UbuntuIRC / 2004 /11 /01 /#ubuntu-devel.txt
niansa
Initial commit
4aa5fce
[12:03] <lamont_r> bbl
[12:03] <amu> lamont_r: do you plan another iso ?
[12:03] <amu> .. just took -16 ;)
[12:03] <lamont_r> yes. Plan to burn one with (hopefully) new artwork for grub screen, and definitely WinFOSS 0.4, in about 6 hours.
[12:04] <chrisa> I'm currently installing ubuntu's gnome 2.8 on my sid system in an attempt to see what explodes. This should prove interesting
[12:04] <lamont_r> right now, must run.
[12:04] <lamont_r> chrisa: you're sick, you know.
[12:04] <lamont_r> good luck
[12:04] <amu> lamont_r: cu
[12:13] <chrisa> wow, it worked
[12:16] <amu> try with woody ;)
[12:17] <chrisa> heh
[12:17] <chrisa> I should probably pin this correctly
=== theantix [[email protected]] has joined #ubuntu-devel
[12:20] <amu> if it runs <1 Week, you win a Beer and become the master of moving bits.
=== chrisa looks at the spare hardware in the office and grins
[12:23] <amu> upgrading from sarge, that manages even my 3 year old daughter:)
[12:24] <amu> .. or sid ;)
[12:25] <__daniel> amu, that's what my dog did last night :-)
[12:27] <amu> Ahh, I think you know also storry with the chicken ;)
[12:27] <__daniel> amu, erm... no i dont :-)
[12:28] <amu> which debian installed....
[12:29] <__daniel> amu, not really, but i hope there's someone who tells it to me as a good night tale :-)
[12:29] <amu> no tale, you should ask joey about it ;)
[12:31] <amu> there are much more interessting things, like sending tcp packages with pigeon ;)
[12:33] <__daniel> amu, hehe... yesterday they talked about the good old times in 1903 were the internet was still cool and they had to carry packages up- and downhill even in the snow - that reminded me of those pigeons :-)
[12:38] <__daniel> amu, haha, found the chickens on http://www.linuxjournal.com/article.php?sid=0172 :-)
[12:40] <amu> live is interesting ;)
[12:42] <mdz> Ng: if you're certain that you are correct about it being a duplicate, then it is very helpful for you to mark it
[12:42] <Ng> mdz: ok, I'll make sure I check it thoroughly
[12:42] <Ng> thanks
[12:43] <mdz> Ng: if you are unsure, then feel free to comment
[12:43] <mdz> and someone will review and confirm
[12:43] <Ng> ok, cool :)
[12:44] <Ng> is there policy yet about modifying things to use gksudo/sudo? e.g. gnome-system-tools uses su, but appears to be easily modifyable to use sudo su as a quick hack. would that be frowned upon?
=== jdahlin [[email protected]] has joined #ubuntu-devel
[12:46] <jdub> Ng: our g-s-t uses gksudo
[12:47] <mdz> Ng: things which use su in the Ubuntu desktop are bugs
[12:47] <Ng> jdub: the desktop entries do, but you can make it use internal routines that call su, e.g. the "Configure" button in Applications->System->Network Tool
[12:47] <mdz> Ng: that bug is already reported (twice)
[12:48] <jdub> oh, that's a different module
[12:48] <jdub> and it is bad that it wasn't fixed
[12:48] <jdub> but i don't think anyone realised that was in the new version
[12:48] <Ng> mdz: I know, fessing up, I didn't actually wait for your answer before marking the duplicate ;)
[12:48] <Ng> and I'm part way to a patch for it
[12:49] <mdz> if it's a simple and safe fix, that'd be a good candidate for warty-updates
[12:50] <jdub> mdz: would we consider it 'high-impact', 'dataloss' or 'security'?
[12:50] <Ng> it is looking like it'll be pretty simple, I'm just tying what is mostly a one liner into autoconf. the only snag I have hit is that there is a string that says "root password" that could use rewording and thus translation :/
[12:50] <jdub> it uses it's own su foo?
[12:51] <mdz> jdub: given that two people have run into it already, apparently it's functionality in the desktop that people use, it's broken, and there's a trivial fix
[12:51] <Ng> jdub: it spawns a pty, runs su and the module with it, feeding it the pw
[12:51] <mdz> ok, maybe not so trivial :-)
[12:51] <Ng> src/common/gst-auth.c (last function)
[12:51] <jdub> i *hate* that
=== jdub is reminded to add permissions elevation to HH
[12:52] <Ng> it seems to work just making the su_args array one larger by putting sudo as arg[0] :)
[12:52] <gma> I'm curious about the development model. Do all developers work for canonical, or are you globally distributed?
[12:52] <jdub> gma: all the developers on warty work for canonical, *and* are globally distributed :-)
[12:52] <Ng> I'm currently arguing with dpkg-buildpackage though ;)
[12:52] <jdub> gma: but now there are heaps of community members helping out
[12:52] <gma> cool.
[12:52] <jdub> Ng: hrm, might want to check if they copied code from gnome-system-monitor
[12:52] <gma> are canonical open to applications?
[12:52] <jdub> Ng: daniels fixed that
[12:53] <Ng> jdub: ooh, cool
=== jdahlin [[email protected]] has left #ubuntu-devel ["Leaving"]
[12:53] <mdz> jdub: HH?
[12:56] <jdub> mdz: hoary page
[12:56] <mdz> ah
[12:56] <mdz> jdub: what about permissions elevation?
[12:56] <jdub> attempting to standardise on something
[12:59] <Ng> jdub: it looks similar, but not close enough to just yoink the code over :(
[12:59] <Ng> I'd say they have a common origin, but the g-s-t one is split up more so it can do ssh auth too
=== gma [[email protected]] has left #ubuntu-devel ["Leaving"]
=== elmo_ [[email protected]] has joined #ubuntu-devel
[01:19] <doogie> btopenworld? does that mean they use opensource? :)
[01:20] <Ng> probably not ;)
[01:20] <jdub> morning elmo_
[01:20] <Ng> BT like things that are slow and bad and expensive ;)
[01:21] <Ng> argh
[01:21] <elmo_> hey jdub
[01:22] <Ng> the gnome-system-tools configure script is telling me I need intltool 0.29 or later, but I just installed ubuntu's, which is 0.31 or something. any ideas?
[01:22] <Ng> if I can just make the thing compile I think my patch is done and trivial enough to be safe
[01:24] <Ng> aha, the source package was missing a file or I didn't run enough auto* stuff
[01:25] <__daniel> Ng, what version of g-s-t did you try to ./configure ?
[01:26] <Ng> __daniel: 1.0.0-0ubuntu7
[01:27] <Ng> I just apt-get source'd it, changed configure.in and a .c file and dpkg-buildpackage'd it
[01:27] <__daniel> Ng, that's strange: "apt-get source gnome-system-tools; cd gnome-system*; ./configure" ran fine at my place
[01:27] <Ng> I did change a string in a glade file too, so that and/or the configure.in change probably needed auto* to run more things
[01:28] <Ng> I had to copy intltool-update.in from intltool's install directory to the "backends" directory in g-s-t
[01:28] <Ng> it could easily be my mistake though, it's probably a couple of years since I last did any of this ;)
[01:29] <__daniel> i don't get why they don't have a proper autogen.sh
[01:32] <Ng> *shrug* :)
=== Ng sticks his patch in bugzilla, hopefully the first of many :)
[01:34] <Ng> g-s-t isn't going to work doing remote module without further patching though
[01:35] <Ng> well, if the target machine is ubuntu at least
[01:35] <Ng> since it ssh's as root :/
[01:37] <__daniel> Ng, ssh-as-root should never really be an option in a program
[01:37] <Ng> well it isn't a great idea to spawn a pty and use it to run su to get root, but they do ;)
[01:42] <jdub> ugh
[01:42] <jdub> ok
[01:42] <jdub> totally have to figure otu sound on this machine
[01:46] <sabdfl> lamont: live cd -16 seems to have solved the "see-through desktop" issue on the tosh
[01:59] <mjg59> Oh, wow
[01:59] <mjg59> lamont's been flamed on Advogato
[02:01] <kylem> it's a valid point.
=== lupus_ [[email protected]] has joined #ubuntu-devel
[02:06] <jdub> so what are some of our "all the drivers work but there's no audio output" solutions?
[02:06] <jdub> parallel port is disabled in bios, and modules are not loaded
[02:06] <jdub> audio definitely worked with early ubuntu kernels
[02:12] <mjg59> Is the hardware unmuted?
[02:12] <jdub> yes
[02:13] <jdub> done by ubuntu by default
[02:15] <mjg59> On all channels?
[02:16] <mjg59> Is there a hardware mixer?
[02:17] <jdub> no
[02:20] <mjg59> If you try to play something, does the interrupt number in /proc/interrupts increase?
[02:22] <jdub> yes
[02:22] <mjg59> Any messages in dmesg?
[02:23] <jdub> 130 -> 248 that time ;)
[02:23] <mjg59> Sounds like it's either a mixer issue or you've managed to get two sound drivers loaded...
[02:24] <jdub> there's only the first device's stuff under /dev/snd
[02:24] <mjg59> Hrm.
[02:24] <mjg59> And lsmod only shows one snd-something driver loaded?
[02:24] <jdub> one module in /proc/asound/modules
[02:25] <jdub> no, there's heaps
[02:25] <jdub> but actual drivers could include snd_bt87x
[02:26] <jdub> removed, still the same
[02:26] <mjg59> Uh. bt87x sounds a touch worrying.
[02:26] <jdub> (this machine has a dvb card in it)
[02:26] <mjg59> Yeah
[02:26] <Ng> I've had problems with hotplug loading snd_bt87x before snd_emu10k1 and breaking sound
[02:26] <mjg59> I'd worry that that might have presented a mixer device, and then the wrong stuff may have been loaded
[02:26] <lupus_> what idem Ng
[02:27] <lupus_> idem
[02:27] <lupus_> I mean :)
[02:27] <jdub> i'll try loading snd_intel8x0 in /etc/modules
[02:27] <mjg59> If you remove and then reinsert the correct module and then check the mixer, what does it look like?
[02:27] <mjg59> Ok, that works too :)
[02:27] <Ng> lupus_: eh? ;)
=== jdub did not realise that distrowatch was seen as a big deal
[02:28] <lupus_> ng I mean bttv also broke my sound because it is loaded before the soundcard
[02:28] <Ng> ah :)
[02:28] <Ng> I only had that with debian, for some reason ubuntu gets it right
[02:29] <Ng> I thought it was dependant on PCI ordering because it's hotplug doing it and it works in bus order
[02:29] <lupus_> A friend of my was complaining the other day that he couldn't switch between his 2 soundcards for sound
[02:29] <lupus_> a tool that could do this would also fix this issue :)
[02:30] <jdub> yeah
[02:30] <lupus_> and should be possible on the fly without rebooting :)
[02:33] <jdub> hrmph
=== hornbeck [[email protected]] has joined #ubuntu-devel
[02:42] <__daniel> good night guys
[02:44] <lupus_> are there plans to let hald use fstab-sync to add all the partitions in /etc/fstab (vfat,ntfs etc) ?
=== jamesh [[email protected]] has joined #ubuntu-devel
[03:22] <mdz> mjg59: lamont's been flamed on Advogato?
[03:34] <vorlon> for a documentation-impaired changelog entry.
[03:34] <vorlon> ... by Mathieu Roy.
[03:35] <doogie> ... and the middle-of-the-road "testing" release seems to offer the worst of both "stable" and "unstable."
[03:35] <doogie> (from lwn)
[03:35] <doogie> re: ubuntu and debian
[03:42] <vorlon> that's a curious characterization. I wonder what they think the best parts of unstable are that testing doesn't have -- the RC-buggy packages that will never make it in? ;)
[03:43] <chrisa> Indeed
=== Keybuk [[email protected]] has joined #ubuntu-devel
[04:23] <chrisa> Both hotplug and discover shouldn't need run at the same time, right?
[04:25] <mdz> chrisa: indeed, they _must not_ both run
[04:25] <tseng> 'lo
[04:25] <chrisa> mdz: So which is preferred for a laptop type system?
[04:25] <mdz> chrisa: hotplug is preferred in all cases, and is the Ubuntu default
[04:29] <jdub> morning tseng
[04:30] <jdub> Keybuk: so if we have out of control evolution processes guzzling RAM like nobody's business, what's the most useful tool i can ask a user to run to get some idea of where the problem lies?
[04:36] <lamont> mjg59: url for the flamage?
[04:47] <jamesh> lamont: http://www.advogato.org/person/yeupou/diary.html?start=63
[04:49] <lamont> jamesh: you mean people actually read changelogs.?
=== tseng does.
[04:50] <kylem> apt-listchanges...
[04:50] <jamesh> apparently.
[04:50] <vorlon> lamont: it's all mdz's fault for that fancy apt-listchanges hoowah.
[04:50] <tseng> but im hardly offended by someone refering to a bug #, as a gentoo dev we did that all the time
[04:50] <lamont> heh
[04:51] <lamont> that one was basically one of me going, "well, that one's definitely fixed by now." And there are probably others
[04:51] <tseng> make a keybinding that pipes xclip -out, containing the bug number to the end of a bugzilla search query, and send it to the brower
[04:51] <tseng> "5 minutes" becomes a fraction of a second
[04:51] <jamesh> maybe he expected you to paste the bind release notes into the changelog.
[04:52] <lamont> better yet, I built it with the wrong email address....
[04:52] <tseng> reading on, this guy is a complete tool
[04:53] <tseng> id ignore it
[04:53] <daniels> isn't yeupou mathie roy?
