15:00:00 <bowlofeggs> #startmeeting FESCO (2018-02-16)
15:00:00 <zodbot> Meeting started Fri Feb 16 15:00:00 2018 UTC.  The chair is bowlofeggs. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:00:00 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:00:00 <zodbot> The meeting name has been set to 'fesco_(2018-02-16)'
15:00:00 <bowlofeggs> #meetingname fesco
15:00:00 <zodbot> The meeting name has been set to 'fesco'
15:00:00 <bowlofeggs> #chair maxamillion dgilmore nirik  jsmith sgallagh bowlofeggs tyll jwb zbyszek
15:00:00 <zodbot> Current chairs: bowlofeggs dgilmore jsmith jwb maxamillion nirik sgallagh tyll zbyszek
15:00:00 <bowlofeggs> #topic init process
15:00:07 <jsmith> .hello2
15:00:08 <zodbot> jsmith: jsmith 'Jared Smith' <jsmith.fedora@gmail.com>
15:01:41 <sgallagh> .hello2
15:01:42 <zodbot> sgallagh: sgallagh 'Stephen Gallagher' <sgallagh@redhat.com>
15:03:20 <bowlofeggs> maxamillion, nirik, tyll, jwb: are you able to make today's meeting?
15:03:30 <bowlofeggs> dgilmore said he had a conflict
15:03:38 <bowlofeggs> i'll ask zbysek in #fedora-devel
15:04:24 <nirik> morning
15:04:35 <bowlofeggs> alright, we need just one more for quorum
15:05:04 <zbyszek> Hi, sorry for being late.
15:05:26 <zbyszek> I hope we can find some other time ;)
15:05:28 <bowlofeggs> excellent, we have quorum now
15:05:41 <bowlofeggs> yeah it has been hard for us to settle on a time this round
15:06:06 <nirik> it usually is, but has gotten worse
15:06:08 <bowlofeggs> let's get this meeting started. mostly i think we have faster things today, with only one topic that might take some time
15:06:21 <bowlofeggs> #topic #1767 F28 Self Contained Changes
15:06:22 <bowlofeggs> .fesco 1767
15:06:22 <bowlofeggs> https://pagure.io/fesco/issue/1767
15:06:24 <zodbot> bowlofeggs: Issue #1767: F28 Self Contained Changes - fesco - Pagure - https://pagure.io/fesco/issue/1767
15:06:33 <bowlofeggs> for this one, i'm going to experiment with giving each change its own #topic
15:06:39 <bowlofeggs> #topic #1767 F28 Self Contained Changes: VA-API 1.0.0
15:06:40 <bowlofeggs> https://fedoraproject.org/wiki/Changes/VA-API_1.0.0
15:06:54 <ignatenkobrain> .hello2
15:06:55 <zodbot> ignatenkobrain: ignatenkobrain 'Igor Gnatenko' <ignatenko@redhat.com>
15:07:10 <sgallagh> bowlofeggs: The usual policy is we only bother with separate topics if any are controversial
15:07:21 <bowlofeggs> sgallagh: ah ok
15:07:25 <sgallagh> Otherwise, these are generally just rubber-stamped as a group for expediency
15:07:28 <bowlofeggs> #topic #1767 F28 Self Contained Change
15:07:38 <bowlofeggs> .fesco 1767
15:07:39 <nirik> +1 on all
15:07:41 <zodbot> bowlofeggs: Issue #1767: F28 Self Contained Changes - fesco - Pagure - https://pagure.io/fesco/issue/1767
15:07:51 <jsmith> I'm +1 on all three changes
15:08:00 <zbyszek> So... I'm +1 on all changes, but I want to voice my discontent with the process
15:08:07 <sgallagh> +1
15:08:11 <sgallagh> Which process?
15:08:16 <bowlofeggs> i too am +1 on all three
15:08:24 <zbyszek> Removing ldconfig scriptlets is clearly a system wide change
15:09:00 <ignatenkobrain> it's not
15:09:08 <zbyszek> I think that labelling it as a non-system wide change skirts around the process
15:09:11 <ignatenkobrain> the Change doesn't imply changing all packages
15:09:16 <ignatenkobrain> but giving ability to drop scriptlets
15:09:35 <zbyszek> A significant percentage of packages is being touched.
15:10:02 <bowlofeggs> zbyszek: they dont' *have* to be touched though
15:10:16 <jsmith> I don't see it as trying to skirt any type of oversight
15:10:18 <bowlofeggs> this just gives them a new ability
15:10:36 <sgallagh> Yeah, the change as written describes adding support for these macros.
15:10:39 <ignatenkobrain> mass touch of packages is planned for F29
15:10:48 <sgallagh> The mass-modification is a separate thing
15:10:53 <bowlofeggs> it also doesn't affect end users
15:11:27 <bowlofeggs> i think of a system wide change as being something most end users should pay attention to
15:11:39 <bowlofeggs> i mena, not by definition, but commonly
15:11:43 <zbyszek> No, not really. A systme wide change is something that requires coordiantion of packagers, also.
15:11:52 <bowlofeggs> true
15:11:54 <bowlofeggs> this one doesn't though
15:12:15 <zbyszek> OK, so let's leave it at that.
15:12:34 * bowlofeggs counts the votes
15:12:41 <bowlofeggs> looks like 5
15:12:45 <nirik> well, it might affect end users by making installs faster. :)
15:12:48 <nirik> but yeah.
15:12:56 <bowlofeggs> #agreed All three approved (+5, 0, -0)
15:13:00 <bowlofeggs> true
15:13:11 <ksinny> yay!
