I don't think we've talked about that, but I think we should at Stockholm
23:14:22 <teor4> it migt also be worth considering Coverity as a long-term role
23:14:49 <nickm> yeah
23:14:52 <catalyst> as a separate long-term role? possibly useful
23:15:11 <catalyst> do we have people formally assigned to triage HackerOne as well?
23:16:03 <nickm> I think we do, but I don't know how formally.
23:16:08 <nickm> (asn has been doing that)
23:17:05 <catalyst> i kind of did some of that for a while, but have fallen out of the habit of checking it
23:17:05 <nickm> anything on reviews or roadmaps?  if not, let's more to discussion?
23:17:06 <ahf> i think asn have taken a lot of it and asks for other people's help on-demand
23:17:13 <teor4> We should make a list of extra roles. "sbws maintainer" should be one of them, because I'm not sure how much time juga will habe for that role.
23:17:57 <catalyst> also i suggest we try to stick with N>=2 for long-term roles, to reduce bus factor
23:18:04 <nickm> yeah; that one cannot rotate, but it is important
23:19:42 <nickm> ok. on discussion I see a rotation update; I don't see specific topics on the rotations.
23:20:15 <nickm> so the first thing could be Taylor's question about trac workflow -- whether we could use "accepted" to mean "working on it now" and "assigned" to mean "planning to start soon"?
23:20:23 <nickm> catalyst: want to say anything about that one?
23:21:02 <catalyst> this is partly me inferring what would be helpful to gaba and other people in terms of greater transparency about what we're working on
23:21:51 <catalyst> i think gaba has been asking people to remove themselves as "owner" of tickets they're not actively working on; i figure the assigned/accepted distinction is a helpful one on top of that
23:21:58 <nickm> I believe gaba said something to me at some opint about how it would be useful to have a "working on now" vs "planned to work on" distinction
23:22:49 <nickm> so yeah.  I'm cool with this in principle if gaba says she'd like it and if nobody objects.
23:23:16 <nickm> we could also make this some even more explicit status in the Brave New Future of a gitlab-based tracker
23:23:23 <nickm> or in a lightly tweaked trac
23:23:52 <nickm> My only reservation here is that nobody besides us will understand this distinction basd on the names.  But that's nonblocking.
23:24:13 <catalyst> we can publicize it more broadly
23:25:09 <nickm> That's true, but it would be even better IMO to have the "active" status be called "active" and the planned status be "assigned" or something
23:25:20 <ahf> i think if it gives a better overview for PM's and ourself it could be worth it
23:25:21 <nickm> like, to have the statuses named what they mean
23:25:32 <nickm> (like I said, my concern is nonblocking)
23:25:49 <nickm> ok.  Let's plan this for a quick show of hands at stockholm too ?
23:26:08 <nickm> Next item is feedback on prop#306 -- happy eyeballs
23:26:23 <nickm> teor4: looks like that's yours; want to say something there?
23:26:27 <ahf> si
23:26:42 <teor4> not mine, neal's
23:26:55 <nickm> I meant, I thought you added it to the pad :/
23:27:24 <teor4> oh right. Iain and I have responded. I'm waiting for Iain's response to my response.
23:27:42 <teor4> If someone else wants to weigh in, that would be great. But it's not essential.
23:27:46 <nickm> ok
23:28:07 <teor4> Our draft policy is 2 network team members, and in this context,
23:28:10 <nickm> I'll try to look before stockholm but my backlog is horrible and I trust you and Iain to say sensible things.  If anybody else can look, that will rock.
23:28:17 <teor4> Iain's expertise is relevant
23:28:45 <teor4> Anyone with an opinion on feasible changes to tor's network internals would be helpful
23:29:33 <nickm> anybody in particular you're hoping can look?  it might be good to ask them specifically
23:30:44 <teor4> I don't know who has the most experience with tor's network internals
23:31:03 <teor4> Maybe we should re-do the subsystems document at some point
23:31:07 <teor4> I'll check it now
23:31:20 <nickm> ok
23:31:53 <nickm> next question for discussion is where we stand for our open topics about how to handle proposals
23:32:12 <nickm> my sense is that we don't have consensus on the metaproposal, based on dgoulet's message today, and so we need to talk more
23:32:19 <nickm> is that other people's sense too?
23:33:07 <teor4> (Looks like dgoulet is the person I want to review prop#306)
23:33:34 <nickm> makes sense. I don't know if he's free to do so right now, but it would be great to have his feedback
23:34:29 <catalyst> i think the metaproposal is still OK if we are good about taking "hey maybe we need to talk about this some more" as an objection
23:34:33 <teor4> My sense is that we don't know if we have consensus on the metaproposal, and dgoulet is pointing that out
23:34:56 <catalyst> also where "talk about this more" can also mean "make sure enough people read and understand it"?
23:35:18 <nickm> I don't know whether dgoulet is blocking or not-blocking, and for me, that means I should block.
23:35:26 <ahf> i think so too. i haven't commented on it much because i think it sounds reasonable, but i'm also very flexible with how we do this, as long as it gets specced out (which i think it is here)
23:37:08 <catalyst> part of the art of presiding over a deliberative body is to use tools like calling for objections as a way to make forward progress with fewer formalities. you can fall back on more rigourous methods if need be (if people are acting in good faith)
23:37:20 <nickm> right
23:37:32 <nickm> i had hoped that we had consensus on this point
23:37:57 <nickm> but I am also thinking that david's message _is_ an objection.
23:38:46 <teor4> I'm not sure that acting in good faith is sufficient (although it is necessary)
23:39:17 <teor4> If people are overloaded, or can't keep up with the pace of process change, then we need to slow things down
23:39:38 <nickm> i had hoped that people would say "this is too soon for me so I am blocking this till we have time"
23:39:38 <catalyst> i think we might not be able to resolve this here. but we should think about where and when to work on resolving it, maybe?
23:39:43 <nickm> yeah
23:39:50 <teor4> If people don't feel comfortable responding, then we need to address the underlying issues to make progress
23:39:51 <nickm> My immediate thought is that we should block this proposal till we've talked about it more.
23:42:14 <catalyst> i think we might not have all the necessary people in this meeting now, so maybe let's talk more on the mailing list?
23:42:25 <nickm> makes sense.
23:42:30 <nickm> any objections there?
23:42:31 <ahf> wfm
23:42:58 <teor4> seems fine to me
23:43:09 <teor4> we should also be prepared to talk in Stockholm
23:43:12 <teor4> if needed
23:43:36 <nickm> +1
23:43:39 <ahf> yes
23:43:56 <nickm> we are out of listed discussion topics; do we have any more things for today's meeting?
23:44:30 * ahf has none
23:44:58 * catalyst is good for now
23:45:52 <nickm> ok.  I'll stick around for another short while, and see folks online!
23:46:17 <nickm> Tomorrow is a US holiday and Friday I'm not planning to be around, but I will be checking email, and I'll be glad to drop by if I can help anybody with anything
23:46:18 <ahf> see you all after i've had some sleep o/
23:46:24 <nickm> peace all!
23:46:27 <ahf> o/
23:46:28 <nickm> sleep well ahf
23:46:31 <nickm> #endmeeting