=== eeejay_away is now known as eeejay [00:17] any xorg people around? [00:18] (#ubuntu-x) [00:18] thanks dtchen === astechgeek is now known as techgeek === asac_ is now known as asac [00:39] huh. stuff like https://bugs.launchpad.net/bugs/494259 seems like overkill. [00:39] Ubuntu bug 494259 in pulseaudio "pulse audio remembers muted applications without notification" [Undecided,New] [00:40] that would seem, IMO, to be UI clutter (apologies for the namespace collisions) [00:40] dtchen, can I set that to "wishlist/confirmed"? [00:41] rickspencer3: certainly. [00:41] rickspencer3: it just seems like the opposite of what stream memory was designed to do [00:41] yeah [00:41] i.e., I can see use cases for both remembering and forgetting [00:42] well, I guess alerting the user about the setting some how wouldn't be aweful, but just seems you have bigger fish to fry [00:43] done [01:00] ArneGoetje, good morning [01:00] rickspencer3: moin [01:00] ok [01:01] = Arne = [01:01] good '''morning''' [01:01] == call time == [01:01] * is about now [01:01] * should we try empathy? [01:01] rickspencer3: yes [01:02] rickspencer3, just what is commented on the bug [01:08] chrisccoulson, are the issues people are hitting with csd captured in bug #491521 so far as you know? [01:08] Launchpad bug 491521 in ayatana-ubuntu "Decorations and RGBA" [Medium,In progress] https://launchpad.net/bugs/491521 [01:09] I ask because seemed you were discussing with seb128 in #ubuntu-devel [01:09] rickspencer3 - comment 12 describes the issue we were discussing earlier [01:10] chrisccoulson, right, so gtksudo is busted [01:10] is that the main thing? [01:11] well, I guess #11 isn't great either [01:11] that's the main thing, in addition to the weird artefacts in nautilus [01:11] thanks chrisccoulson [01:11] ah, ok, comment #11 might just be related to the artefacts we see [01:11] nautilus is probably working ok, but the contents on the screen aren't getting refreshed correctly [01:20] rickspencer3, gksudo is a real problem [01:20] that is a blocker for sure imho [01:21] right [01:21] * kenvandine must not have used thar recently [01:21] that [01:21] robert_ancell: ping [01:22] kenvandine, I think the artifacts are to be expected, but until the gtksudo thing is resolved, probably shouldn't push it into the distro [01:22] yeah [01:22] rickspencer3, we'll see what others report [01:23] thanks kenvandine [01:23] time for me to log off [01:23] see everyone tomorrow [01:23] later! [01:23] vacation [01:23] :) [01:23] kenvandine, see you Thursday! [01:23] later! [01:23] have a fun day tomorrow [01:24] i'll try [01:36] ArneGoetje, hey [01:37] robert_ancell: about the simple-scan problem... shall we discuss it here or in private chat? [01:37] ArneGoetje, here is good [01:37] robert_ancell: ok. === ___bjf is now known as bjf-afk [01:38] robert_ancell: my scanner is an Epson 3490 Photo and is detected by Xsane under snapscan:libusb:003:004 [01:39] robert_ancell: I remember I had to do some fiddeling with udev and I don't rember what else on my system to get it working. [01:39] robert_ancell: scanning works with Xsane [01:40] ArneGoetje, ok, what happens in simple-scan? [01:40] robert_ancell: simple-scan detects the scanner as "Epson", but when clicking on 'Scan', I get the error message "Failed to scan\nUnable to connect to scanner" [01:41] ArneGoetje, what version are you running? [01:41] robert_ancell: after that even Xsane cannot talk to the scanner anymore [01:41] robert_ancell: 0.6.4 from your ppa [01:41] ArneGoetje, yeah, sane is really bad at locking the scanners, you have to wait for a timeout [01:42] ArneGoetje, ok, you can upgrade to 0.7.0 but they will probably not fix the problem (https://launchpad.net/~robert-ancell/+archive/simple-scan) [01:42] ArneGoetje, can you run simple-scan from a terminal like this: [01:42] $ simple-scan 2>&1 >simple-scan.log and paste that into paste.ubuntu.com [01:42] ok [01:53] robert_ancell: after upgrading to 0.7.0 and resetting my scanner, it displayed two entries for Epson Scanner, one was apparently the old usb address 003:004 and the other one the new one after the