15:00:36 #startmeeting KDE SIG Meeting 15:00:36 Meeting started Tue Apr 29 15:00:36 2014 UTC. The chair is Kevin_Kofler. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:00:36 Useful Commands: #action #agreed #halp #info #idea #link #topic. 15:00:41 #meetingname kde-sig 15:00:41 The meeting name has been set to 'kde-sig' 15:00:44 #topic Role call 15:00:50 * Kevin_Kofler is present, who else? 15:00:55 * jgrulich is present 15:01:20 here 15:01:26 * dvratil present 15:01:33 * rkratky is here 15:01:53 hola 15:02:13 rdieter: -ELANG :p 15:03:04 pino|work: bonjour ? 15:03:22 hi ;) 15:03:49 #chair jgrulich tosky dvratil rkratky rdieter pino|work 15:03:49 Current chairs: Kevin_Kofler dvratil jgrulich pino|work rdieter rkratky tosky 15:03:54 moin (ok I'm done :) ) 15:04:25 * than is presen 15:04:32 #chair than 15:04:32 Current chairs: Kevin_Kofler dvratil jgrulich pino|work rdieter rkratky than tosky 15:04:47 mbriza: Ping? 15:05:17 jreznik: Ping? 15:05:31 I think jreznik is not going to attend 15:05:43 He's quite busy these days. 15:06:02 #info Kevin_Kofler, jgrulich, tosky, dvratil, rkratky, rdieter, pino|work, than present. 15:06:10 #topic Agenda 15:06:54 Meeting chan registration/conflict. 15:07:15 Anything else? Status updates on new upstream releases or stuff like that? 15:08:05 I can give status on 4.12.5, pam-kwallet 15:08:11 OK. 15:08:28 IMHO, moving to GStreamer 1.0 should be a priority, maybe we can also take that up. 15:08:39 #topic Meeting chan registration/conflict 15:08:47 Let's start with the administrative stuff. 15:09:13 So I looked at the meeting chan situation: We are currently registered for 15:00 UTC on the wiki, but not in Fedocal. 15:09:49 Env-and-Stacks is registered on both the wiki and in Fedocal as alternating between 13:00 and 16:00 UTC, but they apparently moved to 15:00 UTC with US DST, causing a channel conflict every other week. 15:10:40 My proposal is: Also register our meeting on Fedocal to match the wiki, and show up on time next week so that Env-and-Stacks cannot steal our chan. 15:10:48 We have always been at 15:00 UTC. 15:11:23 but can you register on Fedocal if there is a conflict? 15:11:25 +1, though I don't mind being flexible and considering the use of another meeting channel 15:11:34 tosky: there is no conflict on fedocal 15:11:43 oh, ok, I misunderstood 15:11:48 Env-and-Stacks is listed at 16:00 UTC on both Fedocal and the wiki. 15:12:18 They're actually meeting at 15:00 it seems, but this isn't documented anywhere. 15:12:27 And they didn't bother checking that the chan is free to begin with. 15:12:55 ok, so it's easy then 15:13:07 if it's not documented, this channel is officially free 15:18:12 #info KDE SIG Meeting now also registered on Fedocal. 15:18:47 #link https://apps.fedoraproject.org/calendar/location/fedora-meeting@irc.freenode.net/ 15:18:57 good :) 15:19:01 thanks 15:19:20 IMHO, we should also drop a note to the Env-and-Stacks folks. 15:19:50 We've been on the wiki for 15:00 UTC since forever. 15:19:59 good idea 15:20:28 Now I also added it on Fedocal for every Tuesday from 2013-01-01 to 2025-12-31. :-p 15:21:44 #topic 4.12.5 releases 15:21:47 rdieter: Your turn. 15:22:24 ok, 4.12.5 is in kde-testing, and ready to go in bodhi as soon as upstream announces it 15:22:58 #info 4.12.5 is in kde-testing, and ready to go in bodhi as soon as upstream announces it. 15:23:03 2 tarball respins to building now: kdelibs, kde-workspace 15:23:09 Great. 15:23:22 I guess it should be fairly uneventful, being a .5 release. 