Extras/SteeringCommittee/Meeting-20061109

From FedoraProject

Jump to: navigation, search

Log

Note: This week's meeting logs are truncated due to server disconnect near the end of the meeting. Due to the time, after reconnecting, the meeting was adjourned.

(10:00:04) thl has changed the topic to: FESCo meeting
(10:00:08) thl has changed the topic to: FESCo meeting -- init
(10:00:19) thl: hi everybody
(10:00:22) c4chris_ is now known as c4chris
(10:00:25) thl: who's around?
(10:00:30) c4chris: thl, hi
(10:00:34) ***rdieter waves
(10:00:39) ***scop is here in a couple of minutes
(10:00:52) RoozbehPournader: heh, my first meeting after ages...
(10:00:56) ***abadger1999 is around
(10:01:09) thl: small warning: the "v" on my notebooks keyboard seems to not work propoerly anymore
(10:01:15) jima: oh, hey, it IS that time again, isn't it?
(10:01:22) rdieter: ha
(10:01:47) ***jima goes to get lunch, since he's rabble anyway ;)
(10:02:04) thl has changed the topic to: FESCo meeting in progress -- Enterprise Extras
(10:02:07) tibbs: I'm here.
(10:02:40) thl: there was something about ppc32 kernels from centos in the buildroots
(10:02:49) jwb: hey i'm here
(10:02:58) thl: there is is: RHEL4 has only a 64 bit ppc kernel the kernel is installed in the default buildroot. we need to get a 32bit ppc kernel or disable ppc for RHEL4. [WWW]  CentOS has a ppc32 kernel using this is the simplest solution.
(10:02:58) jwb: abadger1999, and i'm _not_ late! ;)
(10:03:15) abadger1999: jwb: haha :-)
(10:03:21) jwb: thl, i had a separate question about EPEL...
(10:03:22) rdieter: thl: +1 (just do it)
(10:03:31) f13: I'm still curious as to why you need the KERNEL in the buildroot?
(10:03:52) thl: rdieter, just do it +1
(10:04:07) thl: f13, dgilmore explained it to me, but I forgot why
(10:04:16) jwb: thl, right now users will not be able to do mock builds for EPEL unless they have RHEL locally
(10:04:19) f13: its a silly dep on initscripts
(10:04:28) f13: which ALSO seems odd to need in a buildroot
(10:04:45) warren: f13, fake-build-provides?
(10:04:56) f13: warren: probably what we do internally
(10:05:04) warren: f13, well.... kernel-devel of the appropriate arch is necessary for kmod
(10:05:17) thl: warren, exactly...
(10:05:25) f13: does kernel-devel actually require kernel?
(10:05:56) f13: but thats a more fun question, are we supplying extras to RHEL?  If so, ppc32kernel modules are out of the question anyway
(10:06:22) thl: f13, we are supplying extras to RHEL and CentOS
(10:06:46) ***jwb waits for current discussion to end
(10:07:00) f13: which gets us into some odd places wrt whats available on Centos vs whats avialable to RHEL
(10:07:18) thl: f13, for example?
(10:07:29) f13: a ppc32 CentOS isn't a RHEL clone, its something different
(10:07:59) warren: And are kmod's needed in any specific case on ppc?
(10:08:04) warren: nvidia and ATI don't work there
(10:08:06) rdieter: f13: and that is relevant-to/a-problem-for EPEL how?
(10:08:14) f13: warren: asterisk perhaps.
(10:09:05) thl: hmmmm
(10:09:16) ***thl is unsure how to proceed
(10:09:24) jwb: use CentOS on the builders
(10:09:25) thl: I'd say we install the ppc32 kernel for now
(10:09:32) warren: Why is ppc32 CentOS different from RHEL4?
(10:09:33) f13: rdieter: I'm not entirely sure there is one,
(10:09:39) thl: and don't proide kmod for now
(10:09:47) thl: jwb, I don#t like that idea to much
(10:09:52) thl: jwb, I'd prefer RHEL
(10:09:54) rdieter: f13: good. (:
(10:10:03) thl: jwb, and BTW, we discussed that already ;)
(10:10:05) jwb: thl, i don't like the fact that people can't do mock builds with RHEL on the builders
(10:10:08) f13: jwb: thats not going to fly too well for getting RHEL acceptance of these extra packages, and any kind of internal RH support for the project
(10:10:11) warren: +1 don't provide kmod for ppc at first, figure out what to do later.
