[12:02] <mjg59> sladen: Hrm. Interesting. It's also happening on CRTs, though.
[12:04] <Riddell> mjg59: acpid gets which event?
[12:04] <mjg59> Riddell: The sleep key event
[12:05] <Riddell> mjg59: I have acpid running here, how is kubuntu different?
[12:05] <mjg59> Except in a small number of cases, where the sleep key generates a keycode that's caught by a daemon that calls a command that signals gdm
[12:05] <mjg59> Riddell: Because there's also klaptopthingy that provides functionality to trigger stuff under various circumstances
[12:06] <mjg59> People seem to use it and then wonder why stuff doesn't work
[12:06] <mjg59> It doesn't generate the correct lockfiles, so when the user presses the power button to resume it shuts the machine down
[12:06] <mjg59> sladen: We already have the values in that patch
[12:06] <Riddell> mjg59: but sleeps keys and the like work just the same?
[12:07] <mjg59> Riddell: I'd guess so. I know approximately nothing about KDE.
[12:08] <sladen> mjg59: There's various issues, the first being LFP scalers, the second is that I suspect alot of the minimal 'VGA' BIOSes/hardware just aren't designed to do anything except standard 640x400, the third that 640x480/2 is >128kB and if it's not mapped, that shows up as wrap-around (which is why it would be preferable to use 640x400x4 or 320x400x8)
[12:08] <Riddell> mjg59: ok, thanks for the info, I'll have a go at klaptopdaemon
[12:10] <Riddell> mjg59: what's the command for cpu throtteling?
[12:10] <mjg59> Riddell: ? Powernowd handles that automatically
[12:10] <Riddell> klaptopdaemon has 
[12:11] <Riddell> klaptopdaemon has an option for doing it explicitly
[12:11] <mjg59> Riddell: Strip it
[12:11] <Riddell> ok
[12:11] <mjg59> We don't have any means of doing that
[12:12] <sladen> Riddell: the equivalent functionality (in the CPU applet) is disabled
[12:14] <Riddell> which package is pcm in?
[12:15] <sladen> Riddell: pcm?  or pmi?
[12:15] <mjg59> pmi?
[12:15] <mjg59> powermanagement-interface
[12:16] <pitti> mdz: just got an answer about espa-nol vs. aspell-es; the former should be used, the latter is obsolete
[12:16] <Riddell> oh aye.  I'm too sleepy
[12:17] <pitti> mdz: that means we should remove aspell-es from breezy completely
[12:19] <pitti> elmo: can you please remove the aspell-es package? it is obsoleted by espa-nol
[12:20] <pitti> elmo: i. e. both source packages build the binary aspell-es, but espa-nol is newer
[12:20] <pitti> ok, good night everybody! cu tomorrow
[12:22] <Riddell> lllmanulll: I just got some questions from a CNET reporter too
[12:29] <lllmanulll> Riddell, Ah, right :)
[12:29] <lllmanulll> Well that's good news, seems like they want to talk about Ubuntu :)
[12:29] <seb128> hey lllmanulll
[12:29] <lllmanulll> hey seb128 :)
[12:30] <dholbach> good night everybody
[12:30] <Riddell> I wonder how he picks his interviewees
[12:30] <lllmanulll> Oh, by the way, I'd like to rise a small problem
[12:30] <mdz> will the Gnome Massager give me a massage?
[12:30] <lllmanulll> Ah, nevermind, I'll file a bug :)
[12:30] <dholbach> who's the gnome massager?
[12:31] <lllmanulll> haha
[12:31] <lllmanulll> A guy just sent a mail talking about a "Gnome Messager"
[12:31] <Riddell> dholbach: Maria Blackmore does a mean workout
[12:31] <dholbach> oh nice :)
[12:32] <dholbach> Riddell: maria blackmore...?
[12:32] <dholbach> it seems too late for me :)
[12:32] <ogra_> mdz, the kernel module for the HW is still missing :)
[12:33] <mdz> lllmanulll: and then in the body of the message he wrote "Gnome Massager"
[12:33] <mdz> which sounded like even more fun
[12:33] <lllmanulll> Ah, right :)
[12:33] <ogra_> oh, it seems bsd already is prepared :)
[12:33] <lllmanulll> But that's a great idea all the same :)
[12:34] <lllmanulll> If GNOME can actually give massages, I'm sure it'll get far over KDE :-p
[12:34] <ogra_> but there shouldnt be a separate app... it should just get included into workrave /me thinks
[01:01] <mdz> heh, someone is going around actually trying to run GNOME applications on an alpha and filing grave bugs in debbugs when they don't work right
[01:01] <ogra_> with ubuntu ? o_O
[01:02] <elmo> ogra_: "debbugs"
[01:02] <ogra_> err, yes... i dont get enough sleep...
[01:04] <mdz> ogra_: they are imported into bugzilla, of course; debzilla doesn't have enough AI to filter them
[01:04] <ogra_> heh
[01:09] <jdahlin> mdz: shouldn't 16568 block 12301 instead of duping it?
[01:14] <mxpxpod> jbailey: ping?
[01:15] <jbailey> mxpxpod: Lagging though as I'm wrestling with Exchange so I'm mostly at another computer.
[01:16] <mxpxpod> jbailey: how do I redo my initramfs again (bootsplash and initramfs just updated)
[01:16] <jbailey> mxpxpod: dpkg-reconfigure linux-image-$(uname -r)
[01:16] <ogra_> mxpxpod, dpkg-reconfigure linux-image-`uname -r`
[01:16] <mxpxpod> thanks :)
[01:17] <ogra_> bah, 1 char less and still slower
[01:17] <jbailey> ogra_: Considering I have a cup full of tofu icecream in one hand, you should be ashamed.  You have the timezone working against you though ;)
[01:18] <ogra_> lol
[01:22] <mdz> jdahlin: er, yes, thanks
[01:23] <jdahlin> mdz: np, it just confused me a little
[01:25] <HrdwrBoB> does anyone know the default HP iLO username/passworD?
[01:32] <sladen> HrdwrBoB: is there a luggage tag on the bag with the details on?
[01:33] <HrdwrBoB> sladen: I just got the user guide for it apparenty you have to set it up in the BIOS .. I'll read some more :)
[01:42] <elmo> mdz: are you mid anastacia?
[01:42] <mdz> elmo: no
[01:42] <elmo> oh, well you left some python-qt stuff uninstallable
[01:42] <elmo> and with 2.3 versions in main
[01:43] <mdz> python2.3 is in main for breezy
[01:43] <mdz> I haven't really bothered sorting out the modules
[01:43] <elmo> uhm - ok is there any point in keeping 2.3 python modules in main?
[01:43] <elmo> I've been demoting them out of hand
[01:44] <mdz> my understanding is that people need to use python2.3 for zope?
[01:44] <mdz> and so they might want some modules too
[01:44] <mdz> the qt3 thing was an oversight; I thought it didn't have any deps waiting
[02:10] <doko> elmo, mdz: python2.3 is needed for zope2.x. upstream doesn't do security audits with python2.4 and zope2.x, probably they won't do it for the 2.x release cycle anymore
[02:21] <lamont> mdz: any objections to me lsb-base-ing ipsec-tools (racoon, setkey)
[02:21] <lamont> ?>
[02:24] <mdz> lamont: not if it's an easy one
[02:25] <lamont> it's pretty trivial
[02:29] <bob2> I wonder why my desktop always has the fan on now
[02:29] <bob2> even when doing nothing
[03:28] <Keybuk> quiet tonight
[03:34] <ajmitch> hopefully because people are working hard
[03:40] <Keybuk> mdz: why "obsolete after the release" ?
[03:53] <mjg59> mdz: Around?
[04:05] <lamont> mdz: adding all the xorg-{driver,input} packages for hppa - any objections?
[04:05] <lamont> literally just changing debian/control Architecture: lines, and hppa manifest
[04:09] <seth_k> hmm, daniels, re: that xkbcomp command, after I paste in those lines to xkbcomp it just sits and acts like it wants more input. How do I terminate the input?
[04:14] <daniels> seth_k: hit ctrl-d
[04:14] <seth_k> thanks
[04:14] <seth_k> posting output in the bug now
[04:14] <daniels> ta
[04:22] <Keybuk> mmm... anonymous enums
[04:23] <lamont> anonymity is a good thing
[04:24] <lamont> daniels: you planning a -72 upload?
[04:24] <bddebian> Yeah, then how come we have to sign packages? ;-)
[04:24] <Keybuk> he used to whinge at me to use #define instead for defining numerical constants
[04:25] <daniels> lamont: hppa's broken, eh?
[04:25] <lamont> well, xserver-xorg is uninstallable is all
[04:25] <daniels> lamont: bug reports in diff form are appreciated ;)
[04:26] <lamont> daniels: yeah, testing my change now
[04:29] <lamont> daniels: so far, it's just Arch lines for all the packages that correspond to _drv.o files, plus the manifest file changes... anything else I should be looking at?
[04:29] <daniels> seth_k: sudo dpkg -i --force-confmiss /var/cache/apt/archives/xkeyboard-config_0.6-2_all.deb
[04:30] <lamont> Keybuk: well if you declare them "const foo=2' that's one thing... adding random variables is more scary...
[04:30] <daniels> lamont: that should e it
[04:30] <lamont> daniels: woot
[04:30] <daniels> lamont: no, wait
[04:30] <daniels> lamont: debian/scripts/vars.hppa also, IIRC
[04:30] <seth_k> daniels, dpkg - warning: downgrading xkeyboard-config from 0.6-3 to 0.6-2.
[04:31] <seth_k> daniels, and then I get a TON of Configuration file xxxxxxxxxx does not exist on system, installing new config file as you request
[04:31] <seth_k> daniels, restart X now?
[04:31] <lamont> daniels: does vars.hppa get used before install stage?  taht is, can I just fix the file mid-build, or do I need to kill the testbuild?
[04:32] <daniels> seth_k: yeah
[04:32] <Keybuk> lamont: he used to prefer
[04:32] <Keybuk> #define FOO_LEN 1
[04:32] <Keybuk> #define BAR_LEN 4
[04:32] <Keybuk> I prefer
[04:32] <Keybuk> enum { FOO_LEN=1, BAR_LEN=4 };
[04:33] <daniels> seth_k: and all should be good.  that's an error that only triggers when upgrading through a small set of versions of xlibs.
[04:33] <daniels> lamont: unfortunately, it gets sourced first thing
[04:33] <lamont> sigh
[04:33] <seth_k> daniels, safe to upgrade back to 0.6-3 ?
[04:33] <lamont> does XSERVER_XORG_SPECIAL_DEPENDS need all the -input-* pkgs, or just the -driver* packages?
[04:33] <seth_k> daniels, it gave me a downgrade warning
[04:34] <lamont> ah, all the -input- packages are there.  that would explain it
[04:36] <lamont> daniels: build time is ~1 hour or so... will advise
[04:36] <lamont> daniels: and when do you plan to upload?
[04:36] <daniels> seth_k: yeah, that's fine
[04:36] <daniels> lamont: just -driver; -input- is already taken care of, iirc
[04:36] <daniels> lamont: preferably today
[04:37] <seth_k> thanks for your help daniels, I'll mark the bug as invalid
[04:37] <daniels> seth_k: no worries, ta
[04:38] <lamont> daniels: if all goes well, I'll have a diff for you in about an hour
[04:38] <Amaranth> #15636 needs a sync from pyxdg cvs to fix it
[04:39] <lamont> 642 lines of diff -u
[04:57] <jdahlin> daniels?
[04:58] <daniels> jdahlin: mmm?
[04:59] <jdahlin> daniels: I noticed while upgrading to breezy today that some xorg package is doing something like: find / *_drv*
[05:00] <daniels> it shouldn't be, but apparently it does sometimes
[05:00] <jdahlin> I couldn't upgrade, it was a box with nfsroot and many many files in some nfs mounts
[05:01] <daniels> ouch
[05:01] <daniels> okay, I'll check it out, think the fix should be fairly easy
[05:01] <mdz> mjg59: yep
[05:04] <infinity> Oh, look, it's an mdz.
[05:04] <mdz> s/an/THE/
[05:04] <jdahlin> daniels: should I file a bug?
[05:06] <mdz> infinity: were you able to find out what was going on with 10080, or are we hoping against hope that the new version fixes it?
[05:06] <daniels> jdahlin: it's okay, already sorted locally
[05:06] <infinity> mdz : Due to the fact that I once, in a previous life, cared about VGA on a deeper level than I now do, mjg59 has asked me to petition for a kernel freeze exception to drop vga16fb's default res from 640x480 to 640x400 (which, if I remember naything from the bad old assembly/demo days, should resolve a lot of the "wraparound usplash/installer" issues people are having)
[05:07] <mdz> infinity: sounds awfully risky
[05:07] <mdz> we'd also need to resize the artwork, no?
[05:07] <infinity> mdz : Well, the original submitter of that bug was incredibly unhelpful (as you've noted), but other bug submitters have been more forthcoming about their problems, and the pre-up thing should indeed fix it.
[05:07] <infinity> mdz : He has a patch ready, and the artwork can be resized post-haste (in fact, lots of the artwork was alreayd done in both sizes, apparently)
[05:08] <infinity> mdz : The problem is that 640x480 was an extension to the original VGA specs, and doesn't really work everywhere, and overflows fixed buffers allocated for VGA on lots of really old and really new cards (more of the ones in the middle seem to get it right)
[05:09] <mdz> infinity: but it's the only mode we've tested for the past month :-/
[05:09] <infinity> mdz : We'd also lose 5 lines out of the installer (going from 80x30 back to 80x25) unless we used a smaller font, but that's probably not worth caring about.
[05:09] <mdz> and usplash testing is hard to get because people need to regenerate their initrds manually at the right time
[05:10] <infinity> mdz : Yeah, and if I hadn't heard mdz complaining about problems with it this morning, I wouldn't have spoken up at all.  I didn't realise the issues were as widespread as they seem to be.
[05:10] <mdz> actually we have more than a month of 640x480 testing, since the installer's been using it for ages
[05:10] <mdz> I think we ought to wait until dapper
[05:10] <infinity> There shouldn't be a computer on earth for which 640x400 would be a regression, but it should help a lot of broken ones.
[05:10] <infinity> But yes, it can wait, if you're really nervous about it.
[05:10] <mdz> by using 640x480, we're staying in the same mess we've been in since warty
[05:11] <mdz> with 640x400 we risk a new and unknown mess
[05:11] <mdz> daniels: yeah, with known, documented workarounds
[05:11] <mdz> and on the rest of the world's computers, it works splendidly
[05:11] <infinity> Yeah, the documented workaround of using test-only or vesafb should work fine (one or the other) for most people.
[05:11] <infinity> s/test/text/
[05:12] <mdz> I like the idea; I just think we're too close to release to make the change now
[05:12] <infinity> Yeah, you're likely right.  I wish I'd noticed earlier (would have been "nice" if I had broken hardware here, I would have immediately leapt on it..)
[05:12] <infinity> Oh well.
[05:12] <wasabi> mdz, i've heard you were doing some work to get unionfs support in the standard initramfs but ran into some kernel crashes (source: jbailey)
[05:13] <wasabi> Have a resource to fill me in with?
[05:13] <infinity> 320x200 and 640x400 are the only two modes I know of that work pretty much on all PC hardware within the last 15 years.
[05:13] <mdz> wasabi: not exactly; here's the story
[05:13] <mdz> wasabi: we patched in unionfs early in breezy, with the aim of using it for ltsp and the live cd
[05:13] <infinity> (Well, and some strange pixel-doubling modes, like 320x100, 160x200, and other oddities)
[05:14] <mdz> wasabi: it proved itself to be unusably buggy for ltsp, and so we never even enabled it for the live cd (Mithrandir made the patch but it's unreleased)
[05:14] <wasabi> Ahh.
[05:14] <mdz> wasabi: since it wasn't being used for anything and was not behaving very supportably, the kernel team preferred to remove it
[05:14] <wasabi> I was looking into attaching it to my initramfs so that I can specify unionfs on the kernel command line to have root rw without saving.
[05:14] <wasabi> Would make it more convientent for installing on embedded stuff for me.
[05:15] <infinity> mdz : Anyhow, the wraparound isn't going to blow up anyone's hardware or anything, just look ugly, so I guess if we document it, and make the change as early as possible for dapper to get testing, that'll have to do.
[05:15] <mdz> infinity: i think that's the best plan
[05:16] <mdz> wasabi: the code for that is already in the ltsp package
[05:16] <mdz> wasabi: you just need to provide the module
[05:16] <infinity> mdz : As for 10080, did you get the CC on my followup to 10334?
[05:16] <mdz> infinity: I did
[05:16] <mdz> have you read the diff?
[05:16] <infinity> Yeah, it's sane.
[05:16] <mdz> I've never had to set up pppoe, thankfully, so I don't know what to expect from it
[05:17] <mdz> can you mail me the debdiff?
[05:17] <infinity> It's also been in Sarge for ages with pretty good results, if the BTS is to be believed.
[05:17] <wasabi> Also, I was wondering about some sort of loop-back file support for /
[05:17] <wasabi> For instance:
[05:17] <wasabi> root=/dev/embedded/flash:/rootfs.img
[05:17] <wasabi> See where I'm coming from?
[05:17] <infinity> mdz : Yeah, I'll do the merge manually, and mail you the 1.6ubuntu2->1.8ubuntu1 debdiff.
[05:18] <infinity> mdz : I assume you don't mind pulling in the minor translation fixes from 1.7 and 1.8?
[05:18] <mdz> wasabi: with the initramfs script in ltsp, you just switch it on and say "throw a tmpfs on top of my root before starting up userland"
[05:18] <Keybuk> *giggles* at gcov
[05:18] <mdz> infinity: don't mind that
[05:18] <Keybuk> oh dear, I appear to have not tested --help
[05:21] <wasabi> Basically, that last example, mounts root as a loopback fs.
[05:21] <wasabi> Makes managing a flash-based system easier. You distribute a root file, which is put on the flash device, vs actually having a real partition (lots of scattered writes)
[05:25] <desrt> daniel does stir
[05:25] <desrt> daniels; any way i can get more useful information for you on bug 16122?
[05:28] <daniels> desrt: dunno, sorry; i don't do l-r-m anymore
[05:28] <daniels> (nor do I have any idea about that particular bug)
[05:28] <desrt> ah right.  fglrx is a kernel thing, not x
[05:28] <desrt> kthx.
[05:36] <lamont> daniels: sigh... ~50% of the way through the log file
[05:37] <Keybuk> bloody hell, this is what I call an odd release cycle:
[05:37] <Keybuk> 1.0.0  2005-09-29
[05:37] <Keybuk> 0.1.0  2004-10-31
[05:38] <lamont> Keybuk: clearly a busy year
[05:38] <lamont> or a major re-scoping
[05:38] <Keybuk> neither, just zero bugs all year
[05:38] <Keybuk> I must've been on bloody good form when I wrote it
[05:41] <lamont> heh - hct?
[05:41] <Keybuk> nah, grepmap
[05:41] <bob2> hah
[05:42] <bob2> and now it's almost obsolete
[05:43] <Keybuk> indeed
[05:43] <Keybuk> still, it'll be used for breezy so figured I may as well make life easier for people
[05:43] <Keybuk> reduced it's strictness about map file format a little
[05:44] <jdub> mdz: http://lwn.net/Articles/153187/ <- interesting from the POV of solaris contracts on linux (and SMF) :-)
[05:44] <Keybuk> jdub: the kernel guys are going to run out of letters to put in front of "notify" soon
[05:45] <jdub> Keybuk: that's okay, there's always notifyfs to bring it all together in the linux style :-)
[05:45] <Keybuk> hmm
[05:45] <Keybuk> who broke metacity btw?
[05:45] <Keybuk> when I Tab through windows, it only ever-so-briefly flashes the window border rather than keeping it on the screen all the time
[05:45] <mdz> jdub: oh, is that what SMF does?
[05:46] <jdub> mdz: well, SMF depends on solaris contracts, which you could compare with pnotify
[05:50] <lamont> daniels: building xc-xserver-xorg-dbg
[05:51] <daniels> Keybuk: is the window area filled with black?
[05:51] <daniels> Keybuk: if so, fix your xorg.conf
[05:51] <Keybuk> daniels: no, not that I can see, why?
[05:51] <Keybuk> it just draws the border then draws straight back over the top of it again
[05:51] <Keybuk> fix how, ooi?
[05:52] <daniels> *shrug*, wfm
[05:52] <daniels> add your modules section back so you have SHAPE
[05:52] <Keybuk> I don't have shape in there ...
[05:52] <Keybuk> where did that go?
[05:52] <daniels> no-one ever has, or probably will
[05:52] <daniels> hint: Load "extmod"
[05:52] <Keybuk> extmod is in there
[05:53] <infinity> desrt : ping.
[05:53] <daniels> right.  but that's only relevant to the window-area-is-painted-solid-black problem.
[05:53] <Keybuk> Section "Module" has GLcore, i2c, bitmap, ddc, dri, extmod, freetype, glx, int10, type1, vbe
[05:53] <desrt> infinity; pongtastic!
[05:53] <daniels> right
[05:53] <Keybuk> daniels: when you alt-tab do you get a permanent black border around the window ?
[05:53] <daniels> Keybuk: yes
[05:53] <Keybuk> and are you up to date today?
[05:54] <infinity> desrt : Is that on an intel AGP chipset?
[05:54] <Keybuk> any other ideas what it could be?
