FESCo-2008-08-20

--- 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?
* Kick__ is here
dwmw2hi
* nirik is here.
dgilmore is not really here
dwmw2am I just completely fucking incompetent, or did this meeting change times again?
* dgilmore is a zombie now
jwbdwmw2, it changed times at OLS
* jwb is here
bpeppledwmw2: no, it's been the same since we decided on the new time & day a few weeks back.
jwbdwmw2, you were either sleeping or otherwise occupied for that meeting
dwmw2I'm going with the incompetence explanation.
I can take it.
bpeppledwmw2: ;)
ok, we should probably get started since there's a lot on the agenda today.
--- bpepple has changed the topic to: FESCo-Meeting -- sponsor nominations -- Dan HorĂ¡k
nirik+1
bpepple+1
Kick__+1
f13I'm lurking for schedule talk
bpepplejds2001: +1 on mailing list
bpeppledwmw2, jwb, dgilmore: any thoughts on Dan's request?
dwmw2I don't know Dan.
jwb+1
dgilmorebpepple: ive honestly not even looked at his work
bpeppledwmw2: My e-mail to the sponsors list had links to his work.
dwmw2he maintains the dovecot package, and cyrus
bpepplebut anyway, I believe we've got five '+1' to approving his request.
dwmw2+1 I suppose
bpeppleand I didn't hear any complaints about his request on the mailing list.  So, we've approved him.
ok, before going on to features, do we want to briefly discuss the schedule?
* dwmw2 takes that as a 'no'
bpepplealright, then features it is.
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/30SecondStartup
* nirik doesn't think we can discuss the schedule until the outage is over.
dwmw2+1
bpepplenirik: I'm fine with that.
+1 to 30 second feature, though it probably needs to update the info about upstart.
Kick__+1 for any boot speedups even if we won't get to 30sec without full upstart
nirik+1
bpepplenotes that jds2001 gave a 0 on the mailing list.
jwb, dgilmore: any thoughts on this?
jwbi'm with jds2001 on this one
i'd like to see the upstart status before declaring it a Feature
dgilmorebpepple: im going to have to bow out for today.  im really elsewhere right now and not looked at anything FESCo
bpeppledgilmore: np.
jwb: didn't mclasen say last week that this feature was primary about the readahead & auditd.
nirikbpepple: thats what I recall as well.
jwbhm
mclasenI said that
bpepplemclasen: good, I'm glad I wasn't putting words into your mouth. ;)
nirikmclasen: so how close is the 30 second goal? attainable?
(I guess so, since thats the feature name)
mclasennirik: I haven't seen any recent timings from harald
nirikok, anyhow, +1 from me... any improvement would be good.
mclasenbut afaik, no real improvements have been made, since the adaptive readahead got stuck between auditd and the kernel...
bpeppleok, I see four '+1', and two '0' for this feature.  Do we consider that approved, since it's less than 50% of fesco?
jwbcan i approve it with a "we might not call it 30 second startup if it's not really there" +1?
bpepplejwb: I'm fine with that.
jwbit's hard to declare it a Feature before we know what the improvement over F9 is
dwmw2that's part of the contingency plan
jwbso tentative +1 from me for now, but i'd like to get some measurements before RC to review
dwmw2it kind of raises questions about the process -- if we say 'no' as a feature, do people stop working on improving boot time?
* j-rod tardy... lost track of time...
jwbi doubt it
bpepplej-rod: np.
dwmw2so what is the _practical_ outcome of our decision?
bpeppledwmw2: no, work still continues, we just don't publicize it in our release notes.
anyway, we have five '+1', and one '0' to this feature. So we've approved it.
dwmw2so if we say no now, and they get it to a 10-second boot time, we'll _refuse_ to mention that in the release notes, whjen it comes to time to write them?
jwbi think it's more about readahead sucking it up and making things slower than F9
bpeppledwmw2: I would hope we wouldn't be so inflexible as to do that. ;)
f13perhaps to be determined on a case by case basis?
dwmw2bpepple: indeed. So our decision right now seems fairly meaningless :)
jwbFESCo: Bringing you meaningless decisions since FC6
jwb;)
bpepplejwb: ;)
ok, let's move on....