[04:53] <daniels> mathieu, even
[04:53] <vorlon> yep.
[04:57] <daniels> interesting. only 2 peers on BT for i386, and I haven't served a single powerpc or amd64 torrent; contrast with pushing 1.5MB/s (bytes, not bits) for the preview
[04:57] <vorlon> everybody who downloaded the preview died of shock when they found out how good it was? :P
[05:02] <jdub> all the upgraders ;)
[05:08] <daniels> <Culus> my goal is to have people see the download progress meter and
[05:08] <daniels> kill themselves because they know they will never see
[05:08] <daniels> something more eleet than that
[05:08] <daniels> i think that's a pretty good model to be aiming for
[05:08] <mdz> the download progress meter is pretty 31337
[05:10] <jdub> in bittorrent?
[05:11] <daniels> jdub: in apt
[05:12] <daniels> jdub: (culus being the apt maintainer or something, as well as an admin; in true debian style, he is, of course, almost entirely invisible)
[05:12] <jdub> oh, the experimental versions of apt?
[05:12] <daniels> dunno, don't think there's really been a new major apt for quite some time
[05:12] <mdz> jdub: apt's current download progress meter
=== jdub furrows brow, makes dubious glancing looks around the room.
[05:14] <mdz> jdub: dude, it's rad
[05:15] <daniels> it's ill
[05:15] <daniels> how about 'Ill Network Management' as a Hoary goal?
[05:16] <daniels> or 'Ill Link Beat Detection To Make Your Computer Start Up Like Twenty Minutes Quicker When You're Not Plugged In To A Network'
[05:16] <jdub> we'd start saying things like 'network beat box' and confusing the crap out of everyone
[05:16] <kylem> how about 'Ill 802.1X and WPA Integration'
[05:16] <daniels> network beat box! i love it!
[05:16] <daniels> can we please have that?
[05:17] <jdub> fight the power man, all those wireless security protocols are crackrock
[05:17] <tseng> yay for ssh
[05:17] <kylem> it's not about security, it's about authentication. :)
[05:18] <jdub> yeah, and i want to know who you are before i give you any cookies
[05:19] <daniels> jd not *those* cookies.
[05:19] <daniels> oh man. bong.
[05:19] <daniels> daniels@catsby:~/video% ping 192.168.1.1
[05:19] <daniels> PING 192.168.1.1 (192.168.1.1) 56(84) bytes of data.
[05:19] <daniels> From 210.8.1.21 icmp_seq=1 Packet filtered
[05:19] <daniels> From 210.8.1.21 icmp_seq=2 Packet filtered
[05:19] <daniels> maybe ifplugd wasn't quite as phat as I'd hoped.
[05:19] <bob2> apt should use GL to produce a rendered pie graph for completion
[05:20] <daniels> bob2: and it should use COMPOSITE
[05:20] <daniels> or something.
[05:20] <bob2> hahaha
=== daniels decides to research stuff like ifplugd.
=== Mitario [[email protected]] has joined #ubuntu-devel
[05:22] <tseng> NetworkManager will be cool in about 2 years
[05:23] <daniels> i just want /etc/init.d/networking to say 'hey! no cable! ill!' and go along its merry way and not bother me.
[05:23] <jdub> tseng: two years? no way, it's getting lots of attention
[05:24] <tseng> daniels: iirc redhat does that with mii-tool
[05:24] <tseng> checking for cable.
[05:25] <lamont> daniels: that would be rad
[05:26] <jdub> daniels: (what about localhost?)
[05:27] <daniels> jdub: hm?
[05:28] <daniels> jdub: well, just not spend an hour trying to bring up DHCP and NTP
[05:28] <tseng> ctrl plus c, works for me
[05:28] <tseng> but doing it right would be cool.
[05:28] <daniels> yeah, but I usually restart after X crashes, which means I go off to get another glass of water/cup of tea
[05:30] <lamont> daniels: I imagine there's no binary driver bits for ATI rage mobility, true?
[05:31] <daniels> not at all
[05:31] <daniels> were you after a specific feature, or just wondering?
[05:32] <lamont> wondreing
[05:33] <bob2> can't we just let ifplugd do the whole thing?
[05:33] <lamont> trying to figure out how to maximize my radeon 7500 on the desktop as well.
[05:33] <bob2> or does it not notice when something is already plugged in?
[05:33] <lamont> guess I should read the howto.
[05:33] <fabbione> morning guys
[05:33] <jdub> NM > ifplugd (and friends)
[05:33] <bob2> ah
[05:34] <chrisa> hrm, is there a sane way to completely upgrade a sid system to ubuntu?
[05:36] <bob2> you can use pinning.
[05:36] <lamont> chrisa: apt doesn't consider it a true upgrade
[05:36] <bob2> but it's kinda dodgy.
[05:36] <justdave> back up the important stuff, wipe it out, and install from a CD? sid has newer versions than a lot of what's in Ubuntu. It'll be messy.
[05:36] <lamont> since some packages are newer (higher version) in sid than in warty, and vice versa
[05:36] <chrisa> lamont: Right, I've noticed (due to various epochs and version numbers)
[05:37] <justdave> if you want to go that route, it might be safer to wait until Hoary opens for development, and update from the development repository.
[05:37] <justdave> (it'll be a more-recent snapshot of sid, and not frozen yet)
[05:38] <lamont> chrisa: sadly, it's completey unsupportable. woody->warty, no problem. sarge/sid from before 2004-06-28, should be no problem. sarge/sid from after that, install, or go the masochistic route first, and eventually probably do the install anyway
[05:38] <lamont> or wait for hoary to open
[05:38] <chrisa> lamont: I realize that, I'm just experimenting on a sid box I'm not concerned with for fun really
[05:39] <kylem> heh, force downgrade to <2004-06-28 via snapshot.d.n, and upgrade? ;=)
[05:39] <chrisa> I wouldn't do this on a serious system
[05:39] <lamont> ah, well, there is on sane way to cross grade from current sid to warty.
[05:39] <chrisa> kylem: You're nuts!
[05:39] <kylem> thank you.
=== lamont ^5s kylem
[05:40] <lamont> hrm.. still have about 600MB of free space in my dvd tree. what to add....
[05:40] <chrisa> lamont: what is this supposed sane way?
[05:41] <lamont> s/on/no/
[05:41] <chrisa> ah
[05:41] <chrisa> I think this box needs ruining, I'll try kyle's method
=== lamont hands chrisa some rubber gloves
=== chrisa chuckles as he watches this take place
[05:46] <fabbione> lamont: any more testing for the live?
[05:49] <lamont> fabbione: waiting for artwork from jeff
[05:49] <tseng> artwork..
[05:49] <lamont> planned changes from rc to rc2 are: (1) grub screen, (2) new WinFOSS.
=== tseng yawns
[05:50] <lamont> fabbione: if you grab -16 and want to play with a firmware-needing card to see if that works, that'd be way cool
[05:52] <vorlon> daniels: hum, ifplugd always seems to work for me. ... "Packet filtered"?
[05:53] <elmo_> lamont: have you got a firmware for these aironet pcmcia cards you're fond of? the cisco website wants some username/password and doesn't accept the one it gave me
[05:53] <fabbione> lamont: i don't have any firmware based card.. sorry
[05:54] <fabbione> elmo_: already up?
[05:54] <fabbione> or you didn't go to sleep yet?
[05:55] <lamont> elmo_: not sure - I can poke someone tomorrow to see what version they have
[05:56] <jdub> lamont: do you have the grub.conf for the livecd?
[05:56] <lamont> jdub: I expect so.. let me go look
[05:56] <jdub> lamont: or know if it grub itself is modified in any way?
=== vorlon was actually wondering why Warty didn't use ifplugd, and I guess breaking things would be a good reason. ;)
[05:58] <jdub> lamont: oh
[05:58] <jdub> lamont: hold on, it's using gfxboot fork
[05:58] <elmo_> fabbione: not slept yet
[05:58] <elmo_> lamont: cool thanks
[05:58] <lamont> jdub: yes - gfxboot-grub
[05:59] <lamont> elmo_: what version firmware do you have?
[05:59] <lamont> and 350, I assume?
[05:59] <elmo_> Firmware Version: 5.02.19
[05:59] <lamont> jdub: I have the complete gfxboot-grub tree that built the package, if that helps...
[05:59] <elmo_> yeah, 360
[05:59] <elmo_> err 350
[05:59] <lamont> 360?
[05:59] <lamont> ok
[06:01] <daniels> vorlon: yeah, it bonged up my interface
[06:01] <lamont> jdub: http://people.ubuntulinux.org/~lamont/LiveCD/morphix/source/morphix-iso-grubtheme_0.1-3ubuntu4.dsc
[06:01] <lamont> and tar.gz
[06:02] <lamont> jdub: work for you if I go to bed at this point, and plan to build in about 8 hours?
[06:03] <jdub> ok
[06:03] <lamont> jdub: anything else you need before I crash?
[06:04] <jdub> nup, should be ok :)
[06:04] <jdub> thanks
[06:07] <lamont> thanks. artwork hacking isn't my forte and all that...
[06:07] <jdub> hrm, do you stil lhave a copy of the current pcx?
=== lamont decides to start his warty-release DVD burn before he sleeps
[06:07] <lamont> the current pcx is in that source package
=== KeyserSoze [[email protected]] has joined #ubuntu-devel
[06:07] <lamont> (that _is_ the source package that built the current world...)
[06:07] <lamont> o
[06:07] <jdub> rock, ta
[06:08] <lamont> or did you mean the morphix thing?
[06:08] <KeyserSoze> hello
[06:08] <jdub> ha, the morphix one
[06:08] <fabbione> hey KeyserSoze !
[06:08] <KeyserSoze> hey man
[06:08] <fabbione> guys...
[06:08] <fabbione> KeyserSoze has a problem on ubuntu on amd64
[06:08] <fabbione> perhaps someone here can help him?
[06:08] <KeyserSoze> yeah please
[06:09] <KeyserSoze> trying to get an oracle install going
[06:09] <jdub> there's your problem
[06:09] <jdub> oh
[06:09] <lamont> jdub: if you do wnat the morphix one, it's scp-able from chinstrap:~lamont/morphix.pcx
[06:09] <jdub> ;)
[06:09] <KeyserSoze> but we get this error from the jre:
[06:09] <KeyserSoze> current locale is not supported in X11, locale is set to CX locale modifiers are not supported, using defaultException in thread "main" java.lang.InternalError: Current locale is not supported
[06:09] <fabbione> we already checked the locale
[06:09] <KeyserSoze> and no matter what I try it still thinks its on locale CX
[06:09] <fabbione> and it is set properlyu
[06:09] <fabbione> can it be an amd64 glitch?
[06:10] <mdz> what is the locale set to?
[06:10] <mdz> let's take this to #ubuntu, it's not development-related
[06:10] <fabbione> mdz: i told him to join here
=== ph [[email protected]] has joined #ubuntu-devel
[06:11] <mdz> fabbione: why?
[06:11] <KeyserSoze> ok I'll go to ubuntu I don't care
[06:24] <lamont> jdub: you're done with artwork for me already>
[06:24] <lamont> ?
[06:24] <jdub> i hope so
[06:24] <jdub> might look like poo
[06:24] <jdub> aren't you in bed yet? :)
[06:24] <lamont> feh. Now I have to stay up for a while
[06:24] <jdub> heh
[06:25] <lamont> should I care that you sent it twice?
[06:25] <jdub> geez
[06:25] <jdub> that'd be evo
[06:25] <jdub> stupid thing
[06:27] <lamont> hrm... actually, -4 was my screwing around, I think,.
[06:27] <lamont> -3ubuntu3 was really what we're using
[06:27] <lamont> anyway, building -3ubuntu5 now
=== lamont will test the home edition first, in about an hour or so.
=== casche [[email protected]] has joined #ubuntu-devel
=== Ferry [[email protected]] has joined #ubuntu-devel
=== lamont passes the time by burning his 3.7GB warty install dvd
[07:16] <jdub> heh, nice
[07:19] <lamont> all of main, and a little bit of universe. binary and source
[07:19] <lamont> basically, my full mirror + the udebs
[07:19] <lamont> must kill d-i for not dealing well with having Packages.gz instead of Packages....
[07:20] <lamont> this'll be interesting... burning a dvd+rw and a cd-rw at the same time... cdrw drive is slave on the bus with the DVD...
[07:21] <lamont> burning home edition
[07:21] <jdub> have we disabled uploads yet?
[07:22] <lamont> pretty sure
[07:22] <jdub> should probably do it soon
[07:22] <jdub> cool
[07:22] <lamont> was talk of allowing uploads to fix previously-ftbfs packages in universe/multiverse, and I need to upload a new mplayer to multiverse once I'm done with livecd..
[07:22] <jdub> yeah
[07:22] <lamont> right now, it works for anyone with a Xeon, and no one else.:-(
[07:23] <jdub> ouch
[07:23] <lamont> I gave it about 2 minutes today, but that attempt was FTBFS
[07:23] <lamont> basically, it built customized for the buildd hardware, instead of runtime cpu detection
[07:23] <lamont> dvd write went from 2.4x to 0.6x :-(
[07:24] <lamont> otoh, pio-based cdrw write is chunking along quite happily
=== lamont discovered macadamia toffee this week
[07:28] <lamont> cdrw fixating
[07:28] <lamont> dvd getting squat
[07:30] <lamont> jdub: did you touchup the logo at all on the grub screen?
[07:30] <lamont> btw, timer working perfectly.
[07:30] <lamont> not as cute, but definitely working. :)
[07:31] <jdub> lamont: kinda
[07:31] <lamont> ok
[07:31] <jdub> lamont: not as cute as...? how can i make it better?
