path: root/meeting-logs/20100614.txt

21:01:07 <Betelgeuse> let's start
21:01:10 <dertobi123> so, let's go
21:01:13 <dertobi123> who's around?
21:01:25 <Calchan> I am
21:01:35 -*- dertobi123 is, obviously
21:01:47 -*- wired here, proxying for scarabeus 
21:02:15 <dertobi123> wired: funny, scarabeus didn't sent an email about you being his proxy for today
21:02:30 <wired> he said so here a while back
21:03:40 <Betelgeuse> 15:19  * scarabeus not sure if i will make it here, asked wired to proxy me
21:03:44 <wired> [18:19:47] * scarabeus not sure if i will make it here, asked wired to proxy me
21:04:29 <dertobi123> whatever, it doesn't matter for todays non-meeting
21:05:44 <ferringb> so... no one assembling an agenda/request for err.. requests, means nothing happens, or...
21:06:15 <Calchan> I spent a lot of time and effort last month putting together an agenda and a meeting that some of you said was a waste of your time
21:06:15 <Betelgeuse> ferringb: we can still vote on issues but members might be vary on voting on something they know little about
21:06:23 <Calchan> so I certainly wasn't going to waste any more of your time again this month
21:06:45 <Calchan> and it demonstrates yet again that if I don't do it then nothing happens
21:07:09 -*- dertobi123 sighs
21:07:14 <ferringb> Calchan: don't suppose you've been maintaining a list of things that didn't make it into a specific council meeting?
21:07:15 <Betelgeuse> so missing leio, ulm ?
21:07:18 <ferringb> aka, a backlog?
21:07:29 <leio> I'm here
21:09:23 <Betelgeuse> was this our last meeting?
21:10:06 <dertobi123> this is our last one, yes
21:10:51 <Betelgeuse> Not having an agenda is quite fitting for the year.
21:11:12 -*- ferringb sighs
21:11:20 <Betelgeuse> Hopefully the webapp helps the next council
21:11:20 <tove> you can meet again in two weeks
21:11:24 <ferringb> yep
21:11:27 <Betelgeuse> tove: true
21:11:30 -*- ferringb notes there are two things still outstanding
21:11:43 <ferringb> one is Arfrever's request, although I've indicated my views why it's not ready
21:12:05 <ferringb> there is also REQUIRED_USE that's been sitting for a few months now
21:12:35 <dertobi123> well, none of these items has been submitted to the council
21:12:47 <Betelgeuse> dertobi123: that latter has
21:12:49 <ferringb> dertobi123: check your logs.  REQUIRED_USE sure as hell was.
21:13:01 -*- ferringb notes that discussing something shouldn't require red tape either
21:13:11 <Betelgeuse> It's been our failure that it wasn't in the last meeting
21:13:28 <dertobi123> ferringb: sorry
21:13:44 <Betelgeuse> For Arfrever's request I propose we vote on whether it requires more work:
21:13:47 <Betelgeuse> I vote yes
21:13:54 <dertobi123> yes, too ... (of course)
21:14:14 <leio> uh, abstain
21:14:29 <Betelgeuse> leio: ?
21:14:36 <ferringb> spoke my views.  not opposed to the change (there are other USE syntax changes I'd like for example), but this needs to be done in a way that doesn't cause PMS compliant managers to go boom.  Needs work.
21:15:01 <leio> from voting on that, I don't know if it requires more work or not.
21:15:01 <Betelgeuse> leio: did you read the logs little before the meeting?
21:15:21 <leio> I think so
21:16:21 <Betelgeuse> wired, Calchan ?
21:18:42 <ferringb> additional item that likely is worth discussing; since this is the last meeting, a post mortem of the last year is likely wise, in terms of what worked, what didn't.
21:19:22 <leio> everyone expecting someone else from the council to do certain council work tasks does not work
21:20:07 <wired> abstain, but from a quick read of the bug this does seem to need more work
21:21:30 <Betelgeuse> ferringb: reading your GLEP again what comes to mind is it shouldn't it reserve a cache key?
21:21:47 <ferringb> should, yes
21:22:00 <ferringb> s/should/must/ realistically, from a performance standpoint
21:22:17 <ferringb> http://dev.gentoo.org/~ferringb/required-use.html <-- for people who've not read it.  think we've demonstrated we have time, so go read it if you've not
21:23:32 <ferringb> Betelgeuse: good question though, hadn't thought to clarify in there.  currently there is still space left in the flat_list format, so that's not an issue to expand into one of the slots
21:24:06 <Betelgeuse> ferringb: yes but ideally the GLEP would specify the slot number
21:24:24 <ferringb> Betelgeuse: ideally PMS would, but yes
21:25:06 <ferringb> Betelgeuse: to nail down the exact slot for it requires rechecking what portage jammed in there.  PROPERTIES iirc got slapped in there (without PMS updating afaik), so the enumeration in the spec needs updating either way
21:25:30 <ferringb> tbh, I consider that something of an implementation detail, but yes, for the resultant PMS patch the slotting should be explicitly locked.
