For minutes of previous meetings, please see DesktopTeam/Meeting.

Meeting Minutes

Present

Main Meeting

Eastern Edition

Apologies

Agenda

Discussion

Actions from this meeting

Outstanding actions from last meeting

Partner Update

Kubuntu Update

Mozilla Update

Transitioning from A3

Finish blueprints this week, then:

  1. Find and fix the worst bugs (how to find and track?)
    • - hoping that gravity and JSON searches help
  2. Ensure a smooth upgrade experience (how to test?)
    • - rickspencer3 to follow up with marjo to see if we can be more rigorous earlier
  3. Integrate the latest and greatest

Pitivi

How to define ready to ship?

Release Status

Other Business

Activity reports

Alberto Milone (tseliot)

Arne Goetje (ArneGoetje)

Bryce Harrington (bryce)

Chris Cheney (ccheney)

Didier Roche (didrocks)

Jonathan Riddell (Riddell)

done:

Ken VanDine (kenvandine)

Luke Yelavich (TheMuso)

Martin Pitt (pitti)

Assigned blueprints:

Non-blueprint work done:

Sponsoring:

Robert Ancell (robert_ancell)

Sebastien Bacher (seb128)

Till Kamppeter (tkamppeter)

IRC Logs

[16:31] * rickspencer3 taps gavel
[16:31] <didrocks> o/
[16:32] <tseliot> didrocks: just a sec
[16:32] <rickspencer3> https://wiki.ubuntu.com/DesktopTeam/Meeting/2010-02-16<<BR>> [16:32] <rickspencer3> oops
[16:32] <rickspencer3> I forgot "previous actions"
[16:32] <rickspencer3> an oversite which will save us some time
[16:32] <rickspencer3> ;)
[16:32] <rickspencer3> kenvandine, partner update?>
[16:32] <pitti> rickspencer3: they are on https://wiki.ubuntu.com/DesktopTeam/Meeting/2010-02-09<<BR>> [16:33] <tseliot> didrocks: done, you can edit the wiki now
[16:33] <tkamppeter> hi
[16:33] <bryceh> morning
[16:33] <kenvandine> sure
[16:33] <didrocks> tseliot: thanks :)
[16:33] <rickspencer3> ok
[16:33] <rickspencer3> previous actions seem done
[16:33] <rickspencer3> # Bryce to BP/WIify new work assigned on the sprint
[16:33] <rickspencer3> #
[16:33] <rickspencer3> Alberto to add missing alpha-3 WIs to foundations-lucid-boot-experience
[16:33] <kenvandine> i am still waiting on dbarth to update the weekly releases page
[16:33] <kenvandine> so that should happent oday
[16:34] <kenvandine> today
[16:34] <bryceh> rickspencer3, done
[16:34] <rickspencer3> bryceh, tseliot did you guys get those?
[16:34] <tseliot> rickspencer3: yep
[16:34] <pitti> rickspencer3: I think I'm done with the alpha-3 BP mangling
[16:34] <rickspencer3> (sorry kenvandine)
[16:34] <pitti> rickspencer3: except for one which I'd like to discuss in the meeting
[16:34] <rickspencer3> ok, so now ...
[16:34] <rickspencer3> kenvandine, partner update?
[16:34] <kenvandine> OLS is rocking the desktop, libubuntuone and rhythmbox-ubuntuone-music-store just had releases
[16:34] <kenvandine> i will get the packages ready for review later today for lucid
[16:34] <pitti> we need to rush them through NEW, etc.
[16:35] <pitti> FF in two days
[16:35] <kenvandine> the service isn't quite there, but damn close
[16:35] <kenvandine> pitti, yeah.. hopefully today :)
[16:35] <kenvandine> i have been maintaining the packages locally :)
[16:35] <rickspencer3> pitti, does someone need to be ready to do the NEW rush?
[16:35] <kenvandine> so shouldn't take me long
[16:35] <pitti> rickspencer3: a prod to whoever's archive day is is prudent, I think
[16:35] <pitti> (I'll be away later tonight, sorry; but I can have a look tomorrow morning)
[16:36] <rickspencer3> pitti, ok
[16:36] <Riddell> that'll be me today
[16:36] <kenvandine> pitti, it will be ready for you in the AM :)
[16:36] <kenvandine> or Riddell later today :)
[16:36] <rickspencer3> kenvandine, all done partner update?
[16:36] <kenvandine> the new U1 client is still coming this week, i will harass dobey asap for real code :)
[16:36] <kenvandine> yup
