15:00:11 <nils> #startmeeting modularity
15:00:11 <zodbot> Meeting started Tue Feb 19 15:00:11 2019 UTC.
15:00:11 <zodbot> This meeting is logged and archived in a public location.
15:00:11 <zodbot> The chair is nils. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:11 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:11 <zodbot> The meeting name has been set to 'modularity'
15:00:11 <nils> #meetingtopic Weekly Meeting of the Modularity Team
15:00:11 <nils> #topic Roll Call
15:00:18 <nils> .hello nphilipp
15:00:19 <zodbot> nils: nphilipp 'Nils Philippsen' <nphilipp@redhat.com>
15:01:58 <nils> #topic Agenda
15:01:58 <nils> #info #112 Discussion: Module lifecycles
15:01:58 <nils> #info #115 Discussion: Stream branch ownership for packages & modules
15:02:06 <asamalik> .hello2
15:02:07 <zodbot> asamalik: asamalik 'Adam Samalik' <asamalik@redhat.com>
15:02:18 * asamalik waves
15:02:27 <nils> asamalik, any news on these two tickets?
15:02:44 <nils> otherwise, and if there's no last minute topic, we can shortcut this :)
15:03:05 <asamalik> I wanted to have a quick discussion real-time about those before I send out my proposal
15:03:22 <asamalik> but we'd need more people probably :)
15:03:25 <nils> discuss them together or separately?
15:03:28 <nils> yeah :)
15:03:35 <asamalik> separately
15:04:08 <nils> anybody apart from the usual suspects you wanted to discuss this with?
15:04:29 <nils> I see contyk and langdon, sgallagh, ignatenkobrain
15:04:35 <langdon> .hello2
15:04:37 <zodbot> langdon: langdon 'Langdon White' <langdon@redhat.com>
15:04:53 <contyk> .hello psabata
15:04:53 <ignatenkobrain> .hello2
15:04:54 <zodbot> contyk: psabata 'Petr Šabata' <psabata@redhat.com>
15:04:56 <nils> ok, let's kick this off
15:04:56 <zodbot> ignatenkobrain: ignatenkobrain 'Igor Gnatenko' <i.gnatenko.brain@gmail.com>
15:04:57 * contyk was on a wrong server
15:05:02 <ignatenkobrain> I'm always here ;)
15:05:12 <nils> #topic #112 Discussion: Module lifecycles
15:05:12 <nils> #link https://pagure.io/modularity/issue/112
15:05:12 <nils> .modularity 112
15:05:12 <nils> #link https://pagure.io/fesco/issue/2027
15:05:14 <zodbot> nils: Issue #112: Discussion: Module lifecycles - modularity - Pagure.io - https://pagure.io/modularity/issue/112
15:05:16 <nils> #chair asamalik
15:05:16 <zodbot> Current chairs: asamalik nils
15:05:30 <nils> the floor is yours :)
15:06:33 <asamalik> all right, so I posted a proposal about the EOL definition format.. there was some feedback, mostly about the pseudo-syntax, but otherwise it sounds like people are OK with it
15:09:25 <asamalik> thinking about storing it, I was thinking about git vs. fpdc and I'm in favour of the latter because storing it in a database will hopefully prevent typos, make it easier for other tools to consume the info, and potentially make migration easier if we ever need to do change the format
15:10:02 <langdon> you can do both.. git being the truth and fpdc being the cache (and validator)
15:10:18 <nils> querying things is also easier if it's not in git
15:10:21 <ignatenkobrain> what is fpdc?
15:10:35 <asamalik> ignatenkobrain: it's a PDC replacement nirik mentioned in the ticket
15:10:39 <nils> the post-PDC PDC :)
15:10:57 <asamalik> it's still being developed afaik
15:12:01 <ignatenkobrain> I think it would be great to have this info in fpdc and write plugin for pagure to handle it
15:12:06 <ignatenkobrain> so it would show who is maintainer of which branch
15:12:09 <asamalik> langdon: I'd rather do just one.. and the information won't be changed very often
15:12:11 <ignatenkobrain> and allow/reject pushes
15:12:41 <asamalik> ignatenkobrain: that's the second topic... this is about module EOL
15:13:00 <asamalik> but yes, we need to have that info somewhere also
15:13:16 <ignatenkobrain> asamalik: yeah, whatever. it has be shown in pagure
15:13:20 <ignatenkobrain> otherwise it is bad UX
15:15:27 <asamalik> ignatenkobrain: good point.. I'd say that when a stream branch reaches its EOL, and is no longer built, it should probably be retired
15:15:35 <asamalik> and that's how you'd know
15:16:36 <asamalik> but that's just for modules
15:17:14 <asamalik> for RPMs, having an ability to see if a specific stream branch is being built somewhere would be also benefitial
15:17:58 * asamalik takes a note
15:18:01 <asamalik> ignatenkobrain: would that work?
15:18:13 <ignatenkobrain> yes
15:18:22 <ignatenkobrain> I disagree with retiring..
15:18:26 <ignatenkobrain> but the rest makes sense
15:18:50 <asamalik> why do you disagree with retiring?
15:19:37 <mizdebsk> pagure UI already can show which branches are EOL and which are not
15:20:43 <mizdebsk> the difference before EOL branch and retired branch is just presence of dead.package
15:21:06 <mizdebsk> so i don't think it makes sense to retire branches which are already EOL
15:21:08 <asamalik> mizdebsk: ah I didn't know that! now I see it, that's good
15:21:14 <asamalik> ok +1
15:22:55 <asamalik> ok, any other notes? I think we can go to the next topic
15:22:57 <nils> anything that should be #info'd?