(10:10:26) jwb: thl, yeah i know we discussed it.  i seemed to miss the fact that mock builds don't work
(10:10:42) ***mmcgrath late
(10:10:47) mmcgrath: catching up
(10:10:47) f13: if we're not going to do kmod, why not do a fake-build-provides and keep any CentOS packages out of the buildroot?
(10:10:56) thl: jwb, and I don't like the fact that we build a a clone of one distribution that should be compatible with the original; thus are packages should work on the original
(10:11:09) thl: jwb, but htere are two "should" in that sentece
(10:11:12) thl: sentence
(10:11:20) thl: and that's one to much for me
(10:11:35) jwb: thl, how can we expect maintainers to support EPEL if they can't build against what's on the buildroots locally?
(10:11:52) rdieter: why not prvode links to centos repos for (local) mock users?  what's the big deal?
(10:11:53) thl: jwb, we agreed on that they can use centos
(10:12:18) dgilmore: warren: RHEL4 only has support in AS for IBM hardware
(10:12:21) f13: also, you might be able to get RH to pony up non-supported access to RHEL packages
(10:12:30) warren: dgilmore, ah
(10:12:38) RoozbehPournader: just a side note. as an extras packager, i can't even get my hand on RHEL even if I pay redhat.
(10:12:40) jwb: rdieter, if we did that, ok.  we currently arent
(10:12:54) warren: I move that we skip kmod on ppc at first, just to get the project off the ground.
(10:12:58) thl: RoozbehPournader, but accessing centos should be no problem
(10:13:02) warren: The usage case for ppc kmod's are far less needed.
(10:13:04) RoozbehPournader: thl: yep
(10:13:07) thl: RoozbehPournader, and you don't *have* to buiuld for centos
(10:13:17) thl: or rhel
(10:13:22) jwb: rdieter, thl: i just want to make it possible for EPEL maintainers to do a mock build without having to have RHEL or CentOS local
(10:13:45) thl: jwb, sratch build targets?
(10:13:49) thl: scratch
(10:13:54) dgilmore: jwb: id  like to see some EPEL mock configs  that we distribute
(10:14:03) jwb: thl, no local mock build
(10:14:11) thl: (seems my whole keyboard is a bit mad today (een more that normally))
(10:14:12) jwb: thl, make mockbuild in an EPEL branch
(10:14:12) rdieter: jwb: won't putting references to centos bits in mock's configs do that?
(10:14:32) thl: rdieter, +1
(10:14:35) jwb: rdieter, yes.  but where is it going to find the repos at?
(10:14:49) f13: the centos mirrors?
(10:14:56) rdieter: f13: +1
(10:15:24) jwb: it's a bit odd, but i'd be ok with that.  if our Extras mock package ships those configs
(10:15:28) dgilmore: thl: ive been asked to branch everything for EPEL
(10:15:42) jwb: dgilmore, huh?
(10:15:43) thl: everything?
(10:15:43) f13: everything?
(10:15:45) f13: what?
(10:15:48) scop: ???
(10:15:49) f13: -EINSANE
(10:15:50) jwb: no...
(10:15:51) rdieter: echo... echo... echo...
(10:16:00) warren: o_O
(10:16:01) delero left the room (quit: "Leaving.").
(10:16:07) c4chris: yikes
(10:16:14) dgilmore: skvidal: ping
(10:16:14) jwb: f13, so can you update the Extras mock package to include EPEL configs that point at CentOS mirrors?
(10:16:28) jwb: f13, or would you accept a patch to do so
(10:16:32) f13: jwb: you can ask clark to do it upstream.
(10:16:40) f13: I'm just a lowly packager (:
(10:17:03) ***jwb sighs
(10:17:09) skvidal: dgilmore: yes?
(10:17:20) dgilmore: skvidal: branching of EPEL
(10:17:27) skvidal: yay
(10:17:27) skvidal: yes
(10:17:29) skvidal: let's do that
(10:17:40) jwb: not for everything
(10:17:40) dgilmore: some people dont want everything branched
(10:17:53) tibbs: Please do not branch my packages.