[07:32] <lamont> jdub: I had officially declared the double bar thing to be "cute".
[07:32] <jdub> ahr
[07:32] <jdub> so now it's just white bars on a brown background?
[07:33] <lamont> http://people.ubuntulinux.org/~lamont/testing/LiveCD/20041021-06/warty-live-i386-20041021-06.iso
[07:34] <jdub> thaytan put release dates on the generated release names :)
[07:34] <lamont> white _bar_ (full width), instead of 2 partially white bars at 1/3 width
[07:34] <lamont> jdub: too much time, I tell ya
[07:34] <lamont> jdub: you have enough bandwidth to test that image?
[07:34] <jdub> no :|
[07:35] <lamont> I'm going to bed, mind you.
=== ..[topic/#ubuntu-devel:lamont] : Ubuntu development -- general discussion on #ubuntu | Happy Warty Day! | 13 (count 'em) major bugs | BE THE SIGNAL | Warty release is DONE | please test http://people.ubuntulinux.org/~lamont/testing/LiveCD/20041021-06/warty-live-i386-20041021-06.iso so it can be warty-rc2-live-i386.iso
[07:36] <jdub> ahr
[07:36] <jdub> okay, will chat to you about making it sexier in the morning
[07:36] <jdub> ooh - could i get a photo? :)
[07:36] <lamont> camera is in the car.
[07:36] <lamont> must I ?
=== calc [[email protected]] has joined #ubuntu-devel
[07:37] <KeyserSoze> can anyone confirm for me that they can run a java app under ubuntu on amd64 using either blackdown or sun jre please?
=== lamont goes to the car
[07:38] <jdub> lamont: nah, don't worry
[07:39] <lamont> ew. bad burn
[07:39] <lamont> have camera, rebooting now
[07:40] <Mitario> wohoo, all trashapplet boogs are fixed
[07:42] <lamont> http://people.u.c/~lamont/dscn1379.jpg
[07:42] <lamont> and the arrows are even where they belong. :-)
[07:43] <jdub> heh
[07:43] <jdub> ok, so, gotta do something with the top left
[07:45] <lamont> "resistance is futile"
[07:45] <lamont> anyway, send me more artwork. night.
[07:45] <jdub> thanks!
[07:45] <jdub> night
=== mdz_ [[email protected]] has joined #ubuntu-devel
[08:17] <Mitario> hmm, it's getting light already :) time to go to bed
=== justdave [[email protected]] has joined #ubuntu-devel
=== jk [[email protected]] has joined #ubuntu-devel
=== x4m [[email protected]] has joined #ubuntu-devel
=== pitti [[email protected]] has joined #ubuntu-devel
[09:12] <pitti> Morning
[09:31] <pitti> mdz_: still here?
=== vorlon_ [[email protected]] has joined #ubuntu-devel
[09:40] <thom> ello
=== |trey| [[email protected]] has joined #ubuntu-devel
=== fabbione kicks X straight in the balls
[09:46] <fabbione> hey thom
[09:46] <fabbione> thom: how was the party yesterday?
[09:46] <thom> it was cool
[09:46] <thom> good bunch turned up
[09:47] <fabbione> nice
[09:49] <pitti> Hi thom
[09:50] <pitti> hi fabbione
[09:51] <fabbione> hey pitti
[09:51] <pitti> fabbione: been at a release party as well? In some LUG?
[09:51] <pitti> Gosh, this thing gets longer and longer
[09:52] <fabbione> pitti: no.. at home sandpapering walls and taking away 20 huge plastic bags of trash from the works
[09:52] <pitti> I mean the channel subject :-)
[09:52] <pitti> fabbione: oh, sounds like exactly the right thing to do after an exhausting release day :-)
[09:52] <fabbione> pitti: of course
[09:52] <pitti> thom: since elmo is not yet here, do you happen to know how and whether the security upload queues work?
[09:53] <thom> fraid not dude
[09:53] <pitti> I just saw that DSA 570-1 and 571-1 are unapplied in Warty
[09:53] <fabbione> pitti: elmo was working on it
[09:53] <fabbione> pitti: i have the packages read for these 2 already
[09:53] <fabbione> pitti: so don't worry
[09:53] <pitti> fabbione: oh fine, I already wanted to prepare some :-)
[09:54] <fabbione> pitti: mdz and I coordinated a while ago
[09:54] <fabbione> before the sec team election
[09:54] <pitti> okay, fine
[09:54] <pitti> then I can throttle my nerves again
[09:54] <fabbione> eheh
[09:54] <pitti> and let my head continue to ache :-/
[09:54] <fabbione> i need to go back to X.org
[09:54] <fabbione> THIS SOURCE IS SO FUCKING INCONSISTENT!
=== pitti does not like to get up at 6 o'clock
[09:54] <pitti> fabbione: oh, speaking of headaches... :-)
[09:55] <pitti> sometimes I already thought that rewriting X from scratch might be faster :-)
[09:55] <pitti> writing some nice drivers to speed up the framebuffer and basically use this :-) </dream>
[09:56] <pitti> fabbione: anyway, I cross my fingers that you and daniel tame the beast
[09:59] <fabbione> pitti: the problem is not the beast itself
[09:59] <fabbione> it's splitting the beast
[09:59] <fabbione> and forward-porting the patches we have
[09:59] <fabbione> to get it to build
[09:59] <pitti> fabbione: I'm curious. Didn't the fd.o version already split components?
[09:59] <fabbione> next step is manage to put everything into nice little tiny debian pacakges
[09:59] <pitti> or was that X.org?
[09:59] <fabbione> pitti: no
[09:59] <fabbione> that's only daniels
[10:00] <fabbione> X.org is still monolitich
[10:00] <pitti> I thought there wer already efforts in this direction
[10:00] <fabbione> and i am going to CRACK IT
=== pitti gives fabbione a huge hammer
[10:00] <fabbione> 'C0Z I 4M 4 L33T H4CK35
[10:00] <pitti> so "crack of the day" has a completely different meaning to you :-)
[10:03] <jamesh> fabbione: would an X display problem that disappears when I boot with acpi=no likely be an X problem, or a hardware problem?
[10:04] <jamesh> (this is on an athlon64)
[10:05] <fabbione> jamesh: yes.. everything can be when it goes to amd64 and X
[10:05] <fabbione> and it can be easily an X problem
[10:05] <fabbione> Xfree86 didn't get much love on amd64 as X.org did
[10:06] <jamesh> fabbione: okay. With ACPI enabled, it displays random garbage, except for the mouse cursor
[10:06] <jamesh> with acpi=no, it works perfectly.
[10:07] <fabbione> jamesh: ok. please open a bug with all the info
[10:07] <fabbione> such as videocard and so on
[10:07] <jamesh> okay.
[10:07] <fabbione> exact models of the laptop, logfiles with both working and nonworking X
[10:08] <fabbione> jamesh: make it a normal severity
[10:08] <fabbione> there is really nothing i can do to make it working at the moment
[10:08] <fabbione> we will have to see with X.org
[10:08] <jamesh> is there an easy way to get the log file from a previous run of the X server?
[10:09] <fabbione> jamesh: try checking /var/log
[10:09] <jamesh> I had to restart the machine when X screwed up
[10:09] <fabbione> afaik there is a backup of the log
[10:09] <fabbione> well please send both
[10:09] <fabbione> i need them to check
[10:10] <jamesh> yes there is.
[10:12] <jamesh> fabbione: interesting. the .old log file says it finds an AGP card, while the current one says it found a PCI card ...
[10:12] <jamesh> I'll restart to make sure I've got the right logs though.
=== TMiegel [[email protected]] has joined #ubuntu-devel
[10:25] <pitti> sjoerd: here?
=== seb128 [[email protected]] has joined #ubuntu-devel
[10:29] <thom> pitti: so that firefox javascript crasher? is apparently much harder to trigger if your LOCALE is en_US *sigh*
[10:30] <seb128> morning
[10:30] <pitti> thom: what, is it still in 0.9.3? Never occurred for me any more since we downgraded
[10:30] <pitti> Hi seb128
[10:30] <thom> no, not in 0.9.3
[10:30] <pitti> thom: or do you prepare packages for 1.0?
[10:30] <thom> just explaining why mdz/lamont couldn't trigger it
[10:31] <thom> (this is from upstream)
[10:31] <pitti> thom: ah, nice idea
[10:31] <pitti> thom: I use de_DE.UTF-8
[10:31] <thom> yeah
[10:31] <seb128> hello pitti
[10:31] <thom> and i'm on en_GB
[10:31] <pitti> they don't like non-Americans :-)
[10:33] <pitti> jdub: can we still add feature goals for Hoary to the Wiki?
[10:33] <pitti> jdub: we sort them out later anyway, but I'd like to drop some ideas there
=== Sledge [[email protected]] has joined #ubuntu-devel
[10:56] <Sledge> morning
[10:57] <thom> hey sledge. get back ok last night?
[10:57] <Sledge> not too bad
[10:57] <Sledge> only got to 2.20am... :-/
[10:58] <Sledge> so I'm feeling really bright and with it this morning
[10:58] <Sledge> :-)
[10:59] <thom> heh
=== lupus_ [[email protected]] has joined #ubuntu-devel
=== sivang [[email protected]] has joined #ubuntu-devel
[11:07] <sivang> morning all
=== dyn [[email protected]] has joined #ubuntu-devel
=== tuo2 [[email protected]] has joined #ubuntu-devel
[11:42] <sabdfl> morning all
[11:42] <dyn> morning :)
[11:42] <seb128> hello sabdfl
[11:43] <tuo2> god morgan, sabdfl
[11:43] <fabbione> hey sabdfl
[11:43] <sabdfl> fabbione!
[11:44] <fabbione> sabdfl: how was the party?
[11:44] <fabbione> sabdfl: i have a good news and a bad news.. which one first?
[11:44] <sabdfl> good news today!
[11:45] <fabbione> sabdfl: the work on X.org is progressing pretty good
=== sabdfl normally prefers the beef before the ice cream
[11:45] <sabdfl> ok
[11:45] <sabdfl> when's daniels due in cph?
[11:45] <fabbione> the bad news is that is much more than what I expected.
[11:45] <fabbione> sabdfl: 1st nov.
[11:45] <sabdfl> ok
[11:45] <fabbione> manly because they reorganized a good portion of the tree
[11:46] <sabdfl> do we have x.org in arch yet?
[11:46] <fabbione> that makes some stuff more complex
[11:46] <fabbione> sabdfl: nope
[11:46] <fabbione> we need xfree86 and x.org in arch
[11:46] <fabbione> for the patch forwarding
[11:46] <fabbione> that's what is actually taking more time than expected
[11:50] <lupus_> not debrix? :)
[11:52] <sjoerd> pitti: pong
[11:52] <pitti> Hi!
[11:52] <pitti> sjoerd: back from class?
[11:53] <pitti> sjoerd: bad news! modifying g-v-m as I thought yesterday does not work
[11:53] <pitti> sjoerd: gnome exports an interface for unmounting a device, but not for mounting it
=== mvo_ [[email protected]] has joined #ubuntu-devel
[11:53] <sjoerd> pitti: just arrived at the uni..
[11:54] <pitti> sjoerd: it has a function for listing all connected drives, but new USB devices don't appear there
[11:54] <pitti> sjoerd: this might get better if g-vfs is compiled with hal support, but until then g-v-m has to call mount on its own
[11:55] <sjoerd> pitti: we'll see
[11:55] <sjoerd> pitti: donno if gvfs with the new hal patches shows all volumes or only the ones in fstab
[11:56] <pitti> sjoerd: BTW, I'm currently ubuntu-fying your hal package; you still conflict to g-v-m << 0.9.10, but it should be << 1.0.2
[11:56] <pitti> sjoerd: will you change that for Debian
[11:56] <pitti> sjoerd: the older g-v-m expected a different storage semantics, which don't work with the newer hal
[11:56] <sjoerd> pitti: for debian 0.9.10 is good enough.. that's why it's still that way
[11:56] <pitti> sjoerd: oh, okay
[11:57] <sjoerd> pitti: debian's 0.9.10 was a cvs, that's why it worked
[11:57] <pitti> sjoerd: you cheated :-)
[11:57] <pitti> sjoerd: no, just kidding
[11:58] <sjoerd> pitti: i'll probably do << 1.0.2 for the new packages then.. people should be using that anyway
=== __daniel [[email protected]] has joined #ubuntu-devel
[12:03] <__daniel> hai
=== elmo_ [[email protected]] has joined #ubuntu-devel
=== seb128_ [[email protected]] has joined #ubuntu-devel
=== carlos [[email protected]] has joined #ubuntu-devel
=== kOoLiNuS [[email protected]] has joined #ubuntu-devel
=== sivang [[email protected]] has joined #ubuntu-devel
[12:59] <kOoLiNuS> hi to everyone!
[12:59] <__daniel> hi kOoLiNuS
[12:59] <pitti> Hi kOoLiNuS
[01:00] <kOoLiNuS> one quick question, can I ?
[01:00] <pitti> you can do everything, kOoLiNuS :-)
[01:00] <pitti> EMISSINGFULLVERB
[01:01] <thom> if you have a question please just ask it
[01:02] <pitti> thom: btw, you hacked on hal a bit, right?
[01:02] <thom> a bit
[01:02] <kOoLiNuS> yes
[01:02] <pitti> thom: does "libselinux-dev build dependency" ring any bell?
