16:00:01 #startmeeting api sig 16:00:01 Meeting started Thu Sep 20 16:00:01 2018 UTC and is due to finish in 60 minutes. The chair is cdent. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:00:02 Useful Commands: #action #agreed #help #info #idea #link #topic #startvote. 16:00:04 The meeting name has been set to 'api_sig' 16:00:14 o/ 16:00:16 #chair cdent elmiko edleafe dtantsur 16:00:17 #link https://wiki.openstack.org/wiki/Meetings/API-SIG#Agenda 16:00:18 #topic previous meeting action items 16:00:18 Current chairs: cdent dtantsur edleafe elmiko 16:00:23 \o 16:00:59 #link last meetings mintues http://eavesdrop.openstack.org/meetings/api_sig/2018/api_sig.2018-09-06-16.00.html 16:01:27 o/ 16:01:56 it's been a long time since the last in-irc meeting 16:01:58 how was ptg? 16:02:04 so not much in the way of action items 16:02:05 haha 16:02:07 so yeah ptg 16:02:10 Lonely without you! 16:02:13 =( 16:02:19 i missed you all as well 16:02:43 both ed and I made some ptg reports that include api-sig bits: 16:02:55 #link cdent ptg report https://anticdent.org/drafts/openstack-stein-ptg.html 16:02:57 Didn't we assign the task of merging MOnty's stuff to dtantsur ? 16:03:09 and I did update the looooong patch 16:03:19 #link edleafe ptg report https://blog.leafe.com/stein-ptg-recap/ 16:03:21 yeah, i've been re-reading the review, but it's a lot 16:03:32 yeah, now you have to read it, poor souls 16:03:36 ++ cdent and edleafe for writing up reports =) 16:04:16 elmiko: it helps me justify IBM paying for me to go :) 16:04:53 ++ 16:05:33 wow, this linked tweet from Alex Xu just sounds horrible =( 16:05:37 I guess the main event at the monday session was people re-joining the healthcheck effort 16:05:45 ooh, nice! 16:05:49 elmiko: oh yeah, my post is full of joy 16:06:39 it's not clear how much of the healthcheck enthusiasm will expand beyond the boundaries of the ptg 16:06:43 it's just a shame to hear someone say how they don't want to attend another ptg but have to for work, i always found the community events to be a joyous part of openstack for me 16:06:59 keep reading! 16:07:01 ack 16:07:06 X2 16:07:17 #topic new biz/opens 16:07:25 #link https://etherpad.openstack.org/p/api-sig-stein-ptg 16:07:35 we've already started talking about that. anything else to add 16:07:38 i think the healthcheck idea is really nice, but yeah tough if no one has the time to work on it 16:08:19 did you actually get to talk about microversions? 16:08:29 not that I recall 16:08:51 saw it on the agenda and just wondered out of sick curiosity XD 16:09:13 not a single microversion was discussed 16:09:17 ++ 16:09:32 also cool to see the action about the IETF HTTP protocols 16:09:55 we didn't talk about that much, other than to say it is a useful additional resource 16:10:10 yeah 16:10:18 it seemed pretty uncontroversial 16:10:44 looks like some great discussions, were there a lot of people who stopped by? 16:11:23 It was much thinner attendance than in the past 16:11:36 i like the comment on L50 as well, that sounds like a good call to action for the sig 16:11:38 overall the entire prtg was thin 16:11:44 gotcha 16:11:45 but our room especially so 16:12:19 The healthcheck discussions did pick it up a bit 16:12:29 seems like it from the notes 16:12:50 the stuff on L50 was a theme that I heard a few times during the week. overall openstack needs to be better about shouting about the coolness 16:13:03 makes sense 16:13:04 brb 16:13:54 Trouble is, APIs just aren't that sexy or cool 16:13:59 right 16:14:17 yeah 16:14:17 unless we overhaul everything to use grpc! XD 16:14:26 and something about edge 16:16:09 don't forget the far edge 16:17:01 or the fog edge! 16:17:36 yeah, I only learned about far edge today 16:17:42 are there any big actions that we need to take as a group coming out of the ptg? 16:18:24 not that I could tell, but I have a sort of related thing 16:18:34 cool 16:18:42 I guess I should write up a thing about the DELETE concepts 16:19:04 #action