=== asac_ is now known as asac [00:59] hmm.. I remember reading about packaging gnome-shell for ubu 9.10.. who's doing that? [01:12] kklimonda: I believe its in the Ubuntu desktop team PPA. [01:13] heh, only for karmic.. maybe it's time to update. :) [01:13] TheMuso: thanks [01:17] np [01:17] There is a lot of stuff in karmic that is not very easily backported. [01:18] Or so far as I have seen anyway., [07:02] Good morning [07:04] pitti: good morning [07:04] hey ruslanr === tkamppeter_ is now known as tkamppeter [09:18] hey glatzor [09:21] morning mvo! [09:53] seb128: hi :) please sync orc from debian/experimental [09:53] slomo, hey, what is orc? [09:54] seb128: the successor of liboil, it provides a runtime compiler for a assembly-like language, etc that then creates optimized code for mmx, sse, altivec or what else the machine supports [09:54] ok [09:54] seb128: some gstreamer plugins will probably use it soon [09:56] slomo, synced [09:58] thanks :) [10:13] hey robert_ancell [10:13] hey seb128 [10:13] robert_ancell, how is the spanish GUADEC? ;-) [10:13] It's not all Spanish... [10:14] oh good ;-) [10:23] pitti, hey [10:24] pitti, is "you need a password to unmount a CD you just inserted" a known issue? gvfs or devicekit rather? [10:24] seb128: devkit-disks [10:24] seb128: bonjour [10:25] pitti, ok, danke [10:25] brb trying client side gtk [10:26] seems to work! [10:27] pitti, do you want a bug in launchpad or is that something you are already tracking? [10:27] seb128: gtk> yay [10:27] seb128: bug> it's not known to me [10:27] you want me to open the bug upstream directly rather? [10:27] seb128: please attach devkit-disks --dump output with the cdrom [10:27] seb128: if you would? I'd just forward it upstream anyway [10:28] even if I work on them [10:28] seb128: please tell me the bug#, I'll subscribe to it [10:28] " mounted by uid: 0" [10:28] hum [10:29] ah, that would be it, I guess === sabdfl1 is now known as sabdfl [10:29] seb128: can you please add the "mount" output when the CD is mounted? [10:29] pitti, devkit-disks --dump seems to always display uid: 0 [10:30] it does that too for my usb key which I can eject [10:30] "/dev/sr0 on /media/cdrom0 type iso9660 (ro,nosuid,nodev,user=seb128)" [10:30] seb128: ok, then I'm afraid david needs to look at this (he's usually pretty fast with devkit bugs) [10:30] I'll also take a look at the code [10:30] don't bother I will see that with davidz [10:30] but I guess I'll finish wrangling with gdm first, this week [10:31] I'm wondering if that's the same issue than "gvfs asks for passwords to mount an usb key" [10:31] seb128: I'd like to squash bug 395324 and bug 395591 [10:31] Launchpad bug 395324 in gdm "'These windows do not support "save current setup"....' metacity warning when logging in with gdm 2.26" [Medium,In progress] https://launchpad.net/bugs/395324 [10:31] Launchpad bug 395591 in gdm "Installing GDM 2.26 doesn't update /etc/X11/default-display-manager" [Undecided,In progress] https://launchpad.net/bugs/395591 [10:31] seb128: quite possibly; it seems to get confused about them [10:31] seb128: since usually, when I abort this and try a second time, it mounts fine as user [10:32] well the autospawned try seems to fail [10:32] but clicking on the icon next works [10:32] anyway I will talk to davidz, thanks [10:32] seb128: I'm curious what's this "client-side gtk"? [10:33] pitti, [10:33] "GDK now maintains [10:33] its own window hierarchy client-side, and only uses X windows where [10:33] unavoidable. Some of the benefits of this change are [10:33] - Reduced flicker [10:33] - The ability to do transformed and animated rendering of widgets [10:33] - Easier embedding of GTK+ widgets e.g. into Clutter scene graphs" [10:34] pitti, demo on http://blogs.gnome.org/alexl/2009/06/12/the-return-of-client-side-windows [10:35] or http://live.gnome.org/GTK%2B/ClientSideWindows for details [10:35] pitti, it's basically gdk doing the handling directly rather than using the x apis for that [10:37] hah, nice effects [10:37] indeed ;-) [10:37] another gtk test before uploading brb [10:45] hello [10:49] i took a look at that demo a few days ago - looks awesome :) [11:05] nice, http://live.gnome.org/Tracker/Roadmap [11:05] there might be a 0.7 release within the next month according to the ML [11:05] might be time to get an unstable snapshot in Karmic soon [11:10] seb128: new gtk introducies this gdk window management? [11:10] I had to give my password to mount a usb drive too, fwiw [11:10] or isnt that new? [11:10] http://bugzilla.gnome.org/show_bug.cgi?id=581526 [11:10] asac, yes, 2.17.3 just uploaded to karmic [11:10] Gnome bug 581526 in gdk "XID table corruption from reuse of XIDs, resulting in leak, incorrect window destroyed status ("unexpectedly destroyed"), and crash" [Critical,Unconfirmed] [11:11] asac, client side has been merged to git one week ago and the tarball uploaded to karmic 15 minutes ago [11:11] asac, your bug is probably not due to it [11:11] right [11:11] wonder if its fixed by this though ;) [11:11] dunno [11:12] we will see ;) [11:13] Laney - bug 386699? (sorry if I missed part of your conversation ;)) [11:13] Launchpad bug 386699 in gvfs "Mistakes external USB flash disk for system-internal disk" [Medium,Triaged] https://launchpad.net/bugs/386699 [11:14] chrisccoulson, do you still work on the g-c-c update? [11:14] could be [11:14] I didn't dig into it that much [11:14] thanks though, I'll subscribe [11:15] seb128 - yeah, i'm still looking at that [11:15] chrisccoulson, ok thanks [11:15] sorry, i got sidetracked with the metacity issue and some other stuff [11:15] that's ok, I was just looking at what is outdated to clean the table [11:16] i'll hopefully finish that today, as I finish work at lunchtime [11:20] ah yes [11:20] f-spot can be synced for a sexy space saving now [11:24] seb128 - i notice we patch some stuff in g-c-c to use ubuntu-system-service to apply some gconf settings system-wide [11:24] right [11:25] but gconf already provides a way of doing this, which is currently used by g-p-m [11:25] it seems the functionality is duplicated [11:26] should these be ported to use the gconf-defaults mechanism instead? [11:31] chrisccoulson, talk to mvo, there was already the gconf feature when he added that I think [11:31] it might have not been working correctly or something [11:31] I'm not sure about the specifics [11:32] thanks - i'll try and discuss it later with mvo if he's available. i think it works okay at the moment, as the g-p-m preferences dialog and gconf-editor are using it now [11:33] when I played with gconf-editor previous time that didn't work [11:33] that was a month ago or so [11:33] nothing was actually being set [11:37] Laney, f-spot synced [11:37] seb128: thanks, my requestsync was broken [11:47] seb128: oh, did Debian take our patch for the dir selector? [11:48] indeed [11:49] pitti, yes [11:49] pitti, Laney did a good job to get it in sync with debian [11:49] * pitti hugs Laney [11:50] \o [11:50] chrisccoulson: hi, I'm happy to discuss that with you now (or later :) [11:51] hey mvo - i just had a quick look at the u-s-s source, and I think I've answered my own query [11:52] it doesn't really duplicate functionality like i thought [11:53] ok, cool [11:54] sorry ;) [11:54] np :) [11:55] i think i misunderstood what it did with the keyboard settings - i didn't realise it edited /etc/default/console-setup [11:55] which is obviously quite a lot different to what gconf-defaults does;) [12:29] seb128 - do you know if there are any upstream plans for a graphical tool to configure GDM? [12:34] asac: bug 347972 is fixed "upstream", but open in ubuntu; so that just needs an upload, but is otherwise "done"? [12:34] Launchpad bug 347972 in ubufox "Does not work with Shiretoko Web Browser (Firefox 3.5)" [High,In progress] https://launchpad.net/bugs/347972 [12:35] fta: could you please give me a quick update of bug 387042? this initially looked like "needs an upload", but the most recent comment looks bad [12:35] Launchpad bug 387042 in pywebkitgtk "gwibber doesn't display messages" [High,In progress] https://launchpad.net/bugs/387042 [12:36] pitti: ubufox will be uploaded in time for the alpha [12:36] asac: i. e. "no blocker here" [12:36] asac: thanks [12:37] pitti: pywebkitgtk update isnt needed ... at least not for gwibber [12:37] i updated gwibber [12:37] in archive and it works well [12:37] let me look at bug [12:38] pitti: dont see why its actually filed against pywebkitgtk. i think at some point gwibber wasnt compatible because the pywebkitgtk we had was too new [12:39] asac: ah, it should probably be moved back to gwibber itself [12:39] asac: so this bug was fixed by the newer gwibber snapshot you uploaded? [12:39] pitti: the bug is fixed [12:39] asac: and the recent comment is a separate issue? [12:39] asac: cool, thanks [12:40] pitti: i uploaded latest gwibber because it didnt work at all [12:40] (i had exactly this bug in mind, but couldnt find it because it was not filed against gwibber package) [12:40] let me update it [12:40] already done [12:40] thanks for the heads-up [12:40] done [12:52] re [12:52] hate hate karmic [12:52] it keeps crashing while I'm away [12:52] i was thinking of upgrading this weekend;) [12:52] maybe i'll delay [12:52] seb128: crashing what? x/kernel/gnome? [12:53] seb128: when the screensaver kicks in? [12:53] pitti, dunno what, screen doesn't wake up when I come back [12:53] numlock, etc don't work [12:53] neither do vt switch [12:54] seb128: for some time I had the problem that a DPMS blank would kill the machine [12:54] could be that [12:54] apparently that got fixed a while ago (xorg-edgers, at least) [12:54] can you ssh in? [12:54] or is it completely gone? [12:54] anyway, /me -> lunch [12:54] I just walked away for lunch and when I come back no way to get the machine back working [12:55] dunno I've no other box on to ssh and I rebooted [12:55] Do any of the Alt+SysRq keys not work? [12:55] I will try next time [12:55] dunno, I never used alt-sysrq and I don't know the combinaisons [12:55] Alt+SysRq+t will dump some useful information in to the kern.log if the machine is sufficiently alive [12:56] chrisccoulson: Hi, do you have time to process your motu application now in #ubuntu-meeting? the MC is having an impromptu meeting (finally got quorum) [12:56] geser - i'm currently at work at the moment, so it would be a bit difficult for me [12:57] chrisccoulson, you are chatting on IRC though? ;-) [12:57] ok, I just wanted to ask as you seem to be around [12:57] geser - how long will you be there for? [12:58] we just started [12:58] if you give me 5 minutes, i might be able to find myself a quiet room [12:58] sure [12:59] thanks, brb (hopefully) [12:59] trying gnome-do I'm not sure what people like there [13:00] the first dialog has an ugly blurry icon and it seems a slower run command dialog than the default one [13:00] not very easily discoverable either [13:00] it might take some getting used to [13:02] I don't think it's discoverable at all === agateau_ is now known as agateau [13:19] chrisccoulson1, \o/ for being a motu now ;-) [13:20] thanks seb128:) [13:23] congrats! [13:24] thanks:) [13:24] right, it's time for me to go home now [13:26] seb128: can you confirm packages bdeping on scrollkeeper should now build-dep on rarian or rarian-compat? [13:27] lool, rarian-compat indeed if scrollkeeper commands are really required [13:29] there's no rarian anyway ;) [13:30] seb128: thanks [13:39] hey rickspencer3, still on a weird timezone? ;-) [13:40] seb128: yes :( [13:40] I just installed Empathy, and am using it now [13:40] seb128: any tips? [13:41] rickspencer3: which problem do you have with empathy? [13:41] Zdra: none so far [13:41] rickspencer3: you can ask for help on #telepathy [13:41] just works here, as well [13:41] ah :D [13:41] hey rickspencer3 [13:41] rickspencer3, tips about? start it and enjoy? ;-) [13:41] I'm updating my desktop to have all the stuff we plan to ship with Karmic [13:41] so I've got grub2, ff 3.5, etc... [13:42] Zdra: it seems to work fine, but could use some usability tweaks here and there [13:42] I noticed that telepathy-haze didn't show the facebook account type from pidgin-facebookchat without some hacking [13:42] might be worth patching [13:45] seb128: do you know what roughly happens for session saving if you stop a session? [13:45] seb128: i. e. apparently programs have to register to gnome-session, and the gdm simple launcher doesn't, or so [13:45] s/launcher/greeter/ [13:47] pitti, right, you need to register to the session using smclient