2024-02-12 19:30:38 <@sgallagh:fedora.im> !startmeeting FESCO (2024-02-12) !meetingname fesco 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, @mhayden:fedora.im, @tstellar:fedora.im !topic Init Process !hi 2024-02-12 19:30:39 <@meetbot:fedora.im> Meeting started at 2024-02-12 19:30:38 UTC 2024-02-12 19:30:39 <@meetbot:fedora.im> The Meeting name is 'FESCO (2024-02-12) !meetingname fesco 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, @mhayden:fedora.im, @tstellar:fedora.im !topic Init Process !hi' 2024-02-12 19:31:01 <@humaton:fedora.im> !hi 2024-02-12 19:31:02 <@zodbot:fedora.im> TomΓ‘Ε‘ Hrčka (humaton) - he / him / his 2024-02-12 19:31:09 <@conan_kudo:matrix.org> uh oh 2024-02-12 19:31:10 <@salimma:fedora.im> !hi 2024-02-12 19:31:11 <@zodbot:fedora.im> Michel Lind (salimma) - he / him / his 2024-02-12 19:31:14 <@conan_kudo:matrix.org> !hi 2024-02-12 19:31:15 <@marcdeop:matrix.org> !hi 2024-02-12 19:31:16 <@zodbot:fedora.im> Neal Gompa (ngompa) - he / him / his 2024-02-12 19:31:17 <@zodbot:fedora.im> Marc Deop (marcdeop) 2024-02-12 19:31:23 <@nirik:matrix.scrye.com> morning 2024-02-12 19:31:44 <@conan_kudo:matrix.org> Stephen Gallagher: pasting multiple lines into one message causes meetbot to choke 2024-02-12 19:31:45 <@mhayden:fedora.im> !hi 2024-02-12 19:31:46 <@zodbot:fedora.im> Major Hayden (mhayden) - he / him / his 2024-02-12 19:31:57 <@tstellar:fedora.im> !hi 2024-02-12 19:31:57 <@zodbot:fedora.im> Tom Stellard (tstellar) 2024-02-12 19:32:01 <@sgallagh:fedora.im> Damn 2024-02-12 19:32:02 <@mhayden:fedora.im> πŸ‘‹ 2024-02-12 19:32:09 <@sgallagh:fedora.im> !topic Init Process 2024-02-12 19:32:13 <@conan_kudo:matrix.org> !meetingname fesco 2024-02-12 19:32:15 <@sgallagh:fedora.im> !hi 2024-02-12 19:32:16 <@zodbot:fedora.im> Stephen Gallagher (sgallagh) - he / him / his 2024-02-12 19:32:23 <@jonathanspw:fedora.im> !hi 2024-02-12 19:32:25 <@zodbot:fedora.im> Jonathan Wright (jonathanspw) 2024-02-12 19:32:30 <@nirik:matrix.scrye.com> yes, the meeting name is all hosed. 2024-02-12 19:32:43 <@nirik:matrix.scrye.com> !meetingname fesco 2024-02-12 19:32:53 <@conan_kudo:matrix.org> I'm pretty sure the log files are toast again too :( 2024-02-12 19:33:02 <@salimma:fedora.im> we can end meeting and start again 2024-02-12 19:33:05 <@nirik:matrix.scrye.com> it should work now, just need the name set right. 2024-02-12 19:33:05 <@salimma:fedora.im> and just throw away the bad log 2024-02-12 19:33:06 <@conan_kudo:matrix.org> and will need to be manually fixed 2024-02-12 19:33:28 <@nirik:matrix.scrye.com> I am pretty sure it takes that from the meeting name which I just set right. ;) 2024-02-12 19:33:39 <@conan_kudo:matrix.org> okay 2024-02-12 19:33:43 <@conan_kudo:matrix.org> we'll see how it goes 2024-02-12 19:33:55 <@nirik:matrix.scrye.com> buf if you want to end and restart we can. 2024-02-12 19:35:42 <@sgallagh:fedora.im> I count five FESCo members here, yes? 2024-02-12 19:35:54 <@sgallagh:fedora.im> And zbyszek said he'd be here, but a few minutes late 2024-02-12 19:36:04 <@conan_kudo:matrix.org> yes 2024-02-12 19:36:09 <@conan_kudo:matrix.org> we have quorum 2024-02-12 19:36:26 <@sgallagh:fedora.im> Alright, let's begin. We only have two topics on the agenda, but neither are simple. 2024-02-12 19:37:28 <@sgallagh:fedora.im> !topic #3165 Requesting FESCo assistance with issue about plasma-x11 2024-02-12 19:37:34 <@sgallagh:fedora.im> !fesco 3165 2024-02-12 19:37:34 <@zodbot:fedora.im> **fesco #3165** (https://pagure.io/fesco/issue/3165):**Requesting FESCo assistance with issue about plasma-x11** ● **Opened:** 2 weeks ago by siosm ● **Last Updated:** 7 hours ago ● **Assignee:** Not Assigned 2024-02-12 19:38:11 <@conan_kudo:matrix.org> where? 2024-02-12 19:38:12 <@sgallagh:fedora.im> The conversation has (to the best of my ability to follow) dissolved down to the same couple people repeating the same points. 