[12:50] <hughsie> PowerManagement guys -- opinions please.
[12:50] <hughsie> in reference to https://wiki.ubuntu.com/PowerManagementConfiguration
[12:50] <hughsie> the battery critical sliders are per-time, rather than per-percent.
[12:51] <hughsie> now I've got a bit of time, i've been fixing the issues ubuntu has with g-p-m to more match this mockup
[12:51] <hughsie> but, i need some rational about the per-time thing
[12:54] <ogra_ibook> hughsie, i guess nobody of the guys that have written this spec is around ...
[12:54] <ogra_ibook> (and i disagree with it to be honest, i think g-p-m is fine as is)
[12:55] <hughsie> ogra, cool, cheers!
[12:55] <hughsie> ogra_ibook, you mind if I scribble on your wiki?
[12:55] <ogra_ibook> put a comments section at the bottom and make your suggestions there
[12:56] <ogra_ibook> but in the end its mjg59 who will implement it and i think he'll take g-p-m as well as it is now ;)
[12:57] <hughsie> ogra_ibook, I figured if you really dont likethe hal actions thing, you can just wrap g-p-m around PowerManager init.d service like i did before hal had actions support
[12:57] <hughsie> ogra_ibook, but that's hassle :-)
[12:57] <ogra_ibook> gah, got a horrible lag here
[12:58] <hughsie> ogra_ibook, You should see the mount(), unmount(), eject() methods in HAL CVS.
[12:58] <ogra_ibook> mjg59 cares for this part, i think he'll just write a dbus service
[12:58] <ogra_ibook> at least it sounded like it ...
[12:58] <hughsie> they rock for fedora (as gnome can just call a dbus method) but I'm not sure how easily pmount and the ubuntu stuff will work.
[12:59] <hughsie> he can use PowerManager as a start, or just do a quick daemon in python or something
[12:59] <ogra_ibook> currently he calls the gdm functions directly, which works like a charm ...
[12:59] <ogra_ibook> but wont match the security requirements i guess
[12:59] <hughsie> yes, i guess.
[01:00] <ogra_ibook> i'd love to have it like it is now ...
[01:00] <hughsie> :-), well, wait and see
[01:00] <hughsie> you been following g-p-m cvs at all?
[01:00] <ogra_ibook> but somehow there were some people that didnt like the implementation at all and wrote this spec
[01:00] <ogra_ibook> only the releases ... my focus moved to ltsp ...
[01:01] <hughsie> I think Corey Burger has some problems with some stuff, but Jaap sent a few patches to make him happier.
[01:01] <hughsie> Like the Shutdown button now brings up the logout dialogue
[01:01] <hughsie> rather than just blindly shutting down the system
[01:01] <ogra_ibook> yes, Burgwork is a hard bone if it comes to usability, he's a perfectionist ;)
[01:02] <ogra_ibook> cool
[01:02] <hughsie> Yes, he's cool.
[01:02] <hughsie> Exams finsihed last week, and so far I've been committing patches like ther eis no tmw.
[01:02] <hughsie> today I've made 11 commits to gnome cvs
[01:02] <ogra_ibook> heh
[01:03] <hughsie> how's work?
[01:03] <mjg59> hughsie: Hi
[01:03] <ogra_ibook> cool
[01:03] <ogra_ibook> i just get a nightly meal ...
[01:03] <schlomo> Hi
[01:03] <schlomo> sorry to bother
[01:03] <hughsie> mjg59, hi, you got a min?
[01:03] <ogra_ibook> bbl
[01:03] <mjg59> hughsie: Basically, we're going to be using g-p-m as is
[01:03] <ogra_ibook> yay
[01:03] <hughsie> mjg59, with the hal methods?
[01:03] <mjg59> hughsie: I've hacked dbus lightly to get the semantics right for our security model
[01:03] <schlomo> which version has been used to compile kernel 2.6.12.10-386 ?
[01:03] <mjg59> hughsie: Yup
[01:03] <hughsie> mjg59, *sweet*
[01:04] <schlomo> I have to recompile alsa driver for snd-hda-intel
[01:04] <hughsie> mjg59, there's lots of source tidy ups and code reduction in CVS.
[01:04] <hughsie> but CVG g-p-m depends on CVS hal...
[01:04] <hughsie> :-)
[01:04] <schlomo> any idea ?
[01:04] <mjg59> hughsie: I've got a pam module that creates a lockfile when a user logs in, which includes the console
[01:05] <hughsie> mjg59, like the rh at_console thing?
