14:00:49 #startmeeting tc 14:00:50 Meeting started Thu Jul 11 14:00:49 2019 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:51 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:53 The meeting name has been set to 'tc' 14:00:57 Hi everyone! 14:01:01 #link https://giphy.com/gifs/hello-minions-waving-wAVA7WdV2jita/fullscreen 14:01:10 troll call ;) 14:01:10 o/ 14:01:19 Who in tc-members is here for the TC meeting? 14:01:19 o/ 14:01:24 o/ 14:01:26 o/ 14:01:28 o/ 14:01:32 * mugsie is double booked, but wathcing 14:01:32 aloha 14:01:38 * fungi double-booked too 14:01:45 oh, no, not for another hour 14:01:47 * TheJulia is also double-booked 14:01:49 focused then! 14:01:51 mugsie: come over here, I was relying on your GIF game 14:01:59 :D 14:02:04 OK, let's start easy 14:02:09 #topic Follow up on past action items 14:02:13 1/6 Health check changes 14:02:20 asettle to update community (done), fungi to update wiki (done), mugsie to update yaml file with liasons and mnaser to update the tooling 14:02:36 mugsie: picked up the tooling so thanks :) 14:02:51 OK so this is in the pipe, right 14:03:04 mnaser: yeah, I totally forogt you were going to do it, and neeed to populate the yaml to test so I wrote a thing :) 14:03:48 Status: Under review 14:04:05 #link https://review.opendev.org/#/c/668004/ 14:04:06 http://replygif.net/i/716.gif 14:04:14 2/6 Help-most-needed list 14:04:24 AlanClark and zaneb to update investment opportunities document 14:04:33 Not sure what the status is there 14:04:41 A bunch of things were merged for sure 14:05:15 so for 2019, only glance is there. are we going to add others also in that ? 14:05:27 that's part of the rewrite activity 14:05:35 gmann: plan was to convert them and add them all to 2019 14:05:41 glance was just first in the queue 14:05:45 Status: work in progress 14:06:00 3/6 Goal selection 14:06:01 ok. I am going to add that in my slide for openstack day tokyo next week. 14:06:06 lbragstad to prune the community-goals etherpad 14:06:09 #link https://etherpad.openstack.org/p/community-goals 14:06:15 yeah 14:06:15 lbragstad: how is that going? 14:06:17 that's done 14:06:21 Status:Done 14:06:23 #link http://eavesdrop.openstack.org/irclogs/%23openstack-tc/%23openstack-tc.2019-06-06.log.html#t2019-06-06T15:35:13 14:06:31 i removed the obvious ones 14:06:41 4/6 Pop-up teams 14:06:45 ttx to define pop-up teams 14:06:46 lbragstad: it does seem a good deal shorter than the last time i looked 14:06:49 https://giphy.com/gifs/spongebob-spongebob-squarepants-episode-12-3ohzAi9KJLc5Vi9CPm/fullscreen 14:06:59 That is done. However we may want to assign a TC "liaison" for the Image encryption team 14:07:07 Any volunteer? 14:07:44 can put me down as volunteer for that 14:07:55 Great, thanks fungi 14:08:02 i've been trying to keep the security sig in the loop on what's going on there anyway 14:08:15 #action fungi to add himself as TC liaison for Image Encryption popup team 14:08:29 Status: Done 14:08:34 5/6 Explaining governance 14:08:40 ricolin has produced a draft 14:08:42 #link https://review.opendev.org/#/c/668093/ 14:08:48 I posted a -1 because I feel like it's presenting things backward 14:08:50 ricolin: I can help reorganize it to my liking, if you want 14:08:58 I just update the patch according to comments 14:09:08 ttx sure 14:09:12 ah, jinxed 14:09:12 that will be super 14:09:18 Will rereview 14:09:21 Status: Under review 14:09:33 6/6 Review PTL Guide 14:09:38 https://review.opendev.org/#/c/665699/ was merged 14:09:41 Status: Done 14:09:55 Any comment on those past action items? 14:10:57 https://media.tenor.com/images/c32c6ff16cd7fb47769c87f2eb5e95f4/tenor.gif 14:11:07 dammit 14:11:27 https://tenor.com/view/green-cola-no-gif-9810848 14:11:39 #topic Active initiatives 14:11:44 1/3 Python 3 14:11:51 mnaser to sync up with swift team on python3 migration and mugsie to sync with dhellmann or release-team to find the code for the proposal bot 14:12:00 What's up on that? 14:12:13 * mugsie didn't do it 14:12:26 concise! 