=== asac_ is now known as asac === pochu_ is now known as pochu [08:53] seb128: please sync libtheora from debian/unstable, that NMU contains some useful fixes ;) [08:54] slomo, ok, syncing libgdata too [08:54] good morning everyone [08:54] seb128: great, and totem should be merged then :) [08:54] hey chrisccoulson [08:54] hi seb128 [08:54] did you have a good weekend? [08:55] spent saturday travelling but otherwise good [08:55] you? [08:56] yeah, not too bad. it was actually sunny here all weekend, so i cut our grass and did a bit of shopping [09:24] seb128: hi... i think my Bug #410923 is a dup of my own! there are several dups but each seem to be having their own dups! since they are all assigned to different packages... any idea why apport reports the crashes wrong? should i file a bug in apport? [09:24] Launchpad bug 410923 in libmms "totem-plugin-viewer crashed with SIGSEGV in mms_connect()" [Medium,New] https://launchpad.net/bugs/410923 [09:29] seb128 - i've pushed a change to the gconf bzr branch for the polkit-1 migration [09:29] chrisccoulson, ok thanks [09:30] seb128 - if you're interested in testing it to make sure it works, then the g-p-m capplet uses the gconf defaults, and this seems to work ok [09:31] gconf-editor does not work, but i presume that's because it needs porting too [09:31] ok thanks, gconf-editor never worked for me [09:31] I did try on intrepid and jaunty after getting people asking about it [09:31] I might test later but I've a very busy week, new GNOME, alpha4 and pitti is on holidays I cover for him too [09:32] and I need to land new changes before the alpha image since I'm on holidays next week [09:32] seb128 - that's ok:) [09:32] i'll try and help out where i can this week;) [09:32] thanks! [09:33] seb128: Looking at xsplash, the autogen.sh and gtk-doc.make are missing because they aren't EXTRA_DISTed in the upstream Makefile.am [09:34] lut davidbarth [09:34] Nothing I couldn't fix in patches but this source package looks like it relies of being built out of bzr [09:35] lool, right, I would not be surprised, most dxteam work in bzr and build things from there [09:35] Yup; I think the released tarball doesn't pass make distcheck basically [09:35] lool, I would say it's not a blocker for getting the package in, maybe note those and open bugs after the upload? [09:36] + patch if you feel fixing those but don't think you need [09:36] seb128: Ok [09:36] seb128: Do you think it should use the gnome.mk class? [09:37] I don't see any pos but it might benefit from the gnome.mk default configs [09:37] no need I would say [09:37] that's a simple gtk software [09:37] k [09:37] ie no schemas, etc [09:49] seb128 - i took a look at bug 410241 at the weekend, and also one other bug raised by the translations team. [09:49] Launchpad bug 410241 in gnome-session "Merge debian/po-up/patches.pot templates into the main template" [Low,In progress] https://launchpad.net/bugs/410241 [09:49] for the gnome-panel task, pkgstriptranslations is just stripping the contents of debian/po-up, which isn't used [09:49] chrisccoulson, I talked to david at the airport saturday they can filter those on the rosetta side easily [09:49] so we can just delete it [09:49] ah, ok [09:50] chrisccoulson, we just need to make sure that patch strings are in the po templates [09:50] seb128: Hmm I really can't build it at all from the .dsc [09:50] It doesn't have a configure [09:50] lool, wfm, what error do you get? [09:50] urg [09:50] Oh nevermind I'm stupid [09:50] Wrong window [09:50] ok ;-) [09:50] i was actually wondering if we could filter them out in pkgstriptranslations? it seems to strip any folder containing translations (including the hidden .pc/ folder used by quilt) [09:51] should work too I guess [09:51] that's rather a question for pitti though [09:52] yeah, i'll ask him when he comes back. so, i can close that bug report if they're going to be filtered out on the rosetta side can i? [09:53] yes, just make sure we have those string in the po template [09:53] seb128: salut [09:54] seb128: xsplash dist-check issues? [09:54] seb128 - in the gnome-session case, those strings are merged in the main po template. the extra folders (po.saved and debian/po-up) just contain the original template and the new merged-in strings, which aren't needed === tkamppeter_ is now known as tkamppeter [09:55] in the gnome-panel case, those strings in po-up aren't used in ubuntu, and aren't merged in (but are still there in the source) [09:55] davidbarth, that's mostly ok, lool being picky rather than real blockers [09:55] chrisccoulson, ok good so the bugs can be