[12:03] <shaya> has something changed in breezy recently?
[12:03] <shaya> its insanely faster
[12:04] <jbailey> shaya: We added extra gerbles in prep for RC next week...
[12:05] <shaya> I'm serious
[12:05] <ajmitch> so is jbailey, I'm sure :)
[12:05] <shaya> it's so much snappier
[12:05] <shaya> could the libgnomevfs upgrade have done that?
[12:05] <seb128> no
[12:06] <shaya> I doubt gdm update did it
[12:06] <seb128> maybe you have changed your box? :)
[12:06] <shaya> nah
[12:06] <shaya> just did a big update for the day
[12:06] <shaya> ctrl-alt-backspaced
[12:06] <shaya> and logged back in
[12:06] <shaya> new splash screen was barely seen
[12:06] <ogra> so it was xorg apparently
[12:06] <shaya> as it flew so fast
[12:07] <shaya> perhaps
[12:07] <shaya> weird
[12:07] <shaya> anyways, off I go
[12:09] <mvo> pitti: single-reboot-notification stuff is prototypish ready. I'll take care for the rest tomorrow
[12:13] <seb128> Nafallo: around?
[12:17] <Lathiat> mjg59: work with vesafb = rock
[12:17] <Keybuk> oh gods, not another special virtual filesystem
[12:17] <Keybuk> securityfs?!
[12:17] <Keybuk> but we only just killed usbfs :'(
[12:19] <jcohen85> jbailey, well, it attempted to start the splashscreen after only 2-3 seconds but it failed. it went to the gnome brown screen at 34 seconds and gnome was completely started up with all menus by 59 seconds
[12:20] <Nafallo> seb128: pong :-)
[12:20] <Lathiat> Keybuk: haha
[12:20] <Lathiat> Keybuk: wtf is that for
[12:20] <Lathiat> devfs, debugfs, sysfs, procfs, securityfs, what next? )
[12:20] <seb128> Nafallo: why did you revert the xchat-systray changes?
[12:21] <jbailey> jcohen85: Lagging, I'm on the phone.
[12:21] <Nafallo> seb128: the bugreporter and patch contributer asked me to since it hanged xchat-gnome per default.
[12:22] <seb128> Nafallo: oh, k
[12:30] <jcohen85> where did boot-admin go? after the latest update in breezy, it disappeared
[12:30] <mdz> who broke metacity again?
[12:30] <mdz> hmm, behaving now
[12:31] <jcohen85> um, gnome-system-tools removed boot-admin again...damn't
[12:31] <mdz> jcohen85: it was disabled due to bugs which won't be resolved in time for se
[12:31] <seb128> jcohen85: it's broken, it has been dropped
[12:31] <mdz> s/se/release/
[12:31] <jcohen85> every time my kernel is updated, the windows xp grub option is deleted
[12:31] <jcohen85> boot-admin atleast allowed me to recreate it easily
[12:31] <mdz> jcohen85: that's because of boot-admin
[12:31] <jcohen85> *easily
[12:31] <mdz> it created the problem in the first place
[12:31] <jcohen85> jcohen85, how do i stop it from happening?
[12:32] <mdz> restore the backup of menu.lst that you made before modifying it
[12:32] <mdz> if you need specific instructions, please use the support channel
[12:33] <jcohen85> ok
[12:35] <jcohen85> heh, well the only response i got was, " i got the perfect answer!!! dont use windows ;-)"
[12:41] <jcohen85> since the last update i'm not seeing the splash screen anymore. Anyone else seeing this issue?
[12:41] <jcohen85> jbailey, i restarted again. it tries to load the splash screen after 2-3 seconds so the startup is signifigantly faster than before- only a few seconds slower than hoary now
[12:43] <jbailey> jcohen85: Still on phone, but can you please do "update-alternatives  --display usplash-artwork.so | grep status"
[12:43] <jbailey> or not..
[12:44] <michele> I don't see the splash screen either
[12:44] <michele> I just get a black screen until X starts
[12:45] <michele> usplash-artwork.so - status is auto.
[12:46] <jbailey> michele: All black, like no text at all?
[12:46] <michele> yes
[12:47] <michele> and the console stays black also if I switch back from X
[12:47] <jbailey> mjg59: ^^ any guesses?
[12:47] <mjg59> None
[12:48] <michele> this is an old breezy install upgraded (it started with colony 2 I think, before usplash got in)
[01:12] <Keybuk> well, there's some good news
[01:12] <Keybuk>  15: inputmap finds module                        FAILED (inputmap.at:12)
[01:12] <Keybuk> etc. for all the inputmap modules
[01:12] <Keybuk> so make check doesn't even pass on amd64
[01:14] <ogra> hmmm
[01:14] <ogra> could that cause keyboard problems ? 
[01:14] <mjg59> keyboard is built in statically
[01:15] <ogra> i just thought about 16539
[01:15] <Keybuk> only if you're on amd64
[01:15] <ogra> (would be nice to have a pipe and at sign again :) )
[01:15] <Keybuk> this is 64-bit specific
[01:15] <Nafallo> ogra: I don't have my synaptics atm ;-)
[01:15] <ogra> yup, i am on amd64
[01:16] <ogra> and it doesnt occur on my x86 desktop
[01:16] <Nafallo> mjg59: usb-keyboards to?
[01:16] <Keybuk> does "modprobe <something>" fix the problem?
[01:16] <mjg59> Nafallo: They're under USB, not input
[01:16] <ogra> Keybuk, what wouls <something> be ? 
[01:16] <Keybuk> no idea
[01:16] <ogra> *would
[01:16] <Nafallo> mjg59: ah, nice catch :-)
[01:16] <ogra> heh
[01:17] <Keybuk> I'm not aware of any input keyboard module
[01:17] <ogra> me neither, since keyboard is built in statically
[01:17] <Nafallo> ogra: cat /var/log/syslog | grep segfault
[01:17] <Nafallo> :-)
[01:17] <ogra> Nafallo, i cant :p
[01:17] <Nafallo> then you know ;-)
[01:17] <ogra> no pipe sign :)
[01:17] <Nafallo> hmm
[01:17] <Nafallo> copy-paste then! :-P
[01:18] <ogra> no segfault 
[01:18] <Nafallo> baah
[01:18] <ogra> its very unlikely that the kernel sends segfaults to syslog ;)
[01:18] <Nafallo> I'm the only one with the bug ;-(
[01:19] <Nafallo> grepmap does for me ;-)
[01:21] <Keybuk> heh
[01:21] <Keybuk> I'm surprised it ever worked for you, tbh
[01:22] <Keybuk> even the hoary version segfaults for me on amd64 when I poke it with obscene information
[01:22] <Nafallo> baah
[01:23] <Nafallo> must have been a change in the last version AFAIK
[01:23] <Keybuk> nah
[01:23] <Keybuk> the other version is just slightly more prone to segfaulting as a result
[01:23] <Nafallo> well, the reboot after that install triggered it... ;-)
[01:23] <Keybuk> because I put an extra sanity check into it
[01:24] <Nafallo> hmm
[01:24] <Keybuk> the basic problem is that sizeof (unsigned int) is the same as on i386
[01:24] <Nafallo> bad sanitycheck then... ;-)
[01:24] <Keybuk> but the array needs to be twice as big on amd64
[01:24] <Keybuk> I don't have a clue *WHY* it needs to be twice as big yet
[01:25] <Nafallo> 32*2=64 :-P
[01:25] <Keybuk> yes, but why do amd64's input devices need twice as many flags as i386? :p
[01:26] <Nafallo> hmm
[01:26] <Nafallo> could it be a kernel bug or something?
[01:26] <Nafallo> or wait... I haven't upgraded that one :-P
[01:27] <Nafallo> MUST be grepmaps fault ;-)
[01:27] <Keybuk> I think it's a "the input subsystem was written by monkeys" bug
[01:27] <Nafallo> hehehe
[01:27] <Keybuk> basically on amd64 these need to be unsigned hung int rather than unsigned int
[01:27] <Keybuk> and the array needs to be the same size
[01:28] <Keybuk> (number of elements wise)
[01:28] <Keybuk> which means I then can't use scanf() anymore :'(
[01:30] <Keybuk> looks like depmod has two different bits of code entirely, one for i386 and the other for amd64 *sigh*
[01:31] <Nafallo> joy! :-)
[01:32] <Keybuk> mjr: not especially, it's better than most people's roll-their-own line parsers
[01:32] <Keybuk> if you just want a number from a string, it's the right thing (tm)
[01:32] <mjr> now that's sscanf(), a completely different story :)
[01:32] <Keybuk> that's what I mean :p
[01:34] <Keybuk> ahh, no, I understand now
[01:34] <Keybuk> the table has to be the same size in memory
[01:34] <Keybuk> the 32-bit one is made up of an array of 4-byte unsigned ints
[01:34] <Keybuk> the 64-bit one is made up of an array of 8-byte unsigned hung ints
[01:34] <Keybuk> the 64-bit one has half the number of elements
[01:34] <Keybuk> right
[01:35] <lamont> Keybuk: strtol() is your friend for ripping a number from a string...
[01:35] <lamont> Keybuk: sure they're not 'long' s?
[01:36] <lamont> ah, nm on the longs thing
[01:36] <Keybuk> no, definitely unsigned int on i386 and unsigned long long on amd64
[01:36] <lamont> uh, unsigned long long has a good chance of being 128 bits on amd64...
[01:37] <Keybuk> no, just 64
[01:39] <Keybuk> right, time to add some autofoolery to check the platform
[01:39] <lamont> well, I guess that's good.
[01:39] <lamont> but it's really unsigned long, no?  (long is 64-bits there...)
[01:39] <Keybuk> is it?
[01:40] <Keybuk> is that true for all 64-bit platforms?
[01:41] <lamont> I know that amd64 and ia64 are both LP64.  I expect that ppc64 is too
[01:41] <lamont> that is, longs and pointers are 64-bit.
[01:41] <lamont> int is some random size between 16 and 64 bits
[01:41] <Keybuk> I think I might just do what depmod does and use long long
[01:41] <lamont> although 64-bit int's would make you ILP64, which they toyed with for ia64 at one point.
[01:41] <jbailey> lamont: I think GNU Systems are guaranteed to have at least a 32bit int 
[01:42] <lamont> jbailey: that is, int [01:42] <jbailey> >=
[01:43] <lamont> Keybuk: you could just use uint64_t...
[01:44] <lamont> Keybuk: which needs #include <sys/types.h>, iirc
[01:45] <lamont> hrm... "hppa RC bugs"  that's kinda amusing
[01:46] <Nafallo> lamont: oh?
[01:46] <ogra> Keybuk, http://gcc.fyxm.net/summit/2003/Porting%20to%2064%20bit.pdf
[01:46] <ogra> ;)
[01:47] <lamont> mdz: thoughts on http://people.ubuntu.com/~lamont/xorg.diff - fixes hppa, with minimal change/risk to everyone else
[01:47] <ogra> there are the mappings for variables for 32 vs 64 bit arches :)
[01:47] <lamont> ogra: and sys/types.h gives you the sized types you need - what do you care what the underlying name is...
[01:48] <ogra> true
[01:48] <lamont> but Keybuk has a thing for auto*, you see...
[01:48] <mjg59> Now, that is *really* odd
[01:48] <lamont> mjg59: ??
[01:48] <mjg59> If I start irattach after a suspend/resume cycle, this laptop freezes
[01:48] <ogra> but i found the doc quite helpful anyway for the Cxx transition :)
[01:48] <lamont> ogra: yes
[01:48] <lamont> mjg59: score!
[01:48] <Nafallo> mjg59: nice :-)
[01:49] <mdz> lamont: why is it that all these SCC fixes are coming in at the 11th hour?
[01:49] <lamont> mdz: because after midnight would be too late?
[01:49] <lamont> truthfully, work has had me hammered on a couple of assigned tasks that I finally got a bit of freedom from this week
[01:49] <lamont> based purely on the deadline
[01:50] <lamont> the xorg patch makes ubuntu-desktop installable on hppa
[01:50] <mdz> lamont: it's just that fabbione is doing the same thing with sparc :-P
[01:50] <lamont> I don't know anything about sparc.
[01:51] <ajmitch> lamont: are you able to kick a rebuild for scummvm on powerpc? bugreports indicate the problem behind it failing was fixed
[01:51] <lamont> ajmitch: universe?
[01:51] <ajmitch> yeah
[01:51] <mdz> lamont: did you do a full test build on non-hppa?
[01:51] <ajmitch> gs was segfaulting last time it tried to build
[01:51] <lamont> mdz: built on i386
[01:52] <lamont> I suppose I should do amd64 and ppc for a full 'yes', though.
[01:52] <mdz> lamont: with all the same binaries produced as in -72?
[01:52] <mdz> one arch should be enough
[01:52] <lamont> mdz: given xorg's manifest, I expect that's a yes.
[01:52] <lamont> the scripts/vars is a cut/paste from debian/control in -72
[01:53] <lamont> but I'll go do a slew of interdiffs before I consider making it -73 and uploading.
[01:53] <mdz> lamont: I mean all the same .debs, since that's what you changed
[01:53] <lamont> mdz: actually, all I changed was the dependencies of xserver-xorg
[01:54] <mdz> lamont: if you verify that it's a no-op for one other arch, I'm fine with it, though I'd prefer at least one generic bugfix (there's a long list to choose from)
[01:54] <lamont> since hppa doesn't deliver xserver-xorg-input-*
[01:54] <mdz> oh, so you did.  the changelog should say that then ;-)
[01:54] <lamont> mdz: I'll review the OB-bug fix list
[01:55] <lamont>   * xserver-xorg Dependencies on xserver-xorg-input-* need to be optionally 
[01:55] <lamont>     removed for some architectures.  (Specifically, hppa has never delivered 
[01:55] <lamont>     them, and that made xserver-xorg uninstallable.)
[01:58] <lamont> mdz: looking at the major bugs, I'm not really comfortable with the risk level of having me try to fix them...
[01:59] <lamont> I'll stare at it some, but in the end, I'll try to piggy back my upload onto one of daniels's, or just upload it before bedtime tonight
[01:59] <mdz> lamont: send it to daniels I guess; he's supposed to make an upload today anyway
[01:59] <lamont> right
[01:59] <Nafallo> hehe
[01:59] <Nafallo> one xorg a day... ;-)
[02:00] <Nafallo> ...keeps the bandwidth away :-P
[02:00] <Nafallo> but then again. it's by far more better than OO.o2 :-P
[02:01] <Nafallo> should we split that for dapper? ;-)
[02:01] <lamont> Nafallo: esp since the oo.o* uploads tend to be new upstream releases
[02:01] <lamont> Nafallo: is it split upstream?  I thought it was monolithic
[02:02] <Nafallo> lamont: I wouldn't know :-). I just saw xorg's impact on my bandwidth ;-).
[02:02] <lamont> Nafallo: dapper xorg will be much nicer
[02:02] <lamont> since it'll be split up upstream, instead of in the .diff.gz
[02:03] <Nafallo> :-)
[02:03] <Nafallo> 7.0 I guess?
[02:05] <lamont> right
[02:05] <lamont> well, 6.8.99
[02:06] <Nafallo> hehe, when did 6.0 came out? :-)
[02:09] <elmo> lamont: did you report the smooth engines thing?
[02:09] <lamont> I don't remember if I filed a bug on it, but ISTR I did... seb was going to upload something.
[02:10] <elmo> anyone feel up for a trivial merge to fix an RC bug?  if so, please deal with dict-gcide - there's a corresponding bug in bugzilla
[02:14] <lamont> elmo: ok.
[02:15] <lamont> on another note, if someone wants to try to duplicate 16519 from stock hoary, I'd appreciate it
[02:15] <lamont> sadly, it looks like 16609 is a dup, which would make it kinda reproduced.
[02:16] <Lathiat> bugzilla or malone?
[02:16] <jdong> Lathiat: bugzilla
[02:17] <jdong> lamont: I've seen similar errors in the past before... usually successions of dist-upgrades and -f installs will fix it
[02:17] <lamont> elmo: please sync dict-gcide_0.48-4.1.  kthxbye
[02:18] <lamont> elmo: the ubuntu change is subsumed by debian - but let me double check that
[02:18] <elmo> lamont: ubuntu changes ok to override? 
[02:18] <elmo> k
[02:18] <Nafallo> lamont: I just reached the same conclusion :-)
[02:19] <Nafallo> fwiw :-)
[02:19] <lamont> elmo: from the change log, -4.1 has the fix we have, and the fix we need.
[02:19] <lamont> but the build should be quick
[02:21] <lamont> elmo: go for sync
[02:22] <elmo> lamont: thanks
[02:22] <lamont> np
[02:34] <jdong> grr, just had nautilus-cd-burner segfault on me
[02:40] <blueyed> I've just dist-upgraded to breezy but still cannot set dma on my sata hdd. It still says "HDIO_SET_DMA failed: Inappropriate ioctl for device". This is a nForce4 chipset mobo.
[02:41] <jdong> blueyed: (1) wrong channel (2) hdparm is for setting IDE parameters
[02:42] <jdong> blueyed: SATA DMA is enabled through BIOS (though theoretically it shouldn't be off period)
[02:44] <blueyed> Thanks, jdong.
[02:45] <jdong> blueyed: in the future, ask in #ubuntu.. there's just as many breezy fanatics in there. This is really for developers working on Breezy or stuff of that nature
[02:45] <jdong> (not to play moderator outside my domain...)
[02:46] <ogra> jdong, but you are right :)
[03:16] <ogra> mdz, you said you'll be out of town ? how log ? 
[03:16] <ogra> *long
[03:16] <mdz> ogra: I mailed allhands
[03:17] <mdz> ogra: and updated StaffCalendar
[03:17] <mdz> ogra: (through Sunday)
[03:17] <ogra> oki
[03:20] <lamont> mdz/elmo: I was wondering how anastasia output with hppa is looking, and if it can come into the fold....
[03:21] <lamont> mdz: and I'd like to do an ubuntu-meta upload sometime this weekend to finish fleshing out hppa/sparc packages... with a discussion with Kamion planned if it changes anything in the other architectures
[03:21] <lamont> thoughts?
[03:22] <mdz> ubuntu-meta is no problem
[03:22] <mdz> I don't know how anastacia looks; I'm not sure that elmo germinates for hppa currently
[03:22] <lamont> hppa is not currently generated
[03:23] <lamont> which makes it un-debootstrappable (libgcc2) and no CD's (palo)
[03:23] <lamont> http://buildd.mmjgroup.com/hppa-hacks is the collection of universe packages that will move into main if hppa is germinated
[03:23] <lamont> (the others are build-deps)
[03:23] <lamont> hppa64 cross-compiler, and expect-tcl8.3
[03:24] <lamont> the latter works around a really annoyingly difficult to reproduce/find kernel bug where sometimes expect (and others, to be sure) fail to correctly wake up on child death...
[03:24] <lamont> _why_ tcl8.4 hits it and tcl8.3 doesn't is, well, part of the mystery
[03:27] <Keybuk> lamont: the problem is that I need the code to be different on i386 to amd64
[03:27] <Keybuk> on i386 it has to be an array of 8 unsigned ints
[03:28] <Keybuk> and amd64 an array of 4 unsigned long longs
[03:29] <lamont> Keybuk: ew
[03:29] <Keybuk> otherwise I would just do this the easy way
[03:30] <Keybuk> yes, you can tell this code is dealing with the kernel
[03:30] <Keybuk> "we could do it the same way for each architecture, but we won't"
[03:32] <lamont> "...that would be cheating"
[03:33] <Keybuk> they're not that bad though
[03:34] <Keybuk> I mean, it's not as if the kernel has a different entry point table for each architecture, is it?
[03:34] <lamont> mdz: 16526 may want a new adduser, or may be NOTWARTY - I'll dig through it some tonight
[03:55] <Keybuk> meh, that still segfaults *sigh*
[03:58] <ogra> night all
[03:58] <lamont> g'night ogra
[03:59] <Keybuk> yes, this string is so sadistic, even gdb fears it:
[03:59] <Keybuk> parse_array (
[03:59] <Keybuk>     str=0x7fbffffdf3 "402000000:3802078f840d001:f2ffffdfffefffff:", 'f' <repeats 15 times>, "e", array=0x7fbffff960, array_len=8) at util.c:136
[03:59] <bddebian> heh
[04:03] <Keybuk> maybe it's not this code at all
[04:03] <Keybuk> this code is very bounds-safe
[04:03] <Keybuk> it's probably been busted all this time, but the worst it would've done is give people joysticks a little too often
[04:03] <Keybuk> so it's good that I've fixed it, but it's something else
[04:04] <Keybuk> lamont: don't suppose you know what gcc's type promotion is on amd64?
[04:04] <lamont> Keybuk: I don't, but I bet both doko and jbailey do
[04:04] <Keybuk> I bet they're both asleep
[04:05] <Keybuk> the only other thing I can see in this code which is "unusual" is that I use ...
[04:05] <Keybuk> so maybe it's promoting the type differently on amd64 to i386
[04:05] <lamont> Keybuk: I haven't heard anything from jbailey in oh, say about 11 minutes
[04:06] <lamont> Setting up gstreamer0.8-jpeg (0.8.11-0ubuntu4) ...
[04:06] <lamont> Fontconfig error: Cannot load default config file
[04:06] <lamont> seb128: ^^^ what's up with that?
[04:07] <lamont> Setting up linux-image-2.6.12-9-itanium-smp (2.6.12-9.18) ...
[04:07] <lamont> FATAL: Could not load /lib/modules/2.6.10/modules.dep: No such file or directory
[04:07] <lamont> WTH??
[04:08] <lamont> ah, doh.
[04:08] <lamont> the _machine_ is running 2.6.10
[04:09] <Keybuk> whoah
[04:09] <Keybuk> so it's definitely promotion-related
[04:09] <Keybuk> I just got 545460846593 from va_arg not 1
[04:09] <lamont> make it a union.
[04:10] <Keybuk> ahh, I bet I know what it is
[04:10] <Keybuk> size_t's going to be 64-bit, isn't it
[04:10] <Keybuk> and enum is 32-bit
[04:10] <lamont> yepo
[04:12] <Keybuk> aha!
[04:12] <Keybuk> yes, that fixed it
[04:12] <bob2> Keybuk: what are you doing, anyway?
[04:12] <bob2> the hct kernel api?
[04:12] <Keybuk> so it wasn't the bogus kernel structure after all, it was different type promotion combined with va_args
[04:12] <Keybuk> but along the way, I've fixed the kernel structure so it'll actually dtrt on amd64 <g>
[04:13] <Keybuk> bob2: amd64 hardware detection stuff
[04:13] <bob2> ah
[04:14] <Keybuk> right
[04:15] <Keybuk> let's check I didn't break i386 in the process
[04:18] <Keybuk> nope, is all good
[04:31] <lifeless> Keybuk: thats a familiar bug
[04:32] <lifeless> Keybuk: tla hit that, oh every time tom changed stuff that uses varags
[04:37] <Keybuk> yeah, it was pretty much the second thing I looked for
[04:39] <Keybuk> hmm
[04:39] <Keybuk> what's a good group to put mice permission in?
[04:40] <desrt> elmo; ping :)
[04:40] <jbailey> plugdev?
[04:41] <jbailey> That way potentially someone could add some sort of additional hid and let it get used on the fly/
[04:47] <Keybuk> jbailey: that's for removable disks only, isn't it?
[04:47] <Keybuk> it seems like we stick everything in plugdev, and I wonder whether that's wise
[04:48] <lamont> desrt: I'm gonna go out on a limb and bet that he's not awake at 03:47
[04:49] <Keybuk> jbailey: also could you comment on #14669 for me
[04:49] <Keybuk> it looked like you were expecting that result and wondered if you could shine some light on it
[04:54] <jbailey> Keybuk: The kernel doesn't provide us any notification of when it's done scanning the USB bus, so we do a sleep 2 right now.
[04:54] <jbailey> For a USB Harddrive, it needs to be more like sleep 10, it seems.
[04:54] <jbailey> I've asked various kernel folks, and haven't found an answer on how to know when it's safe to proceed.
[04:55] <lamont> jbailey: because why would you ever care to know when udev is done... you just need to be told that it's starting...
[04:55] <lamont> yeah, that'll work

[04:55] <jbailey> It's not even udev.  I want to know when the bus scan is done.
[04:56] <lamont> ah, ok
[04:56] <jbailey> udev can sit on it and rotate for all I care.  At least with it I can trust that it will handle things as soon as possible or block until it's time.
[04:56] <bddebian> hehe
[04:56] <jbailey> (And cause various race conditions, but that's an aside)
[04:56] <jbailey> SOrry, did I just date myself? =)
[04:56] <lamont> jbailey: that aside is the source of my <bitter-mode> reaction.
[04:57] <lamont> sit-n-spin... ISTR that's trying to make a comeback
[04:57] <bddebian> My kids have one
[04:57] <jbailey> As is Air Supply, apparently.
[04:57] <jbailey> At my last job, a coworker went to see them in concert.
[04:57] <jbailey> It's my childhood, all over again.
[04:57] <lamont> welcome to the 70's
[04:58] <bddebian> Air Supply?  No freakin' way :-)
[04:58] <lamont> but _STILL_ no clackers
[04:58] <jbailey> Ooo, that would take us back to when Kenny Rogers sang *good* country music. =)
[04:58] <bddebian> heh
[04:58] <Arrogance> Reuben James?
[04:59] <Arrogance> Reuben Jane
[04:59] <Arrogance> sorry
[04:59] <lamont> http://cgi.ebay.com/KLIK-KLAK-CLACKERS-BNIP_W0QQitemZ6002743746QQcategoryZ30QQssPageNameZWDVWQQrdZ1QQcmdZViewItem
[05:00] <lamont> Keybuk: find me a couple pair of those and I'll buy you a pint or 3
[05:00] <lamont> something tells me they're harder to find in the US...
[05:00] <Keybuk> lamont: wtf are they?
[05:01] <lamont> Keybuk: THE KILLER toy of my childhood
[05:01] <lamont> two glass balls on the end of a string... and you smash them back and forht
[05:02] <bob2> crap
[05:02] <lamont> hrm.. I can get them in the US from one site for a mere 35 USD
[05:02] <bob2> "short read" kernel errors always mean physical disk fuckage, right?
[05:03] <lamont> s/glass/heavy acrylic/
[05:04] <lamont> Keybuk: I'll look around and see what I can get them for here - I might be able to beat ebay
[05:05] <jbailey> Almost like the land-the-ball-in-the-cup game that I was also never able to do. =)
[05:05] <jbailey> But I'm the kid who destroyed like 3 of the plastic tee's in T-ball in elementary school.
[05:05] <jbailey> Noone was allowed a pinch hitter but me. =)
[05:06] <lamont> from a random forum: "They were called clackers, if memory serves. The memory isn't too good, though, ever since I fractured my skull playing with the stupid thing.
[05:06] <lamont> "
[05:06] <jbailey> Turned out my problem that I was scared of the ball exploding or something, so I closed my eyes everytime I swung the bat.
[05:08] <lamont> *FOR ADULTS ONLY. These were taken off the market because of injuries they caused to kids. Please: for collectible use only. .or adults! 
