20:01:45 #startmeeting tc 20:01:46 * mestery consoles edleafe 20:01:46 Meeting started Tue Apr 12 20:01:45 2016 UTC and is due to finish in 60 minutes. The chair is ttx. Information about MeetBot at http://wiki.debian.org/MeetBot. 20:01:47 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 20:01:49 The meeting name has been set to 'tc' 20:01:52 ttx: sup? 20:02:00 Welcome to the Newton membership ! 20:02:14 * mordred looks around and is scared by all the new faces 20:02:22 dims, mtreinish, morgan, thingee: welcome 20:02:28 mordred: they're not that new :) 20:02:31 This is the agenda for our first meeting: 20:02:34 ^_^ 20:02:34 thanks ttx 20:02:35 #link https://wiki.openstack.org/wiki/Meetings/TechnicalCommittee 20:02:36 agentle: inorite? 20:02:36 ttx: thanks 20:02:43 #topic Welcome Newton TC members 20:02:47 have you told them about the initiation ritutal yet? 20:02:50 or is that a surprise? 20:02:53 surprise 20:02:53 mordred: you're no longer the only one sitting up front with blue hair. 20:02:58 * Update member roster 20:03:00 thingee: WOOT 20:03:06 I did update voting rights in Gerrit and the -tc ML moderation bits 20:03:14 We just need to approve the new roster, which was +1ed by election officials 20:03:18 #link https://review.openstack.org/303119 20:03:23 thingee: the hair-posse has doubled in size (and tripled in awesome) 20:03:46 no wonder jay left 20:03:47 if anyone else wants to boost their +2 stats before I approve 20:03:49 ttx: what happens if we do not approve it? 20:03:58 mordred: then I approve it 20:04:07 ttx: hehe all about the boosting of +2 stats 20:04:09 we're all about gaming governance voting stats, huh? 20:04:10 and we kick you off the island 20:04:13 russellb: totally 20:04:14 ttx: I was wondering if it was lke the US Debt Ceiling or something 20:04:21 russellb: It's in our DNA 20:04:21 :) 20:04:27 alright it's in 20:04:32 * Select chair 20:04:40 The chair is responsible for making sure meetings are held (and decisions made) following our charter rules 20:04:49 I'll be happy to continue to serve as the guardian of the rules and meeting organizer 20:04:55 hurrah ttx 20:04:57 If another member is interested in such duties they should file a change similar to https://review.openstack.org/303308 20:04:57 ++ 20:05:02 ++ 20:05:03 yay for ttx! 20:05:03 edleafe: :) 20:05:08 If nobody else is interested, we should just approve https://review.openstack.org/303308 20:05:16 o/ 20:05:18 ttx 4ever! 20:05:41 Alright 20:05:42 ++ttx 20:05:46 Thanks everyone ! 20:05:54 * Confirmation of change approval rules & agenda buildup process 20:06:03 Next in the boring things we need to cover today... 20:06:10 * mordred puts on boring hat 20:06:12 Quick reminder of the rules we followed in Mitaka to build agenda and approve governance changes 20:06:14 thanks for all you do, ttx. 20:06:18 * flaper87 steals mordred's hat 20:06:24 * mordred puts on flaper87 20:06:24 Anyone can propose a change to openstack/governance or a topic to be discussed at the following TC meeting (by editing the agenda wiki page directly) 20:06:34 Things need to be proposed before EOD Thursday to be considered for next meeting, to give time for evaluation and/or community discussion 20:06:44 For openstack/governance changes, there are 4 types. 20:06:46 mordred: mmh... you won... 20:06:55 1. typo fixes, where the content we publish is obviously wrong, and I'm authorized to fast-track them. 20:07:06 2. code changes, which affect the tools or doc publication in the repository, for which we use the classic "two votes in addition to the author" code review rule 20:07:22 * dims pays attention 20:07:27 3. project team updates (like a new repository addition or the assertion of a tag), where as long as the PTL approves the change (+ security team for security: tags, release team for release: tags, stable team for stable: tags...), we do lazy consensus approval 20:07:42 which means if nobody objected after one week the change has been proposed, it is automatically approved 20:07:57 4. formal votes (everything else), where we'll discuss the topic at the next TC meeting, and require motion approval 20:08:12 which generally means reaching 7 votes in favor (or if that fails, at least 5 votes in favor and more votes in favor than votes against after calling for a final vote) 20:08:20 Changes to our charter text are a special case requiring 9 votes in favor. 