[00:27] RenatoSilva: https://wiki.ubuntu.com/MarkShuttleworth#Why%20is%20the%20default%20desktop%20in%20Ubuntu%20BROWN? for you answer about brown. Essentially, it has to do about being comforting... I guess it could also be argued that kind of like orange, it's supposed to help with productivity, IIRC. [00:27] Just read that before you pasted [00:27] as for GDM, I think you're supposed to be able to change the colors and wallpaper and everything by changing the theme for the 'gdm' user. [00:28] kind of late, yes :) [00:28] can't find any gdm customization option anymore [00:32] the gdm options were totally reduced to a single option of automatically loggin in [00:32] * RenatoSilva can't get rid ot the brown login screen [00:33] RenatoSilva, this is because gdm has been pretty much been rewritten [00:33] I haven't tried personally, but perhaps if you look at this page, it explains how to do the customizations now : http://www.ubuntumini.com/2009/09/hack-karmics-gdm-login-screen.html [00:34] cyphermox, frankly It's not a real answer - users don't care about it being rewritten, just about not being able to configure it as before.. I've had a discussion about it today... and yesterday.. and few days ago.. [00:35] so, file a bug, wishlist, with the configuration you want to be able to make - and *why*. [00:37] lifeless, so the current gdmsetup is considered done? [00:37] kklimonda: I don't know. [00:37] But I don't recall hearing about any major planned work. [00:38] kklimonda: 'done' for karmic, I'm hoping that something happens to it for lucid :) [00:39] Well I guess that the answer to *why* is pretty obvious [00:39] RenatoSilva: no, its not. [00:39] for you it seems [00:40] RenatoSilva: if it was obvious, I would know. I don't, so its not obvious enough that I know; and I'm fairly representative of Ubuntu developers, I think. [00:41] lifeless: system > admin > login screen. Before karmic it was full of options that *users* were using. In Karmic, they all were reduced to a single one. Users of those options like me will obviously miss them [00:41] RenatoSilva: thats not a reason to have the options [00:42] RenatoSilva: what value did they offer, was it worth the cost of the options - both code and confusion for other users [00:42] etc [00:42] lifeless, the most common complain I remember was that in 9.10 there is a list of users instead of simple dialog for usename [00:42] kklimonda: there is 'other', where you can type in the username [00:42] then the inability to change background [00:43] lifeless, sure - but they don't want list at all [00:43] kklimonda: You don't need to convince me, you need to make the case, either in a bug or on the list [00:43] RenatoSilva, at the very least this gives (I think) an opportunity to clean up the GDM config. and make it better.. hence why if you open a bug and explain which things you would like to be able to customize it would be immensely useful for developers to make them available if they are often requested [00:43] lifeless, I'm against it [00:43] ;) [00:43] kklimonda: particulary, I don't want some users to be displayed there [00:44] kkilmonda, how you can change the background: see above for my link :) [00:44] but I've dealt with a lot of complains in the last few days about it [00:45] cyphermox, I know how to change it, I even believe that the gdmsetup isn't needed. But other users don't think like that and I've run out of arguments to defend the current gdmsetup ;) [00:45] I wonder wherer my gdm theme is still in the system, but unused [00:45] * whether [00:47] the original way or simpler, I think it should allow customization of the login screen, at least because of the brown color [00:47] RenatoSilva, why? [00:48] how can you tell a user if he does not like the brown thing he can "change it easily" when in fact the login screen will still keep brown? [00:48] * RenatoSilva is a sample user [00:49] RenatoSilva, but it's still an option - just not though the gui [00:49] RenatoSilva, if the way to customize the GDM look was to show up the usual gnome-appearance-properties GUI when you click a button, would that make you more comfortable? [00:49] kklimonda: remember that ubuntu is form humam beings [00:50] RenatoSilva, the question is how much customization is "enough" for human beings and where should we draw the line [00:52] my point is: don't tell the user he can get rid of the brown when in fact he can't (the "human" users) [00:53] RenatoSilva, do we tell that? [00:53] (seriously, I don't keep tabs on visual side of development :) ) [00:53] * kklimonda don't even change wallpaper [00:54] when the user is chaning the theme in gnome-appearance-properties , he's not changing some specific techincal details about the ui. He just wants to change "the look", and that's abstract. So the login screen should be included in the "change look" option, whatever it is (gnome-appearance-properties or another application/meny) [00:55] * menu [00:55] RenatoSilva, but he's changing the look for the logged in user [00:58] I believe regular users may not be aware of that, I believe often they just want to get rid of the brown [00:59] and ubuntu should allow getting rid of the brown [00:59] RenatoSilva, what I have in mind is that users really dedicated to changing the looks of their system are going to dig as deep as needed (based on amount of work needed to change various visual settings in Windows) and the rest won't care enough... I know, "Linux is about customization" but I don't believe this slogan (just like "Linux runs on any hardware") [01:00] RenatoSilva, the question is - Do we have enough data to support the thesis "the just want to get rid of the brown" [01:01] they* [01:01] anyhow, I think it was not good to remove gdm customization *completely* [01:02] RenatoSilva, it's still there - just not though cluttered user interface [01:02] for the majority of people, it may as well be removed [01:02] RenatoSilva: it wasn't removed, nobody ever wrote it for the completely new gdm [01:02] it's certainly not obvious how to change it [01:03] but of course, that's not how users think about it ;) [01:03] ajmitch, give uf.o few days and they will create a script that does it for users ;) [01:03] JanC: customization of the login screen was removed completely [01:03] RenatoSilva, it wasn't ported [01:03] JanC: how many users would know or care that gdm had been rewritten? :) [01:03] JanC: not easily available if you prefer [01:03] RenatoSilva, the new gdm is more or less complete rewrite of the old gdm and gdmsetup just wasn't rewritten [01:04] I think it would be best to add a tab to the "appearances" dialog (only visible for users of group admin) to customize some basic things [01:04] JanC: that's a nice idea [01:04] as I mentioned earlier, just throwing this in the air in case: would it make people feel more comfortable with customizing the GDM look and feel if there was a button in the new gdmsetup, for example, to start a gnome-appearance-properties instance as the gdm user? ;) [01:05] cyphermox: that dialog has too many options that aren't useful in the context of gdm [01:05] cyphermox: it'd help, though it'd need to be trimmed down a lot to not confuse [01:05] cyphermox, I don't like it - looks like a hack [01:06] ajmitch, totally agree that there's too much stuff there though [01:07] why can't we just run the old version of gdm setup without those steps mentioned in the article? [01:07] gdm was replaced [01:07] its not the old gdm [01:08] nothing was deliberately *removed*, it just was not *added back in* [01:08] the old version of gdm doesn't do the xsplash stuff & various bits & pieces [01:09] * ajmitch doesn't really follow along with what happens in the boot process, so can't say what those bits are [01:09] * RenatoSilva didn't understand that article, will the steps make the old gdm options appear somehow? [01:10] RenatoSilva: no, it shows the "Appearances" dialog for the gdm user, as the new GDM uses standard GNOME/Gtk theming now [01:11] so no more separate theme