closed [09:55] thanks:) [09:55] davidbarth, no fusa though which is annoying, do you know where ted put it? [09:59] Laney, hi, thanks for working on the fspot update [10:00] seb128: I'm truly picky in general, and I don't mind packages not passing distcheck, but I don't think it's being particularly picky to reject a .diff.gz used to ship upstream files ;-) [10:01] seb128: no problem [10:01] having some autotools issues if you could take a look though? [10:01] Laney, btw did you look at the package or REVU too? [10:01] Laney, yes sure [10:01] no not yet, it's on the stack [10:01] http://paste.debian.net/43751/ [10:01] lool, right, do you think you can fix it or do you need to wait on the other guys to know what is the source used for packaging etc [10:02] seb128: I'm ok with reporting it and uploading that for now [10:02] Laney, they have been advocated apparently so they just need upload now if you can do that [10:02] seb128: (Which I did) [10:02] oh alright [10:02] lool, ok thanks! [10:02] let me look then [10:02] Almost at the uploading part now [10:02] * lool tries to figure out how to pass -k to bzr bd [10:03] lool, bzr-buildpackage -- -kNNNNNNN [10:03] OK I'll do the second review in exchange for you looking at my pastebin ;) [10:03] Just found that, thanks! [10:03] you're welcome [10:03] Laney, ;-) [10:03] Laney, well desktopcouchdb evolution-couchdb and mago are advocated [10:04] Laney, which means they got the ack and just need dput I guess [10:04] need two acks [10:04] Hmm nice CC lines; I see which teams this is :) [10:04] well, mago has a ack from me but I'm nobody for REVU apparently [10:04] unless the packager is a developer [10:04] seb128: oh you need someone with admin powers to promote you somehow [10:05] Laney, right [10:06] Laney, is your f-spot source online somewhere? [10:06] Laney, my autofoo at not strong enough to debug from pastebin ;-) [10:06] seb128: git clone git://git.debian.org/pkg-cli-apps/packages/f-spot.git [10:06] Laney, thanks [10:08] i didn't push my change to update gnome-doc-utils.make yet though [10:08] but it made no difference anyway [10:09] seb128: chrisccoulson, I've just read it, thanks for having a look at the po-up templates thing. I'll paste the conversation in an e-mail from the thread in the ubuntu-translations-coordinators list where this is also being discussed. [10:14] kenvandine: I'd love if you could review and merge lp:~lool/xsplash/packaging-fixes-for-0.1-0ubuntu1 into the tip of packaging [10:14] seb128: uploaded [10:14] lool, you rock, thanks! [10:14] dpm, ok thanks [10:15] seb128: shall I remove the firefox recommends? [10:15] Laney, feel free to do it yes [10:15] ok [10:23] Laney, not sure what is the issue there, renaming SOURCES to something else workaround the bug [10:23] Uploading mago_0.1-0ubuntu1_source.changes: done. [10:23] Successfully uploaded packages. [10:24] Laney, thanks [10:24] no worries [10:26] Laney, easy way to workaround the bug is to make the if not dynamic [10:26] ie if you don't enable_test in the package build just comment everything in the package [10:31] i'll try it [10:33] asac: Hey there, I tried looking for a TB 3 package (perhaps in a PPA) but didn't find one; did I just look at the wrong places? [10:34] morning everyone [10:34] lut huats [10:34] hello se [10:34] seb128: [10:35] lool: yes. checkout https://edge.launchpad.net/~ubuntu-mozilla-daily/+archive/ppa [10:36] Excellent thanks! [10:36] np [10:36] Ah now I find it in https://launchpad.net/ubuntu/+source/thunderbird-3.0 I had checked thundebird3 [10:36] oh. so launchpad creates packages pages for ppa things now?` [10:36] interesting [10:36] one even can file bugs [10:37] i guess that answers the question: "where to file bugs against those packages not yet in ubuntu for real" [10:37] Indeed [10:38] I'm not sure I like the fact that mistakes in random archives can clutter LP pages but I guess that's ok [10:38] yeah [10:38] asac, hey, did you manage to catch your flight connection? [10:38] the bugs should be removed from all-bug-counts [10:38] seb128: ack ;) ... 