16:00:14 <cdent> #startmeeting api-wg
16:00:15 <openstack> Meeting started Thu Jun  8 16:00:14 2017 UTC and is due to finish in 60 minutes.  The chair is cdent. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:00:16 <openstack> Useful Commands: #action #agreed #help #info #idea #link #topic #startvote.
16:00:18 <openstack> The meeting name has been set to 'api_wg'
16:00:33 <cdent> who has joined us today for api-wg meeting?
16:00:42 <edleafe> \o
16:00:58 <cdent> #chair cdent elmiko edleafe
16:01:00 <openstack> Current chairs: cdent edleafe elmiko
16:01:13 <cdent> no dtantsur today, it seems
16:01:22 <cdent> mordred this is a ping in case you want to join the party
16:01:49 <cdent> #topic previous meeting action items
16:02:01 <cdent> #link http://eavesdrop.openstack.org/meetings/api_wg/2017/api_wg.2017-06-01-16.00.html
16:02:18 <cdent> the only action item from last week was for everyone to review monty's stuff, which people did
16:02:32 <cdent> some of the patches at the beginning of the stack might be ready for freeze
16:02:59 <cdent> #topic open mic and new biz
16:03:07 <cdent> no items on the agenda, anyone have anything they'd care to discuss?
16:03:33 <edleafe> politics?
16:03:36 <edleafe> :)
16:03:58 <cdent> I have to go vote this evening. I'm dreading it.
16:04:18 <cdent> screaming into the void
16:04:26 <cdent> #topic guidelines
16:04:36 <cdent> #link https://review.openstack.org/#/q/status:open+project:openstack/api-wg,n,z
16:05:06 <cdent> edleafe: you have opinions on what of the monty stack is ready to freeze?
16:05:12 <edleafe> yes
16:05:25 <edleafe> I thinkk that mordred is the domain expert here
16:05:36 <edleafe> most of the comments have been to clarify the language
16:05:42 <edleafe> not the substance itself.
16:06:05 * mordred waves
16:06:08 <edleafe> So what about freezing at least the first few? We can always address updates as patches to them
16:06:22 <cdent> I'd say the first two have sufficient votes and review, but after that it gets thin?
16:06:35 <edleafe> yeah, people get too exhausted :)
16:06:39 <mordred> I thnk it's good up through cloud-profile from my end - cloud-profile needs a decent amount of more discussion
16:06:46 <mordred> but happy with however ya'll think
16:07:09 <edleafe> I was thinking the first 3 were looking solid
16:07:47 <mordred> cdent: oh - I agree with your comment on patch 4 for sure
16:07:53 <edleafe> haven't looked at the 4th in a while, but it was also pretty solid the last time I reviewed
16:08:26 <cdent> I have a vague hueristic of "does it have more than two +1 votes"
16:08:30 <mordred> I've got most of it implemented in the keystoneauth stack too, fwiw - in case people would rather tlook at code results
16:09:32 <cdent> I'll go ahead and do the first three, it's not like they are being merged, just opened for wider review
16:09:43 <mordred> ++
16:09:49 <edleafe> +1
16:09:58 <edleafe> or should I say, +2 :)
16:13:29 <cdent> okay that's done
16:13:57 <cdent> any comments on other stuff under review?
16:14:12 <edleafe> the naming issue
16:14:24 <cdent> ah right
16:14:26 <edleafe> on the "no change until..." patch
16:14:43 <cdent> I put the -1 on there because nobody else seemed willing to do so, despite not liking the name
16:15:04 <edleafe> How do we resolve this?
16:15:35 <cdent> we could expressly ask for naming hepl in the newsletter?
16:15:48 <cdent> or we could just choose something and not worry about it so much?
16:16:18 <cdent> we could make elmiko decide, since he's not here
16:17:57 <edleafe> Let's try the newsletter. If we don't get a response, we can just pick one
16:18:02 <edleafe> rock-paper-scissors
16:18:03 <cdent> fair
16:18:19 <cdent> #topic bug review
16:18:29 <edleafe> We should include the top options
16:18:33 <cdent> #link https://bugs.launchpad.net/openstack-api-wg
16:18:36 <cdent> that's a good plan
16:19:11 <cdent> I'm assuming that I'll be doing the newsletter given your time constraints today, I'll make sure to include all the best ones.
