20:02:53 #startmeeting tc 20:02:54 Meeting started Tue Dec 18 20:02:53 2012 UTC. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:02:55 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:02:57 The meeting name has been set to 'tc' 20:02:59 Agenda for today is: 20:03:03 #link http://wiki.openstack.org/Governance/TechnicalCommittee 20:03:12 #topic Update on the "Future of Incubation / core" joint committee 20:03:17 markmc: quick update ? 20:03:39 sure 20:03:50 had quite a productive meeting actually 20:03:58 got more into the incubation process 20:04:07 how projects are "trained" to follow our processes 20:04:25 how CI starts taking more of an interest in helping incubating projects 20:04:39 also raised an interesting question about trademark analysis 20:04:42 how incubating projects being part of the devstack gate should become more of a hard requirement 20:04:53 on project codenames 20:04:56 indeed, trademark 20:05:12 how the legal committee could provide kind of consultative help when choosing new project names 20:05:32 how part of a project graduating is a qualitative analysis 20:05:44 i.e. we should have some sort of quality level we require 20:05:51 and how we go about making that call 20:06:09 and then we started getting into the whole "what does core mean" 20:06:23 jbryce had a nice summary of core vs "part of the release" 20:06:41 that core is a label the board can attach to projects in the integrated release 20:06:53 and then we were out of time 20:07:02 we have one more this Thursday. 20:07:06 we'll get more into the "what does core mean" on thursday 20:07:10 20:07:20 Thanks! 20:07:26 questions before we move on ? 20:07:43 some of that seems a little tangential, but good discussion anyway, i guess 20:08:06 thanks for the update 20:08:09 markmc: were there notes? 20:08:19 markmc: was there another invite sent out for this Thurs? 20:08:25 annegentle_, yes to both 20:08:27 one sec 20:08:31 annegentle_: https://etherpad.openstack.org/IncUp 20:08:31 actually, yes and no 20:08:49 ttx, notes moved from etherpad to http://wiki.openstack.org/Governance/Foundation/IncubationUpdate2013 20:08:56 oh, great 20:09:08 #link http://wiki.openstack.org/Governance/Foundation/IncubationUpdate2013 20:09:14 any other question ? 20:09:16 no invite for this week yet 20:09:41 I'll go check to see why the invite didn't go out 20:09:48 #topic Discussion: Potential Grizzly schedule changes to match probable summit date 20:09:57 The Foundation events staff is busy securing a location for the next summit... 20:10:05 On the dates side, 40% chances it's Apr 15-18, 60% chances it's Apr 29 - May 2 20:10:13 The issue for us is that it's a week before or later than the "ideal" week 20:10:29 (ideal week being the one which leaves two full weeks between release week and summit week to prepare, like we did in Folsom) 20:10:39 In the case it's one week before, no big deal, we'll just have less time to prepare for summit 20:10:49 But if it's the week after, we have 3 full weeks between release week and summit week. 20:10:57 I think that's too late in the cycle to have good design discussions at the summit. 20:11:07 People would either get started way earlier in implementation, or lose too much time waiting for summit sessions to happen. 20:11:19 ideally I would like to limit the amount of time between Grizzly feature freeze and H summit to 8-9 weeks max to avoid that 20:11:39 If we don't change the release schedule (and the summit happens Apr 29-May 2, that will be 10 weeks for Grizzly. 20:11:46 (in comparison, we had 7 weeks in Essex and 9 weeks in Folsom) 20:12:04 So I would like us to look into potential adaptations, so that if we need any, we can push them during the holiday period rather than wait for next year's meetings 20:12:17 +1 for adjusting to fit the summit 20:12:26 Those options are mostly around pushing back final Grizzly release date, in the case Apr29-May 2 ends up being chosen 20:12:43 Note that it's probably a one-off thing, next times we should know the date well before the release schedule is set up 20:12:49 sounds good to me ... and push the freeze out a week too? 20:12:50 I have a PDF showing various options for