<@james:fedora.im>
16:00:06
!startmeeting fpc
<@meetbot:fedora.im>
16:00:07
Meeting started at 2024-06-13 16:00:06 UTC
<@meetbot:fedora.im>
16:00:07
The Meeting name is 'fpc'
<@james:fedora.im>
16:00:10
!topic Roll Call
<@decathorpe:fedora.im>
16:00:24
!hi
<@zodbot:fedora.im>
16:00:25
Fabio Valentini (decathorpe) - he / him / his
<@limb:fedora.im>
16:00:36
!hi
<@zodbot:fedora.im>
16:00:41
Gwyn Ciesla (limb) - she / her / hers
<@james:fedora.im>
16:00:45
!hi
<@zodbot:fedora.im>
16:00:47
James Antill (james)
<@jsteffan:fedora.im>
16:01:05
!hi
<@zodbot:fedora.im>
16:01:06
Jonathan Steffan (jsteffan)
<@nhanlon:beeper.com>
16:01:58
!hi
<@zodbot:fedora.im>
16:01:59
Neil Hanlon (neil) - he / him / his
<@carlwgeorge:matrix.org>
16:02:35
!hi
<@zodbot:fedora.im>
16:02:37
Carl George (carlwgeorge) - he / him / his
<@james:fedora.im>
16:10:11
!topic Open Floor
<@james:fedora.im>
16:11:03
Four official people this week ... but no real new things have happened. Anyone have anything they want to talk about, or anything devconf related?
<@jsteffan:fedora.im>
16:12:04
i have something if there is no other business
<@limb:fedora.im>
16:12:08
Not I.
<@carlwgeorge:matrix.org>
16:12:29
i have a rough something, but daMaestro can go first
<@carlwgeorge:matrix.org>
16:12:45
not rough like bad, just a rough idea of something that might be good for the guidelines
<@jsteffan:fedora.im>
16:13:06
i have a PR up for the vulkan and openxr stuff https://pagure.io/packaging-committee/pull-request/1368
<@jsteffan:fedora.im>
16:13:26
if we could do a vote or whatever the process is for these changes i'll update the commit message and then we can merge
<@limb:fedora.im>
16:14:13
We lack the votes today. We could discuss and possibly vote in-ticket.
<@jsteffan:fedora.im>
16:15:13
i asked for feedback from mesa and openxr maintainers and didn't hear anything over the past week. the PR basically encodes what is already happening. the main change is guidance on how to lay out the filesystem and the explicit use of a `%{_sysconfdir}/ld.conf.d/%{name}.conf` to bring visibility for drivers and runtimes to the linker
<@carlwgeorge:matrix.org>
16:15:15
i need more time to read through it anyways, i've been on pto and haven't gotten to the email notification for this one yet
<@james:fedora.im>
16:16:05
also, people will be busy with devconf prep/travel/recovery ... so probably shouldn't rush it this week or next.
<@carlwgeorge:matrix.org>
16:18:04
at a quick glance, it looks like this is a bigger diff than necessary by switching from semantic line breaks to wrapping. i would recommend keeping the semantic line breaks so the diff is smaller and for consistency with the rest of the docs. i'm sure there other examples of inconsistency, but i think we're trending towards more semantic line breaks, not less.
<@jsteffan:fedora.im>
16:20:11
ah, so that's just line breaks after each sentence? (more nuanced than that) https://sembr.org/
<@jsteffan:fedora.im>
16:20:22
i can do that update because i have zero preference there
<@james:fedora.im>
16:20:52
Pretty much. Also it's kind of weird that the patch link has two patches in it ... I thought it merged things for easier reading.
<@jsteffan:fedora.im>
16:21:54
that might be because i have multiple commits. i'll rewrite the history when i change to sembr and maybe that will help
<@jsteffan:fedora.im>
16:22:18
is the lack of MUST, SHOULD, etc okay? it' felt odd trying to wedge that language in there
<@conan_kudo:matrix.org>
16:24:02
!hi
<@zodbot:fedora.im>
16:24:04
Neal Gompa (ngompa) - he / him / his
<@jsteffan:fedora.im>
16:24:19
it would also be great if i could get guidance that i'm on the right path for this so i can get to work on packages that i needed this policy for :-)
<@jsteffan:fedora.im>
16:24:57
when the actual policy is published i have less of a preference for... but i'd like to get unblocked for some XR stuff
<@james:fedora.im>
16:25:49
It looked great to me ... but I know nothing :)
<@jsteffan:fedora.im>
16:25:57
i'll need to do patches for monado, wivrn, and monado-vulkan-layers and work with upstream(s) to get those merged
<@jsteffan:fedora.im>
16:26:17
i *think* i have buy in for this policy, based on casual conversations about it
<@jsteffan:fedora.im>
16:26:29
we'll see when i punt some MRs ;-)
<@carlwgeorge:matrix.org>
16:26:55
i would try to use MUST, SHOULD, etc if possible where you think it makes sense. those terms are really useful for clarity.
<@james:fedora.im>
16:26:58
Would be really nice to have at least a "it looks fine to me" from a maintainer of the couple of packages you explained how they work.