20:08:25 this is a good review :) 20:08:34 The idea being to limit the number of changes we actually need to discuss 20:08:43 We use the topic "formal-vote" for all formal votes, so you can quickly filter using 20:08:46 makes sense 20:08:49 #link https://review.openstack.org/#/q/status:open+project:openstack/governance+topic:formal-vote 20:09:03 In all cases the chair applies W+1 when the rules are met. Don't hesitate to call me out if you feel I'm not following those rules. 20:09:07 agentle: there will be a test at the end 20:09:08 * mtreinish likes having less things to worry about 20:09:13 Will this work for you in Newton ? Any change you'd like to make to that ? 20:09:18 * morgan agrees with mtreinish 20:09:33 this all sounds pretty spot on an reasonable. 20:09:34 * agentle is furiously taking notes 20:09:35 this all seems fine 20:09:36 ttx : sounds good to me 20:09:39 ttx: wfm 20:09:40 LGTM ttx 20:10:14 I don't have suggestions, it's working fine to me. 20:10:28 Alright, we can always change that after some time, nothing is set in stone here 20:10:33 * Austin reminders 20:10:41 * morgan puts away chisel. 20:10:43 We'll have a joint Board/TC meeting, Sunday starting at 2:30pm 20:10:52 at the JW Marriott, Level 3, Salons G/H 20:11:03 ttx: Is it 2:30-5:30? 20:11:23 checking 20:11:45 unclear... 20:11:50 ttx: Ack, thanks for checking :) 20:11:51 Level 3 of the JW Marriott is quite nice 20:11:57 heh has it been on time before? 20:12:01 agentle: nope 20:12:02 mordred: It's pretty lush indeed 20:12:12 No closure hour announced, but there is WOO at 6pm after that 20:12:14 it was on time in vancouver 20:12:26 There is also optional dinner with the BoD, UC, Gold Member Reps and Foundation Staff, Saturday before the event at 7pm 20:12:33 note the *Saturday* 20:12:41 You should all have received the link to RSVP if you're around / interested. 20:12:41 Boo to Saturday :) 20:12:42 sdague: good memory. Vancouver was so awesome. 20:12:59 yeh, no go on the sat for me, I shall be on the plane 20:13:01 Please RSVP asap since they were supposed to close teh RSVP one week ago 20:13:16 so I have no idea how much longer they will wait 20:13:34 Questions on Austin ? 20:13:37 * morgan has RSVPd and made sure to book appropriate flight to not miss these things. 20:13:59 i most likely won't be in person this time around 20:14:15 russellb: for the board meeting or for the summit? 20:14:22 thanks ttx i won't make it to the Dinner. should be in time for the 2:30 meeting 20:14:27 whole thing ... having another kid due summit week 20:14:30 russellb: You'll be there in spirit 20:14:35 russellb: congratulations 20:14:38 thanks 20:14:40 russellb : congrats! 20:14:43 russellb: good reason to not be there, congrats! 20:14:43 russellb: keep us posted, with pictures 20:14:44 russellb: congrats, and will miss you 20:14:45 russellb: healthy experience to your family 20:14:47 russellb: You're naming the new child Newton, right? ;) 20:14:54 Isaac 20:14:57 thanks for the kind words, everyone :) 20:14:58 :) 20:14:58 russellb: Cognrats! heh, you know how to time things :) 20:15:08 lol 20:15:10 * Rockyg thinks Austin is a great name for a kid 20:15:27 Rockyg: better than Bexar for sure 20:15:32 ha 20:15:37 ok, next topic... 20:15:42 #topic Propose topics for Joint board / TC meeting agenda 20:15:59 Alan Clark reached out to me to build an agenda for the joint meeting, I replied we'd wait for the elections to conclude and for us to have our first meeting 20:16:05 DCO status? 20:16:08 So... What should we be discussing there ? 