format etc. [01:12] JanC: and does the "Appearances" dialog allow all the same options from the old gdm options? [01:15] it's only about the visual properties, and it's a "hack" (not really meant for Joe Enduser) anyway [01:15] * RenatoSilva still does not understand [01:16] RenatoSilva: it's exactly the same as the "appearances" dialog for your own user [01:16] let me try, be right back [02:03] hi all [02:04] JanC: better than nothing, we still get the brown background [02:17] You actually can't change much of gdm anymore [02:18] It's got a gnome-panel with applets on it, a background image, and a GTK+ widget showing the user list [02:18] It's not technically possible to do much to it outside of changing the GTK+ theme [02:18] I wonder if I can remove the user list, and change the nbackground somehow [02:19] sudo -u gdm dbus-launch gconf-editor [02:19] there is something in there for removing the user list [02:19] sudo -u gdm dbus-launch gnome-appearance-properties [02:19] that'll let you change the background and gtk theme [02:21] brb [03:09] setting '/apps/gdm/simple-greeter/disable_user_list' to 'true' for user 'gdm' supposedly removes the user list [03:13] thanks, brb [03:14] what is dbus-daemon and why it is needed in the above commands Amaranth [03:15] RenatoSilva: dbus-launch [03:15] sorry, dbus-launch [03:15] and it makes sure gconfd-2 gets launched for the gdm user so you can actually edit settings [03:19] brb [03:19] Amaranth: unfortunately, it also loads other stuff that you don't need, and doesn't unload anything--at least, last time I tried... ;-) [03:33] Is it possible to convert ubuntu desktop to the netbook remix version without having to format and everything? [03:35] the option in gconf-editor is not saved when you exit the app [03:36] and you can change the background, but the default one is shown after logging in, as a splash screen [03:37] I suggest adding backgrund change (including the spash screen) and Gnome theme change in the appearance preferences [03:38] maybe, when you select a user theme and you have admin rights, then there should be a check box like "apply this theme also for the login screen" [03:43] That wouldn't be good, the gdm theme needs special consideration [03:43] that which [03:43] and since the widget set is static you can do tricks in gtk themes that would break something in your normal desktop [03:46] http://brainstorm.ubuntu.com/ needs creating an account. No OpenID either :( [03:48] anyway, something should change, the login screen should be considerably mutable [04:02] RenatoSilva: if you have a Launchpad account, then you have an OpenID too [04:04] of course, I mean brainstorm does not let you loggin in with that OpenID though [04:04] huh? [04:04] oh right, they use the QA logins IIRC [04:04] weird! [04:05] brainstorm x LP bugs and blue prints are weird you know [04:05] s/e p/ep [04:06] I think brainstorm was designed before LP had OpenID [04:06] hopefully, a future version will implement it ;) [04:33] bug 470026 [04:33] Launchpad bug 470026 in ubuntu "Users should be able to change the brown look of login screen in Karmic" [Undecided,New] https://launchpad.net/bugs/470026 [04:43] bug 470008 ... I want this in Karmic, but people are saying its too big of a fix :P [04:43] Launchpad bug 470008 in gnome-control-center "The Custom option in "Visual Effects" tab of Appearance Preferences should be shown even if simple-ccsm is not installed" [Undecided,Confirmed] https://launchpad.net/bugs/470008 [05:06] eboyjr: s/fix/change/ [05:06] It's not a bug, it's a feature request :P [05:07] in LP, bugs can be feature requests (Wishlist importantce) [05:07] RenatoSilva: Right but there is still a difference [05:07] eboyjr: also you shouldn't confirm your own bugs [05:08] eboyjr: a second person should instead [05:08] eboyjr: so that developers can know that the bug is valid, because it was reproduced by more than one person [05:09] Amaranth, Alright. https://wiki.ubuntu.com/Bugs/HowToFix should be changed to not say: Assign the bug to "Nobody", and the status to "Confirmed". At this point, seek out a developer to review your patch for inclusion in the next upload of the package. [05:09] * RenatoSilva registered the bug 470053 [05:09] Launchpad bug 470053 in ubuntu "Option for disabling user list of login screen in Karmic" [Undecided,New] https://launchpad.net/bugs/470053 [05:11] eboyjr: not sure, but isn't it about downgrading to confirmed status? [05:14] iono [05:17] gtg === bjf is now known as bjf-afk [07:43] mvo: https://code.edge.launchpad.net/~amaranth/compiz/no_wrapper [07:44] hi, btw :P [07:47] hey Amaranth [07:48] Amaranth: cool! let me have a look? have you measured how much startup time this buys us? [07:48] mvo: hard to measure [07:49] my startup times are inconsistent [07:49] * mvo nod [07:49] I know compiz starts faster even with a warm cache [07:49] just eyeballing it [07:50] nice and smaller than I thought [07:50] yeah, compiz already does most of the checks [07:51] sounds like the first branch to merge when lucid opens [07:51] :) [07:52] I think it actually does a better job of doing the texture checks too [07:53] Should make sure we never see the white screen (improperly installed driver) or the black windows (texture larger than size limit) [07:54] Good morning [07:55] mvo: It also does a better job of telling you what went wrong because compiz is actually fairly detailed about things that make it fail :) [07:56] Amaranth: sweet. do you think its stable enough to just stick into the compiz PPA ? a lot of people run this, it will ensure some testing/exporure [07:57] mvo: did you follow the discussion with slangasek about bug 465619 ? [07:57] Launchpad bug 465619 in update-manager "apport enabled after a distribution upgrade from Jaunty to Karmic" [High,In progress] https://launchpad.net/bugs/465619 [07:58] mvo: he uploaded a fix straight to -updates; could you merge this into the proposed branch and reupload? [07:58] (with -v again, please) [07:58] pitti: yes, I can do that [07:59] thanks; accepting the -updates one then [07:59] pitti: a proper fix is in trunk/ but the minimal one is sufficient [07:59] pitti: thanks [07:59] mvo: any idea how we can fix this for people who already upgraded? [08:00] pitti: not a straightforward one, grep "enabling apport" /var/log/dist-upgrade/main.log is my best one currently [08:00] I guess I'd need to upload a hack to apport to disable it in the conffile again [08:00] pitti: if we want to be fancy, compare the debug timestamp to the mtime [08:00] mvo: oh, that grep sounds good [08:01] logging.debug("enabling apport") [08:01] that is the exact line [08:06] mvo: btw, if we clear out the blacklist for lucid compiz will only launch one shell (compiz-decorator) [08:09] Amaranth, are you a compiz dev? [08:09] eboyjr: yeah [08:10] That's really cool [08:27] morning [08:56] hey there [09:00] salut seb128 [09:00] lut baptistemm [09:01] weird, when nautilus is running I only see the desktop wallpaper, not the panel and applications ... [09:03] hello seb128 :) [09:04] lut didrocks [09:06] hello seb128, baptistemm and didrocks [09:07] lut huats [09:07] hey slomo_ chrisccoulson njpatel [09:08] hey seb128 [09:08] how are you? [09:08] hey seb128, hey chrisccoulson [09:08] had a nice weekend? [09:08] hey pitti [09:08] hey pitti [09:08] chrisccoulson, ok, a bit stressed by the number of ***** bug emails [09:08] salut les frenchies [09:08] got over a thousand of those during the weekend [09:09] seb128 - ouch, that's a lot ;) [09:09] spent 2 hours yesterday and 2 hours saturday cleaning to not spend my day today on that [09:09] yeah, i noticed you touched a few yesterday. [09:10] seb128: karmic bug reports? [09:10] sabdfl, new bugs or comment on existing bugs after karmic yes [09:10] the firehose, huh [09:11] i need to spend some more timing doing bug triage, but my weekend was already quite busy [09:11] I wish we would spend the lucid cycle stabilizing things [09:11] so many annoyed users by things that should be working and doesn't [09:12] * pitti scares people with http://www.piware.de/2009/11/hello-ween/ [09:13] seb128: we plan to :) [09:13] lol [09:13] pitti - is it safe to click on from work? ;) [09:13] pitti, I'm waiting for uds to see how much tractions we get for new features too [09:13] (before i click on it with my boss hanging around) [09:14] chrisccoulson, it's an halloween decoration [09:14] safe-for-work [09:14] hey mvo [09:14] pitti, you'd have drawn a panda in the pumpkin :) [09:14] had a good weekend? [09:14] hey seb128! [09:14] seb128: yes :) [09:15] heh:) [09:16] chrisccoulson: yes :) [09:57] \o/ I managed to bring my bug email backlog to 0 [09:57] congrats seb128 :) [09:58] ok, now time to start on some srus [09:58] that was quick! [09:58] chrisccoulson, thanks [09:59] pitti, if you need any more informations for the usb stick bug do not hesitate :) [09:59] huats: still catching up on over-weekend SRU and bug mail; I'll get to it [09:59] huats: will probably faster to debug this on IRC [09:59] sure [10:00] (I mean do it when you ave time...) [10:00] I know you are not daydreaming :) [10:00] thanks already for the time you take pitti [10:00] you're welcome, it seems to be a major issue [10:01] I fear that I am not the only one facing it... [10:02] chrisccoulson, well, as said I spend 2 hours on saturday and 2 hours on synday [10:02] sunday [10:02] ie I was almost to 0 when I went to bed yesterday [10:02] seb128 - that's still pretty impressive though ;) [10:02] considering the amount of e-mails you get [10:03] yeah, what is not impressive is the number of issues we have in the software we ship there [10:03] and we being opensource not especially ubuntu [10:03] we still have some way to go [10:05] reading emails from the weekend, usb key mounting is broken for quite some users, the audio and video players are being crap and can't read youtube videos or dvd or do visualization or crossfading correctly [10:05] the email client is not better, hanging when open files from emails [10:05] or crashing when adding tasks or when pressing enter for some users [10:06] quite some keyboard layout issues too [10:06] evince is not able render or print correctly quite some pdf apparently [10:06] *shrug* [10:08] * mvo hugs seb128 [10:08] * seb128 hugs mvo back [10:08] I share your pain, u-m give bogus error messages when the upgrade mirror is overloaded :( [10:10] seb128: a friend of mine upgraded as well and reported problems with policykit, pulseaudio, and docking station handling [10:11] pulseaudio is a frequent complain too... [10:11] and empathy has quite some issues [10:11] anyway let's see the good side, we have one cycle to make what we have now solid [10:12] I'm just going to fight hard against any new feature proposal at uds ;-) [10:12] we really need to stop adding things this cycle and try to make what we have work [10:12] +1 [10:12] but still I think it was good to squeeze it all into karmic [10:13] Hi. My mother is running ubuntu intrepid. When I run "update-manager -d" it suggest to upgrade to 9.04 and not 9.10. Is there a reason? [10:13] I just wouldn't recommend everyone to upgrade; it's the Fedora of Ubuntu, after all [10:13] Zdra: yes, we don't support upgrades with skipping releases [10:13] Zdra: except LTS->LTS (e. g. 8.04 -> 10.04) [10:13] pitti, it worked on another laptop [10:14] I'm almost sure it was intrepid->karmic [10:14] Zdra: Upgrade to 9.04, then upgrade that to 9.10. You are not supposed to be able to upgrade from 8.10 to 9.10 directly. [10:15] Hm :/ [10:15] I think I'll make a clean install then [10:23] pitti, you might want to reopen bug #463347 [10:23] Launchpad bug 463347 in udev "Usb Media not automaticaly mounted" [Undecided,Incomplete] https://launchpad.net/bugs/463347 [10:23] * seb128 closes another 2 duplicates of this one [10:23] seb128: have you seen something like bug #469399 before ? see comment #4 gtk-update-icon-cache fails with undefinied symbol g_mapped_file_unref [10:23] Launchpad bug 469399 in update-manager "Could not install 'gnome-accessibility-themes' & 'gnome-themes-more'" [Undecided,New] https://launchpad.net/bugs/469399 [10:24] pitti - bug 470433 seems to be another issue with not correctly detecting the filesystem type [10:24] Launchpad bug 470433 in gvfs "no usb automount in Karmic Koala (dup-of: 463347)" [Low,Invalid] https://launchpad.net/bugs/470433 [10:24] Launchpad bug 463347 in udev "Usb Media not automaticaly mounted" [Undecided,Incomplete] https://launchpad.net/bugs/463347 [10:24] mvo, ldd `which gtk-update-icon-cache` [10:25] chrisccoulson, yeah, see what I just wrote, same for bug #463347 [10:25] ups bug #/469722 [10:25] gnagnagna [10:25] seb128 - yeah, i just noticed you touched that one ;) [10:25] bug #469722 [10:25] Launchpad bug 469722 in nautilus "Karmic Koala does not automount usb devices (dup-of: 463347)" [Low,Incomplete] https://launchpad.net/bugs/469722 [10:25] Launchpad bug 463347 in udev "Usb Media not automaticaly mounted" [Undecided,Incomplete] https://launchpad.net/bugs/463347 [10:26] chrisccoulson, well the first bug was the one I just pointed [10:26] seb128: you assume some old stuff in /usr/local ? I will followup with that [10:26] I asked the submitter to use ubuntu-bug and use the symptom based filing [10:26] mvo, either that or gtk updated before glib but that should not happens with the correct depends right? [10:27] mvo, it basically means the glib version seems too old [10:27] those issues are usually local installs [10:27] seb128: yeah, thanks [11:03] hey pedro_ [11:03] salut seb128 [11:03] pedro_, how are you? [11:04] seb128, good, thanks, what about you? how was that long weekend? [11:06] pedro_, quite good thanks, though I spent some daily hours triaging bugs [11:06] got over a thousand email and I didn't want to have a depressing monday doing triaging only [11:06] pedro_, what about you? [11:07] seb128, ah yeah noticed that on Friday :-P [11:07] seb128, was good and quiet this time :-) [11:07] excellent [11:08] oh, ubuntu open week this week [11:08] see reading dholbach on facebook has some use ;-) [11:10] seb128, have you seen bug 448153 ? [11:10] Launchpad bug 448153 in nautilus "Folder settings (metadata) lost after upgrade" [High,Triaged] https://launchpad.net/bugs/448153 [11:11] * pedro_ being nudge by gpoo about that [11:11] pedro_, yes, I will do srus this afternoon [11:11] \o/! [11:11] it show how much we don't test upgrades [11:12] well we test mechanic updates [11:12] but not that user settings, configs etc are handled as they should [11:15] hum [11:16] how do we deal with user cleaning on uninstall breaking due to running softwares? [11:16] ie gdm uninstall can break because the gdm user has an active session [11:17] I don't remember that we ever did that [11:18] pitti, that being? [11:18] doing any particular check whether an app is running for package removal [11:18] well the thing is that the maintainer script is failing right now [11:18] if gdm is still running? [11:18] that sounds correct to me? [11:19] ok, that was my question is some sort [11:20] users should be stopping gdm before uninstalling it [11:59] seb128: does gdm start X? If so, in which file? I'm seeing this root 1040 0.9 0.8 14584 8184 tty7 Ss+ 12:50 0:01 /usr/bin/X :0 -br -verbose -auth /var/run/gdm/auth-for-gdm-H9rYlX/database -nolisten tcp vt7 [11:59] yes gdm does start X [11:59] what do you mean by in which file? [12:00] in the source code of gdm [12:01] dunno [12:01] I need to change this command: /usr/bin/X :0 -br -verbose [12:01] why? [12:01] into /usr/bin/X :0 -nr -verbose [12:01] I'm working on a smooth boot process for Lucid [12:02] and the -nr option allows X to create a root window with no background [12:02] daemon/gdm-server.c: argv[len++] = g_strdup ("-nolisten"); [12:02] gdm_server_init (GdmServer *server) [12:02] server->priv->command = g_strdup (X_SERVER " -br -verbose"); [12:02] tseliot, ^ [12:03] seb128: perfect. Thanks a lot :-) [12:03] you're welcome! === MacSlow is now known as MacSlow|lunch [12:31] mvo: fyi