2024-02-12 19:38:23 <@nirik:matrix.scrye.com> https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/PHKMRCL7IP4OP44PSVFE4PAKHXEOVRLA/ 2024-02-12 19:38:26 <@sgallagh:fedora.im> So I think it's probably time for FESCo to come to a decision 2024-02-12 19:38:55 <@zbyszek:fedora.im> Yeah, I think the proposal we have is good enough 2024-02-12 19:39:09 <@conan_kudo:matrix.org> nirik: as things are currently setup, there are very little in the way of advantages 2024-02-12 19:39:13 <@nirik:matrix.scrye.com> and... I can't find the other one but... what does it mean for the change 2024-02-12 19:39:17 <@zbyszek:fedora.im> I'd like to just approve it and move on 2024-02-12 19:39:26 <@sgallagh:fedora.im> Let's take 5 minutes for any last-minute comments, then vote on the proposal 2024-02-12 19:39:29 <@nirik:matrix.scrye.com> what is 'the proposal' ? 2024-02-12 19:40:04 <@conan_kudo:matrix.org> the implications of allowing the plasma-x11 packages in Fedora is almost the entirety of the point of the KDE Plasma 6 Change is overidden 2024-02-12 19:40:24 <@sgallagh:fedora.im> Current proposal: "KDE packages which reintroduce support for X11 are allowed in the main Fedora repositories, however they may not be included by default on any release-blocking deliverable (ISO, image, etc.). The KDE SIG should provide a notice before major changes, but is not responsible for ensuring that these packages adapt." 2024-02-12 19:40:26 <@conan_kudo:matrix.org> it might as well have not been submitted 2024-02-12 19:40:47 <@nirik:matrix.scrye.com> and under that proposal, no changes are made to the change? 2024-02-12 19:41:01 <@nirik:matrix.scrye.com> so it's misleading and wrong? 2024-02-12 19:41:32 <@salimma:fedora.im> can FESCo require that the packages reintroducing support for X11 not play epoch games? 2024-02-12 19:41:46 <@tstellar:fedora.im> I think that proposal is fair. If someone wants to maintain a package they should be allowed to do it as long as it's not creating a large burden on others. 2024-02-12 19:41:48 <@conan_kudo:matrix.org> with Kevin removing the deprecated() status from kwin-x11 and plasma-workspace-x11 and introducing an Epoch bump, there is no longer an indicator that this is lesser or unsupported too 2024-02-12 19:42:04 <@salimma:fedora.im> that being said, requiring the original change to be updated to advise users on the existence of these compat packages might be useful 2024-02-12 19:42:23 <@conan_kudo:matrix.org> Tom Stellard: it does though, we will get _all_ the bug reports, and we get the support burden without the code 2024-02-12 19:43:36 <@tstellar:fedora.im> Conan Kudo: So you are saying that you expect to get bug reports for the new compat packages filed against the main package? 2024-02-12 19:43:46 <@conan_kudo:matrix.org> Yes 2024-02-12 19:44:07 <@conan_kudo:matrix.org> moreover, the situation around plasma x11 upstream will continue to degrade as upstream works to untangle, split out, and eventually archive the x11 session code 2024-02-12 19:44:22 <@salimma:fedora.im> no doubt. I get bugs against the defunct 0install package (still in Bugzilla because it's in one epel) that are normally Anaconda or distribution bugs 2024-02-12 19:44:51 <@zbyszek:fedora.im> Maybe. But that's the nature of working in a distro.. sometimes other people do things which increase your burden. 2024-02-12 19:44:53 <@conan_kudo:matrix.org> I get bugs about fedora container images because I held a "containers" package (which was a D library that doesn't exist in fedora anymore) 2024-02-12 19:45:13 <@conan_kudo:matrix.org> it's a pretty common thing 2024-02-12 19:45:26 <@tstellar:fedora.im> Are there popular packages that will pull these compat packages in as dependencies? 2024-02-12 19:45:54 <@zbyszek:fedora.im> I hope not. 2024-02-12 19:46:02 <@conan_kudo:matrix.org> previously, they were marked as deprecated() so nobody was allowed to 2024-02-12 19:46:16 <@nirik:matrix.scrye.com> having this in a copr would make it pretty clear that it's not supported by the kde sig/would require no changes to the change... 2024-02-12 19:46:18 <@sgallagh:fedora.im> Honestly, I rather wish the X11 proponents would just fork KDE Plasma, give it a different name and submit it as a Spin... but that's definitely too much for us to demand. 2024-02-12 19:46:34 <@conan_kudo:matrix.org> the KDE SIG had worked for the past couple of years to cut off the dependency web around the x11 session stuff upstream and in fedora 2024-02-12 19:46:36 <@tstellar:fedora.im> So what kind of user would be installing these packages? 2024-02-12 19:47:01 <@stevenfalco:fedora.im> I would install them for KiCad support. 