[05:55] <desrt> yes
[05:55] <Keybuk> I've only noticed it in the last few days, so it's a recent breakage
[05:55] <desrt> 0000:00:01.0 PCI bridge: Intel Corp. 82865G/PE/P PCI to AGP Controller (rev 02) (prog-if 00 [Normal decode] )
[05:55] <infinity> desrt : Can you try 'Option "UseInternalAGPGART" "no"' (or "yes", I'm not sure which way is which) in your conf?
[05:55] <desrt> i have it set to no
[05:55] <Keybuk> jdub: hmm, but unotify is the netlink event socket for the sysfs file system already <g>
[05:55] <desrt> i've tried both... they give different errors :)
[05:56] <Keybuk> (as apposed to using dnotify on sysfs)
[05:56] <daniels> Keybuk: yes
[05:56] <desrt> power is acting up a bit here (bit storm)
[05:56] <desrt> so if i disappear that's why
[05:56] <infinity> desrt : Ahh, joy.  Can I get the errors from both in the bug log?
[05:56] <Keybuk> daniels: spill the beans :p
[05:56] <infinity> desrt : Clearly marked, so I know which is which? :)
[05:57] <daniels> Keybuk: ask the metacity maintainer
[05:57] <Keybuk> already filed a but
[05:57] <Keybuk> though it hasn't been changed since August
[05:58] <Keybuk> oh, no, wait, new version a few weeks ago
[05:58] <Keybuk> 2005-09-07 13:38:16 status installed metacity 1:2.12.0-0ubuntu1
[05:58] <Keybuk> hmm, no, I've definitely noticed the change _since_ then
[05:59] <Keybuk> it basically looks like something's redrawing the window
[06:00] <desrt> infinity; done
[06:08] <mdz> infinity: ok, pppoeconf++
[06:10] <infinity> Uploaded.
[06:11] <desrt> infinity; do you have any more questions/stuff to try?
[06:11] <desrt> if not, i'm gonna hit the proverbial sack
[06:12] <infinity> desrt : I'm looking at it in the background, but not dedicating too many cycles to it, so go have a happy nap. :)
[06:12] <desrt> awesome.  cheerio :)
[06:13] <infinity> desrt : Erm, wait.
[06:13] <infinity> desrt : Which kernel is this on :)
[06:13] <desrt> 686-smp
[06:13] <infinity> I had a feeling it would be SMP.
[06:13] <infinity> Out of curiosity, can you see if it goes away with -686?
[06:14] <desrt> arf
[06:14] <desrt> sure
[06:14] <infinity> (obviously not a solution, just curious..)
[06:14] <desrt> downlaod eta 4m
[06:14] <infinity> mtrr on SMP kernels is slightly different than on UP kernels, and it's possible something's not accounting for that.
[06:15] <desrt> it's not real SMP
[06:15] <desrt> just HT
[06:15] <desrt> so if push came to shove i'd probably suck it up
[06:15] <desrt> i think i'm going to buy a new PC circa january anyway
[06:18] <desrt> k.  reboot.  brb.
[06:18] <Keybuk> gnargh
[06:18] <Keybuk> s/unstable/breezy/ la la la
[06:20] <desrt> infinity; absolutely no impact
[06:21] <desrt> infinity; still getting the same mtrr errors in the dmesg and direct rendering disabled
[06:21] <infinity> \o/
[06:21] <infinity> Oh well.
[06:21] <infinity> Thanks anyhow. :)
[06:23] <desrt> night.
[06:27] <fabbione> morning
[06:28] <Keybuk> morning
[06:35] <lamont> daniels: ping
[06:36] <daniels> lamont: diff?
[06:36] <lamont> well, wondering what happened to my changes to control...
[06:36] <lamont> do you regenerate it or some such silliness?
[06:36] <daniels> ... no
[06:37] <lamont> hrm..
[06:38] <lamont> Package: xserver-xorg-driver-ati
[06:38] <lamont> Architecture: alpha amd64 arm hppa i386 ia64 hurd-i386 m68k powerpc sparc
[06:38] <lamont> and yet we didn't build that
[06:38] <lamont> or rather, never installed it
[06:38] <lamont> I added the driver/input files to MANIFEST.hppa.in, but they didn't get installed, so it hates me
[06:39] <lamont> daniels: for where I'm at: http://people.ubuntu.com/~lamont/xorg.diff
[06:41] <infinity> Is it actually building the drivers?
[06:47] <lamont> infinity: I got the list by saying 'find . -name '*_drv.o' and adding the arch to debian/control, and the destination files to MANIFEST.hppa.in
[06:54] <jdub> mdz: cliff got in touch
[06:56] <daniels> lamont: so xserver-xorg-driver-ati*.deb is empty?
[06:57] <mdz> jdub: what's the status?
[06:57] <lamont> daniels: given that we never get around to building the .debs....
[06:57] <infinity> The MANIFEST check fails?
[06:57] <lamont> because the manifest check fails
[06:57] <daniels> so MANIFEST.hppa.new doesn't have ati_drv.o?
[06:57] <jdub> mdz: going back and forth on the splash atm, but background is sorted
[06:57] <lamont> the string 'xorg-driver-ati' does not appear in the output
[06:58] <bob2> does hppa even have pci?
[06:58] <mdz> jdub: on track to upload tomorrow?
[06:58] <lamont> daniels: I preadded it to hppa.in
[06:58] <lamont> and it's not in .new, so we fail
[06:58] <jdub> mdz: very much so (have plan b artwork as well, just in case)
[06:58] <daniels> lamont: so make install isn't carrying it over, if it's built at all
[06:58] <lamont> ./build-tree/xc/programs/Xserver/hw/xfree86/drivers/ati/ati_drv.o
[06:59] <lamont> daniels: I GOT THE LIST BY FINDING ALL OF THE _DRV.O FILES THAT WERE BUILT
[06:59] <daniels> lamont: ... okay ...
[06:59] <lamont> and that's correct, install isn't installing it
[06:59] <lamont> and I freely admit to having NFC why
[07:00] <infinity> Did hppa have an ati driver in the xorg packages in the old world order?
[07:00] <infinity> (obviously it was built, but was it ever installed?)
[07:01] <daniels> i have no idea why it would occur
[07:01] <fabbione> lamont, infinity: you could compare xorg -10 with what's now
[07:02] <fabbione> in -10 the MANIFEST files were ok and updated
[07:02] <daniels> that'd be getting ObjectModuleTarget with InstallObjectModule
[07:02] <daniels> and hppa is totally SEP for me
[07:02] <fabbione> daniels: on the good side sparc works.. except the keyboard layouts :(
[07:02] <daniels> fabbione: what's wrong with them?
[07:03] <daniels> lamont: oh yeah, welcome to the world of no loadable servers
[07:03] <fabbione> daniels: i can't say exactly.. i know jbailey did play with it yesterday.. X even manage to autoconfigure :)
[07:03] <fabbione> daniels: at least with an ATI card
[07:03] <fabbione> but it's like you press A and comes out something else
[07:03] <daniels> fabbione: and you know your discover1 changes were completely pointless, right?
[07:03] <daniels>       # must be Discover 1.x
[07:03] <daniels>       DISCOVERED_VIDEO=$(discover --disable-all --enable=pci \
[07:03] <daniels>                                   --format="%V %M\t%S\t%D\n" video 2>/dev/null)
[07:04] <infinity> lamont : There are nodriver modules in xserver-xorg in Debian/hppa/sid, perhaps that's just the correct thing? :)
[07:04] <fabbione> daniels: without my changes discover1 can't see the SBUS -> PCI bridge
[07:04] <daniels> correct for hppa, yes, as no-one's bothered to write an elfloader bit for it
[07:04] <lamont> infinity: well, having the xserver-xorg package not installable is what I was trying to fix.
[07:04] <fabbione> daniels: so it can't detect PCI video cards
[07:04] <fabbione> daniels: they have been tested dude 
[07:04] <infinity> lamont : So, the answe may just be to not have it depend on any drivers for hppa.  That would make it match sid's package, for functionality.
[07:05] <fabbione> daniels: the issue was above PCI.. and we still need SBUS to detect esp and some lan cards
[07:05] <daniels> ah, okay
[07:06] <daniels> lamont: well, it'll probably be broken for a while
[07:06] <fabbione> daniels: + the code we had was segfault-o-rama :P
[07:06] <daniels> that does sound like discover, yes
[07:07] <lamont> daniels: atm, I really don't care if the xserver loads... I just want ubuntu-desktop installable, and xserver-xorg's deps are the only thing in the way
[07:07] <daniels> lamont: basically, you need to find a way to not build -input- on hppa, and to have -core provide -input-* and -driver-*
[07:08] <lamont> hrm... OK. but not tonight
[07:41] <infinity> mdz : ping.
[07:46] <ivoks> 'morning
[07:50] <jsgotangco> morning ivoks, sabdfl
[07:51] <Mithrandir> hi mdz
[07:51] <sabdfl> allo allo
[07:51] <ivoks> rene :)
[07:52] <Mithrandir> hi sabdfl
[07:53] <ajmitch> hello sabdfl 
[08:03] <fabbione> daniels: when do you plan the next X upload?
[08:06] <fabbione> daniels: did you also get the diffs from people.u.c/~fabbione ?
[08:07] <fabbione> (i did publish both 70-to-71 and 71-to-72)
[08:09] <daniels> you mean 69-to-70 and 70-to-71
[08:10] <fabbione> oui..
[08:11] <fabbione> daniels: i might have the fix for the keyboard layout stuff pretty soon
[08:12] <fabbione> that's why i was asking when you plan to upload X
[08:12] <daniels> sometime about now
[08:12] <fabbione> ok
[08:12] <fabbione> no .. it won't be THAT soon
[08:12] <daniels> okay
[08:12] <jdub> http://static.flickr.com/29/46304339_cb5ec248e9_o.gif <- X
[08:13] <fabbione> just go ahead.. if they will show up, i will do a porter upload
[08:13] <fabbione> jdub: hahaha
[08:13] <daniels> fabbione: gnnrg
[08:13] <fabbione> gnnrg????
[08:13] <daniels> fabbione: i'd really prefer if you just gave me the diff, unless you need it in the archive YESTERDAY GODDAMNIT
[08:13] <daniels> it makes my life about a bajillion times easier
[08:14] <daniels> jdub: x is love
[08:14] <fabbione> daniels: i needed them in the archive yesterday
[08:14] <daniels> fabbione: ... sparc is release-critical?
[08:15] <fabbione> daniels: no, but the other 3 fixes are
[08:17] <jdub> that image c/o a dude in jordan who digs ubuntu :)
[08:20] <tepsipakki> jdub: I didn't quite get your post regarding dependency init.. do you mean that Sun SMF is a better bet than initng etc?
[08:20] <daniels> fabbione: which other three fixes?
[08:20] <jdub> tepsipakki: absolutely
[08:20] <tepsipakki> jdub: me too
[08:20] <fabbione> daniels: the 3 bug fixes.. they were all Major or >
[08:21] <fabbione> daniels: + yesterday you said yourself that it was ok for you that i did the uploads, because you had nothing in the queue
[08:21] <daniels> actually, that's not quite what I said
[08:22] <fabbione> daniels: i did ask to you if you had pending changes and you said no.. -71 is your
[08:28] <fabbione> daniels: can i fix xfonts-core? or do you have any objection with it?
[08:30] <daniels> fabbione: i said -71 is yours, yeah
[08:30] <daniels> and that I'd catch up with -72 today, which I've done
[08:30] <daniels> what else are you doing to xfonts-core?
[08:31] <fabbione> daniels: the fix i did yesterday (missing simlink) is not enough
[08:31] <fabbione> see the last comment on 15611
[08:31] <fabbione> we need to change /usr/X11R6/lib/X11/fonts/* from dirs to symlinks
[08:35] <fabbione> hey JanC 
[08:35] <fabbione> meh
[08:35] <fabbione> JaneW
[08:36] <Mithrandir> mdz: pongpong?
[08:36] <daniels> fabbione: no, that will break universe font packages
[08:36] <daniels> fabbione: right now, the failure mode is that you don't get fonts from some packages when you upgrade
[08:36] <daniels> fabbione: i'd really, really, rather not complicate that by another transition
[08:37] <fabbione> daniels: the problem is that right now it doesn't work for main either
[08:37] <fabbione> because xorg.conf is not always regerated (and can't be) with the new paths
[08:37] <fabbione> regenerated
[08:38] <fabbione> a solution would be to symlink the contents of /usr/share/X11/fonts/100dpi/* in /usr/X11R6/lib/X11/fonts/100dpi
[08:38] <fabbione> or something like that
[08:38] <fabbione> that would maintain the dirs s they are
[08:39] <fabbione> and make both happy
[08:40] <daniels> if people make a config file with xorgcfg or whatever, they're on their own anyway
[08:40] <daniels> noting that we don't even ship that tool anymore
[08:40] <daniels> it works for all configurations we've ever generated from dexconf
[08:42] <fabbione> daniels: it's enough that the config doesn't match the md5sum by just adding a whitespace
[08:42] <fabbione> echo "" >> /etc/X11/xorg.conf
[08:42] <fabbione> and bam
[08:46] <dholbach> good morning
[08:46] <daniels> fabbione: so?
[08:46] <dhonn> theres a bug in gnome search where I cant right click and choose a program to open a file
[08:46] <daniels> fabbione: if you take an XF86Config-4 from warty and drop it into a breezy system, it will work
[08:46] <daniels> fabbione: this is without dexconf doing rewriting tricks
[08:47] <daniels> fabbione: the only place you get /usr/X11R6/lib/X11/fonts from is external configuration tools
[08:47] <daniels> i think the costs here outweigh the benefits by far
[08:57] <pitti> Good morning
[08:57] <dholbach> pitti: how was the moving?
[08:58] <mpool> hi
[08:58] <zyga> hello folks
[08:59] <fabbione> daniels: if i take a xorg.conf or XF86 with just an extra empty line (no need of external config tools) X will fail to start
[09:00] <mpool> hi, i'm seeing the ipv6 dns slowdown too
[09:00] <pitti> hi zyga 
[09:00] <fabbione> daniels: not making this fix will kill all users that have a modified config
[09:00] <mpool> i understand what's happening and how to work around it for myself
[09:00] <mpool> but i think the default will be pretty bad for many users
[09:00] <zyga> can we modify ATI's fglrx stuff anyway?
[09:00] <fabbione> mpool: hoary or breezy?
[09:00] <dholbach> hi carstenh 
[09:00] <carstenh> hi dholbach 
[09:01] <zyga> I wanted to use fglrx_config recently and it works semi-okay but spits out a config file for XFree86 not for x.org
[09:01] <mpool> fabbione: breezy
[09:01] <mpool> it wasn't this bad in hoary
[09:01] <fabbione> mpool: that's really weird..
[09:01] <daniels> fabbione: dude
[09:01] <mpool> my default nameserver just ignores AAAA requests
[09:01] <daniels> fabbione: take an XF86Config-4 from warty
[09:01] <mpool> so there's a long timeout before breezy's resolver backs off to A requests
[09:01] <daniels> fabbione: drop it into breezy
[09:01] <hunger> zyga: Jut copy the section for the graphics card into your existing xorg.conf.
[09:01] <fabbione> because breezy glibc don't resolv ipv6 as first
[09:01] <daniels> fabbione: and it will work
[09:02] <daniels> fabbione: it doesn't matter if you modify it or not, things will just work
[09:02] <mpool> fabbione: ok, i thought so
[09:02] <mpool> fabbione: this happens even if i disable the ipv6 kernel module
[09:02] <daniels> fabbione: this is because I already added symlinks in /usr/lib/X11, which we used in warty and hoary
[09:02] <mpool> it's pretty yucky
[09:02] <mpool> i'm concerned that people who have nameservers that don't handle ipv6
[09:02] <daniels> fabbione: the only tools which generate anything referring to /usr/X11R6/lib/X11 are external tools like xorgcfg, which we cannot and do not support
[09:02] <mpool> which is probably fairly common
[09:03] <mpool> will just think hoary is really slow
[09:03] <zyga> hunger: yes I know but this could be easily fixed
[09:03] <daniels> fabbione: so, just to be clear -- if you sit on warty or hoary and modify your config file by exactly one character, and nothing in it gets changed -- things will still work
[09:03] <fabbione> daniels: whatever.. i reassinged the bug to you.. handle it as you prefer
[09:04] <hunger> Is there a way to add custom lines to sections of xorg.conf via dexconf?
[09:04] <mpool> fabbione: what do you think? is there anything we can do?
[09:04] <fabbione> mpool: with breezy glibc ipv4 is always resolved before ipv6, but if the apps request ipv6 first, than you might encounter that problem
[09:04] <fabbione> mpool: the general thing is ok
[09:04] <fabbione> so it depends from what apps is giving you problem
[09:05] <mpool> what do you mean by "the general thing is OK"?
[09:05] <sivang> morning all
[09:05] <fabbione> if the DNS you are using is not RFC compliant, fix the dns :)
[09:05] <fabbione> we do the right the thing
[09:06] <fabbione> we are RFC and posix compliant
[09:06] <fabbione> if an application specifically asks for ipv6.. well it has the rights to do so
[09:06] <mpool> ok
[09:06] <mpool> maybe more apps are built with ipv6 turned on?
[09:06] <mpool> i see that if i just call gethostbyname it does only the ipv4 request
[09:06] <mpool> which is good
[09:06] <fabbione> there is no need to tell the apps to enable/disable ipv6
[09:07] <mpool> so
[09:07] <fabbione> it's enough they use gethostbyname2 instead of gethostbyname
[09:07] <mpool> i note your smiley, but many people are not able to fix their dnses very easily
[09:07] <fabbione> yes, but neither we are going to be non-{rfc,posix} compliant
[09:07] <fabbione> the timeout is a perfectly allowed condition
[09:08] <fabbione> normal i would say
[09:08] <mpool> i thought the previous behaviour was that if there were no ipv6 routes/addresses, it wouldn't do the lookups
[09:08] <mpool> that seems like a reasonable heuristic
[09:08] <mpool> isn't it?
[09:08] <fabbione> nope..
[09:08] <fabbione> you are still allowed to perform dns queries
[09:09] <fabbione> the algo is complex.. but in few words it does all the queries to figure out the matching source -> dst ip
[09:09] <fabbione> the fact that ipv6 is not loaded or configured will be checked at a later stage
[09:09] <fabbione> and this is ok, because the apps might be doing looks up to enable ipv6..
[09:09] <fabbione> you cannot know that in advance
[09:10] <mpool> right
[09:10] <mpool> it's a reasonable approach
[09:10] <mpool> however
[09:10] <mpool> there doesn't seem to be any way for people with broken DNS servers to turn off ipv6 queries 
[09:11] <fabbione> you can't
[09:11] <fabbione> a dns query is nothing more than translating strings into strings
[09:11] <fabbione> and the point is that you can do ipv6 queries on an ipv4 network
[09:11] <fabbione> and viceversa
[09:12] <mpool> sure
[09:12] <mpool> i understand why the default is compliant
[09:12] <mpool> but i think this will bite a number of users
[09:12] <mpool> don't you agree?
[09:12] <fabbione> note that breezy is much better than hoary...
[09:13] <fabbione> even if i agree on the general annoyance of a timeout, i won't agree with solutions that will disable ipv6
[09:14] <fabbione> not even configurable ;)
[09:14] <mpool> ok
[09:14] <mpool> so what do you think should happen to people with servers like this?
[09:14] <fabbione> ISP's need to become compliant to RFC's :)
[09:14] <mpool> mm
[09:14] <fabbione> mpool: after the first lookup, the resolver learns
[09:15] <fabbione> so it's only at the first contact the problem
[09:15] <fabbione> after that it gets faster
[09:15] <mpool> fabbione: no, it doesn't
[09:15] <fabbione> it shoul
[09:15] <fabbione> +d
[09:15] <mpool> when i first installed Breezy, I had a ~6s delay on every tcp connection
[09:16] <mpool> it makes a pretty bad first impression i have to say
[09:16] <mpool> and not everyone will be able to figure out what's going on
[09:16] <fabbione> mpool: what apps are you using to time this things?
[09:16] <mpool> apt?
[09:16] <mpool> every apt-get request pauses for ~6s
[09:16] <mpool> ssh
[09:16] <mpool> telnet
[09:16] <ivoks> ?
[09:16] <fabbione> ssh no
[09:17] <fabbione> i don't believe it
[09:17] <fabbione> because it uses gethostname
[09:17] <fabbione> and not gethostbyname2
[09:17] <mpool> firefox has an option to turn off ipv6 dns, because it's broken for a signficant fraction of users
[09:17] <mpool> what i would like to see is a similar system-wide setting
[09:17] <fabbione> that's how i figured the difference between hoary and breezy resolver
[09:17] <mpool> i don't think we can ignore the fact that many users do have DNS servers that don't handle ipv6
[09:17] <mpool> i hear RH and Fedora have a setting like this
[09:18] <fabbione> apt uses gethostbyname too
[09:18] <fabbione> so it defaults to ipv4
[09:18] <mpool> fabbione: anyhow, resolv.conf(5) implies that AAAA records are only consulted if 'options inet6' is set
[09:18] <fabbione> telnet yes.. it does ipv6 first
[09:18] <mpool> but it seems to be always on
[09:19] <fabbione> it's not
[09:19] <fabbione> i am sure
[09:19] <mpool> if it worked as the manpage says, or if we had, say, 'options noinet6' 
[09:19] <mpool> then that would be OK with me
[09:19] <fabbione> because all my networks are ipv6 compliant
[09:19] <fabbione> and i still keep going on ipv4 with ssh & co.