15:13:17 <bowlofeggs> #topic #1820 Adjust/Drop/Document batched updates policy
15:13:18 <bowlofeggs> .fesco 1820
15:13:18 <bowlofeggs> https://pagure.io/fesco/issue/1820
15:13:18 <puiterwijk> Also, I'd like to point out that the s390x Atomic/Cloud/Docker page is missing the "List of deliverables" information. Which is kind of critical for that one.
15:13:20 <zodbot> bowlofeggs: Issue #1820: Adjust/Drop/Document batched updates policy - fesco - Pagure - https://pagure.io/fesco/issue/1820
15:13:55 <zbyszek> We were supposed to discuss this more on fedora-devel, but that never happened. We have no new information.
15:13:57 <puiterwijk> Oh, I guess I'm too late now to bring up that change. Ah well.
15:14:05 <zbyszek> I propose to table this for another week.
15:14:33 <sgallagh> zbyszek: Do you want to revive the thread with new questions?
15:14:42 <ksinny> puiterwijk: you mean listing explicitly "List of deliverables" in wiki page?
15:14:48 <nirik> puiterwijk: I think thats a good point, we should note it to the change owner...
15:14:50 <zbyszek> I'm happy with status quo, mostly.
15:15:07 <sgallagh> I think ideally we'd ask mattdm for a list of problems he wants solved and then see if changes to batched help solve any of them
15:15:15 <puiterwijk> nirik, ksinny: yes. And especially given that the template seems to imply that any change to the deliverables is a system-wide change (which makes sense to me, as it needs more time for releng and infra)
15:15:27 <puiterwijk> "List of deliverables: N/A (not a System Wide Change)
15:15:33 <bowlofeggs> let's try to focus on the batching ticket for a second, we can return tot he previous topic if we need to
15:15:33 <puiterwijk> (in the template)
15:15:37 <puiterwijk> okay.
15:15:50 <nirik> FYI, sort of related to this: there's a proposal for delta repodata (see infrastructure list).
15:16:42 <bowlofeggs> proposal: We solicit mattdm for requirements, and we suggest a tie in with delta repodata for now, revisit ticket later
15:16:42 <jsmith> The more I think about it, I think we need three repos --- "updates-slowlane" and "updates-fastlane" and "updates-testing"
15:17:10 <bowlofeggs> jsmith: i think that might be a problem for infra and mirrors re resources
15:17:18 <zbyszek> jsmith: delta repodata would obsolete the need for batched or "slowlane"
15:17:18 <bowlofeggs> though it does sound nice
15:17:43 <nirik> zbyszek: only the part about downloading repodata a bunch
15:17:47 <jsmith> zbyszek: OK... I haven't read the details on that yet, so I'm not sure how that solves the problem.
15:17:57 <jsmith> zbyszek: I'll go do some reading :-)
15:17:58 <puiterwijk> zbyszek: I don't think that delta repodata fixes all of the problems. Since one of the gaols of batching was to get users to get a less constant stream of updates
15:18:13 <puiterwijk> all of the problems the batching is trying to solve*
15:18:35 <bowlofeggs> yeah batching was about more than stopping people from downloading metadata files
15:18:41 <jsmith> Or, just thinking outside the box here, maybe just generate "slow-lane" metadata vs. "fast-lane" metadata...
15:18:47 <bowlofeggs> though, making the delta metadata files would be an improvement for sure
15:18:48 <sgallagh> And also to be able to run the batched content through automated tests before putting them out as a unit
15:18:56 <sgallagh> Which hasn't happened yet, but is *enabled* by batching
15:19:37 <jsmith> Anyhoo -- sounds like we need more discussion on the list.
15:19:56 <jsmith> +1 to proposal to solicit more feedback
15:20:15 <zbyszek> I can volunteer to write to fedora-devel with a summary (I also need to this for myself) and reopen the discussion.
15:20:28 <bowlofeggs> zbyszek: excellent
15:20:31 <bowlofeggs> i'm +1
15:20:52 <nirik> sounds good
15:20:54 <bowlofeggs> i take zbyszek as a +1
15:21:01 <bowlofeggs> sgallagh: ?
15:21:29 <sgallagh> I missed the proposal
15:21:42 <sgallagh> Oh I see it now
15:21:42 <sgallagh> +
15:21:43 <sgallagh> +1
15:21:51 <bowlofeggs> cool
15:22:24 <bowlofeggs> #agreed zbyszek will solicit more feedback on fedora-devel (perhaps also from mattdm specifically) (+5, 0, -1)
15:22:28 <bowlofeggs> ugh
15:22:29 <bowlofeggs> #undo
15:22:29 <zodbot> Removing item from minutes: AGREED by bowlofeggs at 15:22:24 : zbyszek will solicit more feedback on fedora-devel (perhaps also from mattdm specifically) (+5, 0, -1)
15:22:38 <bowlofeggs> #agreed zbyszek will solicit more feedback on fedora-devel (perhaps also from mattdm specifically) (+5, 0, -0)
15:22:39 <bowlofeggs> haha
15:22:52 <bowlofeggs> ok, let's return to the previous topic since there was confusion around deliverables
15:23:02 <bowlofeggs> #topic #1767 F28 Self Contained Changes: Atomic, Cloud and Docker images
15:23:02 <bowlofeggs> for s390x
15:23:03 <bowlofeggs> https://fedoraproject.org/wiki/Changes/Atomic_Cloud_and_Docker_images_for_s390x
15:23:11 <bowlofeggs> puiterwijk: ^
15:23:23 <puiterwijk> Well, my main problem is that that one also feels like it tried to bypass the process. Since the change template implies that changes to deliverables is a system wide change
15:23:28 <puiterwijk> (which I would agree with)
15:23:44 <puiterwijk> And as I already mentioned: the page entirely leaves out that part of the template that describes the list of deliverable changes
15:23:58 <puiterwijk> This is the specific part in the template I meant: List of deliverables: N/A (not a System Wide Change)
15:24:15 <zbyszek> Agrred.
