19:03:34 <bcoca> #startmeeting ansible core public irc meeting
19:03:34 <zodbot> Meeting started Tue Jul 17 19:03:34 2018 UTC.
19:03:34 <zodbot> This meeting is logged and archived in a public location.
19:03:34 <zodbot> The chair is bcoca. Information about MeetBot at http://wiki.debian.org/MeetBot.
19:03:34 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
19:03:34 <zodbot> The meeting name has been set to 'ansible_core_public_irc_meeting'
19:03:53 <bcoca> #topic https://github.com/ansible/ansible/pull/41197
19:04:01 <bcoca> @akasurde?
19:05:01 <bcoca> ok since this is 3rd noshow, im just going to remove from agenda
19:05:20 <bcoca> #topic https://github.com/ansible/ansible/pull/39960
19:05:30 <Tzur> Hi, I'm here
19:05:33 <bcoca> hi
19:06:02 <Tzur> I've updated the code as requested almost 2 months ago, and I wanted to know what can we do to get it reviewd? IBM wants to contribute more code but blocks us.
19:06:14 <bcoca> wasnt mkrizek reviewing?
19:06:16 <Tzur> but thiis*
19:06:22 <bcoca> @mkrizek ping?
19:06:52 <bcoca> its late for him
19:06:56 <Tzur> I'm guessing he was? but he didn't comment when I tagged him..
19:07:04 <bcoca> did you ping him in irc?
19:07:30 <Tzur> no,  only on the PR because I did not know what is the apropriate way to reach a core member.
19:07:31 <bcoca> many of us don't get notifications from github or we drown in them, so it is easy to miss those
19:07:48 <Tzur> oh,ok. So I'll try to catch him on IRC?
19:08:16 <bcoca> pinging on irc as a followup is ok, #ansible-devel is probably best place, try early in the day as i think he is in european time
19:08:27 <bcoca> if no luck, bring it back up on thrusday
19:08:35 <Tzur> ok, sounds good. Thank you for your time.
19:09:27 <bcoca> np, hopefully we can get you merged soon
19:09:42 <bcoca> #topic https://github.com/ansible/community/issues/291#issuecomment-355016399
19:10:05 <bcoca> @abadger1999 @dag you guys were going to rename this a while back?
19:10:13 <bcoca> @itdependsnetwork
19:10:57 <dag> Not sure what the status is...
19:11:17 <bcoca> well, i had rename PR you guys vetoed, but then never followed up with new name PR
19:11:27 <bcoca> its still causing same problem
19:12:46 <dag> I think gather_facts as module name was ditched because it conflicted with play-directive of same name
19:13:37 <bcoca> no, mostly you wanted shorter, iirc you also wanted to rename the keyword
19:13:55 <dag> that was my reason, but wasn't withheld
19:14:04 <dag> IIRC
19:14:51 <bcoca> well, several peopple disliked teh name, it was agreed to kill my PR and then submit alternatives .. but only the first part happened
19:14:53 <dag> I didn't ask for a rename of the directive
19:15:20 <bcoca> i remember otherwise, but willing to concede
19:15:35 <dag> I guess we need the relevant logs, having the same discussion again is pointless if we can't learn from the previous debate
19:15:35 <bcoca> not important, theimportant part is renaming the module so it doesnt conflict with setup.py
19:16:45 <agaffney> is "conflicting" with the play directive really even a problem?
