[12:12] <lamont> GAH  where is the / key on a german keyboard???
[12:12] <azeem> shift+7
[12:13] <lamont> maybe booting in german wasn't such a good idea...
[12:14] <lamont> seb128: computer menu seems to be lacking some translations, maybe? (livecd)
[12:15] <seb128> in which language ?
[12:16] <seb128> computer menu has a very bad translation level since we changed it and almost nobody sent translations
[12:17] <lamont> de_DE@euro
[12:17] <seb128> the de one should be ok
[12:17] <seb128> what string is missing ?
[12:17] <lamont> and I kinda suspect that 'Terminal' isn't german either
[12:17] <lamont> I see nothing but english in the Computer pulldown
[12:17] <lamont> Desktop Prefs looks good
[12:17] <seb128> I've not tested the livecd
[12:18] <lamont> as does system config
[12:18] <seb128> but for sure that works on a standard warty install
[12:18] <seb128> nothing, are you sure ?
[12:18] <lamont> seb128: not surprising, since I'm still playing with getting the damn thing more warty
[12:18] <seb128> even the screenshot entry ?
[12:18] <lamont> "Take Screenshot..."
[12:19] <seb128> this is an upstream translation
[12:19] <lamont> of course, OO.o startup bitches that the i18n system doesn't support locale "", so I'm thinking it may be something beyond gnome...
[12:19] <lamont> and comes up english
[12:19] <seb128> you have a problem with your environment
[12:19] <lamont> yep
[12:20] <seb128> I've a full german menu here, and the screenshot one is not a canonical change but an old upstream one
[12:20] <seb128> dpkg -L gnome-panel-data | grep locale/de
[12:20] <seb128>  ?
[12:21] <lamont> locale bitches that it can't set LC_{CTYPE,MESSAGES,ALL} to the default locale: no such file
[12:21] <lamont> and then dumps 'de_DE@euro' for all of them (except LC_ALL)
[12:21] <seb128> weird
[12:29] <mdz> sabdfl: so you're saying alex gathered everything from upstream rather than using the patches we have in our kernel source ready to go?
[12:39] <doko> lamont: "Terminal" is german (at least that's the translation used in the Gnome menu)
[12:39] <lamont> doko: ok
[12:39] <doko> lamont: what exactly is the problem (with the Live-CD?)
[12:42] <lamont> at this point, looking at various mostly minor things.
[12:42] <lamont> getting the artwork to be ubuntu artwork is my current test nemesis
[12:42] <lamont> after that, international language issues seem to be the more common ones
[12:42] <lamont> is there an ubuntu-ship package that depends on everything we ship on the CD??
[12:42] <doko> ok, please email me where I can help. I'm going to bed now.
[12:42] <lamont> doko: ok.  I hope to have an ISO on people.ubuntu.com in a few hours (before you wake up, in other words()
[12:44] <doko> fine, good night.
[01:11] <calc> lamont: can dpkg handle a depends line that long?
[01:11] <calc> lamont: a couple years ago when i tried to make a kde-i18n depend on ~ 50 packages it wouldn't work
[01:11] <lamont> calc: so you split it up into pieces.. :)
[01:12] <lamont> but probably not
[01:13] <daniels> clearly your Binary lines weren't long enough
[01:13] <daniels> should've added more binary packages.  lamont really likes it when you do that.
[01:13] <lamont> daniels: remind me to teach you a few new pressure points next time we're in the same room......
[01:14] <calc> daniels: heh
[01:15] <daniels> lamont: hey man, I'm the one working on breaking it up :)
[01:15] <daniels> lamont: last I checked, my xlibs/xfonts/xorg packages ran to about 57 source packages
[01:15] <calc> daniels: is there a list of what still needs work on that part?
[01:15] <daniels> now, if I managed to sensibly get a Binary line >= 1024 from that, I think I should be commended, not killed
[01:16] <daniels> calc: yeah -- these were just preliminary packages.  i still need to beat on the upstream xlibs and do releases of all the modules, then we need to sit down and do a patch audit.
[01:16] <daniels> (i managed to get 'massive vendor merge' on the x11r7 agenda, so this pain will dissipate.)
[01:16] <calc> ok i am about 3/4 way through the arch tutorial then I can start trying to help out with that stuff
[01:17] <daniels> fabio and I haven't actually got it into arch yet
[01:18] <daniels> the split is that we have one source package for every upstream (fd.o) xlibs/apps package, one xfonts source package, and one 'xorg' source package that just contains the X server and Mesa, basically
[01:18] <daniels> (plus a couple of other libraries where breaking the build out would require massive, painful, code changes, e.g. Xfont)
[01:18] <daniels> oh, and a 'kdrive' source package
[01:19] <calc> ok
[01:19] <daniels> that's the split I originally did, and the one I still favour doing.  fabio has been exploring a different option, and his stuff is in XSF SVN, in the xorg module.
[01:19] <daniels> we're getting together for a fortnight in denmark, in a fortnight
[01:19] <daniels> and someone will dangle a bloodied scrap of meat, and whoever emerges the winner ...
[01:19] <calc> heh
[01:20] <calc> i think i've spent all the spare money i have until i get a job
[01:20] <calc> hopefully i'll find one soon
[01:32] <lamont> mdz: empty partial files are bad juju for apt...
