--- 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, dgilmore, dwmw2, jwb, notting, nirik, kick_, jds2001, j-rod
Hi everybody; who's around?
* hansg is present
* nirik is here
jds2001 semi-here
hansgQuestion (I know I'm a PITA), could the BetterWebCam support feature be discussed first, I don't have much time
bpepplehansg: I don't have a problem with that.
dwmw2oh, that's what was going to be happening on Wednesday evening
bpeppleLet's wait a few more minutes for some more FESCo members to show up, then we can start with that.
bpeppleok, I see 6 FESCo members so we can probably get started.
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- http://fedoraproject.org/wiki/Features/BetterWebcamSupport - all
bpeppleAnyone have any questions about this feature for hansg?
jwbhansg, the kernel patches are all upstream now, right?
or are there additional ones that might need to be carried?
dwmw2this looks good to me
gspca is upstream; there's more to be merged
jds2001i read it as there are additional ones, but so long as there's a clear path upstream, i'm not inherently opposed to that.
nirikhansg: are there other supported models you don't have/can't test with? adding them and asking for testers with that hardware might be good.
* dgilmore is here
dwmw2we will defer to davej and his minions, of course, but it seems reasonable enough to merge new drivers if they're going upstream imminently.
* nirik thinks this looks great. Should dust off some old webcams and test it.
jds2001btw, great job on the test matrix :)
hansggspca which was my main target for F-10 is upstream now, other drivers will get cleaned up and merged upstream over time as time and hardware access allows
notting+1 from me, although the LD_PRELOAD hack is kind of meh
jwbi have no inherent opposition to this.  just trying to make sure whatever patches are required get where they need to be ASAP
bpepple+1 to this feature.
dwmw2hm, LD_PRELOAD is just for binary-only crap, isn't it? Not for real software?
hansgnirik there are tons of other models support by gspca, see: http://mxhaard.free.fr/spca5xx.html
nirik+1 here too.
jwbdwmw2, that's how i read it
hansgToo much to be able to try and get them all tested
dwmw2+1 then
nirikhansg: might add that link and note testers are welcome?
dgilmorehansg: h[B[B[B[B[B[B[B[B[Bhas
hansgdwmw2, yes
jds2001hansg: yeah, but getting a few more tested would be good.
by no means do we have the resources/capacity to test every webcam that exists :)
hansgI'm currently focussing on getting the necessary support (for conversion from custom cam formats to normal rgb format) in userspace
ekiga patches have been accepted upstream, gstreamer patch is submitted upstream, no feedback sofar
* dgilmore is ok with it
niriksure, but if people have those cameras, they could join in and test.
jds2001nirik: yeppers
* dgilmore has a camera to test
hansgMy plan for getting more testing is to announce this on planet and fedora-devel and ask for testing once userspace support is in place
* nirik has some old cameras... one is hopeless, but the other might work
nirikhansg: sounds good.
* dwmw2 looks around to see what weird shit he can plug into his PS3
dgilmorehansg: this replaces the old spca50x driver that was always outside of the kernel tree?
jwbdwmw2, you have to fix the kernel first
jds2001you could also plead on f-t-l
hansgdgilmore, yes
jds2001, good idea will do
dgilmorehansg: cool
hansgAbout people having old cams, if you don't use them I very much welcome donations of cams. Drop me a mail for my snail mail address
nirikthere is some ps3/sony camera...
* nirik doesn't have one.
dgilmore has a few quickcam 5000's and an old intel one
hansgI've solicited for this before but so far no donations from outside the Netherlands (one Dutch guy mailed me a box with 5 cams)
jds2001pretty expensive to ship overseas....
nirikI have a linksys one you are welcome to... but it's unlikely to ever work with linux... it needs IE to view steams from it.
hansgjds2001, probably I have no idea
herlohansg: I have one, but its kinda broken
jwbshould we move on?
nirikanyhow, are we at enough + to pass this?
hansgherlo, well if the hw is broken its of no use to me
dgilmorehansg: i might s endyou my intel one
herlohansg: no, the cam works, just missing the stand...
hansgherlo it would be welcome then :)
herlohansg: I'll see what I can do
hansgdgilmore, I already have one intel, see the feature page, if it isn't that one I would be very happy with it
hansgAnyways any more questions about the feature and the plan to complete it?
bpeppleok, I'm back.  stupid coffee shop wifi. :(
* dgilmore says +1 to the feature
jwbbpepple, we all passed it
bpepplegreat. anything else or should we move on?
dgilmoremove on
--- bpepple has changed the topic to: FESCo-Meeting -- http://fedoraproject.org/wiki/RahulSundaram/CollectiveMaintenance -- all
bpepplemether wanted this added to the agenda.
jds2001this looks reasonable, but who enacts this procedure?
it also seems a little cumbersome
bpeppleI'm a little reluctant, since it seems to add unnecessary bureaucracy to the process.
