[16:01] * fmccor is here representing Kugelfang [16:01] * wolf31o2 is mostly here [16:01] I'm here [16:02] kloeri, robbat2|na SpanKY vapier Uber [16:02] moo moo mr cow [16:02] hiya all [16:02] anyone have the agenda for today? [16:03] that would be nice [16:03] none were posted [16:03] I'm here sort of. [16:03] so i figured we just go over things from last [16:03] unless someone saw something i didnt [16:03] * KingTaco defers to vapier [16:03] i meant to post something, but i didnt get a chance ... been busy at work [16:03] only new thing I know of is the multiple suffixes stuff [16:04] from that adhoc meeting [16:05] I've been pretty busy as well [16:05] brb 5 mins [16:06] was a log posted from said adhoc meeting ? [16:06] http://thread.gmane.org/gmane.linux.gentoo.devel/48381 [16:07] I don't recall a log [16:07] iirc it was Kugelfang, kloeri and robbat2|na [16:08] I don't think the log was ever posted, just the notice kugelfang sent to -dev and following discussion [16:08] that'd be useful to have ... i dont think the decision made sense and the logic for banning it didnt cut it for me [16:08] yup [16:08] but let's start on something simpler first [16:08] anyone voluntering to run this meeting ? [16:09] since Danny isn't here, I say we make Ferris do it [16:09] *grin* [16:09] gets my vote :P [16:09] --> bazik (n=bazy@gentoo/developer/bazik) has joined #gentoo-council [16:09] fmccor: you want to take it ? [16:09] heh [16:09] --- vapier sets modes [#gentoo-council +m] [16:10] --- vapier sets modes [#gentoo-council +v fmccor] [16:10] --- vapier sets modes [#gentoo-council +v fmccor|home] [16:10] help if he could speak huh :) [16:10] guess not... ok... how about I do it [16:10] :) [16:11] now I just have to find last month's [16:11] http://www.gentoo.org/proj/en/council/meeting-logs/20070412-summary.txt [16:11] ok... let's go over that stuff, then tove reminded me that we have grobian's keywording glep [16:11] http://article.gmane.org/gmane.linux.gentoo.devel/48017 [16:12] you get your docs merged ? [16:12] k, the log from that impromptu meeting is available at http://dev.gentoo.org/~kloeri/multiple_suffix.log now [16:12] so... let's start with the old stuff... [16:12] vapier: thought robin was taking care of the actual commits [16:13] crap [16:13] it was me [16:13] I apologize [16:13] I guess that makes it a no. I didn't. [16:14] did we ever decide where exactly they were to go? [16:14] my spf doc is all xml'ed up and available on infrastructure.g.o [16:14] have been for a while now [16:14] under the infra project [16:14] since they manage the mail [16:15] yes, but were we just putting them up as-is or were we adding it into another doc? [16:16] for now, as is [16:16] k [16:16] let em be folded back as infra deems they care enough to do so [16:17] done, then... heh [16:17] what doc is this? [16:18] mail server docs [16:18] spf and reply to [16:18] ok... everything is committed [16:18] I thought robbat2 put those up somewhere [16:18] we're all good right? [16:18] are now [16:18] ok [16:18] <-- ian (i=need_hel@gentoo/developer/pdpc.active.ian) has left #gentoo-council [16:19] I think we should file a bug for the trustees [16:19] a bug for what purpose? [16:19] for the social contract thing [16:19] --> rbu (n=rbu@gentoo/developer/rbu) has joined #gentoo-council [16:19] I saw the comment in the last summary, but I wasn't sure exactly what was needed. [16:20] basically, the two should be in sync and things should be clarified... I can't comment further since I didn't see what needed to be changed, personally [16:21] sync Social Contract with Gentoo Foundation (external entities) [16:21] the Gentoo Foundation statement has a section about external entities influence [16:21] the Social Contract does not [16:21] vapier: Oh, that helps enormously. Thanks. [16:22] i proposed that we sync the wording from Foundation to Social Contract, but wolf31o2 wanted to clarify the Foundation statement first [16:22] i'm guessing he hasnt had a chance to drop an e-mail to the trustee list, so he'll file a bug instead for tracking [16:23] yeah... it's another of those cases where what was written is so generic it has no real meaning [16:23] wolf31o2: Yeah. Especially since the Foundation _is_ reigned by a company, the Gentoo Foundation, Inc. In any event, we'll take care of it. Thanks. [16:24] right [16:25] bug #177966 [16:26] ok... about the CoC... I poked amne earlier, but didn't get a response... I think he's away at the moment [16:27] I'm not aware of any actual changes to the document, but do know that the proctors were working on it [16:27] yeah they posted a few status tidbits previously [16:27] wolf31o2, On the