path: root/meeting-logs/20080214.txt

20:00 < Flameeyes@> we're up in a minute
20:00 <      amne@> wheeeeeee
20:00 <    jeeves > dooooooork
20:00 < Flameeyes@> actually, we're up, I think
20:00 < Flameeyes@> jeeves, you act like a jerk ;)
20:00 <   lu_zero@> hi
20:00 <     jokey@> ho
20:00 < Flameeyes@> roll-call!
20:01 <      amne@> let's roll
20:01 <   lu_zero@> seems to be the time now
20:01 < Flameeyes@> amne, dberkholz, jokey, lu_zero, SpanKY, Betelgeus (e)?
20:01  *  lu_zero rolls
20:01 < dberkholz@> yep
20:01 <     jokey@> yup
20:01 < Flameeyes@> lu_zero, how many d20?
20:01  *    jokey rolls as well
20:01 <   lu_zero@> I'm cooking so I'll be here and there for the next 10 min
20:01 <     jokey@> 18
20:02 <   lu_zero@> time to have other people take care of the stuff
20:02 < dberkholz@> Betelgeus, vapier, SpanKY: ping
20:04  *  lu_zero is watching people cooking takoyaki while he is preparing some sort of soup and some venus rice is in the boiler
20:04  *   araujo gets some tea
20:05 < Flameeyes@> lu_zero, where on the world are you now?
20:05 -!- rbrown [n=rbrown@gentoo/developer/paludis.rbrown] has joined #gentoo-council
20:05 <      amne@> Flameeyes: i guess it's because of valentines day
20:06 <    araujo > that's a good question .. don't you guys have a life?
20:06 < dberkholz@> it's only noon here.
20:06 < Flameeyes@> araujo, I think I do, but it's almost as good as new
20:06  *   araujo thought he was the only one with no life
20:06 < dberkholz@> this was a bad day to schedule a meeting. we should've moved it.
20:07 <    araujo > yeah
20:07 <      amne@> why that?
20:08 <   lu_zero@> dberkholz right
20:08 <    araujo > because it's valentines
20:08 <    araujo > well, for those with a life
20:08  *     amne has a life but valentine's day = no care :-P
20:08 <   lu_zero@> ehm
20:08 <   lu_zero@>  lupercalia you mean
20:08 < dberkholz@> alright, let's give the slackers till :15 and then get rolling. go for a walk or something
20:09 <      amne@> heh
20:09  *   araujo walks in circle around the room
20:09 < dberkholz@> that's about 5 minutes for the time-challenged
20:09 <   Halcyon > Pretty sure vapier isn't going to make it, he told me he wasn't feeling well last night, so I am going to fill in for him.
20:10  *  lu_zero gots some time to cook other weird stuff 
20:10 < dberkholz@> Halcyon: i just want to check whether he actually asked you to proxy, or you're just offering to fill in now
20:11 < dberkholz@> ok, just got some queried goodness from Halcyon 
20:11 < dberkholz@> that leaves Betelgeus with 4 minutes to show up
20:11 <   lu_zero@> ^^;
20:12  *    jokey moves quickly upstairs
20:12 <    araujo > well, we have 30 minutes yet
20:12 <    araujo > enough time for lu_zero to prepare sushi for everyone here
20:13 <   Halcyon > I already had sushi today...no more please.... :P
20:13 <     jokey@> heh
20:13 <     jokey@> sushi
20:13 <     jokey@> mmmmm
20:14  *   araujo should get some for dinner
20:14 <    araujo > hiho jokey
20:15 < dberkholz@> ok, let's get going.
20:15 < dberkholz@> everyone's got the agenda?
20:15 < dberkholz@> if not, it's in /topic
20:15 <     jokey@> yup
20:16 <    araujo > dberkholz, shouldn't we wait 30 more minutes?
20:16 < dberkholz@> araujo: how long do you want to wait? it's not that hard for people to be on time, or even within 15 minutes of on time
20:16 <   lu_zero@> I'd wait other 15 and then call the slackers
20:16 <    araujo > just saying ... since it says is at 2000 UTC
20:17 < dberkholz@> and it's 20:17 utc
20:17 <      amne@> who's effectively missing now?
20:17 < dberkholz@> Betelgeus
20:17 <      amne@> well, i think that's enough people and if he shows up late that works for me too
20:18 <      amne@> as long we don't have a vote where his voice is resolving a tie or something...
20:18 < dberkholz@> it really shouldn't be that tough for people to show up on time or ask someone to proxy. this lack of respect for the rest of us is not acceptable
20:18 <   lu_zero@> =/
20:18 <     jokey@> just start
20:18 <      amne@> yeah
20:18 <   lu_zero@> dberkholz I could think about N situations that are cause of incidents
20:18 <   lu_zero@> including network outage
20:18 <   lu_zero@> anyway
20:19 <   lu_zero@> let's go
20:19 <    araujo > :-P
20:19 <     jokey@> we have slacker marks damnit ;)
20:19 < dberkholz@> first, progress on last meeting's topics
20:19 < Flameeyes@> dberkholz, you sounded like Darth Vader, but keep on going :)
20:20 < dberkholz@> only one with any discussion worth speaking of is the CoC thing, again.