19:16:57 <bcoca> no it was never due to 'conflict'
19:17:04 <bcoca> it was people not liking the 'gather_facts' name
19:17:31 <dag> agaffney: confusing and to be avoided
19:18:00 <bcoca> dag: i would say the contrary, since it was the module backing the keyword
19:18:03 <agaffney> dag: but the same is true of 'user' without causing significant confusion
19:18:15 <bcoca> he, i remember using that exmaple also
19:18:48 <bcoca> still, we are not rehashing the old discussion, but trying to get the action item done
19:18:50 <dag> agaffney: to be avoided (maybe that was unavoidable ? :-))
19:19:35 <agaffney> I wasn't part of the original discussion, so I have no idea which horses have been beaten to death
19:19:36 <bcoca> i really dont want this to have died by bikeshed, if no one else does an alternative, i'll just merge the original, which fixed the issue
19:19:51 <bcoca> @agaffney they all were
19:20:48 <bcoca> @thaumos you were also involved in that one, thought you had a favorite, but it never materialized
19:20:57 <dag> if we are deciding now on whatever basis regardless of the previous discussion, we might as well not have any debate with a lot of people
19:21:34 <bcoca> im fine with debate, what im not fine is with people shooting it down and promising alternatives and later abandoning it
19:21:48 <dag> In that case i'd like to bring up https://github.com/ansible/proposals/issues/74 again
19:21:50 <dag> :)
19:21:55 <bcoca> im even fine with gather_facts not used, i just added it as a 'logical' name, just got mired in bikeshed
19:22:27 <dag> we should have a PEP process where ideas are refined, and eventually all sub-options are voted on
19:22:29 <bcoca> that had an agreement, you just didnt like it and then declined to implement
19:22:46 <dag> bcoca: there was no agreement on that one, only a veto of abadger
19:23:03 <bcoca> he veteod this, not the full idea, unless im misremebering
19:23:12 <bcoca> 'this' as the keyword, but not the concept
19:23:16 <dag> bcoca: there was no decision
19:23:22 <dag> there was no vote
19:23:29 <bcoca> wasn't it ansible_task?
19:23:31 <dag> only a veto for "this"
19:23:41 <dag> nope
19:24:19 <bcoca> its tangent anyways, not going to look at logs now
19:24:30 <jhawkesworth_> is this the PR: https://github.com/ansible/ansible/pull/20717
19:24:48 <bcoca> yep
19:24:59 <dag> jhawkesworth_: discussion was in a meeting
19:25:15 <bcoca> well, discussion was all over, but decision was made at meeting
19:25:41 <jhawkesworth_> right
19:25:57 <bcoca> and there were discussions after also
19:26:22 <agaffney> I really like the idea/implementation and have no real opinion on the name. it's not like most people will ever type it explicitly anyway
19:27:22 <jhawkesworth_> with my 'ansible user' hat on I have like 3 places I do explict fact gathering so it would not be a big issue to fix my playbooks whatever new name is
19:27:37 <bcoca> old name will still work for a while
19:27:50 <bcoca> the whole point was adding an alias to setup and a deprecation period
19:28:07 <jhawkesworth_> oh sure, I just try to keep on top of syntax changes
19:28:37 <bcoca> even if we alias to 'setup' w/o .py it would 'still work' and not conflict with the setuptools expectation
19:29:47 <bcoca> https://github.com/ansible/ansible/pull/20717#issuecomment-277231501 <= dag you did propose rename, on ticket
19:30:17 <dag> bcoca: I did, I never said I didn't
19:30:40 <bcoca> [15:13] <dag> I didn't ask for a rename of the directive
19:30:57 <dag> the directive != the module
19:31:05 <agaffney> that confused me too. I think he meant the play directive
19:31:22 <agaffney> I just had to go back and make sure I wasn't crazy
19:31:52 <dag> That was a response to: (21:13:38) bcoca: no, mostly you wanted shorter, iirc you also wanted to rename the keyword
19:31:56 <dag> *also*
19:32:32 <dag> I do not like gather_fact indeed, we have _facts modules but they all are about a specific topic
19:32:35 <bcoca> either im reading the post wrong, or that is exactly what you ask for, rename parameter AND module