20:16:10 sdague: ++ 20:16:10 as that's still not thing yet 20:16:11 * mordred will give russellb a billion dollars to name his kid bexar 20:16:16 yeah, that sounds like a regular 20:16:25 sdague: ++ 20:16:26 sdague: it's blocked on us 20:16:32 no need to discuss with board 20:16:36 sdague: ++ 20:16:40 mordred: define "us" 20:16:47 mordred: he'll need the whole billion to pay for the therapy 20:16:53 sdague: well, bug tracking is the main thing 20:16:54 mordred: we could still give an update, esepcially if that depends on "us" 20:16:56 mordred: i'd call you on your bluff, but that sounds like a lot of work on my part 20:17:01 ttx: ++ 20:17:20 anteaya, you got that right! 20:17:22 like "we are still working on it" 20:17:35 sdague: we still do not have an option to get bug tracking off of launchpad openid, which means we don't have a solid 'so now we can move gerrit off of launchpad openid' story 20:17:40 mordred: yeh, I think it's still appropriate, as it would be good to track to completion 20:17:49 sdague: which is the thing we need to do to fulfill ourside of the "move to dco is ok" agreement 20:17:54 sdague: ok 20:18:08 and, maybe is now a TC ask back to infra to make that a priority 20:18:08 I also wanted the board to know that we would likely kick out from the "big tent" some projects that do not follow the rules during this cycle, if only to warn them that this can happen 20:18:15 sdague: it is 20:18:18 already 20:18:28 but this may not require a full topic 20:18:40 ttx: sounds like we might need a technical discussion on container v vms? 20:18:41 ttx: I think it makes sense to keep them aware 20:18:44 mordred: but would be fair to keep the tc updated on status, yes? 20:18:45 agentle: ++ 20:18:47 esp. at API levels 20:18:50 agentle: TOTALLY 20:18:50 ttx: that sounds like a reaonable bit 20:18:56 agentle : ++ 20:19:02 I recall at the Tokyo joint TC/board meeting there was confusion on the big tent still from some board members 20:19:03 let me etherpad this 20:19:08 I'm not opposed to talking about DCO/openid at all - I just want to make sure we're going in to the discussion with the riht context 20:19:19 i think there's a lot of big tent concern in the board 20:19:20 Please throw thoughts to https://etherpad.openstack.org/p/KAZvzrNkoM 20:19:20 mestery: there is always confusion on the topic :) 20:19:28 mordred russellb: Indeed :) 20:19:35 in general, about the picture of openstack being totally unclear to people 20:19:53 russellb: that is a concern i've had echoed to me privately 20:20:06 might be worth hearing concerns and brainstorming 20:20:10 thing is ... to be blunt 20:20:11 russellb: ++ 20:20:25 every time it comes up in the board meeting, it's the same set of concerns as last time it was brought up 20:20:33 mordred: pretty much 20:20:38 I'm sure we'll talk about it again this time 20:20:49 it's about lack of focus and clarity 20:20:49 but it woudl be really great if we could nudge people to get up to speed 20:20:50 mordred: that doesn't sound too surprising 20:20:53 mordred: Is there a way out of that quagmire other than not discussing it? 20:20:56 russellb: I'd say it's not ours to put in the agenda, if they have questions they should add it 20:20:56 honestly, that was a theme in TC candidacy messages, too 20:21:01 ttx: fair enough 20:21:08 ttx: ++ 20:21:10 * russellb doesn't desire to have it on the agenda 20:21:17 russellb: you being a Board member can switch hats and add it to Alan's portion of the agenda :) 20:21:25 * mordred realizes he's being grumpypants - apologizes 20:21:31 russellb: I see what ttx did there ;) 20:21:37 ninja move :) 20:21:41 mordred: anything we could discuss to push/encourage interoperability or end user experience ? 20:21:49 oh god. 20:21:51 dims: lol :) 20:21:51 * russellb wearing TC hat firmly toeday 20:21:58 mordred: heh 20:22:00 today, not toeday. toeday is a weird idea. 20:22:04 ttx: so many things - but I'd actually be wearing my board hat on that one 20:22:27 ttx: I believe the only direcitonal ask would be for the board to ask the TC to do more about it 20:22:35 Like, "now that the mission statement is changed, what can we actually do (TC or board) to make it happen" 20:22:45 ttx: see - you say things so much better than I do 20:22:55 ttx: api docs clarity 20:23:22 and yes, I'm willing to report latest and how hellish the path has been 20:23:24 there was discussion at the last board meeting about defcore trying to decide if they want to own their own test suite 20:23:28 agentle: on your containers vs vms discussion - at the last board meeting we talked about openstack as a single platform for baremetal, vms and containers ... 