<@zbyszek:fedora.im>
17:00:11
!startmeeting FESCO (2024-12-03)
<@meetbot:fedora.im>
17:00:12
Meeting started at 2024-12-03 17:00:11 UTC
<@meetbot:fedora.im>
17:00:12
The Meeting name is 'FESCO (2024-12-03)'
<@zbyszek:fedora.im>
17:00:15
!meetingname fesco
<@meetbot:fedora.im>
17:00:16
The Meeting Name is now fesco
<@zbyszek:fedora.im>
17:00:23
Chairs: @conan_kudo:matrix.org, @ngompa:fedora.im, @nirik:matrix.scrye.com, @humaton:fedora.im, @zbyszek:fedora.im, @sgallagh:fedora.im, @jistone:fedora.im, @dcantrell:fedora.im, @decathorpe:fedora.im, @salimma:fedora.im
<@zbyszek:fedora.im>
17:00:29
!topic Init Process
<@jistone:fedora.im>
17:00:34
!hi
<@zbyszek:fedora.im>
17:00:35
!hi
<@zodbot:fedora.im>
17:00:36
Josh Stone (jistone) - he / him / his
<@zodbot:fedora.im>
17:00:36
Zbigniew Jędrzejewski-Szmek (zbyszek)
<@nirik:matrix.scrye.com>
17:00:41
morning everyone
<@conan_kudo:matrix.org>
17:00:43
!hi
<@zodbot:fedora.im>
17:00:44
Neal Gompa (ngompa) - he / him / his
<@jkonecny:fedora.im>
17:01:08
!hi
<@zodbot:fedora.im>
17:01:09
Jiří Konečný (jkonecny)
<@sgallagh:fedora.im>
17:01:17
!hi
<@zodbot:fedora.im>
17:01:19
Stephen Gallagher (sgallagh) - he / him / his
<@dcantrell:fedora.im>
17:01:21
!hi
<@zodbot:fedora.im>
17:01:22
David Cantrell (dcantrell) - he / him / his
<@zbyszek:fedora.im>
17:01:23
OK, we have quorum.
<@zbyszek:fedora.im>
17:01:31
!topic #3272 Change: Anaconda Web UI partitioning
<@zbyszek:fedora.im>
17:01:31
!fesco 3272
<@zbyszek:fedora.im>
17:02:03
Pfff, I thought this was supposed to be fixed.
<@zbyszek:fedora.im>
17:02:12
!fesco 3272
<@zodbot:fedora.im>
17:02:13
**fesco #3272** (https://pagure.io/fesco/issue/3272):**Change: Anaconda Web UI partitioning**
<@zodbot:fedora.im>
17:02:13
● **Opened:** 3 months ago by amoloney
<@zodbot:fedora.im>
17:02:13
<@zodbot:fedora.im>
17:02:13
● **Assignee:** jkonecny
<@zodbot:fedora.im>
17:02:13
● **Last Updated:** 7 minutes ago
<@nirik:matrix.scrye.com>
17:03:18
so, where are we here? There was some test days, but also has there been more work landing?
<@nirik:matrix.scrye.com>
17:03:36
but yeah, a summary in the ticket.
<@jkonecny:fedora.im>
17:03:43
correct
<@jkonecny:fedora.im>
17:03:50
mainly the test days are done and we have the result
<@jkonecny:fedora.im>
17:03:56
some features were also delivered
<@jkonecny:fedora.im>
17:04:06
but we are mostly working on fixes from the test days now
<@bookwar:fedora.im>
17:04:27
!hi
<@zodbot:fedora.im>
17:04:40
Aleksandra Fedorova (bookwar) - she / her / hers
<@sgallagh:fedora.im>
17:06:18
My takeaway: feedback from the Test Days was mostly positive and the reservation most noted by FESCo members wasn't of concern to most of the people trying it out. Is that an accurate summary?
<@jistone:fedora.im>
17:07:37
the summary says feedback was neutral
<@jistone:fedora.im>
17:07:54
was there deviation in that? (i.e. positive and negative balancing out?)
<@bookwar:fedora.im>
17:08:49
Neutral feedback on the UI change counts as positive :)
<@jkonecny:fedora.im>
17:09:10
- we made quite a few improvements which are not available in the GTK UI
<@jkonecny:fedora.im>
17:09:10
- no complains about the immediate actions in the Cockpit Storage
<@jkonecny:fedora.im>
17:09:10
- no blocking bugs (from my perspective at least) were found - most of the bugs are not really about the UI but issues in Anaconda which are there for a quite a while
<@jkonecny:fedora.im>
17:09:10
I have my statements in the ticket but a few main points from me:
<@jkonecny:fedora.im>
17:09:10
- based on the voting and on the comments the outcome is a bit neutral (there are people praising the UI and some were confused by specific parts - no strong complain at least from my perspective)
<@conan_kudo:matrix.org>
17:10:04
it counts as not-negative, not positive
<@jkonecny:fedora.im>
17:10:18
Josh Stone: it's hard to said
<@jkonecny:fedora.im>
17:10:18
- a few people were happy about it
<@jkonecny:fedora.im>
17:10:18
- a few people didn't said they are unhappy nor happy, just pointing about the issues found
<@jkonecny:fedora.im>
17:10:32
and also based on the testing it went neutral
<@sgallagh:fedora.im>
17:11:01
Conan Kudo: I disagree; changes to UX always historically elicits more complaints than praise due to the "moved cheese" effect.
