path: root/meeting-logs/20130212.txt

20:49 -!- grobian changed the topic of #gentoo-council to: Meeting today (2013-01-08 20:00 UTC) | http://dev.gentoo.org/~grobian/agenda-20130212.txt | http://www.gentoo.org/proj/en/council/utctolocal.html?time=2000 | http://www.gentoo.org/proj/en/council/
20:57 <+dberkholz> good timing, grobian. was just looking for that to make sure i didn't miss something
20:57 <@  grobian> :)
20:58 <+dberkholz> besides you, at fosdem..
20:58 <@  grobian> oh, I was in the back, and at the side
20:58 <@  grobian> when you were still with your head with website redesign ;)
20:59 *** WilliamH is here
20:59 <+dberkholz> i mean during the important part.
20:59 <+dberkholz> beer.
20:59 <@  grobian> I was there :)
21:00 <@  grobian> mostly outside, though
21:01 <@  grobian> it's 20:00 UTC, isn't it?
21:01 <+dberkholz> yep
21:01 <@  grobian> where's that chairman of ours
21:02 <@      ulm> who's going to chair? chainsaw?
21:02 <@  grobian> at least I thought so
21:02 <@  grobian> don't mind doing it in case he is MIA
21:02 <@  grobian> but it seemed he had a plan
21:02 <@  grobian> so I'd prefer waiting a bit for him
21:03 <+dberkholz> someone got his # handy? i would check but i'm mobile atm
21:03 < scarabeus> hello
21:04 <@  grobian> w00t, even scarabeus is on time!
21:04 <@  grobian> :P
21:04 -!- TomJBE [~tb@gentoo/developer/tomjbe] has joined #gentoo-council
21:04 < scarabeus> hey, i usually do not forget
21:04 -!- mode/#gentoo-council [+v scarabeus] by grobian
21:04 <+scarabeus> i only once fuckup the conversion utf->local
21:04 <+scarabeus> :-)
21:04 <@  grobian> yeah
21:04 <@  grobian> you should try utc instead
21:04 <+scarabeus> s/utf/utc/
21:04 *** scarabeus sobs
21:04 <@  grobian> :)
21:05 -!- mode/#gentoo-council [+v WilliamH] by grobian
21:05 <@      ulm> grobian: I've sent a text message to chainsaw
21:05 <@  grobian> Betelgeuse: ping
21:05 <@  grobian> ulm: thanks
21:05 -!- Chainsaw [~chainsaw@gentoo/developer/chainsaw] has joined #gentoo-council
21:06 -!- mode/#gentoo-council [+o Chainsaw] by ChanServ
21:06 <@ Chainsaw> Well that's a great start. Thank you ulm, and I do apologise for that poor show.
21:06 <@      ulm> hi :)
21:06 <@  grobian> cool
21:06 *** grobian hands over keys to Chainsaw again
21:06 <@ Chainsaw> Thank you grobian. Have you done the roll call already?
21:07 <+dberkholz> missing Betelgeuse
21:07 <@  grobian> Chainsaw: nope
21:07 <@  grobian> ^^^ what dberkholz said
21:07 <+dberkholz> everyone else has spoken in the past 6 min
21:07 <@ Chainsaw> ulm: Could you do Betelgeuse the same service?
21:07 <@  grobian> WilliamH: still present?
21:07 <+dberkholz> actually WilliamH was 2 min early but i presume he didn't disappear.
21:08 *** WilliamH is still here
21:08 <@      ulm> Chainsaw: will do
21:08 <@  grobian> Chainsaw: I've put up http://dev.gentoo.org/~grobian/agenda-20130212.txt for online editing
21:09 <@Betelgeus> hello
21:09 <@  grobian> complete
21:09 <@Betelgeus> all thanks to ulm
21:09 <@ Chainsaw> Excellent. Now that we're all here (thanks again ulm), we shall begin.
21:09 <@      ulm> Betelgeuse: yw
21:09 <@ Chainsaw> Do we have any news from jmbsvicetto on bug #383467 please?
21:09 <@ Chainsaw> I gave a courtesy ping about a week ago.
21:10 <@  grobian> ask ulm ;)
21:10 <@      ulm> should be all done by now
21:10 <@      ulm> jmbsvicetto has uploaded all files
21:10 <@ Chainsaw> ulm: Excellent. Could you officially close that bug please?
