[gentoo-commits] gentoo commit in xml/htdocs/proj/en/council/meeting-logs: 20071108-summary.txt 20071108.txt

[Donnie Berkholz (dberkholz)]

dberkholz    07/11/08 22:22:25

  Added:                20071108-summary.txt 20071108.txt
  Log:
  Add 8 November meeting.

Revision  Changes    Path
1.1                  xml/htdocs/proj/en/council/meeting-logs/20071108-summary.txt

file : http://sources.gentoo.org/viewcvs.py/gentoo/xml/htdocs/proj/en/council/meeting-logs/20071108-summary.txt?rev=1.1&view=markup
plain: http://sources.gentoo.org/viewcvs.py/gentoo/xml/htdocs/proj/en/council/meeting-logs/20071108-summary.txt?rev=1.1&content-type=text/plain

Index: 20071108-summary.txt
===================================================================
amne        here
betelgeuse  absent (1 hour late)
dberkholz   here
flameeyes   here
lu_zero     here
vapier      absent (no show)
uberlord    resigned
jokey       here (replacing uberlord)

Agenda:
http://thread.gmane.org/gmane.linux.gentoo.devel/52772

Also continuing discussion on CoC enforcement. Proposal:
http://thread.gmane.org/gmane.linux.gentoo.council/82

==============================================================================

Empty council slot: vote for Jokey to happen on gentoo-council list
-------------------------------------------------------------------

Jokey is the candidate to replace uberlord [1], and it requires a 
unanimous council vote [2]. Since not all council members are present, 
we'll do this vote on the gentoo-council list. All 6 present council 
members supported Jokey's addition.

1. http://www.gentoo.org/proj/en/council/meeting-logs/20070208-summary.txt
2. http://www.gentoo.org/proj/en/council/voting-logs/council-2007-vote-distribution.txt


Daylight savings change: no slacker marks
-----------------------------------------

In the US and Europe, 2000 UTC shifted by an hour in local time. The 
email announcement was wrong, so we will not give slacker marks to the 
two absent council members.

vapier needs to fix his script before the next announcement.


EAPI=1 approved for use in the main tree
----------------------------------------

Stable portage version 2.1.3.12 supports EAPI=1. It's now officially OK 
to start using it in the main tree. From the ebuild ChangeLog for 
portage:

  This release is the first to have support for EAPI-1 (#194876), which 
  includes SLOT dependencies (#174405), IUSE defaults (#174410), and 
  ECONF_SOURCE support for the default src_compile function (#179380). 
  Package maintainers should carefully consider the backward compatibility 
  consequences before defining EAPI="1" in any ebuilds, especially if
  other packages depend on those ebuilds. See the ebuild(5) and emerge(1) 
  manual pages for EAPI related documentation.

EAPI=1 features are documented in PMS as well as the man pages, but they 
are not yet documented in the devmanual or the dev handbook.

Code of Conduct enforcement proposal: generally positive feedback
-----------------------------------------------------------------

dberkholz sent out a proposal this morning [1], so we just discussed it 
today instead of voting. Initial feedback from council members was 
positive. Some concerns came up on the list about time zone differences 
and coverage, which were brought up again during the meeting.

People generally agreed that although the environment is better now, it 
hasn't been before and won't always be good.

lu_zero brought up the point that since things are fairly good now, we 
don't need to rush through this process and we can take our time and do 
things right.

Council support for the team's actions should not be as controversial 
with the requirement that all actions be private.

The team will need to create the tools to deal with the actions it needs 
to take (short-term moderation on IRC, mailing lists, and Bugzilla). 
This could happen during the initial training period suggested on the 
gentoo-council list.

If there's already existing moderation somewhere (for example many of 
the #gentoo-* IRC channels or the forums), the team will first liaise 
with them rather than preempt them. All official Gentoo forums must 
adhere to the CoC, however; having their own moderation does not exclude 
them from following Gentoo's standards as a whole.

The expectation is that successive actions against the same person will 
increase in length, eventually reaching the 3-day cutoff that requires 
council approval and forwarding to devrel/userrel. The idea is that if 
someone keeps violating the CoC after a given length of moderation, it 
apparently wasn't enough so it shouldn't be repeated.

Next month, we will vote on a concrete proposal.

1. http://thread.gmane.org/gmane.linux.gentoo.council/82


Baselayout-2: uberlord will continue to maintain it
---------------------------------------------------

lu_zero asked whether we had anything to do about baselayout-2 since 
uberlord resigned. He's continuing to maintain it in a git repository 
and will remain upstream for it. More details will emerge over time.

kingtaco raised the question of trusting external releases and hosts. 
Some responses suggested that using git may prevent the malicious host, 
because of the possibility of GPG-signed tags. He mentioned the 
possibility of the infra team hosting Gentoo-critical repositories with 
access by non-Gentoo developers. It's just an idea at this point, but 
he's going to talk to the rest of the infra team.