14:12:36 i've personally been looking from the sidelines, it looks like things are progressing well and it seems like some effort/help from rh is coming in to help iron all these out 14:12:49 ok, let's carry that over 14:13:03 Status: In progress 14:13:09 2/3 Forum follow-up 14:13:13 ttx to organise Milestone 2 forum meeting with tc-members 14:13:16 That will happen in the coming weeks 14:13:22 Status: In progress 14:13:28 3/3 Make goal selection a two-step process 14:13:32 We need reviews at 14:13:34 #link https://review.opendev.org/#/c/667932/ 14:14:01 Any comment on active initiatives? Anything missing? 14:14:14 Status: under review 14:14:23 I like the 2 step process doc 14:14:50 i have not reviewed it yet. need to get more background on that. 14:14:59 I hope it will solve the approval/refinement/selection bottleneck 14:15:12 doing everything in a single review has proven... inefficient in the past 14:15:14 we need to make sure we graduate some idea into the proposed bucket when we merge 14:15:52 I think that was part of the point of having lbragstad prune the existing etherpad, wasn't it? 14:15:59 gmann: we discussed it at the forum/ptg but it's basically a way to keep us from getting stuck on full quorum voting of implementation details for goal approval 14:16:06 we need some intermediary sandbox for ideas to mature before being selected basically. And separate goal submission from goal selection reviews 14:16:34 shouldn't be a different "kind of topic" with more lenient reviews? I thought we agreed on that 14:16:37 since we select a set of goals together, not just individual proposals independently 14:17:09 evrardjp: yes we said that refining a goal should be approved leniently 14:17:10 evrardjp: yeah, the implementation plan was going to be handled under documentation review process while the goal ideas would still be under formal vote 14:17:33 fungi: funny I thought it was otherwise 14:17:43 Anyway, feel free to ask further questions on the review for everyone to benefit 14:17:52 I thought the proposal of the idea was lenient, but the validation was to be a strict proposal 14:18:11 evrardjp: depends on what you mean by "validation" 14:18:12 evrardjp: selecting cycle goals is the formal vote, deciding how many widgets a thingamabob needs to meet the goal requirement under some circumstance is just documentation 14:18:43 ultimately you select them with a strict vote 14:19:06 on that we agree, those selected for a cycle are following a strict vote 14:19:07 so it should be easier to fix/amend the implementation details, after the goal itself is approved 14:19:15 I feel without having implementation details it will be difficult to have clear pic of goal . example osc-client goal. 14:19:54 gmann: that is why you wait until the implementation is more detailed before selecting them. 14:20:10 right, i stated them in reverse chronological order 14:20:13 but improving the goal becomes an incremental process 14:20:45 OK, onto our two topics of discussion for today 14:20:51 #topic Update on U release naming process 14:20:58 sorry. improve/amend implementation details in the repo (as documentation reviews), prior to goal selection (which is the formal vote) 14:21:03 (sorry for being a complete pita in that process) 14:21:03 There is a bit a confusion with the plan here -- in particular the WeChat activity to select China-friendly names 14:21:13 Can anyone give us an update? 14:21:45 fungi: +1. 14:22:05 (personally at this point I would go with any name the China community likes) 14:22:21 I am like ttx on this 14:22:39 yes, i'm inclined to just approve a non-conforming name that has some community consensus and is unencumbered and non-offensive 14:23:10 i agree with ttx 14:23:15 i think right now the best thing to do is let the process happen on its own 14:23:18 we're fast coming up on a cycle which will otherwise end up being called "unnamed" 14:23:24 agreed with fungi, that's so elegantly said, who can refuse? :D 14:23:25 mnaser: is the wechat poll supposed to come up with candidate names ? 