| *** mtreinish has quit IRC | 00:07 | |
| *** mtreinish has joined #openstack-tc | 00:08 | |
| *** lbragstad has quit IRC | 00:12 | |
| *** lbragstad has joined #openstack-tc | 00:14 | |
| *** harlowja has quit IRC | 00:34 | |
| *** annabelleB has quit IRC | 00:39 | |
| *** mriedem_lawnboy is now known as mriedem | 01:04 | |
| *** mriedem has quit IRC | 01:10 | |
| fungi | zaneb: i have never agreed more strongly with anything in my life | 01:42 |
|---|---|---|
| *** ricolin has joined #openstack-tc | 01:58 | |
| *** annabelleB has joined #openstack-tc | 02:10 | |
| *** jamesmcarthur has joined #openstack-tc | 02:23 | |
| *** jamesmcarthur has quit IRC | 02:27 | |
| *** jamesmcarthur has joined #openstack-tc | 02:32 | |
| *** jamesmcarthur has quit IRC | 02:36 | |
| *** annabelleB has quit IRC | 03:34 | |
| *** jaypipes has quit IRC | 04:04 | |
| *** jaypipes has joined #openstack-tc | 04:04 | |
| *** Bhujay has joined #openstack-tc | 04:18 | |
| *** jamesmcarthur has joined #openstack-tc | 04:46 | |
| *** jamesmcarthur has quit IRC | 04:50 | |
| *** Bhujay has quit IRC | 06:20 | |
| *** e0ne has joined #openstack-tc | 06:50 | |
| *** e0ne has quit IRC | 06:52 | |
| *** jamesmcarthur has joined #openstack-tc | 07:34 | |
| *** jamesmcarthur has quit IRC | 07:38 | |
| *** Bhujay has joined #openstack-tc | 07:41 | |
| *** jpich has joined #openstack-tc | 07:41 | |
| *** Bhujay has quit IRC | 08:17 | |
| ttx | Heh, funny that they consider they are fully-elected, while deciding on a make-up of "represented groups" | 08:20 |
| ttx | Overall, ODL's system sounds like a great way to ensure politics is present in every technical discussion | 08:22 |
| ttx | While we are trying to separate those as much as we can | 08:22 |
| ttx | mnaser: thx for the pointer | 08:23 |
| *** e0ne has joined #openstack-tc | 09:01 | |
| *** dtantsur|afk is now known as dtantsur | 09:07 | |
| -openstackstatus- NOTICE: Jobs using devstack-gate (legacy devstack jobs) have been failing due to an ara update. We use now a newer ansible version, it's safe to recheck if you see "ImportError: No module named manager" in the logs. | 09:58 | |
| *** cdent has joined #openstack-tc | 10:21 | |
| *** Bhujay has joined #openstack-tc | 10:21 | |
| *** Bhujay has quit IRC | 10:26 | |
| cdent | ttx i add many words and a bit of historical perspective on https://review.openstack.org/#/c/598380/ | 10:47 |
| cdent | there are quite a few people who pretty badly want a planned deadline | 10:48 |
| *** e0ne has quit IRC | 11:44 | |
| *** ricolin has quit IRC | 11:48 | |
| *** tonyb has quit IRC | 12:02 | |
| *** tonyb has joined #openstack-tc | 12:19 | |
| dhellmann | cdent : to be clear, I'm asking for a *timeline* not a *deadline* | 12:39 |
| *** jaypipes is now known as leakypipes | 12:44 | |
| cdent | dhellmann: I know. But there are people who have asked for a deadline, so I'm trying to represent that. | 12:45 |
| dhellmann | ok, that's fine | 12:46 |
| dhellmann | just making sure I'm being clear, since melwitt also seemed to think I was asking for a deadline | 12:46 |
| cdent | dhellmann: it's difficult, to say the least, to balance the many inputs | 12:46 |
| cdent | I've tried to be fair and not let my own position overly impact the infra and governance gerrit changes while still reflecting the goals people have expressed | 12:48 |
| * dhellmann wonders how long it will be before someone remembers to teach stackalytics about stein | 12:48 | |
| dhellmann | yeah, that's another reason I want to see a plan written down and agreed to -- some of the folks involve seem reluctant to hold a position strongly | 12:49 |
| dhellmann | or at least strongly and publicly | 12:49 |
| cdent | As far as I can tell, when I express my position strongly and publicly, that makes other people entrench, so I'm reluctant to do so, often. I wonder if other people have the same concern. | 12:50 |
| * cdent has to disappear in ten minutes | 12:50 | |
| *** cdent has quit IRC | 12:55 | |
| *** e0ne has joined #openstack-tc | 12:55 | |
| *** EmilienM is now known as EvilienM | 12:57 | |
| *** tosky has joined #openstack-tc | 13:02 | |
| *** lbragstad is now known as elbragstad | 13:37 | |
