*** ulm (~ulm@gentoo/developer/ulm) has changed mode for #gentoo-council to +m let's start agenda is here: https://archives.gentoo.org/gentoo-dev-announce/message/5661c77c0017ec48fa97eb62eb7db316 [19:01] roll call * tamiko here !proj council (council@gentoo.org) dilfridge, k_f, mgorny, slyfox, tamiko, ulm, williamh * K_F here o/ * WilliamH here * mgorny here dilfridge should be back momentarily. [19:02] * dilfridge present slyfox is missing too !seen slyfox [19:03] tamiko: slyfox was last seen 2 hours, 4 minutes and 17 seconds ago, saying "right" in #gentoo-toolchain Someone has a phone number? I don't [19:05] I think we should start [19:06] Status of nios2 and riscv https://archives.gentoo.org/gentoo-project/message/a5127b9dfdd5e6537f8293d204befe12 https://bugs.gentoo.org/644754 Is there anything new there? yes and no [19:07] only that the riscv glibc maintainer supposedly turned up at the fosdem booth and asked about becoming a gentoo dev think the only thing new is nothing has happend since last time, except for glibc interest in riscv for a prospective developer.. but I'd say removing it doesn't change that as it opens up clean slate for anyone actually wanting to implement riscv there was no reply from vapier or anyone interested in working on this I'd say remove them for now and doing it right, with team etc at that point [19:08] can be readded quickly if necessary but I wasnt there at the time, and nothing else happened in the meantime they're merely copies of some other old profile arrgh there's something wrong with my internet connection so if someone wanted to do it, he can copy more recent version of that profile ulm: ++ mgorny: can you come forward with a motion to vote on? [19:09] sorry I might drop out suddenly [19:10] motion: 'remove nios2 & riscv from profiles/, providing a clean slate for anyone wishing to readd it'? my internet connection is going on/off all the time, with any tcp connection breaking down within a minute or so [19:11] there are no keywords to be removed, so i guess 'profiles/' covers it all mgorny: "if anyone wishes to readd it" ? * slyfox is back anyway \o/ removing works for me wfm K_F: sure, maybe with 'properly' I prefer without properly actually too much of a juddgement on current state, which isn't necessary in this context [19:12] anything that removes them works for me ;-) mgorny: how about "remove nios2 and riscv from profiles/ due to inactivity and missing keywords in the tree, providing a clean slate if anyone wishes to readd them" * dilfridge wonders if running a wlan router with firmware date (Jul 31 2008 22:32:30) is a good idea ulm: wfm dilfridge: very safe and stable ulm: not sure if it doesn't remove focus from the fact that it has no arch team to contact [19:13] the only place I see riscv mentioned except in profiles is actually eclass/toolchain-funcs.eclass right Let's just vote on removing the current profile stuff. the problem is primarily "no team" yeah, without extra rationale in vote not sure if we want to mention that or restrict removing the profile (maybe without the dir reflection but generic profile mention?) rationale can be added to summary/log yep, rationale will be in the summary [19:14] K_F: the dir was supposed to catch arch.list which is not strictly part of profile so please vote: "remove nios2 and riscv from profiles" * tamiko yes * dilfridge yes * K_F yes * slyfox abstains * mgorny yes * ulm yes * WilliamH yes [19:15] anything else on this topic? not as far as I can see next then, Mailing list posting restrictions i don't think so actually, same e-mail from mgorny [19:16] cancel the ml restrictions. what is the status there? I stand by the previous vote at least, going back and forth sets a bad precedence nothing has happened since the last meeting K_F: +1 only a few people requested +v on list I voted against them originally and think they send a bad message. [19:17] and there were no new major incidents on -dev mgorny: for now, yes decision on restriction can't be based on the mood of the day and actual behavior that may change at any moment, though We saw the reaction against the restrictions when they were announced. * mgorny doesn't have a strong opinion either way [19:18] The decision was based exactly on that. in the past, it tended to come in bursts, with long quiet periods inbetween that way we'll always be reactive , and things likely have calmed again.. I believe the whitelisting is a good balance