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/Provers
bpepplejds2001: voted that this was not a feature.
Kick__I think that's just a bunch of new packages and as the target audience looks very limited to me I'd say it's not really a feature we can advertise with in our release notes.
dwmw2agreed
nirikyeah.
dwmw2but that doesn't mean it's not a good thing
bpeppledwmw2: agreed.
Kick__right
* nirik sees no mention of a comps group... perhaps thats something they could consider... still doesn't make it more of a feature tho
bpeppleDoes anyone disagree that this doesn't meet the requirements for being a feature?  Otherwise, we can probably move on.
j-rodI'm trying really hard to reconsider this as a Feature... but I think I still say no. Its just a few new packages.
dwmw2indeed. -1
j-rodwith a very limited target audience
bpepplealright, let's move on to the next feature then.
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/SbinSanity
nirik+1
j-rod+1
nottingdidn't we vote on this a long time ago? or just discuss it ad nauseam?
dwmw2fuck yes
+1
notting+1 from me
Kick__-1
bpepple+1
jds2001: +1 on mailing list.
* nirik notes that we have a autoanwer thing on #fedora for this question. It fires many many many times a day.
j-rodKick__: care to expound on why you're saying no?
bpepplealright, I see six '+1', and one '-1' to this feature.
Kick__why do we keep  /sbin /usr/sbin anyway ?
j-rod(or do we just move on w/the six +1...)
bpeppleno, I would like to hear Kick_'s concern also.
dwmw2Kick__: because it would be too painful to remove them now :)
Kick__j-rod:  I'd rather cope with some easy questions about PATH problems than explaining why some programs in the PATH won't work as a normal user
dwmw2this way, people _can_ still remove /sbin/ from the user path if they really want ti
and when people type /usr/sbin/tcpdump it still works.
etc.
Kick__removing something from PATH isn't as easy as adding something
dwmw2we force people to type the full path at the moment; are you suggesting we should kill /sbin/ now, so that what we've been _forcing_ them to type in full is no longer working?
nirikthis solves a real support issue, and I don't think causes any... I doubt people are going to run things as their user that won't work, and if they do they will be less confused than not finding what they want to use.
dwmw2it's not particularly hard to remove something from the path
bpepplealright, so the vote looks to be six '+1', and one '-1'.  So this feature has been approved.
Kick__dwmw2:  not at all, but adding /sbin to every users PATH removes the destiction between /bbin and /sbin so there's no need for that anymore
j-rodlegacy
loads and loads of legacy
dwmw2including a metric shitload of scripts with hard-coded paths
the pain of moving stuff just isn't worth it.
let's just add /sbin/ to the path, which is what people have been doing for themselves for ages.
Kick__lets move on, I have no problem with being overruled
* nirik notes it looks like about 5/day for people in #fedora running into this issue. Sometimes more, sometimes less, but thats 5 confused users a day
bpepplemoving on....
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/VirtRemoteInstall
j-rodyeah, it'd be nice if you could do away with sbin and/or get everything really user-runnable into bin, but the effort just isn't worth it
* j-rod tardy with his comments too
bpepplejds2001: +1 to this feature on the mailing list.
Kick__+1
nirik+1 here, I assume it implys +1 to virtstorage as well.
bpepple+1
notting+1
j-rod+1
jwb+1
bpepplealright, I see six '+1', so we've approved this feature.
bpeppleanyone have anything they want to add before moving on?
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/VirtStorage
bpepplejds2001: +1 to this feature on the mailing list.
j-rod+1
nirik+1
bpepple+1
dwmw2+1
j-roda bit hard to separate this and virtremoteinstall, really
notting+1
bpepplej-rod: agreed.
jwb+1
Kick__+1 except that the test plan should make use of F10
bpepplealright, that's eight '+1', so we've approved this feature.
Kick_: I'll add that to the discussion page.