16:19:26 <cdent> on bugs there's no new bugs and no progress on existing bugs
16:19:55 <cdent> such is life
16:21:15 <elmiko> what am i deciding? bagel flavors for the next bof?
16:21:48 <cdent> in addition to that, you're deciding the name on https://review.openstack.org/#/c/446138/
16:22:10 <cdent> it seems there is sufficient dislike of the name that we ought to try a bit harder to get a better one
16:22:25 <elmiko> gotcha
16:22:35 <edleafe> I still like "not_before"
16:22:36 <elmiko> i loved the suggestion dtantsur had the other day
16:22:37 <edleafe> simple
16:22:50 <edleafe> if anyone is paying attention to it, they know what it means
16:22:57 <elmiko> it was like "broken_after" or some such
16:23:00 <elmiko> XD
16:23:29 <cdent> If I remember right I raised the objection to "not_before" because it has no association with the min_version
16:23:35 <edleafe> EIther that, or "the_min_version_wont_be_raised_before"
16:23:56 <cdent> I like that it is short and simple but it makes me go "not before what?"
16:24:01 <edleafe> cdent: that was my point: anyone parsing the response would know
16:24:15 <elmiko> "not_raised_before" ?
16:24:21 <cdent> however: if the document has a known structure I'm not sure why it matters. It could just as easily the the 24th and 25th bytes in a binary blog...
16:24:33 <elmiko> haha
16:24:37 <edleafe> "drop_dead_date"
16:24:39 <cdent> blob...
16:24:50 <elmiko> a binary blog would be fun though
16:25:03 <edleafe> "update_your_lazy_ass_software_by"
16:25:14 <elmiko> +1
16:26:53 <cdent> instead of not_raise_min_before how about min_raise_not_before ?
16:27:24 <cdent> that addresses dmitry's concern and is not as complicated as dean's suggetions
16:27:26 <elmiko> that actually seems more salient
16:28:04 <edleafe> how about "before_raise_min_not"
16:28:13 <cdent> /kick edleafe
16:28:13 <edleafe> since we're shuffling word order :)
16:28:18 <elmiko> yoda speak, i like it
16:28:27 <edleafe> elmiko: exactly
16:28:29 <cdent> but guuuuuys, I was being serious
16:28:45 <elmiko> i don't have an issue with min_raise_not_before
16:28:52 <edleafe> earliest_min_raise_date
16:29:19 <elmiko> that seems effective too
16:29:44 <edleafe> We can set up a survey monkey page with all the top choices
16:29:54 <cdent> both of those are better than any of the other options thus far
16:30:24 <cdent> I'd rather we just pick one of those, change the doc now, freeze it, and see what the world says?
16:30:54 <elmiko> out of the two, i think earliest_min_raise_date is clearer for me
16:31:04 <edleafe> dunno - I'd rather wait a week
16:31:22 <edleafe> getting feedback from the wider community seems like a good thing
16:31:26 <elmiko> +1
16:31:45 <cdent> that's what freeze means: we have agreed in the group and now want input from the wider community
16:31:47 <elmiko> maybe we could seed the conversation with those 2 as our top choices, but we are open to suggestions?
16:32:10 <cdent> but if we haven't agreed then yeah, asking people is also fine
16:32:12 <cdent> which is it?
16:32:21 <edleafe> I haven't seen much feedback from a freeze
16:32:36 <edleafe> Many of the liaisons are not current
16:32:41 <cdent> long ago and far away there was, but that was when we had active liaisons
16:32:43 <elmiko> i'm ok with giving it a week and asking the wider community their opinions
16:32:44 <cdent> jinxish
16:32:50 <elmiko> lol
16:32:53 <cdent> k, I'll put a few options in the newsletter
16:33:20 <cdent> anything else or shall I start on the newsletter and ping for a proof in a little while?
16:33:59 <edleafe> not from me
16:34:03 <elmiko> were there any action items i missed earlier?
16:34:21 <cdent> elmiko: no sir
16:34:34 <elmiko> ack
16:34:37 <elmiko> thanks
16:35:05 <cdent> cool, thanks for being here
16:35:10 <cdent> #endmeeting