20:23:28 that was slightly alarming 20:23:29 agentle: is that another item, or in reference to the interop story? 20:23:36 agentle: it seems we should maybe tlak about what that means with the TC 20:23:41 but I don't want to dilute the other topic 20:23:49 sdague: for interop and end user both 20:24:06 sdague: as a general indicator of "this is difficult in execution with volunteer staffing" 20:24:07 i don't think it needs to be on the agenda though ... 20:24:14 russellb: yeah I heard the tail end of some of that discussion. It doesn't seem to have any traction though 20:24:17 i'm just going to keep bringing up things i don't think need to be on the agenda 20:24:19 so I think ttx's openning sounds like the right place to open that up 20:24:32 russellb: oh yeah that's good. I was attending the last defcore meeting and also was alarmed. 20:24:35 agentle: meh. I just added it as a sub bullet 20:24:37 russellb: weirdly enough it came up in qa spec review too 20:24:41 /meeting/midcycle/ 20:24:46 as long as we time box interop, we can ask questions, and also highlight some things going on that could use more hands 20:24:57 mtreinish: well good that it's on QA radar. i think people are aware and talking about it, so no need to raise alarm yet i suppose ... 20:25:02 sdague: +1 on timebox 20:25:12 sdague: I suppose it's also about scope -- big tent drowns the middle such as API docs 20:25:26 russellb: https://review.openstack.org/301879 20:25:27 sdague : critical technical debt that needs man power like we talked about oslo.messaging/rabbitmq etc 20:25:46 russellb: I wrote "some people" ... there was disagreement at the midcycle 20:25:51 dims: so "areas needing more resources" ? 20:25:52 * russellb thinks we should just use rabbitmq as little as possible isntead 20:25:57 ttx : yep 20:26:20 I'm a little afraid that's going to turn into a giant laundry list 20:26:26 that's a blog post i need to get out of my system ... 20:26:39 yeah, we should have a target 20:26:40 sdague: perhaps it will but maybe it is also true 20:26:46 to be clear, when I talk about volunteer staffing I specifically mean web dev 20:26:48 ttx: ++ 20:26:54 if we want to have a specific tack on this around interop, we could maybe do something about it 20:26:56 like determine top 3 areas and discuss that 20:27:09 ttx: ++ 20:27:22 which would be the top 3 areas struggling at this point that critically need resources 20:27:32 Technical writers working on docs ? 20:27:52 With dims and dhellmann helping with release management, we are covered there 20:28:06 ttx: no more about web delivery issues, we have much of the content and can spread out the content dev 20:28:40 I feel like what we really need is more people working cross-project from within vertical teams 20:28:41 agentle: I'd love to hear more about the web delivery issues - although maybe not in the middle of the TC meeting :) 20:28:42 ttx: sdague's efforts of late are about web dev and nova only 20:28:44 ttx: ++ 20:28:46 mordred: heh 20:28:47 like cross-project liaisons 20:28:47 right 20:28:47 ttx: yes 20:29:10 ttx: that actually came up at the last board meeting too, amusinglyenough 20:29:11 and peopel ready to tackle cross-project work but having a foothold in a project already 20:29:16 ttx: but is that a board discussion? I guess I don't know how the board helps fix that? 20:29:21 we are also struggling with things like quotas and scheduling across projects 20:29:46 sdague: I was thinking influencing them to provide resources there. But I agree it's mostly shouting to the wind 20:29:49 +1 to all cross-project work, but maybe not for board discussion 20:30:14 At the first BoD/TC meeting there was this big push for providing tech writers already 20:30:18 so what can we ask or tell them? (from previous meetings?) 