(10:18:08) scop: who *does* want everything branched, and why?
(10:18:09) mmcgrath: tibbs: why not?
(10:18:17) thl: we should only branch stuff we have maintainers for
(10:18:27) tibbs: I do not intend to maintain all of them them in EPEL.
(10:18:28) ***awjb says hi to everybody
(10:18:30) jwb: mmcgrath, because it should be an explicit request from a maintianer
(10:18:30) f13: and who want to maintain somethign for EPEL
(10:18:40) dgilmore: scop: the CentOS guys  who currently porvide everything
(10:18:46) mmcgrath: Aren't they the community's package?  Not tibbs?
(10:18:51) thl: sure, I *hope* we have eerything in EPEL later when it's working properly, but we should not start with all of it (that would be insane)
(10:19:01) tibbs: So you have a maintainer lined up for them?
(10:19:12) jwb: mmcgrath, if an EPEL maintainer for them steps up, sure.  until then, fsck no
(10:19:14) mmcgrath: I don't think we're saying build everything.
(10:19:25) thl: dgilmore, well, will the "CentOS guys  who currently porvide everything" help us with maintaining that stuff?
(10:19:28) f13: why create a branch that will languish?
(10:19:31) mmcgrath: in fact everything won't build, a lot of them will require alterations.
(10:19:32) f13: and bitrot?
(10:19:44) dgilmore: thl: we are trying to get them to do exactly that
(10:19:45) mmcgrath: Its just what we do with devel->FC6
(10:19:52) mmcgrath: just the way its always been right?
(10:19:55) jwb: mmcgrath, no, it's not
(10:20:05) thl: dgilmore, well, that needs to be coordinated
(10:20:16) mmcgrath: jbw: can you point me to a package that has no FC-6 branch?
(10:20:26) ***mmcgrath is just playing devils advocate here.  Its causing chatter so I'll stop.
(10:20:34) abadger1999: We could put a dead.package type file into the EPEL branch unti a maintainer steps forward....
(10:20:45) jwb: mmcgrath, in Extras?  yes.  none of them have FC-6 branches.  they're FE-6
(10:20:46) tibbs: Why not just not branch in the first place?
(10:20:55) mmcgrath: symantics.
(10:20:57) f13: branch from where?
(10:21:08) jwb: mmcgrath, and it's not like doing devel->FE-6
(10:21:13) dgilmore: RHEL4  from FC-3 and RHEL5 from devel
(10:21:15) mmcgrath: fine FE-6.  The point is everything in devel becomes FE6, so it seems to reason everything in devel would b ecome EL->4
(10:21:25) warren: the cvs branching script can make a branch from anything, not just devel
(10:21:32) jwb: mmcgrath, it's more like doing auto-branches for FC-3 on a new package import
(10:21:41) jwb: mmcgrath, which we don't do eithre
(10:21:52) thl: jwb, +1
(10:21:52) f13: branching devel to be EL4 doesn't make sense.
(10:22:05) warren: a one-time branch of FC-3 to EL4 makes sense
(10:22:06) c4chris: what's the advantage of doing a "branch everything" ?
(10:22:06) f13: there are chagnes in devel that will not be compatible with EL4.  branching from FC-3 makes more sense.
(10:22:15) warren: one-time branch of FC-6 to EL5 makes sense.
(10:22:16) dgilmore: f13: i was going to branch FE-3 to RHEL4
(10:22:54) awjb: I think it would be better to let maintainers step up an then create the epel branches that the packager requests... no default branching from fe
(10:22:57) mmcgrath: Even then many packages will be broken.
(10:22:58) tibbs: Even when FE-3 branches are unmaintained and haven't been updated?
(10:23:00) f13: er, the branch names are FC not FE
(10:23:00) c4chris: why not only branch the packages that will actually be built and maintained?
(10:23:05) warren: Should we branch EL5 later, when RHEL5 becomes closer to release?
(10:23:18) rdieter: warren: +1
(10:23:33) warren: tibbs, that's why we shouldn't do a blanket branch, but rather only things that maintainers request.
(10:23:45) thl: okay okay okay
(10:23:45) tibbs: Yes, that is my argument.
(10:23:48) rdieter: branch on demand/request only.