edleafe to write a guideline on API fault tolerance 16:19:04 i was gonna ask about that, it sounds like an interesting topic (fault tolerance) 16:19:26 I want to finish the book first :) 16:19:34 ack =) 16:19:38 * edleafe forgot to pack it for the flight 16:19:42 doh! 16:20:10 PTG was a bit of a come to jesus moment for me with regard to the extent of my too-thinly-spread attention, so I'm considering stepping back from (but not leaving entirely) the SIG and I'd either like your blessing or to be convinced I should not 16:20:33 oh wow, that's a big one 16:21:09 I think that this is a good time, as we really don't have very much on our plate 16:21:24 yeah, that's kind of what I was thinking too 16:21:37 well, our work is pretty low bandwidth, but if you feel it's negatively impacting your time with other stuff then i certainly encourage you to make the decision that puts you working on the stuff you want to 16:22:00 * dtantsur wanted to say the same ^^^ 16:22:02 s/want/have 16:22:07 :-P 16:22:11 it's more in terms of the mental gesture: "it's okay to set this aside" 16:22:12 fair ;) 16:22:39 cdent: it is okay to set this aside =) 16:22:41 cdent: totally ok 16:22:48 cool, thanks 16:23:09 * cdent has a bit of a good will hunting moment 16:23:20 hehe =) 16:23:21 Now we'll have someone to gossip about! 16:23:26 LOL 16:24:08 yay! gossip! 16:24:23 huzzah! 16:24:40 next topic: do we need forum space in berlin? 16:24:50 i got a mail about that, not sure if others did 16:24:55 i saw it too 16:25:04 i doubt i will be in berlin 16:25:11 Right now, no. If the healthcheck or GraphQL projects take off suddenly, then perhaps 16:25:18 ++ 16:25:37 I can lead any sessions if needed 16:25:49 cool, that's good to know 16:26:15 Yeah, I'm still 50-50 on whether I can go 16:26:20 if I catch mordred, we may end up finishing The Catalog Discovery Book 16:26:22 maybe at the least we would have a brief update session if dtantsur wanted to give it 16:26:35 elmiko: have we done anything? :D 16:27:17 aye, there's the rub 16:27:32 yeah... 16:27:43 * dtantsur imagines: Hi all, I'm from API-SIG. cdent is too busy, edleafe is huzzah! elmiko and I are just slackers. Bye! 16:27:45 well, this naturally kinda makes me wonder about the sig and our future 16:27:50 I'm not a fan of "let's have a space because we can" 16:28:00 edleafe: totally agree 16:28:17 elmiko: it's not really a consolation but many/most sigs are wondering the same thing 16:28:38 i'm just kinda wondering if there is some signal we put up that we are migrating to maintenance mode 16:28:42 ? 16:28:52 I wonder if something like "Ops feedback on OpenStack API design" can fly 16:28:54 that's a good question 16:28:57 indeed 16:28:58 elmiko: I don't think that's necessary 16:29:16 edleafe: ok 16:29:25 We still need to be available for things that come up, even if we aren't churning out new guidelines 16:29:32 definitely 16:29:43 and i think our biggest impact of late has been facilitating others 16:29:49 just a reminder: finishing the discovery guidelines will probably occupy us for a cycle more 16:30:07 but like, should we consider migrating our meetings to like an open office hours type of thing? 16:30:33 So how about this: let's keep this weekly meeting, but more as a simple check-in with each other. If there isn't anything to do, then end it quickly, and no need for a newsletter 16:30:56 that sounds reasonable to me 16:31:02 If some event does demand attention, we can revert to the format we've been using for a while 16:31:07 I'm definitely in favor of it, esp. since the meeting starts at 6pm 16:31:20 I really don't like newsletters that say "We didn't do crap" 16:31:22 of course it is also an opportunity for folks to reach out to us as well, but that hasn't happened during meeting hours in a while 16:31:33 edleafe: ++ 16:31:51 I wonder if we should have something in Berlin, even if only to highlight that we exist 16:32:07 elmiko: well that's why we should keep this regular meeting, in case someone does want to reach out 16:32:20 edleafe: right, that was kinda my point about office hours 16:32:48 dtantsur: i think if you are willing to host something, i am not opposed to that. do they birds of feather meetups at the forums? 