=== braiam is now known as guampa_ === guampa_ is now known as Braiam [00:53] a friend of mine had massive problems with gnome 3.0 on his ati 5650 [00:53] :( [01:02] basso: On natty, or oneieric? [01:03] RAOF: it was natty using the gnome 3 ppa [01:05] RAOF: we tried getting new drivers, and turning off alot of options, but we got like a.. blocky tear or that appared everytime gnome shell did something, like using activites, but the artifacts werent there when shell is not used (just using a progam) [01:07] basso: Well, that's not been well-tested on natty, obviously. We might be missing some mesa patches that fix those problems. I don't think Unity runs into those issues. [01:09] RAOF: no, unity worked quite fine on his computer :) [01:09] Atleast its working quite well with nvidia on natty :D [01:11] I'd expect it to work significantly better in oneiric (once that's settled down, of course) - not only will we have a newer mesa, but we'll also have actual testing :) [01:14] having actual testing might be beneficial :) [01:20] At least we'll know when it's borken! [01:21] well, maybe ;-) [01:21] We'll have a fighting chance! [01:23] ^^ [01:23] RAOF, have you looked at all at mesa >7.10.2? [01:23] Not seriously yet. [01:27] RAOF, I was just talking with keithp. It sounds like there's fixes in >7.10.2 we may want for natty, you should give it a look [01:28] Oh, right. You were referring to the stable branch rather than 7.11. I'll take a butchers. [01:28] it might be far too crazy, but an sru of 7.10.3 might be worth considering... depends on how much performance work is in there [01:28] Yeah. [01:29] It's a pity that the point releases aren't exclusively bugfixes. [01:29] actually, I talked with him about that exact issue [01:29] * RAOF pulls mesa while emacs is starting up. [01:30] he said if we are open to shipping it post-release it gives him extra ammo to argue for keeping the point updates to fixes-only [01:31] I explained a bit about the ubuntu sru policies for doing point release updates, he and eric seemed especially keen on that [01:31] If it was fixes-only we could concievably have an SRU policy like GNOME. [01:31] right [01:31] That would be excellent. [01:31] yup [01:31] (Particularly since we could reasonably run piglit on a bunch of testing hardware as the test-suite) [01:32] I didn't see how that “make the testing hardware lab more available” UDS session went, but from browsing the etherpad it sounds like we could concievably get access to spawn piglit on each gpu in there. [01:33] yeah that session got a bit bogged down [01:34] That, plus a bugfix-only policy, would make aggressive point release SRUs pretty palatable. [01:34] yep [01:34] Hey, emacs! Why you crashin'? [01:35] I mentioned to intel what the cert team was doing. I'd love to be able to get that list of hardware that they're testing to Intel, so they have the opportunity to obtain the hardware too if they want [01:35] I find it strange that the hw cert lab might have more Intel hardware available for the Intel developers than, you know, /Intel/ [01:35] :) [01:35] yeah [01:40] RAOF: bryce: hi [01:40] two of my favourite victims^Wpeople [01:40] lifeless, hi [01:40] Oh, oh! Sounds like a bug report in progress. [01:41] Quick! To the coffee machine! [01:41] I'm using natty classic mode w/o effects [01:41] unless he's here to tell us about fixing some of our launchpad woes? [01:41] and I'm getting remains-of-windows-left-behind [01:41] bryce: thats scheduled for Monday [01:41] lifeless, driver? [01:41] so a program closes and its border is left behind on my xterm [01:41] x201s so intel [01:41] sounds familiar [01:42] It sounds like a familiar damage(?) problem? [01:42] Not that I could point to any bug or anything useful like that. [01:45] 777447 775936 782855 768587 622068 [01:46] none of those are good matches, but those are the corruption bugs I know of [01:47] fwiw, I'm running natty classic / no effects on two intel systems and don't see that, but they're dells. [01:47] bryce: You've got to remember, lifeless