http://lists.freedesktop.org/archives/packagekit/2009-November/005453.html [12:32] seb128: ^^^ in the gnome-packagekit bits, are there some useful to ubuntu? [13:12] seb128, one question, would you have an item called "Système de fichiers ..." on your nautilus sidebar ? (I have a problem specically setting an emblem on) [13:13] or any frenchy around, I would need to reproduce something [13:16] hello baptistemm [13:16] hi chrisccoulson [13:17] hi chrisccoulson1 rather [13:17] heh, my name keeps changing because of my unreliable connection ;) [13:17] how are you? [13:17] fine, not a lot of thing to do at job, I'm a bit bored [13:18] i feel the same actually ;) [13:18] it means i have plenty of time to perfect my CV though! [13:19] but as I just arrived I feel unconforable to do something else than the job I have to do [13:19] (ie; doing some gnome or ubuntu things) === MacSlow|lunch is now known as MacSlow [13:19] yeah, i know the feeling ;) [13:20] hopefully now I work in a OSS environment [13:21] what job do you do? [13:21] I'm a contractor for Orange [13:21] for the web portal [13:22] huats: hey! got some minutes to debug your usb key? [13:22] excellent, that sounds interesting [13:22] complex architecture but I just managed the front part (the one that use the user) [13:22] s/ed/e/ [13:23] pitti, always ! [13:46] baptistemm, yes there is a filesystem entry in nautilus disks for several cycles [13:47] seb128, once you've mounted it, could you try to set an emblem on it? [13:47] baptistemm, there is no way to unmount it, it's your linux install [13:47] ie your ubuntu disk [13:48] ah sorry I wasn't clear enought [13:48] I mean't an additional volume [13:48] no [13:49] each time I try to set an emblem on the icon on the desktop of the volume I see the message "Erreur GConf : Le texte contient de l'UTF-8 non valide" in .xsession-errors [13:49] fredp, I still have difficulties to understand what packagekit-gnome is [13:50] fredp, I will let mvo comment on that [13:51] seb128: there is a bunch of stuff in there, some might be useful (such as helpers to install a package handling a given mime type) but others are more full fledged applications [13:51] baptistemm, same here [13:52] fredp, how would those benefit GNOME? [13:52] fredp, as said the other day I think what we should have is a set of API apps can use [13:53] seb128, okay nice to know I'm not alone :) [13:53] seb128: I believe they act as small wrappers around dbus calls to packagekit; at the moment applications have to do the dbus call by themselves [13:53] that's probably useful to have [13:53] mvo, ^ [13:54] from a gnome point of view we need known points that applications can use, be it a dbus api, or helper binaries [13:54] (I only had a quick look at what is being done as "packagekit support" in nautilus and file roller) [13:56] seb128, I can open a bug if you want, but I don't know what and where it fails [13:57] I tried to create a foler with è in the name and it workds fine [13:57] baptistemm, if you open one open a GNOME bug ;-) [13:57] :) [14:00] mvo, https://wiki.ubuntu.com/SoftwareCenter#Package%20icon,%20title,%20summary,%20and%20description (in case you hadn't found it already) [14:07] fredp: hm, afaik there is the session dbus api for that, that is provided by the packagekit update applet [14:08] fredp: the trouble is that we have a applet like this already so we would have to do some work on making the PK one invisible (or merging the functionatliy from ours into the PK one) [14:10] fredp: we will have a bigger discussion about PK and ubuntu 10.04 at UDS, if the debconf problem can be solved (and everything looks good) there is no conceptual reason for us anymore not to use it. then it just becomes a lot of work to adjust our tools [14:11] hey rickspencer3 [14:11] fredp: when is the deadline for the 2.30 discussion? [14:11] hi seb128 [14:12] mvo: the module discussion happened on Sunday; and we were not really sure about packagekit, hence the mail to the list. [14:13] how will it work, is there voting? or will the release team make a decision at some point based on the feedback? [14:14] mvo: the release team decides [14:14] * mvo nods [14:14] at the moment distributions (not just ubuntu) are mostly using their own tools; sometimes combining them with a few of the packagekit ones [14:15] but (especially if the debconf issue is solved) it looks like everybody is fine converging to packagekit, backend wise. [14:15] now, if some other distros were starting to contribute to the packagekit tools... [14:15] just like it was proposed earlier for the status icon :) [14:15] everybody would love it :) [14:21] mclasen: there has been a misunderstanding of some sort in the past, the debconf issue was interpreted as a policy decision (hughsies law in the faq), that is why ubuntu/debian was so hesitant with it [14:54] * kenvandine wishes he had noticed his laptop was failing to suspend before we released karmic === jcastro_ is now known as jcastro [15:30] seb128: do you know what file specifies the parameters (e.g. --display-id) that are passed to gdm? [15:43] seb128: re bug 460710 -- can I propose a SRU for it? [15:43] Launchpad bug 460710 in evolution "Evolution hangs when double-clicking on attachment" [Medium,Confirmed] https://launchpad.net/bugs/460710 [15:58] hggdh, hey, what change is that? [15:58] tseliot, to what process? [15:59] seb128: /usr/lib/gdm/gdm-simple-slave --display-id /org/gnome/DisplayManager/Display1 [15:59] I need to pass an additional option [15:59] seb128: murrine (and, it seems, clearlooks) both can be set to draw a rectangle with height=0 -- which causes a loop where all memory is allocated [16:00] seb128: *how* this is driven from Evo is still unknown, but the fix adds an assertion to block it from happening [16:00] I cannot reproduce, so far [16:11] tseliot, the display seems to be passed by a previous dbus query [16:11] /org/gnome/DisplayManager/Display1 seems to be a dbus "path" (not sure about the accurate word) [16:12] tseliot, not sure about this one [16:12] hggdh, if you have a debdiff let me know I can sponsor it as a sru [16:13] baptistemm: ok but I'm trying to pass an additional argument such as --force-active-vt [16:13] seb128: ok, thanks anyway [16:13] seb128: ok, will prepare one [16:15] hggdh, it's not the same issue than bug #393534? [16:15] Launchpad bug 393534 in gtk+2.0 "gtkiconview issue, evolution chockes when attaching files sometimes" [Low,Fix committed] https://launchpad.net/bugs/393534 [16:20] seb128: hum. [16:23] hggdh, it's worth asking users to try the gtk update [16:24] seb128: *might* be. This would fix GTK *before* drawing, while on 460710 we are just blocking an invalid call [16:24] but will ask them to try [16:24] hggdh, in any case I'm fine sponsoring those theme fix as sru === james_w` is now known as james_w [16:39] hggdh, there seem to be quite some bugs about new task crashing evolution [16:39] hggdh, there seem to be quite some bugs about new task crashing evolution, do you know if it's being worked? [16:45] seb128: you mean the eds crashes we have been having? [16:46] do you have one bug # as a sample? [16:46] hggdh, no, the "crash when pressing enter in a task" [16:47] you forwarded it to bugzilla.gnome.org [16:47] oh, OK, yes, easy to reproduce [16:47] on memo [16:47] it's getting quite some duplicates [16:48] did you point upstream guys to it yet? [16:48] I will look at it, and see if either I or upstream can come up with a fix [16:48] thanks [16:48] no, I just opened it upstream, and then got busy elsewhere. I will follow it up [16:49] hggdh, thanks [16:50] what is the so-called upstream exactly? [16:51] RenatoSilva, the people writing the software [16:53] so when people say the bug was reported upstream, it's because it was reported to the source developers, using the source bug tracker, right [16:55] I was thinking about it, sabdfl said once that collaboration between projects is more important than binary compatibility. I wonder how is the work of ubuntu developers regarding this [16:55] RenatoSilva: correct [16:56] RenatoSilva: most of the time we do not have either the expertise or time available to fix it; anyways, it does not help the FLOSS community to just fix it locally [16:56] so we report it to the developers (upstream), and -- if we found a fix -- we also report the fix [16:57] end result is everybody benefits (and we can keep our delta from upstream to a minimum) [16:57] I mean, when you want to change ubuntu, do you create customizations for the used projects, or do you work together with the 'upstream' to get new features there and then integrate the changes into ubuntu? [16:58] ideally we integrate with upstream [16:59] I mean, the work of creating a distro is picking up software out there, bundling all of them together, unless you want to change the software, in which case it'd be better to do it in the upstream if your feature is worth or if you have a bug. [16:59] pitti, the various reports on the bug I have opened seems quite encouraging ! [17:00] RenatoSilva: yes indeed. Sometimes we will carry our own changes out of upstream (due to specific localisations on Ubuntu, or because the fix is critical, or whatever === asac_ is now known as asac [17:01] hggdh: what is the overall "percentage of success" in working with upstream? like, ~80% of the time you get success or what. And what is the overall percentage of commit rights ubuntu developers have in the other projects? [17:03] RenatoSilva, those are not easy metrics since it depends of the project, some are actively worked some other not [17:04] and it depends mostly on how good are the individual's relations with upstream [17:04] you also don't need commit right, upstream are usually happy to commit changes for you [17:05] but -- at least from my own petite contact with upstream -- I usually can get it done/accepted/discussed [17:05] and, as Seb points out, I do not have commit rights upstream, [17:05] seb128: I'm not saying you need, I'm curious about how many ubuntu developers are actually upstream developers [17:06] to an open source project? a good part [17:06] seb128: I know it's not easy to get a number, I was more like the percentage according to your feeling [17:06] but could be things which are totally different from their distro work there [17:06] seb128: like 1/5, or 9/10 etc [17:07] no idea, depends of what you call ubuntu developers [17:07] just packages, distro team, desktop team, bug triager, translators [17:07] seb128: good part == most of them? [17:08] depends of the category of people you consider [17:08] RenatoSilva: notify-osd , messaging menu is upstream developement and is from canonical [17:08] translators are not hackers [17:08] it's one example [17:09] seb128: developers == people that write actual code [17:09] ubuntu is a distribution, most of the work is distributing code not writting it [17:09] RenatoSilva: those are to name a few , there are others developers who contribute to non-canonical upstream too , it is a thin line cant be easily defined ,,, the question is quite abstract [17:09] ie testing, bug triaging, documentation, translation, packaging, etc [17:09] ie testing, bug triaging, documentation, translation, packaging, etc [17:09] ie testing, bug triaging, documentation, translation, packaging, etc [17:09] ups [17:09] sorry [17:10] it's a whole ecosystem you can't really do easy stats [17:11] the desktop team members have almost all commit rights to some upstream projects there but that's not a metric for how active they are on those either... [17:11] mac_v: I know it's abstract, but the notion wheter it is 1/4 or 9/10 is kown to you I think. I have NO notion. Trying to know your process you know [17:12] RenatoSilva, depends of the group of people you talk about [17:12] seb128: that's what I wanted to know, so actual written code is just a samll bit of the development of ubuntu? [17:12] the canonical desktop team has most of its member being upstream contributor to at least one open source project [17:13] RenatoSilva, yes, there is some 15000 softwares in ubuntu we can't write all of that [17:13] the team is like on hundred people [17:13] one [17:13] seb128: exclude translators, bug reporters and triaggers, testers, documentation etc [17:14] the ubuntu distribution team is distributing software [17:15] ie taking things writen by the community, putting those together, looking at bugs, fixing some, adding feature, etc [17:15] we do test, participate to feedback, send fixes and new codes etc to lot of projects [17:15] we are not main code writers for most though [17:16] seb128: I don't mean write the softwares, I mean change them. Ubuntu is a popular distro just because you choose and package the softwares and work with upstream very well? Or there's also code writing to make ubuntu 'different' from the other distros? How much of code write? Most of the time or just a few times? That code is written directly in upstream or in forked versions in Ubuntu? That's what I mean.... [17:16] RenatoSilva: there are no stats like that and can never be , because , if someone knows about the code they can contribute anywhere and any fix in Ubuntu almost always goes upstream too , so the desktop team contributes to upstream development too , but as seb128 said the initial project might not have started in Ubuntu [17:16] RenatoSilva, not sure what you are aiming at though and if that replies to that questions [17:17] RenatoSilva, ubuntu is mainly based on open source project, we do write most of the package management tools we use though [17:17] RenatoSilva, and things like the installer [17:18] the desktop is GNOME which is an opensource project with lot of companies and individuals contributing [17:18] RenatoSilva, not sure what make ubuntu popular as a distro or if it has a higher popularity than some others, you should ask users about that [17:19] RenatoSilva, the softwares used are mostly the same as in other distribution, the specific parts are package management and installer [17:19] maxb: "any fix in Ubuntu almost always goes upstream too " ---> this answers one of my questions :) [17:19] and then some bits like theme, settings, etc [17:20] the ayatana team is aiming at improving user experience too and those are project maintained by canonical people [17:20] seb128: we do write most of the package management tools we use though ---> humm [17:20] ie notify-osd or the indicator message [17:21] RenatoSilva, update-manager, gnome-app-install, software-center, software-properties, gdebi [17:21] seb128: is there a list of ubuntu software for which canonical and/or ubuntu committers are the upstream? [17:21] RenatoSilva: upstart to improve the boot speed is also from Ubuntu [17:21] RenatoSilva, I doubt so [17:21] RenatoSilva, it's hard to define upstream, some project are made of contribution from lot of different people [17:22] it's not easy to weight how much one contributor part count [17:22] canonical has a page that lists its originated projects [17:22] but the ubuntu distribution team focus is to integrate software and make it work fine for its users [17:22] not to write new software [17:22] seb128: we do have such a list somewhere [17:22] * pitti -> dinner [17:22] I just can't find it right now ;) [17:23] jcastro might know [17:23] seb128: by upstream I mean the main developers, the drivers of the project, the ones owning the home page, repository etc [17:24] RenatoSilva, lot of project don't have a static upstream team [17:24] but the ubuntu distribution team focus is to integrate software and make it work fine for its users, not to write new software ----> hummm [17:24] the projects have infrastructures set and people working on those come and go [17:25] seb128: don't have a static upstream team? [17:25] some people work a year on something and stop for whatever reason, some stay [17:25] yes, that's like any project or company [17:25] people change job [17:25] they get bored and want to do something else [17:25] seb128: then they "give" the project to other people? [17:25] they got busy in life and