15:23:15 <asamalik> nils: I'll send out a complete proposal
15:23:16 <asamalik> so maybe that
15:23:34 <nils> #action asamalik sends out a complete proposal
15:23:35 <nils> ^^?
15:23:53 * asamalik was just writing something similar
15:23:55 <asamalik> thanks nils !
15:24:00 <nils> next
15:24:04 <nils> #topic #115 Discussion: Stream branch ownership for packages & modules
15:24:04 <nils> #link https://pagure.io/modularity/issue/115
15:24:04 <nils> .modularity 115
15:24:04 <nils> #link https://pagure.io/fesco/issue/2028
15:24:06 <zodbot> nils: Issue #115: Discussion: Stream branch ownership for packages & modules - modularity - Pagure.io - https://pagure.io/modularity/issue/115
15:24:07 <langdon> maybe #action asamalik to send out a complete proposal and close the ticket?
15:24:15 <nils> mhm
15:24:18 <nils> #undo
15:24:18 <zodbot> Removing item from minutes: <MeetBot.items.Link object at 0x7fc466914750>
15:24:20 <langdon> and /or open a new ticket for final approval?
15:24:22 <nils> #undo
15:24:22 <zodbot> Removing item from minutes: <MeetBot.items.Link object at 0x7fc465c322d0>
15:24:24 <nils> #undo
15:24:24 <zodbot> Removing item from minutes: <MeetBot.items.Topic object at 0x7fc465c32310>
15:24:25 <nils> #undo
15:24:25 <zodbot> Removing item from minutes: ACTION by nils at 15:23:34 : asamalik sends out a complete proposal
15:24:27 * langdon says sorry
15:24:39 <nils> #action asamalik to send out a complete proposal and close the ticket
15:24:41 <contyk> "asamalik does all the things"
15:24:50 <asamalik> contyk++
15:24:50 <nils> #topic #115 Discussion: Stream branch ownership for packages & modules
15:24:50 <nils> #link https://pagure.io/modularity/issue/115
15:24:50 <nils> .modularity 115
15:24:50 <nils> #link https://pagure.io/fesco/issue/2028
15:24:52 <zodbot> nils: Issue #115: Discussion: Stream branch ownership for packages & modules - modularity - Pagure.io - https://pagure.io/modularity/issue/115
15:25:55 <asamalik> ok, so we definitely need per-branch ownership for both modules and packages, and a way to make bugzilla understand it as well
15:26:23 <nils> contyk, asamalik, https://i.imgflip.com/2u27e0.jpg
15:26:26 <asamalik> again, there are people proposing fpdc which I'm in favor with
15:26:31 <mizdebsk> per-branch ACLs should be easily doable in pdc/fpdc, bugzilla will be more difficult
15:27:16 <mizdebsk> especially for users - how they know which component/version to file bugs against?
15:29:14 <langdon> i don't think you want to force that on users
15:29:18 <asamalik> mizdebsk: I think they should be able to file a bug against a package or a module, and the version to be either a fedora release or a module:stream
15:29:36 <asamalik> but it should be flexible, and yes, as langdon says not overwhelming
15:29:44 <mizdebsk> asamalik, i agree, and this is what i proposed in the ticket
15:29:53 <mizdebsk> but some people disagreed
15:31:15 <asamalik> and there would be more ways to report a specific bug... either to the module, or to the package... depends on how much the user understands the issue... again, lowering the barrier
15:31:45 <asamalik> the maintainers can always reassign stuff to each other I'd say..
15:34:40 <nils> yeah
15:37:01 <nils> asamalik, do you want the same #action here? :)
15:37:02 <asamalik> ok cool... again, there's a lot of stuff in the ticket as well, so I'll just follow up again
15:37:27 <asamalik> nils: sure :)
15:37:38 <nils> #action asamalik to send out a complete proposal and close the ticket
15:37:41 <nils> :D
15:38:02 <nils> or is this not yet ready to close the ticket after the proposal?
15:38:25 <asamalik> I'm not sure either are, I'll use my best judgement
15:38:53 <mizdebsk> i think we wanted fesco approval for this?
15:39:08 <asamalik> mizdebsk: yes, I'll make sure both proposals are on the agenda
15:39:12 <mizdebsk> so i think the ticket should remain open we get a definitive answer from them
15:39:13 <asamalik> once this group agrees that is
15:39:20 <mizdebsk> until*
15:39:25 <asamalik> yeah that's my thinking, too
15:39:39 <asamalik> it'd rather close it when it's agreed upon rather than when it's proposed
15:39:46 <nils> ok
15:39:48 <nils> #undo
15:39:48 <zodbot> Removing item from minutes: ACTION by nils at 15:37:38 : asamalik to send out a complete proposal and close the ticket
15:39:49 <asamalik> but that's fine
15:39:54 <nils> #action asamalik to send out a complete proposal
15:40:17 <nils> good, anything else?
15:40:22 <asamalik> not on this one
15:41:52 <nils> I meant as in open floor topics :)
15:43:37 * asamalik has nothing for the open floor
15:44:34 <nils> good
15:44:56 <nils> Thanks everybody!
15:45:11 <nils> Especially asamalik for doing everything! :D
15:45:12 <langdon> thanks all
15:45:18 <nils> #endmeeting