[01:02] <kOoLiNuS> i've tried the very first relase of Warty
[01:02] <pitti> thom: I don't know what it was good for and hal builds fine without
[01:02] <kOoLiNuS> and i was disappointed by the "crippled" Gnome-System-Tools
[01:02] <pitti> thom: the Debian package does not have it either
[01:02] <thom> pitti: the reason for the build-dep was to work around a bug in the selinux packages at the time, iirc
[01:02] <pitti> kOoLiNuS: what's missing?
[01:03] <mjg59> kOoLiNuS: That's an upstream decision
[01:03] <mjg59> The tools that aren't supplied are no longer supported
[01:03] <pitti> thom: what has hal to do with these packages?
[01:03] <thom> pitti: (they shipped the .so in the main package, so configure would pick it up and enable selinux support, which would then blow up cos of the lack of headers)
[01:04] <kOoLiNuS> I'm still in the devel ml, so I was wondering if I can "force" the installation of the "complete" package from the testing repos
[01:04] <pitti> thom: darn
[01:04] <mjg59> kOoLiNuS: If you download the source, you can (I /think/) change the configure options in debian/rules and the rebuild the package
[01:04] <mjg59> That's assuming that you can still force the build of the unsupported tools - I'm not sure about that
[01:05] <kOoLiNuS> mjg59: too far difficoult for me :-/
[01:05] <pitti> thom: my hald is not linked against libselinux, so I guess I can safely drop the dependency
[01:07] <thom> pitti: yeah
[01:07] <fabbione> elmo_: are you already awake?
[01:08] <pitti> daniels: here?
[01:08] <seb128> kOoLiNuS: apt-get source gnome-system-tools && apt-get build-dep gnome-system-tools && cd gnome-system-tools-1.0.0, add --enable-boot --enable-services --enable-disks in configure options debian/rules and then dpkg-buildpackage to build the package ...
=== __daniel [[email protected]] has joined #ubuntu-devel
[01:09] <kOoLiNuS> seb128: ok, note taken :-D thanks
[01:09] <seb128> kOoLiNuS: but these modules are not supported and bugged so probably not a good idea
[01:09] <kOoLiNuS> on SID the've worked
[01:10] <seb128> they have been removed
[01:10] <seb128> there is a reason :)
[01:10] <seb128> what are you trying to do ?
[01:10] <kOoLiNuS> seb128: yeah ? didn't know
[01:10] <kOoLiNuS> modify the services started at boot time graphically
[01:10] <kOoLiNuS> or
[01:11] <seb128> kOoLiNuS: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=271859
[01:11] <kOoLiNuS> if I install (L)AMP for toying around with my blog I do not want them to start everytime, I am on a Laptop
[01:12] <elmo_> fabbione: yeah
[01:14] <__daniel> kOoLiNuS, i only have a non-graphic variant for you: cd /etc/init.d; update-rc.d <name-of-service> remove but listen to suggestions of the others, too :-)
[01:14] <kOoLiNuS> seb128: perfect .... some fedora testing friend of mine did tell me something (not accurate as your link) on this
[01:14] <kOoLiNuS> __daniel: if I had the competence to do that I would not miss it :-D
=== kOoLiNuS is going to have lunch
=== kOoLiNuS is now known as kOoL-eating
[01:19] <fabbione> elmo_: will you ping me when ready for the upload? or do you want me to ping you?
=== pitti enjoys his shiny new hal_0.4.0-1ubuntu1
[01:23] <carlos> is hoary repository open already?
[01:23] <carlos> could I move to it?
[01:23] <elmo_> fabbione: oh, right, meh, working on it - I'll ping you in a bit
[01:23] <thom> no and no
[01:24] <elmo_> carlos: no, no
[01:24] <elmo_> thom: copycat
[01:24] <carlos> ok
[01:24] <carlos> :-)
=== Ramsed [[email protected]] has joined #ubuntu-devel
[01:27] <fabbione> elmo_: ok :-)
[01:33] <jamesh> how's this look? http://www.gnome.org/~jamesh/images/drive-mount-applet.png
[01:34] <pitti> jamesh: nice!
[01:34] <pitti> jamesh: you finally made an applet?
[01:35] <jamesh> pitti: yeah. It was a bit fiddly to get the sizing right.
[01:35] <jamesh> but it works pretty well now.
[01:37] <pitti> jamesh: any upload plans already?
[01:37] <jamesh> I still need to do a bit more testing
[01:38] <pitti> jamesh: okay, nice. Will you upload this to experimental/sid as well?
[01:39] <jamesh> pitti: I'm planning on getting it merged into gnome-applets (since it is meant to replace one of the existing applets). I haven't considered packaging it separately
[01:39] <pitti> jamesh: oh right, even better
[01:50] <fabbione> daniels: you around?
=== sivang [[email protected]] has joined #ubuntu-devel
[02:01] <carlos> so, finally we are not going to have our final release in the main page of slashdot?
[02:04] <lamont> jdub?
[02:07] <thom> carlos: no
[02:08] <carlos> :-(
[02:08] <thom> *shrug*, they have given us like 4 stories in the last month
[02:09] <sabdfl> carlos: no, they said the main page had had enough ubuntu coverage
[02:09] <carlos> :-?
[02:09] <carlos> so a final release cannot be there but a Xandros betatesting process can
[02:09] <carlos> funny
[02:14] <chrisa> Why isn't trashapplet in gnome-applets?
[02:17] <sabdfl> chrisa: we picked it up before it had yet gone mainstream
[02:17] <sabdfl> but expect it to become part of the main gnome release
[02:17] <chrisa> ah
[02:18] <seb128> it's already in gnome-applets cvs head
[02:21] <fabbione> guys is very familiar with libraries that can help me 2 minutes?
[02:25] <amu> Kamion: btw. lolo synced it :)
[02:25] <Kamion> amu: cool
[02:26] <__daniel> just wrote a mail to heise.de to make them cover the ubuntu release :-)
[02:26] <amu> http://source.rfc822.org/pub/mirror/releases.ubuntu.com/warty/
[02:28] <fabbione> well.. i meant who is very familiar...
[02:28] <seb128> fabbione: you should just ask ...
[02:28] <seb128> usually the "who is very .. with ..." kind of questions don't get a lot of replies
[02:30] <fabbione> seb128: well i need to understand how to split a lib in general
[02:31] <fabbione> `-- lib
[02:31] <fabbione> |-- libFS.a
[02:31] <fabbione> |-- libFS.so -> libFS.so.6.0
[02:31] <fabbione> |-- libFS.so.6 -> libFS.so.6.0
[02:31] <fabbione> `-- libFS.so.6.0
[02:31] <fabbione> libFS.a -> libfs-dev
[02:31] <fabbione> libFS.so.6.0 -> libfs6 ?
[02:31] <fabbione> and the other 2?
[02:31] <Kamion> yes; libFS.so -> libfs-dev, libFS.so.6 -> libfs6
[02:32] <fabbione> ok thanks!
[02:32] <fabbione> that makes it simple
[02:32] <fabbione> :-)
[02:32] <fabbione> i already have enough headackes splitting X to dig into each single piece
=== fabbione does really appreciate
[02:34] <seb128> .a / .la / .so -> -dev
[02:34] <seb128> .so.x and .so.x.y.z -> lib
[02:35] <fabbione> thanks seb :-)
[02:36] <seb128> you're welcome :)
[02:40] <fabbione> now the big question is.. which directory should i start polluting if i kill /usr/include/X11 and /usr/X11R6 ?
[02:41] <pitti> fabbione: you kill /usr/X11R6?
=== pitti praises fabbione
[02:42] <pitti> fabbione: why not just put the stuff in /usr/lib/X11 and the executables in /usr/bin?
[02:42] <pitti> fabbione: IMHO /usr/bin can't be made worse, pollute-wise
[02:42] <Kamion> I thought that was roughly the plan
[02:43] <fabbione> pitti: i need somewhere where to store the includes
[02:43] <fabbione> and nothing is allowed to stick <whatever>/X11
[02:43] <pitti> fabbione: but why do you want to kill /usr/include/X11?
[02:43] <pitti> fabbione: it seems fairly "canonical" to me :-)
[02:43] <fabbione> pitti: it's a synlink to ../X11R6/inlucde
[02:43] <fabbione> pitti: see also policy
[02:44] <Kamion> you can't kill /usr/include/X11 as such; people do #include <X11/foo.h>
[02:44] <pitti> fabbione: right, I mean why not put the includes in /usr/include/X11 directly?
=== lamont takes kids to school. bbiab
[02:44] <fabbione> Kamion: i know that
[02:44] <Kamion> but you're going to run into dpkg hell trying to replace a directory with a symlink
[02:44] <fabbione> Kamion: i need an alternate location
[02:44] <Kamion> er, vice versa
[02:44] <fabbione> Kamion: it is already a symnlink
[02:44] <Kamion> 13:44 < Kamion> er, vice versa
[02:44] <fabbione> http://www.debian.org/doc/debian-policy/ch-customized-programs.html#s11.8.7
[02:45] <fabbione> Packages must not provide or install files into the directories /usr/bin/X11/, /usr/include/X11/ or /usr/lib/X11/.
[02:45] <Kamion> yes, I know
[02:45] <Kamion> I was involved in that policy discussion :P
[02:45] <fabbione> and i don't want to go against policy
[02:45] <fabbione> but neither i can pollute *
[02:46] <Kamion> that bit of policy was taken in order to smooth the path for a future move of XFree86 to /usr
[02:46] <Kamion> s/taken/created/
[02:46] <fabbione> Kamion: correct
[02:46] <Kamion> I do not think that you should regard it as constraining X.org packages
[02:46] <fabbione> but now that i am moving X to /usr
[02:46] <fabbione> Xfree86 or X.org.. still the same stuff
[02:46] <fabbione> it's an Xserver
[02:46] <Kamion> the other alternative is /usr/X.org and keep the symlinks, would be simpler, but you don't get to lose the old hack
[02:47] <fabbione> or X Windows System
[02:47] <Kamion> I *know* :)
[02:47] <Kamion> ok, "I do not think that you should regard it as constraining XFree86 or X.org packages"
[02:47] <fabbione> ahh sorry
[02:47] <fabbione> i misread
[02:47] <fabbione> but neither i want to start another mess creating /usr/x.org
[02:48] <fabbione> makes no sence to me
[02:48] <fabbione> at that point i could just keep /usr/X11R6
[02:48] <fabbione> without messing around
=== fabbione needs to see 2 minutes of sunshine in the hope that God will kiss his forehead
[02:57] <fabbione> i will have to discuss it with Branden
[02:57] <fabbione> i don't see a clean solution to this
=== Ubuntu-Linux [[email protected]] has joined #ubuntu-devel
[03:09] <Ubuntu-Linux> i reported another bug
[03:09] <Ubuntu-Linux> https://bugzilla.ubuntu.com/show_bug.cgi?id=2601
[03:09] <Ubuntu-Linux> is the person who email addy [email protected] here?
[03:09] <Kamion> yes
[03:09] <Ubuntu-Linux> who is it?
[03:10] <Kamion> me
=== Ubuntu-Linux want to poke that person
=== Ubuntu-Linux pokes Kamion
[03:10] <tseng> Ubuntu-Linux: poking on bugs works, bugzilla sends out an email with changes
[03:10] <Kamion> if you're objecting to the severity change, I downgraded it as a routine bug management issue because problems on a single piece of hardware don't justify the 'blocker' severity.
[03:10] <Ubuntu-Linux> Kamion: tree.. had to help me when you where the one in charges shess
[03:11] <Kamion> I can't help everybody individually.
[03:11] <Kamion> there's this small matter of "time"
[03:11] <Ubuntu-Linux> i know
[03:11] <Kamion> I also don't know what your problem is, and I was put off helping you by statements like "Ubuntu messed with d-i".
[03:12] <Ubuntu-Linux> but you where there at the same time that i ask and where active
[03:12] <Ubuntu-Linux> they did
[03:12] <Ubuntu-Linux> it now says ubuntu when you boot up
[03:12] <Ubuntu-Linux> that even called messing
[03:13] <azeem> Ubuntu-Linux: crap, Ubuntu writes 'ubuntu' instead of 'debian' when it boots up?
[03:14] <Ubuntu-Linux> i change the aphla d-i i used to say my name
[03:14] <Ubuntu-Linux> i even messed with d-i
[03:14] <Ubuntu-Linux> 0_0
[03:14] <Ubuntu-Linux> who hasnt?
[03:15] <bob2> people with a job and/or hobby.
[03:15] <azeem> I don't understand your problem (but then, I didn't read the bug-report either)
[03:15] <Ubuntu-Linux> what a geek which doesnt mess with his installer to say his name?
=== tseng doesnt
=== chrisa doesn't
[03:15] <Kamion> please take the chatter off #ubuntu-devel, thanks.
=== __daniel shakes his head in disbelief
[03:15] <Ubuntu-Linux> azeem: read it
[03:15] <tseng> Ubuntu-Linux: hey bud
[03:16] <tseng> Ubuntu-Linux: the developers have a nice, prioritized list of bug reports. so if you could just hang in patiently, someone will get around to yours
[03:16] <tseng> it doesnt help your case much by pestering people directly
[03:16] <Ubuntu-Linux> i have d-i BUG
[03:16] <tseng> i have a kernel BUG
[03:16] <Kamion> so far the pestering means it's at the bottom of my priority list
=== Ubuntu-Linux want to install ubuntu
[03:17] <chrisa> Ubuntu-Linux: You're not listening
=== Ubuntu-Linux doesnt want to d/l d-i and do a net install
[03:17] <Ubuntu-Linux> i know
[03:18] <Ubuntu-Linux> but Kamion is in charge of d-i and person x is in chage of x
=== stevey [[email protected]] has joined #ubuntu-devel
[03:18] <azeem> Ubuntu-Linux: send a patch, like everybody else does
[03:18] <Kamion> Ubuntu-Linux: guess how many bugs that means I have
[03:18] <Ubuntu-Linux> Kamion which is in charge of y doesnt do x
[03:18] <chrisa> On a sidenote, I really hope "ubuntu-linux" is a default irc name for the clients and that he didn't actively choose that nick
[03:18] <chrisa> But that's just me...