20:30:23 dims: and some other bits related to authz/interproject-communication. mostly the same as quotas and scheduling when you bridge across the services 20:30:37 so - at the last board meeting, the topic of cross-project-with-foothold came up in the context of wanting to do more work with other communities 20:30:37 They all said something like 'we'll provide one each' 20:30:56 ttx: did they say when? 20:30:59 russellb and I pointed out that there was some of that, but it usually involved a person being solidly grounded in at least one place and being an ambassador to the other place 20:31:15 I could see that being a topic the board might want to discusss similarly with us 20:31:25 mordred: ++ 20:31:26 that could be a good discussion then 20:31:30 yeh, that could be useful 20:32:04 I think explaining, again, that we need more people working cross-project than we need features, can only help 20:32:30 well it is the truth 20:32:31 yeah, and in genreal, how to encourage more outreach and collaboration with other communities 20:32:37 ttx : +1 20:32:41 yah. but also that that's not about adding more people as much as it's reducing the feature pressure. just having 100 new devs show up to "work on cross project" is unlikely to be helpful :) 20:32:42 how to incentivize that type of behavior 20:32:48 We could give out stats about developer retention and other community insights 20:33:07 ttx: getting that sentiment grounded in the community is something we need to figure out how to incentivize (x-project, feature pressure reduction, etc) 20:33:31 * morgan is typing slow today as others cover that sentence almost exactly as i was typing it ;) 20:33:38 morgan: magic eh 20:33:46 morgan : true :) 20:33:46 ttx: i know! 20:33:56 i can get out only a few words at at time :) 20:34:12 OK, let's keep the etherpad open at least until Thursday, and I'll collect it and discuss it with Alan then 20:34:18 dims morgan: Welcome to the machine :) 20:34:24 mestery: hehe 20:34:29 so if you think of something else just add it there 20:34:31 ha! 20:34:35 ideally with your name 20:36:10 mordred: how are we doing infra-resource wise. Should we ask for more from the board members ? Or are public clouds not that well represented there anyway ? 20:36:28 public clouds are not well represented on the board 20:37:05 I mean, we can always use more resources - would it be useful to present a report on that perhaps though? 20:37:11 just to get people aware and on the same page? 20:37:21 mordred: yeh, I think so 20:37:25 given how low we are on nodes 20:37:28 yeah, infra is always oversubscribed, but the community's done a not-terrible job of keeping us staffed 20:37:30 can ask for foundation funding if you want to cross that bridge 20:37:31 mordred : +1000 20:37:31 mordred: it would be great to say the names of those donating resources publicly 20:37:43 ++ 20:37:45 anteaya: that too 20:37:51 right 20:38:02 k. I'll sign jeblair up to help me with that. he likes it when I do that 20:38:07 but our current node counts are not going to handle milestone 2 or 3 this time around 20:38:09 ha ha ha 20:38:22 it will impact the ability to land features and bug fixes for the release 20:38:24 sdague: that's good data to get out there ahead of the milestones for sure 20:38:24 i'd like to be able to have our root sysadmins and core reviewers around more often (some of them seem to also be mired in internal company stuff), but we'll take what we can get ;) 20:38:53 sdague: yah - I want to give some more detailed reports to folks 20:38:57 sdague: that is a really good thing to highlight 20:38:59 sdague: so that people can know what is and isn't causing problems 20:39:00 fungi : especially off US work hours :) 20:39:04 and what we do and don't have, etc 20:39:08 dims: yep! 20:39:24 the Bod/TC meeting is also about communicating and educating 20:39:31 anteaya: good point. we do recognize the ones that are at or over our target threshold 20:39:53 might be worth putting together a "this is what we need to use a public cloud for nodepool" doc for the board to send out to corporste members 