[12:00] <seb128> hum
[12:00] <seb128> mdz: booting in single user gives a root login without any password according to SurcouF 
[12:00] <seb128> mdz: is that expected ?
[12:00] <SurcouF> yep
[12:01] <mdz> seb128: yes
[12:02] <seb128> hum
[12:02] <SurcouF> normal
[12:04] <seb128> mdz: so if somebody turn my box on, boot on single user he gets a root session ... and that's not a problem ?
[12:04] <mdz> seb128: no, it isn't
[12:04] <mdz> seb128: if he has access to the console, you lose
[12:04] <mdz> he could just as easily use init=/bin/sh
[12:04] <mdz> or boot from CD
[12:05] <seb128> true ...
[12:05] <SurcouF> yep
[12:07] <doko> keybuk: thanks for the explanation!
[12:07] <mdz> Mithrandir: night
[12:08] <doko> mithrandir: good night, so the gXX-3.4 dependencies should be no problem
[12:09] <Keybuk> well, as much as << dependencies have always been useless, sure :p
[12:12] <Keybuk> iirc. versioned conflicts have the same problem, you can always stick the conflicting version in afterwards
[12:14] <mdz> fabbione: hmm, my uxterm colours (for ls and such) are now different in unstable
[12:18] <doko> mdz: btw, do we have something faster than gnome-terminal for desktop?
[12:18] <Keybuk> what's wrong with gnome-terminal?
[12:23] <doko> keybuk: try compiling gcc in xterm and in gnome-terminal. it's slower in gnome-terminal.
[12:24] <Keybuk> yes, iirc. vte deliberately slows down scrolling so you can read it
[12:25] <doko> not something that our target user will notice, but it's a bit annoying.
[12:27] <Kamion> current daily looks good on i386, ok on powerpc except for a few glitches
[12:28] <Kamion> Keybuk: whoa, that's a stupid idea
[12:28] <Kamion> doko: xterm's in desktop ...
[12:29] <doko> kamion: but not accessible from the menu by default.
[12:29] <Kamion> true
[12:30] <Keybuk> people who want it know where they can get it
[12:30] <Keybuk> confusing users with "Terminal", "X Terminal", "Unicode X Terminal" is bad
[12:30] <Keybuk> cf. "Another Clock"
[12:32] <glyph> IMHO cryptic options should look cryptic
[12:32] <Keybuk> no, they should look invisible
[12:32] <doko> sure, but maybe that could be addressed for hoary. we have an alternative for a web browser as well. Simply add/substitute the things from gnome, which are not good enough. it doesn't have to be xterm, maybe something else, as long as it's not kterm.
[12:33] <glyph> Keybuk: novices will ignore things they don't understand; if you have an "advanced" menu that says "uxterm aterm rxvt Eterm", nobody is going to expect those words to open a word processor :)
[12:33] <Keybuk> Advanced menus are evil
[12:34] <Kamion> why?
[12:34] <Keybuk> "Here's a set of things that are different, but we're not telling you how"
[12:34] <mdz> xterm is in desktop, rxvt is in universe
[12:34] <mdz> I don't think they should be in the menu
[12:35] <doko> mdz: yes, there should only one terminal emulator in the menu, but gnome-terminal doesn't seem to be the best choice.
[12:35] <Keybuk> if a user really wants eterm, they'll download and install it themselves and set up their own launcher for it
[12:35] <Keybuk> doko: it's the only one that fits in with the rest of the desktop
[12:36] <mdz> doko: I find gnome-terminal to be a nice balance of unicode support, configurability and usability
[12:36] <mdz> it is a much better choice for the menu than, e.g. xterm or rxvt
[12:36] <Keybuk> uxterm is slower than gnome-terminal isn't it, anyway?
[12:36] <jdub> ahr
[12:36] <Kamion> getting gnome-terminal to use unicode is pain
[12:36] <mdz> Keybuk: no
[12:36] <jdub> Keybuk: especially not if you're using screen ;)
[12:36] <mdz> Kamion: eh?  it seems to do utf-8 by default
[12:36] <Kamion> you have to not only start it in a Unicode locale but also export LANG=whatever once you're inside the new terminal window
[12:36] <Keybuk> Kamion: huh?  it uses it by default?
[12:37] <Kamion> if you LANG=en_GB.UTF-8 xterm then it passes that through to the locale like it's supposed to
[12:37] <doko> mdz: I don't argue about the usability of gnome-terminal, but it's way too slooow.
[12:37] <Kamion> I suspect gnome-terminal defaults to a login shell or something
[12:37] <Keybuk> well, obviously LANG has to be exported, otherwise apps don't know to output utf-8 characters
[12:37] <jdub> doko: that can be fixed, and people who need a fast terminal can fall back on xterm
[12:37] <Kamion> Keybuk: dude, LANG=en_GB.UTF-8 gnome-terminal starts up a terminal in en_GB
[12:37] <Keybuk> Kamion: What is the "Run command as a login shell" option set to?
[12:37] <mdz> I don't think it defaults to a login shell; I remember having to enable that
[12:37] <Kamion> I'm running this from another terminal
[12:38] <mdz> because dammit, it's the right thing to do
[12:38] <Kamion> preferences should be irrelevant
[12:38] <Keybuk> descent scott% export LANG=C
[12:38] <Keybuk> descent scott% LANG=en_GB.UTF-8 gnome-terminal
[12:38] <Keybuk> --
[12:38] <Keybuk> descent scott% echo $LANG
[12:38] <Keybuk> en_GB.UTF-8
[12:38] <jdub> mdz: it is? i always wondered about that - does that make .bash_profile do the right thing?
[12:38] <Kamion> fresh warty installation, I have no idea about descent
[12:38] <doko> keybuk: it runs bash --login
[12:38] <Keybuk> are you sure you're actually running gnome-terminal for the first time?
[12:38] <Keybuk> if you had one open as LANG=en_GB ... that is the process that'll open your new window
[12:39] <Keybuk> (compare gnome-terminal --disable-factory)
[12:39] <Kamion> Keybuk: I started one, then immediately in that terminal ran 'LANG=en_GB.UTF-8 gnome-terminal'. it did not work properly.
[12:39] <Kamion> xterm does.
[12:39] <Keybuk> ah, yes
[12:39] <Keybuk> that won't work
[12:39] <mdz> jdub: yes
[12:39] <Kamion> why on earth not?
[12:39] <Keybuk> you only have one gnome-terminal process
[12:39] <Kamion> oh, *that* crap
[12:39] <Keybuk> the one you ran with LANG=en_GB.UTF-8 just kicked the activation server to tell the running process to open a new window
[12:40] <Kamion> I guess that explains it, but I think it's a fundamental misdesign
[12:40] <Clint> does that mean you can't have multiple tabs with different locales?
[12:40] <Keybuk> Clint: gnome-terminal --disable-factory on the one you want with different settings
[12:40] <mdz> sure you can
[12:40] <mdz> Clint: you just have to set it after you open the terminal
[12:40] <Keybuk> or indeed, different profiles with custom commands, etc.
[12:41] <mdz> running 'gnome-terminal' when there is only one open, opens a new window in the running gnome-terminal process (like mozilla)
[12:41] <mdz> which is actually quite nice, considering the amount of memory that gnome-terminal would require otherwise
[12:41] <Clint> so "LANG=en_GB.UTF8 gnome-terminal" could be made to work
[12:41] <Keybuk> Clint: "LANG=en_GB.UTF-8 gnome-terminal --disable-factory"
[12:42] <Keybuk> will start a new gnome-terminal process
[12:42] <mdz> at the cost of some memory
[12:42] <Kamion> mdz: I tried changing the locale in a new tab, but it doesn't seem to display Unicode characters correctly when I do that
[12:43] <Kamion> tested with "printf '\xa3' | iconv -f ISO-8859-1 -t UTF-8"
[12:43] <Kamion> ah, I have to use "Set Character Encoding"
[12:43] <mdz> Kamion: hmm, I thought there was a profile setting for that, but I guess it sets it based on your locale when you start it
[12:43] <mdz> ah, there it is
[12:44] <Kamion> that's per-tab
[12:44] <mdz> not in the profile, but a menu
[12:44] <Kamion> could probably put it in a profile I guess
[12:45] <mdz> Kamion: any remaining external blockers for sounder 8?
[12:45] <Kamion> no, I'll release it once I reboot
[12:48] <seb128> Kamion: unset LANGUAGE && LANG=en_GB.UTF-8 gnome-terminal ?
[12:48] <seb128> Kamion: LANGUAGE is read before LANG IIRC
[12:49] <seb128> (for GNOME stuff)
[12:49] <Kamion> can't check now I'm afraid, I'll have a look next time
[12:49] <seb128> ok
[12:49] <Kamion> mdz: there's another amusing initscript bug though
[12:51] <mdz> Kamion: *groan*
[12:52] <mdz> Kamion: did you file aa bug and/or tell npmccallum yet?
[12:52] <mdz> jdub: gdict not good enough? :-P
[12:52] <ClintU> damn, 236MB upgrade
[12:52] <jdub> mdz: we have command line equivalents :-)
[12:53] <jdub> mdz: and i have never managed to get used to gdict
[12:53] <mdz> jdub: just ruffling your GNOME feathers
[12:53] <mdz> jdub: I would be happy to have dict in desktop, except that it would apparently pull in other packages
[12:54] <Kamion> mdz: just filed, #1088
[12:54] <jdub> mdz: netbase, recode?
[12:54] <mdz> jdub: recode, apparently
[12:55] <mdz> netbase is of course in base
[12:55] <jdub> yeah
[12:55] <mdz> Kamion: oh, I already fixed that, didn't I?
[12:55] <Kamion> did you?
[12:55] <mdz> hmm
[12:55] <mdz> it looks like perhaps I did not upload it
[12:55] <mdz> and deleted it
[12:55] <mdz> I remember doing it
[12:56] <mdz> that was the reason why therm-windtunnel wasn't getting loaded on my ppc
[12:56] <mdz> it was bailing out after ide-generic loaded
[12:56] <mdz> but I certainly don't see the fix in the archive
[12:57] <Kamion> ide-cd and something else were casualties here
[12:57] <Kamion> oh god, now I have to come up with release notes
[12:58] <mdz> Kamion: I'm available to give amd64 a whirl before you bless it, if you're ready
[12:58] <Kamion> go for it
[12:58] <Kamion> the daily's up
[01:00] <ClintU> I suggest updating xchat.
[01:00] <mdz> rsyncing
[01:00] <mdz> ClintU: RC bugs?
[01:01] <mdz>    490483712 100%    4.26MB/s    0:01:49  (3, 44.4% of 9)
[01:01] <mdz> not a bad hit rate from yesterday
[01:01] <ClintU> mdz: issues with crashing and tabs jumping around
[01:02] <ClintU> are 2.6.8.1 ppc images available yet?
[01:02] <mdz> weird, haven't seen that
[01:02] <ClintU> i've only experienced it on one system
[01:02] <mdz> ClintU: yes, as I posted to the sounder list yesterday
[01:02] <ClintU> doh
[01:03] <ClintU> I completely forgot that the name changed.
[01:03] <mdz> we now have all architectures' kernels built from one source package
[01:03] <mdz> and kernel-image-2.6-* metapackages
[01:03] <mdz> er, linux-image
[01:15] <Kamion> mdz: everything's ready on my end, let me know when you're done
[01:15] <jdub> Kamion: the daily ~= sounder8 ?
[01:17] <Kamion> jdub: ==
[01:17] <jdub> cool
[01:17] <jdub> i'm going to have to invert my timezone or something
[01:17] <jdub> i keep doing big downloads on-peak
[01:18] <jdub> oh man
[01:18] <jdub> http://www.smh.com.au/articles/2004/09/07/1094530605483.html
[01:18] <jdub> "Fisher and Hamill as Skywalker and Princess Leia."
[01:18] <jdub> YOU SIPPERS
[01:18] <mdz> Kamion: CD writer is acting up, working on it
[01:26] <mdz> I've tried it in 3 different machines now, and I'm starting to suspect that the drive may be fucked :-/
[01:27] <Kamion> oops
[01:29] <Kamion> mind if I release it anyway? I want to stop work pretty soon
[01:29] <mdz> Kamion: yeah, don't wait for me
[01:29] <mdz> I thought this would be fast
[01:30] <npmccallum> Kamion: module-init-tools bug is fixed, pcmcia fix is on its way
[01:30] <mdz> I hate computers
[01:30] <Hrdwr_BoB> heh
[01:30] <Hrdwr_BoB> pop down the street and buy a new burner
[01:30] <Kamion> npmccallum: great, thanks
[01:31] <Hrdwr_BoB> $36
[01:31] <jdub> Kamion, mdz: did you want to test amd64 exclusively, or...?
[01:31] <mdz> jdub: amd64, I believe, is the architecture that kamion can't test locally
[01:31] <Kamion> jdub: I've only tested i386 and powerpc (the latter with some problems)
[01:31] <jdub> oh, ok
[01:32] <Kamion> reminds me, really must order that system tomorrow, now that they've taken back the amd32 one I ordered by mistake
[01:32] <mdz> Hrdwr_BoB: about twice that; it's DVD, but thanks
[01:32] <jdub> wow, google is six years old
[01:33] <Hrdwr_BoB> ah cool
[01:46] <jdub> wow, the copying stage doesn't suck
[01:46] <jdub> it's way fast
[01:47] <Kamion> that'd be the one-liner find magic
[01:47] <jdub> for the iso sorting, or the actual copying?
[01:47] <Kamion> I was thinking of only copying Desktop and letting Base just install in the old way, since debootstrap is now fast on its own
[01:47] <jdub> (cd is running at a consistent speed)
[01:47] <Kamion> neither, the "get all the directories into the dentry cache first" change to cdrom-detect
[01:47] <jdub> oh
[01:47] <jdub> niiiice
[01:48] <jdub> haha
[01:48] <jdub> that's rad
[01:48] <Kamion> there's ISO sorting as well now but the cdrom-detect change was about an order of magnitude more effective by itself
[01:48] <jdub> oh, you're copying shipseed too, interesting
[01:48] <Kamion> yeah, was unsure about that
[01:48] <Kamion> inclined to turn that off for the preview release
[01:49] <jdub> mdz: copy shipseed?
[01:49] <Kamion> note that at the moment base-config runs 'apt-get clean' after installing Desktop anyway :-)
[01:49] <jdub> heh
[01:49] <jdub> i see ;)
[01:49] <Kamion> that could be changed, of course, but I'm waiting 'til we decide what to copy
[01:50] <ClintU> 2.6.8.1 doesn't work either
[01:50] <Kamion> anyhow, time for !work; eating something might be a good start
[01:50] <Kamion> ClintU: on the blue+white?
[01:50] <ClintU> yup
[01:50] <Kamion> d'oh
[01:51] <ClintU> same initrd panic
[01:51] <Kamion> um, later :-)
[01:51] <ClintU> yup
[01:51] <ClintU> bon appetit
[01:51] <alextreme> hmm, ttf-arabeyes isn't in the archive?
[01:53] <mdz> Kamion: cdrom-detect change?
[01:53] <mdz> ClintU: what panic is that?
[01:53] <mdz> alextreme: hmm, that's strange
[01:53] <alextreme> mdz: (not sure if you got this one earlier) i've set up everything to build a new warty overnight. it still has a 2.6.7 kernel though, would that be an issue?
[01:54] <mdz> alextreme: is it problematic to put 2.6.8.1 on it?
[01:54] <mdz> we need as much testing for the new kernel as we can get
[01:55] <alextreme> well, the kernel would have a different config and be patched, but if you can wait an extra hour or two... :)
[01:55] <mdz> alextreme: for this, I have all night :-)
[01:55] <mdz> alextreme: anything I can do to help?
[01:56] <alextreme> (unfortunately I don't, need to get up in 6 hours)
[01:57] <alextreme> oh yeah, i had a whole list for him, and he splits
[01:57] <ClintU> I was just about to answer his question too.
[01:57] <alextreme> bummer
[01:57] <ClintU> He's a devious one.
[01:58] <alextreme> no shit
[01:58] <alextreme> oh well, back to building, testing, packaging and yelling I guess...
[01:59] <alextreme> dpkg: error processing ubuntu-artwork (--configure):
[01:59] <alextreme>  subprocess post-installation script returned error exit status 1
[01:59] <alextreme> *yell*
[02:00] <ClintU> tsk, tsk, tsk
[02:02] <Uclintu> mdz: #496
[02:02] <jdub> alextreme: what was the error? grub-related?
[02:03] <alextreme> yup
[02:03] <alextreme> hope it sorts itself out, but afaik grub was already installed
[02:06] <mdz> dammit, I think my burner is toast
[02:07] <mdz> that, or it's a problem with 2.6.8.1
[02:08] <Hrdwr_BoB> what is the error?
[02:08] <Hrdwr_BoB> there was a problem with burning CDs with 2.6.8
[02:09] <Uclintu> mdz: #496 is the panic
[02:10] <mdz> Uclintu: curious...what happens if you boot with devfs=mount?
[02:11] <Uclintu> mdz: i'll tell you in a few minutes
[02:11] <jdub> argh, can we get rid of that grub summary?
[02:12] <mdz> jdub: grub summary?
[02:12] <jdub> mdz: boot's much faster ;)
[02:12] <jdub> when the screen clears after the 'press esc to run grub' bit
[02:12] <jdub> it says 'starting ubuntu <kernelversion> '
[02:12] <jdub> (with that annoying space)
[02:12] <jdub> and then prints a summary of grub config
[02:13] <mdz> oh, I believe that's actually grub printing the commands as they're executed
[02:13] <jdub> ahr
[02:16] <jdub> the extracting templates stage -> if there's nothing to preconfigure, can we skip that?
[02:16] <lifeless> jdub: we still on today?
[02:16] <mdz> jdub: BUILTIN_NO_ECHO (for grub)
[02:16] <mdz> jdub: try building with that
[02:17] <jdub> mdz: build var?
[02:17] <jdub> lifeless: hrm, kinda prefer not unless we have an agenda
[02:17] <mdz> jdub: bitfield
[02:18] <jdub> lifeless: going full steam ahead, don't really want much distraction atm (apart from my lappy arriving, but playing with that will also involve serious testing, yes, uh huh)
[02:19] <Keybuk> jdub: not to mention packing it back up in disgust and sending it back
[02:19] <jdub> i've played with one, they're nice
[02:19] <mdz> jdub: what that step is doing is unpacking the bits which decide whether or not there is any preconfiguration to be done
[02:20] <jdub> yes, but we know in advance that there isn't
[02:21] <mdz> jdub: we don't
[02:21] <mdz> xfree86, for example, does its probes to determine whether or not it needs to ask questions
[02:22] <jdub> bum
[02:22] <Uclintu> mdz: no dice
[02:22] <mdz> oh well
[02:23] <Uclintu> it did say "Mounting devfs on /dev" twice though
[02:24] <Uclintu> is there a known XKB problem?