[03:18] <Ubuntu-Linux> Kamion: 2601?
[03:18] <Kamion> chrisa: it's not, as far as I know
[03:19] <bob2> chrisa: it's not
=== Ubuntu-Linux own ubuntu-linux
[03:19] <bob2> chrisa: he was in #debian as GNU-Debian and shimon for a while, too
[03:19] <chrisa> bob2: sigh
=== tseng sighs and goes to class
[03:19] <tseng> g'day boys and girls
[03:19] <thom> cya tseng
=== kOoLiNuS [[email protected]] has joined #ubuntu-devel
[03:22] <Ubuntu-Linux> ok so how long till i can get ubuntu installed?
[03:23] <__daniel> Ubuntu-Linux, TRY to be patient - you won't achieve anything by pestering people
[03:23] <Ubuntu-Linux> or should i just d/l d-i and upgrade to ubuntu?
[03:23] <Ubuntu-Linux> __daniel: i know
[03:23] <Ubuntu-Linux> i am waiting
[03:23] <Ubuntu-Linux> i just want to know if i should goto sleep tonight and forget about it till sunday
[03:23] <azeem> Ubuntu-Linux: no, you are pestering. Waiting is without the 'how long till...?' part
[03:24] <chrisa> Yes, do that
[03:24] <chrisa> If you really cared, you'd just install woody and upgrade
[03:27] <Ubuntu-Linux> i cant install woody
[03:27] <Ubuntu-Linux> i need to install from 2.4.26 or later
[03:27] <chrisa> Then go to sleep
[03:27] <Ubuntu-Linux> because of drivers
[03:28] <Ubuntu-Linux> so should i d/l d-i net install?
[03:28] <pitti> Ubuntu-Linux: why you can't download the ISO?
[03:28] <Ubuntu-Linux> and add the ubuntu sources and upgrade?
[03:28] <__daniel> Ubuntu-Linux, try to
[03:28] <chrisa> Do what you want to do, staying here and pestering Kamion will get you nowhere
[03:28] <Kamion> Ubuntu-Linux: please do not upgrade the severity of that bug report again.
[03:28] <Ubuntu-Linux> which one pitti
[03:28] <pitti> Ubuntu-Linux: recently I installed a CD-ROM less laptop with the netboot images, that went fine
[03:28] <Ubuntu-Linux> Kamion: i didnt
[03:29] <pitti> Ubuntu-Linux: which one? the release ones?
[03:29] <Ubuntu-Linux> i just repily
[03:29] <Ubuntu-Linux> i got no network
[03:29] <Ubuntu-Linux> but i can do debian d-i netinst
[03:29] <pitti> Ubuntu-Linux: then go ahead with that
[03:30] <Ubuntu-Linux> ok will there be any problem if i install from debian sarge d-i
[03:30] <Kamion> [email protected] changed:
[03:30] <Kamion> What |Removed |Added
[03:30] <Kamion> ----------------------------------------------------------------------------
[03:30] <Kamion> Severity|normal |blocker
[03:30] <Ubuntu-Linux> pitti: i am froced to use manbrake
[03:31] <bob2> you're not forced to do anything
[03:31] <bob2> if you can install sarge, you can move to ubuntu
[03:31] <pitti> Ubuntu-Linux: you can't download the ISO from anywhere?
[03:31] <Ubuntu-Linux> Kamion: what setting should it be on my side
[03:31] <Kamion> Ubuntu-Linux: I changed it to normal. Please leave it there.
[03:31] <Ubuntu-Linux> pitti: my hdd died and it was the only cd i had i d/l and burned ubuntu
[03:32] <pitti> Ubuntu-Linux: so you _have_ an Ubuntu CD?
[03:32] <Ubuntu-Linux> Kamion: yea but on my side its still blocker and p5
[03:32] <Ubuntu-Linux> yes
[03:32] <pitti> Ubuntu-Linux: an older one perhaps?
[03:32] <Ubuntu-Linux> the final
[03:32] <Ubuntu-Linux> no latest
[03:32] <pitti> Ubuntu-Linux: ah, and this doesn't boot for you?
[03:32] <pitti> Ubuntu-Linux: you are the G5 guy?
[03:32] <Ubuntu-Linux> it a fscking driver problem
[03:33] <Ubuntu-Linux> it does boot for me
[03:33] <Kamion> Ubuntu-Linux: reload.
[03:33] <Ubuntu-Linux> ptti MIND reading my bug
[03:33] <Kamion> 00:1f.2 IDE interface: Intel Corp. 82801EB Ultra ATA Storage Controller (rev 02)
[03:33] <Kamion> that's strange, I can't see that in /usr/share/misc/pci.ids
[03:33] <Ubuntu-Linux> ohh
[03:33] <Ubuntu-Linux> wait
[03:34] <Ubuntu-Linux> i enabled emation of sata as pata
[03:34] <Ubuntu-Linux> does that matter?
=== Ubuntu-Linux reboots goes from bios and see if it helps
[03:34] <Kamion> hm, maybe it's 808624db
[03:35] <Kamion> which is ide/piix in discover
[03:35] <Ubuntu-Linux> *because i was using a 2.4 kernel and no FULL sata support
[03:35] <Ubuntu-Linux> Kamion: so should i disable emalation
[03:35] <Kamion> worth a try
[03:35] <Ubuntu-Linux> ok
[03:35] <Ubuntu-Linux> brb
[03:35] <Kamion> also check that the piix and ata_piix modules are loaded
[03:35] <Ubuntu-Linux> if i am not back in 4min then its working
[03:36] <Ubuntu-Linux> ok
[03:37] <dyn> please.. dont..
[03:38] <__daniel> Kamion, i admire your patience :-)
[03:38] <Kamion> I don't have much of it left
[03:38] <__daniel> Kamion, i think you did quite well
=== Sledge applauds Kamion for restraint
=== AndyFitz [[email protected]] has joined #ubuntu-devel
[04:09] <fabbione> i think we will live with X11R6 until there is a better solution
[04:10] <fabbione> killing it now isn't an option without polluting *
[04:10] <azeem> better solution than X11R6 or better solution than what you thought up till now?
[04:10] <fabbione> azeem: read above :-)
[04:11] <fabbione> by policy we can't install into */X11
[04:11] <azeem> oh, missed that, sorry
[04:13] <fabbione> azeem: if you have better ideas, please say so
[04:13] <azeem> then change policy :)
[04:13] <fabbione> it's not like i am closed mind
[04:13] <fabbione> azeem: that will take too long :-)))
[04:13] <azeem> well, I don't know about the dpkg limitation for using /usr/include/X11 directly, but that would be the cleanest IMHO
[04:14] <fabbione> and few flames all over
[04:14] <azeem> /usr/bin for binaries and perhaps a subdir for /usr/lib, whatever the name
[04:14] <fabbione> azeem: there are no limitations in using X11 in general
[04:14] <fabbione> but if i start creating subdirs around..
[04:14] <azeem> well, the part about replacing a symlink
[04:14] <fabbione> i can just live with X11R6
[04:15] <fabbione> see.. either we integrate everything inside the FHS (considering X11R6 not fHS compliant)
[04:15] <fabbione> or otherwise it's not worth the mess to just rename 2 directories/symlinks
[04:15] <azeem> having subdirs in /usr/{include,lib} is common practise for libraries
[04:15] <fabbione> (considering the amount of packages that use then)
[04:15] <azeem> having subdire in /usr is not
[04:16] <fabbione> true.. i don't disagree on this
[04:16] <azeem> perhaps you could swap it around and live compatiblity symlinks in /usr/X11R6 (pointing to /usr/{include,lib}/X11 for a while?
[04:17] <azeem> and then kill them off eventually
[04:17] <azeem> (like /usr/doc, hahaha)
[04:17] <fabbione> azeem: eventually = undetermined amount of time.
[04:17] <fabbione> no i am not that nice with other maintainers :-P
[04:17] <azeem> one release
[04:18] <azeem> which means, half a year for ubuntu and an undetermined amount of time for Debian
[04:18] <fabbione> azeem: another release = 6 months here and something between 4 to 10 years in debian
[04:18] <fabbione> exactly :P
=== azeem ^5s fabbione
=== AndyFitz [[email protected]] has left #ubuntu-devel []
[04:18] <fabbione> no i think i will leave X11R6 for now
[04:18] <fabbione> the changes are too deep to be done by a single/two persons
[04:18] <fabbione> we need a team working on it.
[04:19] <fabbione> (without considering the amount of changes that the code requires)
[04:19] <azeem> what's wrong with swapping the symlinks around? Is there a technical problem with that?
[04:19] <fabbione> this definetly has to be done upstream
[04:19] <azeem> at least, that would set a signal that X11R6 is deprecated
[04:19] <fabbione> azeem: probably swapping them no.
=== __daniel [[email protected]] has left #ubuntu-devel ["Verlassend"]
[04:19] <Kamion> swapping the symlinks around requires some very hairy preinst code and probably has undetermined compatibility implications ...
[04:19] <fabbione> but i wonder when X12R1 will be out...
[04:20] <fabbione> and a big API change will take place
[04:20] <fabbione> it's going to be the hell storing includes all in /usr/include/X11
[04:20] <fabbione> we need to be able to differenciate them
[04:20] <fabbione> at least...
[04:21] <azeem> well, X12R1 will have /usr/include/X12, no?
[04:22] <Kamion> X11 and X12 will *have* to be co-installable
[04:22] <Kamion> at least the libraries and preferably development packages
[04:22] <azeem> yeah
[04:23] <azeem> if you want the static libs to be parallel installable, but them in a subdir of /lib
[04:23] <azeem> eh, /usr/lib
[04:23] <bob2> X12 is on the cards?
[04:23] <azeem> how do the other distributions handle this? Most of them have switched to x.org by now, havent' they?
[04:23] <fabbione> uh true :)
[04:24] <fabbione> azeem: yes. but i doubt they have remove X11R6
[04:25] <azeem> we're at X11R8 or so right now?
[04:25] <fabbione> 6.8.1
[04:25] <azeem> eh, aren't we at X11R8 or so right now?
[04:25] <azeem> ah
[04:25] <fabbione> X11R6
[04:25] <azeem> sorry
[04:25] <azeem> so, X12 on the radar, or X11R7?
[04:25] <fabbione> nothing on the radar atm
[04:25] <fabbione> i am just pondering for the future
[04:26] <azeem> what compatibility promises have the X guys taken?
[04:26] <azeem> will they break binary compatibility with X11R7?
[04:26] <fabbione> they shouldn't...
[04:26] <fabbione> afaik
[04:26] <azeem> godd
[04:26] <azeem> eh, good
[04:27] <fabbione> i can't remember if RX was for the binary compatibility and X11 the version of the protocol
[04:27] <fabbione> so new version of the protocol breaks world
[04:27] <fabbione> and binary compatibility kinda
[04:30] <bob2> when was the last time X broke compatibility?
[04:30] <bob2> er, ABI.
[04:31] <fabbione> can't remember
[04:31] <fabbione> probably 3.3 -> 4.0
[04:31] <bob2> fabbione: did you have much to do with X before joining the XSF?
[04:32] <azeem> fabbione: what about you ask keithp for his opinion? Isn't he a DD now?
[04:32] <fabbione> bob2: no
[04:32] <fabbione> anyway removing /usr/include/X11 is not an option
[04:32] <bob2> hah, wow, you got sucked in quick :-)
[04:32] <fabbione> some x packages build-deps on packages that pulls in other x packages
[04:33] <fabbione> including stuff in /usr/include/X11
[04:33] <fabbione> azeem: i am still thinking... and more i think, more i am convinced that upstream is the first one that should do it
[04:33] <azeem> I don't see a need to remove /include/X11. Would you just dump everything in /include?
[04:34] <fabbione> azeem: if i dump everything in /usr/include it will be a mess
[04:34] <Kamion> azeem: you'd have to change every X program
[04:34] <fabbione> i still need a X11 symlink
[04:35] <Kamion> getting rid of /usr/include/X11/ would be broken
[04:35] <fabbione> #include <X11/foo.h> 0wns you
[04:35] <fabbione> and switching it from a symlink to a dir will make X or X.org unbuildable
[04:35] <fabbione> wow
[04:36] <pitti> fabbione: right, so what exactly is wrong with /usr/include/X11? Just because it's a symlink?
[04:36] <fabbione> i guess we will live with X11R6 :-)
[04:36] <fabbione> pitti: no
[04:36] <pitti> fabbione: okay, you just answered.
[04:37] <azeem> fabbione: at least mandrake still uses X11R6, I just asked
[04:37] <azeem> probably the others as well. So I guess it's the right way for now
[04:39] <fabbione> yeps
=== fabbione reverts the changes
[04:41] <azeem> fabbione: anyway, why would switching /usr/include/X11 from a symlink to a dir make X or X.org unbuildable?
[04:41] <fabbione> azeem: because almost all the documentation requires groff
[04:41] <fabbione> and groff pulls in several X stuff
[04:42] <fabbione> including things in /usr/X11R6/include
[04:42] <azeem> oh, circular Build-Depends?