(10:23:56) awjb: +1
(10:23:57) jwb: i say we vote on it
(10:23:59) warren: Branch FC-3 to EPEL4 only on request.
(10:24:00) thl: we need to order the discussion a bit more
(10:24:05) f13: warren: RHEL5 should branch from FC-6
(10:24:09) RoozbehPournader: auto-branching is also scary. packagers will think they are expected to take care of the new branches also.
(10:24:12) warren: Branch FC-6 to EPEL5 on request, but no today.
(10:24:25) ***jwb tries something new
(10:24:32) jwb: VOTE: branch EPEL on request only
(10:24:34) jwb: +1
(10:24:37) warren: +1
(10:24:38) rdieter: +1
(10:24:39) ***f13 has no vote
(10:24:39) scop: +1
(10:24:43) ***f13 isn't in FESCO (:
(10:24:43) c4chris: +1
(10:24:43) abadger1999: +1
(10:24:48) awjb: +1
(10:24:49) thl: jwb, +1
(10:24:58) thl: jwb, (you beat me with that ;-) )
(10:24:59) jwb: tibbs, ?
(10:25:06) warren: OK ratified?
(10:25:09) tibbs: +1
(10:25:11) thl: settled
(10:25:17) jwb: VOTE END
(10:25:28) warren: VOTE: Branch EPEL5 on request only sometime before RHEL5 is released, so packages are ready upon release day?
(10:25:28) c4chris: cute
(10:25:33) c4chris: :-)
(10:25:36) thl: meoment
(10:25:38) thl: stop please
(10:25:49) thl: that won't work if eerynbody calls out VOT blahg
(10:25:55) RoozbehPournader: not enough discussion.
(10:25:56) thl: sorry
(10:25:57) f13: jwb: and just to nit a pick, the branches in extras CVS _are_ FC-6 and FC-5, not FE
(10:26:24) thl: can we do something like "proposed vote: blahh"
(10:26:27) jwb: f13, fsck.  ok.  sorry mmcgrath.  was caught up in the moment and temporarily lost my brain
(10:26:33) abadger1999: mmcgrath: gnome-theme-clearlooks http://cvs.fedora.redhat.com/viewcvs/rpms/gnome-theme-clearlooks/?root=extras
(10:26:34) mmcgrath: ;-)
(10:26:37) thl: and I'll call out the "ote now please on foo" ?
(10:26:48) jwb: thl, yeah
(10:27:11) thl: thx
(10:27:13) abadger1999: thl: Sounds good.
(10:27:19) thl: so, where are we now
(10:27:27) thl: the "branch EPEL on request only" is settled
(10:27:45) Rathann [n=rathann]  entered the room.
(10:27:49) warren: We should agree upon a request method?
(10:27:50) thl: do we want to ote on warrens idea?
(10:27:56) dgilmore: what can i branch to do testing with?
(10:28:05) f13: dgilmore: mock
(10:28:13) rdieter: dgilmore: feel free to touch anything of mine.
(10:28:14) warren: f13, +1
(10:28:27) c4chris: dgilmore, you can branch my packages
(10:28:31) thl: well, when do we start to build for RHEL5?
(10:28:35) warren: wait now
(10:28:37) thl: now? after beta2?
(10:28:39) warren: we should agree upon a formal request method.
(10:28:47) dgilmore: currently there is rpmdevtools, konversation and ytalk branched
(10:28:48) f13: thl: I'd say after public beta2
(10:28:56) jwb: warren, CVSSync isn't good enough?
(10:28:58) thl: f13, when is that due?
(10:29:05) rdieter: f13: +1
(10:29:09) f13: thl: -EIDONTKNOW
(10:29:20) thl: f13, k :)
(10:29:22) f13: thl:  given the harried looks of a lot of folks around here, soon.
(10:29:34) thl: f13, okay
(10:29:44) thl: then I'd propose this vote:
(10:29:47) rdieter: coincedentally, centos won't have anything until at least after rhel5 beta 2 either.
(10:29:48) warren: Proposed Vote: Request branches on CVSSyncNeeded in the current standard way.
(10:30:15) Belegdol left the room (quit: "Leaving").
(10:30:18) jwb: warren, +1
(10:30:26) thl: VOTE: let us build some small amount of pacakges for RHEL5 now, and realy start after beta2
(10:30:28) f13: oh dear, now we're voting on what to vote on?