[02:24] <mdz> hmm, writer isn't working any better after going back to 2.6.7
[02:24] <mdz> I will be unhappy if it is toast
[02:25] <npmccallum> mdz: on your thinkpad?
[02:25] <mdz> Uclintu: for macintosh/dvorak, yes
[02:25] <mdz> npmccallum: on every system I have tried today
[02:26] <Uclintu> mdz: i'm not using dvorak; reporting
[02:26] <npmccallum> mdz: I had a strange problem with my writer not working... I wanted to test a clean install and it worked fine after that... it was very strange though
[02:27] <mdz> npmccallum: it keeps claiming buffer underruns, regardless of conditions
[02:27] <mdz> after a SCSI error
[02:27] <mdz> Sense Code: 0x21 Qual 0x02 (invalid address for write) Fru 0x0
[02:27] <mdz> tried different media, different machines, USB2 vs. USB1.1
[02:27] <mdz> different kernels
[02:27] <mdz> the thing worked yesterday
[02:28] <mdz> trying a different cdrecord version now
[02:29] <npmccallum> mdz: is it a scsi drive or atapi?
[02:29] <mdz> npmccallum: an atapi drive in a USB enclosure
[02:29] <mdz> which translates roughly to scsi/usb-storage
[02:29] <npmccallum> right
[02:30] <npmccallum> did you try a 1x burn?
[02:30] <mdz> I have, but it doesn't seem to accept it
[02:30] <mdz> the drive has never really liked having the speed set
[02:30] <mdz> it just goes ahead and burns at whatever speed it wants
[02:30] <npmccallum> ah
[02:30] <mdz> even with driveropts=forcespeed
[02:30] <mdz> I've tried 4x and 16x media
[02:31] <mdz> makes no difference whatsoever
[02:31] <npmccallum> did the speed drop to 4x on the 4x media?
[02:31] <mdz> yes
[02:31] <mdz> the drive chooses reasonable speeds; it just doesn't let me override it if I want something else
[02:32] <elmo> npmccallum: <whine>please do proper changelog entries, i.e. describe what you fixed, not just "fixed #nnnn"</>
[02:32] <Uclintu> are there no CJK input methods in warty?
[02:32] <npmccallum> elmo: ok
[02:33] <mdz> the only other thing I can think of to try would be to rip out the bare drive and use it as ATAPI
[02:34] <mdz> but that would be extremely inconvenient at this time
[02:34] <mdz> and it seems rather unlikely to help
[02:34] <npmccallum> mdz: did you try a new usb cable?
[02:35] <mdz> npmccallum: no, but I will
[02:35] <npmccallum> I'm off to study... Just started a new class today "Attic Greek 1", so I'm off to learn all the eccentricities of a new language
[02:38] <mdz> and if that doesn't help, I'll try 1394
[02:42] <npmccallum> mdz: btw, http://pydirectfb.sourceforge.net/ if we want to do usplash in python and directfb
[02:45] <mdz> same behaviour with 1394.  I think the drive is done
[02:47] <alextreme> mdz: I'm off for the night, I'll be back online tomorrow evening and hope to have something useful by then. if not, thursday is my day off so that's the latest possible
[02:48] <mdz> alextreme: ok, good night, thanks
[02:48] <alextreme> cheers
[02:49] <daniels> fabbione: i'm not doing ati
[02:50] <daniels> mdz: where's it looping?
[02:50] <mdz> daniels: looping?
[02:50] <daniels> oh, duh, it's probably looping at fooish
[02:50] <daniels> mdz: mail
[02:50] <daniels> hm, no, it's accepting that
[02:52] <mdz> oh
[02:52] <mdz> daniels: I forwarded messages to daniels@d.o
[02:52] <mdz> did those get through?
[02:52] <daniels> no
[02:53] <daniels> oh, wait, they'd be in +debian
[02:53] <daniels> i'll tell you in a few minutes, when i've cleared my inbox
[02:54] <mdz> daniels: /msg'd
[02:57] <mdz> this drive reads discs with no problem, but doesn't burn for shit anymore
[02:57] <mdz> that seems like a weird failure condition
[02:59] <daniels> mdz: ifxed, thanks
[03:01] <daniels> bbiab, downloading xfree86 ubuntu14 diff
[03:06] <Kamion> 00:53 < mdz> Kamion: cdrom-detect change?
[03:07] <Kamion> mdz: what exactly did your question mean? :)
[03:11] <mdz> Kamion: you said that your cdrom-detect change had a larger effect on performance
[03:11] <mdz> Kamion: is that where you did the find?
[03:11] <Kamion> yep
[03:12] <Kamion> made sense to do it immediately after mounting the CD-ROM, and doing it there meant that it sped up the "loading installer components from CD" step
[03:12] <Kamion> (the latter's quite a noticeable improvement on my powerbook)
[03:12] <mdz> Kamion: why is archive-copier a net speed loss?
[03:12] <Kamion> mdz: in the first stage
[03:12] <Kamion> mdz: you're doing more work, obviously it's a speed loss :)
[03:12] <mdz> oh, naturally
[03:13] <Kamion> I think it's a net loss anyway, it's doing more I/O by copying to the hard disk
[03:13] <Kamion> 500MB is non-trivial even on a disk
[03:14] <Kamion> and if you're reading the packages while installing them you probably get more benefit from I/O parallelization </handwave>
[03:14] <mdz> hard drive seeks are orders of magnitude faster, though
[03:14] <mdz> it'd be an interesting thing to measure
[03:14] <Kamion> sure, but that doesn't seem to be subjectively noticeable in the tests I've done
[03:14] <mdz> but it's a better experience even if it is a net loss
[03:14] <Kamion> I think we should copy only Desktop packages
[03:15] <Kamion> that way we can do the better-experience thing but with less of a speed hit
[03:15] <Kamion> installing Base without archive-copier enabled seems pretty quick to me following the cdrom-detect change
[03:16] <Kamion> and if we copy less we'll have lower hard disk space requirements
[03:24] <daniels> mdz: if you remove the offending file, and add in a link /etc/X11/xkb/symbols/macintosh/dvorak->/etc/X11/xkb/symbols/dvorak, does that help?
[03:32] <daniels> any mac uers in da house?
[03:32] <daniels> da hizzouse
[03:32] <Uclintu> i'm on a b&w g3 atm
[03:32] <mdz> daniels: doesn't seem to fix it
[03:32] <mdz> oh, sorry
[03:32] <mdz> I misread
[03:32] <daniels> mdz: arse
[03:32] <daniels> mdz	copying the file?
[03:33] <mdz> daniels: I already have a symlink, /etc/X11/xkb/symbols/macintosh/dvorak -> ../dvorak
[03:33] <mdz> which I added myself
[03:33] <mdz> and yes, that fixes it
[03:33] <daniels> mdz: er yeah, that's what I asked you to try
[03:33] <daniels> cheers
[03:33] <mdz> for some reason, having the broken symlink there prevented my fix from working
[03:34] <daniels> yeah, i can see that
[03:36] <daniels> it's pretty typical for that sort of code
[03:36] <daniels> 'let's try to compile all the mac stuff' 'how about no' 'good plan'
[03:40] <jdub> mdz: mouse fix was good
[03:43] <jdub> yeah
[03:43] <jdub> so
[03:43] <jdub> now i want to run ubuntu on my mipsel
[03:44] <jdub> have build farm, will sing
[03:45] <jdub> Kamion: heh
[03:45] <jdub> Kamion: custom install copies all packages ;-)
[03:45] <jdub> Kamion: file bug?
[03:46] <mdz> I _almost_ wish I had a Windows system so I could rule out all remaining software-related possibilities
[03:47] <Kamion> jdub: uh, it's not supposed to do that
[03:47] <Kamion> jdub: yes, that's a bug
[03:47] <Kamion> # In the custom install, only copy Base (this is purely for CD-reading speed).
[03:47] <Kamion> if db_get ubuntu/install-type && "$RET" = custom; then
[03:47] <Kamion>     baseonly=yes
[03:47] <Kamion> else
[03:47] <Kamion>     baseonly=no
[03:47] <Kamion> fi
[03:47] <Kamion> oh
[03:47] <Kamion> me stupid
[03:48] <Kamion> I don't know what kind of shell syntax '"$RET" = custom' (without [ ]  or test) is, but it ain't gonna work
[03:48] <jdub> don't worry about filing? :)
[03:48] <Kamion> go ahead and file anyway, I'm about to go now and won't have time to upload until tomorrow
[03:49] <daniels> OH FUCK X CAN EAT A BAG OF ARSE
[03:49] <jdub> Kamion: ok
[03:49] <lifeless> daniels: welcome to the light
[03:49] <daniels> make[6] : Leaving directory `/home/daniels/canonical/xfree86/xfree86-4.3.0.dfsg.1/build-tree/xc/lib/font/X-TrueType/BIG5'
[03:49] <daniels> cleaning in lib/font/X-TrueType/BIG5HKSCS...
[03:49] <daniels> make[6] : Entering directory `/home/daniels/canonical/xfree86/xfree86-4.3.0.dfsg.1/build-tree/xc/lib/font/X-TrueType/BIG5
[03:49] <daniels> HKSCS'
[03:49] <daniels> make[6] : *** No rule to make target `clean'.  Stop.
[03:49] <daniels> i didn't even touch that part of the world!
[03:50] <Hrdwr_BoB> daniels, my vnc cpu usage issue makes using VNC almost entirely pointless as it grinds the whole thing to a halt - should I file a bug or ...?
[03:50] <daniels> Hrdwr_BoB: tempting to say 'don't do that, then'
[03:50] <daniels> i mean, you can file a bug, but i can't realistically see us fixing it, sorry
[03:50] <Hrdwr_BoB> that's fine for me, I don't really care
[03:51] <daniels> lucky, then ;)
[03:51] <Hrdwr_BoB> haha
[03:51] <Hrdwr_BoB> :p
[03:51] <jdub> Hrdwr_BoB: until we have a DAMAGE-enabled server, you're kinda stuck with it
[03:51] <Hrdwr_BoB> yeah
[03:52] <Hrdwr_BoB> well, I'm not stuck, as I said, doesn't bother me in the slightest
[03:52] <Hrdwr_BoB> but as a 'user'
[03:52] <Hrdwr_BoB> I didn't expect to be unable to use the machine once I VNCed in
[03:52] <daniels> arse!
[03:56] <daniels> HULK SMASSSHHHH
[03:57] <Keybuk> ooh, angry-daniels
[03:58] <jdub> lamont, elmo, Kamion: did you guys get my process mails?
[03:58] <daniels> imake heisenbugs :\
[04:15] <elmo> jdub: process mails?
[04:17] <jdub> elmo: about packages from ftp to the archive?
[04:17] <jdub> if you have no idea, it was probably killed by your spaminator
[04:17] <elmo> jdub: no, don't think so
[04:18] <jdub> ok
[04:18] <jdub> i'
[04:18] <jdub> i'll resend without using my dead domain ;)
[04:19] <jdub> elmo: just bounced it
[04:19] <jdub> elmo: 'Package source to archive walkthrough + timing'
[04:21] <elmo> yeah, got it this time
[04:21] <elmo> you know the normal queue -> archive stuff has little bearing on the actual release mechanics, right?
[04:21] <elmo> but anyway, I'll answer you in email
[04:48] <mdz> ok, either this drive is failing in a _very_ peculiar way, or all my CD-RW media is fucked, or cdrecord and/or cd writing under Linux just blows
[04:48] <mdz> I can burn DVDs fine
[04:49] <mdz> at ~21x CD speed
[04:49] <mdz> and yet cdrecord bitches endlessly about buffer underruns
[05:13] <daniels> mhmmm
[05:13] <daniels> can someone please try building xfree86 4.3.0.dfsg.1-6ubuntu15? sources are on chinstrap:~daniels
[05:13] <daniels> i can't get x to build here for love nor money
[05:17] <Keybuk> daniels: I'll do it for love and money
[05:18] <jdub> wow-wow-wacka-wacka
[05:21] <daniels> i'll buy you a beer
[05:21] <daniels> wrong dan :P
[05:21] <Keybuk> since when do you buy people drinks? :)
[05:21] <daniels> i've bought a few people drinks
[05:32] <daniels> Keybuk:  it got past makeafle,you won
[05:32] <daniels> gah!
[05:32] <daniels> if it got past making makefiles, you won
[05:37] <jdub> is the sources.list typo known in sounder8?
[05:37] <jdub> no slash on the universe lines
[05:38] <mdz> no, I didn't notice
[05:38] <mdz> I only used the main/restricted lines
[05:38] <jdub> what should that be filed against?
[05:38] <jdub> base-config?
[05:38] <tvon|x31> eh?
[05:38] <tvon|x31> where is a slash missing?
[05:38] <jdub> deb http://ftp.no-name-yet.comno-name-yet warty main restricted universe
[05:39] <mdz> between 'com' and 'no-name-yet'
[05:39] <tvon|x31> ah
[05:39] <tvon|x31> heh
[05:39] <justdave> that reminds me, I burned a sounder8 earlier and forgot about it :)  probably still in the burner
[05:40] <tvon|x31> Is my computer the only one that beeps annoyingly when pcmcia starts/stops?
[05:40] <tvon|x31> s/computer/lappy/
[05:41] <jdub> mdz: 1094, not sure who/what it should be on
[05:41] <mdz> jdub: assign it to fabbione
[05:41] <jdub> ohmansudowithnotimeoutissoannoying
[05:41] <mdz> he's doing those changes
[05:41] <mdz> jdub: yep
[05:41] <jdub> mdz: what's his bugzilla login?
[05:42] <mdz> jdub: 'fabbione' in the box will do the job
[05:42] <jdub> oh, handy
[05:45] <bdale> jdub: sudo with timeout is even more annoying...
[05:45] <bdale> jdub: of course, it depends on which timeout we're talking about
[05:46] <elmo> bdale: I think he means zero timeout which ours has/had
[05:46] <mdz> has
[05:46] <mdz> and I really think we should re-enable it
[05:47] <mdz> (the default timeout of 15 minutes or whatever)
[05:48] <bdale> my packaging uses --with-timeout=15 --with-password-timeout=0
[05:50] <mdz> right
[05:51] <mdz> and we add timestamp_timeout=0 to sudoers
[05:51] <daniels> Keybuk: how's the build?
[05:51] <mdz> it seemed harmless enough at the time, but turned out to be very annoying
[05:52] <mdz> jdub: care to propose on the list that we revert it?
[05:52] <mdz> jdub: I'll back you up :-)
[05:52] <mdz> jdub: or, just upload base-config, fixing the slash typo and this at the same time
[05:53] <jdub> that's okay, i'll just be the fallguy asking first, not shooting first
[05:53] <elmo> eh, who asked for it to be 0?
[05:53] <jdub> oh, i'll throw in the abiword/gnumeric thing too
[05:53] <Keybuk> daniels: whirring ... got tripped by the lack of an .orig.tar.gz <g>
[05:54] <daniels> heh :)
[05:54] <daniels> has it passed make Makefiles?
[05:54] <mdz> elmo: colin and martin, as I recall
[05:54] <daniels> for me, it bombs out in xc/lib/font/X-TrueType because imake seemingly doesn't create BIG5HKSCS's Makefile
[05:57] <mdz> bdale: the idea is to add the initial user to sudoers, and use gksudo to let them run GUI admin applications
[05:57] <Keybuk> /usr/bin/make -C build-tree/xc WORLDOPTS="" IMAKE_DEFINES="-DXFree86CustomVersion='\"Ubuntu 4.3.0.dfsg.1-6ubuntu15 20040908035354 scott@netsplit.com\"'" World > logs/make_world.build.log 2>&1
[05:57] <Keybuk> is what it's doing now
[05:58] <daniels> tail -f logs/make_world.build.log?
[05:58] <Keybuk> looks like a make clean
[05:58] <daniels> right
[05:58] <daniels> hopefully that fails
[05:59] <bdale> mdz: oh, I see.  hrm.
[05:59] <Keybuk> I think it's "including" now
[05:59] <bdale> mdz: so you're trying to get one password typing per login session as a behavior?
[05:59] <mdz> bdale: with a password, of course
[06:00] <mdz> bdale: no, I think the argument was that it would be confusing when it prompted for the password sometimes and not other times
[06:00] <daniels> Keybuk: fuck
[06:00] <mdz> there was also a vague security question, but I think I successfully defended that one
[06:00] <bdale> mdz: oh, right, silly me, that makes it always prompt for a password, doesn't it... ick
[06:00] <mdz> bdale: no big deal in the GUI, but very irriatting on the command line
[06:01] <daniels> that's what 'sudo zsh' is for ;)
[06:01] <mdz> that entirely misses the point of sudo
[06:01] <bdale> sudo su -
[06:01] <Keybuk> Writing 192 characters into file 'UTB___12-ISO8859-4.bdf'.
[06:01] <Keybuk> rm -f UTB___14-ISO8859-4.bdf
[06:01] <Keybuk> perl  ../../../fonts/util/ucs2any.pl UTB___14.bdf ../../../fonts/util/map-ISO8859-4 ISO8859-4
[06:01] <bdale> mdz: true
[06:01] <Keybuk> daniels: sudo -s
[06:01] <Keybuk> bdale: sudo -s
[06:02] <Uclintu> sudo -s -H, if you're going to miss the point of sudo
[06:02] <Keybuk> Uclintu: how does that miss the point?
[06:02] <mdz> sudo -s without -H gets a bit weird
[06:02] <daniels> keyJ@#$()U*
[06:02] <Keybuk> mdz: I quite like it <g>  it uses the right arch, tla, aptitude, etc. options
[06:03] <Keybuk> vi, emacs, etc. work right
[06:03] <mdz> Keybuk: it also writes files into your homedir AS ROOT :-P
[06:04] <fabbione> morning guys
[06:04] <fabbione> jdub: #1094, cdrom install or netinstall?
[06:04] <fabbione> in the latter i get the /
[06:05] <fabbione> bah i started not to sleep again..
[06:05] <jdub> fabbione: cdrom, custom (probably the same for desktop)
[06:07] <Keybuk> daniels: is building shit now
[06:08] <daniels> Keybuk: sigh, thanks
[06:08] <daniels> you can kill it ifyou like
[06:10] <fabbione> jdub: with the tests i did i always got the /, but i will check again later today
[06:11] <Keybuk> daniels: I've never built X before ... I'm going to leave it running
[06:11] <Keybuk> so one day, I can tell my grandkids, "I built X once"
[06:12] <daniels> heh
[06:12] <jdub> Keybuk: and they'll say, "grandpoppy, we do that every day when our computer boots gentoo 3000"
[06:13] <jamesh> 3000.2, you mean.
[06:16] <Hrdwr_BoB> I hope to god that in 60 years we've built something better than X
[06:17] <jamesh> Hrdwr_BoB: current X is already better than X of 5 years ago.