has an arrandale :/ [01:48] yeah arrandale's pretty fubar [01:48] should I ubuntu-bug it up ? if so what package [01:49] 745112 and 754799 are the arrandale bugs on my list to look at [01:49] As always, it's “ubuntu-bug xorg” :). However, my guess is that the bug's in xserver-xorg-video-intel. [01:49] lifeless, of course, and 'ubuntu-bug xorg' [01:49] just remember xorg. the bot will move it [01:50] and in the dialog that pops up, say, "yes, someone told me to report it" or whatever [01:50] :P [01:51] cyphermox: I hadn't heard of contributing developer before, is that new? [01:54] RAOF, why is it that everyone *except* us has arrandale? seems that's where all the fun bugs live. :-/ [01:54] bryce: We mentioned this at UDS. Have you done anything about asking for an arrandale system? (I know I haven't) [01:56] yeah I mentioned it to Rick. I also asked Keith (well, his manager) to put me on the list for SDP's [02:02] honestly though I got so many WI's from UDS not sure how much HW testing I have the capacity for doing. [02:02] I just thought there were the 3 levels: core, motu, & ppu, I guess I'll need to get an application together [02:04] bryce: Well, that's what the HW cert labs *should* be for, right? :) [02:07] jbicha: not new at all, but it effectively doesn't give you any upload, it's recognizing your contributions at the development level with Ubuntu membership [02:09] oh, no extra prileges then [02:09] well, except for ubuntu membershio === asac_ is now known as asac [02:42] RAOF: whats the 'sensitive info' in the gdm logs ? [02:42] lifeless: I don't actually know; it's concievable that gdm could leak something in to them and not care too much, as they're not world-readable. [02:45] leaks username [02:45] rest looks ok [02:49] RAOF: https://bugs.launchpad.net/ubuntu/+source/xorg/+bug/785497 [02:49] Launchpad bug 785497 in xorg "remnants of window getting left behind when apps quit" [Undecided,New] [02:49] lifeless: Ta. [02:50] I *think* it tends to be the borders [02:50] IMBW so I haven't said that in the report [02:51] RAOF: bryce: ask for an x201s w/8gb and an ssd as a test machine [02:51] RAOF: bryce: they rock :) [03:08] Hm. There's no safe way to trigger the reconfiguration of package $FOO from $BAR's postinst, is there. [03:54] RAOF: aaahhhhhhh no [03:55] lifeless: Well, there kinda is; $FOO could have a trigger, which $BAR frobs. That's no help when $FOO doesn't have a trigger, though :) [04:00] (That doesn't help me, though, where $FOO doesn't have a trigger and doesn't have any good reason to grow one). [05:21] Good morning [05:21] cyphermox: strip-schema.pl is obsolete in cdbs, it's now been done by dh_translations [05:21] gnome.mk now calls that [05:22] cyphermox: uh, strip-schema.pl was meant to be an internal implementation detail -- there are packages calling that directly? [05:23] bigon: uh, that's what the backwards compat pygtk.pth file was supposed to avoid? [06:21] pitti, http://www.resplect.com/?q=node/30 [06:23] I'm sure fedora will be interested to learn that radeon cards use software rendering to draw gnome-shell on their livecd :) [06:24] I was quite surprised by that as well [06:24] I don't quite believe that [06:25] but the memory usage difference is quite significant indeed [06:25] it pretty much seems to be a constant 200 MB overhead, though [06:25] then again, unity actually allows you to find and use your files (*shocking*), which g-s very much doesn't [06:25] so of course all that zeitgeist magic comes with a price [06:25] I *think* something leaks over time, too. [06:25] for sure [06:26] glib/GNOME programs are notorious leakers, as long as people keep using C [06:27] if only computers could keep track of references to an object themselves ... oh, a man can dream.. [06:27] It's surprisingly easy to leak in managed environments, too. [06:27] GS has some sort of file searching capability built in & their app searching is better than Unity's atm [06:28] I'll be interested to test GS then; I'm not entirely sure how I'd improve Unity's app search. [06:28] (As