(10:30:37) jwb: thl, +1
(10:30:42) warren: f13, +1
(10:30:43) warren: =)
(10:30:49) rdieter: +1 (for everybody)
(10:30:50) c4chris: thl, +1
(10:30:52) awjb: could we sort thisout a little bit
(10:30:55) thl: small amount; rpmdevtools, mock, and four or fie other packages
(10:31:08) warren: thl, seems reasonable, because they are useful
(10:31:16) warren: however, where are public repos of RHEL5?
(10:31:19) abadger1999: +1
(10:31:27) awjb: +1 here as well for both
(10:31:32) dgilmore: warren: there is only the iso's i beleive
(10:31:44) jwb: warren, we just went over that.  public repos for EPEL is CentOS
(10:31:47) ***thl gets more and more confused, the old style worked better
(10:31:55) warren: jwb, RHEL5?
(10:31:57) thl: sorry guys
(10:32:00) rdieter: sounds like a blocker for ehel5 work to me.
(10:32:11) jwb: warren, yeah.  so doesn't exist yet
(10:32:32) dgilmore: warren: ftp.redhat.com has RHEL beta iso's
(10:32:37) warren: if CentOS makes a CentOS5 pre-release repo, then EPEL5 can get started sooner.
(10:32:56) warren: EPEL5 developers need repos for mock
(10:32:59) jwb: if CentOS goes away for some reason, EPEL is screwed
(10:33:03) f13: internal testing doesn't need external public repos
(10:33:05) thl: centos will start working after beta2 iirc
(10:33:12) awjb: dgilmore, but you need a key to install them :*(
(10:33:16) f13: jwb: the epel builders would use rhel5 for packages, not centos
(10:33:23) f13: just external mock configs would use centos
(10:33:23) mmcgrath: jbw: they should be interchangable for our purposes.
(10:33:34) dgilmore: awjb: oh  i havent tried to install it yet
(10:33:39) ***f13 laughs
(10:33:44) f13: the key would be like um 'V'
(10:33:50) f13: or 'C'
(10:33:56) jwb: f13, mmcgrath: so?  i don't have RHEL.  i can't build against it locally without a public repo.  if CentOS goes away, i'll _never_ do EPEL
(10:33:57) warren: I think we should deal with EPEL4 for now, and begin EPEL5 only after CentOS5 pre-release is in public repos.
(10:33:57) awjb: f13, :P
(10:34:24) jwb: see my point?
(10:34:24) f13: jwb: if CentOS goes away, we'd just use whitebox Linux, or Unbreakable (:
(10:34:37) rdieter: jwb: centos isn't going away (anytime soon)
(10:34:41) awjb: jwb, yes same thing I have been thinking about
(10:35:00) jwb: f13, rdieter: yeah, but my point is we're relying on a 3rd party to provide repos for EPEL maintainers to build against
(10:35:27) jwb: f13, rdieter: do i think it's a huge deal?  no.  but i _can_ be a problem
(10:35:28) rdieter: jwb: by that logic we shouldn't be doing epel at all. (silly, if you ask me)
(10:35:29) f13: so we mirror what CentOS has and offer itup.
(10:35:33) jwb: s/i/it
(10:35:36) f13: then we're not relying on them, we're relying on ourself.
(10:35:45) abadger1999: jwb: In a way we'd be doing it even if we somehow supplied RHEL to the developers... Fedora != RHEL...
(10:35:55) jwb: f13, mirror CentOS on fedoraproject.org?  fine with me
(10:35:57) ***z00dax looks in
(10:36:03) ***scop needs to run in 5 or so minutes
(10:36:08) devrimgunduz [n=Devrim]  entered the room.
(10:36:11) ***rdieter waves to z00dax.
(10:36:34) z00dax: a FYI: Centos-5beta within 6 days of public release of EL5b2 ( i386/x86_64 )
(10:36:40) mmcgrath: jwb: yeah, thats valid.
(10:36:48) rdieter: z00dax: thanks.
(10:36:54) mmcgrath: err a valid concern I mean.