<@nhanlon:beeper.com>
16:28:06
RFC 2119 is love, RFC 2119 is life
<@james:fedora.im>
16:28:28
Note that you don't need to do them all at once, we can click merge on multiple PRs fairly easily ;)
<@jsteffan:fedora.im>
16:29:11
oh, i mean the actual packages. i already have them ready to go to review, but i want to update them to adhere to this policy
<@decathorpe:fedora.im>
16:29:16
sorry, I need to leave
<@decathorpe:fedora.im>
16:29:31
let me know when there's something that needs my vote in ticket :)
<@jsteffan:fedora.im>
16:29:54
when i do the sembr update i'll work in language from this rfc for each requirement/instruction when it makes sense
<@jsteffan:fedora.im>
16:31:32
i've emailed the maintainers, is there something more i need to do? can the FPC help bridge any review request from vulkan/mesa/openxr people/
<@jsteffan:fedora.im>
16:31:37
i've emailed the maintainers, is there something more i need to do? can the FPC help bridge any review request from vulkan/mesa/openxr people?
<@carlwgeorge:matrix.org>
16:33:27
i'm not sure what you mean by bridge, let help find the people familiar with those softwares?
<@james:fedora.im>
16:34:24
You can @foo them ... or ping them on matrix. But, again, devconf might make this a bit slow right now.
<@jsteffan:fedora.im>
16:34:53
yeah, i used the $package-maintainers@ email alias with a request to review. so i should ping people directly? is the FPC able to help me find the right people and poke them?
<@carlwgeorge:matrix.org>
16:35:53
the email alias is the right first step, then escalate to direct pings. that's exactly what i'd do.
<@jsteffan:fedora.im>
16:36:39
okay, once i update the PR with all of this feedback and update my packages as a show and tell i'll poke people directly if we don't hear from them before
<@jsteffan:fedora.im>
16:37:30
is it acceptable for me to push my pending packages through review after i update them to adhere to this pending policy?
<@jsteffan:fedora.im>
16:38:09
i'll be happy if i can unblock the review queue again, i've been stalled for a while now
<@james:fedora.im>
16:39:11
IMO, yeh, that's fine.
<@limb:fedora.im>
16:39:36
I concur
<@jsteffan:fedora.im>
16:40:44
great, thanks for the feedback. i'll work on the requested updates and plan to attend FPC meetings until we get this merged :-)
<@jsteffan:fedora.im>
16:41:17
Carl George: go ahead, thanks
<@carlwgeorge:matrix.org>
16:42:24
yesterday a friend was asking me if packages should use `/etc/sysconfig/$name` files, and i couldn't find anything in the guidelines about it
<@carlwgeorge:matrix.org>
16:43:34
i believe those files are a holdover from sysvinit scripts, which were difficult to modify. but with systemd units and overrides, i don't think they should be used anymore.
<@carlwgeorge:matrix.org>
16:44:12
anyone feel strongly about this one way or the other?
<@limb:fedora.im>
16:45:04
No. Some older programs still expect them and they can be used by unit files as well.
<@carlwgeorge:matrix.org>
16:45:56
right, they are certainly still in use, and unit files often use them with the environmentfile directive
<@james:fedora.im>
16:46:32
I think they are all legacy though? Like there shouldn't be new ones?
<@carlwgeorge:matrix.org>
16:46:51
the question came up because my friend got a bug report that the user was confused about why the config was in two places (main config file and a sysconfig file)
<@carlwgeorge:matrix.org>
16:47:56
and really it's three places if we count systemd overrides
<@carlwgeorge:matrix.org>
16:48:19
i think disallowing new packages from doing this would be a good step
<@carlwgeorge:matrix.org>
16:48:37
and perhaps a SHOULD for existing packages to phase out the usage over time
<@carlwgeorge:matrix.org>
16:48:59
here's the bug report if anyone wants the full context on the user confusion https://bugzilla.redhat.com/show_bug.cgi?id=2283798
<@james:fedora.im>
16:49:08
Not sure we need to go as far as disallowing new packages from doing anything.
<@carlwgeorge:matrix.org>
16:50:06
how about just a blanket SHOULD NOT on this pattern, or is even that too strong?
<@james:fedora.im>
16:50:21
I guess we probably don't want new files in that directory that have never been there before, which might be what you meant.
<@carlwgeorge:matrix.org>
16:50:28
not sure if that rfc has something softer than SHOULD NOT
<@james:fedora.im>
16:50:34
Yeh, I'm fine with should not around all of this, I think.
<@limb:fedora.im>
16:51:45
Same.
<@carlwgeorge:matrix.org>
16:51:59
i'll work with my friend on writing up a pr for this. the overall goal is to express a preference in the guidelines towards systemd overrides instead of special fedora/rh-specific sysconfig files.
<@carlwgeorge:matrix.org>
16:52:17
that's all for my open floor item for now
<@james:fedora.im>
16:53:37
Okay, unless there's anything else I'll give everyone 5 mins back
<@limb:fedora.im>
16:53:45
Thanks all!
<@james:fedora.im>
16:55:24
!endmeeting