opposed to app *browsing* which is currently pretty terrible) [06:28] I tried the final fedora based live CD, and g-s doesn't even have a way to open nautilus, much less links to your home dir or XDG files [06:29] RAOF: just download the live CD; it comes with a script to put it on an USB stick, so it's very easy to test [06:29] RAOF: http://www.gnome3.org/tryit.html [06:30] and it's quite an enjoyable exercise as well :) [06:30] I don't think using a livecd is a particularly good test; I'd probably be significantly less satisfied with Unity if I'd just tried as a livecd. [06:30] I really like the shell integration of network-manager [06:30] Although I guess for quick impressions it can work. [06:31] it's both informative ("Ethernet: no cable", "Wireless: firmware missing"), and also doesn't show unnecessary clutter [06:31] * RAOF downloads fedora livecd. [06:31] I'm running gnome-shell on Ubuntu now & I'm not sure what would have been wrong with Nautilus [06:32] jbicha: well, it might be entirely possible that this is just an f15 design decision [06:32] from my 10 minutes of playing, the app search is indeed quite similar, also the general concept [06:32] but I'm using the one from ricotz' ppa & I'm not completely sure what version it is [06:33] so it should be pretty easy to use the other if you know one already [06:33] jbicha: I also discussed that with some GNOME guys the other day, and file search is planned for 3.2 AFAIK [06:33] yes, GS & Unity are rather similar, I think the notification/system tray thing is one of the big diffs [06:33] * pitti waves to seiflotfy :) [06:34] *nod* [06:34] the version I'm using might be a dev git version [06:34] 3.0.1+git20110512 , maybe it's lying and is actualy a pre-3.1 [06:35] jbicha: OOI, what do you like more in gs app search? [06:35] right now I wouldn't know how to make it better in unity, short of adding loads of extra aliases to programs [06:35] but I can open pretty much all my favourite apps with just windows, one letter, enter [06:35] searching for "office" in Unity doesn't display all the LO apps and it should [06:35] or windows+number in the launcher bar [06:36] jbicha: ah, that might be a case for aliases, if desired [06:36] GS can still find results even if you don't start at the beginning of the word [06:36] RAOF: I agree, I hope app browsing gets some love this cycle. [06:36] jbicha: heh, fun -- searching for "libre" does :) [06:36] Unity used to briefly but their algorithm didn't weight things correctly & it was disabled, I think [06:36] searching for -f still finds d-feet, I don't think that's a general limitatino [06:37] interesting -- "fox" doesn't find firefox either [06:37] there's certainly some cutoff point here [06:37] if I search for "f" I want apps starting with f, not apps having an f anywhere [06:37] Yeah, I think it splits on words - which is why -f works, because ‘-’ is a word-splitter. [06:37] and I hate the category switcher in 11.04, the one in 10.10 was better or GS's implementation works [06:37] but for more than say 3 letters it should do a full string match [06:38] jbicha: full ack on that; it was brought up at the design session, should get a lot better indeed [06:38] bbl [06:38] but I expect the dash will be better in 11.10 [06:39] I'm actually looking forward to the real interesting stuff, like searching through music/photo tags [06:39] windows + "floyd wall" + enter -> default music player starts playing the album [06:40] Just like Do! :) [06:40] I Like GS's workspace management. [06:41] For example, we could add an X to the corners of the windows in scale mode. [06:41] I'm quite comfortable with middle-click to close in scale, but a friendly “close” button would indeed be more friendly. [06:43] I didn't know you could close by middle clicking at first, I found out after trying to switch to my browser a few times only to find it asking me if I want to close the window. [06:45] * RAOF hugs his schroots. Easy rollback for upgrade testing! [08:41] argh, scrolling the email folders list in evolution3 