reset 003:005. Clicking Scan with the first device failed, but with the second one it succeeded. [01:56] ArneGoetje, that makes sense - it remembers the previously scanned devices so there's no delay on startup. When you re-open simple-scan the old one will dissapear [01:56] ArneGoetje, good news then! [01:56] robert_ancell: ok, anyways, it works now :) [01:57] robert_ancell, when will simple-scan land in main? :-D [01:57] kenvandine, a2 for universe at least :) [01:58] * robert_ancell is setting up two new computers at the same time - a dell mini for testing and a new dell studio to replace my old laptop [01:59] robert_ancell: how is 'continuous scan' supposed to be used? [02:00] ArneGoetje, yes it will probably be renamed - it's for automatic document feeders - it keeps scanning until it runs out of paper [02:00] robert_ancell: ah... [02:02] robert_ancell: feature request: 1. scan multiple pages (even on flatbed scanners) and bind them into a pdf, 2. pipe the scanned text through a OCR software. :) [02:02] 1. already works [02:03] 2. not in scope for lucid, but plan to have that feature in the future [02:03] robert_ancell: yay! === astechgeek is now known as techgeek [02:47] robert_ancell, continuous scan = "Scan from document feed" [02:47] ? === eeejay is now known as eeejay_away [03:13] rickspencer3: yeah, name is awful. Unfortunately I don't seem to be able to get SANE to tell me if the source is document fed [03:13] rickspencer3: the whole UI is up for changing before 0.8 === techgeek_ is now known as techgeek [03:39] kenvandine. any workarounds for the gtk testing issues? === robert_ancell_ is now known as robert_ancell === bob is now known as Guest26677 === bob__ is now known as Guest34652 [05:37] Good morning === ArneGoet1e is now known as ArneGoetje [05:53] dobey: hm, so what exactly causes ubuntuone-client-applet to be started with my session automatically? the nautilus extension? (I don't have an autostart .desktop anywhere) === eeejay_away is now known as eeejay === eeejay_away is now known as eeejay [07:07] hey pitti [07:07] the gdm new default session seems to work well :) [07:08] yay didrocks [07:09] pitti: I'm wondering about setting the default session now. [07:09] I guess the rational is: [07:09] There is already a default session -> don't touch [07:09] There is none -> set this session related package as default [07:11] also, I think I'll write a dummy tool in gdm package to enable that. (called in *-default-settings packages postinst) [07:11] The thing is that gdm is only interacting by dbus (there is no way to ask the daemon to reload the configuration without stop it and reload it) [07:11] the this little tools would: [07:12] 1. Try to update (if needed) the setting by dbus [07:12] 2. if it doesn't work, the daemon is probably stop, change the value directly in the file [07:12] make sense? [07:13] oh, there's no reload any more, as in the old gdm? [07:13] didrocks: I had hoped that this would be done by the greeter, i. e. next time it starts it reads the new custom.conf [07:13] but if not, then oh well [07:13] no, I tried and it failed. I asked upstream and they confirm [07:14] so the little tool as to be a little bit smarter, not a big deal :) [07:14] has* [07:14] didrocks: so if you install UNE image, it should of course be the default session, but if you just install the package, it shouldn't automatically become teh default session IMHO [07:14] OTOH, xubuntu-/mythbuntu-default-* have to change the default session [07:15] I think the latter just need to change custom.conf, and probably whatever was installed last wins [07:15] for UNE, the image build process could change custom.conf [07:16] Don't you think we can integrate that in a tool in gdm package? [07:16] just to do the right thing (as changing directly custom.conf needs a gdm restart to take the new value into account) [07:17] can be quite easy, an option --keep-default to the tool involved in changing the setting [07:17] we will call it that way in une post-install [07:17] and don't use that option in xubuntu, mythbuntu... [07:21] didrocks: sure, sounds fine [07:21] /usr/lib/gdm/set-default-session or something like that [07:21] thanks for the name suggestion :) [07:22] I'll try to give some love to that today :) [07:31] * pitti unbreaks desktop installability and CD builds due to the libmetacity breakage [07:34] good work, appreciable before alpha1 release ;) [07:34] and necessary :) [07:35] yay ^^ [07:46] Amaranth: FYI, I upgraded lp:~compiz/compiz/ubuntu to bzr 2.0 (couldn't push any more, and it's easier to upgrade than to downgrade) [07:47] meh, compiz is FTBFS due to some KDE changes [07:52] hello [08:03] * pitti weeps [08:03] gnome-python-desktop is FTBFS, too [08:06] gloups [08:07] too bad that ssh ports are closed at my company, I could have given an hand on that package :/ [08:09] that one seems reasonably easy [08:09] the compiz one seems more worrying [08:10] Amaranth: do you happen to have a minute to fix http://launchpadlibrarian.net/36614923/buildlog_ubuntu-lucid-i386.compiz_1%3A0.8.4-0ubuntu7_FAILEDTOBUILD.txt.gz ? this is blocking alpha-1 [08:34] good morning everyone [08:35] hey chrisccoulson [08:35] hey pitti, how are you today? [08:35] I got up at 6:30 :) [08:35] that's quite early :) [08:35] right now I'm fighting gnome-python-desktop and compiz FTBFS [08:35] we need to get them to build to unbreak alpha-1 [08:36] ah, that's not good [08:36] due to the libmetacity0 -> libmetacity-private0 renaming [08:37] ah, i didn't know about that. that's a change from debian right? [08:37] right [08:38] hello chrisccoulson ;) [08:38] hey didrocks, how are you? [08:38] chrisccoulson: fine, and you? [08:38] yeah, not too bad. although, i didn't get a lot of sleep last night [08:39] I went to bed at 22:00 already [08:39] too much Gluehwein at the Christmas fair yesterday :) [08:41] * didrocks reminds when he sold Gluehwein during "la fête des lumières" in 2005 to finance his non profit organization action in Africa [08:41] i need to go to our christmas market at some point :) [08:42] more than 300 liters in 7 days :) [08:42] but it's quite difficult with a 3 week old baby now [08:42] didrocks - you drank all that? [08:42] ;) [08:42] chrisccoulson: sure, especially for having a real night. Should be difficult [08:42] chrisccoulson: yeah, I auto-founded everything \o/ ;) [08:53] good morning desktopers! [08:55] hello seb128 :) [08:56] lut didrocks [08:56] ca va ? [08:57] good morning seb128 [08:57] ça va bien, et toi? [08:57] bonjour seb128 [08:58] didrocks, bien, merci! [08:58] hey chrisccoulson pitti [09:00] gnome-menus crasher already fixed, pitti as quick as usual ;-) [09:03] phew, I think I bent gnome-python-desktop to my will now [09:05] now off to beating on compiz [09:05] pitti on strike ;) [09:05] *nng* need ... build ... alpha1 *nng* [09:11] building alphas is overated I would say [09:12] well, but having buildable packages isn't.. [09:13] right now the desktop is uninstallable [09:13] right, I was just joking there [09:13] ah :) [09:13] not yet fixed in compiz git head either :-( [09:16] pitti: I can have a look when I finished with apt if you want [09:16] mvo: do you know a bit about KWD (KdeWindowDecorator)? [09:17] mvo: KWD:Window is abstract now, and it tries to instantiate it [09:17] pitti: I think its just a matter of adding the pure virtual functions with NOPs, iirc we had something similar in the past [09:17] pitti: yeah, I saw the failure log [09:17] pitti: if it goes (much) beyond that, we need to ask for help in #compiz-dev [09:18] * mvo asks there anyway [09:18] mvo: is there an example commimt for this? [09:20] * mvo looks [09:22] mvo: ah, there are a couple of almost-empty functions in kde/window-decorator-kde4/window.cpp [09:22] http://paste.ubuntu.com/337877/ [09:23] that was the one we used, I guess only the lower half is needed now [09:25] ugh, what was f() = 0 again? [09:25] mvo: I'll try this [09:25] pitti: that makes it pure virtual, but I think this is no longer needed [09:25] pitti: I mean, this bit of the patch [09:28] mvo: this seems to work (adding empty stubs), but of course nobody guarantees me that these functions aren't actually called anywhere.. [09:29] I can grep the source, of course [09:29] but KDE itself could call them (they are virtual) [09:29] well, I'll just give it a try [09:30] pitti: right, I have no idea if stubs are enough or not, but I think its worth a try. alternatively we could simply disable compiz-kde until upstream fixes it [09:31] right, it's universe after all [09:34] yay, it builds [09:35] feedback from upstream already (they rock!) [09:35] maniac103: do you think its enough to just add stubs? [09:35] mvo: yes, I'm just unsure about the virtual KDecorationDefines::WindowOperation KDecorationBridgeUnstable::buttonToWindowOperation(Qt::MouseButtons) [09:35] mvo: do you think it's okay for me to just commit that patch now and upload? [09:35] pitti: yes [09:36] seb128: that reminds me, did my patch for u-m had a positive impact on the performance? [09:37] mvo: http://pastebin.com/f3ab952f1 [09:37] horrible, and prone to fail (not returning real values, etc.), but builds :) [09:37] it should work if these functions aren't actually called [09:37] * pitti rather won't forward this to upstream like that, though [09:37] mvo, oh, dunno, I didn't get the dist-upgrade prompt today and I did normal upgrade [09:37] pitti: I'm sure upstream will come up with a good fix soonish [09:39] * pitti cranks buildds to finish before next publisher [09:39] * pitti hugs mvo, thanks for handholding [09:40] my pleasure! you did all the real work :) [09:42] die, libmetacity0, die! [09:42] * pitti wonders what will fail next [09:48] ok, now the ports FTBFS is just due to kde library build wreckage [09:58] pitti: how much churn is ok at this point? I have a apt upload pending now but nothing in it is really criticial, but it would be nice to get the exposure and testing [09:59] mvo: no library transitions, no ABI/API changes, no stuff that can potentially wedge archive/installability [09:59] apt sounds a bit on the dangerous side right now? [10:00] compiz built \o/ [10:03] pitti: no abi/api issues, but of course with apt there is always a risk [10:20] hmm, I forgot what you use for setting default values for gconf schemas in packages, what was it please? :) [10:24] rodrigo_, a file named binary.gconf-defaults in the debian dir [10:24] ah ok! [10:24] where binary is the name of the binary package you use [10:25] the format being "key value" [10:25] one on each line [10:25] seb128: time sink> indeed, I spent 50 minutes on the gvfs merge now, and most of the changes are wrong for us or cause unnecessary pain :-( [10:25] rickspencer3: seems like people are supportive of solang, shotwell, and even gthumb, but there still doesn't seem to be a lot of discussion happening [10:25] pitti, oh, please ignore such changes then [10:25] right, I'll revert them [10:26] what did they do? [10:26] I'm just interested in the monitor move to gvfs to be honest [10:26] seb128: cool, thanks! [10:26] lots of changes for building on hurd, etc. (hal), patches to remove the libgvfscommon0 library and build it without a soname, etc. [10:26] dpic, supportive of getting those in universe? [10:26] seb128: the only thing that moved to gvfs is the burn backend [10:26] pitti, ok, please drop all those, we don't want complexity [10:27] it just makes upgrades harder [10:27] pitti, they don't install the remote monitor in gvfs? [10:27] yeah, it's a pain [10:27] seb128: supportive of one of them being used as the default in lucid instead of f-spot [10:27] or lucid+1 [10:27] dpic, where did you see that? [10:27] seb128: right now I actuall ponder dropping the merge completely [10:27] dpic, have you read my reply on lists? [10:27] pitti, please do [10:27] pitti, sorry I made you waste time on that [10:28] seb128: one piece that's worth keeping is fixing /etc/profile.d/gvfs-bash-completion.sh [10:28] which should be in /etc/bash_completion.d/ [10:28] I'll keep that [10:28] seb128: i did, what about it? [10:29] dpic, I'm not sure how you count that as being supportive of the change... [10:29] seb128: what do you meant about the monitor? [10:29] pitti, we should not merge but cherrypick fixes for those [10:30] seb128: we have the gdu monitor in gvfs, and the gphoto monitor in -backends [10:30] just