[12:02] <ivoks> hello
[12:02] <ivoks> pitti: i'm almost done... :)
[12:03] <pitti> ivoks: cool!
[12:03] <ivoks> cups part is done
[12:03] <ivoks> i had one bug in g-c-m
[12:03] <ivoks> but now i'm rebuilding fixed veresion
[12:03] <ivoks> version
[12:04] <ivoks> argh... i forgot one detail :)
[12:04] <ivoks> patch for cupsd.conf...
[12:04] <ivoks> ok, tomorrow you'll have everything in your mail
[12:31] <pitti> good night everyone
[12:31] <ogra> night pitti
[12:31] <ivoks> 'night
[12:36] <jordi> ok, so I have new Catalan dictionary pacakges which fix a good deal of problems the current dapper version has.
[12:41] <zul> heylo
[01:05] <mdke> mako, around by any chance?
[01:10] <jmg> hi all
[01:18] <jordi> mdz: ok, I think this is all I need to do, #41678
[01:18] <jordi> mdz: TELl me if you need anything else.
[01:29] <mdz> jordi: that's fine
[01:29] <jordi> mdz: cool
[01:29] <mdz> jordi: there is nothing more to TELl
[01:30] <jordi> mdz: that was a CAPSLOCKPROBLEM I TELL YOU
[01:30] <mdz> oh sure
[01:31] <jmg> gah
[01:31] <Kamion> right, that had better be the last ubiquity upload I need for beta2
[01:31] <mdke> as long as the internet is working, right?
[01:31] <jordi> Kamion: ?
[01:31] <Kamion> jordi: !
[01:31] <jordi> ubiquity upload?
[01:32] <jordi> I thought you meant ispellcat. nm :)
[01:32] <Kamion> a popular beat combo^W^W^Wpackage, m'lud
[01:33] <jordi> Kamion: will you be in mx?
[01:33] <jordi> or you, mdz?
[01:33] <mdz> jordi: ispellcat is slightly lower priority than "beta eats disks" ;-)
[01:33] <mdz> jordi: I will, yes.
[01:33] <jordi> oh man
[01:33] <jordi> I'm fearing that my tactics to do a last min plan to go there will fail.
[01:33] <jordi> You'll have to pop the trunk w/o me :|
[01:33] <Kamion> jordi: unfortunately not
[01:34] <jordi> mdz: what do you mean lower priority? This is the end of ubuntu! Ubuntu will be as relevant as Caldera after this
[01:34] <jmg> "beta eats disks"?
[01:34] <jordi> Kamion: that sucks :/
[01:35] <jmg> mehico?
[01:36] <mdz> jordi: I expect to be working most of the time anyway
[01:36] <jordi> mdz: yes, it's surprising that you'll be there actually, the release will be so near
[01:37] <jordi> k this is embarrasing
[01:37] <jordi> can't find my way to the d-i translation template url
[01:38] <Kamion> jmg: launchpad.net/bugs/40464
[01:38] <jordi> aha
[01:38] <Kamion> sometimes you just have bug numbers memorised ...
[01:38] <mdz> jordi: I was unsure about going, but in the end I think I need to represent us there
[01:39] <jordi> yes
[01:39] <jordi> is mark not going?
[01:46] <infinity> Hrm.  Does anyone else think that the flat version of our vendor logo looked much cleaner and more professional than the sketchy 3d-ish version I now have in the corner of my panel?
[01:47] <infinity> (Nevermind that the official Ubuntu logo /is/ flat, not "puffy")
[01:47] <ogra> infinity, it was never flat, it was punched in
[01:47] <infinity> ogra: The logo itself was flat.
[01:47] <infinity> ogra: The fact that the flat logo was then punched into the background is irelevant.
[01:48] <ogra> the vendor logo we used in breezy was punched ;)
[01:48] <ogra> or since breezy 
[01:48] <infinity> Yes, punched into the background.  Flat logo.  Unless I can't remember yesterday.
[01:48] <infinity> Which is possible.
[01:49] <infinity> I just know that today's looks less clean.  Too much dither in an attempt to make a (very) tiny icon 3D.
[01:49] <ogra> it had a slight black shadow on the upper left side of the circle (you would have noticed in 48px size)
[01:50] <infinity> Erm, yes.  We're talking past each other.
[01:50] <infinity> The LOGO was flat, and was then "punched in" to the "background".
[01:50] <ogra> i'd rather liked to have seen that effect more worked out than a blurry shadow around it
[01:50] <infinity> That's different from the current logo, which is now "puffy" and doughnut-shaped. :)
[01:50] <ogra> yep
[01:51] <ogra> what i was after was to point out that the punching doesnt affect the sharpness ;)
[01:51] <ogra> while a shadow does
[01:53] <infinity> A shadow would have little effect either, it's the insistence in turning the flat logo into a torus that affects sharpness. :)
[01:53] <infinity> (The shadow under it isn't spectacular either, but pales in comparison to the horror that is the torus.
[01:54] <infinity> )
[01:56] <ogra> hmm i really need to update my other machine ... the edubuntu logo doesnt/wont have a shadow 
[01:56] <ogra> i only saw the proposals for the shadow stuff on ubuntu-art
[02:00] <jmg> hmm
[02:01] <jmg> anyone know a way to compare file1 and file2, and not display a line from file1 if it is in file2?
[02:02] <mdz> jmg: sort + comm
[02:03] <mdz> jordi: yes, he is
[02:25] <mako> mdke: i'm around now
[02:55] <jsgotangco> does the dvd still follow the install/live structure we had on breezy?
[03:00] <zul> heylo
[03:03] <jmg> jsgotangco: are we even making a dvd?
[03:04] <jsgotangco> jmg: we have a dvd for 5.10 (install and live in one  disc)
[03:04] <jsgotangco> mako: nice to see you online again
[03:05] <jmg> ah i remember
[03:05] <jmg> i should go try uninstall kubuntu now
[03:15] <Surak> bug #30701 should be marked as upstream, but how? the package list does not show anything related to zodb.
[03:15] <Ubugtu> Malone bug 30701 in zodb "python2.4-zodb should depend on python2.4-zopeinterface?" [Normal,Unconfirmed]  http://launchpad.net/bugs/30701
[03:16] <Surak> it is related with debian bug #360493 , but I don't know on which package I should add a watch on this specific case.
[03:16] <Ubugtu> Debian bug 360493 in python2.3-zodb "Subject: python2.3-zodb: Conflicts with python2.3-zopeinterface" [Normal,Closed]  http://bugs.debian.org/360493
[03:17] <Surak> perhaps ubuntu-bugs is best suited for asking this.
[03:22] <jmg> hmm
[04:38] <bddebian> Heya
[04:56] <TheMuso> /c/
[08:11] <netstar> what's wrong with xorg?
[08:12] <HrdwrBoB> it doesn't like you
[08:12] <netstar> oh no
[08:12] <ivoks> question of the week
[08:12] <netstar> I see no problem
[08:12] <netstar> Though I haven't rebooted in 3 days
[08:12] <ivoks> netstar: it's normal for packages to break in development version
[08:13] <netstar> ivoks: sure
[08:13] <fabbione> netstar: nothing you need to worry about
[08:13] <fabbione> if you had the problem you would have come here screaming
[08:21] <Mithrandir> (removed xorg; it has been fixed for > 12 hours)
[08:21] <fabbione> Mithrandir: i am taking a lock on X
[08:22] <Mithrandir> fabbione: please do.
[08:22] <ivoks> you can put ubuntu+1 for questions about dapper :/
[09:30] <pitti> Good morning
[09:30] <netstar> guten tag
[09:30] <netstar> Still having issues with hostap over orinoco-cs
[09:33] <freeflying> pitti: hi, how about split language-support-zh into language-support-zh-cn and language-support-zh-tw?
[09:33] <netstar> it hangs the machine on boot, until I remove the PCMCIA card, the only solution is remove the hostap kernel modules
[09:34] <pitti> freeflying: that's not how we designed them to work unfortunately
[09:47] <zakame> hi all
[09:53] <Kamion> more verbosely, the current Ubuntu/Kubuntu/Edubuntu daily-live builds are beta2 candidates; please test them
[09:53] <Kamion> Xubuntu will be forthcoming later this morning
[09:54] <zakame> yay \o/
[09:55] <zakame> hmm has kdrill been synced? re, malone 28810
[09:55] <Ubugtu> Malone bug 28810 in kdrill "undeclared dependency on libXp6" [Major,In progress]  http://launchpad.net/bugs/28810
[09:59] <Kamion> zakame: syncs are broken at the moment; will hopefully be repaired soon
[10:02] <zakame> Kamion: ah, yes I remember now from qprocd...
[10:03] <kagou> hi
[10:04] <zakame> hello kagou 
[10:08] <pitti> Kamion: do you have any objections against putting sqwebmail (and with it courier-maildrop) into supported? we have to support its source 'courier' anyway, and it would give us a webmailer (I was often-asked for that)
[10:13] <dholbach> pitti: you better ask the support department :-p
[10:14] <dholbach> but I agree that it might be nice to have
[10:15] <jdub> sqwebmail is a bit... scary
[10:16] <dholbach> it might be a bit tough to support hula  :)
[10:16] <jdub> yeah, someone should totally finish or release that or something ;)
[10:17] <dholbach> finish as in what? as in remove all the old crufty code from it?
[10:17] <dholbach> i mean... it *looks* nice
[10:17] <jdub> it's in pretty thick devel atm, replacing chunks, heaps of stuff
[10:18] <dholbach> yeah and debian do a good job in packaging it
[10:18] <jdub> even recent stuff?
[10:18] <dholbach> yeah, more than we do
[10:19] <jdub> hrm, last update was november though
[10:19] <jdub> of an svn revision first uploaded in september
[10:19] <zakame> hm, which packge?
[10:19] <jdub> hula
[10:20] <dholbach> oh?
[10:20] <jdub> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=360245
[10:20] <jdub> ha ha
[10:20] <Ubugtu> Debian bug 360245 in hula "Subject: hula: Please package newer versions" [Wishlist,Open]  
[10:21] <kagou> jdub: :D
[10:21] <dholbach> but still there was more action in the debian hula team
[10:24] <jdub> i think that lines up fairly closely with the massive crazy changes going on in hula
[10:24] <jdub> As of today, the latest is r1212, while
[10:24] <jdub> version in unstable is r379.
[10:24] <jdub> september vs. march
[10:24] <jdub> funy
[10:24] <jdub> funny
[10:25] <kagou> pitti: i think that in near futur we must support a groupware tool too. Like http://www.phpgroupware.org/ http://mirror.open-xchange.org/ox/EN/community/ or http://egroupware.org/
[10:26] <pitti> kagou: arrrgh php web applications
[10:26] <kagou> these tools including webmail
[10:26] <dholbach> Gloubiboulga: I uploaded goffice patch
[10:26] <dholbach> Gloubiboulga: thanks for your work
[10:26] <Gloubiboulga> dholbach, thanks! I'll prepare the gnumeric patch then
[10:26] <dholbach> Gloubiboulga: rock on
[10:27] <highvoltage> gnumeric, yeah!
[10:27] <dholbach> Gloubiboulga: do you have any other stuff in the pipeline?
[10:27] <dholbach> Gloubiboulga: or janimo? wrt building more versions of stuff
[10:28] <dholbach> Gloubiboulga: did you guys want to look into updating abiword too?
[10:28] <Gloubiboulga> dholbach, you mean for gtk/gnome packages ?
[10:28] <dholbach> Gloubiboulga: yeah
[10:28] <dholbach> bug 39612 is still open about abiword
[10:28] <Ubugtu> Malone bug 39612 in abiword "Request for UVF exception - Release of Bugfix-only AbiWord 2.4.4" [Normal,Confirmed]  http://launchpad.net/bugs/39612
[10:28] <Gloubiboulga> abiword has no gnome dependencie iirc
[10:28] <dholbach> yeah
[10:28] <dholbach> i rather meant the update
[10:29] <dholbach> anf uvf exception etc
[10:29] <dholbach> s/anf/and
[10:29] <Gloubiboulga> we didn't plan to work on this, but I guess we could
[10:29] <Gloubiboulga> hub doesn't package this new release?
[10:30] <dholbach> Kinnison: could you tell me about your gpm plans? would it be ok, if I'd patch some icons in? (patch adding new files, modifying rules,control) or do you want me to wait until you uploaded the new stuff?
[10:30] <dholbach> Gloubiboulga: hub is upstream, but never was much involved in packaging, I guess.
[10:30] <Gloubiboulga> dholbach, ah ok, I thought he was the package maintainer too
[10:31] <nomed> Gloubiboulga, i'll be happy to help you on that later 
[10:32] <Gloubiboulga> nomed, great! and hi ;)
[10:32] <nomed> for testing i 'm available even now :)
[10:35] <Gloubiboulga> nomed, could you /join #xubuntu?
[10:35] <nomed> Gloubiboulga, i'll not be really active but yes :)
[10:37] <siretart> is dapper/main frozen currently?
[10:37] <dholbach> siretart: no
[10:37] <siretart> ok. thanks
[11:13] <fabbione> mdz: FYI: x11-common already Conflicts with xorg-common.. but as well Replaces it...  xorg-common gets removed, but not purged on breezy -> dapper upgrade
[11:15] <doko> pitti: please could you have a look at g-c-m before an upload?
[11:15] <pitti> doko: yes, of course, if I can be of any help...
[11:16] <pitti> doko: maybe vuntz can take a look at it, too, he certainly knows gtk :)
[11:16] <doko> pitti: thanks, http://people.ubuntu.com/~doko/ 
[11:16] <pitti> doko: can you put a debdiff there?
[11:16] <doko> pitti: mvo already helps with gtk ...
[11:17] <Keybuk> fabbione: what are you wittering on about depmod for in bug 30241 ?
[11:17] <Ubugtu> Malone bug 30241 in sysvinit "Sometimes fails to mount nfs directories" [Major,Confirmed]  http://launchpad.net/bugs/30241
[11:17] <Keybuk> it has *nothing* to do with depmod
[11:17] <pitti> doko: oops, why does your patch remove lots of stuff in po/Changelog?
[11:18] <pitti> doko: in debian/control:
[11:18] <pitti> -Replaces: libgnomecupsui1.0-1
[11:18] <pitti> +ReplaceS: libgnomecupsui1.0-1
[11:18] <pitti> :)
[11:18] <doko> ?
[11:18] <vuntz__> pitti, doko: mmmh?
[11:18] <fabbione> Keybuk: so explain me why if i boot on a new install kernel it does not work. I reboot again it does not work. I run depmod -a and reboot and it works.
[11:18] <pitti> doko: wow, that changelog looks impressive
[11:19] <mvo> fabbione: the upgrade tools purges it btw
[11:19] <pitti> doko: that debian/control change was certainly not deliberate
[11:19] <fabbione> mvo: yes i am trying to understand why mdz did ask for a Conflict that was already there
[11:19] <mvo> ah, o
[11:19] <mvo> k
[11:19] <Keybuk> fabbione: no idea, little voodoo pixies
[11:19] <Keybuk> what doesn't work?
[11:19] <fabbione> Keybuk: mount /home over nfs
[11:19] <fabbione> exactly as reported in the bug
[11:20] <Keybuk> that's just a race-condition, exactly as described in the bug
[11:20] <fabbione> or for the matter.. anything over /nfs
[11:20] <pitti> doko: oh, funny, I debdiffed it against 1.1ubuntu5, but I just noticed that I never uploaded that version
[11:20] <Keybuk> network cards are brought up in the background, so may not be finished yet by the time the boot sequence gets to S45mountnfs.sh
[11:20] <fabbione> Keybuk:  i can reproduce it regularly..
[11:20] <Keybuk> when you install a kernel, depmod is run in the postinst
[11:20] <pitti> doko: this version set custom PPD dir to /usr/share/ppd/custom
[11:20] <pitti> doko: and also fixed 'ReplaceS' in debian/control.in
[11:21] <fabbione> Keybuk: why do you think i did underline in the running kernel?
[11:21] <pitti> doko: can you integrate my changes into yours? I'll send you a debdiff
[11:21] <Keybuk> fabbione: that makes no difference
[11:21] <fabbione> Keybuk: i don't know what diff it can make, but apparently it does
[11:21] <Keybuk> then prove to me that running depmod against an installed kernel produces different output to running it in a running kernel
[11:21] <Keybuk> (it doesn't)
[11:21] <fabbione> Keybuk: ok, let's put it in another way..
