path: root/meeting-logs/20190721.txt

<@ulm> time                                                             [21:00]
<@ulm> !proj council
<+willikins> (council@gentoo.org) dilfridge, gyakovlev, patrick, slyfox, ulm,
             whissi, williamh
* gyakovlev here
* dilfridge here
* Whissi here
<@ulm> anyone else wants the chair? otherwise I'll take it
<@slyfox> go ahead
* xiaomiao here
<@dilfridge> go ahead :)
<@ulm> agenda:
       https://archives.gentoo.org/gentoo-project/message/6106d52ec7a6c75b068cd3c487c26baf
* WilliamH here                                                         [21:01]
<@ulm> 1. roll call
* slyfox here
* Whissi here
* WilliamH here
<@dilfridge> everyone's here!
<@ulm> yep :)
<@ulm> gyakovlev: xiaomiao: welcome :)
<@ulm> 2. Constitute the new council                                    [21:02]
<@ulm> time of meetings, evryone ok with 2nd Sunday of every month at 19:00
       UTC?
* Shentino pays attention
* dilfridge is in favour of the bavarian constitution (lots of freibeer)
<@WilliamH> fwm
<@slyfox> 19:00 UTC ok for me
<@dilfridge> ok for metoo                                               [21:03]
<@Whissi> OK for me.
<@ulm> I don't see any objections
<@gyakovlev> works for me as well, a bit in the middle of the day but fine.
<@xiaomiao> wfm
<@ulm> any objections against continuing last council's workflow?
<@WilliamH> none here                                                   [21:04]
<@ulm> i.e., call for agenda items two weeks in advance, agenda 1 week in
       advance
<@dilfridge> it works
<@dilfridge> so why change it
<@ulm> major discussions on -project ML prior to the meeting
<@slyfox> sounds good
<@Whissi> Not in general but see my mail I wrote today, maybe something for
          open floor.
* WilliamH thinks major discussions applies to us as the council too. ;-)
<@gyakovlev> Whissi sent out an email with some suggestions, consider it
             later?
<@dilfridge> link?                                                      [21:05]
<@Whissi> council@ only
<@dilfridge> ah
<@ulm> Whissi: let's postpone to open floor                             [21:06]
<@dilfridge> Whissi: reading your mail, that's how it's usually supposed to
             be... so, wfm
<@Whissi> ulm: OK, move one
<@ulm> chairmen for this term
<@ulm> any volunteers?                                                  [21:07]
<@Whissi> Maybe you explain charmen first, we have new members ;)
<@slyfox> i can take next two meetings
<@WilliamH> I'll chair some meetings. I'm not really picky when other than I
            don't want nov or dec.                                      [21:08]
<@gyakovlev> I can chair at winter time or spring time, fall/autumn is pretty
             busy and I'd like to learn a bit how you do it.
<@ulm> Whissi: should be obvious?
<@WilliamH> Normally what we do is each of us takes two meetings in a row but
            we all don't have to
<@ulm> August/September: slyfox
<@slyfox> ACK
* WilliamH Jan/Feb I guess                                              [21:09]
<@ulm> k
<@dilfridge> I can do something, but I'm not too eager volunteering... would
             rather restart working on the summary document
<@ulm> gyakovlev: March/April?
<@gyakovlev> wfm
<@dilfridge> can do Nov/Dec                                             [21:10]
<@dilfridge> october is bad for me
<@ulm> I'll do October then
<@Whissi> I'll take May/Jun?                                            [21:11]
<@ulm> xiaomiao hasn't got any
<@dilfridge> there's always one lazy guy... was me last year :)         [21:12]
<@ulm> are you fine with this?
<@xiaomiao> yes
<@ulm> I'll update the table later
<@xiaomiao> my time planning is chaotic enough as it is :)
<@ulm> 3. GLEP 81 approval
<@ulm>
       https://archives.gentoo.org/gentoo-project/message/5d649766eebc4b8550555a66df8c700c
<@ulm> mgorny: are you there?
<@gyakovlev> he mentioned he may not make it.                           [21:13]
<+mgorny> Kinda
<+mgorny> I'm via phone
<@ulm> mgorny: want to say anything before we vote on it?
<@dilfridge> so, I like the whole thing a lot, just have one question / remark
             (which I already asked mgorny on the channel)
<@dilfridge> how can we make sure that "re-enabling a user or group" doesnt
             open any security problems?                                [21:14]
<@dilfridge> like, admin had it disabled by hand, then it gets enabled
<@Whissi> Touching existing users is always a problem. Even disabling... you
          can't know if someone else is using...                        [21:15]
<@dilfridge> "Appropriately, the packages must be able to reenable users when
             they are installed again."
<@dilfridge> ^that's the glep sentence
<+mgorny> We provide explicit override option via local overlay
<@WilliamH> Is there a place a developer can look to see which uids/gids are
            used in case their package requires a specific one?         [21:16]
<@WilliamH> I don't know if that affects the glep or not, but it is a question
            that has run through my mind on this subject.               [21:17]
<+mgorny> We collect them on wiki now
<+mgorny> And i still remember your request about different format      [21:18]
<+mgorny> You can also grep ebuilds
<@gyakovlev> WilliamH:
             https://wiki.gentoo.org/wiki/Project:Quality_Assurance/UID_GID_Assignment
                                                                        [21:19]
<@gyakovlev> mgorny: what about installing users to alternative ROOT, for
             example if crosscompiling, was it adressed? I haven't got a
             chance to use reference implementation yet.                [21:20]
<+mgorny> Works same as user eclass                                     [21:21]
<@ulm> gyakovlev: not different from what ebuilds are doing now
<+mgorny> Ie i dunno
<@ulm> are we ready to vote?                                            [21:22]
<@dilfridge> well, worst case things just dont improve, so...