1.1                  xml/htdocs/proj/en/council/meeting-logs/20071108.txt

file : http://sources.gentoo.org/viewcvs.py/gentoo/xml/htdocs/proj/en/council/meeting-logs/20071108.txt?rev=1.1&view=markup
plain: http://sources.gentoo.org/viewcvs.py/gentoo/xml/htdocs/proj/en/council/meeting-logs/20071108.txt?rev=1.1&content-type=text/plain

Index: 20071108.txt
===================================================================
20:01 <@dberkholz> anyone got a web link to the agenda thread? not a whole lot on it
20:01  * jokey ties amne to $chair
20:01 -!- nox-Hand [i=johnhand@unaffiliated/nox-hand] has joined #gentoo-council
20:01 -!- Netsplit anthony.freenode.net <-> irc.freenode.net quits: cruxeternus, peper, koxta
20:01 < nox-Hand> Am I too late to talk?
20:01 < nox-Hand> And HAI by the way if I can talk
20:02 <@dberkholz> http://thread.gmane.org/gmane.linux.gentoo.devel/52772
20:02 -!- Netsplit over, joins: cruxeternus, koxta, peper
20:03 <@dberkholz> alright, roll call
20:03 <@dberkholz> who's here?
20:03 <@amne> amne:
20:03  * Flameeyes present
20:03  * jokey me
20:03 <@amne> uberlord is absent due to his resignation, which also makes it a point for the agenda
20:03 <@amne> ah, hi jokey :-)
20:04 <@dberkholz> amne: yes, that's on the linked thread
20:05 <@amne> dberkholz: yupp
20:06 <@amne> so where's the rest?
20:06 <@dberkholz> lu_zero, vapier, SpanKY, Betelgeuse: where are ya?
20:07 < nox-Hand> Late reply to the roll call, I am here, but not sure if that matters any..
20:07 <@dberkholz> nox-Hand: just council members; feel free to chip in later on, when the discussion you're interested in is happening
20:07 <+jokey> nox-Hand: erm nop, council people now...
20:07 -!- NeddySeagoon [n=NeddySea@gentoo/developer/NeddySeagoon] has joined #gentoo-council
20:08 < nox-Hand> jokey: dberkholz I Figured as much, cheers, I'll start chirping at my allocated chirpy time :]
20:09 -!- uberpinguin [n=uberping@unaffiliated/uberpinguin] has joined #gentoo-council
20:09 <@amne> hm
20:10  * igli gags nox-Hand
20:10 <@amne> due to the change to daylight savings in europe the meeting is an hour earlier today
20:10 < NeddySeagoon> amne, everyone beaten by the hour change last week ?
20:10 <@amne> i guess that could explain why lu_zero and Betelgeuse aren't here (both europeans, aren't they?)
20:10 < fmccor> Even US has changed now.
20:10 <@Flameeyes> amne, yeah they are, but as fmccor, us already catched up
20:11 <@amne> ah
20:11 <@dberkholz> can someone help me find the council summary link where it says the next highest voted person will fill an empty spot
20:11 <+jokey> so should we delay another bit? ;)
20:11 <@Flameeyes> dberkholz, should be around february last year
20:11 <@amne> council.g.o doesn't work for me right now, anyone else got the same problem?
20:12 <@amne> http://c.g.o to be more accurate
20:12 <@Flameeyes> dberkholz, http://www.gentoo.org/proj/en/council/meeting-logs/20070208-summary.txt
20:12 < NeddySeagoon> dberkholz, I don't think it was ever written down in the process but its established as custom and practice now
20:12 <+jokey> amne: WORKSFORME
20:12 <@Flameeyes> NeddySeagoon, that link above
20:12 -!- Netsplit anthony.freenode.net <-> irc.freenode.net quits: musikc, mpagano__, genone, zxy_64
20:12 <@amne> jokey: strange, but as long it works for everyone else :-)
20:12 <@lu_zero> cough...
20:13  * lu_zero is alive
20:13 <@lu_zero> distracted but alive...
20:13 < NeddySeagoon> Flameeyes, thank you
20:13 -!- Netsplit over, joins: musikc, mpagano__, zxy_64, genone
20:13 <@Flameeyes> [I remembered it easily, it was just before I left :)]
20:13 <@amne> gentoofan23 just pointed out there still was some confusion: gentoofan23> amne, the time is 15:12 EST right now whereas the e-mail said 16:00 EST so maybe that explains it..
20:14 <@dberkholz> well, that means we'll have 5 of 7 people
20:14 <@lu_zero> legal/solar time mismatch?
20:14 <@Flameeyes> [21:10] [Whois] Betelgeuse has been idle for 20 hours, 0 minutes, and 20 seconds.
20:14 <@Flameeyes> Betelgeuse seems not around
20:14 <@dberkholz> the other two should've paid closer attention
20:15 <@Flameeyes> mike will probably appear midway through the meeting as usual
20:15 <@dberkholz> we might as well move on, today will be primarily discussion
20:15 <@lu_zero> ok
20:15 <@amne> works for me
20:15 <@dberkholz> first item requires no voting
20:15 <@Flameeyes> fine
20:15 <@dberkholz> we're welcoming jokey to the council, filling the spot left vacant by uberlord
20:16 <@Flameeyes> and I lost my liason :)
20:16 <+jokey> poor Flameeyes ;)
20:16 <@lu_zero> eh...
20:16 <@dberkholz> so, welcome jokey!
20:16 <@dberkholz> how do we get the ops status fixed?
20:17 <+jokey> tomaw should be able to
20:17 < tomaw> Hi
20:17 <@Flameeyes> /tomaw access #gentoo-council add jokey 30
20:17 <@Flameeyes> :P
20:17 < tomaw> You people should really equip yourselves to do this on your own.
20:18 < tomaw> Let me check, as I probably shouldn't be doing it without a nod from the right people
20:18 <@lu_zero> hmm
20:18 <@amne> vapier is listed as channel contact, he should be able to do it
20:18 < tomaw> yeah, he can do it, but noone else (bar staff) can
20:18 <@dberkholz> tomaw: aren't we the right people? =P
20:19 <@lu_zero> what about having all of us marked as contact/with an higher level?