move to other things [17:26] there is no strong project ownership [17:26] seb128: I mean regarding companies [17:26] it's just collection of contributions by people interested to get the thing working [17:27] seb128: what software is from canonical (as the original or main driver, the one who maintain the project site, wiki, repo etc), and what is from community, that's what I meant [17:28] RenatoSilva, https://wiki.ubuntu.com/Upstream [17:28] seb128: I don't think companines abandon projects that often [17:28] note, that page needs alot of work [17:28] RenatoSilva, right but most projects are not owned by a company [17:28] RenatoSilva, they are community efforts, worked on my student, after hour hackers, etc [17:29] my -> by [17:29] or distribution people fixing issues they run into [17:29] jcastro: thanks [17:30] RenatoSilva, you know about wikipedia for example? [17:30] RenatoSilva, who do you think is maintainer for wikipedia or own it? [17:30] wikimedia foundation? ;) [17:30] ;) [17:31] mpt: hi [17:31] seb128: not a company, I know that [17:32] mpt: would you be okay with having a fullscreen menu item in software centre? [17:32] and471, hm, why? [17:32] RenatoSilva, same for GNOME for example then [17:32] seb128: so you write code for the package management tools like synaptic? but isn't this and other stuff like apt from debian community? [17:32] mpt: I just included one in my branch [17:32] and471, it's not really the sort of program you want to spend ages in [17:33] mpt: meh, if it pleases one person, then I suppose it was worth it (or is that a bad design ethic?) [17:33] RenatoSilva, apt and synaptic were written before ubuntu yes, the ubuntu team wrote update-notified, update-manager, gdebi, gnome-app-install, software-properties, software-center, etc [17:34] and471, yes, it is. :-) For a movie player or a photo editor or a word processor, I could understand [17:34] mpt: so remove it? [17:34] seb128: so you meant these new tools, not apt and synaptic whose mainstream is debina and for which you just collaborate [17:35] and471: hey ;) indicator-messages [17:35] RenatoSilva, apt and synaptic were there before ubuntu, why would have rewriten those scratch? [17:35] and471, by "why" I meant "what sort of person would use it, and when" [17:35] RenatoSilva, we based what we did on existing solutions yes [17:35] RenatoSilva, those software I've listed have been written mostly by ubuntu people though [17:35] mac_v: ah thankyou :-) [17:36] RenatoSilva, I'm still not sure what you try to figure there [17:36] mpt: I was just thinking whether maximus needs it to work [17:36] RenatoSilva, our web browser is firefox which is made by the mozilla foundation for example [17:36] lool: does software centre behave corrected with maximus currently? [17:36] RenatoSilva, it's not owned by any distribution [17:36] seb128: I just mean , you said canonical is the main maintainer for the package tools, then I wondered whether it has 'adodped' apt/synaptic from debian guys [17:37] and471, njpatel is free to correct me, but I don't think maximus cares whether a particular application has a full-screen mode. [17:37] RenatoSilva: i think you caught seb128 in his free time or a really good mood ;) ... [17:37] seb128: You are answering my curiosity pretty well [17:37] RenatoSilva, why would be want to take those project over? they already have a website, bug tracker, etc [17:37] RenatoSilva, we contribute back where tehy are [17:37] we don't aim at taking over other people work, that's not how opensource work [17:37] if the project is already set and working we work together [17:37] mac_v: the icons aren't in that package (well not the humanity styled ones) [17:38] mac_v: does humanity ship them? [17:38] and471: the name is indicator-messages and it is in humanity [17:38] mac_v: ah sorry, I thought you meant the package name :-) [17:38] ;p [17:39] mac_v: when there is a new message the indicator icon is still not obvious enough to indicate this change. maybe adding a small ball like in jaunty [17:39] but in gray color [17:39] for lucid [17:39] tgpraveen1: no ;) [17:39] why? [17:39] seb128: I just mean when you said "package tools", it soudned like apt/synaptic was included in them, but ok I've understood now [17:40] mac_v: is it going to pulse tastefully? :) [17:41] Ng: that would be the only next step , >>> go against the HIG ;) [17:41] mac_v: i think you caught seb128 in his free time or a really good mood ---> :) [17:41] mac_v: it needs to do something :) [17:41] lol [17:41] ok guys thank you very very much for answering my curiosity [17:41] tgpraveen1: the notification is not to notify [though an oxymoron;) ], that would be the the job of the notify-osd or the app itself , to draw attention to the new event [17:41] RenatoSilva: welcome. [17:42] notificatio area* [17:42] RenatoSilva, you're welcome [17:42] mac_v: but it's an indicator applet. how can its job not be to indicate? ;) [17:42] lol ;) [17:42] it's not a notification area item, it's a panel applet [17:42] mac_v: was the suggestion that I gave discussed earlier and rejected. because I feel that it would draw attention better and not annoy [17:43] and I get your point but it seems many people including me seem to be unsatisfied with current behaviour [17:44] especially one can't see the change form a little distance or unless one actually very observantly tries to look for it [17:44] tgpraveen1: there was a bug even when there was a star used , [human theme] [17:44] a small ball or something though in gray colour so as to match the current theme would be nice [17:44] tgpraveen1: a grey ball on a grey icon on a grey-ish panel? that seems like it wouldn't help [17:44] mac_v: so even with star it wasnt' enough. further proves my point ;-) [17:45] tgpraveen1: yes, only blinking will do ;0 [17:45] well the grey icon of the envelope is darked [17:45] mac_v: instead of going against the design principle, could you not change the colour to something that is more distinguishable [17:45] if it's supposed to indicate to the user that there are messages waiting it needs to do it a) prominently, b) persistently [17:45] darker than the panel when new msgs come so it will be ok [17:45] do it prominently not persistently [17:46] that would annoy like blinking as mac_v said [17:46] tgpraveen1: we can make the no new sate lighter ;) [17:46] state* [17:47] hmm maybe do both :-) [17:47] tgpraveen1: if it's not persistent how will I notice new messages when I come back from making a drink? [17:47] Ng: well when ever you leave pc and come back you have to check if there [17:47] are new msgs [17:47] that is how the indicator was designed for [17:47] rickspencer3, 15 minute courtesy warning! [17:48] jcastro, thanks, I'm ready [17:48] the problem with too much persistency is that if a user is sitting at pc [17:48] Ng, you do what I do, check the thing over and over all day. :-/ [17:48] and471: the indicator is just to indicate new events , but for an IM the app needs to bring the new IM window forward , email == new snail mail so no need grab user attention ... imagine the icon is colored , then it would draw undue attention to the user who knows about the new message , but wants to do other work , [17:48] and working on say some doc an sees the notification and knows he will attend to it later [17:49] then also if indicator is persistent then it will casue annoyance [17:49] then the icon would be a nuisance* [17:49] anyways time to sleep. good night [17:49] mac_v: ah I see [17:49] mac_v: what you're approximately saying is that the applet serves no purpose [17:49] Ng: on the contrary ;) , the applet is just a supplement [17:50] the app needs to do the main work [17:50] Ng: don't say that on ayatana channel ;-) [17:50] mac_v: but the only useful thing it can do is tell you about new stuff, but you don't want it to do that in a way that draws attention. if I was going to go and specifically check that applet I would just flip to evolution and see bold folders [17:51] Ng: the indicator app is