2011-05-24

Differences between revisions 9 and 10
Revision 9 as of 2011-05-24 15:57:13
Size: 1720
Editor: port-20249
Comment:
Revision 10 as of 2011-05-24 16:01:57
Size: 11099
Editor: p549B1F3F
Comment: notes from western edition
Deletions are marked like this. Additions are marked like this.
Line 37: Line 37:
 * Strawpoll in Western edition: Nobody was in favor of moving to the UDD branches, everyone preferred the status quo.
 * To be re-discussed in Eastern edition, particularly with Luke and Robert (as the X.org guys use git anyway)
Line 40: Line 43:
17:32:03 pitti I'm not sure why didrocks added the "new session system" twice, presumably to underline its importance :)
17:32:22 pitti so, let's start with the usual reports, then get to the discussions
17:32:25 didrocks pitti: of course :-) I think I conflicted with my edit (wikimoinsmoins is slow :p)
17:32:28 didrocks fixing
17:32:54 pitti Partner update -- kenvandine, I suppose you're still fighting with the flu?
17:33:16 pitti so let's skip this -- I don't think we are supposed to land anything at this early time anyway
17:33:43 pitti didrocks: anything noteworthy wrt. unity/2d/qt/etc?
17:33:56 didrocks pitti: will try to get unity-2d on the CD for alpha1
17:34:08 didrocks well CD/default image
17:34:12 rodrigo_ hello?
17:34:13 pitti that'd be good, to see early what CD overflow we are up against, and also for testing
17:34:15 didrocks I doubt we will be < 700 Mb
17:34:17 didrocks :)
17:34:25 pitti didrocks: np, alpha-1 can be oversized
17:34:38 didrocks for unity itself, the SRU will be this week, bit a little bit delayed
17:34:39 bcurtiswx_ yes kenvandine is still under the weather
17:34:54 didrocks should it -proposed and maybe be copied before to oneiric for alpha1?
17:35:01 rodrigo_ ugh, too much lag here
17:35:32 pitti didrocks: I think at this point we want a real upload; the toolchain and half of the GNOME/GTK infrastructure changed
17:35:42 didrocks pitti: sure, will do that way then
17:35:58 pitti didrocks: if it poses any problem, you can confirm in the bugs that all fixes are in upstream trunk as well :)
17:36:21 pitti tremolux: thanks for the s-c report in the wiki; anything we need to discuss?
17:36:45 tremolux pitti: you're welcome, I don't think so unless somebody has a question, etc.?
17:36:59 * pitti leaves some time to raise questions
17:37:11 * tremolux whistles
17:37:14 pitti tremolux: do you know if anyone already looked into the gtk3 migration? how bad is it?
17:37:39 tremolux pitti: it's not going to be very straightforward
17:37:48 tremolux mainly due to our custom components
17:38:00 pitti that's expected; I meant did you already stumble over major bugs in pygobject/gtk etc. which will make this hard?
17:38:19 tremolux nope, not that I know of
17:39:01 pitti Kubuntu -- as Riddell is in the launchpad camp this cycle, we don't have a representative here, or do we?
17:39:33 didrocks I don't think we have, indeed
17:39:42 pitti X.org -> eastern edition
17:39:53 pitti Let's talk about specs, baby!
17:39:57 pitti https://blueprints.launchpad.net/ubuntu/oneiric/+specs?searchtext=desktop-o
17:40:03 pitti nice progress last week
17:40:10 seb128 specs, yummy ;-)
17:40:13 pitti didrocks, cjwatson, ev, and me have one spec each for approval
17:40:36 pitti and 9 are still to be drafted
17:40:50 pitti so, consider this a gentle reminder :)
17:41:05 didrocks "gentle for now"? :-)
17:41:11 pitti the thing I'd like to discuss here (as we didn't get around to it at UDS) is https://blueprints.launchpad.net/ubuntu/+spec/desktop-o-packaging-branches
17:41:22 pitti didrocks: yes, rick's whip is for after-Thursday
17:41:47 pitti this is roughly whether we shoudl continue to use our debian/ only custom branches for now, or switch to UDD branches
17:41:54 pitti we made some experiments with UDD
17:42:38 seb128 pitti, "here" being the meeting now?
17:42:39 pitti can we do a quick straw-poll from everyone whether they'd prefer moving to UDD ("UDD"), or keep debian/ only ("old")?
17:42:49 pitti and then discuss the pro/con?
17:42:51 seb128 robert_ancell had some opinion about that, should we rather take that to the list?
17:43:09 pitti seb128: more interactive that way, but sure
17:43:15 seb128 ok, we can do a round there and then take it to the list
17:43:24 pitti so, everyone, shoot
17:43:27 seb128 so I'm for keeping the debian dir only
17:43:35 pitti debian/ only
17:44:02 cyphermox hmm... debian/ only, even if it's more trouble for merging
17:44:04 didrocks debian/ only where we are not upstream, merge-upstream udd/workflow otherwise
17:44:50 * rodrigo_ for keeping debian/ only
17:44:51 pitti didrocks: hm, the ones where we have an upstream in LP are the ones where the auto-imported UDD branches make least sense to me?
