From Fedora Project Wiki

< Extras‎ | SteeringCommittee

Revision as of 16:38, 24 May 2008 by Admin (talk | contribs) (1 revision(s))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

(10:00:05 AM) thl has changed the topic to: FESCo meeting -- Meeting rules at [WWW] -- Init process
(10:00:11 AM) thl: FESCo meeting ping -- abadger1999, awjb, bpepple, c4chris, dgilmore, jeremy, jwb, rdieter, spot, scop, thl, tibbs, warren
(10:00:14 AM) thl: Hi everybody; Who's around?
(10:00:15 AM) ***dgilmore is here
(10:00:17 AM) ***bpepple is here.
(10:00:18 AM) ***awjb is here
(10:00:21 AM) ***abadger1999 is here
(10:00:22 AM) tibbs: I'm here.
(10:00:24 AM) ***rdieter is here
(10:00:24 AM) amitdey []  entered the room.
(10:00:28 AM) ***jwb is here
(10:00:32 AM) ***nirik is in the rabble seats.
(10:00:41 AM) jeremy: thl: I'm mostly here
(10:00:53 AM) dgilmore: thl: i beleive spot is on the road today
(10:00:57 AM) thl: k, then let's start
(10:01:07 AM) thl has changed the topic to: FESCo meeting -- EPEL - where to upload stuff (dgilmore, mmcgrath)
(10:01:08 AM) ***mmcgrath pong
(10:01:15 AM) thl: mmcgrath, dgilmore ?
(10:01:20 AM) dgilmore: thl: /pub/epel
(10:01:26 AM) mmcgrath: We've been requesting builds and they've been working.
(10:01:41 AM) mmcgrath: dgilmore: did notting ever get back to you about that?
(10:01:48 AM) ***scop is half here
(10:02:07 AM) dgilmore: mmcgrath: not yet i need to ping him again on getting the sync process setup so packages can hit the master mirror
(10:02:14 AM) thl: dgilmore, did the jeremy, f13 and notting ack /pub/epel ?
(10:02:30 AM) dgilmore: thl: notting did
(10:02:38 AM) mmcgrath: All in all though I we're ready to announce that people can begin requesting branches.
(10:02:55 AM) thl: dgilmore, k
(10:03:05 AM) dgilmore: We are ready for branches and builds
(10:03:06 AM) thl: if anyone dislikes /pub/epel please yell now
(10:03:18 AM) mmcgrath: The plan for now is to branch from FC-3 unless it doesn't exist, in which case branch from devel.
(10:03:26 AM) warren: back
(10:03:32 AM) thl: mmcgrath, sounds like a good idea
(10:03:43 AM) warren: +1 /pub/epel
(10:04:00 AM) thl: dgilmore, mmcgrath, shall the FESCo members the beta testers? for one week before we annouce the stuff to a wider audeience?
(10:04:10 AM) thl: audience?
(10:04:14 AM) dgilmore: thl: :D  im ok with that
(10:04:18 AM) thl: s/one week/some days/
(10:04:20 AM) mmcgrath: yeah, thats probably a wise idea.
(10:04:21 AM) ***c4chris is here now...
(10:04:32 AM) mmcgrath: It'll be easier to have the FESCo people yell at us then the community at large :D
(10:04:39 AM) warren: thl, Yeah, I need stuff from EPEL personally.
(10:04:44 AM) dgilmore: yeah
(10:04:53 AM) dgilmore: warren: its being built right now
(10:04:55 AM) thl: okay; one week?
(10:05:00 AM) mmcgrath: Should I move EnterpriseExtras to /wiki/Extras/EPEL and start filling it with content?
(10:05:02 AM) thl: or a shorter timeframe?
(10:05:12 AM) bpepple: a week seems reasonable.
(10:05:12 AM) dgilmore: one week
(10:05:14 AM) ***mmcgrath likes the week intervals
(10:05:22 AM) thl: mmcgrath, well, we should leave the schedule page where it is
(10:05:24 AM) dgilmore: we can say we are good to go next FESCo meeting
(10:05:35 AM) tibbs: Is the procedure for requesting branches the same?
(10:05:45 AM) thl: mmcgrath, and please use wiki/EnterpriseExtras (or something like that)
(10:05:49 AM) mmcgrath: k, I'll just create a new one.
(10:05:53 AM) dgilmore: tibbs: yes  but you dont need a bugzilla number
(10:05:58 AM) mmcgrath: do you want it out of Extras namespace or in it?
(10:06:00 AM) thl: as Extras might not exists anymore in the future ,)
(10:06:02 AM) warren: what is the name?
(10:06:05 AM) warren: EPEL-3?
(10:06:06 AM) mmcgrath: ahhh, good point
(10:06:13 AM) dgilmore: EL-4
(10:06:17 AM) dgilmore: EL-5
(10:06:17 AM) warren: ah
(10:06:19 AM) amitdey left the room.
(10:06:20 AM) warren: k
(10:06:41 AM) dgilmore: warren: there is no EL-3
(10:06:44 AM) thl: mmcgrath, dgilmore, will we hae a spepareate owners.list for epel?