there as 1.notification ysstem now has no action support 2. to act as a logger for events when user is not at his computer [17:52] Ng: traditionally we have been trained to view only the color changes , but are the two icons similar > http://launchpadlibrarian.net/34598451/IM.png [17:53] tgpraveen1: 1) is an interesting point - nothing uses it for that atm. 2) would mean it should only appear after coming out of idle/screensaver? [17:54] mac_v: they're similar enough that you have to specifically look at the applet to know if it's trying to tell you something, and that's my fundemental problem. I don't want to have to think "hey do I have an IM/email/identica reply waiting for me" [17:54] ime applets which appear when you have new email are hugely popular on all OSes [17:56] Ng: users even complained for the star icon when used for new ,mail > http://launchpadlibrarian.net/28378360/Mail%20Icons.png and wanted to used red to grab more attention ;) [17:57] mac_v: that red icon is *perfect* [17:57] Ng: why does a user ignore the notify-osd and rather depend on the panel icon ? [17:57] mac_v: notify-osd is transient [17:57] pitti, btw I pushed the nautilus sru for folder settings migration too if you want to accept it today still [17:57] pitti, I should be done for the day almost time for sport [17:58] seb128: thanks [17:58] seb128: btw, for this I guess we need a small code patch to delete the stamp file? [17:58] np, shame that we didn't notice this one before karmic though [17:58] pitti, I'm not sure, do you want to overwrite changes made since the upgrade? [17:58] I would rather let that choice to users [17:58] tricky question [17:59] because the conversion will copy previous settings back [17:59] seb128: could you add some documenting sentence into the bug description how to force the migration? [17:59] * pitti processes SRUs [17:59] pitti, in the launchpad bug you mean? there is that in a comment bug I can update the description too yes [18:00] Ng: if the user is at the desk and misses it then the notify-osd isnt doing the job... it needs to notify better ... if the user is away and returns the icon is different[though you dont agree the change is too different]... but if the color is to be used , as i said earlier for the user who does not want to interact immediately it would be a nuisance :) [18:00] seb128: yes; thanks [18:00] mac_v: hey you know skype might be open-sourced? [18:00] pitti, thank you [18:00] mac_v: no it wouldn't [18:01] mac_v: that's a fallacious argument. I've witnessed people go months with the orange update-notifier icon visible in their panel on hardy. we took it out because a great big orange star isn't annoying enough [18:01] Ng: it would , when the rest of the icons are not having a color then this will stick out :) [18:01] mac_v: sorry, it wouldn't [18:02] Ng: even if it wouldnt... that[update-notifier] brings back that no amount of color or icons change is going to notfiy the user immediately unless the icon blinks ;) [18:03] mac_v: I would notice it, which is good enough for me if blinking is really unthinkable [18:03] Ng: good enough for me isnt a good argument ;) [18:03] * mac_v hides [18:04] mac_v: I don't see anyone producing actual usability/annoyance studies on this, so we're all just throwing opinion around until it sticks [18:04] njpatel: and471, njpatel is free to correct me, but I don't think maximus cares whether a particular application has a full-screen mode. [18:04] ^ from mpt [18:06] I'm off for the evening as well; still doing some debugging, then TKD [18:06] cu tomorrow! [18:07] same here [18:07] bye pitti [18:09] and471: he's right, maximus will only maximise. It does have a fullscreening feature, but it's hardly used since jaunty [18:09] and471: so maximus doesn't require a fullscreen functionality? [18:10] njpatel: so maximus doesn't require a fullscreen functionality? [18:10] * and471 needs to stop talking to himself [18:11] and471: lol, nope. At the moment, in karmic, it doesn't do anything at all for fullscreen...only maximising [18:12] njpatel: thankyou :-) [18:12] and471: the 'feature' was that you could assign a key, say the windows key, to toggle fullscreen on/off. You could then add .desktop files for maximus, so it knew which key combiniation to send (i.e. F11 for Firefox, but Ctrl+Alt+J for OpenOffice) [18:12] np [18:14] njpatel: does software-centre need to include one of them? or is it not worth it [18:14] and471: it doesn't hide much else in fullscreen [18:15] and471: nope, it desont [18:15] er, doesn't [18:15] njpatel: hehe thanks [18:15] Ng: actually i had proposed a better solution for this , make the indicator applet icon be more relevant to the new event , > https://blueprints.launchpad.net/indicator-applet/+spec/differentiate-notification [18:16] that way the user will be notified even quicker about the new event rather than just a generic icon for all types messaging [18:18] and471: lol! , how in the world do you manage to talk to yourself so many times ;p [18:19] mac_v: I don't know, it is becoming an annoying habit .. [18:19] and471: isn't that right and471... [18:19] mvo: are you here, if so could you look at merging? [18:20] https://code.launchpad.net/~rugby471/software-center/software-store-andrew === MacSlow is now known as MacSlow|break [19:28] * pwnguin just wants to point out that the mute icon in the default scheme is hard to distinguish from a full volume indicator [19:32] i turned up my speaker volume before double checking the system wide volume. ouchie [19:33] well, i guess it's somewhat different in shape, not just color. while i'd argue color would improve cognition speed, i can appreciate learning curve arguments, so i guess i'll live with it for a while and see how i acclimate [20:50] is anyone else getting piles of bugs being filed via checkbox? [20:51] bryce, not me [20:55] kenvandine, it appears people are using the checkbox system test thingee as a cheap way to file complaints, but aren't really writing valid bug reports [20:55] but I'm getting dozens a day, it's time consuming to go through and close them [20:57] :/ [21:01] well there is something about logging bugs now and the way to do them on in -classroom so hopefully they'll get beter bryce [21:07] bryce, how does checkbox bug look like? :) [21:08] kklimonda, it has a checkbox tag normally [21:08] (i think) [21:08] tagged 'checkbox-bug' [21:08] ah [21:08] i knew it was something like that ;) [21:09] bug 455691 is an example of one which was filed against X [21:09] Launchpad bug 455691 in checkbox "a" [Undecided,Incomplete] https://launchpad.net/bugs/455691 [21:10] title: a, description: d :D [21:10] bryce - how are X events mapped to X keysyms? (i know that's a bit of an open-ended question). I've just bought myself a new mouse, and it has some extra buttons which produce events in xev, but have no keysyms [21:11] ah, I see now what checkbox actually is [21:12] kklimonda, there's also some like bug 448913 [21:12] Launchpad bug 448913 in checkbox "." [Undecided,Incomplete] https://launchpad.net/bugs/448913 [21:12] most are > 1 char, but not by much [21:13] bryce: I like your comment on that one :) [21:13] chrisccoulson, xkeyboard-config [21:13] wow! [21:13] they said '.', you said '?' :D [21:13] bryce, I like your comment :D [21:13] bryce - thanks:) [21:14] there seem to be several bugs with just '.' in the title o.0 [21:15] chrisccoulson, mouse button mappings might be a bit different, I don't remember. You can use 'xinput' to fiddle the button mappings, and I think there's a gnome tool to configure what they actually do [21:15] bryce - thanks, i'll take a look, and see if i can figure out how to map them [21:16] mac_v, yeah what I think is happening is for however the system test thing is implemented, people either don't realize it's causing a bug report to be filed, or are assuming that the system test knows what's gone wrong better than them, so they don't need to provide a description [21:16] if i figure it out then i could perhaps add it to