<@arthurzam> who should chair? I don't care, but ulm did most of the job before now [21:01] <@ulm> yeah, I can do it <@ulm> !proj council (council@gentoo.org) arthurzam, dilfridge, mgorny, robbat2, sam, soap, ulm <@ulm> agenda is here: https://archives.gentoo.org/gentoo-project/uwm8k8thi@gentoo.org/ <@ulm> 1. roll call * @arthurzam here * @soap here * @robbat2 present * @dilfridge here * @mgorny here * @ulm here [21:02] <@ulm> sam_ is missing <@arthurzam> attempting call to signal <@soap> also tried [21:04] <@arthurzam> no reply :( <@ulm> let's wait until 20:07 (5 min after roll call) <@dilfridge> swallowed by a large red dragon <@arthurzam> he replied [21:05] <@arthurzam> soon *** sam__ (~sam@82.8.138.118) has joined channel #gentoo-council [21:06] *** sam__ (~sam@82.8.138.118) is now known as sam^2 hello <@ulm> moving on :) <@ulm> 2. Constitute the new council <@ulm> - Decide on time of meetings. The previous council had its meetings on the 2nd Sunday of every month at 19:00 UTC [21:07] <@arthurzam> Good for me <@dilfridge> let's just continue sure <@ulm> - Vote for continuing last council's workflow considering sending a call for agenda items (two weeks in advance), sending the agenda (one week in advance) and have the meeting focussed, i.e. have major discussions on the gentoo-project mailing list prior to the meeting <@ulm> do we have to vote? <@dilfridge> no [21:08] <@ulm> - Appoint chairmen for this term's meetings <@ulm> aug/sep? <@arthurzam> ulm: you already have jul :) <@ulm> I could do August <@arthurzam> I'm fine with any month <@ulm> sep/oct then [21:09] <@soap> I can't until october <@mgorny> nov/dec for me? <@ulm> noted <@ulm> jan/feb? march/april for me please <@arthurzam> dilfridge: you have any preferences? [21:10] * @sam_ here <@sam_> sorry, it's too hot upstairs so was away from PC <@dilfridge> the last ones for me please <@dilfridge> may/june <@robbat2> I can take Jan, but not the feb <@ulm> sam_ mar/apr, dilfridge may/jun <@arthurzam> soap: does feb work for you? <@soap> sure <@ulm> robbat2 jan, soap feb <@mgorny> we should prolong council term to 14 months to have equal share * @mgorny hides [21:11] <@arthurzam> ulm ulm arthurzam arthurzam mgorny mgorny <@arthurzam> robbat2 soap sam sam dilfridge dilfridge <@dilfridge> (expect random timezones from me, but I'll definitely have time :) <@ulm> good :) <@ulm> - Exchange contact information after the meeting <@sam_> march/april for me please <@ulm> new item, just so that we remember <@arthurzam> sam_: we did it as such (03 & 04) <@ulm> 3. GLEP 83 update [21:12] <@ulm> https://public-inbox.gentoo.org/gentoo-dev/20250619173354.22787-1-ulm@gentoo.org/ <@ulm> anybody wants to discuss this? <@ulm> sam_, arthurzam? <@arthurzam> 10 years ago feels excessive, considering it is around half life of Gentoo [21:13] <@mgorny> my feeling is that we can't enforce this, so it feels kinda intrusive <@mgorny> but i don't exactly mind having it <@dilfridge> the eternal unmerge support is kinda pointless when the package manager itself is delivered with an eapi that the old state didnt understand it's not about enforcement, just a guideline for when it's okay to remove like <@arthurzam> we have like 2+1 package managers: portage will always support I guess, pkgcraft supports EAPI>=7, pkgcore lost package manager official support [21:14] this sort of makes the decision for us if We Don'T Know and for portage I've been struggling to come up with a policy for eventual removal this fixes that for us ;) <@soap> I find the horizon a bit too long <@arthurzam> Therefore if portage team (sam) agrees, I don't think we have any issue? <@ulm> the times were chosen so we could drop the oldest EAPIs now <@soap> ulm: which ones exactly? soap: it's longer than the periods you and I discussed in the past, I think, yeah [21:15] <@ulm> 0 to 4 IIRC <@robbat2> i wish we had the popcon-related project running still; so we can estimate how many users still have packages with very old EAPIs installed <@soap> or rather, when can we drop EAPI 6 then? <@arthurzam> We still have profiles which are EAPI=5 I think <@ulm> we just removed its last ebuild less than a year ago the more important question for me is: [21:16] + +Package managers may