<@ulm> motion: accept GLEP 81                                           [21:23]
* slyfox yes
* gyakovlev yes
* dilfridge yes
* Whissi yes                                                            [21:24]
* xiaomiao yes
<@ulm> WilliamH: ?
* WilliamH yes
* ulm yes
<@ulm> unanimous
<@ulm> next                                                             [21:25]
<@ulm> 4. Unrestrict gentoo-dev mailing list
<@ulm>
       https://archives.gentoo.org/gentoo-project/message/3883fd8a106a0655f412e7c770dfce4e
* WilliamH yes
<@ulm> mgorny again :)
<@ulm> WilliamH: we don't vote yet
<@WilliamH> heh ;-)
<@ulm> so the motion is to revert the previous council decision and open the
       gentoo-dev mailing list again                                    [21:26]
<@slyfox> \o/
<@ulm> anyone wants to discuss?
<@Whissi> Is there anything to discuss? I don't think so: Let's unrestrict
          mailing list again. If there will be someone *spamming* we will take
          action like one is spamming in IRC, bugzilla...  the decision from
          the past was wrong from my POV so let's fix it.               [21:27]
<@WilliamH> Whissi++
<@ulm> yeah, we could give it a try, and revert againif it doesn't work
<@WilliamH> ulm: we shouldn't revert again, this is a comrel/proctors issue.
<@ulm> than means infra should save the whitelist, for the time being   [21:28]
<@gyakovlev> haha, yeah, just don't delete whitelist repo right away
<@ulm> WilliamH: if it doesn't work at all, then we have the option to
       restrict again
<@dilfridge> let's give it a try, we have the proctors now
<@WilliamH> ulm: from my pov we should have never restricted to begin with.
<@Whissi> ulm: GDPR... :D
<@dilfridge> we know
<@dilfridge> you told us                                                [21:29]
<@ulm> WilliamH: we all know your opinion, I guess :)
<@ulm> ok then
<@slyfox> time to vote?
<@ulm> motion: removing posting restrictions from gentoo-dev mailing list
* slyfox yes
* gyakovlev yes
* Whissi yes
* WilliamH yes
* dilfridge yes
<@ulm> xiaomiao: ?
* xiaomiao yes                                                          [21:30]
* ulm yes
<@ulm> unanimous
<@ulm> 5. Real name requirement
<@dilfridge> sigh
<@ulm>
       https://archives.gentoo.org/gentoo-project/message/85de6190bd27693bed07744e04855911
<@ulm> discussion? anybody?                                             [21:31]
<@WilliamH> I'm all for the real name requirement, it hasn't really affected
            us that much.
<@gyakovlev> WilliamH: it did affect proxy-main project
<@ulm> yeah, but we have a procedure                                    [21:32]
<@Whissi> +1. I don't see a reason to change. When we voted for this, we
          wanted real names. I don't see that a majority is disagreeing and
          changed mind.
<@ulm> proxy maintainer can signoff
<@xiaomiao> I don't see how we can verify names
<+mgorny> it's internal proxy-maint decision not to accept fake names
<@xiaomiao> so to me it looks like an empty gesture that creates lots of
            friction
<+mgorny> as it's been already established, individual developers can decide
          differently but it's their choice                             [21:33]
<@Whissi> xiaomiao: We don't really have to. If we will learn that contributor
          makes fun of us we will probably stop working with him/her just
          because the attitude of this person doesn't match what we expect.
<@xiaomiao> I mean - I have no idea if the entity present here as Whissi is
            using the name their passport(s) have or not, and I don't even
            know how to verify that
<@xiaomiao> so it's all just faith and assuming we can "common sense" our way
            through it                                                  [21:34]
<@WilliamH> xiaomiao: Sure, but you at least have a reasonable idea that he
            his, whissi is more than likely Thomas.
<@WilliamH> I can say that here because he is listed as such on our dev page.
<@slyfox> who knows what is the subtlety in legal vs. preferred name :)
<@ulm> ok, in order to keep this focussed, I suggest than someone come up with
       a motion                                                         [21:35]
<@ulm> otherwise, we move on
<@dilfridge> motion: no changes
<@xiaomiao> WilliamH: I don't even know if it's a single person driving the
            entity
* WilliamH yes
* slyfox abstains
* xiaomiao no                                                           [21:36]
* dilfridge yes
* Whissi yes
* gyakovlev no                                                          [21:37]
* ulm yes
<@ulm> 4 yes 2 no 1 abstention
<@xiaomiao> people have more faith than I do :)
<@ulm> no changes to real name policy
<@Whissi> gyakovlev / xiaomiao: Let me ask you a different way: Just the way
          that we can never 100% be sure that an identity is real, is that
          enough for you to stop requiring real names?
<@Whissi> *fact                                                         [21:38]
<@xiaomiao> Whissi: it's an unenforceable policy, so I don't see why it should
            exist
<@ulm> the same would be true for linux                                 [21:39]
<@WilliamH> ulm++
<@ulm> and all other projects that require signed-off-by
<+mgorny> xiaomiao: we have a lot of those, and yet we don't disband gentoo
          just yet
<@xiaomiao> would anyone even notice a chinese person with the name "spicy
            chicken soup" ?
<@ulm> anyway, let's move on
<@ulm> 6. Proctors policy
<@ulm>
       https://archives.gentoo.org/gentoo-project/message/854d484eca8664e7ee6678bf79d63976
<@gyakovlev> Whissi: I personally don't care if it's real or not. If I see a
             person submitting quality ebuild/code written by them which was
             accepter to other projects under same pseudonym I'm ok
             signing-off this as my name. my position is we still need to
             require names for developers (committers) but authors in some
             cases.