2024-02-12 19:47:11 <@sgallagh:fedora.im> Tom Stellard: Might want to rephrase that into a way that's less likely to get personal... 2024-02-12 19:47:31 <@conan_kudo:matrix.org> a copr has existed made by me since last fall, but there is now one by the kde sig that automatically tracks our packages in fedora and builds them 2024-02-12 19:47:35 <@nirik:matrix.scrye.com> I'm sympathetic to someone wanting to package the thing they like... I am just not sure what advantages they get from it being in the main collection, and I am not sure how we would reword the change if we did allow them in. 2024-02-12 19:47:37 <@tstellar:fedora.im> Sorry I mean what are the uses cases for these compat packages if no other Fedora packages depend on them? 2024-02-12 19:47:42 <@conan_kudo:matrix.org> KiCad works in Plasma Wayland as an X11 app through XWayland 2024-02-12 19:48:07 <@nirik:matrix.scrye.com> Tom Stellard: to allow a x11 session instead of a wayland one 2024-02-12 19:48:15 <@conan_kudo:matrix.org> And the app can force itself to always run as an Xwayland app 2024-02-12 19:48:21 <@zbyszek:fedora.im> It's easier for users is it's in the main repos... 2024-02-12 19:48:29 <@stevenfalco:fedora.im> Not really. There are many restrictions in XWayland, as the upstream KiCad team have discussed. 2024-02-12 19:48:54 <@nirik:matrix.scrye.com> zbyszek: but... they still would need to read some document on what to install, etc... adding a copr is just one more step in such a document. 2024-02-12 19:48:55 <@sgallagh:fedora.im> Tom Stellard: There are some situations (hardware limitations in particular) where users CANNOT run KDE on Wayland. Upstream feels that these are bugs that should be fixed in KDE 2024-02-12 19:49:41 <@sgallagh:fedora.im> Certain downstream users believe that this means that we must not retire the decaying X11 session as a result. 2024-02-12 19:50:06 <@zbyszek:fedora.im> I just feel that's it's against the spirit of fedora to disallow a package. 2024-02-12 19:50:23 <@zbyszek:fedora.im> (When there's no good enough reason.) 2024-02-12 19:50:26 <@sgallagh:fedora.im> zbyszek: I feel the same way, but it wouldn't be the first time. 2024-02-12 19:50:35 <@sgallagh:fedora.im> We've made opinionated decisions like this before. 2024-02-12 19:50:50 <@conan_kudo:matrix.org> There is only one hardware limitation I'm aware of, a specific aspeed graphics driver that is in poor shape in Linux and unmaintained in x11. 2024-02-12 19:51:02 <@sgallagh:fedora.im> For example, we don't allow alternative libc implementations 2024-02-12 19:51:26 <@nirik:matrix.scrye.com> kernels is the one I know of. :) 2024-02-12 19:51:30 <@zbyszek:fedora.im> Yeah, but leaf packages are different. 2024-02-12 19:51:37 <@tstellar:fedora.im> Stephen Gallagher: What about musl? https://src.fedoraproject.org/rpms/musl 2024-02-12 19:51:53 <@conan_kudo:matrix.org> Musl is explicitly not supported as a system libc 2024-02-12 19:51:58 <@conan_kudo:matrix.org> I deliberately made sure it wasn't 2024-02-12 19:52:02 <@stevenfalco:fedora.im> Upstream KDE still list showstoppers. Why would we pull the plug on X11 until upstream is satisfied? 2024-02-12 19:52:08 <@conan_kudo:matrix.org> It is only in Fedora specifically for wasm and enarx 2024-02-12 19:52:29 <@salimma:fedora.im> so actually this will be a better experience than anything any distro -x11 package can provide right? 2024-02-12 19:52:34 <@conan_kudo:matrix.org> There is no showstoppers page anymore. There are a pair of significant issues pages for both X11 and Wayland. 2024-02-12 19:52:40 <@salimma:fedora.im> COPR can auto rebuild, you can't do that with Koji/Bodhi 2024-02-12 19:52:56 <@marcdeop:matrix.org> I don't think that statement is accurate. 2024-02-12 19:52:59 <@conan_kudo:matrix.org> https://community.kde.org/Plasma/Wayland_Known_Significant_Issues 2024-02-12 19:53:04 <@conan_kudo:matrix.org> https://community.kde.org/Plasma/X11_Known_Significant_Issues 2024-02-12 19:53:07 <@salimma:fedora.im> maybe a halfway point is providing a transition package, *in* Fedora, that just enable the COPR 2024-02-12 19:53:26 <@stevenfalco:fedora.im> Fine. Wrong terminology. But I feel the point is still valid. 