(10:37:16) thl: okay, we discussed a lot of stuff, but we didn't get much done in the past 30 minutes
(10:37:29) thl: well, what we did get done was: cofuse each other
(10:37:39) thl: I'd like to stop here and now
(10:37:44) jima: wow, i guess i didn't really miss anything :(
(10:37:47) rdieter: thl: +1 to that
(10:37:48) thl: we don't get much further now
(10:37:52) jwb: thl, ok
(10:37:59) c4chris: right
(10:38:12) z00dax: considering EL5 is going to be similar stock as FC6, which should stabalise now, I dont think there will be much porting / patching required for the first few months, therefore doing EL5 only after a public beta2 makes sense to me
(10:38:12) thl: page in the wiki is there
(10:38:14) thl: http://www.fedoraproject.org/wiki/Extras/Schedule/EnterpriseExtras
(10:38:29) thl: if you want to get a topic discussed next wee at some info there
(10:38:42) thl: and/or "proposed vote: foo"
(10:39:00) z00dax: right
(10:39:00) thl: be sure to give enough background infos to make it a smooth meeting
(10:39:23) thl: wand there is also the priate and public mailing lists
(10:39:34) jwb: thl, yep
(10:39:41) thl: if we want to discuss someting, before it's brought to the wiki
(10:39:48) thl: moing to the next topic now
(10:40:12) thl has changed the topic to: FESCO meeting -- what's the status of maintainer-announce-list ?
(10:40:20) thl: warren, ?
(10:40:33) thl: all I know was told me on fedora-adisory-list
(10:40:42) thl: why don#t we get a status update during the week?
(10:41:09) thl: that would be helpfull, as we could moe on / find other solutions before the next meeting
(10:41:16) thl: all I know is that f13 blocks the list
(10:41:22) thl: (sorry f13 )
(10:41:29) f13: I'm not the only one (:
(10:41:33) f13: more lists suck more.
(10:41:41) thl: f13, sure
(10:41:58) thl: but as I said on fedora-adisory: we have extras members that don#t want to discuss
(10:42:04) thl: they want only the annoucements
(10:42:18) f13: thats tough.
(10:42:20) thl: we need a solution
(10:42:30) f13: because the discussion is usually important to them as well.
(10:42:35) c4chris: could put something on the subject line for easy filtering...
(10:42:35) thl: and I'm try my best to get a mailinglist for that purpose
(10:42:35) tibbs: Revise the charter for fedora-maintainers to discourage discussion there.
(10:42:43) thl: een if that's for extras contributors only
(10:42:52) jwb: tibbs, i don't think that is the right answer
(10:42:53) f13: tibbs: no, discussion often needs to happen.
(10:42:56) abadger1999: f13's argument makes some sense -- Can we address the automated addition of cvsextras to current fedora-maintainers?  That would be a start.
(10:42:57) thl: tibbs, no, I think we need a maintainers only list
(10:43:13) tibbs: fedora-maintainers isn't a maintainers-only list?
(10:43:21) abadger1999: tibbs: There has to be a place for discussion... so we'd end up with another list either way.
(10:43:22) thl: abadger1999, we really need some "automated addition" to some list
(10:43:33) thl: abadger1999, but I don#t think it should be a discussion list
(10:43:49) thl: because people will route that to /de/null after the first flamewar
(10:44:15) thl: (my "v" key really does behave bad today... sorry)
(10:44:32) thl: f13, is another list with the reply-to set to fedora-maintainers that bad?
(10:44:42) jwb: why can't we use fedora-announce?
(10:45:11) rdieter: jwb: fedora-announce is for end-users, not contributors/maintainers, I'd say that's the wrong place.
(10:45:17) thl: rdieter, +1
(10:45:22) jwb: ok
(10:45:30) thl: (otherwise they might route that list to /dev/null)
(10:45:31) jwb: i'm confused then...
(10:45:46) jwb: what sort of stuff is going to this new list then?
(10:45:50) ***c4chris will be afk in about 5-10 mins...
(10:46:01) thl: jwb, policy changes
(10:46:06) thl: jwb, annoucements for mass rebuilds
(10:46:07) warren: thl, sorry, I got smashed by a pile of work again
(10:46:22) warren: thl, maintainers-announce-list will be dealt with real soon.
