path: root/meeting-logs/20121009.txt

--- Log opened Tue Oct 09 21:09:46 2012
21:09 -!- Irssi: #gentoo-council: Total of 45 nicks [4 ops, 0 halfops, 1 voices, 40 normal]
21:09 -!- Irssi: Join to #gentoo-council was synced in 0 secs
21:09 <   grobian> sorry folks
21:09 <@      ulm> good :)
21:09 <@ Chainsaw> Good, that did it.
21:10 <@ Chainsaw> Good evening grobian.
21:10 <   grobian> nickserv don't give me identify
21:10 <+dberkholz> this is beautifully ironic
21:10 <+dberkholz> anyone got his number?
21:10 <  WilliamH> dberkholz: All of us should have it; he sent it to the alias.
21:10 <+dberkholz> ah, there it is.
21:10 <+dberkholz> i'll shoot him a text
21:10 <   grobian> thanks all for texting me
21:10 -!- mode/#gentoo-council [+o grobian] by ChanServ
21:10 <@ Chainsaw> Excellent. Identified and everything.
21:10 <@  grobian> really crap, me sending a reminder and stuff
21:10 <@ Chainsaw> You are not an imposter.
21:10 <@  grobian> ah
21:11 <@  grobian> it's jsut incredibly slow
21:11 <@  grobian> and me too impatient
21:11 <@      ulm> Chainsaw: sounds unlikely, given his sms reply ;)
21:11 -!- floppym [~quassel@gentoo/developer/floppym] has joined #gentoo-council
21:11 < scarabeus> :D
21:11 < scarabeus> he sent it to the phone numbers
21:11 < scarabeus> but i cant find MY phone now :/
21:11 <@  grobian> ok, do you guys prefer me to quickly put the agenda thing on www.g.o?
21:12 <@ Chainsaw> grobian: ulm linked me to http://article.gmane.org/gmane.linux.gentoo.project/2163
21:12 <@ Chainsaw> grobian: Which works for me.
21:12 <@  grobian> Chainsaw: yeah, that is it
21:12 <+dberkholz> there's the in_iuse thing too
21:12 <+dberkholz> do you want to talk about that?
21:12 <@      ulm> and we really need our own archives back in working condition
21:12 <@  grobian> dberkholz: I shot it ;)
21:13 <+dberkholz> yes sir chair sir
21:13 <@  grobian> dberkholz: you or anyone disagree with that? we can put it on
21:13 -!- radhermit [radhermit@gentoo/developer/radhermit] has joined #gentoo-council
21:14 <@  grobian> ok
21:14 <@  grobian> rollcall
21:14 <@ Chainsaw> I am present and accounted for.
21:14 <@  grobian> (sorry all for me being late)
21:14 -!- alexxy [~alexxy@gentoo/developer/alexxy] has quit [Ping timeout: 272 seconds]
21:14 <@      ulm> \o/
21:14 <  WilliamH> here
21:14 <@  grobian> scarabeus: dberkholz seen you two
21:14 <@  grobian> Betelgeuse: you around too?
21:14 < scarabeus> here
21:15 <+dberkholz> yep.
21:15 <@Betelgeus> grobian: yes
21:15 <@  grobian> ok, do you want me to report myself as being late?
21:15 <@  grobian> vote
21:15 <@      ulm> no
21:15 *** Chainsaw votes no
21:15 < scarabeus> no
21:15 <  WilliamH> no
21:15 <@Betelgeus> yes but no missing marker
21:15 <@ Chainsaw> grobian was responsive on SMS.
21:15 <@ Chainsaw> And promptly appeared. I have done this kind of thing in the past, and not been shot for it.
21:16 <@  grobian> lol
21:16 <@  grobian> ok
21:16 <@  grobian> good, just for the record
21:16 <@  grobian> thanks
21:16 <@  grobian> Ok, I'd like to go on to point 2, the EAPI5 usage in tree
21:16 <@  grobian> are we prepared to vote on that?
21:16 <@ Chainsaw> I would like for that to proceed with immediate effect please.
21:16 <@ Chainsaw> Yes.