between openness for non-devs and the expectations listed by devs in particular given it is only for the -dev list prometheanfire was working on moderation support that was also discussed during last meeting, so nothing new there [19:19] imo this is more under comrel and the council shouldn't be doing things that affect the entire community. but there's not much progress since personally I don't believe in moderation K_F: only the old ways for you? :P except that there's no real agreement on how would moderation work WilliamH: defining scope of mailing list is council territory prometheanfire: if old ways is responsible behavior by controlling who can post instead of per-message cencorship, yes [19:20] that is just a work burden, and an oasis of conflict the new email system would allow for moderation, emails would have to be coppied by the mail server and sent to the old backup system as mailman3's backups are currently not compatible with our archiving * dilfridge reboots router and modem anyone wants to come up with a motion? [19:21] in any case, I don't see any pertinent new information since last vote, so no reason to do another vote on the matter [19:22] Let's postpone this discussion for now. postponing won't help the problem is that users are misinformed that they can't post when they can I won't impliment a system that won't be used * WilliamH motions that we cancel the restrictions so i think we should do another confirmation yes/no vote We decided that the mailing list will be developer-posting only with whitelist. I have no problems with implementing the posting restriction now if 'yes', then we enforce the restrictions and explicitly request infra to enable them Why haven't we put at least the first part of it into action? without any new information, a re-vote sets bad precedence if 'no', then we revoke it what has changed since the last vote? should we expect re-votes on other issues? K_F: my vote ;-P decided is decided.. mgorny: confirmation of which vote? there were several in the december meeting K_F: decided wrongly can be fixed with a new vote [19:23] I agree with K_F - we have discussed the matter and decided how to move forward. i have cooled down since then mostly, what didnt happen was that someone took care of it Let's not revisit it faster than the German social democrats change their mind. mgorny: so you are suggesting that your vote last time was a knee-jerk reaction? tamiko: right, we may look as inconstant as the man with the beard :) [19:24] i'm saying that my vote was based on the assumption that comrel can't solve problems I motion that we cancel the mailing list restrictions. "restrictions" :) [19:25] and that my vote was based on the general goal of setting equal rights on both lists we had 3 accepted votes wich eventually didn't pass 3 passed ones plus 2 that didn't pass I suggest that we postpone this [19:26] as mgorny said that won't help [19:27] no ulm: if we want to postpone this, we should give a clear message to users we're not the spd I don't like postpoining, I propose we drop it, if new information comes up we can bring it up again mgorny: we have a decision but as far as I can see no new information has showed up since last vote, so it stands we should actually stand by our decision +1 Even if it is a bad one? then you need to implement it :) [19:28] WilliamH: you disagreed in the vote, it was outnumbered, so yes, even if you believe it is and while things have been quiet, the discussion will just restart when the next troll comes up WilliamH: even if we vote again, i expect 5:2, so no difference this is a very light form of moderation, given users are permitted to post with watching Could it not be argued that closing the list violates the social contract? [19:29] "the council confirms its decisions taken in the 20171210 meeting about mailing list restrictions"? K_F: I thought you didn't believe in moderation which doesn't introduce significant workload, or direct censorship of content WilliamH: there are several open lists This is default deny, with whitelisting prometheanfire: per-message content moderation I think a default accept with blacklisting would be better its not prometheanfire: ++ so let's just have someone declare he'll take care of it and push our users out of confusing state we don't have to go with per-message moderation ulm: wfm [19:30] so yeah, before we vote, any volunteer for implementing the restrictions? it