[01:05] <schlomo> don't tell me gcc-3.5 because it's virtual package right now :)
[01:05] <mjg59> hughsie: It's the at_console thing, except we check that the user is at the foreground console
[01:05] <mjg59> Rather than just at /a/ console
[01:05] <schlomo> please any tips ?
[01:05] <hughsie> mjg59, so better than at_console :-)
[01:05] <mjg59> So if there are two users at the console, only the current one can trigger suspends and play with the network and so on
[01:06] <mjg59> hughsie: Yes :)
[01:06] <hughsie> mjg59, makes sense
[01:06] <hughsie> i ight steal the idea and bugzilla rh about it.. :-)
[01:06] <hughsie> mjg59, I think i've solved the "gpm won't run at the login screen" problem
[01:07] <schlomo> hum
[01:07] <mjg59> hughsie: So the only thing I need to add now is a suid app that hal can call
[01:07] <schlomo> thanks anyway
[01:07] <mjg59> hughsie: Oh, coo
[01:07] <schlomo> :)
[01:07] <hughsie> using dbus activation, one instance of g-p-m can replace another.
[01:07] <hughsie> and one instance can be gnome-power-console and have no GUI.
[01:07] <mjg59> schlomo: cat /proc/version
[01:07] <mjg59> hughsie: Ah, excellent
[01:07] <hughsie> mjg59, that sound likea plan?
[01:07] <hughsie> mjg59, one problem tho,
[01:08] <mjg59> hughsie: Yes
[01:08] <schlomo> gcc version 3.4.5 20050809  :)
[01:08] <schlomo> mjg59 : thanks for the tips ;-)
[01:08] <hughsie> mjg59, g-p-m has to use the system dbus connection, not the session connection
[01:08] <mjg59> hughsie: Ah
[01:08] <hughsie> as session gets created and destroyed all the time with login and logout
[01:08] <mjg59> Yes
[01:08] <mjg59> Hmm
[01:08] <mjg59> We want to be able to have more than one g-p-m running, though
[01:09] <hughsie> no big problem, same shit, different connection that's all
[01:09] <hughsie> mjg59, what for?
[01:09] <mjg59> Since more than one user may have a session
[01:09] <mjg59> Or are we back to a system-wide g-p-m now?
[01:09] <hughsie> mjg59, no -- :-)
[01:10] <mjg59> With fast-user-switching, there may be two or more users with gnome sessions at once
[01:10] <hughsie> mjg59, when the user is no longer at_console, the instance of g-p-m attached to that console quits, and the at_console instance gets started
[01:10] <hughsie> but with fusa, only one should be at_console, no?
[01:10] <mjg59> hughsie: With RH, multiple people can be at_console
[01:10] <mjg59> With us, only one
[01:10] <hughsie> mjg59, yes, but that is busted and wrong in my opinion
[01:10] <hughsie> the rh case that is
[01:11] <mjg59> But users can switch just by hitting ctrl+alt+F?
[01:11] <mjg59> How do you catch that?
[01:11] <schlomo> configure: error: You have built-in ALSA in your kernel. ?
[01:11] <schlomo> I have to get source ?
[01:12] <schlomo> why I can compile a module ?
[01:12] <mjg59> schlomo: What are you trying to do?
[01:12] <schlomo> on Hoary kernel I can
[01:12] <mjg59> schlomo: You should probably be asking this on #ubuntu
[01:12] <hughsie> mjg59, not sure of the details at the moment, i'm waiting for somebody to fix https://bugs.freedesktop.org/show_bug.cgi?id=4637
[01:12] <schlomo> sorry
[01:12] <hughsie> so I can test it myself and get all the corner cases out
[01:12] <schlomo> Ihave to recompile alsa module for snd-hda-intel
[01:12] <hughsie> but at the moment session/system dbus is one #define change ;-)
[01:13] <schlomo> on 2.6.12.10-i386
[01:13] <crimsun> schlomo: ask in #ubuntu, I'll help you. This is the wrong channel.
[01:13] <mjg59> hughsie: Ok, cool
[01:13] <schlomo> ok sorry
[01:14] <hughsie> mjg59, whats the release date for dapper?
[01:14] <mjg59> hughsie: April
[01:15] <hughsie> no problem. new hal is expected in jan, so my big release is going to be a bit after that
[01:15] <hughsie> Assuming fd#4637 gets fixed in time....
[01:18] <hughsie> mjg59, one last question (then i sleep) for the critical low sliders, currenly per-percent, in the discussion with Corey, he wanted then per-time (i.e. 5 minutes remaining, rather than 5% remaining) -- what's your take on this?