[05:08] <bddebian> hehe
[05:08] <bddebian> Like Jarts
[05:09] <desrt> lamont; a reasonable assumption :)
[05:15] <bddebian> Gnight lamont 
[05:16] <LaschW> Is there a Chance to find a maintainer for Eiciel to get Eiciel into Breezy? Eiciel is a posix ACL editor for Gnome. IMO well done and fits seamless into Nautilus.
[05:17] <bob2> #ubuntu-motu
[05:17] <LaschW> Eiciel only needs a rebuild on breezy. There are Debian Sarge deb's which are linked against an c102 deb from sid.
[05:18] <LaschW> bob2: #ubuntu-motu is the place to ask for?
[05:18] <bob2> yes
[05:18] <LaschW> bob2: Thank you. 
[05:19] <lamont> LaschW: so it has circular build-deps or something?
[05:20] <lamont> LaschW: or is it that it entered debian post-UVF?
[05:21] <ajmitch> entered debian only a couple of weeks ago, from what I can see
[05:21] <lamont> that'd do it
[05:22] <lamont> anyway, home
[05:23] <fabbione> morning
[05:25] <Keybuk> gnargh.  another source package, another "guess the rule to unpack the tarball and apply the patches" game
[05:25] <bddebian> heh
[05:25] <bddebian> Heya fabbione 
[05:26] <fabbione> ya
[05:26] <jbailey> Keybuk: Yeah, you should beat on the dpkg maintainer to get wig&pen going... ;)
[05:27] <Keybuk> meh, this caps lock bug is silly
[05:27] <Keybuk> there's a whole discussion to get the patch right which ends in an upload
[05:27] <Keybuk> and I can't find the patch in that upload

[05:35] <fabbione> hey Keybuk .. any news on that little apps we were talking about?
[05:37] <Keybuk> which?
[05:38] <fabbione> mdz: was it a good ride?
[05:38] <fabbione> Keybuk: F1
[05:38] <Keybuk> ah yes, not yet
[05:38] <mdz> fabbione: so far the fix/workaround seems to be successful
[05:38] <Keybuk> still got another week before the race
[05:39] <mdz> I can now use my sound device without irqpoll
[05:39] <fabbione> Keybuk: ehhee
[05:39] <fabbione> mdz: cool
[05:40] <Keybuk> have I ever mentioned how much I love apps that don't even compile if you enable debugging?
[05:40] <fabbione> Keybuk: ahaha
[05:40] <fabbione> that's because they don't need debugging 
[05:42] <fabbione> mdz: do you think 16661 is worth fixing?
[05:42] <fabbione> imho it can wait dapper..
[05:46] <Keybuk> oh dear, this only seems to work because the broken code is usually optimised out
[05:46] <Keybuk> or something
[05:47] <mjg59> mdz: acpi-support is in good shape now - pretty much everything that was on my todo list is done
[05:47] <mdz> mjg59: hurrah
[05:47] <mdz> mjg59: given that we need to tighten down on the kernel now, are there any patches we should try to squeeze in on that side?
[05:48] <mdz> fabbione: not for breezy, no
[05:48] <mjg59> mdz: I've hit one kernel problem, but I haven't figured out /what/ and we can work around it, so it can be ignored
[05:48] <mjg59> mdz: The only other thing is that it's possible that the sata IDE passthrough patch is broken
[05:49] <mjg59> I don't seem to be able to get hdparm to do anything useful to SATA devices
[05:49] <mjg59> (noticed this while going through acpi-support bugs)
[05:49] <mdz> mjg59: 16646 perhaps?   sounds safe enough; i've run into that once or twice myself
[05:49] <mjg59> Oh, that's a userspace issue
[05:49] <mjg59> We can extend the timeout before loading the module and then drop it back to normal
[05:50] <mjg59> mdz: Have you got access to a SATA system?
[05:50] <mdz> mjg59: not personally, no
[05:50] <mjg59> Hm.
[05:50] <mdz> my test machines are all ATA and my desktop SCSI+ATA
[05:51] <mjg59> Mixed is fine
[05:51] <mjg59> Oh, sorry, SCSI+ATA
[05:51] <infinity> mjg59 : Is this is "HDIO_DRIVE_CMD(null) (wait for flush complete) failed: Inappropriate ioctl for device" thing you're talking about?
[05:51] <mjg59> infinity: Yeah
[05:51] <mjg59> Hmm. Actually, maybe we don't have that patch at all. I was under the impression we did.
[05:51] <mjg59> Shit.
[05:53] <fabbione> mjg59, jbailey, jdub: 16663
[05:53] <fabbione> this must get fixed yesterday
[05:54] <mjg59> The usplash thing should be fixed
[05:54] <jbailey> Wow, looking that ticket up managed to kill ephy.
[05:55] <infinity> mjg59 : Is it possible to ask the IDE drivers to shut up about re-probing ports?  (or lower the severity of that printk?)
[05:55] <mjg59> infinity: Oh, probably, yes
[05:55] <infinity> mjg59 : Ever since we got ahci loading, my laptop now prints 4 lines of annoying junk before boot.
[05:56] <mjg59> Ah, so that's what it's from
[05:56] <infinity> (Because I then go on and load ata_piix afterward, I assume)
[05:56] <mjg59> What's the precise error?
[05:56] <infinity> [4294672.575000]  ide0: I/O resource 0x1F0-0x1F7 not free.
[05:56] <infinity> [4294672.575000]  ide0: ports already in use, skipping probe
[05:56] <infinity> [4294672.575000]  ide1: I/O resource 0x170-0x177 not free.
[05:56] <infinity> [4294672.575000]  ide1: ports already in use, skipping probe
[05:57] <mjg59> infinity: Yeah, just knock up a patch to drop the KERN_ERR in ide-probe.c to KERN_INFO
[05:57] <infinity> Actually, not sure if it's ahci's fault, the timing in dmesg doesn't make it right for that... But the date when I statred seeing the errors seems to coincide.
[06:00] <mdz> Keybuk: did you track down or file a bug about that metacity thing?
[06:00] <mdz> I can't see which window is selected when alt-tabbing at all
[06:00] <infinity> mjg59 : Does ACPI retrigger an ide bus scan?
[06:00] <mjg59> infinity: Not as far as I know
[06:01] <infinity> mjg59 : In dmesg, I get this order: ahci fails to load, ata_piix loads successfully, ACPI bus registered, then the ide I/O probe errors.
[06:01] <mjg59> infinity: Hrm. Weird.
[06:01] <infinity> Oh, I can't read.
[06:01] <mjg59> infinity: Oh - it'll probably be ide-generic loading
[06:01] <infinity> That's "ACPI: bus type ide registered", from ide-generic, yeahp.
[06:01] <mjg59> mdz: Ok. We're missing the ata-passthru patch
[06:01] <mjg59> Which I thought we had, but, hrmph.
[06:02] <infinity> Yeah, that settles it, then.  If we want to unconditionally attempt ide-generic, we need to lower the severity of that.
[06:02] <mjg59> infinity: It's a two line patch. I'm already in bed, so I'm not doing it :)
[06:02] <infinity> Yeah, I'll do it later today and pass it to BenC.
[06:02] <infinity> I'll put your name on it. ;)
[06:02] <mjg59> Heh
[06:02] <fabbione> infinity: just push it here.. i can commit to baz
[06:02] <infinity> (Cause, you've got more street cred, yo)
[06:03] <mjg59> mdz: So we don't have hdparm support for SATA disks, which means we can't spin them down
[06:03] <mjg59> Whether this matters or not, well.
[06:03] <mdz> mjg59: the complementary question would be how risky the patch is
[06:04] <mjg59> mdz: It's in common use, but it's a 21K diff
[06:04] <mdz> I'm not sure we can risk it
[06:04] <mjg59> But that's almost entirely adding new ioctls, so it's code that won't be run unless hdparm is run on the device
[06:05] <mdz> can someone mail me a copy?
[06:05] <mjg59> http://www.kernel.org/pub/linux/kernel/people/jgarzik/libata/old/2.6.12-git4-passthru1.patch.bz2
[06:07] <mdz> mjg59: not entirely horrifying
[06:07] <mjg59> Alternatively, it may be possible to hack something up with sdparm
[06:07] <mjg59> But I really don't know about that
[06:07] <mdz> mjg59: this isn't a regression from hoary, is it?
[06:08] <mjg59> Nope
[06:08] <fabbione> mdz: nope.. no regression
[06:08] <mdz> ok, I prefer to leave it alone then
[06:08] <mjg59> Ok
[06:08] <mjg59> I'll see what I can do with sdparm
[06:09] <Amaranth> yay, i got my breezy system working again
[06:09] <Amaranth> just had to know the math to find a backup superblock
[06:09] <Amaranth> too bad i've got this crappy windows only dialup
[06:12] <Keybuk> anyone know what the KDGKBMODE ioctl is supposed to do?
[06:14] <Keybuk> no, ignore me, I'm on crack ... helps if I put the printf after the ioctl <g>
[06:17] <Amaranth> this is probably more of an #ubuntu question but does anyone know if there is any support for conexant winmodems in the ubuntu kernel?
[06:18] <Keybuk> right, so I fixed the caps lock bug
[06:18] <Keybuk> but I don't really understand why
[06:18] <Keybuk> heh
[06:19] <fabbione> Amaranth: there is a bug open for them.. but i don't think they were included in breezy
[06:19] <Amaranth> wtf, the linuxant drivers for them aren't Free?
[06:20] <fabbione> that wouldn't surprise me either
[06:20] <Keybuk> right, so, err, does anyone have an exotic keyboard ?
[06:20] <Keybuk> I want to see whether I've fucked everything up, or made an inspired guess
[06:20] <fabbione> Keybuk: exotic to what level?
[06:20] <Keybuk> fabbione: non-latin keys on it
[06:21] <fabbione> nope.. dk-latin1 here
[06:21] <mjg59> Amaranth: No,  because they're not under a license that allows us to distribute them
[06:21] <mjg59> They're not Free, and they're not free
[06:21] <Amaranth> yes, i see that now, that sucks
[06:23] <bob2> am I imagining things, or is it impossible to unlock the screen on the live cd?
[06:23] <Keybuk> bob2: "ubuntu"
[06:23] <lifeless> its undocumented
[06:23] <Keybuk> iirc
[06:23] <lifeless> not impossible
[06:24] <lifeless> Keybuk: be nice if the screensaver told us that;0
[06:24] <bob2> lifeless: well, I can't seem to change to a vt, either
[06:25] <bob2> hm, not possible if you didn't set a password, afaict
[06:29] <Keybuk> oh, now, that's quite entertaining
[06:29] <Keybuk> with caps lock on, even the number keys get capitalised
[06:30] <Keybuk> no, I broke that
[06:30] <Keybuk> oops
[06:30] <infinity> fabbione : You have dpatch mail.
[06:30] <fabbione> infinity: gotcha
[06:30] <Keybuk> I did fix the _original_ bug though :p
[06:31] <Keybuk> ok, no, that makes no sense
[06:32] <Keybuk> according to this, I didn't actually change anything
[06:32] <fabbione> infinity: can i just queue it at the end of 00list ?
[06:33] <infinity> fabbione : I assume so, I hate dpatch, so I have no idea what it did.
[06:33] <infinity> fabbione : Let me add it to 00list and test-apply it.
[06:34] <fabbione> infinity: it's ok.. i am doint it locally
[06:34] <Amaranth> err, is  gnome-app-install 0+20050927 a native package?
[06:35] <infinity> Seems to go fine if I put it in 00list after all the other drivers-ide patches (to keep things tidy)
[06:35] <Keybuk> ahh, I understand what I did
[06:35] <Keybuk> I turned every random character into "can take capslock"
[06:35] <Keybuk> meh
[06:41] <ajmitch> hm
[06:41] <ajmitch> it's interesting coming across bugs in malone that have been in main since january or so
[06:41] <fabbione> infinity: committed
[06:44] <lifeless> Keybuk: haha
[06:46] <Keybuk> meh, I need the dumpkeys off someone european
[06:50] <lifeless> ask ddaa
[06:50] <ajmitch> what's the main upload policy at the moment?
[06:50] <lifeless> hes european dvorak
[06:55] <Keybuk> is he awake?
[06:57] <infinity> ajmitch : Itty bitty bugfixes for obvious bugs are probably still okay without approval, anything bigger should probably go through mdz/Kamion for a quick once-over.
[06:57] <ajmitch> infinity: this is fairly minor fix to a postinst for xbel-utils
[06:57] <infinity> ajmitch : That would fit category 1 then. :)
[06:58] <ajmitch> I'll see if I'm in the keyring yet then.. :)
[06:58] <infinity> ajmitch : The more hideous/obvious the bug, and the more simple the fix, the more likely that it's explicitely approved.
[07:06] <infinity> If you guys need me, queue it up in mail or /msg, I'll be back in an hour or two.
[07:07] <ajmitch> if someone could check & upload what is in http://ajmitch.dyndns.win.co.nz/debuild/ubuntu/tmp/uploads/ it would be appreciated :)
[07:07] <ajmitch> bbiab
[07:28] <fabbione> daniels: people.u.c/~lamont/xorg.diff <-
[07:29] <daniels> fabbione: yes, I saw his email
[07:30] <daniels> was that all?
[07:31] <fabbione> daniels: from him yes..
[07:31] <fabbione> i am trying to fix 14973
[07:36] <daniels> fabbione: it'll be the matrox driver rounding down in one place when it does its default mode, but then not rounding down in mode validation
[07:37] <fabbione> daniels: i am checking the DCC code of the mga driver, but i get the feeling that it doesn't recognize that a certain resolution set is not valid
[07:38] <fabbione> because the mode is validated later 
[07:38] <fabbione> the (WW) comes from libdcc
[07:38] <fabbione> and i2c
[07:38] <fabbione> not from the mga driver
[07:39] <daniels> radeon is about the only driver with its own mode validation
[07:39] <daniels> the rest comes from common/xf86Modes.c and friends
[07:52] <lamont-away> drwxr-xr-x  132 hpojlp lp     8192 2005-09-29 23:52 /
[07:52] <lamont-away> I wonder if that's still the case on a fresh breezy install with hpoj
[07:53] <lamont-away> or was it hpijs
[07:55] <fabbione> daniels: the problem looks like that the mga driver is using 2 different polling methods on the monitor
[07:56] <fabbione> daniels: the first one is the normal ConfiguredMonitor = vbeDoEDID(pVbe, NULL);
[07:56] <fabbione> that seems to return proper values
[07:56] <fabbione> but later it does a MGAdoDCC
[07:56] <fabbione> using libddc and i2c
[07:56] <fabbione> where "somehow" fails
[08:00] <lamont-away> fabbione: I was thinking we'd see if the breezy version of adduser had the shadow bug or not, before we closed that...
[08:02] <daniels> fabbione: well, first it tries through the VBIOS, then through I2C.  sounds about normal.
[08:03] <fabbione> lamont-away: we do enable shadow by default.. also mdz closed similar bugs already
[08:04] <lamont-away> yeah, np
[08:06] <fabbione>     /*
[08:06] <fabbione>      * xf86ValidateModes will check that the mode HTotal and VTotal values
[08:06] <fabbione>      * don't exceed the chipset's limit if pScrn->maxHValue and
[08:06] <fabbione>      * pScrn->maxVValue are set.  Since our MGAValidMode() already takes
[08:06] <fabbione>      * care of this, we don't worry about setting them here.
[08:06] <fabbione>      */
[08:06] <HWolf> Who is responsible for the ubuntu-artwork?
[08:06] <fabbione> ^^^this smells so badly
[08:06] <daniels> fabbione: so check out the difference between sync validations in MGAValidMode and xf86ValidateModes (common/xf86Modes.c)
[08:13] <robitaille> HWolf,  that would be jdub
[08:14] <HWolf> jdub, you've got a bug on your hands. new artwork is blocky and ugly. :P
[08:16] <dholbach> good morning
[08:16] <HWolf> jdub, and it installed the 4:3 version by default on my 16:10 widescreen.
[08:17] <jdub> that's unavoidable
[08:17] <jdub> but a respin to soften up the noise was always pretty likely
[08:18] <HWolf> jdub, I had the widescreen version installed, so have it replaced by the new widescreen version should be possible.
[08:19] <HWolf> *having
[08:19] <HWolf> Anyhow, got to run, math workshop.
[08:19] <jdub> if you had selected the ws version, this one would simply replace it
[08:19] <HWolf> it didn't, apperantly
[08:19] <dholbach> does anybody use a kvm switch (two boxes - one monitor/mouse/keyboard)?
[08:20] <fabbione> daniels: you are right, the mga driver doesn't make anything fancy by itself, but it uses the all the standard xf86modes.c even to prune the invalid resolutions.. so i wonder if it is actually a bug in xf86modes.c
[08:23] <jdub> W: GPG error: http://katia breezy Release: The following signatures were invalid: BADSIG 40976EAF437D05B5 Ubuntu Archive Automatic Signing Key <ftpmaster@ubuntu.com>
[08:23] <jdub> whee!
[08:23] <Amaranth> ook, that's not good
[08:28] <fabbione> AHHHHHHHHHH
[08:29] <Keybuk> I think the isapnp subsystem must've been written by the same great people that gave us the input subsystem
[08:29] <Keybuk> for each module you can get the vendor and device id of each device it supports
[08:30] <Keybuk> it's just a shame that there's no way to get the vendor or device id of any connected device
[08:30] <Keybuk> "if I told you what the vendor/device pair were, you could use this file to find the module you need to load"
[08:34] <mjg59> Keybuk: There's PNPdump
[08:34] <mjg59> And for pnpbios devices, they're exposed in /sys/bus/pnp/devices
[08:34] <mjg59> Oh, wow
[08:35] <mjg59> There's actually an almost moderate number of devices with entries in modules.isapnpmap now
[08:35] <Keybuk> mjg59: they're exposed, but not their vendor/device ids :p
[08:36] <Keybuk> the kernel only gives us it's strange PNPid string in the /id file
[08:36] <tepsipakki> why is ACPI_SLEEP not yet enabled in acpi-support?
[08:36] <mjg59> tepsipakki: Because there are many machines that it doesn't work on
[08:36] <tepsipakki> mjg59: ok, fair enough
[08:36] <mjg59> And giving people a "Crash my computer" button is not a good way of making friends
[08:37] <tepsipakki> mjg59: is there a way to make it enabled for known good machines?
[08:37] <mjg59> Keybuk: I have vague memories of there being a mapping from the strings to the pnp IDs
[08:37] <mjg59> tepsipakki: Yes. See ubuntu-devel
[08:37] <tepsipakki> oh my.. quite recent =)
[08:37] <Keybuk> mjg59: yeah, there's stuff in modalias
[08:38] <Keybuk> but really not much
[08:38] <Keybuk> 9 entries in modules.alias
[08:38] <Keybuk> compared to 226 in modules.isapnpmap
[08:38] <tepsipakki> mjg59: my stinkpad T23 works fine, I'll send you mail soon
[08:39] <mjg59> Keybuk: No, I mean ISTR that there's a mechanical transformation between the two
[08:40] <Keybuk> mjg59: wouldn't that be rather lossy?  the full vendor/device/function set for a pnp device can be quite long ... the ids are just PNPxxxx 
[08:40] <Keybuk> *shrug*
[08:40] <mjg59> Keybuk: The Vendor ID (EISA Product ID) is defined by the Plug and Play ISA specification as the combined two-byte, three-character compressed ASCII EISA ID, and the two-byte, model-specific number (32 bits total)
[08:40] <mjg59> If it's PNPxxxx then it's just a class ID, and you don't need to worry about it
[08:41] <mjg59> Well, not in the same way - there's a strict mapping of PNPxxxx to hardware classes. They're all serial ports or parallel ports or motherboards or something.
[08:41] <Treenaks> Keybuk: you're looking at the crystalsound mess?
[08:42] <Keybuk> mjg59: I still have to worry about them, because I need to be able to load modules for them
[08:44] <Keybuk> the theory is that the kernel is supposed to provide a "modalias" file under sysfs which maps to modules.alias
[08:44] <Keybuk> cause it knows perfectly well what modules go with what :p
[08:45] <jdub> mjg59: i so want to sit down with you and figure out why my laptop, docked, works completely differently wrt special keys and so on
[08:46] <mjg59> Keybuk: But you don't need to worry about the IDs in the same way. There's a very small static table for dealing with them.
[08:47] <Keybuk> mjg59: but there isn't that table, is kinda my point
[08:48] <mjg59> Keybuk: But it's very easy to write
[08:48] <mjg59> jdub: Nngh. I wish I knew.
[08:49] <mjg59> Keybuk: Sound cards and stuff won't have PNPxxxx ids
[08:50] <Treenaks> mjg59: some do
[08:50] <Treenaks> mjg59: oh wait, not PNPxxx
[08:50] <mjg59> tepsipakki: Works fine with ACPI, or does it use APM?
[08:51] <tepsipakki> mjg59: both, I've been using ACPI for a year now
[08:51] <mjg59> tepsipakki: Cool
[08:52] <tepsipakki> mjg59: but the hotkeys don't trigger it now.. used to work though, somehow the sleep.sh isn't run?
[08:52] <tepsipakki> logs show that the hotkey is pressed
[08:52] <Keybuk> mjg59: you mean things like PNPB003 (Sound Blaster 16) are figmets of someone's imagination? :p
[08:52] <mjg59> tepsipakki: Erm. Odd.
[08:52] <pitti> Good  morning
[08:52] <mjg59> Keybuk: Bah. They're supposed to be.
[08:52] <tepsipakki> similar to 16555
[08:54] <Keybuk> see, the way this is _supposed_ to work is that every device, for every subsystem under /sys/bus has a "modalias" file
[08:54] <Keybuk> that file contains a string that looks a bit like
[08:54] <jdub> "Ubuntu" is an ancient African word, meaning "I can't configure Slackware".
[08:54] <Keybuk> pci:v00001002d00004337sv00000E11sd0000005Abc03sc00i00
[08:54] <Keybuk> which is some magic that we don't have to care about
[08:54] <Keybuk> modules have a table of device information they support, which depmod can parse and turn into wildcard-matches for those magic strings which it puts in /lib/$uname/modules.alias
[08:54] <mjg59> Yes, but you don't have that
[08:54] <Keybuk> right, but we're supposed to have that :p
[08:55] <mjg59> So you're going to have to do it manually
[08:55] <mjg59> Which is, thankfully, possible with pnp
[08:55] <Keybuk> the problem is that there's nowhere to do it manually anymore
[08:55] <Keybuk> it seems the effort would be better spent fixing the kernel to actually produce a modalias for isapnp devices
[08:55] <mjg59> tepsipakki: Bah. Older Thinkpads seem weird.
[08:56] <tepsipakki> mjg59: some update broke it recently
[08:57] <Keybuk> given depmod has gone to all that trouble to make the wildcard matches for it
[08:58] <mjg59> Keybuk: isapnp_parse_id is the transformation from vendor string to letters
[08:58] <fabbione> dholbach: i use a KVM here
[08:58] <mjg59> In drivers/pnp/isapnp/core.c
[08:59] <dholbach> fabbione: does it happen to kill the mouse when you switch a couple of times?
[08:59] <Keybuk> mjg59: but we don't want those ones anymore, we want the ones that match things like:
[08:59] <fabbione> no
[08:59] <dholbach> fabbione: under windows it didnt move any more, under linux it felt as i had specified the wrong mouse protocol
[08:59] <dholbach> FUCK
[08:59] <Keybuk> alias pnp:cOPT0924dOPT0000dOPT0002* snd_opti92x_cs4231
[08:59] <dholbach> then i'll send that stupid thing back
[08:59] <Keybuk> (from modules.alias)
[08:59] <fabbione> dholbach: buy a good KVM :)
[09:00] <fabbione> you won't regret the expense
[09:00] <dholbach> fabbione: this d-link thing doesnt seem to be remotely a good one
[09:00] <mjg59> Keybuk: But how are you going to generate that without knowing that transformation?
[09:00] <Keybuk> mjg59: the kernel generates those
[09:01] <mjg59> Except it doesn't
[09:01] <Keybuk> right, my point is that it should
[09:01] <mjg59> Right
[09:01] <Keybuk> because most of the code is clearly there already, because it's generating them for modules, just not devices
[09:01] <mjg59> But we're in kernel freeze
[09:01] <Keybuk> yeah, I'm talking dapper
[09:01] <mjg59> Ok
[09:01] <Keybuk> for breezy I just stick my fingers in my ears and go "la la la isapnp la la la"
[09:01] <pitti> Hi Goonie 
[09:02] <Goonie> hi pitti
[09:02] <pitti> Goonie: I'm eagerly awaiting the workaround test :-)
[09:02] <mjg59> So there's no way of doing a slightly more awkward hotplug for the pnp case?
[09:02] <Goonie> pitti: I tested it half an hour ago... see the bugzilla entry...
[09:02] <Keybuk> mjg59: we already have /etc/modprobe.d/isapnp which is a list of known pnp ids to module names
[09:02] <Keybuk> it's quite small
[09:02] <mjg59> (Since it's blindingly easy to get from the information that the kernel provides to a vendor and device string, and you have a table between those and modules)
[09:02] <pitti> Goonie: ah, cool, thanks
[09:03] <mjg59> Keybuk: Uhm. Yes. It's missing huge amounts of stuff.
[09:03] <Keybuk> if you know any way to massively populate that, I'm all ears
[09:03] <fabbione> dholbach: i have a DANBIT SW-KEYMON8A
[09:03] <fabbione> dholbach: but it's an expensive one
[09:03] <mjg59> Keybuk: Take modules.isapnpmap and mechanically transform it
[09:03] <dholbach> fabbione: i will watch out for it
[09:03] <mjg59> The algorithm required is in the code I pointed you at
[09:04] <Keybuk> mjg59: what do we do with the function bits?
[09:04] <mjg59> Keybuk: Ignore them?
[09:04] <mjg59> The drivers will set those up themselves
[09:04] <Keybuk> those seem to be the active bits?
[09:04] <Keybuk> unless I'm misinterpreting this file
[09:05] <mjg59> The string is built from cardvendor and carddevice
[09:05] <mjg59> That's enough to give you the module you want
[09:06] <mjg59> Hrm. I believe, anyway. Let me check that again.
[09:06] <mjg59> Oh, right, I see what you mean
[09:06] <Keybuk> I think those "function" bits are actually the device ids, you see
[09:06] <mjg59> Yeah
[09:06] <Keybuk> or maybe they're all pairs of vendor/device
[09:07] <mjg59> I'm sure I had a copy somewhere
[09:08] <dholbach> thank you fabbione 
[09:08] <pitti> infinity: ah, tbird 1.0.7 is out
[09:08] <fabbione> dholbach: no problem
[09:08] <infinity> pitti : Cool, I shall start getting all the uploads ready, then.
[09:09] <infinity> pitti : I assume we need to audit the diff for warty/hoary?