20:19 < tomaw> No, freenode requires that the group contact ask staff.  group contacts are currently phreak, christel and whoever your recruitment lead is now
20:19 <@dberkholz> you can only add people to levels below your own
20:19 <@Flameeyes> lu_zero, beside owner and staff, anyone else can add up to (hislevel-1)
20:19 < fmccor> tomaw, Betelgeuse is.
20:19 <@amne> uhm
20:20 <@amne> dberkholz: If they accept and the current Council unanimously accepts the new person, they get the position with a 'reduced' term such that the yearly elections still elect a full group.
20:20 < tomaw> fmccor: that's the guy!
20:20 <@Flameeyes> fmccor, let's get someone who's around, don't we? :P
20:20 <@amne> dberkholz: doesn't that mean we actually need to vote on jokey?
20:20 < fmccor> tomaw, phreak might be.
20:21 < tomaw> I pinged him already :)
20:21 <@dberkholz> amne: it certainly reads that way. i wonder if we already effectively did so on the mailing list thread about it
20:21 <+jokey> just do again so we're set ;)
20:21 <@Flameeyes> if we were all around we could vote on it, but I doubt Betelgeuse is around now
20:22 <@lu_zero> we could just use the email...
20:22 <@amne> works for me
20:22 <@lu_zero> next item?
20:23 <@dberkholz> the daylight savings change mentioned
20:23 <@dberkholz> announcement email was wrong, so i say we don't give slacker marks this week
20:24 <@amne> yupp
20:24 <@lu_zero> fine
20:25 <@Flameeyes> oky doky
20:25 <@amne> vapier: plz fix your announcement mail kthx :-)
20:25 <@dberkholz> alright, next item
20:26 <@dberkholz> approving EAPI=1 for use in the main tree
20:26 <@dberkholz> i wish Betelgeuse were here for this because he had some points
20:27 <@lu_zero> dberkholz which portage version won't be able to handle it?
20:27 <@dberkholz> portage has supported the EAPI variable for more than a year, and stable portage (2.1.3.12) now has EAPI=1 support.
20:27 <@lu_zero> and how old would it be?
20:27 <@lu_zero> fine to start using it then
20:28 <@dberkholz> documentation is in PMS, ebuild and emerge manpages, but it should be added to the devmanual and dev handbook
20:28 <@lu_zero> I see
20:28 <@Flameeyes> what were the finalised features in EAPI=1?
20:28 <@dberkholz> This release is the first to have support for EAPI-1 (#194876), which includes SLOT dependencies (#174405), IUSE defaults (#174410), and ECONF_SOURCE support for the default src_compile function (#179380).
20:29 <+jokey> slot deps and iuse defaults being the more important parts
20:29 <@Flameeyes> I'm fine for it just at slot deps
20:29 <@dberkholz> what EAPI=1 means is already finalized, we're just saying it's fine to start using it in gentoo-x86.
20:30 <@Flameeyes> yeah
20:30 <@dberkholz> i talked to zmedico yesterday and he thought it would be a good idea for the council to approve it
20:30 <@lu_zero> it's fine
20:30 <@amne> ++
20:30 <+jokey> ack
20:31 <@Flameeyes> dberkholz, your vote?
20:31 <@dberkholz> yes from me
20:31 <@Flameeyes> so it's fine for five out of seven (two missing)
20:33 <@dberkholz> alright, last agenda item is discussion of the CoC enforcement proposal
20:33 <@dberkholz> i got it together too late for a vote
20:33 -!- phreak`` [n=phreak``@gentoo/developer/phreak] has joined #gentoo-council
20:33 <@lu_zero> dberkholz I read it and seems fine
20:34 <@lu_zero> probably I should take a bit of time to rethink it
20:34 <+jokey> the idea of it is good imho
20:34 <@amne> i also like it
20:35 <@Flameeyes> I join lu_zero
20:37 -!- mode/#gentoo-council [+o jokey] by ChanServ
20:37 <@lu_zero> oh =)
20:37 <@amne> duck and cover!
20:38 <@amne> so, which points from the proposal need specific discussion?
20:38 -!- phreak`` [n=phreak``@gentoo/developer/phreak] has left #gentoo-council ["................. do I need to say anything else ?"]
20:39 -!- Netsplit anthony.freenode.net <-> irc.freenode.net quits: musikc, mpagano__, genone, zxy_64
20:39 <@dberkholz> amne: you're probably in the best position to talk about whether my understanding of the situation was correct
20:39 <@jokey> I think one valid point was the timezone problem
20:39 -!- Netsplit over, joins: musikc, mpagano__, zxy_64, genone
20:39 <@dberkholz> yeah, but there are distinct times of day that are much more active than others
20:40 <@amne> yeah, i think that's the main point, i'll take a quick look at the mails agani
20:40 <@dberkholz> i don't think we need to try for 100% coverage of people watching everywhere every second
20:40 <@amne> i also like the idea of having a test run as suggested - during that time we can also evaluate how the coverage works
20:40 <@dberkholz> we just need to show that we will enforce it if it's violated
20:40 <@lu_zero> there were violations lately?
20:41 < fmccor> lu_zero, off and on --- some things devrel sees now are really CoC issues in my opinion.
20:42 <@amne> fmccor: agreed
20:42 < fmccor> lu_zero, but rather low volume to us.
20:42 <@dberkholz> i'm not sure how closely y'all read it for devrel/userrel connections, i did add a line about that
20:42 <@lu_zero> fmccor so there isn't a dire need for them
20:43 <@lu_zero> _so_ nobody would start talking about us putting them up in hurry because of crisis
20:43 <@amne> lu_zero: still things may explode again some day, so we shouldn't wait until then
20:43 <@jokey> lu_zero: just that history has proven we could use it from time to time
20:43 <@Flameeyes> agreed with amne
