FESCo-2008-05-08

--- bpepple has changed the topic to: FESCo meeting -- Meeting rules at http://www.fedoraproject.org/wiki/Development/Schedule/MeetingGuidelines -- Init process
bpeppleFESCo meeting ping -- bpepple, caillon, c4chris, dgilmore, dwmw2, f13, jeremy, jwb, notting, spot, nirik, tibbs, warren
Hi everybody; who's around?
* jeremy
nirik is here.
tibbs here
warren here
jwb is here
c4chris here
dwmw2 eventually finds the #fedora-meeting channel
f13
* dgilmore is here
bpeppleok, it looks like everyone (except spot, who said he couldn't make it).
c4chrisall accounted for :)
--- bpepple has changed the topic to: FESCo meeting -- Any objection to this week's report from FPC at https://www.redhat.com/archives/fedora-devel-list/2008-May/msg00637.html
bpeppleanyone object to either proposal from FPC?
* bpepple doesn't have any objections to either proposal.
jeremythe first feels a little bit like over-specifying, but since it's just a SHOULD, I won't argue that hard
dwmw2looks good to me
caillonthe only comment i have is:
jeremythe second is obvious enough
dwmw2well, the PatchUpstreamStatus one does; the wiki hates me now
cailloni prefer putting comments about the patch in the actual patch, not the spec file.
tibbscaillon: I agree; I hope to amend the guidelines to allow that in the future.
jwbi'm fine
tibbsUnfortunately when I posted about it I didn't get much traction, so I went with the current proposal.
jeremycaillon: *nod*  especially for upstreams using a sane scm ;)
* nirik is fine with it.
dwmw2hm, should the wiki's "fuck off; I'm too busy to talk to you right now" page be marked no-cache?
or is that counter-productive? :)
caillontibbs, please do.  i'll +1 with that in mind.
* dgilmore is ok with them
c4chrislooks fine to me
dwmw2caillon: proper changelog comments in the patch is a great idea; I quite like having a one-line status comment in the specfile too, though.
jwbboth are good.  move on?
bpepplealright, I don't hear any objections to the proposal, so the FPC can consider them approved.
tibbsdwmw2: Perhaps dig up my message titled "Patch metadata" from fedora-devel two weeks ago and reply to it?
caillon as well.
bpeppleok, moving on.........
--- bpepple has changed the topic to: FESCO-Meeting -- sponsor nominations - Denis Leroy (denis)
caillon+1
bpepple+1
tibbs+1
c4chris+1
jeremy+1
f13+1
jwb+1
nirik+1
dwmw2wot no + ?
1
notting+1
f13dwmw2: ?
dgilmore+1
dwmw2why are we all saying '1' instead of '+1'?
jwbdwmw2, we aren't
caillonwe aren't
f13dwmw2: we all said +1
caillonyour client is buggered
f13dwmw2: your irc client is busted.
jwbyou said 1
bpeppleok, I see eleven '+1', so denis has been approved to be a sponsor.
jwbyay for denis!
--- bpepple has changed the topic to: FESCo-Meeting -- sponsor nominations - Jef Spaleta (spoleeba)
bpepple+1
caillon+1
jwb-1000000
just kidding
+1
notting+1
c4chris+1
f13+1
* dwmw2 tries to work out whether it's xchat or FESCo who are fucking with his head
dwmw2+1
caillon(that's a total of -999999 from jwb)
dgilmoredwmw2: its xchat
c4chrisor there is a conspiration :)
tibbs+1
jwbcaillon, heh
* nirik wishes he could run xchat. :)
dwmw2 /capab identify-msg fixed it :)
nirik+1
bpeppleok, I see see eight '+1' to Jef request, so he to has been approved.
--- bpepple has changed the topic to: FESCo-Meeting -- sponsor nominations - Douglas Warner (silfreed)
nirikI would like to see more (or any) reviews from Douglas...
nottingdwmw2: in this specific case, xchat
dwmw2notting: :)
c4chrisnirik: agreed
nottingno reviews == -1 for now
bpepple-1, I would like see some more reviews also.
dwmw2who's Douglas?
caillon0
nirik-1 for now, come back after some reviews...
jwb0
dgilmore0
c4chris-1
f13-1
nirikHe's a good guy, maintains packages nicely, but hasn't done much in the way of reviewing.
jwbas an aside, we should really start separating sponsorship from reviews at some point
caillonjwb, hear hear!
tibbsI have to agree with nirik.  I mean, reviews aren't the only thing we go by, but I don't recall being given much to go by except the packages he maintains.
jwb: Yes, but what other criteria do we have?
bpepplejwb: agreed.  some of the desktop folks don't do reviews,  but they know the packaging guidelines.