[06:17] <Hrdwr_BoB> this is true
[06:18] <Hrdwr_BoB> but if in 60 years, we can't come up with something that totally changes computing as we know it, that's pretty poor
[06:18] <jdub> such a crime
[06:19] <Hrdwr_BoB> exactly
[06:19] <Hrdwr_BoB> jdub, though I consider still using base10 a worse crime :/
[06:19] <jamesh> binary could be trivially extended to include an "unknown" state
[06:19] <jamesh> like SQL does.
[06:23] <fabbione> Keybuk: why are you rebuilding X?
[06:24] <Keybuk> fabbione: because daniels smiled sweetly
[06:25] <fabbione> ohhh
[06:39] <daniels> it's OK, I think the problem is a badly screwed-up orig
[06:39] <fabbione> OH GREAT
[06:39] <fabbione> gdm killed X
[06:39] <daniels> fabbione: huzzah
[06:40] <fabbione> a gdm restart killed X and gdm didn't restart
[06:47] <fabbione> daniels: can you reproduce it for me?
[06:47] <fabbione> just do a gdm restart within X
[06:47] <fabbione> and it should kill X
[06:48] <fabbione> jdub: bug fixed
[06:59] <fabbione> daniels: can you send me the diff between ubuntu14 and ubuntu15?
[06:59] <fabbione> daniels: please just keep me in the loop.. it makes my life easier to keep track of the changes
[07:01] <daniels> fabbione: sure
[07:01] <daniels> fabbione: it's pretty small - fix up the link, and add a patch to build XKB with -DNOISY
[07:02] <daniels> so it'll actually log faults
[07:02] <daniels> fabbione: um, I can't reproduce that gdm problem btw
[07:02] <fabbione> i can without any problems on 2 machines
[07:05] <fabbione> well i need to go and talk to the electrician for a rad plant in the new house
[07:05] <fabbione> bbl
[07:07] <daniels> fabbione: http://people.no-name-yet.com/~daniels/xfree86-ubuntu14-to-15.diff
[07:12] <mdz> fabbione: works fine for me
[07:16] <justdave> heh, the wifi status icon looks like it was blatently stolen from a motorola cellphone :)
[07:19] <justdave> nah, it's a little different.  just enough to get away with it :)  looks nice though
[08:29] <daniels> lifeless: why is arch broken !!AGAIN!! for debrix, and can you fix it?
[08:29] <daniels> (committing --patch-14)
[08:29] <daniels> Error calling `vu_lstat' for "/home/daniels/x/debrix/debrix/{arch}/,,inode-sigs/daniel@fooishbar.org--2004%debrix--devel--0.1--patch-8" (No such file or directory)
[08:29] <daniels> PANIC: I/O error
[08:29] <daniels> daniels@nanasawa:~/x/debrix/debrix% ls {arch}/,,inode-sigs 
[08:29] <daniels> daniel@fooishbar.org--2004%debrix--devel--0.1--patch-10  daniel@fooishbar.org--2004%debrix--devel--0.1--patch-13
[08:30] <daniels> daniel@fooishbar.org--2004%debrix--devel--0.1--patch-11  daniel@fooishbar.org--2004%debrix--devel--0.1--patch-14
[08:30] <daniels> daniel@fooishbar.org--2004%debrix--devel--0.1--patch-12  daniel@fooishbar.org--2004%debrix--devel--0.1--patch-9
[08:30] <daniels> daniels@nanasawa:~/x/debrix/debrix% 
[08:30] <lifeless> daniels: give me some context.
[08:31] <lifeless> what command had you just run?
[08:31] <lifeless> do you have a revision library?
[08:31] <lifeless> is this a lical archive or a remote one?
[08:31] <jdub> have you eaten fish today?
[08:31] <lifeless> troll
[08:31] <jdub> * postinst configure most-recently-configured-version 
[08:32] <jdub> ^ that being the one before the currently-being-configured-version?
[08:32] <lifeless> daniels: and where you doing two things to the same working dir from different shells ?
[08:34] <mdz> jdub: correct
[08:34] <pitti> morning
[08:34] <mdz> jdub: on an upgrade, it'll be the version being upgraded from
[08:34] <mdz> pitti: morning
[08:34] <jdub> ah, good
[08:42] <jdub> ooh, new folder icons are nice
[08:43] <mdz> night
[08:43] <jdub> 'nacht mdz
[08:44] <pitti> night!
[08:50] <daniels> lifeless: tla commit -s '...', yes, local, used to live on my laptop, but rsync -a'ed across
[08:50] <daniels> mdz: you can't sleep yet, i was going to ask you if we had any plans to support em64t
[08:52] <lifeless> daniels: try again
[08:53] <jamesh> daniels: buy an EM64T and see if it works ...
[08:53] <daniels> lifeless: now patch 9 isn't found; should i just hit this commit in a for loop?
[08:53] <daniels> oh fuck
[08:54] <daniels> the commit succeeded, despite the 'OH SHIT SON THATS AN ERROR' uppercase SCREAMING ERROR MESSAGES at the end of the ATTEMPT TO COMMIT
[08:54] <lifeless> oh, rotfl.
[08:54] <daniels> patch-14
[08:54] <daniels>     merge from js bringing in new X.Org and DMX sync
[08:54] <daniels> patch-15
[08:54] <daniels>     merge from js bringing in new X.Org and DMX sync
[08:54] <daniels> patch-16
[08:54] <daniels>     merge from js bringing in new X.Org and DMX sync
[08:54] <lifeless> what tla are you running ?
[08:54] <daniels> ha ha ha! tee hee! i love tla!
[08:55] <daniels> integration
[08:55] <lifeless> hmm, thats v. strange then.
[08:57] <daniels> back to the wget
[09:11] <pitti> lamont: still here?
[09:21] <pitti> sabdfl: Good morning!
[09:21] <pitti> sabdfl: Say, could you life with having icons at the desktop for automounted volumes? (See sounder ML)
[09:22] <sabdfl> pitti: no thanks
[09:23] <pitti> sabdfl: icon in the panel?
[09:23] <sabdfl> i don't like the mac metaphor at all
[09:23] <sabdfl> pitti: yes, much better
[09:23] <sabdfl> notifier for automounted items
[09:23] <pitti> sabdfl: IMHO this is the only other sane approach
[09:23] <sabdfl> single item, click it and it shows a list of all automounted items
[09:24] <pitti> click to the item to umount it
[09:24] <sabdfl> if you have two items mounted?
[09:24] <pitti> click on the particular list item, not on the icon, I meant
[09:24] <sabdfl> agreed
[09:25] <pitti> okay, then we should grind the wheels to get that implemented
[09:26] <sabdfl> would it be more important to get that done, or fix rc bugs?
[09:26] <jdub> sabdfl: icons on the desktop does not imply dragging to the trash :)
[09:26] <sabdfl> jdub: understood, nonetheless, we've already had this discussion for warty
[09:27] <pitti> sabdfl: IMHO its quite important to have an obvious and safe unmounting possibility, but bug fixing is important as well. No precedence on my side
[09:27] <jdub> sabdfl: this is not important enough to write code or diverge for
[09:27] <pitti> sabdfl: since I'm quite unfamiliar with gnome programming, I would rather concentrate on bug fixing and security review and have e.g. seb or npmccallum do this thing
[09:27] <sabdfl> pitti: if mdz is happy for you to do this then fine by me, but i think the team as a whole should be rc-focused
[09:28] <sabdfl> pitti: agreed, and seb needs to do other things first
[09:28] <pitti> sabdfl: okay, then bugs get highest prio; I think it is not the end of the world if we don't manage this icon in time
[09:29] <pitti> sabdfl: because now we can umount in the computer window, so at least there is _one_ possibility
[09:32] <sabdfl> yes
[10:10] <fabbione> re
[10:14] <debianist> morning all
[10:14] <pitti> debianist: Morning!"
[10:16] <debianist> morning pitti 
[10:16] <debianist> :)
[10:17] <fabbione> hey pitti 
[10:17] <fabbione> hi debianist 
[10:20] <pitti> fabbione: Morning fabio!
[10:22] <seb128> morning
[10:22] <fabbione> hey seb
[10:24] <seb128> hello fabbione 
[10:26] <debianist> hey fabbione
[10:26] <debianist> and seb128
[10:27] <seb128> hi debianist 
[10:28] <ddaa> Hi folks.
[10:28] <debianist> hi ddaa
[10:29] <debianist> so, does the daily symlink is supposed to point to cd 8 currently?
[10:47] <fabbione> debianist: no, dailys are not sounders
[10:48] <fabbione> dailys are build every single day
[10:48] <fabbione> there might be a moment in which a daily matches the sounder
[10:53] <Oskuro> Hey, I just installed Sounder 8.
[10:54] <Oskuro> X started up ok, but using the wrong resolution, so the desktop was bigger than the display.
[10:55] <Oskuro> Changing the resolution the biggest one configured with ctrl-alt-+ worked ok
[10:55] <seb128> Oskuro: is the ca translation ok in the panel ? 
[10:55] <Oskuro> let me see
[10:55] <fabbione> Oskuro: thanks... will fix soon
[10:56] <Oskuro> fabbione: known issue?
[10:59] <daniels> Oskuro: arse. are you using an lcd?
[10:59] <Oskuro> daniels: nope, crt mon
[11:00] <fabbione> daniels: we need to kill the resolution sort thingy
[11:00] <daniels> fabbione: no
[11:00] <fabbione> daniels: i have the same problem here as everybody else
[11:00] <fabbione> or kill the "biggest" one
[11:00] <daniels> fabbione: most monitors i've seen have an optimum resolution, and n+1, which is the highest it goes, which looks like arse
[11:00] <daniels> we need to kill the sorting for the lcd, which i'll take care of
[11:01] <fabbione> daniels: dude.. read carefully what Oskuro wrote
[11:01] <daniels> sure, it doesn't work for every case
[11:01] <daniels> i personally think it's the best thing to do, however
[11:01] <fabbione> daniels: we just need to kill the highest resolution probed on crt
[11:02] <fabbione> that's what we were talking about
[11:02] <fabbione> or otherwise use the highest resolution probed
[11:02] <fabbione> either one or the other
[11:03] <HrdwrBoB> it worked for me.. detected 1800x1440
[11:03] <HrdwrBoB> and ran it in 1600x1200, with 1800 selectable by R&R
[11:03] <Kinnison> Morning
[11:04] <daniels> fabbione: what's so dangerous about having it at the end of the list?
[11:07] <fabbione> daniels: it's not dangerous, X uses the first entry in mode as resolution, but the highest one as size for virtual desktop
[11:07] <fabbione> daniels: that is rather annoying
[11:07] <fabbione> daniels: i also explained that to the mailing list a while ago
[11:07] <daniels> fabbione: er, how's it ending up with a virtual desktop size?
[11:07] <daniels> hm, which thread?
[11:07] <daniels> that's complete arse, yeah
[11:08] <fabbione> daniels:      66 Aug 25 To: sounder@lists.no-name-yet.com    (5507) . about X resolutions on non-laptops (xresprobe)                       
[11:13] <daniels> fabbione: got it. yeah, that's total arse. i'll upload a new xresprobe later tonight, 'kay?
[11:13] <fabbione> daniels: sure..
[11:13] <fabbione> i wasn't rushing...
[11:13] <fabbione> i need to work on another thing atm...
[11:13] <daniels> does need to be fixed tho
[11:13] <fabbione> daniels: yup
[11:14] <daniels> whoa
[11:14] <daniels> on my 8500, i'm seeing slight, increasing, graphical corruption with use of render
[11:14] <daniels> nice
[11:15] <daniels> thom: wake up, kid
[11:15] <Oskuro> daniels: ok, thanks
[11:15] <daniels> ARSE
[11:16] <daniels> what is up with my X builds failing??
[11:16] <Oskuro> oh daniels, ping-libburn
[11:16] <HcE> god morning Mithrandir :)
[11:16] <seb128> Oskuro: the ca translation is ok ?
[11:17] <Mithrandir> hi hce
[11:17] <Oskuro> seb128: sure, sorry
[11:17] <seb128> cool, np :)
[11:18] <thom> daniels: i'm awake, dumbass
[11:18] <thom> :-)
[11:18] <seb128> morning thom 
[11:18] <seb128> morning rburton :)
[11:20] <rburton> morning seb128
[11:22] <jdub> yo rburton 
[11:23] <rburton> hey jdub
[11:24] <seb128> jdub: !!
[11:25] <seb128> jdub: what's missing on the desktop plan before facing the 2.8 tarball releases ? :)
[11:36] <HrdwrBoB> is ubuntu supposed to detect my scsi CDROM and load the drivers for it?
[11:41] <debianist> Oskuro : that thing with the resolution happened to me also, though it was rather quick and slick fixing it using the screen resolution resizer
[11:48] <Oskuro> debianist: but is that remembered the next time you log in?
[11:48] <Oskuro> debianist: and doesn't work for all users I guess
[11:50] <debianist> Oskuro : yes it does, for the other folks - it's ok to use the highest reolution possible, just make the virutal desktop use the same size :)
[11:50] <debianist> can it be made to match virtual and physical desktop sizes?
[11:53] <debianist> fabbione : for serving the testing best, would you recommend testing a daily or the sounder 8 ?
[11:54] <Mithrandir> they should be fairly equal today
[11:54] <fabbione> debianist: both?
[11:55] <fabbione> debianist: but usually after 1 or 2 days a sounder is released, you should go for dailys
[11:55] <debianist> fabbione : k, sounds fair enough
[11:56] <debianist> who's in charge of desktop layout? I really like the new "Add to panel" big, intuitive highly readabel window
[11:59] <cef> *sigh* another MS web browsing patent
[12:00] <debianist> cef : they patented that?
[12:01] <cef> parts of web browsing
[12:01] <cef> http://www.theregister.co.uk/2004/09/07/microsoft_patents_keyboard_navigation/
[12:02] <cef> tabbing between links on a page is now apparently patented.. they applied back in '97, using art from '94. only just got approved *sigh*
[12:02] <seb128> debianist: "in charge of desktop layout" ? The new add dialog is from Vincent Untz IIRC
[12:03] <debianist> cef : :((
[12:04] <debianist> how would I add ubuntu's apt sources to .list ?
[12:04] <seb128> vi /etc/apt/sources.list ?
[12:05] <HrdwrBoB> debianist: they're in the wiki
[12:08] <rburton> mailto: links in ephy are not working for me
[12:08] <rburton> should i file a bug, or is this just me?
[12:09] <seb128> gnome-open http//www.google.co
[12:09] <seb128> gnome-open http//www.google.com
[12:09] <seb128> does it work ?
[12:09] <seb128> oups
[12:09] <seb128> gnome-open email@...
[12:09] <rburton> $ gnome-open  mailto:foo@bar.com
[12:09] <rburton> Error showing url: There was an error launching the default action command associated with this location.
[12:10] <rburton> ah, fixed it
[12:10] <rburton> gnome-default-apps was setting evolution-1.4
[12:11] <seb128> ok, please open a bug so I remind to change the default mailer :)
[12:12] <rburton> this might be an old preference of mine
[12:14] <thom> seb128: fixed gconf yet? :-)
[12:15] <seb128> thom: doing a sounder 8 installation right now
[12:15] <seb128> thom: no way to get a system installed with daily iso in 2 previous days
[12:18] <daniels> thom: can you please throw me ddcprobe output from your desktop?
[12:19] <daniels> HrdwrBoB: people still have those? :P
[12:19] <debianist> what is the rationale behind keeping the desktop so empty and clean?
[12:20] <daniels> because it's better than having it cluttered and messy?
[12:20] <HrdwrBoB> daniels: I  have one :P
[12:20] <HrdwrBoB> goddamnit.
[12:23] <debianist> daniels : is there a way to enlarge desktop icons and fonts in a sweep ? like in one click set up? (rather than working each icon and stretching it)
[12:24] <HrdwrBoB> debianist: options, default zoom
[12:34] <fabbione> thom: i am checking again gdm if we can make restart | reload)
[12:34] <thom> fabbione: reload doesn't do anything useful if it's not running/
[12:35] <fabbione> ok...
[12:35] <fabbione> hmmm
[12:35] <fabbione> i have an idea
[12:35] <fabbione> let's switch to kubuntu and use kde
[12:36] <debianist> fabbion : huh? :)
[12:36] <debianist> hahah
[12:36] <thom> EWWWWWWWWWWWW
[12:39] <HrdwrBoB> on another note, I was reading the warty wiki and I think bz2 is such an incredible CPU hog that it's not worth it for 10%
[12:39] <HrdwrBoB> law of diminishing returns and all that
[12:40] <fabbione> HrdwrBoB: same comments have been done in the last conference
[12:41] <Mithrandir> CPU isn't as much a problem as disk speed, IME.
[12:41] <Mithrandir> CPU is cheap.
[12:41] <thom> time for the amd64 daily
[12:42] <HrdwrBoB> I download gzip kernel sources, because I can have them downloaded and unzipped quicker than I can the bz2
[12:42] <HrdwrBoB> even though it downloads quicker
[12:42] <fabbione> HrdwrBoB: that's the same reason X switched back from .bz2 to .gz
[12:42] <debianist> pitti : bon appetit
[12:42] <fabbione> (at least the debian packages)
[12:49] <Mithrandir> 1:47,79 (gzip) vs 1:27,62 (bz2) here, for download + uncompressing (not untarring)
[12:53] <seb128> pitti: here ?
[12:53] <debianist> ahh what a great bw on the ftp server..I am upgrading at my maaximum connection speed :)
[12:56] <debianist> do we gtk-engines on the ubuntu rept. ?
[01:08] <debianist> i can't unmount my usb drive, there isn't no icon or anything else to work it, besides the nautilus window which I close ages ago..
[01:08] <Kamion> daniels: we've got kernels for em64t (-amd64-xeon or some such), and base-installer should pick them automatically
[01:08] <Kamion> daniels: whether it'll actually work is something I can't speak to
[01:08] <HrdwrBoB> debianist: this was covered recently, check out the mailing list
[01:09] <daniels> Kamion: rad
[01:09] <daniels> ta
[01:13] <Kamion> jdub: I did actually get your process mail, I was just temporarily too busy executing the process to answer it :-)
[01:15] <pitti> seb128: I'm back
[01:15] <thom> amd64 looks pretty good here
[01:15] <thom> besides the framebuffer horkage
[01:16] <Mithrandir> thom: fb horkage?
[01:16] <fabbione> i386 looks good too
[01:16] <fabbione> gdm sucks
[01:16] <fabbione> but that's another problem ;)
[01:17] <thom> Mithrandir: in the installer
[01:17] <thom> Mithrandir: so some languages don't work right
[01:17] <fabbione> thom: what about makeing the init script slightly more clever.. so that reload will start gdm if it is not running, and let base-config call reload?
[01:18] <thom> fabbione: seems reasonable to me
[01:18] <pitti> seb128: I will eat my pizza now, returning in 15 minutes
[01:18] <seb128> pitti: "Go to the Desktop folder" -> de ? and "Ubuntu Help and Documentation" -> de ? please :)
[01:18] <seb128> pitti: ok, no hurry, latter
[01:18] <thom> fabbione: i'll test and do
[01:18] <Mithrandir> thom: weird utf8-langs or normal, sane ones?