is really slow [08:41] clearly a gtk3 regression here :/ [08:41] is that known? [08:43] Evolution's installable on oneiric? :) [08:44] using natty gnome3 ppa [08:44] evolution just got uploaded [08:56] morning [09:04] RAOF: I think cyphermox did that...was working on it during UDS.... [09:05] Well, upgrading one of my systems to oneiric this morning made evolution go away :). I guess it could have been temporary. [09:05] good morning everyone [09:06] hi chrisccoulson [09:07] hi rodrigo_, how are you? [09:07] Hey there chrisccoulson! [09:07] RAOF: I don't think I'll be upgrading for just a bit yet...I mean, maybe the wife's machine just because [09:07] jasoncwarner1 - you're not running oneiric yet? ;) [09:08] hi RAOF, how are you? [09:08] pitti: did you see LP #785323? [09:08] Launchpad bug 785323 in tomboy "Failed to install tomboy 1.6.1-0ubuntu1 - /usr/share/gconf/schemas/tomboy.schemas:1: parser error : Document is empty" [Medium,Fix released] https://launchpad.net/bugs/785323 [09:08] chrisccoulson: Fridayed! [09:08] I'm trying to reproduce with the pkgbinarymangler testsuite, but it takes a loooong while to run now [09:08] and it fails when one uses ecryptfs :-) [09:08] self.assertTrue(os.path.getsize(png_ship) < os.path.getsize(png_orig)) [09:10] hey rodrigo_ jasoncwarner1 [09:10] hey RAOF [09:10] hey chrisccoulson [09:10] morning seb128 [09:10] Good morning seb128 [09:10] hi seb128, how are you? [09:10] how is everybody? [09:10] oh, unity didn't like me docking my laptop :( [09:10] hi seb128, jasoncwarner1, RAOF [09:10] I'm fine [09:11] i've only got a small area drawn in the bottom left corner of my primary monitor ;) [09:11] brb [09:11] is anyone using the new evolution? [09:11] got back a background and a theme for gtk3 application in oneiric after fiddling a bit around [09:11] that is, is it safe to use it? [09:11] rodrigo_, not yet, should we stay away from it? [09:11] It's installable now? [09:11] seb128, I hope not, I really want to upgrade to test the improvements that have been made [09:12] RAOF, seems so, let me see [09:12] rodrigo_, do it, we will fix it if it's broken [09:12] seb128, yeah, let me finish reading my mail just in case :) [09:13] RAOF, it is installable, yes, it just removes all the 2.32 libe* packages [09:13] Hm. amd64's evolution still requires e-d-s < 2.33.0 here. [09:13] lool: hm, I use ecrytpfs as well, and it does succeed here [09:14] lool: seems this was fallout from directly using the old strip-schemas.pl from cdbs? [09:14] bonjour seb128 [09:14] hey pitti, how are you? [09:14] seb128: quite fine, thanks! [09:15] * pitti cut a small tree yesterday and built yet another bookshelf, making progress :) [09:15] pitti, yeah, strip-schemas.pl was used from the source since the package is not using cdbs and there was no dh equivalent [09:15] pitti: well with ecryptfs /home/lool, test/run failed with this png file check [09:15] ISTR the dpkg testsuite also failed [09:15] pitti: This is ecryptfs on top of ext4 BTW [09:15] lool: weird; exactly the same for me here, ecryptfs ~ on top of ext4 [09:16] * pitti runs it again right now on latest oneiric just to b sure [09:16] I'm trying again too, might have been a race that I only got once [09:17] I also get warnings related to file descriptors which remained open === tkamppeter_ is now known as tkamppeter [09:21] Before I lose it, http://paste.ubuntu.com/610495/ is the output I got on the first failure [09:25] seb128: oy [09:25] seb128: tomboy's build triggers many Unknown option: with [09:25] lut lool, ca va ? [09:25] seb128: bien et toi? :-) [09:25] lool, oui ca va ;-) [09:25] seb128: I think it's a matter of passing --with foo instead of --with=foo [09:25] lool, "known issue" I think and it's coming from Debian [09:25] Ok [09:25] check with Laney [09:26] he's the maintainer [09:26] lool: hm, still runs fine here [09:26] Yes, it's coming from Debian. [09:26] lool: the open fds are a rather silly new python 3 feature [09:26] It's rather annoying; it comes from including cli.make. [09:26] pitti: failed again