14:01:47 #startmeeting kde-sig -- http://fedoraproject.org/wiki/SIGs/KDE/Meetings/2010-12-14 14:01:47 Meeting started Tue Dec 14 14:01:47 2010 UTC. The chair is jreznik. Information about MeetBot at http://wiki.debian.org/MeetBot. 14:01:47 Useful Commands: #action #agreed #halp #info #idea #link #topic. 14:01:59 #meetingname kde-sig 14:01:59 The meeting name has been set to 'kde-sig' 14:02:07 #chair jreznik Kevin_Kofler ltinkl than rdieter_work rnovacek thomasj 14:02:07 Current chairs: Kevin_Kofler jreznik ltinkl rdieter_work rnovacek than thomasj 14:02:16 #topic roll call 14:02:23 Who's present today? 14:02:29 * thomasj present 14:02:29 Present. 14:02:40 * rnovacek is here, but leaving early (to school) 14:02:59 * than present 14:03:34 #info jreznik thomasj Kevin_Kofler rnovacek than present 14:04:20 rdieter_work: ? (we can probably move meeting time, looks like this early one does not fit you very well...) 14:04:35 #topic agenda 14:04:47 Isn't it something like 6AM where Rex is? 14:04:51 * ltinkl present too, ofc :) 14:05:06 * Sho_ too 14:05:21 #info ltinkl and Sho_ present too 14:05:22 tibbs: -rdieter_work- TIME Tue Dec 14 08:05:13 2010 14:06:17 He's most likely on the road to work 14:07:09 thomasj: yep 14:07:21 #topic KDE intergration (oxygen-gtk, GTK+ 3 issues, libreoffice-kde) 14:07:45 So packaging oxygen-gtk is the easy part. :-) 14:07:54 rdieter has prepared an SRPM for review. 14:07:59 Is it now allowed to be packaged? 14:08:00 Ah 14:08:08 Binary packages are available from kde-redhat. 14:08:15 thomasj: They released 1.0.0. 14:08:22 perfect 14:08:30 (And FWIW, the license has always allowed for it to be packaged. ;-) ) 14:08:39 (But now it's officially released.) 14:09:07 The problem there is that so far it's only available for GTK+ 2. 14:09:09 Yeah, i was waiting for the official release after Sho_mentioned i should wait 14:09:16 And F15 will use GTK+ 3 for most stuff. 14:09:28 Reminds me to package qtcurve-gtk3 14:09:28 Which brings us to GTK+ 3 issue #1. :-( 14:09:31 >_> 14:09:45 thomasj: Chances are it won't work anymore with the next gtk3 update (2.91.6). 14:09:52 great 14:09:57 They merged a branch which redid all the theming API. 14:10:07 I've been talking to Hugo (upstream developer) about the GTK+ 3 situation, but he's currently fairly adamant about remaining in bugfixing mode on the GTK+ 2 version for a while, he wants to get a handle on that first 14:10:10 we should take a look to all gtk2 based kde related packages 14:10:12 So all the themes which are already ported have to be ported again, with many more changes. 14:10:26 Awesome.. 14:10:45 we need at least the list of packages we are shipping and that would need port? 14:10:59 anyone taking? :) 14:11:22 * thomasj gets his butt kicked from RL currently 14:11:24 oxygen-gtk, kdebase-workspace (the krdb.cpp stuff), maybe xsettings-kde (it SHOULD work as is, but…). 14:11:27 * jreznik knows - it's risky, this "create list" can end with "port it" :) 14:11:36 And qtcurve-gtk*. 14:11:44 Those are all I can think of. 14:12:30 not so bad 14:12:34 Well, there's Amarok using GTK+ (at least GDK) through libgpod, but it's really libgpod's business what version it uses there, Amarok just has to be linked to whatever libgpod is using (for image retrieval). 14:12:54 nspluginviewer? 14:13:11 AFAIK, nspluginviewer doesn't link GTK+. 14:13:24 It dynamically checks if a symbol called gtk_init has been linked in, and if so, calls it. 14:13:34 (to work around broken plugins which expect it to be called for them) 14:13:45 #info potential issues - oxygen-gtk, kdebase-workspace (the krdb.cpp stuff), maybe xsettings-kde, qtcurve-gtk* 14:16:00 We really need to look into what to do with the kdebase-workspace gtkrc writing code. 14:16:19 Kevin_Kofler: could you provide more info for gtkrc? 14:16:24 I don't know what the right thing to write out is now for gtk3, another copy of gtkrc or the newfangled CSS stuff. 