[01:19] <fabbione> thom: ok.. otherwise i can do it..
[01:19] <fabbione> up to you
[01:19] <thom> Mithrandir: utf-8
[01:19] <Mithrandir> ok
[01:19] <thom> fabbione: *shrug*
[01:19] <Mithrandir> it's not an amd64-specific problem?
[01:19] <thom> Mithrandir: it is
[01:19] <Mithrandir> hm, ok
[01:19] <pitti> seb128: what do you mean by "go"? Open a window? or change the directory in a browser?
[01:19] <Mithrandir> I'll take a look, then
[01:19] <fabbione> thom: i just finished with freenet6.. so i got a bit of spare time...
[01:19] <pitti> seb128: Help&Doc -> "Ubuntu-Hilfe und Dokumentation"
[01:19] <thom> go on then
[01:19] <thom> :-)
[01:20] <fabbione> thom: ok
[01:20] <fabbione> ;)
[01:20] <seb128> pitti: same as "Go to the Home folder" and "Go to the Documents folder" that you have already translated
[01:20] <fabbione> food first
[01:20] <seb128> pitti: ie: open a window on the folder
[01:20] <pitti> seb128: ah, "Den Desktop-Ordner anzeigen"
[01:20] <seb128> ok, thanks
[01:20] <pitti> seb128: not literally, but sounds much better than "go"
[01:20] <seb128> good pizza :)
[01:20] <seb128> ok
[01:21] <pitti> seb128: I did not start eating it yet :-) But I will do now.
[01:21] <rburton> oh my
[01:21] <rburton> is this a .ics button in bugzilla i see before me
[01:23] <rburton> oooohhhhhh that is good
[01:27] <debianist> fixed sound problem. apparently a post install reboot did it, and the gnome mixer is superb :)
[01:33] <thom> what the bloody hell is a .ics?
[01:34] <daniels> fabbione: don't joke about kubuntu, i'm serious
[01:34] <thom> daniels: it's a serious joke, you mean? :P
[01:34] <thom> daniels: sup?
[01:34] <daniels> thom: bah!
[01:34] <pitti> seb128: did you already upload? I'd like to change two translations
[01:34] <daniels> thom: a joke like 'i'll take you to fabric on friday'
[01:34] <daniels> thom: could you please email me ddcprobe output from your desktop (specifically, for the lcd)?
[01:35] <thom> i did, already
[01:35] <thom> daniel@fooishbar
[01:35] <pitti> seb128: the bubble help of "Home" should be "Den persnlichen Ordner anzeigen"
[01:36] <seb128> pitti: I'll upload in 1-2 hours, so you have time for changes
[01:36] <daniels> thom: er
[01:36] <pitti> seb128: I think the umlauts in IRC are broken, I'll send you a mail
[01:36] <seb128> pitti: no, that's fine
[01:36] <daniels> thom: take a card, dude
[01:36] <pitti> seb128: the Umlauts work?
[01:36] <daniels> (eithr that or my mail's broken)
[01:37] <seb128> pitti: yes, my xchat is in UTF-8 charset
[01:37] <pitti> seb128: Great! To make the translations consistent, I would like to change the previous two translations:
[01:37] <pitti> seb128: Home -> "Den persnlichen Ordner anzeigen"
[01:37] <seb128> msgid "Go to the Home folder"
[01:37] <seb128> msgstr "Den persnlichen Ordner anzeigen"
[01:37] <seb128> done
[01:38] <pitti> seb128: Desktop -> "Den Desktop-Ordner anzeigen"
[01:38] <seb128> ok, done
[01:38] <pitti> seb128: Documents -> "Den Dokumenten-Ordner anzeigen"
[01:38] <pitti> seb128: this does not mix up go/display/whatever
[01:38] <pitti> seb128: thanks a lot!
[01:39] <seb128> thanks to you for the translations :)
[01:39] <pitti> seb128: np
[01:39] <daniels> ahr, seven minutes ago
[01:39] <daniels> thom: your mta needs to be quicker to retry 450s ;)
[01:40] <daniels> lulu: yo :)
[01:40] <lulu> howdi! :o)
[01:40] <thom> i'm not sending you mail if you're greylisting :P
[01:41] <daniels> thom: i'm not greylisting, fd.o is
[01:41] <daniels> thom: if you're really desperate, daniel@the.real.fooishbar.org goes straight to tycho
[01:41] <thom> i'm not sending them email then, either
[01:41] <thom> :P
[01:41] <daniels> score! i'd been wondering how to do that
[01:41] <HrdwrBoB> my mail is worse, it's hosted on a windows box :(
[01:42] <daniels> run by you, worse luck ;)
[01:42] <HrdwrBoB> haha, not by my choice, and not really :(
[01:42] <Kamion> thom: oh, hey, it's just occurred to me why the framebuffer might be horked on amd64
[01:43] <thom> oh?
[01:43] <thom> daniels: resent
[01:44] <daniels> thom: thanks mate
[01:44] <Kamion> thom: I assume it requires roughly the same setup as on i386? vga16fb/vesafb module, that kind of thing?
[01:44] <Mithrandir> probably
[01:44] <Kamion> and the fbcon module
[01:44] <thom> daniels: if you wish me to rerun, the output is easy:
[01:44] <thom> 12:42 ~% sudo ddcprobe
[01:44] <thom> VESA BIOS Extensions not detected.
[01:45] <Kamion> thom: well, that's only done on i386 at the moment ;)
[01:45] <thom> Kamion: heh
[01:45] <Kamion> thom: what does 'dpkg-architecture -qDEB_HOST_GNU_CPU' say on amd64?
[01:45] <thom> x86_64
[01:45] <pitti> seb128: I just see that "About Ubuntu..." is still untranslated. I though that I did it, but it does not seem to be applied. It should be "ber Ubuntu..."
[01:46] <seb128> pitti: I forgotten to apply 2 strings in the previous upload, that's already fixed but thanks
[01:46] <daniels> thom: ... huh? what sort of card do you have?
[01:47] <thom> daniels: amd64 ;-)
[01:47] <daniels> thom: oh, sweet mother of god
[01:47] <daniels> thom: feel like connecting it to the i386 and ddcprobing it? :)
[01:47] <thom> no, since you have the results
[01:48] <thom> and that means rebooting
[01:49] <Mithrandir> daniels:
[01:49] <Mithrandir> : tfheen@golem ~ > sudo ddcprobe 
[01:49] <Mithrandir> VESA BIOS Extensions not detected.
[01:49] <Mithrandir> that's a ATI Radeon 9200SE
[01:49] <daniels> Mithrandir: ... on amd64
[01:50] <thom> daniels: that's what happens on amd64
[01:50] <Mithrandir> yes
[01:50] <daniels> yeah, i don't really care about amd64
[01:50] <daniels> unless someone feels like getting ddc working properly on it
[01:50] <daniels> Mithrandir: buy me one and suddenly not only will I care, it'll become really well-supported
[01:50] <Mithrandir> I can get you an account on my home box?
[01:50] <daniels> what I want is the output of ddcprobe for common desktop LCDs, so I can make xresprobe detect them right and thus use the top resolution, not n-1, so your displays don't look like arse
[01:51] <Mithrandir> I'm not using an LCD
[01:51] <daniels> jdub: hmm, how do we configure this stuff? i see no advanced menu or capplet
[01:51] <daniels> Mithrandir: oh, right. well, wanna give me root? i hoope you don't mind the occasional lockup :)
[01:52] <Mithrandir> daniels: it's just a desktop box, so I wouldn't really mind you locking up the box.
[01:52] <Mithrandir> daniels: and please don't poke too much about in my media and stuff, please?
[01:52] <jdub> daniels: what stuff?
[01:52] <Mithrandir> if so, I can give you root, if you fix ddcprobe, no problems.
[01:52] <seb128> jdub: dude you've not replied to my question some hours ago :)
[01:52] <daniels> jdub: the freedb stuff that was mentioned in #g-h
[01:53] <daniels> Mithrandir: um yeah, I'm fine with that; I don't really have anything else I'd be doing on it other than compiling ddcprobe in my ~ and testing the result
[01:53] <daniels> Mithrandir: that would be great, thanks
[01:53] <thom> daniels: but anyway, ddcprobe output you have is for my lcd
[01:54] <daniels> thom: still not coming through, ho hum
[01:54] <daniels> thom: it'll be there eventually, i suspect (wanna sudo runq?) 
[01:54] <Oskuro> Kamion: hmm, I saw a few "de Ubuntu" typos in d-i and base-config in my install
[01:54] <thom> nah, you chose to greylist, you can suffer
[01:54] <daniels> thom: the.real.fooishbar.org isn't greylisting, and i'm not even seeing a connection
[01:55] <Kamion> rootskel (0.84ubuntu4) warty; urgency=low
[01:55] <Kamion>   * Install S33framebuffer-module-linux-i386 on amd64 too.
[01:55] <Kamion>  -- Colin Watson <cjwatson@canonical.com>  Wed,  8 Sep 2004 12:46:27 +0100
[01:55] <Kamion> thom: let me know if that works; may take a day or two to make it into dailies
[01:55] <Kamion> Oskuro: mail me with where they are, if you would
[01:57] <Oskuro> Kamion: sure, I'll fetch the Ubuntu d-i source
[01:57] <Mithrandir> daniels: ok, mail sent.
[01:58] <daniels> Mithrandir: thanks dude
[01:58] <Mithrandir> my pleasure
[01:58] <Mithrandir> it's on my dsl, so the connection sometimes goes down if the router freaks out, if so, the hostname will most likely change
[01:59] <Mithrandir> but I'll tell you, then.
[01:59] <lamont> pitti: waking up at 0530 these days means going to bed a bit earlier...
[01:59] <pitti> lamont: no problem :-) Good morning!
[01:59] <jdub> Kamion: thanks for that mail! :)
[01:59] <Mithrandir> lamont: mozilla-thunderbird built on amd64?
[01:59] <pitti> lamont: what on earth are you doing right after midnight?
[02:00] <jdub> daniels: the only thing that uses freedb is the cd player
[02:00] <jdub> daniels: and you can get to the freedb config thingy from it
[02:00] <jdub> seb128: what was that?
[02:00] <daniels> jdub: ah, right
[02:00] <daniels> jdub: musicbrainz isn't seeming to be doing anything from s-j, i'll check it out later
[02:00] <seb128> jdub: <seb128> jdub: what's missing on the desktop plan before facing the 2.8 tarball releases ? :)
[02:00] <daniels> thom: am not a fascist
[02:00] <daniels> thom: ... you've got a 9800?
[02:00] <jdub> seb128: will send a mail tonight
[02:00] <seb128> jdub: I'm working on a panel upload to fix some strings, update translation and add the trashapplet again
[02:00] <HrdwrBoB> daniels: I'm sure bruce would also give you an account
[02:01] <seb128> jdub: ok, thanks
[02:01] <jdub> seb128: oh, while you're there, can you make the clock applet run 'gksudo time-admin'?
[02:01] <seb128> ok
[02:01] <jdub> seb128: (warning: it checks the path before displaying the menu item)
[02:02] <pitti> lamont: actually I wanted to ask you about postgresql 7.4.5-3. According to the logs it built correctly on m68k, but the package did not go to the archive; so it cannot migrate to sarge
[02:02] <seb128> jdub: the path ?
[02:03] <jdub> seb128: it checks to see if the program is in the path, so unfortunately, we can't just change the 'time-admin' string to 'gksudo time-admin' :-)
[02:03] <lamont> Mithrandir: try adding build-depend: gcc-3.4 :-(
[02:03] <jdub> lamont: did you get my mail about archive process?
[02:04] <lamont> jdub: yeah - and it was To: elmo, so I was going to let him answer it.. :-)
[02:04] <lamont> just got one mail from you on the subject recently, though.
[02:04] <Mithrandir> lamont: AAAAAAAAAAAAARGGGGGGGGGGGHHHHHHH.  Something fucks me over and _EATS_ my build-deps changes.
[02:04] <Mithrandir> lamont: it's a conspiracy, I'm sure.
[02:05] <lamont> always check for debian/control.in :-(
[02:05] <Mithrandir> there's none
[02:06] <fabbione> thom: can you apply the itaglish / english filter?
[02:09] <fabbione> * GNOME Display Manager is not running: trying to load it.
[02:09] <fabbione> the "trying to load it" really sucks
[02:09] <fabbione> i can't come up with anything better..
[02:09] <fabbione> suggestions?
[02:09] <HrdwrBoB> start?
[02:09] <HrdwrBoB> attempting to restart it
[02:09] <HrdwrBoB> ?
[02:09] <fabbione> trying to start it.
[02:10] <Mithrandir> I. am. stupid.
[02:10] <daniels> HrdwrBoB: hm, true dat
[02:10] <daniels> HrdwrBoB: reckon he'd give me his machine for a few days? :)
[02:13] <Mithrandir> lamont: new version uploaded
[02:15] <fabbione> ah hummmm
[02:15] <fabbione> thom: there is another small problem.. base-config calls all the possible display managers with restart
[02:16] <HrdwrBoB> daniels: heh, possibly, maybe if you put him in stasis
[02:17] <Kamion> fabbione: what's the problem with that? (bear with me, I missed it in scrollback)
[02:17] <fabbione> Kamion: the problem is that gdm restart inside X, kills X
[02:18] <Kamion> that's not a problem for base-config though
[02:18] <fabbione> Kamion: so if you run base-config new inside X, even for testing.. it will kill you out
[02:18] <fabbione> Kamion: solution: fix gdm init script so that reload will load gdm if not running and change base-config to use relaod instead of restart
[02:19] <Kamion> fabbione: don't do that then :-)
[02:19] <Kamion> fabbione: I don't think this is a bug, really ...
[02:19] <fabbione> well.. a user can run base-config inside X
[02:20] <fabbione> and he doesn't know he will be kicked out as soon as gdm/whatever will restart
[02:20] <fabbione> i think it's a bug
[02:20] <Kamion> base-config does all kinds of stuff, people who do 'base-config new' deserve whatever they get
[02:20] <HrdwrBoB> how many users do you think would do that?
[02:20] <Kamion> reload
[02:20] <Kamion>     cause the configuration of the service to be reloaded without actually stopping and restarting the service,
[02:21] <Kamion> please don't "fix" this, it really is a don't do that then
[02:21] <fabbione> Kamion: that's not what we are doing
[02:21] <fabbione> Kamion: i am not stopping the service
[02:21] <Kamion> but you're restarting it
[02:21] <fabbione> i start it if it is not running
[02:21] <fabbione> nope
[02:21] <Kamion> restarting includes starting
[02:21] <fabbione> restart = stop -> start
[02:22] <Kamion> also if you change this you have to change xdm and kdm (maybe wdm?) in universe too
[02:23] <Kamion> I think you and I are the only people likely to run 'base-config new', and we should do so from outside X
[02:23] <fabbione> let me check what xdm and kdm do
[02:23] <UbuntuBoB> *bing sounder 8 install time, tata
[02:24] <Kamion> I really do not think this should change; 'base-config new' is fundamentally a dangerous thing to do on an existing system, and restarting X is a perfectly reasonable thing to do in that context
[02:24] <fabbione> Kamion: i don't remember if it was you. or someone else that told me: "never never never do any assumption"
[02:24] <fabbione> ;)
[02:24] <Kamion> if users run it with new, they'll have much worse problems
[02:24] <Kamion> it wasn't me
[02:24] <fabbione> (or similar)
[02:25] <fabbione> i think it was you when were discussing the apt-setup thingy.. but it's not important anyway
[02:25] <Kamion> the base-config man page says 'If this is a fresh install onto a new debian system, the program receives "new" as its first argument.'
[02:25] <Kamion> I'd have no objection to extending that to say 'any other use is at your own risk and may do unexpected things to your system'
[02:26] <fabbione> ok the behaviour is consistent across login manager
[02:26] <fabbione> but it still sucks
[02:26] <fabbione> so i guess i will have to accept that it is a non bug
[02:26] <Kamion> I'd also have no objection to base-config new crashing out if $DISPLAY, personally
[02:27] <Kamion> or, if you like, just guarding that bit of code with if [ -z "$DISPLAY" ] 
[02:27] <fabbione> Kamion: if you run it as root (su -) DISPLAY is not set
[02:27] <fabbione> so it would be almost a useless check
[02:28] <Kamion> remember we don't even support 'su -' by default in warty
[02:28] <Kamion> sudo preserves $DISPLAY by default; not useless in the least
[02:28] <fabbione> yeah.. that's why i said su -
[02:29] <fabbione> and you can still do sudo su -
[02:29] <fabbione> that's easy ;)
[02:29] <Kamion> you're really trying to find ways to break your system at that point
[02:29] <Kamion> at some point we just have to accept that users who want to break their system will, and stop providing cotton wool
[02:29] <fabbione> Kamion: i think that's what we have to expect from users :P
[02:30] <Kamion> this change just isn't worth the fork
[02:30] <fabbione> Kamion: you already convinced me at least 40 lines above.. bug is closed, upload deleted...
[02:30] <fabbione> ;)
[02:30] <Kamion> speaking of base-config bugs, if you run apt-setup again you get another bunch of lines in sources.list
[02:32] <Kamion> I wonder if that needs to be 'apt-setup new' only
[02:32] <fabbione> Kamion: hmmm yeah
[02:32] <fabbione> apt-setup new doesn't exist
[02:32] <Kamion> (or equivalent, I'm sure it isn't called in quite that way)
[02:32] <fabbione> it only accept probe
[02:32] <daniels> WOO! I ROCK
[02:32] <Kamion> sure, yeah
[02:32] <Kamion> maybe do it in lib/menu/apt-setup instead?
[02:33] <daniels> fabbione: ok, we're eliminating the top resolution completely now
[02:33] <fabbione> daniels: yes.. and i am santa
[02:33] <daniels> fabbione: but we're doing it sanely - taking frequencies into account
[02:33] <daniels> so 
[02:33] <fabbione> Kamion: yes.. i was thinking already :-)
[02:33] <daniels> 1600x1200 still shows on meine, because 85Hz gets knocked out, but 75Hz is still there
[02:33] <thom> fabbione: you have the belly for it :L-)
[02:33] <Kamion> after apt-setup is run
[02:33] <daniels> thom: ouch!
[02:33] <fabbione> thom: true ...
[02:33] <fabbione> ehhee
[02:33] <Kamion> it doesn't feel quite right somehow though
[02:34] <fabbione> Kamion: no i was thinking that apt-setup (menu) accept the new option.
[02:34] <Kamion> could also grep to see if it's already there and if so leave it out
[02:34] <fabbione> Kamion: i can just talk to apt-setup via TMPFILE
[02:34] <Kamion> I'd prefer not to muck about with that, it was really really hard to merge base-config last I tried
[02:34] <Kamion> look at how security.debian.org is added ...
[02:35] <fabbione> Kamion: yes i went trough security...
[02:35] <fabbione> perhaps we can use a similar approach..