as Debian [10:30] seb128: i don't. it wasn't very much against the change, except the fact that things might not be as bad as i made them out to be. everyone else who posted seemed to like the idea [10:30] brian and caleb, that is [10:30] wouter seemed neutral [10:30] those replies are not realistic [10:30] you don't replace a project running for years and working by something starte 2 months ago and with no user and testing [10:31] pitti, hum, let me look at this build failure again [10:31] seb128: of course, i was trying to make a call for testing [10:31] dpic, your email was rather a suggestion to consider to change the default [10:32] GVFS-RemoteVolumeMonitor-WARNING **: cannot open directory /usr/share/gvfs/remote-volume-monitors: Error opening directory '/usr/share/gvfs/remote-volume-monitors': No such file or directory [10:32] i suppose i should've started the email where i ended it and ended it where i started it [10:33] pitti, ^ does debian create this dir in gvfs? [10:33] in the gvfs binary I mean [10:34] dpic, you object on f-spot because you are against C#? [10:34] seb128: yes, seems so [10:34] dpic, just curious [10:34] pitti, ok, could you cherrypick that too? [10:34] seb128: seems we don't install that file at all [10:34] seb128: sure [10:34] pitti, that would fix this epiphany build issue [10:34] pitti, it's not a file but a dir? [10:34] seb128: oops, we install that in -backends [10:34] seb128: /usr/share/gvfs/remote-volume-monitors/gdu.monitor [10:34] $ dlocate /usr/share/gvfs/remote-volume-monitors [10:34] gvfs-backends: /usr/share/gvfs/remote-volume-monitors [10:34] gvfs-backends: /usr/share/gvfs/remote-volume-monitors/gdu.monitor [10:34] gvfs-backends: /usr/share/gvfs/remote-volume-monitors/gphoto2.monitor [10:35] seb128: no, not at all, i've just never met anybody that liked or used f-spot... [10:35] seb128: yep, will do [10:35] seb128: the other projects seemed to be on a level of F-Spot even though they were much younger [10:35] dpic, I do use and like it [10:35] seb128: i'm not saying nobody does, i just have never met anybody [10:35] dpic, they are not, they get no testing, no timeline, no export to web services [10:36] you should try using f-spot and compare features [10:36] things like export to flickr should be working [10:36] or picasa [10:36] or to galleries [10:38] seb128: that was for solang, which hasn't done it because it will be doing more than just exporting to other services. shotwell is already there and i'm not sure about gthumb [10:38] dpic, what project will do or not is not revelant for this cycle choice [10:38] f-spot will do ton of cool things too if you look at their roadmap [10:39] seb128: okay... [10:39] as long it's not done it's not revelant though, it's easy to say you will do something it's harder to do it properly [10:39] they might never do it [10:39] or get bored and stop working on the project [10:40] pitti, btw do you want me to look at the afc backend today? [10:40] and do the libiphone mir? [10:40] if you want, sure [10:40] I'm afraid I'm a bit busy today with the release stuff [10:40] well I want those to land before end of year and since we are archive frozen and I start getting bored [10:40] seb128: want me to upload above two gvfs fixes now, though? [10:40] please do [10:41] okay [10:41] I've spent almost a day on nautilus without luck [10:41] it's not wasting time [10:41] or not in obvious way [10:41] seb128: you seem to have just focused on solang there. and i wasn't saying it was relevant to the cycle choice, i was simply explaining why. similar to why empathy hasn't implemented metacontacts (because they're going to use People or Soylent which is better than just mushing people together like pidgin). [10:42] dpic, I didn't focus on anything, I just say that we have good reason to use a known codebase maintained for years rather than a new project which might crash in 2 months because hackers get bored [10:43] seb128: right...you seem to think i want this forced into lucid. [10:43] no, I'm just very surprised you consider that users are in favor of the change when all you got is 2 replies