[09:19] <fabbione> that means that ipv6 is not used as first
[09:19] <fabbione> firefox and mozilla are the only 2 apps that are hutterly broken in that way
[09:19] <fabbione> and it has been known for ages
[09:20] <fabbione> because they implement their own variant of the resolver
[09:20] <maswan> mpool: what kind of brokeness do you have?
[09:20] <maswan> I mean, I have no v6 connectivity at all, and I have no problems.
[09:20] <ivoks> me too
[09:20] <fabbione> maswan: the issue shows up when the DNS you are using is broken and can't resolv ipv6 at all
[09:20] <mpool> maswan: as i said above, apps like apt or ssh pause for ~6s when opening a connection
[09:20] <mpool> whcih is pretty broken
[09:21] <mpool> maswan: i suspect you have no ipv6 but you do have a DNS server that responds to AAAA
[09:21] <fabbione> mpool: can you try to use dns.fabbione.net for a test?
[09:21] <maswan> fabbione: which in practice means what? a bind4 from the late 80ies?
[09:21] <fabbione> mpool: and see if it works better?
[09:21] <mpool> maswan: yeah, or maybe a windows based server or something
[09:21] <fabbione> maswan: or probably other implementations
[09:21] <mpool> or also many embedded routers
[09:21] <mpool> for myself, i can just fix it by using a local server or something
[09:21] <maswan> mpool: Oh, right, embedded routers. Or firewalls.
[09:22] <maswan> firewalls == teh suck.
[09:22] <fabbione> mpool: if that doesn't work.. than there is something hutterly broken on your network...
[09:22] <mpool> the problem is all the other people who will install breezy and say "wow it's really slow"
[09:22] <mpool> on forums and lists you can find lots of people who previously tried to work around this by disabling the ipv6 module
[09:22] <mpool> but for good reasons as fabbione says that's no long enough
[09:22] <maswan> fabbione: I have a wireless nat box thingie, and if the firewall in that is turned on, it mangles dns packets. badly.
[09:22] <fabbione> the kernel module has nothing to do with bind/resolver
[09:23] <fabbione> maswan: yeah.. so do i here
[09:23] <fabbione> maswan: but that doesn't mean it's breezy's fault
[09:23] <mpool> fabbione: it used to
[09:23] <fabbione> mpool: never..
[09:23] <fabbione> the module in the kernel provides the protocol
[09:23] <mpool> i think previous resolvers tried to see if the ipv6 protocol was available before doign the lookup
[09:23] <fabbione> it did never interact with the resolver
[09:23] <mpool> which i agree is broken, but at least it gave people a way to control it
[09:23] <fabbione> than it's a glibc thing
[09:24] <fabbione> resolver = glibc
[09:24] <maswan> fabbione: Of course not, problem is that you need to work around the bug. :/
[09:24] <fabbione> maswan: yes.. i know..
[09:24] <fabbione> i have to do the same here for a few more things..
[09:24] <mpool> so
[09:24] <fabbione> mpool: can you do the test i did ask you?
[09:24] <mpool> i think there should be a workaround for people with broken servers
[09:24] <mpool> agree/disagree?
[09:24] <mpool> oh, yes
[09:25] <mpool> going straight to dns.fabbione.net works fine
[09:25] <mpool> it's just my router that's broken
[09:25] <fabbione> see.. the point is.. we can't workaround all possible brokness on the internet
[09:25] <Treenaks> the workaround is fixing the servers, imho
[09:25] <fabbione> Treenaks: exactly
[09:26] <ivoks> that's bugfix, not workaround :)
[09:26] <Treenaks> either reply to all AAAA requests with an error/nxdomain/whatever... or reply normally.. don't drop the request :)
[09:26] <mpool> i'm not asking you to work around all possible brokenness, just this one case which can be fairly easily worked around
[09:26] <mpool> Treenaks: ok, well, you go and upgrade every dns server in the world
[09:26] <mpool> in the meantime
[09:26] <fabbione> ivoks: there is no real workaround.. other than patching the resolver and make it an extra thing to carry around
[09:26] <mpool> for people on such networks, breezy is unusable
[09:26] <fabbione> mpool: so was hoary and warty
[09:26] <mpool> hang on, back in a bit
[09:27] <fabbione> even worst in hoary and warty
[09:27] <fabbione> i need to go to the doctor
[09:27] <fabbione> bbl
[09:28] <fabbione> mpool: i need to go the doc
[09:28] <fabbione> i will be glad to keep talking about it later
[09:28] <ivoks> fabbione: i hope you're ok
[09:28] <mpool> OK
[09:28] <mpool> yeah i hope you're ok too
[09:28] <fabbione> remember that the easiest workaround is to install a local bind
[09:28] <fabbione> i am fine
[09:28] <fabbione> it's only a check
[09:28] <fabbione> :)
[09:28] <mpool> ok, let's talk later
[09:28] <ivoks> fabbione: i hope not The Check :)
[09:29] <fabbione> ivoks: in the worst case they will tell me that i will die soon..
[09:29] <fabbione> so i mean.. what's the big deal :P
[09:29] <maswan> actually, that's a fairly good suggestion, assuming a caching-only bind isn't hairy to install (it wasn't at all last time, so...)
[09:29] <ivoks> fabbione: i'm sure that won't be the case...
[09:30] <torkel> maswan: install networkmanager and you will get bind9 for free :-)
[09:30] <ivoks> :)
[09:31] <dholbach> good morning mvo
[09:31] <mvo> hey dho
[09:31] <mvo> good morning dholbach 
[09:31] <infinity> maswan : The default install of bind is cacheing only, so yeah, it's pretty simple.
[09:32] <infinity> I think it might even offer to munge your resolv.conf in postinst to put 127.0.0.1 at the top, but not sure.
[09:32] <infinity> bind8 used to.  Don't recall what bind9 does.
[09:32] <mpool> 2
[09:32] <infinity> 3
[09:35] <lifeless> 4
[09:35] <bob2> 5
[09:35] <Treenaks> 6
[09:35] <Mithrandir> 7
[09:35] <lifeless> 7
[09:36] <Mithrandir> tsktsk, lifeless blew it
[09:47] <Kamion> fabbione: er, openssh doesn't use gethostbyname, except in like two unimportant places; it uses getaddrinfo nearly everywhere
[09:51] <mpool> hi Kamion
[09:51] <mpool> i see you replied to one of these requests previously
[09:52] <Kamion> mpool: hmm, which?
[09:52] <mpool> asking for a way to disable ipv6 dns queries
[09:54] <mpool> in hoary, it was possible to do this by disabling the ipv6 module
[09:58] <Kamion> did I reply to such a request specifically in the context of openssh, or somewhere else? sorry, brain like a ... thing with holes in
[09:58] <infinity> sieve
[09:58] <bob2> php?
[09:59] <mpool> no, more generally
[09:59] <mpool> mm
[09:59] <mpool> let's wait for fabbione
[09:59] <mpool> or you can read the scrollback if you have it and want to
[10:04] <Kamion> I glanced through the scrollback on arriving, yes, but I honestly don't remember having replied to a request like this before
[10:04] <Kamion> although I might've commented on a glibc bug, not sure ...
[10:05] <mpool> sorry, i can't find the link
[10:05] <mpool> it's not important
[10:14] <Kamion> ok
[10:20] <zyga> mvo: hey
[10:20] <zyga> mvo: I'll send you pl.po for synaptic later today
[10:21] <pitti> Hi carlos 
[10:22] <carlos> pitti, hi
[10:22] <mvo> zyga: a updated one? your current one went in with the upload yesterday
[10:27] <zyga> hmmm
[10:27] <zyga> mvo: it's missing some stuff 
[10:27] <zyga> mvo: I dont remeber if I translated that in rosetta
[10:28] <zyga> mvo: I did: make -C po update-po
[10:28] <zyga> and it gave me about 60 untranslated stuff
[10:29] <mvo> zyga: can you upload the result po into rosetta?
[10:30] <mvo> today is NonLanguagePackTransationsDeadline
[10:31] <zyga> mvo: ?
[10:31] <zyga> mvo: sure
[10:31] <zyga> mvo: what's that?
[10:31] <mvo> zyga: no translation updates outside of rosetta 
[10:31] <zyga> ah
[10:31] <zyga> okay
[10:32] <mvo> zyga: still CC me, because I think today it's still possible and I will put the update into my svn repo
[10:32] <zyga> okay
[10:34] <mvo> thanks
[10:37] <HiddenWolf> pitti, ping
[10:41] <Kamion> crimsun: please revert your keychain change; it was a bug in ssh-askpass-gnome which has been fixed
[10:42] <crimsun> Kamion, will do.
[10:45] <Kamion> crimsun: (although if it was reported in Ubuntu more recently than 8th September or so, I'd like to know ...)
[10:50] <mvo> pitti: around?
[10:50] <pitti> HiddenWolf, mvo: pong
[10:53] <fabbione> re
[10:53] <fabbione> Kamion: yes sorry.. i mean getaddrinfo
[10:53] <fabbione> but that's changed too in glibc to be posix compliant
[10:53] <mvo> pitti: there seems to be a issue with nautilus-cd-burner and gnome-volume-manager ... if I insert a cdrw (with stuff on it) and trying to write a iso on it, I get "insert a rewritable or blank cd". but there is one (mounted by g-v-m) already. the question keeps poping up unless I manually umount the cd
[10:54] <pitti> mvo: oh, that's odd - n-c-b is supposed to unmount the drive first...
[10:55] <seb128> and is supposed to ask if you want to clear it then
[10:55] <pitti> bah, why does Mandrake do such an exceptionally good job of hiding their SRPMs?
[10:55] <mvo> pitti: I tested it with writing a iso (daily)
[10:55] <mvo> seb128: yes, it asks after I unmounted it manually
[10:55] <seb128> pitti: you need a patch from them?
[10:55] <pitti> mvo: I always use n-c-b, never had a problem with it...
[10:55] <pitti> seb128: yes
[10:56] <seb128> pitti: http://cvs.mandriva.com/cgi-bin/cvsweb.cgi/SPECS/
[10:57] <pitti> seb128: that does not have security update patches
[10:57] <pitti> seb128: but thanks, it's interesting
[10:57] <seb128> pitti: ah, so good luck :)
[10:57] <crimsun> Kamion: reverting it works fine, tested from console and from within GNOME. Thanks!
[10:57] <seb128> you're welcome
[10:59] <mvo> pitti: hm, if I'm the only one with this problem then it may be my system hating me or something :)
[10:59] <seb128> mvo: does hal list the CD as rewritable?
[11:00] <mvo> seb128: I guess so, after I "umount /media/cdrom" it works fine
[11:00] <seb128> oh, interesting
[11:00] <pitti> seb128: so the problem is that n-c-b does not unmount the CD, or it doesn't recognize it as rewritable?
[11:00] <seb128> pitti: does not umount it
[11:01] <seb128> mvo: try settings gconf /apps/nautilus-cd-burner/debug
[11:01] <seb128> gnome-session-remove nautilus
[11:01] <seb128> nautilus
[11:01] <HiddenWolf> pitti, hotplug is yours, right?
[11:01] <jdub> http://mail.gnome.org/archives/networkmanager-list/2005-September/msg00055.html
[11:01] <jdub> ^ interesting for n-m fans
[11:01] <seb128> then you have the verbose of what happen ... dunno if it has something about that
[11:01] <seb128> hey jdub
[11:02] <mvo> seb128: thanks
[11:03] <pitti> HiddenWolf: depends; Keybuk and me usually touch it
[11:04] <HiddenWolf> pitti, my hotplug is taking ages to taking ages to complete since about 2 days. How do I go about creating a decent bugreport about that?
[11:04] <HiddenWolf> pitti, as in taking long enough to drop out of usplash.
[11:05] <ivoks> hello
[11:05] <fabbione> mpool: still around?
[11:05] <ivoks> problems with i386 dbuild
[11:06] <Kamion> crimsun: ok great, thanks
[11:07] <ivoks> i386 build daemon can't authenticate packages for install and fails to build source cause of that...
[11:07] <fabbione> ivoks: what buildd is that?
[11:07] <fabbione> there is the name on top of the build log
[11:08] <ivoks> vernadsky
[11:08] <pitti> HiddenWolf: you can boot in the single user mode, where you can see which module takes so long
[11:08] <fabbione> infinity: ^^
[11:08] <fabbione> ivoks: thanks
[11:08] <ivoks> fabbione: np
[11:08] <pitti> HiddenWolf: please do that, look where it hangs, then you can just write that into the bug report
[11:08] <HiddenWolf> pitti, I will. brb
[11:09] <fabbione> pitti: i think the sparc did finish the build a long time ago :)
[11:10] <pitti> fabbione: it did, I know
[11:10] <pitti> fabbione: I debugged a bit, and then I crashed to bed
[11:10] <fabbione> cool
[11:10] <mpool> fabbione: hi
[11:10] <fabbione> mpool: yo..
[11:10] <pitti> fabbione: when initdb calls postgres, postgres SIGBUSes, but if I call postgres myself in gdb, it doesn't
[11:10] <mpool> fabbione: i fixed my problem by running bind on another machine here
[11:10] <mpool> which is fine for me
[11:10] <mpool> however i think some users will experience this bug
[11:10] <pitti> fabbione: and if I have initdb call postgres through a gdb wrapper, it does not crash either
[11:11] <pitti> fabbione: so I guess I'll wait for Florian to update his mips box for debugging
[11:11] <mpool> i think it's pretty clear from reports in previous releases that there are broken servers which are not easily fixed
[11:11] <fabbione> pitti: you should consider trying only strace
[11:11] <mpool> therefore i would like very much if there's a workaround
[11:11] <fabbione> mpool: yes, but running bind locally is light 
[11:11] <fabbione> given that it will only do cache
[11:11] <mpool> OK
[11:11] <fabbione> + you get a real DNS server implementation
[11:12] <mpool> well, if it's easy to install that and have resolvconf point to it,t hat would be ok
[11:12] <mpool> might need a faq entry
[11:12] <fabbione> imho it's a much better solution than mangling the resolver
[11:12] <fabbione> see this setup:
[11:12] <fabbione> box connected to crappy ISP
[11:12] <fabbione> ipv6 tunnel to whatever free ipv6 isp
[11:12] <fabbione> box dns points to crappy isp
[11:13] <fabbione> you can't use ipv6 at all because it doesn't resolve
[11:13] <fabbione> you install local bind
[11:13] <fabbione> you win over your crappy ISP
[11:13] <fabbione> and you can use ipv6
[11:13] <fabbione> for real
[11:15] <HiddenWolf> pitti, usb, device descriptor/8 - error 110 bunch of errors while my *many* ports are checked.
[11:15] <pitti> HiddenWolf: then please attach the dmesg log, too
[11:17] <mloskot> hi
[11:18] <mpool> fabbione: ok, sounds good
[11:18] <mloskot> Q: I've reported a bug on Bugzilla related to alexandria from Univers, but as I was told on #launchpad, I should report it to the Malone. So, should I report the bug again, to Malone, or leave it as is on Bugzilla?
[11:18] <fabbione> there are other advanteges too..
[11:18] <fabbione> mloskot: move it to malone please
[11:19] <fabbione> mpool: such as you have full control of the cache.. like to kill negative poisoning :)
[11:19] <fabbione> mpool: without having to wait for your ISP cache to die
[11:19] <fabbione> but that goes down to my insane paranoia
[11:20] <infinity> ivoks : It's been fixed.
[11:20] <mloskot> fabbione: but it will stay on Bugzilla too, or how to remove it from bugzilla?
[11:21] <mloskot> fabbione: https://bugzilla.ubuntu.com/show_bug.cgi?id=16602
[11:21] <fabbione> mloskot: just close it..
[11:21] <mloskot> ok
[11:21] <fabbione> i can close it for you if yuo want
[11:21] <mloskot> ok, do it please.
[11:22] <fabbione> done
[11:22] <mloskot> thanks
[11:22] <fabbione> np
[11:22] <mloskot> As I see, there is no selection for release i.e. Breezy and amd64 on Malone, right?
[11:23] <mvo> Kamion: still no usplash on i386-daily-install?
[11:23] <fabbione> dunno really.. i didn't look into malone details yet
[11:23] <mloskot> ok
[11:23] <HiddenWolf> pitti, Attachment #4197  to Bug #16603 Created
[11:24] <HiddenWolf> pitti, chances are it chokes on my printer.
[11:24] <mloskot> Can I use structured or re-structured text (Zope/Plone gadgets) in Malone report bug form?
[11:25] <Kamion> mvo: ok, will have a look later today if nobody beats me to it; though if you're there, I'd appreciate it if you'd investigate
[11:25] <pitti> HiddenWolf: use lsusb to find out what usb 2-5 is
[11:26] <HiddenWolf> pitti, not listed, should be printer then.
[11:26] <pitti> HiddenWolf: did you switch it off after boot?
[11:27] <HiddenWolf> pitti, no. It won't even power on for some reason. I've swithed some cables tho, had some pc trouble earlier.
[11:28] <HiddenWolf> pitti, let me reboot.
[11:29] <JaneW> mdke: ping
[11:32] <mvo> Kamion: I'll have a look now
[11:33] <HiddenWolf> pitti, seems my hardware is troubled.
[11:34] <HiddenWolf> pitti, there was a loose usb cable, when I plug it in any port, my pc won't go past POST.
[11:34] <HiddenWolf> All usb-hardware works fine tho.
[11:34] <Treenaks> HiddenWolf: that was a country-wide ADSL outage.. not your fault (probably)
[11:35] <HiddenWolf> Treenaks, what?
[11:35] <Treenaks> HiddenWolf: seriously though, my laptop won't go POST if I plug in my multi-card-reader-thingy
[11:35] <Treenaks> HiddenWolf: it could be something like that
[11:35] <HiddenWolf> Treenaks, guess it is, seems to be the multi-cardreader in my monitor.
[11:36] <Treenaks> HiddenWolf: Try turning off "USB Legacy" in the bios
[11:37] <HiddenWolf> Treenaks, thanks, I'll try that later.
[12:00] <fabbione> there
[12:00] <fabbione> edubuntu-artwork fixed
[12:01] <ogra> fabbione, ??
[12:01] <pitti> fabbione: will we see naked Fabios as wallpaper now?
[12:02] <ogra> hey, but not for the children
[12:02] <fabbione> ogra: http://people.ubuntu.com/~lamont/buildLogs/e/edubuntu-artwork/0.1.0-7/ <-
[12:02] <ogra> you'd scare them away :p
[12:02] <fabbione> pitti: ahah
[12:02] <ogra> fabbione, bah, that was jbailey
[12:02] <fabbione> ogra: i didn't point the finger to anybody
[12:02] <fabbione> i just took a lock on the package and released
[12:03] <ajmitch> ogra: do you have time to check an edubuntu-specific malone bug?
[12:03] <ogra> ajmitch, there are no edubuntu malone bugs ;) all of edubuntu is in main :p
[12:03] <ogra> ajmitch, tell me :)
[12:03] <ajmitch> ogra: well it was misreported against the wrong package anyway :)
[12:03] <ajmitch> https://launchpad.net/distros/ubuntu/+sources/scsi-idle/+bug/2315
[12:04] <ogra> thats not edubuntu specific at all
[12:04] <ajmitch> no, but he said it worked on a plain breezy box
[12:04] <ajmitch> which is strange :)
[12:04] <ogra> and i asked elkner several times to please repot bugs in bugzilla
[12:05] <ogra> thaere is nothing edubuntu specific that could cause it
[12:05] <ajmitch> I'll reject the bug then
[12:05] <ajmitch> I know
[12:05] <ogra> yup
[12:05] <ogra> make a note that it should go to bugzilla
[12:05] <ogra> *sigh* 
[12:05] <ogra> i even offered him he could mail his bugs to me so i can put them in bugzilla
[12:14] <mvo> mjg59: around? 
[12:16] <Kamion> infinity: just a thought about that vga16fb 640x400 thing from last night; perhaps it would be possible to add a kernel option (disabled by default) that runs vga16fb in 640x400? That way, we could get people to test it with not quite so much risk.
[12:17] <mvo> Kamion: usplash works now, I had to regenerate the initrd so that it included the new usplash-artwork.so
[12:18] <Kamion> mvo: oh, you mean you were upgrading?
[12:18] <mvo> Kamion: no, it was a fresh install, but the cd included 0.1-12. no idea why
[12:19] <Kamion> mvo: you rsynced the image too early this morning
[12:19] <Kamion> 21 8 * * *      for-project ubuntu cron.daily
[12:19] <Kamion> London time, and takes maybe an hour or so to build
[12:19] <Kamion> or a bit less, but anyway
[12:20] <Kamion> today's image definitely has usplash 0.1-14
[12:20] <mvo> Kamion: ok, thanks
[12:36] <mvo> Kamion: can we make termwarp usplash aware? so that it quits usplash and runs console-screen.sh if usplash is runing (to properly init the fonts)? 
[12:38] <Kamion> mvo: lib/menu/intro quits usplash already; it could run console-screen.sh too
[12:38] <Kamion> although maybe termwrap would've been a better place for that, but I'd rather not move it now - maybe in dapper
[12:39] <Kamion> mvo: what do I do, just '/etc/init.d/console-screen.sh'? how do I know if usplash was running, as well as installed?
[12:40] <mvo> Kamion: I'll send you a draft patch, ok? 
[12:40] <Kamion> mvo: sure, great
[12:46] <toresbe> hey guys
[12:47] <toresbe> A problem with launchpad, is this the place to raise that?
[12:47] <daniels> #launchpad, isn't it?
[12:47] <toresbe> I probably should've guessed that ;)
[12:47] <toresbe> thanks :)
[12:56] <Nafallo> Kamion: when are you going to extract the last finished translations? :-) tomorrow morning?