20:43 <@dberkholz> my earnest hope is that we can create a group of people who spend 99% of the time taking no actions
20:43  * jokey resyncs brainlink with amne
20:44 <@amne> jokey: get out of my head! :-P
20:44 <@amne> dberkholz: yes
20:44 <@lu_zero> amne the main point is the second
20:44 <@Flameeyes> dberkholz, and 1% doing drills organised by us councilers?
20:44 <@dberkholz> heh. they could look over old irc logs. =P
20:44 < fmccor> lu_zero, remember, devrel typically does not watch for these very quick things unless we happen to see them in passing.
20:44 <@lu_zero> since there isn't need we can approve/appoint them
20:45 <@lu_zero> s/need/immediate need/
20:46 < NeddySeagoon> My hope is that the council will back the body while they work - and hold a 'lessons learned' after every incident
20:47 <@amne> good idea
20:47 <@jokey> I don't see a reason to not back stuff up atm, but we'll how stuff goes
20:48 <@amne> otherwise the people execute CoC don't know what the council wants in the first place
20:48 < jmbsvicetto> dberkholz: An important point about userrel, Is that userrel hasn't been doing the "user control" role for a long time
20:48 <@lu_zero> why not?
20:48 < jmbsvicetto> dberkholz: userrel can definitely have that role, but it's something that has been lost for some time
20:48 <@dberkholz> that makes me think of a couple of points. first, the council won't back up decisions it disagrees with. second, the council needs to be careful not to overreact
20:48 <@Betelgeuse> hmm it's winter time
20:48 <@Betelgeuse> damn it
20:48 <@lu_zero> hi Betelgeuse
20:48 <@Betelgeuse> stupid UTC
20:49 <@amne> Betelgeuse: welcome
20:49 <@jokey> heya Betelgeuse
20:49 <@dberkholz> Betelgeuse: http://dev.gentoo.org/~dberkholz/summary-20071108.txt
20:49 <@dberkholz> summary so far
20:49 -!- Netsplit anthony.freenode.net <-> irc.freenode.net quits: musikc, mpagano__, genone, zxy_64
20:49 <@amne> dberkholz: agreed, but they still should back up the team doing the work, even if they disagree with a single decision
20:50 <@dberkholz> and on the other side of that, the team needs to handle its actions appropriately
20:50 -!- Netsplit over, joins: musikc, mpagano__, zxy_64, genone
20:50 <@dberkholz> doing the right thing in the wrong way will ruin it
20:50  * amne nods
20:50 < NeddySeagoon> dberkholz, My point is that the council supports any action in progress until its concluded, so we don't have incidents like the one that killed proctors
20:50 < igli> amne's point is critical though, imo
20:50 <@Betelgeuse> dberkholz: Did you want my input on something?
20:50 <@dberkholz> Betelgeuse: nah, was just mentioning some points you brought up about EAPI=1
20:51 <@amne> i guess a --dry-run phase would help a lot with these issues
20:51 < jmbsvicetto> lu_zero: I think that role hasn't been carried by userrel because the current members have focused more on other projects
20:51 < NeddySeagoon> dberkholz, The council 'interfering' in a work in progress will undermine the team
20:51 <@Betelgeuse> dberkholz: Well it's documented in the current PMS version so that's good.
20:52 <@dberkholz> NeddySeagoon: i think that will be difficult if not impossible with the change to only private actions
20:52 -!- Tabasco [i=rhcp@gateway/tor/x-91a87846c415d662] has quit [Read error: 104 (Connection reset by peer)]
20:53 <@dberkholz> all the team would do release is anonymous statistics; X people modded for 6 hours; Y for 12 hours; etc
20:53 <@dberkholz> s/ do//
20:53 < NeddySeagoon> dberkholz, Thats a change I welcome - much procting was done in private anyway
20:55 <@dberkholz> Betelgeuse: while you're here; do you support just adding jokey to council? we need unanimous
20:55 <@Betelgeuse> dberkholz: sure if he is the next one on the list
20:55 <@Betelgeuse> I thought that was already decided by the last counci
20:56 <@amne> i think we should also discuss which tools are available for moderation or even needed to implement
20:56 <@dberkholz> Betelgeuse: the last council said that the next person would be offered the spot, contingent upon unanimous council approval
20:56 <@dberkholz> amne: yes that's very true, i thought about that and didn't add anything to the proposal
20:57 <@Betelgeuse> dberkholz: okay
20:58 <@amne> last proctors had some mailing list interface (which was done by robbat2 iirc) to block emails from people, was there anything else?
20:59 <@jokey> I'm with the mail that stated we don't need another set of moderation for irc and forums where it is in place already
20:59 <@jokey> so we don't need more than this interface imho
20:59 <@amne> other stuff (that's not implemented, but i think was planned) would have been delaying messages until someone approves they are flame free
21:00 <@amne> well, that feature could be potentially useful - otoh if mailing list bans are short term it may not be worth the effort to implement
21:00 <@jokey> ack
21:00 -!- gentoofan23 [n=gentoofa@gentoo/contributor/gentoofan23] has quit [Client Quit]
21:01 <@lu_zero> ++
21:02 <@Flameeyes> [I'm not saying anything because I'm fine for what you're saying :)]
21:02  * Philantrop wonders what jokey and lu_zero ack'ed - "potentially useful" or "not worth the effort".
21:02 <@amne> also the bans would need to be unset manually after the time of ban/silencing/foo is over