[01:52] <lamont> daniels: 2441 - lol
[01:56] <sabdfl> mdz: just realised that i never sent the mail i have written on the new livecd
[01:57] <sabdfl> yes, i think he's got the latest upstream module versions running on 2.6.7
[01:57] <sabdfl> in general i think that's fine but afaics the firmware versions may have change
[01:57] <sabdfl> d
[01:57] <sabdfl> e.g. ipw2100 now requires firmware 1.3, and we are shipping 1.2
[01:59] <sabdfl> personally, i think we should just go "whooooop" and put in the latest ipw2100, ipw2200 and fw, but i am only saying that because i know you'll decline the opportunity
[01:59] <sabdfl> :-)
[01:59] <sabdfl> the EDGE is for hoary, i know
[02:08] <lamont> it must be emphasized that the live CD is a subset of the whole warty experience
[02:08] <lamont> jdub: images I'm testing are the first set I sent you, btw.
[02:24] <daniels> sabdfl: but will it bleed?
[02:25] <sabdfl> no doubt, no doubt
[02:27] <jdub> lamont: trying chocolate with the grub menu?
[02:28] <jdub> n/m
[02:30] <daniels> lamont: just randomly triaging
[02:54] <mdz> sabdfl: I would rather he use our versions
[02:54] <mdz> sabdfl: we've made bugfixes to them
[02:54] <sabdfl> yes, of course :-)
[02:55] <sabdfl> he needs to move the firmware into /lib/hotplug/firmware unless it also checks /usr/lib/hotplug/firmware
[02:55] <sabdfl> he also may need to tag the firmware with the kernel version
[02:55] <sabdfl> then i can test some more
[02:55] <sabdfl> night, i'm toast, an workrave has been insistent for a while
[02:59] <rublind> Can someone help me with my sound card problems?
[03:00] <mirak_> hehe,,,,hi guys
[03:25] <calc> sabdfl: does ipw2200 work good?
[03:25] <calc> sabdfl: aiui everything but monitor mode is in now(?)
[03:26] <jdub> how do you test monitor mode?
[03:27] <calc> don't remember, i don't have working wifi under linux right now, i am waiting for the 2915abg to be released
[03:27] <azeem> if it works with hostap, you can try kismet to see whether monito mode works perhaps
[03:27] <lamont> jdub: http://people.ubuntu.com/~lamont/testing/dscn1377.jpg
[03:28] <azeem> ipw2100 works with kismet, AFAIK
[03:29] <azeem> I have 'source=hostap,eth1,hostapsource' in /etc/kismet/kismet.conf
[03:32] <lamont> jdub: how does that screen look?
[03:32] <lamont> you get 15 indexed colors, and about 640x480 to work in.  I don't think it'll get much better.
[03:32] <jdub> so it looks like i'll have to put the arrows in and stuff
[03:32] <lamont> you want the pcx?
[03:33] <jdub> you sent it earlier
[03:33] <lamont> not this one
[03:33] <jdub> oh
[03:33] <jdub> ok
[03:33] <jdub> thanks
[03:33] <lamont> that one was about 10x too big
[03:34] <lamont> (it's really just warty-final-flat scaled to 640x480, and converted to 15 indexed colors)
[03:34] <lamont> still need to decide what to do with the penguin on the f2 screen.
[03:34] <lamont> getting dragged off to dinner, will publish the CD in a couple hours or so.
[03:36] <jdub> thanks!
[04:02] <mdz> calc: ipw2200 works perfectly for me
[04:15] <calc> mdz: cool :)
[04:16] <jdub> i can even connect to non-broadcast essids now ;)
[04:16] <calc> i hope the 2915abg will work ok in my laptop being its not a centrino one
[04:16] <calc> going to try to stick it in a amd64 laptop
[04:20] <azeem> btw, somebody referenced mjg59's ibm-does-not-want-me-to-change-wlan rant in reader's letter to a german computer magazine on an article about notebook upgrading
[04:20] <azeem> wow, that was a long sentence. Even more so after three beers
[04:20] <jdub> haha
[04:20] <jdub> that's cool
[04:20] <jdub> although
[04:20] <jdub> some times i worry about german magazines
[04:21] <azeem> oh?
[04:21] <jdub> weird shit turning up in them
[04:21] <azeem> heh
[04:22] <jdub> someone mailed me a clipping of an article talking about GARNOME once
[04:22] <jdub> some kind of flame about GARNOME being GNOME-specific or something
[04:22] <jdub> "GNOMEn schiesskopf, Jeff Waugh..."
[04:22] <azeem> hmm, I think I read a brief article about GARNOME in the above-mentioned magazine
[04:23] <azeem> didn't flame you personally, though =)
[04:32] <mjg59> Haha
[04:32] <mjg59> I'm famous
[04:33] <vorlon> s/famous/in&/ ;)
[04:33] <mjg59> (And drunk. Unrelated.)
[04:34] <azeem> well, I'm drunk. And about being famous, my 'the Ubuntu development model' blog entry will hit DWN early next week, so you better point out all its flaws
[04:34] <jdub> mjg59: now that you're a star, you should be more careful about your help
[04:34] <jdub> azeem: that was a great entry, btw
[04:34] <azeem> (I didn't submit it, Joey must have picked it himself=)
[04:35] <jdub> do you perve on dwn cvs or something, and ruin the fun?
[04:36] <mjg59> I expected to be home 6 hours ago, and then got dragged to a club and bumped into some of my students
[04:36] <mjg59> It's very distressing. You're not allowed to pull them.
[04:37] <azeem> hehe
[04:37] <azeem> jdub: I was bored or something
[04:38] <mjg59> I need to find Dave Camp and make him fix the GtkFileChooser Beagle plugin
[05:46] <lamont> what's the lang code for afriakans?
[05:51] <lamont> daniels: you around?