15:24:27 <ksinny> puiterwijk: s390x is not a release blocking arch and it impatcs limited users, so I am not sure if it counts as system wise change
15:24:27 <puiterwijk> That is entirely absent in https://fedoraproject.org/wiki/Changes/Atomic_Cloud_and_Docker_images_for_s390x  - and that's the entire goal of the change
15:24:38 <puiterwijk> ksinny: sure, but it changes the listof deliverables, not?
15:25:09 <ksinny> puiterwijk: yes
15:25:15 <puiterwijk> And as said, the template implies that change of deliverables implies system-wide change. And as said, it impacts work for releng/infra.
15:25:18 <bowlofeggs> i would tend to think of the list of deliverables as being the same as "release blocking", but i'm not sure if i'm right about that
15:25:32 <jsmith> I would tend to agree that a list of deliverable changes would be helpful
15:25:38 <sharkcz> AFAIK the deliverables listed in the fedora-pungi PR, ksinny ?
15:25:41 <bowlofeggs> true that it implies work for releng/infra
15:25:43 <ksinny> I didn't see any objection from infra and releng side for work needs to be done to include this change
15:25:51 <puiterwijk> sharkcz: they're not in the change list.
15:26:04 <jsmith> I wouldn't go so far as to imply that a change in deliverables is "release blocking" -- there can be deliverables that don't block releases.
15:26:33 <sharkcz> puiterwijk: right, but should be easy to fill them into the wiki
15:26:39 <sgallagh> Well, I think the relevant part is that any change that impacts rel-eng needs sign-off from them
15:26:54 <bowlofeggs> so the real question is: is a deliverable that doesn't block a release a "system wide change"?
15:27:02 <puiterwijk> sharkcz: sure. But it is great to have there so releng *can* actually ack/nack
15:27:27 <ksinny> PR for changes to enable these composes are already proposed in https://pagure.io/pungi-fedora/pull-request/496
15:27:32 <puiterwijk> I would like to point out that my main concern I wanted to bring up is the lack of a list, which means that ack/nack for releng/infra is hard
15:27:44 <bowlofeggs> there is a releng ticket, but no response from them yet: https://pagure.io/releng/issue/7286
15:27:58 <puiterwijk> Also, I'd like to point out that the Oz changes needed to make this possible isn't merged yet either
15:28:15 <puiterwijk> So right now, we can't even test it probably.
15:28:38 <ksinny> puiterwijk: we were able to reach maintainer and oz change will be available early next week
15:28:40 <puiterwijk> Which, in my opinion, makes it really hard for releng to say "yes we can add it"
15:28:47 <ksinny> me and imclod are working on this
15:29:16 <puiterwijk> ksinny: and did you get explicit buy-in from releng yet? As Randy said, I have seen a lack of response so far.
15:29:28 <puiterwijk> And I don't like working on "nobody saying no implies yes"
15:29:48 <bowlofeggs> reading https://fedoraproject.org/wiki/Changes/Policy#Self_contained_changes does make it sounds like this might not be a self contained change
15:30:20 <bowlofeggs> specifically the bit about "with limited scope and impact on the rest of distribution/project"
15:30:25 <bowlofeggs> this does affect releng and infra
15:30:52 <ksinny> bowlofeggs: yes and I got ack from infra here https://pagure.io/fedora-infrastructure/issue/6659
15:30:57 <puiterwijk> I do think it's a limited scope on the distribution. Just not sure about the project in general, since I haven't yet seen any composes with any of the new deliverables working
15:31:29 <puiterwijk> ksinny: well, I would not say "let us know when you get to that point" is an explicit ack that infra buys in
15:31:54 <puiterwijk> That sounds like a "let us know when the oz/imgfac changes needed are in and we can see further".
15:32:25 <bowlofeggs> yeah i think i'm convinced that this should be a system wide change
15:32:30 * nirik isn't against the change at all, but would prefer the list of deliverables to be clear and dependencies noted (oz, etc)
15:32:42 <puiterwijk> nirik: yes, that was my main point from the start indeed.
15:33:15 <puiterwijk> since the "Scope" does not list any of the Oz/Imgfac changes needed
15:33:19 <zbyszek> I understand the the contingency mechanism would be "drop those images from the list of deliverables". Is this something we can do easily at any point?
15:33:59 <nirik> it would take pungi-fedora changes...
15:34:26 <puiterwijk> Also, one other thing I'd like to point out about new deliverables is that we would want to inform QA as well. They may not be able to test things, but I'd still like to hear their thoughts too about anything we ship with the official Fedora branding.
15:34:36 <ksinny> pungi-fedora changes are ready on this PR https://pagure.io/pungi-fedora/pull-request/496
15:34:56 <puiterwijk> zbyszek: I think it's easy enough to back out the changes from pungi-fedora, yes.
15:34:59 <ksinny> But we need oz and infra ready before getting it merged
15:35:43 <puiterwijk> Oh, the Oz work is listed under the Contingency plan... That should definitely be under Scope. Just like any imgfac changes needed
15:36:16 <zbyszek> So I think we should upgrade that to a "system wide change", and require the list of deliverables to be filled, and a contingency mechanism and deadline to be specified.
15:36:29 <bowlofeggs> zbyszek: yeah that's what i'm thinking too
15:36:46 <puiterwijk> Agreed with that, if "the scope filled in fully" is also added to that
15:36:59 <puiterwijk> as said: neither oz nor imgfac are listed under "Scope" for this change.
15:37:00 <bowlofeggs> nirik, sgallagh, jsmith: what are you thoughts re making this a system wide change?