20:20 <   lu_zero@> sigh
20:20 <   lu_zero@> fine with your latest idea
20:21 <     jokey@> yup, that looked okay to me as well
20:21 < dberkholz@> i'm going to just move on with making this happen. if it goes fine, no need to bother with anything further
20:21 <      amne@> good
20:21 <   Halcyon > Sounds like a plan.
20:21 < Flameeyes@> nice to hear
20:22 <     jokey@> dberkholz++
20:22 <      amne@> if there's some place found, i'd be interested hanging out there a bit just to see what's going on and probably giving some advice if people want to hear it
20:22 <      amne@> other than that i have not too much interest to get involved deeper though, nor the time
20:23 < dberkholz@> various folks have brought up the idea of groups not working and such, so might just have some public IRC backchannel for discussing stuff before posting on -dev. not sure about that
20:23 < dberkholz@> anyway, i think we're set on that point
20:23 <      amne@> yeah
20:25 < dberkholz@> nothing more to talk about in old topics
20:25 < dberkholz@> in new topics, we've got glep 46
20:25 < dberkholz@> i'll give ya a couple minutes to review it again
20:26 <      amne@> is anyone from the glep folks around?
20:26 < dberkholz@> do you have specific questions?
20:27 <      amne@> yeah, 2 minor things
20:29 < dberkholz@> vanquirius isn't on irc, dev-zero's idle 8 hours, ciaran's away
20:29 <      amne@> well, actually it's just one thing:
20:29 < dberkholz@> you'd think that people with an agenda item would attend the meeting
20:29 <      amne@> changelog should contain a URL prefixed with http:// or https:// where the location of the upstream changelog can be found.
20:30 <      amne@> same for doc - not really likely for most packages, but perhaps this should be extended for ftp://
20:31 < dberkholz@> i'm curious why it needs to specify the allowed protocols
20:31 <      amne@> there are some projects that only host on ftp, iirc vsftpd is one of them
20:32 <      amne@> e.g. ftp://vsftpd.beasts.org/users/cevans/untar/vsftpd-2.0.6/Changelog, which is linked on http://vsftpd.beasts.org/
20:32 <      amne@> dberkholz: following my thoughts further that leads to your question, yeah
20:33 < Flameeyes@> dberkholz, I suppose for the software to be prepared to read that protocol
20:33 < Flameeyes@> but I admit a generic URI would be nicer
20:33 < Flameeyes@> for instance it would be easier to point to an svn:// url
20:33 < dberkholz@> just reject stuff you don't know how to handle
20:35 <      amne@> ah right, the other thing that may be a bit confusing (at least to me): there's maintainers and maintainers
20:35 <      amne@> one of them is upstream (as specified inside the upstream tags) and the other one the gentoo maintainer
20:35 <      amne@> still, this may be a bit confusing (or is it just me)
20:35 < Flameeyes@> not just you for sure
20:36 < dberkholz@> being part of the upstream hierarchy makes that reasonably clear to me
20:36 <   Halcyon > In the structure its going to be put into, it seems sane enough to me as well.
20:37 < Flameeyes@> ambiguos no, a bit confusing yes to me, but I ca cope with it :)
20:37 < dberkholz@> if i'm looking at an appropriately indented metadata.xml with that, it's pretty straightforward
20:37 <      amne@> ok, works for me then, just wanted to make people aware before they notice it later and cry about it ;-)
20:39 <      amne@> ok, so we're happy with that then i think
20:39 < dberkholz@> right. so what do we want to do with this?
20:39 < dberkholz@> it's pretty much approveable, but there's the protocol question
20:40 <   lu_zero@> I'd just update and approve
20:40 < Flameeyes@> if possible, I'd say approve it without protocol limitation
20:40 < dberkholz@> i suppose we could approve it with that caveat attached, which must be resolved before it can be officially "approved glep"
20:40 <      amne@> yupp
20:40 < dberkholz@> there might be a reason for that, which we don't know
20:40 <   Halcyon > So, no protocol limitation at all, or do we want to come up with a sane set?
20:41 < Flameeyes@> that's a good reason for the proposers to be around when the council discuss a glep :P
20:41 < Flameeyes@> Halcyon, I'd say "valid URI" and that would be it
20:41 < dberkholz@> Halcyon: i think that's a question for the authors to answer, not us
20:41 < Flameeyes@> if we need a proper valid uri list...
20:41 <   Halcyon > The only reason I can think of is that you'd need who knows how many tools to be able to access the various SCMs (for example)
20:41 < Flameeyes@> dberkholz, does freedesktop have anything on that?