21:16 *** Chainsaw votes yes
21:16 < scarabeus> ^_^
21:16 <@Betelgeus> fyi there's no late concept in GLEP 39
21:16 <@  grobian> good, votye for usage of EAPI5 in tree
21:16 *** ulm votes yes
21:16 *** scarabeus is for yes
21:17 *** WilliamH votes yes
21:17 -!- alexxy [~alexxy@gentoo/developer/alexxy] has joined #gentoo-council
21:17 <+dberkholz> sure
21:17 <@Betelgeus>  I don't think a separate vote is required but yes
21:17 <@      ulm> maybe we should clarify that EAPI 5 isn't allowed for stable yet
21:17 <@  grobian> grobian: yes
21:17 <@Betelgeus> ulm: then we should clarify a lot of other QA things as well
21:17 < scarabeus> Betelgeuse: we actually wrote it to the roll call :D
21:17 <@      ulm> only when the Portage version supporting it goes stable
21:17 <@  grobian> ulm: good point
21:18 <@ Chainsaw> ulm: Does repoman not check for that situation already?
21:18 <@      ulm> but it was handled like that in the past, so maybe it's obvious
21:18 <@ Chainsaw> ulm: And if it does not, can that be implemented please?
21:18 <@      ulm> Chainsaw: I don't think it does
21:18 <+dberkholz> that should be a technical check, not a policy
21:18 <@  grobian> ulm: do you want a boundary condition? (e.g. when portage goes stable, or X months after that)
21:19 <@      ulm> grobian: as soon as portage goes stable
21:19 <@  grobian> ulm: got that now
21:19 <@ Chainsaw> The second after, if need be.
21:19 <@  grobian> http://dev.gentoo.org/~grobian/agenda-20121009.txt
21:19 <@ Chainsaw> If repoman prevents any accidents... that should do nicely. If devs want to use --force and break the tree, that is when policy can kick in.
21:19 <@  grobian> "common sense"
21:20 <@  grobian> ok, shall we move onto point 3?
21:20 <@  grobian> Package name specification
21:20 <@  grobian> basically, ulm outlined it pretty clearly, IMO
21:20 <@ Chainsaw> Option C please. The rules were voted in.
21:20 <@      ulm> grobian: you've messed up the indentation :p
21:20 <@ Chainsaw> Portage does not get a free pass here.
21:21 <@  grobian> ulm: yeah, tabs, I need to fix my .muttrc
21:21 <@  grobian> ok
21:21 <@  grobian> let's vote for a, b, c or d
21:21 *** Chainsaw votes C
21:21 <@      ulm> b or d
21:21 <+dberkholz> B
21:21 <+dberkholz> i prefer fixing docs to code.
21:21 <@  grobian> ulm: how do I have to interpret that?
21:21 <@Betelgeus> scarabeus: note for next time
21:21 < scarabeus> b
21:22 <@      ulm> grobian: b then ;)
21:22 <@  grobian> ulm: ok :)
21:22 <@ Chainsaw> Looks like a majority for B then?
21:22 *** Chainsaw can live with that :)
21:22 <@  grobian> WilliamH: your vote?
21:22 *** WilliamH votes b also
21:23 <@      ulm> well, both portage and the tree comply with b already
21:23 <@  grobian> B: ulm, dberkholz, scarabeus, grobian, williamh
21:23 <@  grobian> C: chainsaw
21:23 <@  grobian> so, B wins
21:23 <@Betelgeus> b+c
21:23 <@ Chainsaw> grobian: Not unanimous for a change. This is good. I have had complaints that meetings were getting boring.
21:23 <@  grobian> so, ulm, what did you want to do with d?
21:23 <@Betelgeus> If we only change a future EAPI Portage should comply with the existing ones
21:24 <@  grobian> Betelgeuse: what does that mean?
21:24 <@      ulm> grobian: d would have been my second choice
21:24 <@  grobian> Chainsaw: pardon my faulty spelling
21:24 <@Betelgeus> grobian: or was b meant to be retroactive?
21:25 <@  grobian> b means fit the spec to what portage does, IMO
21:25 <+dberkholz> boring is probably good, because it means most issues were hashed out in advance
21:25 <     _AxS_> ..so to clarify that means change PMS everywhere applicable ?
