[12:30] <jdub> mdz: ping
[12:51] <Burgundavia> what the heck is Colin Watson's irc nick?
[12:51] <jsgotangco> Kamion
[12:53] <wasabi_> Slight problem understanding the .install files used by dh_install.
[12:54] <wasabi_> What is the root path they are copied from?
[01:01] <Kamion> Burgundavia: yo?
[01:02] <Burgundavia> Kamion, someone on the forums, asking about OEM installer
[01:02] <Kamion> wasabi_: root of the unpacked source package
[01:02] <Burgundavia> so I mentioned to come and talk to you
[01:02] <Kamion> Burgundavia: *shrug* will be ready when it's ready :)
[01:02] <Burgundavia> indeed
[01:02] <wasabi_> Kamion, Hmm. I'm a bit confused why this used to work.
[01:02] <wasabi_> I thought I could just put usr/share/foo in the .install file.
[01:03] <wasabi_> Without preceeding it with debian/tmp
[01:03] <Kamion> wasabi_: do you mean the left-hand side or the right-hand side of the line?
[01:03] <wasabi_> Left handed side.
[01:03] <wasabi_> Without a right handed side
[01:03] <Kamion> wasabi_: you're using --autodest?
[01:04] <wasabi_> I am using cdbs.
[01:04] <Kamion> or --sourcedir?
[01:04] <wasabi_> Doesn't look like that's being passed.
[01:04] <Kamion> oh, who knows what that does
[01:04] <wasabi_> Heh.
[01:04] <wasabi_> dh_install -peclipse-platform
[01:04] <wasabi_> cp: cannot stat `./usr/share/eclipse/eclipse.ini': No such file or directory
[01:04] <wasabi_> This is odd because I swear I've done this, with cdbs, before.
[01:04] <Kamion> use DH_VERBOSE to see what it's doing
[01:05] <Kamion> perhaps you need to use --sourcedir if ./usr/share/eclipse/eclipse.ini doesn't exist relative to the root of the source package you're building in
[01:07] <Duck_busy> coin
[01:08] <wasabi_> Heh.
[01:08] <wasabi_> It does a bunch of files.
[01:08] <wasabi_> But just doesn't do this one. This one is a symlink.
[01:08] <Duck_busy> wasabi_: use .links then ?
[01:09] <wasabi_> Yeah. That just means part of the logic for creating the symlink is in rules and part isn't.
[01:10] <wasabi_> This thing always beats my ass.
[01:13] <wasabi_> Duck_busy, okay... how about this. I have removed all the complicated parts. I have only usr/share/eclipse/eclipse.ini listed in .install.
[01:13] <wasabi_> And it's saying not found.
[01:14] <wasabi_> This has worked before without me having to tell it what the sourcedir is.
[01:15] <Duck_busy> hum
[01:15] <Duck_busy> was in debian/tmp/ but is now in debian/<pkg>/ ?
[01:15] <Jerrac> I was told on the forums to talk to Kamion or mdz about my request. Are they here?
[01:16] <wasabi_> jhaltom@station-1:/dev/shm/eclipse/eclipse-3.1~RC2$ ls debian/tmp/usr/share/eclipse/eclipse.ini
[01:16] <wasabi_> debian/tmp/usr/share/eclipse/eclipse.ini
[01:16] <Kamion> Jerrac: I'm here, but about to be dragged off to bed
[01:16] <Jerrac> ah, ok. maybe I can be quick then.
[01:16] <Kamion> so if it's not a one-line-answer thing it'll probably have to wait :)
[01:16] <Duck_busy> wasabi_: then using debian/tmp/usr/share/eclipse/eclipse.ini should strip debian/tmp and be ok
[01:16] <Jerrac> Here is my post:
[01:16] <Jerrac> I see that there is going to be an OEMInstaller: http://udu.wiki.ubuntu.com/OEMInstaller
[01:16] <Jerrac> I am working on a Ubuntu based computer to sell on ebay, the one thing I am having problems with is an easy way for the buyer to change the default username and password I have to put on with Hoary.
[01:16] <Jerrac> Is there a way I could use part of the current OEMInstaller to create a script that the buy could run when they first start the computer? As in they log in under the default username and then click the script to run it.
[01:16] <Jerrac> Thanks!!
[01:16] <wasabi_> Duck_busy, I have to put debian/tmp in .install?
[01:17] <wasabi_> See, I've just never had to do that before.
[01:17] <Kamion> Jerrac: there isn't a "current OEMInstaller"
[01:17] <Duck_busy> wasabi_: yes sir
[01:17] <wasabi_> I used ot be able to just plain list "usr/share/eclipse/eclipse.ini"
[01:17] <Jerrac> Is there some code I could use to do what I want?
[01:17] <wasabi_> wihtout a leading /
[01:17] <Duck_busy> did u used .files instead of .install ?
[01:17] <wasabi_> NOpe.
[01:17] <Jerrac> Or do you have any other suggestions?
[01:18] <Duck_busy> wasabi_: or dh_install --autodest ?
[01:18] <wasabi_> I've done this previously with cdbs
[01:18] <Kamion> Jerrac: not really, if I were you I'd install without creating a user (you can do this by setting the root password instead) and run dpkg-reconfigure passwd afterwards
[01:18] <wasabi_> without touching the dh_install logic at all
[01:18] <Duck_busy> wasabi_: why stop using cdbs ?
[01:18] <wasabi_> I am using cdbs. I haven't stopped.
[01:18] <Kamion> Jerrac: but the work basically just hasn't been done yet, so you'll have to do a certain amount of stuff yourself
[01:18] <wasabi_> Actually I'm converting a non-cdbs package to cdbs
[01:18] <Duck_busy> perhaps cdbs logic changed then
[01:19] <Jerrac> yah, i figured that.
[01:19] <Duck_busy> but is am not aware of such changes
[01:19] <Jerrac> How do I bypass creating a user in the install?
[01:19] <Kamion> Jerrac: like I say, set the root password
[01:19] <Duck_busy> wasabi_: ask jbailey tomorrow about it
[01:19] <Duck_busy> he will be able to tell you if anything changed
[01:19] <Jerrac> ok, thanks. I will try to figure that out. :D
[01:20] <Duck_busy> he is cdbs main author
[01:20] <Kamion> Jerrac: if you're working interactively, use expert mode and you'll see it
[01:20] <wasabi_> yeah
[01:20] <wasabi_> I'm looking at this package right here... java-common
[01:20] <Kamion> Jerrac: otherwise, preseeding/kickstart can do it
[01:20] <wasabi_> In it's .install it lists etc/jvm
[01:20] <wasabi_> in it's rules it lists dh_install -i
[01:20] <Jerrac> ok, thanks! 
[01:20] <wasabi_> and that's it
[01:20] <Kamion> anyway, got to go ...
[01:20] <Jerrac> I will go work on it more. :D
[01:20] <Jerrac>  Have a good evening!
[01:21] <Kamion> ta
[01:21] <Duck_busy> wasabi_: compat level is ?
[01:21] <wasabi_> oooh. ones doesn't have it listed.
[01:22] <wasabi_> the one that is broken is 4
[01:22] <Duck_busy> here it is
[01:22] <Duck_busy> old behavior was .files compat stuff with like --sourcedir=debian/tmp i'm sure
[01:22] <wasabi_> ahh.
[01:22] <wasabi_> okay. that probably explains it.
[01:24] <Duck_busy> wasabi_: try using : DEB_DH_INSTALL_ARGS := --sourcedir=debian/tmp
[01:24] <Duck_busy> if you don't want to modify your .install files
[01:24] <wasabi_> I'll just put debian/tmp in front of everything
[01:25] <wasabi_> my install files need to copy from debian/extra also.
[01:25] <Duck_busy> ok
[01:27] <thom> whiprush: feh, feh, feh
[01:28] <Nafallo> thom: hi :-)
[01:28] <thom> Nafallo: you're welcome to, ensure that your dhcp server gives out the right details and you will
[01:29] <Nafallo> thom: it's way more complex than that I'm afraid ;-)
[01:29] <Nafallo> thom: my domain exists both outside and inside my network (with different ips assigned) 
[01:30] <Nafallo> thom: I want to use the internal dns when I'm at home and whatever dns I get when I'm away.