<@jkonecny:fedora.im>
17:11:03
I can't remember about any really annoyed feedback or such
<@conan_kudo:matrix.org>
17:11:27
I think jkonecny's assessment is fair
<@jistone:fedora.im>
17:11:28
overall, neutral to users and positive to maintainers is still a good outcome
<@jkonecny:fedora.im>
17:12:31
just for the record - there was one vote for the strongly not preferred and this vote was from Conan Kudo (not saying it's wrong, just want to point it out)
<@zbyszek:fedora.im>
17:12:46
The lack of complaints about immediate actions is important. This is really the hardest part, and if accept it, the rest is stuff that can be fixed.
<@dcantrell:fedora.im>
17:12:50
it's worth thinking about the users that don't exist yet. we're always talking about growing Fedora'
<@dcantrell:fedora.im>
17:12:59
s numbers and will this UI confuse new users or help them
<@zbyszek:fedora.im>
17:13:25
I expect that it'll help new users.
<@sgallagh:fedora.im>
17:13:28
dcantrell: With the redesigned "guided mode", I think that's a big win for that camp
<@jkonecny:fedora.im>
17:13:56
also we even have a feedback from Thorsten Leemhuis about the immediate action as it is "somewhat nice" https://social.linux.pizza/@knurd42/113468275840518337
<@jistone:fedora.im>
17:14:17
ok, that "strongly not preferred" is what I didn't want to lose in the average. Conan Kudo do you want to speak to that?
<@bookwar:fedora.im>
17:14:53
I think he was the only one who really noticed the difference and also was commenting when explicitly asked.
<@jkonecny:fedora.im>
17:15:08
good point bookwar
<@salimma:fedora.im>
17:15:52
!hi
<@zodbot:fedora.im>
17:15:53
Michel Lind (salimma) - he / him / his
<@conan_kudo:matrix.org>
17:15:56
it's mostly a problem if you have to juggle multiboot stuff, and to be honest, if we're going to lose it by going to cockpit storage, I'd like for us to push them to add plan-then-apply to cockpit storage
<@conan_kudo:matrix.org>
17:16:06
it's mostly a problem if you have to juggle multiboot stuff, and to be honest, if we're going to lose it by going to cockpit storage, I'd like for us to push them to add plan-then-apply to cockpit storage roadmap
<@dcantrell:fedora.im>
17:16:20
my statement was more about the neutral feedback users in the tests. the ones who were confused at different points. for someone entirely new to Fedora, that could indicate how people will react. but if everyone thinks things are fine, then whatever
<@jkonecny:fedora.im>
17:16:23
about the voting - I was expecting more from that but when I started to read the feedback that was much more important than the votes. So the results doesn't seem that important at the end honestly
<@conan_kudo:matrix.org>
17:16:59
these changes don't immediately affect me with Fedora KDE because it's nonfunctional without a DE firstboot wizard, but I envision that I will eventually have to consider it
<@jkonecny:fedora.im>
17:17:30
dcantrell: the new UI is designed in a way that the confusing parts shouldn't be reached by not advanced users. It's "hidden" on purpose
<@conan_kudo:matrix.org>
17:17:31
GNOME is the only desktop with a firstboot wizard, so Anaconda's new UI is nonfunctional outside of Workstation anyway
<@jkonecny:fedora.im>
17:17:39
in a way of if you really need it you can find it
<@jkonecny:fedora.im>
17:18:49
Conan Kudo: we are not targeting outside of Workstation right now but we already have a code for language selection, we just want to polish it
<@jkonecny:fedora.im>
17:18:54
also add keyboard control
<@conan_kudo:matrix.org>
17:18:58
the only other aspect is that I'm not really pleased about using Firefox as the launcher, as I'd like to have a dedicated launcher instead
<@jkonecny:fedora.im>
17:19:29
Conan Kudo: why is FF an issue?
<@conan_kudo:matrix.org>
17:19:38
because Anaconda shows up as Firefox
<@conan_kudo:matrix.org>
17:20:01
there's some brittle stuff to work around this in GNOME, but it doesn't work everywhere
<@sgallagh:fedora.im>
17:20:03
You would prefer to see something in Kiosk mode?
<@conan_kudo:matrix.org>
17:20:26
I'd rather we use the WebKitGTK or QtWebEngine thingies to have a simple application window wrapper
<@zbyszek:fedora.im>
17:20:43
Conan Kudo: I think this is a valid issue, but let's not discuss this here. This is an implementation detail and the discussion can be done in some ticket.
<@conan_kudo:matrix.org>
17:20:47
one that correctly identifies as Anaconda and doesn't look funny to the desktop
<@sgallagh:fedora.im>
17:20:48
Right, that's what I mean
<@conan_kudo:matrix.org>
17:21:34
anyway, Anaconda Web UI is not in a good enough state for all of Fedora, and I'd like us to divert to requesting cockpit-storage to add plan-then-apply to their roadmap
<@conan_kudo:matrix.org>
17:21:50
it doesn't have to happen right now, but I'd like it to happen sometime before we have broad adoption of the web UI
<@jkonecny:fedora.im>
17:21:53
I don't see this as a problem, honestly. For Workstation it works, if we hit an issue when we will adapt other deliverables we can resolve that. We can always start using webKit if necessary for specific variants
<@jkonecny:fedora.im>
17:22:12
however, we have to have FF support as the base because of CentOS / RHEL decisions
<@jkonecny:fedora.im>
17:22:26
that doesn't mean that we are blocked on this for Fedora too....