20:39:58 yup, good to say thanks often 20:39:58 ttx : so update about how we are streamlining release processes and scaling up would help? 20:39:59 #link https://www.openstack.org/foundation/companies/#infra-donors Infrastructure resource donors 20:40:00 OK, we need to move on, but keep on working on that etherpad 20:40:01 wha? 20:40:08 dims: not sure 20:40:16 jeblair: you thought you would get rid of us 20:40:20 jeblair: nice try 20:40:38 Rockyg: http://docs.openstack.org/infra/system-config/contribute-cloud.html 20:40:47 Rockyg: share it liberally 20:40:47 OK, let's cover a few more topics now 20:40:49 wait are we talking about cloud resources? 20:41:02 jeblair: yes, we want more 20:41:12 jeblair: yah. well, specifically about a report at the board/tc meeting 20:41:20 jeblair: so taht we can get people up to speed on the actual situation 20:41:24 great 20:41:29 i wasn't planning on being there 20:41:36 mordred can do it 20:41:40 or fungi 20:41:45 i planned to show up 20:41:49 i've done a lot of work in this area 20:41:54 jeblair: yah. I can do it - I just mostly wanted some help compiling data 20:42:03 but am surprised by this 20:42:15 i would like to be involved 20:42:27 but i did not know we were at the stage of going to the board 20:42:35 jeblair: it's brainstorming BoD topics time, maybe we're not at that stage 20:42:36 we're not going to the board asking for resources 20:42:44 i do suspect that some of the impression we're low on cloud resources is running in a time lag and hasn't caught up with the present 20:42:48 i thought this was something we were going to start talking about amongst ourselves 20:42:57 fungi: that is definitely the case 20:43:16 how about giving just a status update 20:43:34 "how we dealt with the loss of hpcloud" 20:43:43 i will attempt to help how i can, but i would also like to express my strong desire to be involved in this 20:43:59 if that means showing up for meetings like this, it would be nice to have some advance warning 20:44:20 jeblair: we can scrap the idea if that creates more issues than it solves 20:44:29 it doesn't seem to me like it's something so urgent we have to bring it to the board (it's basically solved for now?) 20:44:31 who else is involved? 20:44:35 I'll let you discuss that offline with fungi and mordred 20:44:37 okay 20:44:45 #topic Tidy of item 5 of the vulnerability:managed tag 20:44:49 will do -- we'll see what's necessary, thanks 20:44:51 #link https://review.openstack.org/300698 20:45:13 I'd like to have the VMT's blessing on that one, since this tag is theirs 20:45:29 Feels like it needs to bake a bit more before getting their blessing 20:45:54 so let's revisit that once it has their approval ? 20:45:58 ttx: I think that's reasonable, it's got enough votes once that happens too 20:46:01 it does look like there is still a bit more conversation there. 20:46:11 but i don't think it's going to materially change much 20:46:14 yep, not ready for us yet 20:46:34 stick it back in for a bit longer :) 20:46:36 #agreed wait until it gets VMT's approval 20:46:40 #topic Cross-project workshops at the Austin summit 20:46:44 moreover there is a session at Austin to discuss that item 5 20:46:48 sdague: Could you give us a quick update on that ? All set ? No complaints on the scheduling ? 20:47:00 sure 20:47:02 tristanC: ok, maybe mention htat we should defer that post-Austin 20:47:03 #link https://www.openstack.org/summit/austin-2016/summit-schedule/global-search?t=Cross%20Project%20workshops%3A 20:47:05 sdague: i just want to say the x-project sessions look amazing :) 20:47:22 cross project sessions all scheduled there, I've yet to get hard complaining about conflicts 20:47:38 as we already did a bunch of conflict dodging to get that together 20:47:49 yeah nice work sdague! 20:47:51 yeh, I'm pretty excited by the content in the schedule, this should be good 20:48:09 thanks to all TC members voicing opinions, and dtroyer and ttx for helping on the scheduling 20:48:09 thanks sdague 20:48:11 agentle : morgan : +1 20:48:15 thanks sdague 20:48:23 sdague: I could complain about the overlap between stable and devstack keystone v3 :) 20:48:37 mtreinish: you could :) 20:49:23 questions on that ? 