17:45:05 pitti didrocks: or do you mean "once you switch lp:ubuntu/ over to the custom branch"?
17:45:34 pitti mterry: any opinion?
17:46:00 pitti I put my reasons into the whiteboard of the bp
17:46:05 didrocks pitti: I don't care if we use the "canonical" (lp:ubuntu/ address or not), I just mean, the one where we are upstream (and so, upstream in bzr) should use full source branch
17:46:32 mterry pitti, no, both work fine for me
17:46:34 pitti didrocks: right, full source as a real derivative of trunk is the only thing that really makes sense for upstreams in bzr
17:46:41 didrocks like indicators, unity (what we already have), just want to make the disctinction between the two kind of branches we have :)
17:46:48 pitti didrocks: but the auto-imported ones aren't compatible to anything else
17:47:09 didrocks pitti: no, I though you meant "udd" as "full source" as opposed to "debian/" only
17:47:09 pitti ok, so it seems folks here want to keep the status quo then
17:47:12 didrocks but then, we agree :)
17:47:36 pitti I'll summarize on the wiki page and then ask jasoncwarner to bring it up in the eastern edition as well
17:48:00 pitti well, the X.org guys are using git anyway, but Robert and Luke touch desktopish stuff
17:48:21 pitti that's my agenda
17:48:28 pitti anything else from anyone?
17:48:38 * Sweetshark just read what "udd" is.
17:48:39 pitti seb128: anythign which we should be aware of on the GNOME 3 front?
17:48:48 Sweetshark dont think i want that for LO.
17:49:00 pitti Sweetshark: https://wiki.ubuntu.com/DistributedDevelopment/
17:49:08 pitti Sweetshark: no, you really REALLY don't want these :)
17:49:15 seb128 pitti, no, the transition is doing fine so far, we got most of the GNOME3 basis done
17:49:25 seb128 remaining part is the gnome-panel, gnome-menus,indicators
17:49:45 seb128 the indicators transition is a bit tricky since it requires quite some source updates and involve both GNOME and unity
17:50:20 pitti i. e. a big lockstep transition?
17:50:29 seb128 we have still a bunch of updates to get done but I will start building lists of things using deprecated libs on the CD soon
17:50:50 pitti seb128: would it be easier if we sent out a warning to the lists to not upgrade for a few days, and break oneiric for a couple of days?
17:51:07 pitti or can we prepare the lockstep transition in a PPA? (might be better)
17:51:30 seb128 we should check with dx first how it plays for them
17:51:37 seb128 i.e with njpatel, ted, didrocks
17:51:53 cjwatson for the record, I'd like to get at least *one* oneiric CD daily build out this week
17:51:53 seb128 we probably should land ido and libindicate gtk3 builds first
17:52:01 cjwatson so it would be nice to allow a bit of time for that
17:52:09 seb128 cjwatson, we should not destabilize things before a1
17:52:12 pitti ok, so let's stage this up in the ubuntu-desktop PPA
17:52:42 seb128 pitti, I want to see if building an indicator on the "wrong" gtk version leads it to be ignore or lead to unity exiting because it gets 2 gtk in process
17:52:45 cjwatson I don't mind churn, now's the time in the cycle for it, but the kernel guys want a CD image to test with
17:53:01 didrocks seb128: feeling adventurous? :)
17:53:02 cjwatson and I can only do that if I get a window when the desktop's installable
17:53:04 seb128 cjwatson, ok, we will be careful until we get a build
17:53:08 -- smspillaz ist nun als smspillaz|zzz bekannt
17:53:13 seb128 cjwatson, is there anything not installable now?
17:53:23 seb128 i.e anything we need to fix?
17:53:34 pitti wow @ http://people.canonical.com/~ubuntu-archive/testing/oneiric_probs.html
17:53:38 pitti I had expected it to be a lot worse
17:53:39 cjwatson it's OK *right now*, although that's only as of extremely recently
17:53:44 fta pitti, seb128: did you fix dbus already? or not yet?
17:53:46 cjwatson pitti: I've been working on it ...
17:53:52 pitti cjwatson: great job!
17:54:00 cjwatson TBH it was mostly the perl transition
17:54:09 seb128 cjwatson, ok, so we will keep it this way for a few days, let us know when you got a CD build
17:54:14 pitti fta: fixed for installability, but SpammapS really meant to upload a newer netbase; he's doing that now, and re-fix dbus
17:54:26 cjwatson seb128: will do, thanks. should be soon, I just need to take safekeeping copies of the images that weren't released with natty
17:54:36 pitti dbus broke the world this morning, so I quick-fixed it
17:54:41 fta pitti, ok; thanks.
17:55:17 pitti bah, esound, is that not quite dead yet
17:55:35 pitti oh, it's NBS
17:55:57 pitti I'll look a bit into this, and the language-support-* stuff
17:56:03 pitti ok, thanks
17:56:04 pitti AOB?
17:56:38 seb128 thanks
17:56:44 pitti thanks everyone, adjourned
17:56:46 didrocks thanks everyone
17:57:16 tremolux thanks! good day all