[16:37] <seb128> kenvandine, ping me if you want review and sponsoring
[16:37] <kenvandine> seb128, will do :)
[16:37] <seb128> I can probably do a first check and uploads
[16:37] <pitti> kenvandine: do you know when we'll drop the U1 applet?
[16:37] <seb128> then Riddell and pitti can NEW
[16:37] <dobey> pitti: it's already dropped in trunk
[16:37] <pitti> dobey: \o/
[16:37] <kenvandine> dobey, release eta?
[16:38] <dobey> kenvandine: this evening or tomorrow. i'll be poking statik to upload when ready
[16:38] <kenvandine> ok
[16:38] <kenvandine> thx
[16:38] <kenvandine> only new dep is libu1?
[16:38] <dobey> no, libu1 isn't a dep for ubuntuone-client any more
[16:38] <kenvandine> oh... ok
[16:38] <dobey> webkit will be though
[16:38] <kenvandine> ok
[16:39] <dobey> python-webkit i guess. whatever the package for the py bindings is
[16:39] <kenvandine> moving along
[16:39] <pitti> kenvandine: pywebkit is already on the CD, no prob there
[16:39] <kenvandine> yeah
[16:39] <dobey> yay
[16:39] <kenvandine> yay for gwibber too :)
[16:40] <dobey> and music store :)
[16:40] * dobey goes off to lunch
[16:40] <rickspencer3> kenvandine, all done?
[16:40] <kenvandine> yup
[16:40] <rickspencer3> thanks kenvandine
[16:40] <rickspencer3> Riddell, Kubuntu status?
[16:40] <Riddell> * 4.4 final is in
[16:40] <Riddell> * daily images are looking good for i386/amd64 except for blocking on mysql fixed, amd64 should fit in CD size tomorrow
[16:40] <Riddell> * kdebindings not compiling on ARM (the important bits do compile so we can just work around the unimportant bits)
[16:40] <Riddell> * Qt 4.6.2 due today, and then we're in a good position for feature freeze
[16:41] <rickspencer3> hmm
[16:41] <rickspencer3> thanks Riddell
[16:42] <rickspencer3> ccheney, mozilla update?
[16:42] <ccheney> rickspencer3: i got everything that i think belongs into soup done, there are about 4 symbols left that it uses that i understand probably belong in glib/gnutls directly and that is what i am working on next
[16:43] <rickspencer3> ccheney, how long until you complete it?
[16:43] <ccheney> rickspencer3: i was off thu - mon, so haven't progressed since my last update with you
[16:44] <ccheney> rickspencer3: hopefully not more than a day or two for that part, will need to discuss with asac how to make sure i don't break anything in the process since i will be modifying glib directly it is the most dangerous part of the change
[16:44] <rickspencer3> ok
[16:44] <rickspencer3> moving on
[16:44] <pitti> as long as we just add new code/symbols, it should be okay
[16:44] <asac> the glib changes really looked confined well
[16:44] <pitti> and glib has a large test sutie
[16:44] <asac> but yes, lets look together
[16:45] <ccheney> asac: ok
[16:45] <rickspencer3> ok
[16:45] <asac> (just convenience funcs for thread-local stuff)
[16:45] <rickspencer3> so next team meeting we should hear that libsoup is complete
[16:45] <asac> \o/
[16:45] <rickspencer3> and work has started on the next item
[16:46] <ccheney> wrt OOo 3.2.0 is now out upstream but hasn't been updated by ooo-build and debian yet, hoping to get the final uploaded to ubuntu in the next week
[16:46] * rickspencer3 crosses fingers
[16:46] <ccheney> yes soup should be done by next week
[16:46] <asac> yes. webkit and ephy is next
[16:46] <asac> webkit -> easy (just the rename is painful); ephy require a few symbols (not that many from what i saw)
[16:46] <rickspencer3> ccheney, thanks for the update on OOo, too bad it won't be ready for A3, but not a real problem
[16:46] <ccheney> yea, hopefully no more cascading effects like with soup/glib, heh :)
[16:47] <rickspencer3> ok, thanks ccheney and asac
[16:47] <rickspencer3> moving on ...
[16:47] <ccheney> rickspencer3: yea i doubt the ooo-build/debian stuff will be done by thu
