14:00:20 #startmeeting javascript 14:00:21 Minutes: http://eavesdrop.openstack.org/meetings/_javascript/2016/_javascript.2016-08-31-14.00.html 14:00:22 Minutes (text): http://eavesdrop.openstack.org/meetings/_javascript/2016/_javascript.2016-08-31-14.00.txt 14:00:23 Log: http://eavesdrop.openstack.org/meetings/_javascript/2016/_javascript.2016-08-31-14.00.log.html 14:00:24 Meeting started Wed Aug 31 14:00:20 2016 UTC and is due to finish in 60 minutes. The chair is krotscheck. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:00:26 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 14:00:28 There we go 14:00:28 The meeting name has been set to 'javascript' 14:00:35 Good morning everyone! 14:00:36 hi 14:00:38 Hello ! 14:00:44 Wooo good progress this week! 14:00:59 #link Agenda: https://etherpad.openstack.org/p/javascript-meeting-2016-08-31 14:01:09 OUt of curiosity, who's been putting together the agenda each week? 14:01:14 * krotscheck wants to thank them. 14:01:43 #topic Action Followup: jsdoc-sphinx 14:01:48 Yeah, I totally didn't do that. 14:01:57 #action krotscheck Create story to fix bugs in jsdoc-sphinx 14:02:09 #topic Action Followup: MVP 14:02:18 Ok, so lots of meat here. 14:02:24 o/ 14:02:52 We've landed most of keystone. There's a bunch of things that shook out as work on glance and neutron started. 14:03:12 Yeah almost started 14:03:15 cardeois put up a patch to genericize version negotiation. https://review.openstack.org/#/c/363304/ 14:03:32 For glance DSVM jobs are failing but I found the issue and will fix before the end of the meeting 14:04:00 cardeois: Oh? Nice! I was trying to figure out what's going on with this patch -> https://review.openstack.org/#/c/363301/ 14:04:28 Oh yeah, and we've got a patch up that will turn on voting for DSVM jobs. 14:04:47 Glance has been built out all the way to imageList 14:05:10 sorry I'm late 14:05:17 Nova's a bit stuck until we have cardeois' version negotiation patch, because nova registers its services differently 14:05:26 * krotscheck needs to build on that 14:05:49 But with nova, glance, and neutron, that's really all we need to get the functional portion of the MVP done 14:05:58 (We may need to add key management? 14:06:18 The only outstanding part there then is documentation 14:06:35 I do have a design discussion point on the MVP for later. 14:07:03 Any questions on the MVP? 14:07:44 Oh! I forgot. We ripped out all the requestInterception code, as it wasn't being used, and was causing our tests to fail. Once we need it we can always add it back 14:08:31 * msmol is sharing his laptop with cardeois, his keeps crashing 14:09:20 See, now I'm thinking of this: http://vignette3.wikia.nocookie.net/forgottenrealms/images/e/e1/Ettin_-_Sam_Wood.jpg/revision/latest?cb=20070221031301 14:09:30 But I digress, moving on 14:09:52 #topic Action Followup: Documentation 14:09:58 Not complete, will punt 14:10:02 #action krotscheck split design-by-documentation into to-implement segments 14:10:22 I think that's it for actions. Anyone have smoething they did last week that they want to highlight to the rest of us? 14:11:35 well not much, just want to remind that all jobs should be green before merging now. As DSVM jobs were fixed 14:12:06 Right, until https://review.openstack.org/#/c/363200/ forces it 14:12:11 Yes 14:12:11 Alright, moving on. 14:12:12 yeah I've merged lots of stuff with red dsvm jobs, by bad 14:12:24 sure no pb, now you know 14:12:57 #topic Action Followup: Brainstorm Outreach Ideas 14:13:07 So, who's got ideas? 14:13:29 My biggest idea this week was when I put to the list that I'm departing OpenStack, and suddenly everyone was interested in the Javascript SDK. 14:13:53 In particular, TravT from searchlight. 14:14:10 so what you're saying is, you should announce your depature on a more regular basis to keep interest levels high? 14:14:22 msmol: Indeed. Keep coming back and leaving again :) 14:14:50 Anyway- he wants to play with the SDK to see whether he can use it for the all-javascript components in the searchlight-ui. 14:15:06 And he ran smack into the lack of a Readme or documentaiton 14:15:14 So, yeah, we kinda need that before we ship 14:15:20 good point 14:15:47 Also, he mentioned that he'd be able to contribute engineering resources to build out searchlight and murano. 14:16:44 And advocate on our behalf inside of HPE. So I'm going to keep that converstaion going, and walk him through any and all pain points that he may encounter as he explores the SDK. 14:16:53 i.e. a first user :) 14:17:13 Another idea I had is sending out release announcements to the dev list 14:17:22 Did anyone else have ideas? 14:17:41 mhh not really 14:17:59 betherly, anything from you? 14:18:12 * krotscheck remembers that betherly was struggling with a check box yesterday. 14:18:51 not really at the moment sorry! been pretty deep in ironic-ui stuff getting ready for next release 14:19:13 Alrightey! 14:19:21 checkbox issue solved thank goodness. more a case of looking at the same code for too long and not seeing wood for the trees 14:19:51 Coolio. 14:19:54 Moving on 14:20:06 #topic Keystone Configuration object 14:20:18 So, this is something that I encountered while I was building out glance and nova. 14:20:34 Those two services are, by necessity, configured from their entries in the service catalog. 14:20:54 Except... keystone's configured using the clouds.yaml format. 14:21:28 I'd like to propose that keystone also is configured using the same catalog configuration entry, and that that clouds.yaml format be used for the OpenStack SDK wrapper. 