needs to be implemented by infra if infra refuses to implement it we have another data point and another discussion altogether [19:31] prometheanfire: isn't blacklisting what comrel does now? (sorry, i seem to have some 10s lag) blacklisting is what we have effectively now mgorny: and that won't change, that is always an option right, anyone contacting infra then? mgorny: ya, I don't see what that's insufficient for our needs I'm with prometheanfire ulm: I believe we have a bug already, if not I can liaise with infra K_F: noted I don't see any new facts, so not sure if discussing this further makes much sense [19:32] please vote: "the council confirms its decisions taken in the 20171210 meeting about mailing list restrictions" * WilliamH no [19:33] * ulm yes * K_F yes * mgorny abstains mgorny: why abastain? why abstain? * tamiko yes WilliamH: because i neither want to implement this, nor i want to overrule council decisions [19:34] dilfridge, slyfox: *ping* * slyfox abstains iow, leave it for people smarter than me to decide dilfridge: still here? * dilfridge yes [19:35] 4 yes, 1 no, 2 abstentions yes to the vote or yes to still here? so the decision stands prometheanfire: that is why we use different signal channel for votes (/me) K_F: ah, k prometheanfire: as opposed to aye /nay prometheanfire: /me yes is a vote always can we move on? [19:36] prometheanfire: or ,not opposed, but rather, as an alternative to not using words from regular talk for votes ulm: by all means open bugs bug 635344 ulm: https://bugs.gentoo.org/635344 "[TRACKER] manifest-hashes replacement"; Gentoo Council, unspecified; CONF; mgorny:council [19:37] any news? ulm: i think you know more than we do ;-) [19:38] ulm: I don't see any needs for changes as thing stand does this need to be assigned to the council? or can we reassign to e.g. qa? I believe we already discussed whether it should be re-assigend in december meeting, and already then concluded it can be reassigned [19:39] the only thing we discussed bringing up again previously was going to only 1 hash, but I don't see that as pertinent for that bug nor something we need to decide on atm we can close it and keepthings as they are now to be honest, i think we could close that bug k [19:40] it's not really something needs to be tracked anymore by the council wfm " 99 This is well underway and can be closed or reassigned. is from december summary bug 637328 ulm: https://bugs.gentoo.org/637328 "GLEP 14 needs to be updated"; Documentation, GLEP Changes; CONF; mgorny:security right, we're working on that, security lead has requested to step down, so it will likely be after a new security lead is elected we have a meeting discusssing things including the GLEP next weekend [19:41] k bug 642072 ulm: https://bugs.gentoo.org/642072 "Joint venture to deal with copyright issues"; Gentoo Council, unspecified; CONF; mgorny:council this is in progress too [19:42] there were several updates to the policy draft more at the joint meeting, I guess [19:43] For the record, I will be appealing the ml restrictions to the trustees asking them to determin whether this violates our social contract. determine thanks! WilliamH: I am at a loss what you try to achieve here? tamiko: if it does violate the social contract I think they could probably revoke it. [19:44] WilliamH: Using dedicated communication channels for development (like #gentoo-dev or gentoo-dev@) is exactly orthogonal to the quesiton of a social contract. WilliamH: it is out of scope for trustees, social contract isn't impacted by this decision yeah, also it's all in the open because it can be read publicly If that's the case, I'll definitely respect it. [19:45] the social contract only says that we shall be open and the mailing list is still public [19:46] doesn't say what open means, I think it means read-only WilliamH: it's not open floor yet, can we end the bugs topic first? prometheanfire: ok, if that's the case I guess that makes sense. ulm++ bug 644754 ulm: https://bugs.gentoo.org/644754 "nios2 & riscv: arches added without a backing arch team"; Gentoo Linux, Profiles; CONF; mgorny:vapier WilliamH: Further, I would like to avoid any further incident where we put the question of responsibility at test. Council and Trustees should work as a team, not trying to play power games. that bug is covered by previous vote that one we had already, in fact besides, WilliamH, I do not like the attitude "our majority