* jds2001 lost on the point of the tracker bug
nottingwe only have one feature today?
* nirik isn't sure this solves the problem...
jwbwe started with webcams due to a request from hansg
bpeppleespecially since the maintainer containment should be added soon.
nirikthe problem is that some maintainers are overworked. Adding more work/bugs/nagmail to them doesn't seem like it would help.
jwbi have no idea how this is going to work?
jds2001i think webcam is the only one
nirik: exactly
bpepplenotting: yeah, we only had one since poelcat is on vacation. I looked over the list, and that seemed the only one ready for today.
jds2001but frankly they should seek co-maintainers if they're overworked
nirikwe should be encouraging them to get comaintainers or find new maintainers for the packages they don't have time for... or comaintainers to handle easy bugs when they work on harder ones, etc.
nottingbpepple: i can't help but think we're running behind there
dgilmorei say we just let maintainers fix bugs
i tthiink adds to much overhead
* dgilmore hates X right now
bpepplenotting: I don't disagree with you.
here's what's in the proposed queue. Most of them seem to be spins. http://fedoraproject.org/wiki/CategoryProposedFedora10
dgilmorenirik: thats the current recomendation AFAIK
jds2001what we're talking about here is a patch provided, or something trivial to fix
if i understand correctly
* nirik would say -1 to this for now, lets see how the newmaintainer containment and opening acls to packagers helps first, and if not, we can revisit something along these lines after that.
jds2001the patch SHOULD be reviewed and ACK'ed or NAK'ed
* bpepple agrees with nirik.
dgilmorenirik: indeed
notting-1 from me
dwmw2hm, I like it. It encourages teamwork.
jds2001-1, because as i mentioned this process is too complex. I like the idea, though
jwbno it doesn't
dgilmoredwmw2: so does the comainter policy we are implementing
nirikdwmw2: I think we should encourage teamwork, but we shouldn't legislate it.
jds2001how not
jwbwell, it does but in a very awkward way
LyosNorezeldwmw2: there are many better ways to encourage teamwork... without introducing another stick to beat people with
dwmw2yeah, I agree that it's awkward/complex
yeah, fair enough
bpepplealright, I see five '-1', so this proposal has been rejected for now.
dgilmoredwmw2: i think this is an akward subset of whats already been worked on
bpeppledgilmore: agreed.
dwmw2yeah, ok.
bpeppleanyone have anything else to add? otherwise we can move on.
nirikspeaking of newmaintainer containment/acl opening... is there any status on that?
bpepplenirik: I believe sadmac is still working on our changes.
--- bpepple has changed the topic to: FESCo-Meeting -- Clarification of spins as features - http://fedoraproject.org/wiki/Extras/SteeringCommittee/Meeting-20071206 - bpepple, all
* jds2001 tested the group rename worked and i could cimmit
jds2001commit even
bpeppleok, I looked at the features in the queue yesterday, and noticed most of them seem to be spins.
nirikbpepple: yeah.
bpepplewe decided back in December that we didn't consider spins to be a feature.  Has something changed?
nirikhttp://bpepple.fedorapeople.org/fesco/FESCo-2007-12-06.html is the meeting
I think the thought there was that since the board had to approve spins, why should fesco also have to do so.
dgilmorenirik: spins is all the board
* jds2001 read that meeting
nottingthe spin process has been through enough revisions to make my head, well, ...
jds2001same sentiment - -1 to feature, +1 to pimp em
jwbwait wait wait
nottingi distinctly remember a discussion of why the needed to be tracked as spins, but i don't recall whether that was in fesco, the board, rel-eng, or the spins sig
jwbthere are reasons Rel-Eng wants the spins treated as features
namely, we want the same sort of information (scope, test plan, etc) applied to spins that we do features
jwbwe also want Spins to be treated as a _release_ item
jds2001but does it have to be the same process?
ldimaggi_wwoods, ping
dgilmorejwb: thats the spins-sig and releng's job
jwbe.g. they follow the same release process that Features do
dgilmoreboard says ack/nak
ldimaggi_wwoods, any sign of Slick these days?