[16:47] <asac> rickspencer3: next week upload would hit a3, wouldnt it?
[16:47] <asac> but please dont upload ooo in a3 week ;)
[16:47] <rickspencer3> Transitioning from A3
[16:47] * asac with armel hat on
[16:47] <ccheney> asac: yea i won't be uploading it unless it is done by friday
[16:47] <asac> good
[16:47] <asac> thx
[16:47] <rickspencer3> ok, Transitioning from A3
[16:47] <rickspencer3> ;)
[16:47] <rickspencer3> so, this is pretty much the last week for blueprint work
[16:48] <rickspencer3> so anything that you don't think will be done by Thursday, consider it more or less postponed to Lucid + 1
[16:48] <rickspencer3> if you have any concerns about specific work items, please discuss with pitti, and if he is not available, you can ask me as well
[16:49] <rickspencer3> after this week, we have three areas of focus
[16:49] <rickspencer3> 1. Find and fix the worst bugs (how to find and track?)
[16:49] <rickspencer3> 1. Ensure a smooth upgrade experience (how to test?)
[16:49] <rickspencer3> 1. Integrate the latest and greatest
[16:49] <rickspencer3> in terms of the first two items, I feel there is some ambiguity
[16:49] <rickspencer3> hmm, pedro is not around :/
[16:49] <desrt> rickspencer3: all tied for first place priority? :)
[16:49] <rickspencer3> ?
[16:49] <pitti> on a tangent, http://people.ubuntu.com/~mvo/automatic-upgrade-testing/<<BR>> [16:49] <rickspencer3> desrt, yes, of course, I am an manager
[16:50] <rickspencer3> note that the list is subject to change :/
[16:50] <pitti> http://people.ubuntu.com/~mvo/automatic-upgrade-testing/2010-02-16-12:16:25/<<BR>> [16:50] <rickspencer3> just kidding, that's just Moin formatting
[16:50] <pitti> in short, defualt install upgrades are all working now
[16:50] <rickspencer3> thanks pitti
[16:51] <rickspencer3> I would like to discuss with QA team how we can better organize upgrade testing with the community
[16:51] <rickspencer3> this is ofter a sore point, and since we have created a good amount of time to focus on quality, I am hoping we can do a good job there
[16:51] <mvo> I'm working on a stepmaker based upgrade test (in addition to the one that pitti showed)
[16:51] <rickspencer3> mvo, are you working with marjo on this?
[16:51] <rickspencer3> or ara or pedro?
[16:52] <rickspencer3> also, we probably need to define some upgrade behaviors and make sure they work well
[16:52] <mvo> rickspencer3: no, I discussed it with soren at the sprint and created a profile. but I will and hope they can take it over
[16:52] <rickspencer3> like, different cases if the user removed or moved an indicator, etc...
[16:52] <rickspencer3> mvo ok
[16:53] <rickspencer3> anyone particularly passionate about upgrade experience who would like to work on planning for this aspect?
[16:53] <pitti> the trouble with us developers is that we don't have good real-life test cases for upgrades
[16:53] * rickspencer3 listens to crickets chirping
[16:53] <pitti> we upgrade every day
[16:53] <pitti> so we do them in small steps
[16:54] <rickspencer3> yeah, I think we need to organize some test cases and some community efforts around this
[16:54] <rickspencer3> ACTION: rickspencer3 to engage QA wrt upgrade testings
[16:54] <pitti> we need to receive lots of feedback at beta-1, when all Canonical folks upgrade
[16:54] <rickspencer3> ^sound okay?
[16:54] <pitti> likewise with community feedback, of course (around beta-1)
[16:55] <rickspencer3> pitti, right, but I would like to see some more organized, rigorous testing, since we have created a long period of time to focus on quality
[16:55] <pitti> well, synthetic upgrades are already tested automatically
[16:55] <rickspencer3> ok, I have an action item
[16:55] <rickspencer3> tseliot, bryceh have you anything set up to test upgrades wrt the new alternatives system?