21:26 <@  grobian> Betelgeuse: I'd prefer if you'd choose one option
21:26 *** ulm understands it in this way
21:26 <@  grobian> _AxS_: I also understand it that way
21:26 <@Betelgeus> grobian: b if Portage has always behaved like that
21:26 <@      ulm> Betelgeuse: it has since 2009 at least
21:26 <@Betelgeus> grobian: I don't think we should fit PMS if Portage has changed in the recent history
21:26 <@      ulm> I haven't checked earlier versions
21:27 <@Betelgeus> ulm: 2009 was when PMS was in effect so they should not go about changing things
21:27 <@  grobian> ok
21:28 <@  grobian> do you want to change your vote then, Betelgeuse?
21:28 <@  grobian> if not, I'd like to finish this topic, and move on to the next
21:29 <@Betelgeus> grobian: I would like to see how others understood the option
21:29 <@  grobian> ok, go ahead
21:29 <@Betelgeus> grobian: b is still fine if we note what I said
21:29 <@      ulm> apply b to all EAPIs
21:29 <+dberkholz> +1
21:30 *** WilliamH is fine with that
21:30 <@  grobian> ok, shall we move on then?
21:30 <@Betelgeus> grobian: yes
21:30 <@  grobian> great
21:30 <@  grobian> the open bugs
21:31 <@  grobian> I think the only one is the one we have on the agenda for a while
21:31 <@  grobian> I'll try to sort it out with jmbsvicetto in prague
21:31 <@  grobian> no guarantees
21:31 < scarabeus> thats what i plan to do as he stays at my place
21:31 <@  grobian> cool]
21:31 < scarabeus> as i said last meeting :-)
21:31 <@  grobian> we will both doo it
21:32 *** grobian updated it
21:32 <@  grobian> ok, open floor then
21:32 *** grobian opens the floor
21:32 <@Betelgeus> Any other people going to Prague?
21:32 <@      ulm> +1
21:32 <@Betelgeus> I booked flights yesterday
21:32 <@  grobian> cool!
21:32 <@Betelgeus> So good we can get drunk and do a meeting
21:32 <@  grobian> yeah, hahahaha
21:32 <@  grobian> ok
21:32 <@  grobian> Open Floor!
21:32 <@  grobian> anyone who wants to raise an issue to the council?
21:33 <@  grobian> I take that as a no
21:33 <  WilliamH> Not really an issue, but a comment. It is going to definitely be interesting to see what happens with udev... Is everyone aware of the debate on lkml?
21:34 <@  grobian> no, would you like to share a summary with us?
21:34 <@ Chainsaw> Go ahead WilliamH. Those last two links I sent you should be rather informative.
21:34 <  WilliamH> Basically the kernel guys are looking into taking over some or maybe all of the udev functions...
21:34 <@ Chainsaw> (The initial Linus posting and his response to Kay Sievers)
21:34 <@ Chainsaw> Perhaps Al Viro's take.
21:34 <  WilliamH> Chainsaw: can you post the links here again?
21:35 <@ Chainsaw> WilliamH: That was on a different computer I'm afraid.
21:35 <@  grobian> interesting, so that means udev will be just kernel built-in?
21:35 <  WilliamH> ok folks give me a second to find them...
21:35 <  WilliamH> grobian: I'm not really sure yet.
21:35 <  WilliamH> grobian: but changes there are definitely happening.
21:36 <@  grobian> while you're searching
21:36 <@  grobian> one issue people
21:36 <     _AxS_> Since in_iuse was mentioned earlier -- i believe there was quasi-consensus on that one that the best way to deal with it will be in a future eapi; is that the take Council has on it too?
21:36 <@  grobian> next meeting, 20:00 UTC again?
21:36 <@  grobian> (iso 19:00)
21:36 <+dberkholz> again?
21:36 <@  grobian> _AxS_: I would vote yes
21:36 *** ulm doens't care if it's 19 or 20 UTC
21:36 <@  grobian> dberkholz: daylight savings thing here in europe
21:37 <     _AxS_> grobian: ok so work will be done and it can get added to the agenda for eapi=6 whenever that rolls up.