[11:21] <Keybuk> it's an easy thing to prove.  Install kernel, backup modules.dep, modules.alias, etc. boot into it, run depmod -a, compare
[11:22] <fabbione> sorry for the noise
[11:22] <fabbione> i was trying to give you more input
[11:22] <fabbione> it's your bug anyway. fix it
[11:22] <doko> pitti: sure.
[11:22] <Keybuk> it's a difficult fix, sadly
[11:22] <fabbione> well -server needs it
[11:22] <doko> pitti: yeah, there are other changes ...
[11:22] <Keybuk> sure, it will be fixed
[11:22] <Keybuk> I'm not sure how to cope with the /usr-on-NFS problem though
[11:23] <pitti> doko: http://people.ubuntu.com/~pitti/tmp/g-c-m.ppd-dir.diff
[11:23] <pitti> doko: so you should name your version ubuntu5, too
[11:26] <doko> pitti: hmm, yes, there was an -ubuntu5 version on p.u.c for some time. better name it ubuntu6
[11:26] <pitti> ok, no big deal
[11:27] <doko> pitti: the cups_dir part seems to be wrong, it will _only_ find the custom printers then
[11:30] <pitti> doko: I think it's correct, it just uses the value for installing new PPDs
[11:30] <pitti> doko: I just checked the code again
[11:30] <pitti> doko: the retrieval of the PPD list is not done by g-c-m, it's done by libcups through libgnomecups
[11:31] <pitti> doko: (CUPS-Get-PPDs IPP command)
[11:31] <fabbione> who does load alsa/sound modules today?
[11:31] <fabbione> who/what
[11:31] <pitti> udev normally
[11:31] <fabbione> ok
[11:31] <fabbione> thanks
[11:31] <pitti> apart from e. g. the snd-powermac
[11:31] <pitti> this has to be in /etc/modules
[11:31] <fabbione> ok thanks
[11:33] <fabbione> pitti: do you know if we are using modprobe.d or modutils to add parameters to modules on load?
[11:33] <fabbione> i am not in the mood to reboot to figure it out :)
[11:33] <pitti> fabbione: I'm 90% sure that it's /etc/modprobe.d/*
[11:33] <fabbione> ok thanks
[11:35] <Keybuk> modprobe.d
[11:35] <Keybuk> see the bottom of /etc/modprobe.d/alsa-base
[11:36] <Keybuk> pitti: do you have a powerbook handy?
[11:36] <pitti> Keybuk: just an iBook
[11:36] <Keybuk> pitti: can you "modinfo snd-powermac" for me?
[11:37] <Keybuk> my powerbook is out of action until lunchtime when I can go get a CD lens cleaning kit for it
[11:37] <pitti> I usually have it on STR, but I did an espresso test install last night
[11:38] <Keybuk> yeah, I did a test install which failed due to being unable to read the CD
[11:38] <Keybuk> and now it won't read *any* CD
[11:43] <pitti> Keybuk: I have the modinfo here, what do you need?
[11:43] <Keybuk> just the alias list, if there is one
[11:44] <pitti> Keybuk: there's none at all
[11:45] <Keybuk> ok
[11:45] <Keybuk> that's why it isn't loaded automatically then
[11:45] <Keybuk> wondered whether it was like the bmac module, which does have an alias list, but still isn't loaded
[11:49] <ivoks> pitti: ping
[11:49] <pitti> hi ivoks
[11:49] <pitti> thanks for the patch
[11:49] <ivoks> np
[11:50] <ivoks> i have an idea for samba too :)
[11:50] <ivoks> for sharing printers over samba
[11:50] <ivoks> maybe not dapper material, but could be for edgy
[11:53] <Keybuk> fabbione: what was the bug you filed and rejected again? :)
[11:55] <fabbione> Keybuk: the via module and the uart did change name in .15
[11:55] <fabbione> via82cxx to via82xx
[11:56] <fabbione> you should probably recheck that the ones mentioned in alsa-base still match on .15
[11:57] <Keybuk> ahh
[11:57] <Keybuk> the alsa-base file is generated by code in the alsa drivers :)
[11:57] <Keybuk> so I always figure that one's right
[12:29] <jdub> pitti, carlos: would you regard the launchpad translation stats for dapper as accurate in real terms (ie. would users agree with them)?
[12:29] <pitti> jdub: unless they have an error in their counting algorithm, why should they be wrong?
[12:30] <pitti> jdub: you mean that the missing ones are often translations which normal users don't see anyway?
[12:30] <pitti> jdub: (like translated gcc error messages and that stuff)
[12:30] <jdub> well, the stats take *everything* into account right? so might be skewed by lack of translations for random cli things compared to desktop things, stuff like that
[12:31] <pitti> right
[12:31] <carlos> jdub: yeah, the stats are for all main packages
[12:31] <carlos> not just the desktop
[12:31] <jdub> 'cos the top ten in the list seems really odd compared to gnome top ten, things like that
[12:32] <jdub> (heh, never noticed that dapper's release date in launchpad is "when it's done")
[12:33] <Coyctecm> what's the status of cnr warehouse thing in ubuntu?
[12:34] <jdub> Coyctecm: "it was a random comment from kevin carmony - no real status"
[12:35] <Coyctecm> jdub: ok, and I really hope nobody ever even think about that kind of stuff..
[12:35] <jdub> pitti: 8)
[12:36] <jdub> Coyctecm: run gnome-app-install - it's similar enough in concept :)
[12:36] <Coyctecm> jdub: :P
[12:37] <carlos> jdub: gcc and openoffice adds a bunch of strings
[12:37] <carlos> jdub: also, you have there KDE
[12:38] <jdub> carlos: mmm, so it's hard to use those stats to get a feeling for impact
[12:38] <sivang> re all
[12:39] <carlos> jdub: we could define a better way to get those stats like Kubunut, XUbuntu and Ubuntu ones taking into account just the ones we have inside the CDs
[12:39] <carlos> and things like that
[12:40] <jdub> carlos: not meaning to dump work on you though, just getting a feel for what it means myself :)
[12:40] <jdub> carlos: but figuring stuff out like that would be pretty cool
[12:41] <jdub> perhaps splitting the stats into seeds would be helpful
[12:41] <carlos> jdub: I don't think I would do it before dapper release, but we could add it as a wishlist to implement when we have time...
[12:41] <carlos> jdub: do we have such information already inside soyuz?
[12:41] <jdub> yeah
[12:41] <jdub> probably not
[12:41] <jdub> well, maybe
[12:41] <jdub> kamion's offline atm
[12:41] <jdub> Kinnison: ping?
[12:42] <jdub> i didn't think seed management was in soyuz yet, but maybe it knows about it through other means
[12:43] <jdub> heh
[12:43] <jdub> "Choose "Select languages..." to ensure that the languages you speak are included in all translation pages."
[12:43] <jdub> s/speak/understand/ ? ;-)
[12:43] <carlos> jdub: we would need that information in soyuz to be able to generate those stats
[12:44] <carlos> jdub: well, Rosetta is to do translations ;-)
[12:44] <carlos> you need more than just understand to do translations...
[12:44] <jdub> speaking and reading are different though ;)
[12:44] <carlos> but yes, perhaps 'speak' is not the best term
[12:44] <jdub> man, that serbian team is rocking pretty hard
[12:45] <jdub> 39 contributors
[12:45] <jdub> and kicking the crap out of everyone else
[12:46] <jdub> oh, the column sorting is client-side, hey?
[12:46] <carlos> right
[12:58] <spacey> whats serbian?
[12:59] <Treenaks> spacey: 'from Serbia'
[01:02] <neutrinomass> Is the preferred place for temporary files in /var/tmp or in /tmp ?
[01:02] <spacey> Treenaks: whats the dutch name for that?
[01:03] <Treenaks> spacey: Servisch
[01:04] <spacey> aaaah! Servisch
[01:04] <spacey> home of molosofiets
[01:07] <spacey> quite cool to see countries you don't really think about rock much harder then us lazy dutchies
[01:09] <Treenaks> spacey: you know this is #ubuntu-devel?
[01:09] <jdub> spacey: well, you know, it's much easier when you really have a language you can call your own
[01:23] <dsas> isn't the national language of Australia English?
[01:25] <Treenaks> dsas: As a precaution?
[01:25] <dsas> just a check-up :-)
[01:27] <thom> dsas: well, they speak something that sounds similar
[01:43] <fabbione> hey thombot!
[01:43] <doko> pitti: bug 21722 is this a permission problem?
[01:43] <Ubugtu> Malone bug 21722 in gnome-cups-manager "A4 paper size unalterable in printer setup" [Normal,Needs info]  http://launchpad.net/bugs/21722
[01:45] <pitti> doko: whoa, g-c-m modifies /etc/pagesize???
[01:45] <StevenK> Heh
[01:45] <fabbione> EEEEK
[01:45] <pitti> erm, papersize even
[01:45] <fabbione> it's a configfile for libpaper
[01:45] <pitti> fabbione: ... if run as root
[01:45] <StevenK> pitti: Heh, I was about to correct you.
[01:45] <fabbione> if g-c-m does that is really BAD BAD BAD
[01:46] <StevenK> What should it do instead? Yet another dotfile?
[01:46] <pitti> fabbione: well, it's not a conffile, so it's not as evil as it seems
[01:46] <fabbione> it is a config file
[01:46] <pitti> fabbione: but still, this should go into the cups config, not /etc
[01:46] <fabbione> check libpaper sources
[01:47] <pitti> yes, I know
[01:47] <StevenK> fabbione: Not on breezy, anyway.
[01:47] <fabbione> StevenK: it has always been
[01:47] <fabbione> a conffile for libpaper i mean
[01:48] <StevenK> dpkg -s should spit out conffiles, no?
[01:49] <fabbione> StevenK: i can have a conf file that's generated by postinst
[01:49] <StevenK> Oh, duh.
[01:49] <pitti> fabbione: that's not a conffile then, just a configuration file
[01:49] <StevenK> Correct.
[01:49] <StevenK> Which means g-c-m can fiddle with it.
[01:49] <pitti> i. e. it does not belong to the package (for dpkg's sake)
[01:49] <fabbione> configuration -> conf...
[01:49] <fabbione> comen on :)
[01:50] <StevenK> fabbione: No, they are different.
[01:50] <pitti> conffile is special :)
[01:50] <_ion> Methinks there should be metapackages such as "ubuntu-desktop-printing", "ubuntu-desktop-pcmcia", "ubuntu-desktop-bluetooth" which would be installed by default, but a user _would_ have a chance to not have e.g. Bluetooth and PCMCIA related stuff installed while still having ubuntu-desktop.
[01:50] <fabbione> pitti: well dpkg sake ... we can argue about ar sake :P
[01:50] <fabbione> StevenK: no, really???
[01:51] <StevenK> fabbione: I'm trying to help. If you snap back at me, I'll stop.
[01:51] <StevenK> fabbione: I've had a crap day, so .....
[01:51] <fabbione> StevenK: add a ;) to the above..
[01:51] <StevenK> Ah
[01:51] <fabbione> it wasn't meant to be offensive
[01:51] <fabbione> sorry
[01:51] <StevenK> It's fine, adding the smiley makes it unoffensive.
[01:52] <StevenK> fabbione: Point is, if it's a conffile g-c-m can't touch it, and if it's a configuration file, g-c-m can touch it.
[01:52] <fabbione> StevenK: yes i remember the rule.. i did mix conffile with configurationfile
[01:52] <fabbione> they sound toooooo similar
[01:53] <fabbione> it's all gtk fault
[01:53] <dholbach> pffft
[01:53] <StevenK> Heh.
[01:53] <StevenK> Sure, blame the poor library.
[01:53] <StevenK> ;-)
[01:54] <Kamion> hey folks; how's beta2 testing going?
[01:54] <StevenK> Kamion: .... testing ...?
[01:54] <StevenK> Kamion: I'm sorry, what does that word mean? :-P
[01:59] <thom> StevenK: it means you run lintian before you upload
[01:59] <StevenK> Hah
[02:06] <ogra> Kamion, found one minor utf8 bug http://people.ubuntu.com/~ogra/edubuntu/live-install/Screenshot-Install.png
[02:07] <Kamion> ogra: I noticed that too. Feel free to file it; nobody else has yet
[02:07] <ogra> Kamion, and the guys in #edubuntu had probs with cancelling the install from the progress bar (install.py hangs) bugs are filed
[02:08] <Kamion> I saw those bugs, yes, although the person who filed it neglected to tell me what version of anything he was using :P
[02:08] <ogra> heh
[02:08] <ogra> the current beta2 candidate of edubuntu
[02:08] <Kamion> ok
[02:08] <Kamion> not sure whether to hold for that or not
[02:08] <ogra> he has also probs with the resolution 
[02:09] <ogra> his laptop only detects 640x480 the app window is to big to reach the buttons in that mode
[02:13] <doko> Diziet: font ping
[02:17] <ogra> Kamion, they both test in vmware... might be a vmware bug... *real HW* tests are in progress
[02:18] <pitti> Kamion: do you need a ppc ubiquity test?
[02:18] <ogra> yep :)
[02:21] <Coyctecm> why ubuntu's default gtk1.2 fonts are not configured to look clean? they are huge
[02:22] <_ion> Nobody uses gtk1.2. :-)
[02:22] <tseng> because ubuntu's "defaults" don't use gtk1.2 to the best of my knowledge
[02:22] <tseng> maybe something in edubuntu
[02:22] <Coyctecm> well i use gtk 1.2 :)
[02:22] <Coyctecm> well xmms only
[02:22] <ogra> tseng, ??
[02:23] <tseng> ogra: dont you have some silly gtk1 apps?
[02:23] <ogra> nope
[02:23] <tseng> cool
[02:23] <ogra> gtk1.2 should have died long ago :)
[02:24] <pitti> unfortunately we can't demote it :(
[02:24] <Coyctecm> that could be nice touch if they were probeply configured by default
[02:24] <fabbione> pitti: what's left with it?
[02:24] <ogra> pitti, at some point we'kk be able to
[02:24] <ogra> *we'll
[02:24] <Kamion> pitti: always
[02:25] <pitti> fabbione: imlib, libdv, libiodbc2, smpeg, xmms
[02:25] <Coyctecm> there are few apps that people use and they are gtk 1.2 based
[02:25] <Kamion> ogra: the 640x480 thing is a known bug, we can't easily fit ubiquity into that screen size (in fact we decided not to try at the UI sprint) and I haven't much looked at trying yet
[02:25] <Coyctecm> like xmms, mplayer-gui etc.
[02:25] <ogra> Kamion, i dont think its a ubiquity bug anyway
[02:25] <pitti> fabbione: xmms coudl be demoted, but kdegraphics is the only thing that requires us to keep the old imlib (another dup)
[02:26] <ogra> Kamion, the laptop doe higher resolutions usually
[02:26] <Coyctecm> nerolinux uses gtk 1.2
[02:26] <fabbione> pitti: ok
[02:26] <ogra> pitti, but that could be solved for eft
[02:26] <fabbione> Coyctecm: -ENOCARE.. nerolinux is non-free
[02:26] <Surak> what 640x480 thing? which video board is that? via unichrome did this in breezy, but it works now
[02:26] <pitti> ogra: I truly hope so
[02:27] <ogra> and mplayer-gui is in multiverse :)
[02:27] <pitti> ogra: keeping them (imlib and gtk1.2) in dapper just hurts security-wise
[02:27] <ogra> yes
[02:27] <Coyctecm> fabbione: yes, i don't use it myself, but many many users use it
[02:27] <ogra> and clutter wise :)
[02:27] <fabbione> Coyctecm: demoting a library does not mean killing it
[02:27] <jdub> pitti: how much do we care about kdegraphics?
[02:27] <ogra> Coyctecm, but that doesnt mean it needs to be in the default install
[02:27] <pitti> jdub: well, I don't :)
[02:28] <Riddell> jdub: lots
[02:28] <ogra> edubuntu neither 
[02:28] <jdub> Riddell: what's in it?
[02:28] <pitti> it's a central KDE component
[02:28] <jdub> bummer
[02:28] <jdub> imlib1
[02:28] <jdub> that's pretty scary
[02:28] <Coyctecm> ogra: no, but what i mean is that new users could better "picture" of ubuntu that way
[02:28] <pitti> they should just port it to imlib2
[02:29] <Riddell> pitti: actually it might not be
[02:29] <Coyctecm> ogra: but well, gtk 1.2 indeed is waste of time nowadays :)
[02:29] <pitti> Riddell: a central component, or using imlib1?