[09:09] <pitti> infinity: yes, the diff shouldn't be too big; it looked sane for moz/ffox
[09:09] <pitti> infinity: and it should be audited for breezy, too :-)
[09:14] <Keybuk> mjg59: yeah, they are all pairs
[09:14] <Keybuk> not really sure the significance of having more than one vendor/device pair though
[09:19] <mjg59> Keybuk: There's something very odd about how they're generated
[09:20] <mjg59> .id seems to end up as the cardvendor/cardid unless there are any .devs, in which case it's on the left and the devices fill up the right hand column
[09:20] <mjg59> Keybuk: So I'd guess (to be on the safe side) just transform them all
[09:21] <Keybuk> there's two forms, if cardvendor/carddevice are 0xffff then the real id is on the right hand side
[09:21] <Keybuk> (that's for stuff from pnp_table)
[09:21] <mjg59> Yeah
[09:21] <Keybuk> for stuff from pnp_card_table there's just lots of pairs
[09:21] <mjg59> Ok. It's because cardvendor/cardid defines a specific *card*, and the other IDs define functions on that card
[09:22] <mjg59> If there's only one function, the cardvendor/cardid define that uniquely
[09:22] <Keybuk> ok, so parse them all?
[09:22] <Keybuk> in which case we get http://people.ubuntu.com/~scott/isapnp.auto
[09:22] <mjg59> Yeah, that looks reasonable
[09:23] <Keybuk> alias pnp:dtBA03b0 sb ... looks a bit odd ?
[09:23] <mjg59> CTL is certainly Creative Labs
[09:23] <mjg59> Hmm. Some of those are wrong, yeah
[09:23] <mjg59> The first three have to be alphabetical
[09:24] <Keybuk> I've seen the @@ things before in modules.alias
[09:24] <Keybuk> alias pnp:ctBA03b0dPNPb003* sb
[09:25] <Keybuk> that's in modules.alias too
[09:25] <Keybuk> alias pnp:cALS0007d@@@0001d@X@0001d@H@0001* sb
[09:25] <Keybuk> and that
[09:25] <Keybuk> *shrug*
[09:25] <mjg59> Keybuk: Yeah, they do seem right. How weird.
[09:25] <Keybuk> probably deliberately busticated
[09:25] <mjg59> They're declared that way in the code
[09:25] <mjg59> But:
[09:25] <mjg59> We ought to prefer snd-foo over sb and cs4232
[09:26] <Keybuk> right, sb is blacklisted anyway
[09:26] <Keybuk> are all these modules capable of auto-detecting options and stuff?
[09:26] <mjg59> If they're declaring PNP IDs, then they ought to be doing so
[09:28] <Keybuk> Treenaks: about?
[09:29] <daniels> yo sebarino
[09:30] <Treenaks> Keybuk: ?
[09:30] <mjg59> Keybuk: What's weird is that the ne driver has PNP IDs in it, but doesn't get them generated into the alias list
[09:30] <Keybuk> Treenaks: you have one of those crazy soundcards, right?
[09:30] <Treenaks> Keybuk: yes
[09:30] <Keybuk> Treenaks: want to try dropping that isapnp.auto into /etc/modprobe.d and seeing whether good things happen?
[09:30] <Treenaks> Keybuk: and a weird irda adapter too
[09:30] <mjg59> irda is more awkward
[09:30] <Keybuk> you can probably just rmmod the sound card and sudo /etc/hotplug/isapnp.rc start
[09:31] <mjg59> Most of the drivers aren't PNP aware
[09:31] <mjg59> I should tidy that up some time
[09:31] <Treenaks> Keybuk: I don't have it here atm, I'll get it from my parents this afternoon
[09:31] <Keybuk> ah ok
[09:31] <Keybuk> mjg59: there's a lot of duplicates in that file ... a few things declare love for CSC0010 for example
[09:32] <mjg59> Keybuk: That's probably a joystick port or something...
[09:32] <Keybuk> heh
[09:33] <mjg59> Oh, no, it's a control port
[09:33] <Keybuk> so having one will mean you get three different sound-card drivers loaded?
[09:33] <mjg59> Keybuk: I think that probably means that we should just be using the left hand column unless it's 0xffff, in which case it should be the right-hand column
[09:33] <Keybuk> yeah
[09:33] <Keybuk> that's what I'm thinking
[09:35] <Keybuk> so that means Treenaks will get cs4232 because of his CSC0000 id
[09:36] <Keybuk> which is blacklisted
[09:36] <Keybuk> because the snd-cs4232 doesn't declare that
[09:36] <mjg59> snd-cs4236?
[09:36] <Keybuk> or cs* in fact
[09:36] <mjg59> Hm. There's something very odd about all this.
[09:37] <Keybuk> yup
[09:37] <Keybuk> that CSC0000 snd-cs4236 line came from the end of the line
[09:37] <Keybuk> alias pnp:dCSC0225 snd-cs4236  is the resulting "front"
[09:37] <Keybuk> this all seems a mess <g>
[09:37] <Treenaks> the card needs snd-cs4231
[09:38] <Treenaks> (see #4787)
[09:38] <Keybuk> Treenaks: that driver doesn't declare *any* ids
[09:38] <Keybuk> unless that's also known as snd-opti92x-cs4231
[09:39] <Treenaks> Keybuk: I'll try them all :)
[09:39] <Keybuk> but that only declares support for OPTxxxx not CSCxxxx
[09:39] <mjg59> CSC0000 seems to be WSS
[09:41] <Keybuk> what's IBM0071 ?
[09:41] <Keybuk> hmm, modem of some kind apparently
[09:41] <mjg59> IBM specific thing
[09:41] <Keybuk> IBM0071  IBM Thinkpad infrared port
[09:42] <Keybuk> heh
[09:42] <mjg59> It ought to have a PNP0511 with it as well
[09:42] <Keybuk> no, has a PNP0510 though
[09:42] <mjg59> Close enough
[09:42] <infinity> It's well-hidden.
[09:43] <Treenaks> SMCf010 is irda as well
[09:43] <Treenaks> (for your irda list, mjg59)
[09:43] <mjg59> Treenaks: Got that one
[09:43] <Treenaks> mjg59: ok
[09:43] <mjg59> NSCfoo is, too
[09:44] <Treenaks> whoa, TPM chips are isapnp too
[09:44] <mjg59> And WACblah is (unsurprisingly) a Wacom tablet
[09:44] <Treenaks> mjg59: wacom tablets? on their own proprietary interface card I hope?
[09:45] <mjg59> On tablet PCs
[09:45] <Treenaks> ah
[09:45] <mjg59> So generally a serial UART in an odd bit of io space
[09:46] <Treenaks> hardware is scary
[09:47] <Treenaks> hm.. if I can find some ancient ISA system.. I could test ATI Mach32 8)
[09:48] <Keybuk> ok, so the "slimmed down" version of this file isn't really much use
[09:48] <Keybuk> there's a few things in here, but mostly its oss drivers
[09:50] <Treenaks> Keybuk: I have a TCM5051 also
[09:50] <infinity> mjg59 : Oh, do you already have a full list of working T43 IDs for your auto-enabling of sleep?
[09:50] <infinity> mjg59 : Mine's a 2687DVU, I'd assume all 2687s should work fine.
[09:51] <infinity> (I could be dead wrong, but who knows..)
[09:59] <mjg59> infinity: Yup, got that one
[09:59] <Treenaks> mjg59: got my Asus one too?
[10:01] <mjg59> Treenaks: Yup
[10:01] <Treenaks> mjg59: then it works completely, out of the box.. except for irda and winmodem :)
[10:01] <mjg59> Treenaks: irda-utils doesn't work?
[10:01] <Treenaks> mjg59: I can get it to work, but I have to whack it
[10:02] <Treenaks> mjg59: and "have to whack it" != out-of-the-box
[10:02] <mjg59> Treenaks: With the latest irda-utils?
[10:02] <Treenaks> ("serial" overrides the nsc irda device)
[10:03] <mjg59> Oh, is it one of those irritating things where the nsc driver loads but doesn't actually work?
[10:04] <Treenaks> mjg59: hm, no irda-utils installed
[10:04] <Treenaks> mjg59: is that normal?
[10:04] <mjg59> Nope
[10:04] <mjg59> Well, yes
[10:04] <mjg59> In that it's not part of ship or desktop
[10:04] <Treenaks> ah ok
[10:04] <Treenaks> installing now
[10:04] <mjg59> But if you want IR to actually do anything, you need it
[10:05] <Treenaks> mjg59: the module doesn't get loaded by default
[10:05] <Treenaks> mjg59: maybe now I've installed irda-utils it will be
[10:05] <Treenaks> trying
[10:07] <mjg59> You'll need to run /etc/init.d/irda-setup start and then /etc/init.d/irda-utils start
[10:08] <Treenaks> mjg59: ok, so a reboot should do it :)
[10:08] <Treenaks> (I just finished booting :))
[10:09] <Treenaks> ok module loaded
[10:09] <mjg59> Heh
[10:09] <mjg59> Rock
[10:09] <mjg59> Right. Bit of sleep now, I think
[10:10] <Treenaks> mjg59: "nsc-ircc: unable to allocate dma=-1"
[10:11] <mjg59> Bloody nsc driver
[10:11] <Treenaks> mjg59: good luck with sleep :) try not to dream of nsc drivers ;)
[10:12] <pitti> mjg59: btw, do you think we can sneak you into the nvidia programming lab for one day to fix suspend with their drivers? :-)
[10:12] <Keybuk> pitti: introducing mjg59 to people we dislike is a bad idea
[10:12] <Keybuk> you may find that there's nobody left alive in the nvidia programming lab if we send him
[10:12] <Treenaks> Keybuk: so you're volunteering?
[10:12] <Keybuk> he's a bit of a last-resort weapon for things like that
[10:12] <pitti> Keybuk: I know that Linux is about choice - but choosing between broken suspend and broken video playback is not too easy :-)
[10:13] <Keybuk> pitti: buy a TV and watch your porn on that? :p
[10:13] <pitti> Keybuk: bah, no room for another box :-)
[10:13] <Keybuk> add another room? :p
[10:13] <infinity> Is anyone still seeing "Window List", "Show Desktop", and "Workspace Switcher" quit on login with (nearly) every session login?
[10:14] <pitti> btw, daniels, is a bug "Xv is broken with the nv driver on GeForce 5200" worth filing? or would you ignore it anyway?
[10:14] <robitaille> infinity,  I see it maybe once every 10 logins...
[10:14] <infinity> I used to see it that infrequently, now it's pretty much every login.
[10:14] <pitti> infinity: this never happened to me...
[10:15] <robitaille> I also started sseing it from time to time on my Hoary desktop in last few weeks
[10:16] <Keybuk> hmm, BUG!  DCC SEND -> X-Chat crash
[10:17] <infinity> robitaille : Oh, really?... Hrm.  That's even more curious.
[10:17] <infinity> robitaille : Anything change on the hoary desktop, other than security updates?
[10:17] <infinity> seb128 : Is there a sane and reasonable way to trace/debug these "(panel applet) has quit unexpectedly" on login bugs?
[10:17] <pitti> robitaille: 3v1l backports maybe? :-)
[10:18] <robitaille> infinity,  no.  it's a pretty stable desktop, only security updates.  no backports
[10:18] <pitti> robitaille: before anyone accuses me, I didn't break it :-)
[10:18] <infinity> pitti broke it!
[10:18] <robitaille> but it's a warty PR --> warty -- hoary dist-upgrade  so it could be messy
[10:18] <seb128> infinity: you don't get a bug-buddy dialog to send a bug upstream with a backtrace?
[10:18] <infinity> robitaille : Is it those same 3 panel applets, or just "some random applet crashing"?
[10:18] <seb128> infinity: if not, not really ...
[10:19] <zyga> morning hacker
[10:19] <zyga> s/$/s/
[10:19] <infinity> seb128 : No, I just get "Don't reload" "Reload"... And reload always works fine.
[10:19] <robitaille> but it started doing it semi-recently; maybe in the last 2-3 weeks.     And I believe  it's always these 3 applets
[10:19] <janimo> daniels, should installing x-window-system-core from a server install end up having the same xorg.conf as it would have with default install?
[10:19] <janimo> or is it missing discover1 magic
[10:19] <infinity> seb128 : But I've seen a few people complain about it, I'm not alone, so I'd classify it as reasonably important to hunt down.
[10:19] <pitti> infinity: I didn't upload panel security updates - that would be hopeless :-)
[10:20] <pitti> Hey zyga 
[10:20] <janimo> it asked about resolution but had to do a dpkg-reconfigure to one of the x metapackages to ask about all parameters, otherwise it used the vesa driver
[10:20] <zyga> how do you feel on the verge of release? :)
[10:22] <seb128> infinity: some Debian guys have the same issue, but without a backtrace ..
[10:22] <daniels> janimo: if you don't have discover1 or xresprobe, you'll get a different config, yeah
[10:22] <daniels> pitti: well, as of two days ago, nv is actually maintained
[10:22] <daniels> pitti: so I'll file it in the BZ upstream
[10:23] <pitti> daniels: it works somehow, but the picture gets totally disturbed as soon as I zoom the video window to more than 2x, or fullscreen
[10:24] <daniels> pitti: so scaling is totally broken
[10:24] <pitti> daniels: well, scaling up until 1.9x works even
[10:24] <pitti> daniels: (maybe it's not exactly 2x, just an estimation)
[10:24] <daniels> ah
[10:24] <zyga> hmm what are you talking about?
[10:24] <zyga> scaling video?
[10:24] <daniels> pitti: what specific resolution?
[10:24] <pitti> zyga: Xv on the nv driver
[10:24] <daniels> zyga: with the nv driver, yes
[10:24] <pitti> daniels: my screen is 1280x1024
[10:25] <daniels> pitti: of the video, sorry
[10:25] <pitti> daniels: hm, that happens with any video
[10:25] <dooglus> hi.  is it possible to find the source for old versions of packages?
[10:25] <zyga> I don't know if this is related but totem cannot scale even basic .mpg's on fglrx driver
[10:25] <dooglus> the most recent libgtk2.0-0 broke something in metacity - how can I find source diffs?
[10:25] <zyga> (mplayer scales them fine though)
[10:26] <daniels> zyga: it's probably falling back to the x11 sink instead of xv, because xv isn't enabled by default
[10:26] <pitti> daniels: I can switch to nv (I'm on nvidia now) and try some, if you need
[10:26] <daniels> pitti: as much specific information as I could get would be great
[10:26] <daniels> pitti: you can use xwininfo to find window dimensions to try to work out exactly where it breaks
[10:26] <pitti> daniels: ah, ok, I'll do that
[10:26] <daniels> so does anyone here have working DRI on a Radeon?
[10:27] <pitti> daniels: where is the upstream bz?
[10:27] <pitti> daniels: on my ppc, 3D acceleration works OOTB
[10:27] <pitti> daniels: Radeon 9200
[10:27] <pitti> daniels: tuxracer works great at least
[10:27] <infinity> daniels : fglrx, or radeon?
[10:27] <infinity> (or either?)
[10:28] <daniels> radeon
[10:28] <zyga> daniels: I'm talking about totem, just to be sure - right
[10:28] <Treenaks> daniels: I have working DRI on 2 radeons, but one with a broken resolution
[10:28] <daniels> zyga: right
[10:28] <daniels> pitti: bugs.fd.o
[10:28] <Treenaks> daniels: using fglrx
[10:28] <zyga> daniels: where does totem/gstreamer keep its config?
[10:28] <dooglus> is there a CVS repository with the package sources in it?
[10:28] <zyga> I've checked gconf 
[10:28] <daniels> pitti: could you please test whether DRI worked with -driver-ati 6.8.2-71, and whether -72 broke it?
[10:28] <daniels> dooglus: no, the morgue is broken
[10:28] <infinity> Treenaks : With the open source driver, he meant (I just clarified that a few lines up)
[10:28] <pitti> daniels: yes, as soon as I have my main internet back (ISDN at the moment)
[10:28] <dooglus> daniels: is there an archive of old package versions?
[10:28] <pitti> daniels: I have a slightly older xorg on my iBook
[10:29] <infinity> dooglus : No.
[10:29] <Treenaks> infinity: meh, my radeons are too new for that :(
[10:29] <infinity> dooglus : But you might try pinging the last guy who uploaded it, and see if he has the previous source.
[10:29] <dooglus> so - how do I find out what changed between 2 package versions?
[10:29] <pitti> dooglus: there is the morgue.ubuntu.com
[10:29] <pitti> dooglus: but it's hard to find stuff there
[10:29] <infinity> pitti : The morgue has ben dead for eons.
[10:29] <daniels> dooglus: that's at morgue.ubuntu.com, but currently broken
[10:29] <daniels> pitti: okay, thanks
[10:29] <infinity> s/ben/been/
[10:29] <pitti> infinity: oh, IC
[10:29] <zyga> daniels: I've got radeon 9000 if that helps
[10:29] <dooglus> ew.  shame!
[10:30] <daniels> zyga: if you could do what I asked pitti, that'd be great, thanks
[10:30] <Mithrandir> ogra: the text "someone else" seems to extend outside the button?
[10:30] <zyga> daniels: I'll be back in one hour - I'll check it out then
[10:30] <infinity> mjg59 : BTW, the usplash alternatives bug isn't really fixed.  My alternative had been shoved into manual mode by a previously-broken package, I had to --auto it to unbreak it.
[10:30] <daniels> zyga: thanks
[10:31] <infinity> mjg59 : Once I did, though, got to see it working on vesafb.  yay.
[10:32] <dooglus> ooh.  last guy to upload was seb128...  seb, do you have diffs for libgtk2.0-0 from 2.8.3 to 2.8.4?
[10:34] <pitti> dooglus: oh, diffs between upstream versions? you can certainly use cvs diff -r -r to get them from the gnome cvs
[10:34] <seb128> dooglus: download both upstream tarball and diff -ur between both?
[10:34] <infinity> dooglus : What's the actual bug you're seeing?
[10:34] <seb128> infinity: windows borders on alt-tab
[10:34] <seb128> they used to be solid on screen, they flick now
[10:35] <dooglus> with no change to the window manager code
[10:35] <seb128> that's due to a change between 2.8.3 and 2.8.4 and it's going to be fixed
[10:35] <dooglus> just libgtk changed
[10:36] <seb128> dooglus: you are going to debug it? Just to know if I need to work on it or just have to wait
[10:36] <seb128> :)
[10:37] <dooglus> seb128: I was going to take a look
[10:37] <seb128> cool
[10:37] <seb128> let me know if you figure something
[10:41] <dholbach> good morning mvo
[10:41] <daniels> Zomb: neither
[10:41] <seb128> dooglus: I would say it either http://bugzilla.gnome.org/show_bug.cgi?id=316180 or http://bugzilla.gnome.org/show_bug.cgi?id=316871
[10:41] <Zomb> daniels: I need them however to buy the complete printer list to mozilla
[10:42] <Zomb> s/buy/pass/
[10:42] <daniels> Zomb: if you're using firefox, it'll work, but if you really need the original mozilla, use xprt
[10:42] <dooglus> ok.  i'll also let you know if I don't, and give up...
[10:42] <pitti> bbl
[10:42] <mvo> hey dholbach, good morning all
[10:43] <seb128> hey mvo
[10:52] <dooglus> ok.  first stumbling block:  the GNOME CVS doesn't seem to have labels for releases.  should I use timestamps instead?  if so, where is it recorded what time a release was made?
[10:52] <seb128> it has tags
[10:53] <seb128> dooglus: GTK_2_8_4
[10:55] <dooglus> ok.  I see tags if I use the cvs command.  I was expecting to see them in the web interface, for example http://cvs.gnome.org/viewcvs/gtk%2B/gtk/gtktexttag.c?rev=1.79&view=log
[10:56] <dooglus> ok.  i'll shut up now.  they are in the web interface...
[10:56] <seb128> on the bottom
[10:56] <seb128> you can pick the branch, etc
[10:56] <seb128> the tag
[10:57] <dooglus> yes i see.  thanks :)
[10:57] <seb128> np
[11:09] <JaneW> Kamion: ping
[11:10] <JaneW> Kamion: for the sake of being thorough can we move your remaining BreezyGoals to Completed? https://wiki.ubuntu.com/BreezyGoals
[11:10] <JaneW> mjg59: ping
[11:11] <JaneW> mjg59: can we move Laptop testing to implemented now? as it is well underway (it is a slightly unusual goal)
[11:12] <JaneW> Riddell: ping
[11:13] <Kamion> JaneW: BrandingForDerivatives: remaining string changes in the installer are deferred; please ask jbailey for the status of the rest
[11:13] <JaneW> Riddell: you wanted me to ping you about updating the Kubunut BreezyGoals again at the end of the week... so consider yourself *pinged* ;)
[11:13] <JaneW> Kamion: ok thanks, will do
[11:14] <Kamion> JaneW: I've moved OEMInstaller to Tested; I still want to give it a bit more going-over
[11:15] <JaneW> Kamion: ta
[11:15] <Kamion> JaneW: there are serious bugs in InstallerSimpleResize that I'm working on right now, so it cannot be moved to Completed yet
[11:16] <Kamion> JaneW: bug #16233 and multiple reports of same make me a little wary of moving MountingHDDFilesystems beyond Implemented just yet; I'll be working on those right after the showstopper InstallerSimpleResize bits
[11:18] <Kamion> JaneW: I've left notes against those two specs on the goals page to that effect
[11:19] <JaneW> Kamion: super
[11:19] <Kamion> JaneW: GraphicalInstaller deferred, the base system won't fit (we're having trouble getting everything in as it is)
[11:21] <Keybuk> I really should stop using my bugzilla "ubuntu" keyword as a short-cut to get into bugzilla
[11:22] <Kamion> JaneW: I think most of the metapackage handling and expansion pack stuff from PackageSelection is deferred, but ask mvo for details on that; the base split and debootstrap maintenance simplification are completed
[11:22] <Kamion> (I'll let you figure out how to represent that in the big table, I guess)
[11:22] <JaneW> Kamion: ok
[11:23] <Keybuk> '%s' is not a valid bug number.
[11:24] <Treenaks> Keybuk: maybe %d is :)
[11:24] <Keybuk> Treenaks: don't think firefox knows %d
[11:24] <Treenaks> good point
[11:29] <dooglus> bug 13724 (non-UTF8 fallen out of NLS data files) has been marked as 'fixed', but it's not fixed for me.  Should I make a comment in the bug?  Or is the update somewhere waiting to be uploaded (and if so, shouldn't the bug be 'pending upload' rather than 'fixed'?)
[11:30] <doko> Setting up console-data (2002.12.04dbs-49ubuntu5) ...
[11:30] <doko> Looking for keymap to install:
[11:30] <doko> [and hangs ...] 
[11:30] <seb128> anybody here on the translators list?
[11:31] <Treenaks> seb128: yes
[11:31] <pitti> jdub: ping
[11:32] <pitti> something broke my box - no usplash any more, "Creating initial dev nodes" hangs indefinitively, a mess afterward
[11:32] <pitti> bah
[11:32] <seb128> Treenaks: could you mail them pointing https://wiki.ubuntu.com/DeskopTranslations ... we will do these uploads today but if some people want to update their menu entry translations, they can put them here today
[11:33] <seb128> dooglus: http://bugzilla.gnome.org/show_bug.cgi?id=316180 is the issue
[11:33] <Riddell> JaneW: ping recived thanks :)
[11:34] <JaneW> BreezyGoals Owners: Please can all BreezyGoal owners with goals that are not yet listed as completed, please update the status and note section of the Goal ASAP. The monthly report is going in today, and there'll be questions about any goals that aren't completed.
[11:34] <JaneW> Riddell: :)
[11:34] <dooglus> seb128: I thought it wouldn't be the other one, 'cos I tried metacity without the shape extension and it still failed.
[11:35] <dooglus> seb128: what tells you 316180 is the issue?
[11:35] <seb128> dooglus: I've built a package with this one reverted and that fixes the issue
[11:35] <seb128> dooglus: my build without it :p
[11:35] <Treenaks> seb128: sent
[11:36] <seb128> Treenaks: arg, mvo renamed it "DesktopTranslations"
[11:36] <seb128> there was a typo
[11:36] <seb128> but thanks :)
[11:36] <Treenaks> argh!
[11:37] <seb128> Treenaks: could you reply to your mail with the correct URIs? You can blame me/mvo for the change :) Thanks
[11:37] <Treenaks> ok
[11:37] <dooglus> seb128: how did you revert just that fix?  I don't see a patch attachment.
[11:37] <seb128> Treenaks: or can we do redirect?
[11:37] <seb128> dooglus: I grabbed it from the CVS
[11:37] <Treenaks> seb128: should be possible, too
[11:37] <seb128> Treenaks: let's do that
[11:37] <Treenaks> seb128: go ahead
[11:38] <dooglus> seb128: there's a tag for each bugfix?  or what?
[11:38] <carlos> pitti, hi
[11:39] <Burgundavia> ajmitch, it seems if we are both saving the same content, it doesn't complain
[11:39] <seb128> dooglus: no, but there is a comment for every commit
[11:39] <seb128> dooglus: you just have to pick the commit with the comment described by bugzilla
[11:40] <ajmitch> Burgundavia: hm?
[11:40] <Burgundavia> ajmitch, we both hit deskoptranslations at the sametime, both with redirects
[11:40] <ajmitch> hh
[11:40] <ajmitch> heh
[11:41] <pitti> jdub: here?
[11:42] <dooglus> seb128: it's been a very long time since I used CVS for anything other than 'cvs co'.  How are you finding the commit?
[11:42] <seb128> dooglus: http://cvs.gnome.org/viewcvs/gtk%2B/gtk/gtkwindow.c?rev=1.310&view=log
[11:43] <seb128> dooglus: click on "Diff to previous" ... for the line with the comment you want
[11:43] <dooglus> that's OK if you know which source file was patched - but if you don't?
[11:43] <seb128> you read the Changelog other way
[11:44] <dooglus> ok, of course.  thanks.
[11:44] <seb128> np
[11:44] <dooglus> I'm used to working with ClearCase rather than CVS, and it's totally different...
[11:46] <bob2> woah
[11:48] <ajmitch> evening bob2 
[11:54] <tepsipakki> hmm, firefox segfaults on me if esd is on
[11:55] <tepsipakki> works if I either kill esd or comment out the FIREFOX_DSP-things from the wrapper
[11:55] <pitti> jdub: ping
[12:00] <pitti> mvo: any news about update-notifier?
[12:01] <mvo> pitti: did you got my message from last night?
[12:01] <mvo> pitti: it's working, I need to do the i18n now
[12:01] <pitti> mvo: ah, cool! no, I didn't
[12:01] <seb128> if there is something to translate to french please point it :)
[12:01] <pitti> mvo: I didn't see a new upload
[12:01] <seb128> jordi: around?
[12:02] <pitti> mvo: I guess you can do the .de translation yourself :-) 
[12:02] <mvo> pitti: it will be uploaded today (when I have finished the i18n bit of the generic notification)
[12:02] <mvo> pitti: heh, yeah :)
[12:02] <pitti> cool
[12:02] <pitti> mvo: so I keep myself ready for updating dbus and hal, since this should happen ASAP
[12:02] <pitti> carlos: hey Carlos, good morning!
[12:02] <pitti> carlos: I had big trouble with my ISP, but it seems to have stabilized again
[12:03] <mvo> pitti: normal notifications are like desktop files, not available through rosetta, but I want this one handled via gettext
[12:03] <mvo> pitti: yes, I'll hurry
[12:03] <pitti> mvo: maybe you can use po-debconf to merge translations?