Actions from previous meeting

Actions from this meeting

Weekly Summary

Partner update

Unity

  • second unity SRU slip for a couple of days, should be available in -proposed by end of week.
  • unity-2d targetted to be in the default image for alpha1 (with Qt and co)

Software Center

Kubuntu

X.org

bzr usage

See desktop-o-packaging-branches

  • Strawpoll in Western edition: Nobody was in favor of moving to the UDD branches, everyone preferred the status quo.
  • To be re-discussed in Eastern edition, particularly with Luke and Robert (as the X.org guys use git anyway)

IRC log Western edition

17:32:03        pitti   I'm not sure why didrocks added the "new session system" twice, presumably to underline its importance :)
17:32:22        pitti   so, let's start with the usual reports, then get to the discussions
17:32:25        didrocks        pitti: of course :-) I think I conflicted with my edit (wikimoinsmoins is slow :p)
17:32:28        didrocks        fixing
17:32:54        pitti   Partner update -- kenvandine, I suppose you're still fighting with the flu?
17:33:16        pitti   so let's skip this -- I don't think we are supposed to land anything at this early time anyway
17:33:43        pitti   didrocks: anything noteworthy wrt. unity/2d/qt/etc?
17:33:56        didrocks        pitti: will try to get unity-2d on the CD for alpha1
17:34:08        didrocks        well CD/default image
17:34:12        rodrigo_        hello?
17:34:13        pitti   that'd be good, to see early what CD overflow we are up against, and also for testing
17:34:15        didrocks        I doubt we will be < 700 Mb
17:34:17        didrocks        :)
17:34:25        pitti   didrocks: np, alpha-1 can be oversized
17:34:38        didrocks        for unity itself, the SRU will be this week, bit a little bit delayed
17:34:39        bcurtiswx_      yes kenvandine is still under the weather
17:34:54        didrocks        should it -proposed and maybe be copied before to oneiric for alpha1?
17:35:01        rodrigo_        ugh, too much lag here
17:35:32        pitti   didrocks: I think at this point we want a real upload; the toolchain and half of the GNOME/GTK infrastructure changed
17:35:42        didrocks        pitti: sure, will do that way then
17:35:58        pitti   didrocks: if it poses any problem, you can confirm in the bugs that all fixes are in upstream trunk as well :)
17:36:21        pitti   tremolux: thanks for the s-c report in the wiki; anything we need to discuss?
17:36:45        tremolux        pitti: you're welcome, I don't think so unless somebody has a question, etc.?
17:36:59         *      pitti leaves some time to raise questions
17:37:11         *      tremolux whistles
17:37:14        pitti   tremolux: do you know if anyone already looked into the gtk3 migration? how bad is it?
17:37:39        tremolux        pitti: it's not going to be very straightforward
17:37:48        tremolux        mainly due to our custom components
17:38:00        pitti   that's expected; I meant did you already stumble over major bugs in pygobject/gtk etc. which will make this hard?
17:38:19        tremolux        nope, not that I know of
17:39:01        pitti   Kubuntu -- as Riddell is in the launchpad camp this cycle, we don't have a representative here, or do we?
17:39:33        didrocks        I don't think we have, indeed
17:39:42        pitti   X.org -> eastern edition
17:39:53        pitti   Let's talk about specs, baby!
17:39:57        pitti   https://blueprints.launchpad.net/ubuntu/oneiric/+specs?searchtext=desktop-o
17:40:03        pitti   nice progress last week
17:40:10        seb128  specs, yummy ;-)