[02:35] <elmo> ":L-)" is that thom with a broken nose? ;)
[02:35] <Kamion> # * already have it in sources.list, uncommented: don't ask about it
[02:36] <Kamion> if we lose the ", uncommented" condition, then it's easy ...
[02:36] <daniels> could everyone please try xresprobe 0.4 from http://people.no-name-yet.com/~daniels/xresprobe/ to check i haven't screwed anything else up?
[02:36] <daniels> it should use the top resolution on lcds, and an n-1 heuristic that takes frequency into account on crts
[02:36] <Kamion> daniels: oh, by the way, do you still need the account on my laptop and the stuff in your home directory there?
[02:37] <daniels> Kamion: er, nope, thanks
[02:37] <elmo> kamion: hmm, the i386 and powerpc dailies failed, btw
[02:37] <daniels> Kamion: last i checked, xresprobe was doing ok on your machine
[02:37] <lucas_> hi
[02:37] <daniels> Kamion: 'sides, it's not nearly as entertaining when I can't see you die of anxiety :)
[02:37] <Kamion> elmo: in what way? I haven't tried them yet
[02:37] <Kamion> daniels: heh :)
[02:39] <elmo> kamion: as in, they weren't built
[02:39] <Kamion> elmo: um, were so
[02:39] <elmo> debian-installer | 20040801ubuntu10.0.20040908 | amd64 | 7 hours old
[02:39] <elmo> bah ^-- those dailies
[02:39] <Kamion> cjwatson@little:~/cdimage/www/daily/current$ ls
[02:39] <Kamion> MD5SUMS  report.html  warty-amd64-1.iso  warty-amd64-1.list  warty-i386-1.iso  warty-i386-1.list  warty-powerpc-1.iso  warty-powerpc-1.list
[02:39] <Kamion> oh, right
[02:39] <fabbione> daniels:
[02:39] <fabbione> gordian:~# xresprobe nv
[02:40] <fabbione> Riddell: CPD-E500E
[02:40] <fabbione> res: 1600x1200 1280x960 1280x1024 1280x1024 1152x864 1024x768 1024x768 1024x768 1024x768 1024x768 832x624 800x600 800x600 800x600 800x600 800x600 720x400 720x400 640x480 640x480 640x480 640x480 640x480
[02:40] <fabbione> freq: 30-109 48-160
[02:40] <fabbione> wtf is that?
[02:40] <Kamion> elmo: any idea why?
[02:40] <HrdwrBoB> hm, with sounder CD 8, I get some isolinux messages on my screen
[02:40] <Riddell> hmm?
[02:40] <Kamion> $ sudo xresprobe ati
[02:40] <Kamion> id: Color LCD
[02:40] <Kamion> res: 1280x854
[02:40] <Kamion> freq:
[02:40] <Kamion> (powerpc)
[02:40] <HrdwrBoB> though I can't read them because as soon as they display, it reboots
[02:40] <Mithrandir> lamont: is the new m-thunderbird happier?
[02:41] <spiv> daniels: Seems to work.
[02:41] <daniels> fabbione: looks like we need another sort -u
[02:41] <daniels> spiv: ta. how's the tour?
[02:41] <spiv> res: 1024x768
[02:42] <spiv> daniels: Good, so far.  It should be less hectic from now on, which will be a relief.
[02:42] <elmo> kamion: hmm, never mind, looks like it built, there's some sort of buildd screw up.. I'll pester lamont
[02:42] <daniels> fabbione: wtf? could you please send me the output of sh -x /usr/share/xresprobe/ddcprobe.sh ?
[02:42] <spiv> And hopefully my luggage will arrive soon.
[02:42] <daniels> spiv: rad
[02:42] <daniels> eep. where are you?
[02:43] <james_> hey spiv. Just read about your troubles. Hope your luggage turns up.
[02:43] <Kamion> elmo: version number maybe?
[02:44] <fabbione> daniels: the sort -unr that you call is useless
[02:44] <spiv> daniels: Thom's.
[02:44] <daniels> ...
[02:44] <daniels> spiv: neat
[02:44] <fabbione> daniels: sort uses \n as separator while the string has " "
[02:44] <elmo> kamion: doesn't seem to be
[02:44] <daniels> thom: want to send spiv home with trance nation 2001? ;)
[02:44] <daniels> fabbione: on, arse
[02:44] <daniels> s/on/oh/
[02:44] <elmo> kamion: it's got what looks like the w-b error for "ssh to the w-b host" died, but I'm sort of hoping it's not that
[02:45] <Kamion> elmo: eww
[02:45] <fabbione> elmo: can we add security to base-config?
[02:46] <thom> daniels: does the right thing for x40
[02:46] <HrdwrBoB> is 'Configure the Logical Volume Manager' supposed to take you back to the same screen you started on... because it does
[02:47] <elmo> fabbione: yeah
[02:47] <fabbione> elmo: ok thanks
[02:47] <Kamion> HrdwrBoB: may be broken, file a bug and assign it to me please
[02:47] <HrdwrBoB> done
[02:47] <daniels> thom: rad
[02:47] <daniels> fabbione: hm, any way to change this? i'd really prefer not to have to do space->newline, sort, then newline->space
[02:48] <fabbione> daniels: man sort?
[02:48] <Kamion> jdub: what do you use spreadsheets for other than finance and science? :)
[02:48] <fabbione> dunno
[02:48] <Kamion> daniels: xargs?
[02:48] <daniels> fabbione: already done
[02:48] <daniels> Kamion: hm
[02:49] <Kamion> xargs -n 1 is a pretty good space->newline, and you don't even have to go to much effort to do newline->space in shell
[02:51] <daniels> wow, ddcprobe.sh now looks like complete shit
[02:52] <thom> daniels: at least it's not you trying to write perl :-)
[02:52] <Kamion> jdub: so, what should I be using now for the isolinux splash on the CD?
[02:53] <daniels> thom: your activity reports should all include '* torment daniels'
[02:54] <rburton> daniels: i'll torment you as well unless you tell me where your x.org debs are ;)
[02:55] <daniels> fabbione: new version
[02:55] <daniels> rburton: http://fooishbar.org/daniel/canonical-x/, i'm not even going to tell you what the deb line is because HERE BE DRAGONS, seriously
[02:56] <daniels> rburton: if you really, seriously need them, i'll upload all the source packages form my laptop; expecting those ones to even install is a bit of a stretch as i was in the middle of redoing all of them when i started exclusively working on warty concerns
[02:56] <rburton> heh
[02:56] <rburton> i wouldn't really say I "seriously need" them
[02:56] <rburton> if those debs are likely to break, i'll wait
[02:57] <fabbione> Kamion: ok.. i am taking a lock on base-config.. do you have anything pending before i fix the apt lines?
[02:57] <daniels> if your machine works with xfree86, then stick with that
[02:58] <thom> daniels: i don't have TN 2001
[02:58] <rburton> daniels: i keep on thinking about playing with composite et al, that is all
[02:58] <thom> i have Trance Nation
[02:58] <thom> (from 1999)
[02:58] <daniels> thom: oh, right
[02:58] <daniels> rburton: ah. probably best to just bust out jhbuild and build kdrive, then?
[02:58] <thom> mixed by ferry corsten
[02:58] <thom> (and bright orange)
[02:59] <daniels> thom: i think 2001 was red
[02:59] <rburton> daniels: yeah
[02:59] <fabbione> daniels: looks ok here
[02:59] <elmo> I thought daniels was more into David Hasselhof.... music....
[02:59] <Kamion> fabbione: not for the moment, nope
[02:59] <daniels> fabbione: sensational
[02:59] <rburton> elmo: don't mention that, it brings me out in hives
[03:00] <HrdwrBoB> haha
[03:01] <daniels> elmo: yeah, the hoff is my #1 love; dnb is just a secondary and prog trance a tertiary concern
[03:02] <elmo> daniels: have you seen Dodgeball, yet?
[03:02] <daniels> elmo: ... no ...
[03:02] <daniels> wasn't planning to
[03:02] <rburton> hasselhof hasselhof hasselhof!
[03:02] <rburton> daniels: its great
[03:02] <daniels> am I missing out on some Hoff action?
[03:02] <rburton> hell yeah
[03:03] <rburton> daniels: hoff and shatner are in it
[03:03] <rburton> what more could you want?
[03:03] <rburton> well, apart from Dodgeball: The Musical
[03:04] <daniels> rburton: DUDE! no shit?
[03:04] <daniels> shatner's just one of the common people
[03:04] <HrdwrBoB> any reason why australia defaults to sydney?
[03:04] <rburton> daniels: http://imdb.com/title/tt0364725/fullcredits
[03:05] <HrdwrBoB> dodgeball comes out here tomorrow
[03:06] <rburton> i found it suprisingly good
[03:06] <rburton> daniels: isn't that against some international law?
[03:07] <daniels> they appreciate the humour
[03:14] <fabbione> elmo: can you confirm: LINE="http://security.no-name-yet.com/no-name-yet $DIST-security $COMPONENTS" ?
[03:15] <fabbione> Kamion: do you think we should disable the test on security and just add the lines? 
[03:15] <fabbione> Kamion: or do the test and show the error if the connection didn't succeed?
[03:16] <james_> dudes, I'm about to try installing sounder on a box where there's a couple of other OSes. Is there anything I need to do to make it not hose my partition table etc?
[03:17] <HrdwrBoB> is loading toshiba_acpi the default and it expects to fail?
[03:17] <HrdwrBoB> james_: don't select them and delete them :)
[03:17] <james_> Hrdwr_BoB: ok, last install I did was a nightly into qemu. I tried doing manual partitioning but it wouldn't let me. I assumed it was just a bad night.
[03:18] <james_> Hrdwr_BoB: will current sounder let me do manual partitioning?
[03:18] <HrdwrBoB> I was playing with it 5mins ago, seemed to be fine
[03:18] <james_> Hrdwr_BoB: ok great. Thanks.
[03:18] <Kamion> fabbione: we definitely need some conditions, because it wants to be commented-out on CD installs with no network connectivity and uncommented if the connection works
[03:18] <HrdwrBoB> * garauntee not actual garauntee
[03:19] <Kamion> james_: manual partitioning is pretty safe because I always test it
[03:19] <james_> Kamion: ok, awesome.
[03:19] <elmo> fabbione: err, we don't have a security alias yet
[03:19] <james_> incidently, the qemu installed decided it was a laptop. I wasn't sure if that was expected behaviour or not.
[03:20] <HrdwrBoB> james_: if you mean it put the wifi and battery applets on there
[03:20] <HrdwrBoB> that's the default, I questioned that also :)
[03:20] <elmo> do we want a separate security alias?
[03:20] <fabbione> Kamion: ok ..i will do a silence test and take appropriate action
[03:20] <elmo> given it'll likely always be on the same machine
[03:20] <fabbione> elmo: that's what i was asking before
[03:20] <fabbione> elmo: if i could add security as we discussed...
[03:20] <james_> Hrdwr_BoB: I recall it saying something about laptop stuff when it booted. I've still got the hdd image around if you want me to be more precise.
[03:20] <Kamion> fabbione: I'd appreciate the code changing as little as possible, given aforementioned merge pain
[03:21] <Kamion> although admittedly translations are far more of an issue there
[03:21] <elmo> fabbione: blah, I just forgot about  the security.nny.com bit - it does exist on ftp.nny.com
[03:21] <fabbione> elmo: ok
[03:22] <fabbione> Kamion: yes i am trying to keep changes to minimal
[03:23] <HrdwrBoB> if we're trying to keep to one app per task, why is there mozilla and firefox?
[03:24] <elmo> anyone have any thoughts on whether we need the security alias?  I suppose we should have it on least-surprise principles, but whether we should default to it?
[03:25] <HrdwrBoB> it makes people feel happy
[03:25] <HrdwrBoB> they think that security is special
[03:28] <daniels> fabbione: the other change in 0.4 which i think i forgot to changelog is that i cleaned up all the copyrights
[03:29] <daniels> fabbione: no changes, just putting them on all the files (novel!)
[03:32] <HrdwrBoB> daniels: woo X tried to make my 15" monitor run at 1024x768 at 75Hz
[03:32] <daniels> spiv: wow, just reading mary's trip wrap-up
[03:32] <HrdwrBoB> .. it can't do that
[03:32] <daniels> HrdwrBoB: sensational
[03:32] <daniels> HrdwrBoB: er, what?
[03:33] <daniels> HrdwrBoB: please email the output of ddcprobe to daniel@the.real.fooishbar.org
[03:33] <HrdwrBoB> cool
[03:37] <fabbione> elmo: personally i would like to see security.nny.com
[03:37] <fabbione> elmo: but it's just a psycological factor or reading "security" somewhere in my sources.list
[03:37] <fabbione> elmo: tho it means absolutely nothing
[03:38] <elmo> ok, I'll request it from the ISP, it might not get done today though
[03:38] <elmo> so you can either wait till tomorrow or use ftp.nny.com and change it when we move to the final non-nny.com name
[03:38] <fabbione> elmo: ok. and it will save me time unfuzzying all the translations again :)
[03:38] <fabbione> elmo: i can wait... i need to test several other things in the meanwhile
[03:40] <HrdwrBoB> daniels: sent
[03:40] <HrdwrBoB> bed time!
[03:48] <daniels> SleepBoB: thanks dude
[03:49] <daniels> oem: NVidia                                                                                                             
[03:49] <daniels> vendor: NVidia Corporation                                                                                              
[03:49] <daniels> product: NV10 Reference Board Chip Rev A6                                                                               
[03:49] <daniels> SleepBoB: um, sorry, but tough shit. your monitor's loudly claiming it can do 1024x768@75Hz; either get a monitor that doesn't lie, or ... i dunno, something else
[03:51] <daniels> jesus, bizzare. anyone texish around?
[03:52] <daniels> if so, could you please download scalable-cyrfonts, change generate_fonts and any2sfd to use fontforge instead of pfaedit, then build from source? it ftbfs for me because fmtutil can't find fmtutil.cnf, but it's generated ok in /var/lib/texmf/web2c
[03:55] <daniels> Hrdwr_BoB: and your monitor advertises 1280x960 as well, god knows how it deals with that
[03:55] <daniels> Hrdwr_BoB: i mean, we do n-1 so people have a good-looking display, but we really can't protect against monitors blatantly lying
[04:00] <Kamion> ... after only a little installer hacking ...
[04:00] <Kamion> (and a hideous UI, but can't really help that at this stage)
[04:00] <elmo> id: DPro 2070SB
[04:00] <elmo> res: 1920x1440 1600x1200 1280x1024 1152x864 1024x768 832x624 800x600 720x400 640x480
[04:00] <elmo> freq: 30-140 50-160
[04:00] <elmo> daniels: is 1920x1440 meant to be n-1 ?
[04:00] <daniels> elmo: yeah
[04:01] <daniels> elmo: (the list is generated from ddcprobe | egrep '^c*timing' as its base)
[04:01] <daniels> s/from/with/
[04:01] <daniels> elmo: ... unless you have an lcd, in which case 1920x1440 is n
[04:02] <elmo> ctiming: 2048x1536@85
[04:02] <elmo> ctiming: 1600x1200@75
[04:02] <elmo> mm
[04:04] <elmo> why would it be lower hz for lower  resolutions?
[04:04] <lamont> Mithrandir: chunderbert installed
[04:05] <Mithrandir> lamont: yay!
[04:05] <daniels> elmo: because they wrote the ddc code on boxing day?
[04:05] <daniels> or the day after their christmas party or something'
[04:05] <daniels> elmo: monitors like that are the reason we use n-1 :P
[04:06] <daniels> elmo: what sort of monitor is it, out of interest?
[04:06] <elmo> those aren't in order, btw
[04:06] <elmo> ctiming: 1600x1200@75
[04:06] <elmo> ctiming: 1920x1440@85
[04:06] <elmo> ctiming: 2048x1536@85
[04:06] <elmo> those are the last three
[04:06] <elmo> daniels: mitsubishi
[04:06] <elmo> I had no idea it could do such a high res.. I might have to try X in that res :)
[04:07] <daniels> elmo: heh!
[04:07] <daniels> elmo: 21", i presume?
[04:07] <elmo> yeah, 21" or... 22" I think, technically
[04:07] <elmo> yeah, it's 22"
[04:08] <elmo> what's the easiest way to force X to do it's auto-configure stuff, given this is an ex-sid box?  purge x* and reinstall?
[04:09] <fabbione> elmo: not all of it
[04:09] <daniels> purge xserver-xfree86 and reinstall, yah
[04:09] <fabbione> elmo: apt-get --purge remove xserver-common xserver-xfree86
[04:09] <daniels> make sure xresprobe, mdetect, and discover1 are installed
[04:09] <fabbione> apt-get install xresprobe mdetect laptop-detect discover1
[04:09] <elmo> discover1 ?
[04:10] <daniels> fabbione: ... why -common?
[04:10] <daniels> elmo: yeah, beacuse discover is a bag of arse
[04:10] <fabbione> apt-get install xserver-xfree86
[04:10] <fabbione> daniels: dexconf
[04:10] <daniels> elmo: you can not have it if you really want to tell x what sort of video card you have :P
[04:10] <elmo> oh, it was discover2 that got moved to universe
[04:10] <daniels> fabbione: ho hum
[04:10] <daniels> elmo: yah
[04:10] <elmo> can I tell it to do n-1 and not n, btw?
[04:10] <elmo> err, or vice versa, rather
[04:10] <daniels> elmo: ... it's doing n-1
[04:10] <daniels> elmo: no
[04:10] <elmo> aww, but I want shiny 2048x1536 :)
[04:10] <daniels> ber
[04:11] <fabbione> elmo: we need your brain in one piece
[04:11] <daniels> it's on my todo list, k?
[04:11] <fabbione> elmo: not cooked by monitor radiation
[04:11] <elmo> daniels: sure, only kidding
[04:11] <fabbione> elmo: :-)
[04:11] <daniels> elmo: i need to implement proper option parsing in xresprobe (there's another feature stalled on that also) in order to do that
[04:11] <daniels> elmo: in the meantime, take pingudance.mp3 as your hold music
[04:12] <elmo> dude, I've tried your music crack before, and it's _bad_ crack.. ;-P
[04:12] <daniels> pfft
[04:12] <Mithrandir> heh
[04:12] <daniels> i have worse crack, like my perl :P
[04:13] <fabbione> it's the most overabused string in apt-setup
[04:13] <fabbione> it's repeated like 4 times in each po file
[04:22] <thom> hey k!
[04:22] <kfish> yo thom, how's it going?
[04:23] <thom> pretty damn good. how 'bout you?
[04:23] <kfish> fiiine :)
[04:24] <kfish> just mucking about with ubuntu on my powerbook ...
[04:24] <kfish> checking out bugzilla to log my critical "fortune is missing" bug :)
[04:25] <jame1> guys, is there any reason I should expect grub-install to take > 1 minute in the ubuntu install process?
[04:25] <Kamion> jame1: what filesystems are you using?