<@Whissi> Thank you two for explanation.                                [21:40]
<@dilfridge> somehow my joke about the spicy chicken soup took on a life of
             its own...
<@ulm> is anybody from proctors present?
<@ulm> !proj proctors
<+willikins> ulm: (proctors@gentoo.org) dolsen, leio, rich0, tamiko, zlogene
<@slyfox> "Lead(s): none" :)                                            [21:41]
<@dilfridge> well they dont
<@WilliamH> If they aren't here, should we even vote on this?           [21:42]
<@ulm> not even sure what the motion would be
<@dilfridge> I'm not fully sure what the whole thing is about           [21:43]
<@dilfridge> but I am *very* reluctant to immediately start dictating policy
             to the proctors
<@ulm> we had delegated authority to them in a CoC update
<@Whissi> I agree with desultory that proctors project must change policy. See
          my bug... I still cannot believe that there was _no_ team decision.
          I.e. proctors is a group of people but two people are enough to
          issue public statements. That's _WRONG_. That's a statement of a
          single person acked by a second one. It isn't more.
<+mgorny> i think he's just flaming for the sake of it
<+mgorny> he kinda does that since i told that forum mods are isolating
          themselving from distro life                                  [21:44]
* WilliamH tends to agree with mgorny on this
<@dilfridge> yeah
<+mgorny> so he now tries to make us wish they did that again
<@Whissi> No. He is not flaming.
<@WilliamH> Whissi: I feel like any posts I've seen from him tend to be pretty
            combative.                                                  [21:45]
<@xiaomiao> what do you expect to happen ...                            [21:46]
<veremitz> they're aggressive, but thats hardly unusual *cough* sorry ..
<@Whissi> WilliamH: Interesting. I read most of them differently. Yes, they
          are hard... but he is dealing with people using the same wording
          style.
<@WilliamH> Whissi: there's some pretty crazy stuff on the forums and from
            what I've seen which is very little they just tend to let it go
            rather than call people out.                                [21:47]
<+leio> Lets try to adhere to the CoC, e.g., not accuse people of flaming for
        the sake it
<+mgorny> Whissi: could you elaborate on why you believe he's not flaming? in
          particular, if *you* have any suggestion on what should happen?
<+mgorny> Whissi: do you consider yourself to be a victim of bad proctor
          action?                                                       [21:48]
<+leio> Regarding team decisions vs 2 people, it's about reaction speed and
        the whole current theory "small slap on the wrist quickly", which
        can't be achieved with a big team decision and waiting for all votes
        and whatnot
<@ulm> I tend to agree
<@slyfox> I read the question as "should proctors have a publushed policy"
<@Whissi> mgorny: Not a bad one but proctors have lost all my credits and I
          can't take the project serious anymore given that two people are
          enough when you hear they didn't even discuss.                [21:49]
* WilliamH tends to agree with ulm
<@ulm> slyfox: if they have policies then they should publish them
<@Whissi> The policy is published.                                      [21:50]
<@slyfox> https://wiki.gentoo.org/wiki/Project:Proctors this one?
<@WilliamH> Whissi: if someone doesn't like a proctors action they go to
            comrel.
<+rich0> leio: intent is to be able to react to flame war threads while
         they're still going, not two weeks later.  IMO that is the worse
         approach - re-starting a debate after the original debate is long
         over
<@WilliamH> Whissi: and if that doesn't satisfy them they come to us.
<@Whissi> WilliamH: No, my problem is that 2 people can speak for proctors
          project. That's my problem. Like said, one person is proposing
          something and a second will just ack. BOOM. Enough. That's wrong
          from my POV. You should have at least a public vote.          [21:51]
<@Whissi> (=require a majority)                                         [21:52]
<+rich0> Whissi: all votes are public, and there is a required delay to
         prevent action.  Also, all actions DO require a majority of those
         voting
<@ulm> Whissi: that would be slow decision making
<@WilliamH> Whissi: that's why comrel is so slow to do anything and things
            have gotten out of hand in our community.
<@Whissi> ulm: No. I don't take that argument. If that's true, proctors
          consists of the wrong people.
<+rich0> In any case, do we really have too many proctors decisions in the
         last year?
<+mgorny> Whissi: so how does the alternative work: a person knows he's going
          to be banned but the actual ban is getting delayed by waiting on
          votes from remaining team members?                            [21:53]
<@dilfridge> the "two people rule" was part of proctors setup from the very
             start                                                      [21:54]
<+rich0> mgorny: my issue with the delay is that it just leads to endless
         flame progression.  People go back and forth on lists because they
         feel like there is no alternative.  IMO proctors might as well exist
         if decisions take more than 24h to be decided from the time an issue
         comes up
<@Whissi> Again, if it will take so much time for proctors, the members are
          the problem. And this should get fixed.
<@Whissi> Don't lower requirements just because you don't get enough people to
          vote in time.
<+rich0> I guess we could require a vote one way or another (abstentions being
         allowed but not counted as no), within some period like 12h, and if
         too many votes are late we boot proctors out, but I suspect we won't
         have many proctors left after long.  I doubt any project in Gentoo
         consistently has all its members take actions within 12-24h.
                                                                        [21:55]
<@WilliamH> Whissi: The same thing is true in QA I think. two people can ask
            for a ban, which can be put in place immediately, but the rest of
            the team or the lead can override within 72 hours.
<@Whissi> It's like the police will require a judge. But no judge is
          available. No problem... just ask another cop... if he/she acks, do
          whatever you wanted to do which normally will require a judge... no.