[12:03] <pitti> mvo: the format should be similar
[12:03] <carlos> pitti, ok
[12:04] <carlos> pitti, phone, just give me sometime
[12:04] <pitti> carlos: oh, sure
[12:17] <ogra> Mithrandir, erm, there is no such string anymore ?
[12:18] <JaneW> **BreezyGoals Owners**: Please will all goal owners with goals that are not yet listed as completed (i.e. still at implemented or testing), update the Status and Notes sections of the Goal ASAP. 
[12:20] <mvo> has anyone seen that our init-scripts doing a fsck twice on the same device? I just got two fscks for "/"
[12:25] <pitti> mvo: I can't reproduce #16678, can you please check what breaks there?
[12:25] <carlos> pitti, hi (again)
[12:25] <mvo> pitti: did you used italian?
[12:26] <pitti> mvo: french
[12:26] <pitti> mvo: ok, I try italian
[12:26] <mvo> pitti: I did it with italian and I was very puzzled that a reboot fixed it
[12:31] <pitti> mvo: after installing the langpack, I can instantly do "$ LANGUAGE= LANG=it_IT.UTF-8 pmount --help" and get italian
[12:31] <pitti> mvo: same for you?
[12:32] <pitti> mvo: I try to relogin now
[12:35] <mvo> mjg59: do you mind if I do another usplash upload that (hopefully) fixes the flicker-before-gdm problem. should I send you patches? or create a branch from your bzr tree?
[12:38] <pitti> mvo: indeed, I get it with Italian - but restarting gdm helps
[12:38] <pitti> mdke: ^
[12:38] <mvo> pitti: so, "iz gtk bong" !
[12:40] <pitti> mvo: still puzzled why it works for French, but not for Italian
[12:40] <pitti> I'm sure that is seb128's secret plan to convert the world to French
[12:41] <pitti> mvo: so that he can always close as RESOLVED/WORKSFORME :-)
[12:41] <mvo> haha
[12:41] <mvo> :)
[12:42] <mvo> pitti: you made my morning :)
[12:52] <Mirv> so translators were informed about translating menu entries during the last day that it is possible.. erh. but what about this "notification-daemon", translations for which are also due to be final today? I've never seen anything about translating it... it's not in rosetta so in which way it's supposed to be translated anyway?
[12:52] <Mirv> (supposing I have time today to do something about it)
[12:53] <pitti> Mirv: now would be a good time to coordinate with mvo about translating the new "reboot" notification
[12:53] <pitti> Mirv: if you have a translation for the language pack upgrade, please send it to me ASAP
[12:54] <pitti> Mirv: for langpacks, I have French and German ATM
[12:54] <pitti> Mirv: getting Finnish would be nice :-)
[12:55] <Mirv> pitti: yes, but is there a po file or something else for the notifications?-) I haven't found anything...
[12:55] <pitti> Mirv: no, it's not using gettect
[12:55] <Diziet> usplash seems to have stopped happening on my main breezy testbed.
[12:58] <pitti> Diziet: for me too, I guess jdub broke it with the new artwork
[12:59] <pitti> Diziet: $ sudo dpkg-reconfigure linux-image-2.6.12-9-amd64-generic
[12:59] <pitti> cpio: ./usr/lib/usplash/usplash-artwork.so: Datei oder Verzeichnis nicht gefunden
[12:59] <pitti> ("file or directory not found")
[12:59] <pitti> I guess that's what breaks it
[12:59] <pitti> Diziet: did you experience an udev hang at boot? I did
[12:59] <torkel> pitti: update-alternatives --auto usplash-artwork.so
[01:00] <Diziet> No, no hang.  It just boots normally without udev.
[01:00] <pitti> torkel: but that should be taken care of automatically
[01:00] <torkel> pitti: i agree
[01:00] <pitti> torkel: I don't see a reason to change the path at all
[01:01] <torkel> pitti: a .so should not be in /usr/share/... ?
[01:04] <torkel> pitti: shouldn't the eq in 'dpkg --compare-versions "$2" eq 0.1-12' usplash.postinst be a ge?
[01:07] <Mirv> pitti: I will try to ping mvo about the reboot message
[01:08] <mvo> Mirv: here
[01:08] <mvo> the current text is: 
[01:08] <mvo> Description: Reboot required.
[01:08] <mvo>  To complete the system update a reboot is required so that the new
[01:08] <mvo>  services get activated. We strongly recommend that your machine is
[01:08] <mvo>  restarted as soon as possible
[01:08] <JaneW> **BreezyGoals Owners**: Please will all goal owners with goals that are not yet listed as completed (i.e. still at implemented or testing), update the Status and Notes sections of the Goal ASAP. 
[01:08] <Mirv> mvo: ah, okay, I will msg it to you
[01:08] <mvo> suggestions are welcome, it's a generic message 
[01:08] <JaneW> that was the last time promise ;)
[01:09] <mvo> Mirv: wait a bit please, there may be suggestions/typos etc
[01:09] <Mirv> mvo: okay
[01:09] <mvo> how does the above text sound to native speakers?
[01:09] <Kamion> "get activated" is the wrong tone
[01:09] <Kamion> "are activated" would do
[01:10] <Kamion> I'd drop the full stop from the end of "Reboot required"
[01:10] <Kamion> and add one at the end of the extended description, which is (unlike the short description) composed of sentences
[01:11] <Robot101> did anyone fix the "There are 1 item(s) of post-update informations." string? :)
[01:11] <ogra> is it expected behavior that my system switches to console between usplash abd gdm now ?
[01:11] <ogra> s/abd/and
[01:12] <Diziet> Ha, you have a usplash :-).
[01:12] <Mirv> pitti's string had the whole text after Description: (not new line after a short two words' description)
[01:13] <mvo> ogra: yes
[01:13] <mvo> new: Description: Reboot required
[01:13] <mvo>  To complete the system update a reboot is required so that the new
[01:13] <mvo>  services are activated. We strongly recommend that your machine is
[01:13] <mvo>  restarted as soon as possible.
[01:15] <segfault> "System->About Ubuntu", where does that string come from?
[01:15] <ogra> Diziet, yes, after a manual reconfigure of linux-image :)
[01:16] <Mirv> segfault: I guess it's https://wiki.ubuntu.com/DesktopTranslations now :P I've seen it in Rosetta somewhere too, and it's already correct in Finnish
[01:17] <segfault> humm, didn't know about that link
[01:17] <segfault> still in time to hit breezy final?
[01:18] <Mirv> segfault: I didn't know until two hours ago, and this is the last time to submit those
[01:18] <Mirv> time=day
[01:19] <segfault> ahh, nice
[01:19] <segfault> thanks man
[01:20] <segfault> you saved my life
[01:20] <segfault> :)
[01:22] <Mirv> np :)
[01:23] <nakee> what's ubuntu policy about local markets?
[01:24] <tseng> nakee: how do you mean
[01:24] <nakee> in promotion of ubuntu
[01:24] <nakee> any intrest in sponsoring local linux events?
[01:25] <tseng> definately through ship-it (free cds) and promotion
[01:25] <tseng> if you expect more, youd have to mail info@canonical.com
[01:25] <nakee> what do you mean by promotion?
[01:26] <tseng> http://fridge.ubuntu.com/
[01:26] <tseng> "upcoming events"
[01:26] <tseng> i doubt it would be a problem to add events to that
[01:28] <mdke> nakee, if you are interested in setting up a local group, also have a look at the various wiki pages starting here: https://wiki.ubuntu.com/LoCoTeams
[01:28] <mdke> that is quite comprehensive
[01:30] <nakee> mdke: yea, I ment more investing money in promotion
[01:30] <nakee> but I  guess for that I should email canonical:)
[01:31] <mdke> yes
[01:36] <\sh> ogra_: ping
[01:36] <\sh> ogra_: after reinstalling the sony (latest daily breezy iso) the problem with the keyboard went away
[01:37] <ogra_> hrm
[01:37] <ogra_> its still not working here
[01:38] <\sh> ogra_: did u reinstall all your stuff?
[01:38] <ogra_> nope
[01:39] <\sh> ogra_: looks like we have a problem during upgrade from warty and/or hoary to breezy
[01:39] <ogra_> i cant, my DVD is broken... and its my main development sysytem
[01:39] <\sh> eek
[01:50] <ogra_> \sh, you saw keycodes on the console by just pressing altgr ?
[01:50] <sistpoty> ping infinity
[01:51] <\sh> ogra_: yes
[01:51] <\sh> ogra_: even before the reinstall..so it can be a gnome upgrade problem as well
[01:52] <ogra> \sh, then its a HW problem on my side i guess.... i dont see any keycodes
[01:53] <\sh> ogra: hmmm....
[02:01] <ogra> grmpf
[02:01] <ogra> @@@@
[02:01] <ogra> \sh, it works with USB keyboard :(
[02:15] <dooglus> seb128: just to let you know, I have completely failed to fix the alt-tab bug (16589).  Please feel free to work your magic.
[02:16] <Mirv> pitti: did you get/save/add the Finnish translation before your irc died?
[02:16] <Mirv> ok, thanks :)
[02:21] <pitti> Mirv: I committed them now, thanks a lot
[02:23] <\sh> hach pgra
[02:23] <\sh> aeh ogra
[02:23] <pitti> Diziet: after a hoary->breezy upgrade ffox yells at me: "/usr/share/ubuntu-artwork/home/index.html" not found
[02:24] <\sh> seb128: do u have the possibilty to do an upgrade from warte/hoary to breezy? 
[02:24] <pitti> Diziet: is this a ffox upgrade bug or a bug in the current artwork package?
[02:24] <pitti> \sh: I just did a hoary->breezy upgrade, FWIW
[02:24] <\sh> pitti: and? no problems with umlauts and/or meta-q/e/7/0/ ?
[02:25] <seb128> dooglus: k, thanks anyway
[02:25] <pitti> \sh: I got the dreaded X/Gnome keyboard question again
[02:25] <seb128> \sh: I've planned to do one this afternoon for some bugs, why?
[02:25] <pitti> \sh: no, my keyboard works fine
[02:25] <\sh> pitti: hmmm....I didn't get any question because of the keyboard...but in gnome no special meta-keys were working
[02:26] <\sh> pitti: it was warty -> hoary -> breezy upgrade 
[02:26] <\sh> pitti / seb128: after installing breezy from scratch it disappeared
[02:26] <\sh> so we could have an issue with those upgrades
[02:26] <seb128> hoary to breezy?
[02:26] <seb128> what keymap do you use?
[02:26] <\sh> seb128: warty to hoary to breezy
[02:27] <\sh> seb128: german, so de with dead keys
[02:27] <\sh> seb128: not me...a colleauge of ogra and me...
[02:27] <seb128> gconftool-2 -R /desktop/gnome/peripherals/keyboard/kbd 
[02:27] <seb128> that before and after update would be nice
[02:28] <\sh> seb128: now it's too late...I will try to do this again with vmware
[02:32] <pitti> Is anyone here interested in translating the langauge pack upgrade notification? I already have en, de, fr, and fi
[02:32] <segfault> pitti: i do
[02:32] <mdke> pitti, i can post to the -it mailing list, if you like
[02:33] <pitti> http://people.ubuntu.com/~pitti/langpacks/langpack-upgrade-note.txt
[02:33] <pitti> please mail to martin.pitt@ubuntu.com or just /msg it to me
[02:34] <Diziet> pitti: ffox> How odd.  Obviously my machine doesn't do that.
[02:34] <pitti> Diziet: is this path supposed to be the correct one?
[02:34] <mdke> pitti, awesome, also maybe you can mail it to ubuntu-translators@lists?
[02:35] <Diziet> No, I tell a lie, it does.
[02:35] <pitti> Diziet: i. e. did jdub break the last artwork package or did the path change?
[02:35] <mdke> pitti, or I can
[02:35] <Diziet> The path in ffox didn't change.  Or at least, I didn't intend it to.
[02:35] <Diziet> And obviously I'm pretty sure of that since it worked fine with my test build of ffox ubuntu18.
[02:35] <pitti> $ dpkg -L ubuntu-artwork|grep html
[02:35] <pitti> $
[02:35] <mdke> Diziet, jbailey will explain, essentially the file is moving package
[02:35] <pitti> jduuuuub!!!
[02:35] <Diziet> mdke: So why isn't it present ?
[02:36] <mdke> Diziet, i don't know I am afraid
[02:36] <Diziet> Is there a bugzilla bug about this yet ?
[02:36] <mdke> Diziet, yes let me dig it out
[02:36] <pitti> mdke: I need to subscribe to it, I guess?
[02:37] <pitti> Diziet: btw, nice work, a hoary->breezy upgrade does not have a single question any more
[02:37] <mdke> http://bugzilla.ubuntu.com/show_bug.cgi?id=3985
[02:37] <mdke> i think that is the right one
[02:37] <segfault> pitti: done
[02:38] <Diziet> pitti: question> Good.
[02:38] <Diziet> 3985 ?  Sounds a bit old.  /me reads ...
[02:38] <mdke> Diziet, it started off as something else
[02:39] <pitti> segfault: thanks
[02:39] <mdke> pitti, want me to mail that translation to the translators list?
[02:39] <pitti> mdke: would be nice, thanks
[02:40] <mdke> ok
[02:40] <segfault> is there anything that needs to be translated?
[02:40] <mdke> pitti, done, you in cc
[02:43] <Diziet> That bug is against firefox and assigned to jbailey.  Does that mean he's planning to upload a firefox with a different start URL ?
[02:43] <mdke> Diziet, not sure, but that was what I'd assumed
[02:44] <Diziet> jbailey: ping
[02:44] <pitti> Diziet: why should the file be moved in the first place? u-a seems about right?
[02:44] <Diziet> And why would the file need to move in the filesystem just because the package has changed ?
[02:44] <mdke> the reason was that the file is also in ubuntu-docs already, and is updated there
[02:44] <mdke> Diziet, perhaps the sane solution is to put the file back in the same place, in the new package
[02:44] <Diziet> Oh, well, clearly it ought to be in only one package.
[02:44] <pitti> mvo: in update-notifier, will "Description-pt_BR work" or just -pt"?
[02:45] <mdke> Diziet, the move has also caused issues for ogra i think, from what he said yesterday
[02:45] <jbailey> Diziet: pong
[02:45] <Diziet> Hello.  See scrool re firefox start page and 3985.
[02:46] <jbailey> Diziet: I can upload a fixed FF and Ephy easily enough.
[02:46] <jbailey> Can someone tell me what it should point to? =)
[02:47] <mdke> jbailey, it should point to an html version of the about-ubuntu document
[02:47] <Diziet> No, no, I'm not asking you to do an upload.  I'm just (at the moment) trying to figure out what's going on.
[02:47] <Diziet> I can upload easily enough.  I'm just trying to have some coordination.
[02:47] <Diziet> Do you have any idea what's going on ?
[02:47] <jbailey> Diziet: Lovely.  I'd rather not have the "touched it last" tag on FF. =)
[02:47] <Diziet> :-)
[02:48] <Nafallo> hehe
[02:48] <jbailey> Diziet: Only that it's lived in -artwork for a long time, and apparently stopped yesterday.
[02:48] <Diziet> The proximate symptom is that current breezy's ffox gives a file doese not exist dialogue box.
[02:48] <jbailey> Lovely.
[02:48] <jbailey> Lemme see if I can quickly cobble together the HTML page.
[02:49] <Diziet> Um, if the file has just moved package, I don't understand why it needs to change name.
[02:49] <fabbione> Diziet: did you notice by any chance that mozilla and firefox are crashorama on amd64?
[02:49] <Nafallo> jbailey: btw, I runned update-alternatives --auto usplash-artwork.so and it changed instantly. is that something we should run somewhere in the postinst? :-)
[02:49] <jbailey> Well, the first file wass in /usr/share/doc/ubuntu-artwork. =)
[02:49] <Diziet> fabb: No, I have no idea about that.
[02:49] <Nafallo> fabbione: are they?
[02:50] <jbailey> Nafallo: No, because it should've been auto in the first place.  I thnk you stumbled on a bug from the 6 hours it pointed to the wrong place.
[02:50] <fabbione> Nafallo: yes.. 
[02:50] <Diziet> When did this start happening, and what do you do to trigger it ?
[02:50] <fabbione> Diziet: ok
[02:50] <fabbione> Diziet: i can't say when it started.. the syntom has always been there since i bought the amd64
[02:50] <Diziet> OK ?  Doesn't sound OK to me :-).
[02:50] <Nafallo> jbailey: seems like it ;-).
[02:50] <Diziet> I see,.
[02:51] <fabbione> Diziet: it's enough to hit the reload button sometimes..
[02:51] <fabbione> if i use shift and do it slowly it's ok
[02:51] <Diziet> Are you sure this machine is OK ?
[02:51] <fabbione> most of the time just crashes
[02:51] <infinity> jbailey : Yeah, it went manual on mine too, which generally points ast a serious fuckup in how maintainer scripts are calling update-alternatives.
[02:51] <fabbione> Diziet: yes
[02:51] <fabbione> Diziet: everything else works fine
[02:51] <Diziet> Ha ha.
[02:51] <Diziet> Umm, I've not heard any other reports.
[02:51] <Nafallo> infinity: yay! I'm not alone! :-)
[02:51] <infinity> jbailey : If all is well now, we can't really do much except declare it a casualty of tracking breezy daily, I guess.
[02:51] <Diziet> Anyone else here with an amd64 ?
[02:51] <jbailey> infinity: Right.  I made a thinko and put the wrong path in.
[02:52] <jbailey> infinity: I had put a bit in the postinst that would fix it if it ddn't flip to manual
[02:52] <jbailey> (remove the bad one, put in the good one)
[02:52] <jbailey> But for some readon u-a sets it to manual if it's a bad path.
[02:52] <Nafallo> Diziet: yes, and I don't have those symptoms :-).
[02:52] <jbailey> Stupid.
[02:52] <Diziet> inf: Right.  Perhaps a mail to ubuntu-devel saying `if this happened to you, we know about it, here is how to make it right, users won't see it'.
[02:52] <infinity> jbailey : Try remove the bad, check where it currently points, if still bad, flip back to auto.
[02:52] <Diziet> naf: Thanks.
[02:53] <jbailey> infinity: I think probably not worth touching at this point, but I know that now.
[02:53] <Diziet> fabb: Firefox is, ahm, unusually large.  So it might well crash more readily on bad hardware.
[02:53] <infinity> jbailey : s/bad/known bad/
[02:53] <jbailey> And consider it to probably be an 'important' bug on u-a
[02:53] <infinity> It's been how u-a has behaved for eons.
[02:53] <jbailey> Sure, doesn't mean it's right.
[02:53] <jbailey> It makes it a situation where I might override a users chosen settings by forcing to auto.
[02:54] <Diziet> This reminds me of some bug in the Debian BTS about u-a reverting.
[02:54] <jbailey> Probably.
[02:54] <infinity> Well, it's not likely the user would chose to a) point at the same broken path you used, and b) point at it if it's a dangling link (two easy things to test for)
[02:54] <jbailey> I'm pretty certain that all paths lead to policy violation there.
[02:54] <Nafallo> jbailey: I would have blamed the kernel if it had not worked after next upgrade ;-)
[02:55] <mdke> jbailey, any more help needed on about-ubuntu, cos otherwise I am gonna get back to work
[02:55] <jbailey> Nafallo: Sure, most poeple don't realise the splash is separate.
[02:55] <jbailey> mdke: I don't think so, thanks.
[02:55] <mdke> jbailey, great, thanks
[02:55] <Nafallo> jbailey: well, I would have blamed it cause of the dpkg-reconfigure bailing out with cpio-errors ;-).
[02:55] <jbailey> Nafallo: Eh, did you see cpio errors?
[02:56] <Nafallo> jbailey: ehm, IIRC yes. I'll check the private log from yesterday :-).
[02:56] <jbailey> Nafallo: Doesn't matter now, it's fixed.
[02:56] <jbailey> I just don't remember anyone mentioning that to me. =)
[02:57] <jbailey> Diziet: Are trnaslations useful to you, or should it just always be in English for now?
[02:57] <jbailey> I don't remember if FF/ephy will hunt for a .LANG.html in a directory
[02:57] <infinity> jbailey : Anyhow, yes, as a casualty of breezy tracking, you should probably mail -devel with a "if usplash artwork doesn't work, and update-alternatives --display upsplash-artwork.so says it's manual, then 'update-alternatives --auto usplash-artwork.so && dpkg-reconfigure linux-image-`uname -r`', have a nice day."
[02:57] <Nafallo> "cpio: ./usr/lib/usplash/usplash-artwork.so: File or directory doesn't exist"
[02:58] <pitti> Nafallo: right, same here
[02:58] <Diziet> 308838, 285686, 87677
[02:58] <jbailey> infinity: Right, there's one related bug where the postinst hasn't run the alternative yet that I need to fix first.
[02:58] <jbailey> infinity: So I wonder if I should just look at it in the postinst.
[02:58] <Nafallo> pitti: hehe. you don't have usplash anymore now :-)
[02:58] <infinity> jbailey : Easy enough to do.
[02:59] <jbailey> Just test it with readlink?
[02:59] <Diziet> jbailey: Translations of what ?  Like error messages ?  Depends on the language, but I can probably cope with error messages in at least many languages with Latin script.
[02:59] <jbailey> Hmm, readlink, -e the destination.  Is there a sane way to check for manual or auto?
[02:59] <Diziet> Or do you mean translations of the start page ?
[02:59] <jbailey> Diziet: Of the about ubuntu page.
[03:00] <Diziet> Um, you know, I hadn't thought about that at all.
[03:00] <Diziet> But obviously they should be translated and we should use the right one.
[03:00] <Diziet> I have no idea what it does if you install non-English.
[03:00] <infinity> jbailey : --display tells you if it's auto or manual.
[03:00] <Diziet> I should do a non-English install to dsee what it's like.
[03:00] <jbailey> Diziet: So far it hasn't flipped to French for me.
[03:01] <Diziet> But you agree that it shoudl ?
[03:01] <jbailey> Diziet: Right, so I have to parse the first line.  suck.
[03:01] <jbailey> infinity: ^^
[03:01] <Diziet> Parse the first line ?
[03:01] <jbailey> I wonder if I care?  If it's pointing to /usr/share, and it doesn't exist.
[03:01] <jbailey> ..
[03:02] <Diziet> I think I'll install a Dutch version of hoary and see what it looks like.
[03:02] <jbailey> Diziet: I agree that it should.  I also think that since it never has before, it's a bit late to worry about it for Breezy.
[03:02] <Diziet> Is it ?  It's probably not very hard to make the start URL depend on language somehow.
[03:02] <jbailey> Your call.  We have other bugs to chase too, and by any sane measurement all of the text-related freezes have now passed.
[03:03] <Diziet> I had no idea that it didn't work.
[03:03] <Diziet> Don't you think it's very shoddy to have that page only in English ?
[03:03] <janimo> ogra, who/what starts xscreensaver?
[03:03] <jbailey> 'k, I have the HTML now.   I just need to figure out how to turn <img src="../../images/C/UbuntuLogo.png" /> into a better path
[03:03] <jbailey> Diziet: Oh sure.  Most things that show up on my desktop in English remind me how far we have to go.
[03:03] <Diziet> `You have the HTML'.  I'm getting worried here.
[03:04] <infinity> jbailey : Define "better"....
[03:04] <jbailey> Diziet: mdke said that the page should point to an HTML version of 'About Ubuntu' from the docteam.
[03:04] <Diziet> I thought this HTML was already in some package or other.  ubuntu-doc, or something.
[03:04] <Diziet> So the docteam have it in HTML already ?
[03:04] <Diziet> Or it has to be converted ?
[03:04] <jbailey> infinity: One where you can expect to find the image.  Probably an absolutely path to an images directory.
[03:04] <jbailey> Diziet: Dude, it's docbook. =)
[03:05] <Diziet> Ahhh.
[03:05] <infinity> jbailey : But if that relative path is correct when docbook generates it, what are you doing to make it incorrect?... Moving files around willy-nilly?
[03:05] <jbailey> If you want it in texinfo, I can twist a knob.. =)
[03:05] <Diziet> So you're making ubuntu-doc spit out HTML ?
[03:05] <jbailey> infinity: The layout in the repository does not match the final installed layout right now.
[03:05] <Diziet> Obviously it should be HTML for ff :-).
[03:05] <jbailey> infinity: It's the same problem as the entity problem you whined about earlier.
[03:05] <jbailey> Diziet: Bah!  We want inline PDFs!
[03:05] <Diziet> :-P
[03:05] <infinity> jbailey : So do the docbook to html generation AFTER moving it to an appropriate layout, not before.
[03:06] <bddebian> Howdy
[03:06] <infinity> jbailey : That seems saner than running sed accross everything to munge paths.
[03:07] <infinity> cp foo build-tree/stuff/lives/here && cp bar build-tree/stuff/lives/there && docbook magic && install from build-tree to debian/
[03:07] <infinity> (bonus points, cause it buys you a lovely build-tree to rm -rf in clean)
[03:07] <Diziet> This mixture of Dutch and English in the installer is a bit odd, indeed.  Unfortunately my Dutch is really only up to comprehension and not helping with translation.
[03:07] <slomo> seb128: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CAN-2005-2964
[03:08] <jbailey> infinity: eh?  There's still a sed pass or something in there.  The source docbook has:                 <imagedata fileref="../../images/&language;/UbuntuLogo.png" format="PNG"/>
[03:08] <pitti> \sh: here?
[03:09] <jbailey> infinity: Or perhaps an entity that can be either "../.." or "/usr/share/doc/ubuntu-docs"
[03:09] <jbailey> infinity: Anyhow, it's a fix-today kind of bug.
[03:09] <Diziet> These big disks on modern computers are very nice.  My testbed machine has room for a further 4 roomy installs :-).
[03:09] <jbailey> Diziet: Which timezone are you awake on?  I need to do morning preppish things and can hack after.
[03:10] <mdz> morning
[03:10] <Diziet> GB
[03:10] <Diziet> mdz: Hello.
[03:10] <jbailey> g'm Matt.
[03:10] <Diziet> I'll be around until exactly 7 today, ie a little under 5 hours from now, and then I have to go.
[03:10] <pitti> Hi mdz
[03:10] <mdz> it is wholly unreasonable to be awake before the sun is up
[03:11] <jbailey> mdz: That depends if you're still awake or not. ;)
[03:11] <pitti> mdz: so why you are?
[03:11] <Diziet> Indeed so.  Go back to bed :-).
[03:11] <bddebian> mdz: Yes
[03:11] <mdz> can't sleep
[03:11] <jbailey> ... clowns will eat me!
[03:11] <Diziet> *sympathy*  You need more late nights.
[03:11] <Robinho_Peixoto> pitti: And this bug http://bugzilla.ubuntu.com/show_bug.cgi?id=15820 ?
[03:12] <Diziet> But, since you're here, I was wondering if you had an opinion about this conversation we're having about the firefox start page.
[03:12] <Diziet> See scrollback, or I can summarise.