here, with a fresh apt-get source; test/run [09:26] you can safely ignore them [09:26] lool, what do you try to do on tomboy? [09:26] seb128: extract the .schema which breaks with dh-translations [09:26] lool: does "debclean" succeed, or does it complain about missing buil deps? [09:26] lool: i. e. do you actually have optipng and advcomp installed? [09:26] lool, is it still broken with mterry's upload? [09:27] seb128: this is with 1.6.1-0ubuntu1; I guess it might not be the newest as publisher is down [09:27] lool, no, mterry fixed it in 0ubuntu2 [09:27] pitti: oh gosh, indeed, a missing bdep; sorry about that :-( [09:27] * pitti hugs lool [09:27] pitti: the failure looked to ecryptfsish that I blamed it immediately [09:28] lool: the failure basically just means "it didn't reduce the .png" [09:28] lool: ecryptfs shouldn't affect (visible) file sizes -- did you see such a case somewhere else? [09:28] pitti: Some months ago, the dpkg testsuite wouldn't pass if I'd build it under my home [09:29] much longer before that debootstrap wouldn't work in my home, but that one was fixed in the kernel [09:29] this was a runtime dpkg check [09:30] jasoncwarner1, seb128, kenvandine, chrisccoulson: FYI, I only do 3/4 of a day today, and leave at 11:30 UTC (have some bureaucracy appointments in the afternoon); I'll do some work tomorrow morning, so if you want me to work on versions.html bits, please tell me [09:31] pitti, work on version is always welcome ;-) [09:31] i might do some work from that today as well :) [09:31] seb128: working on udev now, will grab pygtk and pygobject as well, ok? [09:31] pitti, do you want some guidance on what to do? [09:31] it feels like it's been ages since i did some desktop work! [09:32] pitti, excellent [09:32] chrisccoulson, welcome back! ;-) [09:32] heh :) [09:32] seb128: after that, some guidance would be appreciated, yes; but above should keep me busy for a while right now [09:32] 06:23 < pitti> bigon: uh, that's what the backwards compat pygtk.pth file was supposed to avoid? << well gnome-python-{,desktop,extras} are not all using dh_python2 it will breaks [09:32] pitti, well consider "1- just pick anything you like, 2- python*, 3- things in light orange (sync in debian but both outdated)" [09:32] bigon: what breaks in particualar? [09:33] pitti, orange -> totem-pl-parser libsoup libgnome-keyring gnome-keyring libgail-gnome consolekit [09:34] seb128: ah, the "fix in debian" bits [09:34] pitti, we have few people who can do updates in debian so those are less likely to get done, we could also wait on the #debian-gnome guys to do them though [09:35] pitti, but otherwise we are mostly through the first round, a group of updates will be blocked until we get the indicator stack and unity on gtk3 [09:37] seb128: don't we also need to lift the remaining gnome-* bits to 3.0? -panel, -menus, etc.? [09:38] pitti, we can't [09:38] pitti, gnome-panel needs to the on the same gtk version than the applet, which means same version of gtk for the indicator, which means same for unity [09:38] ah, I see [09:38] gnome-menus goes with gnome-panel [09:39] pitti, well I guess otherwise you can start tackling gnome-icon-theme [09:39] Cool, the dpkg testsuite passes under ecryptfs now [09:39] since you wanted to look at cleaning it [09:39] and on tomboy, it is likely that mterry's upload fixes it; I ran dh-translations by hand, and it didn't empty the .schema [09:40] lool, ;-) [09:41] I shouldn't be looking into issues when latest packages have not been published :-) [09:42] lool, you should start by looking at the launchpad package page before starting debugging rather ;-) [09:43] seb128: I did, but I failed to notice that I was installing an old version [09:43] I even reassigned some dups to the bug :) [09:44] ok, evolution 3.0 seems to work ok [09:45] eagerly waiting for the mirrors to mirror again, to try it [09:47] seems as slow to start as before, but it feels a bit quicker [09:48] well, it had to migrate all mail to the new format, Maildir, so will need to restart it to see if it's indeed slow to start [09:50] indeed, expunging is fast now! [09:53] it's because cli.make sets DH_OPTIONS [09:53] possibly compat 8 fixes that? [09:54] pitti: for example http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=625657 could happends [09:54] Debian bug 625657 in python-gnome2 "AttributeError: 'module' object has no attribute 'PARAM_APP_DATADIR'" [Grave,Open] [09:55] bigon: hm, so /usr/lib/python2.