[01:18] <mjg59> hughsie: I think per-time is sensible
[01:18] <mjg59> But I think I've said that before :)
[01:19] <hughsie> mjg59, i'm thinking so at the moment, but some batteries give wildly inaccurate times at low charges... is that not a worry?
[01:19] <mjg59> hughsie: Hmm. How is the time less accurate than the percentage?
[01:20] <hughsie> mjg59, if the resolution is say 2 minutes, and the battery dischange is non-linear (old/knackered batteries) then the time goes completely non-linear less than about 10 minutes
[01:21] <hughsie> my new laptop battery is okay, but my old battery is accurate for the first 30 minutes, then exponentially crashes!
[01:21] <mjg59> hughsie: Oh - so when we say "10 minutes" that might be entirely inaccurate? Hrm.
[01:21] <hughsie> yes
[01:21] <mjg59> I wonder if hal ought to learn
[01:21] <hughsie> i normally hibernate at 1% (critical low)
[01:21] <mjg59> Since spec-compliant batteries provide a model number and serial number
[01:21] <hughsie> but one minute is not accurate enough at all.
[01:22] <hughsie> mjg59, my batteries only differ by age.
[01:22] <mjg59> hughsie: The serial number is identical?
[01:22] <mjg59> Tch.
[01:22] <hughsie> you can do some fancystuff in hal for non-linear discharges, but then you need a backing store for the data
[01:22] <hughsie> not serial no, but model etc
[01:22] <mjg59> Ok. Percentage based, but set the default based on the battery information?
[01:22] <mjg59> hughsie: Right, that was what I meant
[01:23] <mjg59> hughsie: hal could record the curve for individual batteries
[01:23] <mjg59> That's quite advanced, though :)
[01:23] <hughsie> mjg59, yes, but that requires a backing store for about 16k of data. I've thought about this one before
[01:23] <mjg59> hughsie: Yeah. We have no mechanism for that now?
[01:24] <hughsie> the percentage change should be set on the capacity, you're right.
[01:24] <hughsie> i'll do that.
[01:24] <hughsie> at the moment, g-p-p gives a "esimated time value" in small italics under the percentage slider.
[01:24] <mjg59> Ok, that sounds good
[01:24] <hughsie> you can see it in the glade file, but it only works (obvioulsy) when discharging.
[01:25] <mjg59> As long as it doesn't default to warning me at a ridiculous time when I've got a big battery :)
[01:25] <hughsie> yes, i'll fix that. defaults to 5% at the moment
[01:27] <hughsie> mjg59, okay, thanks for the input. I sleep now. Happy Christmas all you Ubuntu'ers!
[01:27] <mjg59> hughsie: Happy christmas!
[02:36] <schlomo> Hi
[02:36] <schlomo> I have take some quotes from an article
[02:36] <raphink> yop schlomo
[02:36] <raphink> faut dormir il est tard ;)
[02:37] <schlomo> about friendly documentation 
[02:37] <schlomo> and support options
[02:37] <schlomo> yep
[02:37] <schlomo> I think it very interesting
[02:37] <schlomo> http://paste.ubuntu-nl.org/6100
[02:37] <schlomo> just 3 rows
[02:37] <schlomo> can you read please
[02:38] <schlomo> does launchpad.net has been done for that purpose ?
[02:38] <raphink> where is that from ?
[02:38] <raphink> schlomo: demande sur #launchpad si tu veux parler de launchpad c'est mieux, c'est fait pour :)
[02:38] <schlomo> ok :)
[02:39] <schlomo> ACM
[04:20] <lunitik> Is there any work around for the locales issue?  (bugzilla #21436)  - I'd ask in #ubuntu but I'm banned....
[04:22] <lunitik> k... thanks for your help  :/
[04:38] <Amaranth> liar, he isn't in the ban list
[07:45] <pef> hello
[11:41] <pvanhoof> I recently upgraded to ubuntu dapper, but list my cute X11 mouse cursor
[11:41] <pvanhoof> how do I regain it? :)
[11:41] <pvanhoof> s/list/lost
[11:41] <pvanhoof> I'm addicted to that cute thing!
[11:42] <pvanhoof> oh, and there's some problems with the cups packages. Some rc.invoke script fails
[11:42] <pvanhoof> and the latest kernel has it's include files installed wrongfully. But I simply installed the kernel source package, changed the version in version.h and used that to build my VMWare modules (which works)
[11:43] <pvanhoof> would be nice if it would get fixed ;0
[11:43] <pvanhoof> ;)
[11:57] <mpt> pvanhoof, this is a developer channel, and most of the developers are on holiday
[11:57] <mpt> Besides, dapper is not an "upgrade" as long as it's not finished :-)
[11:58] <pvanhoof> ok :)
[11:58] <mpt> You'd probably have better luck reporting bugs in bugzilla.ubuntu.com
[11:58] <pvanhoof> yeah, was going to do that
[11:59] <Pygi> pvanhoof: after all, you can always try to uninstall it first. then install it
[11:59] <pvanhoof> oh, but I just installed it. Why would I want to uninstall it? :)
[11:59] <Pygi> I meant the mouse cursor :)
[12:00] <pvanhoof> ah, is it a package?