[03:12] <pitti> Robinho_Peixoto: I hope to build new langpacks today
[03:12] <Nafallo> pitti: yay! :-)
[03:13] <mdz> Diziet: I believe firefox looks for .lang.html
[03:14] <mdz> if it doesn't, it probably should, but if not, it's too late
[03:14] <Diziet> Right.  I was hoping you'd say that.
[03:14] <Diziet> I'm checking it now.
[03:15] <Diziet> Standard rules for locale resolution apply, I hope.
[03:15] <infinity> If it does, that's news to me.  That sort of negotiation is generally done by webservers, didn't think firefox would emulate an http server when using file:/// URLs.
[03:15] <Diziet> The default preferences are done by piece of javascript, which AIUI is run when it creates a new profile.
[03:16] <Nafallo> I agree with infinity. why is that option in apache if ff could do it?
[03:16] <jbailey> Nafallo: Because it would mean an extra round trip over the wire.
[03:16] <jbailey> I'd expect it to only do it on file:/// URI
[03:16] <jbailey> s
[03:16] <Diziet> No, no, no, I don't want to make it do the equivalent of content negotiation on file:/// URI's.
[03:16] <Diziet> Not unless it already does :-), which I doubt.
[03:16] <jbailey> Or possibly several round trips over the wire given locale search orders.
[03:17] <jbailey> Diziet: How else would it do it? =)
[03:18] <Kamion> jbailey: ssh-askpass-gnome does that readlink trick you were suggesting, because I couldn't find any other way to avoid the ssh-askpass alternative ending up in manual mode
[03:18] <Diziet> `Download the required packages from the Internet?  <Terug>  <Ja> <Nee>'> LOL
[03:19] <Diziet> Oh dear, the timezone setting part of the installer has crashed.
[03:19] <Kamion> if dpkg --compare-versions "$2" lt-nl <foo> && [ -h /etc/alternatives/<bar> ]  && [ "$(readlink /etc/alternatives/<bar>)" = <baz> ] ; then update-alternatives --auto <bar>; fi
[03:19] <Kamion> Diziet: hoary?
[03:19] <Diziet> Oh, well, I'll let it believe I'm in Amsterdam.  K: Yes.
[03:20] <Kamion> Diziet: fixed in breezy, though I can't remember the bug number right now
[03:21] <Kamion> was a translation handling problem somewhere in the hideous base-config/debconf/cdebconf stack that's involved there
[03:21] <Diziet> Sure.  I'm starting with hoary because I want to check a few things with the ffox upgrade.
[03:21] <Kamion> right, just letting you know in case you were going to file a bug :-)
[03:21] <Diziet> :-)
[03:22] <jbailey> Kamion: 'kay, thanks.  I'll do that then and make sure it's solved.
[03:22] <mdz> jbailey: what do we lose if we revert that whole mess?
[03:22] <jbailey> Oo, reminds to check.  My dad said that the installer from a week or so ago gave him grief when he selected UTC for a server.
[03:23] <jbailey> mdz: The artwork page for Firefox?  Nothing.  If we'd rather have the HTML version of about-ubuntu, we can drop it in place int he ubuntu-artwork package anyway.
[03:23] <mdz> jbailey: oh, I thought you were talking about the usplash alternative
[03:23] <mdz> jbailey: which seems to continue to generate bug reports
[03:24] <jbailey> mdz: Sorry, two threads at once, hard to track which I'm on at the moment.
[03:24] <Diziet> The init.d script startup messages aren't translated, are they ?
[03:24] <mdz> Diziet: not
[03:24] <mdz> at all
[03:25] <Diziet> Maybe for dapper.  It shouldn't be too hard.
[03:25] <jbailey> mdz: The alternative can be reverted, we lose usplash brnading for kubuntu and edubuntu.
[03:25] <Kamion> mdz: btw, I thought the non-langpack translation deadline was for translations to arrive in Rosetta, not for uploads, which is why the corresponding installer uploads haven't arrived yet; it takes me some hours of clicking on stuff in Rosetta, waiting for mails, and downloading tarballs by hand to merge installer translations from Rosetta
[03:25] <mdz> jbailey: http://bugzilla.ubuntu.com/show_bug.cgi?id=16676
[03:25] <mdz> Kamion: waiting for mails?!
[03:26] <jbailey> mdz: Yeah, race condition between kernel install and the postinst running in usplash to put the file in the right place.
[03:26] <Kamion> mdz: uh-huh. have you downloaded translations from Rosetta lately?
[03:26] <mdz> jbailey: this is too much complexity to drop in at the laast minute
[03:26] <mdz> Kamion: never
[03:26] <Diziet> mdz: Also, your message about artwork freeze.  How does that relate to the Firefox fonts issue ?  (Which I will look at a proper fix for later today hopefully, but I think we'll have to go with the `just furtle the aliases' plan.)
[03:26] <mdz> Kamion: but  involving email in that process sounds unreasonable to me
[03:26] <jbailey> mdz: 'k, I'll pull that outo then.  I wanted the artwork flexible enough and had been testing it here for a while.
[03:27] <mdz> Diziet: grey area. please get  the changes in ASAP
[03:27] <Kamion> mdz: apparently the old useful po-export functionality hung the launchpad appservers for ages while it processed them, so now you have to send an asynchronous request and it e-mails you with a librarian URL when it's done.
[03:27] <Diziet> mdz: Right.
[03:27] <Kamion> which is an utter pain in the arse
[03:27] <jbailey> Kamion: It still loses the requests occasionally.
[03:27] <chmj> quit --> home 
[03:27] <Kamion> jbailey: even better!
[03:27] <jbailey> Kamion: I had to ask a few times for one of the pofiles yesterday.
[03:28] <Kamion> Diziet: timezone bug => #9160
[03:28] <ogra> pitti, how is the web management tool (eq. phpmyadmin) for postgres called ? i know there was one... got a user in #edubuntu asking me...
[03:28] <Diziet> k: 'sok, I'm just going to ignore it.
[03:28] <pitti> ogra: phppgadmin IIRC
[03:28] <pitti> ogra: yep
[03:28] <pitti> ogra: I never saw it, though
[03:29] <mdz> jbailey: why does replacing the artwork require a new .so and alternative?  the existing artwork branding worked by swapping one package for another
[03:29] <Diziet> (phppgadmin ?  Is that what it sounds like ?  A really bad idea, and a strange gargling noise ?)
[03:29] <ogra> pitti, me neither... but it seems there is a demand for gui admin tools... do we have any other ?
[03:29] <Nafallo> pitti, ogra: I've used it a bit. it's like mysql with the added complexity of postgresql :-)
[03:29] <janimo> jbayiley can't the splash be a simple image file?
[03:29] <jbailey> mdz: The .so is just a compiled version of the pngtobogl output.  It has no xpm loader, and putting all the png loader stuff in the binary is a bit heavy for early userspace.
[03:29] <pitti> ogra: pgadmin3 is a really nice one
[03:30] <jbailey> mdz: So /sbin/usplash has the artwork built into it.
[03:30] <pitti> ogra: depends on whether you want a web-based or local application
[03:30] <mdz> eeeeek
[03:30] <Nafallo> s/mysql/phpmyadmin/
[03:30] <mdz> sorry I asked
[03:30] <bddebian> heh
[03:30] <mdz> libpng is all of 146k uncompressed
[03:31] <jbailey> IIRC, it also wants libgd., another 200k
[03:32] <mdz> what does?
[03:32] <mdz> not libpng, certainly
[03:32] <jbailey> Ugh, 12 degrees, that would be why.
[03:32] <sladen> IIRC, libpng only wants zlib
[03:32] <Diziet> Even so, if it has to be a .so, why can't it be just a different .so in the different packages ?
[03:32] <jbailey> Diziet: It can be, but something has to provide the default.
[03:32] <jbailey> Diziet: That's why it's an alternative.
[03:33] <sladen> jbailey: probably for drawing and annotations
[03:33] <Diziet> I don't understand.  Why can't it be a different .so with the same name in different packages, which conflict with each other ?
[03:33] <Diziet> No need for an alternative.
[03:33] <Diziet> You get the one you have installed and if you have none installed you get no usplash or it crashes or something.
[03:33] <jbailey> Diziet: edubuntu-artowkr and ubuntu-artwork don't conflict.
[03:33] <jbailey> One depends on the other.
[03:34] <Kamion> ped_geometry_set_start (geom=0x804a8b0, start=The value of variable 'start' is distributed across several
[03:34] <Kamion> locations, and GDB cannot access its value.
[03:34] <Diziet> So make a new udeb ?
[03:34] <Kamion> er, go gdb
[03:34] <jbailey> Diziet: it's not an installer bit, it's runtime.
[03:34] <Diziet> Or a new .deb ?
[03:34] <jbailey>   png = gdImageCreateFromPng (pngfile);
[03:34] <Diziet> This doesn't seem like rocket science to me.
[03:34] <ogra> Diziet, edubuntu-artwork reuses lots of ubuntu-artwork... it would duplicate the package contents ...
[03:34] <Diziet> ubuntu-artwork-base.
[03:35] <jbailey> Ah, and libgd.so.2 wants freetype png12, jpeg and pthread.  Joy.
[03:35] <Diziet> Or you could have the executable rpath an extra directory where the foobuntu version lives, so the dynamic linker does it for you.
[03:35] <Diziet> Then fooubuntu-artwork would conflict with barubuntu-artwork and you could leave ubuntu-artwork as is.
[03:36] <sladen> jbailey: oh jeeze.  Just write an XPM loader.  Steal it from isolinux/grub/whatever.  It'll be about 20 lines
[03:36] <jbailey> Diziet: This doesn't *Actually* sound easier...
[03:36] <Diziet> Doesn't it ?  Compared to racing around with alternatives at runtime ?
[03:36] <Diziet> rpath is hardly difficult.
[03:36] <sladen> jbailey: or a PCX loader which isn't much bigger
[03:36] <jbailey> You still have the same problem that the artwork has to be in and working before the kernel installs.
[03:37] <Diziet> Come to think of it, are you dlopening it or doing it at executable startup ?
[03:37] <jbailey> Diziet: At least in this case, you can guarantee a sane startup.
[03:37] <jbailey> dlopen
[03:37] <jbailey> The idea was originally to make it so it could take a path.
[03:37] <Diziet> So you could try dlopen(whateverubuntu) and if that fails dlopen(ubuntu-artwork).
[03:37] <Diziet> No need for rpath or alternatives.
[03:38] <Kamion> I don't think we can encode the names of all possible Ubuntu derivatives into usplash!
[03:38] <Diziet> Just make the whateverubuntu-artwork packages put the file in a fixed overriding location.
[03:38] <Diziet> k: No, you misunderstand.
[03:38] <Kamion> oh, fixed override, sure
[03:38] <Kamion> though then kubuntu-artwork-usplash and edubuntu-artwork-usplash would have to conflict - but that's not *so* bad
[03:38] <jbailey> Diziet: Keep in mind that a company IT admin might want to add their own branding to this.
[03:38] <jbailey> Diziet: So it has to be 3 layers.
[03:39] <jbailey> (Isn't this the problem that alternatives were *designed* to solve?)
[03:39] <Diziet> jb: So they can edit the edubuntu-artwork package locally.
[03:39] <jbailey> No, why should they have to?
[03:39] <jbailey> They should be able to load a localised branding package on top.
[03:39] <jbailey> IF they edit packages, they lose upgrade ability.
[03:39] <Diziet> No, alternatives are for if you want to have multiple things installed at once and have a name the refers to the sysadmin's local default.
[03:39] <Diziet> It makes no sense to have multiple usplash screens at once, since only one will ever be used.
[03:39] <jbailey> Which you do in this case.  Default, Distro branding, local branding.
[03:40] <Diziet> You don't have boot arguments saying what branding to have today.
[03:40] <Diziet> If you think that's important have a list and call dlopen on each one.  This is about 5 lines of code and no fighting with complex configuration management.
[03:40] <sladen> Diziet: that's always a possibility :)
[03:41] <Kamion> sladen: with respect, no it's not
[03:41] <Kamion> I'm not going to hack that sort of horribleness into the installer
[03:41] <Kamion> (boot args)
[03:41] <jbailey> Well, more to the point, I don't see what problem it solves.
[03:41] <infinity> jbailey : Local branding can be done with dpkg-divert.
[03:42] <jbailey> infinity: All of the gnome branding works by just dropping files in place that are preferred over the regular ones.
[03:42] <infinity>  /etc/usplash/override-splash.so
[03:42] <jbailey> Yeah.
[03:42] <infinity> (not like it's easy to override until that .so becomes a .png anyway)
[03:43] <jbailey> (ew... You suggested a sofile in /etc)
[03:43] <infinity> So almost no one will do it.
[03:43] <apokryphos> \sh_away: that SUSE userlist theme, btw: giannaros.org/SUSE 
[03:43] <infinity> jbailey : Hey's it's only an .so in name.  I look at it as a PNG with a stub. :)
[03:43] <Diziet> This is very odd.  This hoary install didn't set up sudo for some reason.
[03:44] <jbailey> infinity: Oh, it's really not.  pngtobogl converts into an array of numbers.  Basically, you're loading in a struct with the image in it.
[03:44] <jbailey> Diziet: No root for you!
[03:44] <sladen> Diziet: is it %admin vs. explicit users in /etc/sudoers?
[03:45] <infinity> sladen : Hard to tell without root to look at /etc/sudoers. ;)
[03:45] <Diziet> It's a 100% fresh install.  I haven't touched anything.  Maybe I didn't read one of the install questions closely enough but I don't remember there being one about that.
[03:45] <tseng> is usplash broken universally, or just me?
[03:45] <Diziet> It set a root password instead.
[03:45] <infinity> Diziet : Were you in expert mode?
[03:46] <jbailey> tseng: Depends in which way it's broken.  There are some ways that are known, and others that are specific.
[03:46] <infinity> tseng : Dexribe your brand of breakage.
[03:46] <tseng> jbailey: it seems to change vt
[03:46] <infinity> Describe, too.
[03:46] <Diziet> inf: No.
[03:46] <sladen> Diziet: reboot and select (recovery mode)
[03:46] <tseng> jbailey: then go blank for a bit, then show normal lsb init type stuff
[03:46] <jbailey> Mm, that's a new one for me.
[03:47] <jbailey> I think.  It was up for a little bit, right?
[03:47] <tseng> it was working until yesterday
[03:47] <Diziet> Oh well, I don't really want to debug hoary now.  The display is crap too - it clearly has no idea how to drive this graphics card.
[03:47] <infinity> tseng : When was the last time you upgraded, and when's the last time you regenerated your initramfs?
[03:47] <tseng> infinity: regenerated after 0.1-17
[03:47] <infinity> tseng : Much goofiness with VT switching and tty* device creation has been going on in the last few uploads.
[03:47] <jbailey> There's a -17?
[03:47] <tseng> infinity: it broke after regenerating on -16
[03:47] <mjg59> mvo: Go ahead and do the upload
[03:47] <mjg59> JaneW: Yeah, that's fine
[03:48] <tseng> -15 work, -16 bork
[03:48] <jbailey> Dear Scott, please bring us hct soon.
[03:48] <mvo> jbailey: yes, I uploaded it to get rid of the console-switch flickering
[03:48] <infinity> Hrm, 16 worked for me, I haven't tried 17 yet.
[03:48] <jbailey> mvo: I was only surprised because my working tree was still -14 =)
[03:49] <mvo> jbailey: that's so 24h ago :)
[03:49] <Nafallo> hehe
[03:50] <jbailey> *lol*  Sorry.  Yesterday was evo-exchange, ubuntu-docs, locales troubleshooting, and notification-applet discussion.
[03:50] <jbailey> I'll try to do better next time. =)
[03:50] <azeem> Anybody know whether Ubuntu will be present at the Systems expo in Munich at the end of October?
[03:50] <jbailey> azeem: Unlikely unless a loco team has picked it up.
[03:50] <jbailey> azeem: With UBZ right after it, I doubt anyone has time.
[03:50] <Robinho_Peixoto> galera
[03:50] <mvo> jbailey: heh :) for some reason there is still a problem with the console-font seting. but it's unreleated to usplash it seems, fgconsole --next has only two terminals. do you have any idea about this (assuming it's something with our different early boot stuff)
[03:51] <azeem> who is in charge of the german LoCo team?
[03:51] <mvo> azeem: dholbach is probably a good person to talk to and smurfix
[03:51] <jbailey> azeem: (We have a similar expo here in Montral, but I will be in minisprint with doko)
[03:51] <Nafallo> azeem: https://wiki.ubuntu.com/LoCoTeamList
[03:52] <jbailey> mvo: I don't, sorry.  I've never looked into how the whole console font thing works at all.
[03:52] <Kamion> Diziet: the timezone error probably caused the debconf priority in the installer to be dropped so that you ended up in expert mode, and then you were asked for a root password, which makes the installer decide you probably didn't want to have sudo configured. #9832 coupled with the same kind of thing as #7694.
[03:52] <Diziet> Oh, yes, the installer did stop being wizard-like.
[03:52] <mvo> jbailey: thanks. I'll dig into it (sigh) 
[03:52] <Diziet> That's what you mean by expert mode.
[03:52] <Diziet> It makes sense now.
[03:53] <Kamion> Yes, I think that's probably a bug but one has to be a little careful when fixing it because the debconf-priority-dropping is critical to avoiding infinite loops in some cases.
[03:53] <Kamion> (it means you at least end up at the main menu on error, rather than repeatedly trying the same menu item and failing)
[03:54] <segfault> In the logout screen of GNOME, which package stores the "Hibernate the computer" string?
[03:55] <ogra> segfault, gnome-session
[03:57] <ogra> Kamion, i can confirm 15244 also with a normal X tunnel after logging out and in again...
[03:57] <ogra> ogra@edubuntu:~$ pgadmin3
[03:57] <ogra> Error: Unable to initialize gtk, is DISPLAY set properly?
[03:57] <ogra> updating the bug now
[03:57] <segfault> thanks.
[03:57] <Kamion> ogra: you're in a far better position to debug it than I, I expect
[03:57] <ogra> Kamion, i have no clue about ssh internals 
[03:58] <ogra> but i'll try
[03:58] <\sh> ogra:  want to have a di-524 wlan router? ,-)
[03:58] <Diziet> So this artwork homepage bug.  Who knows what the new filename is ?
[03:58] <Diziet> And is there any explanation for why it has to change ?
[03:59] <ogra> \sh, do you change against a aspire 1520 keyboard without altgr key ? just ordered a new one (3 weeks shipping time :((( )
[03:59] <\sh> ogra: u give up your acer?
[04:00] <ogra> \sh, then i wouldnt order a new keyboard for it
[04:00] <\sh> ogra: Oh it sounds like u ordered a new laptop ,)
[04:00] <ogra> \sh, but probably i'll do that too... to have something thats lighter than 4kg for travelling :)
[04:00] <\sh> hmmm...do we have an ubuntu booth on the systems in munich?
[04:01] <Diziet> jbailey: ping
[04:01] <\sh> ogra: read the announcement of the 100 $ laptop? mako mentioned it, when he moved to his new location :) it's real ;)
[04:01] <jbailey> Diziet: pong
[04:01] <Diziet> jbailey: You say you've made me an html file, but what's its filename ?
[04:01] <Diziet> And are you the person who removed it from ubuntu-artwork ?
[04:02] <Nafallo> ogra: wow. you laptop is even heavier than mine :-P.
[04:02] <ogra> Nafallo, desktop replacement...
[04:02] <ogra> but annoying to travel with...
[04:02] <Nafallo> indeed
[04:02] <Nafallo> 3,6KG here
[04:03] <pitti> ah, \sh: you touched ntlmaps last
[04:03] <jbailey> Diziet: No, it was jdub who removed it.  The file isn't in your current ubuntu-docs package, I wired it up while we were talking this morning.
[04:03] <jbailey> Diziet: Lemme grab the filename.
[04:03] <pitti> \sh: the changelog says you merged ubuntu changes, but there are no ubuntu changelogs
[04:03] <Diziet> So why did jdub remove it ?
[04:03] <Diziet> And why is the filename changing ?
[04:03] <pitti> \sh: can you please look into this and either sync or merge the new Debian version to fix CAN-2005-2962?
[04:03] <Diziet> Can't you not change the filename ?
[04:04] <jbailey> Diziet: We only share a name, dude...  Peering into his thoughts requires possibly more alcohol then I have handy.
[04:04] <\sh> pitti: ntlmaps?
[04:04] <Diziet> jb: :-)
[04:04] <\sh> pitti: yeah will do
[04:04] <pitti> \sh: thanks
[04:04] <jbailey> Diziet: Sure.  It's just that it exists in ubuntu-artwork's space, so it would be a bit weird.
[04:04] <jbailey> IIRC It's /usr/sahre/doc/ubuntu-artwork/ or something like that.
[04:04] <pitti> \sh: please don't trash changelogs on merging, otherwise it is hard to decide whether we can sync
[04:05] <\sh> pitti: I learned it not to drop it anymore :(
[04:05] <Diziet> jb: Um, I think stable names for files are more important than some bureaucracy to do with who owns which namespace.
[04:06] <Diziet> What boggles me is that he seems to have just deleted this file without really thinking that perhaps he should talk to anyone about it.
[04:06] <Diziet> So I assume someone must have talked to him about it, or he wouldn't have done it.  But who ?
[04:06] <jbailey> Diziet: Oh, I don't mean to be bureaucratic.  I just don't feel like fending off the lartings without support from someone else. =)
[04:06] <mdke> Diziet, i saw the discussion
[04:06] <Diziet> mdke: Oh, there was discussion ?
[04:07] <ogra> Diziet, it struck me heavily too yesterday...
[04:07] <mdke> Diziet, yes, it was when sabdfl rewrote the document,
[04:07] <ogra> (edubuntu-artwork replaces this file)
[04:07] <Diziet> mdke: Right ... and that meant it had to move package because ... ?  And move filename because ... ?
[04:07] <Diziet> And was there any explanation for them not communicating with anyone ?
[04:07] <mdke> he expressed surprise that the same doc was in multiple places
[04:07] <Diziet> Or were we just supposed to be hanging around on IRC ...
[04:08] <Diziet> I didn't know it was in multiple places.
[04:08] <ogra> Diziet, obviously :)
[04:08] <jbailey> Diziet: Hanging out on IRC is part of our jobs, mind you.  We're just never permitted to leave =)
[04:08] <mdke> the discussion was not totally full iirc
[04:08] <ogra> (but i hang around 24h here and didnt see the discussion)
[04:08] <mdke> and it was not in this channel
[04:08] <mdke> it was in -doc
[04:08] <jdub> pitti: pong
[04:08] <mdke> ah good
[04:09] <Diziet> The scrollback is less than informative.
[04:09] <Diziet> Ah, not in this channel.
[04:09] <ogra> imho a mail to -devel would have been nice
[04:09] <mdke> jdub, i was trying to explain the absence of about-ubuntu oin the artowk package, you remember the discussion?
[04:09] <Diziet> Well, I don't hang around in -doc.
[04:09] <Diziet> Ah, jdub, hello !
[04:09] <ogra> me neither
[04:10] <jdub> mdke: yeah
[04:10] <jbailey> But there's fun and excitement in there!
[04:10] <jbailey> All the cool kids are doing it...
[04:10] <Diziet> jdub: So can you explain what was done and why ?
[04:10] <jdub> Diziet: the about ubuntu html is no longer shipped in ubuntu-artwork, it's built and shipped as part of ubuntu-docs
[04:10] <Diziet> And why no-one mailed ubuntu-devel with a quick not to tell us what to expect :-).
[04:10] <pitti> jdub: the last u-artwork package seems to make Breezy fall apart into pieces
[04:10] <Diziet> jdub: By `is' you mean `will be'.
[04:10] <pitti> jdub: the usplash artwork is missing as well
[04:11] <infinity> jbailey / mvo : Either of you doing a usplash update in the next while?
[04:11] <Diziet> Since it's not currently on anyone's system.
[04:11] <jdub> Diziet: it is done that way because there's no point maintaining it in two places, and it's owned by the doc team
[04:11] <Treenaks> hm, EKEYBUK
[04:11] <jbailey> infinity: I need to.
[04:11] <ogra> pitti, not here
[04:11] <Diziet> Certainly I don't mind which package it's in.
[04:11] <jbailey> infinity: It's just not obvious to me what update to make atm.
[04:11] <jdub> pitti: ubuntu-artwork has never owned the usplash artwork, so if jbailey's depending on me to put it in, then i need to know about it ;-)
[04:11] <pitti> jdub: oh, ok
[04:11] <infinity> jbailey : Can I get you to patch mvo's chvt magic a bit, to get rid of one icky case of text flicker?
[04:11] <Diziet> jdub: Well, wasn't it in a previous version of ubuntu-artwork ?
[04:12] <jdub> pitti: i was of the understanding that usplash had the ubuntu artwork in it by default
[04:12] <jdub> Diziet: yes
[04:12] <Diziet> So you just deleted it.
[04:12] <ogra> pitti, reconfigure your linux-image ;)
[04:12] <jdub> Diziet: if you mean the about ubuntu html
[04:12] <jdub> Diziet: yes
[04:12] <jbailey> infinity: Sure, mind sending it by email?  This channel is a bit fast paced at the moment to track TODOs here.
[04:12] <pitti> ogra: that didn't help
[04:12] <ogra> oh
[04:12] <ogra> helped here
[04:12] <jbailey> pitti: Eh?  usplash contains the default ubuntu artwork.
[04:12] <Diziet> I see.  Did it occur to you that something might be using it ?
[04:12] <pitti> jbailey: ah, I had to call update-alternatives for the artwork file
[04:12] <jdub> Diziet: i knew very well that it was the default homepage, that was its entire purpose in life
[04:13] <pitti> jbailey: one of the recent updates broke the alternatives link
[04:13] <pitti> jbailey: (I wasn't the only one)
[04:13] <Diziet> Right.  So you deleted the default homepage.  How were you expecting the default homepage to still be provided ?
[04:13] <jbailey> pitti: The usual problem is that the kernel installed before the update-alternatives ran in the the usplash postinst.
[04:13] <jdub> Diziet: by ubuntu-docs
[04:13] <jbailey> pitti: Did you still have to run it yourself after?
[04:13] <pitti> jbailey: I see
[04:13] <Diziet> So of course you edited ubuntu-docs to make it provide it ?
[04:13] <pitti> jbailey: yes, I had; without running it, reconfiguring the linux image didn't help at all
[04:13] <jdub> Diziet: no - jbailey maintains ubuntu-docs
[04:14] <jbailey> Diziet: Objection leading the witness, when he clearly did not. =)
[04:14] <jbailey> pitti: Argh.
[04:14] <Diziet> So you just decided you'd break it, not tell anyone, and leave the rest of us to figure it out ?
[04:14] <Diziet> Can you tell yet that I'm annoyed ?
[04:14] <seb128> slomo: ?
[04:14] <jbailey> pitti: Do you still have a system in that state?  I'd love to know why the usplash postinst didn't do it?
[04:14] <jdub> i'm not sure your 50 questions method is really very useful when the answers are pretty obvious, and we can just get on with our lives and fix things
[04:14] <pitti> jbailey: btw, is it really the right thing to manually reconfigure the linux image to update the ram disk? can't some postinst do this automatically?