[67]/dist-packages/gtk-2.0-pysupport-compat.pth seems to fail in some cases :/ [09:56] I tried with with a few modules which install stuff in /usr/lib/pymodules/python2.7/gtk-2.0/, and it worked for them [09:56] but it was a hack from day one, so moving everything over and adjusting dependencies indeed sounds like the better long-term solution [09:57] in the case of the bug, some of the gnome-python where switched to dh_python2 the other where not [09:57] pitti: well this will requires to add Breaks on a lot of packages for lot of packages [09:58] bigon: so many? I thought there were only a magnitude of 10 which install stuff into /usr/lib/pymodules/python2.*/gtk-2.0/ [09:58] yes but they have to breaks each others [09:59] pygobject has to break them, and the dh_python2 converted packages depend on >= converted pygobject, yes [09:59] well not only [09:59] that's why I added that .pth file to avoid this mess [10:00] all the pkg generated by gnome-python gnome-python-extras and gnome-python-desktop also have to breaks each other [10:00] if one of them is not using dh_python2 it breaks the import for all [10:00] IIRC [10:00] oh, han gon [10:01] s/all/some [10:01] bigon: was the problem that the reporter installed a newer python-gnome-something with dh_python2, didn't yet upgrade to a newer pygobject with dh_python2? [10:01] no [10:01] wait [10:02] I was able to build vinagre successfully with Debian's gtk-vnc so I just have to figure out what we did to break gtk-vnc [10:02] pitti: it has python-gnome2 using dh_python2 and python-gtk2 and [10:03] python-gconf still using pysuport [10:03] ah, so perhaps python only considers the .pth file for the first time a namespaced module is imported, and then caches the path [10:03] dunno [10:05] hey jbicha [10:06] pitti: for example I had to add a breaks between python-gtk2 and python-gtkgl [10:06] jbicha, weird, it build fine there with the oneiric one [10:07] seb128: what do you mean? it didn't build fine for me as vnc support didn't work [10:08] jbicha, well for some reason I don't get the issues you described the other day there [10:09] well, it failed in an oneiric PPA too [10:13] ok, maybe it's some of the packages I didn't update yet [10:56] seb128: so pygobject and pygtk were easy, we could sync both (done) [10:56] pitti, \o/ [10:58] * pitti moves to consolekit, glibmm2.4, and gnome-keyring [10:58] seb128: btw, is there much point adding a (pitti) everywhere on the pad? the colors already show the author? [10:59] pitti, if you know who is what color which is not obvious [10:59] not obvious if the person is not connected [10:59] oh, the top right field only shows connected users? [10:59] ok, keeping it then [10:59] right [11:00] pitti, btw not sure if " - python-gtk2 suggests not depends on python-numpy" was still needed? [11:01] ah blergh, that was the thing that escaped me [11:01] debian runs dh_numpy in their rules [11:01] will that add a depends? [11:01] I'll re-merge that then [11:01] pitti, well I'm not sure I didn't try to build it [11:02] seb128: but python-numpy is in main, and we already had it [11:02] the debian->ubuntu diff was [11:02] - python-numpy-dbg (>= 1:1.4.1-4~), [11:02] + python-numpy-dbg (>= 1:1.0.4-1~), [11:02] pitti, it's not on the CD though and it's not trivially small [11:02] which seemed fine to me [11:02] pitti, well, we moved the Depends to Suggests back then but debian moved to use dh_numpy in between [11:03] which might lead to the depends [11:03] seb128: it doesn't source-depend on numpy, that's why it escaped me apparently [11:03] apparently ${python:depends} drags it in [11:03] I'll try to