20:42 <   Halcyon > dberkholz: yea, I'm just kind of throwing it out there, for what is going to be asked of them.
20:42 < dberkholz@> i don't know
20:44 <      amne@> just looked through the old and new threads, seems this wasn't discussed
20:45 <     jokey@> maybe people didn't even think about that or assumed it
20:46 <      amne@> (or i'm not seeing it of course)
20:46 <      amne@> jokey: i'd guess so
20:46 <     jokey@> at least I would assume it without further thinking
20:47 <      amne@> well, i consider it rather a minor detail that needs to be figured out while the rest of the glep looks good to me
20:47 <     jokey@> same here
20:47 < dberkholz@> alright. let's vote on approved with caveat
20:47 <     jokey@> I'm all for it
20:48 <   Halcyon > Sounds good
20:48 < dberkholz@> yes from me
20:48 <      amne@> so if it's fine with you folks we could just approve it under the condition it'll be discussed
20:48 <      amne@> i type too slow
20:48 < Flameeyes@> good for me
20:48 <      amne@> ++ from me if not clear already
20:48 < dberkholz@> great.
20:48 < dberkholz@> next topic, another one that keeps coming up, is eapi
20:48 < dberkholz@> Halcyon wants a spec
20:49 <   Halcyon > s/Halcyon/lots of people :) /
20:49 < dberkholz@> ok, 3 people
20:49 < dberkholz@> one of whom is here to discuss his desires
20:49 <   Halcyon > I definitely want it from a QA perspective as well.  Its hard to enforce something when there is no official spec.
20:50 < dberkholz@> Halcyon: is there some particular part that's unclear?
20:50 <   lu_zero@> the fact it isn't on a single paper?
20:51 <   Halcyon > Correct.  The fact that we don't have a clear specification of what EAPI=0 is, and what was added to it, in a document that can be easily referenced for new developers or people writing package managers or qa.
20:51 <   Halcyon > PMS seems to fill this void, but we don't have an approved version of it that defines EAPI=0 and 1.
20:51 < dberkholz@> it's in ebuild(5) in two places if you search for /EAPI 1
20:52 -!- wolf31o2|work [n=wolf31o2@gentoo/developer/wolf31o2] has joined #gentoo-council
20:53 <   Halcyon > And I just dragged wolf31o2|work in here since he has the same feelings as I do.
20:53 <wolf31o2|w > ehh... hi
20:53 < dberkholz@> Halcyon: ok, we don't have EAPI=0 specification approved. so there isn't anywhere that we can point to that has a spec of it.
20:54 <   Halcyon > So, if the ebuild manpage is our specification, what purpose does PMS have?  I'm just trying to understand where we are going to have all of this documented.
20:55 <   Halcyon > And it would be nice for other package managers to be able to point to a document that they are implementing (I don't think paludis or pkgcore want to use our ebuild manpage)
20:55 < dberkholz@> anyone else think we almost have too many sources of ebuild documentation?
20:55 <     Opfer > Here!
20:55 <      amne@> dberkholz: exactly my thoughts
20:55  * Flameeye  
20:56 <wolf31o2|w > for sure
20:56 <wolf31o2|w > heh
20:56 <   Halcyon > dberkholz: definately, which is why I'm trying to figure out where all of this is going to be documented.
20:56 <     jokey@> thing is
20:57 <     jokey@> specs don't help people writing ebuilds, they just help PMs
20:57 < dberkholz@> Halcyon: i'd say put it into the devmanual
20:57 <   lu_zero@> eh
20:57 <   Halcyon > I guess the real thing we need to figure out is, what are the purposes of these pieces of documentation, and what should they contain? 1) ebuild manpage 2) Devrel dev handbook 3) QA's devmanual 4) PMS
20:57 < dberkholz@> have a section that will list EAPI additions
20:58 <     solar > EAPI= additions are being done on the fly and are illy defined
20:58 <wolf31o2|w > 2 should be development policy-only, IMO...
20:58 <      zlin > jokey: it certainly helps ebuild devs when there's a dispute about whether a bug is an ebuild bug or a pm bug..
20:58 < dberkholz@> the ebuild manpage should be a reference, and i think it does a reasonably good job of that. it also pretty much duplicates a chunk of the dev handbook (not devmanual)
20:59 <   Halcyon > jokey: the spec helps everyone to know what is valid (especially QA)
20:59 -!- tsunam [n=tsunam@gentoo/developer/tsunam] has joined #gentoo-council
20:59 <   lu_zero@> and a way to enforce QA includes have the PM helpers
20:59 <   lu_zero@> so it's fine having a sane spec
21:00 <   Halcyon > lu_zero: the problem is that we do not have a specification document that is approved right now for what should be implemented for an EAPI=1 compliant PM.
21:00 <wolf31o2|w > a specification also defines how the PM should act... so one knows what functions to use/not use, given a desired outcome...