<@conan_kudo:matrix.org>
17:22:32
yup
<@jkonecny:fedora.im>
17:22:44
if it is necessary
<@conan_kudo:matrix.org>
17:22:55
I just don't want to accept hacks like that which can screw with desktop automation and navigation
<@conan_kudo:matrix.org>
17:23:15
e.g. implicit/explicit window rules to make anaconda window behave properly don't work if it's actually just firefox
<@jkonecny:fedora.im>
17:23:19
we don't want to add more maintenance and work for if it will be a problem
<@conan_kudo:matrix.org>
17:23:31
like what we just had to add for wayland-native anaconda liveinst
<@conan_kudo:matrix.org>
17:23:39
to make it behave properly on non-GNOME
<@jkonecny:fedora.im>
17:23:44
good point about FF is that we are getting nice support, that won't happen with webkit
<@bookwar:fedora.im>
17:23:48
(disclaimer: I am the manager of the Cockpit and Anaconda teams in Red Hat nowadays) we will discuss with the Cockpit team the idea to use blivet as a backend for Cockpit Storage in the future. We are open to other suggestions. I can not promise the outcome of the discussion though. The feedback, which we get from Fedora 42 Workstation with the Web UI, may contribute to it.
<@m4rtink:fedora.im>
17:24:04
Conan Kudo: I remember we got it to look like Anaconda in the overview name, icon, verything a while ago - maybe it redressed since then. We do use the Firefox kiosk mode, custom user chrome CSS & custom profile overrides.
<@conan_kudo:matrix.org>
17:24:33
mkolman: I can discuss the exact details of the problems later in #anaconda:fedoraproject.org, but my testing a couple months ago indicated that it's insufficient with Firefox
<@m4rtink:fedora.im>
17:24:38
As for Webkit GTW - IIRC there were some serious performance problems in some configurations.
<@m4rtink:fedora.im>
17:24:45
OK :)
<@conan_kudo:matrix.org>
17:24:49
Firefox's lack of PWA support means it's unfixable with Firefox
<@jkonecny:fedora.im>
17:24:52
mkolman: to my understanding it doesn't. There is more a worry about how it will behave outside of Gnome Shell
<@conan_kudo:matrix.org>
17:25:55
GNOME is a considerably "simpler" Wayland environment with specific quirks that the current implementation of the webUI abuses
<@sgallagh:fedora.im>
17:26:00
So, just to bring the conversation back to the main point: does anyone want to call something out as a BLOCKER for approving this Change, or can we just go ahead and vote?
<@bookwar:fedora.im>
17:26:24
I think in the current state of the Web UI we are already adding value to Fedora Workstation by covering the most simple cases in a much simpler and accessible way. We will keep the GTK interface available and accessible from the same installation media, thus I think we shouldn't block the rolling out of the already improved experience by the need of the future work.
<@conan_kudo:matrix.org>
17:26:30
honestly I still don't like it as is, but I don't see a point in doing anything about it
<@conan_kudo:matrix.org>
17:27:02
just being able to extract interest in improving cockpit-storage would be enough for me
<@conan_kudo:matrix.org>
17:27:11
because I really do think that's important for the long term
<@zbyszek:fedora.im>
17:28:03
I think the improvements that were done in the WebUI should be acknowledged. Things _are_ much better than a few months ago.
<@zbyszek:fedora.im>
17:28:07
Proposal: Change: Anaconda Web UI partitioning is approved for F42
<@sgallagh:fedora.im>
17:28:13
patch
<@jkonecny:fedora.im>
17:28:18
just want to be clear and open here to avoid surprises in the future
<@jkonecny:fedora.im>
17:28:18
I just asked today mcataranzo about the GIS support in Live ISO and he told me that it is possible they will get it there but they might not be able to do that. We will try to go that way but if it won't go well I think we are able to finish the code for language and keyboard selection in web UI (it's just finishing what we have)
<@jkonecny:fedora.im>
17:28:46
I want to be clear that the proposal might change a bit based on the Workstation capacity
<@jkonecny:fedora.im>
17:29:13
I'm not expecting this will be a blocker or serious issue in other way - as it is getting us closer to what we had since now
<@conan_kudo:matrix.org>
17:29:49
if you wind up implementing a frontend for blivet, then it can be integrated directly in the web UI anyway :)
<@sgallagh:fedora.im>
17:30:17
Sorry, ignore my "patch" comment. I misremembered the scope of this one vs. the WebUI in general.
<@sgallagh:fedora.im>
17:30:29
+1
<@conan_kudo:matrix.org>
17:30:38
reluctantly +1
<@dcantrell:fedora.im>
17:30:51
0
<@zbyszek:fedora.im>
17:30:56
I was trying to figure out what you meant ;) Thanks for the clarification.
<@nirik:matrix.scrye.com>
17:31:18
+1 here
<@sgallagh:fedora.im>
17:31:26
zbyszek: In Blocker bug meetings, "patch" is a note that you want to amend the phrasing of the proposal.
<@salimma:fedora.im>
17:31:28
+1
<@jistone:fedora.im>
17:31:36
+1
<@zbyszek:fedora.im>
17:31:39
Ah, OK.