decision was bad, so I try to find someone else to override it" [19:47] *** ulm (~ulm@gentoo/developer/ulm) has changed mode for #gentoo-council to -m open floor ulm: i'm ready to pull the plug on them as soon as the meeting ends ulm: Thanks for your patience :-) np :) I feel like a bad decision was made. Mgorny was pushing it I think, and he has implied that he changed his feeling about it now. [19:48] btw the one new information that might be relevant is that somebody mentioned that freebsd is working on a closed ml working as in actually using open floor item, thank you mgorny for OpenPGP work for gentoo repository K_F: np +1 ya, that's been a long time coming :D [19:49] mgorny: thats not new info, it was already part of available information at time of vote K_F: which reminds me you were planning to audit the new changes (to glep) mgorny: now speed up gemato by a factor of 10 :p I will be quiet in the meetings about it I guess now. mgorny: right, I'll see if energy level spikes a bit next week, this week got hit by a flu, badly ulm: it's rather fast. disks are slow mgorny: but you're referring to lifting single filesystem restriction (to have that on log since we're discussing it) [19:50] K_F: yes ulm: in fact, it's much faster than i expected it to be. and my disk isn't fast there's also expected gain when portage starts to verify on-the-fly instead of checking whole repo mgorny: I like checking whole repo consistency, actually [19:51] yeah, that would be a big step but the problem with doing that is that portage has no real vfs impl, and just accesses files directly in a lot of places *** [Arfrever] (~Arfrever@apache/committer/Arfrever) has quit: Ping timeout: 240 seconds in reguards to council and trustees working together... [19:52] you realize the optics of a council member calling for a trustee to be removed right? so we would have to actually secure every single place when that happens removing PORTDIR and ECLASSDIR is also a good step towards making this actually secure (no-the-fly can't help much if ebuild can refer to any other file that wasn't tested) it's something that the council should be made aware of prometheanfire: i'm going to withdraw this, i wasn't aware that you've actually pushed the matter forward [19:53] prometheanfire: its done as a single member, not on behalf of council, so if anyone objects to it, it will be a matter for next election prometheanfire: I don't think that mgorny has acted with his council hat there however, i should point out 'the optics' of trustee insulting council member with his trustee hat on, and then trustees requesting to work together with council [19:54] Who was calling for a trustee removal? that's all fine, just wanted to bring it up mgorny: ya,it's just bad all around really :| looks pretty hypocritical to me ok, before this derails completely, any other topic for open floor? [19:55] its not something we need to spend time on in council meeting, in any case, please discuss this in other channel I want to say thanks to all devs helping making FOSDEM stand great this year I think *if* we discuss this we should probably also discuss zlg's mail K_F: noted [19:56] and +1 of course making Gentoo visible at various conferences is important for perception and recruiting even if I wasn't there we should encourage that also outside of europe (alicef is doing a good job in japan etc) iirc we are sending people to scale ulm: https://blog.sumptuouscapital.com/2018/02/gentoo-at-fosdem-2018/ / https://www.facebook.com/media/set/?set=a.10156877297098812.1073741830.8059228811&type=3 [19:57] prometheanfire: do we have a stand there? yes prometheanfire: nice :) we generally do every year iirc so who's going? I think nerdboy and someone else, don't remember off the top of my head [19:58] we can follow up on that outside of meeting, but someone sould take responsibility and do a quick writeup to document it I think dberkholz is leading it also if they want promotional material we have the source for flyers etc prometheanfire: afaik he is not a current gentoo dev :) [19:59] anything else? [20:00] * ulm bangs the gavel meeting closed thanks all! ulm: thanks for chairing *** ulm (~ulm@gentoo/developer/ulm) has set the topic for #gentoo-council: "Next meeting: 2018-03-11 18:00 UTC | https://www.timeanddate.com/worldclock/fixedtime.html?iso=20180311T18 | https://wiki.gentoo.org/wiki/Project:Council | https://dev.gentoo.org/~dilfridge/decisions.html"