14:21:51 That way, Clouds,yaml is used to configure a cloud, and the service catalog is used to configure services. 14:21:56 Any thoughts? 14:22:06 Yeah so I agree keystone should be contructed the same way glance and nova are. That means clouds.yaml is used somewhere else, or is optionnal for advanced users. 14:23:06 vkramskikh? ^^ 14:23:07 sounds reasonable... +1 14:23:43 +1 from me 14:24:16 Alrightey 14:24:25 #agreed use catalog entry to configure keystone 14:24:44 #topic switch to babel-transform-runtime 14:24:52 Who put that on the agenda? 14:25:11 it was me 14:25:25 I think it's necessary before cutting MVP 14:25:32 #link https://www.npmjs.com/package/babel-plugin-transform-runtime 14:25:35 because web bundle currently is about 300kb in size 14:25:54 because it includes the whole babel-polyfill 14:25:57 So, just to clarify.... 14:26:02 oh wow 14:26:15 What this does is it takes all our polyfills, and makes sure they don't end up in the final package. 14:26:27 That'd then require our end users to add them as dependencies themselves? 14:26:38 (Which should be automatic) 14:27:01 no, it just polyfills only necessary parts which are actually used in the porject 14:27:20 that's super cool, good idea vkramskikh 14:27:23 Oh, gotcha 14:27:24 Nice! 14:27:30 for example, we don't use generators, so it won't include polyfills for generators in the resulting bundle 14:27:49 +1 ! 14:27:55 +1 14:28:01 have you tried including it yet? by how much can we expect to shrink? 14:28:13 (Also, generators don't work that well, because apparently babel doesn't play well with yield 14:28:28 no i didn't do this yet for js-openstack-lib, though it's used in fuel 14:29:03 vkramskikh: What was the impact on fuel? 14:29:45 krotscheck: I doubt I can remember exact number, but I think it was about 30-40 kb, but that' because fuel is quite large project 14:29:54 for js-openstack-lib it should reduce the size dramatically 14:29:55 RIght, and we're still pretty small 14:30:15 vkramskikh: Are you able to do that work? 14:30:22 * krotscheck doesn't know what the effort would be. 14:30:34 krotscheck: I think I am, shouldn't be too hard 14:30:41 Awesome! 14:30:51 #agreed swithc to babel-transform-runtime 14:31:05 #action vkramskikh Swithc to babel-transform-runtime 14:31:26 Any othequestions? 14:31:44 nope 14:31:51 ALright. 14:31:56 #topic Open Discussion 14:32:25 So, as of next week I'm volunteer-time-only. 14:32:31 Also, I will not be at the summit, as my talk was not accepte.d 14:32:41 And new-employer will only send me if I have a talk 14:33:05 what new adventures will you be embarking on krotscheck if you don't mind answering ;-) 14:33:36 :( 14:33:38 msmol: I will become an App-On-Openstack developer 14:33:48 Or, well, an App-On-Cloud developer 14:34:06 Cool so that means, you "could" have to use the js-openstack-lib sdk 14:34:10 then I think we should elect more cores, since I cannot spend much time on this project 14:34:13 * msmol googles what that means 14:34:13 "Could". 14:34:30 haha yeah "could" 14:34:36 cardeois: Yes, continuing to work on it willd efinitely be part of my new job, only it won't be part of it immediately 14:34:55 * krotscheck can't name his new employer for legal reasons 14:35:15 vkramskikh: I agree. I nominate cardeois, since he's been the most vocal on reviews, and has submitted patches ;) 14:35:34 +1 14:35:52 +1 14:35:56 cool thanks ! I agree to volunteer 14:36:03 cardeois: Sucker. 14:36:05 ;) 14:36:11 #agreed cardeois elected to core. 14:36:14 :) 14:36:21 should we organize a formal vote or we can just add him to the core group? 14:36:51 vkramskikh: I'll doublecheck with fungi, since we're still technicall under infra. 14:36:52 depends, who has voting rights? current cores only or open to the public 14:36:57 Also, I don't apparently have the rigst to add someone 14:37:10 msmol: Everyone has +1 rights, only cores can +2A and tag a release. 14:37:40 to elect another core is the same? 14:37:55 for infra projects, because we have multiple core review groups, the ptl is the group owner and controls addition/removal of core reviewers 14:38:02 Thought I have no idea what cores means except for the +2. Is there any other responsability that that? 14:38:22 if you want to add core reviewers on one of the js repos in infra, just hit me up 14:38:25 msmol: Ya know, we never settled on that. I always thought that election-by-consensus-of-cores was the thing, unless someone on the public list had a big big problem. 14:38:32 fungi: Danke, willdo :) 14:38:54 fungi: Actually, could you add cardeois as a core to js-openstack-lib? :D 14:38:59 (we just elected him ^^) 14:39:07 For now I think this is fine. 14:39:20 in a general openstack developer community sense, core review teams are basically delegated by ptls. the ptl is in turn elected by all contributors to all repos under their project 14:39:34 krotscheck: sure, just a sec and i'll take care of it 14:39:46 Awesome. 14:40:02 fungi: Thank you! 14:40:09 Anything else for open discussion 14:40:11 ?? 14:40:28 nope 14:40:36 #link https://review.openstack.org/#/admin/groups/1408,members 14:40:43 cardeois has been added 14:40:49 cardeois: Congratulations! 14:40:59 great, thanks fungi and krotscheck 14:41:02 welcome to added responsibility! ;) 14:41:15 Oh gee I hope this new power doesn't get to his head. I still have to sit next to him all day :o 14:41:26 msmol: It's your job to keep him in chEKC 14:41:29 Argh baby fingers 14:41:38 Alright, i'm calling the meeting. Thanks everyone! 14:41:40 #endmeeting