[04:26] <Kamion> kfish: hm, I thought we added fortune
[04:26] <jame1> Kamion: XFS.
[04:27] <Kamion> wasn't there a warning about that?
[04:27] <Kamion> fortune-mod is in the desktop seed, has been for a little while
[04:28] <kfish> Kamion, ah, wasn't installed by default, that's all
[04:29] <jame1> Kamion: you're right. There was, I'd forgotten. Why does it not work with XFS?
[04:30] <thom> jame1: go to vt2, ps auxww|grep grub, kill that process, it restarts and works
[04:30] <thom> (as a work around)
[04:31] <jame1> thom: there are two grub processes. Kill the grub-install one or the grub one?
[04:32] <thom> grub
[04:32] <jame1> thom: you're a ninja.
[04:32] <thom> heh :-)
[04:33] <Kamion> kfish: hm, should've been installed by default; is this an older installation that you've upgraded?
[04:33] <Kamion> jame1: it's a grub bug, reported in Debian, I don't think anybody's figured out how to make it work yet
[04:33] <Kamion> something to do with having to xfs_freeze things first
[04:34] <jame1> Kamion: interesting. Thanks for the info.
[04:34] <Mithrandir> Kamion: I think maybe daniels knows?
[04:35] <kfish> Kamion, installed from warty-powerpc-1.iso a few days ago
[04:36] <fabbione> NO PHEAR! base-config new!
[04:36] <fabbione> :P
[04:37] <Kamion> weird that it can't do LVM over RAID, I wonder why
[04:38] <lamont> Kamion: so I'm hacking on the BuildDI script again...
[04:38] <daniels> Kamion: you need to invoke xfs_freeze -f then xfs_freeze -u just before grub is called, or maybe during
[04:38] <daniels> Kamion: either way, it's a xfsprogs udeb for one, and yeah
[04:38] <daniels> pain
[04:38] <lamont> wrt the date and that extra digit...
[04:39] <jame1> Ooh. It's broken a raid volume for me.
[04:39] <fabbione> Kamion: i noticed that too, but probably it's just a flag to set
[04:40] <daniels> fabbione: sync may do it
[04:40] <daniels> xfs_freeze essentially flushes
[04:40] <jame1> phew. no it hasn't. Exciting times over here.
[04:40] <fabbione> daniels: i was talking about LVM on RAID
[04:40] <Kamion> daniels: no
[04:40] <fabbione> daniels: i don't use fancy filesystems
[04:40] <daniels> Kamion: ...
[04:40] <Kamion> daniels: (a) the Debian maintainer tried that, and it didn't work for some reason; see the bug report
[04:40] <fabbione> daniels: i leave that tasks to kids ;)
[04:40] <daniels> ah
[04:41] <Kamion> daniels: (b) you don't need a udeb since you can chroot into /target
[04:41] <daniels> fabbione: you're getting dull in your old age, dad :P
[04:41] <daniels> Kamion: yeah, I fixed that by chrooting into /target and running xfs_freeze; seemed kind of icky tho
[04:41] <fabbione> daniels 
[04:41] <fabbione> ehehe
[04:42] <Kamion> daniels: nah, I think that's probably better
[04:42] <daniels> Kamion: i'll take your word for it :)
[04:42] <Kamion> daniels: no need for udebs for things you only need after the base system's installed
[04:42] <Kamion> well, in general anyway
[04:43] <Kamion> daniels: also, somebody (joeyh?) told me that grub or update-grub or something itself has to run xfs_freeze, so a udeb wouldn't help anyway
[04:43] <daniels> Kamion: eeyow
[04:45] <fabbione> elmo: is warty-updates like stable-proposed-updates ?
[04:46] <Kamion> fabbione: flag> could be, I'd have to understand parted :-)
[04:47] <jame1> I need a little more grub help, guys. I told the installer to install grub to /dev/hdg1. I've setup lilo on the other OS on the box to have an entry like this:
[04:47] <jame1> other=/dev/hdg1
[04:47] <jame1>         label="ubuntu"
[04:47] <jame1>         table=/dev/hdg
[04:47] <jame1> but booting ubuntu from the lilo prompt gets me to the OS on /dev/hde1 (windows xp).
[04:48] <jame1> Am I missing something obvious?
[04:48] <jame1> (yes, I ran lilo)
[04:48] <fabbione> Kamion: there must be a flag of some sort that say that a certain device can be used for lvm or not
[04:48] <Kamion> different disk order? lilo resolves names at the point when you run lilo
[04:48] <Kamion> fabbione: there is, in parted
[04:49] <fabbione> Kamion: perhaps the one for RAID is not set properly or forced
[04:49] <Kamion> I could well believe it, it isn't set properly on powerpc
[04:49] <elmo> fabbione: yeah, it definitely shouldn't be on by default
[04:50] <fabbione> elmo: yup.. i just wanted to be sure
[04:50] <fabbione> and no .. i was NOT going to add it
[04:50] <fabbione> i swear on thom's ppc

[04:50] <thom> heh
[04:50] <thom> fortunately i still have one ;)
[04:51] <elmo> so _that's_ where the 4th Xserver went, eh?
[04:51] <thom> heh
[04:51] <fabbione> wasn't an ipod? ;)
[04:51] <daniels> i was trying to work out how many X servers there were
[04:51] <Mithrandir> fabbione: yeah, tube ipod.
[04:51] <fabbione> ok security is done...
[04:52] <fabbione> now i need to fix the other crap
[04:52] <jame1> Kamion: assuming it was the disk order, is there a way to fix it?
[04:52] <fabbione> but first a little break
[04:52] <fabbione> i am getting old
[04:52] <fabbione> :(
[04:53] <Kamion> jame1: figure out what the disk is called in the system where you're running lilo (e.g. mount disks to find out), then call it that in lilo.conf
[04:54] <jame1> Kamion: I have targets for both /dev/hde and /dev/hdg (which are both visible from lilo-os). Booting either of them sends me to /dev/hde.
[05:01] <lamont> much better response with 1.125GB than with .125GB of RAM. :-)
[05:02] <thom> strange that
[05:03] <elmo> omg, far too much sunshine
[05:03] <lamont> thom: I _had_ DDR RAM, so I didn't buy more than the minimum when I bought the box.  sadly, that'd be registered vs unregistered or some such - my DDR was no good in the box...
[05:03] <lamont> elmo: I thought they banned sunshine in .uk, no?
[05:04] <lamont> Kamion: the new and improved daily build script only adds the extra digit when you have me do a second build in the same day, and resets it back to just YYYYMMDD whenever you do a sourceful upload
[05:04] <lamont> oh, and don't do any source-NMU's...  It won't like that.
[05:05] <Kamion> lamont: heh
[05:05] <Kamion> cool, thanks
[05:05] <pitti> seb128: Now I need some translations from you :-)
[05:05] <seb128> yeah ?
[05:06] <pitti> seb128: Administrative tasks have been disabled for security reasons. Please use Menu Computer &gt; System configuration &gt; Printing
[05:06] <pitti> seb128: this is the new warning for the CUPS web interface
[05:06] <lamont> pitti: you gonna fix sudo back, or do I get the privilege? :-)
[05:07] <pitti> lamont: what do you mean? the timeout?
[05:07] <lamont> yeah
[05:07] <pitti> lamont: since I have to go in about five minutes, I would appreciate if you could do that...
[05:07] <seb128> Les tches administratives ont t dsactives pour des raisons de scurit. Veuillez utiliser le menu Poste de travail -> Configuration systme -> Impression
[05:07] <lamont> GAH!
[05:07] <pitti> seb128: thanks!
[05:07] <lamont> mount /media/cf
[05:07] <lamont> mount: special device /dev/sda1 does not exist
[05:07] <seb128> np
[05:08] <lamont> but only if I'm silly enough to boot the machine without CF plugged in, and then try to use it...
[05:08] <lamont> wonder what I need to poke.
[05:08] <lamont> CF reader is a usb device.
[05:10] <pitti> seb128: the syst?me is broken. Can you please send me this by mail?
[05:10] <lamont> boot found the device (and allocated sd[a-d] , but no device files were created, it appears...)
[05:10] <daniels> mdz: ping
[05:10] <seb128> pitti: ? is that a new string to translate ?
[05:10] <lamont> hrm.. sda exists, but is unreadable
[05:11] <pitti> seb128: no, just the second menu part of the first string
[05:11] <pitti> seb128: all other characters worked, but not the one in "syst...me"
[05:11] <seb128> systme
[05:11] <seb128> `e
[05:12] <seb128> pitti: do you want a mail, or that's ok now ?
[05:12] <pitti> seb128: it still does not work. Maybe this character cannot be displayed in latin1
[05:13] <pitti> seb128: although it should. Wait, I just enter it by hand, `e should work
[05:13] <seb128> pitti: /charset utf-8
[05:13] <seb128> in xchat
[05:13] <kiko> the ubuntu kernel from yesterday's dist-upgrade is giving me a crc error :-(
[05:13] <pitti> seb128: okay, I did. Can you pleas send again the word?
[05:13] <seb128> systme
[05:13] <pitti> seb128: ah!
[05:14] <pitti> seb128: thanks!
[05:14] <seb128> you're welcome
[05:15] <pitti> bye
[05:17] <kfish> is it a bug for a ubuntu package to Suggest an unsupported package?
[05:20] <thom> don't think so
[05:20] <thom> certainly we've not been checking :-)
[05:20] <kfish> ok :)
[05:21] <seb128> thom: FYI I've the gconf problem on a fresh install with en_GB
[05:21] <thom> yay :-)
[05:22] <seb128> trying to find the problem for 2 hours now, and still nothing, grrrrr
[05:22] <lamont> Kamion: what does 'Add 'Defaults !lecture,tty_tickets' to initial sudoers file.' mean?
[05:24] <lamont> so we still want '!lecture', but drop tty_tickets, it appears
[05:24] <thom> lamont: eh? we don't want the Defaults <user> timeout=0 bit from base config
[05:24] <lamont> thom: that's not what was actually done.
[05:25] <lamont> ah, so that's it.
[05:25] <lamont> sudo also has per-tty tickets turned on, which I think we want to eliminate
[05:25] <thom> no, i think we're keeping those
[05:25] <thom> i thought we were, anyway
[05:25] <lamont> grumble..
[05:26] <lamont> prolly no jdub here either, eh?
[05:28] <jame1> 'night all.
[05:28] <jame1> thanks for the bootloader help.
[05:31] <Kamion> lamont: the timestamp thing was done in base-config, not in sudo
[05:32] <Kamion> lamont: that change is the bit that wants reversed
[05:42] <mdz> daniels: pong
[05:43] <mdz> lamont: I like the tty_tickets; we want to get rid of the timeout=0
[05:44] <mdz> morning all
[05:44] <Kamion> lamont: check with fabbione, he took the lock on base-config a while ago
[05:44] <fabbione> Kamion: almost done :-)
[05:45] <fabbione> Kamion: security is done.. i am testing the fix for the duplicates
[05:45] <fabbione> Kamion: we only need to wait for the dns to be updated
[05:46] <HcE> anybody got a tip to how to replace gconfd with gnome-settings-daemon in openbox? Now I'm doing a hack in .xsession, but it should be a nicer way?
[05:48] <fabbione> oh god.. for a second i had a panic attack
[05:48] <debianist> do we have any cd writing software outofthebox?
[05:48] <fabbione> i read "dentist" instead of 2debianist"
[05:48] <thom> debianist: nautilus-cd-burner and cdrecord
[05:48] <thom> fabbione: rofl
[05:48] <debianist> there rather unintuitive to find
[05:49] <thom> debianist: right clicking on an iso image and clicking write to cd is unintuitive?
[05:49] <rburton> debianist: n-c-b opens when you put a blank CD in... how much more do you want?
[05:49] <thom> and ... yeah
[05:49] <debianist> rburton : oh oops sorry. my cdrw media is not blank. ok thanks
[05:50] <Kamion> fabbione: wanna reverse the timestamp_timeout change while you're at it, then?
[05:50] <debianist> rburton : cool! it's like on Mac OS X
[05:50] <rburton> n-c-b rocks
[05:51] <thom> rburton: shame about it's author
[05:51] <thom> ;-)
[05:51] <debianist> rburton : does it empty the cdrw before writing the iso? (it's full with the 2 days daily)
[05:51] <rburton> debianist: there is an option, yeah
[05:51] <rburton> thom: the no-armed frenchman you mean? i just package it
[05:51] <debianist> rburton : i really ought to pay more attention, it's doing it right and reports it nicely.
[05:51] <thom> ah, my point still stands :-)
[05:51] <debianist> haha
[05:52] <debianist> going to test sounder8 on my dell inspiron 8200
[05:52] <rburton> thom: actually, there is probably still RossCode in it 
[05:52] <rburton> so both points stand
[05:52] <thom> *g*
[05:52] <thom> have you made m3us work yet? :P
[05:53] <debianist> thom : what's m3us ?
[05:53] <rburton> thom: i'm still waiting for the patch
[05:53] <thom> debianist: a running joke, but also the mp3 playlist file format
[05:53] <thom> rburton: pffft
[05:54] <rburton> thom: thought i've nearly got quality sorted!
[05:54] <thom> rburton: cool :-)
[05:54] <thom> now you just need to turn the whole thing into a rhythmbox plug in...
[05:55] <rburton> i keep on telling them that the extracting bit it is standalone, but no-one listens
[05:56] <thom> rburton: hrm, care to file a bug to remove ephy web bookmarks for us, too?
[05:56] <rburton> i was about to propose that
[06:05] <thom> odd that
[06:12] <fabbione> Kamion: deb cdrom:[Ubuntu 4.10 _Warty Warthog_ - Unofficial i386 Binary-1 (20040908)] / unstable main restricted
[06:12] <fabbione> is it normal?
[06:12] <fabbione> that it adds restricted at the end
[06:16] <mdz> I believe so
[06:16] <mdz> since we'll be shipping some restricted on the CD
[06:17] <fabbione> ok
[06:18] <lamont> fabbione: wanna remove the timeout=0 from base-config's sudo hackery while you're in there?
[06:19] <fabbione> too late
[06:19] <lamont> fabbione: does that mean that you're done with the lock, and ubuntu13 is in-process?
[06:19] <fabbione> ubuntu13 should be accepted in less than 30 secs
[06:19] <fabbione> lamont: correct
[06:20] <lamont> np - I'll do it.
[06:20] <fabbione> i was waiting for katie to release the lock
[06:20] <lamont> this is a virtual-lock, yes?
[06:20] <fabbione> mdz, elmo: ubuntu13 will add security.nny.com if the archive is reachable, otherwise it will add it commented out
[06:20] <fabbione> lamont: of course
[06:21] <mdz> fabbione: sounds good
[06:21] <mdz> fabbione: is there any reason not to do the same for main?
[06:21] <fabbione> mdz: yes. main and cdrom are supposed to be the same
[06:21] <mdz> we should make all of supported available if they are connected to the internet
[06:22] <mdz> fabbione: the cdrom only has desktop+ship
[06:22] <mdz> fabbione: oh, you are saying you did it already
[06:22] <mdz> thanks
[06:22] <fabbione> mdz: i think that's more than enough.. also.. people that wants more, usually knows what to look for
[06:22] <Kamion> fabbione: ha, it started doing that without my intervention
[06:22] <fabbione> Kamion: ?
[06:22] <Kamion> fabbione: (largely, anyway) - because debian-cd has started to include /dists/warty/restricted/ now and restricted is in /dists/warty/Release
[06:23] <Kamion> so it all happens automatically
[06:23] <fabbione> mdz: security is enabled automatically if you are connected to the network, otherwise it is added but commented.
[06:23] <mdz> fabbione: right. but what about warty/main?
[06:23] <fabbione> mdz: but not for main/restricted/universe
[06:23] <fabbione> mdz: main/restriced are enabled automatically on netinstalls
[06:23] <mdz> I think it should do the same thing for main, and possibly restricted
[06:23] <fabbione> mdz: but not on cdrom
[06:24] <fabbione> mdz: i don't think we should do that
[06:24] <mdz> I think they should be enabled for cdrom installs if network is available
[06:24] <fabbione> mdz: because that stage happens before the install
[06:24] <Kamion> mdz: has anything from restricted been seeded yet?
[06:24] <mdz> Kamion: no, because there is only one package there, and its name is going to change
[06:24] <Kamion> mdz: I'd like that to happen soon if it's going to happen so that I can shake out any bugs in debian-cd it exposes
[06:24] <fabbione> mdz: so you might end up downloading stuff from the net because it's "newer" than CD.
[06:24] <seb128> hum, we still don't have bash_completion activated on default installation ?
[06:24] <mdz> fabbione: did you hear from herbert, btw?
[06:24] <Kamion> seb128: bash_completion is very slow to load for me
[06:24] <fabbione> mdz: yes. he will come back to me in one or two days
[06:25] <mdz> hmm, that rename should have been trivial
[06:25] <mdz> we cannot wait on this stuff, it needs testing
[06:25] <mdz> I'll email him
[06:25] <fabbione> mdz: that's what i was expecting
[06:25] <seb128> Kamion: how slow ?
[06:25] <fabbione> mdz: but he said that the new packages will be ready in 1/2 days
[06:25] <Kamion> seb128: about a second
[06:25] <Kamion> seb128: which is pretty painful at the rate I fire up new shells
[06:25] <Kamion> this is on a newish box, too ...
[06:26] <fabbione> mdz: i didn't push any further because i don't know exactly the releation with Xu
[06:26] <Kamion> well, ok, it's more like half a second once it's in the cache
[06:26] <mdz> seb128: if you want fancy completion, we provide zsh :-)
[06:26] <seb128> Kamion: I've no problem with it here, but ok, if that's slow, I'll just keep activating on my boxes :)
[06:26] <fabbione> Kamion: ok.. gotcha (about restricted)
[06:27] <seb128> mdz: no, I just like to apt-get ins<TAB> nauti<TAB> 
[06:27] <seb128> that's fine with the bash one :)
[06:31] <mdz> fabbione: re: "so you might end up downloading stuff from the net because it's "newer" than CD.", the same is true of security
[06:31] <mdz> fabbione: if we update things in Warty, it is for good reason, and they should supersede what is on obsolete CDs
[06:32] <mdz> fabbione: granted, the user could be behind a modem, so we may not want to install them immediately
[06:32] <fabbione> mdz: sorry but i don't completly agree
[06:32] <mdz> fabbione: but when the user does their first upgrade, they should certainly receive the updates
[06:33] <mdz> fabbione: what is the basis of your objection?
[06:33] <fabbione> mdz: modem is one reason, the other one is that we want to be able to install from cdrom 100%. If we enable stuff around is going to be bad
[06:33] <kagou> hi from ubuntu 8 :)
[06:33] <mdz> fabbione:  but when the user does their first upgrade, they should certainly receive the updates
[06:33] <fabbione> mdz: i am dealing to push security updates as "mandatory" but not all the rest
[06:34] <fabbione> mdz: if they want updates they will have to uncomments the lines.
