FESCo-2007-10-25

--- bpepple has changed the topic to: FESCo meeting -- Meeting rules at http://www.fedoraproject.org/wiki/Extras/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 is here
dgilmore is here
warrenmeh
f13here
* notting is here
bpepplef13: was the answer to mdomsch question in the wiki, or is that something we still need to add?
* tibbs here
nottingabout when to stop adding new packages for a release?
bpepplenotting: correct.
* spot is here
bpeppleI thought we had already decided that, but I wasn't sure if it is on the wiki.
f13bpepple: i thought we had already decided that, but I didn't get around to searching the minites
f13minutes
bpepplef13: ok, I do a search after the meeting, and update the wiki.
* abadger1999 watches and learns
poelcat here
--- bpepple has changed the topic to: Misc - Features - http://fedoraproject.org/wiki/Releases/FeatureNetworkManager - poelcat
bpepplepoelcat: you want an update for this correct?
poelcatbpepple: please :)
* f13 tries to get a dcbw online
caillonf13, i just phone him
here he is
f13wewt.
dcbwsorry, got sidetracked fixing evo bugs this morning
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
dcbw 5542 1.2 45.4 739512 470504 ? Tl 09:26 2:49 evolution --component=mail
f13dcbw: heh, we were looking for an update on the NetworkManager feature
dcbwf13: want me to update the page then?
bpeppledcbw: yeah, that would be great.
poelcatdcbw: and is there anything we need to know here?
dcbwpoelcat: we'll have feature parity with 0.6.x from F7; we won't have multiple device support for F8 gold but that is the next item on the list and can go in as an F8 update since the API won't have to change
nottingdcbw: are we going to have future abi ties/breaks like nm/nm-vpnc has had recently?
dcbwnotting: I don't want to rule that out, but maybe I'll just pad the public structures with some longs to be safe
notting: there may be one with multiple device support if something unexpected comes up
notting: but I don't forsee anything
notting: at some point we can break the hard dep between NM-gnome and NM itself
f13dcbw: at the point you split them at the srpm level right?
'cause otherwise, what's the point
dcbwf13: true; I'd like to break the applet out for F9 too
f13: well actually, even if they were in the same SRPM there's still a point
f13: evo could pull in NM-gnome but not have to pull in NM itself
f13: which is perfectly legal
f13ah
caillondcbw, what use would NM-gnome without NM be?
dcbwcaillon: it has the previous symbols from libnm_glib in it
caillon: which is what evo and a few others use rather than the D-Bus interface
caillon: those should be deprecated though
caillon: there's a pretty clear path forward at least
caillondcbw, anyway, isn't the plan to ship/enable NM by default?
warrenin F9?
dcbwcaillon: warren: yes
caillonyes
warrenI hope only if we have a list of pre-conditions
dcbwwarren: fine by me
warren: if there's a list then there's a goal to work towards and a good way to measure progress of fixing that list
f13warren: yes yes, that's the reason why it isn't default now.
bpeppleIs there anything else to discuss regarding NM?
poelcatdcbw: when you  update the NM page can you make it reflect feature complete for F8?
dcbwpoelcat: definitely
poelcatdcbw: thanks
bpeppledcbw: thanks for the quick update.
dcbwnp
bpeppleok, we can probably move on then.
--- bpepple has changed the topic to: FESCo meeting -- Free discussion around Fedora
tibbsmarek, didn't you have a feature to propose?
bpeppleThere's nothing else formally on the schedule, but is there any F8 items we need to discuss?
caillonwhat's the status of the blockers?
any change?
mether12 or so bugs left
caillonare they actual blockers that we will stop ship for if they still exist in two weeks?
or just "blockers"
metherwe reviewed them yest. what is left appears to be real blockers
f13caillon: each time I go through (as we get closer to release) I shave more off that aren't "blockers".
caillon: some we really need more information about them, they are potentially blockers.
caillonf13, okay.  are people working on them or do we need to bonk some heads?