21:27:50 <ferringb> leio: wired: comments?
21:27:56 <ferringb> dertobi123: Calchan: y'all?
21:28:19 <leio> ferringb: nope, but I'll take a portage patch :)
21:28:32 <ferringb> read the glep.  a patch already exists.
21:28:47 <leio> stop referring to it as a glep :)
21:28:49 <ferringb> under the implementation section specifically, sebastian luther implemented it
21:29:01 <ferringb> s/glep/text/ whatever.  point is, the doc covers this already
21:29:40 <dertobi123> ferringb: no comments, looks good to me.
21:29:49 <dertobi123> though ^^ looks rather funny
21:29:58 <dertobi123> but that's just a detail
21:30:04 <ferringb> ^.^ was my first preference.
21:30:08 <dertobi123> heh
21:30:11 <wired> ferringb: id like to see that in portage asap :)
21:30:19 <ferringb> switching it to voting specifically
21:30:28 <leio> while we don't have this, pkg_pretend had issues doing this in a temporary fashion?
21:31:12 <Betelgeuse> leio: see motivation?
21:31:19 <ferringb> already covered.
21:32:12 <ferringb> expounding a bit... there is no 'temporary' in EAPI.  once it's there, it's there till we start stating xyz EAPI's are no longer supported (I don't expect that for at least a half decade either, if that).  once it's in, it screws up doing this properly long term
21:32:17 <leio> yeah, wish Calchan made an agenda as volunteered in lack of others :)
21:33:28 <Betelgeuse> ferringb: well that's not obsoleting pkg_pretend
21:33:29 <leio> so not optimal, but temporarily possible to achieve something better than current with pkg_pretend too
21:34:13 <ferringb> leio: no... pkg_pretend is added in eapi4.  this doc modifies eapi4 (which isn't released so we can pull this) to fix a flaw in using pkg_pretend for use validation like this.
21:34:21 <leio> I'm hoping for some markup to express that a USE flag takes effect only in case some other one is enabled
21:34:37 <ferringb> no REQUIRED_USE, you have to use pkg_pretend, and you suffer the issues laid out in motiviation.  this is why it needs fixing before eapi4 goes live.
21:34:44 <leio> would this fit in the syntax style somewhere too? :P
21:35:43 <ferringb> currently it's limited to just this new metadata key.  if someone wants to use ^^ elsewhere, they need to put forth a proposal (solar has some notions, but nothing concrete last I knew)
21:36:16 <leio> same syntax like that but in IUSE could be expressing what I have in mind?
21:36:51 <ferringb> no
21:37:27 <leio> anyway, lets discuss this additional feature later in private perhaps.
21:37:27 <ferringb> IUSE is an enumeration.  you'd have to combine effectively a tree syntax in w/ an enumeration... it's not pretty to even try it (have, hence the seperate key)
21:38:04 <leio> just hoping that once all these new syntax features get added, that they all fit in consistently
21:38:19 <ferringb> wired: dertobi123: fine with this, and moving on to post mortem?
21:38:33 <ferringb> Calchan: your thoughts?  you've been fairly quiet
21:40:05 <wired> ferringb: yeah
21:40:34 -*- dertobi123 nods
21:41:16 <ferringb> good enough.  at this point since there is logging, might as well start laying out the perceived/outstanding issues w/ the council over the last year... what went wrong, what can be improved, what went right, etc.
21:41:35 <ferringb> standard post mortem (if you've not participated in one of these just pm and I'll go through details)
21:41:40 <ferringb> sound good?
21:42:43 <Betelgeuse> yes
21:44:00 <Betelgeuse> I think this council suffered from the usual lack of "drive"
21:44:20 <Betelgeuse> People do show up but effective work would require plenty of hours outside meetings
21:44:21 <ferringb> Betelgeuse: lack of drive, or burnout?