[02:29] <Riddell> I think imlib is only used by some mostly obsolete part of kdegraphics
[02:29] <Surak> is nmapfe still using gtk1.2 on ubuntu? just tested on fedora and the answer is yes...
[02:29] <Riddell> let me see what happens if I compile it without
[02:29] <ogra> Coyctecm, we dont want to make it disappear, we just want to demote it to universe 
[02:30] <pitti> go, Riddell, go!
[02:30] <jdub> Riddell: yay!
[02:30] <Coyctecm> ok :)
[02:30] <pitti> getting rid of imlib would be worthwhile on its own
[02:30] <ivoks> hi
[02:30] <ivoks> pitti: i can confirm that share my printers work (with that Allow @LOCAL in cupsd.conf)
[02:31] <pitti> ivoks: with Browsing off?
[02:31] <ivoks> i didn't test that... 
[02:31] <ivoks> just a sec
[02:32] <ivoks> works with browsing off too :)
[02:32] <ivoks> nice
[02:32] <ivoks> so, *sharing* scripts need rewrite
[02:32] <pitti> ivoks: that's contrary to upstream's documentation and the comment ...
[02:33] <pitti> but it makes much more sense
[02:33] <ivoks> yeah, i tought browsing must be on too :/
[02:33] <Surak> Guys, I think my contributions to Ubuntu are enough for become a Ubuntu member. However, I have no testimonial from actual Ubuntu members yet. Can someone which saw some of my work help me on that? My wikipage is https://wiki.ubuntu.com/AlexandreOttoStrube  and I am registered in launchpad as https://launchpad.net/people/surak
[02:33] <ogra> pitti, CUPS upstrem documentation ????
[02:33] <ogra> HAHAHAHA
[02:33] <pitti> ivoks: hm, there's a bug about it
[02:33] <pitti> ogra: ?
[02:34] <ogra> pitti, you never read ESR's rant about cups and theor docs ?
[02:34] <ogra> *their
[02:34] <pitti> ogra: no, I didn't
[02:34] <ivoks> ok, at least we now know how it really works :)
[02:34] <pitti> ogra: I know the docs a bit, though
[02:35] <ivoks> pitti: one sec
[02:35] <ivoks> this is how it works:
[02:35] <ivoks> if browsing is off
[02:35] <ogra> pitti, http://www.catb.org/esr/writings/cups-horror.html
[02:35] <ivoks> clients can print on it (but can't see it on the list)
[02:35] <ivoks> if browsing is on
[02:35] <ivoks> clients can see and print on it
[02:36] <pitti> ivoks: yes, sure
[02:36] <ivoks> so, i guess we should enable browse for sharing
[02:37] <pitti> ivoks: I meant, the documentation suggests that you need to enable Browsing for exporting your printers over the browse protocoll
[02:37] <pitti> ivoks: well, I'll try that out myself when I can get my attention back to cups
[02:37] <ivoks> i'm here with 4 cupses and few printers :)
[02:37] <pitti> ivoks: in the meantime, thanks a lot for your investigations and patches :)
[02:38] <pitti> ivoks: if you find out anything, please note it in the bug reports, so that it doesn't get lost
[02:38] <ivoks> pitti: do you prefere any special bug (allready opened) or want me to open a new one?
[02:38] <pitti> ivoks: bug 41403 for example
[02:38] <Ubugtu> Malone bug 41403 in cupsys "Network printers are visible even though Browsing is off" [Normal,Unconfirmed]  http://launchpad.net/bugs/41403
[02:38] <ivoks> ok :)
[02:39] <pitti> or whatever
[02:39] <pitti> I didn't look at this one at all so far
[02:39] <pitti> I really need to catch up with security before continuing printer stuff, sorry
[02:39] <ivoks> np
[02:39] <ivoks> i'll try to help as much as i can
[02:39] <ivoks> stop guys :)
[02:40] <ivoks> first dholbach, then pitti...
[02:41] <ivoks> those dev guys... :)
[02:41] <ogra> :)
[02:44] <_ion> Hm. Looks like persian text is rendered uglier than before (with the Sans font alias).
[02:44] <_ion> "Before" being something like 12 weeks ago.
[02:53] <sivang> hmm, who can check if a apcakge went out or needs a kick out of NEW ?
[02:57] <tseng> sivang: if it left NEW you got a mail that says ACCEPTED
[02:57] <tseng> sivang: if you didnt, it didnt.
[02:58] <sivang> tseng: okay, so I got the accepted email, I'll wait then until it's went out of NEW.
[02:58] <tseng> sivang: if its accepted it is out
[02:59] <sivang> tseng: but I can't install it or find it in the archive, can you? (pkg name: upbackup)
[02:59] <tseng> sivang:  i didnt say it built or hit the archive
[03:00] <tseng> https://launchpad.net/distros/ubuntu/+source/upbackup
[03:00] <tseng> https://launchpad.net/+builds/+build/187860
[03:01] <tseng> the source package is in the mirrors
[03:02] <StevenK> The binaries are in NEW
[03:02] <StevenK> I suspect.
[03:02] <tseng> there is only one binary, and presumably the source passed new
[03:02] <pitti> Kamion: ppc ubiquity success
[03:02] <tseng> but i guess i agree with StevenK 
[03:02] <StevenK> The binary .deb's are in NEW
[03:02] <StevenK> (I guess)
[03:03] <StevenK> Kamion or infinity could tell you for certain.
[03:04] <nomed> dholbach, around ?
[03:04] <fabbione> how has a ppc here around to run one command for me?
[03:05] <dholbach> nomed: about to go outside
[03:05] <sivang> StevenK: okay, thank you gues.
[03:05] <sivang> guys
[03:05] <nomed> i see there is a patch within icon-naming-utils
[03:05] <ogra> fabbione, ?
[03:05] <fabbione> ogra: what ppc do you have?
[03:05] <ogra> ibook G4
[03:05] <dholbach> nomed: what about it?
[03:05] <zul> heylo
[03:05] <nomed> do u think upstreamer could accept it ?
[03:05] <fabbione> ogra: ati or nvidia card?
[03:05] <ogra> ati
[03:05] <ogra> (100% bugfree btw)
[03:05] <nomed> dholbach, i'll need to send a patch to him to add xfce icons
[03:06] <dholbach> nomed: I talked to them about it, they thought it wasn't necessary
[03:06] <fabbione> ogra: can you please logout from X and tell me the output from xresprobe ati ?
[03:06] <nomed> ok
[03:06] <dholbach> nomed: there should be a bug report about it
[03:06] <ogra> fabbione, it works from within X as well ...
[03:06] <Amaranth> fabbione: I can too, if needed.
[03:06] <dholbach> nomed: although it might have been a irc discussion as well
[03:06] <fabbione> Amaranth: yes please
[03:06] <dholbach> nomed: you can ask about it, if you like
[03:06] <Amaranth> G4 Mac mini, Radeon 9250
[03:06] <fabbione> ogra: please do it from outside
[03:06] <Amaranth> ok, brb
[03:06] <fabbione> Amaranth: yes that's fine
[03:06] <nomed> dholbach, ok
[03:07] <dholbach> nomed: thanks
[03:07] <ogra> fabbione, then it will take a minute, i khave several open unfinished mails 
[03:07] <fabbione> ogra: ok thanks
[03:11] <sivang> tseng: is there a reason why the binaries wouldn't get out of NEW ?
[03:11] <sivang> tseng: do thay need speical manual approval ?
[03:11] <StevenK> sivang: They do
[03:11] <sivang> StevenK: ah, I see
[03:11] <StevenK> (Just like source)
[03:12] <sivang> StevenK: okay then, I just need to wait for it to come out of NEW again I guess, funny thing I already have a fix for the previous package....I hope it wouldn't take longer after the package has already went out one time from NEW for both bin and src.
[03:12] <tseng> StevenK: ive never seen a source with a single binary go through new (source) and not new (bin)
[03:13] <StevenK> tseng: Tell sivang, not me. :-)
[03:13] <Amaranth> fabbione: http://rafb.net/paste/results/SMP54Q62.html
[03:13] <fabbione> Amaranth: thanks
[03:13] <Amaranth> funny, i'm running at 1152x864
[03:13] <sivang> StevenK: should I ask in #launchpad maybe? ;-)
[03:13] <tseng> why there
[03:14] <sivang> tseng: possible buildd bug ? I don't know...
[03:14] <sivang> tseng: or publishing one, rather.
[03:14] <StevenK> They will be able you if unseen binary packages go into NEW, but not what's in there, I suspect.
[03:14] <StevenK> Er, s/unseen/unknown/
[03:17] <fabbione> woo
[03:17] <fabbione> i was this >< close to add a regression in X
[03:18] <thom> don't do that then
[03:18] <fabbione> i am not going to :)
[03:19] <Amaranth> fabbione: whew :)
[03:20] <ogra> Mithrandir, id the liveCD supposed to have working hibernate ? 
[03:20] <Mithrandir> ogra: no, and it won't for dapper.
[03:20] <ogra> ok
[03:20] <Mithrandir> ogra: it's on my list of stuff to add for eft, though
[03:21] <ogra> Mithrandir, fine then i just have a user in #edubuntu trying it and wondering why it doent work :)
[03:21] <ogra> i'll tell him to stop testing suspend/hibernate ;)
[03:21] <Mithrandir> ogra: it should work if you pass resume=$partition, though.
[03:21] <Mithrandir> but I haven't tried it and I suspect it won't work, but it's worth a try.
[03:23] <ogra> Mithrandir, he's asking if he should file a bug (he cant come over here, chatting through a restricted cgiirc scrip t from work)
[03:23] <Mithrandir> ogra: https://launchpad.net/distros/ubuntu/+source/casper/+bug/23882
[03:23] <Ubugtu> Malone bug 23882 in casper "Hibernate option should be suppressed on the live CD" [Wishlist,Unconfirmed]  
[03:24] <ogra> thanks
[03:24] <Mithrandir> so no need to file a bug.
[03:24] <ogra> yep
[03:28] <janimo> dholbach: hi
[03:28] <janimo> talked about the gnumeric build hack with Gauvain
[03:28] <janimo> it is the cleanest solution we could come up with
[03:32] <Kamion> tseng: source and binary NEW happen at minimum an hour apart, so it's pretty common
[03:32] <Kamion> sivang: relax, it's in NEW, I'll deal with it in a bit
[04:07] <dholbach> janimo: cleanest solution involves building libgoffice-gtk-something-dev?
[04:08] <janimo> dholbach: that will make the two libraries not installable in the same buildd
[04:08] <janimo> which is our goal if we want to build from the same source package
[04:08] <dholbach> arg arg arg arg arg arg
[04:08] <janimo> this is why I specifically talked to gnumeric upstream
[04:09] <janimo> and they guarrantee that -gtk and -gnome libs are going to be
[04:09] <janimo> API/ABI compatible at any version
[04:09] <dholbach> yeah
[04:09] <janimo> so the sed hack takes advantage of that
[04:09] <janimo> buold agains the -dev, because we _know_ it will be ok at runtime
[04:09] <dholbach> i'd like to hear some views on that on the mailing list
[04:09] <dholbach> is that ok for you?
[04:09] <dholbach> i don't like either idea, but I want to hear some concerns before we do it
[04:10] <janimo> I am thinking about this since december :) and talking to debian/gnumeric upstream just to make sure :)
[04:10] <dholbach> (if there are any)
[04:10] <dholbach> yeah
[04:10] <dholbach> i definitely see your need for it
[04:10] <janimo> dholbach: sure np with the getting other idea
[04:10] <dholbach> that's out of question
[04:10] <dholbach> Thank you.
[04:10] <janimo> thank _you_ for taking care of this :)
[04:10] <dholbach> I don't want to be a pain in the ass, I'd just like to hear more views from people being more apt than I am.
[04:11] <janimo> after how much I talked about this with gnumeric/debian upstream you'll have to try hard before becoming a pita :)
[04:11] <janimo> relative to them :)
[04:11] <janimo> indeed I am curious if there's some cleaner trickery to do this thing
[04:12] <dholbach> I'm sure we'll get relevant opinions on ubuntu-devel.
[04:12] <dholbach> ... and get over with it.
[04:14] <janimo> ok, so are you going to post?
[04:14] <janimo> there's probably more madness on the same scale that will be for evince and gnome-system tools :)
[04:15] <janimo> that's the price for getting no duplicates for these in main
[04:15] <dholbach> janimo: no, you do that. :)
[04:16] <dholbach> janimo: I'm not good at explaining other people's ideas.
[04:16] <janimo> ok
[04:18] <janimo> hmm there's no patch to revisit
[04:19] <janimo> dholbach: was there a LP bug on this, or did Gauvain point you to a diff?
[04:19] <dholbach> janimo: not yet
[04:19] <janimo> dholbach: just talking about the idea?
[04:20] <dholbach> yep
[04:23] <dholbach> Kinnison: could you tell me about your gpm plans? would it be ok, if I'd patch some icons in? (patch adding new files, modifying rules,control) or do you want me to wait until you uploaded the new stuff?
[04:24] <janimo> dholbach: ah I remember now that you mention icons and patching
[04:25] <janimo> we have a situation where an app uses an icon name which is only in the gnome icon theme but not in tango
[04:25] <janimo> a screenshot panel plugin for xfce
[04:25] <janimo> what is the best way to make sure it gets an icon at runtime?
[04:25] <Kamion> Riddell: any test results for today's Kubuntu daily-live? (sorry if I missed them earlier)
[04:26] <janimo> patch the source to look for a fallback? install the hardcoded icon along with it?
[04:26] <Riddell> Kamion: yes, they're all good
[04:26] <Kamion> great
[04:26] <Kamion> topicdiff: beta2 candidates available for Xubuntu too
[04:27] <Kamion> Riddell: hope my KDE frontend changes looked reasonable
[04:27] <janimo> dholbach: is it polite to just reject bugs aginst breezy which are not dataloss/security?
[04:27] <Kamion> I think the manual partitioner should be a good bit better now
[04:27] <janimo> Kamion, thanks. just discovered on today's live that install or any other .desktop file is not runnable from the desktop
[04:27] <dholbach> janimo: I think that's ok. If people complain say that we don't have the manpower or the changes are too intrusive
[04:27] <pitti> janimo: if they are fixed in dapper, they should be 'Fix released' instead
[04:27] <Kamion> janimo: oops. is a fix on its way?
[04:27] <ogra> Kamion, apart from the reported bugs edubuntu should be good to go as well (didnt test the installer on ppc yet)
[04:28] <janimo> so it would not be too good of a beta, unless we tell users to launch it from terminal
[04:28] <janimo> Kamion, just discovered it so not yet I ;m afraid
[04:28] <dholbach> janimo: as for the icon question, you might want to install an icon to hicolor (make sure you don't overlap with others)
[04:28] <janimo> if it's ok I can delay till tomorrow?
[04:28] <pitti> Kamion: does that mean you just released new images? or does that still refer to tonights/this morning's ones?
[04:28] <Kamion> janimo: yeah, that's ok
[04:28] <Kamion> pitti: still this morning's images
[04:29] <janimo> pitti, ack, although this one looks like needinfo or invalid even for breezy. I'll reject it with a polite wording then
[04:30] <Lure> Kamion: new Kubuntu live available? I can do another test run...
[04:30] <pitti> janimo: sure
[04:30] <Riddell> Lure: yeah, please do
[04:30] <Riddell> Lure: rsync against yesterdays
[04:30] <Lure> Riddell: will do
[04:32] <Kamion> Lure: new from this morning
[04:32] <Kamion> as in about ten hours ago
[04:33] <Lure> Kamion: got it - need 1-1.5 hours to rsync...
[04:33] <Kamion> stuff's looking OK though - the cancellation bugs reported on Edubuntu (which will apply to Ubuntu and Xubuntu as well) are not good, but I don't think they're a regression from breezy either
[04:33] <Kamion> testing i386 and amd64 Ubuntu here
[04:33] <ogra> they shouldnt block beta2
[04:33] <Kamion> yeah, I agree
[04:34] <pitti> Kamion: not sure whether you saw it, ppc was fine here (reiserfs install OOTB)
[04:34] <Kamion> er, when I said regression from breezy I obviously meant regression from beta
[04:34] <ogra> :)
[04:34] <Kamion> pitti: great, thanks - I think I saw it
[04:34] <pitti> my amd64 is compiling mozilla like mad, and that'll still take a while
[04:34] <pitti> so I can't test amd64 right now
[04:34] <janimo> dholbach, so putting the icon into hicolor is ok then? no icon them should put stuff there. I don't know how to avoid a name clash besides changing the icon name but that is the same as patching the sources
[04:35] <dholbach> icon theme lookup is safer than hardcoding a path
[04:36] <sivang> Kamion: k, thanks, sorry for the speculations, I just wanted to make sure everything's right with it.