<@sgallagh:fedora.im>
17:31:41
I was going to, then re-checked my assumptions before I made a change.
<@sgallagh:fedora.im>
17:33:01
I see (+6, 1, -0) so far, including zbyszek's implicit +1, which means this will pass. Anyone else want to chime in?
<@zbyszek:fedora.im>
17:33:03
?
<@zbyszek:fedora.im>
17:33:18
~~?~~
<@zbyszek:fedora.im>
17:33:30
nvm.
<@sgallagh:fedora.im>
17:33:49
I was going to suggest a change to the proposal, then re-checked my assumptions before I made a change.
<@zbyszek:fedora.im>
17:35:33
Hmm, Element is showing me a strange message.
<@zbyszek:fedora.im>
17:35:52
jkonecny++, thank you for coming.
<@zodbot:fedora.im>
17:35:53
Sorry, %40jkonecny%3Afedora.im does not look like a valid matrix user ID (e.g. @username:homeserver.com )
<@jkonecny:fedora.im>
17:36:06
thank you everyone
<@zbyszek:fedora.im>
17:36:18
@jkonecny++
<@jkonecny:fedora.im>
17:36:26
I know this was a hard one but I think thanks to everyone we were able to get the project into much better shape
<@zodbot:fedora.im>
17:36:30
sgallagh gave a cookie to jkonecny. They now have 8 cookies, 1 of which were obtained in the Fedora 41 release cycle
<@zbyszek:fedora.im>
17:36:45
!topic #3290 Preset Discussion: fips-crypto-policy-overlay.service
<@zbyszek:fedora.im>
17:36:49
!fesco 3290
<@zodbot:fedora.im>
17:36:50
● **Last Updated:** 2 hours ago
<@zodbot:fedora.im>
17:36:50
**fesco #3290** (https://pagure.io/fesco/issue/3290):**Preset Discussion: fips-crypto-policy-overlay.service**
<@zodbot:fedora.im>
17:36:50
<@zodbot:fedora.im>
17:36:50
● **Opened:** 2 weeks ago by sgallagh
<@zodbot:fedora.im>
17:36:50
● **Assignee:** Not Assigned
<@sgallagh:fedora.im>
17:37:09
zbyszek: You forgot to !agree
<@zbyszek:fedora.im>
17:37:35
I didn't. Element decided to stop sending messages.
<@zbyszek:fedora.im>
17:37:50
I switched to nheko for a moment.
<@nirik:matrix.scrye.com>
17:38:06
wonder if it will be eventually consistent. ;)
<@sgallagh:fedora.im>
17:38:07
OK, it didn't reach me at least
<@zbyszek:fedora.im>
17:38:30
!info #3272 Change: Anaconda Web UI partitioning is APPROVED (+6, 1, 0)
<@zbyszek:fedora.im>
17:38:45
Sorry, also !undo doesn't work properly. I'll adjust the summary manually.
<@zodbot:fedora.im>
17:38:55
sgallagh has already given cookies to zbyszek during the F41 timeframe
<@sgallagh:fedora.im>
17:39:46
Regarding this ticket, I think I'm good with just going ahead and providing the preset. I misunderstood the original scope when I brought it to FESCo, but it still just needs at least a rubber-stamp from us, per policy.
<@zbyszek:fedora.im>
17:39:53
Anyway, for 3290, the status is that some folks asked for a formal Change. But the maintainers argue that this is not a user visible change and doesn't need one.
<@zbyszek:fedora.im>
17:40:45
I agree with Stephen Gallagher
<@nirik:matrix.scrye.com>
17:40:50
I have not read any of the piles of comments this morning that arrived while I was asleep. ;)
<@salimma:fedora.im>
17:41:03
Yeah, given that this is just a corner case that most users shouldn't need to know about I now think we can just approve this
<@zbyszek:fedora.im>
17:41:12
nirik: those comments can be ignored. They are parenthetical to the main topic.
<@jistone:fedora.im>
17:41:12
are there any non-RHEL Fedora derivatives that might care about this?
<@nirik:matrix.scrye.com>
17:41:33
ok, fair enough.
<@sgallagh:fedora.im>
17:41:34
Josh Stone: None that aren't *also* RHEL derivatives, I believe
<@zbyszek:fedora.im>
17:41:52
Unlikely. Without certification FIPS is not useful.
<@conan_kudo:matrix.org>
17:41:53
Ultramarine might?
<@conan_kudo:matrix.org>
17:42:28
and maybe Amazon Linux?
<@jistone:fedora.im>
17:42:28
zbyszek: others could seek such certification though
<@sgallagh:fedora.im>
17:43:10
Anyone going through FIPS certification will have people directly involved in this effort
<@sgallagh:fedora.im>
17:43:22
So I'm happy to just let them deal with it
<@jistone:fedora.im>
17:43:26
fair
<@zbyszek:fedora.im>
17:43:28
Yeah.
<@nirik:matrix.scrye.com>
17:43:38
Josh Stone: are you asking for if it should need a change or not? or ?
<@conan_kudo:matrix.org>
17:43:59
I think that if FIPS is "pointless" on Fedora, why are they expending any effort to support it properly?
<@jistone:fedora.im>
17:44:01
yes, for Change significance
<@conan_kudo:matrix.org>
17:44:24
FIPS certification is separate from FIPS mode
<@sgallagh:fedora.im>
17:44:25
Conan Kudo: Because it's still an integration point upstream of RHEL?