[12:59] <Kamion> Nafallo: no idea
[01:00] <Kamion> whenever it winds up being convenient, I guess
[01:00] <Nafallo> hehe, we want to know the explicit deadline... ;-)
[01:00] <Nafallo> if we get this night we would be happy :-P
[01:00] <Nafallo> I would work to atleast 3 UTC ;-)
[01:04] <Diziet> Oh, bugger, I'm a total idiot.
[01:05] <Diziet> I uploaded from the wrong tree and it had random #error's strewn through it.
[01:06] <Kamion> Nafallo: sure, you can have this night no problem
[01:07] <infinity> Diziet : Already been fixed.
[01:07] <Diziet> Oh, has it ?
[01:07] <infinity> Yes.
[01:07] <Mithrandir> jbailey: shouldn't 14239 be closed?
[01:08] <Diziet> That's nice.  My mirror is obviously out of date.
[01:08] <infinity> Also made an executive decision about versioning, as a guy who occasionally does security uploads, and changed the c/r to (<< ${Source-Version}).... Sure, it's not technically true, but it serves the same purpose, makes life a lot less painful, and the only drawback is that it forces the packages to be in version lockstep.
[01:08] <infinity> Which isn't really a big deal.
[01:09] <Diziet> inf: Errr.
[01:09] <Nafallo> Kamion: thanks :-)
[01:11] <fabbione> Diziet: you can keep close #16460
[01:11] <Diziet> So I'm told.
[01:11] <Diziet> Gah, apologies for incompetence.
[01:12] <Kamion> Nafallo: (it's not like extracting translations from Rosetta is a quick, painless, or automatable process)
[01:14] <infinity> Diziet : Since we have a pretty loose concept of package ownership here, it pays to subscribe to breezy-changes (or subscribe to an rss feed of it, or peruse the archives occasionally) to see if people have uploaded things you care about. :)
[01:16] <Nafallo> hihi :-)
[01:16] <Nafallo> anyone else scroll stopped working with the new xorg?
[01:17] <daniels> nothing changed there
[01:17] <daniels> so I'd be looking at the kernel
[01:17] <Nafallo> daniels: the kernel haven't changed since this morning.
[01:17] <Nafallo> baah! coward! ;-)
[01:18] <Lathiat> haha
[01:21] <Nafallo> hmm
[01:22] <Nafallo> what does Option "HorizScrollDelta" "0" do? :-P
[01:24] <tepsipakki> is it intentional that /etc/nanorc has "set mouse" that breaks middle-button paste?
[01:24] <Nafallo> tepsipakki: THANKS! :-)
[01:25] <tepsipakki> nafallo: you hit the same?-)
[01:25] <Nafallo> tepsipakki: irritated me for the last couple of months atleast ;-)
[01:26] <tepsipakki> yeah, me too, but not enough to actually do something about it.. the version in sarge apparently does not have this feature
[01:26] <tepsipakki> oh it does, but not on by default
[01:36] <Kamion> elmo: please sync nano from unstable (tepsipakki's bug above, plus line count bug that I think might be one that's been annoying me for ages on amd64)
[01:37] <Nafallo> hihi
[01:37] <Kamion> last version just missed UVF, but looks safe
[01:37] <ogra> Nafallo, uuuh... how unclean...
[01:37] <ogra> use dpkg :)
[01:38] <tseng> dpkg --purge
[01:38] <Nafallo> ogra: ehm... rm nano* after apt-get source ;-)
[01:38] <ogra> heh
[01:38] <tepsipakki> =)
[01:39] <tepsipakki> hrr, adduser fails if nscd is running.. but, maybe thats why nscd is crap and supported ;)
[01:39] <tepsipakki> UNsupported..
[01:40] <maswan> is there something like nscd that is supported?
[01:40] <maswan> (cache name/group lookups)
[01:40] <tseng> something like that will be supported in dapper for edubuntu
[01:40] <tepsipakki> heh, no
[01:40] <tseng> see ogra 
[01:41] <maswan> tseng: ah, neat. then we might not have to live with nscd for that much longer.
[01:41] <tseng> hm im thinking network auth
[01:41] <tseng> not nessecarily "group/name lookups"
[01:48] <sivang> does anybody know if we have anything that can fix zip files in main/universe?
[01:49] <infinity> 'fix'?
[01:49] <ogra> we have zipfiles in main/universe ?
[01:49] <Lathiat>  assume he means he has a corrupt .zip and wants soemthign to try recover it
[01:50] <ogra> i think there are only debs
[01:50] <maswan> tseng: well, that's the lookups I'm doing that needs nscd here
[01:50] <ogra> ah
[01:50] <ogra> :)
[01:51] <infinity> sivang : Not sure if it does the same thing as the oldskool pkzipfix (was that what you were looking for?), but 'zip' has a '-F' switch.
[01:52] <infinity> sivang : Which can be doubled-up (-FF) for really broken archives.
[01:54] <Nafallo> hmm, where is Keybuk? :-)
[01:55] <Nafallo> or is it common practice for grepmap to segfault a lot? ;-)
[01:56] <Kamion> Keybuk works notoriously strange hours; I'm sure he'll be along eventually; file a bug in the meantime
[01:57] <Nafallo> just made sure the bug was new first :-)
[01:58] <Nafallo> couldn't find anything in /var/log/syslog.*, so it has to be new :-P
[02:04] <sivang> infinity: THANK YOU
[02:12] <jbailey> Mithrandir: Yup, thanks
[02:17] <Nafallo> hmm, synaptics seems to be Keybuks fault :-P
[02:22] <mjg59> mvo: Hello?
[02:23] <segfault> morning.
[02:23] <sivang> yo mjg59 
[02:24] <zyga> hello seb128 
[02:24] <zyga> ah
[02:24] <zyga> hello segfault 
[02:24] <zyga> darn tab, why doesn't it work like in the shell?
[02:25] <seb128> zyga: xchat?
[02:25] <mvo> mjg59: thanks, alreday solved (some usplash stuff)
[02:25] <seb128> zyga: /set completion_amount 0
[02:26] <mjg59> mvo: Ah, ok
[02:27] <zyga> seb128: mmm :-)
[02:27] <zyga> thanks
[02:28] <mvo> mjg59: the recent changes make it now run on vt1? it used to be vt8? but the changelog indicates that's jbailey change
[02:29] <backports-r-us> seb128: does GAIM have completion of some sort?
[02:30] <pitti> Diziet: shall I look into the firefox backporting issues (the three things that backports-r-us mentioned that need to be changed)?
[02:30] <jdub> elmo: ping
[02:31] <backports-r-us> sweet, firefox progress :)
[02:31] <seb128> backports-r-us: what sort of completion? gaim for IRC? 
[02:31] <mjg59> mvo: You'd have to ask jbailey
[02:38] <ogra> seb128, i guess he means logout completion :p
[02:46] <HiddenWolf> pitti, ping
[02:46] <segfault> when will the next language pack be generated?
[02:48] <pitti> Hi HiddenWolf 
[02:48] <pitti> segfault: ask carlos :-)
[02:49] <carlos> pitti, I ping you this morning, did you see my private msg?
[02:49] <Diziet> pitti: Sure.  I'm afraid I don't have any useful opinions about it but do keep me posted.
[02:49] <pitti> carlos: I did, I answered you :-)
[02:49] <HiddenWolf> pitti, hi
[02:49] <pitti> Diziet: me neither, I'd just apply the changes and test them
[02:50] <carlos> pitti, are you identified?
[02:50] <carlos> pitti, I didn't get any answer...
[02:50] <pitti> carlos: yes
[02:50] <pitti> carlos: got my ping?
[02:51] <carlos> pitti, freenode says you are not identified
[02:51] <pitti> hm?
[02:51] <pitti> carlos: now I am, sorry
[02:51] <carlos> and no, I didn't get your ping
[02:51] <carlos> pitti, no problem ;-)
[02:52] <segfault> carlos: any idea?
[02:53] <carlos> segfault, we should have one at the end of this week
[02:53] <carlos> pitti, right?
[02:53] <pitti> carlos: depends on the next tarball, but I hope so :-)
[02:54] <segfault> and it'll include all the purple and blue modified strings?
[02:54] <HiddenWolf> pitti, seems that bug I reported earlier has something to do with usb legacy support.
[02:56] <seb128> carlos, pitti: how long does it take from a package upload to a rosetta pot update?
[02:56] <carlos> segfault, yes
[02:56] <carlos> seb128, usually, the worst case is 8-9 hours
[02:57] <segfault> nice
[02:57] <carlos> seb128, atm the queue is a bit busy so could take one or two days
[02:57] <seb128> carlos: I've uploaded a new control-center yesterday, can you look what happened with it?
[02:57] <carlos> seb128, sure
[02:57] <seb128> carlos: it was not shipping the pot before, this upload is supposed to fix that
[02:58] <carlos> seb128, so that explains it... cool, thanks!
[02:58] <seb128> what explains it?
[02:59] <seb128> https://launchpad.net/distros/ubuntu/breezy/+sources/control-center/+pots/control-center-2.0/ should be updated right?
[02:59] <pitti> HiddenWolf: yep, I read that; can you please followup on the bug?
[02:59] <carlos> seb128, it explains that control-center is completely outdated...
[02:59] <HiddenWolf> pitti, excuse me, hadn't checked mail
[03:00] <carlos> seb128, yeah
[03:01] <HiddenWolf> pitti, it's solved if i turn off legacy support, any idea what can cause it?
[03:01] <seb128> carlos: just let me know if it's queued or if there is an issue :)
[03:02] <pitti> HiddenWolf: I guess your BIOS tries to find an USB keyboard and somehow is buggy
[03:04] <HiddenWolf> pitti, ok
[03:04] <carlos> seb128, sure
[03:08] <carlos> seb128, it's queued to be imported
[03:08] <HiddenWolf> pitti, probably not worth fixing then?
[03:08] <seb128> carlos: thanks
[03:09] <pitti> HiddenWolf: no idea, if your BIOS hangs, then not :) if the kernel hangs, the report shuold stay open
[03:09] <HiddenWolf> hotplug hangs'
[03:12] <pitti> Hi jdthood 
[03:12] <jdthood> pitti: I just noticed that #5813 hasn't been fixed in Breezy and easily could be with a little patch.
[03:13] <pitti> jdthood: oh, that patch looks easy enough, thanks for pointing out
[03:32] <nomed> just a question ..
[03:32] <bddebian> Hello
[03:32] <nomed> why when i generate kernel-headers using make-kpkg the scripts dir is not included?
[03:33] <Diziet> Why when people report firefox bugs do they always use bugzilla or forums pages as their test cases ?  So far I haven't had one where it WFM but not for them just because the page had changed, but it's only a matter of time.
[03:55] <pitti> elmo: can you please sync alsa-driver? it's just a small build fix
[03:57] <elmo> pitti: ok to override ubuntu changes?
[03:57] <pitti> elmo: yes
[03:57] <pitti> elmo: (there are no actual changes)
[03:58] <elmo> pitti: pls say so, so I don't have to ask
[03:58] <elmo> done
[03:58] <pitti> elmo: yep, next time; thanks
[04:00] <pitti> jdthood: patch applied, thanks!
[04:00] <jdthood> pitti: Yer fast, dude.
[04:24] <bddebian> Can someone please clarify for me?  Can I add new .desktop files or not?
[04:27] <Riddell> bddebian: why would you want to?
[04:28] <Diziet> Anyone here know anything about defoma ?  16599 shows a missing font file.
[04:28] <bddebian> Riddell: To close Malone bugs
[04:30] <Riddell> bddebian: any example?
[04:30] <bddebian> Malone 2690
[04:32] <Rotund> What is the proper procedure to go about requesting a package gets updated to a newer version before Breezy is released?
[04:32] <Rotund> is it a bug report?
[04:32] <Mithrandir> Rotund: if it's in main: you're probably too late, if it's in universe, ask on #ubuntu-motu
[04:32] <bddebian> Rotund: A main or universe package?
[04:32] <Rotund> libtheora.  I'm assuming main
[04:33] <mvo> Riddell: if you haven't uploaded kdm already, please hold on a bit. I have a second version of the console-screen patch
[04:33] <bddebian> Rotund: Yep, it's main so you probably won't see it.
[04:34] <Rotund> darn.  they've fixed the memory leaks and bitrate management issues
[04:34] <Rotund> That's kind of major in my mind
[04:34] <Mithrandir> Rotund: is it a clean and nice patch?
[04:34] <Rotund> I don't know.
[04:35] <Rotund> alpha4 was 14 Dec 2005 and alpha5 was 20 Aug 2005
[04:35] <Rotund> I have to assume there were major changes
[04:36] <Rotund> Well.  they call it a "incremental update"
[04:36] <janimo> Rotund, the changes summarized on their site suggests there aren't so many
[04:37] <Rotund> The site suggests that, but 8 months between releases would make me think that's unlikely
[04:37] <Kamion> depends on how active the project is.
[04:37] <Kamion> anyway, only way to actually check is to look at the changelog or the source diff
[04:38] <Riddell> bddebian: I don't see what's wrong with adding that, having an untranslated .desktop file is better than none at all in my opinion
[04:38] <Rotund> I'd hope theora (the great hope for a good open source no patent video codec) is fairly active
[04:38] <Riddell> bddebian: but don't take my word for it
[04:39] <jdthood> I have been looking at bugs filed at bugs.freedesktop.org for the radeon driver.  I find that some of the issues reported there are also mentioned in the xserver-xorg changelog, sometimes as fixed.  Are patches being sent upstream?
[04:39] <bddebian> seb128: Do you have an opinion on that?
[04:43] <highvoltage> elmo: i want to arrange a mirror for ubuntu & edubuntu at a local isp. how much space would it take up? and which paths should ideally be mirrored?
[04:43] <highvoltage> elmo: if you're not arround, please mail me back: jonathan@tsf.org.za
[04:46] <Rotund> The changes in theora alpha5 are mostly comments
[04:47] <Rotund> 1 added feature and a little renaming
[04:54] <jdub> elmo: ping
[04:55] <bddebian> Anyone know much/anything about erlang?
[04:56] <Rotund> another question about packages.  What if there's a newer version of the package in Debian (same maintainer) than in Universe?
[04:56] <Rotund> (Ross Burton and Meld)
[04:57] <elmo> jdub: ?
[04:57] <jdub> Rotund: we stop syncing at UpstreamVersionFreeze - that's quite normal
[04:57] <jdub> hey elmo 
[04:57] <ogra> Rotund, main packages need a real good reason to et upgraded in this state of the release cycle, added features dont fall in this category...
[04:58] <ogra> Rotund, but inuverse upgrades are possible as long as it is proven that a upgrade doesnt affect any other packages or introduces new bugs
[04:58] <sivang> Rotund: mainly bugfixes
[04:59] <ogra> Rotund, so if it doesnt introduce new breakage, meld seems like a good candidate
[04:59] <Rotund> Well, I see the libtheora as bugfixes.  The meld is a much larger change (going version 1.0) but there's no interlinking
[04:59] <ogra> Rotund, for universe stuff #ubuntu-motu is the right place
[05:00] <Rotund> ogra: and what should I say?
[05:00] <ogra> Rotund, which crasher or dataloss bugs does the theors upgrade fix ? 
[05:00] <ogra> *theora
[05:00] <Rotund> bad encoding/decoding and memory leaks.
[05:00] <Kamion> ogra: don't confuse him by talking about universe - libtheora is in main
[05:01] <ogra> Kamion, he asked about meld
[05:01] <Rotund> Kamion: meld is in universe
[05:01] <Kamion> oh, but meld, ok
[05:01] <Kamion> fair enough, sorry
[05:01] <Rotund> I've found 2 packages =)
[05:02] <ogra> Rotund, go to -motu and ask for a update there, the MOTUs will care for it
[05:02] <Rotund> libtheora I expected to see a new version when it was released in August and didn't notice it hadn;t gotten upgraded until now
[05:02] <ogra> for theora ... that needs proof that it really fixes stuff...
[05:03] <Diziet> IWBNI people who wrote shell runes had some idea about how.
[05:04] <Diziet> It Would Be Nice If.
[05:05] <ogra> Diziet, btw, you would be my hero if you could fix the "the cert window pops up on the wrong (even minimized) ff win"
[05:05] <sivang> lol
[05:05] <sivang> mine too :)
[05:05] <ogra> dunno a bug # though... but its quite annoying and thim never managed to fix it
[05:05] <ogra> thom even
[05:06] <sivang> I guess we should use epiphany , as seb128 says :)
[05:06] <Diziet> I haven't looked at it at all.  It gives me the impression of being very nontrivial.
[05:06] <Rotund> Hmmm.  libtheora on Debian hasn't been updated at all (even in unstable)
[05:06] <Rotund> That's interesting as there wasn't many changes
[05:06] <Diziet> Is it upstream ?  It's the kind of thing you would think would be in upstream's bugzilla.
[05:06] <ogra> i think its a question of missing metacity/wnck integration
[05:07] <Diziet> Has _no-one_ noticed that if you have two printers of the same type, you can't print to the second one in firefox ?  (Because the second one ends up with a space in its name.)
[05:07] <ogra> i guess the lack of two printers prevents most people of notcing :)
[05:08] <Rotund> ogra: so for theora, the change list from Xiph isn't good enough.
[05:08] <Diziet> But making lots of different printers (all of which don't actually go anywhere) is the only sane way to test printing.
[05:08] <sivang> Diziet: firefox specific? not CUPS thing?
[05:08] <Diziet> sivang: It's a bug in the default print command.
[05:08] <ogra> Rotund, if its not in debian yet its unlikely anyway that we can easily upgrade...
[05:08] <Diziet> Of course ffox doesn't bother putting up a dialogue saying `your print command died', let alone capture the stderr and show it to you.
[05:08] <Rotund> I guess I don't really care as I'm thinking of grabbing a CVS version of theora anyways.
[05:09] <Diziet> So if you run ffox from the panel, you print and nothing comes out and you wonder why.
[05:09] <Rotund> thanks a lot
[05:09] <ogra> elmo, could you sync meld from sid please ? i dont think there were ubuntu changes
[05:10] <Rotund> ogra: meld is only changed in unstable... not sid
[05:10] <Diziet> cert window bug> I agree it's a bug that really ought to be fixed.  But it feels like one with a big patch.
[05:10] <Rotund> oh wait.  That is sid =)
[05:10] <ogra> err, sid != unstable O_O ? did i miss something ?
[05:11] <Rotund> sorry.  I think they'd say "okay now we bump sid to testing and make a new release name for unstable"
[05:11] <ogra> Diziet, i think its only a bt fiddling with wnck/metacity... to find out which win is requesting the popup and let it appear there
[05:11] <ogra> s/bt/bit
[05:11] <ogra> but might be time consuming...
[05:11] <ogra> probably to late for tha
[05:11] <ogra> t
[05:12] <Kamion> Rotund: no, not true; Debian releases testing as stable and then makes a new codename for testing
[05:12] <Rotund> I was thinking sarge as well. not sid.  double oopsy
[05:12] <Rotund> Kamion: yeah.  I forgot that notation.
[05:12] <Kamion> sid is (nowadays) a permanent codename for unstable
[05:12] <Rotund> anyways.  Thanks a lot and off to work
[05:13] <Diziet> ogra: You think it's a metacity bug ?  I'm not very familiar with post-1995 wm protocols :-).
[05:14] <dholbach> bbl
[05:15] <ogra> Diziet, i think its missing integration ... i bet epiphany does it right
[05:15] <Diziet> `Missing integration'?  Sounds like someone has invented a scheme that means every app has to know about every wm ...
[05:16] <Diziet> Wahey !  10293 happens to me too, all by itself.
[05:16] <Diziet> I mean, with my actual printer when I just try to print something.
[05:16] <Diziet> Strange that it only happens sometimes.
[05:18] <ogra> Diziet, metacity uses timestamps for windows since hoary, it prevents windows from stealing the focus.... i think there we would have to look...
[05:18] <ogra> (stealing the focus and popping up in fron of the other wins ...)
[05:19] <Amaranth> seb128: http://bugzilla.ubuntu.com/show_bug.cgi?id=15636 is fixed in pyxdg cvs
[05:19] <Amaranth> /whois seb128 
[05:19] <Amaranth> err
[05:19] <Amaranth> oops
[05:20] <Diziet> ogra: Oh, this is that launch timestamp thing that I remember seeing somewhere.
[05:20] <ogra> yup
[05:20] <janimo> is powermanagement-interface available on all platforms?
[05:20] <ogra> iirc the prob appeared with this change
[05:20] <Diziet> If you think you know roughly where to look etc. I'd be happy to consider a patch if it's _really small_ and I can see that it's correct.
[05:20] <pitti> elmo: please sync py2play (ubuntu override ok) and phpmyadmin from Debian (security updates)
[05:21] <ogra> Diziet, i really think its time consuming to find the right place
[05:21] <ogra> lets keep it for dapper, but look into it early
[05:21] <Diziet> Right.  More so for me - I'd have to start to understand it all first.
[05:21] <Diziet> Sure.
[05:21] <ogra> i'll happily help :)
[05:21] <pitti> elmo: oh, and gopher and imp3, too, please (all universe)
[05:21] <Diziet> ogra: Thanks, I'll probably take you up on that.
[05:21] <ogra> :)
[05:23] <elmo> pitti: done
[05:23] <pitti> elmo: thanks
[05:26] <ogra> elmo, did you get my request for meld above ? 
[05:27] <Diziet> Anyone see my request for help with defoma and fonts ?
[05:27] <Diziet> 16599
[05:29] <Diziet> ogra: If you know something about metacity, do you have an opinion about the correctness of the patch http://bugzilla.gnome.org/attachment.cgi?id=51679&action=view which is supposedly for our 15995 ?