i'm in vampyre country around halloween. i can summon up all kinds of hell :)
jeremythey seem to be moving along for the most part
caillonok
nottingjeremy: if we can 'fix' things in the anaconda upgrade remove list with packaging, would oyu consider that worth doing?
f13dcbw: what happens if you check vpnc connection to autoconnect?
dcbwf13: It should get ignored
marektibbs: houps, I thought we start in 30 minutes
jeremynotting: well, I'm probably not going to remove things from the upgrade remove list at this point...
nottingjeremy: right, but it would fix non-anaconda upgrades
dcbwf13: I don't want to bet too many $ on that though
marekhello everybody, indeed, I have a feature to propose
http://fedoraproject.org/wiki/Features/FedoraAstronomy
here, please
dcbwf13: I'll add to my check it out list
f13dcbw: ok.
jeremynotting: if we can fix it in packaging, it's definitely better to do so
dcbw: it does get ignored
marektibbs: (it says that the meeting is at 18h UTC during winter time)
jeremydcbw: I already tried :)
tibbsmarek: It's not winter time in the US.
notting...yet
marektibbs: hm, I misunderstood the page in that case, sorry.
tibbsWe're in that in-between time, unfortunately.
nottingmarek: seems fine as something to host on spins.fedoraproject.org, although i'd put any astronomy stuff in the 'engineering and scientific' group rather than creating a new one
mareknotting: ok, understood
methermarek: you might want to post to fedora-devel list so more people can participate. Do you have a wishlist?
mareknotting: hmm, after thinking now, you're right, it's better to group them under scientific group
mether: not official yet, but we are already working with Sergio on it -- yes, I will bring this to devel list, if you find my feature good
methermarek: i think we need more details to determine that
marekmether: http://fedoraproject.org/wiki/MarekMahut/Packaging -- this is a draft I drow few months ago, but it's only a 1/3 of packages I would like to include
mether: which kind of details? if it's the package list, I would prefere to start working on it, if it's approved
metherI just wanted to know if there are enough Free software packages to warrant a separate spin
nottingit's stuff that probably wouldn't ever be in the main fedora spin, i would think
metherthats likely but you need a few dozen packages in that category that we can show off
marekmether: yes there is -- I'm using a old knoppix CD I've build two years ago for my astronomy trips
marekmy estimation for that spin is about 20-30 packages
metherthe custom spin process is outlined in http://fedoraproject.org/wiki/Infrastructure/CustomSpins
bpepplepoelcat: when are we planning to start the features process for F9?
metheressentially you need a kickstart file with the complete list of packages and then send it to rel-eng and board for approval
poelcatbpepple: after F8, but we could start earlier
marekmether: yes, I read that but it's still a draft
methermarek: the one that we currently follow though
bpepplepoelcat: we might want to d do quick review of our process, and see what failed and what worked, and adjust accordingly.
warrenAre we still needed here?
marekmether: ok
metherone important feature that we didn't track in that list in system-config-firewall
marektibbs: what's the normal process? any votting or something like that?
poelcatbpepple: that is my plan
tibbsmarek: I don't think we're at the point of voting on F9 features yet.
poelcatbpepple: need to prepare for that a little in advance
bpepplepoelcat: cool.  In that case I think we should start the F9 feature process sometime soon after F8.
poelcat: agreed.
marektibbs: ok, when should I return? :)
caillonbpepple, can we do that next week?  that might take a while :)
tibbsI'm really not the one you should be asking.
marektibbs: who then?
f13bpepple: poelcat: yeah, something poelcat and I were talking about, getting an actually schedule layed down of when "planning" phases start for the next release
f13which would include FESCo approving features
bpepplecaillon: definitely.
caillonbpepple, (review what worked, what didn't, etc)
poelcati laid out of my initial thoughts here: http://poelcat.wordpress.com/2007/10/16/fedora-8-feature-retrospective/
tibbsFolks, is there anything marek needs to do regarding the spin he wants to make?