21:10 <@      ulm> and I've just fixed the last missing link 10 minutes ago
21:11 <@ Chainsaw> grobian: You have the floor for your two items.
21:11 <@  grobian> ok, these are FOSDEM discussions of me, Betelgeuse and graaff
21:11 <@  grobian> mainly
21:11 <@      ulm> bug 343467 closed. finally.
21:11 <+scarabeus> yay
21:11 <@      ulm> 383467 that is
21:11 <@  grobian> first, we think that with the lookout for git, we shoudl start requiring devs to commit with gpg
21:12 <@  grobian> as in, get their systems setup so they know how to get it working
21:12 <@  grobian> as with git it will be mandatory
21:12 <@  grobian> and getting it right isn't really trivial, if you don't know where to begin
21:12 <+dberkholz> so we might want to set up a commit hook with a warning now, then a cutover date where it's mandatory
21:12 <@  grobian> so getting everyone on siginig their commits should be a good preparation for the git migration
21:12 <@      ulm> grobian: "commit with gpg" means FEATURES=sign?
21:12 <@  grobian> ulm: yes
21:13 <@  grobian> so, any questions on that?
21:13 <+scarabeus> nope i agree with the requirement
21:13 <@  grobian> like dberkholz says, we probably have to go through infra at some point
21:13 <@      ulm> it's long overdue that we require this
21:13 <@  grobian> we can start with officially poking/encouraging people to do it
21:14 <@ Chainsaw> I already sign my commits, yes.
21:14 *** WilliamH agrees, but we can't really force it without git.
21:14 <@  grobian> ulm: it's a bit pointless to be honest
21:14 <+ WilliamH> I sign mine also.
21:14 <@ Chainsaw> The main thing you should do is not try and overregulate.
21:14 <@  grobian> WilliamH: exactly, but those people will be in a shitty position if they don't invest now
21:14 <@  grobian> so, let's just try to push people
21:14 <@  grobian> not forcing anything
21:14 <@ Chainsaw> The last time mandatory signing was suggested, there was key length between X and Y, must be a subkey, blah this, blah that.
21:14 <@  grobian> I don't like that
21:15 <@  grobian> Chainsaw: indeed, and we should sort that out, help that discussion, but at least get people to have the setup ready
21:15 <@  grobian> I don't care how many bits people sign with
21:15 <@ Chainsaw> grobian: Main thing I'm missing is a discussion on -dev.
21:15 <@  grobian> or what key
21:15 <@  grobian> Chainsaw: yes, but this isn't a vote, so it's an action point for us (council)
21:15 <+dberkholz> i don't think anyone will really care how complex the requirements are as long as there's a couple of simple one-liners to check your existing key and make a new one that fits them.
21:16 <@ Chainsaw> grobian: But from the floor, it's looking like a "keep it simple, yes, put it on the next agenda".
21:16 <@ Chainsaw> I for one use my main key.
21:16 <@  grobian> Chainsaw: nono, not for the next agenda
21:16 <@  grobian> this can be done now
21:16 <+ WilliamH> I use my main key also.
21:16 <@  grobian> I think dberkholz is completely right
21:16 <@  grobian> we just have to create awareness
21:17 <@  grobian> is there someone who wants to take the lead on this process?
21:17 *** WilliamH would rather not have to make a new key unless it is absolutely necessary...
21:17 <+dberkholz> so why doesn't someone interested write up a -dev-announce post that quickly and easily summarizes what and why, and send it out
21:17 <@  grobian> dberkholz: you just volunteered?
21:17 <+dberkholz> i didn't put the topic on the agenda.
21:18 <@ Chainsaw> WilliamH: Agreed. By just saying "all devs, please sign your commits. no complex requirements, no draconic rules. Pretty please"
21:18 <@  grobian> hah
21:19 <@      ulm> Chainsaw: would be a good point for people to reconsider their key length, though
21:19 <@      ulm> but still, I wouldn't enforce anything at this point
21:19 <@ Chainsaw> ulm: I have had this key since 1/1/2004. I am rather attached to it.
21:19 <@  grobian> ulm: agreed
21:19 *** WilliamH agrees with Chainsaw on this one... I don't want to be forced to make a new key.
21:20 <@ Chainsaw> The less red tape you put up, the more likely people are to agree with it.