[05:53] <fabbione> morning guys
[05:54] <lamont> fabbione: does the xf86 autodetection require logic outside of xserver-xfree86?  (That is, if I just have xresprobe and friends installed, does the right thing just happen)?
[05:55] <fabbione> lamont: no. everything is optional
[05:55] <fabbione> it always check if a detection tool exists before using it
[05:55] <fabbione> otherwise it switches to internal default
[05:55] <lamont> fabbione: the other direction.
[05:55] <lamont> if it's installed, will it always use it?
[05:55] <fabbione> lamont: if they are installed the magic should happen yes
[05:55] <lamont> and what packages need to be there for magic to happen?
[05:55] <lamont> xresprobe and?
[05:56] <fabbione> discover1 mdetect laptop-detect xresprobe
[05:56] <lamont> what if discover1 isn't there...?
[05:57] <lamont> (liveCD has issues with discover1.)
[05:58] <fabbione> lamont: it would ask for the video driver
[05:58] <fabbione> but i was told that the livecd uses another detection system
[05:58] <lamont> ok.  could be that the liveCD is doing a bunch of the autodetecting for X.
[05:58] <lamont> yeah
[05:58] <lamont> doesn't work on as much hardware as yours does, though
[05:58] <fabbione> it doesn't use the same logic we do on plain installs
[05:59] <lamont> I'll leave yours absent then
[06:00] <fabbione> lamont: what is the problem with discover1 on livecd?
[06:00] <tseng> lamont: is it going to go around on its own and rebuild gtk-sharp and then get the mono apps?
[06:00] <lamont> <!--   <packagereq>discover1</packagereq> problems as it tries to -->
[06:00] <lamont> <!--   autodetect -->
[06:00] <lamont> tseng: it should, but I've been swamped dealing with lvieCD issues.
[06:00] <tseng> cool
[06:00] <lamont> I'll kick it soon.
[06:01] <lamont> and get ppc through the mcs build and others.
[06:01] <fabbione> lamont: well that's problably because there was a init script in it or something.
[06:01] <lamont> fabbione: probably
[06:01] <fabbione> lamont: X uses only detect "video"
[06:01] <lamont> tseng: if it looks like one package needs a kick, telling me the name (in email), is the best way to go
[06:02] <tseng> ok
[06:03] <fabbione> lamont: if you can you should give it a kick
[06:03] <fabbione> lamont: (discover1)
[06:03] <fabbione> and see what happens
[06:07] <lamont> fabbione: step 1) rcc liveCD.  step 2) artwork from jdub,  step 3) add more packages, step 4) kick discover1
[06:11] <fabbione> roger :-)
[06:12] <fabbione> i need to test the last bug fixes on X
[06:12] <lamont> hrm.. I also need to sync the latest kernel from Alex
[06:23] <fabbione> mdz: ping
[07:03] <lamont> tseng: kicking gtk-sharp and monodoc
[07:04] <lamont> tseng: any chance you built amd64 bits too?  powerpc?
[07:04] <lamont> or just i386?
[07:07] <lamont> yeah, thought so.
[07:08] <lamont> 687736832 bytes will fit on a CD, yes? :-)
[07:09] <vorlon> lamont: well, more because of the Debian maintainer's current attitude towards portability concerns than due to the portability problems in mono itself. :)
[07:09] <lamont> vorlon: what concerns? :-(
[07:10] <vorlon> lamont: see the changelog for the last mono upload, where rather than providing minimal support for !pet archs, Zomb changed the Architecture line.
[07:12] <fabbione> that's lazyness
[07:12] <lamont> what do I have to kill to get my firefox bookmarks back?
[07:14] <lamont> ah, import is your friend.
[07:15] <lamont> vorlon: good thing they plan to keep mono out of testing, eh?
[07:15] <lamont> unstable (interpreters): The Mono .NET development environment
[07:15] <lamont> 1.0.2-1: i386 powerpc s390
[07:15] <lamont> 1.0.1-1: alpha arm sparc
[07:16] <lamont> : alpha not in arch list: i386 powerpc s390 -- skipping
[07:16] <vorlon> lamont: yeah.  I did what I could to point them in the right direction, but if this is their attitude, they're on their own. :P
[07:16] <lamont> yeah
[07:18] <lamont> fixating
[07:27] <lamont-live> LiveCD, home edition seems happuy
[07:28] <lamont> needs another 's'
[07:38] <lamont> although I suppose I should go into town and download the real test CD..
[07:51] <fabbione> HMMMM
[07:52] <lamont> fabbione: why are you HMMMMing?
[07:57] <fabbione> i found another error in X autoconfig logic but that's easy to fix
[07:58] <lamont> fabbione: yep.  right after warty
[07:58] <fabbione> it's a suboptimal fix but it works fine
[07:58] <fabbione> lamont: nope.. it will be in warty
[07:59] <lamont> g'luck
[07:59] <fabbione> it's part of a previous fix that mdz allowed to fix
[08:00] <fabbione>   * Respect selection-method user value on reconfiguration.
[08:00] <fabbione> it's like:
[08:00] <fabbione> if ! reconfiguring; then
[08:00] <fabbione>  blabla
[08:00] <fabbione> fi
[08:01] <fabbione> but it solves a bunch of interesting user horror experiences
[08:01] <fabbione> like people that were running resolution XxY
[08:01] <fabbione> and all of sudden they were bumped back to 640x480 after an upgrade
[08:02] <lamont> oops. :-(
[08:12] <fabbione> lamont: either today or during this week i will get my sparc up and running.
[08:12] <fabbione> lamont: after the release do you mind to help me setting up the buildd for warty/hoary?