niriksurely, but will need to figure out other criteria
dgilmoresponoring you need to see how well they work with others in the community
jwbbpepple, sponsorship should be more than just packaging, but that's a separate discussion :)
bpeppleok, going back to douglas's request..
nirikif he is sponsoring people based on reviews, he needs to know how to do them...  if we are sponsoring based on other things sponsors should know how to do those too. ;)
tibbsWell, we need to have that discussion.  Just not during this meeting.
dwmw2we don't have enough people actively doing reviews. I don't see any particular problem making people do reviews before they're allowed to {become a sponsor,submit new packages,eat their own lunch}
jwbtibbs, right
bpeppleI'll send him a message that we would like to see him do some reviews, and come back to us in a few weeks.
abadger1999Isn't sponsorship primarily useful for reviewers?
nirikabadger1999: right now, yeah.
bpeppleabadger1999: mostly.
jwbabadger1999, at the moment.  which leaves other classes of contributions out
abadger1999Oh.  Sorry.  I think I misread that.  sponsor vs sponsorship.
jwbwhich is a problem
tibbsWell, we're specifically talking about cvsextras sponsors here.
* bpepple thinks we might be going off into the weeds. This might be better to discuss on the mailing lists.
jwbyes
dwmw2bpepple: you're probably right
c4chrisyup, let's move on for now
bpepplealright, let's move on.
--- bpepple has changed the topic to: FESCo-Meeting -- Proposal to block old version of automake & autoconf - Karsten Hopp
caillon-1000000
Kick__I'd like to cancel this proposal as it is obvious after that discussion on the list that we can't do that. I'd rather work with upstream of the affected packages to move them to more recent autofoo and keep the old compat stuff for development outside Fedora. Maybe have reviewers check if old autofoo is really required, but that's it
dwmw2I don't really see the point in removing things which people are using
caillonbpepple, btw, what happened to mclasen?
nirikKick__: fair enough... do you just not want to maintain the older ones? perhaps find others interested in that if you don't want to?
bpepplecaillon: I never got any links to some examples of his work, so I could send out his request to the mailing list.
dwmw2I just find it amusing that autocrap itself is one of the main problems for those who can't just write proper Makefiles :)
caillonbpepple, i bet you're using an example of his work right now...
f13Kick__: yeah, you could orphan the older ones and let somebody else maintain them
jwbcaillon, people have to be motivated enough to actually want to do that
Kick__nirik:  maintaining isn't the problem, having developers still use them for new projects is the problem
caillonjwb, to do what?
nirikKick__: or perhaps we could rename them? compat-autofoo / ancient-autofoo, etc... ?
Kick__;-)
nirikanyhow, more discussion on list, etc... move on?
jwbcaillon, gather up examples of their work.  mclausen is excellent, we all know it.  but he's not special enough to ignore what everyone else has to do...
Kick__ok
caillon(I didn't have to do it when I became a sponsor)
tibbsIf someone wants to start up a project to see about porting some packages off of old autotools, I'd certainly support it.
bpepplecaillon: I agree, it seems sorta pointless to ask mclasen for examples, but other FESCo members felt we shouldn't give him an exception.
jwbcaillon, me either.  i think that's a failure
* dwmw2 doesn't even know whether he's a sponsor or not.
jwbbpepple, i just think exceptions are a slipper slope.  then you have to deal with people saying "but he didn't have to do this.."
jwbslippery
caillonhttp://koji.fedoraproject.org/koji/userinfo?userID=78
his work ^^
dgilmoredwmw2: you dont have those keys
bpepplecaillon: I'm fine with using that, since mclasen most likely doesn't have time for package reviews.
caillonactually, he even does those.
https://bugzilla.redhat.com/show_bug.cgi?id=442693 for one example
buggbotBug 442693: medium, low, ---, Nobody's working on this, feel free to take it, NEW , Review Request: geoclue - Geoclue is a modular geoinformation service
caillonand why he should be a sponsor
dgilmorecaillon: no one is working on that  ;)
tibbsLooks to me like he should assign that ticket to himself.
It's one of the 37 reviews that our good friend Nobody has done.
bpepplecaillon: I'll send out an e-mail about mclasen after the meeting, and we can vote on him next week. does that work?
c4chrisfair enough
caillonthat's another week our anxious contributor doesn't get to do fedora work
we're not screwing mclasen here...
we're screwing out contributors
jwbwhich anxious contributor?
tibbsHonestly I don't care who supplies the info about mclasen, but I personally don't know much about what he has done so I have nothing to go on when deciding how to vote.
caillonthe one submitting the review i just posted
* dgilmore thinks that mclasen will make an excellent sponsor, he works well with others and does good work
jwbthat can be dealt with in 30 seconds
bpeppledgilmore: agreed,
caillonjwb, if we deal with it in 30 seconds, that's a wrong way to do so.  the person willing to walk through with them when they have problems should be the sponsor
jwbsponsor-by-proxy has been done before
caillonsponsoring people we aren't going to interface with is an insult to the sponsorship process, really.