[01:30] <thom> Nafallo: so? if your dhcp server gives out the correct nameservers, then NM will configure bind9 to use them
[01:30] <Nafallo> thom: this new setup breaks that :-/
[01:30] <thom> Nafallo: i don't see how, at all
[01:30] <Duck_busy> good night folks
[01:31] <elmo> man, mdz's got his joeyh-game-face on for ubuntu uploads
[01:31] <thom> Nafallo: all bind9 is being used for by NM is as a caching nameserver that it can reconfigure based on the instructions it recieves from dhcp
[01:31] <wasabi_> oh darn dh_install can't rename
[01:31] <thom> yes, it totally sucks that it has to be bind9; there're moves afoot to fix that. but that's not the issue here
[01:33] <Nafallo> thom: hmm, and the instruction is not the options domain-name-servers and domain-name?
[01:33] <thom> Nafallo: if it's not working, you're probably seeing the result of 11905
[01:33] <thom> which i will fix in the morning
[01:34] <Nafallo> thom: I've recompiled with that package as build-dep :-)
[01:34] <Nafallo> thom: ehm. let me check something ;-)
[01:35] <thom> well, what does "grep forwarders /var/lib/NetworkManager/NetworkManager-named.conf" give you?
[01:36] <Nafallo> thom: right now I've screwed the config a bit. but that has indeed the correct forwarder :-)
[01:37] <thom> and your /etc/resolv.conf has 127.0.0.1?
[01:37] <Nafallo> thom: yes
[01:37] <thom> then i don't see what the problem is? i'll fix the two bugs whiprush reported in the morning
[01:38] <Nafallo> thom: before I didn't got the correct forwarder set, cause I couldn't find my own domain.
[01:39] <sabdfl> http://help-info.de/en/Help_Info_Longhorn/longhorn_help_pane.htm
[01:39] <sabdfl> "Ask Your Community"
[01:39] <sabdfl> can you say... launchpadIntegration?
[01:40] <Nafallo> thom: right now I'm happy you came online so that I can test it again :-)
[01:40] <Nafallo> thom: thanx for explaining :-)
[01:42] <thom> np
[01:55] <thom> mdz: i'm not keen on bind9 being necessary; i'm going to chase down what changes are needed for lwresd to be suitable and work on them tomorrow
[02:57] <Unfrgiven> what does "CCDEPMODE = depmode=gcc3" mean in makefiles? im trying to build a package right now that seems to want gcc-3.4 and I have a feeling its because of the depmode
[02:58] <mako> http://finaperf.com/annuaire/
[02:58] <mako> nice logo
[02:59] <mako> http://finaperf.com/images/logo-finaperf.png
[02:59] <tseng> mako: circle of ripoffs
[02:59] <mdke> mako, did you see the msn one too?
[02:59] <mako> dude, people mail me about the MSN one at least 4-5 times a week
[02:59] <mdke> oh sorry
[02:59] <mako> ALL THE TIME
[03:00] <mako> everybody else did too apparently :)
[03:00] <mako> well, it's fine.. you don't read my email :)
[03:00] <mdke> that's what you think
[03:00] <mako> i get trademarks@ubuntu.com
[03:00] <mdke> ah that explains it
[03:00] <mako> so i see it all 
[03:00] <mako> Clint: fine with me.. answer some of it will you
[03:00] <mdke> mako, one day i will apply to be a lawyer at Canonical ;)
[03:01] <mako> mdke: then i get to forward to this stuff to you :)
[03:01] <mdke> *grins*
[03:01] <Unfrgiven> anyone have a link for the msn one?
[03:01] <mako> better yet, my email will BE your email
[03:01] <mdke> Unfrgiven, hang on, it is on someone's blog
[03:01] <azeem> it's in mako's mailbox
[03:01] <jsgotangco> i love the backports logo
[03:01] <jsgotangco> reminds me of bizarro
[03:01] <Unfrgiven> jsgotangco: link?
[03:02] <mako> Unfrgiven: it's msn-spaces
[03:02] <jsgotangco> its basically the ubuntu logo in blue and the dots are inside
[03:02] <mako> azeem: my inbox is a scary scary place
[03:02] <azeem> I can imagine
[03:02] <mdke> Unfrgiven, http://blog.carthik.net/vault/2005/04/14/msn-spaces-and-ubuntus-logo/
[03:02] <azeem> mine already looks quite scary
[03:02] <mdke> whoops
[03:03] <mdke> Unfrgiven, sorry, follow the link to http://people.warp.es/~jorge/blog/?p=38
[03:03] <Unfrgiven> woah!
[03:04] <jsgotangco> heh
[03:05] <jsgotangco> mdke, become Ubuntu's favorite Barrister
[03:05] <Unfrgiven> jsgotangco: the backports logo is pretty neat!
[03:05] <mdke> jsgotangco, what can i say, it would be my dream job
[03:06] <mdke> mako, so what is the answer to the msn-spaces trademark thing? i'm curious
[03:06] <jsgotangco> Unfrgiven, yeah, totally Bizarro
[03:09] <mako> mdke: the answer is we're not doing anything about it now
[03:09] <mako> i've told mark and jane
[03:09] <mako> and i guess we're sitting on it
[03:09] <mako> we're registering the mark as many places as we can
[03:09] <mdke> hmm
[03:10] <mdke> you also have copyright in it
[03:10] <mako> we were there first
[03:10] <mako> well, that's going to really impact any dilution suit
[03:10] <jsgotangco> Barrister, please give your pro bono opinion
[03:10] <jsgotangco> heh
[03:10] <mdke> i'm not great on copyright
[03:10] <mako> personally, i'm happy with MSN using their little pudgy friends thing
[03:10] <mdke> that's why I'm interested
[03:11] <mako> what i'm worried about is that they will sue *us* just as a way to hurt us
[03:11] <mdke> mako, well if they do you can respond
[03:11] <mako> ultimately, it's up to mark and/or his lawyers
[03:11] <mako> if they think we're safe, great
[03:11] <mako> i suspect it's a "how long is this string" sort of thing
[03:11] <mdke> jsgotangco, its basically a question of whether they WANT to do anything about it, its pretty clear that they can do if they want
[03:11] <mako> we'll be able to find folks on either things
[03:12] <mako> i mean, MS can sue for stupid things with or without the logo
[03:12] <mdke> are mark's lawyers SA based?
[03:12] <mako> some of them (e.g., patents, etc) probably more promising than this :)
[03:12] <mako> i have no idea
[03:12] <mako> mark doesn't talk about his lawyers very much :)
[03:12] <mdke> LOL
[03:13] <mako> i mean, even in situations where i think he would, he doesn't
[03:13] <mako> like, his bankers.. sure.. lawyers not so much
[03:13] <mdke> interesting
[03:13] <mdke> we're not a popular race
[03:14] <mako> i have a disproportionately large number of friends among your ilk
[03:14] <mdke> yeah i am not surprised
[03:14] <mdke> i've read your blog
[03:14] <mako> at least one of them is trying to recruit me
[03:14] <tseng> mdke: you should read it daily
[03:14] <mdke> tseng, alright
[03:14] <tseng> heh, he stopped updating it
[03:15] <tseng> but its always great
[03:15] <mako> tseng: no way dude.. 1 week break
[03:15] <mako> i'm recouping my strength
[03:15] <mako> i posted yesterday
[03:15] <tseng> ill await your triumphant return to the blogosphere
[03:15] <mako> oh man.. it's gotta be good
[03:15] <mdke> *grins*
[03:15] <mdke> pressure!