21:01 <wolf31o2|w > we don't even have it for EAPI=0
21:02 <     jokey@> Halcyon: thing is, both paludis and pkgcore have hit a bunch of corner cases which haven't been defined and won't be unless you either reverse-spec the portage code or hit those cases
21:02 -!- Pesa [n=Pesa@151.16.87.94] has joined #gentoo-council
21:03 < dberkholz@> the devmanual feels like an extended reference, with the same reference material but also additional descriptions. still not a tutorial, though.
21:03 <   Halcyon > jokey: which is why we should have defined EAPI=0 before we decided to tack things on.  We don't even completely know how Portage is supposed to behave is what I'm reading in your statement.
21:03 <   Halcyon > But, that's beside the point since we are where we are now.  Working on, and getting a complete specification that is approved by the council should be the end goal.
21:03 < dberkholz@> well, we can define a difference in behavior as EAPI=1
21:04 <      zlin > dberkholz: same reference material as what?
21:04 <     solar > Halcyon: has the topic of GLEP's for EAPI >0 additions been discussed here yet?
21:04 <wolf31o2|w > dberkholz: only after EAPI=0 is defined
21:04 < dberkholz@> zlin: as ebuild(5), as the dev handbook's section on the same content on ebuild(5)
21:04 <   Halcyon > solar: no, I was just going to raise that.
21:04 < dberkholz@> wolf31o2|work: i think we can define a set of conditions that were not guaranteed before but are after without also specifying everything else that happens before and after
21:05 <   Halcyon > dberkholz: those conditions could interact with thigns that are not specified as well though, in that case.
21:05 <wolf31o2|w > dberkholz: then why even bother? why not just tack it all onto EAPI=0 and say that there's no specification yet and everything is EAPI=0
21:05 <wolf31o2|w > exactly
21:05 < dberkholz@> maybe EAPI=0 should even be built up from smaller chunks like that
21:05 < dberkholz@> just so we can make some progress on getting it approved
21:06 <wolf31o2|w > *nothing* is specified at this time... so any conditions would be against what, exactly? a best-guess?
21:06 <   Halcyon > So, just to get this somewhat focused again, here are a few things I would like answers to (and I think we can answer them)
21:06 <NeddySeago > EAPI-0 is history with EAPI-1 being in use. Why spend effort documenting EAPI-0 ?
21:06 <    tsunam > erm...
21:07 < dberkholz@> EAPI=1 isn't a complete replacement of everything in EAPI=0, NeddySeagoon 
21:07 <   Halcyon > 1) What should be the purpose of PMS?  Is this going to be the document that gives the complete spec and is approved before introducing new EAPIs.
21:07 <    tsunam > eapi=1 is seeking to expand on eapi=0
21:07 <NeddySeago > dberkholz my understanding is that its a superset
21:07 <   Halcyon > NeddySeagoon: so what did it build upon?  That's why we need to know.
21:08 <    tsunam > NeddySeagoon: as well believe that eapi=0 eapi=X if that's the way it goes...are to be interoperable so that one doesn't violate another...etc
21:08 <   Halcyon > But, before we go all over the place, lets take this on a point by point basis please :)
21:08 <NeddySeago > tsunam, Ah yes ... I had overlooked the interoperability
21:09 <   Halcyon > So, does anyone have an answer for my #1? :)  I have more to follow.
21:09 <     jokey@> Halcyon: +1 and even over to -dev ML :)
21:10 <   Halcyon > jokey: yes, it will need to go there as well, but I'd like to atleast get intial input from the council.
21:10 -!- Opfer [n=Opfer@gentoo/developer/opfer] has quit ["Leaving."]
21:11 < dberkholz@> i think it's important to approve PMS as EAPI=0 specification. i'm not convinced that needs to happen before we can define other EAPIs.
21:11 <      zlin > wolf31o2|work: we bother to avoid breaking every portage that doesn't support slot deps or iuse defaults...
21:11 < dberkholz@> i am more interested in not blocking new features we need
21:11 < dberkholz@> the problem is that not having PMS of EAPI=0 approved makes it difficult to refer to it as a good source for anything else
21:12 <   Halcyon > dberkholz: I agree we shouldn't impede upon progress, but we need to know the foundation we are building upon before piling more on top of it.
21:13 <   Halcyon > But, we do agree that PMS should be the PM agnostic specification (resources, reference, whatever) that is to be approved before EAPI features are introduced in the tree? (once we clean up where we are at now and get PMS up to speed)
21:14 <     solar > once the PMS is final.. No additions really should be added to it w/o hitting the GLEP process. Then there should be a fixed period of time before ppl just start using it.
21:15 <   Halcyon > solar: that's kind of what I"m getting at.  I'm just trying to see if everyone else agrees too :)
21:15 < dberkholz@> the whole fixed period of time thing is the reason for EAPI...