[05:30] <ogra> Diziet, looks good to me... but a quich look from seb128 wouldnt do any harm i think ;)
[05:31] <ogra> hes far deeper in metacity...
[05:32] <Diziet> I'm not wholly convinced, in particular because it seems to count UTF-8 chars rather than bytes (which is probably what the X server is complaining about when it sends the error that makes metacity die), and there's no justification for the number 512.
[05:32] <Diziet> I mean the X server probably counts bytes.
[05:33] <Diziet> Also, I'm not sure if it's in the right place - that it catches all bugs of this form.
[05:33] <Diziet> It seems unlikely that there would be just the one, but of course ICBW.
[05:34] <Diziet> seb128: ping
[05:34] <ogra> nope, it wont catch all bugs of this form... 
[05:34] <Diziet> OTOH we might be better off fixing this one which can be exploited by web pages.  Others probably can't.
[05:34] <ogra> ask Riddell if konqueror in KDE can see the site without crash :)
[05:35] <Riddell> which site?
[05:35] <seb128> Diziet: pong
[05:35] <ogra> if he can, X shouldnt matter
[05:35] <ogra> Riddell, 15995
[05:35] <Diziet> seb: See scrool re 15995.
[05:35] <hunger> Is wiki.ubuntu.com down?
[05:35] <seb128> Diziet: it can't hurt, I would use it
[05:36] <hunger> No, works with another browser:-( Sorry for the noise.
[05:36] <Mithrandir> ogra: well, it doesn't crash with openbox.
[05:36] <Diziet> riddell: Use my test case, not the original site, and save yourself the unpleasant experience :-).
[05:36] <ogra> Mithrandir, thast what i expected
[05:36] <ogra> :)
[05:36] <Diziet> seb: Um, but I would like to fix the bug properly and so forth.
[05:36] <ogra> Diziet, see Mithrandir 
[05:36] <Diziet> If you don't know, that's fine, I'll investigate myself.
[05:36] <ogra> so X shouldnt matter here
[05:36] <ogra> its a WM thing
[05:37] <seb128> Diziet: is there any reason to put a limit to the title length?
[05:37] <Diziet> The way it fails is that metacity sends some X request that's too large and the X server says `no way' and then metacity's X error handler kills it.
[05:37] <Riddell> Diziet: no crash in konqueror
[05:37] <Mithrandir> openbox limits the title to ~4k, though
[05:37] <Diziet> Clearly a title that's too long to display is pointless and ought to be truncated.
[05:37] <Diziet> Display anywhere, I mean.
[05:38] <seb128> Diziet: the patch seems fine to me
[05:38] <Diziet> Where did this `512' come from ?
[05:38] <Riddell> Diziet: no crash in konqueror when using metacity either
[05:39] <Diziet> Riddell: konqueror probably (sensibly) truncates the title itself.  I don't think, though, that it's a _bug_ in ffox that it doesn't.
[05:40] <Diziet> Is that the only place where the title is set ?  I grepped for \btitle\b and got other stuff which I don't currently understand but seems like it might have similar problems.
[05:40] <Diziet> Are there other strings analagous to the title which metacity also mishandles ?
[05:40] <Diziet> These are the kind of questions I'm looking for answers to.
[05:41] <seb128> comment on the gnome bug
[05:41] <seb128> they will probably reply
[05:41] <Diziet> Just `the patch is fine' isn't so reassuring :-).  I can see that the patch doesn't make it worse ...
[05:42] <mjg59> Kamion: New installs mount my other Linux partitions by default, but don't mount my Windows partition  by default. Deliberate?
[05:43] <seb128> Diziet: I'm not upstream for that and doesn't know the code for it, better to comment on bugzilla.org for a GNOME guys :)
[05:43] <Diziet> seb: :-).  OK.
[05:43] <Diziet> If they're good about replies, I'll do that before I get stuck in myself.
[05:49] <Kamion> mjg59: #16233, I imagine
[05:49] <Kamion> mjg59: but send me /var/log/partman if you could
[05:49] <Kamion> or /var/log/installer/partman if you've finished the install
[05:50] <mjg59> Ok
[05:50] <mjg59> It didn't involve autopartitioning, BTW
[05:51] <Kamion> hmm
[05:51] <mjg59> I just selected some existing partitions and formatted them
[05:51] <Kamion> it's supposed to preselect your Windows partition to be mounted on /media/hda<n> or whatever
[05:53] <mjg59> Yeah
[05:55] <pitti> carlos: ok, I have created a reasonably useful blender.pot now :-)
[05:57] <carlos> pitti, wow, how?
[05:58] <bddebian> elmo: Should we morgue manderlbot since neither we nor Debian have source for libxmerl-erlang which manderlbot depends on?
[05:58] <pitti> carlos: some blender guy sent me 64 KB python code that grabs the strings and builds a pot
[05:58] <pitti> carlos: it contains some strings that shouldn't be translated, but that's good enough
[05:59] <bddebian> seb128: Did you happen to see my question about new untranslated .desktop files?
[05:59] <ogra> elmo, ta for meld...
[06:00] <carlos> pitti, ok
[06:02] <pitti> doko: ping
[06:03] <elvirolo> hi all
[06:04] <bddebian> Hello elvirolo
[06:04] <elvirolo> is the relaese of breezy gonna be delayed ? because it still has serious bugs
[06:04] <bddebian> infinity or lamont-away: Can you please clear the dep-wait for ghc-cvs?
[06:04] <elvirolo> like this one : http://bugzilla.ubuntu.com/show_bug.cgi?id=16066
[06:07] <Yagisan> elvirolo: thats a normal bug - if you want serious try to install hoary on linux software raid with /usr as 0 or 5- and see if it starts
[06:07] <elvirolo> I understand
[06:07] <Yagisan> elvirolo: I'm sure it will get fixed at some stage though
[06:07] <elvirolo> but still it definitely needs to be fixed ...
[06:08] <elvirolo> but breezy will be released soonush
[06:08] <elvirolo> soonish*
[06:08] <doko> pitti: pong. while you are here, mind checking OOo1 on you iBook? *duck*
[06:09] <pitti> doko: is the "sys" module really included in python-minimal? dpkg -L does not reveal it
[06:09] <pitti> doko: why ooo1?
[06:10] <doko> pitti: it's a builtin module
[06:10] <pitti> doko: ah, thanks
[06:10] <lamont-away> bddebian: doing it now, with a vengence
[06:10] <torkel> elvirolo: follow the hint in comment #3 (setting the loglevel to debug) and attatch a log to the report and someone will probably take a look at it
[06:11] <doko> pitti: OOo1 does have more translations than OOo2, does even run on machines like iBooks ...
[06:11] <elvirolo> torkel: thank you, i shall
[06:11] <bddebian> lamont-away: With a vengence? :-)
[06:11] <bob2> what is the selling point of 2.0 over 1.0?
[06:12] <tseng> bob2: i think the gnome integration is actually part of 2.0
[06:12] <tseng> bob2: and not a massive fork/patch
[06:12] <elvirolo> torkel: in fact, comment #3 was mine :-D
[06:12] <Mithrandir> bob2: slower, fatter and makes even more people cry.
[06:13] <doko> bob2: writing OpenDocument format, better MS compatibility (although there are some regressions)
[06:13] <bob2> haha
[06:13] <bob2> tseng: I always thought OO 1 didn't link against enough libs ;)
[06:14] <lamont-away> bddebian: I um, cleared 'em all
[06:14] <bddebian> lamont-away: Ahh :-)
[06:14] <tseng> bob2: http://www.equalmusic.com/photos/coheed3.jpg in other news, you could double for the singer of coheed and cambria, i think
[06:17] <bddebian> lamont-away: How far behind is: http://people.ubuntu.com/~lamont/buildLogs/Lists/breezy.all.i386 ?
[06:18] <lamont-away> That file gets created and pushed to that URL every 20 minutes
[06:25] <seb128> bddebian: nop, say it again ?
[06:30] <janimo> mjg59, is pmi susp/hib supposed to work in console mode too?
[06:30] <bddebian> lamont-away: OK, thx
[06:30] <bddebian> seb128: Should I not add new untranslated .desktop files either?
[06:31] <pitti> doko: FWIW, neither OOO1 nor 2 are particularly nice on a 256 MB iBook, they are both slow as hell
[06:31] <pitti> doko: ok, where are the debs?
[06:31] <seb128> bddebian: I've no opinion on that, that doesn't make a regression ... but I'm not sure if we prefer to lack some entries or to have some english ones 
[06:31] <seb128> maybe you could mail the mailing list about that?
[06:32] <janimo> ogra ping
[06:32] <ogra> janimo, pong
[06:33] <janimo> ogra mailing list admin questions again got 2 minutes?
[06:33] <ogra> sure
[06:33] <lamont-away> pitti: you should try them on a 128MB system.. :-)
[06:33] <lamont-away> my wife needs more memory...
[06:33] <pitti> lamont-away: no thanks, /me uses LaTeX and gnumeric :-) (my gf, too)
[06:33] <PzyCrow> Since upgrading to Breezy each try to start X with the nvidia driver results in a system lockup. Is this a known problem? I couldn't find anything in bugzilla.
[06:36] <pitti> elmo: loop-aes-utils sync, please
[06:36] <bddebian> seb128: OK
[06:37] <doko> pitti: in the archive (1.1.5)
[06:37] <pitti> doko: ah, that one; ok
[06:37] <pitti> doko: is it utterly urgent? or is tomorrow ok?
[06:37] <doko> pitti: end of week/weekend would be nice
[06:37] <pitti> doko: yep, ok
[06:40] <pitti> BenC, mvo: btw, has there been any solution to the kernel upgrade notification?
[06:44] <mvo> pitti: it should be a matter of adding the needed key to the notification file. but I don't want to upload a kernel with just that change in :)
[06:44] <mjg59> janimo: If you're lucky
[06:44] <pitti> mvo: so there is such a key now?
[06:44] <pitti> mvo: jbailey wants to add a reboot notice to glibc; I guess it is high time to factorize those
[06:45] <jdub> elmo: ping
[06:46] <mvo> pitti: there has been the idea around to have a "reboot" notification that can just be touched if a pkg needs rebooting
[06:46] <mvo> pitti: I find this idea nice and appealing
[06:46] <janimo> mjg59, also why is vbestate not saved on prepare (commented in the script) on resume it prints out errors
[06:47] <mjg59> janimo: Because doing that breaks things
[06:47] <mjg59> It's saved on boot isntead
[06:47] <pitti> mvo: yes, but we need to make it a bit more flexible, maybe the user is interested in the reason
[06:47] <pitti> mvo: otherwise just touching user.d/reboot is easy
[06:47] <janimo> mjg59, which script saves it?
[06:48] <janimo> also is pmi platform indep, can I depend on it in a package?
[06:48] <janimo> I am rying to get susp/hib into xfce-session
[06:48] <mjg59> janimo: /etc/init.d/vbesave
[06:48] <jbailey> pitti: Ah, you're already talking about it. =)
[06:48] <mjg59> Yes, pmi should be platform independent
[06:48] <infinity> pitti : The kind of people those notes are targetted at don't care about reasons, they care about "some updates you installed require a reboot, please do so"
[06:49] <pitti> infinity: well, true
[06:49] <infinity> pitti : People who want reasons have changelogs, can watch the packages installing, etc, etc.
[06:49] <janimo> is pmi an ubuntu only thing for now? Can I send patched sources calling pmi upstream?
[06:49] <infinity> pitti : We're not really targetting people who should have to know or care what glibc is in the first place.
[06:49] <pitti> mvo: ok, so what about having u-d just ship a well translated generic reboot notification?
[06:49] <jbailey> mvo: My thought was to have a notification in /usr/share/doc of some sort.
[06:49] <jbailey> Nice, generic and translated.
[06:49] <pitti> mvo: then hal, dbus, kernel, etc. just create a symlink to it
[06:49] <jbailey> Then just ln -sf it when you need it
[06:50] <jbailey> On reboot rm the created symlink.
[06:50] <infinity> Looks like many people have all reached the same page at the same time.
[06:50] <infinity> Always good. :)
[06:50] <jbailey> infinity: Watch it page fault now.  WE'LL ALL BE SWAPPED OUT!  OH NO!!
[06:50] <jbailey> err.
[06:50] <jbailey> dont mind me. =)
[06:50] <elmo> jdub: ?
[06:51] <elmo> pitti: done
[06:51] <pitti> thanks
[06:51] <infinity> jbailey : Nerd jokes don't suit you.
[06:51] <bddebian> heh
[06:51] <jbailey> infinity: What would you rather I tell, dear?
[06:52] <infinity> jbailey : Try something vaguely sexually suggestive and oddly creepy instead.
[06:52] <jbailey> Oh, you mean go back to my usual.
[06:52] <bddebian> hahaha
[06:52] <jbailey> infinity: It's a bit harder without jblack sitting beside me, but I can try. =)
[06:52] <mvo> pitti: yes, that sounds good to me
[06:53] <pitti> mvo: it would really help to fix this nasty "4 notifications in a row" mess, so I hope mdz ack's this for Breezy :-)
[06:53] <janimo> mjg59, ah I did not have vbestate becauseI forgot to uncommend ACPI_SLEEP.Will uncommented remain the default for breezy?
[06:53] <jbailey> pitti: My favourite is getting the notices after I reboot. =)
[06:53] <infinity> pitti, mvo : Well, the fact that it's "very translateable" will score brownie points for approval, I'm sure.  And if you get the string written and frozen in the next day or two, we may even have 15 or 20 languages by release. :)
[06:54] <mvo> pitti: I'll attack it right after coming back from playing hockey
[06:54] <pitti> mvo: I love you :-)
[06:54] <bddebian> hmm
[06:54] <pitti> infinity: and it's not even intrusive
[06:55] <lamont-away> elmo: please take a gander at anastasia w/hppa and see how it looks...
[06:55] <lamont-away> getting at least palo and libgcc2 into main would mean we could build CD's and actually test things and such
[06:56] <pitti> infinity: sleep well
[06:59] <LaschW> Kamion: Is there any CVS access to have a closer look at OEM mode install related programms?`
[07:02] <mdz> pitti: ?
[07:02] <pitti> Morning mdz
[07:02] <bddebian> Hello mdz
[07:03] <pitti> mdz: the idea was to unify the current plethora of "please reboot" update notifications of hal, dbus, kernel, and glibc to one common and well translated generic reboot notification
[07:04] <pitti> mdz: u-n would ship this in /usr/share somewhere, packages could just drop a symlink into user.d/ and u-n's init script coudl properly clean up the symlink
[07:04] <pitti> mdz: so user see one nice message instead of 4 untranslated one
[07:04] <mdz> pitti: that would definitely be an improvement
[07:04] <pitti> mdz: what do you think?
[07:04] <mdz> pitti: but what changes are involved?
[07:04] <pitti> mdz: u-n needs to ship that file, and we need to update hal, dbus, kernel, and glibc to set a symlink instead of creating their own notifications
[07:05] <mdz> I think it should be abstracted in u-n
[07:05] <pitti> mdz: (well, glibc does not yet do it, jbailey is just about to add the notification, that's why we discussed about it in the first place)
[07:05] <pitti> mdz: a small u-n script that sets the symlink is even nicer, right
[07:06] <pitti> mdz: maybe /usr/share/u-n/flag-reboot, or whatever
[07:06] <mdz> so packages would do if ...exists...; then notify-reboot-required; fi
[07:06] <mdz> right
[07:06] <pitti> sounds good
[07:06] <mdz> rather than worrying about symlink issues
[07:06] <pitti> mdz: I think that is unintrusive enough to be breezyable, and it would improve user experience a lot
[07:06] <mdz> that's fine with me if it goes in this week
[07:06] <mdz> next week we start rolling RC candidates
[07:06] <pitti> mdz:<mvo> pitti: I'll attack it right after coming back from playing hockey
[07:07] <pitti> mdz: changing hal and dbus is a piece of cake, kernel probably too
[07:08] <pitti> mdz: technically, this would be subject to NonLanguagePackTranslationDeadline, though :-)
[07:10] <mdz> pitti: well, if we create these common templates in u-n, we can include them in u-n's po file
[07:10] <mdz> pitti: and export to rosetta and langpacks
[07:10] <mdz> which would be great
[07:11] <pitti> right
[07:11] <mdz> Diziet: any resolution on the firefox font situation?
[07:11] <pitti> we can't just translate them automatically since the files don't use gettext
[07:15] <ogra> grrrr 5th evo crash today
[07:15] <Nafallo> ogra: sounds fun :-)
[07:15] <pitti> ogra: "mutt sucks less" (SCNR)
[07:15] <pitti> ogra: seriously, any useful backtrace?
[07:16] <ogra> pitti, ENOTIME
[07:16] <Mithrandir> pitti: mutt sucks differently, you mean?
[07:16] <pitti> ogra: my last evo crash produced a nice bt, which was fixed quickly
[07:16] <pitti> Mithrandir: definitively much less for me at least
[07:16] <ogra> pitti, it just locks up... after about 3-4h usage
[07:16] <pitti> Mithrandir: although I agree that it is a matter of taste
[07:16] <ogra> no crash or backtrace
[07:16] <pitti> ogra: strace?
[07:16] <jbailey> strace on evo for two hours would be too painful to use.
[07:17] <pitti> jbailey: why for two hours?
[07:17] <pitti> jbailey: just attach strace to it when it hangs
[07:17] <ogra> pitti, i'm running 2 pbuilder simultaneously compiling xscreensaver, i have no ressources for 4h evo stracing
[07:17] <pitti> ogra: ^
[07:17] <pitti> ogra: no need to
[07:17] <jbailey> pitti: Yeaeh, true.
[07:17] <pitti> ogra: also, if evo hangs, just gdb-attach on it and bt it
[07:17] <pitti> ogra: if you have -dbg installed, you should get a good result
[07:18] <Mithrandir> ogra: you should have more than one computer. :-)
[07:18] <ogra> pitti, will do, next time it crashes
[07:18] <pitti> ogra: "gdb /usr/bin/evolution $(pidof evolution)" -> you get the idea
[07:18] <ogra> sure
[07:18] <pitti> ogra: and "strace -p $(pidof evolution)"
[07:19] <ogra> i wonder if its mdz's fault, it happened 3 out of 7 times when i answered mails from him :P
[07:19] <pitti> ogra: shhh - that's still s3kr3t
[07:19] <Mithrandir> too much hot mdz&ogra action?
[07:20] <Nafallo> lol
[07:20] <pitti> ogra: check the source for strcmp(s, "mdz") || sleep(10000000)
[07:20] <Diziet> mdz: Hi.  Fonts, yes:
[07:20] <Nafallo> lol
[07:20] <ogra> *g*
[07:20] <Diziet> The discussion on the mailing list seems rather inconclusive.
[07:21] <Diziet> I can change the default from serif to sans (indeed I've just uploaded that).
[07:21] <ogra> Diziet, i find the debian bug very clear
[07:22] <Nafallo> elmo: did you miss Kamions syncrequest for nano earlier? :-)
[07:23] <doko> elmo: please install lib64z1-dev on davis/breezy, the concordia stuff can wait, but this one would be very nice to have
[07:25] <Diziet> ogra: You mean you agree with Keith Packard's message of the 3rd of April ?
[07:25] <Diziet> It certainly seems unclear to me at the moment that the right fix is some change in firefox to request a different font.
[07:25] <ogra> Diziet, i agree with robot101's fix of th wrong mapping for helvetica, tahoma, verdana
[07:26] <ogra> #its fonconfigs mapping of the MS fonts
[07:26] <Diziet> Hello.
[07:26] <elmo> doko: done
[07:27] <elmo> Nafallo: yes
[07:27] <Diziet> Um, where is that fix ?
[07:27] <Diziet> Are you referring to those two URL's in Romain Francoise's message in the Debian BTS ?
[07:27] <ogra> yup, to Robot101's blg
[07:28] <ogra> blog
[07:28] <Diziet> 403
[07:28] <doko> elmo: thanks
[07:28] <ogra> there is a new url attached to the mail
[07:29] <ogra> sorry, me evo just freaked out..
[07:29] <Diziet> What mail ?
[07:29] <Nafallo> 13:36:44 Kamion  elmo: please sync nano from unstable (tepsipakki's bug above, plus line count bug that I think might be one that's been annoying me for ages on amd64)
[07:30] <elmo> Kamion: done
[07:30] <ogra> Diziet, http://www.robot101.net/2005/03/16/fontconfig-fun/
[07:31] <ogra> from "Ming Hua" 
[07:31] <Diziet> Oh, wow, actual running prose :-).
[07:33] <Diziet> Right, that's a nice summary of the problem.
[07:33] <Treenaks> You don't have permission to access /files/fonts.conf on this server.
[07:33] <Diziet> _However_ it doesn't answer my question about metric equivalence.
[07:33] <Treenaks> uh
[07:33] <Treenaks> Robot101: 
[07:33] <Robot101> oh, sorry
[07:34] <Robot101> hmm
[07:34] <Robot101> thats interesting
[07:34] <ogra> Diziet, dont touch the sizes... some webdesigners rely on that ... you'll produce a lot of anger among them :)
[07:34] <Robot101> er, it works for me
[07:34] <Treenaks> ogra: those web designers should be taken out, shot, shot again, cut to pieces, and dumped in a fast-flowing river
[07:34] <Diziet> ogra: We had that conversation on ubuntu-devel.  See 
[07:35] <ogra> Treenaks, its the majority
 from James Livingston <jrl@ids.org.au>
[07:35] <Treenaks> ogra: time to organise a party then ;)
[07:35] <ogra> Diziet, see mpt's comment there
[07:36] <Diziet> mpt> Don't have that one yet I think.