15:37:12 <nirik> +1 to zbyszek's proposal
15:37:26 <jsmith> Sure, I guess I can agree with that -- +1
15:37:28 <sgallagh> +1
15:37:39 <bowlofeggs> does makignt his a system wide change imply that it will wait for F29?
15:37:56 <bowlofeggs> or is it still good for F28?
15:37:56 <nirik> I don;t think so.. we can still add it now?
15:37:59 <puiterwijk> bowlofeggs: I'd say that given the "simple" contingency, I think we could pull it off for f28
15:38:09 <bowlofeggs> i'm fine with f28
15:38:10 <zbyszek> No, I think it should go in for F28, too.
15:38:11 <bowlofeggs> ok
15:38:55 <bowlofeggs> #agreed Atomic, Cloud and Docker images for s390x needs to be upgraded to a system-wide change with deliverables documented, and a contingency mechanism and deadline specified (+5, 0, -0)
15:39:09 <zbyszek> scope?
15:39:09 <bowlofeggs> ok anything else before we move on?
15:39:30 <zbyszek> puiterwijk> Agreed with that, if "the scope filled in fully" is also added to that
15:39:40 <bowlofeggs> ok i'll update it
15:39:41 <bowlofeggs> #undo
15:39:41 <zodbot> Removing item from minutes: AGREED by bowlofeggs at 15:38:55 : Atomic, Cloud and Docker images for s390x needs to be upgraded to a system-wide change with deliverables documented, and a contingency mechanism and deadline specified (+5, 0, -0)
15:40:04 <bowlofeggs> #agreed Atomic, Cloud and Docker images for s390x needs to be upgraded to a system-wide change with deliverables documented, scope filled in fully, and a contingency mechanism and deadline specified (+5, 0, -0)
15:40:34 <bowlofeggs> alright, let's move on to the next topic
15:40:45 <bowlofeggs> #topic #1838 Nonresponsive maintainer: joost
15:40:45 <bowlofeggs> .fesco 1838
15:40:45 <bowlofeggs> https://pagure.io/fesco/issue/1838
15:40:46 <zodbot> bowlofeggs: Issue #1838: Nonresponsive maintainer: joost - fesco - Pagure - https://pagure.io/fesco/issue/1838
15:41:18 <bowlofeggs> i also CC'd joost in the ticket and in the fesco meeting notes
15:41:26 <zbyszek> Joost built the package in koji, so I think orphaning is not warranted. I'd propose to require him to add a comaintainer or comaintainers.
15:42:04 <zbyszek> https://koji.fedoraproject.org/koji/buildinfo?buildID=1038094
15:42:05 <sgallagh> zbyszek: Unless you are volunteering, I don't think we can actually do that
15:42:30 <zbyszek> sgallagh: weren't there some people who wanted to have build rights for it?
15:43:02 <zbyszek> (I'm not volunteering, I haven't used Pascal in 15 years at least.)
15:43:39 <bowlofeggs> the ticket filer isn't asking to maintain the package
15:43:40 <nirik> the submittor says they want the packages fixed, but they don't want to maintain them
15:43:54 <zbyszek> Artur Iwicki wrote "I'd be willing to join as a co-maintainer."
15:44:02 <nirik> but it sounds like joost might be around, just not super active?
15:44:18 <sgallagh> That's all basically irrelevant. We are only being asked to address the non-responsive packager request
15:44:49 <nirik> the email in the recent changelog isn't the same as the one in the email to devel
15:45:16 <nirik> oh, it is.
15:46:28 <nirik> so, yeah, proposal: add co-maintainer and hope joost comes back to help them?
15:46:46 <sgallagh> works for me
15:46:47 <sgallagh> +1
15:46:58 <zbyszek> I think might help the issue, +1
15:47:03 <bowlofeggs> nirik: which co-maintainer, artur?
15:47:09 <bowlofeggs> or the reporter?
15:47:22 <bowlofeggs> it's weird that the reporter doesn't want ACLs haha
15:47:35 <bowlofeggs> +1 to the general proposal from me though
15:47:38 <nirik> Artur
15:47:51 <bowlofeggs> works for me, +1
15:47:54 <zbyszek> Artur, he's the only one who volunteered.
15:48:06 <bowlofeggs> jsmith: ?
15:48:10 <jsmith> +1
15:48:51 <zbyszek> sorry, one more thing. I think we should do this for two packages: fpc and lazarus.
15:49:00 <nirik> yes. +1
15:49:15 <bowlofeggs> yeah agreed
15:49:20 <jsmith> agreed
15:49:23 <bowlofeggs> #agreed FESCo will add  Artur Iwicki as a co-maintainer for fpc and lazarus (+5, 0, -0)
15:49:37 <bowlofeggs> anyone volunteer to make that change in ACLs? i dont' think i have perms
15:50:17 <nirik> I can do it
15:50:29 <bowlofeggs> #action nirik will assign ACLs to Artur Iwicki
15:50:43 <bowlofeggs> #topic #1839 Nonresponsive maintainer: radford
15:50:44 <bowlofeggs> .fesco 1839
15:50:44 <bowlofeggs> https://pagure.io/fesco/issue/1839
15:50:45 <zodbot> bowlofeggs: Issue #1839: Nonresponsive maintainer: radford - fesco - Pagure - https://pagure.io/fesco/issue/1839
15:50:57 <zbyszek> +1 to orphan, seems a pretty clear case
15:51:10 <sgallagh> +1 to orphan
15:51:28 <nirik> +1 to orphan
15:51:58 <bowlofeggs> do we want to orphan it, or transfer it to till?
15:52:27 <zbyszek> Transfer to till, otherwise he'll have to open an infra ticket.
15:52:43 <zbyszek> We can just skip this extra manual step.