this scenario: 20:12:54 #link http://ubuntuone.com/25a7M1o3jyNAjBUTOfZkHx 20:12:58 I don't love that are schedule is driven by event planning considerations, in general 20:13:07 but no real issue with pushing it out a week or two 20:13:12 if planned well in advance 20:13:20 Option A is the "no change" option, just accept having three full weeks between release week and summit week, and 10 weeks between feature freeze and summit 20:13:37 Option B1 is pushing back G3 and release one week away. Main issue with this one is that I'm skiing that week. 20:13:49 (so someone else would have to handle the final push toward feature freeze and publish that milestone) 20:13:58 Option B2 is the same as B1, but taking the opportunity to push back G2 as well 20:14:09 Option C is solving the skiing issue by pushing G3 *two* weeks off and have "only" 5 weeks of RC period between G3 and release (like we did for Essex) 20:14:21 (means 8 weeks between feature freeze and summit and also pushing back G2 one week off) 20:14:29 Finally, option D pushes back *both* G3 and release two weeks off. 20:14:30 ttx: and others is it your instinct that people would appreciate a Jan 17th milestone based on how much they're trying to get into G2? 20:14:54 annegentle_: that's one nice side-effect of B2, C and D 20:15:09 (option D leaves only one full week between release week and summit week, and I know we all appreciated having two full weeks after Folsom release) 20:15:16 Thoughts ? 20:15:22 yeah, I'm sure there'd be general happiness with pushing g2 out a week :) 20:15:24 My slight preference would be to go with option C with a pretty anal feature freeze 20:15:24 B2 or C seem good to me 20:15:36 +1 to annegentle_ 20:15:46 so the shortebed Rc period does not have adverse effects 20:15:52 shortened* 20:16:14 B2 or C seem fine ... but it's C of those 2 unless someone wants to drive the release of grizzly-3, right? 20:16:21 did much happen in the last week of rc in folsom? 20:16:22 C it is! 20:16:22 russellb: yes 20:16:34 bcwaldon: +1 :) 20:16:37 markmc: not so much 20:17:06 markmc: I think if we are more careful with the feature freeze and landing all stuff by G3... we have less of an issue with 5 weeks of RC 20:17:32 6 weeks was kinda nice but we also had a lot of Feature Freeze abuse after Folsom-3 20:17:39 yeah, C it is - don't imagine anyone too eager to be ttx for grizzly-3 :) 20:18:16 OK, so we don't change anything if summit is the week of Apr 15 20:18:28 And we use option C if the summit ends up on the week of Apr 29 20:18:38 hm, really? what about shorten RC to 5 weeks if summit moves up? 20:18:42 if we think 5 is enough? 20:18:47 2 weeks before summit is really nice ... 20:18:54 Ah. 20:19:21 That's one option yes. I'd like to only change the release date for a good reason though 20:19:33 ttx: when would we *know* WRT the summit dates? 20:19:49 so.. are we more attached to keeping release date or to two full weeks of summit prep ? 20:20:01 jgriffith: they promised me in the next 72 hours 20:20:12 * jgriffith votes 2 weeks of summit prep 20:20:13 jgriffith: but last week they promised the final decision for today 20:20:16 2 weeks of summit prep personally 20:20:19 ttx: hehe 20:20:34 russellb: that means releasing on March 28 20:20:54 and rewriting all those slides that say "in April" 20:21:08 who ever got in trouble for finishing a project early 20:21:14 I'm fine with it, I like to confuse everyone 20:21:26 vishy: opinion ? 20:21:35 how about we aim for a very quite last week of rc 20:21:47 quiet, dammit 20:21:49 danwent: same -- you're the most RC-busy projects 20:22:03 I don't like shortening the schedule for sure 20:22:19 markmc: +1 20:22:35 Let's aim for a quiet release week if summit is on Apr 15 20:22:39 extending g2 1 week and g3 1 week seems fine 20:22:50 which option was that? 20:23:07 vishy: oh, we are exploring the case if the sumit is on Apr 15 now 20:23:13 o/ 20:23:27 We settled for option C in the case the summit is on Apr 29 20:23:37 Apr 15 i say no change 20:23:37 vishy: +1 20:24:01 sorry i was behind :) 20:24:08 vishy: OK -- and we'll try to make a calm release week to have a bit more time to prepare summit 20:24:27 * annegentle_ lights calm candles 20:24:41 #info If summit is on Apr 15 -- no change in schedule 20:25:24 any info you can leak on location? :) 20:25:38 #info If summit is on Apr 29 -- push back G2 one week, G3 two weeks, release one week. 20:26:00 russellb: not yet :) 20:26:12 I heard it's in europe 20:26:19 it's in the US for sure 20:26:24 (the April one) 20:26:24 heh :) 20:26:25 ttx: also, they're working on getting the I summit scheduled/booked so that we know where it is by the H summit, right? 