14:17:59 jreznik: call tbzatek in? :) 14:17:59 Kevin_Kofler: we have to check if the code still works with gtk3 14:18:13 ltinkl: yep, I was thinking about asking tbzatek :) 14:18:21 The info is that kdebase-workspace writes out a ~/.kde/share/config/gtkrc-2.0 with some color settings and some other stuff (e.g. gtk-alternative-button-order = 1). 14:18:28 ask him to come here for the meeting :) 14:18:50 And we need to make this work for GTK+ 3 as well. 14:19:07 (The same code also writes a gtkrc for GTK+ 1.) 14:19:08 thanks, just to make it clear for all people here 14:19:44 #info problem with gtkrc - kdebase-workspace writes out a ~/.kde/share/config/gtkrc-2.0 with some color settings and some other stuff (e.g. gtk-alternative-button-order = 1) 14:20:43 hi, sorry I'm late 14:20:46 * jreznik is asking tbzatek right now 14:20:50 rdieter_work: hi! 14:21:03 Since upstream doesn't care about GTK+ 3 yet, if we want to ship oxygen-gtk for F15, it'll likely be up to us Fedora folks to do the port and contribute it upstream. 14:21:15 Kevin_Kofler: yep 14:21:21 Otherwise we'll likely have a GTK+ 3 port for F16 at the earliest. 14:23:00 Kevin_Kofler: so the question is - should we package it now in Fedora if we know that it's going to break/not work soon with Gtk3? 14:23:37 jreznik: rdieter already packaged it, we could ship it for testing purpose? 14:23:47 well, it's in kde-redhat/testing 14:24:05 we can ship the gtk2-based oxygen-gtk for now, for testing. 14:24:07 yes 14:24:18 I'll submit it for review after meeting 14:24:29 Kyril: yes, it can be in kde-redhat but is it worth to put it to Fedora repos and then listen to users complaining it does not work in Gnome? 14:24:54 Likely not worth the stress yet 14:25:09 there's going to be a lot of transition, where fedora ships a lot of gtk2 + gtk3 stuff. 14:26:27 Does Qtcurve is compatible or not with GTK3? 14:26:27 gtkrc seems like still supported but tbzatek is not completely sure... 14:26:48 There's a version which works with up to 2.91.5. 14:26:48 jreznik: nope ^_^ 14:27:02 http://kde-look.org/content/show.php?content=40492 14:27:05 But it likely doesn't work with 2.91.6 anymore. 14:27:08 QtCurve (KDE4, KDE3, Gtk3, & Gtk2 Theme) 1.8 14:27:15 Kevin_Kofler: ok 14:27:21 There's a new theme API in 2.91.6. 14:28:30 http://www.lanedo.com/~carlos/gtk3-doc/gtk-migrating-GtkStyleContext.html 14:28:37 sounds, like we'll just have to wait-n-see, there's a possibility we won't have anything for gtk3 (QtCurve or oxygen-kde) 14:28:41 I will ask upstream later how far they are with the new API (wrt QtCurve-GTK*) 14:29:06 thomasj: thanks 14:29:17 np 14:29:20 I'm sure there are going to be some swear words uttered at the idea of having to port AGAIN. 14:29:33 #info thomasj to ask upstream how far they are with the new API (wrt QtCurve-GTK*) 14:29:44 Kevin_Kofler: that's the life :) 14:30:51 And the last integration topic: getting libreoffice-kde built… 14:31:14 All the necessary code should be already there in the tarballs libreoffice* is built from. 14:31:39 Yep 14:31:41 I guess it needs a few BRs added, some compilation options and a subpackage. 14:32:01 So we just need our libreoffice packagers to actually build it 14:32:06 One of us needs to take up libreoffice comaintainership and finally get that stuff built. 14:32:07 ltinkl: you're openlibreoffice expert here, could you take a look? 14:32:32 thomasj: I think they want one of us to sign up as comaintainers to take care of the KDE stuff. 14:32:33 as the guy trying it last time, probably the most experienced one :) 14:32:42 Kevin_Kofler: that make sense 14:32:58 Kevin_Kofler, that's good news if they accept one of us 14:33:00 (At least that's what they said last time we asked them to enable the KDE stuff.) 14:33:07 jreznik: I could, first I'd ask our LibreOffice maintainers if I could co-maintain that package and create the libreoffice-kde subpackage 14:33:20 ltinkl, ftw! 14:33:20 ltinkl: That'd be great. 14:33:24 Please talk to them! 14:33:28 ok 14:33:52 jreznik: is dtardon still working on it? 14:33:53 #info ltinkl to ask our LibreOffice maintainers if he could co-maintain that package and create the libreoffice-kde subpackage 14:34:04 ltinkl: yes 14:34:18 libreoffice is to replace openoffice for f15? 