<@conan_kudo:matrix.org>
17:44:39
and FIPS mode is the whole point we have a bunch of crazy rules around crypto libraries
<@nirik:matrix.scrye.com>
17:44:42
I think it might actually be bad to advertise it... or confusing at least
<@sgallagh:fedora.im>
17:44:43
How many Fedorans use Fiber Channel?
<@jistone:fedora.im>
17:44:59
it kind of feels like "upstream first" is going too far upstream :)
<@nirik:matrix.scrye.com>
17:45:07
so, I am +1 to just allowing and leaving it at that.
<@conan_kudo:matrix.org>
17:45:11
you're missing my point... I'm saying that it's a bad reason to botch it just because we don't get certified
<@salimma:fedora.im>
17:45:18
Yeah, carrying a change in fedora even though it's inert for us is probably still better than having it in ELN only
<@conan_kudo:matrix.org>
17:45:32
nobody should be using a justification that "it is pointless" as a reason to accept a change
<@conan_kudo:matrix.org>
17:46:04
and yes, that was one of the comments in the ticket
<@conan_kudo:matrix.org>
17:46:37
if the crypto team is doing FIPS enablement work to prepare for RHEL, then they should treat it as if people would use it on Fedora too
<@sgallagh:fedora.im>
17:46:42
I'm in favor of enabling the preset on the grounds that it lessens the delta between Fedora and RHEL.
<@sgallagh:fedora.im>
17:47:28
It's an improvement
<@conan_kudo:matrix.org>
17:47:30
basically, I don't want to see a Change or a ticket land at my desk that says "we're doing this for rhel and expect nobody to use this in fedora"
<@sgallagh:fedora.im>
17:47:40
(something something perfect vs good)
<@conan_kudo:matrix.org>
17:48:03
I'm fine with accepting the preset, I'm not fine with their justifications to people's questions
<@conan_kudo:matrix.org>
17:48:16
so this is my finger-wag to them about this
<@sgallagh:fedora.im>
17:48:24
Which are not relevant to the actual request here.
<@zbyszek:fedora.im>
17:48:27
Not everything that is developed upstream in various projects is directly usable in Fedora. For example, systemd has many features for immutable images and stuff. We don't want to advertise this in Fedora because it's not really usable without additional integration. But removing those features would be a lot of unnecessary and pointless work.
<@zbyszek:fedora.im>
17:48:53
OK, Josh Stone do you want us to vote on a requiring a Change Proposal?
<@zbyszek:fedora.im>
17:49:13
(Or anybody else who thinks that this should be required.)
<@jistone:fedora.im>
17:49:25
I'm satisfied with no Change
<@zbyszek:fedora.im>
17:49:33
OK, in that case:
<@zbyszek:fedora.im>
17:49:35
Proposal: approve the request and move on
<@jistone:fedora.im>
17:49:40
+1
<@nirik:matrix.scrye.com>
17:49:43
+1 to the proposal
<@zbyszek:fedora.im>
17:49:45
+1 FTR
<@sgallagh:fedora.im>
17:49:52
+1 (and I'll implement this today if approved)
<@conan_kudo:matrix.org>
17:50:04
+1
<@conan_kudo:matrix.org>
17:50:18
my complaint is I don't like the attitude that came back when people asked questions
<@dcantrell:fedora.im>
17:50:21
+1
<@conan_kudo:matrix.org>
17:50:23
the actual request in itself is fine
<@zbyszek:fedora.im>
17:50:57
!agree APPROVED (+6, 0, 0)
<@zbyszek:fedora.im>
17:51:24
Michel Lind 🎩 UTC-8 until Dec 7: mentioned that he needs a short break, so I'm not expecting a vote.
<@zbyszek:fedora.im>
17:51:47
!agreed APPROVED (+6, 0, 0)
<@zbyszek:fedora.im>
17:51:56
!topic #3292 Skipping FESCo Interviews for F41 Election?
<@zbyszek:fedora.im>
17:52:01
!fesco 3292
<@zodbot:fedora.im>
17:52:02
<@zodbot:fedora.im>
17:52:02
● **Opened:** 6 hours ago by amoloney
<@zodbot:fedora.im>
17:52:02
● **Last Updated:** 4 minutes ago
<@zodbot:fedora.im>
17:52:02
**fesco #3292** (https://pagure.io/fesco/issue/3292):**Skipping FESCo Interviews for F41 Election?**
<@zodbot:fedora.im>
17:52:02
● **Assignee:** Not Assigned
<@zbyszek:fedora.im>
17:52:22
The issue of requiring candidates that can fail was raised.
<@zbyszek:fedora.im>
17:52:28
Is this rule written down somewhere?
<@nirik:matrix.scrye.com>
17:52:42
yes, I added a link to the policy
<@sgallagh:fedora.im>
17:52:45
It used to be, but it appears to have vanished at some point. So 🤷
<@nirik:matrix.scrye.com>
17:52:49
https://docs.fedoraproject.org/en-US/fesco/FESCo_election_policy/
<@zbyszek:fedora.im>
17:53:01
Oh, right, sorry, I didn't see the comment.
<@sgallagh:fedora.im>
17:53:10
Oh there we are
<@jistone:fedora.im>
17:53:15
> A minimum number of candidates are necessary in order to hold an election. This will be the number of open seats + 25%.