I think [13:47] seb128: so we either need to do that, or properly exit the greeter before shutting down the session? [13:48] I don't know the specifics though you might want to ask mvo or vuntz [13:48] ok, thanks [13:48] pitti, if you don't register your program will probably just not be session handled [13:49] pitti: I've fixed the greeter a while ago; I think those fixes should be in git master [13:50] mclasen: ah, thank you; will build and try that [13:50] maybe I can convince someone to roll a gdm release... [13:51] mclasen, could you also convince somebody to review the stack of patches in bugzilla.gnome.org? ;-) [13:51] not so sure about that... [13:52] do you know who could be pinged about that? [13:53] http://git.gnome.org/cgit/gdm/commit/?id=758666242f97ff02c826ee37f2965ac5a828402d \o/ [13:53] the most likely candidate is halfline [13:53] pitti, http://git.gnome.org/cgit/gdm/commit/?id=51443645f2e394c1f138fd285fbbe56962d34779 too maybe [13:53] pitti: yeah, that is the patch [13:54] seb128: right, while I'm at it [13:54] pitti, I mean that might be worth backporting too [13:54] pitti, thanks! [13:54] mclasen, ok thanks [13:55] mclasen: I wonder why nobody else seems to hit gnome bug 572765; I sent a patch for this, but this bug is too obvious to go unnoticed [13:55] Gnome bug 572765 in general "always overrides keyboard layout to US" [Normal,New] http://bugzilla.gnome.org/show_bug.cgi?id=572765 [13:56] pitti: there is some nondeterministic behaviour in keyboard handling :-( [13:58] pitti: the patch makes some sense to me, but it only treats a symptom, no ? [13:58] mclasen: it's not a symptom AFAICS [13:58] I mean, if the users layout combo is empty, there is a more serious issue... [13:58] gdm doesn't have a keyboard selector and always sets $GDM_KEYBOARD_LAYOUT=us [13:58] and even if it had a keyboard selector, why should the default layout be "us"? [13:58] why does your gdm not have a keyboard selector ? [13:59] we have X read the default layout from hal, etc. [13:59] mclasen: I don't know, should it? [13:59] ours does [13:59] pitti, keyboard selection works correctly there [13:59] I only have a locale selector [13:59] mclasen: ok, that's interesting; I'll investigate that [13:59] it needs one, so people can type their password [14:00] seb128: in gdm? [14:00] the combo is only displayed when an user is selected [14:00] but it's there [14:01] seb128: where should that be? next to the locale selector? [14:01] but either way, if you don't use/select this, it shouldn't hardcode "us" and overwrite hal [14:02] in fact ti's a locale selector right [14:02] seb128: okay, *phew*; I already wondered which s3kr1t gdm you are using :) [14:03] mclasen: so, I found http://live.gnome.org/GDM/Screenshots?action=AttachFile&do=view&target=simple-greeter-user-selected-fedora.png [14:03] pitti: we used to have a patch in fedora to pick the system layout from /etc/sysconfig/keyboard [14:03] mclasen: this has a "sessions" selector (which we miss as well, looking) [14:03] pitti, I've locale and session combos there [14:04] pitti: nowadays, we read the system default from hal [14:04] mclasen: right, we have something similar, hal picks the layout from /etc/default/console-setup [14:04] mclasen: so gdm is supposed to read the layout from hal and export it as $GDM_KEYBOARD_LAYOUT? [14:05] mclasen: and wrt. the screenshot, you have a third selector somewhere for keyboard? [14:05] pitti: yeah, there's a third combo [14:05] oh, there's one on http://live.gnome.org/GDM/Screenshots?action=AttachFile&do=view&target=gdm-a11y-dialog.png [14:05] pitti: it is all a bit complicated, the gdm daemon has no X connection, so it can't fall back to the layout that X picked [14:06] pitti: that is why we fall back to hal (where X got its layout in the first place...) [14:06] pitti, http://bugzilla.gnome.org/show_bug.cgi?id=521100 [14:06] Gnome bug 521100 in general "keyboard layout detection and selection" [Normal,Resolved: fixed] [14:07] and all this is only for the initial situation, before the user picked the right layout once [14:07] once he did that, it is remembered in .dmrc [14:07] mclasen: ah, perhaps it's http://cvs.fedoraproject.org/viewvc//devel/gdm/gdm-system-keyboard.patch?view=markup [14:08] yes [14:09] ok, sooner or later that will become obsolete anyway, AFAIK Peter Hutterer is working on a hal -> udev migration in X [14:09] mclasen: thanks for your help! [14:10] np [14:18] seb128: ok, so above patch didn't help for the greeter session bug [14:18] do you get a similar issue if you use compiz? [14:19] seb128: in gdm? [14:19] yes [14:19] could be a bug on the wm side [14:19] (gnome-settings-daemon:26519): GLib-CRITICAL **: g_propagate_error: assertion `src != NULL' failed [14:19] or a gnome-session one [14:19] Fenstermanager-Warnung: Gespeicherte Sitzungsdatei /var/lib/gdm/.config/metacity/sessions/10f5e667d3a443982124723185079248100000265130000.ms konnte nicht gelesen werden: Datei »/var/lib/gdm/.config/metacity/sessions/10f5e667d3a443982124723185079248100000265130000.ms« konnte nicht geöffnet werden: No such file or directory [14:19] that might be related [14:20] "couldn't read saved session file" [14:20] (/var/lib/gdm/.config/ is in fact empty) [15:00] chrisccoulson: congratulations to your MOTU ranks! finally! [15:00] * pitti hugs chrisccoulson [15:01] * chrisccoulson hugs pitti [15:01] thanks:) [15:06] seb128: ah, got the keyboard selector now; missing xklavier b-dep [15:06] pitti, oh, good catch! [15:12] chrisccoulson: congrats :) [15:12] thanks pochu:) [15:27] * seb128 is away for some errants bbl [15:28] why is xchat no longer in the tray when started with the gnome session? === chrisccoulson is now known as chrisccoulson_ [15:32] fta, dunno xchat is an universe software we don't work on it [15:32] I use xchat-gnome [15:32] and other people there is command line utilities or their im client [15:34] * hyperair uses irssi in screen [15:34] i guess it's more a tray issue === chrisccoulson_ is now known as chrisccoulson [15:40] fta: perhaps xchat started before the panel =\ [15:41] hyperair, that's my guess too, but it's a regression [15:41] in that case, it would be because gnome-panel's starting later or something =\ [15:41] either way, i believe the bug would lie with xchat for not automatically detecting the tray's appearance [15:42] e.g. when you kill the panel, and the panel restarts, i dont think the xchat icon comes back [15:51] fta - how do you have xchat start with the session? did you add it to startup applications or is it via a saved session? [15:52] startup applications [15:52] in that case, it's guaranteed to start after the panel [15:53] the panel starts in an earlier phase, and gnome-session doesn't progress to the application phase (where xchat will be) until all stuff in the panel phase finished loading [15:53] at least, it should work like that;) [15:54] if i restart xchat, it appears in the tray as before. [15:54] hmmmm:-/ [15:55] when did it stop working? [16:36] fta: there was a similar bug reported against amule IIRC [16:37] chrisccoulson: not sure, i'd say no more than 2 weeks old [16:50] fta - i wonder if xchat should connect to the size-changed signal on the GtkStatusIcon? I notice that other applications which work ok already do this (eg, network-manager-applet redraws the icon on this signal), but xchat does not [16:51] (i'm just trying to think of why other applications work correctly, but xchat does not) [16:56] hmmm, it's not that [17:05] chrisccoulson, are you able to reproduce? === dpm is now known as dpm-afk [17:08] fta - i haven't actually. but i've been trying to look at the differences between apps which work and which don't [17:08] you're using karmic right? [17:08] yes [17:08] heh. might help if i try with the same version as you;) [17:12] fta - i can reproduce it on karmic too [17:12] so i'm not alone, good :) [17:13] i've got to disappear now, but i might take a look later to see if i can figure out whats going on [17:14] thanks [17:14] you also need to make your inaugral upload ;) [17:15] Laney - i do ;) [17:55] have a great weekend everyone [18:08] Laney: Are you working on a tomboy SRU or do you know who would be? [18:08] lool: what for? [18:10] Laney: In 345166 someone mentions a probable SRU [18:10] bug 345166 [18:10] Launchpad bug 345166 in tomboy "Tomboy tries to load fuse module in Jaunty...and can't" [Medium,Fix