[04:14] <Diziet> So another question: why does the filename have to change ?
[04:15] <pitti> jbailey: no, unfortunately I already updated-alternatives
[04:15] <infinity> mvo : What's your one outstanding bug with console setup?
[04:15] <infinity> mvo : I'm touching that area right now, so maybe it'll jump out at me.
[04:15] <torkel> jbaily: shouldn't the eq in 'dpkg --compare-versions "$2" eq 0.1-12' in usplash.postinst be a ge?
[04:15] <Diziet> My point with the 50 questions is to try to suggest that there might be less, erm, confusing, ways of going about things.
[04:15] <jbailey> pitti: Ther'es an update-initramfs tool, but it requires that it was used to create the alternative in the first place.  The kernel folks weren't comfortable with dropping it in after preview, so it's deferred until dapper.
[04:15] <ogra> Diziet, just more communication is enough
[04:16] <jdub> Diziet: perhaps saying, "hey, there are less confusing ways of doing things" would have been more effective communication
[04:16] <jbailey> pitti: For amusement you can call "update-alternatives -u -t".  The -t is for "takeover" (as in, don't care  that I didn't make t his).  Just -u works after that.
[04:16] <jdub> regardless, we break and fix, it's not like it's a crisis
[04:16] <Diziet> Well, I didn't want to diss you until I'd heard your excuse.  Anyway, I've made that point.
[04:16] <Diziet> Thank you.
[04:16] <Diziet> What about the filename ?
[04:16] <Diziet> I think the filename should not change.
[04:16] <pitti> jbailey: ok, so Dapper won't require us to manually build the ram disk any more? nice
[04:17] <ogra> Diziet, i was trapped several times by such things in this release cycle, i think we should have a BOF at UBZ to sort out such things in advance for dapper
[04:17] <jbailey> pitti: Right. =)
[04:17] <jbailey> pitti: And in fact, the idea is that tools that involve the initramfs can just do the updates themselves all they want and safely.
[04:17] <jdub> Diziet: that's up to jbailey - perhaps a symlink to the /u/s/doc location would make sense
[04:17] <jbailey> pitti: It does a sha1sum check to make sure the user didn't alter the file without the tool.
[04:17] <Diziet> jdub, jbailey: Are you both happy that ubuntu-docs can ship the file at its previous location ?
[04:18] <jbailey> Diziet: Sure.  Am I shipping the old file, or the HTMLized about-ubuntu?
[04:18] <pitti> Diziet: this is almost a requirement, given that we can't change user profiles
[04:18] <jdub> jbailey: html-ised about ubunut
[04:18] <jdub> jbailey: a symlink would do it
[04:19] <ogra> jdub, there is a logo in the html file...
[04:19] <ogra> that needs to be placed properly then
[04:20] <jdub> ogra: hmm - if it's a directory symlink, that's no problem; if it's not a directory symlink, the logo would have to use a fully qualified path - that might be
[04:20] <ogra> yup
[04:21] <pitti> seb128, _mvo_: I moved cdrdao into the "Approved, but not yet ready to promote" queue
[04:21] <Robinho_Peixoto> pitti: and the update-manager. This don't has in Rosetta
[04:21] <jdub> jbailey: whatcha think?
[04:22] <seb128> pitti: why "not ready to promote"?
[04:22] <pitti> seb128: it needs to be germinated
[04:23] <pitti> seb128: i. e. either a reverse-dependency or a seed
[04:23] <seb128> pitti: right, every package does
[04:23] <Diziet> jb: You should ship whatever file ought to be the default startup page.
[04:23] <seb128> pitti: you said that like it was a special case :)
[04:24] <pitti> seb128: in this case we should seed it, a dependency does not make sense (this is more for libs)
[04:25] <infinity> mvo : unping, I found your problem in backscroll, looking into it now.
[04:26] <seb128> pitti: agreed
[04:26] <_mvo_> infinity: my problem?
[04:27] <mvo> infinity: sorry, my network is playing silly bugger
[04:27] <jdub> http://news.com.com/Ubuntu+carves+niche+in+Linux+landscape/2100-7344_3-5886194.html
[04:28] <Treenaks> carves?
[04:28] <Treenaks> not blasts?
[04:28] <jdub> haha
[04:29] <jbailey> carved with a pile driver.
[04:29] <jbailey> It's a different sort of art.
[04:29] <Treenaks> jbailey: yay artwork-team
[04:29] <bddebian> Heh
[04:29] <ogra> jdub, bah, they mention ltsp but not edubuntu... boycott !
[04:30] <lamont-away> Setting up scrollkeeper (0.3.14-10ubuntu1) ...
[04:30] <lamont-away> Rebuilding the database. This may take some time.
[04:30] <lamont-away> //usr/share/gnome/help/about-ubuntu/C/about-ubuntu.xml:5: I/O warning : failed to load external entity "/usr/share/gnome/libs/global.ent"
[04:30] <lamont-away> %globalent;
[04:30] <lamont-away>            ^
[04:30] <lamont-away>  %globalent; 
[04:30] <lamont-away>             ^
[04:30] <lamont-away> neato
[04:30] <jbailey> lamont-away: Known bug.
[04:30] <mjg59> Could anyone with working laptop suspend who hasn't already done so email me the output of dmidecode?
[04:31] <jbailey> mjg59: To which address?
[04:31] <jbailey> do you have an @ubuntu.com yet?
[04:31] <mjg59> Nope
[04:31] <mjg59> mjg59@srcf.ucam.org is good
[04:33] <jbailey> mjg59: Is it worth notign that I had to enable it in /etc/defaults/acpi* to see it in gnome?
[04:34] <pitti> jbailey: I thoughth the latest libc would support Kurdish locale now?
[04:35] <jbailey> pitti: Remind me the locale string of it?
[04:35] <pitti> jbailey: ku_SOMETHING
[04:35] <pitti> jbailey: what I aim at is that there is no ku_* in /usr/share/i18n/SUPPORTED
[04:36] <mjg59> jbailey: No, that's normal
[04:36] <mjg59> jbailey: And it works fine?
[04:36] <jbailey> mjg59: Yup.
[04:36] <jbailey> mjg59: I used it all through UDU even.
[04:36] <Treenaks> mjg59: what info did you and keybuk need for the isapnp junk on the 380ed (re: crystal sound)
[04:37] <wasabi_> hey jbailey, i was thinking about something else that would be cool in the init scripts. I think I'll work on it this weekend.
[04:37] <wasabi_> root=/dev/hda:/root.img   
[04:37] <jbailey> pitti: Can you add a note to that effect to 13871 ?
[04:37] <ogra> jdub, how does Sam Pohlenz get into that interview ? 
[04:37] <mjg59> Treenaks: The contents of /sys/bus/pnp/devices/*/id would be good
[04:37] <jbailey> wasabi_: You know that random changes are not-for-breezy at this point, yes? =)
[04:37] <wasabi_> Oh sure.
[04:37] <mjg59> Treenaks: But I suspect that it's not going to be complete
[04:37] <jbailey> Oh good. =)
[04:38] <wasabi_> But I want to implement it anyways. ;0
[04:38] <Treenaks> mjg59: that's in the bug already
[04:38] <wasabi_> Just wanted to see what you thought about it.
[04:38] <mjg59> Treenaks: Yeah
[04:38] <wasabi_> Since you're in that space a lot.
[04:38] <jbailey> wasabi_: So explain to me what this does?
[04:38] <Treenaks> mjg59: I'm going to do a -current install on it
[04:38] <lamont-away> seb128: you around?
[04:38] <Treenaks> mjg59: (*zzz*)
[04:38] <lamont-away> gtk-smooth-engine
[04:38] <wasabi_> The idea is to mount / as a loopback file on another fs. My specific use case is for my embedded flash system.
[04:38] <wasabi_> I want to be able to "flash" the system with a new image, and I want it to be as simple as on a Cisco or something. 
[04:38] <wasabi_> You just drop a new .bin file into a directory on the flash disk and reboot.
[04:39] <jbailey> wasabi_: Okay.  Try hooking that in /usr/share/initramfs-tools/scripts/local-top
[04:39] <wasabi_> Also prevents the scattered writes that would be caused by upgrading it by extracting an entire file system.
[04:39] <wasabi_> Which is hard on flash.
[04:39] <jbailey> Right.
[04:40] <jbailey> The trick is going to be that you'll need to somehow feed back an altered root variable.
[04:40] <seb128> lamont-away: yeah, I'll do that now
[04:40] <wasabi_> Yeah, I'm very unsure how that works. ;)
[04:40] <lamont-away> danke
[04:40] <jbailey> wasabi_: Well, right now it doesn't.  You're in a subshell.
[04:40] <jbailey> I didn't want all of the variables from everyone's scripts messing things up.
[04:40] <wasabi_> The pivot_root process is pretty unknown to me.
[04:41] <jbailey> wasabi_: So instead, maybe set root= to something like /dev/loop0 and just set that up?
[04:41] <jbailey> There's no pivot_root anymore. =)
[04:41] <pitti> jbailey: ah, sorry, the Kurdish locale was about installing belocs, not adding to libc locales
[04:41] <wasabi_> Oh.
[04:41] <jbailey> pitti: Right.
[04:41] <jdub> ogra: who *is* sam pohlenz?
[04:41] <jbailey> pitti: BTW, I added your name to LocalesThatDontSuck
[04:42] <wasabi_> Yeah, good idea.
[04:42] <jbailey> pitti: For a UBZ bof. =)
[04:42] <ogra> jdub, my SoC student...
[04:42] <wasabi_> root=/dev/loop0 loop0=/root.img loopfs=/dev/hda
[04:42] <wasabi_> something like that
[04:42] <jdub> ogra: aha
[04:42] <pitti> jbailey: thanks, I don't want to miss that
[04:42] <jbailey> pitti: I need to fill in the details, but we have two conflicting proposals.
[04:42] <Treenaks> wasabi_: sounds UML-ish
[04:42] <jbailey> pitti: 1) Use belochs.  2) Feed belochs into rosetta, and use rosetta.
[04:42] <ogra> jdub, he wrote a 100 line pygtk tool to load win drivers into ndiswrapper
[04:43] <jdub> ogra: yeah, well, supposedly i'm "head" of b&cd :-)
[04:43] <ogra> heh
[04:43] <ogra> i love well researched articles
[04:43] <ogra> lol
[04:43] <jbailey> wasabi_: Right.  It's easy enough to get those variables out.
[04:43] <jdub> ah, it was easy to misinterpret
[04:43] <wasabi_> Cool.
[04:43] <jdub> i only did the interview with stephen last night
[04:43] <wasabi_> That'll be neat to have built in.
[04:43] <jdub> i just don't want jane to drive over me in her tank
[04:43] <jbailey> wasabi_: And then the only thing you have to do is setup that the root device is mountable in local-top
[04:44] <wasabi_> Yeah.
[04:44] <jbailey> wasabi_: You can do that all without altering initramfs-tools in anyway.
[04:44] <wasabi_> Yeah, just some additional local-top scripts
[04:44] <wasabi_> what does local-top stand for anyways?
[04:44] <jbailey> And in fact if you're quick, clever, and beg ogra and dholbach, you might be able to do it for breezy. =)
[04:44] <pitti> jbailey: you mean like ship the locales itself into the langpacks? Highly interesting and flexible idea
[04:45] <wasabi_> jbailey, it's not totally useful without ro file system support.
[04:45] <pitti> jbailey: although installing a locale should not require installing the langpack
[04:45] <jbailey> wasabi_: There are two layers of hooks.  The init- series, which is always run for everyone.  Then the local- and nfs- series which is run depending ont he mode you select.
[04:45] <wasabi_> Hmm. Mode?
[04:45] <wasabi_> How do you change modes?
[04:46] <jbailey> wasabi_: Change the BOOT= line in /etc/mkinitramfs/initramfs.conf
[04:46] <jbailey> wasabi_: But they're just sourced in scripts.  You could write your own without altering the initramfs-tools package.
[04:46] <wasabi_> oh interesting... never saw this file.
[04:46] <wasabi_> Yeah agreed.
[04:46] <jbailey> wasabi_: So each of those then has hooks where you can just drop things and expect it to work.
[04:46] <wasabi_> That's pretty neat.
[04:47] <jbailey> pitti: Right.  We could generate the entirety of locales from the rosetta export, though.
[04:47] <pitti> jbailey: yes, that'd rock
[04:47] <jbailey> The only place where I don't like the idea is in a place that doesn't matter for us: Default character encoding.
[04:47] <wasabi_> I'll get this working, then figure out someway to do unionfs or other fake-writable support.
[04:47] <jbailey> Since we set it to utf-8 for everyone, we've pretty much sidestepped the question.
[04:47] <pitti> doko: I gave ooo1 on ppc a quick test. All apps start and don't entirely fall apart into pieces; anything I should check for in particular?
[04:48] <jbailey> wasabi_: Right, ask mdz about unionfs love.
[04:48] <wasabi_> I'm interesting in getting this working, allowing a rw file system, and then having a save script which saves only certain files in /etc.
[04:48] <pitti> jbailey: yes, and we should aim to get rid of non-UTF-8 locales whereever possible
[04:48] <jbailey> wasabi_: But *afteR* breezy releases. =)
[04:48] <wasabi_> For an embedded firewall for instance.
[04:48] <wasabi_> You'd save specific config files, but reload the rest from scratch
[04:48] <jbailey> pitti: I'm not convinced that's correct for most places in the world, but it's a vision I can buy into. =)
[04:48] <doko> pitti: maybe open a document you did save with OOo2 in the OpenDocument format, but nothing more.
[04:49] <pitti> doko: alright, will do
[04:49] <infinity> mvo : Dude, I'm looking for those missing VCs, and I can't figure out WTF...
[04:49] <jbailey> wasabi_: For extra fun, remember that cpio.gz files can be overlayed.
[04:49] <infinity> mvo : I'm trying to figure out if I should blame jbailey (cause that's fun in general), or sysvinit (which is responsible for spawning them..)
[04:49] <jbailey> wasabi_: So you can ship a generic piece, and then a configuration file on top of it for choosnig the image.
[04:49] <wasabi_> Woh.
[04:49] <wasabi_> Yeah... also, you could allow third party addons
[04:49] <jbailey> So you can switch root images cheaply.
[04:50] <wasabi_> But overlaying an addons/ dir
[04:50] <wasabi_> by
[04:50] <wasabi_> Or something nifty.
[04:50] <jbailey> What bootloader does the device use?
[04:50] <wasabi_> grub or lilo. it's a regular PC.
[04:51] <jbailey> Lovely.  The hack you want to make this sexy is about 15 lines to grub. =)
[04:51] <wasabi_> I just need to find some mini-itx board or something which supports ECC ram.
[04:51] <wasabi_> Because I think that it needs at least that.
[04:51] <jbailey> mini-itx boards are frigteningly expensive.
[04:52] <jbailey> And all the ones I've looked at recently still have VGA on them.
[04:52] <jbailey> Kinda annoying.
[04:52] <wasabi_> Yeah that's unfortunate.
[04:52] <wasabi_> Well, I'm building a variaty of little devices for this company.
[04:52] <jbailey> Ah, cool.
[04:52] <wasabi_> First the presentation thing, second probably a firewall.
[04:52] <jdub> oh, we are on the front page of news.com.com too
[04:52] <mvo> infinity: right, thanks! I banged my head against it this morning too, I have no clue right now
[04:52] <wasabi_> We have a ceiling mounted projector, I'm setting a mini itx case with it up on the ceiling.
[04:52] <wasabi_> no fan, no hd
[04:52] <mvo> infinity: I think iz gtk bug
[04:52] <jbailey> wasabi_: Ah, cool.  On ubuntu?
[04:53] <wasabi_> Yup
[04:53] <wasabi_> It'll boot, and display a "choose the computer you want to see" list from the interface
[04:53] <wasabi_> And anybody can remote to their own desktop
[04:53] <jbailey> jdub: wasabi might have a success story for you.
[04:53] <wasabi_> Not yet. ;)
[04:53] <infinity> mvo : I wish. :)
[04:53] <wasabi_> My first big task with it will be getting it working with bluetooth in some way.
[04:54] <wasabi_> So the CEO and sales can walk into the conference room, and have their computer name pop up on the projector, and just choose it.
[04:54] <wasabi_> and see their desktop
[04:54] <wasabi_> laptop, bluetooth, etc.
[04:54] <zyga> re
[04:54] <zyga> artwork-people: neat background :)
[04:55] <Diziet> OK.  I now know where to change firefox so that we can have a translated start page.
[04:55] <wasabi_> jbailey, http://www.logicsupply.com/product_info.php/cPath/79/products_id/444    cute little rack mount
[04:56] <Diziet> These translations are in langpacks, aren't they ?
[04:56] <pitti> doko: I tested ooo2 -> ooo1 with .odt: three fonts, various formatting, a picture; works flawless
[04:56] <Diziet> Anyone familiar with langpack stuff here ?
[04:56] <zyga> Diziet: pitti should be :)
[04:56] <pitti> Diziet: that's mine, mine, mine!
[04:57] <zyga> pitti: yourrr precioussss
[04:57] <zyga> ;] 
[04:57] <Diziet> Right.  So.  The file /usr/share/ubuntu-artwork/home/index.html is going to be part of ubuntu-docs.
[04:57] <pitti> Diziet: there are no translations of the ffox start page ATM
[04:57] <jbailey> I was more thinking the seagulls from Finding Nemo. =)
[04:57] <Diziet> I have a scheme in mind for supporting translations.
[04:57] <jbailey> pitti: There are. =)
[04:57] <Diziet> It works like this:
[04:57] <Diziet> You put a translation of this file in the relevant langpack, called /usr/share/ubuntu-artwork/home/index-LOCALE.html
[04:58] <doko> pitti: I didn't do more on i386 as well. that should be sufficient.
[04:58] <pitti> Diziet: where do I get them from?
[04:58] <Diziet> Well, translators translate things, right ?
[04:58] <pitti> doko: ok, thanks; glad to get to know your extensive and sophisticated hard test suites :-)
[04:59] <jbailey> Diziet: They're in rosetta as ubuntu-docs package, pofile about-ubuntu
[04:59] <zyga> pitti: I'd love to contribute index-pl.html for you
[04:59] <Diziet> I don't understand how all that stuff works, but you must have ways of getting stuff from translators.
[04:59] <Diziet> Sorry, I don't mean they want to go into langpack-nl.
[04:59] <Diziet> They want to go into mozilla-firefox-locale-nl-nl
[04:59] <pitti> Diziet: we don't touch these automatically ATM
[05:00] <pitti> Diziet: for now, just putting them into ubuntu-artwork as well wouldn't be too wrong, would it?
[05:00] <Diziet> And the file /usr/lib/mozilla-firefox/extensions/{83532d50-69a7-46d7-9873-ed232d5b246b}/chrome/nl-NL.jar (!) contains a member locale/browser-region/region.properties which should be changed too.
[05:00] <jbailey> Diziet: They can only really come from ubuntu-docs for now.
[05:00] <Diziet> Yes, unfortunately having firefox automatically find these files at runtime is quite hard.
[05:00] <jbailey> Diziet: All the stuff is stored in there, it's not langpack driven right now.
[05:00] <doko> pitti: dude! ;-)
[05:01] <wasabi_> wonder what the best way to create/maintain this flash based image is.
[05:01] <wasabi_> guess a chroot.
[05:01] <zyga> pitti: that can be a symlink right?
[05:01] <Diziet> The reason it has to go into the same package, mozilla-firefox-locale-nl-nl, is so that the correct start URL can always be provided.
[05:01] <Diziet> The start URL is in that package.
[05:02] <zyga> pitti: ah - no wait - it's per-user
[05:02] <Diziet> That is, if you run in the nl locale, you get the start page from that package instead of the one in firefox.
[05:02] <jbailey> Diziet: What about people who use multiple langpacks and locales on the system.
[05:02] <Diziet> That's OK.
[05:02] <Diziet> It'll work just right.
[05:02] <jbailey> My wife and I share a computer and don't use the same language.
[05:02] <pitti> Diziet: right, I know, but for now we manually have to update all the m-f-locale-* packages
[05:02] <Diziet> If the translated document lives together with the configuration in the right .jar in the right -locale- package.
[05:03] <pitti> Diziet: well, actually, luckily there is m-f-locale-all, so it's only one source package
[05:03] <pitti> Diziet: and there is already a script in debian/rules which updates the start page
[05:03] <Diziet> Which takes tarballs from rosetta, or something ?
[05:03] <pitti> Diziet: should be easy to adapt that to per-lang
[05:03] <pitti> Diziet: no, it's not connected to Rosetta right now - we plan that for Dapper
[05:03] <Diziet> Ah.
[05:03] <jbailey> Diziet: Does epiphany also inherit that same thing?
[05:03] <Diziet> Right.  So the question remaining is, do we want to try to support this for breezy at all ?
[05:03] <Diziet> jb: I don't know.
[05:04] <Diziet> It may be that epiphany reads a different file.
[05:04] <Diziet> In which case firefox would see the right one and epiphany the English one.
[05:05] <Diziet> The main question I suppose is do we have the ability to get enough translations done, and incorporated, to make it worth faffing with the  machinery in m-f-locale-all ?
[05:06] <Diziet> jb: Have you uploaded that new ubuntu-docs yet, BTW ?
[05:06] <jbailey> Diziet: I have 28 translations of about-ubuntu
[05:06] <pitti> mvo: ping
[05:06] <Diziet> And that's the file we're using for English ?
[05:06] <jbailey> Diziet: No, because I'm not clear that everyone has agreed on the right thing.
[05:07] <Diziet> OIC.
[05:07] <jbailey> Diziet: That's the file that jdub has asked me to use.
[05:07] <mvo> pitti: pong
[05:07] <Diziet> Right.  Is that the one that sabdfl rewrote ?
[05:07] <janimo> jbailey, is usplash+pluggable artworks going ahead?
[05:07] <infinity> mvo : Debian #271013
[05:07] <pitti> mvo: in the update notifications, will "Description-pt_BR" work?
[05:07] <jbailey> Diziet: Not that I see.
[05:07] <mvo> pitti: yes
[05:07] <infinity> mvo : That's the bug, and it's nothing new, apparently.
[05:07] <Diziet> jb: So we have a disagreement between jdub and sabdfl, or what ?
[05:08] <jbailey> janimo: I'm also not clear where that ended up.
[05:08] <infinity> mvo : init fires up the consoles AFTER it runs the init scripts, not before.
[05:08] <jbailey> Diziet: Where I think that it's left is that you were trying to figure out how to make FF find the translated files by default.
[05:08] <jbailey> Diziet: If you have a way, I'll upload all te translations.
[05:08] <wasabi_> jbailey, how are kernel parameters accessed from these scripts?
[05:08] <jbailey> Diziet: If not, then I will upload English only.
[05:08] <wasabi_> I notice the use of the ROOT env var
[05:08] <Diziet> Right.  I know how to do that, but the translations have to be in the m-f-locale-all package.
[05:08] <wasabi_> Are they all exported like that?
[05:09] <jbailey> wasabi_: Look at init-top/usplash
[05:09] <jbailey> Sorrry, scripts/init-top/usplash
[05:09] <mvo> infinity: gosh, 1y old? 
[05:09] <jbailey> wasabi_: Only special ones are extracted for common use.  You can fetch your own this same way.
[05:09] <pitti> fabbione: what is "Language pack reorganization" on Italic?
[05:09] <jdub> jbailey: the html output of the doc team's about ubuntu (which is what mark wrote) is what should be used
[05:09] <infinity> mvo : Well, the bug is older, that's just the first one filed. :)
[05:09] <Diziet> Because m-f-locale-all contains a file which specifies statically the start URL.
[05:10] <jbailey> wasabi_: If you're feeling extra sexy, please feel free to write a function to extract them in a pretty way.
[05:10] <infinity> mvo : It's always been like this, this isn't a regression.
[05:10] <infinity> mvo : It's just "the way it is".
[05:10] <Diziet> We can easily make m-f-locale-all generate the file right depending on whether there's a translation at build-time.
[05:10] <infinity> mvo : So, it's hardly a breezy blocker, though irritating.
[05:10] <Diziet> But it's much harder to make all this happen on the user's system at runtime.
[05:10] <jdub> Diziet: that'd be kinda cool, but if it's too hard, don't worry about it
[05:10] <wasabi_> Ahh, proc command line.
[05:10] <jbailey> jdub: Ah, okay.  I didn't know who wrote it. =)  All I have is what's sitting int he breezy branch of the docteam's svn. =)
[05:10] <jdub> mdke: around?
[05:11] <wasabi_> Might make sense to have them all exported as env vars by default, no?
[05:11] <jbailey> So there's just the upload all of the translations, or just the English question left.
[05:11] <jbailey> And then I'll send 'er off.
[05:11] <jbailey> (Well, I'll fix some other bugs on the way first, probably)
[05:11] <infinity> mvo : If we're willing to assume our users don't mess much with /etc/inittab, I can just do the console setup on vc1 through vc6, without first verifying that the consoles exist.  It doesn't hurt anything, and when getty spawns, the console should be in UTF-8 mode.
[05:11] <Diziet> jb: Um, did you see what I said up there ^ ?  These translations are currently in ubuntu-doc, you say ?
[05:11] <Diziet> But I'm saying they have to be in m-f-locale-all.
[05:12] <jbailey> Diziet: I did.  I'm telling you that they're docbook files in ubuntu-docs.
[05:12] <jbailey> I have no way of getting them into that other package.
[05:12] <jbailey> Not in any sort of sane way.
[05:12] <jbailey> And I haven't seen pitti buy into the idea yet.
[05:12] <jbailey> This is why I haven't randomly uploaded. =)
[05:12] <mvo> infinity: that sounds good to me, I think we should fix it (even if the fix is a bit hackish)
[05:12] <jdub> Diziet: doesn't m-f-locale-all just care about urls?
[05:12] <Diziet> Right.  Just a mo, realconv.
[05:13] <infinity> mvo : Definitely hackish, but I don't think it'll hurt.
[05:13] <pitti> mvo: btw, can you please ping me when I can upload the new hal and dbus with the reboot notice?
[05:13] <infinity> mvo : The flicker will be more noticeable, but I'll test that.
[05:13] <infinity> mvo : Do you have a bug number for this, or just fixing it cause it drives you batty?
[05:13] <mvo> infinity: thanks. I have a greek install here that I can use for testing too
[05:13] <pitti> mvo: btw, I don't know whether you saw yesterday's discussion - can you provide a small wrapper around this?
[05:13] <mvo> infinity: we don't have a bugnumber for that particular problem, but you could import the deb bug
[05:14] <infinity> mvo : No point importing it just to close it.
[05:14] <mvo> pitti: yes, the wraper will be called "notify-reboot-required"
[05:14] <mvo> infinity: right
[05:14] <pitti> mvo: so that postinsts merely do "if [ -x /usr/share/u-n/trigger-reboot ]  && /u/s/u/t-r"
[05:14] <pitti> mvo: ah, nice
[05:14] <infinity> mvo : Want to mail me some UTF-8 text, so I know my consoles are working as advertised? :)
[05:14] <pitti> mvo: in /usr/share/u-n?