(10:06:54 AM) thl: mmcgrath, dgilmore, and where is bugzilla for it?
(10:06:54 AM) mmcgrath: hmmm
(10:06:56 AM) dgilmore: thl: i think we need it
(10:07:03 AM) mmcgrath: yeah, we'll need it.
(10:07:17 AM) thl: dgilmore, mmcgrath can you work this stuff out until next week?
(10:07:19 AM) dgilmore: we will need a bugzilla component for EPEL with the same syncing as extras currently
(10:07:21 AM) ***mmcgrath notes there's bugzilla integration that will need to be done.
(10:07:28 AM) dgilmore: thl: dure
(10:07:32 AM) bakers left the room (quit: "Leaving").
(10:07:38 AM) mmcgrath: with a new owners, warren: who's in charge of that?
(10:07:38 AM) thl: dgilmore, mmcgrath, and please take a close look at the schedule page; there are probably other issues that we need to solve sonn
(10:07:55 AM) dgilmore: thl: yeah there is
(10:08:17 AM) warren: dgilmore, you mean Enterprise Extras with components from Extras owners.list?
(10:08:27 AM) warren: even though Extras owners.list will have significantly  more stuff than EPEL?
(10:08:28 AM) mmcgrath: we can discuss some of the technical details at the fedora-admin meeting today
(10:08:31 AM) warren: k
(10:08:50 AM) dgilmore: warren: Enterprise extras with components from owners.el.list
(10:08:57 AM) dgilmore: or something to the effect
(10:09:13 AM) warren: dgilmore, should be possible to do.  we'll talk during infrastructure meeting.
(10:09:13 AM) thl: mmcgrath, dgilmore, we also need discuss who get's allowed for EPEL before we annouce it to the real public
(10:09:14 AM) ***nirik sees the owners.list.el there and thinks about a emacs lisp module. ;)
(10:09:22 AM) thl: mmcgrath, dgilmore, can you prepare that soon?
(10:09:31 AM) mmcgrath: thl: yeah, we'll have something soon.
(10:09:36 AM) dgilmore: warren: we should start a new owners.list  it will allow different owners straight away
(10:09:37 AM) warren: what happened to that centos extras guy?
(10:09:43 AM) thl: mmcgrath, dgilmore thx
(10:09:46 AM) ***cweyl wakes his rabble self
(10:09:50 AM) warren: dgilmore, agreed
(10:10:02 AM) rdieter: warren: z00dax?  He's waiting on us to actually *do* something. (:
(10:10:08 AM) dgilmore: warren: hes still kinda onboard
(10:10:16 AM) BobJensen-Away is now known as BobJensen
(10:10:26 AM) dgilmore: rdieter: correct
(10:10:47 AM) thl: k, anything else regarding epel?
(10:11:08 AM) dgilmore: thl:  just request branches and test test test
(10:11:30 AM) thl: dgilmore, send a reminder to the fesco list if we don#t test enough ;)
(10:11:34 AM) thl has changed the topic to: FESCo meeting -- Opening Core - (warren, jeremy, rdieter)
(10:11:40 AM) thl: warren, jeremy, rdieter, any news?
(10:12:09 AM) warren: jeremy and jkeating are in private meetings explaining all this with management.  I am not privy to the details just yet, but I hear it is going surprisingly good.
(10:12:33 AM) f13: meetings to continue Friday
(10:12:34 AM) thl: jeremy, f13, thx for your work; I don#t want to do your job ;)
(10:12:34 AM) jeremy: thl: nothing yet...  soon.  hopefully at least some news tomorrow
(10:12:53 AM) warren: We should just proceed figuring out what we want for our part.
(10:13:03 AM) jwb: all of it
(10:13:04 AM) thl: okay, what about hte "future for FESCo" stuff that was discussed on fab?
(10:13:05 AM) jwb: we want it all
(10:13:13 AM) thl: do we want to discuss this further here?
(10:13:20 AM) thl: or wait for a signal from the Board?
(10:13:31 AM) jeremy: thl: it's on at least my list of things to discuss in the board meeting tomorrow
(10:13:51 AM) thl: jeremy, k, thx
(10:14:18 AM) thl: did you like the stuff besides the part "50% ratio for community"?
(10:14:22 AM) abadger1999: thl: I liked the direction it was going.
(10:14:27 AM) thl: or did I forget anything important?
(10:15:06 AM) ***thl takes that as no
(10:15:25 AM) warren: I somehow feel that the discussed was a little overdesigned, but no strong feelings.
(10:16:11 AM) thl: just for the reference a quick question here: please say with "-1" "0" and "-1" if you like the "50% ratio for community" stuff
(10:16:16 AM) thl: just out of interest
(10:16:23 AM) bpepple: -1
(10:16:25 AM) ***jeremy abstains
(10:16:28 AM) jwb: 0
(10:16:33 AM) abadger1999: thl: -1
(10:16:34 AM) thl: + 0,5
(10:16:38 AM) c4chris: 0
(10:16:42 AM) tibbs: 0
(10:16:53 AM) warren: -1
(10:17:02 AM) rdieter: -1
(10:17:11 AM) thl: okay, thx :)
(10:17:23 AM) warren: thl, the approach of asking each specific point like this is probably good though.