(10:46:33) thl: jwb, *maybe*  FESCo meeting summaries
(10:46:37) thl: (but probably not)
(10:46:57) jwb: we need LDAP, not another list for that
(10:47:12) ***thl wonders how LDAP could help
(10:47:23) jwb: bah, wrong acronym
(10:47:53) f13: thl: yes, as they miss the discussion in which good things and good information happen.
(10:47:53) jwb: this is what i'm thinking: have something that automatically just sends that stuff directly to each maintainer.  directly.  not via a list
(10:47:55) jima: RSS?
(10:48:05) f13: thl: trying to remember to send back all the 'important' stuff to the announce list is just silly and won't work.
(10:48:05) jima: no, guess not
(10:48:27) thl: f13, I disagree strongly
(10:48:30) warren: LDAP for what?
(10:48:32) f13: thats great.
(10:48:38) warren: oh crap
(10:48:39) thl: f13, a mass rebuild just happens
(10:48:47) thl: and we need a place to annouce that
(10:48:48) jwb: thl, so like: maintainers@fedoraproject.org sends an email to everyone that is a maintainer
(10:48:50) ***warren realizes that he didn't post the fedora directory server thing...
(10:49:05) thl: and a lot of people just will simply do what was annouced
(10:49:08) warren: thl, maintainers-announce-list isn't that?
(10:49:24) f13: warren: I'm vetoing yet another announce list, as is Jeremy
(10:49:24) warren: maintainers-announce-list already exists, we just didn't set it up yet.
(10:49:32) warren: most important part is automating subscription.
(10:49:40) thl: f13, you can't eto a mailinglist that's for extras only
(10:49:42) thl: f13, sorry
(10:49:45) warren: f13, you can't veto it for Extras-only purposes. I understand why you don't want it for Core, and that's fine.
(10:49:46) thl: f13, not your buisiness
(10:49:57) f13: warren: maintainers isn't extras only
(10:49:58) f13: sorry
(10:50:13) f13: and this whole core vs extras thing ends soon.
(10:50:15) thl: f13, then we start maintainers-annouce as extras-maintainers-annouce
(10:50:26) thl: f13, then we have to have the discussion again
(10:50:35) f13: all Fedora package maintainers still must be on fedora-maintaines list.
(10:50:49) ***jwb detects circular arguing
(10:50:55) warren: f13, it is INFEASIBLE to expect VOLUNTEERS to follow a discussion list and not miss important things.
(10:51:03) thl: let's just stop here, for today, too
(10:51:08) thl: we don#t get any further
(10:51:13) warren: f13, it is FEASIBLE to expect ENGINEERS who are PAID to follow lists if they are told to do so as part of their jobs.
(10:51:15) thl: I'l bring that to fedora-maintainers list
(10:51:20) warren: VOLUNTEER and ENGINEER are two different things.
(10:51:26) thl: and we'll discuss that there
(10:51:31) thl: f13, that okay for you, too?
(10:51:39) f13: thl: I suppose.
(10:51:44) jwb: warren, well no.  VOLUNTEER and PAID EMPLOYEE are two different things
(10:51:46) thl: f13, thx
(10:51:51) warren: jwb, ok
(10:51:54) jwb: :)
(10:51:59) f13: warren: if you're signing up to be a package maintainer, you're signing up for some responsibilities.
(10:52:02) f13: end of ostory
(10:52:17) awjb: f13: but not to read n+1 ml's
(10:52:19) thl has changed the topic to: FESCo meeting in proress -- Packaging Committee Report
(10:52:27) thl: was there any report?
(10:52:30) tibbs: Nothing to report this week.
(10:52:37) craigt left the room (quit: "Leaving").
(10:52:39) thl: okay, that was easy
(10:52:47) abadger1999: Except what spot mailed to fesco-list.
(10:52:48) thl has changed the topic to: FESCo meeting in proress --  Sponsorship nominations
(10:52:53) f13: awjb: reading a mailing list created specificially to communicate with all package maintainers should be one of those responsibilities.
(10:53:03) thl: abadger1999, well, let's continue with that on the list for now
(10:53:06) abadger1999: k
(10:53:24) thl: nominated was: Paul F. Johnson (pfj/nodoid)
(10:53:37) thl: bpepple gave his +1
(10:53:41) tibbs: +1
(10:53:44) warren: +1
(10:53:51) jwb: abstain
(10:53:51) awjb: f13: I agree but I am currently subscribed to about 10 +/- fedora ml's and thats a lot... not everybody wants that
(10:53:56) thl: the sponsor of pfj agreed also
(10:54:05) f13: awjb: nothing says you have to be on fedora-devel or fedora-test or any other list.