21:15 < dberkholz@> s/the/a/
21:16 <      amne@> Halcyon: i think that makes sense, yes
21:16 <    tsunam > We're talking about something that effects a LOT of people
21:16 -!- Betelgeuse [i=betelgeu@a88-114-31-12.elisa-laajakaista.fi] has joined #gentoo-council
21:16 -!- mode/#gentoo-council [+o Betelgeuse] by ChanServ
21:16 < dberkholz@> i just took a look at PMS, and it defines the EAPI=1 features similarly to ebuild(5) -- all mixed in with the other text. it also adds a mention of ECONF_SOURCE, the part of it most people don't care about.
21:16 <   Halcyon > 2) Okay, and if PMS is not up to par by the time EAPI=2 comes around, the council will still be approving that before people add it to the tree, correct?
21:16 < Flameeyes@> hi Betelgeuse
21:17 <     solar > who defined it and when? Who approved those additions?
21:17 <   Halcyon > dberkholz: I'll be actively working on PMS soon, so I'll be cleaning it up and making it easier to look through.
21:17 <      amne@> Halcyon: can you re-word your question? not sure if i understand it correctly
21:18 <      amne@> Halcyon: you mean eapi=2 needs to be approved before it will be used in the tree?
21:18 <   Halcyon > amne: yes.
21:18 <   Halcyon > Regardless of the status of PMS at that point in time.
21:18 <      amne@> ah
21:18 < dberkholz@> i agree that non-zero EAPIs have to be approved by the council before they can be used in the tree
21:19 <   Halcyon > Okay, and we agree that PMS should be the single point of reference for EAPIs?
21:19 <Betelgeuse@> Flameeyes: Stupid connection stuff.
21:20 <Betelgeuse@> Well better late than never.
21:20 <  g2boojum+> I assume that the concern is not that the various PMs are getting out of sync w/o a well-defined EAPI, but that features are going in w/o some sort of official channel?
21:20 <Betelgeuse@> Flameeyes: can you upload logs so far somewhere?
21:20 <   Halcyon > g2boojum: partly that is my concern.  My other concern is to know where the problem really lies when something doesn't work as expected.
21:21 < Flameeyes@> Betelgeuse, http://rafb.net/p/OwynUu65.html
21:22 <   Halcyon > So, I will begin actively working on PMS and hopefully by next month I will have made some progress on getting it closer to being "complete".  At that point in time we can assess where we are at and take it from there?
21:23 < dberkholz@> that sounds reasonable
21:24 <      amne@> Halcyon: if there is a complete PMS (or at least one that can be expected to be complete soon) as a basis for future eapis, that surely would be the best way to do things
21:24 < dberkholz@> since this is another issue tied to package managers, i'd certainly want to hear from their devs before approving EAPI=0
21:24 <   Halcyon > amne: it should have been how we started before we moved forward, but we can't go back in time now, so I want to make it better :)
21:25 <      amne@> Halcyon: agreed - sounds great
21:26 <Betelgeuse@> Halcyon: I don't see a problem in defining EAPI 1 before 0 is finished
21:26 <Betelgeuse@> as it only add things
21:27 <   Halcyon > Betelgeuse: we already discussed why that isn't a good way to look at it.  What are you adding onto?  How does it interact with what we started with?  Etc
21:27 <wolf31o2|w > wait... you see no problem with allowing adding to a specification that's undefined?
21:27 <Betelgeuse@> wolf31o2|work: If the other option is waiting for months to get the spec finished, then no.
21:27 <     solar > Halcyon: Re: take it from there. So does that mean that ebuild-devs need to stop using the undefined EAPI=1 stuff in the tree?
21:27 <    tsunam > Betelgeuse: because eapi0 isn't finished...we don't have a solid basis on which to base anything off of. Its like having 1 brick as a base then building horizontally on top of it...You'll get top heavy and topple over
21:28 <wolf31o2|w > Betelgeus: rather than spending resources to actually... you know... finish the specification and make it usable, thereby actually *resolving* the issue rather than working around it
21:28 <   Halcyon > solar: I don't think we are going to be able to convince people of that.
21:28 <Betelgeuse@> wolf31o2|work: That would work if we could make people work on it.
21:29 <Betelgeuse@> solar: Why?
21:29 < dberkholz@> it's not undefined. it's just not defined in precisely the way some of you apparently want it to be.
21:29 <    tsunam > wow...
21:29 <     solar > uhh
21:29 <wolf31o2|w > Betelgeus: how about "You won't be able to use EAPI=* where * != 0 until the spec is finished and approved and if you want the features, help get the spec complete" ?
21:29 <wolf31o2|w > you know... that works just *great* for professional software development... why should we work differently than the established best practices for software engineering?
21:30 <Betelgeuse@> wolf31o2|work: Well tons of Java stuff use EAPI 1 already so don't see going back.
21:30 <wolf31o2|w > ...and?
21:30 <wolf31o2|w > actually... never mind
21:30 <Philantrop > Betelgeuse: As does KDE 4. We won't go back either.