[04:38] <sivang> Kamion: after the first time it's out of NEW (both bin and src) it will be no hussle to have subsequent uploads without you needing to authorize them ?
[04:38] <Kamion> sivang: correct
[04:38] <Kamion> if we needed to approve every upload by hand we'd never get anything done
[04:38] <sivang> Kamion: okay, sorry for the stupid question :)
[04:39] <ogra> Kamion, we could hire a monkey for NEW then 
[04:39] <mvo> Kamion: if you have a moment I would appreciate your opinion on #41297
[04:40] <ogra> (they work for bananas and you can easily teach them to hit the big NEW button :) )
[04:40] <Kamion> mvo: I've got it open in my browser already but haven't quite dealt with it yet :)
[04:41] <mvo> Kamion: ok :) 
[04:41] <infinity> ogra: As it turns out, processing NEW actually requires real people to check packages and make sure they're A) not a complete mess, and B) legally distributable.
[04:41] <infinity> ogra: Most monkeys aren't terribly good at determining either.
[04:42] <ogra> infinity, damned, youre right 
[04:42] <ogra> :)
[04:42] <Kamion> mvo: you can always set DEBIAN_FRONTEND=dialog in the environment if you like
[04:42] <ogra> infinity, how's the health situation over there ? 
[04:43] <infinity> ogra: Pretty crap, but I'm working out of guilt at this point.
[04:43] <infinity> Quite the motivator.
[04:43] <ogra> working out of guilt and being ill even longer is no good plan ...
[04:45] <ogra> brb
[04:45] <mvo> Kamion: sure, but if I do this, I would like to make sure that the internal terminal is expanded as soon as a debconf question is asked. so I need a way to detect it. I was wondering if I could do some magic for this
[04:46] <Kamion> oh, hmm
[04:46] <Kamion> watch for the ANSI codes that indicate dialog switching screen mode?
[04:46] <mvo> thats a pretty nifty idea :)
[04:47] <Kamion> would be cleaner to implement some kind of passthrough frontend
[04:47] <Kamion> but then you'd need a second terminal to put debconf questions in
[04:47] <ogra> fabbione, id: COLOR LCD
[04:48] <ogra> res: 1024x768
[04:48] <ogra> freq:
[04:48] <Kamion> see /usr/share/doc/debconf-doc/passthrough.txt.gz if that approach sounds interesting - though it would probably be a fair bit of work
[04:48] <ogra> disptype: lcd/lvds
[04:48] <Kamion> using debconf again to display the dialogs would be complicated because you'd have to make it use separate databases to avoid locking pain
[04:48] <mvo> Kamion: thanks, a great deal of work sounds bad. anything in espresso I could reuse? (I kow it has a new name, but I can't pronounce it)
[04:48] <Kamion> and you'd be running into the less well-tested bits of debconf
[04:49] <Kamion> mvo: not usefully, no
[04:49] <mvo> *meh*
[04:49] <Kamion> I suspect the horrible hack of watching for specific escape sequences on the terminal is actually easier
[04:49] <Kamion> if you can do that
[04:49] <mvo> a lot easier
[04:49] <mvo> and its less horrible than the pre-depend that is mentioned in the bugreport ;)
[04:50] <Kamion> hmm, German for ubiquity is Allgegenwart apparently - think I prefer the English version ;)
[04:51] <fabbione> ogra: ok thanks
[04:51] <ogra> Kamion, but i know from the top of my hat how to write Allgegenwart :)
[04:51] <ogra> er
[04:51] <ogra> and from my head as well
[04:52] <ogra> pronouncing ubiquity is easy after trying it once, but keeping in ming where all these q's and u's have to be placed isnt easy :)
[04:52] <ogra> *mind
[04:52] <dholbach> mvo: Allgegenwart sounds a bit too sacral for my taste :)
[04:53] <ogra> yeah
[04:54] <mvo> allgegenwart/from today looks good btw, it seems to fix at least the keyboard issue I had earlier
[04:55] <Riddell> pitti: well, we'll see how many complains we get :)
[04:55] <infinity> mvo: Can you record yourself pronouncing "ubiquity" (or failing to do so) for me? :)
[04:56] <pitti> Riddell: what breaks without it?
[04:56] <Riddell> pitti: it means kuickshow no longer gets compiled, but it's obsolete as far as I'm concerned
[04:58] <bddebian> Morning peoples
[05:03] <mdz> fabbione: I was asking because both my laptop and my desktop had both x11-common and xorg-common installed
[05:03] <mdz> fabbione: ah, perhaps only the init script was there, and the init script didn't check whether the package was installed
[05:03] <mdz> fabbione: this is starting to sound familiar, perhaps we discussed it before
[05:03] <fabbione> mdz: i did a test here.. fresh breezy -> dapper and xorg-common is removed, but not purged
[05:03] <mdz> fabbione: and its init script still runs
[05:03] <infinity> And init scripts are conffiles, hence the percieved problem.
[05:04] <fabbione> meeeehhh
[05:04] <mdz> mvo's upgrade tool should work around this by purging it
[05:04] <fabbione> mdz: yes that does work...
[05:04] <infinity> mvo's upgrade tool should really have a more generic "list everything I just removed, and now that the upgrade is done, purge it" thing or something, I suspect.
[05:05] <fabbione> mdz: but more than conflicts, i am not sure what i can do
[05:05] <fabbione> infinity: it does have it
[05:05] <infinity> apt-get --purge dist-upgrade can be dangeours in rare corner cases, where you don't want the purge to happen before the new install, but it should be safe to purge as a last step.
[05:05] <Kamion> infinity: would need to be strictly optional
[05:05] <mdz> fabbione: well, 1) have the upgrade tool purge it (should be done), or 2) upload to breezy-updates fixing the init script
[05:06] <infinity> Kamion: Probably.  Though we get more bugs from people not purging packages then from people breaking their system by purging stuff. :)
[05:06] <Kamion> purge-by-default is pretty scary still
[05:06] <jdub> pitti: only a few days to go :-)
[05:06] <fabbione> mdz: the init script is not dangerous is it?
[05:06] <pitti> jdub: mozilla/warty is building...
[05:06] <infinity> Just verbose.  it doesn't actually DO anything particularly scary, AFAIK...
[05:06] <mdz> fabbione: no, but it prints *two* spurious messages which clutter the much prettier boot in dapper
[05:06] <fabbione> mdz: hmm ok
[05:07] <mdz> so 1) is probably the best we will do
[05:07] <ogra> cant we keep a transitional empty package of it to remove the initscript  ?
[05:07] <infinity> mdz: Well, fixing the upgrade tool fixes it for people who upgrade "the right way", and I suspect others can just keep the slightly uglier boot, until we give them a "find unpurged packages and purge them" tool.
[05:07] <fabbione> mdz: i can do a workaround on it in x11-common to check the md5 of that script and disable it if it is the same as in breezy
[05:07] <pitti> infinity: after I finished mozilla, I'll attack tbird for stables; however, do you think I can grab you for doing the corresponding enigmail update?
[05:07] <Keybuk> ogra: empty package would need a postinst that removed the init script
[05:07] <pitti> infinity: (it won't happen today, proably next Tuesday)
[05:07] <infinity> pitti: Sure thing.  I can do tbird itself too, if you're too busy.
[05:07] <mdz> fabbione: I don't think it's worth the trouble
[05:07] <ogra> Keybuk, yep, but if we dont need it at all, thats fine, no ?
[05:08] <mdz> fabbione: better to fix the real bugs in X
[05:08] <fabbione> mdz: also.. i couldn't reproduce the X issue you mentioned...
[05:08] <pitti> infinity: I guess after your sick days you are horribly overloaded, too
[05:08] <mdz> fabbione: X issue I mentioned?
[05:08] <Kamion> infinity: (find unpurged packages> we call it dselect :P)
[05:08] <infinity> pitti: Well, this is true. :)
[05:08] <pitti> infinity: and I'd rather have you for php
[05:08] <fabbione> mdz: the one about the complex md5sum calculation that was bailing on you
[05:08] <fabbione> mdz: with cat $something missing
[05:08] <infinity> Kamion: I've heard good things about this dselect thing, but people keep telling me it's obsolete, and I should use package tools that hide useful info from me at every turn. :)
[05:09] <fabbione> mdz: i *think* that people that are experiencing it have been going trough a broken x11-common 
[05:09] <infinity> fabbione: The one about /var/whatever/X11 versus /var/whatever/xfree or something, and crazy md5 madness?
[05:09] <mvo> infinity:  (find unpurged packages> we call it synaptic :P)
[05:09] <fabbione> mdz: because breezy -> dapper is ok
[05:09] <mdz> fabbione: mizar:[~]  ls /var/lib/x11
[05:09] <mdz> X.roster                Xwrapper.config.roster  xorg.conf.roster
[05:09] <mdz> Xwrapper.config.md5sum  xorg.conf.md5sum
[05:09] <fabbione> infinity: yeah exactly
[05:09] <Keybuk> aptitude purge "~c"
[05:09] <Keybuk> easy
[05:09] <infinity> fabbione: I think I may have a handle on that.
[05:09] <mdz> fabbione: note there is no X.md5sum
[05:09] <fabbione> mdz: yes that is right.. but i can't get to that situation
[05:10] <fabbione> mdz: so if i don't understand why .md5sum was missing
[05:10] <fabbione> i have no way to fix it
[05:10] <mdz> fabbione: does it matter?  if it is missing, it should be created based on the existing symlink
[05:10] <fabbione> infinity: did you reproduce it?
[05:11] <fabbione> mdz: i will need to recheck that part of the code.. but yeah it's doable that way assuming there is only one xserver installed
[05:11] <fabbione> mdz: otherwise you are doomed :)
[05:11] <mdz> fabbione: but it doesn't seem to fail the maintainer script, so unless you know otherwies it is cosmetic
[05:11] <infinity> fabbione: No, I only noticed it in passing, but I have some fair ideas of how it could happen, and how to make it happy.
[05:11] <fabbione> mdz: i did read that code quickly.. i was hoping to reproduce the problem and go down to a set -x
[05:11] <mdz> fabbione: so if it is a corner case which doesn't affect upgrades from breezy, and is a cosmetic error, then the other bugs are more important
[05:12] <infinity> mdz: BTW, the "missing GL headers" bug (if you recall that one) is a bug in breezy, only triggered on upgrade to dapper.
[05:12] <fabbione> mdz: i will check it again
[05:12] <infinity> mdz: Would it be appropriate to upload a new mesa to breezy-updates to fix that, and just pray that people will upgrade before the upgrade? :)
[05:13] <mdz> infinity: we already instruct them to do so
[05:13] <mdz> if there is no way to work around it in dapper, then yes
[05:13] <infinity> mdz: (unversioned "Replaces", leading to a forward-replace during the upgrade, and the world explodes, cause those files just moved too many effin' times)
[05:13] <infinity> So, version the replaces in breezy, and we're gold.
[05:14] <mdz> sounds safe enough
[05:14] <infinity> I'll pop up a fix later on.
[05:15] <bddebian> *cough* ivtools *cough*
[05:18] <fabbione> mdz: ok you got some ati love today...
[05:18] <fabbione> mdz: i will look at the X.md5sum later today
[05:18] <fabbione> i need a break now
[05:18] <mdz> fabbione: forget about X.md5sum as above; there are 500 bugs on ubuntu-x-swat which are worse
[05:19] <mdz> bugs which make X unusable for users
[05:19] <fabbione> mdz: do you have a customized xorg.conf?
[05:19] <mdz> mizar:[~]  cat /var/lib/x11/xorg.conf.md5sum
[05:19] <mdz> 9102f2a981a8a0af1e6e33840804388d  /etc/X11/xorg.conf
[05:19] <mdz> mizar:[~]  md5sum /etc/X11/xorg.conf
[05:19] <mdz> 9102f2a981a8a0af1e6e33840804388d  /etc/X11/xorg.conf
[05:19] <fabbione> mdz: not anymore.. down to 490 or so
[05:20] <fabbione> mdz: than there is a regression there too. The error message was claiming a customized config when it's not
[05:20] <fabbione> mdz: that means something is not happening there too
[05:21] <fabbione> don't worry
[05:21] <fabbione> we will get as much as possible sorted
[05:21] <ogra> +++
[05:21] <ogra> ++
[05:21] <ogra> \o/
[05:22] <mdz> what will happen is that we will face difficult decisions about making disruptive changes to fix those bugs, because we're already post-beta
[05:23] <fabbione> mdz: 
[05:23] <fabbione> the major issue is one
[05:23] <fabbione> you don't want to ask for at least the resolution on livecd
[05:23] <fabbione> without that question and no decent info from the system, we fail
[05:23] <fabbione> 50%/50%
[05:23] <fabbione> and there are other corner cases like this one where user A want foo and B != foo
[05:23] <ogra> nope, we fall back to the least sucky resolution
[05:24] <fabbione> ogra: no, we fail.
[05:24] <ogra> its not completely failing
[05:24] <mdz> for me, 100% of the systems which asked the question work fine with the default answer to the question
[05:24] <mdz> every single one
[05:24] <pitti> I get a sucky resolution, and I know two people who suffer the same
[05:24] <pitti> and there's no apparent method to fix it
[05:24] <fabbione> ogra: sucky resolution = fail for me
[05:24] <ogra> for me none of my laptops work right apart from the ibook
[05:24] <pitti> 1024 just looks awkward on a 1280 TFT
[05:25] <mdz> pitti: what's the cause for autodetection failing in your case?
[05:25] <ogra> fabbione, fail == no X for me
[05:25] <pitti> mdz: it doesn't work through DVI, works if I use the VGA connector
[05:25] <ogra> fabbione, my 1280x800 display is still usable at 1024x786, it just looks odd
[05:25] <pitti> mdz: no idea about the cause at my friend's, I couldn't debug it so far
[05:25] <fabbione> ogra: that is a failure
[05:25] <fabbione> ogra: users complain about stuff like that
[05:25] <fabbione> a lot
[05:26] <pitti> the point is, defaulting to 1024 is fine as long as the user can switch to the right resolution in the gnome tool
[05:26] <ogra> fabbione, yes, but not easily solveable 
[05:26] <pitti> using the resolution chosen in gfxboot would also be nice
[05:26] <ogra> yeah
[05:26] <fabbione> that cannot be done easily either.. pitti
[05:26] <pitti> but we discussed that already, it's not completely adequate
[05:26] <Kamion> gfxboot gives you VESA modes, I'm not at all convinced it's appropriate
[05:26] <pitti> fabbione: right, I know
[05:26] <ogra> pitti, thats my suggestion since months
[05:27] <ogra> Kamion, there are some widescreen modes ...
[05:27] <mdz> fabbione: what makes it difficult?
[05:27] <fabbione> Kamion: sometimes it is.. others is not
[05:27] <pitti> I'd still prefer setting up more resolutions in xorg.conf, default to 1024 and let the user choose in System -> Settings -> Resolution
[05:27] <Kamion> ogra: then your VESA BIOS is announcing them
[05:27] <pitti> but fabbione told me that this gives regressions, too
[05:27] <mdz> pitti: yes, I was thinking the same thing
[05:27] <ogra> fabbione, when Kamion and i talked the last time about it we came to the conclusion to have a monitor db as fallback, like xfree 3.x did
[05:28] <ogra> fabbione, Mithrandir also said we should test edid v2
[05:28] <fabbione> mdz: when you have situation in which: Card: foo with monitor bar works best without HorizSync/VertRefresh, and Card: foo with monitor baz wants HorizSync...
[05:28] <Kamion> ogra: don't assume I actually have any authoritative knowledge about X though
[05:28] <ogra> so combining both might gain us the results
[05:28] <ogra> Kamion, nor have i, but i've seen dexconf (still washing my eyes)
[05:28] <fabbione> ogra: the monitor db is built using output from ddcprobe
[05:29] <mdz> fabbione: what does "works best" mean there?