(10:17:36 AM) tibbs: I have to agree.
(10:17:58 AM) thl: well, do we want to go thourgh the whole proposal now?
(10:18:18 AM) thl: that would take quite some time...
(10:18:23 AM) tibbs: I need to have it in front of me.
(10:18:32 AM) thl: I don#t have it in front of me either
(10:18:34 AM) bpepple: tibbs: agreed.
(10:18:52 AM) thl: if we want to do that let's do it next week
(10:18:53 AM) warren: thl, let's focus on the proposal next week?
(10:19:06 AM) c4chris: warren, yes
(10:19:08 AM) bpepple: warren: +1
(10:19:11 AM) thl: anything else regarding opening core?
(10:19:21 AM) warren: thl, set a time limit that everyone knows, so everyone is familiar and knows THAT is when they must like or dislike parts.
(10:19:39 AM) thl: well, let's talk a bit about hte size now maybe
(10:19:51 AM) thl: two weeks ago there was the idea to make FESCO bigger
(10:20:00 AM) thl: now the plan seems to be to make it smaller
(10:20:07 AM) warren: eh?
(10:20:24 AM) thl: s/plan seems/the idea/
(10:20:27 AM) warren: I understand why some people want that, but I think it is a bad idea. =)
(10:20:38 AM) c4chris: warren, what is bad ?
(10:20:45 AM) tibbs: I think the size is pretty good as it is.
(10:20:46 AM) warren: c4chris, further shrinking FESCo.
(10:20:58 AM) c4chris: k
(10:21:01 AM) warren: I would prefer the current size or slightly bigger.
(10:21:22 AM) c4chris: I wouldn't like it smaller
(10:21:31 AM) abadger1999: I think I agree with sopwith that "less is more" on a logical level.  But emotionally I have misgivings about getting smaller.
(10:21:31 AM) rdieter: it was argued on the list that reducing FESCo size could improve productivity, provided FESCo delegated things more. (:
(10:22:02 AM) thl: well, can everyone just through in his prefered number please here?
(10:22:03 AM) bpepple: abadger1999: I agree.
(10:22:03 AM) abadger1999: We need to delegate more as it is.
(10:22:12 AM) thl: My vote: 11 +/-2
(10:22:19 AM) jwb: abadger1999, you might be faced with a smaller FESCo sooner than you think
(10:22:21 AM) warren: abadger1999, less is more, when you can count on everyone to always be there.  But for a volunteer org, I would prefer to have qualified individuals in FESCo and whoever is available at the time to push forward decisions.
(10:22:49 AM) bpepple: thl: Keeping current size.
(10:23:13 AM) warren: Size of FESCo could be its own thread.
(10:23:19 AM) rdieter: what *is* FESCo's current size (dunno off the top of my head)?
(10:23:26 AM) c4chris: thl, 13 is fine with me
(10:23:28 AM) thl: rdieter, 13
(10:23:30 AM) bpepple: rdieter: 13.
(10:23:44 AM) warren:
(10:23:46 AM) rdieter: lucky 13, sounds good to me. (:
(10:24:11 AM) warren: even with 13, we struggle to have enough people respond and make decisions happen at any given time.
(10:24:26 AM) thl: and do we want to continue as we areuntil F7  or a new election quite soon?
(10:24:42 AM) abadger1999: warren: "Silence is consent"?
(10:24:43 AM) jwb: i think a new election soon would be good
(10:24:49 AM) c4chris: wait before the future is a bit clearer
(10:24:52 AM) thl: warren, then we need to get the community more involved instead of makeing FESCo bigger
(10:24:53 AM) bpepple: thl: I'd say wait until FC7.
(10:24:56 AM) warren: thl, figure out how the new governance with open core will work first.
(10:25:29 AM) rdieter: warren: +1
(10:25:32 AM) jeremy: I definitely think we need to see what the future holds... at that point, election/moving things around starts to be more interesting
(10:25:44 AM) thl: okay, then let's move on now
(10:25:53 AM) warren: we don't need to decide size of FESCo until that point
(10:26:00 AM) warren: move on
(10:26:11 AM) thl has changed the topic to: FESCo meeting -- MISC -- broken deps
(10:26:16 AM) thl: ther are quite some broken deps
(10:26:41 AM) tibbs: Some of it is due to rawhide churn, which is OK as long as the bustedness doesn't persist.
(10:26:44 AM) mether left the room (quit: Remote closed the connection).
(10:26:47 AM) thl: I was told even FESCo members own packages where the deps are broken for quite some time
(10:26:52 AM) bpepple: tibbs: agreed.
(10:27:02 AM) warren: thl, rawhide only?
(10:27:07 AM) thl: well, yes, the last two reports where quire big
(10:27:13 AM) thl: warren, not only iirc
(10:27:25 AM) thl: does nobody read those reports?
(10:27:34 AM) dgilmore: the last one was a rawhide curn.  i need to bump and rebuild snort which is about 10 of them
(10:27:54 AM) warren: big reports send to a list are less effective than individualized reports sent to each contributor for their specific problems.