[06:34] <mdz> fabbione: the net result is that cdrom installs do not get access to additional supported software by default, and that is wrong
[06:34] <fabbione> mdz: perhaps they don't want to update from us
[06:34] <mdz> forget about updates for a moment
[06:34] <mdz> all of Supported becomes invisible for those users
[06:35] <fabbione> mdz: i understand what you mean, but at that point a user that wants more, will know what to do
[06:35] <fabbione> mdz: in the beginning Mark said: "There is only cdrom installation"
[06:35] <mdz> fabbione: a Debian user, yes
[06:35] <mdz> a power user, yes
[06:35] <mdz> but everyone else should have it, too
[06:35] <fabbione> ah here is ;)
[06:36] <kagou> seem strange that i had to reboot for access my sound card 
[06:36] <mdz> sabdfl: question: should users installing from CD have access, by default, to all of warty/main via http by default?
[06:36] <sabdfl> yes absolutely
[06:36] <mdz> I agree
[06:36] <sabdfl> so the default sources.list should not refer to the cd, just the web site
[06:37] <lamont> sabdfl: and should there initial install fetch newer bits from the net if they're there, or just install from CD?
[06:37] <sabdfl> main and restricted
[06:37] <fabbione> sabdfl: also for people installing from a modem?
[06:37] <mdz> _not_ the CD?  that would defeat the point of ShipSeed
[06:37] <mdz> I do not think that the initial install should download anything for a CD install
[06:37] <sabdfl> mdz: it's copied into their cache already, isn't it?
[06:37] <mdz> however, after the install is complete, all of Supported should be available
[06:37] <mdz> sabdfl: hmm, good point
[06:38] <fabbione> sabdfl: not necessarely.. if we add the http stuff no...
[06:38] <sabdfl> fabbione: it would be nice to have the "fetch latest updates" as an option during the install
[06:38] <fabbione> sabdfl: option = question?
[06:38] <sabdfl> in fact, that could happen during the Great Copy, couldn't it?
[06:38] <sabdfl> fabbione: would have to be
[06:39] <Kamion> sabdfl: the cdrom: URI needs to be in sources.list otherwise the cached copies don't work
[06:39] <mdz> Kamion: they should work fine for http as well
[06:39] <sabdfl> ah, so they are "caches of the cdrom packages"
[06:39] <Kamion> sabdfl: I ran into this just yesterday
[06:39] <sabdfl> ok
[06:39] <Kamion> mdz: only if you can contact it, though ...
[06:39] <sabdfl> will it always prompt for the cd then?
[06:39] <Kamion> it's more complicated to get rid of the cdrom: URI and it's not necessary
[06:39] <mdz> Kamion: we already agree that we should only uncomment it if it can be contacted
[06:39] <fabbione> sabdfl: as it is now after the install we suggest the http lines for main/restriced and universe as we agreed via email and we force security updates
[06:39] <Kamion> sabdfl: no
[06:40] <Kamion> sabdfl: basically we do an update from the cdrom before reboot, and that sets up apt's database post-reboot properly
[06:40] <mdz> as it is now, if you uncomment main, then it will still always prompt for the CD if the package was on the CD
[06:40] <mdz> which is in many situations much less convenient than downloading
[06:40] <Kamion> mdz: oh, I see. really?
[06:40] <Kamion> I'd *much* rather have the CD
[06:40] <mdz> Kamion: it goes roughly in sources.list order, first-match
[06:40] <mdz> oh, that
[06:40] <Kamion> mdz: we could always put http: above cdrom: ...
[06:41] <mdz> when I finish installing, I put the CD away
[06:41] <mdz> usually I want to grab something small
[06:41] <mdz> which takes 5 seconds via network
[06:41] <mdz> and much longer for me to go get the CD and put it in the drive
[06:41] <Kamion> mdz: then the ship seed is entirely pointless
[06:41] <sabdfl> what about creating a repository on the hdd... i suppose that's a space issue
[06:41] <mdz> Kamion: my concept of ShipSeed was to have things close by that you might need to get on the network and such
[06:41] <sabdfl> getting the cd is a right pain. imagine a year later...
[06:41] <fabbione> sabdfl: that's more complicate yes..
[06:41] <Kamion> sabdfl: yeah, also an automatic cleaning issue, using /var/cache/apt/archives means that 'apt-get clean' gets rid of it
[06:42] <Kamion> mdz: shipseed has stuff like alternate browsers at the moment
[06:42] <Kamion> sabdfl: I wasn't suggesting making the CD mandatory, merely usable ...
[06:42] <mdz> Kamion: those are huge; if the user really wants them, they might be grateful to have them on the CD
[06:42] <mdz> I'm not sure we can remove the cdrom: line without putting low-bandwidth users in a bad position
[06:42] <fabbione> mdz, sabdfl: switching apt lines after the install means creating another hook for base-config or a deep modification to apt-setup...
[06:42] <Kamion> mdz: but they won't be able to if the cdrom: URI is zapped
[06:42] <sabdfl> Kamion: how are we for cd size at the moment?
[06:43] <Kamion> sabdfl: 'bout 100MB short of the wall
[06:43] <sabdfl> ok
[06:43] <debianist> are we preloading firefox, or better yet epiphany? this also has slow response times for the first fire ups
[06:43] <debianist> ?
[06:43] <Kamion> maybe a bit more
[06:44] <sabdfl> if we created an apt repository on the hard disc, then that would take up an extra 600MB of space, but you'd never need the cd again
[06:44] <Kamion> sabdfl: but you wouldn't be able to clean it automatically
[06:44] <mdz> Kamion:    * Install S33framebuffer-module-linux-i386 on amd64 too
[06:44] <mdz> Kamion: is that the only reason the framebuffer didn't work for me on amd64?
[06:44] <mdz> (potentially)
[06:44] <thom> mdz: that was the thought
[06:44] <Kamion> sabdfl: and there's a debootstrap deficiency that means it'll probably actually be 1.2GB extra during the install
[06:44] <mdz> I think the /var/cache/apt/archives approach is the right one
[06:44] <sabdfl> ah
[06:45] <Kamion> mdz: in theory, can't test yet :)
[06:45] <fabbione> sabdfl: our installation is already big enough...
[06:45] <mdz> though we should probably clean out the stuff we actually _installed_ already
[06:45] <Kamion> mdz: agreed
[06:45] <Kamion> sabdfl: I was thinking of making archive-copier not bother to copy Base (since with the cdrom-detect change, debootstrap is now pretty fast by itself)
[06:46] <sabdfl> hmmm... so in an ideal world, we'd have the desktop stuff copied to cache, which you can remove instantly, and the ship.seed stuff copied to a mini-repo
[06:46] <Kamion> copying Desktop is clearly necessary to make the no-CD-in-second-stage work
[06:46] <mdz> but we're straying from the important issue, I think, which is that users with Internet access should, by default, have access to all of Supported
[06:46] <sabdfl> Kamion: agreed
[06:46] <Kamion> that leaves the question of Ship
[06:46] <mdz> when they fire up aptitude and synaptic and look for an application, they should find it
[06:46] <mdz> if we had the infrastructure to warn them about unsupported packages, I'd say universe should be added by default as well
[06:46] <sabdfl> the effect we want, the use case, is:
[06:46] <sabdfl> (a) install with CD, it pops out.
[06:46] <sabdfl> (b) throw away the CD
[06:47] <mdz> sabdfl: nono, (b) give the CD to a friend :-)
[06:47] <Kamion> oh, I should make the CD pop out on powerpc too now, shouldn't I?
[06:47] <sabdfl> (c) install anything that was on it (in ship.seed) without needing the cd
[06:47] <mdz> Kamion: yes
[06:47] <sabdfl> (d) and anything newer, or outshide desktop+ship, comes over http
[06:47] <mdz> sabdfl: ok, currently that doesn't actually work
[06:47] <Kamion> that was turned off because powermacs don't boot from CD without manual intervention danyway
[06:47] <sabdfl> ok
[06:47] <sabdfl> what works?
[06:47] <mdz> removing cdrom and adding http :-/
[06:48] <Kamion> mdz: if we turned off base-config's cache-cleaning, that would be fine ...
[06:48] <mdz> apt doesn't look in the cache for stuff it expects to find on cdrom
[06:48] <mdz> cdrom is already 'local'
[06:48] <mdz> (I think)
[06:48] <mdz> Kamion: oh, I didn't realize we were still cleaning the cahce
[06:48] <mdz> so I haven't actually tested it
[06:48] <Kamion> the mini-repo for Ship would only be 100MB or so, which is not so bad
[06:48] <fabbione> mdz: today's crack didn't ask me for a CD and everything was copied in cache
[06:48] <Kamion> mdz: after aptitude -y install blahrant runs, yes
[06:49] <mdz> fabbione: yes, but then it's all deleted afterward
[06:49] <mdz> hmm, I wonder if aptitude diverges there
[06:49] <Kamion> diverges?
[06:50] <mdz> I'm going to test it right now
[06:50] <mdz> Kamion: aptitude reimplements significant pieces of apt
[06:50] <Kamion> so, we could have archive-copier create /var/cache/archive-copier/desktop/ and /var/cache/archive-copier/ship/
[06:50] <debianist> mdz : aptitude is better than apt, right?
[06:50] <Kamion> when base-config runs, it moves /var/cache/archive-copier/desktop/ into /var/cache/apt/archives/ and then something sticks /var/cache/archive-copier/ship/ into sources.list
[06:50] <mdz> debianist: aptitude is a frontend to apt
[06:50] <mdz> ok, so the cache thing does in fact work
[06:51] <mdz> it will look in the cache for cdrom: sources
[06:51] <Kamion> mind you, I'm a bit leery of having sources.list refer to anything in /var/cache/
[06:51] <mdz> (both apt-get and aptitude)
[06:51] <Kamion> mdz: what does apt do if a file: URI disappears? does apt-get update fail?
[06:51] <mdz> Kamion: yeah, that wouldn't be right. apt gets unhappy if it disappears
[06:51] <mdz> apt-get update would fail, yes
[06:51] <Kamion> ok. it would be nice to have something which can kill that archive on demand, though
[06:52] <Kamion> apt-setup feels like kinda the right place but we're overloading it already ...
[06:52] <fabbione> Kamion: if stuff needs to be done AFTER the first install apt-setup is not the place to look at.
[06:52] <Kamion> mdz: /var/lib/?
[06:53] <mdz> what can we do about the ftp.no-name-yet.com issue?  we can either do it in apt-setup, which has the problem that the install may take a very long time, or we can do it later, which means adding a new base-config bit, right?
[06:53] <Kamion> fabbione: this is a reconfigure thing
[06:53] <fabbione> Kamion: we will need another "hook" for base-config or something
[06:53] <Kamion> fabbione: it's acceptable to run apt-setup after the install; it's got a man page and everything
[06:53] <fabbione> Kamion: it will require a switch or something in any case = extra diversion from Debian
[06:54] <Kamion> and there isn't really any other tool that munges sources.list, apart from dselect's configurator, unless there's something in synaptic
[06:54] <Kamion> fabbione: well, it would have to be a menu item; but it's true that apt-setup doesn't ever remove sources.list entries at the moment
[06:55] <Kamion> ok, I suppose we could be overanalysing this, we could just say to people "if you want to get that space back, rm -rf /var/lib/ubuntu/ship/ [or whatever]  and edit the line out of sources.list"
[06:55] <Kamion> remembering that it is warty
[06:56] <fabbione> Kamion: exactly and it does comment out the old ones only when called via menu/apt-setup new
[06:56] <fabbione> that doesn't make anything easier
[06:56] <Kamion> mdz: the second stage already takes a long time; I'm not sure that it's a problem to have "downloading <foo> from security.no-name-yet.com" show up there
[06:56] <Kamion> fabbione: it's not easier, but it works :-)
[06:57] <fabbione> Kamion: well.. it removes ALL of them ;)
[06:57] <Kamion> mdz: gives people "secure by default" warm fuzzies, contrasted with the Windows approach of "install system then get on the net as fast as you can to download updates before your system is cracked"
[06:58] <fabbione> well...
[06:58] <fabbione> i need to go and prepare dinner
[06:59] <fabbione> later (hopefully)
[06:59] <Kamion> also, we'll be releasing update CDs fairly frequently as I understand it, so it shouldn't get too bad provided you have a recent CD (which, granted, is not going to be everyone)
[06:59] <fabbione> (otherwise it means that i will be doomed by something i don't want to do)
[07:01] <mdz> Kamion: it takes a long time, but not nearly as long as installing xfree86 over a modem or ISDN
[07:02] <Kamion> do we want to deliberately install systems without security updates, though? I'm betting that most users will not install a system and then immediately think "ooh, I must update it"
[07:06] <mdz> Kamion: ideally, the security updates would be a separate step, which could be run from X while they do other things
[07:06] <mdz> for warty, though...
[07:11] <Kamion> that's the $64000 question
[07:18] <thom> gvm has the build-deps of doom
[07:19] <lamont> what other sudo-like packages are there in debian>?
[07:20] <Oskuro> super
[07:20] <lamont> thanks
[07:21] <lamont> http://equal-rights.no-name-yet.com/patches/dvd+rw-tools.268308.diff
[07:24] <glyph> I am lame; I get some leet 0-day warez and I don't even have time to test them :-(.
[07:24] <lamont> but I really don't like unplugging wires internal to the computer...
[07:24] <glyph> does anyone have a suggestion for me to install ubuntu on one of my machines without destroying the existing debian install?   I don't have room anywhere for a full backup of my existing drives.
[07:25] <lamont> glyph: do you have a spare couple of gig in a partition of it's own?
[07:25] <glyph> lamont: if I had that, I wouldn't be asking this question ;)
[07:25] <lamont> rumor has it you can force ubuntu to just install in that partition, instead of trashing the machine...
[07:26] <lamont> then you're probably stuck with installing another hard drive... :-(
[07:26] <glyph> lamont: Rgargh
[07:26] <lamont> or you could install vmware under debian, and then install ubuntu in a virtual machine...
[07:26] <glyph> lamont: I meant "warez" in the figurative sense
[07:26] <lamont> or s/vmware/selinux/
[07:26] <glyph> selinux?
[07:26] <lamont> sorry.  uml
[07:26] <glyph> you mean UML?
[07:26] <glyph> yeah, thought so
[07:27] <glyph> I guess it's about time I bought a new hard drive anyway
[07:27] <glyph> unless you guys want to send me one? :-)
[07:27] <lamont> glyph: I fear it's too far to walk. :-)
[07:27] <glyph> lamont: you could just mail-order me one
[07:29] <Kamion> glyph: how big's your swap partition? :-)
[07:31] <glyph> Kamion: On the machine where I have one, it's just about big enough for software suspend :_)
[07:32] <Kamion> not big enough to install into, then ...
[07:32] <Kamion> lamont: such as?
[07:32] <lamont> if the source/binary disappears before my mirror script's rsync gets to it, then it thinks it's done, when it's not.
[07:38] <spiv> glyph: do you have a large swap partition..? :)
[07:39] <spiv> Oh, I see Kamion just asked that.
[07:39] <lamont> sigh.  sometimes you can over kill something.
[07:40] <lamont> --size-only is good for debs (especially a debain mirror, where datestamps get screwed up in the mirroring), but not so good for Packages files.. :-(
[07:41] <mdz> gah
[07:41] <mdz> today is Debian "let's clone every RC bug into three" day
[07:41] <lamont> mdz: le huh?
[07:42] <mdz> lamont: http://bugs.debian.org/270619 http://bugs.debian.org/268165
[07:43] <lamont> Merged 266228 268143 268163 268164 268165 268225. 
[07:43] <lamont> Now _that_ is funny
[07:44] <Kamion> rootskel (0.91) unstable; urgency=low
[07:44] <Kamion>   * Joey Hess
[07:44] <Kamion>     - Add S05acpi-i386 to load fan and thermal modules if available
[07:44] <Kamion>       as early as practical in boot to avoid machine meltdown.
[07:44] <Kamion>  -- Joey Hess <joeyh@debian.org>  Thu, 19 Aug 2004 20:39:33 +0100
[07:44] <Kamion> ooh, must merge that
[07:44] <spiv> Kamion: Oh, which reminds me...
[07:45] <spiv> Kamion: Does the fan module get loaded before fsck runs?
[07:46] <Kamion> dunno, I was kind of assuming it'd be in the initrd
[07:48] <spiv> I'll try to pay attention next time I boot, I guess :)
[07:51] <mdz> spiv, Kamion: https://bugzilla.no-name-yet.com/show_bug.cgi?id=489
[07:51] <mdz> which was fixed ages ago
[07:51] <mdz> fan and thermal are loaded from the initrd
[07:53] <spiv> mdz: Ah, cool :)
[07:54] <Kamion> mdz: we should load those on amd64 too, I take it
[07:54] <mdz> of course, if the kernel panics before then or something, you're screwed
[07:54] <mdz> if your hardware is buggy like Keybuk's
[07:57] <lamont> my USB CF reader (2.6.8.1 kernel) only auto-mounts the CF when the USB cable is connected, removing CF and re-inserting gives no happiness...
[08:00] <thom> mdz: *g*
[08:01] <SteveA> weird... totem-xine is working now.  I changed nothing.
[08:08] <kiko> is there a way to avoid all these device mapper errors at bootup beyond disabling the script?
[08:09] <Kamion> kiko: mdz said that was fixed in recent kernels
[08:09] <Kamion> kiko: are you using 2.6.8.1 in warty?
[08:10] <kiko> I'm using warty, this morning's dist-upgrade.
[08:10] <kiko> "ubuntu debugging artwork" basically.
[08:10] <thom> mdz: i still got the device mapper spew on amd64
[08:10] <Kamion> kiko: dist-upgrade won't install the new kernel
[08:11] <kiko> Kamion, okay, a new package name, just picking it up.
[08:11] <Kamion> should do from now on, provided you have the linux-image-2.6-{386,686,whatever} package installed
[08:11] <Kamion> but it won't work across the kernel-image -> linux-image rename
[08:11] <mdz> thom: during evms activation?
[08:12] <thom> mdz: i imagine during devicemapper activation, but i'd have to reboot to check
[08:12] <kiko> and rhythmbox just randomly crashes with more than 1,400 songs loaded
[08:14] <lamont> kiko: I think there's a trivial workaround for that...  who needs 1400+ songs, after all.???
[08:15] <lamont> 3days + playing time.  heh
[08:16] <kiko> I have 4 days something, but..
[08:16] <glyph> I have 3000 songs just in my electronic downloaded collection
[08:17] <glyph> closer to 15,000 total :)
[08:24] <thom> so, um, the hal in unstable does a lazy umount by default
[08:24] <mdz> oh?
[08:25] <thom> when you remove a device
[08:25] <mdz> I thought hal didn't implement actions like that
[08:25] <mdz> I've certainly yanked out usb devices and not had them unmounted
[08:25] <thom> 19:12 < sjoerd_> thom: hal runts lazy umount if a device is gone
[08:26] <mdz> huh
[08:26] <thom> and that's the debian maintainer
[08:26] <mdz> hald/linux/block_class_device.c:force_unmount
[08:26] <kiko> Kamion, why was the rename done, though?