| zaneb | ttx: a few countries have had constitutions like ODL's system, and they pretty regularly seem to end up having coups, if not outright civil wars | 14:02 |
| leakypipes | zaneb: ODL or OpenStack? | 14:05 |
| zaneb | ODL | 14:05 |
| zaneb | leakypipes: https://slides.com/dfarrell07/appointed-to-elected-gov | 14:06 |
| ttx | leakypipes: you'll love this model | 14:07 |
| ttx | building vendor fights into the governance model :) | 14:08 |
| *** jamesmcarthur has joined #openstack-tc | 14:08 | |
| ttx | zaneb: the other weird thing is that they insist that candidates should be regular contributors. So they take good people and force them to wear company hats in governance discussions | 14:09 |
| zaneb | yeah, in other circumstances that would have been a _good_ move | 14:10 |
| zaneb | I liked that they're trying to encourage the people doing the work to be elected, instead of managers | 14:10 |
| ttx | I think it's a like they know what a good model looks like, but are set up with weird initial constraints | 14:11 |
| zaneb | yes, I imagine this was the compromise needed to get those companies to give up autocratic control | 14:11 |
| ttx | do what you want, but we need "represented groups". | 14:12 |
| leakypipes | zaneb: I agree with his Tips slides. However, I think he's built an overly complex and bureaucratic "electioneering" system, IMHO. | 14:13 |
| zaneb | I'm sure they're doing the best they can with what they have, but I'm really glad to be a part of _this_ community <3 | 14:13 |
| *** mriedem has joined #openstack-tc | 14:15 | |
| openstackgerrit | Merged openstack/governance master: make the board repo list file format consistent with others https://review.openstack.org/598350 | 14:17 |
| *** cdent has joined #openstack-tc | 14:26 | |
| *** jaosorior has quit IRC | 14:43 | |
| *** annabelleB_ has joined #openstack-tc | 15:08 | |
| *** annabelleB_ has quit IRC | 15:08 | |
| *** e0ne has quit IRC | 15:14 | |
| *** ricolin has joined #openstack-tc | 15:37 | |
| *** e0ne has joined #openstack-tc | 15:45 | |
| *** annabelleB_ has joined #openstack-tc | 15:50 | |
| *** e0ne has quit IRC | 15:56 | |
| *** annabelleB_ is now known as annabelleB | 15:57 | |
| *** jpich has quit IRC | 15:59 | |
| *** openstackgerrit has quit IRC | 16:06 | |
| mnaser | fwiw: I think the TC is stepping in with regards to the placement topic n | 16:10 |
| mnaser | Because it looks like we’re overriding the decision of the PTL | 16:10 |
| mnaser | so my question is are we being asked to step in and intervene and vote something against the will of the PTL of this project, because it looks like we’re doing that in this case. | 16:11 |
| smcginnis | Well, counter argument - we're allowing the PTL of one project to dictate the actions of another project. A bit chicken and egg though. | 16:12 |
| mnaser | I still think we should move with the technical deliverable change and I would like to have a conversation with all those parties at Denver to help Nova vocalize what they think the best way to go about it is, and then moderate the conversation between the two teams | 16:12 |
| mnaser | If the two teams can’t come to an agreement, we step in based on the information provided | 16:12 |
| mnaser | It would circle back to the things dhellmann asked about timeline and steps needed | 16:13 |
| zaneb | mnaser: the thing we're voting on *is* the technical deliverable change | 16:13 |
| mnaser | If melwitt has a very rational reason that she’s deciding X results need to be met for the split to happen, then I think we should listen to her and the Nova team | 16:14 |
| mnaser | Well the deliverable change shouldn’t have anything about timeline in it then. | 16:14 |
| smcginnis | The splits not going to happen if things aren't in place. | 16:14 |
| zaneb | but I do think we need to have an in-person discussion with the Nova/placement team at the PTG | 16:14 |
| mnaser | And they are going in place slowly | 16:14 |
| smcginnis | Whether they say they want it to happen next week or next year. | 16:14 |
| mnaser | The repo has been created | 16:15 |
| mnaser | That’s why I’m proposing to do an actual sit down | 16:15 |
| smcginnis | No ones going to say today's the day and pull the rug out. | 16:15 |
| persia | Semantics are tricky. If there are two teams, then there are two teams. If there is one team, tehre is one team. Talk of two teams being one team needs more thought. | 16:15 |