(10:54:07) c4chris: +1
(10:54:08) thl: + 0.75 from me
(10:54:09) jwb: awjb, f13: later please
(10:54:14) f13: nod
(10:54:17) abadger1999: abstain
(10:54:19) thl: awjb, f13, please stop now
(10:54:28) rdieter: +1
(10:54:32) dgilmore: +1
(10:54:33) awjb: +1
(10:54:35) tibbs: Can folks who are not fully positive towards pfj give him some suggestions?
(10:54:38) thl: we are on a disfferent topic now (sorry, but otherwise the meeting will neer finish)
(10:54:47) tibbs: He deserves that much.
(10:55:01) jwb: tibbs, "fully positive"?
(10:55:07) warren: awjb, maintainers-announce-list exists as an ultra-low traffic alternative that maintainers are required to be subscribed to, because we cannot realistically expect VOLUNTERS to follow the many discussion lists.  If maintainers fail to respond to requests on maintainers-announce-list, then disciplinary action can happen.
(10:55:16) thl: warren, please STOP
(10:55:18) thl: warren, thx
(10:55:26) thl: otherwise we'll neer finish
(10:55:38) thl: warren, we'll continue on the list
(10:55:54) tibbs: jwb: obviously when someone writes 0.75 they have some form of reservation.
(10:55:54) awjb: warren, did not say anything against it... =)
(10:56:00) mmcgrath: thn's just mad because warren said VOLUNTERS instead of OLUNTERS ;-)
(10:56:01) tibbs: Otherwise they'd save three keystrokes.
(10:56:03) dgilmore: awjb: stop
(10:56:04) mmcgrath: s/thn/thl/
(10:56:19) thl: mmcgrath, ;))
(10:56:24) jwb: tibbs, oh i see.  so you're not including abstains in that
(10:56:29) abadger1999: He's a lot better know... I'm just worried that he doesn't always understand the reasoning behind the rules which is hard to quantify... (this impression is from his mass opening of bugs on core mono libs related to pkgconfig having to go in -devel)
(10:56:59) tibbs: Plus, I recall that there was a negative vote during list discussion.
(10:57:10) thl: tibbs, this time?
(10:57:19) ***thl seems to have missed that
(10:57:28) abadger1999: thl: I remember one too.
(10:57:31) c4chris: thl, yup, I saw a -1
(10:57:35) tibbs: It may take me some time to find it....
(10:57:42) thl: there was some discussion about Steve iirc
(10:57:58) craigt [n=cthomas]  entered the room.
(10:58:11) tibbs: It was bpepple's -1.
(10:58:12) thl: well, we got + 7.75 for Paul F. Johnson (pfj/nodoid)
(10:58:21) tibbs: In regards to the sponsorship nominations for Steve Pritchard & Paul
(10:58:21) tibbs: Johnson:
(10:58:21) tibbs: +1 for Steve
(10:58:21) tibbs: -1 for Paul.
(10:58:33) thl: ohh, sorry
(10:58:37) thl: my fault it seems
(10:59:05) thl: the meeting today is ery confusing; sorry for the trouble
(10:59:28) ***c4chris needs to run now
(10:59:30) tibbs: We can postpone the vote until we understand bpepple's objections if folks like, but pfj deserves an explanation.
(10:59:37) c4chris: +1 for Steve from me
(10:59:39) abadger1999: Steve +1
(10:59:45) c4chris: ttyl
(11:00:03) thl: tibbs, let's get back to Paul next week when bpepple is around
(11:00:07) thl: tibbs, that okay?
(11:00:18) tibbs: I have no objections.
(11:00:30) dgilmore: +1 for steve
(11:00:36) thl: okay, seems people want to ote about steve
(11:00:40) thl: +1 from me
(11:00:45) rdieter: +1 steve
(11:00:47) dgilmore: thl: +1 for next week
(11:00:49) awjb: +1
(11:00:59) jwb: abstain
(11:01:29) warren: I don't know enough about Steve, abstain.