21:02 <@dberkholz> jokey: yep, mailing lists and #gentoo-dev are the main things
21:02 <@amne> this means people should also be there to unban people in a timely manner
21:02 <@dberkholz> might be able to script that sorta thing
21:03 <@amne> dberkholz: atm the mailing list has a web-interface for that
21:03 <@amne> but that's all technicalities that i think can be solved. i think it would be best to ask robbat2 about these things, too
21:03 <@jokey> if needed, scripting it shouldn't be too hard iirc
21:03 < jmbsvicetto> dberkholz: I think that would require some sort of cron for that
21:04 <@amne> perhaps he can easily implement a self-destruct timer for bans
21:04 < igli> at would be better
21:04 <@amne> ban amne@g.o for 12 hours -> done
21:04 <@dberkholz> that's really just a detail that can be figured out after we decide whether to do this at all
21:04 <@amne> nod
21:04 <@dberkholz> let whoever's doing the work deal with it
21:04 <@jokey> yup
21:05 < NeddySeagoon> amne, it directs mail to /dev/null just now
21:05 <@amne> NeddySeagoon: ack
21:05 <@amne> is anyone aware of any other big issues that we haven't discussed yet?
21:06 <@dberkholz> regarding CoC, or in general?
21:06 <@amne> CoC
21:08  * jokey assigns silence to None ;)
21:08 <@dberkholz> i wrote down all my ideas, so i'm waiting for others to respond. =)
21:08 < fmccor> Is bugzilla covered by it?
21:09 <@dberkholz> anywhere without existing moderation should be, so i'd say so
21:09 <@dberkholz> which i would interpret as "anything gentoo except for #gentoo and forums"
21:09  * fmccor had guessed that, but did not recall.
21:09 <@amne> dberkholz: i like the proposal (including the things that have been discussed now)
21:09 < igli> i don't think having one strong leader is a good idea. better a strong collective imo.
21:09  * fmccor wanted to hear "yes" :)
21:10 <@amne> i think it would be good to continue in that direction and finalize it
21:10 < NeddySeagoon> dberkholz I was expecting your paper to be discussed on -project rather than -council.  As an ex proctor, I intend to provide the benefit of my experiance.  I've just been reading the thread
21:11 <@amne> discussing it further on -project sounds like a good idea to me
21:11 <@lu_zero> we could try -project if the noise ratio increase we'll fallback on -council
21:11 <@lu_zero> deadlines for it?
21:11 < NeddySeagoon> amne, That was what the last council meetting decided
21:11 < igli> project is pretty quiet ;-)
21:12 <@dberkholz> project list had an opportunity to express its interests, which it did
21:12 <@dberkholz> from those interests i created a concrete proposal
21:12 <@dberkholz> i don't think that concrete details are well-designed by committees
21:13 <@dberkholz> i do think basic interests are, though
21:13 < NeddySeagoon> dberkholz, agreed
21:13 <@dberkholz> so if any of the basic interests are incorrect or need refinement, i'd be happy to hear about that wherever
21:13 < jmbsvicetto> dberkholz: One word of caution, your comment of everything gentoo except #gentoo and forums caused a lot of grievance for proctors
21:13 <@dberkholz> jmbsvicetto: could you expand on that?
21:14 < jmbsvicetto> dberkholz: Most #gentoo-* irc channels feel they already have a moderation team and are not to keen on having outside interference
21:14 <@dberkholz> if they want to be an official channel, they have to deal with the CoC
21:14 < jmbsvicetto> dberkholz: Also, no one has ever been able to come up with a consensous list of official #gentoo-* channels
21:15 <@dberkholz> whatever's on the IRC channels page
21:15 <@amne> jmbsvicetto: all #gentoo-* is official
21:15 <@Flameeyes> ##gentoo-* should be the unofficial ones
21:15 <@jokey> amne: I don't think so, some are driven by users
21:15 <@amne> jmbsvicetto: if you want it to be non-official, it's ##gentoo-*
21:15 <@dberkholz> but certainly this team would need to work with the existing moderation, it doesn't need to take direct action to preempt them
21:15 <@Flameeyes> [like we have ##gentoo-anime]
21:15 < jmbsvicetto> dberkholz: I agree, but for instace, #gentoo-userrel, #gentoo-devrel, #gentoo-sparc or #gentoo-pt, just random examples, are not likely to welcome that interpretation
21:16 <@lu_zero> Flameeyes we should have #gentoo-anime
21:16 <@dberkholz> are they part of gentoo? if so, the CoC affects them.
21:16 <@Flameeyes> lu_zero, let's keep it unofficial ;)
21:16 <@dberkholz> we just need to figure out the best way to deal with that
21:16 <@Flameeyes> dberkholz, regional channels are a problem though, that I agree with
21:16 <@Flameeyes> [it so happens I'm one of the #gentoo-it moderators btw]
21:17 <@dberkholz> that's why the team would liaise with existing moderation
21:17 <@amne> yes
21:17 < jmbsvicetto> amne: ok, though question then, is #gentoo-xeffects an official gentoo channel? Also, has freenode changed their policy? Last time we had this discussion the gentoo contacts left it very clear that is was impossible to refuse anyone from registering #gentoo-my-channel
21:18 <@jokey> tomaw: have some up-to-date details there ^^
21:18 <@lu_zero> jmbsvicetto if they have the gut to register it as different project
21:18 <@amne> jmbsvicetto: yes; not that i am aware of; i've seen channels being removed because they were not official and used # instead of ##
21:18 <@lu_zero> and we aren't knocking them with a trademark hammer