[05:29] <fabbione> ogra: edid fails too on some hw.... no matter what version
[05:29] <mdz> so long as the HorizSync/VertRefresh is big enough, why should it matter?
[05:29] <fabbione> mdz: because you cannot set arbitrary HorizSync/VertRefresh
[05:30] <ogra> fabbione, yep, i know i have 2 laptops here that just report edidfail 
[05:30] <fabbione> mdz: it does matter in 2 cases: when the driver can't probe HorizSync/VertRefresh from the monitor
[05:30] <mdz> fabbione: if it works fine without HorizSync/VertRefresh, that means that X is able to autodetect the correct values, yes?
[05:30] <fabbione> mdz: and when the monitor returns wrong value
[05:30] <fabbione> mdz: it depends from the monitor...
[05:30] <fabbione> mdz: sometimes the monitors return crap
[05:30] <fabbione> it's really complex
[05:30] <mdz> fabbione: then presumably it would not work fine
[05:30] <ogra> many cheap widescreens do that
[05:31] <fabbione> and it doesn't... that's why you override them manually.... with "work best with/without"
[05:31] <fabbione> so that you can tell the driver: hey driver, don't trust this crappy monitor.. do as i say
[05:31] <fabbione> that means we can't just set whatever large enough value
[05:32] <fabbione> or people will get bed refresh rates out of monitor syncs
[05:32] <mdz> but where do we get the values in that case?
[05:32] <fabbione> we calculate them statistically
[05:32] <fabbione> according to the probed resolution
[05:32] <fabbione> or the resolution that comes back from user input
[05:33] <fabbione> and the calculation is accurate enough given that people complains that we are too "safe"
[05:33] <mdz> I am saying that for the case where we are writing out sync ranges, we should write out sync ranges sufficient for a wider range of modes
[05:33] <mdz> and  include more modes in the config
[05:33] <mdz> so that the user can select them
[05:33] <fabbione> mdz: you can't
[05:33] <mdz> forget about the case where we are not writing the sync changes for a moment
[05:33] <mdz> s/changes/ranges/
[05:33] <mdz> fabbione: this seems very much like what Windows does, and it seems to work there
[05:33] <fabbione> mdz: we need to write syncs in cases where we know that the combination driver/monitor is broken
[05:34] <fabbione> to set a upper limit
[05:34] <fabbione> otherwise we go out of sync
[05:34] <pitti> hi zyga 
[05:34] <mdz> even at 1024x768?
[05:34] <fabbione> mdz: windows has better drivers
[05:34] <fabbione> yes even at 1024x768
[05:34] <zyga> hey guys :)
[05:34] <zyga> I finally work from 8-16 :)
[05:34] <fabbione> i can give you use case for 800x600
[05:35] <zyga> pitti: how are you?
[05:35] <fabbione> mdz: windows has a huge monitor db to match against.. we don't
[05:35] <fabbione> it's really impossible to compare the two
[05:35] <jdub> o/` we built this kitty on rock and roll! o/`
[05:35] <mdz> fabbione: we cannot reasonably expect to autodetect everything, but we should be able to provide a safe fallback *and* allow the user to choose after installation
[05:36] <pitti> zyga: quite fine, and you?
[05:36] <mdz> without a monitor db
[05:36] <ogra> mdz, that will need a gui then
[05:36] <ogra> or do you really want to bother users with dpkg-reconfigure ?
[05:36] <Lure> Kamion, Riddell: Kubuntu ubiquity results for today in bug 41683
[05:36] <mdz> ogra: we didn't have a gui for the mode question ever before
[05:36] <infinity> mdz: The problem with writing high sync ranges is that X calculates the refresh rate based on resolution and total bandwidth, so if your monitor can only do 1024x768@60, and the sync range is too high, it'll head up to 1024x768@75 on boot, and you get no display.
[05:37] <mdz> infinity: yes, I understand now
[05:37] <ogra> mdz, but it was asked automatically
[05:37] <infinity> mdz: So, the only "safe" way to write a sync range is to write one low enough that it forces you to use 1024x768@60.
[05:37] <mdz> we'd have to write different modelines
[05:37] <fabbione> EEEEEKKKKKKKK
[05:37] <zyga> pitti: good, new big monitor to play with :)
[05:37] <fabbione> mdz: we have been carrying around this problem since warty
[05:37] <mdz> actually...aren't there default modes named like that?
[05:37] <mdz> 1024x768@60?
[05:37] <fabbione> it didn't change since than
[05:38] <mdz> why can't we write out a larger list of modes and make low-refresh-rate modes the default?
[05:38] <fabbione> the problem only become more complex with the increasing amount of users
[05:38] <fabbione> mdz: that won't work at the first time you upgrade because modes need to be sorted
[05:38] <simira> what's claire's nick again?
[05:38] <fabbione> some users wants the best resolution at the first shot
[05:38] <fabbione> simira: cvd
[05:39] <infinity> fabbione: Those users can cope.  This is exactly how Windows does it.
[05:39] <simira> thanks fabbione 
[05:39] <mdz> infinity++
[05:39] <fabbione> infinity: ok, than we can close about 489 bugs in X
[05:39] <ogra> mdz, it might roast your monitor since you can select the possibly broken modes from the resolution selector in the desktop
[05:39] <fabbione> we force a default config for 640x480 and we are done
[05:39] <infinity> It starts out at 640x480, then at the end of the install, asks if you'd like it to attempt to make it "nicer", which leads to a whopping 800x600 (or maybe 1024x768 in XP, don't recall), then leaves it there until you mnaually change it to something better still.
[05:39] <mdz> ogra: that is the user's own fault
[05:40] <mdz> where we can protect them, we do
[05:40] <pitti> ogra: well, are there really monitors which immediately break then?
[05:40] <ogra> mdz, if we offer the modes there ?
[05:40] <mdz> ogra: yes
[05:40] <Keybuk> ogra: bah, any decent monitor (ie. ones capable of *doing* high modes/refresh rates) also have anti-roast stuff
[05:40] <mdz> Keybuk: agreed
[05:40] <Keybuk> it's more a "black screen" problem
[05:40] <infinity> ogra: I haven't seen a monitor explode on out-of-sync since 1985.  Seriously.
[05:40] <pitti> ogra: I used wrong resolutiosn starting from my ancient 15 years old CRT up to my shiny TFT without any damage
[05:40] <Keybuk> and they can solve that themselves
[05:40] <ogra> pitti, probably only pretty old ones, no idea how many would break nowadays
[05:40] <pitti> the old CRT made funny noises, but it coped well, at least for some seconds
[05:40] <mdz> ogra: I think it's a non-problem
[05:40] <infinity> ogra: I have an old SVGA 14 inch monitor from 1989 (don't ask) that just emits a high-pitched whine and SOUNDS like it'll blow up when out of sync, but has never actually done so.
[05:41] <ogra> ok
[05:41] <pitti> ogra: if they are that old, then the user will be grateful to have an excuse to buy a shiny new TFT
[05:41] <ogra> was just a thought :)
[05:41] <fabbione> ok
[05:41] <pitti> a good CRT will cope with that :)
[05:41] <fabbione> since it's so simple...
[05:41] <fabbione> you are welcome to fix it that way with your name on it
[05:41] <mdz> I'd be surprised if they couldn't fry their monitor in the same way by selecting the wrong mode in windows
[05:41] <mdz> fabbione: you are the one saying this is a big problem
[05:41] <infinity> fabbione: I played with the resolution munging in dexconf during the breezy release, so I don't mind having a poke at it this cycle too.
[05:42] <fabbione> mdz: you are asking me what is the biggest problem and i did picture one example.
[05:42] <ogra> mdz, the thing is that they wont have wrong modes in windows i guess
[05:42] <fabbione> mdz: the outcoming solution is to set everything to a sane default and let the user decide
[05:42] <infinity> The biggest win in doing it this way is a fully-populated RandR list in GNOME, so people can pick whatever they want, not what we (poorly) detected they could do.
[05:42] <fabbione> if that was this simple we would have done it for warty
[05:42] <jcole> what's the ubuntu server meta package name? is there one?
[05:42] <mdz> fabbione: I didn't ask, but I appreciate your offering an example anyway :-)
 fabbione: what makes it difficult?
[05:43] <infinity> ogra: Windows lets you pick any mode that your video card claims to be able to do.  They gave up on "PnP monitors" long ago, despite being instrumental in writing the spec, because so many of them are broken.
[05:43] <fabbione> and i did explain what makes difficult to fix xorg
[05:43] <ogra> infinity, ah, k, i'm really not up to date on windows :)
[05:43] <mdz> fabbione: you said the major issue was that we don't ask the question anymore
[05:43] <infinity> jcole: There isn't one, since ubuntu-server doesn't install any "default" stuff.
[05:43] <mdz> fabbione: (which has been true for months)
[05:43] <infinity> jcole: It's just a base install, a package manager, and your imagination to make it what you want.
[05:44] <fabbione> mdz: yes because from that answer we can (in chain) calculate all the other stuff we need... including ranges
[05:44] <Kamion> Lure: OK, I think that's due to the known bug that sometimes manual partitioning doesn't get written into partman right the first time round; I suspect that if you said no to the partman confirmation dialog and then tried again, it'd work fine
[05:44] <seb128> doko: when you change a package like gnome-cups-manager could you comment what you change from the changelog (cf any GNOME package upload)?
[05:44] <mdz> fabbione: I am suggesting a way that we can accomplish the same thing, but without asking a question during BOOT
[05:44] <ogra> why dont we just add a debconf mode to usplash and ask the question again ?
[05:44] <fabbione> mdz: and yes.. i know it has been true for months.. like it has been true for a release that X sucks
[05:44] <Kamion> Lure: I've seen the bug a number of times myself, but I've not quite managed to track it down yet; it's fairly high on my list though
[05:44] <Kamion> however, this isn't a regression from beta 1, which is good
[05:45] <ogra> the shutdown stuff mdz did seems to work very well, i dont see why the same shouldnt work at boot as well
[05:45] <fabbione> mdz: you are asking to skip a question and set some defaults to allow users to change afterwards..
[05:45] <mdz> fabbione: there are hundreds of bugs in Malone right now from users where X completely fails to start up; I refuse to believe that this is a bigger problem
[05:45] <fabbione> mdz: while we do still set a default, we can't easily make it "changable" afterwards
[05:45] <infinity> ogra: Are you being serious?  Kill usplash, run debconf, re-start usplash?.. That seems less than ideal. :)
[05:46] <jcole> infinity: is it just a server kernel without pkgsel/install-pattern=~t^ubuntu-standard$|~t^ubuntu-desktop$ ?
[05:46] <mdz> ogra: and it was Keybuk who did those changes, btw
[05:46] <ogra> infinity, run dewbconf *in* usplash ... like the "press enter" message mdz added
[05:46] <ogra> oh
[05:46] <fabbione> mdz: most of them are related to broken configs... go figure...
[05:46] <Kamion> ogra: mdz's change had no actual extra UI
[05:46] <ogra> s/mdz/Keybuk then
[05:46] <infinity> ogra: That wasn't an addition at all, dude.
[05:46] <fabbione> or broken upstream drivers.. for which we can't do much about
[05:46] <Keybuk> ogra: usplash still gets killed by init, we just restart it again :)
[05:46] <ogra> ah, k
[05:47] <infinity> ogra: That's just taking advantage of the fact that usplash doesn't trap the keyboard, so if you press [enter] , it passes through the the underlying process.
[05:47] <mdz> casper had been trying to display its message through usplash for ages, but failing because it was dead
[05:47] <Kamion> I mean, I'm sure it would be *possible* to write a debconf frontend that draws on usplash, but it's a huge boatload of work for one question
[05:47] <Keybuk> Kamion: plain text | pipe usplash_write? :p
[05:48] <infinity> Yeah, actually, it wouldn't be too hard.
[05:48] <infinity> But that doesn't make it right.
[05:48] <Kamion> not sure the debconf text frontend is, uh, entirely optimal for this
[05:48] <ogra> you have to display a list, i think thats the only real prob here
[05:48] <infinity> The text frontend is hard to parse, but wouldn't be a difficult starting point.
[05:49] <Kamion> ogra: I await your patch
[05:49] <infinity> But you'd be limited to question types that involve simple answers.
[05:49] <mdz> debconf at boot is madness
[05:49] <Kamion> mdz: agreed
[05:49] <infinity> Doing lists would be seriously difficult.
[05:49] <ogra> mdz, but thats what we did until now :)
[05:49] <mdz> ogra: no, we did it in the installer
[05:50] <mdz> not during a standard system boot
[05:50] <Kamion> jcole: see /preseed/ubuntu-server.seed on ubuntu-server CD images
[05:50] <ogra> an easy fix would be having another menu in gfxboot and preseed the res. even if you had to boot a second time after noticing its wrong
[05:50] <Mez> hola all ;)
[05:50] <Kamion> Lure: (thanks for the retest though, it's appreciated)
[05:51] <ogra> the only prob is that dexconf doent accept *any* preseeding for resolutions
[05:51] <jcole> Kamion: right, thanks :)
[05:51] <Kamion> there's no more room on the screen in gfxboot for additional menus
[05:51] <Kamion> the horizontal space at the bottom is full, and already overflows in some languages
[05:51] <infinity> Kamion: Oh, speaking of the res selector in gfxboot, can we do away with that turning into vga= options in the installed system (and breaking suspend/resume for users who have no idea why)?
[05:51] <Kamion> wrapping that menu bar onto another line is seriously non-trivial theme hacking
[05:52] <infinity> Kamion: Or has that already happened when I wasn't looking?
[05:52] <ogra> Kamion, nothing seems trivial with the current problem ...
[05:52] <Kamion> ogra: then maybe saying "an easy fix" isn't the best idea
[05:52] <fabbione> cya all at the meeting later
[05:52] <ogra> i guess we'll need to find the least sucky solution
[05:52] <Kamion> infinity: are we certain that no systems anywhere require vga= to display a console any more?
[05:53] <Kamion> because that's basically why it's propagated
[05:53] <infinity> Kamion: Not certain, no, but we /are/ certain that it's breaking other things due to being too easily discoverable. :/
[05:54] <janimo> Kamion, ok found the xubuntu live bug, will upload the fix in a few minutes
[05:54] <infinity> In the old days, no one would boot the installer with "vga=foo" unless they knew what they were doing, now it's a simple navigation to go "ooo, more resolution is better!" and they've just broken suspend.
[05:54] <janimo> shared-mime-info package was missing as a thunar dep, and we were saved so far by the extra gnome stuff on the CD which brought it in :)
[05:55] <dholbach> fabbione: I got kudzu removed from the archive in Hoary or something :-p
[05:55] <ogra> dholbach, i dont think he was even remotely serious :)
[05:55] <dholbach> I noticed. :-p
[05:56] <Mez> woah, people have vbeen spamming my blog like mad
[05:57] <Kamion> infinity: ok, fixed in debian-installer-utils 1.22ubuntu8 (won't take effect until the next debian-installer build)
[05:57] <pitti> carlos: hm, I don't see a significant increase in the number of domains today
[05:58] <pitti> carlos: if that requires more time, can we just use yesterday's langpacks? they are tested and built
[05:58] <infinity> Kamion: Thanks, dude.
[05:58] <infinity> Kamion: I'm trying to figure out if I still have time to hack isolinux to fix the default boot res, but I may be running out of time there..
[05:58] <janimo> fabbione: is there a way of making the R300 cards which die on resume with DRI not use it by default (short of taking their pci ids out of the kernel)?
[05:58] <Lure> Kamion: I went now through manual partitioning again, no changes (partitions are there from previous run) selected what I want and got confirmation w/o any mention of partition - if I confirm, it should be fine, no?
[05:58] <carlos> pitti: you used yesterday's tarball, if you see yesterdays' one, you had exactly the same values...
[05:59] <carlos> I think we had a race condition again...
[05:59] <Kamion> infinity: it's still on my list too
[05:59] <Kamion> Lure: the confirmation screen should display the partitions to be formatted
[05:59] <pitti> carlos: hm, I already moved the cronjob back by half an hour...
[05:59] <infinity> s/feault/default/
[05:59] <carlos> pitti: let me see the time it finished...
[05:59] <pitti> carlos: shall I refetch the tarball?
[05:59] <Lure> Kamion: will try it then... ;-)
[05:59] <Kamion> Lure: if it doesn't, then either you forgot to check the reformat box on the mount point screen, or you need to say no and then press next on the mount point screen again
[06:00] <carlos> pitti: perhaps the new domains did it slower
[06:00] <Kamion> (to work around the aforementioned bug)
[06:00] <Lure> Kamion: reformat is shaded in KDE frontend
[06:00] <Kamion> disabled?