<@jistone:fedora.im>
17:53:15
<@jistone:fedora.im>
17:53:15
> If not enough candidates have signed up by the deadline, the election may be delayed waiting for more candidates to appear, in coordination with the schedule for combined Fedora elections. If there are still not enough candidates, the candidates who are present will be voted upon (or merely confirmed if there are less candidates than open seats.)
<@zbyszek:fedora.im>
17:53:17
!info The rules say "A minimum number of candidates are necessary in order to hold an election. This will be the number of open seats + 25%."
<@sgallagh:fedora.im>
17:53:19
I was wrong about the percentage
<@jistone:fedora.im>
17:53:44
we're already 6 days past the deadline, but no extension was announced AFAIK
<@nirik:matrix.scrye.com>
17:54:26
yeah, so we need to decide if we want to extend the deadline, or just magically elect everyone who nominated.
<@zbyszek:fedora.im>
17:54:30
Our election wrangler is out sick.
<@nirik:matrix.scrye.com>
17:54:40
voting is supposed to start friday...
<@amoloney:fedora.im>
17:54:52
I have retuned from the ded this week :)
<@amoloney:fedora.im>
17:55:01
happy to follow your preference!
<@amoloney:fedora.im>
17:55:10
better change my name
<@zbyszek:fedora.im>
17:55:42
I think we should follow the procedure and ask for more candidates. For example with an additional week to propose. And then vote next week.
<@nirik:matrix.scrye.com>
17:55:45
proposal: reopen nominations, re-announce that, close nominations thursday and start voting friday with other groups?
<@sgallagh:fedora.im>
17:55:51
Proposal: Announce an extension until next Friday for nominations. Note that all current nominees will be accepted without a vote if no one else chooses to run.
<@zbyszek:fedora.im>
17:56:05
If no additional candidates show up, skip the election interviews.
<@zbyszek:fedora.im>
17:56:27
+1 to Stephen Gallagher's proposal
<@sgallagh:fedora.im>
17:56:29
nirik: 's proposal is basically mine, better worded. So I withdraw mine
<@nirik:matrix.scrye.com>
17:56:33
so then fesco elections are out of sync with the others?
<@amoloney:fedora.im>
17:56:43
This is my only election for this cycle
<@nirik:matrix.scrye.com>
17:56:53
I was proposing this thursday... ;) before the voting date
<@jistone:fedora.im>
17:56:56
next Friday, the 13th? 😱
<@zbyszek:fedora.im>
17:56:57
Sorry, I'm having Matrix/Element/Internet/DNS troubles today, things are showing up not in order.
<@amoloney:fedora.im>
17:57:04
there is also mindshare, but they had one open seat and a nomination so they dont require a vote
<@amoloney:fedora.im>
17:57:15
council moved to once-per-year elections
<@nirik:matrix.scrye.com>
17:57:20
ah right, and councile is once a year
<@zbyszek:fedora.im>
17:57:22
It'd be wrong to miss that opportunity.
<@nirik:matrix.scrye.com>
17:57:30
ah right, and council is once a year
<@conan_kudo:matrix.org>
17:57:36
it'll be fiiine
<@nirik:matrix.scrye.com>
17:57:56
ha. so, sure, I'd be fine with another week and moving voting back, etc.
<@conan_kudo:matrix.org>
17:58:16
we don't really have much room left
<@jistone:fedora.im>
17:58:26
plus interview time
<@conan_kudo:matrix.org>
17:58:29
so I'd be fine with just adding a couple more days
<@zbyszek:fedora.im>
17:59:15
nirik: please restate your proposal without the "other groups" part
<@nirik:matrix.scrye.com>
17:59:19
I'd be fine with that too. ;) I suspect anyone who might want to run already has had time to think about it..
<@nirik:matrix.scrye.com>
17:59:41
proposal: reopen nominations, re-announce that, close nominations this thursday and start voting friday.
<@sgallagh:fedora.im>
17:59:50
+1
<@zbyszek:fedora.im>
17:59:53
+1 to nirik
<@nirik:matrix.scrye.com>
18:00:04
I guess that means droppping interviews, but not sure how much people read them.
<@nirik:matrix.scrye.com>
18:01:08
I'm +1 to my own proposal too... any other votes?
<@amoloney:fedora.im>
18:01:19
If you get more candidates, you really should do interviews...
<@zbyszek:fedora.im>
18:01:25
Josh Stone, Conan Kudo, dcantrell votes?
<@conan_kudo:matrix.org>
18:01:29
+1
<@jistone:fedora.im>
18:01:39
+1 I guess, though it feels a bit self-serving
<@dcantrell:fedora.im>
18:01:46
+1
<@sgallagh:fedora.im>
18:02:11
Josh Stone: How so?
<@zbyszek:fedora.im>
18:02:31
Josh Stone: Just skipping the elections would be self-serving. We're going out of our way to follow the procedure and entice additional candidates.
<@nirik:matrix.scrye.com>
18:02:34
If we do interviews, voting would need to get pushed out... there really wouldn't be time
<@jistone:fedora.im>
18:03:25
I am a candidate deciding the rules of my own election. But that's the majority of us here, so be it.
<@salimma:fedora.im>
18:03:29
+1
<@jistone:fedora.im>
18:03:33
yes, giving it a little more time is better than nothing
<@zbyszek:fedora.im>
18:03:52
!agreed Reopen nominations until next Thursday and announce the reopening. The voting will start on Friday after that.