dgilmorefesco does nothing with spins
jwbdgilmore, rel-eng is asking fesco to do something.  if we have to go and duplicate the whole feature process for spins, that seems pretty wasteful
dgilmorejwb: what does releng want?
nirikjwb: so spins have to pass thru the spins sig, then the board, then back to fesco, then to rel-eng?
nirikcan we also pass them thru ambassadors and bugzappers so they get the grand tour? :)
jwbwe want Spins treated as Features.  if fesco doesn't want to ack them, then fine.  but we're still going to require a feature writeup
jwbif a spin isn't ready by Feature Freeze, it isn't going to get done
nottingall those steps are different items. board is simply for trademark usage, for example
* jds2001 feels they should make a stop in QA
jwbjds2001, just like features...
dgilmorenotting: its too many steps.
jwbit's really not
jds2001though QA has really delegated that to the spins sig for spins, and pointed them to our release testplans
and encourage/help them to add their own
jwbdgilmore, this is not just rel-eng.  spins sig also sees value in doing this
think of it as a mechanism to sync the spins with the rest of the distro during release
jwbif fesco isn't the reviewing body, someone needs to be
nottingjwb: just clarifying - does spins sig and rel-eng see a need for anything from fesco besides just the tracking *infrastructure* (as opposed to yea/nay votes)
* nirik doesn't care too much... I guess we can review them along with the rest of features... not sure what we will need to look for tho
dgilmorejwb: i have no problem in pimping spins in the release
dgilmorejwb: i just dont think fesco needs to approve them
jwbnotting, i don't believe so
jwbdgilmore, fesco is the only body with enough clout to declare something as "not going to make it this release"
nottingif that's the case, what do we need to discuss here, other than just note that spins will use the same feature reporting infrastructure, with feature wrangler pushing those features to the spins sig, etc. for approval?
jwbthat's what we need to discuss
nottingif that's what you want, i'm fine with that
jds2001if you're not looking for ack/nack, and just process - then that's fine in my book
dgilmorejwb: if the spins sig or releng comes to fesco withhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhhh i isssue with a spin then we can look at it then
jwbproposal: rel-eng has final say in acking spins for a release based on the Feature criteria.  spins sig/owner can escalate to fesco if needed
dgilmorejwb: +1 from me
bpepplejwb: +1
jds2001err, 1
jwbi'll abstain officially, though i'm obviously +1
i see 5 votes
bpeppleok, I see six '+1' to jwb's proposal, so it's been passed.
* bpepple sees jwb beat him to the punch.
nirikok, move on?
--- bpepple has changed the topic to: FESCo meeting -- Free discussion around Fedora
* nirik has an item...
bpeppleok, that was it for what was on the schedule for today.
nirik: floors yours.
* dgilmore will probably not be available for next weeks meeting
nirikShould folks in fedorabugs be able to review/approve package reviews for already sponsored packagers? I thought we required they be packagers already to do that.
dgilmorenirik: no
* jds2001 thought so too
nirikbut spot / the package review webpage says only fedorabugs needed.
dgilmorenirik: they should have a package in fedora first
jds2001technically that's true
however, policy-wise it's not.
nottingi thought the algorithm was that for approved packagers, you need to be a packager. for non-approved packagers, you need to be a sponsor.
jds2001that's been a fear of mine.
notting: my understanding as well.
niriknotting: that was my thought as well, but the wiki/spot do not think so... wanted to clarify.
dgilmorenotting: as did I,  however the wiki reads differently
bpepplenotting: +1
nottinganyone want to 'fix' the wiki/
* nirik was hoping to round up spot to provide his input, but he doesn't seem to be around.
bpepplenotting: doesn't it need to be someone on the packaging committee?
* jds2001 gives 'fedorabugs' freely - I don't want folks in *only* fedorabugs reviewing stuff.
dgilmorebpepple: i think so
bpeppleI thought those pages where locked down.
dgilmorejds2001: i think when it was set as such.  we only gave fedorabugs  to people who had a package in fedora
bpeppleIs anyone here on the packaging committee? If not, I'll contact spot or tibbs.
nirikProposal: FESCo clarifies that you must be in the packager group to review and approve package reviews of already sponsored submitters
bpepplenirik: +1
nirikand if they want to change it back to just fedorabugs, we can revisit.
bpeppleok, that's seven "+1" to nirik's proposal.
abadger1999I'm on FPC.  That was my understanding as well.
jds2001fedorabugs is necessary for triage.
bpeppleabadger1999: could you update the wiki?
abadger1999yeah.  Link?
jds2001perhaps we make a new bz group that 'packager' gives you that lets you set fedora-review
nirikabadger1999: https://fedoraproject.org/wiki/Package_Review_Process
nottingoh feh. so it's a technical thing where we don't have another ACL mechanism in bz?
nirikjds2001: that would be nice, but wouldn't totally solve the issue.
jds2001we can easily make one
nirik: why not?