15:53:10 <bowlofeggs> proposal: transfer ownership of dot2tex to thofmann
15:53:32 <zbyszek> hm, and what about the other three packages?
15:53:42 <bowlofeggs> right
15:53:49 <bowlofeggs> transfer the one, orphant he 3?
15:53:57 <sgallagh> Fine by me
15:53:59 <jsmith> +1 to transfer the one, orphan the other three
15:54:04 <zbyszek> +1 to the same
15:54:14 <bowlofeggs> nirik: good with that too?
15:54:26 <nirik> sure.
15:54:51 <bowlofeggs> #agreed FESCo will transfer dot2tex to thofmann and will orphan radford's other packages (+5, 0, -0)
15:54:58 <bowlofeggs> nirik: would you like to do the honors on this one as well?
15:55:12 <bowlofeggs> i'm not sure anyone else present has the perms
15:55:20 <nirik> sure.
15:55:27 <bowlofeggs> #action nirik will do the deeds
15:55:38 <bowlofeggs> #topic #1840 Better policy for mass cleanups / trivial changes for
15:55:38 <bowlofeggs> provenpackagers
15:55:38 <bowlofeggs> .fesco 1840
15:55:40 <zodbot> bowlofeggs: Issue #1840: Better policy for mass cleanups / trivial changes for provenpackagers - fesco - Pagure - https://pagure.io/fesco/issue/1840
15:55:56 <bowlofeggs> ignatenkobrain: you're up
15:57:02 <ignatenkobrain> I'm still alive!
15:57:12 <bowlofeggs> i personally think the macro thing was done according to my understanding on policy
15:57:22 <ignatenkobrain> tl;dr I want some more explicit words about automatic cleanups
15:57:39 <bowlofeggs> (and i'm +1 to mass clean ups in general, it helps all of us, even those who don't appreciate it ☺)
15:57:45 <ignatenkobrain> because from what I see on ML -- people don't want to get any cleanups
15:57:47 <ignatenkobrain> and they are even against them
15:58:03 <bowlofeggs> yeah i don't understand why people don't want help with stuff like that
15:58:17 <sgallagh> ignatenkobrain: No, just the same noisy people who believe they deserve their own little fiefdom
15:58:24 <nirik> yeah, I also disagree with them and find the cleanups great
15:58:44 <sgallagh> "I will either take what I am given, or I will leave. Why should I bother to spend ym time doing maintenance when others will?"
15:59:02 <sgallagh> Frankly, I don't see it as much of a loss if people who can't appreciate help decide to leave.
16:00:36 <nirik> right.
16:01:13 <bowlofeggs> zbyszek think the existing docs already cover the use case
16:01:24 <zbyszek> OK, after re-reading https://fedoraproject.org/wiki/Mass_package_changes once again, I think it covers the case of automated cleanup enough. ignatenkobrain do you have a specific proposal?
16:01:29 <ignatenkobrain> just sayin that it would be nice if we would have some sentence about automatic cleanups
16:01:35 <zbyszek> but we do!
16:01:42 <bowlofeggs> it does have a section about automation
16:01:57 <bowlofeggs> i guess it doesn't mention the provenpackager bit
16:02:06 <ignatenkobrain> zbyszek: it does cover everything I think
16:02:11 <bowlofeggs> though it'd be hard to do taht without being a provenpackager
16:02:19 <bowlofeggs> so it's kinda implied
16:02:31 <bowlofeggs> though i guess you can do it with PRs and not be a PP
16:02:33 <nirik> yeah, I think we had someone who wasn't confused by that recently
16:02:50 <nirik> they thought they could propose some change, then file a ticket "somewhere" and it would be done for them
16:03:00 <orc_fedo> am I the only one seeing the -devel list posts from lower involvement packagers saying:  I am not getting notifications, and I am being startled?
16:03:25 <jsmith> orc_fedo: No, I saw that too -- but I think that's a separate issue in some regards.
16:04:08 <orc_fedo> pp 'notifications' in the mass that is -devel are clearly not working well enough
16:04:13 <zbyszek> One proposal would be ask people doing the proposal to send the announcement e-mail personally to all packagers whose packages are touched.
16:04:23 <nirik> devel-announce should be something all maintainers are subscribed to
16:04:43 <jsmith> zbyszek: Again, that's hard to do when you're talking about "mass" cleanups -- you might be emailing hundreds of developers
16:04:46 <jsmith> nirik: Agreed!
16:05:08 <zbyszek> jsmith: there's massmail and other things like that
16:05:33 <ignatenkobrain> zbyszek: that would be rejected by mailman
16:05:36 <bowlofeggs> so maybe we can change the text here to say to mail devel-announce instead of devel
16:05:50 <orc_fedo> ... it there is time to do non time critical mass cleanups, there would seem to be time to write tooling to query pkgdb2 / bugzilla and send emails
16:06:00 <orc_fedo> s/it/if/
16:06:03 <ignatenkobrain> nirik: then can we get someone who will approve those emails faster?
16:06:05 <puiterwijk> nirik: agreed, and it's even in our documentation: "You must join the fedora devel-announce mailing list" :)
16:06:17 <nirik> ignatenkobrain: faster?
16:06:20 <ignatenkobrain> IIRC what tibbs sent on 2nd of Feb, got approved only yesterday
16:06:42 <nirik> I don't know how that one was missed for so long. I didn't see it or approve it.
16:06:54 <nirik> but in general if you don't see something in a day or so, please ask
16:06:58 <sgallagh> nirik: Maybe it would be sensible to make all sitting FESCo members moderators of that list?
16:07:06 <nirik> but do give it a bit instead of pinging right after you send it.
16:07:11 <puiterwijk> I approved it because I was looking at another email.