[04:42] <fabbione> yeps
[04:42] <azeem> suck
[04:42] <fabbione> if we fuck up one upload nobody will build anything for a long while
[04:43] <Kamion> uh, the groff source package has documentation of how to avoid the X build-dependency
[04:43] <Kamion> but yes, groff is just one example of the zillion programs that want /usr/include/X11/
[04:44] <fabbione> Kamion: than i will need a xgroff to build x
[04:44] <Kamion> ?
[04:44] <fabbione> groff simply depends on these packages
[04:44] <Kamion> no you don't, you build groff with DEB_BUILD_OPTIONS=groff-no-x11, then you build X, then you build groff properly
[04:44] <fabbione> Kamion: so you need a temporary groff around
[04:45] <Kamion> yes, welcome to bootstrapping
[04:45] <Kamion> it's only gxditview that needs X11
[04:46] <fabbione> Kamion: it would be easier for me to create a anti-x-groff and upload it :-)
[04:47] <Kamion> and die shortly afterwards :)
[04:47] <Kamion> tetex is no different
[04:48] <Kamion> you need it for some of the docs too AIUI, but it build-deps on X
[04:48] <fabbione> yeah i also build-dep on tetex
[04:48] <fabbione> and X build-deps on it
[04:48] <azeem> well, I believe this circular Build-Dep is not a very good reason to keep X11R6 alive, in case we agree it should in principle die
[04:48] <fabbione> cool, isn't it?
[04:48] <Kamion> groff doesn't pull stuff in from /usr/X11R6/include/ directly though, it uses the symlink
[04:48] <elmo_> fabbione: okay, the security stuff is in theory done - if you don't mind, I'd like to wait for mdz to ack it before you start using it
[04:49] <fabbione> elmo_: sure i don't mind
[04:49] <kylem> wow. that installation went well. good work guys.
[04:49] <fabbione> elmo_: i am not sure for how long i can stay around
[04:49] <fabbione> elmo_: in the worst case we will do tomorrow
[04:50] <elmo_> fabbione: well, can you put your upload somewhere so matt could upload, it and try the new mechanism out?
[04:50] <fabbione> elmo_: otherwise i can just handover the packages to pitti since he will take care of security
[04:50] <fabbione> elmo_: sure
[04:53] <fabbione> azeem: anyway i have a very limited amount of time
[04:53] <fabbione> azeem: i need a decision asap
[04:53] <fabbione> and i think that for the next 6 months we can live with X11R6
[04:53] <fabbione> we had it around for ages
[04:53] <fabbione> 6 months more or 6 months less won't kill anybody
[04:54] <fabbione> and etch will not be released by that
[04:54] <azeem> fabbione: well, I'd talk to keithp and the rest of the x.org maintainers whether they ponder changing that general X11R6 prefix
[04:56] <fabbione> azeem: i know daniels did all these changes in his tree
[04:56] <fabbione> azeem: but the problem is i dunno how robusts they are
[04:56] <fabbione> and he only built xc/lib/
[04:56] <fabbione> not xc/*
[04:56] <azeem> maybe it really is the best to wait for the modular X before killing X11R6
[04:57] <fabbione> yeah
[04:57] <fabbione> i agree
=== azeem_ [[email protected]] has joined #ubuntu-devel
[05:01] <mdz_> elmo_: what kind of ack do you need?
[05:02] <elmo_> mdz: err, see your mail? just that your happy with it? also, per your instructions, you're the only one who can run amber atm
[05:02] <mdz_> elmo_: no, just got up, haven't read it yet
[05:02] <elmo_> ok
=== wartylog [[email protected]] has joined #ubuntu-devel
=== Topic for #ubuntu-devel: Ubuntu development -- general discussion on #ubuntu | Happy Warty Day! | 13 (count 'em) major bugs | BE THE SIGNAL | Warty release is DONE | please test http://people.ubuntulinux.org/~lamont/testing/LiveCD/20041021-06/warty-live-i386-20041021-06.iso so it can be warty-rc2-live-i386.iso
=== Topic (#ubuntu-devel): set by lamont at Thu Oct 21 07:35:46 2004
[07:19] (daniels/#ubuntu-devel) the problem with Debian is that they're not progressive enough. if they opened it up to, say, hip-hop artists and dnb producers ...
[07:19] (Kamion/#ubuntu-devel) hence, also, the sounder@ mailing list
[07:19] (daniels/#ubuntu-devel) doogie: but humour nevertheless, and it's what we have.
[07:19] (doogie/#ubuntu-devel) daniels: *the* problem with debian?
[07:21] <azeem> everything else would be solved instantly if we s/etch/rakim/
[07:22] <bob2> mix master etch
[07:23] <lamont> Kamion: apparently d-i doesn't like empty Packages files either...
[07:23] <Kamion> lamont: empty Packages files where?
[07:23] <lamont> my mirror
[07:24] <Kamion> d'oh
[07:24] <doogie> bob2: don't start
[07:24] <doogie> maybe there should be theme music with each release?
[07:24] <lamont> I didn't tell it to mirror anything from multiverse, but did tell it about the component. --> 0 length Packages file. And a coaster. :-(
[07:24] <daniels> doogie: we ship an ubuntu-sounds package with default sounds
=== lamont lunches
[07:24] <doogie> daniels: not what I meant.
[07:24] <bob2> doogie: it would combine the best features of Debian and OpenBSD
[07:24] <doogie> I mean each release has a custom mix track, or some such
[07:25] <Kamion> lamont: cdimage doesn't mirror universe or multiverse at all ...
=== ph [[email protected]] has joined #ubuntu-devel
[07:36] <mdz_> lamont: is 1021-06 the latest live CD candidate?
=== hornbeck [[email protected]] has joined #ubuntu-devel
[07:44] <amu> mdz_: yes
=== T-Bone [[email protected]] has joined #ubuntu-devel
[07:44] <T-Bone> hi
[07:48] <mdz_> downloading it to test
[07:51] <T-Bone> Kamion: ping?
[07:52] <Kamion> T-Bone: yep?
[07:53] <T-Bone> Kamion: do you want to do the "setup" now or tomorrow?
[07:54] <Kamion> T-Bone: now's fine, just going to grab a bite to eat
[07:54] <T-Bone> ok
[07:54] <T-Bone> i'll wait ;)
[07:54] <Kamion> but tell me what you need and I'll catch up
=== x4m [[email protected]] has joined #ubuntu-devel
[07:55] <T-Bone> Kamion: actually i need _you_ to tell me what you need ;)
[07:59] <Kamion> hm, ok
[08:00] <T-Bone> that includes material and tasks you'd like me to work on ;)
[08:00] <Kamion> well, the other day I think I said that I'd like to be able to drop in a new netboot kernel and initrd, reboot remotely from that image, and have remote access to its console.
[08:00] <T-Bone> so you need dhcp server
[08:00] <Kamion> right
[08:00] <T-Bone> ok
[08:01] <T-Bone> i have a server already setup, i'll give you access to it
[08:01] <T-Bone> so if you just need that and the ia64 box, i need a ssh2 public key, and a login
[08:02] <Kamion> the other things we need to put together for d-i are: linux-kernel-di-ia64-2.6 package based on the Ubuntu kernel, ports of all the bootloader installer and partitioner component packages from Debian, and probably additions of ia64 to a few lists
[08:02] <Kamion> login name cjwatson, where can I mail the key?
[08:02] <T-Bone> [email protected]
[08:03] <Kamion> it'll be a while before I can actually start using this for testing of course, so the buildd work is higher-priority
[08:03] <T-Bone> yeah i got that
=== plovs [[email protected]] has joined #ubuntu-devel
[08:04] <T-Bone> unfortunately i need lamont's skills now, cause i hit something strange that _shouldn't_ have happened. We are waiting for him to complete stage1 as well and see if he has the same problem
[08:04] <hornbeck> hey plovs
[08:04] <Kamion> mailed
[08:04] <T-Bone> thx
[08:04] <Kamion> will need to figure out what needs to change in debian-installer to port the ia64 initrds to Ubuntu
[08:04] <Kamion> I only did the changes for the arches we support
[08:05] <T-Bone> k
[08:05] <Kamion> think it should all be fairly easy, anyway, amd64 wasn't hard
[08:06] <Kamion> i386 and powerpc were hard because they were the first ones. :)
[08:06] <T-Bone> hehe
[08:07] <Kamion> partman-efi, efi-reader, and elilo-installer are clean of possible debconf-priority damage, too; good
[08:07] <T-Bone> yeap
[08:07] <Kamion> but elilo-installer will need to be Ubuntu-branded.
[08:07] <T-Bone> right
[08:10] <mdz_> justdave: ping?
[08:11] <Kamion> hm, and partman-efi needs branding too
[08:11] <Kamion> joy and rapture
[08:11] <T-Bone> hehe
=== _danny__ [[email protected]] has joined #ubuntu-devel
=== _danny__ is now known as sivang
=== lupus_ [[email protected]] has joined #ubuntu-devel
[08:22] <hornbeck> plovs: I agree that Mulligan is doing good doc work
[08:23] <mdz_> lamont, amu: 1021-06 looks good to me
[08:23] <justdave> mdz_: pong
[08:23] <plovs> yeah, we might ask hi to join, join the wiki-team page in a how-to use the wiki and wiki notespage
[08:23] <mdz> justdave: is it possible to change the default on the "find a specific bug" page to search all bugs, open and closed?
[08:23] <mdz> justdave: and just have it sort open bugs ahead of closed ones?
[08:24] <hornbeck> plovs: I agree, if we go and split a small team like doc already we are in for trouble
[08:24] <justdave> changing the default on the open/closed state to all is easy. changing the default sort can probably be done, but isn't so easy
[08:24] <sivang> So, a wiki team under doc team? :)
[08:25] <hornbeck> sivang: I am suggesting we just try to get him to work with us, and that the wiki maintanance should just go along with with docs
[08:25] <plovs> under sounds bad, let's say the wiki-team will spearhead the doc-team
[08:25] <hornbeck> I say no wikiteam
[08:25] <hornbeck> there is a doc team
[08:25] <hornbeck> we do docs
[08:25] <hornbeck> wiki is a doc
[08:25] <sivang> hornbeck : who is he?
[08:26] <plovs> agreed
[08:26] <sivang> who are we talking about? :)
[08:26] <hornbeck> sivang: KevinMulligan
[08:26] <plovs> http://wiki.ubuntulinux.org/KevinMulligan
[08:26] <hornbeck> if he is here please speak up
[08:26] <plovs> there was a thread on the mailinglist also let me find it ... brb
[08:26] <hornbeck> ok
[08:26] <plovs> ask in #ubuntu
[08:27] <hornbeck> I will search real quick
[08:27] <mdz> justdave: can you change the open/closed default, and look into the sort?
[08:27] <mdz> justdave: let me know if you'd prefer I filed a bug about this request
[08:27] <sivang> ok, I read his page
[08:28] <hornbeck> plovs: his mail must have come when I changed to my new computer
[08:28] <hornbeck> I have a reply but that is it
=== sivang is way lagged after the list. checking now.
[08:29] <seb128> mdz: have you tested the current liveCD ? The localisation is fine ? I've downloaded it yesterday and the french localisation is broken ...
[08:29] <plovs> i love gmail, found it in 3 seconds
[08:29] <mdz> seb128: I did not test French localisation
[08:29] <mdz> but it works for me in the default english locale
[08:29] <seb128> could you test ?
[08:30] <seb128> but I get such messages "locale: Cannot set LC_MESSAGES to default locale: No such file or directory"
[08:30] <seb128> and the apps (panel, evo, ...) are not in french
[08:30] <plovs> DOC it seems it is only one guy, we might just send him a mail
[08:30] <plovs> or we can write on his page, maybe i'll do that if you guys agree
[08:30] <hornbeck> plovs: do you want to mail him?
[08:31] <seb128> and LANG=fr_FR@euro according to "locale"
[08:31] <hornbeck> plovs: that might be good
=== maskie [[email protected]] has joined #ubuntu-devel
[08:31] <mdz> seb128: /usr/lib/locale is empty
[08:31] <mdz> seb128: no locales are generated
[08:31] <plovs> sure i'll mail him and ask to add himself to the doc-team
[08:31] <hornbeck> plovs: tell him also that we think it might be bad to make multiple teams when it comes to docs
[08:31] <justdave> ok, turns out the sort order was trivial to fix, after all.
[08:31] <justdave> all done
[08:32] <mdz> thanks
[08:32] <hornbeck> man beagle is rocking
[08:32] <justdave> it was specified in the form, not in the cgi :)
[08:32] <mdz> seb128: I get the same errors
[08:33] <seb128> ok
=== justdave tweaks it to sort on resolution instead of status
[08:33] <mdz> seb128: it gets EACCES opening /usr/lib/locale/locale-archive
[08:33] <justdave> so all open bugs will still be sorted by relevance
[08:33] <mdz> seb128: after I've run locale-gen
[08:33] <justdave> regardless of status
[08:33] <seb128> hum, weird
[08:33] <mdz> seb128: it is mode 600
[08:34] <mdz> seb128: so I changed it to 644
[08:34] <mdz> seb128: and now evolution is in French
[08:34] <seb128> ok
[08:34] <hornbeck> sivang: is the gnome-guide mainly what you are working on?
=== inklingx [[email protected]] has joined #ubuntu-devel
[08:34] <mdz> seb128: what procedure did you use to set your locale on the live CD?
[08:34] <mdz> seb128: dpkg-reconfigure locales?
[08:35] <seb128> mdz: no, I've just picked Submenu -> Supported languages in the boot screen
[08:35] <sivang> hornbeck : currently yes, I Have also pending works on the wiki, would appriciate if you could continue with it a bit
[08:35] <seb128> and then french
[08:35] <mdz> or is there something at the morphix boot prompt which sets it?