          That's wrong.                                                 [21:56]
<+mgorny> WilliamH: nope
<+rich0> Whissi: if a violation happens at 5PM EDT, half the proctors won't
         even be awake for 12h.
<+Amynka> WilliamH: qa requires whole team to vote usually
<@dilfridge> Whissi: we designed proctors to be able to act fast this way.
             your ideas go against the entire design.
<@WilliamH> mgorny: Ok, I thought qa was set up that way.
<+mgorny> Whissi: actually, a police can lock you up for short time before
          getting evidence afaik
<+rich0> WilliamH: we already have a policy that actions can be overturned
         after the fact if a new majority emerges
<@WilliamH> rich0: ah ok. in that case, what's the deal Whissi?         [21:57]
<+rich0> as slyfox linked - the process is documented on the page
<+mgorny> Whissi: what you saying sounds like you assume that you need at
          least half of proctors to establish whether discpilinary action is
          necessary
<+rich0> it was intended to be as transparent as possible
<+mgorny> does that imply that individual proctors are unprofessional but as a
          group they suddenly become professional?
<@Whissi> Gentoo will survive if a decision to ban someone will take up to 48
          hours. Srly, about what we are talking here? It's not like there's
          something ongoing requiring immediate action. And in case something
          like this would really happen, like a dev will delete gentoo
          repository... people will apply common sense and take action to stop
          such a behavior IN TIME.                                      [21:58]
<+rich0> fwiw, proctors has yet to take any disciplinary action against
         anybody in the community aside from that spammer a while ago
<+mgorny> i don't really see why do you believe that decision made by 4 people
          is that different from decision made by 2 people
<+mgorny> either the action was justified, in which case it doesn't require
          everyone to sign off it
<+rich0> The intent is to stop the 100 post reply-fests before they get to 100
         posts.  :)  If you wait 48h, then you get a 100 post reply-fest.
         Then you start a new 100 post reply-fest.                      [21:59]
<@ulm> I don't see this discussion going anywhere, and by the workflow we have
       just accepted, it should have taken place on the ML prior to the
       meeting
<@WilliamH> I'm not for dictating proctor policy, especially since the policy
            is already published.
<+mgorny> or it wasn't justified, in which case the people who decided it
          shouldn't be proctors
<@Whissi> ulm: ACK
<@ulm> so unless anyone comes up with a concrete action item, I suggest that
       we move on
<@WilliamH> ulm++
<@slyfox> let's move on
<@WilliamH> let's move on
<+mgorny> and ftr, we have 5 proctors, so we're talking 2 vs 3 people
<@dilfridge> move on
<+rich0> ulm++  btw, we're certainly open to feedback/discussion on the policy
<@ulm> 7. Open bugs with council involvement
<@ulm> bug 637328                                                       [22:00]
<+willikins> ulm: https://bugs.gentoo.org/637328 "GLEP 14 needs to be
             updated"; Documentation, GLEP Changes; IN_P; mgorny:security
<@Whissi> No update.
<@ulm> no news, I suppose?
<@ulm> bug 642072
<+willikins> ulm: https://bugs.gentoo.org/642072 "                                      [Tracker] Copyright policy";
             Gentoo Council, unspecified; IN_P; mgorny:council
<+mgorny> last i heard, security team wants to kill it
<@ulm> that one is just a tracker
<@ulm> bug 662982
<+willikins> ulm: https://bugs.gentoo.org/662982 "                                      [TRACKER] New default
             locations for the Gentoo repository, distfiles, and binary
             packages"; Gentoo Linux, Current packages; CONF;
             zmedico:dev-portage
<@ulm> any news there? catalyst was a blocker IIRC?                     [22:01]
<@Whissi> stage3 are using new /var/db/repos location already
<veremitz> WIP from my observations
<@Whissi> Just installed a new system this week
<@ulm> ok, so some progress there
<@ulm> bug 677824
<+willikins> ulm: https://bugs.gentoo.org/677824 "Deferred decision: Forums
             (specifically OTW)"; Gentoo Council, unspecified; IN_P;
             k_f:council
<@ulm> not sure, what is the status there?                              [22:02]
<+mgorny> some bits of discussion were happening but nothing solid
<+mgorny> i'd suggest closing it until somebody comes with a proper agenda
          item                                                          [22:03]
<@Whissi> Discussion happened... but because there is no clear motion
          formulated (nothing actionable), I would move one.
<@slyfox> yup
<@ulm> RESO NEEDINFO?
<@slyfox> sounds good
<@Whissi> NEEDINFO or CANTFIX :)
<@ulm> k
<@ulm> finally, bug 687938                                              [22:04]
<+willikins> ulm: https://bugs.gentoo.org/687938 "QA lead approval 2019: soap
             edition"; Gentoo Council, unspecified; CONF; mgorny:council
<@ulm> that misses one vote
<@ulm> but was a vote for the previous council
<@ulm> which is no longer in office, so I guess it can be closed
<+mgorny> i think it's WilliamH
<@dilfridge> let's consider it timed out and count the votes
<@Whissi> It's too late. So you have to count the missed vote as absent
<+mgorny> (who didn't vote)
<@ulm> result is clear in any case, 6 yes votes                         [22:05]
<@Whissi> yup
<@ulm> moving on
<@WilliamH> I was the one who missed this I guess, so I can vote, and would
            vote yes.
<@Whissi> You cannot vote anymore for past year. :)
<@ulm> WilliamH: I think you can't, since it's a new council term now   [22:06]
<@ulm> 8. Open floor
<@WilliamH> ok
<@dilfridge> we should calculate the new geometric center of the council!
                                                                        [22:07]
<@dilfridge> :)
* Shentino raises his hand
<@ulm> yes?