[07:36] <ogra> its from tuesday (18:39 german time)
[07:36] <Diziet> In any case, let us suppose for the moment that we accept the idea that the metrics _must_ be right even for web fonts.
[07:37] <Diziet> What's that in UTC ?  Message-ID ?
[07:37] <slomo> Diziet: german is UTC+0200
[07:37] <ogra> Message-Id: <a375865ad4f49036d755b25015008af7@canonical.com>
[07:38] <Diziet> ogra: Um, eh ?
[07:38] <ogra> ?
[07:38] <Diziet> Anyway, back to what I was saying:
[07:39] <Diziet> So if you think that the metrics have to be right, then surely you must conclude that we have to leave the situation as it is.  Since in hoary the metrics were wrong - we used a font with the wrong metrics - and in breezy the situation is more correct.
[07:39] <Diziet> Whereas in fact everyone's complaint seems to be the reverse.
[07:39] <ogra> Diziet, in any case you should discuss it with mtp, why else has canonical a UI specialist who cares exactly for that
[07:39] <ogra> *mpt
[07:42] <mdz> Diziet: my understanding from the discussion is that the defalut fonts, as in hoary, are Vera, but some users are getting a different font from fontconfig for some reason
[07:42] <mdz> Diziet: that's the bit we need to address
[07:42] <Diziet> Yes.
[07:42] <ogra> yup
[07:42] <Diziet> It's a shame no-one is answering my point about metrics.
[07:42] <Amaranth> mdz: fontconfig 2.3 change from what has been said
[07:43] <Diziet> We can't just change fontconfig to hand out fonts with the wrong metrics more often, because that will break typesetting applications.
[07:43] <Amaranth> mdz: the metrics are now correct and good for printing but look like ass on the screen for a website
[07:43] <mdz> Amaranth: I'm talking about the fact that some users are getting an entirely different font, as returned by fc-match
[07:43] <ogra> Diziet, we shouldnt step away to far from upstream with the metrics...
[07:44] <Amaranth> mdz: you mean the ones getting nimbus?
[07:44] <mdz> Diziet: if nothing else we should revert to the Hoary behaviour; the change of fonts in firefox is a regression
[07:44] <mdz> Amaranth: yes
[07:44] <ogra> Diziet, if i design a page for mozilla, i expect it to look the same in all mozillas (at least on the same OS)
[07:44] <Diziet> mdz: AIUI it's not just a regression - it fixes some broken cases for printing.
[07:44] <janimo> elmo, please sync xfce4-notes-plugin, discard our changes, thanks
[07:44] <mdz> Diziet: the screen is more important
[07:44] <elmo> janimo: please read my email
[07:44] <mdz> firefox is a front line app
[07:45] <mdz> does anyone remember that bug about unicode_start only being run for the first console?
[07:45] <janimo> elmo, did not get it was wondering why you synced the tow others
[07:45] <ogra> Diziet, i would follow the blog entry but with the djvu fonts instead of the standard vera fonts, looks like the conclusion for me if i take all answers
[07:45] <mdz> it's been duped as 16607 but I can't find the original bug
[07:45] <Diziet> I'd like to try to find a proper fix before we jump back to hoary.  But if that bug is better than this bug sure.
[07:46] <ogra> Diziet, we dont jump back to hoary
[07:46] <elmo> janimo: sigh.  it's in my email.  xfce4-notes-plugin is newer in breezy than in sid.  I can't sync it.
[07:46] <ogra> Diziet, there were no djvu fonts in hoary...
[07:46] <pitti> mdz: that bug is still present, btw
[07:46] <janimo> epoch?ok thanks
[07:46] <mdz> pitti: what' sthe bug#?
[07:47] <pitti> no idea, /me looks
[07:47] <elmo> janimo: no, don't epoch it
[07:48] <Diziet> We could jump back to hoary by nixing fontconfig to hand out `wrong' but better-looking fonts.
[07:49] <janimo> elmo, of course I am not, I was wonderind if we already had epoch. but I'll look. thanks
[07:50] <ogra> GRRRR
[07:50] <Amaranth> Diziet: or change the firefox preferences to use certain fonts instead of 'serif' and 'sans-serif'
[07:51] <ogra> why is xscrensavers status mapping scattered all over the code... 
[07:51] <ogra> Amaranth, thats not good as default...
[07:51] <pitti> mdz: can't find it either, maybe it was never filed...
[07:51] <Amaranth> ogra: no, it but it does fix both problems, just in a hacky way
[07:52] <Diziet> amaranth: You think that's right ?  I'm going to suggest it in the discussion.
[07:52] <ogra> Amaranth, thats merely evil
[07:52] <Diziet> Will it help when the website asks for a particular font ?  Because that seems to be the case in question.
[07:52] <ogra> nope
[07:53] <ogra> the fontconfigmapping will still bring you Nimbus
[07:53] <Amaranth> Diziet: I don't think it's right, I think it is just less wrong than breaking either printing or screen reading in fontconfig
[07:53] <Amaranth> ogra: if you specify a windows font that you don't have installed, yeah
[07:53] <Amaranth> which most websites do....even mine
[07:53] <ogra> Amaranth, we have no windows fontrs in ubuntu
[07:53] <Amaranth> i know
[07:54] <Diziet> AIUI, the problem is when a site asks for a Windows font.
[07:54] <Diziet> fontconfig recognises the font name and realises that only Nimbus fits the metrics.
[07:54] <Amaranth> i was just showing my change of heart on irc so it wouldn't sound like i'm saying two things at once :)
[07:54] <Diziet> So it uses Nimbus.
[07:54] <ogra> so this must get fixed... its an odd experience if half the sites look blurry by default, even if they print fine
[07:54] <Diziet> Am I right ?
[07:54] <Amaranth> Diziet: yep
[07:54] <ogra> Diziet, not really...
[07:54] <Diziet> Where am I wrong ?
[07:55] <Amaranth> The only real solution I can think of is to make fontconfig understand the difference between screen and print.
[07:55] <ogra> Diziet, it recognizes that keithp selected Nimbus as the right font...
[07:55] <Diziet> keithp ?
[07:55] <Amaranth> but that's why he selected it
[07:55] <ogra> fontconfig doesnt compare metrics, it follows the local..conf file
[07:55] <ogra> Diziet, upstream
[07:56] <Amaranth> keithp == keith packard, super X hacker ;)
[07:56] <Diziet> upstream> Oh, right.
[07:56] <Diziet> Yes, yes, but the _reason_ that font is listed there is because it has the right metrics.
[07:57] <ogra> thats the real bug imho... it *should* cmpare metrics instead of pulling them from a font file and it should know if its for print or screen... but it doesnt now...
[07:57] <Amaranth> wasn't him getting kicked out of the xfree86 team the final straw that made people fork? (offtopic)
[07:57] <Diziet> What ?  Why would it know print or screen ?  It just gets the requested name, right ?
[07:57] <pitti> Lathiat: indeed, I'm on a security rampage today :-)
[07:57] <ogra> Diziet, only for priting which is not as important in a browser as screen
[07:58] <Amaranth> ogra: Yeah, like I said the real solution is to make fontconfig understand the difference between print and screen. That's not happening before breezy releases though.
[07:58] <ogra> Diziet, yes, but it *should* know what for the font is needed...
[07:58] <Diziet> ogra: *should*> Right.
[07:58] <ogra> as long as we have different fonts for print/screen
[07:58] <Diziet> Can we just stick to the facts rather than the shoulds for a moment ?
[07:58] <ogra> as long as it doesnt, concentrate in screen, not on print
[07:59] <Amaranth> Diziet: The facts are we either break printing or break screen reading in ways that will have people bitching.
[07:59] <Diziet> Do we in fact have different fonts ?  I thought the same font files were used for both ?
[07:59] <Diziet> ama: That's not a fact, that's a conclusion.
[07:59] <ogra> type1 vs ttf :)
[07:59] <Diziet> Stop trying to pursuade me and inform me, dammit !
[07:59] <ogra> type1 -> wondeful for printing, they look far better than ttf...
[07:59] <dilinger> wow
[08:00] <ogra> ttf -> best for screen ... 
[08:00] <dilinger> apt just shit itself quite spectacularly
[08:00] <Diziet> So when ffox invokes fontconfig, does fontconfig know it's for the screen ?
[08:00] <ogra> nope
[08:00] <Amaranth> No.
[08:00] <Diziet> Note of course that ffox wants fonts for printing too and then it still wants `nice' rather than `right metrics', unlike most other cases of printing.
[08:01] <ogra> thats place for future improvement 
[08:01] <Diziet> So if you're sayign that type1 is for printing and ttf is for screen, how does fontconfig know which to pick ?
[08:01] <ogra> the main purpose for a browser is screen reading...
[08:01] <Diziet> I mean, how do you get it to pick a type1 for printing ?
[08:01] <ogra> it doesnt, thats the point
[08:01] <Diziet> I'm going to ignore all comments of the form `it is more important that'.
[08:01] <jdub> um
[08:02] <ogra> and currently it picks a type1 for screen
[08:02] <jdub> there's really no qualitative difference between type1 and ttf for printing
[08:02] <Diziet> So is it really the case that you should never pick a type1 for the screen ?
[08:02] <jdub> no
[08:02] <ogra> jdub, scale a ttf to 400pt ;)
[08:02] <ogra> and then print it :)
[08:02] <jdub> ogra: looks fine, no technical reason why it shouldn't
[08:03] <Diziet> jdub: Are you disagreeing with:
 type1 -> wondeful for printing, they look far better than ttf...
[08:03] <jdub> Diziet: it is less optimal to choose type1 for screen because they don't provide as much hinting as ttfs
 ttf -> best for screen ... 
[08:03] <Diziet> jdub: Ah.
[08:03] <Diziet> So you would normally prefer a ttf ?
[08:03] <jdub> this is not wildly relevant to choosing a font for firefox
[08:03] <jdub> of course
[08:03] <jdub> one of the bitstream vera family
[08:03] <ogra> Diziet, so wipe my first sentence then ... i remember times where you shouldnt have used ttf for print
[08:04] <Diziet> Couldn't we have ffox invoke fc with a special flag, eg  --use-font-maps-from=/usr/share/other/directory/with/different/mappings   ?
[08:05] <ogra> Diziet, yeah, in dapper :)
[08:05] <Diziet> Well, is it that hard ?
[08:05] <jdub> Diziet: erk, why?
[08:05] <Diziet> So that ffox can get different answers to gs.
[08:05] <Mithrandir> ogra: ttf has always been ok to print, at least since it was introduced in Mac OS 7 in 1991.
[08:05] <Diziet> Or whatever else uses fc.
[08:05] <ogra> jdub, Diziet wnats perfect metrics for printing
[08:06] <jdub> diziet can't always get what he wants :-)
[08:06] <Diziet> If you use a font with significantly different metrics with a page description language you end up with seriously damaged output.
[08:06] <Diziet> The question is, can I this time ?
[08:06] <ogra> Mithrandir, so bring a ps document with embedded ttf fonts to a print shop... 
[08:06] <jdub> right, so we should not configure fontconfig to match fonts with wildly different metrics
[08:07] <Mithrandir> ogra: print shops usually prefer PDFs, not PS nowadays. :-)
[08:07] <jdub> and, really, you should be choosing specific fonts when you require this anyway
[08:07] <Diziet> Was the previous situation with substituting Vera for the M$ fonts OK ?
[08:07] <ogra> Mithrandir, i'm out of print business since nearly 10 years *shrug*
[08:07] <ogra> Diziet, yes
[08:07] <Diziet> jdub: Yes, but we're about to configure it so that even when the document requests specific fonts we give it different ones, aren't we ?
[08:08] <jdub> Diziet: why would we do that?
[08:08] <ogra> imho you only need 2 fonts in a browser, serif and sans...
[08:08] <Diziet> The problem case is web pages which specify a specific font.
[08:08] <ogra> err 3 ... monospace indeed
[08:09] <Diziet> You'd prefer to nobble it so that ffox only used the defaults ?
[08:09] <jdub> Diziet: why is that a problem case? they get what they ask for
[08:09] <Mithrandir> mdz: re the gnome slowdown; I'm unable to reproduce that on any machine I have available.  The only one which matches the other machines somewhat dies in the middle of the bonnie run, but I was too tired&bored at the end of the day to debug why.
[08:09] <Diziet> Web designers are idiots.
[08:09] <Diziet> Many sites request specific M$ fonts because that's what was on the idiot's system.
[08:09] <Diziet> And they look crap as a result.
[08:09] <jdub> and we have fonts that are like those
[08:10] <Diziet> (This is not my experience, only my understanding.)#
[08:10] <Diziet> We have (a) fonts that would look good if we substituted them and (b) fonts that have the same metrics but not (a)&(b) together.
[08:10] <Diziet> Now in a web browser we want (a) and for printing we want (b).
[08:10] <jdub> ij
[08:10] <jdub> ok
[08:10] <jdub> this is solving a non-problem
[08:11] <jdub> let's avoid the damage by ignoring whoever suggested this
[08:11] <Diziet> People seem to think it's important that web pages look pretty in our browser.
[08:11] <jdub> they do
[08:11] <jdub> if they think web pages rely on precise font metrics, they're wrong
[08:11] <jdub> so
[08:11] <Diziet> Not these ones with the broken font specifications, which are (apparently) fairly common.
[08:11] <Diziet> Yes, yes, the web pages _don't_ rely on precise metrics.
[08:11] <Diziet> They just specify the M$ font.
[08:12] <jdub> and that's fine
[08:12] <Diziet> If you use another font they render fine.
[08:12] <jdub> the browser says "oh, don't have that font, we'll fall back to serif or sans"
[08:12] <jdub> or, more likely, the page specifies a fall back
[08:12] <Diziet> But, the font that breezy's fontconfig picks when you ask for the M$ one, looks crap.
[08:12] <jdub> either way
[08:12] <jdub> it doesn't matter
[08:12] <jdub> if fontconfig's matching list includes crap looking fonts, remove the matchings that refer to crap looking fonts
[08:13] <jdub> better still, remove the fonts
[08:13] <Diziet> Those fonts look OK when printed.
[08:13] <Diziet> And when a document to be printed asks for the M$ font, we have to give them that font because otherwise the metrics will be wrong and it can look very bad (text overwriting itself, etc.)
[08:14] <Amaranth> Diziet: I'm one of those 'idiots' who ask for MS fonts on his website.
[08:14] <Diziet> (By document to be printed I mean something in a page description language as opposed to a reformattable document.)
[08:14] <jdub> specifics would be nice
[08:14] <Amaranth> Diziet: Because falling back to sans-serif has always come out ok.
[08:14] <Diziet> ama: But you specify an alternative, right ?
[08:14] <Diziet> Does your site look bad in breezy's ffox ?
[08:14] <jdub> if 'verdana' matches to 'bitstream vera serif' then yes, that's silly
[08:14] <Amaranth> font-family: Trebuchet MS, Arial, Helvetica, sans-serif;
[08:14] <Diziet> And, URL please :-).  A nice example would help.
[08:15] <Amaranth> Diziet: I dunno, bad superblock on my breezy HD.
[08:15] <jdub> Diziet: dude, specifics
[08:15] <Diziet> OK, tell me the URL, I'll look for myself.
[08:15] <Amaranth> damn netware messages keep popping up and breaking my concentration
[08:15] <Diziet> And *sympathy*
[08:15] <Amaranth> Diziet: http://www.realistanew.com
[08:15] <Diziet> Netware ??
[08:15] <Amaranth> i'm at school
[08:16] <ivoks> Amaranth: ugly fonts :/
[08:17] <Amaranth> ivoks: I blame fontconfig. ;)
[08:17] <ivoks> Amaranth: ah...
[08:18] <ivoks> that's trebuchet
[08:18] <Diziet> OK, so that comes out in Nimbus.
[08:18] <Diziet> Which is what we don't like.
[08:18] <Diziet> And this is not Amaranth's fault, it's our system's.
[08:18] <ivoks> Diziet: true
[08:18] <Amaranth> it should be falling back to bitstream vera sans or whatever :)
[08:18] <ivoks> Amaranth: anyway, trebuchet is ugly :)
[08:18] <Amaranth> it looks great here on windows xp
[08:19] <ogra> Diziet, i'm not sure... it might be Amaranth's fault anyway ;) 
[08:19] <Amaranth> trebuchet is a little off, but i wanted to try something different
[08:19] <Diziet> If I go into the firefox preferences and ask for Bitstream Vera for serif, I see Bitstream Vera.
[08:19] <Diziet> Which I hadn't expected.
[08:19] <Diziet> So obviously I still don't understand what's going on.
[08:19] <Diziet> Ticking `always use my fonts' works too :-).
[08:20] <mdz> Mithrandir: are there any patterns in the hardware configurations of those experiencing it?
[08:20] <Amaranth> Why wouldn't you expect it to give you bitstream vera if you ask for it?
[08:20] <Amaranth> it only falls back on matching and coming up with nimbus if the requested font isn't available
[08:20] <Mithrandir> mdz: a lot of those people seem to have ICH6 IDE bridges.
[08:21] <Mithrandir> mdz: I might have access to such a system, but as I said, it was really unstable today.
[08:21] <mdz> Mithrandir: are they the same chipsets which have this "led always on" bug?
[08:21] <Diziet> amaranth: I guess I don't know how the firefox font preference works.
[08:22] <Amaranth> Diziet: it just asks fontconfig for a font
[08:22] <Diziet> I had imagined that ffox would take the list from the website's css and replace sans-serif with the string you specify in the preferences.
[08:22] <Amaranth> if the font exists it gets exactly what it asked for
[08:22] <Diziet> And would then pass that list to fontconfig.
[08:22] <Diziet> So given that I don't have `always use my fonts' I would have expected my setting to come last.
[08:23] <Amaranth> hmm
[08:23] <Amaranth> perhaps it knows to use your font if it doesn't get exactly what it asked for with the other options
[08:23] <Diziet> So given it finds Nimbus with the sans preference set to sans, why wouldn't it find it with the sans preference set to Bitstream Vera ?
[08:24] <Mithrandir> mdz: it seems to all be intel-based systems at least, but I can't reproduce on my only stable intel system (my x40)
[08:24] <Diziet> IBWNI someone here had some idea what was actually going on.  I certainly don't.
[08:24] <jdub> because you're then saying sans-serif == bitstream vera sans, not "go match!"
[08:24] <Diziet> `no match', you mean.
[08:24] <Diziet> But if you say fc-match sans-serif you get a match, don't you ?
[08:24] <jdub> no, i mean "so go and make a match!"
[08:25] <jdub> only because it's an alias
[08:25] <ivoks> khm...
[08:25] <mdz> Diziet: what we want is for firefox to use bitstream vera, as it did for hoary
[08:25] <ivoks> maybe i'm wrong
[08:25] <ogra> Diziet, fc-match Helvetica
[08:25] <ivoks> but doesn't it say Helvetica first?
[08:25] <ogra> Diziet, fc-match Verdana
[08:25] <ivoks> and... fonts.conf says...
[08:25] <Diziet> root@samual4:~ # fc-match Helvetica 'Bitstream Vera Sans'
[08:25] <Diziet> n019003l.pfb: "Nimbus Sans L" "Regular"
[08:26] <ivoks> alias helvetica Nimbus
[08:26] <ivoks> line 150
[08:26] <ivoks> fonts.conf
[08:26] <Amaranth> delete that ;)
[08:26] <ogra> Diziet, Bitstream is ignored in your line above
[08:26] <Diziet> Because it's an absolute match.
[08:26] <Diziet> I mean, Helvetica is.
[08:27] <Diziet> So what I don't understand is why I get Bitstream Vera by changing the ffox preferences.
[08:27] <Diziet> On Amaranth's site.
[08:27] <Amaranth> *boggle*
[08:27] <Diziet> If that works everywhere perhaps we should just change the ffox prefs.
[08:28] <Amaranth> and the gnome prefs, and the kde prefs, etc
[08:28] <jdub> what's your match for 'trebuchet ms'
[08:28] <jdub> ?
[08:28] <jdub> (mine is trebuchet, having the ms fonts installed already)
[08:29] <Diziet> root@samual4:~ # fc-match 'trebuchet ms'
[08:29] <Diziet> Vera.ttf: "Bitstream Vera Sans" "Roman"
[08:29] <Amaranth> so you should have been getting bitstream vera sans anyway
[08:29] <ivoks> argh..
[08:29] <ivoks> ok
[08:29] <ivoks> i have a fix
[08:29] <ivoks> fc-match Helvetica
[08:29] <ivoks> Vera.ttf: "Bitstream Vera Sans" "Roman"
[08:29] <ivoks> just kill those Adobe lines in fonts.conf
[08:29] <jdub> if we never want nimbus, just kill it from the mathces
[08:30] <jdub> regardless of printing
[08:30] <Diziet> But we do want it for printing, I think.
[08:30] <jdub> alternatively, check out red hat's fonts.conf
[08:30] <Diziet> Now _that_'s a good idea.  Copy DR.  Joy.
[08:31] <jdub> no, 'check out'
[08:31] <jdub> it *always* helps to survey what others are doing
[08:31] <Diziet> I think we should try to understand.
[08:31] <ivoks> so, problem isn't in #150 in fonts.conf?
[08:31] <jdub> that may be assisted by looking at alternatives
[08:31] <Diziet> OMG we are going round in circles.
[08:32] <jdub> red hat have a solid history of doing the right thing regarding fonts
[08:32] <ogra> Diziet, just wipe Nimbus as jdub says
[08:33] <Diziet> And WTF is with the XML config file anyway ?!