19:32:43 <dag> so setup_facts would be a better alternative than gather_facts
19:32:49 <dag> someone proposed that IIRC
19:32:54 <dag> but the meeting logs would be nice
19:33:27 <dag> bcoca: ah, indeed, but I was talking about the meeting
19:33:30 <bcoca> many things were proposed, which is why i gave up and let you guys with the proposals submit your ideas, but none of you did
19:33:33 <dag> this is at least a year earlier
19:33:38 <agaffney> setup_facts seems...wrong somehow
19:34:14 <agaffney> to me, it makes a lot of sense for it to match the play directive
19:34:26 <bcoca> ssshss
19:34:36 <dag> look, it makes no sense to decide now if we're going to disregard the feedback from 20 people
19:35:06 <agaffney> let's just call it Fred and move on
19:35:44 <agaffney> the name seems like one of the least important parts
19:36:04 <agaffney> it matters for UX and not much else
19:36:28 <bcoca> well, it matters to setuptools and it breaks setups that rely on git/submodules for projects
19:36:39 <bcoca> that also have 'buildable' python
19:36:45 <itdependsnetwork> just catching up
19:36:51 <dag> https://meetbot.fedoraproject.org/ansible-meeting/2018-01-09/ansible_core.2018-01-09-19.00.log.html
19:37:03 <itdependsnetwork> I agree, +1 for fred
19:37:15 <dag> who's fred ?
19:37:28 <bcoca> rename of setup.py
19:37:31 <itdependsnetwork> agaffney: 's friend :)
19:37:49 <itdependsnetwork> actually, facebook just came out with the fred package management
19:38:20 <itdependsnetwork> let's go with fredrick
19:38:28 <bcoca> federico
19:39:09 <itdependsnetwork> lol
19:39:15 <itdependsnetwork> in all seriousness, just not setup
19:39:45 <bcoca> well, i tried, still unsure wth to do with it now, since those with opinions on name dont want to submit anything
19:40:00 <agaffney> is there any reason not to just stick with 'setup'?
19:40:09 <dag> bcoca: where was it stated to submit anything ?
19:40:23 <dag> bcoca: the meeting logs doesn't mention anything about new names to be submitted
19:40:38 <dag> IIRC it was just postponed for the release without decision
19:40:39 <bcoca> was followup with thamos, who took point iirc
19:41:12 <itdependsnetwork> since setup.py is a common usage for python package management
19:41:28 <bcoca> i dont remember exactly where things were said, mostly that they were said
19:41:57 <dag> bcoca: same problem...
19:42:16 <dag> it's called: getting old :-/
19:42:21 <itdependsnetwork> gather_facts seems to make sense logicially and namespace wise
19:42:34 <bcoca> nah, i had that problem since i was 6, can remember facts, but not their origin
19:42:52 <dag> let me count the votes in that meeting log, and let's stick with that as a decision
19:43:11 <itdependsnetwork> there was a lot of fred thrown around then
19:43:12 <dag> and put that decision up for review in the meeting notes
19:43:27 <itdependsnetwork> But we agree to change?
19:43:56 <bcoca> i dont think there ever was disagreement, afaik the name was biggest issue, then people wanted  to redo/refactor facts .. and that happened, but they want to do it again now
19:44:25 <bcoca> which i believe is why thamous stepped in, to try to work it into facts refactor .. but that also didnt happen
19:44:32 <itdependsnetwork> let me put it another way, we agree to change for 2.7?
19:44:38 <stoned75> reap_facts
19:44:45 <bcoca> uf
19:46:25 <bcoca> idk what to do at this point to move it forward, im jsut pissed that this is stalled for as long as it has when it is not hard to fix
19:46:35 <itdependsnetwork> I concur
19:46:38 <bcoca> i can  imagine @itdependsnetwork feels the same (kudos on your patience)
19:48:29 <itdependsnetwork> I'll take silence as gather_facts.py is gtg
19:48:35 <itdependsnetwork> ship it
19:48:41 <itdependsnetwork> (not sure how this works)
19:49:12 <bcoca> we have not formalized, but i would say we need at least quorum that includes most of those that voted last time
19:49:59 <jhawkesworth_> I suggest best way to move it forward is to reopen/bring forward bcoca's PR, and ping those to voted last time.