released] https://launchpad.net/bugs/345166 [18:11] "" [18:11] "" [18:11] My understanding is that 0.14.2 is going to be in an upcoming Jaunty update. I'm not really involved in Ubuntu packaging, though. [18:11] Not that I know of [18:12] Laney: Is it worth pushing a SRU just for this bug? [18:12] Laney: I don't run tomboy myself, this is via the main sponsorship queue [18:12] lool: I'll have a look in a min, busy ATM [18:14] Ok [18:15] I have a debdiff ready [18:18] superm1: hi. did you ever find the time to check if blueman has any deficiencies over gnome-bt for the test devices you have? [18:18] asac, so i was able to make all of my devices work with blueman [18:19] there were some UI deficiencies I saw where I got confused at times, but everything was functional [18:19] superm1: UI deficicies? compared to gnome-bt ? [18:19] or general? [18:20] asac, well there are a couple of ways to associate the device in blueman, and i wasn't successful with all of them [18:20] i just think it's a few bugs in blueman that need to be cleaned up [18:20] yeah ok [18:20] but compared to gnome-bt, blueman was more functional for me (since it got me DUN too) [18:20] and after enabling the blueman pulse plugin, that actually worked properly for me too [18:21] superm1: if you scratch network support from the comparison [18:21] do you see any benefit of gnome-bt over blueman? [18:21] well, in it's current state no [18:21] the biggest thing probably is knowing upstream's intent for each of the projects though [18:22] and how well they will be supported and responsiveness to bug fixing [18:23] yeah. so gnome-bt guy says there wont be any changes ... but his mail didnt really sound like he was interested in talking [18:23] also gnome-bt needs this new obexd thing [18:23] which sounds messy as we wont be able to replace obex-data-server completely [18:23] oh yeah, file transfer works perfectly both ways with blueman [18:23] I am at GCDS with upstream, what questions do I need to ask? [18:23] forgot about that [18:24] jcastro: with blueman folks? [18:24] the fork that bastien started [18:24] jcastro: i got what i wanted to know from him [18:24] lool: if you have a debdiff for .14.2 that would be great, seb told me he would look at pushing out .14.2 next week [18:25] lool: upstream is keen on getting those fixes to jaunty users [18:25] asac, so the things that gnome-bt would be missing in the current state are: pulse association when pairing, DUN, and file transfer receiving [18:25] all of which blueman does [18:26] superm1: the question is what would blueman be missing [18:26] i am leaning towards using blueman, because a) its more mature and has a nice UI that even my mother would feel comfortable with [18:26] b) doesnt require obex-data-server, which is pretty new and most likely buggy [18:26] err obexd [18:26] i think you guys will need to lean in on the usability folks yet before i'd say it's good for mom and grandma and stuff [18:27] but it's a lot closer than bluez-gnome and gnome-bt [18:27] superm1: i am sure that my mother would feel more comfortable without asking usability folks ;) [18:27] wait, there's another bluetooth thing? [18:27] we can ask usability folks how to further improve it [18:27] right [18:27] jcastro: blueman [18:27] jcastro: blueman [18:27] so there's bluez-gnome, gnome-bt, and blueman? [18:28] blueman, gnome-bt and blue-gnome ;) [18:28] as for what it's missing, it needs to have pulse support turned on by default [18:28] in that order ;) [18:28] otherewise i think it hits the services/features that we were looking for [18:28] superm1: ok. thats a packaging thing i guess [18:28] i took notes and will check them out with blueman upstream [18:29] as for bluez on the foundations side, i mentioned switching to dynamically spawning bluetoothd when an adapter is detected [18:29] i've got a new package ready for that (all the code is finally upstream), but it is having some weird interactions with udev [18:30] it doesn't work if the adapter is plugged in during boot [18:30] superm1: do you have it somewhere? [18:30] but if you plug it in later it's fine [18:30] superm1: do you use dbus activation through a dbus-send or something? [18:30]