21:37 <@  grobian> _AxS_: from my point of view, yes.  That ferringb said/suggested
21:37 <@  grobian> ok, next meeting will then be 13 November 2012, 20:00 UTC
21:37 <@Betelgeus> the earlier the better for me
21:38 <  WilliamH> https://lkml.org/lkml/2012/10/2/303
21:38 <  WilliamH> That's Linus' original post, and the other things follow it in that thread
21:38 <@  grobian> Betelgeuse: I'd like it too
21:38 <     _AxS_> Rumour has it that infra will be working on rolling out the git tree; does Council know anything about that?
21:38 <  ferringb> _AxS_: wrong forum for asking
21:39 <  ferringb> aka, ask infra, not council
21:39 *** ferringb sent emails detailing current status of it, and areas people nee dto step up (hooks in particular, a helping hand is needed for)
21:39 <     _AxS_> ahok.  wasn't sure if it'd be a Council thing to freeze the tree or whatnot while the conversion happens
21:39 <@ Chainsaw> _AxS_: The council doesn't call the shots on this.
21:39 <  ferringb> we'll sort that when it comes; if a freeze is necessary, it'll be sub 8 hours
21:39 <@  grobian> not a council thing, imo
21:40 <@ Chainsaw> _AxS_: Nothing moves until ferringb says it does.
21:40 <  WilliamH> What we are going to do is not stabilize udev-18x for a wwhile and monitor the upstream situation.
21:40 <  ferringb> the plan involves no freeze however, beyond an hour outage or so
21:40 <  ferringb> Chainsaw: robin moreso.  I'm just his minion
21:40 <  WilliamH> There is already a commit in the kernel to load firmware directly.
21:40 <     _AxS_> wonderful!  (Can this be left in the minutes?)
21:40 <  WilliamH> It looks like that will hit in 3.7
21:40 <@  grobian> Ok, let's end the meeting here, then you can continue here whatever
21:40 <  ferringb> with that said
21:41 <  ferringb> council commentary- subjective commentary- on the rough proposed unified dependencies would be useful.
21:41 <  ferringb> no, I'm not asking for approval.  I'm after a basic headcount of who says no, and or the potential of a dev vote if folks are particularly divided and no clear majority
21:42 <  ferringb> (how's that for chucking a grenade into your quiet meeting? :)
21:42 <@  grobian> I for one would like to be a bit more informed about the issue because I saying anything about it
21:42 <  ferringb> grobian: what do you need to be better informed?
21:42 *** WilliamH agrees  with grobian 
21:42 <  ferringb> glep needs updating, which is on the todo
21:42 <@  grobian> ferringb: like what you're asking me
21:42 <@Betelgeus> ferringb: I like the exherbo approach
21:42 <@  grobian> ferringb: I don't think the git migration thing should EVER be a council topic
21:42 <  ferringb> I'm just looking to see how to get the details/info to y'all *clearly*, w/ less of the trolling on the ml gumming the info up
21:42 <@  grobian> because it simply needs to be done
21:42 <@  grobian> not decided upon
21:43 <@  grobian> the plan is pretty much laid out clearly, IMO
21:43 <  ferringb> grobian: err.  I was asking about unified dependencies.
21:43 <@  grobian> here, see
21:43 <@  grobian> lol
21:43 <  ferringb> git tranition isn't a council topic because y'all aren't doing the work, so nothing to talk about. :)
21:43 <@  grobian> I don't even know what you're talking about
21:43 < scarabeus> ferringb: I like the unified deps idea, but I didnt get my ass to reply there due to all that noise to real stuff ratio
21:43 <@  grobian> please just bring it up on -project, with pointers and all
21:43 <@  grobian> I get lost in the flamewars sometimes
21:44 <@Betelgeus> sounds like a -dev topic
21:44 <@  grobian> we can add it to the next agenda
21:44 <@  grobian> in fact, please do
21:44 <  WilliamH> Yeah me too. I like the concept, but there was so much in that thread it was difficult to follow.
21:44 <  ferringb> ehh
21:44 <@ Chainsaw> I don't think it needs to be on the agenda.
21:44 <  ferringb> grobian: tbh, I think it's better I identify exactly how to make sure y'all know the details/bits involved here, then adding it to the agenda
21:45 <@ Chainsaw> If you'd like a private briefing by the stakeholders sent to council@
21:45 <  ferringb> I don't want discussion w/out understanding in full
21:45 <@ Chainsaw> I think that is more feasible.