20:26:37 * markmc tries to get vicious rumours started :) 20:26:40 mordred: yes. It should be international 20:26:51 and location secured before the H summit 20:26:56 GREAT 20:26:59 #topic Ongoing discussion: Distro & Python 2.6/3.x support policy 20:27:00 that will be quite lovely 20:27:03 * vishy looks around. 20:27:06 mtaylor: You said you would push a clear motion to the ML to reboot the discussion 20:27:12 someone is starting my rumors! 20:27:15 mtaylor: looks like it will be for next year meeting now :) 20:27:19 ttx: I did say that didn't I? 20:27:25 * mordred accepts chastisement 20:27:52 Do you have a teaser version of it we could talk about today ? 20:28:04 vishy, from here on in, I'm gonna think of you as Sid Vishy 20:28:06 it was basically the summary from the last thread 20:28:17 lol 20:28:52 ttx: "we should continue dev focus for the latest ubuntu, but ensure that we don't do anything that fundamentally breaks the latest rhel" 20:28:54 Python 4 three-eva ? 20:29:11 dev focus on latest ubuntu and fedora :) 20:29:31 also, continue to support 2.6.x until it's blocking someone with patches getting 3.x support done 20:29:35 ? 20:29:35 markmc: for all intents and purposes for dev, that's probably the same 20:29:47 markmc: I don't think we need to be that explicit 20:29:57 markmc: I don't think we can drop 2.6 until there is a rhel that has 2.7 20:30:07 markmc: based on the prior statement 20:30:32 and once there is, we'll have an ubuntu lts and a rhel with 2.7 - so there wouldnt' be a compelling reason to keep 2.6 around 20:30:50 heh 20:30:57 * markmc avoids repeating the whole thread again 20:30:59 and we know for a fact that once we start trying to get 3.x patches done, we will immediately run in to 2.6 issues 20:31:21 but, for now at least, I think my statement above doesn't really change anything for anybody, right? 20:31:42 (it's clarification on the current policy, which is "we target latest ubuntu") 20:32:00 not until there's a rhel with 2.7 or someone sooner than that with 3.x patches 20:32:09 and then the whole discussion would start again :) 20:32:15 markmc: I'm open to that 20:32:21 mordred: ok, you can turn that into a proper motion to be discussed on the ML..; and we can decide on it at the next meeting 20:32:23 "we target latest ubuntu/fedora" 20:32:35 markmc: ok. btw ... 20:32:53 markmc: since we're expressing the current justification for supporting 2.6 in terms of liking RHEL... 20:33:12 markmc: you think there's any chance someone at redhat might be interested in helping port our jenkins_slave puppet template to be able to run on a rhel host? 20:33:29 because that's our current blocker in running our 2.6 tests on rhel instead of ubuntu oneiric 20:33:34 mordred, yes, definitely 20:33:37 markmc: great! 20:33:48 check it out - people helping each other :) 20:33:51 mordred: anything more on the subject ? 20:33:57 ttx: nope 20:34:12 ttx: when markmc and I agree on things, I consider the subject closed 20:34:13 questions ? mordred WILL push a thread on the ML for further discussion anyway 20:34:21 right ? 20:34:28 * mordred whistles 20:34:32 #topic Skipping next weeks meetings 20:34:40 So... next Tuesdays happen to be Christmas and New Year's day. 20:34:49 My suggestion would be to hold our next TC meeting on January 8. 20:35:05 ttx: +1 - that's the next keystone meeting as well 20:35:05 Sounds good ? 20:35:09 ++ 20:35:14 school's out! 20:35:21 * jgriffith realizes he needs to go shopping 20:35:25 we'll probably do the same fate to the release status meeting in half an hour 20:35:36 #topic Open discussion 20:35:42 Anything else anyone ? 20:37:03 Alrighty then. 23 minutes recess 20:37:08 #endmeeting