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/TimeZoneAndLocation
bpepplejds2001: voted 0 to this feature on the mailing list.
dwmw2would like to see more details
but it sounds like a good idea
nirikso, this is gnome only?
bpeppledwmw2: agreed.  consolidating the time setting seems like a good thing to me.
dwmw2yeah
that got a lot worse in F9
bpepplenirik: I believe so.  Does kde use gweather?
nirikno, I think it has it's own, as does Xfce.
dwmw2Would be nice to get the location from cellphones too.
bpepplewe probably need dan to add something about how this affects kde & xfce since this feature ties into anaconda.
mclasendwmw2: thats the geoclue integration, more or less, which we split off
since it won't make F1
0
dwmw2ok
nirikyeah, and it might be possible to integrate... not sure.
bpepples/anaconda/first boot/
spstarr_workFirstAidKit is neat, it means people who *gasp* install Xp for dual partitioning can have it repair the MBR :)
mclasenbpepple: how do you think using a different widget in firstboot will affect kde ?
spstarr_workunless anaconda 'rescue mode' is going away (or FirstAidKit will go into it?)
jeremybpepple: it probably shoudl tie in more with anaconda, to be honest
bpepplemclasen: I wasn't sure if it would effect kde, that was my question.
spstarr_workbpepple: not gweather :) there's two things, kweatherservice which is not pluggable, and the plasma weather engines, i wrote which you can load multiple data sources and pick.
jeremythen again, if s-c-date is changed, that will make anaconda pull it in.  but there's a dependency/interaction there which is not captured
mclasenbpepple: I don't think it will; of course, sharing the data would be useful...
j-rodfantastic test plan... :)
spstarr_workif someone wants to write a gweather backend that hooks into my engines then that would be nice
nirikit would be nice for the scope to be better defined... and at least some talk about other DE integration...
nirikcan we add comments and defer? or are we at the last feature approval meeting?
notting+1 from me
j-rodI'm definitely down with deferral, the scope of what's to be done just isn't clear
bpepplenirik: this was supposed to be the last feature approval meeting, but with the slip that might not be the case any longer.
+1 to this feature from me.
Kick__+1, but the test plan needs some changes as I certainly won't approve any expenses ;-)
bpeppleok, I see three '+1', and one '0', and a couple of requests to defer to next week on this feature.
nirikI guess a +1, and I will add comments to the discussion page.
dwmw2+1
j-rod+1 to the general idea, but needs clarification as to what is actually going to be done
dwmw2we might as well do it
bpepplealright, that's six '+1', and one '0', so we've approved this feature but we would like a little more clarification on this feature.
Could folks please add their questions for dan to the discussion page. thanks.
j-rodmost notably, the scope shouldn't be asking "what should we do?", it should be saying what its going to do...
bpeppleok, moving on if there's nothing else on this.
j-rodparticularly this late in the game
spstarr_workyou should only install any firstboot plugin for configuring that for GNOME only not KDE
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/KernelModesetting
notting+1
* Kick__ is worried about 35% completion
bpepplejds2001: +1 on the mailing list to this feature.
dwmw2+1
Kick__+1, though as it will make a visible difference
bpepple+1 here also.
j-rod+1
nirik+1
bpeppleok, I see seven '+1', so we've approved this feature.
anyone have any comments to add before moving on?
ok, onto the next feature...
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/NetBeans
dwmw2bpepple: "woohoo!"
dwmw2netbeans... is it a feature?
just another package, isn't it?
nottingdo we like beans? do we like george wendt?
markg85+1
bpeppledwmw2: I would argue it is since it's a fairly visible piece of software that we haven't had.
notting+1
bpepple+1
Kick__+1
nottingalthough. bleah... another ide
probably needs a comps group.
nirik+1 (also for epel? :)
bpepplejds2001: didn't think this met the requirements for a feature btw.
bpepplenirik: I'm not sure if Victor was planning to do this for epel or not, but I could check.
j-rod+1, if only for the marketing hype
bpepplealright, I see five '+1' and one '0' for this feature.  So we've approved it.
j-rodI still think new packages and updated packages aren't new features so much as things to be expected, but meh.
bpepplej-rod: yeah, I think it's a fine line for new packages, but on high profile packages I think it makes sense to publicize it.
anyway, we should move on...