21:45 <  ferringb> bluntly, the -dev ml already had enough of that
21:45 <@  grobian> I can read, if I know where to find it, and what's the problem
21:45 <@ Chainsaw> Even if it's just ferringb sending you his earlier write-up.
21:45 <  ferringb> grobian: http://dev.gentoo.org/~ferringb/unified-dependencies/
21:45 <@  grobian> I don't like exercises like that lengthy discussion about the sub-slot bogus
21:46 <  ferringb> grobian: http://dev.gentoo.org/~ferringb/unified-dependencies/extensible_dependencies.html <-- glep, http://dev.gentoo.org/~ferringb/unified-dependencies/examples/herds/ <-- herd level view of how it would impact deps
21:46 <     _AxS_> there's some good stuff that came out of the ML arguments too -- like, ome of the primary differences between *DEPEND vars and DEPENDENCIES (that being "authoritative" specification for each phase, i think is the wording?)
21:46 <  ferringb> re: exherbo labels, I addressed the similarity between the two in an email, and via analysis http://dev.gentoo.org/~ferringb/unified-dependencies/labels/
21:47 <  ferringb> authoratitive I need to incorporate fully into the glep, since that's implicit, but not explicitly stated
21:47 <  ferringb> Betelgeuse: presume you're pretty well caught up on the topic, sans potentially my "this is why labels isn't worth it for us" arguments?
21:47 <     _AxS_> ..and possibly important enough to be dealt with/decided upon separately
21:47 <  ferringb> _AxS_: can't be, unfortunately, since going authoritative w/out this matching change to metadata makes devs lives worse
21:48 <     _AxS_> true
21:48 <  ferringb> when is the next meeting?  date, not time
21:49 <@ Chainsaw> The second Tuesday in November, presumably?
21:49 <  ferringb> ok
21:49 <@  grobian> ferringb: please chuck all those links in a mail on -project, preferably in reply to the next call for agenda items mail
21:49 <     _AxS_> but it defines the decision of "yes we need a change" vs "no we dont" , separately from the change itself.
21:49 <  ferringb> will sort the glep, and if necessary, will cc each of your asses (I'm not naming names, but... grobian) to make sure y'all see it :P
21:49 <@  grobian> ferringb: 13 November 2012, 20:00 UTC
21:49 <     _AxS_> <grobian> ok, next meeting will then be 13 November 2012, 20:00 UTC  <-- that?
21:49 <  ferringb> _AxS_: yeah, I'm blind
21:49 <@  grobian> ferringb: and agenda call is sent 30th of october
21:50 <  ferringb> yep
21:50 <@  grobian> agenda is sent out on 6th of november
21:50 <@  grobian> ok, @council: I'd like to close this meeting
21:50 <  ferringb> may put it on y'alls agenda.  not looking for necessarily approval (would be nice, but les be realistic), just discussion
21:50 <  WilliamH> Before anyone takes off, did my link and comment about the kernel commit get lost in the chatter?
21:50 <     _AxS_> WilliamH: <WilliamH> https://lkml.org/lkml/2012/10/2/303   <-- that one?
21:50 <@  grobian> WilliamH: you want it in the summary?
21:51 <  WilliamH> grobian: I'm not sure if it needs to be there or not, It isn't really an issue we decide anything on here, just something to track.
21:51 <@  grobian> WilliamH: w/e http://dev.gentoo.org/~grobian/agenda-20121009.txt
21:52 <@  grobian> ok. thank you all for being productive
21:52 <@  grobian> I'll send out the summary to @council soon
21:52 <     _AxS_> thanks for chairing, grobian !
21:52 < scarabeus> grobian: ack on the summary and chairing
21:52 < scarabeus> s/ack/thanks/
21:52 <@  grobian> and sorry once again for my problematic arrival
21:53 <  WilliamH> grobian: ok, that looks good.
21:54 <  ferringb> WilliamH: fun thread btw
21:54 -!- grobian changed the topic of #gentoo-council to: Next meeting: 2012-11-13 20:00 UTC | http://www.gentoo.org/proj/en/council/utctolocal.html?time=2000 |