(10:27:59 AM) tibbs: I read all of them, but I'm not sure I should try to fix them.
(10:28:05 AM) rdieter: there seem to be some parted brokenness that I noticed.
(10:28:07 AM) nirik: last report:
(10:28:09 AM) c4chris: dgilmore, curn ?
(10:28:13 AM) thl: warren, the maintainers get them, too
(10:28:19 AM) jwb: c4chris, churn
(10:28:23 AM) dgilmore: c4chris: churn
(10:28:32 AM) Rathann: if I may: I prefer to read those reports on the list than have them sent directly to my inbox
(10:28:33 AM) c4chris: oh, sorry...
(10:28:36 AM) warren: thl, ah, didn't realize because I didn't receive any.
(10:28:42 AM) Rathann: warren: ^
(10:28:43 AM) tibbs: I guess I could fix syck-php again; I did it last time.
(10:28:59 AM) warren: Rathann, in the future we may be able to make that configurable
(10:29:04 AM) Rathann: cool
(10:29:16 AM) warren: Rathann, but in the majority case, individual reports when action is needed is more effective.
(10:29:18 AM) nirik: look at how many are > 30 days tho... thats what the complaint was...
(10:29:22 AM) dgilmore: what would be nice  is when someone down in the low level end does a bump like that they give a heads up email
(10:29:29 AM) dgilmore: that way alot of noise can be avoided
(10:29:34 AM) bpepple: thl: maybe if a package remains broken for something like 7 days, FESCo needs to step.
(10:29:38 AM) tibbs: Do we want to cover individual packages now?
(10:30:02 AM) thl: bpepple, that might soon end in a lot of work...
(10:30:03 AM) Rathann: warren: majority? how many people did you ask to be able to say that?
(10:30:04 AM) jeremy: dgilmore: heads up are being sent for things that I've seen; but it's still going to lead to at least a day of broken things in extras as long as extras has to wait for the rawhide sync
(10:30:06 AM) dgilmore: plague can only be fixed by legacy  and its low priority
(10:30:25 AM) rdieter:  gift -    (32 days), NOTMYBUG:
(10:30:32 AM) bpepple: thl: True, but some maintainers don't seem willing to ask for help based on the report.
(10:30:33 AM) jwb: dgilmore, why can't it revert to an older version?
(10:30:52 AM) warren: Rathann, it is unrealistic to expect all participants of the project to read daily reports on a list, when 99% of the time it does not concern them.
(10:30:53 AM) thl: bpepple, I think we need a QA Sig and/or a release manager that should take care of it
(10:31:03 AM) bpepple: thl: That sounds fine.
(10:31:15 AM) warren: Rathann, it is more effective to notify the individual in the rare case where their attention is needed, rather than to expect EVERYONE to watch constantly.
(10:31:20 AM) thl: but it seems nobody want to do the work :-/
(10:31:25 AM) nirik: 14 packages broken more than 7 days.
(10:31:26 AM) thl: maybe we should ask on the list
(10:31:30 AM) dgilmore: jeremy: i never got one for  libpcap update
(10:31:40 AM) Rathann: warren: is it? I wonder... I thought maintainers had some mandatory subscriptions
(10:31:45 AM) dgilmore: jwb: it would require epoch so i guess it could be done
(10:31:56 AM) jeremy: dgilmore: I know I saw mail about pcap somewhere....
(10:32:09 AM) dgilmore: jeremy: maybe it went to core only people
Rathann Rathann|work
(10:32:22 AM) daniel_hozac: dgilmore: fedora-maintainers
(10:32:23 AM) Rathann: warren: well I don't really care that much as long as it doesn't clutter my inbox
(10:32:25 AM) warren: Rathann, in an ideal world yes, but we cannot realistically demand such things from volunteers.
(10:32:26 AM) nirik: pcap update mail heads up was on maintainers...
(10:32:31 AM) jeremy: dgilmore: I don't own anything that links to it... so it was definitely on a list
(10:32:36 AM) thl: is anybody willing to put this topic on his plate and work out a solution?
(10:32:41 AM) warren: Rathann, I'm talking about only RARE notifications when something is wrong and you are responsible for fixing it.
(10:32:47 AM) dgilmore: ok  i missed it
(10:32:52 AM) warren: Rathann, not daily reports
(10:32:58 AM) c4chris: thl, if we decide QA members can step in and fix old broken deps, we migth find som epeople to do the work...
(10:32:59 AM) Rathann: ok then
(10:32:59 AM) abadger1999: Rathann: Mandatory subscription but getting people to read the report when it almost never applies to them is the hard part.
(10:33:01 AM) warren: Rathann, unless you haven't fixed it for days in a row, then you'll get daily reports.
(10:33:09 AM) tibbs: Well, there's one maintainer who has four packages broken for 30+ days.