[08:33] <Diziet> I'm losing it, can you tell !
[08:35] <ogra> which XML file ? 
[08:36] <jdub> fonts.conf
[08:36] <ogra> ah
[08:37] <ivoks> why don't we alias times to serif?
[08:38] <ivoks> helvetica to sans and courier to mono?
[08:39] <ogra> exactly
[08:40] <Diziet> Ah, good, my ffox 18 has built.
[08:41] <Nafallo> is there a list of what packages the installation uses? Rosetta has packages for Debians, Ubuntus and boths installations it seems :-P.
[08:41] <Nafallo> lot of templates with the same strings floating around...
[08:42] <bddebian> Damn, ghc-cvs still hasn't shown up
[08:43] <runeh> Where do I request package updates? Need the newest libipoddevice and such for my nano to work in Banshee.
[08:44] <Nafallo> runeh: #ubuntu-mono, slomo.
[08:44] <runeh> Thanks.
[08:44] <slomo> runeh: i'll do them soon, don't worry ;)
[08:44] <runeh> Ah. Thanks. :)
[08:45] <Nafallo> hihi
[08:45] <Nafallo> hilight deluxe :-P
[08:45] <runeh> Been trying to build them myself, but had no luck. :P
[08:45] <slomo> i just wait for a new banshee release which should be released soon and then do all 3 packages at once
[08:45] <runeh> Yeah. They said monday.
[08:45] <zyga> libipoddevice?
[08:45] <zyga> geez open source has a lib for everything
[08:45] <zyga> libyourmotherinlaw....
[08:45] <Nafallo> zyga: ;-)
[08:46] <slomo> runeh: it's not worth the worries... ;) when nothing happens until saturday i'll upload a cvs snapshot of banshee with the newest release of the 2 libs...
[08:46] <slomo> runeh: the cvs fixes many really ugly bugs
[08:46] <runeh> zyga: For an easy, higher level way to access her?
[08:46] <Nafallo> hmm. would be nifty. apt-get remove --purge libyourmotherinlaw :-P
[08:46] <zyga> runeh: yeah, there's a remote access method that's far safer
[08:47] <zyga> you can send her encapsulated messages 
[08:47] <runeh> slomo: Yeah. And I figured it's nicer to have dpkg handle the libraries installed, than me pushing modified versions in here and there.
[08:48] <runeh> Looking forward to next week then. :)
[08:48] <ivoks> Diziet: change, not delete ;)
[08:48] <runeh> Don't like having to use a friends computer to transfer music to my player. :P
[08:49] <mjg59> jbailey: Hang on. So we unconditionally attempt to load vesafb?
[08:49] <mjg59> jbailey: That would explain why all the other framebuffer modules are already loaded when we try to start usplash
[08:49] <slomo> runeh: ;)
[08:49] <jbailey> mjg59: Yup, because the kernel package asks us to.
[08:50] <jbailey> Although usplash I think runs before that.
[08:51] <mjg59> jbailey: So how on earth are we trying to load modules that are already loaded?
[08:52] <mjg59> Or does usplash do something funny like add vga16fb's depends to the autoload list?
[08:52] <jbailey> mjg59: What bug number are you looking at, so I can follow along? =)
[08:52] <mjg59> #15317
[09:03] <HiddenWolf> pitti, around?
[09:04] <Kamion> you know if LaschW actually stayed around for more than five minutes I might be able to answer his question
[09:05] <jbailey> Anyone here got the insmod: error inserting
[09:05] <jbailey> '/lib/modules/2.6.12-8-386/kernel/drivers/video/console/bitblit.ko': -1 File Exists
[09:05] <jbailey> errors at startup?
[09:05] <Nafallo> jbailey: yes
[09:05] <jbailey> Nafallo: Are you free to do a test that involved rebooting?
[09:06] <Nafallo> s/8-386/9-amd64-k8/
[09:06] <bddebian> Gawd I hate C++
[09:06] <jbailey> Nafallo: Right. =)
[09:06] <Nafallo> jbailey: not at the moment, I'm making Kamion happy ;-).
[09:06] <zyga> jbailey: me
[09:06] <jbailey> Nafallo: I would never interfere with Kamion's happiness. =)
[09:06] <zyga> jbailey: strange, noticed it yesterday
[09:06] <Nafallo> jbailey: I could check if 9-686 is though :-P
[09:06] <Nafallo> hihi
[09:07] <zyga> jbailey: 9-k7
[09:07] <jbailey> bddebian: You keep hacking on *bad* C++.  People who think that casts and CPP macros are good.
[09:07] <Kamion> next time LaschW shows up, somebody point him at 'apt-get source oem-config localechooser kbd-chooser' please
[09:07] <bddebian> jbailey: Aye.  I'm back looking at xgsmlib :-(
[09:07] <pitti> Hi HIdd
[09:08] <pitti> oh, neat timing
[09:08] <Nafallo> jbailey: 9-686 availible :-)
[09:09] <Keybuk> mdz: want to see something even funnier?
[09:09] <Keybuk> with Caps Lock _on_, go onto the console and type "abcdef" with the shift key held down
[09:09] <Nafallo> Keybuk: you have a bug from me :-)
[09:09] <Keybuk> you'll get abcCEf
[09:09] <Keybuk> uh, abCdEf
[09:09] <zyga> Keybuk: that's an old story ;-)
[09:10] <zyga> Keybuk: depending on the locale
[09:10] <Keybuk> zyga: it's still being told today
[09:10] <zyga> Keybuk: I'll get aBcDeF
[09:10] <Keybuk> bet that means your locale has a fancy accented character for 'a' ?
[09:10] <Nafallo> jbailey: did you get my msgs? :-)
[09:10] <Keybuk> ie. you have , ,  or anything like that on your keyboard
[09:11] <zyga> Keybuk: 
[09:11] <zyga> Keybuk: (you didn't think of that - did you ;-)
[09:11] <Keybuk> didn't even know that one existed
[09:11] <zyga> :D
[09:11] <Nafallo> :-)
[09:11] <Keybuk> looks like Salvador Dali's 'a' key
[09:11] <Keybuk> Nafallo: oh?
[09:11] <Nafallo> Keybuk: grepmap segfaults here. 16133 IIRC
[09:11] <zyga> if you are *so* interested I can show you the capital version...
[09:12] <Keybuk> Nafallo: oh, good, can I have your modules.inputmap
[09:12] <Keybuk> (grepmap has never worked on amd64, just nobody noticed)
[09:12] <Nafallo> sure, I'll attach it :-)
[09:12] <Keybuk> the inputmap array size is different on amd64 to i386, and so far we've just been lucky there was nothing important after it in memory
[09:13] <jbailey> Nafallo: I did sorry, suddenly had 3 popups at once. 
[09:13] <Keybuk> zyga: nah, it just fits the pattern of the bug, that's all
[09:13] <Keybuk> any key with anything more than just the basic latin lower and upper case variants gets inverted
[09:13] <Keybuk> (c and e are usually also  and  in most euro locales)
[09:14] <jbailey> Keybuk: Eh.  Don't you use UTF-8? =)
[09:15] <Keybuk> jbailey: yes, afaik
[09:16] <jbailey> Huh, I got a cents sign from you and then some noise for the second character.
[09:16] <Keybuk> unless x-chat in breezy has turned it off by default
[09:16] <mjg59> jbailey: That was utf-8
[09:16] <jbailey> Huh, weird.
[09:17] <spayne> who where is on the CC?
[09:17] <spayne> Community Council i mean
[09:17] <ogra> spayne, elmo, Kamion, mako 
[09:18] <spayne> mako: ping
[09:18] <Nafallo> Keybuk: well, I'm here for your debugging needs ;-).
[09:18] <Keybuk> Nafallo: -> /msg
[09:18] <spayne> elmo: ping
[09:18] <spayne> Kamion: ping
[09:19] <pitti> seb128: can I somehow switch to the Gnome keyboard settings? I believe I currently use X', so the keyboard applet etc. don't work
[09:19] <\sh> zyga: meta+3 i think
[09:19] <pitti> mdz: good news about langpacks - it seems that we'll get new packs tomorrow
[09:20] <pitti> mdz: having fresh packages tomorrow for testing, and another set of fresh packages right before the release - is that fine?
[09:20] <pitti> HiddenWolf: pong
[09:22] <zyga> does anything use ~/.local ?
[09:22] <zyga> apart from autopackage?
[09:22] <HiddenWolf> pitti, all the ports in my cardreader suddenly show up in the gtk filechooser window. Even when there is nothing in them.
[09:22] <mdz> pitti: depends on how they go tomorrow ;-)
[09:22] <mdz> pitti: get them in as soon as humanly possible
[09:22] <HiddenWolf> pitti, not in places or on the desktop. is that seb128's, or something for you?
[09:23] <pitti> mdz: carlos just tested today's tarball, and it just has three obsolete translation domains
[09:23] <seb128> pitti: /desktop/gnome/peripherals/keyboard/kbd/overrideSettings maybe?
[09:23] <zyga> or... does anyone here use/ever-used autopackage
[09:23] <pitti> mdz: so in theory I could even use it, but tomorrow we should get all
[09:23] <Keybuk> zyga: freedesktop-ish stuff uses it too
[09:23] <zyga> Keybuk: hmm
[09:23] <Keybuk> ~/.local/share/applications is where custom .desktop files go, for example
[09:24] <zyga> Keybuk: mine was heavily polluted 
[09:24] <zyga> with autopackage
[09:24] <Keybuk> yes, well, autopackage
[09:24] <Keybuk> aka. "please fuck my system with a chainsaw, no, pleeeease"
[09:24] <zyga> vim: relocation error: /lib/tls/i686/cmov/librt.so.1: symbol __pthread_clock_settime, version GLIBC_PRIVATE not defined in fil e libpthread.so.0 with link time reference
[09:24] <mdz> pitti: we need some time to catch the errors we haven't seen yet
[09:24] <mdz> pitti:  through user exposure
[09:24] <mdz> pitti: which domains are they
[09:24] <pitti> mdz: right
[09:25] <ogra> mdz, that ltsp patch you sent is for the debian package... while i agree that the exporting should be done in the build-client script, i dont agree with the implementation as well as with the rest ...
[09:25] <pitti> mdz: btw, we always review the full diff, so it shouldn't get too bad
[09:25] <mdz> pitti: which domains are obsolete?
[09:26] <mdz> pitti: you review the full diff including all of the changed translations in every language?
[09:26] <pitti> mdz: gdm, control-center and evolution
[09:26] <carlos> mdz, gdm, evolution and control-center
[09:26] <mdz> pitti: oh, those are important ones ;-)
[09:26] <pitti> mdz: well, I can't review every single bit of 22 MB :-)
[09:26] <mdz> ogra: I forwarded it to you because you are working on the same problem
[09:26] <ogra> mdz, not really...
[09:26] <pitti> mdz: but when you look a it, you get a fairly good idea about the differences
[09:27] <mdz> pitti: but there are many updated translations from rosetta, right?
[09:27] <ogra> mdz, i need to generate the dhcp.conf on the fly... these probs are already solved for ue
[09:27] <pitti> mdz: yes, some
[09:27] <mdz> pitti: that is what I mean, we need time to QA the translation changes
[09:27] <ogra> s/ue/us
[09:27] <pitti> mdz: right, exactly my feelign, that's why I want fresh packs ASAP, and not just before release
[09:27] <ogra> mdz, but moving the exports to the script is indeed the better way
[09:32] <janimo> seb128 if there's a way of reminding you of the lpi/libgnome patch without being annoying let me know :)
[09:33] <seb128> janimo: I just have something like 700 bugs to my bugzilla list so you have to be patient
[09:34] <janimo> seb128, ok it is just holding up xubuntu-desktop a bit, but ok if it is done before release I am fine.thanks
[09:35] <seb128> you can't get xubuntu-desktop grabbing libgnome for the moment?
[09:36] <janimo> seb128, I could I guess but that may mean the actual testers will be fewer if they have memory footprint issues
[09:36] <janimo> I'll do that if it takes a while till you get to the fix though
[09:37] <janimo> I just worry not to delay it too much so it becomes too risky for breezy
[09:38] <seb128> janimo: I'll have a look on the patch tomorrow
[09:39] <janimo> sbe128, merci
[09:40] <zyga> jbailey: fixed
[09:41] <jbailey> zyga: \\o/
[09:41] <jbailey> I'll wait for the confirmation from Nafallo too then and then close the bug.
[09:42] <seb128> janimo: de rien
[09:48] <spayne> pitti: ping
[09:48] <pitti> spayne: pong
[09:56] <janimo> which is the program that gets SUSPEND/REBOOT etc commands from gnome-session and calls pmi in turn?
[09:56] <mjg59> janimo: gnome-settings-daemon
[09:58] <ogra> mjg59, ? 
[09:58] <mjg59> ogra: Hello?
[09:58] <ogra> isnt that gdm anymore ? 
[09:58] <mjg59> Oh, sorry, I see 
[09:58] <mjg59> janimo: Yeah, ogra is right. It's gdm that actually calls pmi
[10:00] <janimo> ok thanks 
[10:00] <janimo> I actually manage to call pmi from xfce-session but on resume X is killed and I am dropped into console
[10:01] <janimo> so I have to see what does gnome do besides calling pmi
[10:01] <janimo> does gdm run as root?
[10:01] <ogra> janimo, yup
[10:01] <janimo> so no sudo right
[10:02] <ogra> yup
[10:02] <janimo> hmm that is one difference
[10:02] <janimo> thanks ogra
[10:02] <ogra> do you have a login manager wrapped in your xubuntu package ? 
[10:05] <janimo> ogra, not yet
[10:05] <janimo> haven't decided if we'll use one or not
[10:05] <zyga> Keybuk: how to re-set the keyboard after kbd_mode -k
[10:06] <janimo> 'we' is exagerrated, so far noone comments on xubuntu ;)
[10:06] <janimo> it may be time to announce to users and ask for opinions
[10:07] <zyga> janimo: DD will need a 'select desktop' post-install package that shows screenshots and installs [xg] ?ubuntu-desktop
[10:07] <jdub> seb128: ping
[10:08] <seb128> jdub: pong
[10:08] <jdub> seb128: did you get to chat to vuntz about the menu logo stuff?
[10:08] <janimo> zyga, DD?
[10:08] <janimo> debian developers?
[10:09] <seb128> jdub: yeah, didn't you read what I put to the query before going to bed yesterday?
[10:09] <zyga> Dapper
[10:09] <janimo> aha
[10:09] <janimo> you think they'll fit on the same CD?
[10:09] <zyga> janimo: - no
[10:10] <janimo> although xubu  is mostly a subset of ubu
[10:10] <zyga> janimo: but for network enabled users this could be neet
[10:10] <janimo> sure
[10:10] <zyga> janimo: broadband enabled users ;-)
[10:10] <janimo> so I am out ;)
[10:10] <jdub> seb128: hmm
[10:10] <jdub> seb128: aha!
[10:11] <jdub> seb128: unfortunately, that still requires us to create a new icon theme, unless we can hack it with stock icons
[10:11] <seb128> jdub: why?
[10:11] <seb128> jdub: just ship it to the hicolor folder
[10:11] <jdub> directly in ubuntu-artwork?
[10:11] <jdub> i guess that works
[10:14] <ogra> jdub, ubuntu-artwork needs a autotools update ... i (actually jbailey) fell into this with edubuntu-artwork
[10:14] <jdub> seb128: turns out there's already a gconf hack for this
[10:15] <jdub> ogra: hrm, i don't ship a native pacakge
[10:15] <jcohen85> jbailey: did you ever get a chance to write a new initramfs-tools package?
[10:15] <jbailey> jcohen85: No, but I'd be curious if you see the same pauses now.
[10:15] <ogra> jdub, but its generated with automake1.7 or something similar old...
[10:15] <jbailey> jcohen85: The splash screen starts sooner now.
[10:16] <seb128> jdub: where?
[10:16] <jbailey> jcohen85: It also seems that udevstart might wind up pausing on non-responding devices, but I'm not sure of that yet.
[10:16] <jdub> seb128: /apps/panel/default_setup/objects/menu_bar
[10:18] <jcohen85> jbailey: ok, i'll try again and tell you what happens
[10:18] <jbailey> jcohen85: Thanks!
[10:18] <jdub> ogra: this doesn't seem to have a negative impact on the resulting package, however
[10:19] <ogra> jdub, thats true....
[10:19] <jdub> seb128: can't seem to get any of those keys to work
[10:19] <seb128> jdub: seems to be a leftover of some old key
[10:19] <ogra> even i dont understand the whole makefiel fuss in there
[10:19] <jdub> ogra: it's just a relatively standard gnome style autofoo module
[10:20] <ogra> yes, but cp or dh_install would suffice :)
[10:20] <jdub> seb128: did you ship the distributor-logo patch?
[10:20] <jdub> ogra: not for intltooling
[10:21] <ogra> ah, true
[10:21] <jdub> seb128: which section should it be in?
[10:21] <ogra> thats the missing bit then :)
[10:22] <seb128> jdub: not yet, these gconf key seem to be still used by the panel in fact
[10:22] <seb128> except they don't apply to current config
[10:22] <seb128> just if you create a new panel
[10:22] <jdub> what about the setting under /apps/panel/objects ?
[10:23] <jdub> hrm, though i didn't restart the panel
[10:24] <jdub> nup, not even if i do that
[10:25] <jdub> seb128: ok, if you can push the distributor-logo patch, let me know which section it is, and i'll respin u-a
[10:27] <ogra> jdub, do you have a icon ? i'd have a bumped slightly glowing version here if you need one....
[10:28] <seb128> jdub: 48x48/apps ?
[10:29] <ogra> s/bumped/punched
[10:30] <tseng> jdub: is there more art coming?
[10:31] <tseng> jdub: im getting anxious
[10:31] <Nafallo> hmm
[10:31] <jdub> ogra: ah, worth having a try with that too - send over when you have a minute - thanks :)
[10:31] <jdub> tseng: u-a upload a few minutes ago
[10:31] <Nafallo> jdub: please detect that I use WS and set the background accordingly? ;-)
[10:32] <jdub> Nafallo: heh, i wish :-)
[10:34] <jdub> ha ha
[10:34] <jdub> gar, after all that, i uploaded the wrong one
[10:34] <Nafallo> lol
[10:34] <jdub> didn't include the circle update
[10:35] <Nafallo> circle update? gdm?
[10:35] <jdub> yeah
[10:36] <ogra> grmpf.... and my next evo crash...
[10:36] <tseng> circle-of-elmo?
[10:36] <ogra> #6 today ...
[10:36] <tseng> jdub: 
[10:36] <tseng>      - removed browser homepage
[10:36] <tseng> i love you
[10:36] <Nafallo> I almost hope for a replacement of the gnome foot in the left corner ;-)
[10:36] <jdub> Nafallo: that's coming in the next upload
[10:36] <Nafallo> yay
[10:37] <jdub> bug seb about required panel changes ;-)
[10:38] <Nafallo> seb128: ^ :-)
[10:39] <ogra> seb128, my evo is crashy... i just wanted to gdb backtrace it and recognized about 40 running e-d-s processes... is this a new behavior ? or rather a bug ?
[10:39] <seb128> Nafallo: let's wait for artwork changes so I can test before uploading
[10:39] <Nafallo> seb128: oki :-)
[10:39] <seb128> ogra: bug ... but don't you get a bug-buddy dialog?
[10:39] <ogra> nope
[10:39] <ogra> it just hardlocks
[10:40] <seb128> gdb evolution
[10:40] <seb128> (gdb) handle SIG33 pass nostop noprint
[10:40] <ogra> actually the e-d-s processes might be from other crashes ...
[10:40] <seb128> (gdb) run
[10:40] <seb128> ...
[10:40] <seb128> wiat for a hang

[10:40] <seb128> thread apply all bt
[10:40] <seb128> evolution --force-shutdown 
[10:40] <seb128> to cleanup
[10:41] <ogra> seb128, ok... it an take some hours... happens all 4-5h 
[10:41] <Amaranth> hopefully tonight i'll be able to make my pre-colony 4 breezy install work again so i can work on smeg :)
[10:43] <ogra> wow, what a luck ... now it hang immediately :) this hasnt happened before...
[10:44] <Nafallo> ogra: lol. you have a funny way to define luck ;-).
[10:44] <ogra> Nafallo, nope i'm short of time to donate to such things... but it seems it pulled down gdb whit it... very weird...
[10:44] <Nafallo> "YES! My package hangs! Now I can go eat marshmallows for the rest of my life."
[10:45] <Nafallo> ogra: hehe, joy :-)
[10:45] <Nafallo> ogra: I would go for the marshmallows if I were you :-).
[10:48] <ogra> seb128, (gdb) thread apply all bt
[10:48] <ogra> Cannot find new threads: generic error :/
[10:48] <seb128> urg
[10:48] <seb128> gdb bog ?
[10:48] <ogra> hmm, but also evo boog
[10:48] <ogra> since it hangs again no
[10:48] <ogra> w
[10:49] <seb128> anybody knows about this gdb msg?
[10:49] <ogra> oh
[10:49] <ogra> gdb is stopped and evo still runs 
[10:49] <ogra> now i have a crash dialog
[10:50] <ogra> hrm... but bug buddy hangs too...
[10:50] <ogra> damn
[10:50] <dholbach> wait a bit
[10:51] <dholbach> i guess bug-buddy is working a bit
[10:51] <ogra> it hard locked and my cpu runs full speed...