nirikI can step up and sponsor that guy now if you like, and we can vote on mclasen next week?
jwbcaillon, yes.  which is part of the larger problem, given that there is no defined responsibilities for sponsors
caillonbut it's not the first failure in the sponsorship process we have
nirik, i can too. the point was that mclasen should be.
jwbagreed.  nor is it quite as big a deal as  "hurting our users"
cailloni think hurting our contributors is more of a big deal than hurting our users
jwbsorry, meant contributors
nirikcaillon: sure, as long as he has time to do so...
tibbsSo where is the actual failure here?  We asked for info on mclasen before making a decision and we didn't get it.
cailloni never saw that request.  so it's apparently my failure since i proposed him and i saw bpepple say he added him to the docket.
nirikI suppose we could vote now, and those with enough info could vote and those without could abstain?
dwmw2+1
dgilmore+1
bpepple+1
caillon+1
tibbs0
notting+1
jwb0
nirik+1 (I haven't seen a ton of reviews, but the ones I have seem good)
jwbc4chris, ?
bpeppleok, I see six '+1', and two '0'.
c4chris+1
jwb7
tibbsSo this wasn't passed by the other sponsors, right?
f13+1
jwbtibbs, correct
bpeppleok, we've approved mclasen to be a sponsor.
nottinghis reviews are at http://tinyurl.com/4dkqlk
f13I know that mclausen comes and talks to me or others when he encounters something that doesn't make sense or he doesn't understand.
notting(fwiw(
tibbsWhen we did an identical vote last week we corrected ourselves and went through that process.  I just wish we could stick to one actual procedure.
cailloni'll also note that http://koji.fedoraproject.org/koji/tasksbyuser shows him as the 4th highest builder behind kojira, jkeating (who did the mass rebuilds), and lmacken (due to bodhi stuff)
jwbtibbs, yes that was sort of my point
caillonso if he's doing stuff wrong, we have a much bigger problem.
abadger1999caillon: Reviews are much more important than builds to be a sponsor.
jwbabadger1999, for now
caillonthat's what the process says
tibbsBut honestly I have no problem switching to a "get it done" attitude if that's what we want.
caillonthat's not really true.
nottingabadger1999: ok, then he has the third most reviews of any non-sponsor, currently  (according to package status)
abadger1999notting: That is very relevant and important :-)
tibbsAnyway, it's past the half-hour mark; what's left?
caillonnotting, thanks for the quick stat
f13getting trashed for F9
--- bpepple has changed the topic to: FESCo meeting -- http://fedoraproject.org/wiki/Extras/Schedule/MaintainerResponsibilityPolicy -- bpepple
tibbsGlad to see my old draft getting some love.
nottingcaillon: i just read the page. c4chris does all the hard work behind PackageStatus
* c4chris blushes :)
bpeppletibbs: yeah, I had some free time, figure I would dust that off.
cailloni didn't realize that stat was there
tibbscaillon: It's getting harder to see through all of the mega-reviewers, but there's still some useful bits to be gleaned.
jwbbpepple, i like what you have hear
jwber, here
* nirik likes it... wonder if anything is missing... seems a bit light for some reason.
nottingit's a good start, and might as well go with it and add from there
dwmw2yeah, it looks good to me
tibbs+1 the maintainer responsibility document.  Obviously we'll need to adjust it going forward, but what's there is good.
dwmw2+1
nottingthe 'deal with reported bugs' may not apply equally to all packages. the kernel is still doomed.
tibbsIs there an easy way for a maintainer to know what depends on them?
c4chris+1
caillon+1 with knurd's addition
nirik+1 overall (we can always add)
nottingtibbs: repoquery --whatrequires --alldeps, for both source and binary repos
bpepple+1 here also. ;)
* c4chris afk a few minutes...
nirikhow about something about interfacing with upstream? not sure to what extent, but might be nice to mention that communication with upstream is important.
dgilmore+1
notting+1
jeremy+1
bpeppleok, I count nine '+1', so this has been approved.
And as mentioned we can add to it, as needed in the future.
nottingi suspect knurd's addition would be 'Maintain stability for other developers'. with a large 'don't change the ABI in a released version, please' entry
dwmw2we should add something on making sure we merge patches upstream as well as possible
bpeppledwmw2: not a bad idea.
dgilmoredwmw2, nirik: yep
dwmw2with the obvious corollary that we should only really be shipping patches which are suitable for upstream in the first place, as far as possible
tibbsWho moderates fedora-devel-announce?  Do they know to expect change notices like this?
dgilmoretibbs: f13 does it
f13I moderate it, and can extend said moderation to more people
something I want to do after F9 is consolidate the various -annoucne list passwords and get a group of moderators
bpepplef13: sounds like a good idea.