[03:16] <mako> i need to invent a game
[03:17] <mdke> night all
[03:17] <tseng> cya
[03:17] <mako> oh man.. i thought of a good one
[03:17] <jsgotangco> i want to hear that
[03:18] <tseng> mako: my recent favorite was about eating pennies
[03:18] <mako> not really a game.. but sort of
[03:18] <mako> i am going to write poetry using only the names of packages
[03:18] <mako> this will be great
[03:18] <jsgotangco> oh no
[03:18] <jsgotangco> this reminds me so much of magnetic poetry
[03:19] <mako> jsgotangco: i'm pretty good at magnetic poetry
[03:19] <jsgotangco> "synergistic action"
[03:19] <jsgotangco> "we come"
[03:19] <mako> that was long
[03:19] <mako> it was an epic
[03:19] <mako> move over homer
[03:19] <jsgotangco> hah
[03:32] <mako> there are a lot of packages
[03:32] <mako> i'm still at b
[03:33] <jsgotangco> hmm probably around 17,000
[03:33] <mdz> mako: you should write a poem using each of them exactly once
[03:33] <mdz> except the ones containing the letter 'e'
[03:33] <jsgotangco> go beat dr. seuss
[03:33] <mako> mdz: well, i'm going to have it make sense.. so i'm using the ones that are or could be read as real words out of context
[03:35] <jdub> you *have* to use apt
[03:35] <mako> jdub: i'm gonna to do better
[03:35] <jdub> it's just a good real word
[03:35] <mako> jdub: BETTER.. you'll see
[03:35] <mako> i'm going to write several
[03:35] <mdz> mako: apt-cache dumpavail | grep-dctrl -nsPackage '' | sort | comm -12 /usr/share/dict/words
[03:36] <mako> i'm going to write one in honor of sarge
[03:36] <mako> mdz: nah.. i'm not *just* using the real words
[03:36] <jdub> are you going to ignore numbers in names?
[03:36] <mako> mdz: i'm also using ones that *could* be read as real words
[03:36] <mako> like apoo
[03:36] <mako> "that software is totally apoo"
[03:37] <jsgotangco> how about emacs
[03:37] <jsgotangco> or gpg
[03:37] <jsgotangco> heh
[03:37] <mako> nope
[03:37] <mako> probably not
[03:37] <mako> dude, there are a lot of words to choose from
[03:37] <jdub> mako: zenity :)
[03:37] <jsgotangco> the zenity of it all....
[03:37] <mako> mdz: i will do that to make sure i didn't miss any
[03:39] <mdz> jdub: what's wrong with apt?
[03:39] <mako> hmm.. cccc, maintained by Kamion 
[03:39] <mdz> mako: ah, right, you can use some of them phonetically
[03:39] <mdz> the enthusiastic spaniard said "cccc"
[03:39] <mako> or compound words
[03:39] <mako> exactly
[03:39] <jdub> mdz: i was praising it as an excellent word
[03:40] <mdz> jdub: oh, I misinterpreted you
[03:40] <mako> oh, howa bout "cduce"
[03:40] <jdub> mako: so you couldn't use 'man', for instance?
[03:40] <mako> as in, "cduce me"
[03:40] <jsgotangco> sounds like an sms acronym heh
[03:41] <mako> jdub: no, i'm using both real words and sound-alikes
[03:41] <mako> i will probaly use man
[03:41] <jdub> man isn't a package though
[03:41] <mako> 'an' 'and'
[03:41] <jsgotangco> "cduce me, it feels so apoo"
[03:41] <mako> ah, you're correct
[03:41] <mako> damnit
[03:41] <mako> so, no, i won't
[03:41] <jdub> aha
[03:42] <jdub> it is rad that 'the' exists
[03:42] <jdub> so you could say package names and 1 or 2 letter words
[03:42] <mako> i'm happy about that
[03:42] <tseng> as does wtf
[03:42] <jdub> tseng: haha!
[03:42] <mako> dude, cheesetracker!?
[03:42] <jdub> $ apt-cache show bbq
[03:42] <jdub> W: Unable to locate package bbq
[03:42] <jdub> E: No packages found
[03:42] <jdub> cock!
[03:42] <mdz> CHEESETRACKER??
[03:42] <mako> oh man
[03:42] <mako> totally NOT a cheesetracker
[03:43] <tseng> $ apt-cache show cock 
[03:43] <tseng> whiprush: Unable to locate package cock
[03:43] <tseng> elmo: No packages found
[03:43] <tseng> buh ironwolf you suck ass
[03:43] <tseng> irssi!
[03:43] <mdz> there are 660 package names which are also in dict/web2
[03:43] <tseng> gosh stupid thing.
[03:43] <mdz> that is a pretty big vocabulary by itself
[03:43] <mako> mdz: awesome
[03:43] <tseng> ironwolf: sorry :(
[03:43] <mako> mdz: yes, it is
[03:43] <jdub> mdz: hrm, so that doesn't catch packages that are multiple real words with or without a hyphen. hmm.
[03:43] <mako> mdz: how did you get that list?
[03:43] <mdz> mako: amphetamine
[03:44] <mako> mdz: have it
[03:44] <mdz> mako: apt-cache pkgnames |sort | comm -12 /usr/share/dict/words
[03:44] <jdub> mako: will you let yourself pluralise if it's just a matter of adding 's'?
[03:44] <mako> oh, i already did that
[03:44] <mdz> mako: apt-cache pkgnames |sort | comm -12 /usr/share/dict/words -
[03:45] <mdz> there is a distinct lack of adjectives, though
[03:45] <mdz> predominantly nouns
[03:45] <jdub> how is apt-proxy2 going?
[03:46] <jdub> hrm, ww
[03:46] <mdz> the august zebra thrust straw at the zoo
[03:46] <jdub> *snicker*
[03:46] <mako> i'm going to write three
[03:46] <mako> one about sarge, one about politics and one about sex
[03:46] <mdz> in haiku form?
[03:46] <mako> no
[03:46] <mdz> :'-(
[03:47] <mako> i MAY try to make the sarge one a limerick
[03:47] <mako> (!!!)
[03:47] <jdub> mako: august has gotta be a sarge word ;)
[03:47] <mako> jdub: ok
[03:47] <mako> LIMERICK DUDE
[03:47] <mako> THATS LIKE FUCKING IMPOSSIBLE
[03:47] <jdub> is there a Free rhyming dictionary?
[03:47] <jdub> i have one on my shelf
[03:47] <jdub> but i can't grep it
[03:48] <mako> jdub: not that i've found
[03:48] <mdz> the bamboo beast felt evolution outguess muscle
[03:48] <jdub> ha ha
[03:48] <jsgotangco> jeezz
[03:49] <mako> watch out for hte bambee beast
[03:49] <mdz> mako: I don't know where you got this idea, but I hope there was whiskey involved
[03:49] <mako> i am also not going to repeat words
[03:49] <jdub> mako: 'the'? :)
[03:49] <jdub> $ apt-cache show rules
[03:49] <jdub> W: Unable to locate package rules
[03:49] <jdub> d'oh
[03:50] <mdz> mako: your limerick claim sounds like a challenge
[03:50] <jdub> rocks!
[03:50] <jdub> haha
[03:50] <jdub> Description: Make network sockets reliable in a transparent way
[03:50] <mako> mdz: dude, let me get mine out the door
[03:50] <mako> i'm still on 'c'
[03:50] <mdz> oh, interesting, web2 doesn't have plurals in it
[03:51] <jdub> wow, rocks looks interesting
[03:51] <mako> hahah
[03:51] <mako> jdub: it DOES have 'coq'
[03:51] <mako> COQ
[03:51] <jdub> haha!
[03:51] <mako> and coq-doc
[03:51] <tseng> just in case you needed instruction on proper care of your coq
[03:52] <mako> exactly
[03:52] <mako> my potty poem involves: cpp! apoo!