20:49:26 ttx: ok, unless there are questions, I think that's done on topic 20:49:53 #topic Video with advice for design summit session moderators 20:50:04 Not much of an update there, we are "shooting" tomorrow 20:50:10 nice. 20:50:12 We'll send the link to TC for review before sending to big-tent PTLs to disseminate to their session moderators 20:50:47 For those who missed previous episodes, that would make up a short video with tips specifically targeted to session moderators 20:51:05 since with so many project teams we can't assume everyone knows how this works 20:51:13 #topic Open discussion 20:51:25 Alright... open discussion now. Anything you'd like to achieve during the coming 6 months ? 20:51:38 Other topics to add to the BoD/TC meeting agenda ? 20:52:11 I was wondering if this might be an appropriate TC agenda item: 20:52:12 I think post summit i'll have a better idea of the next 6mo and ideas there :) besides what we already put on the BoD/TC meeting agenda 20:52:16 Maybe the newcomers could tell us what they plan to achieve/push over the Newton session 20:52:29 "unifying application development efforts" 20:52:38 docaedo: explain 20:53:07 * mordred steps back slowly from the can of worms docaedo is opening ... 20:53:10 I joined the app dev WG and have been lurking in product WG and following along with much of the conversations 20:53:23 I'm seeing many people have very different versions of what an openstack application would look like 20:53:29 docaedo: ++ 20:53:31 +1 20:53:33 ++ 20:53:34 * jroll pushes mordred back into the mosh pit 20:53:35 yeahhhh docaedo 20:53:38 and would like help getting some focus on this 20:53:49 docaedo: Seems like a nice can of worms right there :) 20:53:53 I'm trying to get educated to see where I can help 20:54:01 those are definitely words 20:54:03 (and I am all about cans of worms :D ) 20:54:03 I'm curious if "openstack application" is meant to be "an app that runs on an openstack cloud" or? 20:54:09 docaedo: +1 20:54:19 jroll: this is the thing 20:54:21 exactly 20:54:25 people say "app developer" 20:54:27 more developer-centric? 20:54:28 jroll: yeah. 20:54:31 jroll, what mordred said 20:54:31 and pretend that it means something 20:54:41 often a lot of the openstack projects assume user = developer. 20:54:41 jroll: yeah this is about the app catalog, heat templates, containers, all of it 20:54:51 are you an openstack app developer if you deploy a production gerrit on an openstack cloud using ansible and puppet? 20:55:10 we want a world where that assumption doesn't hold true. app devloper provides program though the app catalog, and a 'user' launches it on their cloud. 20:55:13 right, I guess idk what's special about an "openstack application" versus a "software application" 20:55:15 or are you only an app developer if you wrote a from-scratch cloud-native app that stores its data in swift and its keys in barbican 20:55:16 without knowing how the thing works. 20:55:17 but ... APPS! 20:55:17 * morgan dislikes "app developer" as a term 20:55:23 morgan: +10000000 20:55:26 My intention is not just to try to make the app catalog relevant - it's to help define who we are trying to reach, and get better at that as a community 20:55:35 docaedo: +1 20:55:46 containerize all the stuff! then everyone is happy :) 20:55:51 russellb: hehehe 20:55:53 right now we're pretty exclusive to non developers. 20:56:05 jroll: yeah, I'm not sure where the distinction is 20:56:10 docaedo: Sounds like an admirable goal, and I think it would help our USERS a lot to clarify that 20:56:12 I'm still confused, I guess, but maybe that's the point 20:56:17 * docaedo notes only 4 minutes left, and appreciates the grenade he lobbed 20:56:19 jroll: It may indeed be :) 20:56:20 you have to be at least devops to use openstack for the most part. 20:56:20 kfox1111: oh I hadn't thought of an app dev that way, interestign 20:56:46 kfox1111: that's a cloud architect writing templates to me, but what I think continues to be shaped heh 20:56:56 kfox1111: I do not agree with you 20:56:59 sounds like some folks maybe have an end goal for openstack clouds to be "push a button, get a deployment of a FOSS application". is that a valid interpretation? 