16:07:12 <nirik> sure, if you want
16:07:37 <nirik> I think there may also be something going on with notifications... I haven't seen in any a while
16:07:52 <tibbs> Sadly I just didn't notice that the message hadn't actually gone through.
16:07:54 <bowlofeggs> yeah fmn is partially broken
16:07:59 <puiterwijk> sgallagh: a couple of them already are. but sure, that'd make sense. Maybe also the FPC members?
16:08:02 <puiterwijk> bowlofeggs: that's not actually FMN
16:08:03 <sgallagh> OK, in that case I'll mention that I sent a devel-announce message this morning too ;-)
16:08:06 <bowlofeggs> ah
16:08:06 <nirik> bowlofeggs: well, mailman should notify there.
16:08:13 <nirik> sgallagh: I approved it already
16:08:13 <bowlofeggs> right
16:08:17 <sgallagh> Oh ok
16:08:19 <puiterwijk> bowlofeggs: mailman is supposed to inform moderators directly
16:08:28 <sgallagh> Probably a gmail-ism that I didn't get it back.
16:08:40 <nirik> anyhow, I'm fine changing devel to devel-announce there so we get more coverage to maintainers.
16:08:41 <bowlofeggs> proposal: change policy to mail devel-announce instead of devel. mention in automated cleanup that a PP may do it without PRs, and a non-PP can send PRs
16:08:54 <ignatenkobrain> Proposal: Change https://fedoraproject.org/wiki/Mass_package_changes to use devel-announce@ ;)
16:09:04 <ignatenkobrain> bowlofeggs: you are faster ;)
16:09:06 <bowlofeggs> haha
16:09:08 <bowlofeggs> i can +1 either
16:09:25 * ignatenkobrain loves bowlofeggs 's proposal more
16:09:31 <nirik> I'm not sure PRs are great for mass changes tho
16:09:37 <nirik> look at the python2 thing.
16:09:39 <bowlofeggs> yeah it's a lot of overhead
16:09:42 <ignatenkobrain> because it mentions that PR is not required
16:09:46 <ignatenkobrain> (explicitly)
16:09:52 <tibbs> When I did the last big rewrite of the mass package change document I wasn't really sure which list to use.
16:09:59 <bowlofeggs> nirik: yeah that's why i said that a PP can just do it without PRs in my proposal
16:10:30 <sgallagh> bowlofeggs: +1
16:10:31 <zbyszek> +1 to bowlofeggs's proposal
16:10:37 <bowlofeggs> nirik: but a non-PP wouldn't have any other option that i know of than PRs
16:10:45 <nirik> bowlofeggs: sure, but non PP's filing 1000 PR's is likely to end in... disappointment
16:10:48 <bowlofeggs> i guess non-PPs don't do mass changes very often
16:10:51 <sgallagh> bowlofeggs: "Bribe a PP to do it"?
16:10:59 <bowlofeggs> haha yeah
16:11:00 <nirik> convince a PP to yeah...
16:11:02 <zbyszek> Ask to become a PP?
16:11:09 <bowlofeggs> yeah that too
16:11:34 <sgallagh> For the record, I am open to bribery for such purposes ;-)
16:11:46 <ignatenkobrain> sgallagh: +1
16:11:54 <tibbs> My original idea was that people would be able to make a distinction between stuff that needed to go via PR and things that wouldn't.
16:11:56 <nirik> (as a side note, it would be interesting to have some kind of reporting on PRs on src... like how many total, how many merged, how many open, what packages have the most open, etc)
16:11:56 <ignatenkobrain> zbyszek: that's what ischerb did :D
16:12:01 <bowlofeggs> do you want me to take the non-PP bit off and just say PPs can just do it without PRs?
16:12:12 <nirik> bowlofeggs: I would prefer that
16:12:13 <jsmith> bowlofeggs: I'm fine with that.
16:12:14 <sgallagh> bowlofeggs: Yeah
16:12:16 <tibbs> But actually drawing a line is a bit difficult.
16:12:25 <bowlofeggs> proposal: change policy to mail devel-announce instead of devel. mention in automated cleanup that a PP may do it without PRs
16:12:39 <tibbs> So when I wrote that document I didn't really try to state anything explicitly.
16:12:49 <nirik> bowlofeggs: +1
16:12:56 <zbyszek> +1
16:13:21 <ignatenkobrain> sgallagh: we still need someone in Europe too!
16:13:21 <sgallagh> bowlofeggs: +1
16:13:28 * nirik notes he originally wrote that page, but thanks tibbs for reworking it to the nice state it is these days. ;)
16:13:34 <bowlofeggs> ok i'm counting jsmith as a +1 with his "i'm fine with that"
16:13:49 <tibbs> True, wasn't intending to claim credit.
16:13:58 <jsmith> Yes, +1
16:14:07 <bowlofeggs> #agreed  change policy to mail devel-announce instead of devel. mention in automated cleanup that a PP may do it without PRs (+5, 0, -0)
16:14:14 <bowlofeggs> #action bowlofeggs will make the changes
16:14:20 <tibbs> FWIW I agree with the proposal.
16:14:33 <bowlofeggs> well, unless the page is protected or something
16:14:34 <bowlofeggs> haha
16:14:41 <tibbs> No, it shouldn't be protected.
16:14:46 <bowlofeggs> #topic Next week's chair
16:14:47 <puiterwijk> Okay, now short remark: unless Kevin now says "please no", I can add all FESCo members as moderators on devel-announce?
16:14:51 <bowlofeggs> volunteers?
16:15:09 <sgallagh> I'm on vacation next week
16:15:10 <bowlofeggs> puiterwijk: sure i'd help moderate (if it notifies me anyway)
16:15:13 <nirik> puiterwijk: fine with me, but note fesco membership changes each cycle, so do we keep it in sync with that? or ?