[03:53] <jdub> heh
[03:53] <jdub> tseng: "first, remove extraneous packaging" ;)
[03:54] <mdz> we should have packages named after all of the words in Jabberwocky
[03:54] <mdz> the next time I don't know what to name something, I will choose one of those
[03:54] <mdz> we already have bandersnatch
[03:54] <jdub> s/don't know what/need/
[03:54] <mdz> jdub: those are equivalent
[03:54] <jdub> heh
[03:54] <mdz> I _never_ know what to name programs
[03:55] <jdub> casper was inspired
[03:55] <jdub> although it wasn't actually inspired
[03:55] <mdz> I lost like a week of work on that one
[03:55] <mdz> LTSP wouldn't even be started yet if it hadn't already had a name
[03:56] <jdub> is express remotely close to CD testing?
[03:56] <tseng> ltsp puts release-early-release-often into full effect
[03:56] <mdz> jdub: hmm?
[03:56] <mako> i actually wrote a section of a howto on how to name packagesonce
[03:56] <jdub> mdz: ubuntu express, live installer
[03:57] <mdz> jdub: oh, "is ubuntu-express ready for people to test it"
[03:57] <jdub> yeah
[03:57] <mdz> jdub: I read that as "is ubuntu-express a good test of a CD" :-)
[03:57] <jdub> wow, traffic to archive blows from here atm
[03:57] <jdub> mdz: ah, heh, d'oh ;)
[03:57] <mdz> jdub: it doesn't install the boot loader
[03:57] <mdz> but is otherwise functional, if raw
[03:57] <jdub> nice
[03:58] <jdub> we could write a poem about six month release cycles
[03:58] <jdub> to catch the ubuntu-express!
[03:58] <mdz> it would be nice if, when one logs out of GNOME, if all of the processes in one's session exited
[03:58] <mdz> haha
[03:58] <mdz> I did not anticipate the train metaphors when I came up with that name
[03:58] <mdz> the press will be all over it
[03:58] <jdub> heh, yeah
[03:59] <jdub> Catch the Ubuntu 5.10 Express!
[03:59] <tseng> not if lugradio doesnt ruin it first
[03:59] <jdub> mdz: what's not dying for you?
[04:00] <mdz> 21886 ?        Ss     0:00 /usr/lib/bonobo-activation/bonobo-activation-server --ac-activate --ior-output-fd=17
[04:00] <jdub> did we convince daniels about xlibs-dev depending on all the dev packages?
[04:00] <mdz> 21889 ?        S      0:00 /usr/lib/control-center/gnome-settings-daemon --oaf-activate-iid=OAFIID:GNOME_SettingsDaemon --oaf-ior-fd
[04:00] <mdz> 21892 ?        S      0:00 /usr/lib/gamin/gam_server
[04:00] <mdz> 21903 ?        S      0:00 xscreensaver -nosplash
[04:00] <jdub> "transitional package" -> grr.
[04:01] <jdub> mdz: ouch!
[04:01] <mdz> I wonder how gdm handles this
[04:01] <mdz> I bet it resets the X server after the session leader exits
[04:01] <mdz> which kills their connections to the X server, causing them to exit
[04:01] <daniels> jdub: xlibs-dev will never grow more dependencies.  ever.
[04:01] <daniels> jdub: that being said, I haven't actively *removed* any dependencies ...
[04:02] <jdub> daniels: we need a meta package tho
[04:02] <mdz> jdub: we do?
[04:03] <jdub> it would be handy for anyone building against X stuff who doesn't really grok all of X
[04:03] <mdz> having one for compatibility purposes seems OK, but do we really need a metapackage for everyhting going forward?
[04:03] <tseng> jdub: apt-get build-dep, dude
[04:03] <jdub> perhaps a task, but i haven't really bought in to the task thing
[04:03] <mdz> jdub: by that logic, we should have a metapackage of all -dev packages in Ubuntu ;-)
[04:03] <jdub> tseng: outside package land
[04:03] <jdub> mdz: YEAH THAT TOO!
[04:03] <jdub> no
[04:03] <jdub> i mean
[04:03] <jdub> random dude using jhbuild or something
[04:04] <mdz> dude, jhbuild ought to be packaged
[04:04] <mdz> and depend on all the build-deps
[04:04] <tseng> jhbuild should list debian deps
[04:04] <tseng> garnome used to iirc
[04:04] <jdub> suddenly their gnome doesn't have xinerama support
[04:04] <jdub> mdz: i appreciate your enthusiasm, but... ;)
[04:05] <tseng> besides that sebuild has obsoleted jhbuild
[04:06] <daniels> jdub: i think the entire concept of a metapackage is broken
[04:06] <daniels> jdub: user convenience will turn into packages b-d'ing on every single x library in existence because a) they can and b) they can't be arsed working out which libraries they depend on
[04:06] <daniels> (metapackage broken -> in this case, not in general)
[04:06] <jdub> daniels: but you can see the issue with xinerama and friends
[04:07] <daniels> jdub: that's actually a separate issue
[04:07] <daniels> jdub: where it used to be in xlibs-static-dev, and I intentionally broke something that used to work
[04:07] <mdz> jdub: we don't do this for gnome, or jakarta, or anything else I can think of.  why X, now that it's split up?
[04:07] <jdub> mdz: primarily because there are a lot of optional depends on X things
[04:08] <jdub> i don't know if this is the right solution
[04:08] <jdub> i think we need something like it for gnome
[04:08] <tseng>  man we need a monopod package
[04:08] <jdub> but for different reasons
[04:08] <mdz> I think that if anything, we should have a grouping of development libraries for everything in desktop
[04:08] <jdub> tseng: it has such a bad name!
[04:08] <daniels> eh, imo if you want xinerama functionality, just hardcode --enable-xinerama, or whatever
[04:08] <tseng> jdub: dude think of 3 pees in a pod
[04:08] <tseng> jdub: but with 2 less pees
[04:08] <daniels> that way, if libXinerama would ever fall out of a package it was in (hypothetically), you would just have a hard build failure, not a silent failure to use Xinerama
[04:09] <jdub> daniels: most configure scripts check, it says no, they continue on -> optional depends
[04:09] <mdz> would it be worthwhile to load powernowd on thin clients?  what portion of thin client-ish hardware is likely to support frequency scaling?
[04:09] <jdub> daniels: now, we can do things to make sure we don't miss these in ubuntu
[04:09] <daniels> jdub: right, but if you use --enable-xinerama or whatever, you break when these things disappear
[04:10] <daniels> instead of just silently stop using it
[04:10] <jdub> mdz: do those via chips do power scaling?
[04:10] <daniels> i think this is a general best-practice thing, not just for ubuntu
[04:10] <mdz> jdub: dunno
[04:10] <jdub> daniels: Common Man doesn't care to do that
[04:10] <jdub> mdz: mjg might know
[04:10] <infinity> mdz : A lot of smaller all-in-one systems will (and have) ship with CPUs that can frequency scale.
[04:10] <mdz> jdub: yeah, but he also might be asleep
[04:10] <jdub> most of the thin client things i've seen have via c3s or laptopish cpus
[04:10] <infinity> mdz : And for machines that spend a lot of their time doing very little (which thin clients tend to do), it may be a win to have half your network at low power.
[04:11] <mdz> jdub: that, and recycled 10-year-old desktops
[04:11] <jdub> oh, well, yeah
[04:11] <infinity> (What I don't know is if powernowd supports frequency scaling of VIA CPUs, which do support it at the hardware level)
[04:11] <mdz> infinity: well, there's "might be handy in some deployments" and "should be there by default"
[04:11] <daniels> jdub: arguably Common Man shouldn't be maintaining stuff in main?
[04:11] <infinity> jdub : The CPUs definitely do.  I don't know if the package supports them.