19:50:39 <dag> gather_facts: bcoca, jimi, maximilion
19:50:39 <dag> facts: dag, alikins, abadger, nitzmahone
19:50:39 <dag> no change: jimi, sivel
19:50:42 <jhawkesworth_> =those who voted
19:50:45 <dag> that's what I counted
19:50:57 <agaffney> if it's open to new votes, it gets +1 from me
19:51:10 <dag> thaumos and sivel didn't make clear their preference AFAICT
19:51:17 <agaffney> for the PR as it is (gather_facts)
19:51:29 <dag> maximilion was basically fine with both, but preferred gather_facts
19:51:32 <bcoca> they did, but not at time of vote
19:51:48 * bcoca does not have that branch anymore ...
19:51:50 <dag> bcoca: I haven't seen a vote, was that before or after this meeting ?
19:52:03 <dag> because in the meeting minutes there are references to an earlier discussion
19:52:12 <agaffney> bcoca: easy enough to extract from the PR
19:52:46 <bcoca> iirc, discussion went on for days, in ticket. #ansible and #ansible-devel, didnt really need vote to see ticket was going to be rejected
19:53:03 <dag> 19:20:59 <abadger1999> gundalow: change the name there from gather_facts to facts and I think we had agreement from alikins and i that it would be okay.
19:53:03 <dag> -snip-
19:53:03 <dag> 19:22:05 <abadger1999> I'm okay with it happening.
19:53:03 <dag> 19:22:15 <abadger1999> As loing as we can agree on name.
19:53:03 <dag> 19:22:23 <abadger1999> that's where we got stuck last time.
19:53:03 <dag> -snip-
19:53:04 <dag> 19:22:38 <abadger1999> I think everyone except bcoca agreed o nthe name but bcoca had been the one writing the PR?
19:53:29 <bcoca> did we agree on 1 name? thought we had 2 or 3
19:53:45 <dag> that's from the meeting logs about an earlier discussion/vote
19:53:51 <dag> I wasn't involved in that one
19:53:56 <bcoca> i thought the only agreement was 'not gather_facts'
19:53:59 <dag> https://meetbot.fedoraproject.org/ansible-meeting/2018-01-09/ansible_core.2018-01-09-19.00.log.html
19:54:11 <itdependsnetwork> facts is just kicking the can down the road, going to have name space issues
19:54:28 <bcoca> well, facts are now namespaced under ansible_facts
19:54:40 * bcoca needs to add deprecation soon
19:54:57 <dag> (which I would have called 'facts' as well ;-))
19:55:05 <itdependsnetwork> so do I not have the setup.py issue?
19:55:05 <itdependsnetwork> If so, see you guys later
19:55:35 <dag> can't we simply make 'setup' work without calling the file 'setup.py' as was suggested for the alias ?
19:55:57 <dag> at least the issue is unblocked
20:02:26 <dag> https://github.com/ansible/ansible/pull/20717#issuecomment-405709479
20:02:47 <bcoca> sorry, my router seems to be alergic to lightning
20:02:55 <bcoca> did i miss anything?
20:03:04 <itdependsnetwork> yea, that storm was crazy
20:03:14 <bcoca> you in NJ?
20:03:20 <itdependsnetwork> NY right now, but from NJ
20:03:31 <bcoca> he, its like nightfall right now
20:03:56 <itdependsnetwork> yea, turned around and it was dark
20:03:57 <bcoca> ok, so i don think we are resolving anything, i'll try to ping thamos (he took point on this last time) and see what happens
20:04:09 <itdependsnetwork> ok, thanks
20:04:26 <bcoca> sorry about that, really irks me such a simple change is taking sooo long
20:04:30 <bcoca> #endmeeting