<veremitz> geographic*
<veremitz> ;p
<Shentino> I heard that someone on the ml was repeatedly evading bans
<@dilfridge> that too
<@gyakovlev> dilfridge: professors should do that =)
<veremitz> ^ :D
<Shentino> I am curious if consequences for that should escalate beyond just
           repeatedly getting re-banned                                 [22:08]
<+mgorny> Shentino: isn't that trustee business?
<Shentino> particularly if such escalations can avoid collateral damage
           against innocent bystanders
* dilfridge gets some Domina Trocken.
<Shentino> mgorny: I don't know, but if the MLs are being hosted on infra
           managed hardware I think it *should* be trustee business at the
           very least
<Shentino> perhaps the foundation can send a C&D notice to ban evaders on the
           mls
<@WilliamH> Shentino: Yeah, I think that would be a trustee issue (it is
            possibly a legal issue).
<Shentino> I agree                                                      [22:09]
<veremitz> Shentino: perhaps you mean how does that affect white/black-listing
           efforts as prescribed by the council ?
<+mgorny> at least until we take down the foundation and switch to umbrella
* mgorny hides
<Shentino> but the coc and proctors comrel etc who issue the bans answer to
           council
<Shentino> mgorny: perhaps later but for now we work with what we have atm
<@dilfridge> mgorny++
<Shentino> my point is that people who defy bans need stronger handling than
           comrel/proctors atm can provide
<Shentino> at least in my opinion
<veremitz> OT: but shifting the blame to another authority ain't gonna solve
           anything </2cents>                                           [22:10]
<@WilliamH> Shentino: I don't think anyone disagrees with that.
<Shentino> and whitelisting just causes collateral damage against innocent
           bystanders
<+Amynka> Shentino: why?
<@WilliamH> Amynka: that's possibly a CFAA issue at that point.
<+mgorny> oh *beep*, i forgot one thing
<+mgorny> what should we do about people who were explicitly blacklisted (i.e.
          removed from whitelist)?
<@dilfridge> nothing for the moment?
<+mgorny> should they get access back now, or should it be moved to comrel?
<Shentino> mgorny: let me elaborate                                     [22:11]
<veremitz> probably be moved to comrel ..
<Shentino> I'm saying that evading a ban on the ml (or on the forums or irc
           for that matter) should be treated as a separate offense of its own
<@gyakovlev> veremitz: are you blacklisted? =D
<@dilfridge> mgorny: let's just give access back now, and leave the followup
             to comrel/proctors
<Shentino> separate from whatever offense provoked the ban
<+zlogene> mgorny: which ones?
<+willikins> zlogene, you have notes!                                                   [Jul-21 19:21] <dilfridge> beamer 3.55
             seems to have some problems, my presentations fail with it
<veremitz> gyakovlev: no idea...
<Shentino> and I propose that ban evasion get escalated to stronger measures
           that trustees could probably handle
<@WilliamH> Shentino: I don't think anyone disagrees with that.
<+zlogene> dilfridge: oh my                                             [22:12]
<veremitz> Shentino: trustees can't be trusted ...
<veremitz> *cough*
<veremitz> I should go see a doctor :(
<+Amynka> veremitz: i am not sure if they can help you *cough*
<Shentino> william: so what is necessary to actually implement such responses
           to ban evasion?
<veremitz> Amynka: maybe you can ;P
<Shentino> do we need to talk to trustees abou tit?
<@Whissi> Can we move on are we still discussing the ml spammer?
<@WilliamH> Shentino: I would say go to the trustees.                   [22:13]
<@dilfridge> no family feuds during council session!
<Shentino> william: will do
<Shentino> I'll mention that I have council blessing to approach them about it
<Shentino> NOW we can move on i think
<@dilfridge> you have what?
<+Amynka> blessing
<@dilfridge> Amynka: bless you
<veremitz> dilfridge: its a catholic thing :)
<@WilliamH> Shentino: what you have is one member of council suggesting it.
            :-)                                                         [22:14]
<Shentino> that's more than zero
<@WilliamH> Shentino: but yeah let's move on.
<@dilfridge> Shentino: we can think and talk for ourselves, thankyouverymuch
<+Amynka> dilfridge: bless you too
<@Whissi> I wouldn't say that council voted on this. I mean, this isn't our
          business...
<Shentino> sorry dil, no harm meant
<Shentino> I'm just tired of ban evasion
<@ulm> let's move on
<Shentino> and I don't like collateral damage against innocent bystanders,
           that's all
<@ulm> anything else for open floor?
<@Whissi> Yes, I'd like to discuss my mail.
<+Amynka> Recruitment process change
* veremitz just compiling.
<+mgorny> oh u
* mgorny wants to go visit his kittens before it gets too dark          [22:15]
<+Amynka> I believe we should drop quizzes and make it strictly contribution
          based
<@ulm> Whissi: by what you suggest there, you should submit it as an item for
       the next meeting :p
<@WilliamH> Whissi: that's how things are supposed to be actually, so there
            really isn't a lot to discuss, we should just start following it.
<veremitz> Amynka: erm What?!
<@Whissi> It's an internal thing but let's dicuss Amynka first.
<@ulm> Amynka: please hold on, let's discuss Whissi's item first
<veremitz> or is that Amy-sarcarm :P
<@Whissi> Now we have a loop.
* dilfridge checks the Domina Trocken.
<+Amynka> veremitz: deadly serious
<@ulm> *sigh*                                                           [22:16]
<+Amynka> ulm: ok
<@ulm> Whissi: you go first
<+Amynka> sorry
<@Whissi> OK.
