16:59:14 <gaba> #startmeeting Network - June 29th
16:59:14 <MeetBot> Meeting started Mon Jun 29 16:59:14 2020 UTC.  The chair is gaba. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:59:14 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:59:23 <gaba> that pad is at http://kfahv6wfkbezjyg4r6mlhpmieydbebr5vkok5r34ya464gqz6c44bnyd.onion/p/tor-netteam-2020.1-keep
17:00:27 <ahf> o/
17:00:56 <jnewsome> o/
17:01:00 <nickm> hi all
17:01:48 <gaba> dgoulet, asn: we have links in line 40 to reviews, can we update that with the ones from gitlab?
17:02:10 * dgoulet looks
17:02:39 <dgoulet> gaba: ok we will! asn and I have to discuss things asap about this since last week we changed workflow there
17:02:44 <nickm> asn, dgoulet: did you have time to do review assignment today?
17:02:47 <gaba> OK
17:03:02 <nickm> I see tor!9 and tor!10.
17:03:28 <nickm> those are the only merge requests in tor
17:03:32 <nickm> and in core
17:03:44 <asn> i did not do review assignment yet
17:03:45 <nickm> dgoulet is reviewer on tor!9
17:04:05 <nickm> that leaves only tor!10 to be assigned; it's a fix for tor#32622
17:04:07 <asn> yeah the MR workflow still needs to be figured out
17:04:39 <nickm> that one touches tls error reporting to the controller, but is fairly clean. anybody willing to review?
17:04:43 <asn> dgoulet: perhaps we can do that early tomorrow?
17:04:52 <dgoulet> asn: whenever
17:04:54 <asn> ye i can take it
17:05:09 <asn> dgoulet: ye let's do it tomorrow
17:05:29 <gaba> ok. can we start looking at the roadmap? :)
17:05:30 <gaba> https://gitlab.torproject.org/groups/tpo/core/-/boards
17:05:49 <gaba> does it reflect what you are working on?
17:06:02 * asn looks
17:06:13 <dgoulet> that is rather large list..
17:06:17 * dgoulet filters on tor.git
17:06:42 <nickm> yeah i thing so
17:06:46 * nickm filters on nickm
17:06:54 <asn> jeez
17:06:56 <ahf> hm, yeah, #5304 is on my table now, but i think the 044 thing have higher priority right now for network team stuff, no?
17:06:58 <asn> im nowhere in there
17:07:01 <asn> that's a rich kanban
17:07:02 <nickm> dgoulet: I assigned you to a huge bunch of prop#312 stuff on friday, but didn't change your next/doing
17:07:16 <dgoulet> nickm: that is good, I do that as I go
17:07:33 <dgoulet> nickm: my "Doing" tickets are always changing so this is good
17:08:25 <dgoulet> gaba: is "Next" something you use?
17:08:44 <dgoulet> gaba: as in you would benefit from us setting those up?
17:08:51 <gaba> yes
17:08:58 <asn> hmm i need to work on that kanban to add my stuff
17:09:03 <dgoulet> gaba: ack
17:09:03 <gaba> I would like at some point sort it out as the next thing to work in the net week
17:09:06 <gaba> next*
17:09:08 <gaba> so far not yet
17:09:18 <asn> open has 2000+ tickets
17:09:29 <gaba> asn: yes, I need to sort them out
17:10:01 <dgoulet> asn: filter it by Assigne=asn if you want less annoying pile of tickets
17:10:55 <nickm> I suggest sorting by milestone too
17:11:00 <nickm> up to you
17:11:27 <gaba> by name is good as you can see if there is somthing you are not working on or planning to
17:13:34 <dgoulet> gaba: what is the difference between Backlog and Next then?
17:13:47 <dgoulet> gaba: as in I have items in the "Backlog" that are assigned to me ... so isn,t "Next" ?
17:13:48 <nickm> I think "backlog" comes after "next"
17:14:02 <gaba> right, stuff in the backlog should not be assigned to you
17:14:11 <dgoulet> oh
17:14:13 <nickm> oh
17:14:18 <dgoulet> so all my backlog at the moment should be Next then? :)
17:14:24 <gaba> yes
17:14:31 <dgoulet> interesting ok
17:14:31 <nickm> gaba: can we reconsider that?