[04:11] <jdub> mdz: i'd say there by default, removeable :)
[04:12] <jdub> daniels: building stuff, dude
[04:12] <jdub> daniels: nothing to do with ubuntu land
[04:12] <daniels> jdub: i think mdz's idea is a good one
[04:12] <jdub> i'm definitely not saying packages should b-d on a metapackage
[04:12] <jdub> yeah, though i'm not sure about tasks
[04:13] <jdub> but there should be a way for someone to say "I want to build my kernel" -> zap
[04:13] <daniels> er
[04:13] <jdub> "I want to write a GNOME app" -> zap
[04:13] <daniels> 'I WANT LESS SECURITY SUPPORT"
[04:13] <mdz> jdub: that is totally a build-dep use case
[04:13] <daniels> 'make that really easy'
[04:13] <jdub> mdz: i'm not thinking in terms of ubuntu land
[04:13] <jdub> consider dude who wants to build jhbuild
[04:13] <mdz> the one I'm driving at is sort of the traditional unix "if you install something, you have the development bits too" semantic
[04:13] <jdub> or write a gnome app
[04:14] <mdz> jdub: everything should be part of ubuntu land!
[04:14] <mdz> there is no other
[04:14] <jdub> "i appreciate your enthusiasm, but..." ;)
[04:14] <mdz> I'm not keen on optimizing for the case of unpackaged software
[04:14] <mdz> I'd rather it be a pain, to drive people into packaging ;-)
[04:14] <jdub> mdz: whoa, so one bit flip that says "i am a developer" and everything comes with related -dev?
[04:15] <jdub> or mark's idea of making sure related packages are installed
[04:15] <mdz> jdub: the former
[04:15] <jdub> if you have php and mysql installed, you should get the bindings
[04:15] <mdz> I haven't actually evaluated the list of packages which would result; it might be crack
[04:15] <mdz> but I think there's a greatest common factor "development STUFF" that we could do
[04:15] <jdub> ^ expression of idea, not statement ;)
[04:16] <mdz> which is like "what you expect to have"
[04:16] <jdub> mdz: particularly if it stressed the interfaces we support
[04:16] <mdz> we'd just have to add a jennifer check to reject packages which build-depend on it ;-)
[04:16] <jdub> should these things be packages at all?
[04:16] <jdub> what was the spec for fixing meta?
[04:17] <daniels> yeah, if anyone b-ds on it, I will file bugs every hour until it is fixed
[04:17] <jdub> daniels: you are so debian-devel
[04:17] <daniels> and then every two hours afterwareds out of spite, until I get bored
[04:18] <daniels> jdub: i hope the irony of saying that to a canonical employee in #ubuntu-devel hasn't escaped you, given threads of late
[04:18] <daniels> also, I haven't used any profanities in this discussion
[04:18] <jdub> i was counting on those ironies
[04:18] <infinity> (... yet)
[04:19] <daniels> jdub: and it's irony that's why we can't ship a x-development package
[04:19] <daniels> hth, hand, kthxbye
[04:19] <daniels> but, uhm, yeah
[04:19] <daniels> can we not use the task mechanism for this?
[04:19] <infinity> Tasks and metapackages are all about "all or nothing", though.
[04:20] <infinity> So, I say "I'm a developer" and I get every lib*{,-dev} in the archive?
[04:20] <infinity> That's very not keen.
[04:20] <jdub> it's odd that ian thinks glibc is too much divergence, but xorg and gnome aren't
[04:20] <jdub> infinity: for stuff you install
[04:20] <infinity> I want a switch that automatically makes sure I have a matching -dev for every lib I have installed, but not for ones I don't.
[04:20] <infinity> Which would require a bit more magic than tasks or metapackages.
[04:20] <jdub> yes, that's what we were talking about
[04:21] <infinity> To do it elegantly, it would require a new control field.
[04:21] <infinity> Headers: libfoo-dev
[04:21] <infinity> So when I install Package:, I get it's Headers:, if I want them.
[04:21] <infinity> Or something.
[04:21] <infinity> s/it's/its/... I can hear English teachers around the world screaming even now.
[04:22] <daniels> jdub: no dude, there's GOOD divergence (innovation), and BAD divergence
[04:22] <daniels> jdub: afaict, sid has all the bad divergence.  from itself. :)
[04:22] <infinity> (Of course, that control field can be added after the fact with ftpmaster overrides, so we don't have to touch each source package, initially)
[04:23] <daniels> infinity: dude, it's SUPERKEEN
[04:28] <infinity> daniels : Shall we get xorg building on i386 today?
[04:28] <infinity> daniels : I fear version skew and archive oddities until we get it happy again.
[04:28] <infinity> Check. :)
[04:28] <infinity> Yay.
[04:29] <infinity> Thanks.
[04:29] <daniels> ARE WE THERE YET?!? :)
[04:29] <infinity> Can I have an ice cream?
[04:31] <daniels> infinity: Not until you've eaten all your pumpkin, broccoli, and zucchini.

[04:38] <infinity> Like, you can have 12 versions of libdbXX installed, but only one libdb-dev.
[04:39] <daniels> well, it's not like anyone ever uses multiple versions of libdb
[04:39] <daniels> maybe you could have a Prefers header on your -dev, to make it the first of a conflicting set of packages
[04:39] <daniels> either that, or have apt's dependency resolver work out which one will involve removing the least packages.
[04:39] <daniels> which should actually be fairly trivial
[04:40] <infinity> apt's dependency resolver is black magic that no sane man (hi mdz!) wants to touch. :)
[04:40] <daniels> infinity: it's actually reasonably easy to do in this case.
[04:41] <infinity> Yeah, I know.  I just knee-jerk when anyone suggests diving into Culus's code.
[04:41] <daniels> instead of calling markRemove() on every package, you just increment a counter in an array of strings (or packages)
[04:41] <daniels> either that, or just have a list of packages to be removed for every package installed
[04:42] <Keybuk> I thought mvo was the APT maintainer these days? :p
[04:42] <daniels> then just pick the shortest one and install that
[04:42] <infinity> Keybuk : Not by choice, I'm sure. :)
[04:43] <daniels> hm, leaving in 5 hours or so.  probably should investigate this 'packing' theory.
[04:43] <Keybuk> bring back the dselect FTP method
[04:43] <daniels> Keybuk: dpkg-ftp!
[04:45] <daniels> hm.  on that note, wonder how I'll get to the airport.
[04:46] <Keybuk> where you going?
[04:46] <infinity> He's skipping the country to get out of the drinks he owes me.
[04:46] <daniels> eh, don't you still have Coopers in your fridge? :)
[04:47] <daniels> Keybuk: the only possible answer to this is 'your mum's house'
[04:47] <infinity> No.  It kinda went away.
[04:47] <infinity> (What?... it wasn't going to drink itself)
[04:47] <daniels> well, there's your answer
[04:49] <infinity> daniels : If your xorg stuff still isn't building by the time you have to find your way to the airport, care to dump your working directory in my lap, so I can fix it up?
[04:49] <daniels> infinity: don't have to leave for about 5 hours yet, so it would take a fairly tremendous effort, but yeah, will do
[04:52] <Keybuk> CHECKSUM FILE(S) DISAGREE WITH DIRECTORY LISTING ABOUT WHAT FILES SHOULD BE PRESENT IN REVISION DIR OF ARCHIVE
[04:52] <Keybuk> oh yes, I know this one
[04:52] <Keybuk> it means I'm doing more than one thing at once to the same archive
[04:52] <Keybuk> bad me, I should do less
[04:59] <daniels> haha
[04:59] <daniels> dad's sitting downstairs eating tuna out of the tin
[05:03] <mako> daniels: i've done that
[05:11] <mako> oinkmaster
[05:16] <Keybuk> is that a euphemism for something?
[05:21] <daniels> /etc/gdm/Xsession invoves #!/bin/ksh but does not have a depenedency on pdksh
[05:21] <daniels> not any other corn shell. Suggestions would be to add this dependancy or replace
[05:21] <daniels> the script with on that calls /bin/sh
[05:21] <daniels> SSSSEEEEEEEEEEEEBBBBBBBBBBBBBBBBBBBBBBBB
[05:21] <daniels> infinity: seemed to work fine on amd64 here, doing a chroot test now
[05:22] <daniels> infinity: i can throw you a version if you want to do an i386 chroot test?
[05:22] <infinity> daniels : Well, it built on amd64 before, so you proved nothing. :)
[05:22] <infinity> daniels : Unless you turned on the via driver on amd64 in this version..