16:15:23 <jsmith> puiterwijk: I can help moderate (again)
16:15:42 <puiterwijk> nirik: meh, I think just adding the new members is fine, I think that we kind of implicitly trust past FESCo members too
16:15:52 <zbyszek> nirik: I think it can stay, that's not really a sensitive thing, and people can always unsubscribe if they get bored.
16:15:57 <sgallagh> nirik: I'd say aspire to that, but in general if someone gets left on as a moderator, FESCo alumni are generally trustworthy
16:16:04 <nirik> sure, how about we add those people who said ok here?
16:16:15 <puiterwijk> nirik: sure. Just done so :)
16:16:19 <nirik> great
16:16:31 <nirik> anyhow, I haven't chaired in a while, I can do next week.
16:16:35 <zbyszek> nirik: please add me too
16:16:42 <bowlofeggs> #action nirik will chair next week's meeting
16:16:45 <puiterwijk> zbyszek: done
16:16:48 <bowlofeggs> #topic open floor
16:17:00 <zbyszek> puiterwijk: ah, thanks.
16:17:02 <sgallagh> I have one question for open floor
16:17:12 <zbyszek> I have one too, go ahead
16:17:17 <jsmith> I have a comment,  but I'll wait for sgallagh and zbyszek
16:17:23 <sgallagh> When do we consider it an appropriate time to put Tomasz Klozek on moderation?
16:17:39 <bowlofeggs> yeah i've noticed that he is… agressive
16:17:41 <sgallagh> As near as I can tell, the overwhelming majority of his posts are inflammatory and often personal attacks
16:17:45 <bowlofeggs> and that i can't spell aggressive
16:17:52 <jsmith> If we had a working Community Working Group, I'd recommend that route
16:17:55 <bowlofeggs> sgallagh: is that a council thing?
16:18:00 <jsmith> Maybe time to try to resurrect the CWG?
16:18:10 <nirik> he was on moderation and mattdm allowed him back. ;)
16:18:23 <orc_fedo> I have spoken w him by phone, and hee seems, ratehr socially inadept, rather than from a malicious source
16:18:26 <nirik> he did get better for a while.
16:19:19 <sgallagh> I'm all for people having differences of opinion, but his open hostility (I feel) is harmful to the community.
16:19:48 <bowlofeggs> yeah i agree
16:19:59 <zbyszek> I can speak with him too. It might be easier because of shared language. I don't think the hostlity is intended.
16:20:27 <jsmith> sgallagh: With the lack of a working CWG, this is probably something that mattdm or the Council should address formally -- informally, it would be great if zbyszek could chat with him.
16:20:45 <nirik> note (speaking as devel list moderator), I am not in favor of moderation anymore these days. If someone is not behaving nicely, I'd prefer to just remove them/drop all posts.
16:20:47 <orc_fedo> that may hellp -- he did not get 'nuance' in English convesation
16:21:17 <nirik> trying to read stuff and pass through "good" posts is very time consuming and not worth it.
16:21:27 <nirik> zbyszek: if you could do that that would be great
16:21:31 <sgallagh> nirik: Yeah, I can definitely understand that
16:21:37 <ignatenkobrain> kloczek-proxy@zbyszek.pl
16:21:46 <bowlofeggs> haha
16:22:03 <bowlofeggs> zbyszek: will you offer to be a proxy for me too?
16:22:18 <zbyszek> bowlofeggs: sure, please give me your bank password first
16:22:20 <sgallagh> zbyszek++ for offering to intercede
16:22:20 <zodbot> sgallagh: Karma for zbyszek changed to 2 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:22:24 <bowlofeggs> i get carried away in my anti-google rants sometimes…
16:22:35 <bowlofeggs> zbyszek: haha
16:22:40 <bowlofeggs> zbyszek++ indeed
16:22:41 <zodbot> bowlofeggs: Karma for zbyszek changed to 3 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:22:58 <ignatenkobrain> zbyszek-as-a-proxy!
16:22:59 <puiterwijk> bowlofeggs: oh, that one's easy: just make your mail server replace any "I hate Google" with "I love Google and everyone should use Android with Google Play Services"
16:23:05 <bowlofeggs> hahaha
16:23:18 <zbyszek> OK, so I have another bureacratic issue: should I be able to set tags on fesco tickets in pagure?
16:23:24 <ignatenkobrain> zbyszek++
16:23:25 <zodbot> ignatenkobrain: Karma for zbyszek changed to 4 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:23:33 <bowlofeggs> zbyszek: yeah i think you should
16:23:36 <bowlofeggs> zbyszek: like the meeting tag
16:23:43 <sgallagh> zbyszek: Yes; probably your group membership wasn't updated when you joined FESCo
16:23:45 <nirik> sure.
16:23:46 <zbyszek> Yeah, so I don't seem to be able too.
16:24:04 <bowlofeggs> oh yeah you aren't in the fesco group
16:24:08 <bowlofeggs> that's the problem
16:24:16 <zbyszek> who can add me?
16:24:23 * nirik can adjust it
16:24:28 <zbyszek> nirik++ thanks
16:24:28 <zodbot> zbyszek: Karma for kevin changed to 22 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:24:30 <puiterwijk> .admins fesco
16:24:32 <zodbot> puiterwijk: Administrators for fesco:
16:24:36 <bowlofeggs> #action nirik will add zbyszek to the fesco FAS group
16:24:39 <puiterwijk> I guess nobody can :)
16:24:43 * sgallagh snickers
16:24:45 <bowlofeggs> hahah what
16:24:51 <nirik> it's not a fas group
16:24:58 <nirik> it's a pagure.io group
16:24:58 <bowlofeggs> oh really? i didn' trealize
16:24:58 <puiterwijk> Oh, right. pagure.io is not synced via fas
16:25:02 <bowlofeggs> ooooh
16:25:04 <puiterwijk> Yeah, sorry.