21:20 <+dberkholz> i'm fine with incremental advances in security
21:20 <@  grobian> ok
21:20 <@  grobian> done with this topic then?
21:21 <@ Chainsaw> WilliamH: Would you like to draft & send it?
21:21 <@ Chainsaw> WilliamH: To make sure it stays simple and nobody sneaks anything complicated in?
21:22 <@ Chainsaw> grobian: Just about, trying to volunteer someone.
21:22 <+ WilliamH> Chainsaw: Ok, I can do that... Does repoman protest if you try to commit something that isn't signed?
21:22 <@ Chainsaw> WilliamH: Not yet. We need to get the rating of signed commits up before that's feasible.
21:22 <@  grobian> yup, no changes on repoman/cvs side
21:22 <@ Chainsaw> WilliamH: First we ask politely, then we ask, then we plead and then we make binding rules.
21:23 <+ WilliamH> Chainsaw: ok, so something like, "from this point forward, please sign your commits." Is there a guide somewhere that tells how to set up to do that?
21:23 <@ Chainsaw> WilliamH: Yes, I think there's a document on gentoo.org that you can link to.
21:24 <@ Chainsaw> grobian: Now we can move on. The task has been assigned.
21:24 <@  grobian> ok, cool
21:24 <@  grobian> second point
21:24 <@  grobian> we figured that glep 39 doesn't really say council members need to be gentoo devs
21:24 <@  grobian> we might want to have that
21:24 <@  grobian> (at least the foundation will)
21:25 <@ Chainsaw> If a non-dev is vocal and has the popular vote, why should we put barriers in their path?
21:25 <@  grobian> but we can't vote, so we have to suggest changes and organise a vote 
21:25 <@  grobian> well, I disagree on that
21:25 <@ Chainsaw> Go on...
21:25 <+scarabeus> it should not be a problem
21:26 <@  grobian> anyway
21:26 <+scarabeus> if they add themselves to the list it is fine
21:26 <+scarabeus> they still have to get majority of voters to vote for them
21:26 <+scarabeus> and if devs pick somebody out of the project we have way bigger problem
21:26 <@      ulm> we had the issue once with a non-dev as proxy
21:26 <@ Chainsaw> If a majority of the developers votes in ciaranm because he has good points and voices them politely...
21:26 <+scarabeus> ulm: hm, refresh me about that?
21:26 <@  grobian> fair enough
21:26 <@ Chainsaw> Then he gets to be on the council. I don't see this as a problem.
21:27 <+dberkholz> Chainsaw: we can absolutely warn before the rate of signing goes up. can't error, sure.
21:27 <+dberkholz> re repoman.
21:27 <@  grobian> I see a problem in terms of foundation
21:27 <@ Chainsaw> Essentially... if you have lost trust in the majority vote of the developer community... why try to restrict what that untrusted community can vote in?
21:27 <@ Chainsaw> grobian: I'm listening.
21:28 <@  grobian> foundation is legally responsible for us ;)
21:28 <@  grobian> no fun to have random $JOE in there
21:28 <@  grobian> regardless how good they are
21:28 <+ WilliamH> grobian: I don't think the foundation requires its members to be devs does it?
21:28 <@Betelgeus> WilliamH: it doesn't
21:29 <@Betelgeus> And neither are devs required to join the Foundation
21:29 <+dberkholz> it's really implied in the glep that the council members are developers
21:29 <@      ulm> scarabeus: I cannot find it atm, I think it was in 2009
21:29 <+dberkholz> through the statement that they represent all developers
21:29 <+dberkholz> so this is more about fixing a bug
21:30 <+ WilliamH> I would tend to think that the council members should be devs because of the council's role.
21:30 <@Betelgeus> ulm, scarabeus: yeah there was a long discussion and I think the council then was of the opinion that proxies should be devs
21:30 <+scarabeus> actually proxies should not be non-devs as only the council member can pick that one
21:30 <+scarabeus> but electees should be anyone
21:31 <+scarabeus> otoh council member should be able to pick anyone in his place, and as he is already in council he should show up some brain to pick someone good :-)
21:32 <@      ulm> scarabeus: 2009-07 meeting, dev-zero had appointed ciaranm as proxy
21:32 <@      ulm> and there was a long discussion per e-mail
21:34 -!- NeddySeagoon [~NeddySeag@gentoo/developer/NeddySeagoon] has joined #gentoo-council
21:35 <@Betelgeus> Any way I think this is best bundled with a vote on council.