* nirik notes that there have been cases where people not in the group did reviews, said they couldn't set the flag, and someone helpfully set it for them...
jds2001you dont do cvsadmin stuff without fedora-review+, right?
abadger1999Err.. one thing, anyone can review... only a packager can approve.
niriksure, it helps make it more obvious tho
abadger1999: right.
bpeppleabadger1999: correct.
abadger1999k.  Changing
dgilmoreabadger1999: yes.  we want/ed new packagers  to work on reviews first
* nirik notes he is working with the recent person who did a bunch of reviews without being in fedorabugs/packager. Hopefully he will be sponsored by me soon...
dgilmorenirik: :)
bpeppleok, is there anything else folks want to discuss, or should we wrap up for the day?
nottingdo we want to do the non-spin features?
btw, i'll miss the next two meetings
dgilmorenotting: and I likely the next
bpepplenotting: we can, but they seemed to be missing stuff.  Like docs, qa stuff, etc.
* dgilmore starts new job Monday
bpeppledgilmore: congrats on that.
dgilmorebpepple: thanks
jds2001congrats dgilmore :)
nottingbpepple: if they're in proposedfedora10, aren't they by definition ready? or only if they get the featurewrangler magic stamp of approval?
nirikdgilmore: hopefully your new company will let you work on fedora some in your spare time. ;)
dgilmorenirik: i would hope so
at least one goal is to get the product in fedora
bpepplenotting: poelcat tends to only bring them up if everything is complete, but I have no problem looking at them if you want.
dgilmorenirik: i should get to work on fedora some on work time also
bpepplewe've got about 13 minutes left, so how about we try to look at 1 or 2 before wrapping up for today.
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- http://fedoraproject.org/wiki/Features/BetterStartup
nirikbpepple: sounds fine to me.
nottingbpepple: the worst that can happen is we'll have to do them again
jwbi talked to airlied about this one last night a bit
bpepplenotting: agreed.
jwbthe kernel drivers aren't quite there.  intel will be first, then ATI
dgilmorejwb: want to fill us in
jwbi don't see any reason to _not_ approve this, but one has to wonder if all the bits will be in place in time for beta
nottingactually, the policy says that the feature wrangler is supposed to raise the feature at the meeting. so technically we shouldn't do this
* nirik hopes this lands at least somewhat soon... needs lots of testing I suspect.
jwbnirik, right
nottingplymouth's in rawhide. the modesetting bits are not
jwbnotting, correct
* nirik tried intel mode setting in f9... it works, but doesn't survive suspend/resume at all.
niriknotting: fine, we could adjorn... or just look at them "informally" without voting?
bpepple+1 to this feature, but the documentation & release notes needs to be completed.
jwbbpepple, i think the kernel drivers should come first.. :)
nottingnirik: i suppose by the rules of the process we should adjourn
bpepplenotting: yeah, that might be for the best.
sticksterSeriously?  :-\
* nirik is looking forward to the SaveToBugzilla feature. thats pretty darn cool.
jwbone could always spend the time asking questions of the Feature itself
* jds2001 too
bpepplestickster: most of the feature pages left aren't complete, not to mention I'm a little leery on stepping on poelcat's toes.
sticksterbpepple: I'm pretty sure he'd be happy if it progressed -- but incomplete is another matter then.
* stickster backs out humbly
bpepplestickster: the only one that was complete was the webcam feature which we approved at the beginning of the meeting.
sticksteryeah, saw that one -- fair enough.
* stickster tips hat to FESCO in thanks
bpeppleDoes anyone have anything else to discuss?  Otherwise we can wrap up, so the next meeting isn't late.
* nirik has nothing.
* bpepple will end the meeting in 60
bpepple will end the meeting in 30
* bpepple will end the meeting in 15
bpepple-- MARK -- Meeting End
Thanks, everyone!
abadger1999nirik: I recall why the PackageReview page was written that way but it's obsolete now.
It was before cvsextras automatically granted fedorabugs.
bpepplebtw, does anyone know when poelcat is back from vacation?
nirikabadger1999: ah ha. Makes sense.

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