2024-02-12 19:53:35 <@tstellar:fedora.im> I'm ready to vote fwiw. 2024-02-12 19:53:42 <@salimma:fedora.im> (not sure if policy wise we can recommend COPR from an official package, but since this is owned by the SIG...?) 2024-02-12 19:53:57 <@conan_kudo:matrix.org> We have a SIG wiki page created to point to the COPR as needed 2024-02-12 19:54:05 <@sgallagh:fedora.im> marcdeop: Anyone can speak in a FESCo meeting. 2024-02-12 19:54:09 <@sgallagh:fedora.im> But only FESCo members vote 2024-02-12 19:54:12 <@conan_kudo:matrix.org> marcdeop: you are allowed to comment 2024-02-12 19:54:19 <@conan_kudo:matrix.org> feel free to speak up and say what you like 2024-02-12 19:54:59 <@conan_kudo:matrix.org> yes, additionally the copr has a higher priority than distro packages and won't get uninstalled 2024-02-12 19:56:43 <@davide:cavalca.name> are there usecases where going with the copr vs the in-distro packages _wouldn't_ work? 2024-02-12 19:57:05 <@jistone:fedora.im> !hi 2024-02-12 19:57:06 <@zodbot:fedora.im> Josh Stone (jistone) - he / him / his 2024-02-12 19:57:06 <@nirik:matrix.scrye.com> I asked on list, but I did not see a reply from the x11 proponents. 2024-02-12 19:57:13 <@jistone:fedora.im> (sorry I'm late, I'll try to catch up...) 2024-02-12 19:58:07 <@nirik:matrix.scrye.com> only one I can think of is bugzilla... no bugzilla components for copr and harder for kde sig to reassign bugs. 2024-02-12 19:58:07 <@farchord:matrix.org> It may be a question of visibility... having the packages in the official repos add credibility to the packages' legitimacy I think. And it's harder (in theory) to find in the copr repos. 2024-02-12 19:58:08 <@humaton:fedora.im> I feel like there is no good choice here, either we maintain legacy technology and "undo" KDE SIG work or we forbid some packages. Both choices seem against the spirit of fedora to me. 2024-02-12 19:58:31 <@conan_kudo:matrix.org> I believe the main concern was that we wouldn't keep the copr running 2024-02-12 19:58:50 <@sgallagh:fedora.im> Farchord [Fedora]: That's part of the argument in favor of the COPR: that it ISN'T official and is less credible. 2024-02-12 19:58:52 <@conan_kudo:matrix.org> but I have committed in the discussion that we'd keep it going for at least 2-3 years and re-evaluate then 2024-02-12 19:59:06 <@sgallagh:fedora.im> Because the KDE SIG does not want to support it. 2024-02-12 19:59:10 <@farchord:matrix.org> Correct. I agree. 2024-02-12 20:00:08 <@conan_kudo:matrix.org> that said, strictly speaking, anyone can build a copr rebuild with the x11 flag turned back on, none of us intend to remove it unless the code is actually gone from the packages first 2024-02-12 20:00:10 <@marcdeop:matrix.org> let's try to make something clear: the _support_ we mention is not only bugzilla tickets. It's channel support, it's direct emails, it's mailing lists, it's fedora discussions, it's reddit.... it's waaaay more than simply getting tickets assigned 2024-02-12 20:00:18 <@sgallagh:fedora.im> As a straw poll (non-binding), how many FESCo members would be in favor of asking the X11 supporters to fork the deskop and rename it? 2024-02-12 20:00:31 <@conan_kudo:matrix.org> +1 2024-02-12 20:00:36 <@nirik:matrix.scrye.com> I suppose I could support adding the packages with the proposal, but... it would need some changes to the change... and I am not sure what that would look like... "x11 session support has been removed, but a unrelated goup of packagers have created support for it again, see this doc' reads really... not great. 2024-02-12 20:01:10 <@conan_kudo:matrix.org> the biggest problem with this is that it also ruins the SIG's credibility with the wider community 2024-02-12 20:01:16 <@jistone:fedora.im> I'm in favor of softly floating the idea -- but without implying that we're going to force that 2024-02-12 20:01:23 <@conan_kudo:matrix.org> we had been actively communicating we would do this with Plasma 6.0 for almost 4 years 2024-02-12 20:01:25 <@farchord:matrix.org> And Fedora's as a whole 2024-02-12 20:01:32 <@nirik:matrix.scrye.com> that seems like a ton of work for... 2 packages? 2024-02-12 20:02:08 <@marcdeop:matrix.org> it's not 2 simple packages I guess. They are the _main_ package of the KDE experience 2024-02-12 20:02:29 <@conan_kudo:matrix.org> plasma-workspace and kwin make up almost the entirety of the core shell 2024-02-12 20:02:44 <@nirik:matrix.scrye.com> sure, but forking everything and making a new name and all that... 2024-02-12 20:03:10 <@sgallagh:fedora.im> nirik: It's analogous to GNOME vs. Mate once upon a time. 2024-02-12 20:03:16 <@farchord:matrix.org> So essentially, they'd have to create a "KDE X11 SiG"? 