21:30 <   Halcyon > wolf31o2|work: solar: I don't think we are going to be able to go back, but I hope that everyone here does see what happened here, and learns from it atleast.
21:31 < dberkholz@> well, imagine people are saying this about EAPI=2 instead of EAPI=1, and it has USE deps.
21:31 <   Halcyon > Making progress is what we all want, but not at the cost of shooting ourselves in the foot down the road.
21:31 <    tsunam > in essense we already are shooting ourselves in the foot
21:31 <Betelgeuse@> But any way I did raise these concerns when zmedico initially brought EAPI 1 to gentoo-dev
21:31 <      amne@> yeah, let's focus on how we can do that
21:31 <      amne@> making progress, not shooting into our feet
21:31 <    tsunam > progress would be having a full spec
21:31 <wolf31o2|w > indeed
21:31 <      amne@> as far i see it, Halcyon seems to be pretty motivated to finishing it
21:32 <     solar > lets see. Head of releng, head of the x86 team, head of QA, head of hardened/embedded are all sharing our concerns and ppl are still putting undefined shiny little features ahead of everything else that matters.
21:32  *    solar is proud to be apart of this team sometimes
21:32 < dberkholz@> things that matter to you are not things that matter to everyone else in gentoo...
21:32 < dberkholz@> if they matter to you, work on 'em. i know y'all do that anyway
21:32 <Betelgeuse@> solar: I don't think I said that I would agree with how things went about.
21:32 <    tsunam > dberkholz: same goes for you
21:32 <      amne@> so if there are some people actively working on pms, do you see it finished within a reasonable timeframe?
21:32 < dberkholz@> tsunam: precisely. so people should work on things they care about. that's how oss goes
21:32 <Betelgeuse@> But I don't see any reason to stop using EAPI 1 now.
21:33 <    tsunam > dberkholz: we can run that one into the ground if you want
21:33 -!- ferdy [n=ferdy@gentoo/developer/ferdy] has joined #gentoo-council
21:33 <wolf31o2|w > solar: that's pretty much the way I see it... and what I'm hearing is that best practices for software engineering... you know, the stuff designed to allow for this sort of collaboration, should take a back seat to "progress"
21:33 <   Halcyon > Well, we could go at each other all day, but it isn't going to get us anywhere (fortunately or unfortunately).
21:34 <   Halcyon > To move forward here (in a sane way), I am going to get PMS done as quickly as I can.  Everyone that is interested in helping me, please let me know.
21:34 <NeddySeago > We are where we are ... how do we get to where we need to be ?
21:34 <   Halcyon > I am going to be vehemently opposed to introducing any new EAPI's into the tree until we have EAPI=0 and EAPI=1 defined in a full spec.
21:34 -!- hparker [n=hparker@gentoo/developer/hparker] has quit [Remote closed the connection]
21:34 <   Halcyon > We can't go back from where we are (again fortunately or unfortunately), so lets all try to work together to ensure that things that you want to work on, don't make things I want to work on a living nightmare.
21:35  *    jokey can live with not introducing new stuff but we have to keep eapi=1 now
21:35 <wolf31o2|w > why?
21:35 <Betelgeuse@> Halcyon: sounds good
21:35 <wolf31o2|w > why do we have to keep it?
21:35 < Flameeyes@> I agree with Halcyon
21:35 <wolf31o2|w > can anyone actually explain that one?
21:35 <      amne@> i think Halcyon's suggestion is good
21:35 <NeddySeago > Halcylon ... timescales ?
21:35 <   Halcyon > wolf31o2|work: I'm saying we have to keep it just because I don't want to get into a long drawn out fight that isn't going to get me anywhere.
21:35 <   Halcyon > NeddySeagoon: when it gets done.  The more help I have, the faster I can get it done.
21:36 <wolf31o2|w > Halcyon: so we're allowing the proliferation of EAPI=1 to continue into the tree?
21:36 < dberkholz@> we're also allowing the proliferation of EAPI=0 to continue into the tree...
21:37 <   Halcyon > wolf31o2|work: I don't see anyway of stopping it at this point.  Not without causing many many people to get into a useless flamewar.
21:37 <      zlin > wolf31o2|work: have you really forgotten which council decided that pms wasn't important?
21:37 <wolf31o2|w > dberkholz: without a specification of EAPI=0 and EAPI=1, there is *NO* EAPI, at all... so your point is completely moot
21:37 <Betelgeuse@> Why is there an open floor?
21:37 < dberkholz@> if there's no eapi, you aren't arguing about anything
21:38 < Flameeyes@> Betelgeuse, we decided not to moderate unless needed
21:38 <wolf31o2|w > zlin: is there a point you're trying to make or are you just trying to point fingers because you don't have anything useful to contribute to the conversation?