| mnaser | Hoping that both parties are present and we talk it out openly and honestly so both teams can come to terms | 16:15 |
| mnaser | (taking off soon so might fall off but be back in a bit) | 16:16 |
| mnaser | smcginnis: i worry that pulling the rug is what might happen. That’s why we need to | 16:16 |
| mnaser | Be clear about the technical challenges early. | 16:16 |
| mnaser | Come to agreement on those | 16:16 |
| smcginnis | I'm a mildly shoked by that concern. | 16:17 |
| mnaser | Let’s be honest: given how all of this has went down | 16:17 |
| mnaser | It hasn’t been good for either “sides” and I put sides in quotes. | 16:17 |
| mnaser | So again, we move with adding a deliverable and setup a time to sit down with both teams and help them | 16:18 |
| mnaser | Come to an agreement | 16:18 |
| mnaser | And if we feel that one side is being too far off, we can talk then but I don’t think any of that will happen and we’ll have a criteria to say this is the right time to do it | 16:18 |
| mnaser | Because I don’t think the criteria is time but *actual* technical solution of the split | 16:19 |
| zaneb | mnaser: to be fair, I thought there was an agreement to for an agreed timeline for the split, but the comment from melwitt appears to indicate that the Nova team is not willing to compromise even the tiniest bit to throw the placement folks a bone | 16:20 |
| *** mriedem is now known as mriedem_away | 16:21 | |
| zaneb | the other interpretation is that melwitt's comment is a minor nitpick over wording of the commit message | 16:21 |
| melwitt | I don't think it's fair at all to accuse me of not compromising through this whole situation | 16:21 |
| mnaser | I think the team starting to split things off is a step from the Nova team that was TC blessed. | 16:26 |
| mnaser | Err PTL blessed | 16:26 |
| zaneb | melwitt: ok, but I'm feeling very blindsided here because it was my understanding that a compromise had been agreed, but apparently it had not and the official position of the Nova project does not appear to have budged at all from the pre-compromise position | 16:27 |
| dansmith | zaneb: seriously? | 16:28 |
| zaneb | alternatively, it's just a disagreement over whether 'expected' means 'carved in stone' | 16:28 |
| melwitt | zaneb: I think I was very clear in my ML posts. I said that I thought the governance switchover should be done after we get critical work items done with nova/placement teams as one group | 16:28 |
| melwitt | I never said or agreed to "we expect to switch governance at chosen milestone s-2" | 16:29 |
| zaneb | my understanding of the existence of the compromise agreement was based on IRC discussions | 16:29 |
| zaneb | it appears that those discussion were not representative of the opinion of the whole group | 16:29 |
| melwitt | my lack of comment was not to be taken as agreement. this whole situation has been extremely stressful for me and I'm having a hard time keeping up with the constant pressuring, despite all of the time I've taken to write detailed emails about my position | 16:30 |
| dhellmann | I think I've reached the point where I need something more definitive than another mailing list thread or IRC conversation where not everyone is participating in order to understand what the agreement actually is. I don't know if that means a spec, or something else. That's why I'm asking for "a plan" to be written down, so we get get everyone to sign off and stop going back and forth about the details. | 16:36 |
| melwitt | yeah, I think that's a good idea. we usually do things like that via etherpad | 16:40 |
| melwitt | I'm not sure how we could do it in gerrit, we haven't used a spec in that way before | 16:41 |
| scas | etherpad was my armchair thought. specs tend to introduce a lot of latency due to being embedded in the code review process | 16:42 |
| scas | maybe a spec somewhere as a means to codify the resolution itself, if one were so inclined, but that's more work | 16:44 |
| zaneb | I guess the advantage of Gerrit is we can see exactly who has and has not reviewed it, and thus avoid the non-representative-subset-in-the-discussion problem | 16:45 |