17:14:39 <gaba> nickm: how so ?
17:14:48 <nickm> gaba: I think it's useful to have a difference between what I do next, and what I do later.
17:15:05 <gaba> yes, next is what you do next
17:15:09 <gaba> backlog what you do later
17:15:14 <dgoulet> but unassigned?
17:15:20 <gaba> the backlog issues could be taken by other people
17:15:29 <nickm> I think it's okay to have assigned stuff in backlog though?
17:15:38 <ahf> can't we have assigned things in backlog?
17:15:39 <ahf> ye
17:15:44 <nickm> like, can we say "unassigned stuff in backlog is fine to take; for assigned stuff, ask first?"
17:15:50 <gaba> sure
17:15:55 <ahf> i have some PT stuff that i think is my domain, but it might not be the Next thing
17:16:01 <gaba> ok
17:16:07 <gaba> that change sounds ok
17:16:21 <dgoulet> oki
17:16:29 <nickm> great
17:17:14 <dgoulet> and Icebox is unassigned stuff?
17:17:18 <dgoulet> or must be assigned?
17:17:35 <gaba> totally unassigned, we have no idea when we will do it
17:17:44 <dgoulet> ack
17:18:01 <gaba> next in the agenda we have reviews. You all talked about it, right? Anything that should be moved?
17:19:34 <dgoulet> not at the moment I would say
17:19:37 <asn> +1
17:19:53 * asn loves the kanban now
17:19:56 <asn> but needs some care and love
17:20:04 <nickm> I need tor!10 to be assigned
17:20:12 <asn> ok
17:20:15 <nickm> dgoulet: I think you are reviewing tor!9
17:20:27 <nickm> I need a reviewer to ack #33346.
17:20:44 <asn> what is the link to !0?
17:20:45 <asn> what is the link to !10?
17:21:07 <dgoulet> asn and I will allocate those yeah
17:21:25 <asn> nickm: u need second reviewer in #33346?
17:21:55 * asn kinda lost in gitlab right now
17:22:34 <nickm> asn: https://gitlab.torproject.org/tpo/core/tor/-/merge_requests/10
17:22:48 <nickm> asn: i'm not reviewer on #33346, i wrote the patch
17:23:33 <gaba> we are not assigning any label to MRs ? Would it work if they also get into the backlog/next/doing stacks?
17:23:35 <dgoulet> asn: we just need to sync up here for all this
17:24:07 <nickm> so, MRs are all in theory "needs review" or "merge ready" or "backport"
17:24:10 <nickm> they can't have other states
17:24:18 <asn> dgoulet: ye let's do it tomorrow
17:24:22 <asn> dgoulet: it's too late today for me
17:24:26 <asn> after this mtng i mean
17:24:46 <asn> nickm: ack ack
17:24:51 <nickm> ok
17:24:57 <nickm> we ready to talk about 044?
17:24:59 <asn> nickm: noted all this stuff. will do it tomorrow.
17:25:11 <nickm> I've added a bunch of links that I hope will work for 044.
17:25:21 <nickm> first thing to do is note the tickets assigned to you in the 044 milestone
17:25:26 <gaba> ok. About 044 I see that those tickets are not in the kanban board. Again, would it be usefull to include them? All the ones assigned to you in next or doing.
17:26:02 <nickm> I have mine in next or doing.  It would be a good idea for everybody to put their 044 tickets in a kanban column.
17:26:05 <dgoulet> an 044 kanban would be good yeah ...
17:26:17 <gaba> yes, and that way you can sort out your own tickets
17:26:21 <nickm> dgoulet: you can make one by filtering the regular kanban on milestone
17:26:21 <gaba> and have everything in one place
17:26:24 <dgoulet> I do
17:26:32 <dgoulet> but I mean a board pre-set in core/tor
17:26:39 <nickm> I'll move the other tickets to backlog in 044.