from random mailing lists guys [10:44] that doesn't seem a solid metric [10:44] looking at popcon to start would be something [10:56] re [10:56] other side of the split [10:57] dpic, sorry got cut by the split [10:57] dpic, http://popcon.ubuntu.com/by_inst [10:57] dpic, solang and shotwell got some 160 and 60 user votes [10:58] f-spot got almost 41 000 [10:58] so I was saying too [10:58] [10:58] well, that's hardly a meaningful number [10:59] since we install it by default [10:59] pitti, well I looked at the votes column not the installed one but right f-spot number is not revelant [10:59] it just shows that the other ones don't have a solid userbase [11:00] gthumb has some 27000 votes [11:00] and it's not installed by default [11:00] anyway enough on that topic [11:00] [11:00] * pitti hugs gthumb [11:00] so moving conf.d away wins some 2 seconds at login [11:00] half of nautilus start time is due to fontconfig [11:00] ! [11:00] we have too many configs and fonts there... [11:00] which conf.d? [11:00] how do we address that? [11:00] pitti, /etc/font/conf.d [11:01] the fontconfig dir [11:01] pitti, /etc/fonts/conf.d [11:01] ah [11:01] sorry [11:01] np, found it [11:02] also we have fontconfig 2.6 and upstream is 2.8 now for a while [11:02] we lack a fontconfig maintainer [11:02] seb128_: do I need anything else apart from the gconf-defaults file? It seems it doesn't take my change [11:02] hey pedro__!!! [11:03] hola rodrigo_! [11:03] rodrigo_, no, is the .defaults installed in the binary? [11:03] bonjour seb128_ [11:03] rodrigo_, dpkg -c binary.deb | grep gconf [11:03] ola pedro__ === pedro__ is now known as pedro_ [11:04] seb128_: ah, yes, it's there: [11:04] -rw-r--r-- root/root 141 2009-12-09 11:50 ./usr/share/gconf/defaults/10_rhythmbox [11:04] I was looking at the original .schemas file [11:04] what is the file content? [11:04] are you sure the value is not set? [11:04] in the .schemas file after installing the package, it's not [11:04] right but the .schemas is not used [11:04] /apps/rhythmbox/plugins/power-manager/active false [11:05] /apps/rhythmbox/plugins/power-manager/hidden false [11:05] /apps/rhythmbox/monitor_library true [11:05] defaults take over schemas [11:05] I added the monitor_library line [11:05] try in a guest session [11:05] should be working [11:05] we layout the gconf dirs to have defaults used first then schemas [11:06] oh, right, it works indeed [11:06] cool [11:06] I was looking in the wrong place then [11:07] rodrigo_, /etc/gconf/2/path has the directories layout [11:07] if you are interested [11:07] ok [11:07] the 2 bottom entries are those [11:08] the debian default is built from the defaults dir [11:08] using update-gconf-defaults, which is called a package installation [11:10] pitti, how do you suggest we process to get fontconfig updated for lucid? [11:10] pitti, talk to the foundation team about it? [11:10] it's a desktop package [11:10] :-( [11:10] I though it was a doko sort of package [11:10] closest maintainer is probably ArneGoetje [11:11] seb128_: I see one commit in git log which talks about speedup === seb128_ is now known as seb128 [11:12] http://cgit.freedesktop.org/fontconfig/commit/?id=31e0f0321057a7612ed5a7fa890dad09e6a53ee6 [11:12] in fact asac did most of the recent work [11:12] but no idea how much that brings, or whether it's relevant at all [11:13] ArneGoetje wanted to make a package for testing with latest upstream release [11:13] FYI I did a fontconfig packaging for 2.8.0 and update the big patch inside [11:14] baptistemm: that big patch should be reviewed/stripped down [11:14] baptistemm: but cool. do you have that in some PPA for testing? [11:14] yep [11:14] hey baptistemm [11:14] baptistemm, are you still working on the gnome-bluetooth update? [11:15] I was not able to produce a debdif as it said to me the expected size problem I don't recall [11:15] ? [11:16] asac, https://edge.launchpad.net/~bmillemathias/+archive/ppa?field.series_filter=karmic [11:16] hmm [11:17] baptistemm: there is a diff.gz there. not sure why debdiff wouldnt work [11:17] asac, I was responding to seb128 sorry [11:17] ah ok [11:17]