[05:14] <mvo> pitti: I can upload it now, but I want to fix some smaller issues along the way
[05:15] <pitti> mvo: well, if I know the exact path, I can already prepare packages
[05:15] <mvo> pitti: /usr/bin/
[05:15] <pitti> mvo: and even upload them
[05:15] <pitti> mvo: uh, why /usr/bin/? that's not really supposed to be called by admins
[05:15] <pitti> mvo: erm, users
[05:15] <Kamion> pitti: say what?
[05:15] <Kamion> I'd have some difficulty using a system as an ordinary user without using /usr/bin/
[05:16] <mvo> pitti: right, /usr/sbin/? /usr/share/u-n?
[05:16] <infinity> Kamion : Shh, English isn't always easy.
[05:16] <jbailey> infinity: I can forward you korean spam. =)
[05:16] <Diziet> jb: no sane way> Hmmm, I see, yes.
[05:16] <seb128> pitti: so dholbach is telling me that you took over gamin and its bugs because you like it? :)
[05:17] <Kamion> infinity: /usr/share/doc/console-tools/examples/ ?
[05:17] <pitti> Kamion: "that" == the update notice trigger script
[05:17] <Kamion> infinity: oh, I thought he was confusing /usr/bin with /usr/sbin
[05:17] <pitti> seb128: only if he takes firefox
[05:17] <jbailey> Diziet: For dapper, all of this will be langpack driven.  Martin and I sorted that out yesterday.
[05:17] <dholbach> hahaha
[05:17] <dholbach> no :)
[05:17] <pitti> mvo: I'd prefer /usr/share/u-n
[05:17] <seb128> pitti: deal
[05:17] <jbailey> Diziet: I even now have a basic script that does the extract and generates all of the needed pieces.
[05:17] <Diziet> jb: Right.
[05:17] <mvo> infinity: send you some nice 's
[05:17] <dholbach> simple question, simple answer :)
[05:17] <seb128> dholbach: you won firefox
[05:17] <Diziet> m-f-locale-all does indeed only care about URLs.
[05:17] <seb128> pitti: you won gamin
[05:17] <jbailey> 
[05:17] <seb128> seb128: you won some sleep :p
[05:18] <Kamion> infinity: try http://www.cl.cam.ac.uk/~mgk25/ucs/examples/UTF-8-demo.txt
[05:18] <Diziet> But we can't just have m-f-locale-all generate those URLs for every language because we'd have to keep them in synch.
[05:18] <infinity> I haven't the first clue how to type or tab-complete that...
[05:18] <Diziet> I mean, the list of translations.
[05:18] <Kamion> infinity: that's why there's a README.strange-name in the same directory :)
[05:18] <Diziet> Oh, just a mo, here's an idea:
[05:18] <pitti> seb128: but you know that I then will replace gnome with fvwm, OOo with latex, nautilus with mc, and X with a nice fb console? :-)
[05:18] <Diziet> This file (the index.html) is quite small.
[05:18] <jbailey> Diziet: Right, this might be unsolvable in this timeframe.
[05:18] <Diziet> And we could use symlinks.
[05:18] <Diziet> Here's my idea:
[05:18] <HiddenWolf> mvo, is xscreensaver your package?
[05:18] <infinity> Kamion : Ahh. ;)
[05:18] <pitti> seb128: talking to yourself?
[05:18] <jbailey> Diziet: FF could also do with some hacking to make it do language negotiation on file:/// URIs
[05:18] <Diziet> We make a complete list of all locales we've ever heard of.
[05:18] <seb128> pitti: GetRideOfTheDesktop? :)
[05:19] <Diziet> jb: You must be joking.
[05:19] <seb128> pitti: yeah, that's friday :p
[05:19] <pitti> seb128: exactly
[05:19] <Diziet> Then, we have the m-f-locale-all package always rewrite the homepage URL to include the locale, iff the locale is in that list.
[05:19] <Diziet> And we have the ubuntu-docs package provide _all_ of those files for every locale.
[05:19] <Diziet> If there is no translation, it provides a symlink to the English.
[05:19] <pitti> Diziet: see? breezy's design is hopelessly deprecated even before it is released :-)
[05:20] <wasabi_> jbailey, what is the appropiate place in the initramfs to mount to device that holds the image? Is there a /mnt or what?
[05:20] <jbailey> wasabi_: ${root}, I think. =)
[05:20] <infinity> Hrm.  Our default console faunt is pretty shit for UTF-8...
[05:20] <infinity> faunt?
[05:20] <infinity> font.
[05:20] <wasabi_> ${root}?
[05:20] <Diziet> Since this design is ephemeral, for this `complete list' we could just take the list of translations that are currently available plus a few that look plausible, or something, and copy it about manually this once.
[05:20] <infinity> Wow, what happened there, I wonder.
[05:20] <mako> \sh_away: the $100 laptop isn
[05:20] <pitti> infinity: sure that you started unicode_start?
[05:21] <mako> \sh_away: the $100 laptop isn't real yet.. but it has a design and work is certainly progressing on it here
[05:21] <wasabi_> I thought ROOT was the path to /dev/loop0
[05:21] <wasabi_> I mean, the temporary fs, holding the .img file itself.
[05:21] <jsgotangco> hey mako nice seeing you again
[05:21] <lifeless> mako: ola
[05:21] <lifeless> how you going
[05:21] <jbailey> Lemme look, I might be confused.
[05:21] <Diziet> jb,pitti: What do you think of this plan ?
[05:21] <infinity> pitti : Yes.
[05:21] <Kamion> infinity: you can only have 512 glyphs in a console font, so we can't really win
[05:21] <infinity> pitti : Perhaps my using a vesafb has something to do with it using a sucky font?
[05:21] <infinity> Kamion : Oh.  Then why do we even care about unicode consoles?
[05:21] <Kamion> (according to Denis Barbier in $some_bug_I_forgot)
[05:22] <infinity> Kamion : I get pretty much zero codepoint coverage here.
[05:22] <Kamion> infinity: the installer attempts to pick a sensible-ish font depending on your language
[05:22] <pitti> Diziet: that sounds sane; in fact that's what I planned with carlos :)
[05:22] <infinity> mvo : We may be fighting a losing battle here.
[05:22] <jbailey> Diziet: So does ubuntu-docs still provide the base translations?
[05:22] <Kamion> we have to care about Unicode consoles because we only install a Unicode locale by default
[05:22] <Diziet> jb: `base' translations ?
[05:23] <jbailey> Diziet: Sorry, the about-ubuntu translations?
[05:23] <Diziet> Yes.
[05:23] <jbailey> Cool.
[05:23] <jbailey> Then if Martins happy, just lemme know where I should stuff 'em.
[05:24] <mvo> infinity: oh crap :/
[05:24] <Diziet> jb: We need a list of the translations that you (a) have now or (b) might have for breezy.
[05:25] <mako> jsgotangco: hey there! :)
[05:25] <mako> lifeless: i'd like to talk to you
[05:25] <lifeless> mako: sure
[05:26] <pitti> mvo: so, /usr/share/update-notifier/notify-reboot-required ?
[05:26] <Diziet> Then for each locale LOCALE in that list, provide a file /usr/share/ubuntu-artwork/home/index-LOCALE.html which is the translation if it exists or a symlink to the English if it doesn't.
[05:26] <Diziet> Can you do that nicely ?
[05:27] <spayne> yo all
[05:27] <Diziet> And then pitti changes m-f-locale-all to contain a copy of your list in its source, and it writes those URLs into the region.properties files.
[05:27] <Diziet> Are you with me ?
[05:27] <pitti> Diziet: sounds fine
[05:28] <ogra> erm, how do i go with my derivative artwork package then ?
[05:28] <jbailey> Diziet: Yes.
[05:28] <ogra> that means i need to replace *all* m-f-locale packages for edubuntu ? 
[05:28] <jbailey> Diziet: I think the translations are largely fixed now, mod post-Breezy updates.
[05:28] <wasabi_> Hmm. Done. Heh.
[05:28] <wasabi_> Now to test somehow.
[05:29] <wasabi_> Oh, how do I get losetup into the initramfs?
[05:29] <Diziet> ogra: Um, why ?
[05:29] <jbailey> You need a hook script!
[05:29] <pitti> ogra: no, I thought the m-f-all just contain a reference to the files?
[05:29] <jbailey> wasabi_: Look at /usr/share/initramfs-tools/hooks
[05:29] <jbailey> grep for copy_exec
[05:29] <Diziet> That's right, m-f-l-all just contains the filenames.
[05:29] <ogra> Diziet, because about-edubuntu exists only untranslated
[05:29] <wasabi_> Ahh hah
[05:29] <jbailey> that will copy the binary and any libraries it might depend on.
[05:30] <Diziet> Right.  So just ship it as index.html and every index-LOCALE.html that you see in ubuntu-docs.
[05:30] <pitti> Diziet: did I understand you right? m-f-locale-all just changes the homepage location, it does not ship the start pages itself, right?
[05:30] <Diziet> That's right.
[05:30] <pitti> ok
[05:30] <ogra> ok
[05:30] <jbailey> I'll do that.  I just need to know where to put them.
[05:30] <Diziet> But you change the homepage location IF AND ONLY IF the new location is in the list from ubuntu-docs.
[05:30] <pitti> ogra: well, then you just need to have an edubuntu-artwork with the translated (and english) start page and it should DTRT
[05:31] <Diziet> That's why there has to be some minimal coordination.
[05:31] <pitti> Diziet: yes, of course
[05:31] <jbailey> Teamwork! \\o/
[05:31] <mvo> pitti: yes
[05:31] <pitti> Diziet: but if we build all this in a central place, langpack-o-matic can figure this out
[05:31] <Diziet> It would be better to have some actual source-level dependency to shuffle this list about but if it's just for breezy we can just have one blessed list.
[05:31] <jbailey> pitti: Right.  But that's totally not breezy material.
[05:31] <pitti> jbailey: of course
[05:32] <Diziet> OK, glad we've got that sorted.  Do you think I should write it up ?
[05:32] <Diziet> I mean, send an email to ubuntu-devel with details ?
[05:33] <pitti> Diziet: I'd rather write it up on the wiki
[05:33] <pitti> Diziet: better for integrating into UBZ specs, and the like
[05:33] <Diziet> Sensible.
[05:33] <jbailey> pitti, Diziet: That should be linked from LocalesThatDontSuck
[05:36] <jbailey> Diziet: Sending it somewhere so that we can all work from it with a clear list of what will be at what path would be useful.  I'm still not sure where all the varioous HTML files should go.
[05:36] <Diziet> jb: Sure.
[05:36] <Diziet> I thought this might need a proper writeup.
[05:36] <martinhj> jbailey: we talked a while a go about lvm and breezy.. I would like you to know that the usplash works together with lvm (but I suppose you already know)
[05:37] <jbailey> martinhj: I always appreciate success reports. =)  It's a nice change from the pile of failure reports. =)
[05:37] <martinhj> jbailey: yeah, like lilo doesn't work;-)
[05:37] <pitti> jbailey: hmm, is there any chance we can have the Kurdish locale in locales proper?
[05:37] <jbailey> martinhj: Right. =)
[05:37] <jbailey> pitti: Yes, dear.
[05:38] <jbailey> pitti: seb128 asked me to hold off the locales update while we figure out a gtk bug.
[05:38] <Darknight> hello
[05:38] <pitti> jbailey: I just got a question from the Turkish user group, which wants to heavily promote Ubuntu, but can only do that if Ubuntu supports it OOTB
[05:38] <jbailey> I have some first_weekday patches that aren't clear if they should go in or not.
[05:38] <Darknight> sorry, byt can anyone help me with setup of tv card in breeze?
[05:38] <jbailey> They're right from glibc's POV, but gtk seems to behave poorly.
[05:38] <pitti> jbailey: ah, thanks
[05:39] <seb128> jbailey: have you read Denis's reply?
[05:39] <jbailey> seb128: I did.
[05:39] <jbailey> seb128: Alhtough that was several hours ago, and I should reread it. =)
[05:40] <seb128> jbailey: my conclusion is that the locales are screwed anyway
[05:40] <jbailey> seb128: So the French can suck it up and cope with the week starting on Tuesday? =)
[05:41] <wasabi_> Wow that was remarkably easy.
[05:41] <seb128> no, drop the patch
[05:41] <wasabi_> gg initramfs people
[05:41] <seb128> it will make the .fr work fine
[05:41] <seb128> but de_DE is b0rked by example
[05:43] <pitti> seb128: that's release critical then :)
[05:44] <seb128> pitti: say that to mdz :)
[05:46] <lamont-away> seb128: the only thing remaining is for you to email elmo asking for gtk-smooth-engine to be removed from the archive.
[05:46] <lamont-away> assuming that gtk2-$mumble now delivers gtk-engines-smooth
[05:47] <seb128> lamont-away: no, as said the universe package is still useful for the gtk1 binary
[05:48] <lamont-away> except that it has no binaries installed, because the upload is rejected.
[05:49] <lamont-away> the only binary there is  gtk-engines-smooth_0.5.6-4_powerpc.deb
[05:49] <lamont-away> which is probably warty-ish
[05:49] <lamont-away> pitti: heh
[05:50] <pitti> lamont-away: good morning, btw :-)
[05:50] <lamont-away> pitti: g'morning.
[05:50] <lamont-away> and I have to leave for a few hours.
[05:50] <lamont-away> later
[05:51] <Diziet> https://wiki.ubuntu.com/BreezyFirefoxStartPageTranslation
[05:52] <Diziet> jbailey: You need to add the list of Translateable locales to that wiki page.
[05:52] <pitti> Yagisan: too late anyway, it's already out :-)
[05:52] <pitti> Yagisan: well, every sensible German worker is already in the weekend now, so they won't notice anyway :-)
[05:54] <jbailey> Diziet: Done.
[05:57] <Diziet> Thanks.
[05:58] <bob2> I guess you missed the ZWIKI STRIKES BACK episode
[06:18] <Diziet> I suppose I'll just have to do the somewhat-wrong thing for firefox fonts.
[06:19] <infinity> mvo : Alright, console-tools will be uploaded in a while, with increased cleverness.  Took a few (!) reboots to get it doing what I wanted...
[06:25] <mdz> Diziet: actually this is the best wiki we've found; do you have one that you've found to be superior?
[06:25] <Diziet> I much prefer twiki.
[06:27] <Diziet> OTOH it depends how much you want to worry about people putting random arbitrary HTML in.
[06:27] <fabbione> Diziet: the same as the one used to root freedesktop?
[06:27] <jdub> fabbione: yes :)
[06:27] <ogra> heh
[06:28] <Diziet> I didn't hear about that.
[06:28] <fabbione> jdub: i knew the answer :P
[06:28] <jdub> they're now using... moin :-)
[06:28] <bob2> well, it got a local account
[06:28] <fabbione> jdub: i was sitting less than a meter from daniels when that happened
[06:28] <bob2> it took hard work fro mthe kernel team for it to get root
[06:28] <jdub> yay team
[06:29] <lifeless> what the world needs now
[06:29] <lifeless> is just another wiki
[06:29] <jdub> sweet wiki
[06:30] <ogra> couldnt we just run it with thttpd ? so vulnerabilitys go straight through then ? 
[06:33] <mvo> infinity: rock!
[06:34] <pitti> mdz: can you please import Debian #329127 into bz?
[06:35] <pitti> mdz: a friend of mine complained about our "cvs" package crashing all over the place, pro'lly a compiler bug
[06:37] <zyga> could any hacker tell me how to compile coreutils as static stuff?
[06:44] <bob2> zyga: might be easier to use sash?
[06:45] <zyga> bob2: thanks, I'll check it out
[06:45] <Kamion> I wish translators would read the comments left there for the benefit of translators
[06:46] <zyga> Kamion: ?
[06:48] <Kamion> # This menu entry may be translated.
[06:48] <Kamion> # However, translators are required to keep "Choose language"
[06:48] <Kamion> # as an alternative separated by the "/" character
[06:48] <Kamion> # Example (french): Choisir la langue/Choose language
[06:48] <Kamion> _Description: Choose language
[06:48] <Kamion> that one in particular
[06:48] <Kamion> it's depressingly common for the comment to be ignored
[06:49] <jdub> at least the french got it right
[06:49] <Kamion> I've been filling in replacement suggestions in Rosetta, but I have no idea if anyone pays attention to those
[06:49] <jdub> you'd think they'd screw it up on political grounds
[06:49] <zyga> Kamion: who/what chooses final translation from various suggestions?
[06:50] <zyga> last used?
[06:50] <Kamion> zyga: translation team, as far as I know
[06:50] <Kamion> but in the case of what actually gets into the installer, me ;)
[06:51] <Kamion> (I can do a number of sanity-checks even if I don't speak the language, such as "strings with wildly different meanings in English ought not to translate to the same string in another language")
[06:51] <Kamion> and I tend to go with Debian translations where available just because the merging gets insane for me otherwise
[06:52] <zyga> Kamion: I was thinking about a possible scenario where two people keep changing the same translation over and over
[06:53] <Kamion> well, that would be dysfunction within the translation team; Rosetta doesn't let you change a translation directly if you're not in the translation team - you can only make suggestions
[06:53] <zyga> Kamion: I am in the translation team
[06:54] <zyga> bob2: thank's sash looks great!
[06:54] <Kamion> presumably you're in contact with your fellow team members then
[06:54] <\sh> gnarf
[06:54] <\sh> anybody knows which version of madwifi drivers we're shipping in the restricted modules?
[06:55] <_mvo_> infinity: can you please ping me after you uploaded the console-tools? I would like to test it here too 
[06:59] <Kamion> Rosetta is going to give me RSI at this rate
[07:01] <doko> RSI?
[07:01] <Kamion> http://en.wikipedia.org/wiki/Repetitive_strain_injury
[07:02] <Mirv> can anyone pick from today's ubuntu-devel e-mails the command that was supposed to be run on a laptop that has working suspend-to-ram, and the e-mail address of the person who asking for it :) [also, an _up-to-date_ archives link is okay] 
[07:03] <mjg59> Mirv: sudo dmidecode, mail to mjg59@srcf.ucam.org
[07:03] <Diziet> I've had to switch to mousing left-handed when I'm having a bugzilla nightmare day, because it's so awful.  That way I spread the load.
[07:04] <seb128> Diziet: are you going to do a firefox upload soon?
[07:06] <Mirv> mjg59: ok, thanks
[07:18] <Diziet> seb: No, why ?
[07:18] <Diziet> I'm going to fix the fonts, yes, but that's in fontconfig, which I'm preparing now.
[07:19] <ogra> yay
[07:25] <segfault> strange, seems that the archives of the lists is not being generated
[07:25] <segfault> err, are.
[07:29] <mvo> pitti: update-notifier uploaded, you can use /usr/share/update-notifier/notify-reboot-required now and it will be translateable via gettext/langpacks
[07:29] <pitti> mvo: yay
[07:30] <seb128> Diziet: because we have some translation updates for the .desktop
[07:30] <mvo> pitti: yeah, sorry that it took so long :/
[07:32] <jsgotangco> good night =)
[07:34] <Diziet> seb: Ah, right.
[07:34] <seb128> Diziet: https://wiki.ubuntu.com/DesktopTranslations has them if you do an upload
[07:34] <Diziet> Can you file a bugzilla bug P1 against firefox ?
[07:34] <seb128> sure
[07:34] <Diziet> Put that URL in.
[07:34] <Diziet> Then I'll fold them into the next upload.
[07:35] <seb128> cool, thanks
[07:35] <Diziet> I'll make sure to do the upload before the RC.
[07:41] <infinity> I WIN.
[07:41] <infinity> mvo : Being uploaded right now.
[07:42] <mvo> infinity: cool, I'm away for ~1,5h now, but I'll test it when I come back
[07:44] <infinity> mvo : Spiffy.
[07:44] <infinity> mvo : Keep in mind that you'll probably still be hindered by the crap font (but you can load/specify another, if that takes your fancy), but all the VCs should be in UTF-8 now.
[07:44] <infinity> mvo : Care to see the ugly-hack debdiff?
[07:45] <mvo> infinity: oh yes!
[07:46] <mvo> infinity: I would have debdiffed anyway :)
[07:46] <tseng> is it possible to get ubuntu.com addresses for launchpad teams
[07:46] <tseng> to act as Maintainer:
[07:46] <mvo> infinity: can you send it to me? I really need to go now
[07:46] <infinity> mvo : It's actually not that bad, just took a bit of thinking to get it right.  Very little typing once the thinking kicked in.
[07:46] <infinity> mvo : http://cerberus.0c3.net/~adconrad/console-tools.debdiff
[07:46] <infinity> mvo : Download it and cherish it.
[07:47] <mvo> infinity: will do :)
[07:47] <mvo> cheers and bye
[07:49] <Diziet> I just love programs which crash when my chroot hasn't got /proc mounted.  Great error handling that.  Not as bad as the hideous lvm tools though.
[07:51] <bddebian> Bah, where's lamont
[07:59] <Diziet> Thank you.
[08:02] <CarlFK> The Etherboot mail list has agreed that making a UNDI net driver for Linux is a good idea, but no one has time. Where is a good place to post the "job" to see if someone wants to do it?
[08:03] <CarlFK> This would probably benefit the Ubuntu LTSP project. and any other linux thin client
[08:07] <slomo> seb128: ? what do you ask? ;) i just gave you the link to a CAN regarding abiword... maybe something we should fix for breezy
[08:07] <infinity> slomo : Another one?... We just fixed a security hole in abiword.
[08:07] <ivoks> g-v-m says run gnome-cups-add on printer connection
[08:07] <ivoks> but there is no g-c-a
[08:08] <slomo> infinity: the rtf thing?
[08:08] <infinity> Yeah, CAN-2005-2964
[08:09] <slomo> exactly... ok, seb128 just ignore me ;)
[08:09] <pitti> slomo: I only know about one current vuln in abiword (the rtf thing)
[08:09] <\sh> pitti: where was your sec list webpage again?
[08:09] <pitti> \sh: http://people.ubuntu.com/~pitti/ubuntu-cve/
[08:11] <\sh> pitti: hmmm...I can't read it :(
[08:11] <slomo> hm, why don't we have eclipse for amd64?
[08:11] <\sh> the official announcement I mean
[08:11] <pitti> \sh: hm? it's http and unrestricted
[08:12] <\sh> pitti: http://cve.mitre.org/cgi-bin/cvename.cgi?name=CAN-2005-2962
[08:12] <pitti> \sh: ah, it might still be unprocessed by mitre
[08:12] <pitti> \sh: https://www.ubuntu.com/usn/usn-188-1
[08:13] <\sh> thats 2964 ;)
[08:13] <\sh> i need 2962 ntlmaps
[08:13] <pitti> \sh: ah, I see; sorry
[08:14] <pitti> \sh: http://www.debian.org/security/2005/dsa-830
[08:14] <pitti> \sh: however, just looking at the sid package should be fine
[08:14] <pitti> \sh: the question is just "merge or sync"
[08:14] <\sh> pitti: I'll check
[08:15] <\sh> pitti: we have 0.9.9-3* and debian's fix is in 0.9.9-2 (sarge)
[08:16] <pitti> "For the unstable distribution (sid) this problem has been fixed in version 0.9.9-4."
[08:16] <pitti> \sh: it's not in -2, but in -2sarge1
[08:16] <\sh> wah...yes
[08:20] <\sh> can someone give me an information about the madwifi version in the restricted modules?
[08:21] <Treenaks> "it exists"
[08:22] <\sh> well...so we can forget that wpa-psk (wpasupplicant) + dhcp is working correctly
[08:22] <\sh> if madwifi is >= CVS snapshot from 20050707
[08:23] <\sh> s/=//
[08:24] <slomo> infinity: do you know why eclipse isn't built anywhere except x86?
[08:26] <\sh> pitti: I think a sync is ok...
[08:30] <\sh> pitti: will you request or should I?
[08:31] <pitti> elmo: please sync ntlmaps (ubuntu override ok according to \sh)
[08:31] <slomo> infinity: it is arch any... so it would be pas?
[08:37] <elmo> pitti: are these two language packs going to get fixed any time soon?
[08:42] <infinity> lucifer:~/build/dak/srcdep/> grep eclipse Packages-arch-specific
[08:42] <infinity> %eclipse: i386
[08:42] <infinity> slomo : ---^
[08:43] <slomo> infinity: ok, thanks... i'll try to get it working on amd64... at least the eclipse guys have amd64 packages ;)
[08:44] <pitti> elmo: the -support packages? doko asked me to leave them as they are since he wanted to fix them soon
[08:44] <CarlFK> \sh, is there a way to simulate a GR keyboard?
[08:45] <elmo> pitti: support-{lt,bs}
[08:45] <elmo> pitti: well I just don't want them to be forgotten for RC
[08:45] <elmo> we really shouldn't be shipping uninstallable packages
[08:45] <segfault> who is responsible for the wiki template?
[08:45] <pitti> elmo: yes, right; ooo2 currently doesn:'t build ooo2-l10n-{lt,bs}
[08:45] <pitti> elmo: if they aren't there by monday, I just drop these dependencies
[08:46] <jdub> segfault: hno73 
[08:46] <elmo> mdz: around?
[08:46] <elmo> pitti: ok - thanks
[08:47] <\sh> CarlFK: a GR? greek? 
[08:47] <CarlFK> \sh, "So no german keyboards (or any other keyboards with special keys via AltGR) "
[08:47] <mdz> elmo: yes
[08:47] <\sh> CarlFK: altgr == meta- ;)
[08:47] <mdz> I suck at this 'day off' business
[08:48] <elmo> mdz: I promoted edubuntu-artwork-usplash as 'obvious' (binary of a source package already in main) - have you looked at the sip stuff at all needed to fix python2.4-qt ?
[08:49] <mdz> elmo: no, I haven't
[08:49] <mdz> elmo: if pitti gives it the OK, it's fine with me
[08:50] <CarlFK> elmo, I think the 2 bt seeds for http://cdimage.ubuntu.com/edubuntu/daily/current/ are behind firewalls or something - im not getting any connections
[08:51] <CarlFK> elmo - Im working on figuring out what the problem really is, so if it isn't obvious I may have more info later
[08:55] <segfault> is there any howto on how to change the default splash? i saw there's a .so separated file now..
[08:56] <jbailey> segfault: Not yet, but I can put one up for next week.
[08:56] <jdub> segfault: easiest way to grok it would be to get the edubuntu-artwork source and have a look :)
[08:56] <elmo> why do our main inclusion report not include a "requested-by" field?
[08:57] <jbailey> segfault: Fundamentally, the .so is the compiled output of pngtobogl turned into a shared library for dlopening.  You have to made sure that it's all named right.
[08:57] <segfault> jdub: thanks, i'll look.
[08:58] <segfault> jbailey: humm, seems easy to do so. i'll try it now.
[08:58] <jbailey> segfault: Take a look at the package, feel free to ask questions if you have them.
[09:02] <jdub> mjg59: ping
[09:09] <dholbach> have a nice evening
[09:14] <jdub> night dholbach 
[09:14] <slomo> hi backports-r-us ;)
[09:14] <dholbach> not going to bed ;)
[09:14] <backports-r-us> slomo: greetings :)
[09:17] <pitti> backports-r-us: hi! btw, http://people.ubuntu.com/~pitti/ubuntu-cve/unfixed.html shows some nice backport targets (gaim and abiword)
[09:17] <backports-r-us> pitti: Cool -- can you email that to me? On a text-only console right now :)
[09:18] <pitti> backports-r-us: the link? sure
[09:18] <backports-r-us> pitti: Wasn't Abiword brought up to 2.2.7 during a recent USN?