17:26:42 <nickm> dgoulet: interesting
17:27:10 <dgoulet> so I would like to revisit what we talked a while back that tickets in the 044-final milestone are all MUST
17:27:47 <nickm> dgoulet: you think we should do that, or you think we shouldn't do that?
17:28:10 <dgoulet> I think we should do that. 52 tickets remain in 044-final at the moment
17:28:23 <dgoulet> are they all "must" ?
17:28:28 <ahf> hm, i need to clean up some of mine, they are not all must
17:28:37 <gaba> are all the unassigned a must?
17:28:43 <nickm> no
17:28:51 <dgoulet> thing is that once we clean all must, then we can attack "should" but not before
17:28:58 <gaba> right
17:29:03 <nickm> can we use sort order for this?
17:29:15 <nickm> I think it's okay to use the milestone for can/should.
17:29:24 <dgoulet> but now I don't have that vision for all "must" except "044-must" which I don't think is that of a good label to have version specific label
17:29:48 <dgoulet> nickm: but then isn't give us a hard time to see completion properly?
17:29:52 <nickm> I've put the must above should and the should above can.
17:30:40 <nickm> when I move tickets vertically, does that affect the board as other people see it?
17:31:03 <gaba> yes
17:31:13 <dgoulet> tbh, -could and -can ... I actually never used that, I do filter on -must
17:31:27 <dgoulet> and so if -can and -could are used, I'm curious how people here use them?
17:31:29 <gaba> we could leave the can tickets in the icebox... and the should in the next or backlog
17:32:02 <dgoulet> gaba: and all of them in the milestone?
17:32:18 <nickm> "can for the icebox" is a neat idea
17:32:24 <nickm> I will move them now?
17:32:27 <gaba> yes
17:32:52 <dgoulet> ok what is happening?
17:33:42 <nickm> great, done
17:33:54 <nickm> I have put all the 044-can tickets in Icebox
17:33:57 <ahf> nice
17:34:11 <nickm> I have put all the remaining unassigned 044 tickets in "Backlog"
17:34:45 <dgoulet> ok but that does _not_ address/answer my initial question about the milestone
17:35:04 <nickm> the milestone as a whole is still for "can", "must", and "should".
17:35:19 <dgoulet> so we never reach 100% completion? ...
17:35:22 <dgoulet> not sure that makes any sense?
17:35:32 <gaba> yes, right
17:35:43 <gaba> it would feel weird not being 100% ever
17:35:45 <ahf> we move them out and into 0.4.5 when we close the 044 one?
17:35:47 <nickm> when we freeze, we dump stuff out of the milestone.
17:35:49 <gaba> do we have many cans?
17:35:51 <ahf> yeah
17:35:57 <gaba> ok
17:36:02 <nickm> we have 14 things in icebox now
17:36:17 <dgoulet> -freeze is easy, it has its milestone...
17:36:19 <nickm> including some that are assigned
17:36:29 <dgoulet> but -final needs a "final" due date
17:36:38 <nickm> as we move forward, we can defer things or do them
17:36:43 <dgoulet> and if we can't see completion for that due date, we constantly triage?
17:37:12 <dgoulet> not sure I agree that milestones are a dumping ground that constantly get triaged tbh...
17:37:20 <dgoulet> but I won't hold up this discussion further...
17:37:33 <nickm> let's try with this for a bit and see if it works better?
17:37:49 <dgoulet> well what we are doing is _exactly_ like we did with Trac? :)
17:37:59 <nickm> except with icebox for can?
17:38:00 <dgoulet> and we were always complaining that too many tickets in the milsetone
17:38:05 <gaba> i kind of agree with dgoulet on it is better to have a clear milestone
17:38:07 <gaba> but we can check later
17:38:09 <nickm> so we can see the backlog for should/must that's unassigned
17:38:31 <nickm> also, ahf, dgoulet: there are some tickets assigned to you in 0.4.4.x that are should or must but are not in backlog, next, or doing
17:38:34 <dgoulet> but should/can are really used by anyone here?