2024-02-12 20:03:23 <@marcdeop:matrix.org> not only communicated but discussed even upstream if they could observe removing X11 support from Plasma 6. They defferred to later point in time of the 6 release because the code base was not ready 2024-02-12 20:03:28 <@conan_kudo:matrix.org> and GNOME vs Cinnamon 2024-02-12 20:03:31 <@nirik:matrix.scrye.com> sure. But in that case there were a lot more differences... more like kf5 vs kde4. 2024-02-12 20:03:40 <@conan_kudo:matrix.org> more due to lack of time rather than anything else 2024-02-12 20:03:49 <@conan_kudo:matrix.org> only so many things to be tackled at once 2024-02-12 20:05:09 <@marcdeop:matrix.org> for kwin in particular, there's a change that building X11 will be optional in 6.1 or 6.2 release 2024-02-12 20:05:20 <@marcdeop:matrix.org> for kwin in particular, there's a chance that building X11 will be optional in 6.1 or 6.2 release 2024-02-12 20:05:26 <@nirik:matrix.scrye.com> Is it worth trying to punt and see if they would be willing to do a copr? Or is that likely non viable for them? (I don't recall if that was said in the long thread or not) 2024-02-12 20:05:33 <@marcdeop:matrix.org> for kwin in particular, there's a chance that building with X11 support will be optional in 6.1 or 6.2 release 2024-02-12 20:05:40 <@sgallagh:fedora.im> A COPR already exists and they don't want to use it 2024-02-12 20:05:49 <@conan_kudo:matrix.org> the copr already exists, they don't seem to be happy with it 2024-02-12 20:06:11 <@conan_kudo:matrix.org> actually two coprs exist because I never deleted the original one I made for the guy using KDE in VMware Workstation 2024-02-12 20:06:35 <@conan_kudo:matrix.org> (though that guy actually contributed and fixed the bugs for Plasma Wayland in VMware, so yay) 2024-02-12 20:06:41 <@farchord:matrix.org> It's more that they are afraid we are going to "yank" it from under them out of spite 2024-02-12 20:06:52 <@farchord:matrix.org> Which..... why!!? 2024-02-12 20:07:29 <@conan_kudo:matrix.org> hell, we made a new one in the kde sig namespace specifically to prove it won't get yanked out of spite 2024-02-12 20:07:51 <@salimma:fedora.im> can the commitment to maintain it for 2-3 years minimum be added to the Change, if it's not already? 2024-02-12 20:07:54 <@nirik:matrix.scrye.com> well, they could make their own. ;) 2024-02-12 20:07:59 <@conan_kudo:matrix.org> right ^ 2024-02-12 20:08:17 <@salimma:fedora.im> and/or to not nuke support from the spec unless that support is gone upstream 2024-02-12 20:08:17 <@conan_kudo:matrix.org> I'd be fine with FESCo binding us to retain the bconds in our spec files for copr rebuilds 2024-02-12 20:08:28 <@nirik:matrix.scrye.com> but I see in the thread: "I do not see how it is a better experience to have to enable an additional repository, for which dnf will warn the user that it is "unsupported" (even if it does not have that in the name as the kde6-x11-unsupported Copr does) than to have those packages available in the standard Fedora repository." 2024-02-12 20:08:32 <@sgallagh:fedora.im> Given that we're burning through our time, can we at least pause and see how people would vote on the CURRENT proposal? 2024-02-12 20:08:39 <@farchord:matrix.org> That's always been the intent afaik 2024-02-12 20:08:40 <@sgallagh:fedora.im> If that fails, we can return to discussing new ones. 2024-02-12 20:08:43 <@salimma:fedora.im> if it's in a document vetted by FESCo and thus require FESCo to vote to be changed, IDK what stronger promise that can be 2024-02-12 20:09:26 <@marcdeop:matrix.org> I have aquestion that might be relevant to the whole thing 2024-02-12 20:09:33 <@conan_kudo:matrix.org> I replied to that I don't mind dropping the suffix, the SIG agreed to have that originally so that it is a constant reminder especially for automated installs that don't see the warning 2024-02-12 20:09:39 <@jistone:fedora.im> can you restate the proposal that you'd like a vote on? 2024-02-12 20:09:50 <@sgallagh:fedora.im> OK, as chair I will recognize one more question from marcdeop and then let's try for a vote. 2024-02-12 20:09:53 <@marcdeop:matrix.org> are we sure that these packages won't be automatically upgraded _right now_ on the systems that have X11 session installed? 