14:34:33 ltinkl: already replaced in rawhide 14:35:04 ok, I'll see what I can do 14:35:07 and caolan and david still maintaining it 14:35:27 https://admin.fedoraproject.org/pkgdb/acls/name/libreoffice 14:35:59 libreoffice is mostly just the new name for the same thing. 14:36:49 * ltinkl asked for the ACLs 14:37:01 Kevin_Kofler: but just mostly 14:37:10 Though they merged (or are still merging?) the go-oo patchset, ending the ages-long split between upstream OO.o (and Fedora shipping it) and go-oo (which everyone else shipped). 14:37:16 ltinkl: probably it would be better to ask the people before asking for ACLs 14:37:26 yup 14:37:57 they are both offline already 14:38:18 Send them an e-mail explaining what you're requesting the ACLs for. 14:38:49 Kevin_Kofler: +1 14:39:16 It's always better to see an explanation rather than just "ltinkl has requested supercowpower ACL rights on libreoffice". :-) 14:39:41 ltinkl: I can try to help too, if you want some. I built ooo-kde integration way back in the ooo-v1 days. 14:39:58 rdieter_work: great! 14:40:35 #info rdieter to help ltinkl with libreoffice integration 14:40:38 * rdieter_work added reviewing the f15/kde46 feature to the agenda, if we have time 14:40:47 rdieter_work: ok 14:41:14 rdieter_work: another agenda topics - meeting time as it looks like current does not suit you very well, do you agree? 14:41:58 jreznik: sorry, it's just been bad these past 2 weeks, should be better now... however, if we can go even 1 hr later, that would make things a little less stressful. 14:42:28 rdieter_work: np, we can talk about it 14:42:36 So, 15:00 UTC? 14:42:44 everything for integration? 14:42:54 Kevin_Kofler: yeah, something like that. 14:42:55 yup, mail sent :) 14:43:09 " everything for integration?" → Yes. 14:43:16 ok, let's move on 14:43:48 For the time, 15:00 UTC is OK for me. Is the chan free? Do we all have time? 14:43:48 #topic Features/KDE46 14:44:06 https://fedoraproject.org/wiki/Features/KDE46 14:44:10 Kevin_Kofler: it's ok for me - at least in winter (CET) 14:44:41 I whipped that up quickly, based on kde45 feature as a template. I'd like to submit this to fesco asap. 14:44:42 but first - let's finish current topic 14:45:54 rdieter_work: thanks for feature page, I think - it looks ok, just some minor updates are needed now - 4.5.85 etc. 14:46:05 I added 2 potentially controversial items: up'ing spin size (1 gb?), and considering new default phonon backend 14:46:20 jreznik: sure, I'll 4.5.85-ize it after meeting 14:46:31 I think these are more meeting topics than something to stick into the feature page… 14:46:36 and we should finally finish "Fedora KDE Desktop User Guide (draft)" but when :( 14:46:48 Kevin_Kofler: yep 14:47:12 We can always sneak it onto the feature page when we decide it. 14:47:38 14:48:07 so, besides, 4.5.85, anything else obvious to add/update/fix? 14:48:48 rdieter_work: I'll take a look, from first look it seems to be ok 14:48:55 anything else we should add? 14:49:12 Re Phonon, I agree the GStreamer backend is worth another try, the blockers we found back then should be solved. But we need to evaluate carefully what we gain and what we lose (e.g. Dragon Player's Xine-only extra features). 14:49:23 I'm not sure we have another nice feature in our viewpoint... 14:49:37 Re spin size, I think it all depends on how much stuff we have to drop to keep within 700 MiB. 14:49:38 Kevin_Kofler: but QtMultimedia... 14:49:46 fwiw, apachelogger (new phonon dev/maintainer), is high on gstreamer now, not so much about xine anymore 14:50:05 Kevin_Kofler: I think it's really todo for next meeting 14:50:33 jreznik: We already have some GStreamer stuff on the spin anyway, so QtMultimedia using GStreamer isn't a problem either way. 14:50:36 let's move on - should I skip kdevelop again as you were talking about it on #fedora-kde? 14:50:52 to be clear, wrt increasing spin size, *if* we do that, I still advocate providing a cd-size kickstart file for folks who still want it 14:51:01 Kevin_Kofler: but it's easier to maintain it then - from users perspective - just one codecs etc... 14:51:03 (We likely won't be able to get rid of xine-lib either unless we completely disable the support for the xine-only features in Dragon Player at build time.) 14:51:09 rdieter_work: +1 14:51:26 Kevin_Kofler: that's something apachelogger was working on too. 14:51:51 great 14:51:59 If apachelogger (or anybody else, really) finally fixes this xine-only stuff, that'll be great! 