17:38:38 <nickm> if you could move them to the right column that would be cool
17:38:43 <dgoulet> we barely finish the -must usually in a milestone
17:38:44 <nickm> dgoulet: I use them, yeah
17:38:47 <dgoulet> ok
17:38:52 <gaba> yes, im curious how many can/should gets done for a specific release
17:38:55 <nickm> should-fix means that we're being bad programmers if we release without it
17:39:01 <nickm> must-fix is a release blocker
17:39:09 <nickm> and can is a nice-to-have
17:39:15 <asn> "bad programmers" more like "bad firefighters"
17:39:16 <gaba> should those be the same? must and should
17:39:19 <dgoulet> I'm aware but we shove them all in the one milestone?
17:39:23 <nickm> not really
17:39:24 <dgoulet> anyway
17:39:29 <nickm> that is,
17:39:40 <nickm> must and should are different because we can release without doing all the shoulds.
17:39:52 <nickm> like, for example at random, #30971
17:39:56 <ahf> must is release critical
17:40:00 <nickm> we should rebuild the fallback directory list
17:40:01 <gaba> does this mean that sometimes you are all bad programmers? :P
17:40:03 <nickm> yes
17:40:15 <nickm> but we can release without it if we must
17:40:27 <nickm> contrast with #33503 -- that's a must-fix memory leak
17:40:37 <gaba> I think we can move forward this way now and revisit later
17:40:48 <dgoulet> anything decided now is the future of our workflow
17:40:59 <dgoulet> because we'll still go on using -must, -can, -should labels then?
17:41:10 <nickm> I suggest not adding any can/should/must for 045
17:41:17 <nickm> let's see how this works for us this time
17:41:22 <dgoulet> ok
17:41:30 <ahf> +1 on experimenting right now with this and figure out what works
17:41:31 <asn> ack
17:41:38 <dgoulet> also 045 has *two* milestones so please keep that in mind
17:41:41 <dgoulet> -freeze and -final
17:41:41 <nickm> dgoulet/ahf: did you see my note about moving your 044 stuff out of the "Open" column?
17:41:51 <dgoulet> no
17:42:05 <nickm> ok. https://gitlab.torproject.org/groups/tpo/core/-/boards?scope=all&utf8=%E2%9C%93&state=opened&milestone_title=Tor%3A%200.4.4.x-final
17:42:14 <nickm> that's the filtered board I'm looking at now for 044
17:42:33 <nickm> the can stuff is in icebox, the  unassigned should/must stuff is in backlog
17:42:39 <nickm> there are also assigned things in open.
17:42:49 <nickm> if you can move them into doing/next/backlog, i think that will help?
17:43:27 <nickm> dgoulet/ahf: also you have a couple of things in Icebox there.  Maybe double check that you are okay with them being in icebox
17:43:44 <ahf> yep, i am gonna go over all of mine tomorrow
17:43:47 <nickm> great
17:45:29 <gaba> ok. Are we done with 044 ?
17:45:42 <nickm> i think so!
17:46:02 <gaba> Not sure if we can remove that announcement. It was from last week
17:46:24 <nickm> i'll take off the review i did...
17:46:40 <nickm> all those reviews are done :)
17:46:47 <gaba> :)
17:46:48 <gaba> ok
17:46:58 <gaba> A reminder to look at the ticket that Antonela is asking for feedback with
17:47:04 <gaba> She is working on dev.torproject.org
17:47:10 <gaba> and needs some feedback there
17:47:38 <nickm> (it's fun stuff and looks good to read)
17:47:51 <gaba> The board of s55 I'm guessing is up to date now after dgoulet and nickm looked at their board.
17:48:21 <dgoulet> gaba: I moved all my things into Next from the Backlog :)
17:48:25 <dgoulet> so I believe it is accurate on myside
17:48:34 <gaba> great
17:49:26 <gaba> It seems nobody is asking for help. Do we have anything else for this meeting?
17:49:32 * ahf is good
17:49:39 * dgoulet is good
17:49:40 * asn good
17:50:03 <gaba> ok. let's close the meeting then
17:50:07 <gaba> #endmeeting