17:40:13        pitti   didrocks, cjwatson, ev, and me have one spec each for approval
17:40:36        pitti   and 9 are still to be drafted
17:40:50        pitti   so, consider this a gentle reminder :)
17:41:05        didrocks        "gentle for now"? :-)
17:41:11        pitti   the thing I'd like to discuss here (as we didn't get around to it at UDS) is https://blueprints.launchpad.net/ubuntu/+spec/desktop-o-packaging-branches
17:41:22        pitti   didrocks: yes, rick's whip is for after-Thursday
17:41:47        pitti   this is roughly whether we shoudl continue to use our debian/ only custom branches for now, or switch to UDD branches
17:41:54        pitti   we made some experiments with UDD
17:42:38        seb128  pitti, "here" being the meeting now?
17:42:39        pitti   can we do a quick straw-poll from everyone whether they'd prefer moving to UDD ("UDD"), or keep debian/ only ("old")?
17:42:49        pitti   and then discuss the pro/con?
17:42:51        seb128  robert_ancell had some opinion about that, should we rather take that to the list?
17:43:09        pitti   seb128: more interactive that way, but sure
17:43:15        seb128  ok, we can do a round there and then take it to the list
17:43:24        pitti   so, everyone, shoot
17:43:27        seb128  so I'm for keeping the debian dir only
17:43:35        pitti   debian/ only
17:44:02        cyphermox       hmm... debian/ only, even if it's more trouble for merging
17:44:04        didrocks        debian/ only where we are not upstream, merge-upstream udd/workflow otherwise
17:44:50         *      rodrigo_ for keeping debian/ only
17:44:51        pitti   didrocks: hm, the ones where we have an upstream in LP are the ones where the auto-imported UDD branches make least sense to me?
17:45:05        pitti   didrocks: or do you mean "once you switch lp:ubuntu/ over to the custom branch"?
17:45:34        pitti   mterry: any opinion?
17:46:00        pitti   I put my reasons into the whiteboard of the bp
17:46:05        didrocks        pitti: I don't care if we use the "canonical" (lp:ubuntu/ address or not), I just mean, the one where we are upstream (and so, upstream in bzr) should use full source branch
17:46:32        mterry  pitti, no, both work fine for me
17:46:34        pitti   didrocks: right, full source as a real derivative of trunk is the only thing that really makes sense for upstreams in bzr
17:46:41        didrocks        like indicators, unity (what we already have), just want to make the disctinction between the two kind of branches we have :)
17:46:48        pitti   didrocks: but the auto-imported ones aren't compatible to anything else
17:47:09        didrocks        pitti: no, I though you meant "udd" as "full source" as opposed to "debian/" only
17:47:09        pitti   ok, so it seems folks here want to keep the status quo then
17:47:12        didrocks        but then, we agree :)
17:47:36        pitti   I'll summarize on the wiki page and then ask jasoncwarner to bring it up in the eastern edition as well
17:48:00        pitti   well, the X.org guys are using git anyway, but Robert and Luke touch desktopish stuff
17:48:21        pitti   that's my agenda
17:48:28        pitti   anything else from anyone?
17:48:38         *      Sweetshark just read what "udd" is.
17:48:39        pitti   seb128: anythign which we should be aware of on the GNOME 3 front?
17:48:48        Sweetshark      dont think i want that for LO.
17:49:00        pitti   Sweetshark: https://wiki.ubuntu.com/DistributedDevelopment/
17:49:08        pitti   Sweetshark: no, you really REALLY don't want these :)