16:25:17 <bowlofeggs> #action nirik will add zbyszek to the fesco pagure group
16:25:54 <nirik> done
16:25:59 <bowlofeggs> jsmith: you had a comment?
16:26:32 <jsmith> RE: the recent thread about helping to make sure Rawhide composes work
16:26:47 <zbyszek> nirik: thanks, I see I can change tags now.
16:27:01 <jsmith> I just wanted to say two things -- one, I don't have time (between ${DAYJOB} and grad school) to be "on-call" per-se, but I'd be happy to try to help when I can.
16:27:26 <jsmith> Second, I think increased communications/visibility is the key.  I know rawhide compose hasn't happened for days, but it took quite a bit of digging to find out why
16:27:32 <nirik> jsmith: happy to have more folks help... #fedora-releng is the place. ;)
16:27:48 <jsmith> Might be worth brainstorming how to increase the visibility / make it easier for smart community members to help.
16:27:49 <nirik> yeah, dustymabe created a issue tracker thing for them
16:28:03 <jsmith> nirik: Yeah, finally found that, and got the answers I was looking for
16:28:12 <zbyszek> link?
16:28:19 <nirik> https://pagure.io/dusty/failed-composes/issue/1
16:28:20 <Kellin> It's unsanctioned nirik
16:28:36 <nirik> sure, it's a proof of concept/first cut at something.
16:28:44 <jsmith> Anyhoo -- that's all I had.
16:28:51 <bowlofeggs> cool
16:28:52 <nirik> I was just going to make a gobby file, but whatever works.
16:28:53 <bowlofeggs> dustymabe++
16:28:53 <zodbot> bowlofeggs: Karma for dustymabe changed to 13 (for the f27 release cycle):  https://badges.fedoraproject.org/tags/cookie/any
16:29:13 <bowlofeggs> anything else for open floor?
16:29:56 <zbyszek> Are we sticking with this meeting time?
16:30:23 <bowlofeggs> zbyszek: this was the time that the most people could do, though it's very unfortunate that dgilmore can't do it
16:30:34 <bowlofeggs> woudl it be crazy if we did an alternating schedule?
16:30:49 <bowlofeggs> like odd numbered weeks we do this time, even numbered weeks we do a dgilmore-friendly time?
16:30:52 <zbyszek> I think it would cause even more issues
16:30:56 <bowlofeggs> yeah
16:31:17 <nirik> this time is not great for me, but I'm willing to do it if it works for (most) everyone else
16:31:27 <bowlofeggs> should we do a new whenisgood?
16:31:37 <zbyszek> Maybe we can do a poll again in a month or so?
16:31:39 <bowlofeggs> i couldn't figure out how to edit my responses on the existing one
16:32:03 <bowlofeggs> we could ask people to "tighten their belts" and be a little more agressive about opening their schedules
16:32:16 <bowlofeggs> i'll get up at 3 am if i have to
16:32:58 <bowlofeggs> yeah let's do another poll
16:33:09 <nirik> when does DST go away?
16:33:17 <bowlofeggs> nirik: depends ont eh country
16:33:45 <puiterwijk> bowlofeggs: I interpreted KEvin's question as "when does it get killed entirely".
16:33:56 <puiterwijk> Which I'm afraid is probably still depending on the country
16:33:56 <zbyszek> also depends on the country ;)
16:34:01 <puiterwijk> Yep :)
16:34:04 <bowlofeggs> hahaha
16:34:08 <bowlofeggs> yeah DST suuuucks
16:34:15 <bowlofeggs> actualyl technically we are not in DST right now in the US
16:34:22 <bowlofeggs> it's standard time
16:34:27 <puiterwijk> Right. It starts in a month or so
16:34:35 <puiterwijk> (CET -> CEST is on March 25)
16:34:39 <bowlofeggs> https://en.wikipedia.org/wiki/Daylight_saving_time_by_country
16:34:57 <puiterwijk> bowlofeggs: I love how in the US you *remove* the S in the timezone abreviation, in Europe we add it :)
16:35:00 <bowlofeggs> for the US it's the second sunday in march, so ~1 monthish
16:35:05 <bowlofeggs> haha
16:35:14 <bowlofeggs> zbyszek: you want to take the action of running the next poll?
16:35:14 <puiterwijk> Central European Time -> central European Summer Time. Mountain Standard Time -> Mountain Daylight Time
16:35:15 * nirik needs to move to hawaii...
16:35:36 <bowlofeggs> nirik: arizona doesn't do it, except several towns and reservations do it
16:35:43 <puiterwijk> nirik: yeah. Let's just move all Fedora activities there. All in one timezone and no DST
16:35:44 <zbyszek> sure, I can do it, but when? End of February?
16:35:54 <bowlofeggs> nirik: there's a road you can drive on there, where you have to adjust your clock 7 times in a 2-3 hour drive
16:36:08 <nirik> yep
16:36:12 <bowlofeggs> zbyszek: end of feb works for me, or you could do it now too
16:36:29 <bowlofeggs> i think we primarily want to ask people to be more generous with the new poll
16:36:33 <zbyszek> OK, I'll do it now, and remind people to keep the link to update.
16:36:44 <bowlofeggs> yeah i wish i'd kept my copy of it :)
16:36:46 * dustymabe reads scrollback
16:36:51 <dustymabe> sorry had stepped away
16:37:00 <bowlofeggs> though i left mine very open last time
16:37:12 <bowlofeggs> dustymabe: we were just singing your praises
16:38:03 <bowlofeggs> any other thoughts?
16:39:29 <bowlofeggs> #endmeeting