[08:26] <mdz> perhaps that just needs to be changed to run pumount
[08:27] <thom> looking now
[08:28] <mdz> updated bugzilla
[08:39] <lamont> bad gdm
[08:39] <lamont> libtool: link: `/usr/lib/gcc-lib/powerpc-linux/3.3.4/../../..//libbonobo-activation.la' is not a valid libtool archive
[08:40] <Kamion> kiko: we were redoing the packages from scratch, and it seemed like a good time to resolve mdz's pet peeve :-)
[08:40] <kiko> heh. 
[08:40] <lamont> seb128: gdm looks to be missing a build-dep or 6...
[08:42] <mdz> kiko: the name of the software contained in the package is 'linux'; that should be reflected in the name of the package :-P
[08:43] <kiko> mdz, this isn't a plan to allow something like hurd-image to pop up, right?
[08:44] <lamont> kiko: but netbsd-image, otoh........
[08:45] <kiko> see, you made mako_ go away.
[08:45] <kiko> err
[08:46] <kiko> so I get some mknod file exists messages
[08:46] <kiko> and my eth0 stops working.
[08:47] <kiko> there's an error in bluetooth.conf..
[08:47] <lamont> mdz: only 6MB of diff between abiword...
[08:48] <thom> mdz: ARH!
[08:48] <thom> mdz: hal runs as user hal
[08:48] <mdz> thom: and rightly so
[08:48] <mdz> thom: hal is, however, in group plugdev
[08:49] <lamont> id hal
[08:49] <lamont> uid=113(hal) gid=113(hal) groups=113(hal),24(cdrom),25(floppy)
[08:49] <thom> so it is
[08:49] <lamont> hrm... wonder if that's it...
[08:49] <thom> 19:47 ~/work/packages/hal-0.2.92% id hal
[08:49] <thom> zsh: 'hal' -> 'hald' [nyae] : n
[08:49] <thom> uid=109(hal) gid=109(hal) groups=109(hal),24(cdrom),25(floppy),103(plugdev)
[08:49] <lamont> yeah -mines an upgrade, hence plugdev is b0rked,.
[08:50] <kiko> OH.
[08:50] <seb128> lamont: gdm is missing a Build-Depends ? Which one ? That's ubuntu14 ... weird that Build-Depends changed 
[08:50] <kiko> mdz, Kamion: this kernel recognizes my pcmcia network card *before* my inbuilt one.
[08:50] <kiko> an, not the kernel, the order modules are loaded.
[08:51] <lamont> ubuntu14 on powerpc.  bitched about libbonobo-activiation, which didn't occur in the log file before the usage...
[08:51] <seb128> weird
[08:51] <thom> mdz: still not umounting. hrm
[08:51] <seb128> bonobo-activation is a part of bonobo
[08:51] <kiko> mdz, Kamion: can I trust the change will be stable, or is this a bug we should be fixing?
[08:52] <seb128> thanks
[08:52] <mdz> kiko: when you install from scratch, the installer writes an /etc/iftab with names for each interface
[08:52] <mdz> kiko: which is then used to assign them the same names in the future
[08:53] <mdz> kiko: please do test it
[08:53] <kiko> let
[08:53] <kiko> 's see.
[08:53] <mdz> kiko: if you yank the card corresponding to eth0, the other one should remain eth1
[08:53] <lamont> have I mentioned that my experience running buildd's on powerpc's leads me towards believing that the kernel on powerpc may not be ready for prime time...
[08:53] <kiko> mdz, no /etc/iftab exists here, I'm upgrading from a very early sounder (5 or so).
[08:53] <mdz> kiko: that'd do it
[08:54] <kiko> mdz, can I kick something to get it created?
[08:54] <mdz> kiko: you can reinstall, which would be an excellent test for us :-)
[08:54] <mdz> or if you can't, you can create the file by hand
[08:54] <kiko> heh. I can on one of the new notebooks we are buying, but not on my production box, not today.
[08:55] <kiko> is there an example on my filesystem?
[08:55] <mdz> it'll look like this
[08:55] <mdz> mdz@max ~ $ cat /etc/iftab
[08:55] <mdz> # This file assigns persistent names to network interfaces
[08:55] <mdz> eth0 mac 00:0d:93:b0:50:f0
[08:55] <kiko> trivial. okidok.
[08:55] <mdz> see also man 5 iftab
[08:55] <mdz> hmm, that comment in the file ought to refer to the man page
[08:55] <lamont> mdz: what package is iftab in??? (/me has no man page...)
[08:56] <kiko> apropos iftab didn't tell me anything, hum hum.
[08:57] <[Clint] > lamont: ifrename
[08:57] <lamont> ah, ok
[08:57] <mdz> kiko: you're probably in the same position as lamont
[08:57] <mdz> kiko: older CDs wouldn't have installed ifrename by default
[08:58] <kiko> oh.
[08:59] <kiko> reboot 2.0
[09:01] <kiko> aha, that works.
[09:02] <kiko> mdz, so about the mknod spew and bluetooth error -- should I be concerned enough to file bugs or are you on this?
[09:03] <mdz> kiko: I don't have any bluetooth hardware, so I've no idea about that one
[09:03] <mdz> kiko: I don't recall seeing any mknod errors, either.  if you could file a bug specifying which init script is producing the error, we should be able to track it down
[09:05] <kiko> I have no bluetooth hardware that I know of either, but it is loading the HCI device layer..
[09:05] <mdz> kiko: what is the error that you see?
[09:06] <kiko> mdz, let me reboot without gdm so I can actually tell you.
[09:09] <kiko> cardmgr is complaining about an error in ./bluetooth.conf line 3
[09:09] <kiko> no driver bindings for AmbiCom BT2000C Bluetooth PC/CF Card
[09:10] <mdz> odd that it would mention line 3
[09:10] <mdz> but apparently you have a PC card device that looks like a bluetooth card
[09:10] <kiko> I have a netgear 11g card, that's it.
[09:10] <mdz> run "cardinfo"
[09:10] <kiko> and prism54 works correctly with it.
[09:11] <kiko> okay, one sec, rebooting again.
[09:11] <kiko> (I want to get you the mknod errors)
[09:11] <kiko> I get
[09:11] <kiko> Starting raid devices: ...
[09:12] <kiko> mknod: /dev/hda1: file exists
[09:12] <kiko> ...
[09:12] <kiko> (perhaps for each of the hda devices, it flips up too fast and dmesg is too short to catch it and ctrl-S doesn't work at that point)
[09:12] <mdz> dpkg -s mdadm | grep Version
[09:13] <kiko> 1.6.0-3
[09:13] <mdz> that's not from Warty
[09:13] <mdz> I fixed that bug in Warty already
[09:13] <mdz> mdadm (1.5.0-2ubuntu3) warty; urgency=low
[09:13] <mdz>   * Don't try to create device nodes which already exist
[09:13] <mdz>  -- Matt Zimmerman <mdz@alcor.net>  Wed, 25 Aug 2004 09:10:38 -0700
[09:13] <kiko> okay, sorry about that one then.
[09:13] <mdz> 1.6.0-3 is from unstable
[09:14] <kiko> you're right, apt-cache says it's my fault too.
[09:15] <kiko> I'll wager I have non-warty pcmcia files here too, let me dig into that.
[09:15] <mdz> if you were running unstable before, you really need to do a fresh install if you want to switch to Warty
[09:16] <mdz> it's worth it! :-)
[09:16] <kiko> no, I just picked up some packages from unstable while upgrading I think.
[09:16] <kiko> this was warty from day zero.
[09:16] <kiko> no other OS was ever installed on it :)
[09:18] <mdz> http://wiki.no-name-yet.com/FrequentlyAskedQuestions number 10
[09:25] <debianist> i've just had to fsck my part under deb_sid, and i noticed that usb mouse and keyboard didn't work - their drivers didn't load, might we inherit that as being debian based?
[09:26] <debianist> this is because the drivers don't load on that stage of boot
[09:26] <kiko> the crash in gstreamer is in gst_atomic_int_read
[09:28] <mdz> debianist: at that point, your BIOS needs to be emulating a legacy keyboard
[09:28] <mdz> likewise for grub
[09:28] <mdz> typically there's an option in the BIOS for it
[09:29] <debianist> mdz : hmmm..it might as well be a bios emulation problem. I had it failing on me when i first got the mobo, but it went away eventually
[09:30] <debianist> mdz : however it was like, it worked for grub, just didn't work for saying "Yes" to fsck
[09:37] <seb128> lamont: how is the gdm retry ?
[09:37] <lamont> seb128: successful.
[09:38] <lamont> "powerpc: bringing you inconsistant failures at random times."
[09:38] <lamont> makes me wonder how many non-fatal errors it's getting...
[09:38] <seb128> ok :)
[09:38] <seb128> but weird
[09:38] <mdz> lamont: is it running an SMP kernel?
[09:38] <lamont> From buildd@adare.warthogs.hbd.com  Wed Sep  8 12:55:51 2004
[09:38] <lamont>  Subject: Log for successful build of gdm_2.6.0.3-1ubuntu14 (dist=warty)
[09:39] <lamont> mdz: cat /proc/cpuinfo lists one processor
[09:39] <lamont> so I think it's not SMP
[09:42] <lamont> 350kbps isn't _bad_, but then it's not great..:-(
[09:44] <mdz> lamont: I don't understand why a local mirror is necessary in order to build and test abiword
[09:45] <lamont> mdz: it's not, but the bits come just as fast - since I'm only mirroring what's currently needed/missing
[09:46] <lamont> the benefit of the mirror is that I can snarf bits at 20kbps all day long and not pay for it, having the bits there when I need them.  Fetching as needed would kill my bandwidth bill
[09:47] <lamont> (I just did an apt-get -udy build-dep abiword and cut/edited the 'failed to fetch' list to be the debs I need, and have the rsync going unthrottled.)
[09:47] <lamont> once that's started, I'll throttle it back down to something affordable, and let it spend the afternoon catching up...
[09:47] <lamont> OTOH, I'm fetching large parts of X, which is the bulk of what's out of date in the mirror...
[09:47] <lamont> damn X
[09:48] <mdz> abiword should build just as well with the last 10 versions of X as with the current one
[09:48] <lamont> not when they're not in the cache anywhere...
[09:48] <lamont> I don't have all the -dev packages installed outside of my chroot
[09:50] <lamont> build running now, fwiw
[09:51] <lamont> and only 42 packages out of date in the mirror now...  almost entirely X packages not needed by abiword's build
[09:52] <lamont> where the issue really creeps in is (1) the mirror is constantly updating, and (2) I optimized it for my partial (what I need for chroot-sid to upload debian packages), and caused it to skip a few things.  Noticed that this morning trying to fetch base-config, and fixed it.  But it was behind on warty mirring as a result.
[09:53] <lamont> for that matter, I expect that I could probably get away with just installing the sid version of abiword, but I want the test to be real... :-)
[09:53] <lamont> I didn't see anything that screamed "SCARY" at me in the changelogs
[09:57] <lamont> hrm.. abiword doesn't need xaw[67]  or libdps - clearly it needs to expand. :)
[10:00] <debianist> Kamion : i am testing again #1707 seems that it still won't accept my changes on partman
[10:00] <debianist> Kamion : under QEMU
[10:01] <lamont> mdz: build finished
[10:02] <debianist> Kamion : that's the sounder8
[10:04] <lamont> mdz: any suggestions on things that we want to test beyond just banging around in abiword?
[10:18] <lamont> 827 is a royal pain.. not sure how or even if that can be fixed...
[10:18] <lamont> well, adding a postfix user to base-passwd would fix it..
[10:18] <lamont> or let me fix it, rather.
[10:20] <lamont> if it's really generating the warnings during preconfig, then postfix hasn't had time to create the postfix user, and then the directories in question..  hence when some idiot process tries to send mail from its preinst, the mail falls on the floor.
[10:20] <lamont> postfix does so noisily..
[10:22] <mdz> base-config could preinstall postfix along with mdetect and xresprobe
[10:22] <mdz> whether that's a good idea or not would be a question for Kamion
[10:22] <lamont> ok.  I'll poke him in email
[10:28] <dieman> lamont: h8 that
[10:28] <dieman> lamont: debconf should be configurable to not mail.
[10:28] <dieman> of course
[10:28] <dieman> i bet theres a package that mails outside of debconf
[10:28] <dieman> to be difficult
[10:28] <lamont> yeah...
[10:29] <lamont> mdz: is there any way to get apt to be more verbose about it's decision that a package is not installable?
[10:29] <dieman> i need to put some bugs in on the remote desktops tuff
[10:29] <dieman> it does not like xinerama
[10:29] <dieman> period
[10:33] <mdz> lamont: it tells you the immediate cause; for the nontrivial case it typically says that it's because of something else that is "not going to be installed", in which case you try to install that thing to see why, etc.
[10:34] <lamont> mdz: re 1066 - media removal part (and indeed the rest of the burning) works as root.  eliminating the error (just print and continue) for mortals just lets them get one step further before they die.
[10:34] <lamont> mdz: 270653
[10:34] <mdz> lamont: ok, so the bug is not that media removal no longer works as non-root, the rest of the process doesn't either
[10:35] <mdz> seb128: ping?
[10:35] <lamont> mdz: right.  The inital write fails, IIRC.
[10:36] <lamont> basically, the mortal has no access to the device
[10:36] <lamont> effectively, that is.
[10:40] <mdz> they do have read access
[10:41] <lamont> but write access is gone, despite perms.
[10:42] <lamont> which leads one to expect that something is happenign that needs roots loving before the first write tries to happen..
[10:44] <thom> it's a shame that we didn't declare dbus/hal to be part of the gnome feature goal :-)
[10:45] <mdz> I haven't actually tried dd'ing onto a dvd+rw to see if it works
[10:54] <debianist> mdz : d-i suggests a small /boot partition for grub, otherwise it says it'd fail. why not having something automagically to compute the size ?
[10:54] <debianist> mdz : when trying JFS
[10:55] <mdz> debianist: JFS is not recommended for root for that reason; it displays a warning
[10:55] <debianist> mdz : it asks you manually to do that. or maybe the layman should not be tamper with JFS?
[10:55] <mdz> it's a corner case that we don't need to optimize for
[10:55] <debianist> mdz : k
[10:55] <debianist> mdz : reizerfs install was sweet :)
[10:56] <mdz> thanks
[10:56] <debianist> mdz : how'd make it so fast? (bootup,shutdown,install)
[10:57] <kiko> mdz, btw, the new bootup is significantly faster on this notebook.
[10:57] <mdz> kiko: we optimized a few things
[10:57] <mdz> just recently I rewrote a bit of hotplug which saved me about 4 seconds on boot
[10:57] <kiko> in perl, too. :)
[10:58] <mdz> I had no choice :-P
[10:58] <debianist> perl? hotplug is in c...
[10:58] <debianist> this patches would reach debian eventually? 
[10:58] <debianist> strange, seems there is no app to handle pdfs
[10:59] <mdz> debianist: er, no, hotplug is a bunch of shell scripts
[10:59] <mdz> which is why it was so slow in the first place
[11:00] <debianist> so perl did it faster?
[11:00] <debianist> or did you compailed it natively afterwards?
[11:05] <mdz> it was faster to call out to perl for this specific bit of work than to do it in shell
[11:09] <dieman> is there a db for laptop support?
[11:09] <dieman> ive got a prof with a d600 that i might put ubuntu on if i have to, he wants suspend really bad
[11:09] <thom> no, we've been talking about doing such a thing tho
[11:11] <dieman> god
[11:11] <dieman> this d600 website is talking about a pile of patches to 2.6.8.1 to get acpi working right
[11:12] <thom> we'd love to know if it works :-) seriously though, we're shipping 2.6.8.1 with current acpi patches, so if it works at all in linux it'll work with us
[11:12] <dieman> hmm
[11:12] <dieman> i should try it then
[11:13] <dieman> yeah, if you already have the acpi patches..
[11:13] <dieman> i think that sucker had a broken dsdt too
[11:13] <thom> joy
[11:13] <thom> well, i think jdub has bought a dell laptop
[11:14] <thom> so there might be some synergy there :-0
[11:14] <dieman> oh no
[11:14] <dieman> they fixed it in a recent bios
[11:14] <dieman> A12 now has a fixed DSDT
[11:14] <dieman> yay
[11:22] <carlos> hi
[11:22] <seb128> hey carlos 
[11:23] <carlos> seb128: I have a funny bug with epiphany
[11:23] <carlos> seb128: I cannot visit https://www.networksolutions.com
[11:23] <carlos> it says that epiphany does not supports SSLv2 :-?
[11:23] <carlos> but it works from mozilla
[11:23] <seb128> "You cannot connect to www.networksolutions.com because SSL version 2 is disabled."
[11:24] <seb128> hum
[11:24] <carlos> and I was able to visit it some days ago
[11:24] <carlos> seb128: I see it's not a know bug
[11:25] <carlos> I will fill now a bug report
[11:28] <carlos> seb128: #1112
[11:28] <seb128> ok
[11:29] <seb128> it was working with epiphany ?
[11:29] <carlos> yes
[11:29] <carlos> I think it was last week
[11:29] <carlos> because I was able to update a domain info
[11:37] <seb128> carlos: still here ?
[11:37] <seb128> about:config
[11:37] <carlos> seb128: yes
[11:37] <seb128> security.enable_ssl2 -> true
[11:38] <carlos> seb128: thanks, it works
[11:38] <carlos> but why was it changed ?
 networksolutions sucks! isn't ssl2, like, insecure ?
 crispin: didn't you want to find a site still using ssl2 ? :)
 I wonder if we should re-enable it by default
 don't they offer v3 or tls ?
[11:39] <seb128> 
[11:39] <carlos> X-)
[11:39] <carlos> ok
[11:39] <seb128> carlos: basically
[11:40] <seb128> I've closed the bug report with the settings mention, if you don't mind
[11:40] <thom> carlos: SSL v2 is really, really broken
[11:40] <seb128> we have no real interest to keep it open
[11:40] <carlos> ok, it's fine for me
[11:40] <seb128> thanks
[11:40] <carlos> thom: I didn't know that
[11:42] <carlos> dinner time, later!!
[11:42] <mdz> seb128: is there currently a way to suppress a menu entry without deleting the .desktop file?
[11:42] <seb128> mdz: there is a ShowSomething option for the desktop files yes
[11:43] <seb128> but I'm not sure that's currently supported ... should check with jdub 
[11:43] <seb128> he looked on this
[11:43] <seb128> why ?
[11:43] <mdz> is there an example in Ubuntu?  I have been looking
[11:43] <mdz> I was considering some changes and wanted to test them out
[11:44] <seb128> for a quick test, just removing Application from the Categories= should work :)
[11:46] <mdz> heh, thanks
[11:46] <seb128> np