<@nirik:matrix.scrye.com>
18:04:00
Josh Stone: yeah, it was even weirder when we drafted and voted on the elections policy used to... elect us.
<@sgallagh:fedora.im>
18:04:04
Josh Stone: I'd argue we're enforcing the rules that the Ghost of FESCo Past established.
<@amoloney:fedora.im>
18:04:18
I see the predicament here, but if you get more candidates than open seats, allowing folks to interview for the seats is the fairest way to run the election. I would suggest we keep the interviews to a 1-2 day turnaround though if it comes to that
<@bittin:matrix.org>
18:04:26
ah was wondering about that and emailed Aoife Moloney earlier this or last week, thx for the answer :)
<@sgallagh:fedora.im>
18:04:44
It had to be bootstrapped somehow. Plus I'm pretty sure we got Council buy-in back then
<@nirik:matrix.scrye.com>
18:05:20
so, we are opening nominations under 2024-12-05 or 2024-12-12?
<@nirik:matrix.scrye.com>
18:05:27
so, we are opening nominations until 2024-12-05 or 2024-12-12?
<@zbyszek:fedora.im>
18:05:56
I understood this to mean 2024-12-12.
<@jistone:fedora.im>
18:06:03
the proposal said "this" Thursday
<@amoloney:fedora.im>
18:06:18
Open them until Friday 6th Dec? I thought you were giving the rest of this week
<@zbyszek:fedora.im>
18:06:21
Oh, right.
<@conan_kudo:matrix.org>
18:06:23
2024-12-05
<@zbyszek:fedora.im>
18:06:33
Sorry, my bad.
<@nirik:matrix.scrye.com>
18:06:34
schedules are anoying when you try and bend one part and the rest breaks
<@zbyszek:fedora.im>
18:06:56
OK, do we all agree with 2024-12-05?
<@sgallagh:fedora.im>
18:07:02
I'm fine with it
<@nirik:matrix.scrye.com>
18:07:18
it's just a few days, but should allow someone who didn't know/might decide last minute I would think.
<@zbyszek:fedora.im>
18:07:50
!agreed Reopen nominations until Thursday 2024-12-05 and announce the reopening. The voting will start on Friday.
<@sgallagh:fedora.im>
18:08:10
Friday == 2024-12-06, yes?
<@zbyszek:fedora.im>
18:08:19
!undo
<@sgallagh:fedora.im>
18:08:30
Just clarifyinh
<@zbyszek:fedora.im>
18:08:34
!agreed Reopen nominations until Thursday 2024-12-05 and announce the reopening. The voting will start on Friday 2024-12-06.
<@sgallagh:fedora.im>
18:08:35
Just clarifying
<@salimma:fedora.im>
18:09:02
just clarifying... we always use EOD UTC? or another timezone
<@amoloney:fedora.im>
18:09:13
Would you like me to send an email now announcing that nominations have reopened?
<@zbyszek:fedora.im>
18:09:20
Aoife Moloney: I think that if there's a completely new candidate, they can just write an interview and we can publish it immediately.
<@sgallagh:fedora.im>
18:09:20
"Sometime that day" :)
<@sgallagh:fedora.im>
18:09:33
Aoife Moloney: Yes, please.
<@zbyszek:fedora.im>
18:09:37
Yes please.
<@nirik:matrix.scrye.com>
18:09:51
and also unprotect the nominations wiki page. ;)
<@amoloney:fedora.im>
18:09:52
and are the rest of the candidates ok to use their previous interview answers?
<@conan_kudo:matrix.org>
18:09:55
UTC typically
<@amoloney:fedora.im>
18:10:14
that ole chestnut
<@zbyszek:fedora.im>
18:10:41
I'm fine with reusing my previous interview.
<@nirik:matrix.scrye.com>
18:11:10
Some things in old answers might be out of date/weird... but sure I guess.
<@nirik:matrix.scrye.com>
18:11:30
Or... we could just push voting to 2024-12-13 and do interviews properly next week?
<@amoloney:fedora.im>
18:12:05
If it comes to interviews needed, I'll double check with each candidate before publishing so if folks would like to update their answers they will have a *little bit of* time
<@zbyszek:fedora.im>
18:12:19
That sounds good enough.
<@zbyszek:fedora.im>
18:12:25
Let's move to the next topic.
<@zbyszek:fedora.im>
18:12:31
!topic Next week's meeting
<@zbyszek:fedora.im>
18:12:46
C'mon zodbot
<@zbyszek:fedora.im>
18:13:07
We decided to hold a meeting on video to discuss a private topic.
<@zbyszek:fedora.im>
18:13:22
So there will be public meeting next week.
<@zbyszek:fedora.im>
18:13:30
*no public meeting
<@sgallagh:fedora.im>
18:13:34
I was going to suggest that; I'm not sure we *decided* it yet. But I'm in favor.
<@zbyszek:fedora.im>
18:13:46
Yeah, OK, so anyone opposed?
<@nirik:matrix.scrye.com>
18:13:54
I have some questions, but can ask them out of band. ;)
<@salimma:fedora.im>
18:14:00
ah, so we'll just use our normal meeting time?
<@zbyszek:fedora.im>
18:14:07
Yes.