[08:12] <fabbione> lamont: just tell me a time that fits you best and i will sleep/not sleep around it ;)
[08:13] <lamont> fabbione: sure.  Once we release, I have a desparate need to finish my wife's van (before I really do get killed), and I need to finish writing up the "so you want to bootstrap an architecture" paper.
[08:14] <lamont> basically step 1 is to use snapshot.debian.org, 2004-06-28 to build a then-sarge chroot, then build all of the build-depends and chroot components.  Then you build a new chroot using the built components, and build everything.
[08:14] <lamont> at each step, preface the instructions with "do whatever it takes"
[08:15] <lamont> if it took very much evil, then plan on a complete clean rebuild of everything at the end.
[08:15] <lamont> then there's di, of course.
[08:15] <fabbione> lamont: yup.. no rush about it. I need to get the hardware installed first ;)
[08:15] <lamont> but once the port is complete, and everyone believes that it'll stay current, then we get to build the beast in the data center.
[08:16] <fabbione> lamont: yes i know about the DC stuff
[08:16] <fabbione> lamont: but i need to get used to admin a real buildd
[08:16] <fabbione> and test everything locally
[08:17] <lamont> new buildd coming for hoary anyway
[08:18] <lamont> and buildd assumes that you have w-b access, which we don't outside the data center.
[08:19] <fabbione> i know
[08:19] <fabbione> but i can still use a local repo to do the first port
[08:20] <lamont> yep.  I'm looking at mini-dinstall to see if that'll work with what I want, or if I get to grow my own.
[08:21] <fabbione> lamont: http://debdev.fabbione.net/cgi-bin/viewcvs.cgi/debarch/kelly?rev=1.4&view=log
[08:21] <fabbione> that's almost a Debian pool
[08:24] <lamont> heh
[12:01] <Keybuk> jdub: I didn't realise the tomboy guy was unemployed
[12:05] <sivang> does anybody know if we already have an eclipse package?
[12:22] <daniels> lamont: sup?
[12:23] <sivang> he's probably asleep :)
[12:24] <daniels> bong
[01:07] <sivang> '
[01:07] <sivang> '
[01:18] <doko> lamont: awake?
[01:36] <sivang> doko : probably still asleep
[01:56] <doko> sivang: he wanted to leave the live-CD for testing, do you know where I can find it?
[02:00] <sivang> doko : see channel's topic
[02:00] <sivang> doko : http://people.ubuntu.com/~lamont/testing/warty-live-20041017-05.iso
[02:01] <sivang> I got it at 98.9KB's , no rsync daemon running there though.
[02:01] <sivang> Am going to test it when I come back..
[02:02] <sivang> daniels : you are still uploading packages by a sponser? (reading through your blog)
[02:03] <daniels> sivang: debian packages, yeah
[02:03] <daniels> haven't been able to bootstrap a chroot yet
[02:04] <sivang> daniels : nither do it, if you'd like to share some experience with me - I'd be glad 
[02:08] <daniels> sivang: debootstrap sid ~/chroot/sid http://http.au.debian.org/debian/ /usr/lib/debootstrap/scripts/sid, should do it
[02:38] <nobse> hi
[02:38] <nobse> I just realized that subversion is in main, but subversion-tools is in universe. They are both from the same source package.
[02:39] <nobse> Is this intended?
[02:56] <Kamion> Mithrandir: yes
[02:57] <Kamion> lamont: I don't think ubuntu-ship is a good idea - it's not necessarily expected that people will install *all* of ship
[02:57] <Kamion> nobse: the main/universe-from-same-source-package is normal and expected; whether subversion-tools should be in main is a different question ...
[02:57] <Kamion> (I think it should - if we support subversion we should support subversion-tools)
[02:57] <Kamion> nobse: please mail ubuntu-devel
[02:59] <nobse> Kamion: alright
[03:00] <Kamion> nobse: it is, however - er, how can I put this - VERY VERY LATE to be adding new packages to main :)
[03:00] <Kamion> nobse: it'll probably be a hoary thing
[03:01] <nobse> hehe, sure :)
[03:01] <nobse> I just wondered...
[03:02] <nobse> But I'll ask on -devel.
[03:21] <amu> warty-live-20041017-5 does not start :(
[03:26] <azeem> I get 'Kernel panic: VFS: Unable to mount root fs on unknown-block(9,0)'
[03:34] <amu> linuxrc,splashes, static compiled bins in miniroot.gz are not there
[03:46] <jdub> hi amu
[03:48] <amu> hi jdub 
[04:01] <amu> jdub: gnoppix mirror on linux.org.au is out of sync, Anand runs the server ? 
[04:01] <jdub> he helps out yeah
[04:20] <tseng> lamont: sorry, i have no {amd64,ppc} for testing anything.
[04:36] <lamont> tseng: np
[04:37] <lamont> just means I have to do them the hard way. :-)
[04:37] <lamont> assuming that amd64 will even build the beast, that is... how is 64-bit support?
[04:37] <tseng> hm well
[04:37] <tseng> the latest devel branch i think is that one that has native 64
[04:38] <tseng> the stable (1.0.1) used a slower runtime interpreter (mint)  on amd rather than -jit
[04:38] <tseng> the mono packages depend appropriately on -jit | -mint
[04:38] <lamont> amu: sigh
[04:39] <lamont> tseng: so I won't even bother trying amd64, sound right?