20:57:03 you should be able to have pure ops as an example. cloud dev provides a heat template to launch a scalable trac website. 20:57:05 docaedo: I bet dfflanders would like to speak to you about that 20:57:10 kfox1111: and I think that is one of the biggest misconceptions about openstack 20:57:14 pure op uses it to launch and maintain track sites using openstack. 20:57:15 openstack is a great tool for traditional ops 20:57:20 I'm happy to discuss in person or on IRC further, but mainly want some attention around the issue because it's important to whether we're able to engage USERS of clouds in addition to people who are building clouds 20:57:21 I would like to increase our quality with cross-project efforts that I have already been part of. 20:57:35 ttx: dfflanders and I are talking muchly about it 20:57:48 the infra team uses a lot of openstack public cloud as a traditional operations workflow enhancer 20:57:52 I want end user scientists, chemists, bioligists to be able to use openstack with the app catalog to run "apps" and consume resource. :) 20:57:53 fungi: ++ 20:58:10 docaedo: i think this also plays into the general definiton of what openstack really is (to deployers, end users, developers storing data in clouds). 20:58:10 Things like installation docs being a necessary thing for big tent projects. 20:58:17 thingee: awesome! 20:58:18 fungi: yep - infra is a good example for sure 20:58:19 kfox1111 : interesting 20:58:22 so looks like we should start a taxonomy of "users" 20:58:29 ttx: ++ 20:58:34 ttx: ++ 20:58:35 because we are lacking the words to describe things 20:58:47 and "app dev" and "end user" won't cut it 20:58:48 openstack is an OS. right now its simliar to linux in its infancy. back then, users = developers. 20:58:56 yep 20:58:58 thingee: I agree installation docs are rather important 20:59:00 that has been a recurring theme lately, needing a common vocabularty to be able to discuss these thingss clearly 20:59:01 ttx: absolutely 20:59:04 it sounds to me like maybe the disconnect is that there should also be an audience of people consuming this stuff without needing to know how to be server sysadmins? that seems like a dangerous idea though 20:59:08 kfox1111, dims that's just not true 20:59:12 we need to seperate the two use cases so more users then just developers can consume resource. 20:59:13 like, not even a little bit 20:59:18 mordred: ++ 20:59:20 it's what peopel keep saying 20:59:24 but it's absolutely inaccurate 20:59:29 and I would like to stop the meme 20:59:33 mordred : my yep was on "app dev" and "end user" :) 20:59:36 in one min 20:59:44 there's been multiple discussions about "is an application built on top of openstack, openstack?" 20:59:50 and I think this loops back to that 20:59:56 jroll: wow. that sounds like a fun one 21:00:01 One min left 21:00:05 Personally in Newton (as discussed in my candidacy email) I'd like to have the time to do some tent cleanup, discuss the limits of the tent (i.e. the Poppy/Tacker type issues), and nail down the split event format 21:00:05 jroll: meep 21:00:14 jroll: hope you've got alcohol for that one 21:00:15 ttx: +1 on tent 21:00:18 jroll: openstack is built using servers running on openstack. ;) 21:00:18 * jroll runs away from the grenade he threw 21:00:27 ttx: +1 21:00:29 ttx: sounds good 21:00:39 its true. whats the difference between a rest api and a linux syscall? abstraction? syscalls? both have it. both let you run programs, both provide datastorage abstractions, etc. 21:00:48 Oh, btw, if new TC members would like to get involved in the project team guide, just ask 21:01:10 Also the comms team (flaper87/agentle) would probably not mind some help/rotation 21:01:17 that's a good idea ttx 21:01:24 so feel free to reach out to them 21:01:27 and we are out of time 21:01:33 Thanks everyone ! 21:01:38 thanks! 21:01:38 #endmeeting