[08:35] <mdz> ah
[08:35] <mdz> I'll check if that has the same problem
[08:35] <hornbeck> sivang: continue with the guide?
[08:35] <seb128> I'm restarting the liveCD on a box right now
[08:35] <sivang> hornbeck : yes, if you could I'd appriciate it.
[08:35] <seb128> mdz: in fact a french user pinged me about this. According to him it was working fine with the version is download some days ago
[08:36] <hornbeck> sivang: just resend to me and I will see what I can do
[08:36] <sivang> hornbeck : ok, do you want diffs also, or do you want to wait for 2.8.1 docs altogether maybe?
[08:36] <hornbeck> sivang: just send me your diffs
[08:36] <hornbeck> or the whole folder
[08:37] <hornbeck> does not matter
[08:37] <sivang> ok
[08:37] <plovs> hornbeck: sivang what do you guys think about the kind of layout for wiki-pages?
[08:37] <hornbeck> plovs: are you talking about a standard?
[08:38] <sivang> the current layout is not good enough?
[08:38] <hornbeck> plovs,sivang: I think we need to standardize all the pages, to use same formats
[08:39] <plovs> pages look all a little different and we just began, it is now or never
[08:39] <sivang> plovs : gemme some examples
[08:39] <plovs> when we have 2000+ pages it might be a little late
[08:39] <hornbeck> plovs: they all need to look the same
[08:39] <plovs> sivang: just a sec
=== vorlon_ is now known as vorlon
[08:40] <mdz> seb128: confirmed, I get English if I boot with the French option
[08:40] <mdz> err
[08:40] <mdz> never mind
[08:40] <mdz> seb128: it works
[08:40] <seb128> oh ?
[08:40] <mdz> it's just that the computer menu is not localised
[08:40] <hornbeck> plovs: it will be impossable to make them identical, but we can at least make the standards the same
[08:40] <mdz> nor is "applications" or "computer"
[08:40] <mdz> but the menu items under applications are localised
[08:40] <seb128> mdz: yes, these are .desktop files
[08:40] <seb128> not po files
[08:40] <seb128> nothing to do with the locales
[08:40] <mdz> ah
[08:40] <mdz> and evolution is not localised
[08:40] <seb128> yes
[08:41] <mdz> and it has given me a French keyboard layout it seems :-)
=== carlos [[email protected]] has joined #ubuntu-devel
[08:41] <plovs> sivang: compare http://wiki.ubuntulinux.org/BasicCommands with http://wiki.ubuntulinux.org/BeagleInstallHowto
[08:41] <mdz> seb128: it is the same bug
[08:41] <plovs> different styles
[08:41] <plovs> not big differences but stil...
[08:41] <hornbeck> plovs: both are my pages :)
[08:42] <mdz> seb128: /usr/lib/locale/locale-archive is -rw-------
[08:42] <hornbeck> plovs: I have not changed the basiccommands page yet
[08:42] <mdz> seb128: please file it in bugzilla
[08:42] <plovs> hornbeck: :) i know
[08:42] <plovs> hornbeck: i like the beaglepage
[08:42] <hornbeck> plovs: I have been working on the beaglepage nonstop because new stuff keeps happening
[08:42] <seb128> mdz: ok, thanks
[08:43] <hornbeck> plovs: I need to remake the whole BasicCommands page
[08:43] <plovs> it rocks, i want to do it all the time, but have no time
[08:43] <hornbeck> plovs: I know how that is
[08:44] <plovs> sivang: what if we make the beaglepage layout the basic layout?
[08:44] <sivang> plovs : I like it better, yes.
[08:44] <hornbeck> plovs: I think that would work
[08:44] <plovs> we can make an empty howto page called DocumentationHowto and point to it
[08:44] <sivang> well, have a look at /HowDoc
[08:45] <sivang> hornbeck : If I recall right your comments about the laytout, you want everything that needs be typed to be in a "code" box right?
[08:45] <hornbeck> sivang: yes
[08:46] <plovs> sivang: it is easier to copy and paste from there
[08:46] <hornbeck> sivang, plovs: it makes it easier to copy and paste sections
[08:46] <plovs> and it looks nice :)
[08:46] <hornbeck> :)
[08:46] <plovs> sivang: but whatever we do it looks best if it is the same
[08:47] <sivang> ok, I suggest we streamline it as so. I will modify HowDoc accordingly
[08:47] <hornbeck> plovs, sivang: is someone going to note all this stuff?
[08:48] <sivang> hornbeck : I am , see the line before :)
[08:48] <hornbeck> sivang: I read right as I was hitting enter
[08:48] <plovs> sivang: nice, will you make DocumentationHowto?
[08:49] <plovs> or should I do it?
[08:49] <hornbeck> HowDoc is basicly a DocumentationHowto
[08:49] <hornbeck> http://wiki.ubuntulinux.org/HowDoc
[08:49] <hornbeck> maybe needs a name change?
[08:49] <hornbeck> since we are mainly going with Howto at the end of howto's
[08:51] <plovs> what we can do is making DocumentationHowto and use [[IncludePage()] ] to put it inside HowDoc
[08:51] <hornbeck> plovs: good idea
[08:51] <hornbeck> plovs: is there a good wiki howto out there/
[08:51] <hornbeck> ?
[08:51] <hornbeck> I am learning wiki markup as we go
[08:52] <plovs> hornbeck: brb
[08:52] <hornbeck> ok
[08:52] <plovs> hornbeck: http://wiki.ubuntulinux.org/HelpIndex
[08:52] <plovs> hornbeck: especially HelpOnMacros
[08:53] <plovs> hornbeck: and HelpOnMacros
[08:53] <hornbeck> plovs: thanks
[08:54] <plovs> hornbeck: the include is a little broken you can only include whole pages
[08:54] <plovs> hornbeck: moin 1.3 will/should solve that
[08:54] <hornbeck> ok
[08:54] <hornbeck> should we just rename that page?
[08:54] <hornbeck> than just add new stuff to it
[08:55] <hornbeck> have a '''wiki howto''' '''docbook howto''' all on the same page?
[08:55] <hornbeck> a stop shop
[08:55] <plovs> look at CategoryCategory , you can just make an automatic list
[08:56] <hornbeck> you lost me
[08:57] <hornbeck> link
[08:58] <plovs> if you look at the page-source you'll see it's just two lines
[08:58] <plovs> so you can make a HowtoPage with an automatic links to all howto's
[08:58] <plovs> something like that
[08:58] <hornbeck> ahh
[08:58] <hornbeck> ok
[08:59] <hornbeck> but the main question is, should we just be HowDoc into DocumentationHowto
[08:59] <hornbeck> sivang?
[08:59] <hornbeck> I think it makes more since
[08:59] <sivang> yes, I think of HowDoc to be both for offline stuff we do, and wiki content guidline. have 2 sections on that.
[09:00] <hornbeck> sivang: a rename though?
[09:00] <sivang> oh, you want it to have HowTo at the end
[09:00] <sivang> yes
[09:00] <sivang> :)
[09:00] <plovs> hornbeck: a howto is usually short and to the point (kernel howto is too long)
[09:00] <sivang> for the catrgorization to work.
=== x4m [[email protected]] has joined #ubuntu-devel
[09:00] <plovs> hornbeck: a document is long and detailed
[09:00] <plovs> hornbeck: i *think*
[09:00] <hornbeck> plovs: DocDoc than?
[09:00] <hornbeck> :)
[09:01] <hornbeck> plovs: kernel howto is long no matter what
[09:01] <plovs> what about ending doc-stuff in Doc or Document?
[09:01] <plovs> hornbeck: there are three kernel-pages already, that needs cleaning up
[09:01] <hornbeck> DocumentationDoc?
[09:01] <hornbeck> plovs: I noticed today
[09:02] <plovs> WritingGuidelinesDoc ?
[09:02] <hornbeck> plovs: the one I made was made real nice
[09:02] <hornbeck> plovs: that works
[09:02] <hornbeck> shoudl the kernelhowto be made KernelInstallDoc?
[09:04] <plovs> yes that's better i think
[09:04] <hornbeck> make it so
[09:04] <sivang> btw, I think the wiki team falls nicely under the "documentation sounder team" accoridng to the outlined tasks it carries
[09:04] <plovs> sivang: what do you think Howto pages and Doc pages?
[09:05] <sivang> plovs : well, actually I think that big docs should be shorty incorporated into PLone CMS, and the wiki should be acting more of a bleeding edge corner and docdevel works
[09:06] <hornbeck> sivang: we are not at that area yet, but what do you think for right now
[09:07] <sivang> hornbeck : yes it's good, then we should have 2 indexes for Guides / Howtos. Like on http://www.debian.org/doc/
[09:08] <hornbeck> sivang: I agree
[09:10] <plovs> sivang: i agree, but in wiki, they can be worked on.
[09:11] <plovs> what about something like: http://wiki.ubuntulinux.org/AlexanderPoslavsky_2fPlayGround
[09:11] <plovs> and then one for the docs
[09:11] <plovs> feel free to mess up that page
[09:12] <plovs> hornbeck: how good are you at python?
[09:12] <hornbeck> plovs: decent
[09:12] <hornbeck> I do alot of it for my jobs
[09:13] <hornbeck> plovs: why you ask?
[09:13] <plovs> hornbeck: i played around with a moin2docbook.py thingie but i'm just starting with python
[09:13] <hornbeck> plovs: did you write it?
[09:14] <plovs> hornbeck: i wrote open file, close file check parameters ... so no, not yet
[09:14] <plovs> hornbeck: i am trying to understand classes
[09:14] <hornbeck> plovs: if that is something you would like to work on together, send me what you have and we can work on it
[09:15] <hornbeck> brb have to order books for my classes
[09:17] <plovs> sivang: yes, like the debian guys
[09:18] <sivang> plovs : actually the python classes are the nicest and most straightforward I've seen compared to Java, C++ etc
[09:19] <plovs> sivang: that says something about my programming :(
[09:19] <sivang> plovs : no, you might no thave any other introduction to OOP langs, so that might explain it :)
[09:20] <plovs> sivang: it is the first time, and i need more time to read it
[09:20] <plovs> sivang: how difficult is it to make yelp docs?
[09:20] <sivang> sivang : but you'll catch on fast, if you programmed before - take "Dive into python" by Mark Pilgream , very good
[09:21] <sivang> plovs : actually, you just learn the DTD and that's it. You're writing XML
[09:21] <plovs> sivang: i have it installed :)
[09:21] <hornbeck> ok, I am back
[09:21] <plovs> sivang: if we could write this converter then we could just convert wiki pages (they are really basic) to XML
[09:21] <sivang> plovs : you can do some work on the manual, it teaches you about the dtd as you see al those tags for representing menu choises, items.
[09:22] <plovs> sivang: *the* manual?
[09:22] <sivang> plovs : Yes, I have talked with Enrico about that - We might have the Plone CMS team to do this for us maybe
[09:22] <sivang> :)
[09:22] <plovs> sivang: duh, what manual?
[09:23] <hornbeck> plovs: we can work on it, if someone else does it, it will at least give you some experiance with python
[09:23] <sivang> plovs : yes, the gnome official needs some adjustments and modification to nicely follow Ubuntu's desktop looks and actions.
[09:23] <hornbeck> plovs: also Docbook is easy
[09:23] <hornbeck> just follow what is already there, for most manuals
[09:23] <plovs> sivang: i am trying to concentrate on two things, my wife is explaining diets and recipes to me from the second computer
=== justdave [[email protected]] has joined #ubuntu-devel
[09:24] <hornbeck> plovs: I understand that :)
[09:25] <plovs> hornbeck: how would you go about converting a file from moin to docbook? i have a look for line in file with lot of re.match
[09:25] <hornbeck> I have wife and two kids, who talk to me while I do all this
[09:25] <hornbeck> plovs: I would have to start hacking on it honestly, I am not a good enough programmer to explain it off the top of my head
[09:26] <hornbeck> I am more of a hit or miss guy, learn as I go
[09:27] <plovs> hornbeck: well, i'll write some more and thn send it, it would be nice to get a converter
[09:27] <hornbeck> plovs: cool
[09:27] <hornbeck> plovs: if you want to learn docbook work on the gnome-guide with sivang
[09:27] <hornbeck> its a good starting place
[09:28] <plovs> sivang: can yelp use non-local =internet files?
[09:28] <hornbeck> yelp just reads DTD as far as I know
[09:28] <sivang> plovs : lemme ask it's developer for a sec :)
[09:29] <plovs> sivang: do you hve some simple task?
[09:30] <sivang> plovs : no it can't, and won't.
[09:30] <plovs> sivang: less is more, i should have guessed, this is not kde
[09:30] <hornbeck> sivang: are you in #docs?
=== hornbeck [[email protected]] has joined #ubuntu-devel
[09:34] <sivang> hornbeck : yes
[09:34] <plovs> sivang: i could take HowDoc from common approach, and put it in DocHowto, and the
[09:34] <plovs> n link it back
[09:34] <sivang> plovs : ok
[09:39] <plovs> hornbeck: sivang what page do we use for suggestions to the documentation team?
[09:40] <hornbeck> hmm
[09:40] <hornbeck> DocSuggestions
[09:40] <hornbeck> hows that sound?
[09:42] <mdz> does anyone here have a copy of vmware?
[09:42] <hornbeck> plovs: sivang: are we renameing HowDoc and the KernelHowto?
[09:42] <hornbeck> I have 4.5 for linux somewhere mdz
[09:42] <mdz> hornbeck: if it is the eval version, is it possible for me to get a copy?