21:45:21 <Betelgeuse> It tells quite a lot that our last to summaries are "To be completed"
21:45:26 <ferringb> yep
21:45:28 <Betelgeuse> s/to/two/
21:45:35 <dertobi123> well, from my pov ... this council was way too passive. we merely just discussed some eapi/pms stuff and became tired while doing so. personally i somehow lost interessed in gentoo and several other things became much more important for me. i should've stepped back from the council half a year ago or so - but seeing who'd take my seat i decided to fullfill my term.
21:45:49 <ferringb> Betelgeuse: think making the council's term 6 months would be good?
21:46:13 <ferringb> dertobi123: your thoughts on 6 months?
21:46:28 <wired> id like to see the council be more decisive
21:46:59 <Betelgeuse> ferringb: if there's a big shift in membership it might be hard to get things done
21:47:08 -*- ferringb notes the purpose of a post mortem, while including wishes/desires for going forward, should also be looking at specific failures in the past to identify/ensure they don't occur again
21:47:12 <dertobi123> ferringb: ambivalent ... might help, but it might be useful to shorten nomination and election periods then, too
21:47:18 <ferringb> Betelgeuse: things aren't really getting done however.
21:47:19 <dertobi123> we should give it a try
21:47:32 <ferringb> Betelgeuse: faster cycle at least means that if people want out, they can bow out quicker.
21:47:47 <Betelgeuse> ferringb: I would hope if people loose interest that they drop out voluntarily
21:48:07 <ferringb> Betelgeuse: not guranteed.  as mentioned by dertobi123, knowing the competition, sometimes they hold on.
21:48:08 <Betelgeuse> Current rules still state that the council has to approve the next one in line
21:48:42 <ferringb> dertobi123: with respect mind you, but my stance is if there isn't interest, then one needs to let the next dude in line handle it... even if they're a schmuck (someone has to do the work one way or another)
21:49:03 <ferringb> Betelgeuse: re: disruption, staggered seat elections comes to mind btw to combat that.
21:49:24 <Betelgeuse> ferringb: we do keep those if council doesn't approve and it has already happened
21:49:32 <Betelgeuse> happened this term already I think
21:50:06 -*- ferringb means that 2/7 memebers have their terms for feb->feb, 2/7 have it for sept->sept, and the 7th lands wherever (under a year term)
21:50:11 <ferringb> *members, pardon.
21:50:28 <ferringb> either way... other failings/issues from the past year?
21:50:55 <ferringb> wired: "be more decisive" is a bit vague.  specifics/examples would be useful here- post mortem should be specific failings rather than "I'd like us to be more xyz" if at all possible
21:52:32 <dertobi123> ferringb: there was interest, i just didn't worked an hour or two a day on gentoo - just like i did a year or two ago. that's what i tried to express with "lost interest".
21:53:08 <ferringb> dertobi123: fair enough.  as said, wasn't trying to pick at you specifically, more looking at the general problem that can arise there
21:54:40 <ferringb> dertobi123: any thoughts on how to help w/ that?  realistically the council does have a semi-hefty req on one's time
21:54:52 <wired> ferringb: a few months back i proxied for leio at a meeting where nothing was decided, no-one had read the agenta items enough and everything was postponed. i don't blame people for not having enough time to spend for the council, but in my eyes the council should be a swift body, quickly resolving things devs can't
21:55:09 <ferringb> similar views for myself
21:55:34 <ferringb> so why is it slow?  ignoring people not reading shit (can't do anything about that aside from come election time)
21:55:54 <dertobi123> well, i was prepared (i.e. reading the agenda and related topics/gleps) every meeting i attended
21:56:05 <dertobi123> but that doesn't matter at all now
21:56:22 <ferringb> "doesn't matter at all now" howso?
21:56:58 <wired> im not pointing fingers here and my experience doesn't mean that you hadn't done your studing, but the result was the same, a month went by without anything getting done
21:57:32 -*- ferringb notes the point of a post mortem here is to not care about who did what.  it's to ensure that next time around, these issues don't re-occur
21:58:08 <dertobi123> ferringb: as in "it doesn't matter if i was prepared, if in the end the result was to just postpone things"
21:58:14 <ferringb> ah, right
21:58:32 <ferringb> dertobi123: any thoughts on how to keep the council body from continually postponing things?
21:58:48 <wired> i also feel that the council is kind-of afraid to take big decisions
21:58:51 <Betelgeuse> ferringb: you need a member actively pushing things to avote
21:58:53 <NeddySeagoon> council meetings should mostly be a dog and pony show - making decisions public and explaining the reasoning. There should be very little discussion.  Its an opportunity for the council to be held to account by the voters
21:59:25 <Betelgeuse> ferringb: I get annoyed many times when actual votes don't happen in meetings
21:59:37 <Betelgeuse> ferringb: We could try a no major discussion rule
21:59:48 <Betelgeuse> ferringb: Get together every two weeks
21:59:49 <NeddySeagoon> Betelgeuse, votes can happen before meetings
21:59:56 <wired> i'd like to see members just vote for things as they see fit, instead of postponing.
