18:59:44 <marga> #startmeeting
18:59:44 <MeetBot> Meeting started Wed Jun 19 18:59:44 2019 UTC.  The chair is marga. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:59:44 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:59:51 <marga> #topic Roll Call
19:00:01 <marga> Who's here for this month's meeting?
19:00:06 <marga> Margarita Manterola
19:00:07 <fil> Philip Hands
19:00:12 <ntyni> Niko Tyni
19:00:12 <bremner> sorta here. on s bahn back to htel
19:01:11 <marga> Ok, I guess that's all of us...
19:01:12 <marga> #topic Review of previous meetings AIS
19:01:29 <marga> So, we closed one bug and the other one smcv sent the update, but didn't close the bug.
19:02:24 <marga> I think it wasn't intentional, though (i.e. everything that's needed is there, only the closing is missing as far as I'm concerned)
19:02:43 <marga> Let's switch into that bug and see if we agree on that :)
19:02:50 <marga> #topic #923450 Requirements for being pre-dependency of bin:init
19:03:16 <marga> I think the only action left is to actually close the bug. Am I missing something or can we just close it?
19:03:55 <fil> I think that's right
19:04:52 <ntyni> me too; might be worth checking with smcv if he had something else in mind though
19:05:18 <marga> Alright, then we can close the bug. I'll take the AI to ping smcv and check whether he had something else in mind.
19:05:35 <marga> #action marga to check with smcv if this can be closed already.
19:05:49 <marga> Ok, so now for the fun topic of the day...
19:05:51 <marga> #topic Sam's thread about "Delegating decisions about Policy"
19:06:05 <bremner> yeah it looked OK toe
19:06:26 <marga> So, we had this thread where Sam asked about future decisions about policy and whether the TC would be the right body to do that and more fun stuff...
19:06:56 <marga> Nobody in the TC replied to the thread. Sean and Ian had a nice conversation, up until the point where Ian started saying that the TC is flawed by (his) design...
19:07:19 <marga> So, I'm wondering what the others think about this thread, whether we should respond something, etc
19:08:55 <marga> I think the question boils down to whether the TC is the authority on technical policy, for things that were not elevated to us through the usual process, but rather to establish new policies for the project.
19:11:04 <ntyni> "The Technical Committee does not engage in design of new proposals and policies."
19:11:17 <marga> Indeed, but Sam's question is not about design
19:11:25 <marga> Rather about establishing policy
19:11:46 <marga> Say that we wanted the whole project to adopt some new standard or something like that
19:12:40 <fil> My gut reaction is that the TC isn't really for this.  I'm happy with the TC being responsible for the things that people cannot sort out amongst themselves, but not if we're expected to poke our nose into some specialist area where everyone is cheerfully working together.
19:12:43 <ansgar> marga: Isn't that "design of new policies"?
19:13:36 <marga> ansgar, I guess it's a bit of a gray area.  If the standard already exists and we say that that's the standard that we should adopt, are we actually designing a new policy or not?
19:13:37 <fil> of course, project wide change is another thing, but I don't really see that we're somehow elevated above others in the project when it comes to deciding that sort of thing
19:13:43 <marga> But, yeah, I kinda feel that this is not the right role for the TC, and personally I think the right way to go is to let the policy maintainers feel empowered to take these decisions and then involve us only when they are not sure.
19:14:38 <bremner> there is a bigger question about policy leading practice imho
19:16:25 <ansgar> marga: And then there is also "Technical Committee makes decisions only as last resort"; (one could change that of course if really wanted)
19:17:58 <marga> Right, that's true
19:18:32 <marga> I'm not sure how that would combine with a separate delegation, which I think was what Sam had in mind
19:18:47 <marga> But, I think we are basically in agreement that the role proposed by Sam doesn't really fit the TC
19:19:09 <bremner> I think tc might have a role if policy editors don't see cocncencus
19:19:16 <fil> bremner: quite -- policy strikes me as being akin to English dictionaries (following common usage, rather than trying to set it)
19:19:25 <ansgar> bremner: Or policy documenting things that never were practice (doc-base for example; or other things)
19:20:58 <ntyni> yeah missing consensus fits in with the "last resort" thing
19:21:41 <marga> Ok, it seems that we are basically in agreement about this: TC can be consulted when policy editors don't see consensus, but we feel it's the role of policy editors to decide on what to do about policy.
19:21:48 <marga> Does that represent the general agreement correctly?
19:22:07 <ntyni> works for me
19:22:20 <bremner> OK for me
19:23:01 <marga> Ok, shall I reply to Sam, or does someone feel strongly inclined to take this?
19:23:30 <fil> something like that -- although I'd expect the best policy to come out of some grouping of interested developers pushing changes at the policy editors, but that's a detail
19:24:04 <ansgar> So, what about issues that won't see consensus?  Say doc-base...
19:24:53 <marga> I think that in that case it makes sense to ask the TC to weigh in.  But we'd only do it after being asked, i.e. "last resort"
19:24:58 <ansgar> I proposed to remove it as I think it never gained much use; mostly to get rid of unused stuff in policy.