[09:43] <mdz> I'm trying to track down this bug: https://bugzilla.ubuntu.com/show_bug.cgi?id=2492
[09:43] <hornbeck> mdz: you can download the eval version
[09:43] <mdz> not without registering on their website, getting an evaluation key, etc.
[09:44] <mdz> I just want to look at some of the files
[09:44] <mdz> and see if it is doing what I suspect it may be
[09:44] <hornbeck> I don't have a eval version
[09:44] <hornbeck> I can get the eval version for you if you want
[09:44] <mdz> can you look and see if it changes the samba startup links?
[09:45] <hornbeck> I will have to install, I have it on disk right now
[09:45] <hornbeck> I got it along time ago, when I still used windows some
=== Mitario [[email protected]] has joined #ubuntu-devel
[09:46] <sivang> hornbeck : what's the software ?
[09:47] <hornbeck> vmware
[09:50] <sivang> mdz : I am running samba on this machine, should I try and upgrade and test it?
[09:50] <hornbeck> mdz: I can send to you with reg code if you would like
[09:50] <hornbeck> I just started download and got registered
[09:50] <pitti> elmo_, mdz: can I put the security packages somewhere, so that either of you can upload them? I'm going to bed soon
[09:54] <sivang> I have upgraded samba while it was still running, wasn't able to reproduce.
[09:55] <hornbeck> mdz: I now have the eval copy with reg code. Do you want me to send to you?
[09:55] <hornbeck> I am not running samba so I would not be able to test
[09:57] <hornbeck> plovs: you still around?
[10:00] <hornbeck> sivang?
[10:01] <sivang> hornbeck : yes
[10:01] <hornbeck> ok, no one was answering so I was wondering if I was still connected
[10:01] <hornbeck> :)
[10:02] <hornbeck> well I am off to work
[10:02] <hornbeck> night
[10:03] <sivang> night hornbeck
=== mxpxpod [[email protected]] has joined #ubuntu-devel
[10:06] <mxpxpod> chrisa: ping
=== ploum [[email protected]] has joined #ubuntu-devel
[10:09] <ploum> I just want to thanks all developpers and Mark for the wonderful Warty release. Good job guys, it's really great !
[10:12] <plovs> hornbeck: sorry, yes
[10:13] <hornbeck> plovs: I wanted to say keep up the good work :)
[10:13] <plovs> hornbeck: goodnight :) !
[10:13] <hornbeck> night
[10:13] <plovs> sivang: good night!
=== maskie [[email protected]] has left #ubuntu-devel ["Leaving"]
[10:13] <sivang> plovs : night
[10:15] <mdz> ploum: thanks :-)
[10:16] <hornbeck> mdz: you want the vmware?
[10:16] <mdz> hornbeck: sure, thanks
[10:16] <mdz> sivang: if you have vmware installed, yes, that would be a good test
[10:17] <sivang> mdz : installing in the background
[10:18] <mdz> hornbeck: dcc won't work; I'm behind a firewall here
[10:18] <hornbeck> how can I get it to you?
[10:18] <mdz> hornbeck: I just finished drafting a short howto on how to use the community support resources, in the wiki
[10:19] <mdz> HowToGetHelp
[10:19] <mdz> hornbeck: I'd be interested in your feedback
[10:19] <hornbeck> ok
[10:19] <sivang> mdz : I'll give it a look also :)
[10:19] <hornbeck> mdz: you get my pm?
[10:21] <hornbeck> mdz: nice use of eric raymond site
[10:22] <hornbeck> mdz: adding a link to the doc howto at the bottom would help when saying "or how about a howto article" and have the link to the dochowto
[10:23] <mdz> hornbeck: by all means
[10:24] <hornbeck> mdz: added
[10:24] <mdz> thanks
[10:24] <hornbeck> well all, I really have to go now
[10:24] <hornbeck> goodnight
[10:25] <mdz> I've tried the vmware eval download about 20 times and it's no good
[10:25] <mdz> I'll keep fiddling :-)
[10:25] <mdz> night
[10:25] <hornbeck> hmmm
[10:25] <hornbeck> keep working at it :)
=== hornbeck is now known as hornbeck|away
[10:25] <theantix> any idea why there is no php4-common in Ubuntu like there is in Debian Sid? some universe packages depend on it... any developer I can talk to about the reasons for that or any workaround?
[10:25] <mako> hornbeck|away: great stuff
[10:26] <sivang> mdz : you tried the tarball?
[10:26] <mdz> sivang: no, I'm trying to download it
[10:27] <mdz> I was hoping to find someone who already had a copy and could poke around and see if it does anything nasty to /etc/rc?.d
[10:27] <sivang> mdz : I suppose it does. It always added stuff to bootup scripts when I Used it on sarge,sid and RH
[10:27] <mdz> sivang: ls -l /etc/rc?.d/K09samba
[10:28] <mdz> if that file exists, that's a good clue
[10:28] <sivang> mdz : after I install it :) I will
[10:28] <mdz> ah
=== lamont_r [[email protected]] has joined #ubuntu-devel
[10:30] <lamont_r> seb128 around?
[10:31] <sivang> mdz : "read the FAQ" is broken on Plone, I opened a bug on it on bugzilla but till isn't corrected
[10:31] <sivang> mdz : the link you pointed from the wikie
[10:31] <seb128> lamont_r: yes
[10:32] <lamont_r> how does /usr/lib/locale/fr get populated?
[10:32] <mdz> sivang: fixed the link, thanks
[10:32] <lamont_r> and btw, the french keyboard is just plain screwed up.
[10:32] <sivang> mdz : no prob. this is also referenced from _within_ the plone somewhere.
[10:32] <lamont_r> I mean layout-wise, that is
[10:33] <seb128> lamont_r: /usr/lib/locale/locale-archive you mean ?
[10:33] <lamont_r> the errors from strace indicate that it's tryting to open /usr/lib/locale/fr/LC_*
[10:34] <seb128> lamont_r: have you read your mails (#2292) ?
[10:34] <seb128> oups
[10:35] <seb128> 2614 I mean
[10:35] <lamont_r> not yet
[10:35] <seb128> the problem is /usr/lib/locale/locale-archive being 600
[10:35] <seb128> (thanks mdz for finding this)
[10:36] <seb128> chmod 644 it and that works fine
[10:36] <lamont_r> on the liveCD?
[10:37] <seb128> yes
=== lamont_r adds a 'umask 0' to the build process, and pushes things to the top of the hill
=== mdz gapes
[10:37] <mdz> lamont_r: that file is created dynamically during boot
[10:37] <mdz> must be
[10:37] <lamont_r> ah, ok.
[10:38] <mdz> not that umask 0 would be a good idea _anyway_...:-)
[10:38] <lamont_r> means I'll have to go find that.
[10:38] <seb128> it is, depending of the locale selected in the menu
[10:38] <lamont_r> just knew that i have 022 , which still wouldn't explain it...
[10:38] <amu> lamont_r: problem ( warning ) with booting, umount: /.dev not found ; rmdir: invalid option --f; can't create /var/lib/dhcp3/dhclient.leases: Read-only file system; alsactl: load_state:1134: No soundcards found ...
[10:38] <lamont_r> is anything getting 077 as a umask?
[10:38] <lamont_r> amu: we're ignoring the leases file issue
[10:38] <lamont_r> but rmdir -f could use some staring at
[10:38] <mdz> that file wouldn't be there after a reboot anyway :-)
[10:39] <amu> lamont_r: i took screenshots
[10:39] <lamont_r> right - that's because we switched from pump to dhclient3...
[10:39] <lamont_r> amu: cool
=== lamont_r is sitting at the mechanics and using dialup.
[10:39] <mdz> something must be calling localedef or locale-gen somewhere in the boot process
[10:39] <lamont_r> hence testing is gonna be a bitch
[10:39] <mdz> based on the lang= boot parameter
[10:39] <lamont_r> mdz: yes.
[10:40] <amu> grubsplash you should wait min. 60sec. instead of 5
[10:40] <lamont_r> right before it starts pcmcia
[10:40] <lamont_r> amu: hell no
[10:40] <lamont_r> use case is enduser, not technical people.
[10:40] <lamont_r> end user doesn't want the 1 min wasted time
[10:40] <amu> lamont_r: process bar is ok now ;)
[10:40] <lamont_r> this was discussed wrt the installed box quite a whileback
[10:40] <lamont_r> amu: jdub fixed that last night
[10:41] <lamont_r> mdz: do we care that gnome takes forever to come up under some situation involving no link?
[10:41] <lamont_r> link beat, that is.
[10:41] <amu> lamont_r: ok, we should note that in a part of the faq bootoptions *ducks*
[10:41] <mdz> lamont_r: depends on how long forever is
[10:41] <lamont_r> < 5 minutes
[10:41] <lamont_r> just long enough to go WTF and reboot.
[10:42] <amu> uploading screenshot
[10:42] <mdz> lamont_r: do you realize that sources.list on the CD contains people.ubuntu.com/~lamont entries?
=== lamont_r notices that elmo used a large club in locking down warty
[10:42] <lamont_r> mdz: yes
[10:42] <lamont_r> but that's ok, because apt-get update exits immediateluy
[10:42] <lamont_r> :-)
[10:43] <lamont_r> http method is busted, alex blames part of the kernel hacks
[10:44] <mdz> SIGSEGV
[10:44] <lamont_r> yeah - in select on a socket
[10:44] <lamont_r> "I was suspicious of that socket code"
[10:44] <lamont_r> --Alex
[10:46] <mdz> what socket code?
=== lamont_r looks
=== lamont_r forgpt the package name
[10:47] <lamont_r> mdz: minifo
[10:48] <lamont_r> or rather, probably minifo-2.6.7
[10:48] <mdz> isn't that the filesystem overlay thing?
[10:48] <mdz> I do not see what that has to do with TCP sockets, which are what that program is using
[10:50] <lamont_r> sockets are fd's, which go through the filesystem...
[10:50] <amu> http://amu.debian.net/tmp/l5.png
[10:50] <lamont_r> and alex really thought the socket handling was, um, suspect
[10:50] <lamont_r> mdz: think unix domain sockets
[10:50] <mdz> I agree, it probably is
[10:50] <mdz> but apt doesn't use unix domain sockets
[10:51] <lamont_r> but the sockets code gets involved anyway
[10:52] <mdz> you're saying that mini_fo mangles generic socket code?
=== mdz runs away
[10:52] <mdz> no wonder that thing will never go near upstream
[10:52] <mdz> where can I get a copy of that patch?
[10:54] <lamont_r> mdz: nfc - I didn't even look at it, other than 'http method dies with segv, and strace, pick on alex'. see p.u.c/~lamont/LiveCD/morphix/source
=== amu detects a bugs bombarding
[10:56] <mdz> I didn't realize you had noticed that major things were broken and weren't filing bugs about them
[10:57] <lamont_r> 1) didn't see apt not working as major - it's a livecd. 2) the last bug I asked someone to file I got told "please dont"
[10:58] <lamont_r> the localization bug was mentioned last night - investingating today.
[10:59] <mdz> lamont_r: that was regarding "the live CD has fundamentally different hardware detection" which is a design issue
[10:59] <lamont_r> true
[10:59] <lamont_r> but still a WONTFIX bug.
[11:00] <lamont_r> or at least NOTWARTY
[11:00] <lamont_r> because we've already declared it to be RC for hoary
[11:01] <mdz> doesn't apt work on morphix?
[11:01] <lamont_r> not recent disks, apparently
[11:01] <mdz> if it doesn't, I really don't see the point of tolerating all this mini_fo madness
[11:01] <mdz> that seems like the only reason
=== ph [[email protected]] has joined #ubuntu-devel
=== sivang [[email protected]] has joined #ubuntu-devel
=== lamont_r adds a chmod to the morphix startup scripts
=== mvo_ [[email protected]] has joined #ubuntu-devel
[11:21] <lamont_r> mdz: looks like minifo might be trivial to fix - want me to see what I can do?
[11:31] <mdz> lamont_r: what's the story?
[11:32] <lamont_r> well, if it's just a null pointer deref issue (like the oops says), it could just be a matter of a check, dunno until I look
[11:35] <lamont_r> fear any code that contains the comment: XXX ... this is wrong?
[11:35] <T-Bone> lol
[11:36] <T-Bone> lamont_r: dude, I write that kind of code!
=== T-Bone ducks
[11:41] <mdz> lamont_r: before the oops, there is an assertion failure down in the mini_fo code
[11:41] <mdz> so I don't know what the root cause is
[11:41] <mdz> it shouldn't be passing a NULL up the stack
[11:42] <lamont_r> mdz: I'll dig into it some
[11:42] <lamont_r> car done at the mechanics, heading back home in a few
[11:42] <mdz> lamont_r: ok, the localisation stuff is higher-priority though
[11:42] <mdz> I imagine it's more fixable, too
[11:43] <lamont_r> trivially.
[11:43] <lamont_r> and I'll fix the rmdir -f issue, since it's in the same package, and equally trivial
[11:43] <lamont_r> (I fixed the locale issue by adding a chmod 644 after the locale-gen call
[11:44] <lamont_r> in theory, wifi drivers are in anything newer than the RC bits, but I have no hardware to verify
=== mxpxpod [[email protected]] has left #ubuntu-devel ["later.."]
=== lamont_r finishes up with the mechanic and heads homeward.
[11:54] <lamont_r> mdz: anything else before I flee?
[11:55] <mdz> lamont_r: new build with locales fixed later this afternoon?
[11:55] <mdz> if so, nothing else at the moment
=== doko [[email protected]] has joined #ubuntu-devel