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/HDTVEnhancements
victorvInitially for Fedora, but late for epel too... Thanks to ALL!
bpepple+1, though it needs some docs.
j-rod+1 for HDTV enhancements
Kick__+1
nirik+1
nottinghey, it saves the planet. +1 from me
do we have a list of other hdtvs that have extended blocks? or just attach and see what happens?
bpepplenotting: I think ajax had a small list included in the test plan.
nottingyeah, but didn't know if we should also test random other ones
bpeppleprobably wouldn't hurt.
jwbif it warrants sending me a new HDTV to test with, i'm for it
Kick__;-)
dwmw2+!
bpeppleok, I see five '+1' to this feature, so we've approved it.
j-rodhm, would the folks at best buy mind if we just went in there and tried hooking up to as many TV's as possible?...
drago01try it ;)
bpepplestickster: you mind if we do two more features before handing the room over to you?
j-rodunfortunately, I'm without a tv w/hdmi out...
s/tv/laptop/
sticksterbpepple: It's quaid's call, but I have no problem with it
bpepplequaid: you about?
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/MinimalInstall
sticksterProceed until you hear otherwise, I guess! :-)
bpepplestickster: works for me.
jeremythere's no one actually attached to doing work there
Kick__minimal installs been requested for ages, but we need someone to do it
nottingj-rod: oh, this only works with hdmi
bpeppleok, this feature seems more like a wish list than anything.  Based on the 0% completed.
notting?
j-rodnotting: dvi as well
drago01"0% (I'm no programmer) " ??
bpepplejeremy: you beat me to it.
j-rodnotting: not sure about vga
bpepple-1 to this feature since no one seems to be doing the work.
* quaid is here
notting-1
nirik-1
nottingj-rod: ok, so i need to buy a new computer *and* a bunch of TVs
kanaripisn't this "feature" more appropriately featured in the appliance-tools/AOS feature also?
quaid+1 we'll start Docs in #fedora-docs then move over, np ev4r
j-rodnotting: bingo. :)
Kick__-1
bpepplequaid: thanks, we shouldn't be much longer since we got just one more feature.
j-rod-1, better luck next time...
bpeppleok, I see five '-1', so we've rejected this feature.
j-rodwould be nice if someone really actually did it, but...
bpepplej-rod: agreed.
bpepplealright, onto the last feature for today.
--- bpepple has changed the topic to: FESCo-Meeting -- Features -- https://fedoraproject.org/wiki/Features/OnlineAccountsService
bpepple+1 here.
drago01maybe we should make it clear that this isn't a wishlist....
nottingis empathy going to tie into this?
also, 'woo, ancient metacity theme' :)
niriklooks interesting, but won't be really that exciting until a lot of apps use it.
bpepplenotting: I'm not sure, I didn't see any mention of empathy but I'm sure down the line it will since it's now the default GNOME IM.
right now it looks like this mainly ties into the online desktop sidebar.
* Kick__ doesn't like those keyword storage applications at all, it's way to easy to lose all accounts when you forget to back up the key ring before a reinstall
bpepplenotting: looks like empathy, f-spot, and twitux support is on the todo list.
bpeppledo people want to defer voting on this?
* nirik got a phone call.
nirikI guess I would say defer this and ask for what apps will be ready in time for f10?
bpepplenirik: that works for me.
j-rodthis is more or less just an extension of the gnome keyring to support non-local creds, right?
(plus the slightly scary thought of being able to store them on a gnome server somewhere)
nottingwell, it appears that online.gnome.org already does this if you ask it to
j-rodaha. well, as long as I can opt out of it, fine by me
seems to generally bring the gnome keyring closer to on par with apple's keyring thingy, esp. once more apps use it
+1 for me, but fine w/deferring too
bpepplehow about we add our questions about this to the discussion page, and defer on voting on this until next week?
Kick__ok with me
bpepplealright, we should wrap up for this week, since we've already eaten 15 minutes of the docs team meeting.
* bpepple will end the meeting in 60
bpepple will end the meeting in 30
bpepple will end the meeting in 15
* abadger1999 missed the Provers Feature :-(
bpepple-- MARK -- Meeting End

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