| scas | from my perspective, i interpreted things as "this was always meant to happen". the opposition, to me, seemed to be more of a "not right now" approach, and not a "thou shalt not" one. but, that's my take. without things being defined as verbosely as, say, constitutional law, interpretation is everything | 16:48 |
| scas | i'm mostly just following along to see how/where i need to adapt my thing | 16:49 |
| melwitt | that's correct. we are in the middle of challenging integration work (first use of a placement feature) where being one group helps us to get the work done | 16:51 |
| melwitt | (we're integrating nested resource providers into nova to enable vGPU enhancements, for one) | 16:52 |
| melwitt | I have been saying, let's extract the code, make sure it works with it being its own package, upgrades work, then let's agree on some critical work items that would be best worked through as one group where we can say, when we complete those, we split the governance | 16:54 |
| dhellmann | an etherpad works for me, as long as it includes a list of all of the folks who should be agreeing and then an explicit ack from each of them somehow | 16:57 |
| dhellmann | that's easier with gerrit, as zaneb points out, but we can make an etherpad work too | 16:58 |
| cdent | I left a comment on the patch asking what people want it to say, so we can at least move that part forward. | 17:01 |
| cdent | I wish that being earnest wasn't such a firestarter. | 17:02 |
| scas | i see the advantage of gerrit, which is why i followed up with an out-loud thought about a spec to ratify the resolution. it's more work for all, but it preserves the spirit of democracy | 17:05 |
| melwitt | I agree on being earnest. I've been earnest that I want to prevent disruption to our ability to deliver enhancements for users and extracting code obviously disrupts it. but since there's never a "good" time to extract code, I've said OK, let's extract it now as a compromise to make some progress. and I say, let's agree on a set of work items to get done before we split governance and not maximally disrupt our ability to deliver for | 17:08 |
| melwitt | users. and then I'm told I'm unwilling to compromise and receive lots of fire upon me | 17:08 |
| cdent | melwitt: I left a pretty direct question for you in my latest comment. I don't understand how changing governance makes us any less "one group". I'm not denying that people feel that way, I'm saying "I don't understad" | 17:08 |
| cdent | from my standpoint I don't think you're unwilling to compromise. I think people watching us are jumping to bigger conclusions than are really there | 17:09 |
| scas | in the information age, perceptions are larger than reality | 17:09 |
| cdent | you're right: the extraction (already begun) is the bigger disruption to being able to deliver than any other potential change | 17:09 |
| melwitt | cdent: you have stated yourself that you want to split groups so that placement can have its own priorities, independent from nova. and I'm saying, let's keep our priorities the same for at least some set of critical work items where close cooperation is needed | 17:10 |
| cdent | that's long term melwitt, and for the most part that is so _nova_ can have its own priorties. I've felt for the the last year that nova has been too placement (server-side) focused | 17:10 |
| melwitt | cdent: I totally agree that it's long term. and that should be when we split governance, is when we are ready for that, IMO | 17:11 |
| cdent | I want use to slow down the server-side and only deal with making the things people want to do with existing functionality (shared disk, nested rp, happen) for _all_ the projects | 17:11 |
| cdent | so you're concerned that the "all the projects" part will be detrimental to nova's goals? I disagree. I think nova will hugely benefit from a more diverse group of people (eventually) exercising placement. fewer bugs is the big one. | 17:12 |
| edleafe | What I don't understand is that if the placement cores are the nova cores, how there could be any divergence of priorities | 17:13 |
| melwitt | I understand that, but currently we are in the position where server side features have _never_ been used before (NRPs) and I think first-ever integration should be done while we are one group. it will likely need fixes on the server side, based on previous experiences we've had | 17:13 |