[09:18] <pitti> backports-r-us: yes, but I patched the current version due to the freeze
[09:19] <pitti> backports-r-us: sent
[09:21] <backports-r-us> pitti: thanks
[09:22] <slomo> wasabi: ping?
[09:22] <mdz> backports-r-us: do you track USNs and make sure to update the versions in backports to incorporate security fixes?
[09:22] <elmo> Riddell: ?
[09:23] <backports-r-us> mdz: roughly manually; what's a good way to keep track of them?
[09:23] <segfault> jbailey: are those image 16 colors only?
[09:23] <mdz> backports-r-us: subscribing to ubuntu-security-announce
[09:23] <jbailey> segfault: Yes.
[09:24] <jbailey> segfault: And take care to keep foreground/background and failed colours separate.
[09:24] <backports-r-us> mdz: k, already do that :)... I'm usually OK with keeping up to date with USNs
[09:24] <pitti> mdz: I just mailed him my tracker page url
[09:24] <mdz> pitti: you have a page? send that to me too
[09:24] <backports-r-us> pitti: just too a look; awesome tracking page :)
[09:24] <backports-r-us> took*
[09:24] <pitti> mdz: you don't know it? http://people.ubuntu.com/~pitti/ubuntu-cve
[09:25] <pitti> mdz: it's automatically generated; it's my "security big brother"
[09:25] <mdz> thanks
[09:28] <backports-r-us> bddebian: am I allowed to summon yet?
[09:29] <backports-r-us> Guys, successful upgrade report from Warty to Breezy :)
[09:29] <backports-r-us> (and it had old-style sf.net backports, too!)
[09:29] <slomo> backports-r-us: depends on what you want to do :)
[09:29] <bddebian> backports-r-us: Sure :-)
[09:35] <segfault> jbailey: is this ok? https://wiki.ubuntu.com/USplashCustomizationHowto
[09:36] <mvo> segfault: nice!
[09:36] <jbailey> segfault: Looks right.  you should add something like the following, though:
[09:36] <jbailey> segfault: The PNG must be: 640x480 16 colours.
[09:37] <jbailey> segfault: The following pallettes are used for the usplash display:
[09:38] <jbailey> segfault: Background colour: 0, Progressbar color: 1, Progressback background: 4, Text background: 0, Text foreground: 2, Failured colour: 13
[09:38] <jbailey> segfault: But this is awesome, thanks.
[09:38] <jbailey> mjg59: (The progress background is what you recently added, right?)
[09:39] <crimsun> jbailey, yes, in -16
[09:39] <segfault> jbailey: ok, will do that.
[09:40] <jbailey> crimsun: Cool, thanks.  I'm wondering how it looks on the kubuntu and edubuntu pallettes. =)
[09:46] <bddebian> Anyone a .mime file expert?
[09:47] <Mirv> hmm, where has the Ubuntu-style lock screen dialog gone? And where are those "Switch user", "OK" and "Cancel" in the new dialog translatable?
[09:48] <seb128> can anybody gives editbugs rights to petermcv@clarkson.edu ?
[09:48] <mvo> infinity: sorry for spoiling the fun, but the console-tools changes seem to not fix the problem here (greek install)
[09:49] <infinity> mvo : That's because I'm retarded, there's another upload just sent.
[09:49] <segfault> just updated: https://wiki.ubuntu.com/USplashCustomizationHowto
[09:49] <bddebian> seb128: Do you or do you know anyone who knows .mime files?
[09:49] <seb128> bddebian: context?
[09:50] <infinity> mvo : Can you manually remove the "!" from the "pidof usplash" test at the top of /etc/init.d/console-screen.sh, then test? :)
[09:50] <bddebian> seb128: gnome-apt doesn't self register .deb files.  It has a gdeb.mime which doesn't seem to get installed but the syntax of it looks wrong
[09:50] <mvo> infinity: sure :) 
[09:51] <seb128> bddebian: /usr/share/mime/packages/glade.xml 
[09:51] <seb128> bddebian: you want to do a such .xml for the current mimetype implementation
[09:51] <infinity> mvo : That's just crazy talk.
[09:52] <mvo> infinity: unfortunately yes
[09:53] <pitti> seb128: do you have time to test the new french langpacks?
[09:53] <mvo> infinity: hm, after chaning the ! in console-tools it works again for vt1, but no luck with vt2-6 :/
[09:54] <bddebian> seb128: That file doesn't exist
[09:54] <seb128> pitti: sure
[09:55] <seb128> bddebian: it's a part of glade-common-2
[09:55] <seb128> bddebian: ls /usr/share/mime/packages/
[09:55] <zyga> hmm :-)
[09:55] <zyga> usplash keeps dissaparing
[09:55] <zyga> is this common for others
[09:55] <zyga> ?
[09:56] <mvo> zyga: disapparing in what way?
[09:56] <mvo> not coming up at all?
[09:56] <infinity> mvo : Erm.  I'll have to reboot more here, then...
[09:56] <jbailey> segfault: Nice, thanks for the update!
[09:57] <infinity> mvo : That init script is the one from the package, right?  Not one you've altered?  (other than reversing that test)
[09:57] <bddebian> seb128: OK, thx.  Do I have to do anything special to update the mime database or just putting the .xml file in the proper directory?
[09:57] <pitti> seb128: http://people.ubuntu.com/~pitti/langpacks/
[09:57] <mvo> infinity: yes. but I will double-check again
[09:58] <infinity> mvo : What does "vt-is-UTF8" return on vc2-6?
[09:58] <seb128> bddebian: the package has to use dh_installmime
[09:58] <mvo> infinity: Single-byte char mode 
[09:58] <bddebian> seb128: OK, I'll give it a try thx
[09:58] <infinity> mvo : Bah.
[09:58] <bddebian> seb128: Oh and the .desktop should use gksudo not gksu correct?
[09:59] <seb128> correct
[09:59] <bddebian> Thanks
[09:59] <infinity> mvo : Worked here on my last boot...
[09:59] <mvo> infinity: let me first double check that thre isn't something bad on my system
[10:01] <zyga> mvo: after some updates usplash is disabled
[10:01] <zyga> mvo: yes 
[10:02] <pitti> ah, update-notifier in German :-)
[10:02] <pitti> seb128: they work fine for me, no obvious breakage and Rosetta goodness :-)
[10:02] <segfault> pitti: how about in pt_BR? :)
[10:02] <mvo> zyga: there was a problem with /usr/lib/usplash/usplash-artwork.so and update-alternatives on some systems
[10:02] <mvo> pitti: yeah!
[10:04] <pitti> jbailey: ah, and new langpacks work fine with belocs, too
[10:04] <pitti> mvo: language-selector is not translated, though
[10:05] <pitti> seb128: I'm waiting for your ok, then I upload the stuff
[10:05] <mvo> pitti: it went into rosetta late, I guess I should do a translation myself
[10:05] <seb128> pitti: playing with them, a sec
[10:05] <zyga> mvo: any idea how to fix this?
[10:05] <pitti> seb128: I can't ignore my gf any longer, otherwise I won't have a gf any more :-)
[10:06] <seb128> pitti: works fine for me, GO
[10:07] <jcohen85> jbailey, i have all the latest updates including the new usplash and i'm still not seeing a splash screen.
[10:07] <pitti> mvo: there is a mo file, but some strings are missing in it
[10:07] <pitti> seb128: yay
[10:08] <mvo> infinity: ok, the problem seems to be that it works fine with usplash, but not without usplash. could you please try booting without splash and see if vt2 is unicode for you then?
[10:09] <mvo> infinity: and with usplash the font seems not to be setup correctly. sorry for being so negative :/
[10:09] <pitti> segfault: uploading now, find it out yourself :-)
[10:10] <bddebian> seb128: Sorry to keep bugging you but hopefully one last question.  Do you know if the mime xml file should be for gnome-apt or for gdeb?
[10:10] <jcohen85> is anyone else not seeing the splashscreen?
[10:10] <Kamion> Nafallo: would it be possible for you to work with the d-i Swedish translation team to get your changes in there? I'm incorporating your changes to Ubuntu-specific strings, but the problem with taking changes to generic strings (ones in both Debian and Ubuntu) in Ubuntu is that I'm then stuck later with having to merge changes in a language I don't speak
[10:10] <jcohen85> it looks as it if is coming up and then goes back to the boot screen
[10:10] <seb128> bddebian: gdeb if that's gdeb which is used for that
[10:10] <Kamion> Nafallo: (as it happens I can make out Swedish well enough, but usually not at the level of being able to decide which string out of two alternatives is better ...)
[10:10] <bddebian> Uhmm OK
[10:11] <mvo> pitti: for language-selector?
[10:11] <infinity> mvo : Negative works.  How do I set a different enough font that I can tell it's working?
[10:11] <lamont> hrm... no daniels
[10:12] <infinity> lamont : Well, it is 6am on Saturday morning here.
[10:12] <lamont> infinity: true
[10:12] <lamont> before I upload it with just my fix.....
[10:12] <mvo> infinity: I used "lat1-08" in /etc/console-tools/config (SCREEN_FONT=lat1-08)
[10:13] <infinity> GAH.
[10:13] <mvo> infinity: it's ugly as hell, but it's definitly differnet from the default one :)
[10:13] <infinity> My last upload was nothing but a changelog entry, I didn't actually invert the test.
[10:13] <infinity> I might be tired.
[10:13] <mvo> infinity: 6am? no wonder!
[10:13] <zyga> hmm
[10:13] <infinity> Anyhow, no big deal, since it looks like I'll need another upload. :)
[10:13] <zyga> how do I run nautilus in gdb?
[10:13] <mvo> infinity: you didn't got up right early, you are still awake?
[10:14] <infinity> Yeah, I'm still awake.
[10:14] <infinity> Lord only knows why.
[10:14] <lamont> seb128: why don't my uxterms get a utmp entry anymore, I wonder....
[10:14] <infinity> I think I just like rebooting.
[10:14] <mvo> heh
[10:14] <\sh> lamont: what?
[10:15] <mvo> infinity: it's fun when you get a "mount-count exceeded" message two times a day
[10:15] <\sh> lamont: uxterm comes from xterm...so it can be me you want to ask ;)
[10:15] <zyga> hmm
[10:15] <lamont> ok.  why doesn't it get a utmp entry?
[10:15] <seb128> lamont: no clue, you may want to ask \sh
[10:15] <\sh> lamont: I'll have to check....
[10:15] <Kamion> lamont: not setgid utmp?
[10:15] <pitti> mdz: new langpacks with Rosetta love on the way :-)
[10:15] <lamont> Kamion: 755 root root
[10:15] <lamont> that'd be it.
[10:15] <zyga> nautilus keeps crashing when I try to reassign program responsible for opening given file type
[10:16] <lamont> pitti: what did you do to me?
[10:16] <Kamion> -rwxr-sr-x  1 root utmp 285272 2005-09-01 01:31 /usr/X11R6/bin/xterm
[10:16] <Kamion> on Debian
[10:16] <pitti> lamont: context?
[10:16] <Kamion> uxterm can be 755 root:root with no problem
[10:16] <lamont> /usr/bin/xterm is 755 root:root
[10:16] <Kamion> \sh: make /usr/bin/xterm's permissions be like the above, and it should work properly again
[10:16] <pitti> lamont: you mean your buildds now have something to chew at?
[10:17] <lamont> Kamion: uxterm is a script that runs xterm
[10:17] <Kamion> lamont: indeed
[10:17] <slomo> elmo: thanks for accepting mosml :)
[10:17] <lamont> pitti: I figured you'd de-rooted xterm
[10:18] <bddebian> lamont: Are you the enlightenment expert? :-)
[10:18] <bddebian> Gnight pitti
[10:18] <pitti> lamont: IZ DANIELS BUG
[10:18] <zyga> night pitti 
[10:18] <elmo> pitti: no, it's not, it's a \sh bug
[10:18] <elmo> daniels isn't the one who breaks xterm these days
[10:18] <pitti> anyway, I wasn't it :-)
[10:18] <lamont> bddebian: I've been known to elighten people, but not even sure what a package called enlightenment might _do_, let alone how.
[10:18] <\sh> pitti: my fault
[10:18] <pitti> night!
[10:18] <lamont> damn he's quick
[10:19] <\sh> lamont / kamion: I'm busy right now...will do a bugfix tomorrow evening when I'm coming home from work :(
[10:19] <bddebian> lamont: Well your name is all over the changelog :-)
[10:20] <lamont> Kamion: I'm figuring it's close enough to that magical 3 or 4 thing I mentioned earlier... unless you'd rather I didn't....
[10:20] <Kamion> it's a bit too much of a Friday evening for me to care
[10:21] <\sh> grmpf vmware is slow...really slow on this 1.6GHz cpu
[10:21] <infinity> mvo : Okay, to make sure I'm getting the same results you are...
[10:21] <backports-r-us> \sh: vmware shouldn't be THAt slow
[10:21] <backports-r-us> \sh: unless you really don't believe in RAM :)
[10:21] <infinity> mvo : With usplash enabled, everything worked (font and UTF)...
[10:21] <infinity> mvo : With usplash disable, font worked, but UTF8 didn't (on 2 through 6)
[10:22] <infinity> mvo : If your system did the same as mine, I know why, and the fix is easy.
[10:22] <\sh> backports-r-us: I'll believe in 256 MB ram ;) so it's slow...when the other 256 are used by the rest of breezy ;-)
[10:22] <infinity> mvo : If your results were different, I hate you and you can die.  A bit.
[10:22] <backports-r-us> \sh: yeah, that shouldn't be too bad then
[10:22] <backports-r-us> \sh: think of it compared to qemu ;)
[10:23] <\sh> backports-r-us: don't let me think of it...I tested livecds with qemu
[10:23] <backports-r-us> \sh: I did that last week :-/... painful even on AMD64
[10:23] <\sh> 26mins to hoary  now :(
[10:23] <elmo> of course it's slow if you don't give it enough RAM
[10:24] <elmo> 'cos all your actually testing is linux's ability to swap and/or thrash
[10:24] <\sh> elmo: please send me add. 512 MB for hp nc6000 ,-)
[10:24] <segfault> pitti: thanks!
[10:25] <\sh> elmo: but anyways...if I can reproduce the bug ... would be great...so someone has more work until the 11th ;)
[10:25] <\sh> daniels or seb128 one of the two ;)
[10:25] <tseng> elmo: is it possible to get @ubuntu.com addresses for launchpad teams
[10:25] <tseng> elmo: for group Maintainer
[10:26] <segfault> tseng: i think you should be approved as ubuntu member first
[10:26] <Kamion> he already is
[10:26] <tseng> segfault: yes :)
[10:26] <Kamion> segfault: see https://launchpad.net/people/ubuntumembers/+members/
[10:26] <segfault> then he alread has an email i guess?
[10:27] <elmo> segfault: that's not what he's asking for
[10:27] <tseng> that wasnt exactly the question
[10:27] <elmo> tseng: ugh
[10:27] <tseng> but close :P
[10:27] <Kamion> segfault: he said "for launchpad teams"
[10:27] <segfault> ah!
[10:27] <\sh> elmo: btw..thx for the latest syncs 
[10:27] <segfault> heh, ignore me
[10:27] <segfault> :P
[10:27] <elmo> tseng: maybe - it brings up some issues; like, e.g. we don't currently police the namespace
[10:28] <bddebian> Shit, I don't know what to do with the <match type> entry
[10:29] <Kamion> elmo: @teams.ubuntu.com?
[10:29] <Kamion> or something like that
[10:30] <tseng> hm
[10:30] <Kamion> maybe lists are better anyway
[10:30] <tseng> i think pkg-mono points at a list..
[10:31] <elmo> actually, yeah, I think lists are a better idea
[10:31] <tseng> yes
[10:32] <bddebian> seb128: Am I allowed 1 more question? :-)
[10:32] <bddebian> lamont: Seriously do you know enlightenment or not?
[10:32] <tseng> elmo: is that still jeff?
[10:32] <seb128> bddebian: sure
[10:33] <bddebian> seb128: Any ide what I would use for the <match type=" ".. /> entry for gdeb?
[10:33] <elmo> tseng: jeff can still do them yeah
[10:33] <bddebian> s/ide/idea
[10:33] <tseng> elmo, Kamion thanks
[10:35] <lamont> bddebian: the software, no.
[10:36] <bddebian> lamont: Then why are you all over the changelog? ;)
[10:36] <seb128> bddebian: they seem to start with "!<arch>
[10:36] <seb128> debian-binary"
[10:36] <lamont> bddebian: because I uploaded it.
[10:36] <lamont> you think I actually _know_ the code I hack on?
[10:37] <bddebian> lamont: Of course :-)
[10:37] <zyga> can usplash theoretically use more than 640x480x16?
[10:37] <lamont> bddebian: nah - I just fix things
[10:37] <bddebian> seb128: This is what glades looks like but I can't make sense out of it:
[10:37] <bddebian>       <match type="string" value="&lt;!DOCTYPE glade-interface" offset="0:256"/>
[10:37] <bddebian>       <match type="string" value="&lt;glade-interface" offset="0:64"/>
[10:37] <lamont> fixing something that's broken does not require understanding the thing, only understanding the problem.
[10:37] <bddebian> lamont: Well it FTBFSs for me on i386 now :-)
[10:37] <seb128> bddebian: mimemagic
[10:37] <jdub> Kamion: how ready-for-prime-time would you regard oem-config?
[10:37] <lamont> bddebian: sounds like it should get fixed... :-0)
[10:38] <seb128> bddebian: if the files has a "!DOCTYPE glade-interface" string between 0 and 256 chars it's a glade file
[10:38] <seb128> bddebian: or glade-interface between 0 64
[10:38] <qbertibur> hi
[10:39] <bddebian> lamont: Aye no shit.  The question I guess was whether to bring in the newer debian version or not :-)
[10:39] <bddebian> seb128: Hmm, do you think I need those entries at all?
[10:40] <qbertibur> Who is working on the new ubuntu init-system?
[10:41] <seb128> bddebian: you can try without them, but that's the way to get an efficient mime
[10:41] <bddebian> seb128: OK, well google ain't helping much. :-)
[10:42] <\sh> guys...what do u say to this:
[10:42] <\sh> Automatically selecting en_US.UTF-8 locale in addition to en_US.
[10:42] <\sh> xserver-xorg config warning: migrating xserver-xfree86 templates to
[10:42] <\sh>    xserver-xorg.
[10:42] <\sh>    xserver-xorg config warning: failed to infer keyboard layout from layout/lang
[10:42] <\sh>       'us--'
[10:43] <mvo> infinity: my system looks like this right now http://people.ubuntu.com/~mvo/00001.jpg
[10:43] <lamont> infinity: -73 inbound
[10:44] <jdub> mvo: looks like ass. :-)
[10:45] <mvo> jdub: yep, console-fonts are a real PITA
[10:45] <janimo> jdub, what are the prerequisites for seeding xubuntu stuff?So I know what I do in the weekend :)
[10:46] <jdub> janimo: given that you've got a metapackage already, you're most ofthe way there
[10:47] <janimo> hmm we'll need to transition those to main?
[10:47] <jdub> janimo: we just have to convince Kamion to eek out some time to set up builds :)
[10:47] <jdub> nup
[10:47] <janimo> ok that's great than I'll put some more stuff in the meta as people on the list tend to agree on what we shoudl have 
[10:48] <seb128> janimo: I'm looking on your patch, will upload rsn
[10:48] <jdub> then someone will have to lead you through setting up the seed lists from the packages you've already selected
[10:48] <qbertibur> Does nobody know who might be working on an improved init?
[10:48] <janimo> seb128,yupee!
[10:48] <jdub> that could be myself or kamion
[10:48] <jdub> qbertibur: no one is atm
[10:48] <jdub> qbertibur: we're stabilising for release! :)
[10:48] <jdub> highly unlikely we'd do something along those lines for dapper, either
[10:48] <qbertibur> There will be a time after the release ;-)
[10:49] <janimo> jdub, whenever you guys will have time let me know a bit in advance and I'll be there
[10:49] <jdub> ok
[10:49] <tseng> jdub: so dude, how do i get a mailing list
[10:49] <jdub> tseng: you email mailman@lists.ubuntu.com with what you want :-)
[10:50] <jdub> what's the list for?
[10:50] <tseng> mono group
[10:50] <tseng> to be the target of the Maintainer: field to start
[10:50] <jdub> wow, rad - you've enough people interested in ubuntu-specific mono work?
[10:50] <tseng> 3 regulars
[10:50] <jdub> ah, maintainer field
[10:50] <jdub> cool
[10:51] <tseng> slomo and ajmitch help out all the time :)
[10:51] <jdub> ubuntu-mono@lists... etc?
[10:51] <tseng> yeah pretty much
[10:51] <janimo> jdub, the motu folks - dholbach especially - have requested a spearte ml a while ago, what do you think?the motu is expanding
[10:51] <tseng> janimo: what would we talk about
[10:51] <janimo> and havinf the same stuff which is discussed on irc
[10:51] <jdub> janimo: yeah, we should definitely do that now
[10:51] <janimo> archived for generations of motus would be nice
[10:52] <tseng> devel is only too much traffic because of people posting bugs and replying to spam
[10:52] <janimo> tseng, the irclogs look very similar to half a year ago
[10:52] <janimo> similar blunders, questions
[10:52] <tseng> janimo: and we documented most of it on the wiki
[10:52] <janimo> tseng, but since you cannot decrease level on devel...
[10:52] <tseng> janimo: im not sure its our fault people keep asking the same
[10:53] <janimo> whether it's our fault or not is besides th epoint :)
[10:53] <janimo> it would definitely help new motus learn
[10:53] <janimo> I had such a break from motu that I could use some ml searching if it was in place ;)
[10:54] <bddebian> seb128: What do you think: <match type="string" value="debian" offset="0:8"/> ?
[10:54] <jdub> that would be a good list to mirror on the forums, too
[10:54] <backports-r-us> p.s. I love the new xscreensaver lock dialog
[10:54] <backports-r-us> looks very professional
[10:54] <seb128> bddebian: "!<arch> debian-binary"  seems better, I bet that other stuff match "debian" here
[10:57] <bddebian> seb128: Use that as string?  At what offset?
[10:59] <seb128> bddebian: as a string with an offset of 20
[10:59] <seb128> 22
[11:00] <tseng> jdub: bwar, have you found your dell batteries to be suck?
[11:01] <tseng> jdub: my second one seems to be dying
[11:02] <seb128> janimo: bash: x-www-browser: command not found
[11:02] <janimo> seb128, hmm
[11:02] <janimo> aren't all browsers register with update-alternatives?
[11:03] <seb128> janimo: I change it for "sensible-browser"
[11:03] <janimo> seb128, ok
[11:03] <janimo> I thought it was a debian standard that all X browsers register that
[11:03] <janimo> do you have epiphany only installed?
[11:03] <janimo> or maybe I am grossly mislead
[11:04] <seb128> hum
[11:04] <seb128> mine was pointing to galeon
[11:04] <seb128> which is not installed
[11:05] <seb128> alternatives are bogus, I don't like them
[11:05] <bddebian> seb128: http://pastebin.ubuntulinux.nl/2674  if you get a second
[11:05] <janimo> I don;t know much about them
[11:06] <seb128> bddebian: have you tried it?
[11:06] <bddebian> seb128: I can't until I get home :-(
[11:06] <jdub> ogra: nice work on the xscreensaver lock dialogue :)
[11:06] <seb128> bddebian: BTW that's not a space between both, but a 0xa
[11:07] <infinity> I WIN!
[11:08] <jdub> seb128: hrm, not getting distributor-logo love
[11:08] <juliux> hi i have a problem with the ppc kernels in breezy can some one help me?
[11:08] <jdub> seb128: oh, should i delete the icon-theme.cache for hicolor?
[11:08] <bddebian> seb128: Sorry, not a space between both what?
[11:09] <seb128> jdub: yeah
[11:09] <seb128> bddebian: no, a 0x0a
[11:09] <jdub> ah, there it is
[11:09] <jdub> vuntz: mmm! thanks for the distributor-logo patch :-)
[11:09] <bddebian> seb128: Where?
[11:09] <jdub> seb128: ooh, about-ubuntu should use that now too ;)
[11:09] <seb128> bddebian: sudo apt-get install ghex
[11:09] <seb128> ghex2 file.deb
[11:09] <seb128> and read the start
[11:10] <bddebian> seb128: Oh, ok
[11:10] <seb128> jdub: right, I'll change that with next upload
[11:11] <joh> Uhm, anything know about any problems with the latest madwifi drivers?
[11:11] <seb128> bddebian: that's a LF, not sure of how to code that with the xml file
[11:11] <jdub> seb128: thanks :)
[11:12] <seb128> jdub: np
[11:13] <jdub> BREEZY IS THE NEW HOTNESS!
[11:13] <jdub> seb128: maybe we should make sure that cache file isn't there
[11:13] <bddebian> seb128: OK, I'm going to try it when I get home.  Thanks for all your time
[11:15] <seb128> jdub: why? Some people have a cron to make them probably
[11:15] <seb128> bddebian: np, thanks for working on that
[11:18] <bddebian> seb128: One other quick thing before I head home.  There is a newer version in Debian.  Do I dare bring that in?
[11:18] <infinity> mvo : Still aorund?
[11:21] <seb128> bddebian: depending of the changes
[11:21] <bddebian> seb128: OK, fair enough.
[11:21] <bddebian> Later folks
[11:23] <seb128> janimo: lpi changes uploaded
[11:24] <janimo> seb128,thanks alot
[11:24] <janimo> so sensible-browser?
[11:26] <seb128> janimo: no, I've used your patch
[11:27] <janimo> hmm, so can't it cause probs if x-www-browser is not there?
[11:27] <janimo> althugh when gnome-open isn't we'll make sure ff isin xubuntu
[11:29] <mae> HA! I made gtk/glib totally flip out by creating a HBox with 500 buttons
[11:29] <mae> had to restart x
[11:29] <mae> nothing would click
[11:29] <tseng> that sounds useful
[11:29] <mae> well the thing is
[11:30] <mae> creating thousands of button objects is no prob
[11:30] <mae> but then when you display them in a container
[11:30] <mae> all hell breaks loose :)
[11:30] <seb128> mdke: bugzilla is not the right place to keep the new stuff to code
[11:30] <mae> my screen flickered a bunch too
[11:30] <HiddenWolf> seb128, is there any reason why gtkfilechooser suddenly shows unmounted cardreader slots?
[11:30] <seb128> HiddenWolf: ask pitti when he's around
[11:31] <seb128> mdke: upstream http://bugzilla.gnome.org/show_bug.cgi?id=154029
[11:31] <HiddenWolf> seb128, ok
[11:45] <slomo> wasabi: you maybe want to look at http://www.backports.org/~mkoch/unstable/. they have working eclipse packages for amd64/ppc/x86... maybe you can get the fixes into our packages