16:32:57 *do they have* 16:33:00 elmiko: hmmm, then maybe that will be the way to go. Office hours in -sdks 16:33:02 I think we've been operating on a premise somewhere in the realm of "if we build it they will come" but we may need to adjust to "if there's a problem they will come" (not just here, but in general) 16:33:08 I'm surprisingly ignorant to the Forum format 16:33:23 the fact that I haven't been to a single one before may contribute 16:33:24 cdent: zactly 16:33:25 it's not just you dtantsur . it keeps moving around 16:33:32 cdent: yeah, i think those days of using drawing a crowd are probably past us 16:33:45 RIP Microversion Controversy 16:33:51 \o/ 16:33:52 hahaha 16:35:40 so, i'm ok with keeping this meeting time as a touch point, but i would also be ok with announcing some sort of office hours in -sdks as a replacement 16:36:25 * cdent has no preference 16:36:25 We could make two office hours, so that dtantsur doesn't have to miss dinner 16:36:25 maybe give an indication of deprecation before we yank this location/time though 16:36:26 a good thing about office hours is that you can have them in different time zones 16:36:31 edleafe: ++ 16:36:32 edleafe: exactly :) 16:36:46 it would also help with the asia times 16:37:36 OK, so how does this sound: we have one more meeting next week, where we will set office hours. We announce that in today's newsletter, and anyone who cares can provide input, either on the ML or at that meeting 16:37:44 what about those of us who don't have an office though? 16:37:59 edleafe: i'm good with that 16:38:01 mordred: you've used up all your attendance tokens 16:38:18 mordred: where we're going, we don't need offices.... XD 16:38:32 elmiko's was better 16:38:44 also, i thought you were calling it quits for today after your previous triumph in -sdks? 16:38:46 mordred: I'm more concerned with those of us who don't have hours 16:38:53 LOL 16:39:10 :D 16:39:42 edleafe: hours and offices are both overrated 16:39:52 elmiko: :) 16:40:49 ed's plan sounds good, should we move on? 16:40:54 ++ 16:40:59 #topic guidelines 16:41:05 dtantsur did some stuff 16:41:12 #link https://review.openstack.org/#/q/status:open+project:openstack/api-sig,n,z 16:41:28 i have not attended to me "design advice" thing 16:41:31 or rather messed with mordred's stuff 16:41:59 * mordred is TOTALLY enjoying reviewing that 16:42:10 i'm still reading dtantsur's update, but things have heated up locally for me so i am slow 16:43:22 so nothing to freeze or anything along those lines, correct? 16:43:38 yup 16:43:41 #topic bug review 16:43:41 #link https://storyboard.openstack.org/#!/project/1039 16:43:52 nothing new 16:44:23 #topic weekly newsletter 16:44:23 #link https://etherpad.openstack.org/p/api-sig-newsletter 16:45:00 I guess I can do it 16:45:05 i have a meeting right after this one, but i can take it if no one else has time 16:45:12 thanks edleafe =) 16:45:18 edleafe++ 16:45:24 huzzah to edleafe 16:45:44 * edleafe loves his huzzahs 16:45:48 also huzzah to cdent for many many great contributions to the sig and openstack =) 16:45:56 huzzah! 16:46:01 I'm not leaving, just fading, but thanks 16:46:14 i know, but still, i feel it's worth mentioning 16:46:53 ✔ 16:46:57 I should write something like that and make it sound like a eulogy 16:47:05 hhahaha 16:47:09 * cdent is full of dread 16:48:24 anything else or shall we call it? if we follow the plan, next week will be the last official meeting for a while. I'll bring virtual booze. 16:48:35 nothing else from me 16:48:49 i will be at a conference next week, but i will do my best to make the meeting 16:48:59 I'm good to go 16:49:03 even if i have to use the web interface from my phone! 16:49:06 rad 16:49:08 thanks 16:49:12 #endmeeting