14:23:29 and if there are proposals that are "out of normal criteria" 14:23:33 they can be proposed like train 14:23:34 so the only way to vote would be the wechat poll? 14:23:34 unnamed would be quite appropriate 14:23:35 ttx I'm pushing that part now but most people from china community wechat group agree on the list I give last time 14:23:54 dhellmann: wait what? 14:24:04 dhellmann: no 14:24:09 evrardjp : I'm not sure, I don't understand what's happening :-/ 14:24:10 ok, good 14:24:16 * ricolin looking for the name list 14:24:17 so the poll is what, then? coming up with suggestions? 14:24:17 i thought the wechat bit was asking the community there for suggestions of names 14:24:17 That would select a shortlist and we'd put that to our usual voting 14:24:22 ok 14:24:33 Just making sure all candidates on the ballot are "good" 14:24:43 ok, got it 14:24:47 So select between their top picks 14:25:18 that works for me, and aligns with a comment I made in channel earlier when I suggested we formalize a "local contributor" selection committee for names in the future 14:25:21 ricolin: my point is, it's also OK to add non-geographic-but-very-popular-in-china options to the list, imho 14:25:33 +1, easy and less controversial way 14:25:35 location： Urumqi Ussri(Ussri River）Ulanqab Ulanhot Ulansu(Ulansu sea) Urad (乌拉特中旗) Ujimqin(东/西乌珠穆沁旗）Ula (Ula nara) 14:25:35 others：Unnamed Undefined Unique Unicorn Undead Uncle Umpire Utopia umbrella ultimate 14:25:42 yeah, please let's just get a list of 5-10 names and start voting 14:26:16 These are raised from WeChat group, but not formal voting yet 14:26:18 I'm just afraid we put 3 chinese names and "unnamed" on the list and our US/EU community selects unnamed out of familiarity 14:26:39 and we miss an opportunity 14:26:52 yeah, I think unnamed and undefined have negative connotations and I would not support including those on the ballot 14:27:08 me too. we can avoid those 14:27:09 ttx: the vote is only an "indicative vote" right? We don't have to take it if we feel there is an issue with the name 14:27:12 unicorn would work, if people in China end up liking it 14:27:14 we should probably limit voting to geographical/place names if we have enough to make a reasonable length ballot with them 14:27:31 mugsie: yes there is still a... marketing filter 14:27:40 + a trademark one 14:27:50 but that is done after the ranking 14:28:16 but who doesn't like a unicorn? 14:28:18 I personally would not put unnamed on the ballot at all 14:28:20 Like "Unupgradeable" would probably not make it 14:28:27 the most popular are Ussri and Urumqi IIRC 14:28:31 ttx: oh surprise?! 14:28:38 one benefit of ranked voting. you can safely eliminate ineligible candidates after the poll closes without significantly mipacting the relative popularity of the remainder 14:28:55 Undead maybe... 14:28:57 ricolin: ussr-i? 14:29:09 i was hoping for unpossible... oh well 14:29:17 fungi: haha 14:29:23 I guess we completely disgressed now 14:29:53 naming the bikeshed 14:29:57 anyway, is there something to decide? 14:29:59 ricolin, mnaser: so.. what is the next step ? Wechat activity ranking a top 5 options and then us vetting that list, and putting it into a vote? 14:30:07 evrardjp, it' Russian 14:30:36 ttx I do plan for such thing, but when will be the deadline for it? 14:30:45 ricolin: I'd say ASAP 14:31:02 ricolin: https://www.youtube.com/watch?v=a1Y73sPHKxw 14:31:12 our technical election officials need to plan our upcoming elections, and need a cycle name to associate with those 14:31:26 which was a big part of the urgency on this 14:31:39 if we put an arbitrary say 14:31:44 next wednesday is that ok? 14:31:53 oh sure 14:32:23 do we want to just do a resolution to say that for this release 14:32:24 ttx Horace said he will use official wechat for the polling activity, not sure how that goes for now 14:32:27 will check 14:32:30 we'll be accepting anything that's "china" related 14:32:39 so that is atleast landed by the time we have suggestions so they're all 'valid' 14:32:43 and popular with the Chinese community 14:32:46 not to delay things any longer? 14:32:57 ++ 14:32:59 thatll give it the week it needs for formal-vote 14:33:02 ok, i'll work on that 14:33:06 o' 14:33:12 +1 14:33:22 i'll use the train thing as the template and urge tc-members to vote asap so we can have its grace period 14:33:30 ++ 14:33:34 +1 14:33:43 +1 14:33:46 ok, sounds like we have a way forward 14:33:57 awesome, thanks! 