[16:55] <pitti> what they miss is customizations in /etc/, wild package combinations, nvidia drivers, broadcom wifi, all the stuff that makes real boxes explode
[16:56] <tseliot> rickspencer3: yes, today I've put my new nvidia card in my testing box so now I should have all the hardware I need
[16:56] <rickspencer3> tseliot, so you are doing all the testing yourself?
[16:57] <rickspencer3> (until the community starts upgrading)?
[16:57] <pitti> I seriously think that we'll get lots of bug reports after beta-1
[16:57] <rickspencer3> pitti, I know
[16:57] <bryceh> rickspencer3, I have hardware set up for it as well, but time is the main limiting factor
[16:57] <rickspencer3> I would like to get some of those earlier if we can
[16:57] <tseliot> rickspencer3: yes but we also have a mailing list for that
[16:57] <seb128> pitti, upgrade bugs or just bugs?
[16:57] <pitti> so far our problem has always been to fix everything, not really not getting enough feedback
[16:57] <pitti> seb128: upgrade problems
[16:57] <pitti> seb128: well, "bugs" as well, of course
[16:58] <rickspencer3> ok
[16:58] <tseliot> bryceh: +1 on lacking the time
[16:58] <rickspencer3> I will talk to marjo and see if we can organize something a tad more methodical than "everyone update for beta-1 and tell us what is broken"
[16:58] <rickspencer3> so, that brings us back to
[16:58] <rickspencer3> find and fix the worst bugs
[16:59] <rickspencer3> I think that with the new gravity system, we might be a tad more effective at picking out actionable and important bugs
[16:59] <rickspencer3> thoughts on how we could do better finding the worst bugs?
[17:00] <rickspencer3> oookay
[17:00] <rickspencer3> moving on
[17:00] <pitti> rickspencer3: btw, ISTR that ara wanted to organize that as well
[17:00] <bryceh> rickspencer3, hehe
[17:01] <bryceh> rickspencer3, I think that same question gets asked every release ;-)
[17:01] <pitti> rickspencer3: "that" > driver upgrade testing, etc.
[17:01] <rickspencer3> bryceh, yes
[17:01] <rickspencer3> and we have taken some action wrt the problem
[17:01] <bryceh> rickspencer3, in fact I was asking myself the same question in the shower this morning
[17:01] <rickspencer3> gravity + JSON searches
[17:01] <rickspencer3> okay
[17:02] <rickspencer3> by next week I will have the perfect bug finding workflow figured out
[17:02] <bryceh> awesome
[17:02] <rickspencer3> ;)
[17:02] <rickspencer3> ok
[17:02] <rickspencer3> speaking of which
[17:02] <rickspencer3> pitivi
[17:02] <rickspencer3> so, we are to decide by beta 1 if pitivi is in or out
[17:02] <rickspencer3> but we didn't really discuss the criteria for that
[17:03] <bryceh> whats the word on the street?
[17:03] <pitti> ideally we'd have a set of use cases that it needs to fulfill
[17:03] <pitti> IOW, a "test plan"
[17:03] <pitti> and then check the bug situation (incoming bugs vs. activity on them)
[17:03] <seb128> I still failed to do anything useful with pitivi
[17:03] <seb128> but maybe that's only me
[17:03] <pitti> I just don't see anyone who currently maintains it on the Ubuntu side
[17:03] <rickspencer3> bryceh, works on the street that the next version will be ready almost on time, fixes a lot of bugs, and has simple fade transitions
[17:04] <rickspencer3> pitti, right
[17:04] <rickspencer3> does someone want to step up and be ubuntu pitivi maintainer?
[17:04] * seb128 hides
[17:04] <rickspencer3> also, of course, the pitivi team has a big say in this, especially if they don't feel it is ready
[17:05] <seb128> I'm fine doing updates
[17:05] <seb128> but I don't want to do extra work on it
[17:05] <pitti> seb128: would not make much sense anyway if you aren't even using it
[17:05] <rickspencer3> there may be some codecy stuff that needs to be tweaked
[17:05] <rickspencer3> yeah