10 minutes before boarding i arrived. thanks! [10:38] good ;-) [10:39] asac, still not recommended to open ppa bugs there they will be mixed with ubuntu bugs [10:39] * asac gets reminded to send a complain - but doesnt bother now ;) [10:39] lol [10:39] asac, thanks for the telepathy-butterfly fix btw [10:39] seb128: i think it depends on the ppas ... our dailies exist to catch regressions early [10:39] and for firefox-3.6 we dont have any bug tracker at all ;) [10:40] asac, right but as lool said you don't want bugs from random crack user ppas you don't know about [10:40] yeah [10:40] thats true. we should be able to maintain a whitelist ... or "ubuntu-team ppas" [10:40] right [10:41] actually i dont like the feature that it displays random crack builds on the frontpage either [10:41] "Other versions of 'firefox-3.6' in untrusted archives. " [10:42] that's handy to see what cracks are available but it's annoying that users can be lead to cracks this way [10:43] seb128: yeah. ototh, we have apport preventing that [10:43] so for packages already in the archive this doesnt change a thing. just for new packages ;) [10:45] hmm. i have problems running update-manager in a chroot [10:45] Failed to fetch http://ubuntu.intergenia.de/ubuntu/dists/karmic/Release.gpg Something wicked happened resolving 'ubuntu.intergenia.de:http' (-11) [10:45] thats odd [10:45] i saw the same bug in a fresh chroot [10:45] even with apt-get [10:46] now updated an existing chroot and its also like that [10:46] too bad [10:46] EOL for karmic chroots i guess [10:47] when will mvo return ;)? [10:48] ftp doesnt work any better :( [10:55] lool, could you push your nautilus changes to bzr? [10:57] seb128: I suck; pushed [10:57] lool, so you don't, thanks ;-) [10:57] seb128: can you promote firefox-3.5 to main please? [10:57] asac, ok [10:57] gratias [11:06] seb128: Hey do you know why clutter-gtk (Source) wasn't API versionned in the last upload? [11:06] hello [11:07] seb128: I emailed didrocks but I didn't get a reply [11:07] That was last Thursday [11:07] I think he was going on holiday [11:07] Ah [11:07] lool, no, is there several versions or a need to version it? [11:07] lool, he's on holidays for 2 weeks I think [11:08] seb128: I just seems inconsistent with clutter/clutter-0.10 [11:08] s/I/It [11:08] well clutter is 1.0 now [11:08] not sure if versionning is still useful [11:08] but no I don't know and I've no strong opinion either way [11:09] clutter became clutter-1.0 to keep the 0.8 versions in archive parallel installable; clutter-gtk wasn't I don't care how we fix it but it'd better match [11:09] Will check with didrocks when he comes back I guess [11:09] Debian didn't push clutter-gtk 1.0 yet [11:12] lool, I guess that's because we had clutter-0.8 clutter-gtk [11:12] clutter-0.9 I mean [11:12] ie clutter was already versioned [11:12] looks like those autotools warnings were around in 0.5.0.3 too [11:12] so I won't worry about them [11:12] I'm wondering if it would make sense to drop any versioning now [11:12] Laney, ok [11:14] clean up build-deps and close bugs and then it should be good [11:14] Laney, \o/ [11:35] seb128 - i just saw the nautilus and gnome-panel patches. surely that change would be better in gnome-session? (sending the DBus message on the transition from Panel -> Application phases, when both nautilus and gnome-panel are already loaded), rather than patching both nautilus and gnome-panel to do the same thing? [11:36] chrisccoulson, no clue about those, I've not been discussing nor working on those just doing sponsoring [11:36] chrisccoulson, robert_ancell is on holidays now though [11:37] but yeah that seems to make sense if gnome-session knows when those are done displaying things [11:37] seb128 - ok, no problem. i just had a look at the xsplash code, and it looks quite hacky there. if the message came from the session manager, then it would work with any combination of filemanager or panel. with the current situation, it only works with gnome-panel an nautilus, and requires them both to be patched [11:37] ie I don't know when gnome-session goes to the next stage, are gnome-panel applets still loading, etc [11:38] seb128 - yeah, i'm not too sure about that [11:38] bratsche, hey, are you there? [11:38] bratsche, ^ that's a topic for you I guess [11:39] chrisccoulson, bratsche is writing xsplash, they discussed those changes with robert_ancekk [11:39] robert_ancell [11:39] doesn't gnome-session send a DBus signal when the phase changes? that would be even better, as it wouldn't even need o be patched at all [11:39] ok, no problem, i'll chat with bratsche [11:39] chrisccoulson, well the idea is to display the splash until screen is stable [11:40] ie if gnome-panel still loads applets after being "done" that's an issue [11:40] yeah, i'm not sure when gnome-panel connects to the session manager [11:40] davidbarth might know === MacSlow is now known as MacSlow|lunch [12:01] how can i see if a package is in "auto-installed" mode? [12:06] does anybody already have think about the upnp support in farsight2 and libnice? do you plan to ship it on the cd? ( [12:06] * libgupnp-igd-1.0-2_0.1.3-0ubuntu2_i386.deb (13.9 KiB) ) [12:11] asac, "auto-installed"? [12:12] bigon, no and no idea [12:12] seb128: well. packages that would become autoremove candidates if the package that pulled them in got removed [12:13] e.g. apt-get autoremove [12:13] ah [12:13] there is a flag in some database ;) [12:13] but i cant find it [12:13] bigon: if there is place, it would be good to ship it imho [12:13] so packages won't have to divert from Debian [12:22] how does thunderbird device what app to use to open a file type? [12:22] On a number of machines, jaunty and karmic, I've seen it recently start trying to use Font Viewer for PDFs [12:23] Ng, what thunderbird version? I would say it's an asac's question [12:25] seb128: at least the very latest 2.0 packages in jaunty [12:27] Ng: I think it uses /etc/mailcap* [12:34] asac, aptitude search ~i | grep ^id something around those lines seem to do the trick [12:35] asac, in fact no, is has a A has digit 3 [12:38] grep -A1 firefox /var/lib/apt/extended_states [12:38] good catch [12:39] seb128: by sladen ;) [12:39] we miss mvo around for such questions ;-) [12:39] ack [12:40] seems firefox-3.0 and firefox-3.0-gnome-support are not auto installed ... even though firefox and firefox-gnome-support is in ubuntu-desktop :/ [12:40] waiting for davmor to confirm that on a fresh install [12:42] asac, they are in the autoremove list once you remove firefox [12:43] or rather firefox and firefox-gnome-support [12:43] and ubufox [12:43] asac, "aptitude why firefox-3.0" [12:44] asac, I had to uninstall firefox firefox-gnome-support ubufox to get it there [12:45] seb128: davmor didnt see ubufox in why [12:45] I get it after removing the others [12:46] asac, you might want to have a | firefox-3.5 there [12:46] yeah [12:46] though i have the feeling that doesnt helpü [12:46] because after moving firefox to 3.5 its not autoremove either [12:46] even though firefox is in depends of ubufox [12:46] what is the issue? refuse to change the default? [12:46] smells like a bug [12:47] seb128: no. its not a big issue. its just that i expected firefox-3.0 go to autoremove after moving firefox and firefox-gnome-support meta package [12:47] ok [12:47] will do migration packages at the end of the cycle then [12:48] mvo said it would work. maybe i should make the whole firefox, etc. stuff a recommends for ubufox [12:52] Hi chrisccoulson [12:53] chrisccoulson: I think robert_ancell did it that way because he couldn't find a reliable way in gnome-session to determine when the session was finished loading. It would send the signal too early or something. [12:54] If you have some other ideas on how to do this it would be awesome. :) [12:56] bratsche - yeah, i think you're right (sort of). gnome-session provides a reliable way to know when the session is finished loading, but gnome-panel seems to connect to the session manager too early (causing the next session phase to begin whilst the applets are still loading) [12:56] i had a look at the gnome-panel code, and it connects to the session manager just before entering the main loop, which I think is way too early [12:57] Well, the way I understood what robert said was that gnome-session would signal once it has asked everything else to load.. but not when they're actually done loading. He had a better idea of how to solve this, but it would require some more work: [12:58] Which is for each thing that gnome-session spawns, have a dbus callback signal from that item to gnome-session saying, "I'm totally done loading" [12:59] bratsche - that's already the case i think. gnome-session starts the session in phases, and only progresses to the next phase when all applications in the current phase are done loading [12:59] Oh I see. What are the different phases? [12:59] clients already tell gnome-session when they've loaded, by connecting to it === MacSlow|lunch is now known as MacSlow [13:00] the phases are Initialization -> Window Manager -> Desktop -> Panel -> Applications (or something very similar to that [13:00] Nautilus starts in "Desktop" phase [13:00] and gnome-panel starts in "Panel" [13:00] So it's not trying to do them in parallel? [13:00] the issue is to know if gnome-panel is done actually loading applets, etc when sending the signal [13:00] bratsche, right that's the idea, hey btw ;-) [13:01] Morning seb128! :) [13:01] no, it does them sequentially, to ensure things like the window manager are done loading before running any applications [13:01] Okay, fair enough. [13:01] so, gnome-panel won't even being loading until Nautilus connected to the session manager [13:01] so, with the current