14:33:57 Did not want to stall it until Tony is back from vacation 14:34:13 Next topic... 14:34:15 #topic What are retired repos ? 14:34:21 evrardjp, https://en.wikipedia.org/wiki/Ussuri_River 14:34:56 context: i started work on a ci job that checked if a repo was properly retired because we had some disagreements on landing a governance change if the repo was retired (or not) 14:35:16 we use the term "retired" to mean several different things in different contexts 14:35:39 yep, that's what surfaced in the ML thread with a bunch of projects that acutally still are somewhat maintained, with commits, but not "official" 14:35:49 being "out of governance" and "retired" are different 14:35:51 ricolin: thanks for the link, TIL :) 14:36:04 our governance does not reflect that difference right now, afaik 14:36:04 there are also repos which have been "retired" from use in master but still have stable branches maintained 14:36:19 should we use different terms then 14:36:29 mnaser : our governance only reflects being in or out of governance, which seems appropriate. 14:36:31 but as per new namespace, are we going to maintain anything non official under openstack/ namespace. that is most confusing part 14:36:35 we likely should start by (don't hit me) coming up with new names for some of those concepts? 14:36:53 * mnaser is going to be playing devils advocate 14:36:55 well I have the impression it's a per branch thing _again_ 14:37:06 that means we no longer -1 things if there is no retirement commit 14:37:13 dhellmann: ^ 14:37:56 * mnaser is just trying to find a thing we can all agree too for project-update retirement changes 14:38:13 mnaser : We could say that. Or we could look at each case, and decide what to do based on what the intent is. If we want to encourage folks to follow the full retirement process, we could keep repos under governance until they have done that. If they don't intend to retire the repo, then we don't need to do that. 14:38:35 dhellmann: +1 14:39:11 in the past (before we switched to just using the openstack namespace for everything) the idea was that if a project became unofficial it would have to be renamed out of the openstack namespace. i think that has some problems of its own now that we have redirects in place, since it becomes a lot less obvious to source code consumers when something stops being a part of openstack 14:39:33 historically OSA for example, has waited for things to be empty on all branches before asking for a retirement, and that was fine 14:39:49 evrardjp : we've seen problems in other projects that didn't do that 14:40:30 dhellmann: could those problems have been solved by the right docs? Saying how to properly retire something over time? 14:40:32 yeah. out of governance means it should be out of openstack/ namespace. because cmg to governance is cmg under openstack/ namespace 14:40:38 being listed in governance means that commits to that repository's stable branches still count as contributions to openstack 14:40:47 I think our lives would be simpler if the retirement process required that, but that's not what's confusing here, right? It's that we have projects that left governance with the intent to retire and then didn't, or that left governance with the intent to keep running and weren't renamed. 14:41:12 dhellmann: I guess I misunderstood the problem :p 14:41:26 evrardjp : the problem you mentioned is related, but not the same 14:42:00 the first question in my mind is, do we have someone willing to go around and fully retire the repos that look like they have been abandoned? 14:42:12 yes, i've in recent years come to the conclusion that we should be forcing projects who want to leave openstack to fork, so we can replace the content in the version which remains behind in the openstack namespace with a notice saying it's no longer part of openstack and where it has gone or what has happened 14:42:23 and the second question is, do we have someone willing to follow fungi's suggestion and enforce a fork for all of the repos that left governance and are still active? 14:42:31 fungi: agreed. 