14:52:12 it would be nice to have one fedora-kde wide framework in use 14:52:16 (We'll want to backport that ASAP then. :-) ) 14:52:19 presumably xine-free dvd support through gst works now (but that needs kde-4.6 now probably... didn't work with kde-4.5.x + phonon-gst last I tried) 14:52:47 I'll schedule it for next meeting 14:52:53 Well, 4.6 is good enough for F15. 14:53:00 #topic split kdepim-mobility 14:53:04 If it'll need 4.7, we'll have to backport it. 14:53:18 #link http://nucleo.fedorapeople.org/lost.png 14:53:30 kdepim-mobility split seems like a no-brainer to me (for various reasons) 14:53:45 it's just not going to be fun doing it. :) 14:53:53 Yeah, get this pre-alpha stuff the heck out of our kdepim packages! :-) 14:54:08 * Kevin_Kofler wonders if it's worth shipping at all in the current state of brokenness… 14:54:20 Kevin_Kofler: yep, it seems like really very pre-alpha stuff 14:54:22 But in any case it ought to be subpackaged. 14:54:39 frankly, -mobility isn't much better/different than the rest of kdepim-4.6beta. :( 14:54:44 * jreznik is for subpackage at least too 14:54:57 * rdieter_work still hasn't gotten kmail to work at all. 14:55:04 pre-alpha stuff should not be shipped! 14:55:22 we may want a backup plan if kdepim-4.6 is a bust. 14:55:46 than: I kinda agree with you there. I wonder why upstream installs this by default from a tarball called "Beta". 14:55:47 there's ongoing discussion on kde (-release) lists about that 14:55:53 nucleo says most of those apps don't work at all. 14:56:05 #agreed to subpackage -mobility apps, ship it??? 14:56:06 Kevin_Kofler: perhaps mistake! 14:56:30 Well, the backup plan is, Epoch bump our way back to kdepim 4.4.x, again. :-( 14:56:34 upstream definitely intends to ship the -mobility pieces, and that's a focus of a lot of their work atm 14:56:45 Kevin_Kofler: yup 14:57:00 ops, it's going to be mess shipping 4.4.x again... 14:57:02 But at least I would like to give 4.6 a chance. 14:57:16 There's quite some time left until F15. 14:57:19 if possible, there's dispute whether kde-4.6 + kdepim-4.4 is a supportable combination upstream too 14:57:37 a bit of a mess. 14:57:40 rdieter_work: I can make anything work. 14:57:54 it should work, KDE 4.6 + kdepim 4.4 14:57:58 I'll find SOME way to make it work. 14:58:07 ok. 14:58:11 Kevin_Kofler: lol :) 14:58:22 If I have to also ship some old kdepimlibs and pass it off as a higher version number to make other packages happy, I'll do it. 14:58:58 speaking of kdepim-4.6, that probably should be mentioned on the feature page explicitly 14:59:01 Say hello to kdepimlibs-4.6.2-0.1.really.4.5.5.fc15 or whatever. ^^ 14:59:14 (Ubuntu-style version hack.) 14:59:18 and our backup plan(s), if it doesn't work out 14:59:43 Though a better plan would be to patch the other packages to accept the old kdepimlibs if that's needed. :-) 14:59:53 rdieter_work: ok, to mention it there 15:00:13 ok, we are out of time now... 15:00:49 everyone agrees on moving meeting to 15:00 UTC (for now)? bugzappers should be rescheduled already too... 15:01:06 15:00 UTC is fine here 15:01:17 Fine for me too. 15:01:18 if it helps rdieter_work... 15:01:21 +1 15:01:32 +1 (yay) 15:01:32 thomasj, than, ltinkl: ? 15:01:42 it's ok for me 15:01:45 Still and again fine ;) 15:02:27 ltinkl: last chance to say no :) 15:02:40 no prob 15:02:49 ok 15:03:01 #agreed to move meeting to 15:00 UTC 15:03:20 at least for now - let's evaluate it again with summer time... 15:03:26 thanks guys 15:03:28 Bugzappers are rescheduled, yes 15:03:47 #endmeeting