[05:22] <daniels> infinity: yeah, good point
[05:22] <daniels> (nope, not yet, but I might on the next CVS pull, since they allege it's hopefully less broken)
[05:22] <infinity> (Which isn't an entirely insane idea)
[05:23] <infinity> And disable it on ia64, while you're at it.
[05:23] <infinity> The idea of an ia64 machine with a VIA embedded video chipset seems laughable to me.
[05:23] <Amaranth> yeah
[05:23] <daniels> stop oppressing ia64 users
[05:24] <daniels> infinity: so ... got a fast i386 around? :)
[05:24] <infinity> It's in the mail.
[05:24] <daniels> wonder if I should feed libxext to the dogs of war^W^W^Warchive
[05:24] <infinity> (I can spin it on a buildd)
[05:24] <daniels> that'd be good if you could throw them both in locally
[05:26] <daniels> libxext is in chinstrap:~daniels (well, scping now)
[05:26] <daniels> there you go, have a libxext
[05:26] <daniels> this time I even used B-D rather than B-D-I
[05:26] <infinity> Good boy.
[05:26] <daniels> xorg uploading now
[05:31] <daniels> infinity: xorg's there too, when libxext's finished building (or not)
[05:41] <robitaille> jdub,  ping
[05:42] <jdub> robitaille: pong
[05:42] <jdub> elmo: planet sync please
[05:43] <robitaille> jdub,  is it possible to put my blog on planet.u.c?
[05:43] <jdub> elmo: s/sync/update/ ykwim
[05:43] <jdub> robitaille: haha! elmo needs to update it for me :)
[05:43] <jdub> robitaille: committed the change only half an hour ago or something
[05:43] <robitaille> ok...wasn't sure you got my email a few days back.  thanks.
[05:44] <Keybuk> daniels: is libx11 known fucked? :p
[05:44] <daniels> Keybuk: i386 is known to be shit
[05:44] <daniels> libx11 works just fine
[05:44] <Keybuk> I can't upgrade it
[05:44] <infinity> libx11 is fine, if xorg is built.
[05:44] <Keybuk> it conflicts on xlibs-data
[05:44] <daniels> Keybuk: i know
[05:45] <daniels> so upgrade xorg :P
[05:45] <infinity> Keybuk : That's cause xorg didn't build on i386.
[05:45] <Keybuk> heh
[05:45] <infinity> Keybuk : Working on that right now.
[05:45] <Keybuk> right, known-fucked
[05:45] <infinity> (but libx11 is fine)
[05:45] <Keybuk> I guess this is also why my X server won't start? ;)
[05:45] <infinity> NEAT.
[05:45] <Keybuk> Fatal server error:
[05:45] <Keybuk> could not open default font 'fixed';
[05:45] <Keybuk> ... sweet
[05:46] <jdub> do a dpkg-reconfigure
[05:46] <infinity> I think that's the single most-reported X error message.  Ever.
[05:46] <daniels> Keybuk: it can be one of two things
[05:46] <Keybuk> on what?
[05:46] <daniels> Keybuk: a) you have a custom config, and need to change all your font paths to /usr/share/X11/fonts/foo
[05:47] <daniels> b) i'm shit, and mkfontdir is broken
[05:47] <daniels> b can happen in a corner case on !i386 at the moment, I'm fixing that
[05:47] <Keybuk> daniels: it was (a)
[05:47] <daniels> Keybuk: ahar
[05:47] <daniels> Keybuk: that'll go away when I fix every universe package
[05:47] <daniels> ha ha ha
[05:47] <daniels> uhm, yeah
[05:48] <Keybuk> "every universe package" ?
[05:48] <daniels> zgrep /usr/X11R6/lib/X11/fonts Contents-foo.gz
[05:48] <daniels> kind of hard to turn it into a symlink :P
[05:48] <Keybuk> heh
[05:48] <Keybuk> why?
[05:48] <Keybuk> dpkg lets you do that
[05:48] <Keybuk> and doesn't whinge like a bitch about it either
[05:49] <daniels> i don't want a 1266-character Replaces line, either :P
[05:49] <Keybuk> turning directories into symlinks is expressly allowed (and the reason for much pain when people try and reverse it)
[05:49] <Keybuk> heh
[05:49] <daniels> Keybuk: by the way, dpkg doesn't have any restrictions on line length in control fields, does it?
[05:50] <daniels> because, hypothetically, if one was to have a 1282-character Build-Depends line ...
[05:50] <daniels> that was expected to grow before it expanded
[05:50] <dilinger> daniels: you could split it up amongst multiple lines
[05:50] <Keybuk> none that I'm aware of
[05:50] <Keybuk> if you find out, let me know ;)
[05:50] <daniels> word
[05:50] <daniels> heh
[05:50] <Keybuk> dilinger: that so doesn't work
[05:51] <Keybuk> though I'm sure evolution would've found it by now
[05:51] <Keybuk> evolution, the primary use case for "Build-Depend: the whole fucking archive"
[05:51] <daniels> Keybuk: it would be really cool if it could just do Build-Depend: headers
[05:51] <daniels> and drag in the whole fucking archive that way :P
[05:51] <Keybuk> which headers? :p
[05:51] <dilinger> Keybuk: eh?  it works for depends
[05:51] <Keybuk> dilinger: FSVO. works
[05:51] <dilinger> heh
[05:54] <Keybuk> dpkg tends to be happy about it, but other things (*cough*katie*cough*) don't
[05:54] <daniels> just like .bz2 debs :P)
[05:54] <Keybuk> there's an open wishlist about dpkg automatically folding them into single lines
[05:54] <daniels> (diveintopython, anyone?)
[05:54] <Keybuk> katie is fine with them?
[05:56] <daniels> when it was uploaded, katie choked
[05:57] <Keybuk> her gag reflex is improving
[05:58] <jdub> on his sparc :)
[05:58] <jsgotangco> opensolaris!
[05:58] <daniels> COMMUNITY EDITION
[05:58] <jsgotangco> not good enough?
[06:05] <Unfrgiven> im having a problem with a package at the moment. when doing a pbuilder build, it fails as it can't find gcc-3.4. but the source doesnt have any reference to gcc-3.4. what gives?!?
[06:06] <Amaranth> what package?
[06:06] <Unfrgiven> Amaranth: gmetadom
[06:06] <Amaranth> (i can't help, but it would help others find the problem)
[06:13] <fabbione> morning
[06:15] <daniels> shooting shoulder-to-fingertip pains -> afk for a bit
[06:16] <Keybuk> daniels: we told you to ease up on the heroin
[06:45] <daniels> infinity: poke
[06:50] <infinity> daniels : ekop
[06:50] <daniels> infinity: any buildd love?
[06:50] <infinity> daniels : Still building.  (xext was fine)
[06:50] <daniels> dude, your buildd is crap
[06:56] <infinity> ../../programs/Xserver/hw/xfree86/drivers/libdriver.a(via_drv.o): In function `ViaInitXVMC':
[06:57] <infinity>  /build/buildd/xorg-6.8.2/build-tree/xc-xserver-xorg-dbg/programs/Xserver/hw/xfree86/drivers/via/via_xvmc.c:390: undefined reference to `xf86XvMCRegisterDRInfo'
[06:59] <daniels> ...
[06:59] <bob2> it would be via, wouldn't it
[06:59] <daniels> nothing even CHANGED there!
[06:59] <daniels> unless they typoed it
[06:59] <infinity> Indeed.
[07:00] <Lathiat> daniels: can you tell me (I can't find any descriptions) what XvMC does?
[07:01] <daniels> Lathiat: it's 'xvideo motion compensation'.  basically, allows you to feed video in at an earlier stage of the pipeline (i think you can basically cram raw MPEG into some cards), so you have to do less decoding in software.
[07:04] <Lathiat> ahh, what the hell does that have todo with 'motion compensation'? :)
[07:05] <Amaranth> http://www.gnome.org/~carlosg/stuff/gst/new-services.png
[07:05] <Amaranth> isn't that what BUM does?