[10:51] <ogra> oh, youre right
[10:51] <dholbach> it's working
[10:51] <ogra> but veeeerrryyy slllooooowwww
[10:52] <ogra> jdub, dont tell me you wiped the homepage from the artwork
[10:52] <ogra> edubuntu-rtwork relies on that
[10:52] <jdub> ogra: ha ha ha
[10:52] <jdub> oh dear
[10:52] <jdub> why's that?
[10:52] <ogra> shit
[10:53] <jbailey> Is that the FF homepage?
[10:53] <ogra> because mozilla has this hardcoded as a home location...
[10:53] <jbailey> Is it moving to -docs now?
[10:53] <jdub> jbailey: i thought that change had already been made?
[10:53] <ogra> i didnt want to fiddle with firefox
[10:54] <jdub> the intent was to use the html-generated about page from ubuntu-docs
[10:54] <mdke> yep
[10:54] <jdub> though ideally we'd have something slightly more useful
[10:54] <ogra> gah, my system burns... its at 80 degree and bug buddy still only moves the load bar 1mm a minute
[10:55] <jdong> ogra: that's not good for your computer.... :)
[10:55] <ogra> haha
[10:55] <jbailey> jdub: I don't think so, but tell me what you need and I'll make it happen pdq.
[10:55] <jdong> ogra: unless it's an athlon tbird
[10:55] <ogra> its an amd64 lappie
[10:55] <jdub> jbailey: just to make sure there's an html version of the about ubuntu page shipped in... something :-)
[10:55] <ogra> it survives until 89 degree 
[10:55] <Nafallo> ogra: !targa? :-)
[10:56] <jbailey> mdke: Do you know if we have a suitable doc for that already, or should I drag up the old one from artwork?
[10:56] <ogra> Nafallo, aspire 1520
[10:56] <jdub> ogra: one fix to ff will sort it out, don't worry
[10:56] <jdong> ogra: my athlon64 desktop burns at 70-ish
[10:56] <Nafallo> ah, acer.
[10:56] <Nafallo> seems you should have gone to lidl instead ;-)
[10:57] <dholbach> wow, i'm on berlinblogs.com and didn't know :)
[10:57] <jbailey> dholbach: They're stalking you ;)
[10:58] <ogra> hmm, probably i should install the 200 waiting updates ... might please evo :)
[10:58] <mdke> jbailey, aboutubuntu is the doc dude
[10:58] <mvo> dholbach: you are famous already?
[10:58] <Nafallo> ogra: OMFG :-P
[10:58] <dholbach> mvo: haha, not really
[10:58] <mdke> jbailey, put it in /usr/share/doc/ubuntu-docs/HTML i guess?
[10:58] <jdong> ogra: MOTU preventing you from doing other stuff? ;)
[10:58] <jbailey> mdke: Lovely. =)
[10:59] <ogra> jdong, i'm not only MOTU 
[10:59] <mdke> jbailey, do you think there is any way of localising the firefox homepage? e.g. HTML/en/ HTML/fr and getting them to appear in the right language?
[10:59] <jdub> jbailey: then i think you can pass the firefox fix to Diziet 
[10:59] <jdub> :-)
[10:59] <jdub> mdke: no
[10:59] <mdke> there are loads of translations for about-ubuntu
[11:00] <mdke> jdub, i bet someone is clever enough to do that
[11:00] <ogra> jdub, but thast really odd, now i have to change ff defaults for custom distros... thats odd, who decided that ? 
[11:01] <mdke> it makes sense ogra, ubuntu-artwork doesn't maintain that page
[11:01] <ogra> i dont want to fiddle with app defaults
[11:01] <ogra> mdke, but hacking ff is no option... so i can forget about about-edubuntu 
[11:02] <ogra> it was perfectly right in the -artwork package ....
[11:02] <mdke> ogra, maybe you can give the file the same name, and ship it in edubuntu-docs
[11:02] <Nafallo> ogra: no. -docs makes sense...
[11:02] <mdke> that way you don't need to play with ff
[11:03] <ogra> Nafallo, its a branding thing, not a doc thing imho
[11:03] <mdke> oh well
[11:03] <ogra> and -artwork are the branding packages
[11:03] <jbailey> In the end, so is docs, really.
[11:03] <mdke> ogra, the doc is maintained b the doc team
[11:03] <Nafallo> ogra: yea, but rather have the -docs conflict and put the page in /usr/share/homepage.html or something ;-)
[11:03] <jbailey> I'm not clear on the separation aside from who's producing it.
[11:03] <mdke> not much point having to syncronise the two packages
[11:04] <jdub> ogra: you would've had to change the url in the old scheme too
[11:04] <ogra> Nafallo, i can only put the page where ff looks ... 
[11:04] <ogra> jdub, nope
[11:04] <mdke> ogra, then agree with jbailey the right place to put the page
[11:04] <jdub> the old url included ubuntu-artork
[11:04] <Nafallo> ogra: well, we have to make ff look at things where it's sensible then...
[11:05] <ogra> jdub, i linked it in postinst and removed the link in postrm... edubuntu-artwork depends on ubuntu-artwork since it reuses most of the ubuntu stuff from there
[11:05] <mdke> ogra, do something similar?
[11:05] <ogra> so it was perfectly right there without much fiddling... now i have to have two packages for a derivative branding
[11:06] <jbailey> You'd want to fiddle with -docs anyway
[11:06] <ogra> mdke, i'll do that anyway since i need -docs for edubuntu... but that way we force everyone to have *two* branding packages for his derivative, thats not nice
[11:07] <mdke> do we?
[11:07] <ogra> sure
[11:08] <mdke> well the package is maintained in our repository
[11:08] <jbailey> The whole fact that the file is in a path that contains the word ubuntu breaks the sabdfl's want to make it possible to remove the word 'ubuntu' from the system, I think.
[11:08] <mdke> do with it what you like
[11:08] <sladen> ogra: could you move human etc into a base-branding and then keep the really specific stuff in a separate package (and then perhaps override that)
[11:08] <mdke> jbailey, iirc mark gave his approval to putting the page in ubuntu-docs
[11:08] <ogra> sladen, yes, that would probably be best...
[11:09] <ogra> but having such chamges now that my packages were in order and i didnt plan time to change them before release is quite hard to manage...
[11:09] <ogra> but well...
[11:10] <mdke> ogra, perhaps it is best if you guys discuss it all together on the mailing list or something
[11:10] <ogra> it would have been nice to know about it before...
[11:10] <mdke> it is not too late to sort it out either way
[11:10] <mdke> ogra, yeah, you said that
[11:11] <ogra> mdke, for me it is... i just lost a week of edubuntu development to the screensaver
[11:11] <ogra> and i dont even know here to get the time to finish edubuntu ... now surprise surprise i have to touch packages that i thought were finished
[11:12] <ogra> it just breaks my schedule..
[11:12] <mdke> ogra, perhaps just write an email to the list and let the people who decide these things sort it out
[11:13] <ogra> mdke, how ? i will make the change to my package... but the communication here is not well it seems, its ot the first time i ran into such a trap.. i'll put a BOF on the BOF list to have it better in dapper...
[11:13] <ogra> s/ot/not
[11:14] <lamont> I wonder if I should worry that xscreensaver-data (or some such) failed to configure correctly during a dist-upgrade to breezy
[11:14] <ogra> mdke, the decision apprently was made already...
[11:14] <mdke> ogra, it is very trivial to un-decide and re-decide, afaics
[11:15] <mdke> since the file just needs to be built and inserted into one or the other package
[11:15] <mdz> mdke: there are translations for About Ubuntu? they aren't in the package.
[11:15] <mdke> mdz, not yet
[11:15] <ogra> mdke, its a questin if i sleep an hour less and change my package or if 5 people discuss it again and lose 1h plus 2 people have to work on 2 packages...
[11:16] <mdke> mdz, jbailey is on it :)
[11:16] <mdz> mdke: are you saying translations exist but are not packaged yet?
[11:16] <mdke> mdz, yes
[11:16] <seb128> mvo, pitti: about cdrdao, is it promoted?
[11:16] <mdz> because that page is going to change
[11:16] <mdke> mdz, oh dear
[11:16] <ogra> mdke, i will just accept it now... but i'm not happy that i didnt know about it
[11:16] <pitti> seb128: the issues are resolved, so it can be promoted
[11:17] <mdke> mdz, what do you mean?
[11:17] <mdz> mdke: I mean it's needed to be replaced for a long time and we are finally getting to it
[11:17] <seb128> pitti: can I upload a ncb Depending on it ?
[11:17] <mdke> mdz, i was copied into a discussion about this today I think...
[11:17] <seb128> hum, that would be wrong
[11:17] <seb128> should we list if for desktop ?
[11:17] <mdke> mdz, it didn't seem like it was going to be changed
[11:17] <mdke> mdz, mark already rewrote it completely once
[11:17] <pitti> seb128: Suggests: sounds more appropriate, and seeding it to supported
[11:17] <mdke>  [22:16:13]  < mdke> mdz, jbailey is on it :)
[11:17] <mdke> argh
[11:18] <dholbach> good night
[11:18] <ogra> lamont, error ?
[11:18] <seb128> pitti: not desktop?
[11:18] <pitti> seb128: or do you want it in desktop? CD space is expensive...
[11:18] <mdke> mdz, translations are at: t:
[11:18] <pitti> night dand 
[11:18] <mdke> https://launchpad.net/distros/ubuntu/breezy/+sources/ubuntu-docs/+pots/aboutubuntu
[11:18] <seb128> pitti: I would like to get "duplicate CD" working out of the box
[11:18] <mdz> mdke: there's already someone working on it.  I didn't consider it a big deal to replace it late because I could see that there were no translations
[11:18] <mdke> mdz, there are like 20 languages
[11:18] <pitti> seb128: fine by me, but Kamion/mdz should approve, too
[11:18] <lamont> *** libsane.usermap (Y/I/N/O/D/Z) [default=N]  ? 
[11:18] <lamont> we know about taht one?
[11:18] <seb128> pitti: universe or supported doesn't make a real difference, people have to install it
[11:19] <seb128> pitti: sure
[11:19] <lamont> empty-> 453 lines
[11:19] <pitti> lamont: I tested upgrades from hoary, I didn't get that
[11:19] <mdke> mdz, janew said today in her email that they had decided not to employ a company to work on the page
[11:19] <ogra> lamont, what was the xscreensaver-data error ? 
[11:19] <lamont> I may have universe crappage as well
[11:19] <pitti> lamont: empty??? how did you get that?
[11:19] <lamont> ogra: the xscreensaver-data error was, um, scrolled off the screen
[11:19] <Keybuk> jdub: no new HumanCircle yet?
[11:19] <ogra> lamont, ok
[11:19] <mdz> mdke: who are 'they'?  please forward me this email, I should have been in the loop
[11:20] <Keybuk> pitti: btw, did libsane.usermap magically start working with the grepmap update?
[11:20] <mdke> mdz, i haven't got my computer on me, i left it at the office, but JaneW will fill you in I'm sure, otherwise I can forward tomorrow sometime
[11:20] <mdz> mdke: she hasn't
[11:20] <mdke> mdz, jdub was in cc though
[11:20] <pitti> Keybuk: no idea, in the last bug report the culprit was the missing libusbscanner script
[11:20] <Keybuk> ahh
[11:21] <mdke> mdz, i'm sure if you mail her she will forward it quicker than I can
[11:22] <mdz> mdke: JaneW is gone for the day and I am leaving town tomorrow
[11:22] <mdz> and this is something that I asked to be done
[11:22] <mdz> so I need to know if something has gone wrong
[11:22] <mdke> mdz, ah ok, listen
[11:22] <mdke> mdz, when I replied, I copied in ubuntu-doc@lists
[11:23] <lamont> ogra: my bad
[11:23] <mdke> mdz, you won't see all of her email, but the general gist is there
[11:23] <mdke> mdz, http://news.gmane.org/gmane.linux.ubuntu.doc/cutoff=3255
[11:23] <lamont> Unpacking replacement xscreensaver-data ...
[11:23] <lamont> dpkg: error processing /var/cache/apt/archives/xscreensaver-data_4.21-4ubuntu14_i386.deb (--unpack):
[11:23] <lamont>  trying to overwrite `/usr/share/man/man1/xscreensaver-getimage.1.gz', which is also in package xscreensaver
[11:23] <lamont> dpkg-deb: subprocess paste killed by signal (Broken pipe)
[11:23] <mdke> mdz, hope that helps
[11:23] <lamont> ogra: amusingly, dpkg --configure -a and the problem did not reoccur
[11:24] <mdke> mdz, (the thread is "About UBuntu to be beautified")
[11:24] <mdz> mdke: that message is about KReleaseNotes...do you remember the subjcet line?
[11:24] <mdz> ok
[11:24] <lamont> hrm.
[11:24] <lamont> another dist-upgrade, wants to install a new xscreensaver-data
[11:24] <lamont> ogra: so missing Replaces?
[11:25] <mdke> mdz, in a later email (not on the list) I made the point about there being loads of translations
[11:25] <mdz> mdke: ok, jane was confused
[11:25] <mdke> right...
[11:25] <mjg59> mdz: Most of the usplash stuff should be sorted
[11:26] <ogra> lamont, which version of xscreensaver was there before, do you know that ? 
[11:26] <lamont> Preparing to replace xscreensaver-data 4.21-4ubuntu12 (using .../xscreensaver-data_4.21-4ubuntu14_i386.deb) ...
[11:26] <ogra> lamont, there is a replaces, but it might contain the wrong version
[11:26] <ogra> lamont, thanks will fix that
[11:26] <lamont> ogra: thanks
[11:26] <ogra> :)
[11:27] <mdke> mdz, anyhow you saw the url with the translations?
[11:27] <mdz> mjg59: which stuff?  I think outstanding were progress bar border, insmod errors, console font issues
[11:27] <mjg59> mdz: Progres bar border, insmod errors should be fixed since the move to init-top rather than init-premount
[11:27] <mdz> mdke: I see it, and it's a shame that they were never used
[11:27] <mdke> mdz, jbailey is working on putting them into the package
[11:27] <mdke> mdz, they are quite recent
[11:27] <mjg59> mdz: The previous issue would have been that vesafb was being loaded first, failing and then we were trying to reload the same modules
[11:28] <mjg59> mdz: Also, it now works with vesafb
[11:28] <mdz> mdke: jbailey should now that the deadline was today
[11:28] <mdz> mjg59: yay
[11:28] <mdke> mdz, i didn't know that :/
[11:28] <mdz> s/now/know/
[11:28] <mdz> mdke: it's been on https://wiki.ubuntu.com/BreezyReleaseSchedule since May
[11:29] <mdke> mdz, yes, but I am just a volunteer
[11:29] <mdke> mdz, anyhow, i understood that translations could be updated
[11:29] <mdke> as with the language-packs
[11:29] <mdz> mdke: I don't follow from BreezyReleaseSchedule -> volunteer
[11:30] <mdz> the schedule applies to the entirety of Ubuntu, not individuals
[11:30] <mdke> mdz, yeah, i wasn't suggesting it didn't apply, just that i wasn't aware of this deadline
[11:30] <mdke> mdz, perhaps the best thing would be to hurry jeff along?
[11:32] <mdke> or leave the translations out, or update them. Whatever you decide
[11:33] <mcella> is there a way to hide usplash progress verbose output?
[11:33] <mdz> mdke: jbailey is gone for the day according to his schedule
[11:33] <mdke> mdz, ok, your call on what you want to do.
[11:33] <jbailey> mdke: I'm still here.
[11:33] <jbailey> Sorry was dealing with puking cat.
[11:34] <mdke> ouch
[11:34] <pitti> Robot101: here?
[11:34] <mdke> jbailey, mdz says that today was the deadline for translations
[11:34] <Nafallo> jbailey: hope you didn't mean me ;-)
[11:34] <jbailey> Nafallo: LOL
[11:34] <Nafallo> jbailey: msg btw :-)
[11:34] <mdz> mdke: after we prepared the schedule for breezy, I thought I remembered sending you the URL personally, even, to verify that it met your needs (since it was in part based on your input)
[11:35] <mdz> mdke: non-language-pack translations, yes
[11:35] <mdke> mdz, yep you did
[11:35] <mdke> mdz, but I have not been keeping up with it, because I am working quite hard at the moment and haven't been able to do much docs work
[11:35] <mdz> pitti: did we succeed in getting oo.o or firefox into langpacks?
[11:35] <mdz> pitti: we need to create a list of what qualifies for NonLanguagePackTranslationDeadline
[11:36] <pitti> mdz: oo.o yes, with ffox there were too many troubles
[11:36] <pitti> mdz: I had a long discussion with carlos about ffox, it's not breezyable
[11:36] <Nafallo> oh
[11:36] <mdz> pitti: ok, so installer, firefox, HTML documents, desktop files...
[11:36] <Nafallo> OO.o in langpacks :-P
[11:36] <Nafallo> damn
[11:36] <jbailey> mdz, mjg59: Nafallo and zyga have both confirmed the errors are gone with the current initrafs/usplash
[11:36] <mdz> pitti: anything else worth mentioning specifically?
[11:37] <mjg59> Ok. I'll go back to concentrating on suspend/resume, then
[11:37] <pitti> mdz: maybe upgrade notes and debconf templates
[11:37] <Nafallo> ehm, HTML Documents is ubuntudocs?
[11:37] <mdke> mdz, ok i'm going to bed, jbailey knows where the translations are if you decide you want them in
[11:39] <mdz> pitti: upgrade notes, as in notification-daemon
[11:39] <mdz> mdke: ok, good night
[11:39] <pitti> mdz: yes
[11:39] <Nafallo> jdub: can we have the circle to please? the splash and wallpaper rocks after half a year ;-).
[11:39] <mdke> mdz, same, sorry for any trouble
[11:39] <pitti> mdz: although they are pretty uncritical
[11:40] <mdz> pitti: created https://wiki.ubuntu.com/NonLanguagePackTranslationDeadline
[11:42] <jbailey> mdz: Is it too late now?  I thought we had the 29th to do this.
 next time LaschW shows up, somebody point him at 'apt-get source oem-config localechooser kbd-chooser' please
[11:43] <mvo> I have some pending desktop files here, but I'm sort of hesitating to upload e.g. a new xsane because of a single i18n update in the desktop file...what is the opionion in this?
[11:43] <seb128> mvo: I usually do quite a bunch of upload for this reason
[11:43] <LaschW> ogra: Nice hint, but already downloaded :-)
[11:44] <ogra> ok :)
[11:44] <seb128> mvo: but ping me before, I've a list of uploads for a 100% french menu :p
[11:45] <mvo> seb128: can we coordinate that tomorrow then :) ?
[11:45] <LaschW> ogra: I see it as a nono criteria for OEM's not to have a propper documentation for OEM mode install...
[11:45] <seb128> mvo: sure
[11:45] <mvo> seb128: thanks
[11:45] <seb128> np
[11:46] <ogra> LaschW, not my area, i only forwarded Kamions request :)
[11:46] <LaschW> ogra: Are there any people here in Germany who work on an OEM-Install documentation?
[11:47] <mdz> mjg59: any suggestions on the console font issue?
[11:47] <ogra> LaschW, glaub ich nich ;) 
[11:47] <mjg59> mdz: We can't change the console font while in graphics mode. We can either fix the kernel so we can (not practical now), do it in text mode (either before usplash or after usplash) or not set the console font at all
[11:48] <LaschW> ogra: Hhhm, up to now I know about 3 companies who asked about input on breezy OEM-Install...
[11:48] <mdz> mjg59: what we do now is do it after usplash, in gdm, immediately before X starts
[11:48] <mjg59> Yes
[11:48] <mjg59> Which means flicking to a text console briefly
[11:48] <mdz> this seems to cause a noticeable vt switch, though, which would be nice to avoid if possible
[11:49] <mjg59> I can't see any way around that
[11:49] <mjg59> Though it's possible we might be able to get it to switch to a blank console
[11:49] <ogra> LaschW, i havent wokred in ths area at all... but i'm sure Kamion does a documentaion an the translation team will translate it... but there is no specific german documentation writer afaik
[11:51] <mvo> mdz, mjg59: I'll try that again tomorrow, I was experimenting with various ways to avoid the flickering today (without much sucess :/)
[11:52] <MK> anyone: is there a way of "upgrading" to Colony5 from the preview release by using apt/synaptic, or do I have to re-install fully? (in relation to: http://bugzilla.ubuntu.com/show_bug.cgi?id=16244#c6) - if I run apt-get update + apt-get dist-upgrade .. will this do the trick?
[11:52] <LaschW> ogra: We had a guy here in Friesland who has been known as some kind of Mr. Linux. But he says he will not give Linux suport anymore due he has been tricked very badly by some OpenSource sectarians who offered him a deceitful job offer...
[11:52] <mjg59> mvo: Calling the usplash init script is almost certainly the wrong thing to do
[11:53] <mjg59> mvo: Try just chvting to an unused console
[11:53] <mjg59> There'll be flicker, but there won't be any text
[11:53] <ogra> LaschW, thats bad :/
[11:54] <mjg59> mvo: You shouldn't need to wait for usplash to exit. Just do the chvt and it'll release the lock when it's shut down
[11:54] <mvo> mjg59: thanks, will do this
[11:56] <jbailey> mvo: Ooo.  The -15 usplash actually quits properly for me now when I break in the initramfs.  Thanks!
[11:57] <LaschW> ogra: In his position I should have done so likewise. They know about his situation (Hartz-IV) and misused his enthusiasm until he was forced to accept a job as a toilet man. :-((
[11:58] <ajmitch> morning all
[11:58] <ogra> bah
[11:59] <mvo> mjg59: a chvt 2 ; console-screen before gdm starts currently does not set the fonts properly. I'm too tired for looking further now. will check tomorrow