[06:00] <Kamion> that's supposed to happen when there's no filesystem there so it *has* to reformat
[06:01] <mdz> infinity: it seems to me that we should be able to add some modes to extramodes so that we can differentiate between the different refresh rates by name, and use those to order the modes in the config
[06:01] <Kamion> anyway, I do know that particular bit of the KDE frontend is still buggy - I fixed it up a fair bit last night, but I think it basically never worked right
[06:01] <carlos> pitti: today, the export has 762 translation domains, and it finished at 15:12 DC time
[06:02] <infinity> mdz: I suspect you may be right.  If you need/want someone to bounce some of this off of, care to attack me post-meeting?  (it's 2am right now, and I'm pondering another nap before the meeting)
[06:02] <Lure> Kamion: actually they were not disbaled this time - I have seen this now... BTW, copying is in progres and fdisk -l still looks good
[06:02] <carlos> pitti: http://mawson.ubuntu.com/~carlos/rosetta-dapper-2006-04-27.log
[06:02] <pitti> carlos: ok, I fetched it at 1500
[06:02] <mdz> infinity: right
[06:02] <pitti> carlos: any possibility to run your jobs earlier?
[06:04] <carlos> pitti: no, that's not possible until we move to use production database directly
[06:04] <pitti> carlos: ok
[06:04] <carlos> the mirror we use finish 30 minutes before my script runs
[06:04] <pitti> carlos: I move my cronjob back another 30 mins
[06:04] <carlos> pitti: anyway the move to production should happen between this week and next week
[06:05] <carlos> and we would agree a better time
[06:05] <pitti> carlos: ok, I re-run the process manually now
[06:05] <carlos> pitti: would you do a manual run now?
[06:05] <carlos> ok, thanks
[06:08] <mdke> mako, it was about wiki:Listiquette, I wondered if you would take a look and see what you think about adding it to ML subscription emails, or posting now and again with reminders about it. If you agree, maybe we could transfer it to the website rather than the wiki
[06:11] <infinity> mdz: I have half a mind (I'll find the other half when I stop being sick, I suppose) to do a total rewrite of dexconf for sid/etch, so we can merge it nearly for free when we open edgy...
[06:12] <infinity> mdz: Anyhow, off to nap, poke me post-meeting if you want to discuss this stuff further.
[06:12] <mdz> infinity: night
[06:20] <bddebian> Ack, meeting..  What meeting?
[06:21] <Keybuk> bddebian: weekly canonical distro team meeting
[06:21] <bddebian> Ah, whew, I keep missing CC and MOTU meetings :-(
[06:36] <fabbione> janimo: not that i know off.
[06:40] <Diziet> Dammit, I primed this ccache yesterday.  Why is it taking all afternoon ?
[06:46] <mdz> pitti: the eject button stuff seems to work fine with my external DVD drive, but not the internal one on my laptop
[06:46] <mdz> pitti: g-v-m receives the notification, but doesn't unmount or eject
[06:47] <pitti> mdz: even with ubuntu6?
[06:47] <mdz> pitti: hmm, is ubuntu6 not in the archive yet?
[06:47] <mdz> I have ubuntu5 but i am up to date
[06:47] <pitti> mdz: I uploaded it maybe 1 hour ago, I doubt it
[06:47] <pitti> mdz: right, ubuntu6 fixed it
[06:47] <Kamion> mdz: beta2 is pre-published for mirrors, I'll do the release and announcement and stuff later tonight
[06:47] <mdz> pitti: ah, ok. I thought the upload was earlier
[06:47] <mdz> Kamion: excellent
[06:48] <ogra> Kamion, do you include derivatives in it ? 
[06:48] <Kamion> ogra: Kubuntu and Edubuntu, yes; Xubuntu isn't ready
[06:48] <ogra> thanks 
[06:48] <ogra> (i'm not after writing another announcement, every two weeks is enough) :)
[06:50] <janimo> Kamion: xubuntu is readier now, after thunar builds the icon launcher thing should be solved
[06:50] <janimo> I uploaded about 30 minutes ago
[06:51] <janimo> but I'll wait for tomorrow to be sure
[06:52] <pitti> carlos: wow, take a look at the new report; that's impressive
[06:54] <pitti> janimo, carlos: it seems that many xfce translations are missing in Rosetta; is that due to not having rebuilt the packages since the main promition, or due to not yet having them manually imported into Rosetta?
[06:54] <janimo> pitti, they are not in rosetta
[06:54] <carlos> pitti: :-P
[06:54] <carlos> pitti: some of them are imported already
[06:54] <janimo> I was under the impression they are not suppoed to be there if upstream does not want
[06:54] <carlos> pitti: there are others that lack a .pot files
[06:55] <pitti> janimo: can you please look at http://people.ubuntu.com/~pitti/langpacks/rosetta-buildd-diff-report.txt
[06:55] <janimo> but that was awhile ago when they were not in main
[06:55] <janimo> requirements probably changed
[06:55] <carlos> janimo: anything that is in main, should be imported into Rosetta
[06:55] <pitti> janimo: in particular in the last section, which shows the domains I have in my buildd tarballs, but not in the rosetta tarball
[06:55] <pitti> janimo: so if these pacakges are still unstripped, and there are no pkgstriptranslation tarballs for them yet, they need a rebuild
[06:56] <janimo> carlos: ok, It was a while ago I asked on the rosetta list and was told that needs upstreams consent. although it makes sense to ignore upstrream if we wanrt a translated ubuntu :)
[06:56] <carlos> janimo: that's only true for products, not for Ubuntu
[06:56] <pitti> janimo: maybe you can look what's still in /usr/share/locales, and rebuild the packages which still have mo files there?
[06:56] <janimo> pitti, ok. will look at them as soon as beta2 is done ok?
[06:56] <janimo> tomorrow probably
[06:56] <pitti> janimo: yes, sure, it's not particularly urgent; it should just be cleaned up for final
[06:57] <janimo> definitely
[06:57] <pitti> great
[06:57] <pitti> carlos: ok, I build new langpacks now
[06:57] <carlos> janimo: please, be sure that the build generates the .pot files
[06:58] <janimo> carlos: ok I'll look at it. probably wil come to ask you guys for details
[06:59] <pitti> janimo: 'use the force, Luke' - or rather, cdbs rules :)
[06:59] <janimo> ;)
[06:59] <janimo> wait, is that in the langpack.mk?
[06:59] <janimo> if so that's included in xfce,mk for a while
[07:00] <KaiL> pitti, do I understand the lastest fix in gnome-volume-manager right, if this allows now to eject the disc with the drives hardware button?
[07:00] <Mez> pitti: cdbs rules? depends on what POV you look at it from
[07:00] <pitti> KaiL: yes
[07:00] <KaiL> cool :)
[07:00] <pitti> KaiL: it has always been supposed to work that way
[07:00] <janimo> yeah, I remember uploading a few weeks ago with changelog along the line: generate pot file
[07:00] <pitti> KaiL: it just broke due to a subtle bug I fixed
[07:00] <KaiL> but never did? ;)
[07:00] <_ion> mez: Yeah, from the left it looks like it rules, but from the right it looks like it's awesome.
[07:01] <Mez> _ion - it's useful I'll give it that - but IMO too many people rely too heavily on cdbs and it makes their packaging awful
[07:02] <pitti> Mez: I meant it as a noun there, but the verb is equally true :)
[07:02] <Mez> pitti: ah :P
[07:03] <pitti> . o O { oh no, not *that* flamewar again }
[07:03] <_ion> :-D
[07:03] <Mez> pitti: lol - I don't see it as a flamewar ... It's just a difference of opinioin
[07:04] <pitti> Mez: indeed; cdbs was incredibly helpful for all this translation stuff; I know that some packages do braindead stuff with it, though
[07:04] <pitti> like automatically generating debian/control *shudder*
[07:05] <Mez> pitti: that and the fact that it sometimes decides to go off on a tangent and do things that you dont want it to do all of it's own accord
[07:05] <pitti> heh :) yes, it's as powerful in helping you as with helping you shoot yourself in the foot
[07:06] <_ion> Does someone actually make it generate d/control automatically? Eww.
[07:06] <Mez> exactly, which is why I prefer debhelper.
[07:06] <_ion> cdbs _uses_ debhelper. :-)
[07:06] <Mez> ok, I may use it myself occasionally ... but meh
[07:06] <pitti> _ion: yes, that's what I meant with 'some packages abuse it'
[07:07] <LaserJock> on debian-mentors there was someone using debian/rules to generate debian/changelog
[07:07] <LaserJock> that is a bit too much automation, I think ;-)
[07:07] <_ion> Next up: using debian/rules to generate debian/rules
[07:08] <azeem> I think I saw debian/patches patching debian/rules
[07:08] <_ion> Hehe.
[07:11] <pitti> janimo: http://paste.ubuntu-nl.org/12979 <- list of packages without a POT file; it could be useful for you to decide which xfce packages need a rebuild with our new cdbs (or need to be fixed to use langpacks.mk)
[07:12] <LaserJock> azeem: that seems scary
[07:20] <Lure> Kamion: Kubuntu beta2 ubuquity install completed (reboot still does not work) - good work and I really LOVE it!
[07:21] <pitti> Keybuk: ok, I need to do an n-m upload anyway to add POT file building. I'll just remove the 'don't call wpasupplicant for unencrypted networks' patch then. or did you happen to know a differnet solution from any discussion I didn't see?
[07:21] <Keybuk> LaserJock: do you know who that was?
[07:22] <Keybuk> LaserJock: and where they live?
[07:22] <LaserJock> Keybuk: what? the automatic changelog thing?
[07:22] <Keybuk> pitti: could you just send me the patch for the POT stuff -- I have an upload half-prepared here
[07:22] <Keybuk> LaserJock: yes
[07:22] <Keybuk> LaserJock: because then I'll pay them a visit
[07:22] <pitti> Keybuk: oh, sure; will you care about removing the patch yourself?
[07:22] <Keybuk> and cut them into little pieces
[07:22] <Keybuk> and feed them to my dog
[07:22] <Keybuk> just to make sure they DO NOT DO IT AGAIN
[07:22] <waylandbill> Lets say a user decides to choose "root" as the initial user thinking it would make them the super user. The system lets them with a UID of 1000. That's only mildly confusing.
[07:22] <Keybuk> pitti: yup, already done
[07:23] <LaserJock> Keybuk: it was on the debian-mentors ML, and it was because upstream "decided" they wanted to do the debian packaging
[07:23] <pitti> Keybuk: erm, *cough*, langpack changelogs are generated automatically (not from debian/rules, though)
[07:24] <Keybuk> pitti: not from debian/rules is fine :)
[07:24] <pitti> *phew*
[07:31] <pitti> Keybuk: http://paste.ubuntu-nl.org/12984 is that enough for you or do you want a mail?
[07:31] <pitti> (it's a trivial patch anyway)
[07:32] <mdz> Keybuk: they could always include a custom changelog format parser which would regenerate it...*duck*
[07:33] <mdz> waylandbill: it does?  I'd expect adduser to refuse to create duplicate users
[07:33] <mdz> waylandbill: mizar:[~]  sudo adduser root
[07:33] <mdz> adduser: The user `root' already exists.
[07:34] <ogra> what gets intresting is to call the initial user admin
[07:34] <mdz> Lure: what goes wrong with the reboot?
[07:34] <ogra> i wonder if that has any hidden sideeffects
[07:34] <highvoltage> ogra: because it creates a group called admin too?
[07:34] <waylandbill> the installation program uses adduser to add the initial user does it?
[07:34] <ogra> highvoltage, yes and the admin group already exists
[07:35] <highvoltage> yeah. i'm going to change our docs to suggest the first user being called sysadmin instead of admin because of that.
[07:35] <Lure> mdz: it does not reboot - you have to do it manually - I have published syslog in bug 41683
[07:35] <Ubugtu> Malone bug 41683 in ubiquity "manual partitioning problems" [Normal,Unconfirmed]  http://launchpad.net/bugs/41683
[07:36] <waylandbill> cause I was chatting to someone who was root, but had a UID of 1000.
[07:36] <highvoltage> waylandbill: perhaps they just edited their /etc/passwd file
[07:36] <mdz> Lure: failure to reboot doesn't sound like a partitioning problem
[07:36] <waylandbill> and they were wondering why they were getting permissions denied all the time. :-D
[07:36] <waylandbill> highvoltage, that's what I'm thinking now.
[07:36] <highvoltage> waylandbill: you can rename your root user to unclebob, then add a new user called root
[07:37] <Lure> mdz: true, I just logged the whole process there as there are still some minor issues. Should I open separate bug for reboot issue?
[07:37] <waylandbill> I just assumed they installed that way, but it's more likely you're scenario is correct as they'd not likely tell me how they managed that.. 
[07:37] <mdz> Lure: yes, but only if there isn't one open already
[07:38] <mdz> waylandbill: yes, it does use adduser
[07:38] <Lure> mdz: will check and do
[07:38] <highvoltage> waylandbill: that so sounds like some kiddie just trying to be clever on an irc channel :)
[07:39] <ogra> highvoltage, i dont know if its a problem or not ... theoretically it shouldnt be a problem to call the admin user admin
[07:39] <waylandbill> highvoltage, that's possible.
[07:39] <ogra> but it might have surprising side effects, you never know
[07:43] <highvoltage> ogra: we've had some strange, unique situations before, where i suspected that the admin user/group might have something to do now, i'm racking my brain to think what happened in those instances, but i just can't remember
[07:44] <ogra> highvoltage, thats fine, on a public holiday you are allowed to leave your brain in the closet
[07:44] <ogra> ;)
[07:44] <highvoltage> hehe
[07:45] <highvoltage> i would've yelled something like, "are you calling my brain gay!?", but i'll behave myself on ubuntu-devel instead.
[07:46] <ogra> are gays supposed to be in closets on public holidays ? 
[07:47] <highvoltage> heh, i don't know.
[07:47] <waylandbill> lol
[07:47] <ogra> hey you are nominated for the edubuntu council 
[07:48] <ogra> nice :)
[07:48] <highvoltage> cool :)
[07:48] <janimo> pitti, great only the xfce apps I expected not to have pot file yet, as I did not convert them to xfce,mk
[07:48] <janimo> they have been in universe till a few days ago
[07:48] <janimo> will fix them after beta2
[07:49] <Keybuk> pitti: yup, that's fine
[07:49] <pitti> Keybuk: too late, I just mailed you :) thanks, though
[07:49] <Keybuk> hehe
[07:49] <janimo> was there a liveCD not rebooting issue with beta1? like waiting fro keypress but not rebooting?
[07:49] <Keybuk> just grabbing a bite to eat :P
[07:49] <janimo> to make sure it was not just xubuntu live
[07:50] <janimo> mdke: hi, are you in charge of ubugtu?
[07:58] <mdz> pitti: confirmed, ubuntu6 fixes it here. thanks!
[07:58] <pitti> cheers :)
[07:58] <mdke> janimo, no, Seveas 
[07:58] <janimo> mdke: ok thanks
[07:59] <janimo> Seveas: if you control ubugtu can you put it on #xubuntu
[07:59] <janimo> thanks
[07:59] <pitti> carlos, Riddell: http://people.ubuntu.com/~pitti/langpacks/ has new langpacks; can you please test them?
[07:59] <pitti> Riddell: I'm especially interested in the KDE ones; first ones EVAR with Rosetta love :)
[08:00] <zyga> pitti: yeah!!! :)
[08:01] <carlos> pitti: ok, downloading it
[08:06] <Riddell> just what is the en_DK locale?
[08:06] <zyga> Riddell: just guessing, denmark?
[08:06] <mdke> dk is denmark, never heard of that locale though
[08:06] <pitti> Riddell: special gift for Fabio
[08:06] <pitti> :)
[08:07] <Riddell> I'd have thougt he would work in it_DK :)
[08:10] <Riddell> pitti: kde-fr translations are looking good
[08:11] <Riddell> seems to have lost k3b but that'll need me to fix it for utf-8 first then carlos can import it
[08:12] <carlos> Riddell: cool
[08:13] <Diziet> *phew*  That 1.5.0.2 seems to work just nicely.