bug, could you please give a reference to the document in question? [16:27] amne couldn't be around for tonights meeting but they don't have any updates yet [16:27] I don't think anything has changed [16:27] * fmccor is behind here. [16:27] --> ferringb (n=bharring@c-67-171-130-60.hsd1.or.comcast.net) has joined #gentoo-council [16:30] KingTaco: robbat2|na: any word on a dev-only archive for -core? [16:30] wolf31o2, I'm not involved in that [16:30] right, but do you know anything about it? [16:30] nope [16:30] k [16:30] the people to talk to are probably robbat2|na and solar [16:30] alright... [16:31] also, 2007.0 is out now, so we can get to working on the survey... Robin said he would look at it [16:31] just wanted to note that [16:32] so I guess all that is left is PMS unless someone has something else [16:32] grobians glep is what's left [16:32] right... sorry [16:33] so PMS then the GLEP? [16:33] I think ferringb had something too [16:33] ok [16:33] --- KingTaco sets modes [#gentoo-council +v ferringb] [16:33] ferringb, what were you riding me about last week? [16:33] heh [16:33] he was [16:34] * ferringb will get out the sadle again damn it :P [16:35] KingTaco: multiple suffix; few things... 1) suggestion to block further usage, instead of forcibly punting whats there now for the future... 2) actually discuss it, since right now it's completely blocked [16:36] aight [16:36] opinion on it's a wee bit divided to say the least, but have seen enough weird versions from upstreams that it's useful as a general mechanism- not all permutations necessarily are sane, which is really what the discussion ought be about [16:36] anyone disagree with what Kugelfang kloeri and robbat2|na did the other week? [16:36] the point was to block usage so we could discuss it on -dev instead of tree policy just randomly changing [16:36] anyone want to talk about it? [16:36] and the proposal to workaround it by leveraging -r### is plain wrong [16:36] ++ to that. [16:36] well, there's two things... first, *should* we limit the suffixes, and if so, how... [16:37] so I guess we'll start simple... should we limit the number/usage of multiple suffixes? [16:37] i agree that some combinations are weird and dont generally make sense, but i dont see any real logic as to what should be kept and what should be tossed [16:37] I'm all for multiple suffixes where it makes sense as long as we actually work that out first [16:37] I don't see a valid case for ever having multiple suffixes [16:38] KingTaco: using _p to track upstream date patches against upstream rc/beta/etc... [16:38] How does what kloeri propose differ from ferringb? [16:38] --> ahf (n=eroyf@gentoo/developer/paludis.minion.eroyf) has joined #Gentoo-Council [16:38] some upstreams suck... so anything we do without multiple suffixes is diverging from upstream versioning [16:38] but it did remind me of an idea I had the other day [16:38] I'll talk about it after this [16:39] ok... like I said, let's keep it simple... [16:39] as genone said, there is no technical reason for limiting the combinations [16:39] vote: should we allow multiple suffixes? [16:39] which is true ... if you support one combination of suffixes, you might as well make it completely arbitrary [16:40] I tend to agree that there's no reason to restrict it... the minute we do, somebody will make a version that does something we didn't expect and we're back to where we are now [16:40] and I vote yes, we should allow multiple suffixes [16:40] right, i'm for it [16:40] * Uber votes yes also [16:40] I'll vote Kugelfang yes. [16:40] if we're going to do multiple suffixes then we need to document the order [16:40] sure [16:40] the order is documented [16:40] Yes. [16:40] the thing about restricting it isn't technical but users should be able to look at 5 ebuilds and see what's newest which can get rather complex with multiple suffixes imo [16:41] i dont understand that argument at all [16:41] I vote no [16:41] kloeri: robbat2|na: ? [16:41] if a > b > c, then why do you need to document a > a.b > a.c > b [16:41] I think some cases might make sense but I don't think we should allow arbitrary combinations [16:41] well, I guess it goes in anyway... so now... should we limit what suffix combinations are allowed? [16:41] heh [16:42] imo we should [16:43] i'm with genone on this ... since there is no technical limitation, imposing our own arbitrary rules on an arbitrary system seems pointless to me [16:43] i'm with vapier, we shouldn't restrict needlessly [16:43] I agree with you/genone... I see no reason to limit it when it isn't a technical limitation we have to worry about [16:44] KingTaco: fmccor: robbat2|na: ? [16:44] wolf31o2, I already voted no to multiple suffixes [16:44] ok... [16:45] Conditional agree. As someone mentioned, they need to be human-decipherable without too much effort. [16:45] (Agree with vapier Uber etc) [16:45] ok.. that makes a majority [16:45] so when someone uses a silly suffix, you let them know [16:45] fmccor, it's a simple yes/no [16:46] "hey you, stop being stupid" [16:46] exactly [16:46] Then yes :) [16:46] I think that's silly [16:46] just inviting flamewars and more infighting [16:47] any more than any other decision we make? [16:47] probably not [16:47] no matter what, somebody isn't going to be happy, or else we wouldn't have to discuss/decide on it [16:47] ok... keywording GLEP... [16:47] yeah, we wouldnt need to exist in the first place [16:47] * fmccor would consider a flamewar over multiple suffixes to be proctor bait. [16:48] *cough* raised two points above actually :) [16:48] ferringb: got more to add ? [16:48] 1) what to do with multi-suffix, 2) how to handle such a decision in the future [16:48] answer to 2 is -dev [16:48] right [16:48] moreso, refering to the "outlaw further usage" vs "it has to be cleaned out of the tree now" [16:48] multi-suffix predates pms [16:48] ferringb: ehh.. we said multi-suffix is allowed... meaning it isn't blocked anymore [16:49] so if you have something that doesnt exist in portage now and requires a change in pms -> gentoo-dev [16:49] wolf31o2: again, difference between blocked for further usage, and making people change their versioning scheme for a 2 week window [16:49] that's actually my biggest complaint about the multiple suffix stuff that tree policy was just changed with no -dev discussion at all - no discussion I'm aware of at least [16:49] kloeri, why did you vote for it then? [16:50] at that meeting [16:50] KingTaco: I'm talking about allowing it in the first place without any discussion [16:50] ferringb: what exactly do you want us to discuss... we reversed the previous decision and allowed it... I'm not following what you want [16:50] wolf31o2: basically asking that in the future, when putting the breaks on something like this, block rather then punt while waiting for a full decision. [16:50] we blocked it to have that discussion [16:50] punted [16:50] kloeri: it was added by any other means; feature request long ago in bug tracker when portage was the only real solution [16:50] so complaining about a breakage in policy when a policy didnt exist is wrong [16:50] look at the original email, and the lack of multi-suffix in the tree now- it's usage was outlawed and folks had to change over right then/there [16:50] ferringb: ahh... ok... so nothing to do with this decision, but rather future ones... [16:50] yep [16:51] vapier: yes but we tree policy changes should be discussed on -dev ML imo [16:51] I see no problem with not requiring things be punted... we probably should have done that anyway... [16:51] and it's always been that way [16:51] breakage has historically been discussed, not new features [16:51] new features have been generally filed in bugzilla and then implemented as portage devs seen fit [16:52] in that case I want portage commit access so I can make some new features without discussing it :) [16:52] features != ebuild support changes :) [16:52] (might want to be specific there) [16:52] just because you're not watching the portage alias and seeing the discussion doesnt mean it didnt happen :P [16:52] ok, so the few devs watching the portage alias decides tree policy? [16:52] ferringb: where do you draw the line ? FEATURES ? RESTRICT ? USE ? [16:53] kloeri: the guys maintaining portage were deciding portage policy [16:53] that's how it has always been [16:53] you cant apply current view of the Gentoo world retroactively and take offense [16:54] important policy changes (and I count multiple suffixes as important) should be discussed on -dev imo [16:54] ok... I've got to run... on the GLEP, I totally agree with it [16:54] we've always required important policy changes to be discussed [16:54] vapier: just pointing out that misc. portage enhancements just affect that manager, not the others [16:55] ferringb: this change was made pre-others [16:55] vapier: well aware; was just correcting terms kloeri was using is all :) [16:55] so again, you cant take the current state of how things are handled and apply it retroactively [16:55] the GLEP reverses GLEP 22 and actually changes the keywording to match current practice, which I think is a good idea... and that's all I have time for, folks... [16:56] ok, I give up on this as nobody seems to agree that it should have been discussed [16:57] no, it definitely should have been