2024-02-12 20:10:12 <@marcdeop:matrix.org> because Kevin seemed to mention that wouldn't happen but... I am not so sure 2024-02-12 20:10:32 <@marcdeop:matrix.org> we wanted to move everybody to the wayland session when upgrading to F40 2024-02-12 20:11:29 <@sgallagh:fedora.im> My understanding is that this is what the KDE SIG intended and that Kevin is trying to bypass with "epoch" tricks. 2024-02-12 20:11:36 <@conan_kudo:matrix.org> yes 2024-02-12 20:12:15 <@salimma:fedora.im> I specifically suggested letting people upgrade and get opted in to Wayland first, with the option to then reenable X11 support, and Kevin Kofler ignored that part in his reply 2024-02-12 20:12:40 <@salimma:fedora.im> so yeah I think he intentionally wanted to make it harder for people to opt out of x11, given his unexplained insistence on using Epoch 2024-02-12 20:12:58 <@jistone:fedora.im> didn't he previously claim that this upgrade did work under his scheme? 2024-02-12 20:13:37 <@farchord:matrix.org> It would, it would take the existing x11 package in f38/f39 and upgrade it to his in f40 2024-02-12 20:13:48 <@conan_kudo:matrix.org> In general, if we allow this Change to be effectively undone in this manner, it sets the precedent for everyone else too as desktops transition to Wayland. 2024-02-12 20:14:14 <@jistone:fedora.im> no, I recall a claim that the obsoletes did work into Wayland even with his packages, and then they could be manually installed back 2024-02-12 20:14:31 <@jistone:fedora.im> this point matters for my vote 2024-02-12 20:14:58 <@marcdeop:matrix.org> Josh Stone: this is the thing I would like to _know_ if it's true or not. Thus why I raised the question πŸ˜… 2024-02-12 20:15:10 <@sgallagh:fedora.im> Josh Stone: Do you have a reference? Because I don't recall hearing that and to the best of my knowledge, his proposed implementation would do the opposite. 2024-02-12 20:15:19 <@zbyszek:fedora.im> Gnome transitioned, even though the x11 was easily available... 2024-02-12 20:15:21 <@jistone:fedora.im> I'll try to find it 2024-02-12 20:15:45 <@solopasha:matrix.org> https://bugzilla.redhat.com/show_bug.cgi?id=2260793#c13 2024-02-12 20:15:48 <@conan_kudo:matrix.org> The transition is not complete. GNOME's completion of that transition is slated within the next couple of releases. 2024-02-12 20:15:52 <@nirik:matrix.scrye.com> https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/message/JUM7XUFO3OZEQA2CHFTMVVIFZ3BMTIAA/ 2024-02-12 20:15:55 <@jistone:fedora.im> https://bugzilla.redhat.com/show_bug.cgi?id=2260793#c13 2024-02-12 20:15:59 <@conan_kudo:matrix.org> That includes the removal of the X11 code. 2024-02-12 20:19:06 <@conan_kudo:matrix.org> It's interesting that he's willing to use COPRs only in some parts but not others 2024-02-12 20:19:30 <@conan_kudo:matrix.org> That makes the motivation squarely to unwind our Change. :( 2024-02-12 20:19:30 <@marcdeop:matrix.org> let's not get into that discussion and stay on technical questions πŸ™‚ 2024-02-12 20:20:09 <@stevenfalco:fedora.im> Sure - I can switch to NEON :-) 2024-02-12 20:20:51 <@salimma:fedora.im> Fedora DOS edition 2024-02-12 20:21:02 <@sgallagh:fedora.im> OK, I'm really putting down the gavel now. 2024-02-12 20:21:14 <@solopasha:matrix.org> How are that packages tracked in the Copr repository with x11 stuff? Distgit autorebuild just starts a build immediately after a new commit appears. So order is not preserved and packages from a side tag aren't present in the buildroot. 2024-02-12 20:21:24 <@sgallagh:fedora.im> We'll vote on zbyszek's proposal. If that fails, we'll accept one more for voting. 2024-02-12 20:21:29 <@sgallagh:fedora.im> If that fails, we'll adjourn to the ticket. 2024-02-12 20:21:49 <@sgallagh:fedora.im> zbyszek's proposal: "KDE packages which reintroduce support for X11 are allowed in the main Fedora repositories, however they may not be included by default on any release-blocking deliverable (ISO, image, etc.). The KDE SIG should provide a notice before major changes, but is not responsible for ensuring that these packages adapt." 