[08:13] <carlos> pitti: I thought your scripts take the .po from buildd if rosetta lacks it
[08:13] <Diziet> Now we just have to wait for my DSL.  I wonder if I'll be uploading 0ubuntu2 before the .orig.tar.gz has arrived ?
[08:16] <janimo> dholbach: seems noone is answering the gnumeric mail :)
[08:17] <janimo> either it's obviously right, or obviously so wrong it cannot be considered :)
[08:18] <pitti> carlos: they do
[08:23] <pitti> carlos, Riddell: maybe they give errors when msgfmt'ed?
[08:26] <Riddell> pitti: maybe what does?
[08:29] <_ion> http://www.audacity.com/
[08:30] <janimo> nomed, colossus said in the mornig uploading to sf.net 'now'
[08:30] <janimo> I guess post beta2
[08:38] <jordi> dholbach: dude
[08:41] <dholbach> jordi: quoi?
[08:42] <dholbach> janimo: will have a look in a bit
[08:43] <janimo> dholbach: thanks. But as you said, other should have a look :)
[08:43] <janimo> too
[08:44] <jordi> dholbach: er, sorry. I thought alacarte was fsckd in Catalan, but I just didn't uplad the translation.
[08:44] <Amaranth> :(
[08:44] <dholbach> jordi: you might want to talk to Amaranth about that :)
[08:44] <Amaranth> Well, glad I didn't do it. :)
[08:45] <Amaranth> jordi: You're going to have to teach the folks going to GUADEC how to order a beer. :)
[08:45] <jordi> Amaranth: ?
[08:46] <jordi> Amaranth: "posa'm una birra!"
[08:46] <jordi> :)
[08:46] <Amaranth> jordi: :)
[08:46] <jordi> but, depends on the size of course.
[08:46] <jordi> 1/3 of a litre is a "mitjana". BUT, if you go a bit to the South, it'll be a "tercio". :)
[08:46] <jordi> and it madrid it has another name iirc
[08:46] <Amaranth> confusing
[08:46] <jordi> heh
[08:47] <jordi> coffee names are confusing
[08:47] <jpatrick> no ell que sap?
[08:47] <jordi> coffee with ice is "caf del temps" here
[08:47] <jpatrick> s/no/i
[08:47] <jordi> but, say that in madrid, and they will think you're silly
[08:48] <jordi> and in other places they will say, "you... err, want *ice* in your coffee??!"
[08:48] <Amaranth> oops, you already did this (reads blog)
[08:48] <Amaranth> err, or not
[09:16] <mako> mdke: yes, let me take a look.. in principle it sounds like a very good idea
[09:16] <Nafallo> is burning busted in current dapper?
[09:16] <mdke> mako: thanks
[09:23] <Hadaka> Hello - are there any live-cd:s for Ubuntu Server?
[09:23] <tseng> no, that isnt a very common use case
[09:25] <Hadaka> Hmh, actually, is Ubuntu Server anything more than a selection of packages to put on an install CD?
[09:25] <tseng> no
[09:25] <tseng> not really.
[09:25] <Hadaka> okay, thanks
[09:27] <Hadaka> next question - if I want to make ubuntu based filesystem images (eg. for virtualization use, pre-installs, that sort) - is it just debootstrap from ubuntu repositories & friends, or does ubuntu have something else?
[09:28] <zul> debootstrap should work
[09:29] <Hadaka> how does the ubuntu live cd -> harddisk install thing work? does it make a fresh install, or does it copy stuff, or?
[09:32] <janimo> Hadaka: it should work. it copies over what's on the live CD to the partition to install
[09:32] <Hadaka> janimo: okay, thanks, that sounds good actually
[09:33] <LaserJock> I finally go the live CD installer to work on my intel iMac (using Parallels virtualization) and it was quite fast.
[09:50] <janimo> mdz: hi, regarding the libgoffice build
[09:50] <janimo> even if we explicitely depend on the -gtk only one for gnumeric-gtk
[09:50] <janimo> the dep added via shlibs is still there no?
[09:51] <janimo> so what we do is remove it using sed. that is the hack
[09:51] <janimo> which we don;t see a better alternative to
[10:00] <mdz> janimo: it should be an alternative dependency as I described in my mail
[10:00] <janimo> mdz, but how do we make sure the good one is chosen at runtime iif gnumeric-gtk will depend on both libgoffice and libgoffice-gtk?
[10:00] <janimo> at least that's what I understood from the mail
[10:01] <janimo> since the non-gtk dep is added automatically via shlibs:depends
[10:02] <mdz> janimo: that's what I described in the mail
[10:02] <mdz> let's talk about it after the meeting
[10:02] <janimo> ok
[10:04] <janimo> infinity: I know you're sick. So if you can't spin a xubuntu live now is entirely understood, and I can wait till tomorrow mornig or cron
[10:04] <janimo> just need a beta2 candidate
[10:05] <infinity> janimo: The current beta2 candidate is no good?
[10:05] <kagou> too tired. see you later
[10:05] <janimo> infinity not good
[10:06] <janimo> infinity we got away with some missing deps which were broght in by gnome-session so far :)
[10:06] <janimo> since that got removed since today morning it was dicovered that deskto picons cannot launch
[10:06] <janimo> but now the fix is in the archive
[10:10] <simira> mvo: the install-button in the update-manager is active even though there's no updates
[10:11] <mvo> simira: oh, thanks
[10:12] <infinity> janimo: Okay, spinning then.
[10:12] <janimo> infinity:  thanks
[10:40] <janimo> Kamion, does your big lock on little affect livecd builds in any way? one build was triggered about 20 min ago, not sure if it should have shown up or not
[10:41] <dAndy_> is there an easy way to test preseeding config info for packages without doing a whole install?
[10:41] <Kamion> janimo: oh, bugger, you're not telling me there was an ISO build in progress
[10:41] <Kamion> janimo: no, that was a livefs build I think, not an ISO build
[10:42] <janimo> Kamion, how would I know what BIG lock means ? :)
[10:42] <janimo> not sure even if little is the build host or not
[10:42] <Kamion> janimo: livefs builds don't go anywhere near little (the cdimage build machine) until ISO image builds start
[10:42] <janimo> ok then
[10:43] <Kamion> so I can't do a CD build for you until I've unbroken the tree that gets synced out to releases.u.c
[10:43] <janimo> although the live would become an iso eventually
[10:43] <janimo> ok
[10:43] <Kamion> it affects you to precisely that extent
[10:43] <janimo> it;s beta2 candidate with the fix
[10:43] <Kamion> I know
[10:43] <janimo> np
[10:43] <janimo> I'll rsync it in the morning if mirrorng takes long
[10:46] <Keybuk> http://www.flickr.com/photos/mojen/135613674/
[10:49] <Keybuk> ^ there's something very wrong about that
[10:50] <tseng> haha
[10:50] <Kamion> (from #ubuntu-meeting) http://xanna.livejournal.com/51821.html
[10:54] <_ion> Windows  humanity to fluffy toys.
[10:58] <jdub> "The name of the author may not be used to endorse or promote products derived from this software without specific prior written permission." - clause from libjingle license
[10:58] <jdub> freeornot.com?
[10:59] <infinity> jdub: That's free, IMO.
[11:00] <infinity> jdub: Doesn't say the name can't be included in distribution (and, indeed, it must, as you must retain the copyright), just that you can't use it for endorsements and promotion, no different than trademark protection clauses.
[11:00] <infinity> jdub: Or, think of it as a "reverse BSD ad clause". :)
[11:00] <_ion> I would consider that free as well.
[11:00] <elmo> err, it's _in_ the BSD license
[11:00] <elmo> 3. Neither the name of the University nor the names of its contributors may be used to endorse or promote products derived from this software without specific prior written permission.
[11:01] <elmo> it's obviously free
[11:01] <infinity> Oh, fair point.  I always forget clause 3.
[11:02] <infinity> elmo: Can you quote from the GPL verbatim too?
[11:02] <Keybuk> I have a drunken party piece of quoting the MIT licence at high speed
[11:02] <Keybuk> needs a lot of beer though
[11:02] <_ion> From memory?
[11:03] <Keybuk> yeah
[11:04] <infinity> Keybuk: With or without the warranty disclaimer?
[11:04] <Keybuk> with
[11:04] <infinity> And if with, do you SHOUT IT AT THE TOP OF YOUR LUNGS?
[11:04] <Keybuk> I shout that bit
[11:04] <Keybuk> heh
[11:04] <Keybuk> yeah :)
[11:04] <_ion> :-D
[11:05] <_ion> Me wants to see it. You should shoot a video.
[11:06] <Kamion> Keybuk: very worried about you
[11:06] <Kamion> Keybuk: my wife says she's heard about you and is not surprised
[11:07] <Keybuk> I'd be worried, except I'd imagine your-wife-knows-fanf-who-knows-flash-who-knows-me
[11:07] <Keybuk> in that Cambridge-people-knowing-everyone-in-IT kind of way
[11:07] <Kamion> true, though the route may be via Kinnison instead
[11:07] <Keybuk> true, true
[11:09] <Keybuk> oh, yay Orange
[11:09] <Keybuk> both my phones now say "Inactive SIM"
[11:10] <Kamion> maybe I should decide how I want releases.u.c to look before munging it
[11:10] <janimo> pitti, got a link again to the list of 16 potless packages?
[11:10] <Kamion> mdz: since I'm going to have *-beta2-live.iso, do you want me to rename the *-beta-install.iso symlinks to *-beta2-install.iso?
[11:10] <Kamion> or leave them be, since they're unchanged?
[11:10] <Keybuk> damn, I picked up the wrong flavour Pringles
[11:11] <mdke> always the green
[11:11] <Keybuk> BLUE!
[11:11] <pitti> janimo: http://paste.ubuntu-nl.org/12979
[11:11] <janimo> thanks
[11:11] <mdke> yeah
[11:11] <Keybuk> I picked up the red ones, which are David's; if I ate those I'd be IN TROUBLE!
[11:14] <mdke> yeah, your taste buds will hate you
[11:15] <mdz> janimo,Keybuk: ok back
[11:15] <danimo> hi pitti
[11:16] <Keybuk> right
[11:16] <pitti> hi danimo 
[11:16] <mdz> Kamion: if we're renaming things anyway, can we go to -desktop and -whatever?
[11:16] <janimo> mdz, ok. I am not sure I understand what to do about the libgoffice dependency which gets added
[11:16] <danimo> pitti: riddell told me you are the cups dude
[11:16] <janimo> how to make it not get added?
[11:16] <danimo> pitti: I am about to report a bug wrt foomatic
[11:16] <danimo> pitti: what details do you need?
[11:17] <Keybuk>  o xfce4-clipman-plugin: xfce4-clipman-plugin
[11:17] <Keybuk>    [Reverse-Depends: Xubuntu-Dapper desktop seed] 
[11:17] <Keybuk>  o xfce4-mount-plugin: xfce4-mount-plugin
[11:17] <Keybuk>    [Reverse-Depends: Xubuntu-Dapper desktop seed] 
[11:17] <Keybuk>  o xubuntu-system-tools: xubuntu-system-tools
[11:17] <Keybuk>    [Reverse-Depends: Xubuntu-Dapper desktop seed] 
[11:17] <Keybuk> those seem pretty obvious given they're seeded? :)
[11:17] <mdz> janimo: what's in the shlibs file?
[11:17] <pitti> danimo: hm, no idea, I know next to nothing about foomatic
[11:17] <Kamion> mdz: this is already giving me nightmares - a little more leisure there would be preferable
[11:17] <mdz> Keybuk: new source needs main inclusion reporst
[11:17] <janimo> mdz, shilibs is autogenerated
[11:17] <pitti> danimo: just file it for now with everything that could be relevant
[11:17] <danimo> pitti: the process constantly dies
[11:17] <Kamion> (yeah, I know I should have done that straight after beta)
[11:17] <janimo> and adds libgoffice 
[11:17] <pitti> danimo: /var/log/cups/error_log with LogLevel=debug shoudl help as first shot
[11:17] <janimo> so we sed that out and put libgoffice-gtk instead
[11:17] <Keybuk> mdz: just looking that up now
[11:17] <mdz> Kamion: sure, I wouldn't have said anything except that you asked ;-)
[11:17] <mdz> Kamion: beta-install vs. beta2-install doesn't make much difference tome
[11:17] <Keybuk> xubuntu-system-tools: MainInclusionReportXubuntuSystemTools: code duplication
[11:18] <mdz> Keybuk: we can do the easy ones first
[11:18] <Keybuk> (needs work)
[11:18] <mdz> Keybuk: new binary promotions are usually no-brainers unless they look funny
[11:18] <Keybuk> the only one of those is git-email
[11:18] <mdz> yeah, it has a crapload of deps though
[11:18] <Kamion> mdz: OK, I think I'll make it beta2-install then on the grounds that having all the images in the directory be the same status confuses stuff less
[11:18] <Keybuk> Depends: git-core (= 1.1.3-1ubuntu1), libmail-sendmail-perl, libemail-valid-perl
[11:18] <infinity> Keybuk: me, pitti, and Kamion went through git-email's dependencies a while back and deemed them "okay".
[11:18] <mdz> Keybuk: who added git-email to supported?
[11:19] <Keybuk> mdz: not me
[11:19] <mdz> Keybuk: there are 5 new sources in the 'source and binary promotions' list which come from git-email
[11:19] <Kamion> infinity: we did? I must have forgotten doing that
[11:19] <infinity> Kamion: It was a week or two back...
[11:19] <Kamion> mdz: IIRC fabbione but I could be wrong
[11:19] <mdz> Kamion: bzr annotate agrees with you
[11:19] <infinity> Kamion: It came up on IRC, we spoke of it, I hunted the Debian bugs, decided we weren't affected by them, etc..?
[11:20] <mdz> janimo: libgoffice should be changed to have a different shlibs, rather than changing the shlibdeps-generated dependency in gnumeric
[11:20] <infinity> pitti: Surely, you recall this conversation? :)
[11:20] <Keybuk> pitti: deny all knowledge
[11:21] <mdz> janimo: look at, e.g., libgl1-mesa
[11:21] <pitti> infinity: yes, of course
[11:21] <janimo> mdz, ok. but since only one lingoffice will be used in the build, it will probably result in the same shlibs generated for gnome and gtk variants of gnumeric
[11:21] <mdz> pitti: were the packages reviewed or no?
[11:21] <janimo> mdz, I'll look at mesa
[11:21] <infinity> pitti: Right, tell them... *waves frantically in the general direction of Keybuk, mdz, and Kamion*
[11:22] <infinity> :)
[11:22] <janimo> I forgot to write main incl report for two plugin
[11:22] <pitti> mdz: quickly, but there are no reports; it seemed to be a bit urgent last time
[11:22] <janimo> will do it
[11:22] <pitti> if it didn't happen until today, we can as well do full reports
[11:22] <infinity> mdz: We did a quickie "on IRC" inclusion review.  But we can do it "properly", if that's preferred.
[11:22] <mdz> pitti: it doesn't seem urgent; nothing is   broken because of it.  it's a new package added to supported
[11:22] <Keybuk> git-email => (libemail-valid-perl => [libnet-dns-perl => {libnet-dns-perl}, libnet-domain-tld-perl] , libmail-sendmail-perl)
[11:22] <mdz> infinity: jbailey might have something to say about it
[11:23] <Keybuk> seems to be the list
[11:23] <pitti> infinity: AFAIR the Debian bugs didn't apply to us, right?
[11:23] <infinity> pitti: Right, the Debian bugs apply to a new upstream version we're not shipping.
[11:23] <infinity> pitti: And will be fixed by the time we open edgy.
[11:23] <infinity> pitti: So we'll smoothly skip right over them.
[11:23] <Keybuk> oh, it's in git-core
[11:23] <pitti> yes
[11:23] <pitti> he just didn't mention the dependencies in git-core's report
[11:23] <fabbione> Keybuk: tsk
[11:23] <mdz> right, it's only a new binary, so it's just the deps
[11:24] <mdz> fabbione: is git-email a new binary or was it just not seeded before?
[11:24] <fabbione> mdz: not seeded befire
[11:24] <fabbione> before
[11:24] <mdz> ah
[11:24] <fabbione> it's part of git-core source that's already in main
[11:24] <Keybuk> Perl script to send e-mail on commits?
[11:24] <fabbione> or should be since i seeded git a while ago
[11:25] <mdz> pitti: cyrus21 is deprecated in favour of cyrus-*-2.2, right?