21:35 <@Betelgeus> So if grobian wants to proceed this then next summer is best.
21:36 <@  grobian> ok
21:36 <@Betelgeus> (vote meaning elections)
21:36 <@  grobian> right
21:36 <+scarabeus> Betelgeuse: that is good point, to bond those
21:37 <@Betelgeus> grobian: Some years back Calchan did rounds for larger changes
21:37 <@Betelgeus> grobian: We could brainstorm some more to see what else is useful
21:37 <@  grobian> now, or next meeting?
21:38 <@Betelgeus> grobian: next meeting or outside meetings is better to check what all was on the table
21:38 <@  grobian> -project discussions are fine
21:39 <@Betelgeus> I still have to kick the ml discussions but I could mention the EAPI opinion gathering from my talk
21:39 <   Calchan> Betelgeuse: say my name another two times and I will appear before you
21:40 <@Betelgeus> Most devs there agreed that it would be a good idea to mandate using new EAPIs with the exception of security bumps
21:40 <@  grobian> Calchan: wanna run for some changes to glep 39? :)
21:40 <@Betelgeus> Calchan, Calchan
21:40 <   Calchan> grobian: you mean run for council again?
21:40 <@  grobian> Calchan: no, adapt the text
21:41 <   Calchan> grobian: I do have a project, partly on paper, but it involves getting rid of you guys
21:41 <@  grobian> Calchan: lol
21:41 <   Calchan> I'm being very serious here
21:42 <@  grobian> Calchan: feel free to open up the discussion on -project?
21:42 <   Calchan> been there done that, years ago
21:42 <@Betelgeus> Calchan: The Calchan appoints the supreme leader plan? :D
21:42 <   Calchan> although the idea has evolved a lot towards simplicity since then
21:44 <+ WilliamH> Calchan: I'm interested in seeing what you are working on. :-)
21:44 <@  grobian> good, is that open floor, chairman?
21:46 <@ Chainsaw> Yes, it is.
21:46 <@ Chainsaw> The microphone is on folks.
21:46 <   Calchan> WilliamH: I may blog about it, I've been slacking on this for about 3 years
21:47 <   Calchan> WilliamH: I actually wanted to do it during that week off when I was not supposed to spend watching my wife in a coma in the ICU
21:47 <+ WilliamH> how is http://bpaste.net/show/76870 for the key signing announcement?
21:47 <+ WilliamH> s/key signing/commit signing/
21:48 <@  grobian> WilliamH: requesting -> strongly suggesting ?
21:49 <+ WilliamH> grobian: fixed
21:50 <@Betelgeus> WilliamH: typo in your name
21:50 <+ WilliamH> hehok I'll fix it.
21:50 <@      ulm> WilliamH: s/sign commits/sign manifests/
21:50 <@      ulm> otherwise there will be confusion after the git migration
21:51 <@      ulm> where the two are different
21:52 <+ WilliamH> http://bpaste.net/show/76874
21:52 <@  grobian> WilliamH: terrific
21:53 <+ WilliamH> grobian: ok cool.
21:54 <@  grobian> ok, so no open floor items?
21:56 <@  grobian> Chainsaw: can we close the meeting?
21:56 <@ Chainsaw> grobian: Please proceed. Who chairs the next one?
21:56 <+ WilliamH> Should this just go to dev-announce, or to dev as well?
21:57 <@Betelgeus> Chainsaw: ulm
21:57 <+dberkholz> hasta luego
21:57 <@ Chainsaw> WilliamH: Just dev-announce with reply-to dev.
21:57 <+ WilliamH> Chainsaw: ok
22:00 <@  grobian> later all, thanks Chainsaw for chairing
22:00 <@ Chainsaw> And thanks ulm for texting!
22:00 <@Betelgeus> WilliamH: both with reply-to
22:00 <@Betelgeus> WilliamH: the archives for gentoo-dev should have the full thread
22:00 <@      ulm> grobian: small correction to the summary, 2009-07 should actually read 20090625
22:01 <@      ulm> whose log is not on the council page, whatever that means
22:01 *** ulm will file a bug
22:01 <@Betelgeus> ulm: you don't have it?
22:01 <@  grobian> ulm: fixed