2024-02-12 20:22:19 <@conan_kudo:matrix.org> For now, we're using distgit autorebuild, but we can change to have our scripts do it as part of submitting to bodhi 2024-02-12 20:22:30 <@conan_kudo:matrix.org> that way we can ensure the side tag repos are included in the buildroot 2024-02-12 20:22:30 <@tstellar:fedora.im> +1 2024-02-12 20:22:37 <@conan_kudo:matrix.org> -1 2024-02-12 20:23:35 <@sgallagh:fedora.im> Actually, may I make one amendment to that proposal before we continue? 2024-02-12 20:23:38 <@conan_kudo:matrix.org> This proposal demolishes the Change almost entirely 2024-02-12 20:23:38 <@sgallagh:fedora.im> (Breaking my own rule...) 2024-02-12 20:24:00 <@jistone:fedora.im> +1, on the condition that the intended Change upgrade path does work. Maybe that should be explicit in the proposal though. 2024-02-12 20:24:29 <@nirik:matrix.scrye.com> The bulk of the change is surely plasma 6? not the x11 thing (although thats what people focus on) 2024-02-12 20:24:38 <@conan_kudo:matrix.org> no 2024-02-12 20:24:41 <@sgallagh:fedora.im> Proposal: "KDE packages which reintroduce support for X11 are allowed in the main Fedora repositories, however they may not be included by default on any release-blocking deliverable (ISO, image, etc.). The KDE SIG should provide a notice before major changes, but is not responsible for ensuring that these packages adapt. Upgrades from F38 and F39 will be automatically migrate to Wayland." 2024-02-12 20:24:46 <@conan_kudo:matrix.org> the bulk of the change was about getting rid of x11 2024-02-12 20:24:58 <@conan_kudo:matrix.org> Plasma 6 itself was not that big of a deal 2024-02-12 20:25:00 <@sgallagh:fedora.im> Proposal: "KDE packages which reintroduce support for X11 are allowed in the main Fedora repositories, however they may not be included by default on any release-blocking deliverable (ISO, image, etc.). The KDE SIG should provide a notice before major changes, but is not responsible for ensuring that these packages adapt. Upgrades from F38 and F39 will be automatically migrated to Wayland." 2024-02-12 20:25:14 <@zbyszek:fedora.im> +1 to Stephen Gallagher 2024-02-12 20:25:19 <@tstellar:fedora.im> I'm still +1 even with Stephen Gallagher amendment. 2024-02-12 20:25:23 <@salimma:fedora.im> to Stephen's modification, how about adding "and the packages reintroducing X11 support cannot interfere with this" 2024-02-12 20:25:38 <@salimma:fedora.im> otherwise it's not clear if it's binding on any such package or not 2024-02-12 20:25:43 <@sgallagh:fedora.im> To follow our normal policy of "Upgrades should try to achieve the same results as fresh installs" 2024-02-12 20:25:55 <@conan_kudo:matrix.org> the "deprecated()" provides must be retained 2024-02-12 20:26:01 <@conan_kudo:matrix.org> and epoch games must not be permitted 2024-02-12 20:26:16 <@conan_kudo:matrix.org> if we're going to allow this at all, it needs to be appropriately constrained 2024-02-12 20:26:28 <@conan_kudo:matrix.org> no epoch: 1, no eliminating deprecation, nothing 2024-02-12 20:26:29 <@sgallagh:fedora.im> Conan Kudo: I'm intentionally not trying to specify technical details. But that would be implicit, yes 2024-02-12 20:26:53 <@conan_kudo:matrix.org> I'm sorry, but I have been burned too much to not be explicit 2024-02-12 20:26:54 <@farchord:matrix.org> Agreed 2024-02-12 20:27:29 <@salimma:fedora.im> if we here all agree that epoch should not be used and deprecated() should not be removed, I think it will save headaches down the road to just make it explicit 2024-02-12 20:27:33 <@nhanlon:beeper.com> I'll have you know that I've never once done that and don't intend to start soon ;) 2024-02-12 20:27:51 <@sgallagh:fedora.im> Michel Lind🎩: It's binding; if they introduce something that breaks that upgrade promise, they're in violation 2024-02-12 20:27:54 <@salimma:fedora.im> the mailing list debate is already full of arguments about what people mean on both sides 2024-02-12 20:27:59 <@zbyszek:fedora.im> "upgrades … will be automatically migrated to Wayland" β€” this should be enough. 2024-02-12 20:27:59 <@nhanlon:beeper.com> just been lurking, but also happy where the conversation has gone. appreciate everyone's candor. 