[04:39] <tseng> hey
[04:40] <tseng> ive never tried it myself, one of the gentoo amd64 guys said it sucked
[04:40] <tseng> on the converse, a few people have bugged me about not nhaving amd64 in there
[04:42] <tseng> the source packages will be there, so they will be able to do the same thing I did
[04:43] <lamont> I may try bootstrapping it at the same time as ppc (otoh, ppc has debian 1.0.1-1 bits)
[04:46] <tseng> lamont: looks like there is another reverse dep on gtk-sharp and monodoc
[04:46] <tseng> er, s/reverse/circular/g
[04:46] <lamont> tseng: yeah, was working on that last night some.
[04:46] <lamont> I'll clear it first thing in the morning.
[04:47] <tseng> :)
[04:47] <lamont> right now I need to run away for the day
[04:47] <lamont> well, with a pass back by in about 20 minutes to upload 20041018-16 or such
[04:54] <amu> lamont: servus ;) 
[05:01] <plovs> /join #python
[05:06] <lamont> where one change is not one change, but rather 3. :-(
[05:06] <tseng> cya.
[05:06] <lamont> well, first another strong glance at the output of the CD build
[05:09] <lamont> nothing leaps out at me.  later
[05:37] <mdz> fabbione: pong
[05:37] <fabbione> mdz: hey
[05:38] <fabbione> you did catch me in a 3 minutes break
[05:38] <fabbione> ;)
[05:38] <fabbione> mdz: i found another error related to the 640x480 problem
[05:38] <mdz> fabbione: I read a bit in scrollback
[05:38] <fabbione> one block was not wrapped to be executed only on install 
[05:38] <fabbione> and it was resetting some values across upgrades
[05:39] <fabbione> it was noted by some users that "after upgrade -> reboot my X was at 640x480"
[05:39] <fabbione> the fix is very simple
[05:39] <mdz> well, since you have an X upload pending already, let's add it to that
[05:39] <fabbione> if [ -z "$2" ] ; then
[05:39] <fabbione> ok
[05:39] <fabbione> i did already
[05:39] <fabbione> i was waiting blessing :-)
[05:39] <mdz> that needs to be uploaded very soon
[05:39] <fabbione> mdz: tomorrow morning or later today
[05:39] <mdz> so that it can see a bit of testing before release
[05:40] <fabbione> i had to wait for daniles and you
[05:40] <fabbione> daniels reported that it is not possible to patch X for the color thing
[05:40] <fabbione> anyway i need to go away now
[05:40] <fabbione> i might take 10 minutes and upload later
[05:40] <fabbione> all changes have been tested
[05:40] <fabbione> cya
[05:44] <pitti> mdz: Hiya! Recently you asked me to upload g-v-m to my unofficial repo (#2370). Do you plan to allow it into Warty?
[05:45] <mdz> fabbione: ok
[05:46] <mdz> pitti: yes, I would like to
[05:46] <mdz> but it is getting quite late now
[05:46] <mdz> I tested it a bit
[05:46] <pitti> mdz: that's why I'm asking. The more we wait, the worse is uploading it
[05:47] <pitti> mdz: it's not the most important thing, though, but it is no new code after all...
[05:47] <mdz> it looks good
[05:47] <mdz> go ahead and upload
[05:47] <pitti> mdz: I can announce it again on -users/-devel and we can see 
[05:47] <pitti> mdz: oh, okay. If it should cause problems, I can revert the patch
[05:47] <mdz> right
[05:48] <pitti> mdz: okay, I'll upload. Thanks.
[06:10] <daniels> mdz: yes, not possible to do this sort of thing properly
[07:11] <mdz> daniels: brilliant
[07:27] <Kamion> to ug?
[07:29] <pitti> mdz: was anacron just forgotten to be included in the seeds, or is that deliberate?
[07:30] <mdz> pitti: see the bug about it
[07:30] <pitti> mdz: oh, thanks
[07:41] <hornbeck> jdub: what is in multiverse?
[07:41] <azeem> hornbeck: http://archive.ubuntu.com/ubuntu/dists/warty/multiverse/binary-i386/Packages.gz
[07:42] <hornbeck> azeem: thanks
[07:51] <pitti> Hi seb128! Nice Sunday evening!
[07:51] <seb128> hello pitti :)
[07:52] <seb128> thanks
[07:58] <Keybuk> mdz: you still haven't actually provided any evidence of these so-called problems with it :)  it's always worked fine for me
[08:04] <fabbione> seb128: what happened to my nice panel setting?
[08:04] <seb128> what ?
[08:05] <fabbione> i got the error about the trash applet
[08:05] <fabbione> i answered to keep it
[08:05] <fabbione> and all my nice little tiny icons just disappeared
[08:06] <fabbione> i had a bunch of them on the panel
[08:06] <seb128> "the error" ? which one .. ?
[08:06] <seb128> I don't have any problem
[08:06] <seb128> you lost your configuration ?
[08:07] <fabbione> the one that says that there was an error opening the trash applet
[08:07] <fabbione> and asking if i wanted to keep it or delete it
[08:07] <fabbione> and basically all my panel settings are gone
[08:07] <Keybuk> kill the panel
[08:07] <seb128> killall gnome-panel
[08:07] <fabbione> done
[08:08] <fabbione> it's the same
[08:08] <fabbione> they are gone
[08:09] <seb128> hum, the icons were some launchers ?
[08:10] <fabbione> seb128: yes
[08:10] <fabbione> xterm, mozilla and a couple of few more
[08:10] <fabbione> anyway.. not a big big deal
[08:10] <seb128> weird
[08:10] <fabbione> just curious if anybody saw this before
[08:10] <seb128> afaik we never got a bug report about a such problem
[08:10] <fabbione> i have /home on NFS if that makes a difference
[08:10] <fabbione> ok
[08:10] <fabbione> than it must be a moon ray hitting my netcable
[08:10] <fabbione> ;)
[08:10] <fabbione> gotta go
[08:10] <fabbione> later
[08:11] <fabbione> thanks
[08:11] <seb128> later
[08:11] <seb128> np
[08:17] <sabdfl> Kamion: to hug but not really mean it
[08:45] <Keybuk> as apposed to an ugh, which is a hug with someone you find repulsive?