situation, you could probably actually drop the patch from Nautilus [13:02] because Nautilus will already be loaded when gnome-panel tells xsplash that it is done [13:02] But you think the signal is emitted before panel has loaded all its applets? [13:02] That's good to know. I'll drop the Nautilus signal fu in xsplash. [13:02] bratsche - in the case of gnome-panel, it connects to the session manager just before calling gtk_main(), which is long before anything is displayed I think [13:03] Oh okay. I'm pulling down the source and I'll try to find this. [13:03] i don't know how early Nautilus connects to the session manager (whether it is before or after drawing the desktop) [13:04] if it connects after drawing the desktop (which it should do really), then the nautilus patch could be dropped [13:05] Do you know where in the panel code it's doing this callback? [13:05] Well, since everything happens sequentially can't we just drop the Nautilus patch altogether? [13:06] Except that some users may not be using panel I guess.. [13:07] bratsche - gnome-panel calls panel_session_init from main() right before gtk_main() [13:07] panel_session_init does the stuff which registers with the session manager [13:08] gnome_client_connect (client); [13:08] I guess [13:09] that's the one:) [13:09] I wonder if it would work to move that until later. [13:09] I'm going to try. :) [13:09] once that has happened, gnome-session then progresses on to the next phase (Application), because there are no other applications starting in the panel phase [13:10] moving it later would work - but it would also stall the session loading until all the applets loaded [13:10] and that might not please some people ;) [13:11] We really should find some way to parallelize certain parts of the session startup it sounds. [13:11] bratsche - that's already mostly the case. these phases are there to make sure that certain things load before others [13:12] eg, gnome-settings-daemon must start before anything begins to draw on the screen [13:12] and the window manager must be started before any windows appear [13:12] i don't know if it is possible to parallelize nautilus and gnome-panel though [13:12] but most stuff starts in the last phase anyway [13:12] and that happens in parallel [13:13] chrisccoulson, bratsche: so [13:13] bratsche - http://live.gnome.org/SessionManagement/NewGnomeSession explains the session management stuff quite well :) [13:13] hi vuntz [13:13] So most of this stuff is new to me, I mostly only know gtk+ very much. So this is kind of cool to me. :) [13:13] Hi vuntz [13:13] very quickly (busy with other things, but I keep being highlighter when you say gnome-session or panel ;-)) [13:14] the panel connects to the SM as soon as it has set the hint about struts [13:15] and that is before the panel or any applets are displayed isn't it? [13:15] this way, when nautilus starts displaying icons on the desktop, nautilus knows where to put icons [13:15] we don't want to wait for applets because it's looooong [13:15] and this would block everything else [13:15] ah, right - that's an important point bratsche - nautilus starts after gnome-panel [13:15] i got that bit wrong :-/ [13:15] So if I can just find the place in the code where all applets are finished loading, I could put robert_ancell's signal there.. [13:15] bratsche: it's trivial [13:16] bratsche: panel_applet_queue_initial_unhide_toplevels() [13:16] bratsche: what would this code do? [13:17] vuntz: So the background here is that we want a splash screen that fades out once the desktop is totally ready to use, we don't want the user to see things loading and such. [13:18] bratsche: full-desktop splash? [13:18] Yes. [13:19] bratsche: here's how I'd like you to do it so it goes upstream [13:19] bratsche: we already have org.gnome.Panel [13:19] create a signal there [13:19] and in panel_applet_queue_initial_unhide_toplevels, send this signal [13:20] that sounds good. bratsche - is xsplash connected to the users session bus? [13:21] (and just want to point out that you should discuss this upstream) [13:21] No, I think right now it's started by gdm.. its own dbus stuff is on the system bus I think. [13:21] vuntz: Thanks very much! [13:22] also, stupid example: [13:22] brb [13:22] gnome-volume-contorl [13:22] will you wait for it to be shown before removing the splash? [13:23] if yes, this means waiting for all autostart apps, but, this might be slower than what you'd like [13:23] the splash will be removed before displaying the volume control i think [13:24] but really, if you want it to do it the right way according to