21:18 <@lu_zero> sure
21:19 <@amne> lu_zero: yeah, it were some special cases where people used the official #gentoo-* to be abusive
21:19 <@amne> lu_zero: don't remember the exact details any more
21:19 <@amne> i don't think anyone who does some good work will get slapped by us for not being official enough or whatsoever
21:19 < jmbsvicetto> So I'm not misunderstood, I do agree that #gentoo channels should be official channels, but that is not the case currently and freenode doesn't seem to be willing to support that claim
21:20 <@amne> jmbsvicetto: how come?
21:20 < jmbsvicetto> amne: well, if they will allow anyone to register any #<whatever> channel, how can you enforce that #gentoo-* channels are official gentoo channels?
21:21 <@jokey> from what I know, at least within europe you can't enforce trademark over channel names, I think there was something with #debian.de in the past...
21:21 <@amne> jmbsvicetto: freenode staff can get it back
21:21 < jmbsvicetto> amne: But I think that working with existing teams and working with people that are doing "good work" is the best way
21:21 <@amne> jmbsvicetto: yes
21:22 <@lu_zero> agreed
21:22 < jmbsvicetto> However, I urge you to have a "crystal clear" definition of where you want/expect that team to work
21:22 <@amne> jmbsvicetto: i think the only problem would be someone managing an official (#gentoo-*) channel and refusing to accept CoC at all by allowing attacks and other nasty stuff
21:23 < NeddySeagoon> amne, like in -dev ?
21:23 < jmbsvicetto> amne: You were on the proctors team, so you know the answer to that ;)
21:23 <@amne> NeddySeagoon: #gentoo-dev?
21:23 <@amne> jmbsvicetto: heh
21:23 < NeddySeagoon> amne, yes. With everone an op, its impossible to enforce
21:24 < tomaw> jmbsvicetto: it's difficult to refuse them being registered, but a contact could reclaim it if they felt they should
21:24 < tomaw> jmbsvicetto: that's freenode policy - I have no idea of gentoo policy
21:24 <@amne> NeddySeagoon: #gentoo-dev has no clearly defined people in charge, does it? (so it's up to the new team to manage it imho)
21:25 <@lu_zero> everybody should be in charge
21:25 < NeddySeagoon> lu_zero, everbody == nobody
21:25 <@dberkholz> in general it's sort of devrel, since they're the ones who add/remove ops
21:26 < jmbsvicetto> tomaw: At least that's what I gathered from my talks with some gentoo contacts while I worked on the proctors
21:26 < NeddySeagoon> Its a detail to think about
21:26 <@jokey> indeed
21:26 < jmbsvicetto> amne: I agree, but you should expect resistance to a "controlled" #gentoo-dev
21:27 <@amne> i think #-dev should fall under the same treatment as the mailing list
21:27 < tomaw> jmbsvicetto: yes, the gentoo project registered with freenode and claimed #gentoo-* channels.  Currently, it's possible for anyone to register a channel in that namespace, but it can still be turned over the to gentoo project on request.
21:27 < kingtaco|laptop> not really, tell them to go somewhere else
21:27 < kingtaco|laptop> freedom of speech(if we even have it) is the freedom to say what you want, not where you want
21:27 < jmbsvicetto> tomaw: Thanks for clearing that
21:27 <@amne> jmbsvicetto: i told you :-P
21:28 < jmbsvicetto> amne: :)
21:28 < tomaw> jmbsvicetto: note that a single project cannot own #gentoo-* and ##gentoo-*, so gentoo has no control over ##gentoo-* channels
21:30 <@jokey> despite talking on details, are there big points we missed?
21:31 <@amne> seems everyone's happy (or fallen asleep)
21:31 <@amne> perhaps it hasn't been said explicitely, but from my understanding, if someone gets into trouble frequently, issues will be escalated to devrel, right?
21:32 <@lu_zero> I think it's sane
21:32 <@jokey> amne: devrel or userrel
21:32 <@amne> jokey: right
21:33 <@dberkholz> my presumption is that successive actions against the same person will eventually extend to the 3-day cutoff that means they need council approval and forwarding to devrel
21:33 <@amne> dberkholz: ack
21:33 <@dberkholz> but i didn't really allow for someone getting a million 6-12 hour blocks
21:33 <@amne> heh
21:33 <@dberkholz> maybe i should expressly disallow that, as it's the same "warnings go on forever" problem that means nothing ever gets fixed
21:34 <@amne> hm
21:35 <@amne> personally i'd prefer loose guidelines and individual treatment depending on the situation
21:35 <@amne> but if someone is warned and repeats his behaviour, action should follow
21:36 <@lu_zero> ok
21:36 <@dberkholz> http://dev.gentoo.org/~dberkholz/summary-20071108.txt has my summary of the discussion so far
21:36 <@dberkholz> we should wrap it up soon and send back to the -council list
21:38 <@lu_zero> do we have anything to discuss about baselayout2 and uberlord?
21:39 <@amne> dberkholz: summary looks good to me
21:39 <@Betelgeuse> lu_zero: Not much if Uberlord contains to maintain it  externally
21:39 <@Betelgeuse> lu_zero: Otherwise we do need to find someone who works on it.
21:39 <@lu_zero> Betelgeuse BSD license switch included?
21:39 <@dberkholz> it wasn't a switch, it was an addition
21:39 <@dberkholz> so dual GPL/BSD
21:40 <@dberkholz> both of which are FSF free, fwiw, so fits our social contract fine
21:40 <@lu_zero> held on git on our infrastructure?
21:40 <@dberkholz> not sure about the infrastructure bit. anyone know?