14:42:40 fungi: That is how I've seen some non-OpenStack projects handle that. 14:42:46 i don't know that there's a ton of benefit to doing that for historical exits, but for future ones i do 14:43:07 what is the workload we are talking about? 14:43:18 I am fine working on cleaning things up 14:43:22 we should write down this new policy as a formal governance document 14:43:38 i sent a mailing list post with all of the things that "should be retired" 14:43:42 ofc, else it would be unpossible to apply said policy 14:43:48 in cases like fuel where they left openstack and then within a year stopped working on the project but still had users who were becoming increasingly out of date and exposed and thought they were still using openstack software, i think that damage is probably already done. cleanup might be nice but doesn't fix things 14:43:48 (see what I did there?) 14:44:18 the change https://review.opendev.org/#/c/669549/ 14:44:31 http://logs.openstack.org/49/669549/2/check/openstack-tox-linters/4084f9c/job-output.txt.gz#_2019-07-06_17_42_15_763974 14:44:34 the list 14:44:41 other example is networking-l2gw which was out of governance (neutron stadium) but in active developement under openstacl/netowkring-l2gw 14:45:13 * mnaser thinks dhellmann approach makes sense of saying if we have someone to go and do this first 14:45:15 networking-l2gw sounds like a case where we should encourage them to fork to create a new repo 14:45:22 ++ 14:45:23 dhellmann: the more we are discussing this, the more I am wondering if this shouldn't be int he project-team-guide instead 14:45:34 for projects which have already been removed from our governance but are still lingering in the openstack namespace, we should probably give them one last opportunity to rename out of the namespace 14:45:36 evrardjp : process docs, yes, but policy docs no 14:45:47 dhellmann: at the time, my intention was to post to the ML in the hopes that the projects/teams themselves pick it up and take care of it 14:45:50 on that we agree. 14:46:26 retroactively imposing a requirement to fork on them is not super friendly 14:46:32 fungi: yes but they are not aware of that. we should communicate them about renaming 14:46:34 rename, then? 14:46:41 fungi, including rename on Pypi? 14:46:48 pypi doesnt inclue a prefix 14:46:55 things on pypi aren't openstack/foo (afaik) 14:46:59 ricolin: there are no namespaces on pypi 14:47:19 are we focusing on the right things for the right people? I want to make sure if we are making a policy, it's for clarity of the users, not to have a policy 14:47:20 (that is a separate conversation the pypi maintainers and users seem to have ~annually) 14:47:56 ok lets take a step back, someone has to retire the abandoned projects first 14:48:15 those have to be done regardless of policy or whatever, they're actually retired projects 14:48:16 evrardjp: i think the "policy" is simple (only current official projects in the openstack namespace). the *process* for ensuring that is what needs fleshing out 14:48:30 so 1) define abandonned project, 2) create a policy about moving them out of governance 14:48:43 does anyone feel like maybe going over those projects and making commits to retire them properly? 14:48:52 or at least following up with the teams to do so? 14:49:16 (some people felt very strongly we don't merge anything that was 'retired' but had code, so perhaps a good time to help enforce this :) ) 14:49:17 i do think that the process for the current cases (historical removals) can and probably should follow a different process from what we want to do going forward 14:50:05 i don't like this, but alternatively, we can enforce retired-on in legacy.yaml and have the checks only check those past a date, but then we'd have two sets of rules :) 14:51:05 i my opinion projects who have already left openstack can be renamed into a different namespace if their maintainers wish it, and then we tell all official projects that any which leave openstack in the future have to fork so we can retire the openstack namespace repo with a prominent notice 14:51:27 that works for me -- let's get that written down as a resolution or something 14:51:44 +1. that will be very clear for everyone 14:52:00 +1 14:52:10 maybe give existing teams a deadline too in case they have some they want to move out of openstack before that rule goes into effect, though that's less important in my mind 14:52:37 Alright, who is taking that todo? 14:52:39 dhellmann: just wanna double check the wfm comment is re what fungi said? 14:52:59 mnaser : yes, I like his proposal but think we should write it down formally before acting on it 14:53:10 i will draft a resolution later today 14:53:21 thanks fungi -- i can try helping with the logistics once it lands 14:53:27 appreciated! 