(10:33:24 AM) jima: istr seeing a pcap warning, it just didn't occur to me that some of my newer inherited packages used pcap ;)
(10:33:33 AM) thl: c4chris, everybody can do that already ; allows it
(10:33:52 AM) ***rdieter pulls out his cluestick.  Who needs to be whacked? (:
(10:33:56 AM) c4chris: thl, oh ok.  I didn't remember that part
(10:34:04 AM) thl: c4chris, np :)
(10:34:13 AM) c4chris: k
(10:34:26 AM) tibbs: I think the point is that if you can't get to your packages you need to think strongly about orphaning them.
(10:35:03 AM) thl: tibbs, I agree, but we need somebody that reminds people about it ;)
(10:35:12 AM) bpepple: thl: Agreed.
(10:35:13 AM) warren: tibbs, it might just be a problem of notification
(10:35:31 AM) thl: shall we set up a official "release manager"?
(10:35:37 AM) warren: thl, there is no reason why private individual notification can't happen in an automated fashion.
(10:35:54 AM) _wart_ []  entered the room.
(10:35:55 AM) nirik: warren: it already does... doesnt it? I get them
(10:36:04 AM) thl: warren, ? contributors get  individual notifications
(10:36:08 AM) tibbs: Yes, I got one the other day because of the libpcap churn.
(10:36:20 AM) tibbs: I'm sure ixs knows he has many packages that need work or rebuilding.
(10:36:25 AM) nirik: ie, p0f broke due to libpcap... I got a email about the broken dep. I updated it.
(10:36:31 AM) warren: thl, kind of like security, release manager can be a tedious and thankless task, less fun when your job is to just poke people.  It might only work with someone accountable to the role.
(10:36:32 AM) thl: we probably should limit the reports to the list to the "long time not fixed" sutt
(10:36:35 AM) thl: stuff
(10:36:38 AM) tibbs: I'm just not sure why he doesn't orphan them or ask for help.
(10:36:49 AM) tibbs: thl: I'd go for that.
(10:36:50 AM) warren: thl, err... yeah, I'm a moron. =)
(10:36:52 AM) thl: warren, sure; but we can at least try
(10:36:59 AM) Belegdol [n=jsikorsk@]  entered the room.
(10:37:00 AM) nirik: how about removing broken > 7 days packages? forced orphan
(10:37:05 AM) nirik: (for on devel)
(10:37:09 AM) warren: for devel, fine
(10:37:10 AM) nirik: non devel. ;)
(10:37:18 AM) tibbs: non-devel is important.
(10:37:26 AM) nirik: agreed.
(10:37:27 AM) warren: for non-devel, X days, WARN, Y days, REMOVE
(10:37:32 AM) thl: how about vacations?
(10:37:34 AM) tibbs: I mean, busted dependencies can break installations.
(10:37:39 AM) thl: how checks the cacation page in the wiki?
(10:37:49 AM) thl: who does the orhan process?
(10:38:05 AM) warren: co-maintainership and granting permissions (even just verbally) other contributors should solve that.
(10:38:19 AM) nirik: isn
(10:38:30 AM) warren: Often contributors are just doing the Right Thing when something is obviously broken.
(10:38:34 AM) nirik: 't it better to remove the broken package and they can fix/push a new one when they get back?
(10:38:44 AM) tibbs: I think this is where trusted members of the community just need to step in.
(10:38:47 AM) warren: Ownership should not be such a strict concept.
(10:38:54 AM) thl: I think we really need to bring this discussion to f-e-l
(10:38:59 AM) thl: any volunteers?
(10:38:59 AM) bpepple: thl: +1
(10:39:00 AM) tibbs: But the question is whether this hides maintainers who have gone away.
(10:39:08 AM) warren: How do you feel with the general idea of...
(10:39:09 AM) c4chris: If there are dependencies, I'd prefer a rebuild
(10:39:17 AM) warren: broken depenency, X days, WARN, Y days, REMOVE
(10:39:35 AM) thl: warren, I tend to agree, but it's not that easy
(10:39:42 AM) thl: we are all on vacation now and them
(10:39:56 AM) warren: thus co-maintainership, grants of permission, etc.
(10:40:15 AM) thl: well, does the co-maintainership stuff from owners.list work these days?
(10:40:18 AM) tibbs: Especially in released distros, these things need to get fixed.  Even seven days is too long.
(10:40:22 AM) thl: or is it still broken?
(10:40:31 AM) thl: and does the script send mails to the co-maintainers, too?
(10:40:37 AM) warren: Bob asks me, "Hey Warren, your foo package is broken."  I say, "Hmm. I'm busy now, do you know how to fix it?"  Bob says, "Sure."  Warren says, "Go ahead."
(10:40:38 AM) thl: tibbs, +1
(10:40:42 AM) bpepple: tibbs: agreed.
(10:40:43 AM) tibbs: It works as long as you understand it doesn't do anything.
(10:40:48 AM) xris []  entered the room.
(10:41:04 AM) ***BobJensen never said any such thing
(10:41:05 AM) warren: thl, jeremy was interested in fixing the initialcc thing, but got stuck.  I need to follow up...
(10:41:08 AM) ***warren sends mail about that...