discussed before being put in the tree, but not before it was put into portage [16:57] that make sense? [16:57] it's a moot point [16:57] we've decided going forward [16:57] wolf31o2: that's all I want [16:57] right [16:57] so glep [16:57] heh [16:58] it doesnt account for kbsd [16:58] but do we care [16:58] it doesn't explicitly list everything, just examples [16:58] How does it not? [16:59] (Like wolf31o2 said) [16:59] it doesnt account in the same way as the old glep [16:59] kbsd is half way between "x86" and "x86-fbsd" [16:59] I don't like it at all [17:00] And? [17:00] it creates confusion and work for almost all of the ebuild maintainers while only giving benefit to fringe arches [17:00] vapier: isn't that x86-kbsd ? [17:00] KingTaco: huh? go read it again... -linux is implied if missing [17:01] right, it just ratifies what we're currently doing [17:01] right... and now I really have to go... doorbell [17:01] oh, I missed that [17:01] been fun [17:01] sure [17:01] it's already being done [17:01] --- robbat2|na is now known as robbat2 [17:01] Uber, that's how I read it, too. [17:02] arr my head :-( sucky to get a cold now [17:02] sorry for being an hour late [17:02] robbat2, Kugelfang already used that excuse [17:02] robbat2, It's OK, you got all the action items. [17:03] lol [17:03] ok... got a free second... yeah, it's already being done... main thing is it reverses glep 22 and applies current practice as policy [17:04] I vote yes [17:04] yes for Kugelfang [17:04] yes [17:04] yes [17:04] this is grobian's glep? yes on that from me [17:04] robbat2: yes, his glep [17:06] yes [17:07] vapier? [17:08] <-- Jokey has quit (Client Quit) [17:09] well, it's got enough votes anyway [17:10] so that's it then? the chair (wolf31o2) has gone, so meeting closed? [17:10] heh [17:10] nope [17:10] vapier had something about PMS [17:11] Not if he's not here, he doesn't. :) [17:11] heh [17:11] ok.... [17:11] anything else? [17:12] Thanks to the PMS authors for a splendid job? [17:12] hahah [17:12] PMS vcs [17:13] still was up in the air from last time around [17:13] * KingTaco points to robbat2 [17:13] * Uber is away for 30 mins or so [17:14] ciaranm seemed to want to reject anything other than his own SVN [17:14] fuck ciaranm [17:14] if he's altered his position since then, there's a git repo waiting on stork [17:14] i just need to hook up whoever's going to be the main person handling it if we are doing an LKML model [17:15] or multiple people if desired [17:15] supposedly that's spb [17:16] also... EAPI1, might want to figure out how that'll be managed. basically, if it's ad hoc, doc writers get to force the format (sane or otherwise manager wise), potentially desired, potentially not [17:17] mainly pointing it out since discussion of each tidbit would be wise- pythons pep approach for example would work well once EAPI=0 is locked down imo [17:17] either way, ad hoc sucks. :) [17:17] sorry [17:17] work IRQ [17:18] i think the PMS state is still broken [17:18] i wont be happy until it's been moved to proper Gentoo hosting with proper teams behind it [17:19] i can let it sit until next meeting since people have taken off [17:19] vapier, see above, i've got a git repo on stork now, just need to give people access to it once they agree on the commit model [17:19] if they're all up with spb being the person to accept commits, then i'll give him write access when he's around, and read for everybody else [17:21] No objection that I can think of. [17:21] robbat2: seems moot dependant on how the EAPI discussions from above are managed frankly; if it's ad hoc, means spb/whoever controls that repo controls the format [17:21] other than git is a lame solution, it does satisfy my current requirements [17:21] robbat2, I don't think that'll satisfy vapier [17:21] no objection from me [17:21] doesn't sit well with me either [17:22] the latex format really makes PMS unaccessible for many to contribute to [17:22] KingTaco, which, git or ferringb's problem? [17:22] robbat2, 1 rw * ro [17:22] unfortunately that's the only way git will actually work here [17:22] I *love* latex, but if it's going to be a gentoo thing, it should be xml like everything else [17:22] grafting a dscm solution into a scm solution [17:23] we made a mistake with the devmanual [17:23] no, it can work for $N rw * ro too [17:23] who was working on xmlifying it ? [17:23] dunno [17:23] I don't think it ever got farther than chitchat [17:24] Does it matter until it's final? [17:24] yes [17:24] wait [17:24] fmccor, what are you refering to? [17:25] latex --> xml [17:25] ah [17:25] ignore my yes [17:25] With latex, I can actually work with the document. [17:25] like