17:49:15        seb128  pitti, no, the transition is doing fine so far, we got most of the GNOME3 basis done
17:49:25        seb128  remaining part is the gnome-panel, gnome-menus,indicators
17:49:45        seb128  the indicators transition is a bit tricky since it requires quite some source updates and involve both GNOME and unity
17:50:20        pitti   i. e. a big lockstep transition?
17:50:29        seb128  we have still a bunch of updates to get done but I will start building lists of things using deprecated libs on the CD soon
17:50:50        pitti   seb128: would it be easier if we sent out a warning to the lists to not upgrade for a few days, and break oneiric for a couple of days?
17:51:07        pitti   or can we prepare the lockstep transition in a PPA? (might be better)
17:51:30        seb128  we should check with dx first how it plays for them
17:51:37        seb128  i.e with njpatel, ted, didrocks
17:51:53        cjwatson        for the record, I'd like to get at least *one* oneiric CD daily build out this week
17:51:53        seb128  we probably should land ido and libindicate gtk3 builds first
17:52:01        cjwatson        so it would be nice to allow a bit of time for that
17:52:09        seb128  cjwatson, we should not destabilize things before a1
17:52:12        pitti   ok, so let's stage this up in the ubuntu-desktop PPA
17:52:42        seb128  pitti, I want to see if building an indicator on the "wrong" gtk version leads it to be ignore or lead to unity exiting because it gets 2 gtk in process
17:52:45        cjwatson        I don't mind churn, now's the time in the cycle for it, but the kernel guys want a CD image to test with
17:53:01        didrocks        seb128: feeling adventurous? :)
17:53:02        cjwatson        and I can only do that if I get a window when the desktop's installable
17:53:04        seb128  cjwatson, ok, we will be careful until we get a build
17:53:08        --      smspillaz ist nun als smspillaz|zzz bekannt
17:53:13        seb128  cjwatson, is there anything not installable now?
17:53:23        seb128  i.e anything we need to fix?
17:53:34        pitti   wow @ http://people.canonical.com/~ubuntu-archive/testing/oneiric_probs.html
17:53:38        pitti   I had expected it to be a lot worse
17:53:39        cjwatson        it's OK *right now*, although that's only as of extremely recently
17:53:44        fta     pitti, seb128: did you fix dbus already? or not yet?
17:53:46        cjwatson        pitti: I've been working on it ...
17:53:52        pitti   cjwatson: great job!
17:54:00        cjwatson        TBH it was mostly the perl transition
17:54:09        seb128  cjwatson, ok, so we will keep it this way for a few days, let us know when you got a CD build
17:54:14        pitti   fta: fixed for installability, but SpammapS really meant to upload a newer netbase; he's doing that now, and re-fix dbus
17:54:26        cjwatson        seb128: will do, thanks.  should be soon, I just need to take safekeeping copies of the images that weren't released with natty
17:54:36        pitti   dbus broke the world this morning, so I quick-fixed it
17:54:41        fta     pitti, ok; thanks.
17:55:17        pitti   bah, esound, is that not quite dead yet
17:55:35        pitti   oh, it's NBS
17:55:57        pitti   I'll look a bit into this, and the language-support-* stuff
17:56:03        pitti   ok, thanks
17:56:04        pitti   AOB?
17:56:38        seb128  thanks
17:56:44        pitti   thanks everyone, adjourned
17:56:46        didrocks        thanks everyone
17:57:16        tremolux        thanks!  good day all

IRC log Eastern edition

DesktopTeam/Meeting/2011-05-24 (last edited 2011-05-24 16:01:57 by p549B1F3F)