21:41 <@Flameeyes> I think the idea was to use our git
21:41 <@Betelgeuse> I think infra does not allow commit access to non devs but I could be wrong.
21:41 <@dberkholz> i'm also unsure how much it matters; i suspect other distros use init systems they don't personally keep on their infra
21:42 <@jokey> imho it doesn't matter where it resides as long as someone packages it for gentoo ;)
21:42 <@Betelgeuse> dberkholz: yeah like upstart
21:42 <@Flameeyes> dberkholz, yeah doesn't really matter, it just means we have to split baselayout (base files) from the rc system
21:42 <@Flameeyes> and that was probably a good idea to begin with
21:42 <@lu_zero> as long it doesn't break for us...
21:43 <@dberkholz> Flameeyes: i'm sure upstream would be glad to include distro files for any distros that wish to use it. =)
21:43 <@Flameeyes> dberkholz, I think Roy's idea was splitting them, anyway I'll certainly follow up with him about it as I do have interest in it
21:43 <@dberkholz> sure, i don't know the details
21:43 <@lu_zero> ok
21:44 <@dberkholz> anyway it seems that baselayout-2 remains in good hands and maintained
21:45 <@dberkholz> anyone got another open floor topic?
21:46 <@jokey> nox-Hand: you wanted?
21:46 < kingtaco|laptop> what about infra?
21:46 <@dberkholz> kingtaco|laptop: is there something in particular about infra you'd like to bring up?
21:47 < kingtaco|laptop> nope, just cought my eye
21:47 < kingtaco|laptop> you guys were talking about infra
21:47 <@dberkholz> kingtaco|laptop: oh, we're just not exactly sure where baselayout-2 repository will end up
21:47 <@amne> kingtaco|laptop: we decided infra needs to provide the ice cream machine i promised in the election :-)
21:47 < kingtaco|laptop> amne, sure, just send me the money
21:47 <@amne> heh
21:47 <@lu_zero> kingtaco|laptop we were thinking about overhauling one of the boxes
21:47 < kingtaco|laptop> dberkholz, yeah, thats a problem.
21:48 < jmbsvicetto> amne: Please put it next to bender. kthxbye :P
21:48 < kingtaco|laptop> lu_zero, ?
21:48 <@dberkholz> kingtaco|laptop: i see it as an open question, certainly
21:48 <@lu_zero> kingtaco|laptop for the icecreams
21:48 <@dberkholz> it's fine on any site that hosts git, as far as i'm concerned
21:48 < kingtaco|laptop> lu_zero, no recycled icecream please
21:49 <@lu_zero> dberkholz so far repo.or.cz
21:49 < kingtaco|laptop> dberkholz, tbh, I don't know if that's good
21:49 <@dberkholz> kingtaco|laptop: why?
21:50 < kingtaco|laptop> well, put something as critical as baselayout on anything other than our hardware and it's hard to trust.  that said the guy who write the stuff is no longer a dev so it's hard to trust him as well
21:50 <@lu_zero> kingtaco|laptop solutions?
21:50 <@jokey> it's open source so you take a look at the diffs
21:50 <@dberkholz> kingtaco|laptop: how about all the other code on your system?
21:50 < kingtaco|laptop> whoever is importing it into gentoo is going to have to keep a close eye on whats going on
21:50 <@lu_zero> beside abducting roy
21:50 < kingtaco|laptop> lu_zero, sadly, no
21:50 < kingtaco|laptop> dberkholz, well, baselayout is critical
21:50 < kingtaco|laptop> other stuff is not
21:51 <@dberkholz> baselayout is just as critical as glibc, gcc, coreutils
21:51 <@jokey> ack
21:51 < jmbsvicetto> I agree with kingtaco|laptop in particular if we take into account the decision regarding the PMS
21:51 <@lu_zero> still we can keep a local git
21:51 <@lu_zero> and pull from roy from time to time
21:51 <@dberkholz> and yes, it's git so we can do whatever we want
21:51 <@dberkholz> we can maintain our own patchset on top, we can track his branch, we can fork off, whatever
21:52 < kingtaco|laptop> I'm not saying not do it, I'm saying whichever gentoo dev is the one who is importing it is going to have to audit each and every import.  there is just too much chance for an exploit.  projects like coreutils/gcc/glibc have built up a reputation which is why they are different
21:52 <@jokey> and not trusting him because he just is no longer a dev tastes very very bad imho
21:52 <@dberkholz> that's disgusting
21:52 <@Betelgeuse> I would trust Uberlord lot more than new devs.
21:53 <@dberkholz> roy's built a reputation too, and quitting gentoo does not affect my view of it in any way
21:53 < kingtaco|laptop> would you trust whatever random host he hosts the repo on?
21:53 < kingtaco|laptop> I know I wouldn't
21:53 <@dberkholz> i trust sha512
21:53 <@jokey> if he uses git ans signs it, then yes
21:53 <@jokey> *and
21:53 < kingtaco|laptop> SCM doesn't matter here
21:53 < kingtaco|laptop> signs it perhaps
21:54 < kingtaco|laptop> however, you're still trusting uberlord to audit his own code before he releases
21:54 <@dberkholz> how's that different? =)
21:54 <@amne> (i have to slack off and take care of some stuff, hope no one minds)
21:54 < kingtaco|laptop> it's where you put the trust
21:55 < igli> isn't that what the testing releases are for?
21:55 -!- desultory [n=dean@gentoo/developer/desultory] has joined #gentoo-council
21:56 < kingtaco|laptop> when it was on gentoo hardware developed by a gentoo dev there was a lot of trust.  now it's on random hardware by a (trusted) ex dev
21:56 <@jokey> kingtaco|laptop: re trust... we had exploitable code for year(s) on packages. and we controlled the repo and had a gentoo dev maintaining it. so I really don't see a point