upstream, at least send a mail to d-d-l so that people can comment [13:25] Okay cool, will do. [13:27] vuntz, hey [13:27] seb128: mon alsacien préféré ! [13:27] vuntz, !!! [13:28] vuntz, do you plan to do patches reviews? [13:28] vuntz, libwnck and gnome-panel patches review ;-) [13:28] seb128: Hey, did you get a chance to put that gtk+ patch into Karmic to test gdm against? [13:28] yep [13:28] gnome-panel first [13:28] bratsche, yes, uploaded that this morning [13:28] woot [13:28] * bratsche checks for updates [13:28] I think I've cleaned up my list of completely urgent things to do [13:29] seb128: so if you have a list of patches, feel free to send it :-) [13:29] bratsche, but alex is back and fixed it in git since [13:29] Oh nice! [13:32] ah, the bug on the root window. Cool, was wondering what the fix was ;-) [13:32] hi bratsche - i don't know if i missed any discussion while i was away. [13:33] are you concerned about things like gnome-volume-control and nm-applet starting after the splash went away? [13:33] chrisccoulson, I don't think we are no [13:33] I'm not concerned about them [13:33] cool:) [13:33] vuntz, I would give you bug numbers with patches if bugzilla was responding [13:34] chrisccoulson: you only missed me complaining that this should be discussed on d-d-l ;-) [13:34] seb128: so you'll be happy with the upgrade :-) [13:34] vuntz: What's the right way to do this? I was going to try to work up a patch and post it to d-d-l as I bring this up. [13:34] vuntz, yes ;-) [13:35] bratsche: well, maybe just say "we want to do that, we think we'll be doing it this way. Patch will be done soon, please comment now if you think the method is wrong" [13:35] vuntz, ok, thanks:) we should probably stop talking about gnome-session and gnome-panel now, else you will be ping'd constantly ;) [13:35] chrisccoulson: nah, that's fine [13:36] it's cool to see good activity! [13:36] vuntz, so bugzilla behaves now [13:36] vuntz: Cool, thanks! [13:36] vuntz, http://bugzilla.gnome.org/show_bug.cgi?id=341441 has users who tried the change on the duplicate [13:36] Gnome bug 341441 in general "Changing screen resolution randomizes applets position in the panel" [Major,New] [13:36] so, bratsche - you probably still want the change in nautilus too, as that starts after gnome-panel anyway [13:36] vuntz, not sure why federico closed the bug with patch as duplicate makes harder to spot the change === james_w` is now known as james_w [13:51] hello [13:51] Hi crevette [13:51] hi bratsche [13:57] lut crevette [13:57] salut seb128 [14:09] chrisccoulson, missed you after you responded yesterday. so with this new setup, do I understand correctly that GSD applies the gtk theme and what not to gdm, and pulseaudio allows the hotkeys (and pulseaudio) to work during gdm too? [14:09] I thought pulseaudio ran as the user's uid once he or she got logged in [14:11] superm1 - g-s-d runs in the gdm session now, so that the theme will be applied. i don't know about media keys or pulseaudio in the GDM session though, as some g-s-d plugins are disabled in the GDM session i think [14:11] but pulseaudio is a recommends of g-s-d now because the media-keys plugin needs it for the volume buttons in the user session [14:12] ok. that logically makes sense. i suppose [14:13] your concern is about pulling in pulseaudio on mythbuntu isn't it? [14:13] yeah [14:13] i can't seem to find a way to block it from getting pulled in since gsd is pulled in now for gdm and it's a recommends there [14:14] yeah, i'm not sure what else to say really - gnome is depending more and more on pulseaudio now, and it's getting difficult to run gnome components without it [14:14] is mythbuntu gnome based? [14:14] xfce based [14:14] didn't xfce go to xdm for the login manager, for this reason? [14:15] so maybe you should look at using a different login manager [14:15] xubuntu is still doing gdm [14:16] kdm would add half of kde, so that's not the best of ideas either [14:16] right the recommend changes were good enough there [14:16] but they are fine with having pulseaudio I think [14:16] is xfce using g-s-d though? [14:17] s/xfce/xubuntu [14:17] somehow it looks like xubuntu doesn't have gnome-settings-daemon in their livefs [14:17] (http://cdimages.ubuntu.com/xubuntu/daily-live/20090810/karmic-desktop-i386.manifest) I'm not really sure how they pulled that off. [14:17] maybe that will be the solution then [14:17] superm1 - gdm depends on g-s-d | xfconf [14:18] so, xubuntu is using xfconf rather than g-s-d [14:18] chrisccoulson, ah well then that is a bit odd because xfconf is in the mythbuntu livefs too [14:18]