(10:41:25 AM) thl: I think we should fix the initialcc thing
(10:41:33 AM) thl: and then encourage com-maintainership more
(10:41:41 AM) c4chris: thl, yes
(10:41:42 AM) thl: s/com/co/
(10:41:46 AM) jeremy: thl: the first thing is to move the script.  then it should be pretty fixable :)
(10:42:12 AM) thl: okay, we discussed a lot of things now
(10:42:12 AM) warren: jeremy, the script is in cvs
(10:42:20 AM) thl: someone really needs to sum it up
(10:42:27 AM) thl: and post it for discussion on f-e-l
(10:43:03 AM) ***gregdek wonders who the secretary is.  :)
(10:43:10 AM) c4chris: I can do that
(10:43:17 AM) tibbs: We don't need a secretary; we have IRC logs.
(10:43:24 AM) thl: c4chris, that would be great; thx
(10:43:37 AM) thl: c4chris, I'll create a sperate page on the schedule for it
(10:43:44 AM) c4chris: thl, k
(10:43:54 AM) thl: c4chris, could you please add the most important stuff there ? tia!
(10:43:59 AM) thl: k, so let's move on
(10:44:00 AM) nirik: I think we should push to fix the existing <=fc6 ones soon...
(10:44:06 AM) tibbs: A lot of this discussion goes for the EVR problems as well.
(10:44:29 AM) thl: tibbs, yes
(10:44:32 AM) c4chris: thl, will do
(10:44:43 AM) thl: nirik, are you intersted to just fix it in cvs?
(10:45:08 AM) thl: nirik, but warning, people might yell...
(10:45:15 AM) nirik: well, someone should if maintainers aren't...
(10:45:21 AM) thl: nirik, exactly
(10:45:49 AM) thl: I always wanted to do it myself, but did not find the time for it
(10:45:56 AM) nirik: I guess I can look at what needs to be done...
(10:46:19 AM) thl: nirik, many thx; fixing the most important stuff would be a great start
(10:46:22 AM) ***c4chris pases an asbestos +2 suit to nirik
(10:46:28 AM) thl: so, let's move on now
(10:46:30 AM) ***nirik is pretty flame resistant.
(10:46:49 AM) thl has changed the topic to: FESCO meeting -- report from packaging committee
(10:47:17 AM) thl: well, there was quite a bit of discussion on the private FESCo list in the past two hours
(10:47:26 AM) thl: we should do that in the public in the future
(10:47:36 AM) tibbs: Yes, sorry for not having that sent earlier and to the proper place.
(10:47:50 AM) tibbs: I'm just going to take responsibility for doing it in the future.
(10:47:58 AM) rdieter: is fedora-extras list more appropriate then?
(10:48:12 AM) tibbs: thl mentioned fedora-maintainers
(10:48:17 AM) thl: sorry, telephone call here
(10:48:23 AM) jwb: fedora-maintainers
(10:48:34 AM) ***thl only partly here for a moment
(10:48:50 AM) tibbs: I still don't think I'll attempt to summarize the discussion, though.
(10:49:07 AM) ***rdieter nods, fedora-maintainers makes a little more sense.
(10:49:24 AM) rdieter: tibbs: I say summarize anyway, screw anyone who whines.
(10:49:36 AM) ***nirik wonders what the general topic was at least...
(10:49:51 AM) bpepple: nirik: group tag & comps file.
(10:50:05 AM) nirik: ah, that can of worms. ;)
(10:50:17 AM) c4chris: nirik, precisely
(10:50:21 AM) tibbs: I am avoiding attempting to restate the opinions of a certain person, since that would surely result in my demotivation through massive flaming.
(10:51:05 AM) rdieter: tibbs: want to borrow c4chirs' asbestos +2 suit? (:
(10:51:08 AM) ***thl still busy on the telephone, sorry
(10:51:18 AM) jwb: tibbs, who?
(10:51:19 AM) thl: jwb ?
(10:51:23 AM) c4chris: tibbs, I'm pretty sure I can guess...
(10:51:34 AM) thl: can you take the meeting over for a moment please?
(10:51:38 AM) jwb: thl, sure
(10:51:40 AM) tibbs: In the end it's not important.
(10:51:52 AM) nirik: in a perfect dream world, I would love to see a web interface where there is a page for each package, and maintainer could add tags/comments, users could add comments, people could rate the package, and an rss feed could be used to show updates to the package.
(10:52:24 AM) nirik: and a search interface could find packages that match tags or descriptions or commets.
(10:52:36 AM) jwb: ok, so the proposal is to make the Group tag optional
(10:52:39 AM) c4chris: nirik, feel like working on the package database ? ;-)
(10:52:53 AM) dgilmore: jwb: i nack it
(10:53:01 AM) jwb: this apparently breaks smart and apt
(10:53:08 AM) tibbs: It doesn't break them.
(10:53:10 AM) nirik: c4chris: no time I fear... ;)
(10:53:16 AM) rdieter: jwb: not necessarily, but does make them less useful.
(10:53:17 AM) jeremy: jwb: break is an awfully strong word to use there
(10:53:32 AM) jwb: ok, sorry
(10:53:34 AM) tibbs: I can put Group: uncategorized on all of my packages and get the same result.