<@Whissi> To summarize my mail: People who put something on the agenda must
          add specific motions; Council will only vote on motions published
          with agenda 1w before meeting; Council will NOT formulate *new*
          motions during meeting (like motion "change foo=X to foo=Y" was
          added to agenda, changing to to foo=Z based on discussion during
          meeting is a no-go -- new motions must be on mailing list before)
<+mgorny> Whissi: what about out-of-meeting votes?                      [22:17]
<@WilliamH> Whissi: Like I said above, that is how it is supposed to be done,
            so we should just start following it.
<+mgorny> Whissi: and how do we prevent from things being stalled forever due
          to council members voting 'no' because of last-minute ideas?
<@WilliamH> mgorny: there was really no  criteria for when an out-of-meeting
            vote could happen.
<@ulm> Whissi: IMHO council is free to vote on any motion brought forward
       during the meeting, as long as it's related to the agenda
<@Whissi> mgorny: Example? Things like new QA lead approval don't have to be
          an agenda item I think.
<@dilfridge> "Council will NOT formulate *new* motions during meeting" makes
             it *very* difficult to get anything done.
<@dilfridge> So this is not a good idea.                                [22:18]
<+mgorny> Whissi: i recall at least few of my ideas being approved after
          adding minor change requests on meeting
<@ulm> basically it would mean to delay decisions until next meeting, or to
       have extra meetings
<@dilfridge> at least we need the flexibility to adapt a motion to discussion
             on the list and to discussion during council meeting
<+mgorny> with your proposal, that would mean they would be rejected and i
          would have to wait another month, and hopefully they wouldn't be
          rejected then...
<@WilliamH> I understand what you are talking about mgorny,  I'm not sure what
            the answer is... maybe what you are talking about would work...
                                                                        [22:19]
<@Whissi> dilfridge: But it's not a good idea to say "Must discuss on meeting"
          and someone interested in X believes his/her idea/motion was
          accepted because nobody raised objections just to learn the day
          after the meeting council voted on a different proposal than he/she
          proposed. That's not good either.
<@ulm> Whissi: the other part of it is fine, agenda items should be specific
<@dilfridge> that is precisely why we usually ask the champions to be present
                                                                        [22:20]
<+mgorny> Whissi: the other idea would be to kill meetings entirely, and just
          vote on mailing lists
<@Whissi> And don't just think about the person proposing the first motion.
          Also think about people not sharing their concerns because current
          proposed motion is fine for them. But when we are going to change
          during meeting and the one who proposed it is fine with that other
          can still disagree...
<@WilliamH> We can't kill meetings without a full dev vote so that's a no-go
<@ulm> mgorny: we can't do that because of the slackers rule
<@dilfridge> well, we can do a 5min meeting "everyone here, let's do the rest
             via bugs"                                                  [22:21]
<@dilfridge> that fulfills requirements
<+mgorny> + open floor
<@xiaomiao> I like the idea, makes the meeting just a simple yes/no/abstain
            vote without complex discussion and confusion
<@WilliamH> ulm: the slacker rule isn't the issue, the glep requires us to
            meet once a month.
* Shentino is contacting trustees                                       [22:22]
<+mgorny> in any case, this is something that needs to be discussed on ml
<+mgorny> in depth
<@ulm> anyway, we won't vote on it today
<@Whissi> No. This is council workflow.
<@Whissi> We already have to follow official rule and this is discussion must
          be on ml before
<veremitz> vote to discuss?!
<veremitz> :D
<@Whissi> So this is not really NEW
<@ulm> Whissi: can we move on?
<+mgorny> Whissi: if it prevents people from having their motion accepted,
          people should have a right to express their opinion           [22:23]
<@dilfridge> we need to make sure that bureaucracy doesnt get too much into
             the way of getting things done
<+mgorny> it's silly when you say you want to change X because people don't
          give a chance to give their opinion, and then don't give their
          chance to give their opinion on this change
<@ulm> ok, let's move on                                                [22:24]
<@ulm> Amynka: you have the floor
<+Amynka> as I said I believe quizzes should be entirely dropped from the
          recruitment process since they bring 0 value to gentoo.. and we
          should focus more to contributions of the developers
<+Amynka> of course some mandatory review and talk with the person will be
          placed                                                        [22:25]
<+Amynka> s/developers/contributors
<+mgorny> quizzes still have some educational value; do you have an idea how
          to reuse that?
<+Amynka> mgorny: they dont
<+Amynka> people dont remember anything
<+Amynka> mostly
<@gyakovlev> just fyi, the new merged completed quiz from a recruit was almost
             3000 lines of text.                                        [22:26]
<+Amynka> mgorny: we will of course use questions in the review
<veremitz> that is onerous ..
<veremitz> for both sides.
<+Amynka> but filling up some quizz is just complete waste of time
<+mgorny> i mean, the questions target specific problems, so i think it would
          make sense to convert them into at least 'please read these things'
<+mgorny> Amynka: for that, i agree
<+mgorny> and i don't think most of the people in this room would have done
          them again if they had too                                    [22:27]
<@WilliamH> Amynka: I don't see a problem with it as long as the interview
            process includes questions about things that were in the quizzes.
<@WilliamH> Amynka: like the council etc.
<+Amynka> WilliamH: that of course is mandatory
<+Amynka> some general knowledge have to be verified of course          [22:28]
<@WilliamH> I don't think the council set the recruitment process right?
<+Amynka> No but this is very big change
<+Amynka> and we do not have agreement with zlogene
<@WilliamH> So, technically Amynka, I think your team can change it.    [22:29]
<+Amynka> well I am not in the team anymore
<+mgorny> though i think you ought to discuss that on mls
<+mgorny> council deciding on it is kinda gray area
<+mgorny> since it effectively means council deciding on who can vote for
          council
<+zlogene> Amynka: we have not discussed it seriously
<+Amynka> mgorny: i think the amount of flamewar that would bring is not worth
          it
<@WilliamH> Yeah, council shouldn't decide that, good point mgorny.