21:38 < dberkholz@> because we just arbitrarily add new features whenever we feel like it
21:38 <Betelgeuse@> Flameeyes: could be
21:38 <wolf31o2|w > dberkholz: that's my point, entirely...
21:38 < Flameeyes@> Betelgeuse, could be needed? :P
21:38 <wolf31o2|w > dberkholz: there's nothing stopping me from adding EAPI=wolf31o2 and doing whatever I damned well please with it...
21:38 < dberkholz@> council never approved your eapi. =P
21:39 <    tsunam > dberkholz: has the council approved EAPI=1?
21:39 <   Halcyon > wolf31o2|work: I'm willing to bend on somethings, so long as we don't screw up again in the future, and it seems to me that we are in agreement on that, mostly....
21:39 <   Halcyon > tsunam: yes, they did actually.
21:39 <wolf31o2|w > tsunam: no, but they approved its *use*
21:39 <Betelgeuse@> Flameeyes: Just saying that I have no recollection either way.
21:39 < dberkholz@> we approved EAPI=1 and cited the 3 features that included
21:39 <Betelgeuse@> wolf31o2|work: well isn't that the same thing?
21:40 <wolf31o2|w > Betelgeuse: show me the specification and I'll agree... otherwise, no it is not the same thing, at all
21:40 <     solar > ok so thats the end of EAPI=1 and now we are on to =2 ?
21:40 < Flameeyes@> I'm going to infringe lu_zero's copyright and just say "yawn"
21:40 <Betelgeuse@> wolf31o2|work: So you argue that EAPI=1 must be written down to exist?
21:40 < dberkholz@> does anyone feel like we're making any progress in this discussion, with the exception of Halcyon committing to working on the PMS?
21:41 <    tsunam > wow...again
21:41 < Flameeyes@> dberkholz, no
21:41 < dberkholz@> thi sis 21:39 <wolf31o2|w > tsunam: no, but they approved its *use*
21:41 < dberkholz@> woops
21:41 < dberkholz@> this is not going anywhere that i can see
21:41 <Betelgeuse@> Yep
21:41 < Flameeyes@> dberkholz, agreed
21:41 -!- solar [n=solar@smtp.gentoo.org] has left #gentoo-council []
21:41 -!- wolf31o2|work [n=wolf31o2@gentoo/developer/wolf31o2] has left #gentoo-council ["Leaving"]
21:43 <   Halcyon > And do we agree that EAPI=2 can be held off until we get an approved specification done?  I don't even think EAPI=2 is on the radar yet, so please just keep PMS and myself in mind when it does come around.
21:43 <      amne@> so. back to constructive?
21:43 <   Halcyon > I'll hopefully be completely done by that point in time.
21:43 < Flameeyes@> Halcyon, agreed
21:43 <     jokey@> Halcyon: +1 on that
21:43 <   lu_zero@> fine...
21:43 <      amne@> Halcyon: as long it doesn't take a trillion years for PMS, i think that's a good plan to stick to
21:43 <   Halcyon > amne: if it does, you can override me (the council is the only group capable over overriding QA anyway :) )
21:44 <   Halcyon > But I don't see it taking that long if I can get some movement going on it.
21:44 <   lu_zero@> btw could we have the specs in rfc format?
21:44  *    jokey fetches one of Halcyon's cars and overrides him just in case ;
21:44 <   lu_zero@> (or a format that isn't yet another one)
21:45 <   Halcyon > lu_zero: I'll see how it is now, how flexible it is, and take it from there.
21:45 <   Halcyon > In the end though, its probably going to be what I'm most comfortable with (like devmanual) since no one else is going to work on it for the most part.
21:45 <      amne@> Halcyon: i wouldn't want to do that as i'd rather like to see someone actively working on it anyway :-)
21:46 <Betelgeuse@> Halcyon: Considering that zmedico hasn't implemented use deps yet it's coming very soon
21:47 -!- kingtaco|laptop [n=kingtaco@gentoo/developer/kingtaco] has joined #gentoo-council
21:47 <   Halcyon > Betelgeuse: I'm sure it can wait a few weeks if we are making progress with PMS.
21:47 <   Halcyon > People have waited this long, another few weeks isn't going to kill anyone.
21:47 <Betelgeuse@> Halcyon: Has someone said something would be coming in few weeks that I missed?
21:47 <   Halcyon > Betelgeuse: I'm not even sure what you are saying now, could you rephrase?
21:48 <   Halcyon > You said zmedico hasn't implemented use deps, so it will be coming very soon?  What's coming, EAPI=2?
21:48 <   Halcyon > What is "soon"?
21:49 <Betelgeuse@> Halcyon: I don't think there is any timeframe for use deps. I remember zmedico saying he would do them last summer :D
21:49 <Betelgeuse@> Halcyon: So I wanted to know where the talk about a few weeks comes from as I don't understand it.