[17:05] <seb128> pitti, well I'm doing updates for lot of GNOME things I don't use
[17:06] <rickspencer3> seb128, right, I think he was referring to the "extra work" part
[17:06] <pitti> seb128: is that an argument that you want pitivi maintenance pushed to you? :-)
[17:06] <seb128> noooo ;-)
[17:06] <rickspencer3> ok, let's carry this over to next week
[17:07] <rickspencer3> see how everyone feels after ff ;)
[17:07] <rickspencer3> perhaps I will have talked someone into it by then
[17:07] <rickspencer3> pitti, did you have an "other business" item?
[17:07] <pitti> anyone at least got a video cam?
[17:07] <seb128> hehehe
[17:07] <pitti> rickspencer3: some stragglers
[17:07] <rickspencer3> ok
[17:07] * rickspencer3 hands mic to pitti
[17:07] <pitti> kenvandine: I saw that power-indicator just got dropped, so that settles question 1
[17:07] <kenvandine> :)
[17:08] <pitti> http://people.canonical.com/~pitti/workitems/canonical-desktop-team-lucid-alpha-3.html<<BR>> [17:08] <pitti> I cleaned up a little today (some people got nagging questions from me, sorry)
[17:08] <pitti> so we are by and large back on track
[17:08] <pitti> I went through the remaining WIs, and they seem sensible and current to me
[17:08] <pitti> if you sort this list by number of "todo", you'll see that 9 specs are just having 1 WI left
[17:09] <seb128> there is probably some that can be moved there
[17:09] <seb128> like speeding xrandr
[17:09] <pitti> also, things like "write test plan", etc.
[17:09] <pitti> which can be done after FF, and in freezes
[17:09] <pitti> startup-speed is obviously the biggest one left
[17:09] <pitti> we still have some stuff in the pipe
[17:09] <pitti> oh, btw
[17:09] <pitti> http://people.canonical.com/~scott/ratchet-netbook-lucid-20100216-23.png<<BR>> [17:10] <pitti> 10.8 s!
[17:10] <pitti> that's with some tweaks of Keybuk which will get uploaded soon
[17:10] <rickspencer3> oh man
[17:10] <rickspencer3> go go go
[17:10] <didrocks> :)
[17:10] <seb128> ;-)
[17:10] <pitti> all other bits are in the budget, just desktop is at 4.8
[17:10] <pitti> but this is awesome
[17:10] <rickspencer3> so where do we get the .8 seconds?
[17:11] <pitti> kenvandine: sfts -> remaining ones is integration (MIR and the like)? or still coding work to do?
[17:11] <pitti> it gets tight for this
[17:11] <pitti> rickspencer3: we still have some stuff in the pipe, like seb128's current work on nm-applet
[17:12] <rickspencer3> great
[17:12] <pitti> (which doesn't influence this particular chart, though)
[17:12] <seb128> I'm done with it
[17:12] <kenvandine> pitti, just one coding item
[17:12] <seb128> it has been handed to upstream and nm-applet packaging team now
[17:12] <kenvandine> and a few docs
[17:12] <kenvandine> pitti, i still hope to be done
[17:12] <kklimonda> chrisccoulson: bug 522726 - builds fine, haven't found any problems
[17:12] <pitti> seb128: rock, thanks a lot
[17:12] <ubottu> Launchpad bug 522726 in transmission "Update to 1.90" [Undecided,Confirmed] https://launchpad.net/bugs/522726<<BR>> [17:13] <seb128> pitti, you're welcome
[17:13] <rickspencer3> pitti, any other wi business to discuss?
[17:13] <pitti> the other bigger bit is the remaining mobile-lucid-une-2d-launcher
[17:13] <didrocks> pitti: so, you think that the 2D/3D detection added last week is still relevant? (this will slow down the boot, without speaking I wasn't able to work on that today)
[17:13] <pitti> didrocks: you had some concern there which is worth discussing?
[17:13] <pitti> didrocks: heh, snap :)
[17:14] <rickspencer3> asac has stepped out
[17:14] <didrocks> well, with the bg cache WI always changing, I don't know if I will have the time to do this (the 3 WI related)
[17:14] <rickspencer3> but I wanted to discuss with him today