19:25:21 <ansgar> But I don't feel like investing much energy it getting it removed (I just ignore it like everyone else)
19:26:02 <bremner> well, probably the TC is also not for things you don't want to invest much energy in ;)
19:26:20 <fil> ansgar: sounds like something to be added to the papercuts list?
19:28:04 <fil> I'd imagine that most people would agree to just drop it at this point, since as you say, it mostly gets ignored
19:28:21 <marga> Yeah, where's the lack of consensus there?
19:30:13 <ansgar> marga: Well, someone wanted to push for doc-base adoption instead.  So Policy is at a standstill.
19:30:47 <ntyni> this is #910783 fwiw
19:31:09 <marga> Ok. So, I think this would be something that the TC could consider, but it would require someone bringing this up
19:31:41 <ansgar> If you want something with more disagreement, there is also #911165 ;-)
19:31:54 <marga> One thing that worried me from one of Ian's messages was that he brought up the fact of how much friction bringing things up to the TC means.
19:32:17 <marga> I'd like for people to be able to bring things up to the TC without it becoming a flamewar, but I don't know if that's actually possible.
19:33:17 <fil> we do _almost_ have a rule that it has to be a flamewar before you bring it to the TC though ;-)
19:34:00 <marga> Well, that's not really it, and maybe there's something for us to change there?  We want people to first try to discuss and agree... Not go through a flamewar
19:34:27 <ansgar> I probably would bring it to the TC if someone asked for a package to be removed because it has no sysvinit script.
19:34:56 <ansgar> But just policy having that requirement and reality disagreeing is not that large a problem ;-)
19:35:01 * gwolf arrives, panting...
19:35:22 <marga> ok
19:35:22 <bremner> marga: we could consider some concensus building exercises
19:35:33 <marga> bremner, what do you mean?
19:35:41 <ansgar> So I guess one could ask if the TC wanted to be proactive about such standstills in policy.
19:35:57 <bremner> marga: well, basically what sam just did for dh.
19:36:21 <bremner> ansgar: I think you should talk to the policy editors about standstills in policy
19:37:05 <marga> yup
19:37:27 <ansgar> bremner: Well, I think there is no consensus... So not sure what policy editors should do?
19:37:38 <bremner> ansgar: it's their process to manage.
19:37:39 <fil> bremner: I think that's a good idea, but on the other hand I don't think that we have any monopoly on that sort of thing, so anyone could do such an effort
19:38:13 <bremner> fil: I was think as alternative to just closing bugs because people haven't tried to build concensus
19:38:20 <bremner> I'm not sure I personally have the tuits
19:38:29 <marga> Ah, I see what you mean now
19:38:53 <marga> Like instead of saying: "go discuss and come back when you haven't been able to agree", to work with them to find consensus?
19:38:59 <gwolf> FWIW I agree with the points I have read, and particularly with marga's conclusions...
19:39:00 <bremner> yeah, something like that
19:39:18 <marga> Yeah, I think that would be valuable, but we'd need to know how to do it :)
19:39:25 <fil> Sam does seem to have done a particularly good job on that one -- I wonder if it would have gone as well if he were not DPL, and what we could do to support non-DPLs in making similar efforts
19:39:28 <bremner> marga: I think moderating discussion doesn't count as detailed design
19:39:41 <bremner> ymmv
19:40:15 <marga> bremner, I agree with that assessment, I just think it's a different skillset than the one usually requested for someone joining the TC
19:40:30 <gwolf> I believe that one of the criteria and reasons for selecting people to TC is to try to compose the TC of non-flamewarists
19:40:53 <bremner> well, it's mainly a process thing, and sam has just had practice. He'd probably help with getting started
19:41:25 <marga> Yeah, that might help
19:42:01 <gwolf> marga: I think the way you frame it ("come back when...") would be just the wrong way to put it :) Maybe we should try to steer things into agreeing instead of requiring proof of flame
19:42:32 * gwolf supports bremner "moderating discussion is not detailed design"
19:42:33 <fil> BTW I think it would have gone pretty-much as well had he not been DPL, in case that wasn't clear
19:43:07 <marga> Yeah, but maybe he wouldn't have been motivated to be involved.
19:43:30 <marga> Anyway, I think we sort of have consensus here, and I'll reply to Sam's email when I don't have a headache like I do right now :-/
19:43:37 <marga> Let's move to AOB
19:43:42 <marga> #topic Any other business?
19:45:48 <marga> I guess not, then :)
19:45:50 <gwolf> Shall we prepare something for our DebConf talk?
19:45:58 <marga> What do you have in mind?
19:46:07 <gwolf> I guess we can basically rehash what we used in previous confs
19:46:14 * fil won't be there I'm afraid
19:46:23 <gwolf> (that's something we can sort out face to face, so no reason in having it in-meeting)
19:46:32 * ntyni neither
19:46:33 <gwolf> I just wanted to say something meeting-worthy ;-)
19:46:37 <marga> I think it would be nice to do something different, btw
19:46:42 <fil> lol
19:46:43 <marga> But we should probably discuss in person
19:46:57 <gwolf> right - but we should start thinking what we want to do
19:47:12 <marga> Agreed
19:47:15 <marga> I'll think about
19:47:25 <marga> #endmeeting