(10:53:36 AM) jwb: just trying to summarize
(10:54:14 AM) tibbs: I think the bottom line is that this should have gone out for public discussion before the packaging committee voted on it.
(10:54:28 AM) jwb: tibbs, perhaps.  but what is done is done
(10:54:30 AM) |DrJef| [n=jefrey@fedora/Jef]  entered the room.
(10:54:30 AM) rdieter: to me it's simple: I fail to see what problem is being solved here (and only see new ones being cause by the solution).
(10:54:42 AM) jwb: rdieter, i agree
(10:54:46 AM) bpepple: rdieter: +1
(10:54:55 AM) rdieter: tibbs: +1 too, it could have been handled better.
(10:54:56 AM) tibbs: So we have this meaningless tag that you have to fill in with something.
(10:55:15 AM) tibbs: I would like to see Comps settle down first.
(10:55:22 AM) dgilmore: tibbs: it need not be meaningless
(10:55:29 AM) tibbs: But it is today.
(10:55:37 AM) tibbs: Anyeay, please let me finish.
(10:55:38 AM) jwb: it is today for some packages
(10:55:38 AM) abadger1999: jwb, rdieter: Did my list of reasons make no sense?
(10:55:44 AM) tibbs: I would like to see comps settle down first.
(10:55:45 AM) jwb: abadger1999, somewhat
(10:55:52 AM) bpepple: tibbs: Agreed.  I think once the fate of comps is decided might be a better time to look at the group tag.
(10:55:57 AM) tibbs: And then figure out how to somehow encode Comps in groups.
(10:56:04 AM) c4chris: Yes, let's settle down comps first
(10:56:18 AM) c4chris: the killing Group will be a no brainer
(10:56:20 AM) tibbs: Either pick the primary comps location or somehow encode all of the comps locations into groups.
(10:56:24 AM) c4chris: s/the/then/
(10:56:26 AM) jwb: tibbs, i like that
(10:56:27 AM) abadger1999: tibbs: +1
(10:56:54 AM) tibbs: But it's still maintaining the same information in two places.
(10:57:03 AM) jwb: for now
(10:57:10 AM) tibbs: Perhaps once our packaging database is advanced enough, some of this stuff can be dealt with more cleanly.
(10:57:17 AM) rdieter: yeah, the sky is falling.  end of the world.  I think we have bigger fish to fry.
(10:57:23 AM) tibbs: I.e. filling the spec from the database, or filling the database from the spec.
(10:57:28 AM) jwb: ok, so how is for letting the comps stuff settle down before ack/nacking this?
(10:57:37 AM) jwb: s/how/who
(10:57:39 AM) tibbs: rdieter: I agree that there are more important things to spend limited committee time on.
(10:57:45 AM) bpepple: jwb +1
(10:57:49 AM) jwb: +2
(10:57:51 AM) tibbs: jwb: +1.
(10:57:52 AM) c4chris: jwb, +1
(10:57:54 AM) jwb: gah, +1
(10:57:57 AM) abadger1999: jwb: +1
(10:58:11 AM) rdieter: +1
(10:58:21 AM) tibbs: I think part of the problem is that the packaging committee got tied up in the discussion and failed to ask the important question:
(10:58:25 AM) tibbs: why are we voting on this now?
(10:58:28 AM) jwb: jeremy, ?
(10:58:39 AM) jwb: warren, ?
(10:59:10 AM) abadger1999: tibbs: I felt like we voted on it because it was the beginning of the FC7 cycle nad so it was appropriate to modify rpm now.
(10:59:33 AM) jeremy: I'm fine with waiting
(10:59:42 AM) c4chris: abadger1999, makes sense
(10:59:48 AM) ***thl is back, sorry again
(10:59:49 AM) abadger1999: tibbs: I just didn't know that the opposition to shifting the relevant information to comps.xml existed.
(10:59:51 AM) jeremy: I also think it's okay if rpm is modified to not choke if group _isn't_ present
(11:00:05 AM) tibbs: jeremy: +1.
(11:00:11 AM) c4chris: jeremy, +1
(11:00:21 AM) abadger1999: jeremy: +1
(11:00:32 AM) jwb: i have no problems with making the technical change
(11:00:48 AM) tibbs: I thought of the vote as "let core rpm change to allow it", but it grew beyond that.
(11:00:49 AM) jwb: it's the policy change i would like to wait on
(11:01:04 AM) tibbs: And no effort was made to distinquish those things.
(11:01:12 AM) rdieter: jwb: +1
(11:01:18 AM) thl: jwb, +1
(11:01:28 AM) c4chris: jwb, +1
(11:01:43 AM) tibbs: I can buy into that.  +1.
(11:01:44 AM) thl: and a discussion on the list should IMHO still be done, too, even if the PC decided on it
(11:01:47 AM) abadger1999: jwb: +1
(11:02:11 AM) jwb: thl, yes
(11:02:16 AM) thl: s/on it/& already/
(11:02:26 AM) thl: jwb, shall I take over again?
(11:02:31 AM) jwb: sure
(11:02:49 AM) thl: so the consensus afaics is: we ask the PC to defer the issue for now
(11:03:05 AM) thl: but the wrok on technical things to get rid of Groups continues?