Or is there any approval he needs from us?
bpepplecaillon: I was thinking maybe we could do the dissection on the mailing list.
tibbsThe spin process indicates that he should talk to rel-eng.
cailloneven better
methertibbs, I dont think FESCo is involved at all
nottingtibbs: once he's got more concrete output, yes. but he's on the right track with what he's doing
marekthe FEL has been approved by FESCo, no?
f13dcbw: a question came up regarding killswitch support.  WHat is that actually supposed to do, that would be different from the module just disappearing out from under NM?
dcbwf13: huh?
f13dcbw: nm detection of killswitch stuff.
what does it actually /do/?
bpepplemarek: yes, it was approved for F8.  We haven't started looking at F9 features yet.
dcbwf13: brings down the network device when you turn off the switch
marekbpepple: so, does FESCo approve these features or not?
f13dcbw: and how is that different than if the module were rmmod'd behind nm's back?
dcbwf13: with a killswitch, the module is still loaded and nothing appears to have changed to NM (besides the IFF_RUNNING flag) so NM keeps going on as though you still have wireless
bpepplemarek: FESCo approves them.
dcbwf13: an rmmod looks like a hotplug
f13: a killswitch looks like _nothing_
marekbpepple: ok thank you, do you have any estimation when should I bring my feature request again?
f13dcbw: oh hrm... it looked like to me the killswitch made the module go byebye on my laptop
I should try it again.
dcbwf13: only for _bluetooth_ because most of those are USB-connected
f13: I've never seen one that hotplugs mini-PCI cards :)
bpepplemarek: not yet.  We'll send out an announcement when we're starting the feature process for F9.
f13dcbw: well, my kernel went all wonky when I last tried to killswitch the wifi.  I'll try later today
nottingmarek: bpepple: i guess the distinction is that not all spins are (or need to be) tracked as features
marekbpepple: ok thank you -- in meantime, I will work on my feature request.
dcbwf13: killswitches are either (a) kill the power amp but don't tell the chip, (b) toggle a GPIO on the chip which the driver can poll, (c) toggle an interrupt for the chip, (d) are handled by ACPI and have to polled by ACPI code, or (e) a "soft" input button
f13: each gets handled differently
bpepplenotting: agreed.
f13dcbw: cute
dcbwf13: no shit
f13: that's one of the reasons it's not better integrated yet
f13: for example, the dell utility for wireless control has to poke ACPI or nvram or something evil like that
bpeppleok, is there any other FESCo business we need to discuss, or should we call it quits for today?
mareknotting: ok, how should I know if my spin proposal should be considered as feature?
f13bpepple: next week I want to vote on the development changes proposal.
bpepple: so this is the last week to get input to me
caillonyeah
bpepplef13: ok, I'll add it to the schedule for next week.
poelcatf13: when is the IRC meeting before to discuss it?
nottingmarek: i'm not sure
f13poelcat: I think I'm just going to take over the FESCo meeting time and do it then.
poelcat: two birds, one stone.
caillonand dinner
bpepple;)
f13poelcat: I may not get to announcing that today, but please tkae that to those that need to know.
poelcatwill do
bpeppleok, anything else?
f13I'm good.
* bpepple will end the meeting in 60
f13caillon: so all that arguing over what compiz should do, and the 0.6.2 build doesn't do the cubespin, even with your new firefox.
* bpepple will end the meeting in 30
bpepple will end the meeting in 15
bpepple-- MARK -- Meeting End
Thanks, everyone!
jeremyf13: I noticed that (and thought I added it to the bug)
bpepplepoelcat: are you on the FESCo mailing list?  I was thinking of having the discussion regarding the feature process there instead of the devel list, since most folks wouldn't be interested in it.
metherbpepple: btw, shouldn't the FESCo list archives be open? It seems to be a invisible list
f13jeremy: maybe.  bugspam, noisy
bpepplemether: I think it was originally created to be a FESCo-only list so there could be candid discussions.  Probably makes sense to open it up, since it tends to be mainly used for fesco admin type discussions.
metherbpepple: then consider this as a request to do so
bpepplemether: I'm fine with that.  Let me check with the other FESCo members to make sure no one has a problem with it (not that I expect anyone to).

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