[07:05] <Lathiat> Amaranth: somewhat,but thats *much* simplerand usable. :)
[07:05] <daniels> Lathiat: well, it makes it more tolerable when you have large videos with lots of movement :) else you can't get the data in fast enough, so your videos end up looking like arse
[07:06] <Lathiat> daniels: oh i see
[07:06] <daniels> Lathiat: i think there's also some code in the hardware-specific libraries to enable actual on-card motion compensation though
[07:06] <Amaranth> Lathiat: Exactly, BUM is more of a poweruser interface.
[07:07] <Lathiat> Amaranth: i like that interface :)
[07:07] <Lathiat> actually
[07:07] <Lathiat> does that mean
[07:07] <Lathiat> 'running'
[07:07] <Lathiat> or does  it mean
[07:07] <Lathiat> 'start on boot'
[07:08] <daniels> infinity: ... i honestly have no idea how this ever built.
[07:08] <infinity> daniels : Blind luck..?
[07:09] <daniels> oh, I see how
[07:09] <daniels> clever.
[07:10] <daniels> infinity: edit debian/patches/000_stolen_from_unichrome.sf.net.diff
[07:10] <daniels> infinity: search for XvMCRevision
[07:11] <jdub> so is usplash going to happen for breezy?
[07:11] <daniels> infinity: change #if foo || bar to #if defined(fnord) && ...
[07:11] <daniels> jdub: it needs some sweet bountying
[07:11] <daniels> jdub: i'm on the verge of overcommitted as it is
[07:12] <jdub> what about that dude who wrote his own usplash?
[07:12] <Amaranth> splashy?
[07:12] <daniels> i thought it wasn't what we wanted?
[07:12] <jdub> no, but the dude might be able to do it
[07:13] <daniels> probably worth a shot
[07:14] <lamont> daniels: what does this mean fro libx11's build?
[07:14] <lamont> ../../include/X11/Xlib.h:3575: error: parse error before '_X_SENTINEL'
[07:14] <infinity> lamont : It's fixed.
[07:14] <Lathiat> daniels: so, xvmc wouldnt help if i was playing xvid?
[07:14] <infinity> lamont : You need the latest x11proto-core
[07:14] <infinity>  ... -dev
[07:15] <infinity> (Yes, maybe I should have fixed libx11 to have a versioned build-dep on x11proto-core-dev, but I figured since it was just uploaded, we can consider it a bugfix and move on)
[07:20] <infinity> daniels : Changing those to ifdefs will change what they were trying to test for...
[07:22] <infinity> daniels : Or, did you want "if defined(XvMCRevision) && XvMCVersion > 1" (which still looks diffrently wrong to me)
[07:28] <daniels> infinity: you want to end up with defined(fnord) && (XvMCVersion > 1) || (XvMCRevision > 0), or whatever
[07:28] <daniels> basically, the protocol headers are now up to scratch, but we haven't got the latest implementation details in the server
[07:28] <daniels> so we never want it to test true
[07:30] <infinity> "Note that this severely brakes libviaXvMC and when this commit
[07:30] <infinity> is transferred to Xorg, libviaXvMC build must be disabled until
[07:30] <infinity> an update is made of that library."
[07:31] <daniels> nah, no-one cares about libviaXvMC
[07:31] <daniels> the whole concept of separate hardware-dependent client-side libraries is UTTER, UTTER, UTTER CRACK
[07:33] <daniels> crikey, it seems vfat is rather buggy
[07:34] <daniels> keeps trying to write beyond the end of my iAudio
[07:41] <infinity> ...
[07:42] <infinity> Generic USB mass storage?
[07:42] <fabbione> daniels: what kernel?
[07:44] <daniels> fabbione: .10 still
[07:44] <daniels> infinity: yeah
[07:46] <infinity> That sounds more likebuggy hardware misreporting its volume/partition sizes.
[07:47] <daniels> it's always been good to me
[07:47] <daniels> i re-mkfs.vfat'ed the thing, upgraded the firmware, and I'm going to try again
[07:48] <infinity> Oh, for the love of...
[07:49] <infinity> daniels : Would you believe that "fnord" actually /is/ defined?
[07:50] <daniels> infinity: you.  are.  not.  serious.
[07:51] <infinity> Well, either that or you suck at cpp operator precedence.
[07:52] <daniels> yeah, just make the whole line #if 0
[07:58] <infinity> FASTER!
[08:04] <fabbione> Kamion: can you please readd usb-modules to sparc64 in d-i? we will start shipping it in 2.6.12
[08:04] <fabbione> (we don't care if it fails with 2.6.10)
[08:05] <infinity> daniels : Oh, you may want to upload xext now to give the buildds a head start on that before we get xorg up.
[08:05] <daniels> infinity: good point
[08:06] <daniels> uploading now
[08:06] <jdub> can i turn the screen back on with a command?
[08:06] <jdub> backlight
[08:06] <jdub> gdm is running, but the screen is still off
[08:06] <jdub> i'd hate to be unable to type *and* unable to see it :-)
[08:09] <infinity> daniels : make install is going... Looks like we're okay with that "if 0" fix.
[08:09] <infinity> daniels : So, you can do that and upload xorg too.  (it'll get auto-dep-waited just fine)
[08:10] <daniels> infinity: thanks
[08:16] <infinity> Build finished fine, BTW.
[08:16] <jdub> aha, new xserver + xset :)
[08:17] <jdub> daniels: still no keyboard love
[08:17] <Burgundavia> jdub, the reports I have seen from splashy on the forum seem to be very positive. It would also be great press
[08:17] <jdub> daniels: what else can i do to help debug?
[08:17] <jdub> Burgundavia: hrm?
[08:18] <Burgundavia> jdub, the ubuntu specific graphical boot thingy
[08:18] <daniels> jdub: do you have new xorg and shit?
[08:18] <jdub> daniels: the very latest
[08:18] <robitaille> smurfix,  ping
[08:18] <daniels> jdub: like, xorg -25, libx11-6 1:6.2.1+cvs.20050615
[08:18] <mdz> is -25+ in the archive now?
[08:18] <daniels> cock
[08:18] <jdub> Burgundavia: graphical boot == great press?!
[08:18] <daniels> mdz: 25 is in for amd64/powerpc, 26 has just gone in to fix a stupid via cock ftbfs
[08:18] <smurfix> robitaille: 
[08:19] <Burgundavia> jdub, no, adopting something that someone on the forums created
[08:19] <daniels> (as an added bonus, I broke another big library out of 26)
[08:19] <jdub> daniels: i have 6.8.2-24 of both
[08:19] <robitaille> smurfix,   is smurflogbot a new permanent feature of all the LoCo channels?
[08:19] <mdz> now is the time for breaking things
[08:19] <smurfix> robitaille: yes
[08:19] <daniels> jdub: ok, wait until new xorg lands
[08:20] <robitaille> smurfix,  great; so we can now turn off our own canadian home-made logger
[08:20] <fabbione> daniels: ah... new more crack??
[08:20] <jdub> mdz: except for my keyboard. now is the time to fix my keyboard.
[08:20] <daniels> fabbione: libXext is gone
[08:20] <smurfix> robitaille: I can import your old logs if you want them to continue being available
[08:20] <fabbione> daniels: is libxaw8 gone?
[08:21] <daniels> fabbione: killed it in -25
[08:21] <fabbione> daniels: (sparc related question)
[08:21] <fabbione> daniels: ok.. that explain
[08:21] <mdz> jdub: now is the time to work around your keyboard so that development marches on ;-)
[08:21] <fabbione> hey mdz...
[08:21] <mdz> fabbione: morning
[08:21] <daniels> jdub: i had the exact same problem, it sort of mystically disappeared with a new libx11 and a couple of symlinks.  um.
[08:21] <robitaille> smurfix,  I'll ask the owner of the logger if he wants to;  but I think the logs are somewhat incomplete
[08:21] <fabbione> mdz: 1.2 is almost done.. i only need to sort a few more modules and do a build test orgy
[08:21] <daniels> jdub: if not, I'll try to dive headlong into the code and work out WHAT THE FRIG IS GOING ON
[08:22] <daniels> (worst.  code.  ever.)