21:59:56 <Betelgeuse> discuss and vote separately
22:00:04 <dertobi123> ferringb: there two importants points ... having a secretary tracking things and looking for things to get done. the other one is people who submitted things to actively ask what the status is and remember to add things to the agenda if something got lost
22:00:07 <ferringb> Betelgeuse: every two weeks is something I've been after for a while, although the "no major discussion rule" I don't totally agree with.
22:00:17 <wired> perhaps there should be a rule that when something is brought up for voting, it must be resolved
22:00:38 <ferringb> wired: "postponed" is a resolved
22:01:17 <dertobi123> not only the council was quite passive, also requests from the developer body were quite rare ... if you want to get things done and/or decieded by the council, mailing them via council@g.o is the way to go atm - but not that many people used that. i'm pretty sure lots of interesting stuff got lost on gentoo-dev@g.o
22:01:38 <leio> I typically don't see any outside meeting discussions initiated either, in some cases there are gentoo-dev threads, which die
22:02:14 <dertobi123> and having items where discussions start within a meeting lead to people looking unprepared and not being able to decide something
22:02:56 <ferringb> from my perspective, there were dev requests
22:03:06 <dertobi123> so in the end it's not the council who failed somehow, but also our developer body
22:03:18 <ferringb> just very, very rarely did the council pick them up and run with them.  if it occured at all, it was typically either ulm or calchan proxying it into their realm of awareness
22:03:35 <ferringb> dertobi123: my personal view, but the council should be actively engaging the dev body
22:03:54 <ferringb> I'm not sure if others agree with that however
22:04:26 <dertobi123> ferringb: probably none (or just a few) of the devs did mail those requests to council@g.o
22:04:30 <ferringb> point is, from my view the council should be doing more than just "it's that time of the month" emails and sifting through the results- actively tracking what was requested, notifying why it was punted, etc.
22:04:47 <ferringb> well, if the request for discussion goes to *dev*, it's not particularly surprising, no?
22:04:48 <dertobi123> discussing things on irc or via mail doesn't automatically put this things on the council's agenda
22:05:36 <ferringb> dertobi123: actually, it should.
22:05:54 <ferringb> not even should, it kind of must considering the emails sent to dev.  one second.
22:06:25 <ferringb> http://www.gossamer-threads.com/lists/gentoo/dev/210446#210446
22:06:29 <wired> i agree with ferringb here, council members are devs, i'd like to see them filter through the dev mail and take decisions without someone explicitly requesting it when devs cant reach a decision themselves
22:07:05 <ferringb> emails of that sort have been going to dev since day one, and very, very little of what people request on those threads ever makes it into the council's purview.  if those email threads won't be watched by the folk requesting discussion topics, the emails need to stop.
22:07:31 <ferringb> explicit requests need addressing (even if it's a statement of "piss off you wanker").  implicit would be nice, but explicit isn't even working.
22:07:41 <wired> i also think that the 6 month term is a nice idea, with meetings closer to each other
22:08:28 <ferringb> any council members got other issues they'd like to bring up?
22:08:36 -*- ferringb notes right now it's mostly the proxies speaking
22:08:50 <dertobi123> we don't have different opinions here - i described how the council worked until now, it's your chance to make the council more active
22:09:10 <ferringb> yeah, pardon- you're right.
22:09:17 <ferringb> dertobi123: other issues/comments?
22:10:12 <ferringb> brb
22:11:36 <Betelgeuse> ferringb: nope
22:11:48 <Betelgeuse> the webapp back log is open for ideas as usual
22:12:14 <dertobi123> no ... just one thing, i'd like to see a more (pro-)active council being elected and from my pov it would help to get a fresh council started with new and fresh people. no offense, but imho everyone should be able to be member of two councils and then have take a break.
22:12:33 *** Mode #gentoo-council -o dertobi123 by dertobi123
22:14:04 <-> dabbott|afk is now known as dabbott
22:14:05 <leio> I think for the whole council to be (pro-)active all individual members need to be proactive and proactively discuss things in this channel and elsewhere with others, so that there isn't one or a couple doing that, others missing drive, and then remaining getting a burnout
22:17:28 <Betelgeuse> I need to go. Thanks for the year and let's see who show up next month.