[12:00] <pvanhoof> that might have slipped during the install then
[12:00] <pvanhoof> err, upgrade
[12:01] <Pygi> heh :)
[12:01] <Pygi> try do that, it may fix it :)
[12:01] <pvanhoof> which package?
[12:05] <Pygi> try xcursorgen with all it's dependencies
[12:16] <Pygi> wb
[12:19] <Yagisan> Is there some lag on the mirrors atm ? I received  pitti's notice about new kernels 2 days ago, and still don't see them @ au.archive.ubuntu.com
[12:19] <jbailey> Yagisan: I would expect them to be on security.ubuntu.com
[12:21] <Yagisan> jbailey: by default apt was set up with a  breezy-security line and my "local" mirror
[12:22] <Yagisan> I'd assume that should have taken care of it
[12:23] <jbailey> *shrug* I don't have good answers, only answer.
[12:25] <Yagisan> jbailey: thanks anyway, being less the 2 hours until christmas, I didn't expect anyone to answer
[03:25] <zul> merry christmas
[03:26] <zakame> merry christmas too zul :) 'tis only an hour and a half away here :)
[03:26] <zakame> in .ph
[03:26] <zul> cool
[03:37] <zakame> hi smurf 
[03:39] <HiddenWolf> jdub, nice job bringing reason to the spatial debate on gnome-desktop-devel. Thumbs up.
[03:52] <alejandro> hi
[03:57] <mhz> hi alejandro 
[05:17] <lamont> configure.in:220: error: possibly undefined macro: AM_PROG_LIBTOOL
[05:17] <lamont> poor dbus
[05:17] <BenC> anyone with a UP amd64 want to test a kernel for me?
[05:18] <lamont> BenC: I have 32-bit user space, but I could test it later today if you want
[05:18] <BenC> lamont: http://people.ubuntu.com/~bcollins/kernels/amd64-alternative-smp/
[05:18] <lamont> I'm assuming that you'reconfident it won't trash my filesystem or anything like that...
[05:18] <BenC> it's an SMP kernel
[05:18] <BenC> yes, all should be safe
[05:18] <BenC> it will either crash or boot :)
[05:19] <lamont> model name      : AMD Athlon(tm) 64 Processor 3500+
[05:19] <lamont> which beast do I want for that?
[05:19] <BenC> after you boot it do "dmesg | grep SMP" and see if you see a message like "SMP alternatives: switching to UP code"
[05:19] <BenC> -k8-smp
[05:20] <lamont> fetching now - holler if someone beats me to testing it..
[05:20] <BenC> the more testers the better
[05:20] <lamont> otherwise, I'll munge it and install once I get back from running to town and such.
[05:20] <lamont> ok
[05:20] <lamont> I'll plan on testing in any case then
[05:20] <BenC> no one has tested it yet
[05:20] <BenC> thanks
[05:21] <lamont> of course, if it comes up and tells me I have 2 processors, that'll be cool too.:-)
[05:40] <lamont> dpkg-deb: building package `linux-image-2.6.15-10-amd64-k8-smp' in `../linux-image-2.6.15-10-amd64-k8-smp_2.6.15-10.15_i386.deb'.
[05:43] <neuralis> BenC: ping
[05:49] <tseng> lamont: is there a default -j to MAKEOPTS in the builders?
[05:49] <tseng> lamont: > 1
[06:06] <BenC> neuralis: pong
[06:07] <neuralis> BenC: can you take a look at wiki/ServerTesting (bottom of page) and see what else you'd like on there?
[06:07] <neuralis> i wrote that in 10 minutes at 4am, so i quite likely missed several things
[06:09] <BenC> reading...
[06:13] <BenC> neuralis: When reporting bugs, tell them to make sure to include atleast dmesg, and also check https://wiki.ubuntu.com/DebuggingProcedures for other things they can do prior to submitting the bug report
[06:15] <neuralis> will do. anything else they should be testing?
[06:18] <tseng> lamont: also, im told mvo can not build mono in even a breezy chroot on ppc buildd, which built fine before.. is something amiss with the system?