15:23:54 #topic pam-kwallet 15:24:36 .bug 1091479 15:24:39 rdieter: Bug 1091479 Review Request: pam-kwallet - PAM module for KWallet - https://bugzilla.redhat.com/1091479 15:24:41 ^^ needs a reviewer 15:24:52 #link https://bugzilla.redhat.com/1091479 15:25:03 its in kde-unstable too (since it needs 4.13.0), been testing it for about a week with good results (kdm) 15:25:23 posted on kde@fpo list for more testing/feedback 15:25:32 I'll do it tomorrow if nobody do it today 15:26:11 jgrulich: thanks 15:26:53 #action jgrulich will do the review tomorrow if nobody does it today. 15:27:33 I suppose it needs 4.13.0 because the old KWallet can't talk to it? I guess the KWallet patches could be backported… 15:27:43 I suppose they should have no effect if pam-kwallet isn't installed. 15:29:08 there was something in kde-runtime too I think 15:29:53 That's where KWallet really lives, the other package is just the kwalletmanager. 15:30:14 ah, ok, maybe that's it then 15:30:40 Anyway, we don't HAVE to support that, I just thought that it'd be nice to allow users to opt into it on F20 and even F19. 15:30:46 once in, will have to modify kde-settings to set kwallet defaults to not autoclose 15:31:12 Though for F20, if we're pushing 4.13 anyway, we don't need to backport. :-) 15:31:26 (and that's another open question there) 15:33:36 ok, so that's all I have on pam-kwallet 15:34:00 anyone else try it yet? (if not, I'd encourage you to do so) 15:35:14 #halp Testing wanted for pam-kwallet. 15:35:41 I guess none of us did. :-) 15:35:47 #topic GStreamer 1.0 migration 15:36:05 So, let's take up the GStreamer thing again. 15:36:27 KDE is kinda late at catching up there, a lot of stuff is still stuck on 0.10. 15:36:34 It'd be nice to get everything moved to 1.0 for F21. 15:36:42 (It was supposed to happen for F20 already.) 15:37:26 rdieter split the qt-mobility packaging, which should at least help avoid the … → qt-mobility → GStreamer 0.10 depchain. 15:37:37 theres a lot of packages that would require patches/snapshots , probably good idea to create a tracker bug 15:37:57 There isn't anything important using the multimedia parts of qt-mobility (anymore), AFAIK. 15:38:01 , qt-mobilities libQtMultiMediaKit (which nothing in the distro currently uses, afaict) 15:38:18 QtWebKit used to use them long ago, but moved to using GStreamer directly. 15:38:27 (using the code that WebKit had for webkitgtk anyway) 15:38:58 And KGranatier used the stuff for a moment, but then switched to OpenAL. 15:39:25 The other stuff that uses GStreamer 0.10 in KDE: 15:39:29 so biggies include: qtwebkit, phonon-gstreamer 15:39:50 qt-gstreamer (used in telepathy stack) 15:39:52 Hmmm, actually, the topic is bad, let me fix it… 15:39:53 #undo 15:39:53 Removing item from minutes: 15:40:02 #topic GStreamer 1 migration 15:40:07 We're at 1.2 already. :-) 15:40:14 close enough 15:40:19 So I was saying: The other stuff that uses GStreamer 0.10 in KDE: 15:40:47 * QtWebKit – WebKit already supports GStreamer 1, and openSUSE has a small patch to backport the support in the Qt parts 15:40:54 (The Qt 5 version already supports it, I think.) 15:41:23 correct 15:41:25 * Phonon-GStreamer – There's a branch that supports GStreamer 1, it should be working. 15:41:28 (about qt5) 15:42:00 ironic that qt5-qtmultimedia still uses gst-0.10 15:42:39 * QtGStreamer – There too, there's a branch with GStreamer 1 support, and also patches for kde-telepathy (that also port it to the current version of farstream, that uses 1.x instead of 0.10 too) and Kamoso to work with the new version. 15:42:51 So in principle, everything should be there to ship. 15:42:59 It should be just a matter of importing it into Rawhide. 