14:53:34 #action fungi to draft a resolution on proper retirement procedures 14:53:36 talking to teams / getting abandonded code properly retired / renaming things (if needed) 14:53:55 #topic Open discussion 14:54:02 i'll attempt to encapsulate the historical context for the decision as well 14:54:07 Next meeting should in theory be on August 1st but that may be a bit too close from now. 14:54:14 hey ttx slow reply. I contributed several suggestions to the Glance investment opportunities document. I also also sent suggestions for the documentation document. Are there others? 14:54:18 I was wondering if having next ones on August 8 and September 5 would not spread them out more evenly. 14:54:39 seems reasonable 14:54:48 AlanClark: no new ones yet i don't think, though expect a few more in coming months 14:54:56 AlanClark: would you mind reiterate on a new one? :D 14:54:57 AlanClark: there will be others we'll propose a rewrite from. I suspect we'll call you for input on those when ready 14:55:03 thanks for being involved AlanClark 14:55:10 thanks 14:55:11 yes that really helps 14:55:12 ++ thanks AlanClark 14:55:18 thanks for the help indeed :) 14:55:21 in other related open topics ,i wanted to ask which tc-members plan/might/will be in shanghai? 14:55:29 thanks AlanClark:) 14:55:34 we did also get that first one promoted in the osf newsletter yesterday 14:55:37 #link https://superuser.openstack.org/articles/osf-newsletter-july-9/ 14:55:38 mnaser: I plan to be there, pending visa acceptance 14:55:40 I am planning on it, looks likely I will be there 14:55:44 * mnaser wanted to fill out the ptg tc form 14:55:49 * jroll will not be there 14:55:50 mnaser: I will be there (yet to start the visa process though) 14:55:52 how much min/max time do we feel like we want? 14:55:55 definitely:) 14:55:56 * lbragstad is TBD 14:56:00 mnaser: I do not know yet 14:56:01 I am not planning to attend 14:56:03 i think last time a day was .. okay 14:56:03 #info next TC meeting: August 8 14:56:13 https://i.imgur.com/kzC6MOx.gif 14:56:36 ricolin: I am glad you're here for sending the gifs :) 14:56:39 mnaser: if my visa is approved (and assuming no natural disasters at home this year) then i'll be there 14:57:06 i'm trying to see if we're going to have enough of us there to actually need a ptg-one-day-meeting like we had 14:57:14 I'd say 0.5 min/1 day max 14:57:16 evrardjp, I can't resist! 14:57:53 i think that's reasonable ttx 14:58:03 10-15 people room probably? 14:58:24 if it's a public meeting, we should allow for observers 14:58:24 round chair one right ? 14:58:28 we might have others come in.. maybe 18-20 14:58:32 or maybe 20ish i guess 14:58:41 That PTG is so full of unknowns 14:58:44 ok, fair nuff, i think i have all that i need to give kendall. 14:58:47 yeah.. 14:58:52 mnaser, 18-20 sounds reasonable, counting observers 14:58:56 yeah, we have in the past had as many observers as tc members present. no idea if it will be the case again in shanghai 14:59:20 who knows, maybe our chinese community are curious about what an elected governing body actually looks like in the flesh 14:59:31 will we be able to get extra chairs if we needed? 14:59:43 i think so 14:59:43 I hope it will encourage more to run 14:59:55 so it's a bit of a publicity exercise too 15:00:00 i agree 15:00:10 not a formal onb-boarding, but a bit like it 15:00:20 voila, sent it over, thanks 15:00:21 "this is what the TC does" by example 15:00:27 show folks how we operate, yeah 15:00:41 Alright, we are done here 15:00:45 #endmeeting