[06:19] <tseng> lamont: its not reproducable on similar systems
[06:20] <slomo> lamont: and it does also happen now for versions that worked before on that machine
[06:32] <BenC> neuralis: that's pretty close to it
[06:53] <decklin> where is the 'component' box in bugzilla? (when entering a new report)
[07:00] <slomo> decklin: for new bugreports you should use launchpad.net/malone
[07:01] <decklin> oh, ok. thank you.
[07:09] <decklin> argh. ok, i created a bugzilla account 10 minutes ago, and that generated password does not work on malone, but malone also says my email is already registered.
[07:11] <Amaranth> decklin: your wiki account should be your launchpad account
[07:11] <Amaranth> slomo: all new bugs are going to malone now?
[07:12] <slomo> Amaranth: afaik yes
[07:12] <decklin> Amaranth: i'm pretty sure i don't have a wiki account.
[07:12] <decklin> this is for an address i use on Debian packages, if that matters
[07:18] <decklin> (well, a reset worked. ok, sorry to bother everyone.)
[07:24] <\sh> evening
[08:07] <lamont> tseng: no.  gcc has certain options (-pipe) forced, and defaults certain others (but not on ppc)
[08:07] <lamont> tseng: what did change is possibly ppc vs ppc64 on the base system
[08:08] <lamont> there could also be toolchain diffs between when it built and when it shipped.
[08:09] <lamont> BenC: hrm... that new kernel requires a dapper initramfs-tools... what else does it need from dapper?
[08:10] <\sh_away> merry christmas to all of you :)
[08:23] <xhaker> \sh, to you too
[08:51] <siretart> merry xmas to you all!
[08:51] <siretart> does anyone happen to have an older version of xserver-xorg-driver-ati than xserver-xorg-driver-ati_6.5.7.2-0ubuntu1_i386?
[08:55] <\sh> whiprush: your christmas present is on its way :)
[08:55] <whiprush> heh
[08:56] <\sh> ah well...dapper-changes message is already there :) 
[08:58] <whiprush> heh
[08:58] <whiprush> neato, thanks!
[08:59] <\sh> ok..guys ... have some nice presents...nice celebrations :) cu later these days :)
[09:17] <xhaker> siretart, i got xserver-xorg-driver-ati_6.5.7-0ubuntu3_i386.deb
[10:12] <tseng> lamont: hm i ruled out toolchain on account of it failing the same on breezy chroot
[10:12] <tseng> lamont: and merry christmas, you have the patience of a saint
[10:12] <tseng> :P
[10:22] <BenC> lamont: probably udev
[10:34] <hub> the locales package does not want to install 
[10:34] <hub> it is a dapper upgrade from flight CD 2
[10:35] <hub> any take?
[10:37] <tseng> dist-upgrade agaain
[10:37] <slomo> yes... a second try fixed it for me too ;)
[10:38] <hub> I tried it several time
[10:38] <hub> I'll try again
[10:38] <hub> still the same
[10:38] <hub> I tried 2 times in a row
[10:38] <hub> all it says is
[10:38] <hub> dpkg: error processing /var/cache/apt/archives/locales_2.3.7-1_all.deb (--unpack):
[10:38] <hub>  subprocess pre-installation script returned error exit status 1
[10:38] <hub> Errors were encountered while processing:
[10:39] <Sepheebear> its the preinst script
[10:39] <Sepheebear> it doesnt check for the target dir before trying to write a file
[10:39] <hub> oh
[10:39] <hub> so?
[10:39] <hub> is there a workaround?
[10:40] <tseng> (make the directory?)
[10:40] <hub> mpitt email on ubuntu announce was not really clear
[10:40] <hub> which directory?
[10:40] <Sepheebear> yeah: sudo mkdir -p /var/lib/locales/supported.d
[10:40] <Sepheebear> then it goes smoothly again
[10:40] <hub> still
[10:40] <tseng> its not clear because its only a problem with dapper -> dapper
[10:41] <tseng> which isnt supported
[10:41] <hub> yeah I knoe
[10:41] <Sepheebear> bugzilla #21419
[10:42] <hub> sure
[10:42] <hub> but the mkdir does not do anything
[10:42] <hub> the directory already did exists
[10:44] <Sepheebear> check the output of "dpkg -i /var/cache/apt/archives/locales_2.3.7-1_all.deb"
[11:21] <Gatorade> I have a problem with my live cd can anyone help?
[11:21] <Gatorade> When I load the cd my mouse does not respon
[11:21] <Gatorade> repond*
[11:22] <Gatorade> and the optical lights inside don't work
[11:24] <Gatorade> help
[11:58] <JanC> Gatorade: this is a channel for the developers to work together, try asking on #ubuntu