(11:03:13 AM) bpepple: thl: +1
(11:03:13 AM) thl: that correct?
(11:03:16 AM) jwb: +1
(11:03:21 AM) c4chris: thl, +1
(11:03:27 AM) thl: k, hen let's move on
(11:03:30 AM) thl: it's getting late
(11:03:43 AM) thl has changed the topic to: FESCo meeting -- Sponsorship nominations
(11:03:47 AM) thl: any new nominations?
(11:03:54 AM) ***bpepple doesn't have any.
(11:04:03 AM) c4chris: nope
(11:04:10 AM) mether [n=ask@fedora/mether]  entered the room.
(11:04:20 AM) thl has changed the topic to: FESCo meeting -- Maintainer Responsibility Policy
(11:04:35 AM) thl: bpepple sent e-mail to f-e-l about EOL for FC3 & FC4. Not many replied, but MSchwendt did ask the following, which should be decided upon: "EOL as in stop-ship? As in close the build servers for FC-3 and FC-4 and make the push script disable FC-3 and FC-4, too?"
(11:04:47 AM) thl: close build servers?
(11:04:56 AM) jwb: i believe EOL should mean EOL
(11:04:58 AM) thl: + 0.75
(11:05:18 AM) tibbs: That's a tough one.
(11:05:18 AM) thl: or are there any good reasons to leave FE3 and FE4 open?
(11:05:21 AM) bpepple: I'm fine with closing the build servers.
(11:05:41 AM) tibbs: One one hand, the security folks (which is mostly scop at the moment) would love to see them go.
(11:05:53 AM) tibbs: On the other hand, epel builds from FC3.
(11:06:07 AM) dgilmore: thl: if legacys closes FC-3 and 4  then we should close FE3 and 4
(11:06:09 AM) dgilmore: not before
(11:06:14 AM) thl: tibbs, it branches from FE3 and that should be no problem
(11:06:15 AM) abadger1999: warren had a good point last week about not slamming the door on a new community legacy.
(11:06:17 AM) kushal left the room (quit: Read error: 110 (Connection timed out)).
(11:06:22 AM) abadger1999: How does that fit in?
(11:06:31 AM) jwb: abadger1999, it doesn't
(11:06:39 AM) jwb: abadger1999, we can't wait around for one to step up
(11:06:46 AM) rdieter: I think we wait until legacy announces something first.
(11:06:49 AM) thl: call out EOL, but leave the builders open until they break ?
(11:06:52 AM) kushal [n=kd@]  entered the room.
(11:07:31 AM) thl: rdieter, are there any plans from legacy to annouce something about it in the near future?
(11:07:47 AM) tibbs: I recall that they're still discussing it.
(11:07:52 AM) rdieter: thl: there were grumblings regarding that on the legacy-list today.
(11:08:02 AM) tibbs: But it's essentially already done.
(11:08:17 AM) rdieter: thl: seems the consensus was when, not if, an announcement will be made.
(11:08:19 AM) thl: wwthen let's get back to it next week
(11:08:23 AM) thl: it#s quite late already
(11:08:36 AM) thl: anything else?
(11:08:50 AM) thl has changed the topic to: FESCo meeting -- free discussion around extras
(11:08:58 AM) rdieter: (anything else we can put off until next week?) (:
(11:09:09 AM) ***c4chris has nothing
(11:09:24 AM) thl: ther are some other things on the schedule, but there is no need to discuss them now afaics
(11:09:29 AM) jwb: at some point in the future we need to get a show of hands on who plans on sticking around until F7 is out
(11:09:36 AM) jwb: in FESCo i mean
(11:10:02 AM) bpepple: jwb: you mean for the next election?
(11:10:25 AM) jwb: bpepple, until the next election, yes.  i mean we need to know if anyone is going to step down
(11:10:43 AM) bpepple: jwb: ah.
(11:10:53 AM) tibbs: I'll be here as long as the community wants me to be here.
(11:11:06 AM) jwb: there have been rumblings from various people.  perhaps we could ask on the list
(11:11:08 AM) thl: bahh, telephone again
(11:11:09 AM) ***c4chris plans to stay
(11:11:19 AM) thl: jwb, could you please end the meeting
(11:11:25 AM) ***bpepple plans to stay ass well.
(11:11:33 AM) bpepple: s/ass/as/
(11:11:35 AM) jwb: thl, yes
(11:11:39 AM) abadger1999: jwb: That sounds like a good idea.
(11:11:50 AM) jwb: ok, i'll start the thread on the list
(11:12:44 AM) thl: sorry, it was the telephone once again
(11:12:49 AM) jwb: is there anything else?
(11:12:58 AM) thl: I don't think so
(11:13:02 AM) thl: let's close for today
(11:13:07 AM) jwb: ok sounds good to me
(11:13:11 AM) ***thl will end the meeting in 30
(11:13:12 AM) bpepple: thl: +1
(11:13:26 AM) ***thl will end the meeting in 15
(11:13:37 AM) thl: -- MARK -- Meeting End