<+Amynka> who should then?                                              [22:30]
<@dilfridge> but who should then?
<@ulm> might be a subject for -core
<@Whissi> The person who want to the such a motion pass. :)
<+mgorny> if you want it really formal, probably whole dev vote
<@Whissi> *see
<+mgorny> if you don't want it that formal, internal recruiters decision
<@WilliamH> dilfridge: Hmm, I think we should be able to trust the recruiters.
<@dilfridge> besides, it's not deciding about "xxx will become developer" but
             deciding about general procedure
<@dilfridge> works for me
<+Amynka> which bring us to another point
<+Amynka> you have one recruiter.. what are you going to do about it?   [22:31]
<@gyakovlev> WilliamH: recruiter(s) is a single person now
<@ulm> mgorny: was the current recruiters' policy decided by an all-devs vote?
<@dilfridge> !expn recruiters
<+willikins> dilfridge: recruiters = amynka,zlogene,
<@Whissi> Amynka: I'll probably join the project next month.
<@dilfridge> !proj recruiters
<+willikins> dilfridge: (recruiters@gentoo.org) zlogene
<+Amynka> Whissi: you cannot
<+mgorny> ulm: doubt it
<+Amynka> Whissi: training of recruiter takes 6 months
<+mgorny> it probably grew out of historical procedures
<+Amynka> zlogene: am i right?
<@ulm> mgorny: might even go back to managers times                     [22:32]
<+zlogene> Amynka: 6 months or longer, you are right
<+mgorny> indeed, quizzes are very old
<@dilfridge> ok so just for clarification: Amynka you first left the team and
             now want the council to reorganize team procedures?
<@Whissi> Show me that policy please... but anyway, at some point this process
          has to be started.
<+Amynka> dilfridge: maybe, I can always come back if I see that we are all
          nice people and stuff :D                                      [22:33]
<+zlogene> to clarify, I have never insisted Amynka left the team, I always
           welcome contrubutions from her
<+Amynka> dilfridge: i wanted to show that the current situation is not
          feasible
<+Amynka> if something happens to me or zlogene
<@dilfridge> So.
<+Amynka> what are you going to do?
<@ulm> so who would train new recruiters if zlogene gets hit by a bus?[22:34]
<+Amynka> seems that council doesnt care till its too late :)
<@dilfridge> I think both of you should immediately start training two new
             recruiter candidates.
<+Amynka> exactly
<+Amynka> dilfridge: oh we did.. except they gave up
<+mgorny> didn't last trained recruiters turned out completely useless?
<+Amynka> zlogene: right? :D
<+zlogene> dilfridge: bad news\
<@dilfridge> Yes I heard about that.
<+zlogene> 4 will not fly
<+Amynka> zlogene: 4 out of 4
<@WilliamH> I've wondered about the value of the quizzes also.  I know we have
            proxied maintainers who have been in that project for years
            because they don't want to mess with the quizzes.
<+mgorny> let's disband recruiters and accept new people via proxy-maint
                                                                        [22:35]
<+Amynka> mgorny: not bad idea
<+zlogene> long story short: there are always only 2 active recruiters
* mgorny hides
<Shentino> you're biased mgorny, you're the pmaint head
<@dilfridge> Shentino: shuddup
<+Amynka> i think it might be wort thinking about setting up some process
<+zlogene> in short, gentelmen and lady
<+Amynka> which would not rely on two people
<@WilliamH> dilfridge: :p
<+zlogene> I propose I discuss the changes wuth Amynka first
<+zlogene> we always find a consensus first                             [22:36]
<+mgorny> wanna hear how recuiting in netbsd works?
<@dilfridge> so in the past we had recruiters too, maybe we could ask them for
             advice?
<@dilfridge> jmbsvicetto: ^
<+mgorny> they have some interesting things we could steal
<+Amynka> either way.. its something to consider.. could I have council vote
          on what they think about abadoning the quizzes?
<+Amynka> not decion one
<+Amynka> i am curious about if people agree or not
<@ulm> Amynka: no vote in open floor, sorry
<+Amynka> so no opinions?
<veremitz> show of hands?
<@WilliamH> Well, I'm not sure there's a need for a vote since the council
            didn't institute the quizzes to start with.
<@gyakovlev> mgorny: they get tried of your patches and give you commit bit?
             that's how openbsd works afaik.
<veremitz> non-binding
<+Amynka> fingers? :D
<@Whissi> To be honest, you haven't shown me enough reasons why I would say
          "quizzes are bad" yet.
<@ulm> yes, we could do a straw poll                                    [22:37]
<+Amynka> Whissi: and you want to join recruiters?
* dilfridge polls a straw
<@Whissi> I still believe in quizzes for the moment.
<@Whissi> Amynka: Yes
<+Amynka> Whissi: they are not bad.. they are useless?
<+Amynka> tell me one usefull thing about them
<@WilliamH> Whissi: see my comment above wrt quizzes
<+mgorny> gyakovlev: nah, they actually have people process your request but
          they also have some interesting things                        [22:38]
<@ulm> Amynka: they may be more objective than an interview
<@ulm> like written vs oral exam
<@WilliamH> Whissi: I know off some proxied maintainers who have been so for
            years because they don't want to mess with them.
<+mgorny> for example, before a new dev is accepted, an rfc is sent to ask if
          anyone is opposed to that
<+Amynka> ulm: not true
<+Amynka> i can still reject anyone
<+Amynka> if i have bad mood
<+Amynka> in theory
<+Amynka> even with perfect quizzes
<+Amynka> you got me that power
* veremitz stifles a cough.