21:56 < kingtaco|laptop> I think you need to consider how it can be attacked before you decide on trivialities like git
21:56 <@dberkholz> git supports gpg-signed tags and the whole thing is based on sha1 sums
21:57 < kingtaco|laptop> jokey, packages is not an attack vector for each and every single gentoo install
21:57 < kingtaco|laptop> baselayout is
21:58 < kingtaco|laptop> dberkholz, I don't know enough about git to comment on it's trustworthness
21:59 -!- leio_ is now known as leio
21:59 < igli> hang on; uber signs it so it's verifiably his code. base-system bring it in and sign off on review, then it goes thru testing. where's the vector?
21:59 <@dberkholz> the suggestion is that nobody will audit the code, either roy or ebuild maintainers
21:59 <@dberkholz> and the git host will somehow be the vector
21:59 <@dberkholz> i'm just asking in #git about that
22:01 < kingtaco|laptop> when it was on gentoo hardware, you could trust the repo as much as you trusted a combination of gentoos security, arch, and infra teams.  when it's somewhere random, you can't trust that git doesn't have a bug or that the host of the git repo isn't doing something malicious
22:01 <@dberkholz> i don't think malicious hosts are possible with how git is implemented
22:02 <@dberkholz> but i'm asking for clarification
22:02 -!- uberpinguin [n=uberping@unaffiliated/uberpinguin] has quit ["Leaving"]
22:02 < Philantrop> kingtaco|laptop: Would infra allow non-devs to commit to the repository if it was on Gentoo hardware?
22:02 < bonsaikitten> apart from bugs in git itself the signing should be as tamper-resistant as any other system
22:02 < igli> sorry i don't see it; only vector is a bug in gpg, which i agree has happened before
22:02 < kingtaco|laptop> dberkholz, this is a package that's so important that infra would entertain allowing a baselayout repo for uberloard to use
22:02 -!- fmccor_ is now known as fmccor|home
22:03 -!- Ingmar^ [n=ingmar@83.101.12.48] has joined #gentoo-council
22:03 < kingtaco|laptop> Philantrop, normally no, this is an extrodonary case where I would bring it up with the other infra folks
22:03 -!- Ingmar [n=ingmar@83.101.12.89] has quit [Nick collision from services.]
22:04 <@jokey> anyway there are no full rewrites near so a diff shouldn't be hard to look into (at least when looking at last updates)
22:05 < Philantrop> kingtaco|laptop: Well, if I had anything to say about it, I'd prefer the Gentoo hardware under that circumstances even though I don't think there are any attack vectors but gpg itself.
22:05 < igli> would be better
22:05 < kingtaco|laptop> Philantrop, on the surface, I would prefer that too, but infra has to have a pow-wow to look deeper
22:06 < kingtaco|laptop> dberkholz, jokey can you table this until next month so I have time to talk to infra peeps?
22:06 <@dberkholz> repo.or.cz is a git hosting site from the git developers, so i guess if you trust git, you can trust the site
22:06 <@dberkholz> kingtaco|laptop: we're not making any decisions so we have nothing to table
22:06 < kingtaco|laptop> if we can't do it, then I'll shut up
22:06 < eroyf> kingtaco|laptop: infra is willing to let uberlord work on baselayout even though he's not a developer?
22:06 < kingtaco|laptop> eroyf, not exactly
22:06 < eroyf> then what exactly
22:07 < kingtaco|laptop> what I proposed is that infra might create a seperate repo for him to use for the sole purpose of his baselayout development
22:07 < eroyf> so he's going to stay as a developer or what?
22:08 < kingtaco|laptop> afaik, he quit
22:08 < kingtaco|laptop> if he's intending on coming back, I don't know about it
22:08 <@dberkholz> basically the proposal is that we'd become a very limited project-hosting site for non-gentoo devs working on critical gentoo packages
22:08 < eroyf> or do you create a repo for him to work on and then let someone with an @gentoo.org commit it to the *right* repo?
22:08 < eroyf> with svn support?
22:09 < kingtaco|laptop> I don't care about the scm
22:09 < igli> which other packages are so critical, and not dev'ed by gentoo?
22:09 < kingtaco|laptop> if I had my way everyone would still use RCS :p
22:09 < eroyf> PMS for example.
22:09 < eroyf> which is not a package.
22:09 < eroyf> but infra simply refused to let non-developers get access to that repo
22:09 < eroyf> so this is somehow interesting.
22:09 < igli> if moot
22:10 < kingtaco|laptop> extremely moot
22:10 < eroyf> well, i'm looking forward to see what you decide
22:10 < kingtaco|laptop> ok...
22:10 <@dberkholz> since we're not going to make any progress at this meeting, let's just adjourn the meeting
22:11 <@dberkholz> feel free to keep talking about it afterwards
22:11 -!- igli [n=igli@unaffiliated/igli] has left #gentoo-council ["Have a good one ;-)"]
22:11 < kingtaco|laptop> I don't have much more to say, I need to talk to my team about it and work out the details.  I'll send you guys an email when infra agrees on something
22:16 < fmccor|home> A final thanks to dberkholz for his CoC proposal, which I view as great progress.
22:17 <@dberkholz> thanks fmccor|home, i'm glad it went over fairly well
22:17 -!- windzor [n=windzor@84.238.69.202] has quit [Remote closed the connection]
22:18 < fmccor|home> In my opinion, better than "fairly well". :)



-- 
gentoo-commits@g.o mailing list