* c4chris is back
bpeppledwmw2: I'll add your suggestion, when I add it to the wiki. thanks.
unless there is anything else, we can probably move on.
--- bpepple has changed the topic to: FESCo-Meeting -- F9 - any issues needing discussion? - all
caillonF9: are we there yet?
bpeppleanything we need to discuss about F9, or are we in celebration mode?
nottingbarring something really bad exploding under testing in the next couple of hours, F9 is done.
f13caillon: QA has a few more hours to hit the big red button
bpepplenotting: sweet.
f13that means rawhide will see F10 content early next week
jwbwe've secretly disconnected the big red button and QA doesn't know it though
jeremyjwb: shhhhh!
caillonjwb, that's ok.  QA is colorblind.
tibbsI downloaded F9 yesterday...  But someone fixed the permissions before I got the boot images.
nottingf13:  are we allowing syncing solely from tier0?
jwbsilence...
nottingso, in light of F9 being mostly done... get your stuff ready for F10 alpha! coming sooner than you think!
jeremyindeed!
next week's fesco topic -- f10 schedule ;-)
wwoodstwo months and a week!
jwbwhich means the F10 name game starts soon
bpeppleok, that's all I got on the schedule for this week. So,
c4chrisacid anyone ?
--- bpepple has changed the topic to: FESCo meeting -- Free discussion around Fedora
nirikso does F-10 get 2 weeks shorter due to f_9 slip? oh well, I guess we can discuss that later. ;)
wwoodsQA hasn't found anything big-red-button-worthy, and at this point I don't think we can find anything bad enough to require cancelling everything
dgilmorenirik: probably
wwoods: we revoked big red button powers last night
:)
bpeppleanything else folks want to discuss, or should we start to wrap up for this week?
wwoodsif someone was like "uh it makes all Gateway machines catch fire" I'd be looking for a workaround.
jwbwwoods, what about powerpc machines?
wwoodsI've tested that
they don't catch fire
jwbyou disappoint.  i was looking for a snarky comeback
c4chrisonly acrid smoke ?
f13so uh, I can't boot from mdraid 1
nottingf13: really?
f13yeah, really :/
jwbf13, you don't count
nirikjwb: aside from my video not working ppc has been doing really well with my testing.
nottingf13: lemme try
jwbnirik, mine as well
f13notting: key might be two actual disks
nirikjwb: r128?
jwbnirik, radeon 9600
which is r350 i think
dwmw2hm, my 9600 works fine (dual head and everything)
dgilmorejwb: whats the other user have?
wwoodsyeah, I think my G5 is 9600 as well
nirikwonder if it's fbdev thats broke
dgilmoreoh sorry dwmw2 forgot you were here :)
nirikhttps://bugzilla.redhat.com/show_bug.cgi?id=445419 is the r128 bug I filed
buggbotBug 445419: low, low, ---, Dave Airlie, ASSIGNED , r128 driver broken on ppc
dwmw2*rsap*
jwb: failure mode?
jwbdwmw2, er... 9600 doesn't do dual head...
dwmw2, i can't recall actually. it's been forever since i tried the actual radeon driver
dwmw2pmac /pmac/git/mtd-2.6 $ lspci | grep Radeon
0000:f0:10.0 VGA compatible controller: ATI Technologies Inc RV350 AP [Radeon 9600]
pmac /pmac/git/mtd-2.6 $ xrandr | grep ^DVI
DVI-1 connected 1920x1200+0+0 (normal left inverted right x axis y axis) 519mm x 324mm
DVI-0 connected 1600x1200+1920+0 (normal left inverted right x axis y axis) 360mm x 270mm
nirikanyhow, not red button worthy. :)
jwbSection "Device"
Identifier "Videocard0"
Driver "fbdev"
Option "UseFBDev" "true"
EndSection
that's what i use
and it works
dwmw2hm, so you're not using the radeon driver at all?
nirikdriver fbdev?
* notting thinks this is slightly off-topic for fesco -> #fedora-devel,maybe?
jwbmoving
* nirik nods
dwmw2bugzilla, even
nottingbpepple: anythign about elections we need to bring up?
bpepplenotting: not really.  I'm going to be sending out a reminder about nominations each week until the election.
if there's nothing else, I think we can wrap up.
* bpepple will end the meeting in 60
c4chriswrap++
* bpepple will end the meeting in 30
bpepple will end the meeting in 15
bpepple-- MARK -- Meeting End

Generated by irclog2html.py 2.5 by Marius Gedminas - find it at mg.pov.lt!