21:49 <   Halcyon > Betelgeuse: okay, so we shouldnt' have a problem for quite awhile (since I'll have plenty of time to get PMS done)
21:50 <   Halcyon > Betelgeuse: you said use deps would be coming soon.  I read that to mean as in any day now, so I was saying that using EAPI=2 could wait a few weeks if that was the case, so I could get documentation done.
21:50 <Betelgeuse@> Halcyon: Seems I missed a not
21:50 <   Halcyon > Betelgeuse: I thought so, which was why I got confused :)
21:51 < dberkholz@> ok
21:51 < dberkholz@> that's the last topic we had, and i'm please that Halcyon committed to some action so we're making progress
21:51 < dberkholz@> pleased*
21:52 < dberkholz@> are there any _new_ topics to bring up?
21:52 <   Halcyon > Okay, and on that note, I have to run.  I will give you guys an update by the next meeting, if not before then.
21:52 <      amne@> Halcyon++
21:53 <      zlin > Halcyon: I think Betelgeuse was trying to say that if pms comes within a month or two it'll likely be very soon compared to use deps..
21:54 <      zlin > or not
21:54 <Betelgeuse@> use deps happens very fast after someone starts working on the
21:54 <Betelgeuse@> m
21:57 < dberkholz@> i've posted a summary at http://dev.gentoo.org/~dberkholz/20080214-summary.txt
21:57 < dberkholz@> take a quick look through it
21:58 < dberkholz@> Betelgeuse: do you think you should get a slacker mark?
22:01 <      amne@> dberkholz: summary looks good, just one detail about glep 46:
22:01 <      amne@> Approved, with a caveat
22:01 <      amne@> did someone count?
22:01  *     amne scrolls back
22:01 < dberkholz@> yeah there was like 5 votes
22:01 <      amne@> yeah, just saw it
22:01 <      amne@> all fine then :-)
22:02 <Betelgeuse@> dberkholz: I would let others decide.
22:02 <Betelgeuse@> dberkholz: I would mark myself as late like with the timezone session.
22:02 <      amne@> Betelgeuse: i guess it depends why you were late
22:03 <Betelgeuse@> amne: Battling with the server.
22:03 <      amne@> Betelgeuse: if you were busy watching chuck norris movies, then you're a slacker
22:03 <Betelgeuse@> amne: You should see from your logs that it has been down for most of the day.
22:03 <      amne@> well, that's fair enough for not being a slacker imho
22:03 <      amne@> Betelgeuse: too lazy to check and i trust people easily
22:04 <Betelgeuse@> amne: good, I counted on that.
22:04 <Betelgeuse@> dberkholz: It should at least reflect that I was here for half of it.
22:05 <      amne@> Betelgeuse: heh
22:06 < dberkholz@> we really need to come up with some kind of rule for what to do when people are crazy late.
22:06 < dberkholz@> this isn't the first time
22:06  *  lu_zero is still cooking takoyaki
22:06 <      amne@> lu_zero: don't starve while cooking
22:06 <   lu_zero@> I'm trying...
22:06 < dberkholz@> anyone else got thoughts or care?
22:06 <Betelgeuse@> dberkholz: No they slack :D
22:06 < Flameeyes@> dberkholz, I unfortunately got used to people being late
22:07 < Flameeyes@> I live in italy after all, being on time is ... rare :P
22:07 <      amne@> dberkholz: not really as long it doesn't makes us unable to work
22:07 <      amne@> dberkholz: as for today, Betelgeuse was late, we started without him and i can live with that
22:07 < dberkholz@> i'm gonna put him as here, since he was around for really the most important bit
22:08 <      amne@> if we want to tighten up the rules, we can do that - not that i care a lot about whether it's done or not
22:08 <Betelgeuse@> amne: I don't think that's important
22:08 <Betelgeuse@> amne: You get what you vote for.
22:08 <      amne@> Betelgeuse: heh
22:08 < dberkholz@> might be a good idea to always have a backup proxy in case you're late
22:09 <Betelgeuse@> dberkholz: That's true.
22:09 <      amne@> Betelgeuse: if someone is always late, he'll get dropped from the council sooner or later, yes
22:09 <Betelgeuse@> dberkholz: But people shouldn't be late that often.
22:09 < dberkholz@> Betelgeuse: individual people, no, but 1 of 7 i think is higher chance
22:10 < dberkholz@> well, i think we're pretty much done with the meeting part after the EAPI bit, so let's officially declare it over
22:10 <      amne@> yeah
22:10 <      amne@> unless anyone has anything left to say
22:11 <      amne@> i'd like to say hi to my parents before the log closes
22:11 < dberkholz@> i asked once already and nobody chipped in
22:11 -!- Pesa [n=Pesa@151.16.87.94] has left #gentoo-council []
22:11 <      amne@> dberkholz: ah, seems i missed that with all the excitement we had today
22:13 < dberkholz@> ok, i'll send out the summary in a few minutes
22:13 <      amne@> great, thanks dberkholz
22:13 <   lu_zero@> ^^