2024-02-12 20:28:00 <@jistone:fedora.im> +1 2024-02-12 20:28:09 <@conan_kudo:matrix.org> As it is, I'm deeply unhappy because now our Change has been unwound if this is approved 2024-02-12 20:28:26 <@nirik:matrix.scrye.com> I'm a bit on the fence here, but I suppose a weak +1 at this point. I'm not liking that the change will seem wrong, but perhaps it's stil right with no official support... 2024-02-12 20:28:28 <@conan_kudo:matrix.org> and at this point, I guess I should expect it to be 2024-02-12 20:28:35 <@jistone:fedora.im> on the technical solutions, has anyone considered *renaming* the reintroduced packages? e.g. kwin-alt-x11 with no upgrade path from before 2024-02-12 20:28:52 <@salimma:fedora.im> I like compat-kwin-x11 myself or kwin-x11-compat 2024-02-12 20:29:04 <@jistone:fedora.im> sure, bikeshed away :) 2024-02-12 20:29:08 <@conan_kudo:matrix.org> it was proposed on list, no one responded 2024-02-12 20:29:13 <@salimma:fedora.im> yeah 2024-02-12 20:29:44 <@salimma:fedora.im> that was part of what I proposed iirc 2024-02-12 20:29:53 <@nirik:matrix.scrye.com> perhaps that should be added to review(s)? 2024-02-12 20:30:59 <@sgallagh:fedora.im> I need formal votes on the current proposal. I see (+4, 0, -1) at the moment 2024-02-12 20:31:28 <@mhayden:fedora.im> Sorry, multitasking here, Stephen Gallagher. I'm okay going +1 here. 2024-02-12 20:31:59 <@conan_kudo:matrix.org> πŸ™β€β™‚οΈ 2024-02-12 20:33:44 <@farchord:matrix.org> I.... understand your point of view.... but at least this will make it known that, while, yes, we do want to get rid of xorg, we also acknowledge that for some people that's just wholly unacceptable. We... I guess... are leaving the door open for this but making it well heard that it's a compat package, and it has restrictions 2024-02-12 20:33:45 <@sgallagh:fedora.im> !agreed KDE packages which reintroduce support for X11 are allowed in the main Fedora repositories, however they may not be included by default on any release-blocking deliverable (ISO, image, etc.). The KDE SIG should provide a notice before major changes, but is not responsible for ensuring that these packages adapt. Upgrades from F38 and F39 will be automatically migrated to Wayland. (+5, 0, -1) 2024-02-12 20:33:54 <@zbyszek:fedora.im> phew 2024-02-12 20:34:27 <@sgallagh:fedora.im> Let's acknowledge that no one here is making this decision lightly or without a great deal of consideration. 2024-02-12 20:34:40 <@nirik:matrix.scrye.com> yeah, but with the change the same... the only way anyone will know this exists is from looking at searches.... 2024-02-12 20:34:40 <@sgallagh:fedora.im> And also that no one is getting everything they wanted from it. 2024-02-12 20:35:00 <@conan_kudo:matrix.org> your agreed statement does not include the necessary restrictions 2024-02-12 20:35:35 <@conan_kudo:matrix.org> please revise it accordingly 2024-02-12 20:35:45 <@aleasto:matrix.org> can you also clarify that _major_ means _every_ change since every minor upgrade is breaking? 2024-02-12 20:36:01 <@zbyszek:fedora.im> Guys, this is the proposal that was voted. 2024-02-12 20:36:20 <@zbyszek:fedora.im> It passed the vote, let's move on. It's been 66 minutes. 2024-02-12 20:36:31 <@sgallagh:fedora.im> Conan Kudo: As stated earlier, I'm not attempting to cover every technical detail here. Violation of the spirit of this agreement will still be addressed accordingly. 2024-02-12 20:36:46 <@conan_kudo:matrix.org> I hope so. 2024-02-12 20:37:05 <@jistone:fedora.im> please apply the same trust that you were hoping to get from others 2024-02-12 20:37:16 <@conan_kudo:matrix.org> I do not apply this trust to Kevin Kofler. 2024-02-12 20:37:22 <@conan_kudo:matrix.org> I cannot. 2024-02-12 20:37:28 <@sgallagh:fedora.im> Now, we're over time. The other topic will be deferred. 2024-02-12 20:37:40 <@sgallagh:fedora.im> !topic Next Week's Chair 2024-02-12 20:37:53 <@sgallagh:fedora.im> I will be unavailable next week. 2024-02-12 20:38:21 <@zbyszek:fedora.im> I can do it. 2024-02-12 20:38:25 <@sgallagh:fedora.im> Thank you 2024-02-12 20:38:36 <@sgallagh:fedora.im> !action zbyszek to chair the 2024-02-19 meeting 2024-02-12 20:38:41 <@sgallagh:fedora.im> !topic Open Floor 2024-02-12 20:40:03 <@sgallagh:fedora.im> Anyone? 2024-02-12 20:40:17 <@sgallagh:fedora.im> I'll wait 120s and then close the meeting. 2024-02-12 20:42:10 <@sgallagh:fedora.im> !endmeeting