drop (or omit) support for an EAPI only when all +of the following conditions apply: you might read this to imply you can't even start to make it "unmerge-only" before 10 years? but removing that unmerging support is that we really need otherwise we get very little benefit <@soap> exactly [21:17] in fact, it's worse because we'd rip out code AND couldn't test it <@ulm> I'd really wished that we' have had this discussion when I posted the draft to the ML :( i think i'll write these bits up on the ML and we'll discuss it there yes, i'm sorry, it's been a busy period for me <@ulm> moving on then? [21:18] <@ulm> 4. EAPI 7 deprecation <@ulm> https://public-inbox.gentoo.org/gentoo-project/uikkcvzsz@gentoo.org/ <@ulm> I believe we can just vote? <@arthurzam> finally :) [21:19] <@ulm> motion: deprecate EAPI 7 * @mgorny yes <@soap> yup * sam^2 yes * @robbat2 aye * @soap yes * @arthurzam yes * @ulm yes * @dilfridge yes <@ulm> unanimous <@ulm> 5. Discussion on leaving GitHub <@arthurzam> Porters can use https://packages.gentoo.org/packages/eapi7 to help us <@dilfridge> it just dropped below 5000 ebuilds btw [21:20] <@ulm> https://public-inbox.gentoo.org/gentoo-project/eb13f46cf4aac83365dff0add53f49cb7c86d05c.camel@gentoo.org/ ztrawhcse: mgorny: ^^ I'll let mgorny (or Eli) give an intro, then I have some misc. comments to make <@mgorny> long story short, we went to github because it was popular and people wanted to contribute which is nice [21:21] <@mgorny> unfortunately, it's being going downhill for a long time now <@mgorny> the site is heavy, reviews are pretty shoddy, and they're trying to push copilot everywhere [21:22] <@ulm> what exactly do you mean by "heavy" <@mgorny> and with pretty much every alternate forge supporting github oauth, the cost of switching to users is relatively small <@mgorny> ulm: like bad & slow UI [21:23] <@mgorny> heavy on the browser <@soap> it's really becoming enshittified <@mgorny> and on top of that, there are more people who just don't want to use github <@arthurzam> We could also remain in both forges, for example using the CI & artifacts storage of github, but all other real actions through new forge <@mgorny> so the idea was to consider codeberg, which is closer to free software principles <@mgorny> arthurzam: we don't use either of that [21:24] <@mgorny> for the record, i'm only talking of gentoo/gentoo <@arthurzam> oh, ok <@mgorny> other repos can follow later, but let's focus on one <@arthurzam> this changes my stance :) <@mgorny> codeberg is pretty similar to github of old in UI terms <@mgorny> the way i see it, the only real problem is porting all the hacky scripts i've made years ago [21:25] <@dilfridge> now we see the difference between you and ago :D <@arthurzam> mgorny: we could port many of them into using codeberg "pipelines" <@mgorny> yeah, that's my thinking -- improve when possible <@arthurzam> dilfridge: let's not open this can now please [21:26] <@robbat2> do those pipelines require dropping files into the repos? <@arthurzam> robbat2: yes <@mgorny> if codeberg lets us define them without files in the repo <@dilfridge> it would already be nice if they lived in a branch of their own <@mgorny> anyway, what i'd like to cover right now is establishing if we want to try this <@mgorny> i.e. 1) contacting codeberg if they don't mind the extra load [21:27] <@mgorny> 2) starting some experimenting, possibly with volunteers <@robbat2> our repo is very expensive on hosting; I remember the days when GitHub couldn't handle it at all <@robbat2> so +100 on asking codeberg first <@robbat2> as an overall strategy however, I think at least a readonly copy of gentoo/gentoo should remain on github [21:28] <@dilfridge> with disabled PR etc then <@mgorny> you can't disable them -- it's github <@mgorny> people usually deploy bots closing them [21:29] <@arthurzam> mgorny: we could add automatic closure of any PR (a new script) <@arthurzam> like those stale bots, but stale after 30 seconds <@mgorny> "this PR has not received any attention in the last 30 seconds" [21:30] <@ulm> where would such a script live? I hope not in the repo itself? <@dilfridge> squirrel <@arthurzam> ulm: in some random gentoo server running this script using API key <@ulm> k [21:31] what i'm more interested in is eventually trying to archive PRs into some text file we attach to corresponding bugs <@mgorny> we can repurpose the assignment script but that's for another time in terms of the initial approach, it's fine for me <@arthurzam> I agree, let's do it step by step, and the initial 2 steps are logical <@ulm> should we vote on this (like a statement of intent)? [21:32] <@arthurzam> I think we should <@mgorny> sure, why not [21:33] <@dilfridge> ++ <@ulm> mgorny: can you word a motion? <@robbat2> -1 on agreeing to intent at this early stage <@dilfridge> well, vote on contacting codeberg as first step? <@soap> +1 on agreeing to intent at this early stage <@arthurzam> robbat2: https://codeberg.org/Codeberg-e.V./requests/issues/868 <@mgorny> sigh [21:34] <@sam_> (not "sigh", it was mentioned in -dev weeks ago) <@robbat2> my 2c on the bigger picture overall; we need to make it easier to get open contributions and for maintainers to review&accept them <@soap> robbat2: somethign we've been discussing for 10 years <@robbat2> pushing external contributors to codeberg INSTEAD of github doesn't solve that [21:35] <@mgorny> having ai slop give bullshit reviews doesn't solve that either <@arthurzam> Enabling both helps contributors, but makes it harder for reviewers (2 places to search) <@mgorny> github is never a solution given that some devs openly rejected it [21:36] <@ulm> plus bugzie, so it's 3 places <@dilfridge> why not add sourceforge too while we're at it <@mgorny> ulm: and mailing list <@robbat2> evil thought: with the API token can we just delete the Copilot reviews? <@mgorny> and private mail (yes, it happens) <@ulm> mgorny: yes, and private mail :) <@dilfridge> but seriously, one well-documented place? <@arthurzam> robbat2: actually an interesting idea <@arthurzam> until they disable the ability to remove/hide them [21:37] <@mgorny> and more developers leave github <@mgorny> s/leave/refuse to interact with gentoo via/ <@dilfridge> so... what are other large projects doing? <@dilfridge> kde, gnome, libreoffice mgorny: that said, the developers in that state currently also don't handle patches via bugzilla or anything <@mgorny> dilfridge: running their own infra usually <@arthurzam> kde & gnome use their own managed gitlab <@ulm> dilfridge: glibc? [21:38] <@dilfridge> sourceware.org sourceware is looking at hosting a forge using the same software as codeberg, it currently has one as an experiment <@robbat2> libreoffice runs their own gerrit <@dilfridge> no gerrit please [21:39] * +Arsen likes gerrit <@dilfridge> we want to make it easy to contribute, not drown people in git <@mgorny> running more infra doesn't sound like a good idea [21:40] <@robbat2> to try and focus this <@mgorny> we can't handle what we're running now yes, we're already overwhelmed i don't want to start more services at this point <@robbat2> proposed vote: "investigate the possibility of accepting contribution via codeberg" [21:41] <@robbat2> if that is successful; followup vote: "decide on gating the accepted contribution mechanisms to gentoo" <@mgorny> why does "investigate" mean? we're investigating right now? <@dilfridge> wat *** sam^2 (~sam@82.8.138.118) has quit: Changing host *** sam^2 (~sam@gentoo/developer/sam) has joined channel #gentoo-council *** ChanServ (ChanServ@services.libera.chat) has changed mode for #gentoo-council to +o sam^2 <@robbat2> get a response from codeberg; because we haven't yet <@robbat2> see that it will actually work [21:42] <@mgorny> so can't we word is as "contact codeberg", like originally proposed? <@robbat2> arthurzam already contacted them <@robbat2> there is no response yet <@arthurzam> robbat2: not me, ztrawhcse <@robbat2> sorry <@dilfridge> well, maybe it's easier to get a response if the request has an official feel [21:43] <@sam^2> worth a go anyway <@mgorny> i mean contact *officially* <@ulm> i.e. backed by council vote? [21:44] <@mgorny> also, that feels like "feature request" <@sam^2> no, that repo is for quota/limit discussion <@mgorny> while i think we should ask whether they are willing to host us <@sam^2> yes <@sam^2> let's ask <@dilfridge> yes <@robbat2> ask and make it clear it's the got the council backing [21:45] <@arthurzam> yes <@ulm> can someone word a motion please [21:46] <@robbat2> draft: "run experiment to accept contributions to gentoo/gentoo via codeberg, inclusive of codeberg's approval" [21:47] <@dilfridge> why so complicated always <@ulm> "officially contact codeberg about the possibility of hosting the gentoo repository"? <@dilfridge> ... and accepting contributions via codeberg" <@robbat2> hosting gentoo/gentoo [21:48] <@mgorny> "officially contact codeberg about the possibility of hosting repo/gentoo mirror for the purpose of accepting contributions" <@ulm> that's the github name though <@sam^2> ulm: actually, it would be the same on codeberg <@sam^2> we got control of the 'gentoo' org by asking the user who had it nicely (eli handled that) <@arthurzam> ulm: we already have gentoo org, so it will remain <@sam^2> mgorny: that works for me, let's just vote on that please [21:49] <@ulm> mgorny: that sounds good <@dilfridge> ++ <@ulm> please vote on the motion as stated by mgorny * @dilfridge yes * @robbat2 aye * @sam^2 yes * @arthurzam yes * @ulm yes * @soap yes * @mgorny yes <@ulm> unanimous <@ulm> anything else for this item? [21:50] <@dilfridge> any more action items? <@robbat2> somebody to start a discussion in -dev & -project on "decide on gating the accepted contribution mechanisms to gentoo" <@mgorny> i think that's it -- just need to decide who will mail them <@arthurzam> robbat2: I think it might be too early, since we didn't get approval from codeberg. And if they approve, we can then send a ML to -dev [21:51] <@robbat2> i want to start that discussion regardless <@arthurzam> no need to vote for bringing conversation to ML <@robbat2> it's going to be long and drawn out <@ulm> moving on [21:52] <@ulm> 6. Foundation dissolution status <@dilfridge> still there <@ulm> robbat2 has filed the trademark for renewal <@robbat2> yes; trademark renewals all filed; now we have to wait probably 6 weeks to confirm it was accepted <@ulm> but ownership is still with the foundation, for the time being? [21:53] <@robbat2> next SPI financial side <@dilfridge> we're getting impressions now how many donations have moved <@robbat2> i have a response from TBM, their current bookkeeper (he is NOT the treasurer at this time, but he was in the past) <@robbat2> re our latest financials at SPI <@robbat2> he hasn't posted the 2025/04..06 formal statements yet <@robbat2> we only have 2 recurring donors at SPI, for $100/mo each; and the lowest month we only got an additional $100 over that; [21:54] <@robbat2> the highest month however we made nearly $2000 in one-time donations <@mgorny> robbat2: perhaps a better metric is seeing how many people stopped contributing to GF [21:55] <+NeddySeagoon> Is there a list of tasks still to be accomplished? <@robbat2> yes tasks are tracking in Bugzilla <@ulm> bug 936211 https://bugs.gentoo.org/936211 "[Tracker] Gentoo Foundation dissolution"; Gentoo Foundation, Proposals; CONF; ulm:trustees <@robbat2> mgorny: of the recurring donors contacted; only 2 moved :-9 [21:56] <@mgorny> sorry, i meant to ask: how many remained? <+NeddySeagoon> bookmarked. <@ulm> ok [21:57] <@ulm> anything else? <@robbat2> 43 recurring donors remaining on paypal [21:58] <@robbat2> sorry; that was not a stat I had ready to go; so I had to build it fast <@ulm> moving on [21:59] <@ulm> 7. Open bugs with council participation <@robbat2> next immediate steps; since the income is good enough; is that I want to switch over all expenses <@robbat2> then come to council for a funding approval to get the bank signatories together in a branch to close the accounts / transfer balances to SPI <@mgorny> robbat2: do you have an idea if it's contact problem, or just people not having time? <@robbat2> which? the donors not updating? <@mgorny> yes [22:00] <@robbat2> the big donors got two emails so far <@robbat2> the small donors got one <@mgorny> these are email addresses from paypal, so they should be up to date, right? [22:01] <@robbat2> yes <@robbat2> paypal does verification on them <@mgorny> no more questions from me [22:02] <@ulm> moving on <@ulm> 7. Open bugs with council participation <@ulm> bug 936211 we just had https://bugs.gentoo.org/936211 "[Tracker] Gentoo Foundation