15:08:26 #startmeeting kde-sig -- http://fedoraproject.org/wiki/SIGs/KDE/Meetings/2012-02-21 15:08:26 Meeting started Tue Feb 21 15:08:26 2012 UTC. The chair is jreznik. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:08:26 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:08:53 who's ready (unzombify rdieter)? 15:08:57 * jreznik is here 15:09:05 * rnovacek is here 15:10:17 Present. 15:10:30 present 15:11:03 * jreznik is semi-ready, my dog has want-a-hug time... 15:13:23 let's wait a moment for rdieter 15:13:42 here (my body is objecting to non-czech food :( ) 15:13:45 * ltinkl present 15:13:52 rdieter: hehe 15:14:58 #info jreznik rdieter rnovacek Kevin_Kofler than ltinkl present 15:15:10 #chair jreznik rdieter rnovacek Kevin_Kofler than ltinkl 15:15:10 Current chairs: Kevin_Kofler jreznik ltinkl rdieter rnovacek than 15:17:05 anything for agenda? 15:17:14 Alpha blockers (there's at least the "llvmpipe is still crap, so enabling KWin compositing on it was premature and breaks the Alpha in VMs" proposed blocker, though we've mostly agreed on a resolution on #fedora-kde). 15:17:54 Oh, also DevConf wrapup. 15:18:13 Kevin_Kofler: we've agreed it should be fixed, but there's contention still on how best to do that. 15:18:37 devconf is good 15:19:09 added 15:19:15 * rdieter can't think of anything else right now 15:19:51 Oh, touch-enabled apps. 15:20:47 ok, touch/mobile versions for plasma-active spin (if we have time) 15:21:18 only in case of time - it's not urgent 15:22:07 #meetingname kde-sig 15:22:07 The meeting name has been set to 'kde-sig' 15:22:09 ah... 15:22:23 ah, changed topic... sorry guys 15:22:28 .bug 794835 15:22:30 rdieter: Bug 794835 kde desktop unresponsive in F17 virt-manager, compositing+llvmpipe too slow, not ready - https://bugzilla.redhat.com/show_bug.cgi?id=794835 15:23:25 so, there's 2 things that concern us here. 1. llvmpipe is not ready for testing, 2. kwin compositing fallback mode is (sometimes?) flaky 15:24:19 and 2 proposals (hrm, I flipped the ordering here from prior discussion in #fedora-kde, sorry). 15:24:34 prop 1: disable/omit llvmpipe kwin patch 15:24:44 propsal 2: disable kwin compositing by default 15:24:58 #topic compositing and llvmpipe in F17 15:25:17 turns out implementing prop 2 will (largely) fix both issues by itself 15:25:56 if llvmpipe is really not yet ready, it doesn't make sense to ship it... 15:25:57 but, I'd like us to consider both proposals independently, if that's agreeable? 15:26:02 yep 15:26:38 I'm for doing 1 for now, and reconsider 2 later 15:26:38 The only proposed Alpha blocker we have is about llvmpipe, and at least I agree that it should be a blocker (we'll see what QA/rel-eng thinks). 15:26:39 anyone interested in further discussion (beyond what we did prior to this meeting)? 15:27:06 and if not, we can jump into some formal voting 15:27:15 I'm voting +1 for prop. 1 (drop llvmpipe patch for now). 15:27:34 * rdieter waits for any discussion requests first. 15:27:37 I'm voting -1 for prop. 2 (disable compositing for now) because we really need to test upstream defaults with the Alpha. 15:27:49 yup, +1 with Kevin 15:27:52 We can change things for Beta or Final if it turns out to still suck too much. 15:28:12 We've been overriding upstream there for way too long, and more and more stuff is going to rely on compositing starting with 4.9. 15:28:51 drop llvmpipe for now, enable compositing for alpha, get some results, disable it for beta+ :) 15:29:09 ok, I guess it looks like we're voting. :) 15:29:11 Kevin_Kofler: with 4.9 we would have to do it for sure, same for Active 15:29:54 if 4.9 would require compositing, we will close the door for it when f17 won't required compositing 15:29:57 me, +1 prop 1, -1 prop 2 (for now) 15:30:27 rnovacek: +1, which is one of the reasons I don't like prop. 2. 15:30:28 #info the only alpha blocker proposal is llvmpipe breakage (794835) 15:30:28 than: still feel the same as before? 15:30:35 rdieter: yes ;-) 15:31:08 ok, if I count right, +5/-1 for prop 1, -5/+1 for prop 2 ? 15:31:28 rnovacek, Kevin_Kofler: it's probably true - we can reconsider effects with 4.9 for f17 (it's going to be huge update) or try to ship without DE 15:33:15 Kevin_Kofler: does it mean 4.9 doesn't work without compositing? 15:34:22 Some parts like the new screen locker won't work without compositing, there will be a fallback (in this case to the old screen locker). 15:34:50 jreznik: I don't think we should change default behavior (enable DE) after release of f17, it can break working systems for a lot people 15:34:50 than: more features will depend on compositing - locking etc. but as Kevin_Kofler there will be fallback 15:35:07 rnovacek: we can use fallback then 15:35:25 I'd support DE if I'd be sure it's working, but it's not... 15:35:31 I agree that this is a setting we should ship F17 with, not change in an update. 15:36:25 so, if there's no contesting voting results, I can offer to implement prop 1 after meeting. 15:36:44 FWIW, it's unfortunate that the llvmpipe feature is advertised as a Fedora 17 feature, but still not in now that we're in feature freeze. :-/ 15:37:11 Why is this stuff not in Branched yet if it's being worked on? :-/ 15:38:35 rdieter: Go ahead. 