<@salimma:fedora.im>
18:14:09
that is probably easier than trying to find yet another time that works
<@sgallagh:fedora.im>
18:14:19
s/probably/unquestionably/
<@conan_kudo:matrix.org>
18:14:29
fine with me
<@salimma:fedora.im>
18:14:31
oh I might have to vote by proxy then
<@jistone:fedora.im>
18:14:45
it's fine with me
<@sgallagh:fedora.im>
18:14:55
A vote in the ticket would be honored, per our rules
<@zbyszek:fedora.im>
18:15:40
!info We'll hold a non-public meeting next week to discuss a private ticket.
<@zbyszek:fedora.im>
18:15:55
A volunteer to organize and hold the meeting?
<@sgallagh:fedora.im>
18:16:00
I'll generate an invite
<@zbyszek:fedora.im>
18:16:15
!action Stephen Gallagher will send out an invite to the meeting
<@zbyszek:fedora.im>
18:16:39
I hope it's shorter than that.
<@sgallagh:fedora.im>
18:16:54
Michel Lind 🎩 UTC-8 until Dec 7: I'm going to put a hard stop on it at the hour mark
<@sgallagh:fedora.im>
18:17:03
If it hasn't been decided by then
<@salimma:fedora.im>
18:17:20
thank you. I'll be in London next week and right after that hour is dinner :)
<@zbyszek:fedora.im>
18:17:33
That's a good motivation to wrap things up :)
<@zbyszek:fedora.im>
18:17:35
!topic Open Floor
<@sgallagh:fedora.im>
18:18:24
Just noting that I'll be largely incommunicado from Dec. 14 - Jan 5
<@nirik:matrix.scrye.com>
18:18:50
I wanted to mention a bit about the riscv sig plans in case folks were not following.
<@zbyszek:fedora.im>
18:19:15
nirik: go ahead
<@sgallagh:fedora.im>
18:19:17
Ooh, please do.
<@nirik:matrix.scrye.com>
18:19:52
Currently there's a external seperate koji hub and builders all over. The next step is for me to setup a secondary hub in fedora-infra (but still using external builders). This should allow maintainers to login and do scratch builds and such.
<@nirik:matrix.scrye.com>
18:20:28
Eventually, once hardware exists, we would then get hardware and bring that up. Also many folks have been working to decrease delta between fedora git and needed patches.
<@conan_kudo:matrix.org>
18:20:43
I will be gone from Dec 15 - Jan 2
<@nirik:matrix.scrye.com>
18:20:48
so in some ideal world future, we would have hardware and no deltas in specs and could try for primary status
<@sgallagh:fedora.im>
18:21:00
Question about the builders: are they reasonably performant or are they underpowered?
<@nirik:matrix.scrye.com>
18:21:14
Hopefully no one minds us using external builders right now (with a secondary setup)
<@jistone:fedora.im>
18:21:16
is there any known timeline for that hardware?
<@sgallagh:fedora.im>
18:21:25
As in, if we were to turn them on in main Koji, would they be far slower than other arches?
<@nirik:matrix.scrye.com>
18:21:29
right now, builders are pretty slow
<@nirik:matrix.scrye.com>
18:21:48
there are promises for hardware thats more performant and DC enabled coming 'soon'...
<@nirik:matrix.scrye.com>
18:22:01
but it's hard to say when that will be. next year sometime... perhaps
<@sgallagh:fedora.im>
18:22:38
Can you mention to the RISCV SIG that the ELN SIG would like to explore that bootstrap as well over the coming year?
<@nirik:matrix.scrye.com>
18:22:59
https://abologna.gitlab.io/fedora-riscv-tracker/ is a tracker for delta between fedora and riscv builds... it's MUCH smaller than it was. Lots of hard work from folks...
<@nirik:matrix.scrye.com>
18:23:14
yeah, there was some talk about doing that indeed.
<@sgallagh:fedora.im>
18:23:36
It just came up in my internal team meeting this morning by coincidence
<@nirik:matrix.scrye.com>
18:24:01
eln -> centos stream should be very doable I would think (given smaller package sets, etc)
<@nirik:matrix.scrye.com>
18:24:37
anyhow, just wanted to bring that up as a FYI, and if anyone has any objections to the plans
<@conan_kudo:matrix.org>
18:25:03
on the contrary, we should make external builders a norm
<@conan_kudo:matrix.org>
18:25:33
we lack performance and capacity on our internal builders, so giving people a way to contribute more performant stuff when fedora can't get them itself would be nice
<@sgallagh:fedora.im>
18:26:05
Conan Kudo: If you've got a Z-series lying around, I'm all ears 😀
<@conan_kudo:matrix.org>
18:26:18
Stephen Gallagher: who knows? 😛
<@nirik:matrix.scrye.com>
18:26:29
In order to really do that we need a bunch of other ponys... like reproducable builds and other stuff.
<@sgallagh:fedora.im>
18:26:54
External builders introduce potential supply chain risks though, so such a plan would need to be carefully considered
<@nirik:matrix.scrye.com>
18:27:11
right. Anyhow, thats all I had on it.
<@zbyszek:fedora.im>
18:27:25
nirik: thank you for the update.
<@zbyszek:fedora.im>
18:27:28
Anyone else?
<@zbyszek:fedora.im>
18:28:17
If nothing, I'll close in a minute.
<@sgallagh:fedora.im>
18:28:29
Thanks for chairing, zbyszek . It was a marathon.
<@zbyszek:fedora.im>
18:28:39
!endmeeting