<+mgorny> WilliamH: i also know proxied maintainers who were along for long
          and i also know that they can't become devs because they're too
          careless
<@ulm> you shouldn't base decisions on your mood :/
<+Amynka> i agree                                                       [22:39]
<+Amynka> and I dont
<+Amynka> i am just saying nothing is objective at this point
<@WilliamH> mgorny: that's true.
<+Amynka> you have to pick objective people
<+Amynka> the process never will be objective enough
<@Whissi> Amynka: The quizzes will demonstrate knowledge. Because everyone has
          to do them, everyone at one time demonstrated same knowledge. If you
          do it via interview and don't follow same process....
<+Amynka> Whissi: its not true
<@ulm> well, I see it mostly as written vs oral exam, so not fundamentally
       different
<+Amynka> i passed quizzes and my knowledge was almost 0
<+mgorny> Whissi: except people who joined before had easier quizzes
<veremitz> people are prone to forget the knowledge ..                  [22:40]
<Shentino> we also require new devs to be mentored, and to some exten tthe
           mentor is on the hook for the mentee's foulups
<+mgorny> so 'everyone has to do them' is not really exactly true
<veremitz> Shentino: fair point
<Shentino> in essence new devs start out as probationary devs
<+Amynka> Whissi: plus from neuroscience point of view its short term memory..
          which means in 3 weeks you have no clue
<+zlogene> mentors is another pain of gentoo
<@WilliamH> The mentoring is fine, I think we should still do that.
<Shentino> I would like mentors to be let off the hook to some extent cuz
           otherwise they could be gun shy about onboarding a noob
<+zlogene> Amynka and I saw totally careless mentors
<@dilfridge> do we still mentor?
<veremitz> dilfridge: I should hope so!                                 [22:41]
<@WilliamH> dilfridge: yes afaik we do. :)
<Shentino> dill: I would guess so, if it's still a formal part of the
           recruiting process
<+Amynka> dilfridge: kinda
<+mgorny> dilfridge: technically yes but we fail to enforce responsibility on
          mentors
<@WilliamH> zlogene: Amynka: If a mentor is careless, I would think that
            person shouldn't be a mentor again for a while?
<+mgorny> Whissi: that's actually a nice idea                           [22:42]
<+zlogene> WilliamH: I tried to rise that time ago, but have not found support
           with jlec in the past
<+Amynka> WilliamH: agreed but if the person has contributions and knowledge..
          and these contributions can be reviewed by various people mentor is
          not such a critical point
<+Amynka> meaning
<+Amynka> if people are trained by the work
<+Amynka> they dont need mentors thatm uch
<@WilliamH> How do other distros bring in new devs?                     [22:43]
<veremitz> zlogene: is jlec still aruond even?
<+zlogene> gyakovlev: gonna pivk up yours tommorrow once home ;)
<+Amynka> WilliamH: i think its mostly contribution based too
<+Amynka> like opensuse
<@Whissi> Well, how many people start becoming a dev without contributing
          first? You say that in a way like they start becoming dev out of the
          nowhere...
<+zlogene> veremitz: no, this was like five years ago
<veremitz> zlogene: ah!
<+zlogene> the main problem with quizzzes is that really skilled people write
           them reluctantly                                             [22:44]
<+zlogene> and bad skilled people try to cite the devmanual
<@Whissi> Someone who already contribute should be able to do quizzes without
          many problems... is that not what you are experiencing?
<+zlogene> literally
<@WilliamH> This is definitely not something we are going to decide today, but
            I"m not opposed to changing the recruitment process.
<+mgorny> quizzes worked for me when i was a student
<+mgorny> today i wouldn't find time for them                           [22:45]
<@WilliamH> We do not control the process specifically at the council level.
<+mgorny> today i don't find time for such prolonged meetings!
<+zlogene> WilliamH: well, and honestly council is not about to control
           recruiters, so everything is right :)
<+mgorny> Amynka: i'd say if you need council to do something, go with a
          motion                                                        [22:46]
<@Whissi> ulm: Let's move on.
<+mgorny> if you don't, just do your thing
<+zlogene> only in case of appeals
<+mgorny> if recruiters break gentoo, council will complain
<@ulm> Amynka: zlogene: can you discuss it, and maybe bring it up on mailing
       lists?
<+mgorny> if things continue working, i don't think there will be a reason for
          council to complain
<+zlogene> ulm: we will I hope
<@WilliamH> mgorny++
<@ulm> any other item for open floor?                                   [22:47]
<@dilfridge> oh somebody always complains
<@WilliamH> zlogene: Amynka: recruitment is your process. :-)
* mgorny notes he had kinda adopted 3 stray kittens
<@dilfridge> and get your shit together please.
<+mgorny> that for open floor summary
<@ulm> I don't see anything else
<@dilfridge> no                                                         [22:48]
<@ulm> next meeting will be on 2019-08-11
<+zlogene> dilfridge: what a rude sentense, but ok :p
<@ulm> slyfox: you'll have the chair
<@slyfox> *nod*
<@ulm> meeting closed
<@slyfox> \o/ thanks all!
* ulm bangs the virtual gavel
<+mgorny> thanks and good night
<@Whissi> Thanks for chairing.
*** ulm (~ulm@gentoo/developer/ulm) has set the topic for #gentoo-council:
    "191st meeting: 2019-08-11 19:00 UTC |
    https://www.timeanddate.com/worldclock/fixedtime.html?iso=20190811T19 |
    https://wiki.gentoo.org/wiki/Project:Council |
    https://dev.gentoo.org/~dilfridge/decisions.html"