everything else, you can say "it'll be done later" but it wont be :p [17:25] I agree with vapier [17:26] Actually, that's a test of whether it matters, too. :) [17:26] *shrug* [17:26] depends on who you ask [17:27] I don't want to have to host more special content [17:27] i think KingTaco has a point with the "Gentoo has settled on xml as a standard for documentation" [17:27] xml is easy as infra can push it to the www nodes [17:27] it can also be branded the same way most of our other content is [17:28] I don't have an opinion one way or the other --- my question should be read exactly as I wrote it, and I think the answer was "yes" [17:29] the answer is yes [17:29] if you expect the majority of gentoo people to contribute [17:29] most of us know gentoo xml [17:29] the same statement can not be made about latex [17:29] weird observation on the site that had the PMS SVN - a bunch of revisions are missing now - i pulled the git when it was r164, but now the SVN only goes to r129 [17:29] you could also say that lots of ebuilds devs hate xml though.. :p [17:30] robbat2: yes, that was a pretty big concern i had as well [17:30] robbat2, either someone got pissy or svn broke for them too [17:30] I've had problems with svn like that at work [17:30] btw is this meeting offically over or what ? :) [17:31] vapier, PMS was your topic [17:32] i got the feeling that people had peaced out and PMS was happening post [17:32] we can table it [17:32] i think that's best [17:32] I don't think anything is going to get done today [17:32] ok [17:32] any thing else? [17:32] speak no or wait 30 days [17:32] i dont believe so [17:32] going [17:32] going [17:32] gone [17:33] --- KingTaco sets modes [#gentoo-council -m] [17:33] KingTaco, I thought everyone spoke latex. :) [17:33] I wish [17:33] latex is neat [17:34] but gentoo's official docs are all in xml [17:35] (lets convert all of gentoo's docs to latex! ;)) [17:35] you're saying that guidexml is neat? [17:35] no [17:35] good. [17:35] i never said that, or even implied it... [17:36] thus people has to use something that they don't like, just because everyone else is using it? [17:36] ahf: the answer to that is yes [17:36] wtf is ahf ? [17:37] eroyf. [17:37] dont be a tool [17:37] stay in school [17:37] sorry, on my workstation [17:37] and shut your word hole [17:37] okdad. [17:37] the usefulness of guidexml has been discussed on the gentoo-dev mailing list [17:37] yes, it's xml ... yes, many people think xml is the devil [17:38] and developers are enforced to write every technical document in it? [17:38] forced* [17:38] however it's done us a huge service [17:38] indeed it has [17:38] for simple pages [17:38] i agree on that [17:38] ahf: if you want to write your own technical document for your own stuff, have at it [17:38] Gentoo wide things there's a standard [17:38] dont ask infra to host it [17:38] but if it has to be official gentoo documentation it has to be guidexml? [17:39] i'm not making an argument saying PMS needs to be converted to guidexml right now [17:39] i'm saying that it has to be seriously considered and have a good reason for it to not be done [17:39] no. why not finish it then see if you accept it at all [17:39] PMS isnt some random Gentoo document [17:39] then start the converting job if you have to [17:39] i know that. [17:39] fewl [17:39] if i wasnt planning on accepting it, i would have punted spb already [17:40] nod [17:40] i actually need to talk to you about something else but let's take that in query [17:40] that's what she said [17:41] i'll be her for today [17:41] hawt [17:41] read your query god dammit! [17:41] stop yapping [17:42] ahf: YAP! YAP! YAP! [17:43] *meh* [17:43] dostrow, how big is that TV you brought to scale? [17:43] 32" [17:43] bah [17:43] fyi, i shot ciaranm an email to ask if his SVN broken [17:43] KingTaco: por que? [17:43] I just got mine (32") and it isn't as big as I remember [17:44] aha [17:44] it's not his asfaik [17:44] robbat2: I think it moved [17:44] <-- FuzzyRay|work has quit (Remote closed the connection) [17:44] zlin, no, that repo had r164 before, and now it only has r129 [17:45] diff'ed them? [17:45] i think they moved server [17:45] * ferringb just assumed that when svn broke they cut off his access to be annoying :) [17:45] i had to change planetpaludis a lot of times and i think they're on the same server [17:46] ahf, got a newer URL then? [17:46] vapier: btw, I've set you up on this hot new alpha box contingent on you helping with some binutils problems :p [17:46] i don't know [17:46] haven't followed pms discussion lately [17:46] http://svn.repogirl.net/pms [17:47] oh, it's on repogirl now [17:47]