| cdent | most of the placement-side functionality for pending stuff is in place (as matt has pointed out many times) | 17:13 |
| *** ricolin has quit IRC | 17:14 | |
| cdent | I guess we'll just have to disagree on "while we are one group". I feel like we always will be but that if you feel like we won't be when governance changes (because, apparently, that will change priorities) then we aren't one group now | 17:14 |
| melwitt | edleafe: because that's the point of having separate governance. it means you are an independent group with the ability to have independent priorities. I think it's premature given the work we're still in the middle of | 17:14 |
| scas | what i'm curious about is what is the urgency on going through the motions? to put a firm-yet-malleable date on the actual motions, seems totally reasonable to me, without doing things so close to a release. this is just my peanut gallery interpretation, but stein is a longer cycle | 17:16 |
| melwitt | cdent: maybe, but it's a big step that we can't come back from, if we take it. I'd rather we just sit down and agree to some set of work items to get done before we call it. we will split the governance, I would like it to be based on a real work milestone | 17:16 |
| edleafe | melwitt: I would agree if we started with a completely different team. The idea of making all nova cores as cores in placement is so that while those critical pieces are in play, the behavior will be the same either way. Over time, though, the number of nova cores actively involved in placement will diminish, and new people drawn to the project will become engaged, and possibly merit core status | 17:16 |
| cdent | melwitt: do you agree with jay and matt that this is a people problem? If not, what are they talking about that you disagree with? | 17:17 |
| melwitt | edleafe: I understand where you're coming from, but I still think it's premature. we've only just started the code extraction and we have important work to do that I think makes most sense while we are still one group, before we split | 17:18 |
| *** annabelleB has quit IRC | 17:19 | |
| melwitt | cdent: I don't know, I've been focused on the work we need to get done, the biggest one being NRPs integration. I'm doing my best to ensure that happens as smoothly as possible, and I think being one group under compute is the smoothest path for it | 17:20 |
| melwitt | ideally, we wouldn't be extracting code in the middle of it, but I am trying to get where you're coming from to make some progress there, so we're doing it | 17:21 |
| * cdent nods | 17:21 | |
| cdent | I'm grateful for that. | 17:21 |
| * leakypipes will have to remember the term "firm-yet-malleable" | 17:23 | |
| melwitt | I'd like for us all to have a serious discussion about splitting the governance and not make arbitrary dates for a cutover. I hope, as a group, we can agree one some minimal set of work items to do and then we split it after that | 17:23 |
| cdent | But I'd like to reiterate: I'm just trying to do my job. This is a thing we'd say we'd do. I've been working on it for going on 2.5 years. Had I my druthers we never would have started in nova in the first place, but we did, and that was okay. My efforts these past couple of weeks have just been trying to bring all the necessary people together to make it happen. It wasn't suppose to be a cluster fuck, and I'm still not s | 17:23 |
| cdent | why it has been. | 17:23 |
| melwitt | I too, am just trying to do my job | 17:24 |
| scas | agreeing past one another: a time-honored technology tradition | 17:24 |
| cdent | melwitt: yeah, and I really do appreciate that. I sometimes feel like we're caught in the crossfire of something that's going on where we can't quite see | 17:25 |
| elbragstad | i can relate to the concern about things taking longer even involving multiple groups or even just repositories (this is something we deal with quite a bit in keystone) | 17:26 |
| cdent | I'm sorry, I have to go, will catch up later | 17:28 |
| melwitt | I think we've pretty much wrapped up here for now. I'd like to get back to work too o/ | 17:30 |