[08:48] <mdz> or a hug rotated one character to the left
[08:52] <sabdfl> mdz: that sort of thinking has already got me into enough trouble with the artwork, thank you
[08:53] <sivang> hey, the artwork is __way__ cool. It's plain art, nomatter the reactions :)
[08:55] <Keybuk> I still don't understand the severity of peoples reactions to it ... (a) it's nothing extraordinary, they're not wearing leather and brandishing whips and (b) you can change it!
[08:56] <sivang> Keybuk : me nither. It's really overrecating 
[08:56] <sivang> *neither
[08:57] <Kamion> you can change lots of things, but we've still been concentrating really hard on sane defaults everywhere
[08:57] <sivang> You imply that this is not a sane defaut?
[08:57] <Kamion> I don't think it is, no. I've said this in more depth elsewhere, though.
[08:57] <sivang> oh
[09:03] <doko> I was once involved in a case at the university. for institutions/organizations it's the point that you _can_ install the "offending" background, you don't have to become actively involved in downloading it. For the community as a target the backgrounds are fine. (didn't sabdfl has a ubuntu-keybuk package for your special needs ;)
[09:04] <sabdfl> ubuntu-keybuk?
[09:05] <sabdfl> that the one with the leather and whips?
[09:08] <doko> sabdfl: yeah, don't know if you can call it artwork
[09:08] <sabdfl> just.... work :-)
[09:09] <sivang> everything is art, it only depends on your prespective
[09:09] <doko> :-)
[09:09] <sabdfl> i'm interested to see tomorrow if the community feeling is that the new art is fine if it's there but not the default
[09:09] <sivang> isn't that right, sabdfl ?
[09:09] <sabdfl> or if the vocal types will insist that it be removed from the cd
[09:09] <sabdfl> and why stop there... perhaps even the entire archive
[09:09] <sabdfl> Kamion?
[09:10] <Kamion> yep?
[09:10] <sabdfl> ^?
[09:10] <sivang> hey hey hey
[09:10] <sivang> let's not get too scared by the vocals
[09:10] <sivang> :)
[09:10] <Kamion> personally, I'd have absolutely no problem with it being there but not the default
[09:10] <sabdfl> well, i was thinking about it
[09:10] <Kamion> dunno what others will think; I think *most* of the comments have been about the default
[09:11] <sabdfl> if these images are unacceptable, then a web site that served these images would be filtered
[09:11] <Kamion> but that'll probably be coloured by my own opinion
[09:11] <sabdfl> so, some of the -user guys would have an argument for filtering out the archive, because it "contains" offensive images
[09:12] <sabdfl> i mean, the saudi arabian mega-proxy-filter couldn't selectively filter a package, right?
[09:12] <sabdfl> but i suppos gpg is illegal there tooo....
[09:13] <Kamion> they might already be filtering it out because of the existence of pornview in universe ;)
[09:13] <sivang> hahah
[09:13] <Kamion> (although I think the actual porn got taken out of that a while back - due to copyright violation, of all things)
[09:13] <sabdfl> as opposed to porn in main?
[09:13] <sivang> there's a package named porn ?
[09:13] <Kamion> no
[09:13] <sivang> (I know pornview)
[09:14] <sabdfl> yes, called ubuntu-artwork
[09:14] <sabdfl> by some definitions, it seems
[09:14] <sivang> :))))
[09:14] <Kamion> sabdfl: oh BTW, changing the subject entirely, I got an ultra-alpha ultra-ugly graphical installer booting on Friday
[09:14] <sabdfl> i think i did
[09:14] <doko> what about one extra question on install: this package contains materials which may be "offending" depending on your point of view. do you want to remove this materiel from you harddisk and do not install newer versions of this material?" Of course the default answer should be no.
[09:15] <sabdfl> can have some interesting consequences, that one
[09:15] <sivang> sabdfl : why don't leave the artwork as it is, and let people change it? we could author an "anti ubuntu-porn artwork" guide :)
[09:15] <sabdfl> Kamion: cool
[09:15] <Kamion> webcollage> grovels through your web cache?
[09:15] <sabdfl> no, it fetches random images from the web
[09:15] <Keybuk> gah, Amaya really pisses me off with her continual rant about 'pornview' and how it degrades women!
[09:15] <sabdfl> i think it does google searches for images based on random words
[09:15] <Keybuk> (tangents briefly)
[09:16] <Kamion> sivang: for some people it's now an issue of trust that upgrades aren't going to cause problems for them :(
[09:16] <Kamion> sabdfl: yow
[09:16] <Kamion> that's terrifying
[09:16] <sabdfl> i've seen some nasties
[09:16] <sabdfl> but the best was when i looked up and saw a pic of an astronaut mate of mine floating across the screen
[09:16] <Kamion> bah, mdz left
[09:16] <sabdfl> with her hair standing on end in zero-g
[09:17] <sivang> ooo
[09:17] <sabdfl> sivang: i think, for everyone's nerves, we're going to have to break the golden rule and force the desktop wallpaper to be the default one, even if it says "calendar"
[09:18] <sabdfl> because, in fact, calendar was the default, and it saying so in the user's settings means the user didn't actually change the default
[09:18] <sabdfl> just looked at it
[09:19] <sivang> i see
[09:19] <Keybuk> you can do that easily enough by changing the filename of the calendar symlink
[09:19] <Keybuk> it'll go "gah! my WALLPAPER vanished" and revert to the default, fwict
[09:22] <sivang> Kamion : ok, I see your point in it.