[11:25] <fabbione> Keybuk: no, use mbox for patch management. wildly used on kernel stuff
[11:25] <Keybuk> fabbione: oh
[11:25] <pitti> mdz: I guess so, but I have no idea whether 2.2 has been tested, or how the transition goes
[11:26] <mdz> pitti: well, it should be moved to universe at any rate
[11:26] <mdz> Keybuk: cyrus21-imapd + binaries can go
[11:26] <Keybuk> mdz: done
[11:26] <pitti> yes, doesn't seem to have any rdeps
[11:27] <Keybuk> dasher went in and out of main quickly
[11:27] <mdz> doko: gcj-4.0?  should it go to universe or be removed entirely?
[11:27] <pitti> mdz: if we don't want it on it's own right (but then, we have dovecot...)
[11:27] <mdz> Keybuk: I think doko's gone back to real life, should email him
[11:27] <Kamion> elmo: any ideas on where warty CD images should be shunted to once we EOL warty?
[11:28] <infinity> We need a "history.ubuntu.com" for old CD images and pool/dists, I suspect.
[11:28] <trappist> kill list?  does that put eterm on the kill list?
[11:28] <Kamion> Keybuk: unfortunately anastacia is currently hacked up rather than being a proper LP app, so it relies on the publisher having run
[11:28] <mdz> Keybuk: imlib definitely ok to move
[11:29] <Keybuk> Kamion: I'm aware
[11:29] <mdz> Riddell: ivman?
[11:29] <Kamion> once it talks to the database directly it'll be easier to regenerate output on the spot
[11:29] <Kamion> trappist: eterm has never been in main
[11:29] <mdz> Keybuk: eximdoc4 smells like it should be added to supported
[11:29] <janimo> mdz, ivman can go as far as xubuntu is concerned
[11:29] <trappist> Kamion: oh, misunderstood the "kill list" nomenclature
[11:29] <mdz> janimo: what do you use instead?
[11:30] <janimo> mdz, thunar talkt to HAL
[11:30] <janimo> talks
[11:30] <mdz> janimo: was ivman in main only for xubuntu?
[11:30] <Keybuk> mdz: "eximdoc4-html" is in supported
[11:30] <janimo> mdz, yes for breezy
[11:30] <janimo> afaik kde no longer uses it
[11:30] <Keybuk> no package exists for that though
[11:30] <mdz> Keybuk: sounds like it was renamed
[11:30] <janimo> and we used it for dapper for a good while
[11:30] <mdz> Keybuk: sounds like ivman can go
[11:31] <mdz> Keybuk: check with Riddell on konserve
[11:31] <mdz> wonder what happened with tspc
[11:31] <mdz> I thought that was seeded
[11:31] <mdz> for some fabbione ipv6 madness ;-)
[11:31] <Keybuk> fabbione: ?
[11:32] <mdz> Keybuk: he's resting
[11:32] <mdz> mail would be better
[11:32] <mdz> Keybuk: lots of binary *-doc to be seeded
[11:32] <mdz> and -dbg
[11:32] <Keybuk> yeah, building a stack of mails to send as I go
[11:32] <mdz> festival-dev should be seeded
[11:32] <mdz> check with doko on the gcc stuff
[11:32] <mdz> nm-applet should be seeded
[11:33] <mdz> and the oo.o stuff
[11:33] <doko> mdz: gcj-4.0, yes, to universe
[11:33] <infinity> nm-applet can probably be dropped entirely, actually.
[11:33] <mdz> gah, wanted to ask janimo about xubuntu-artwork
[11:33] <mdz> infinity: oh?
[11:33] <mdz> oh, it's transitional
[11:33] <infinity> The only reason I added the transitional package was to ease dapper->dapper upgrade for testers.
[11:33] <infinity> We don't really need ot release with it.
[11:33] <mdz> did we have it in breezy?
[11:33] <mdz> ok
[11:33] <danimo> pitti: ok
[11:33] <Keybuk> we skipped a few there ... what about dasher, dbh, drac, libdvd* ?
[11:34] <infinity> Oh, it may have been in breezy/universe.
[11:34] <mdz> Keybuk: dasher should go to supported, I think (confirm with heno)
[11:34] <Keybuk> it wasn't in breezy/universe
[11:34] <Keybuk> I added the nm-applet package, then renamed it
[11:34] <infinity> Kay.  Then we can safely remove it again at this point.
[11:34] <mdz> dbh and drac I'm not sure about; presumably they are no longer deps of whatever used them
[11:34] <infinity> It served its purpose.
[11:34] <mdz> and their -devs were unseeded (or never seeded in the first place)
[11:35] <doko> mdz, Keybuk: ttf-gentium should stay in main, it's quiet a nice font for laser printers. 
[11:35] <mdz> Keybuk: they can both go until/unless we need them again
[11:35] <mdz> doko: it's not in main
[11:35] <mdz> doko: it's in universe but seeded; does it have an inclusion report?
[11:35] <doko> mdz: I'll do it tomorrow
[11:35] <mdz> seems likely to be a trivial package, just needs a cursory review
[11:36] <mdz> doko: if pitti is agreeable, we can skip a formal report
[11:36] <mdz> if someone looks at the packaging and confirms it is sane
[11:36] <dholbach> good night guys
[11:36] <Keybuk> ah
[11:36] <Kamion> dbh was a xubuntu dep at one point; it doesn't seem to be any more
[11:36] <Keybuk> libdvdnav* were deps of gst-plugins-0.8
[11:36] <Keybuk> I assume with the move to 0.10 they fell out
[11:36] <Keybuk> I guess they might be coming back as 0.10 gets more plugins?
[11:36] <pitti> mdz: agreed; I can help you in a few minutes, I just want to finish the mozilla USN
[11:36] <Kamion> yeah, I was avoiding demoting dvdnav and dvdread in case gstreamer dvd support came back in
[11:37] <Kamion> which I understood we were hoping for
[11:37] <mdz> that hope has expired
[11:37] <doko> pitti, mdz: ttf-gentium packaging is ok. defoma-fonts could need some love
[11:38] <mdke> what would you guys recommend for playing dvds in dapper? we recommended gxine for the desktop guide
[11:38] <mdz> we could seed the -devs to keep them around
[11:38] <mdz> mdke: I use totem-xine
[11:38] <pitti> mdz: I'm fine with approving fonts, dictionaries, etc. without a formal report
[11:38] <mdke> mdz: does that coexist with totem-gstreamer?
[11:38] <mdz> mdke: no
[11:38] <mdke> right, we've generally recommended gstreamer for everything except dvds :)
[11:38] <mdz> mdke: yes
[11:39] <pitti> mdke: playing DVD with totem-gst works generally for me, but I cannot search in the stream
[11:39] <mdke> you don't get menus or subtitles either, apparently
[11:39] <pitti> mdke: t-xine works well here, too
[11:39] <doko> pitti, mdz: cyrus: it works fine for me, but you have to manually upgrade your database. just checked very basically with a test installation. but there's no automatic upgrade. otoh, db3 could go to universe as well
[11:39] <mdz> pitti: really?  seb128 said totem-gst was fairly broken for dvd with gst 0.8, and of course with 0.10 there is no dvd support at all in gstreamer
[11:39] <pitti> mdz: hm, apparently there is now
[11:39] <mdke> mdz: there is dvd support, just no menus/subtitles
[11:40] <mdke> it's in -ugly I think
[11:40] <pitti> mdz: if I insert a DVD, totem opens automatically and plays my DVD
[11:40] <pitti> mdz: yes, I have all the -ugly, -bad, -stinking, -eatskittens, etc. gst plugins installed
[11:40] <mdke> heh
[11:40] <mdz> hm, so it is.  I didn't see that go in
[11:40] <mdz> lack of menus is fairly crippling though for many use cases
[11:41] <pitti> doko: hey, indeed, cyrus21 is the only thing that wants libdb3
[11:41] <mdz> I run totem-xine on my dvd playing machine, and -gstreamer everywhere else
[11:41] <pitti> Keybuk: so if you demote cyrus21, make sure to kill db3 as well :)
[11:41] <mdke> ok, gxine seems a happy compromise then
[11:42] <mdz> BenC: what's linux-kernel-devel?
[11:42] <mdz> ah, I see
[11:42] <Keybuk> hmm, should we promote that ?
[11:42] <mdz> Keybuk: linux-kernel-devel -> supported
[11:43] <mdz> it's already in main
[11:43] <Keybuk> pitti: has other reverse deps
[11:43] <pitti> Keybuk: uh, which? I just checked
[11:43] <Keybuk> s/promote/seed/
[11:43] <Keybuk> http://people.ubuntu.com/~cjwatson/germinate-output/dapper/rdepends/db3/libdb3
[11:43] <pitti> Keybuk: oh, sorry, yes
[11:43] <Keybuk> bonobo, gconf, etc.
[11:43] <pitti> librpm-dev
[11:43] <Keybuk> seb would be upset at you
[11:43] <pitti> that stuff is universe
[11:44] <doko> mdz, pitti: looking at the python stuff tomorrow, and seed where appropriate. all OOo help and l10n packages should be sucked in by language packs.
[11:44] <pitti> -- dapper/main build deps on libdb3-dev:
[11:44] <pitti> cyrus21-imapd
[11:44] <pitti> iproute
[11:44] <pitti> rpm
[11:44] <Keybuk> oh, yeah, I always get confused by those
[11:44] <Keybuk> rpm is in main though
[11:44] <Kamion> Keybuk: p.s. checkrdepends in lp_archive's $PATH on drescher is what pitti's running there
[11:44] <pitti> yep, and iproute
[11:44] <Kamion> handy
[11:44] <Keybuk> Kamion: oh, handy
[11:45] <Keybuk>  o brltty-flite brltty-x11                                            {brltty}
[11:45] <BenC> mdz: it's a meta-package that installs everything needed to build a kernel from our source
[11:45] <mdz> pitti: could you take care of language-support-* for the oo.o stuff in anastacia?
[11:46] <Keybuk> those appear to be "optional drivers"
[11:46] <mdz> BenC: I was wondering if we should have a metapackage which installs everything you need to build a module
[11:46] <BenC> that would probably do it too
[11:46] <mdz> BenC: linux-headers, build-essential (is that all?)
[11:47] <BenC> pretty much, yeah
[11:47] <mdz> that's what I thought linux-kernel-devel would be before I looked at it
[11:47] <mdz> pitti: goodness, that is a lot of CVEs
[11:47] <pitti> mdz: openoffice.org-help-br openoffice.org-l10n-br ? sure
[11:47] <mdz> pitti: yes, thanks
[11:48] <pitti> mdz: yes, that explains why there was such a long delay for 1.7.13/1.0.8
[11:48] <infinity> mdz: "Everything you need to build a module" would have to be clever about picking the right headers-$(flavour), or just install all of them (eww) for your arch.
[11:48] <Keybuk> doko: ok, gcj-4.0 demoted -- nearly missed that (you said it to mdz, not me)
[11:49] <doko> Keybuk: thanks, will seed ecj-bootstrap-gcj.
[11:49] <infinity> mdz: I guess "linux-modules-devel-{386,686,k7}" in linux-meta could do it, but I'm not sure how helpful/discoverable that would be.
[11:50] <mdz> infinity: linux-headers-<generic> | linux-headers-{foo,bar,baz} would be appropriate
[11:50] <Keybuk> pitti: libcupsys2-gnutls10 ?  what's that
[11:51] <pitti> Keybuk: transitional package for breezy update, should be seeded
[11:51] <pitti> mdz: l-support-br uploaded
[11:51] <Keybuk> doko: what about: g++-3.3 libstdc++5-3.3-dev g++-3.4 libstdc++6-dbg libstdc++6-dev ?
[11:51] <infinity> Keybuk: libstdc++5-3.3-dev shouldn't be falling out..?
[11:51] <infinity> Keybuk: It's a build-dep of LRM.
[11:52] <pitti> btw, is anyone changing seeds right now? it would be nice to add vim-tiny to supported (source is vim)
[11:52] <mdz> infinity: it only build-deps on libstdc++5
[11:52] <infinity> Oh, silly me.
[11:52] <Keybuk> infinity: no it isn't :)
[11:52] <infinity> And that's fine, actually, cause all I need is the shlibs.
[11:52] <danimo> pitti: in other news, cups can't find my usb printer, but lsusb says it's there
[11:52] <infinity> So, yeah, I take it back.  The -dev can go away.
[11:52] <danimo> any clue?
[11:53] <infinity> Yay binary build-deps.
[11:53] <Keybuk> aren't the shlibs in the -dev ?
[11:53] <infinity> No, shlibs are in the binary package unless someone really messed up.
[11:53] <Keybuk> no, you're right
[11:53] <Keybuk> pitti: I have the seeds open yeah
[11:53] <doko> Keybuk: we need them for fglrx in restricted AFAIK
[11:53] <infinity> doko: See above.
[11:54] <Keybuk> doko: we just did that discussion :p
[11:54] <infinity> doko: We only need the runtime, not the -dev.
[11:54] <pitti> danimo: sudo -u cupsys /usr/lib/cups/daemon/cups-deviced 1 1 `id -u cupsys` '' > /dev/null
[11:54] <Keybuk> infinity: I guess you don't need the g++'s either?
[11:54] <pitti> danimo: please pastebin the result
[11:54] <infinity> Keybuk: I don't use them, so I'll go with "no". :)
[11:54] <Keybuk> going...
[11:54] <Keybuk> going...
[11:54] <doko> did somebody check, if lsb 3 says something about libstdc++5?
[11:55] <danimo> pitti: ok, no usb there
[11:55] <infinity> 'win 26
[11:55] <Kamion> beta2 going out to mirrors now, unless I screwed up (which is unfortunately relatively likely)
[11:55] <Keybuk> do we want them in general for "compiling old shit?"
[11:55] <danimo> pitti: any idea what: "DEBUG: [cups-deviced]  Added device "hp:/no_device_found"..." is supposed to be?
[11:55] <Keybuk> infinity: 'lose 182,124,124
[11:55] <pitti> danimo: supposedly you do not have any HP printer hplip recognizes
[11:55] <Kamion> g++ 3.3 is the older ABI which a fair bit of proprietary stuff still uses
[11:55] <pitti> danimo: that's quite normal, though
[11:55] <danimo> ok
[11:56] <Keybuk> doko: it might say about the binary, not the -dev
[11:56] <Kamion> (afaik)
[11:56] <Keybuk> this is just the -devs and g++
[11:56] <pitti> danimo: does ls -l /dev/usblp* look sane?
[11:56] <Keybuk> the binary lib is staying in main by other things
[11:56] <danimo> pitti: http://pastebin.com/685837
[11:56] <pitti> we need to support packages whose source is in main anyway, so we can as well leave the whole bunch in main
[11:56] <danimo> pitti: yes, the printer even reacts to data sent there
[11:57] <Keybuk> pitti: ok, should I see them then?
[11:57] <pitti> Keybuk: WFM
[11:57] <pitti> danimo: does the result look any different if you use 'sudo' instead of 'sudo -u cupsys'?
[11:58] <danimo> pitti: ah, yes it does. seems to be a permission problem then
[11:58] <Keybuk> I guess people may need to compile things with an older C++ ABI to link against prop. libs
[11:58] <pitti> danimo: ls -l /dev/usblp0 ?
[11:58] <danimo> pitti: crw-rw---- 1 root lp 180, 0 2006-04-27 23:57 /dev/usblp0
[11:58] <pitti> looks good
[11:58] <danimo> pitti: but cupsys is gid p
[11:58] <danimo> gid lp
[11:59] <Keybuk> pitti: cupsys-driver-gimpprint foomatic-db-gimp-print ijsgimpprint ?
[11:59] <danimo> I just double-checked
[11:59] <pitti> Keybuk: transitional packages for the gimp-print->gutenprint renaming
[11:59] <Keybuk> seed or demote?
[11:59] <pitti> need to be seeded for dapper
[11:59] <pitti> for the breezy upgrade
[11:59] <pitti> can go in edgy
[11:59] <Keybuk> ok
[11:59] <pitti> (maybe note that in a comment)
[12:00] <Keybuk> right, now we have: libesd0 libsdl1.2debian-esd libsdl1.2debian-oss 
[12:00] <Keybuk> pitti: there's a whole section for upgrades in the supported seed :p
[12:00] <pitti> Keybuk: we consitently use libesd-alsa0 now, libesd0 uses oss which is evil
[12:00] <pitti> so feel free to drop that
[12:01] <pitti> the other sdl backends can probably stay, we need to support the source anyway