| *** jamesmcarthur has quit IRC | 17:30 | |
| *** jamesmcarthur has joined #openstack-tc | 17:34 | |
| *** annabelleB has joined #openstack-tc | 17:37 | |
| *** jamesmcarthur has quit IRC | 17:45 | |
| *** mriedem_away is now known as mriedem | 17:50 | |
| *** annabelleB has quit IRC | 17:58 | |
| *** dtantsur is now known as dtantsur|afk | 17:59 | |
| * dims catches up | 18:02 | |
| *** annabelleB has joined #openstack-tc | 18:04 | |
| *** annabelleB has quit IRC | 18:30 | |
| dims | does FNTECH have one of the throwable microphones? | 18:35 |
| dims | would be handy when we get everyone in a room and have to pass it around (doubles as a talking stick) | 18:35 |
| leakypipes | dims: heh. | 18:37 |
| smcginnis | Based on the ups and downs here, I'm not sure throwing things is a good idea. ;) | 18:37 |
| dims | hahahaha smcginnis | 18:41 |
| dims | smcginnis : when i was chaperoning some 4th graders, i threw a clementine to my son (during lunch, not too far!) and got yelled at by the teacher-in-charge! | 18:42 |
| *** annabelleB has joined #openstack-tc | 18:46 | |
| smcginnis | dims: Hah, naughty naughty. | 18:47 |
| dims | kids still remind me to this day :) | 18:47 |
| zaneb | dims: lol | 18:48 |
| dims | i clearly don't learn my lessons :) | 18:49 |
| *** tosky has quit IRC | 18:52 | |
| *** openstackstatus has quit IRC | 18:58 | |
| *** annabelleB has quit IRC | 19:32 | |
| *** e0ne has joined #openstack-tc | 19:36 | |
| *** openstackstatus has joined #openstack-tc | 19:39 | |
| *** ChanServ sets mode: +v openstackstatus | 19:39 | |
| *** openstackstatus has quit IRC | 19:55 | |
| *** openstackstatus has joined #openstack-tc | 19:58 | |
| *** ChanServ sets mode: +v openstackstatus | 19:58 | |
| cdent | goodnight | 20:02 |
| *** cdent has quit IRC | 20:02 | |
| *** e0ne has quit IRC | 20:06 | |
| *** openstackstatus has quit IRC | 20:11 | |
| *** openstackstatus has joined #openstack-tc | 20:12 | |
| *** ChanServ sets mode: +v openstackstatus | 20:12 | |
| *** openstackstatus has quit IRC | 20:36 | |
| *** openstackstatus has joined #openstack-tc | 20:38 | |
| *** ChanServ sets mode: +v openstackstatus | 20:38 | |
| *** harlowja has joined #openstack-tc | 20:46 | |
| *** annabelleB has joined #openstack-tc | 20:55 | |
| *** e0ne has joined #openstack-tc | 20:56 | |
| mnaser | In regards to having a discussion, I drafted up an email about this subject and I agree we all just need to sit and discuss this. | 20:58 |
| *** cmurphy|vacation is now known as cmurphy | 20:58 | |
| *** e0ne has quit IRC | 21:00 | |
| jbryce | dims: We've tried those throwable mics before with limited success. I'll see if they still have one lying around | 21:00 |
| dims | thanks jbryce ! | 21:01 |
| *** jamesmcarthur has joined #openstack-tc | 21:44 | |
| mriedem | please no mics | 21:59 |
| mriedem | mics are always a pain in the ass at these things | 21:59 |
| mriedem | if you want to be part of the discussion, speak up and/or sit near the discussion | 21:59 |
| *** annabelleB has quit IRC | 22:00 | |
| mriedem | i think we've consciously tried to spread out the big mouths at the ptg in recent times to try and avoid side discussion | 22:01 |
| *** elbragstad has quit IRC | 22:02 | |
| *** EvilienM is now known as EmilienM | 22:04 | |
| *** tosky has joined #openstack-tc | 22:09 | |
| *** annabelleB has joined #openstack-tc | 22:21 | |
| *** harlowja has quit IRC | 22:24 | |
| *** mriedem has quit IRC | 22:32 | |
| *** zaneb has quit IRC | 22:48 | |
| *** zaneb has joined #openstack-tc | 22:48 | |
| *** annabelleB has quit IRC | 22:51 | |
| *** annabelleB has joined #openstack-tc | 22:58 | |
| *** jamesmcarthur has quit IRC | 23:01 | |
| *** jamesmcarthur has joined #openstack-tc | 23:10 | |
| *** tosky has quit IRC | 23:14 | |
| *** jamesmcarthur has quit IRC | 23:15 | |
| *** annabelleB has quit IRC | 23:18 | |
| mrhillsman | lol | 23:23 |
| fungi | seating charts! | 23:24 |
| *** jamesmcarthur has joined #openstack-tc | 23:31 | |
| *** jamesmcarthur has quit IRC | 23:40 | |
| *** jamesmcarthur has joined #openstack-tc | 23:53 | |
| *** jamesmcarthur has quit IRC | 23:57 | |
| *** jamesmcarthur has joined #openstack-tc | 23:58 | |
| *** jamesmcarthur has joined #openstack-tc | 23:58 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!