[10:07] <lamont> one of the questions wrt the live CD is: "What packages are part of the desktop seed, and not on the Live CD, but should be?"
[10:08] <sivang> hey lamont
[10:08] <shlomil> hi sivang 
[10:10] <shlomil> lamont: what base module did you use ? 
[10:11] <sivang> shlomil : base module?
[10:11] <sivang> shlomil : what do you mean in base module ? :)
[10:11] <shlomil> the liveCD uses Morphix , right? 
[10:11] <sivang> yes
[10:12] <lamont> so the gzip futex bug showed up in apt-extracttemplates?
[10:12] <shlomil> Morphix is a modular distribution, one of the modules is defined as "basemod" 
[10:17] <lamont> shlomil: http://people.ubuntu.com/~lamont/testing/*.xml
[10:18] <sivang> lamont : morphix is used for the livecd right?
[10:20] <lamont> yes.  based on morphix
[10:21] <lamont> both the basemod and the mainmod are warty based
[10:23] <shlomil> nice 
[10:25] <amu> lamont: the problem is sysvinit, *noppix run their "own" init
[10:26] <shlomil> lamont: did you write a script to parse the XML files and create the modules? or is it one of Morphix tools i'm not aware of? 
[10:26] <lamont> morphix-mmaker is your friend
[10:26] <lamont> amu: what problem?
[10:27] <Keybuk> I've always wondered why livecds have to be specially made -- surely a compressed filesystem and a ramdisk is sufficient?
[10:28] <amu> hwdetection, start system, run/stop programs, systemhalt there are modifications compared to a normal init  
[10:28] <lamont> amu: system halt is easy.  just log off
[10:28] <amu> ... and cdeject ? 
[10:29] <lamont> Keybuk: there are other hooks for automatically doing things with the compressed chroot in strange and beautiful [?]  ways.
[10:29] <pitti> Oh, hell^Whal 0.4 is out...
[10:29] <lamont> eject
[10:29] <lamont> eject /dev/hdc worked fine for me here
[10:30] <lamont> that and the *noppix init fails to install (conflicts), and there is no sysvinit-morphix package.
[10:30] <lamont> amu: of course, once you eject the CD, your days are numbered.
[10:30] <sivang> lamont : btw you got over that uninterruptable thingy that you had to reboot for?
[10:31] <shlomil> lamont: so where do you put the eject command ? 
 Keybuk: there are other hooks for automatically doing things with the compressed chroot in strange and beautiful [?]  ways.
[10:32] <lamont> shlomil: I think it's in /usr/bin
[10:32] <lamont> yep.
[10:32] <lamont> just not in the output of ls /usr/bin./
[10:32] <lamont> until you run it.
[10:32] <lamont> maybe.
[10:32] <sivang> :)
[10:33] <lamont> but eject is in both the base and warthog mods, so it is probably in /usr/bin even for ls
[10:33] <shlomil> lamont: no, i ment, one of Morphix problems is it doesn't eject CD after shutdown .. 
[10:33] <amu> lamont: you run a normal init ?  
[10:33] <lamont> yep
[10:33] <lamont> stock warty sysvinit/svsv-rc
[10:34] <lamont> which is why 'reboot' doesn't work, but 'reboot -f' does.
[10:34] <shlomil> really .. hmm 
[10:34] <amu> lamont: well you _must_ run a knoppix-init ;)   
[10:34] <lamont> amu: why?
[10:35] <shlomil> lamont: knoppix hardware detection maybe > 
[10:35] <lamont> we use that.  considering trying to bolt our own hardware detection in post-warty
[10:36] <lamont> (there are 3 knoppix packages in the build)
[10:36] <lamont> hwdata-knoppix_0.107-8_all.deb  usleep-knoppix_0.5-1_i386.deb
[10:36] <lamont> hwsetup_1.0-14_i386.deb
[10:36] <amu> init starts/stops programs, knoppix do it in a special way, if you cange it you get a seroiuse behavoir, some programs will not stop, some die, cdeject will not work, hwdetection will not work probably 
[10:37] <lamont> amu: I'm just doing what the nice morphix guy told me to, using his templates.
[10:37] <amu> lamont: this hwsetup will _NOT_ work with a normal init  
[10:37] <lamont> amu: in other words, give me an existance proof
[10:37] <shlomil> heh
[10:38] <lamont> I know that morphix has solved some stuff in different ways than knoppix did, while still using those knoppix packages.
[10:38] <amu> lamont: hwdetection will not work, test it, change a pci-id, in knoppix-hwdata, rerun and you'll see 
[10:40] <amu> probably lspci informations are from the kernel 
[10:48] <shlomil> lamont: oh, i see you created a bootsplash package.
[10:48] <lamont> yep.
[10:49] <Keybuk> shlomil: delayed until hoary
[10:49] <Keybuk> we've gone for an entirely userspace solution to it
[10:49] <shlomil> oh, you mean , you don't want to patch the kernel  ?
[10:50] <Keybuk> the bootsplash patch only works on i386, iirc.
[10:50] <lamont> Keybuk: which means it's not a patch. :-)
[10:50] <lamont> s/patch/real patch/
[10:51] <shlomil> Keybuk: but that means you'll have some "lost" bootsplash seconds, until the kernel finishe doing it's stuff
[10:51] <Keybuk> heh, it's under a second!