dissolution"; Gentoo Foundation, Proposals; CONF; ulm:trustees <@ulm> bug 948684 [22:03] https://bugs.gentoo.org/948684 "Missing summary for 20240609 council meeting"; Gentoo Council, unspecified; CONF; ulm:dilfridge <@ulm> bug 948686 https://bugs.gentoo.org/948686 "Missing summary for 20241110 council meeting"; Gentoo Council, unspecified; CONF; ulm:dilfridge <@dilfridge> in your inbox <@arthurzam> both were sent by dilfridge for review <@ulm> good :) <@ulm> bug 959226 https://bugs.gentoo.org/959226 "Vote: Change authority for metadata/AUTHORS from Trustees to Council"; Gentoo Council, unspecified; IN_P; ulm:council <@arthurzam> Council: 7 yes votes (unanimous) [22:04] <@arthurzam> Trustees: 5 yes votes (unanimous) <@ulm> we have voted in the bug, and now it's also in the council log <@ulm> then we have two requests for listing in the AUTHORS file <@ulm> bug 730200 ulm: https://bugs.gentoo.org/730200 "metadata/AUTHORS: inclusion of myself and prior employers that may own copyright"; Gentoo Council, unspecified; CONF; robbat2:council <@robbat2> some of that dates to the CVS era, so it's extra hard to dig out [22:05] <@ulm> I have asked for pointers to copyrightable commits so we'll have a proper paper trail <@ulm> robbat2: git conversion of the repo exists <@ulm> bug 959157 [22:06] ulm: https://bugs.gentoo.org/959157 "Request for listing "Eli Schwartz" in metadata/AUTHORS"; Gentoo Council, unspecified; CONF; eschwartz:council <@ulm> so I'd suggest that we wait for replies in these bug reports <@arthurzam> ulm: let's just ping them again in bug? <@ulm> it's not so long ago that I asked [22:07] <@ulm> but I can ping again before the next meeting <@ulm> 8. Open floor [22:08] <@ulm> anyone? <@arthurzam> let's have another good council year? <@dilfridge> time64 [22:09] <@dilfridge> for x86 all stage3 are now available on the download page <@robbat2> who is our codeberg.org/gentoo admin? <@dilfridge> now it's time to test them in chroots and new installations <@sam^2> robbat2: ztrawhcse: <@sam^2> robbat2: ztrawhcse * <@dilfridge> next step is mgorny's migration progam [22:10] <@sam^2> robbat2: he contacted the user who owned the name 'gentoo' and the user was very happy to hand it over <@robbat2> ztrawhcse: can you please reach out to ensure infra has a safe copy of creds etc <@dilfridge> for arm, the time64 stages are under preparation, armv4,5 are already being built and uploaded, armv6,7 are being bootstrapped and need some minor eclass bugfixes <@mgorny> i think it's "backup admins" there <@dilfridge> then follow the other 32bit arches <@arthurzam> robbat2: can you please join codeberg, so you can also become admin in the org later [22:11] <@robbat2> i'm already on codeberg <@ulm> dilfridge: is there an upgrade path for existing systems? <@ulm> like a news item? or is it too early for this? <@dilfridge> mgorny made one, yes, that's kinda the next step after we know the t64 stages themselves work ok <@dilfridge> it's not trivial, similar to the merge-usr scripts [22:12] <@dilfridge> we can prepare and test the official version now <@robbat2> i have to run; errand and then trying to fix bugzie's jobqueue issue <@ulm> that's why I'm asking <@dilfridge> mgorny: can you summarize what the program is doing? <@mgorny> I don't remember anymore xD [22:13] <@dilfridge> eh never mind then [22:14] <@dilfridge> but we should dig it out agin :D <@mgorny> Something to do with libdirs, right? <@dilfridge> right [22:15] <@arthurzam> can we exit this conversation from council meeting? <@ulm> yes, we should :) <@ulm> any other topic for open floor? <@ulm> seems not to be the case [22:16] <@ulm> meeting closed, thanks everyone <@arthurzam> thank you * @arthurzam falls asleep *** ulm (~ulm@gentoo/developer/ulm) has set the topic for #gentoo-council: "264th meeting: 2025-08-10 19:00 UTC | https://www.timeanddate.com/worldclock/fixedtime.html?iso=20250810T19 | https://wiki.gentoo.org/wiki/Project:Council | https://dev.gentoo.org/~dilfridge/decisions.pdf" <@sam^2> thanks! [22:17] *** sam^2 (~sam@gentoo/developer/sam) has quit: Quit: leaving <@ulm> list of chairs updated on the wiki page <@mgorny> thx <@ulm> I'm going to send a message to council@ asking for your contact info [22:18]