[17:14] <pitti> didrocks: I saw another WI in a different spec which just said "if mutter fails to load, present a dialog and explain how to switch" or so?
[17:14] <didrocks> pitti: right, hence the "3" :)
[17:14] <rickspencer3> what is the consequence of not getting those items done?
[17:14] <pitti> didrocks: we certainly shouldn't reintroduce something like the compiz wrapper
[17:15] <kwwii> seb128: lp:~ubuntu-art-pkg/humanity/release is updated with the new message menu icon
[17:15] <didrocks> pitti: that was the plan: a wrapper to choose right "netbook-launcher{,-efl}
[17:15] <seb128> kwwii, danke
[17:15] <didrocks> but well, as it has been added at some kind of last minute effect, I don't know if it is really important
[17:15] <pitti> didrocks: can't we put that as code into n-l itself, with a very cheap test?
[17:16] <pitti> didrocks: anyway, I don't think it's an alpha-3 blocker
[17:16] <pitti> if we need it, it can land by beta-1
[17:16] <rickspencer3> ok, I'll talk to asac about just removing it from Lucid altogether
[17:16] <pitti> if we determine that the current behaviour breaks too much on boxes without 3D support
[17:16] <didrocks> pitti: so, asking for a FFe?
[17:16] <rickspencer3> pitti, but if we keep working on it, that's less bandwidth for:
[17:16] <rickspencer3> 1. find and fix teh worst bugs
[17:16] <rickspencer3> 2. ensure a smooth upgrade experience
[17:16] <pitti> didrocks: well, a fallback to the 2D launcher is a bug fix if starting thhe 3D launcher totally breaks the box
[17:17] <didrocks> pitti: heh, right
[17:17] <pitti> right, it's not even an upgrade problem, is it"
[17:17] <pitti> ?
[17:17] <jpds> 12
[17:17] <pitti> karmic's netbook-laucher totally screws up in kvm
[17:17] <pitti> i. e. on non-3D machines
[17:18] <didrocks> no, nothing was done previously
[17:18] * Amaranth sees compiz mentioned
[17:18] <pitti> didrocks: so if at all, I'd like to see the live system installer make the change (with an alternative perhaps)
[17:18] <pitti> depending on whether the live system detects 3D caps or not
[17:19] <didrocks> pitti: only on the live system so, what if you have a nvidia card and install 3D acceleration?
[17:19] <pitti> didrocks: then you have a postinst to change the alternative
[17:19] <pitti> but still, it seems quite complex to me
[17:19] <pitti> so my question would rather be, how much are we required to get this in the first place?
[17:20] <pitti> didrocks: if we can do something cheap and easy, like a 0.01 s runtime test in n-l which, when fails, executes n-l-2d, that's fine
[17:20] <didrocks> that's not on my call, asac or rickspencer3, maybe Smile :) for me, at least, it's not a regression, just a better experience thing
[17:20] <pitti> didrocks: right, it's a question for asac
[17:20] <didrocks> pitti: right, not sure about the test to do yet TBH
[17:20] <rickspencer3_> my tubes!
[17:21] <pitti> rickspencer3_: wb
[17:21] <rickspencer3_> heh
[17:21] <pitti> seb128 fell off the planet as well
[17:21] <rickspencer3_> my network connection went down, am on 3g now :/
[17:21] <rickspencer3_> anywho,
[17:21] <pitti> didrocks: ok, too complex for meeting topic; let's discuss tomorrow morning with asac, shall we?
[17:21] <seb128> sorry about that
[17:21] <seb128> I'm back
[17:21] <rickspencer3_> pitti, all done?
[17:21] <didrocks> pitti: sure
[17:21] <pitti> rickspencer3_: yes, from my side
[17:22] <rickspencer3_> any other business?
[17:22] <kenvandine> not from me
[17:22] <seb128> no
[17:22] * rickspencer3_ taps gavel
[17:22] <rickspencer3_> thanks all!
[17:22] <pitti> thanks everyone
[17:22] <ccheney> thanks
[17:22] <pitti> dinner o'clock
[17:22] <didrocks> thanks
[17:22] <seb128> thanks
[17:23] * didrocks returns fighthing the deserialize stuff

CategoryDesktopTeam

Back to DesktopTeam.

DesktopTeam/Meeting/2010-02-16 (last edited 2010-02-16 22:11:26 by 203-219-185-1)