gnome-mouse-properties and make it easier for other people too [21:16] sounds good [21:17] mac_v - "." in babelfish translates to "i couldn't be bothered to write a description" [21:17] bryce: all those bugs are from the same user, he might have been testing [21:17] mac_v, I *think* the test could be improved to do some automatic analysis so it does describe what the problem is, but it's far from doing that currently [21:17] mac_v, ah that could be. But there's a bunch from other users which are just one sentence descriptions, basically "something's broken" [21:18] yeah , those too [21:18] I'm wondering if 'system test' should be done like apport, and turned OFF at release time === MacSlow|break is now known as MacSlow [21:31] yay, my extra mouse buttons are mapped \o/ [21:46] session inhibiting is such a mess in karmic :( [21:46] bad totem [21:47] chrisccoulson, excellent [21:47] totem doesn't clean up inhibitors if you close it whilst playing a video [21:48] so those inhibitors remain for the lifetime of the session, preventing the screensaver activating [21:48] :( [21:50] chrisccoulson: gnome-session nukes inhibitors if the owner drops off the bus [21:50] mclasen - it doesn't work for inhibitors that are proxied through gnome-screensaver [21:51] gnome-session *thinks* that gnome-screensaver registered the inhibitor [21:51] so they end up being leaked :( [21:51] that is possible [21:51] I think I asked about that specific issue when we ported totem over to use gnome-session [21:51] and was told that would be handled correctly :-( [21:52] mclasen - it isn't being handled correctly at the moment [21:52] i get an inhibitor leaked every time i watch a video [21:58] i wonder why totem isn't cleaning these up anyway :-/ [22:23] hggdh, still around? [22:24] seb128: yes [22:24] hggdh, looking to your gtk2-engine-murrine debdiff on the bug [22:24] not sure that adding a patch system in a sru is a good thing [22:25] well... what should I do then? [22:25] the debdiff also has config.guess,config.sub noise [22:26] the package was generated from a standard debuild -S -si, from a clean archive. Not sure where the noise came from [22:26] same distro serie? ie karmic? [22:27] yes, I pull-lp-source from karmic [22:27] the rules clean target has a rules to update those [22:27] ok, no problem, I will drop those from the debdiff and sponsor the update [22:27] let's see if somebody complains about the patch system adding [22:27] seb128: but what should I do re. a patch on clean source? [22:28] I guess I would not have said anything if that was no quilt :-) [22:28] hggdh, no don't bother [22:28] (hate quilt) [22:28] thanks for the work on it [22:28] heh. I just copied from the Jaunty version -- it used quilt there, I went with it, minimum changes and all of that [22:28] will you do the gtk2-engines change too? [22:28] yes [22:29] personally, I do not like quilt either ;-) [22:31] hehe [22:31] brb session restart [22:39] hey seb128 [22:39] hello chrisccoulson [22:39] how are you? [22:39] seb128 - i'm not too bad thanks [22:39] and you? [22:40] I'm good thanks [22:40] not to bad is not great either [22:40] seb128: a question re. gtk2-engines: should I open a new bug for it (since it is a different patch)? [22:40] seb128 - could i ask you to test something on your fresh session? [22:40] you should stop overworking yourself with ubuntu tasks ;-) [22:40] chrisccoulson, yes? [22:40] hggdh, no, we have a task for each one on the bug [22:41] seb128 - could you try playing a video in totem, and then closing totem without stopping the video first [22:41] seb128: k, thanks [22:41] seb128 - and then run: [22:42] dbus-send --session ^Cdest='org.gnome.SessionManager' --print-reply --type=method_call /org/gnome/SessionManager org.gnome.SessionManager.GetInhibitors [22:43] and see if it outputs anything? [22:43] i'm just intrigued if i am the only person seeing this issue [22:43] oops, make that: dbus-send --session dest='org.gnome.SessionManager' --print-reply --type=method_call /org/gnome/SessionManager org.gnome.SessionManager.GetInhibitor [22:44] (without the ^C) [22:44] "Must use org.mydomain.Interface.Method notation, no dot in "/org/gnome/SessionManager" [22:44] " [22:44] I hate this new ctrl-C behaviour [22:44] yeah, i tried to copy with ctrl-C [22:45] chrisccoulson, I get the message I copied [22:46] hmmm [22:46] i wonder if there's a copy/paste error? [22:46] it seems to work ok here [22:47] dbus-send --session dest='org.gnome.SessionManager' --print-reply --type=method_call /org/gnome/SessionManager org.gnome.SessionManager.GetInhibitors [22:47] i don't know if xchat mangles anything along the way? [22:49] chrisccoulson, use pastebin? [22:49] seb128: the debdiff for gtk2-engines is in the bug. [22:49] hggdh, thanks, I uploaded the murrine one, I assigned the bug to you too since sru should have an assignee [22:50] ie somebody responsive to read user feedback after upload [22:50] oh, OK, was not aware of that, thanks [22:50] no problem [22:50] chrisccoulson: I'll have a patch in a bit [22:51] seb128 - http://paste.ubuntu.com/307982/ [22:51] mclasen - oh, ok, i was going to work on one too ;) [22:51] but if you're already working on it, then that's ok [22:51] yeah, I pondered letting you work it out yourself....but that would be cruel :-) [22:51] mclasen - i opened gnome bug 600488 fyi [22:51] Gnome bug 600488 in general "Totem is leaking session inhibitors" [Major,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=600488 [22:52] ok, I'll put it there [22:52] brb [22:52] mclasen - thanks [22:53] chrisccoulson, still same messages [22:53] chrisccoulson, d-feet says [] though [22:53] chrisccoulson, so I think it's just that no inhibitor get in action [22:54] seb128 - you don't seem to be seeing the same behaviour as me then. do you get an inhibitor whilst a video is playing? [22:54] yes [22:54] but it goes away when I close totem [22:54] it gets removed it totem is closed normally [22:54] it doesn't get removed it totem falls off the bus [22:55] two different code paths in g-s-s [22:56] mclasen - i'm seeing different behaviour. i see them not being removed if i close totem whilst a video is still playing [22:56] * chrisccoulson wonders it totem is crashing on shutdown [22:58] hmmm, strange, it seems to be closing normally [23:03] hum, I got my left click stucked in selection mode in xchat [23:03] weird [23:03] had to restart xorg to fix it [23:05] seb128 - what was the actual symptom? i only started using xchat a couple of days ago, and noticed an issue this evening where i couldn't click my mouse on any other window, and had to restart xorg to fix it too [23:05] it acted like left click was always on in xchat [23:06] ie moving the cursor would select text and never stop [23:06] left click somewhere else would do nothing [23:06] in most cases just killing xchat will work too ;) [23:06] or whatever application does weird things [23:06] I closed everything open without luck [23:07] brb another quick try to this one [23:12] re [23:14] hey seb128, how do I get a package included in universe? [23:14] I made a package with all the wallpapers we couldn't include on the CD [23:14] hi kwwii, you find somebody who has upload right to package it? [23:14] open a bug against ubuntu and subscribe sponsors [23:14] I packaged it myself and it works fine [23:14] the universe sponsors [23:14] cool, thanks ;) [23:15] you're welcome [23:22] bryce, is there a way for a script to detect that a bug was filed with checkbox? [23:22] * rickspencer3 wonders if we can "auto-ignore" if there are annoying [23:30] rickspencer3: sure, they all get tagged [23:31] re [23:31] sorry got a crappy internet tonight [23:37] mclasen - thanks for the patch [23:38] i will try it out shortly :) [23:40] seb128: I proposed a fix for bug 460050; I will wait for upstream revision before doing a debdiff [23:40] Launchpad bug 460050 in evolution "Pressing Enter or Tab in a new memo line (in memo list) exits Evolution" [Low,Triaged] https://launchpad.net/bugs/460050 [23:40] hggdh, thanks