[10:51] <Keybuk> you know when it says "Starting Ubuntu..." at the moment?
[10:51] <shlomil> really .. hmm 
[10:51] <Keybuk> that's when the kernel's finished and you're in the initrd
[10:52] <Keybuk> all the stuff that takes time, like IDE or SCSI probing is done in initrd, remember
[10:52] <shlomil> oh, isee 
[10:52] <Keybuk> the main reason the kernel seems to take a second or two is just how slow it is to dump two or three pages of printk to the screen
[10:52] <Keybuk> sivang: it's a bit of pretty
[10:53] <Keybuk> that was going to have not-naked people on it as well
[10:53] <Keybuk> imagine the outcry
[10:53] <lamont> and the clueless end user is confused by all that gibberish on the screen
[10:53] <sivang> that's right
[10:53] <vorlon> So the userspace solution is still pending?
[10:53] <lamont> vorlon: known, demonstrated, not ready for warty freeze
[10:54] <lamont> freezes are such a bitch, you know.
[10:54] <sivang> than again, you could direct all boot message to somewhere and priont "Please wait. System loading"
[10:54] <shlomil> sivang: are you kidding ? some people love it and don't want to see anything else, besides, would you really prefer endless log line over the blonde chick ? 
[10:54] <Keybuk> vorlon: yeah, main problem with having conference right before a release
[10:54] <vorlon> conference?
[10:54] <Keybuk> hoaryconf will be great, because it comes right at the point we *can* develop new stuff
[10:54] <lamont> Keybuk: hell of a lot of coding was done at that conf.
[10:55] <sivang> shlomil : ooo, I get my dose of the blond chick on the wallpaper believe me :)
[10:55] <Keybuk> yeah, the two weeks in oxford we had in august
[10:55] <Keybuk> lamont: you don't *want* coding that close to a release 
[10:55] <vorlon> ah, the conference abutted your freeze? :)
[10:55] <Keybuk> you want it at the start
[10:55] <lamont> vorlon: the freeze was pushed out a week to accomodate the conf.. :-)
[10:55] <lamont> Keybuk: very true/
[10:56] <vorlon> heh. :)
[10:56] <Keybuk> bug fixing is what you want, and that's not a good use of a conference; is better done in isolation
[10:56] <vorlon> so where in the timeline does "merging all this yummy stuff back into Debian" fall? ;)
[10:57] <Keybuk> we've sent most the yummy stuff back ?
[10:57] <vorlon> ah, ok.
[10:57] <mdz> Keybuk: hmm?
[10:57] <mdz> we've sent bugfixes back, but all of the customisation stuff needs merging
[10:57] <mdz> and the bugfixes that were sent but didn't get merged in Debian
[10:58] <mdz> shouldn't take more than, oh, the first two months of the release cycle :-P
[10:58] <Keybuk> heh, was there any reason we didn't send those back as well?
[11:00] <mdz> good question; we should send the Ubuntu branding patches upstream
[11:00] <mdz> :-P
[11:00] <Keybuk> lol, wouldn't that be fun :p
[11:00] <mdz> debuntu
[11:00] <Keybuk> that'd put tbm into serious bitch mode *giggle*
[11:00] <Keybuk> almost worth it
[11:04] <sivang> mdz : what about security patches from review? 
[11:04] <sivang> mdz : they also need be merged
[11:04] <lamont> vorlon: they're just the morphix packages, see www.morphix.org/debian
[11:04] <mdz> sivang: the security review involved almost entirely backports of patches from Debian or syncing of new versions from Debian
[11:04] <mdz> it is just a matter of checking that everything has been merged
[11:04] <sivang> mdz : ok
[11:05] <pitti> mdz: hal 0.4 was just released, and I found a security related change in it
[11:05] <pitti> mdz: http://freedesktop.org/cgi-bin/viewcvs.cgi/hal/hal/hald/linux/net_class_device.c?rev=1.19&view=log
[11:05] <pitti> mdz: we do not need that patch for our own purposes, so it's not overly critical
[11:06] <pitti> mdz: the question is, shall we apply it anyway? Users might do custom stuff with hal
[11:06] <pitti> mdz: the patch looks alright, but still it is damn near the release to change something
[11:06] <mdz> pitti: so nothing is listening to the dbus messages currently?
[11:06] <Keybuk> mdz: I can probably give you a good idea what the actual differences are with a tablecloth-trick (blame jdub for the name :p) ... try a warty/hoary jump and see what falls out
[11:06] <seb128> mdz: perhaps we want #2345 fixed for warty ?
[11:07] <mdz> Keybuk: try a *what*?
[11:07] <pitti> mdz: well, the utopia packages don't care for the netlink stuff
[11:07] <mdz> seb128: looks safe to me
[11:07] <vorlon> mdz: off-topic, but while I have you ... :-)  Does the kernel-patch-uml package work with kernels newer than the ones currently suggested?
[11:07] <pitti> mdz: but of course user programs _can_ attach to dbus and query these attributes from hal
[11:07] <mdz> vorlon: perhaps, perhaps not
[11:07] <seb128> mdz: ok, thanks
[11:07] <pitti> mdz: it's not a matter of securing Ubuntu proper, but only securing possible user installed programs
[11:07] <vorlon> Heh.  From a release standpoint, how should k-p-uml be handled?
[11:12] <mdz> vorlon: honestly?
[11:12] <mdz> it wouldn't be entirely unreasonable to remove it from sarge
[11:12] <mdz> upstream is fucked; UML hasn't worked properly on Debian for months
[11:12] <vorlon> Well, I find that pretty persuasive.
[11:37] <jdub> AHR!