fix that in Debian [11:03] well likely dh_numpy? [11:03] it's used in the rules [11:03] pitti, thanks [11:03] well in theory I think numpy is used in some parts of pygtk [11:04] but doko argued that the few users of that part of the code in the archive should have the depends [11:04] right [11:04] to avoid brining numpy on the CD [11:04] rodrigo_, wb [11:04] rodrigo_, is something blocking the g-c-c upload? did you do the accountsservice mir? [11:04] seb128, hi seb128 :-) [11:05] seb128, no, just finishing the patch rebasing and will do upload + mirs [11:05] ok [11:05] let me know if you need a review [11:05] ok [11:28] seb128: ok I think I figured out vinagre, it may have just been missing a libgvnc dependency, so merge proposed [11:28] jbicha, great, thanks, I will review it in a bit [11:32] I got a funky bug in natty banshee refuses to display and is taking out compiz at the same time by the look of it [11:32] seb128, https://bugs.launchpad.net/ubuntu/+source/apg/+bug/785682 and https://bugs.launchpad.net/ubuntu/+source/accountsservice/+bug/785680 <- are they ok? [11:32] Launchpad bug 785682 in apg "[MIR] apg" [Undecided,New] [11:33] seb128, pushing now my changes for a last review, if you want [11:35] rodrigo_, not really ;-) [11:35] rodrigo_, did you read the mir documentation? [11:36] seb128, yes [11:36] rodrigo_, https://bugs.launchpad.net/ubuntu/+source/libpeas/+bug/782958 [11:36] Launchpad bug 782958 in libpeas "[mir] libpeas" [Low,Fix released] [11:36] is a nice example [11:36] ok [11:36] or https://bugs.launchpad.net/ubuntu/+source/seed/+bug/782972 [11:36] Launchpad bug 782972 in seed "[mir] seed" [Low,Fix released] [11:37] rodrigo_, but basically you need some details and specify you checked the build-depends, depends, how well the package is maintained and the security record [11:38] ok, sorry, read the wrong page it seems [11:38] rodrigo_, https://wiki.ubuntu.com/UbuntuMainInclusionRequirements [11:38] https://wiki.ubuntu.com/MainInclusionProcess [11:39] yeah, just read the 2nd one [11:39] rodrigo_, ^ read those, but if you take one of the bugs jbicha wrote they are nice examples [11:39] rodrigo_, mind if i upload http://paste.ubuntu.com/610545/ for g-s-d? [11:39] rodrigo_, well the first point has an url to the other page ;-) anyway just copy what jeremy did [11:39] it stops gsettings-data-convert from crashing on every log in :) [11:40] chrisccoulson, does it need a schema entry, or it's already in some schema? [11:40] fixed made banshee fullscreen issues gone away odd though [11:41] rodrigo_, no, that key is not listed in a schema anywhere (and it doesn't appear to be used anywhere in g-s-d either) [11:42] chrisccoulson, but then apps that access it will fail [11:42] rodrigo_, seems you forgot to subscribe ubuntu-mir to those bugs as well, see 3. on https://wiki.ubuntu.com/MainInclusionProcess [11:42] rodrigo_, but maybe wait to have the bugs updated before doing that ;-) [11:42] rodrigo_, are any apps accessing it? if they access it, then they will fail already (as gsettings-data-convert does) [11:43] seb128, yes, was waiting to see if it's ok, will do as soon as I fix the bug description [11:43] seems like we will need gnome-icon-theme-symbolic as well btw [11:43] chrisccoulson, where does it come from then? [11:43] quite some icons are displayed as broken icons without it [11:43] rodrigo_, no idea. it's just listed in gnome-settings-daemon.convert :) [11:43] seb128, right [11:44] chrisccoulson, ok, upload it then, I'll check upstream if we can remove it [11:46] rodrigo_, thanks [11:48] oh, it's so great to upload a package which isn't 70MB :) [11:48] :) [11:49] hi [11:50] I'd like to test the gnome-shell environment and have used the gnome3-team ppa, but I'm finding some problems [11:50] I had some visual glitches and just killed the session, but now I get a clean desktop without any shell elements, just a background and a terminal window [11:50] do you know which process may be missing? [11:53] libpolkit-gobject-1-0 is in main, right? [11:57] rodrigo_: yes [11:58]