15:39:39 #agreed (+5/-1) proposal 1: disable/omit llvmpipe kwin patch 15:40:23 #info (+1,-5) rejected proposal 2: disable kwin compositing by default (for now, re-evaluate for f17beta) 15:40:48 #action rdieter to implement proposal 1 after meeting 15:41:08 anything else here, or can we move on? 15:41:15 was it really to disable for alpha? I thought it's enable for alpha, re-evaluate for beta 15:42:15 jreznik: your question is about prop 2? rejecting prop2 means keep compositing enabled (per upststream default) 15:42:48 or am I confused? 15:43:20 Yes, we rejected the proposal to disable compositing by default for F17 Alpha, which means we keep compositing enabled by default in F17 Alpha. 15:43:56 Double negatives tend to confuse people. ;-) 15:45:00 :D 15:45:10 definitely doesn't not confuse me... or wait... tee hee 15:45:34 It is not false that this does not confuse me. ;-) 15:45:52 ah, me is - I see rejected proposal :) ok, sorry :) 15:46:13 * jreznik is really confused now, dog gots crazy for some reason, probably wants out... 15:47:00 my "I'm confused, wait, maybe I'm not" t-shirt can be the motto for todays meeting. :) 15:47:31 * jreznik agrees 15:47:38 Next topic please! 15:47:53 #topic brno fad wrapup 15:48:34 so how about we go around the room, everyone give a brief summary of what they worked on @ devconf, and the results (or not :) ) of said work. 15:48:53 (then, for extra homework, blog about it) 15:49:19 So, new colored systray icons for 4.8, done, not uploaded yet. 15:49:50 Update of git-cola in Fedora to latest upstream, done in Rawhide, in updates-testing for F1[567]. 15:50:21 Talking to all of you: priceless. ^^ 15:50:37 from active packaging perspective the only one package is still pending review - share-like-connect another package with invalid-soname - do we care about it? as we have it in many our packages already... 15:51:25 on the Solid udisks2 front, a first compilable version was finished and pushed to kdelibs git (https://projects.kde.org/projects/kde/kdelibs/repository/revisions/aef03f63de422990ffe14f72b9994c8aba3141b6) 15:51:34 Unversioned shlibs are not a review blocker. 15:51:47 * rnovacek looked at the nepomuk stuff, it looks interesting but not many apps uses it (dolphin and bangarang so far) 15:52:38 and it has some issues, sometime the update of metadata fails for some reason 15:53:54 I'm suggesting to wait some time and then reconsider to package it, if it gets properly updated 15:54:19 for the talk - it was recorded and all talks are now under processing... 15:54:46 rnovacek: can package for rawhide(f18) only for now? or you think we should wait even longer? 15:54:51 jreznik: Great. 15:54:58 jreznik: cool! 15:55:16 rdieter: ok, that seems reasonable 15:55:51 for qt 5 packaging we decided to wait at least for alpha... but still I'm tempted to start working on it :D 15:55:53 Your talk was interesting, (ab)using QML as a presentation languageā€¦ (I think it's not the optimal language for that purpose. ;-) But it was fun to have the examples embedded in the presentation.) 15:56:25 * rdieter worked on integrating kde telepathy 0.3 bits, got good/sane defaults now. updated comps-f18 for this, and other related kde-package splits (waiting till after f17alpha, to backport to comps-f17) 15:56:46 (Oh BTW, you said that people don't normally compile a presentation, I do, using LaTeX-Beamer. ;-) ) 15:57:19 (Compiled languages FTW!) 15:57:20 touche 15:57:23 Kevin_Kofler: there's no easier way how to write talks, the issue is export to PDF 15:57:48 Kevin_Kofler: check sources http://qt.gitorious.org/qt-labs/qml-presentation-system 15:57:53 I guess I have one more tweak to make wrt kde-telpathy, and that's to enable it's systray/plasma icon by default (pending upstream feedback) 15:58:22 rdieter: btw: instant messaging present has no icon when added to panel for me 15:58:24 to be clear, the plan is to ship kde telepathy, and not kopete, by default. 15:58:48 rnovacek: you right-click on systray ? and check it? 15:58:54 rdieter: +1, tried it, it's really good (comparing to 0.2) 15:58:57 (similar to how plasma-nm displays) 15:59:14 rnovacek, rdieter: that's the only issue I hit using it 15:59:24 rdieter: that works :) 15:59:29 ok 15:59:36 but adding as widget is problem 15:59:46 rdieter: The trend upstream is to deprecate the use of the systray by applications, at least by defaultā€¦ 15:59:47 interesting, that should work too, but I can test that more later 16:00:18 But I guess it can be discussed in this case. 16:00:58 I'm missing merging of contact, otherwise is ktp great :) 16:01:34 * rdieter also helped cwickert review zanshin (who also worked on lightdm packaging) 16:02:17 and we all pitched in a little on various efforts to move plasma-active spin/reviews forward 16:02:27 I tried zanshin and it works good (+1 karma added) 16:02:28 kudos mostly to rnovacek, jreznik there 16:04:25 * rdieter can't think of anything else, but almost certain we've forgotten to mention *something*. :) 16:05:14 * jreznik just wants to say - it was really awesome to have you guys here, sad than didn't make it :( but still, thanks for coming, thanks for great time spent with you! 16:05:33 oh right, than, hope you're feeling better! 16:05:42 jreznik: +1, I totally agree 16:05:50 +1 :-) 16:06:13 rdieter: it's better now, still have sore throat 16:06:15 and yeah, I hope than is ok and next time... 16:06:42 * rdieter will miss all the nice, small, downstairs czech pubs 16:06:50 :) 16:07:01 sure i will be there next time! 16:07:21 ok, so I think we can end meeting today and next one will be in stuttgart, right? :) 16:07:38 jreznik: why not :) 16:08:14 #endmeeting