15:43:20 last we checked, the rumors of a kamoso patch were exagerated, we couldnt find one 15:43:34 I could eventually find it. 15:44:00 IIRC, it was in a git clone on the github project where the QtGStreamer port is being developed. 15:44:02 ok, so back to my original suggestion, tracking bug, then bugs against all the affected packages... 15:44:03 kamoso has not had a release in years 15:44:34 pino|work: But it works. :-) 15:44:40 We should keep it that way. 15:45:19 well, sure, if you carry your own patches to eg fix kipi stuff (like new api, file conflicts with kde-workspace), ... 15:45:50 The advantages from moving to GStreamer 1: more features, fewer bugs, less user confusion (no more requirement for 2 different sets of patent-encumbered plugins depending on the app), less wasted space on the live images. 15:46:17 And also, automatic plugin installation will keep working. (It got switched from being built against 0.10 to being built against 1.x in F21.) 15:46:42 , gst upstream has no support for 0.10 too 15:48:30 So, what's the plan? Do we start the import of the updated/changed bits into Rawhide now? 15:48:47 ok, so back to my original suggestion, tracking bug, then bugs against all the affected packages... 15:49:10 once we have everything ready, then start importing 15:49:35 -1, unnecessary bureaucracy… 15:50:00 * Kevin_Kofler dislikes the latest trend to try to track every silly thing on Bugzilla. 15:50:16 It just generates lots of useless e-mail traffic and wastes everyone's time. 15:50:26 Kevin_Kofler: no one knows all the necessarily details except you. 15:50:33 We already know exactly what needs to be done, just do it. 15:50:36 unless you want to do all the work to implement this 15:50:39 I listed the necessary details above. 15:50:42 well, if for some reason you cannot finish it off soon, you don't forget it later nor lose track of it 15:50:48 I've no idea where all the patches can be found 15:51:05 if you want help, please *document* it 15:52:19 Whatever… I'm filing the terabazillion bugs that you ask for now… ;-) 15:52:30 (Well, actually, it should be only 5 or 6.) 15:52:43 #action Kevin_Kofler to file tracking bugs. 15:52:59 #topic Open discussion 15:53:02 Anything else? 15:55:03 Closing the meeting in 60 seconds… 15:55:25 dvratil: you did have some answers for the co-installability issue, right? 15:55:36 yes 15:55:41 (maybe for next week, no much time left?) 15:55:46 as you wish :) 15:55:51 We can discuss it now. 15:55:52 talked to dfaure, he they decided to use kf5 prefix for some stuff 15:55:54 co-install of kf5 ? 15:55:56 *they :) 15:56:16 and the generic stuff will have "5" suffix, so /usr/share/kservices5 15:56:19 for instance 15:56:48 good 15:56:49 it will be already in 4.99.0 next week(-ish?) 15:57:44 and there's some pressure on upstreamto make workspace co-installable too, but that's not decided yet 16:00:26 That could help us, it'd allow us to ship it in F21 as a non-default option, avoiding the 4.0 flamewar from F9… 16:00:59 Upstream keeps blaming the distros for the way 4.0 was introduced, but are not willing to make the changes to let us do it better. :-/ 16:01:32 yeah, though having copr and playground repos may make that easier in the long-run too, we'll have to see 16:03:29 anyway, no conflicts for libraries (and if there are, it's an upstream bug) 16:05:04 FYI: 16:05:09 .bug 1092641 16:05:13 Kevin_Kofler: Bug 1092641 KDE GStreamer 1 migration tracking bug - https://bugzilla.redhat.com/1092641 16:05:26 I'm filing the bugs against the affected components now. 16:11:06 Looks like we're really out of topics to discuss now, and also out of time, so thanks everyone for attending, see you next week! 16:11:12 #endmeeting