[08:22] <mdz> fabbione: ok. I probably won't be awake for much longer
[08:22] <fabbione> mdz: you will get it when you wake up
[08:22] <fabbione> mdz: there is no way i can manage to build it in time before you go to sleep :)
[08:23] <daniels> bbiab, got to head down to the post office
[08:23] <fabbione> mdz: and not because i wouldn't love too
[08:23] <mdz> fabbione: when you're ready, ask elmo to move it into main
[08:23] <fabbione> mdz: we don't have machine fast enough for that :P
[08:23] <fabbione> mdz: yup
[08:24] <robitaille> smurfix,  what's the prefered URL for your irc logs.  I see one with "free" and one with "freenode" in their URL; they appear to be the same.
[08:26] <smurfix> robitaille: freenode. The "free" is just there because the IRC server truncated the bot's name and I didn't want to restart it Yet Again.
[08:28] <robitaille> smurfix,  ok; but  "/whois smuflogbot" display a link to the "free" url.
[08:29] <smurfix> robitaille: That's what I said -- that text comes from the Freenode IRC server.
[08:29] <robitaille> oh.
[08:29] <smurfix> My bot logs in with "freenode" at the end when it logs on. :-/
[08:35] <infinity> mdz : Can we get libxext shoved through NEW?
[08:35] <mdz> infinity: not by me right now, but elmo should be awake soon enough
[08:35] <infinity> Alright, I'll pester him in a while.
[08:36] <mdz> night
[08:36] <infinity> 'Night.
[08:38] <fabbione> night mdz
[08:39] <pitti> Good morning
[08:41] <fabbione> morning pitti
[09:09] <fabbione> maswan: ping?
[09:20] <\sh> any ia64 specialists around? ,-)
[09:21] <\sh> http://people.ubuntu.com/~lamont/buildLogs/c/cln/1.1.9-1ubuntu2/cln_1.1.9-1ubuntu2_20050616-2350-ia64-failed.gz
[09:21] <\sh> but the other archs were ok...
[09:30] <robitaille> smurfix,  to quote from your e-mail:  "log a hundred channels during the next ten years".  Pretty impressive future to think of it that way.
[09:30] <smurfix> robitaille: I sure hope so
[09:32] <robitaille> smurfix,  I'm also hoping for it.  But first, we have to do 10x10.
[09:32] <xuzo> hi smurfix, I reply your email to loco maillist. Category "CategoryUserpages" is not created.
[09:33] <smurfix> xuzo: Bah. Thanks.
[09:34] <xuzo> ;)
[09:36] <infinity> \sh : I think I'll defer to an ia64 porter on that one.  The cause doesn't immediately jump out at me.
[09:36] <\sh> infinity: i checked the upstream bugzilla and ml...but they don't mention anything concerning this problem...actually it's really on ia64 :(
[09:37] <infinity> Smells like toolchain to me.
[09:38] <\sh> problem is, i'm not an expert in things like assembler..the last time i programed with assembler was on a z80
[09:38] <Lathiat> i wrote some x86 asm
[09:39] <Lathiat> booted from a floppy toggle bitsonthe parallel port that showed up as 8 leds on this little device.:)
[09:53] <mvo> Lathiat: what sort of http error?
[09:53] <robitaille> elmo,   is it possible for members to get @ubuntu.com addresses now, or it's something that will only be done in the future?
[09:54] <elmo> jdub: done
[09:55] <elmo> robitaille: not right now, but it's going to be done fairly soon
[09:55] <robitaille> elmo,  ok.  thanks
[09:56] <fabbione> Kamion: ping?
[09:58] <pitti> infinity: is pkgstriptranslations 13 already installed in all buildds?
[09:59] <Kamion> fabbione: pong
[10:00] <fabbione> Kamion: scsi-*.udeb question :/
[10:01] <fabbione> Kamion: other than scsi-core... there are 3 other scsi- udebs.. all in priority standard.. do they all get loaded? or there is still a specific selection per arch?
[10:03] <Kamion> anything priority >= standard gets loaded when anna runs
[10:03] <Kamion> however some of those are built into the initrd and some aren't
[10:03] <Kamion> depending on the architecture and the installation medium type
[10:04] <fabbione> hmmm ok
[10:04] <fabbione> i guess i will just stick the new scsi drivers in extra than
[10:04] <fabbione> and review it later on with you
[10:05] <fabbione> Kamion: scsi and nic are the 2 most problematic ones at the moment.. all the others are pretty clean
[10:08] <Kamion> if they're not the sort of SCSI controllers you'd find on huge numbers of machines, scsi-extra's generally sane
[10:10] <fabbione> Kamion: i think i will defer scsi-* nic-* cleanup for the next upload.. we have no rush to add
[10:10] <fabbione> Kamion: and given that they are the last 2 sets of udebs that needs cleaning
[10:10] <Kamion> sure
[10:11] <fabbione> Kamion: did you also read the scrollback about sparc64/usb?
[10:11] <fabbione> Kamion: if you want you can readd it. i will start shipping the udebs with .12
[10:11] <Kamion> no, my machine crashed, haven't read scrollback at all
[10:11] <fabbione> so it's less diff to merge for you
[10:11] <fabbione> ah ok
[10:12] <Kamion> I'll re-add it when we switch to .12 then
[10:13] <fabbione> Kamion: it's ok with me at any time.. we need to move .12 in main today
[10:13] <Kamion> ah, ok
[10:13] <fabbione> so i think we can switch from monday or when you fell more confident with it
[10:18] <infinity> pitti : Is that the one you uploaded over a day ago?... I'm pretty sure it got installed yesterday when I was doing a bunch of manual chroot maintenance.
[10:18] <pitti> infinity: I uploaded it yesterday noon
[10:18] <pitti> infinity: it removed the KDE blacklist so that kde-i18n-* gets stripped now
[10:19] <infinity> pitti : Yeah, I think that's installed on all 12.
[10:19] <pitti> infinity: ok, great. Then I can ask Riddell to reupload the kde-i18n stuff
[10:20] <pitti> let's break KDE translations and annoy KDE users :-)
[10:20] <infinity> I'm all for that.  I did two KDE uploads a few minutes ago too, in hopes of irritating people.
[10:21] <infinity> (Well, that, or to fix bugs)
[10:23] <pitti> elmo: according to the LanguagePackRoadmap we split the packs into KDE/Gnome/other; so this will require some major NEW love in the near future; is there anything we should do to ease this pain?
[10:24] <jdub> elmo: thanks!
[10:24] <elmo> pitti: not do it?
[10:25] <elmo> so we've gone from 150 x 3 to what?  150 x 6?  rock on
[10:25] <pitti> elmo: we don't split the support packages
[10:27] <elmo> pitti: where's this spec?
[10:28] <pitti> elmo: 
[10:28] <pitti> http://udu.wiki.ubuntu.com/LanguagePackRoadmap
[10:28] <elmo> oh, blah udu
[10:28] <pitti> (skip the -support split, we don't do it)
[10:28] <infinity> Surely you remember UDU... Lots of fruit flavoured candies, people speaking with Australian accents?
[10:29] <pitti> lots of blood flowing in our head, standing upside down all the time...
[10:30] <elmo> what's the rational for splitting the non-support stuff?
[10:31] <pitti> elmo: if we didn't, the langpacks would contain all the KDE and Gnome translations, although almost no user will require them both
[10:31] <pitti> elmo: i. e. Ubuntu isntalls l-p-gnome-*, Kubuntu installs l-p-kde-*
[10:31] <jdub> elmo: RATIONALE
[10:31] <elmo> jdub: IT WAS A TYPO
[10:31] <daniels> elmo: ARE YOU SAYING IT'S IRRATIONAL?!?
[10:31] <elmo> pitti: dude, by definition the language packs are going to contain things most users will never use
[10:31] <jdub> elmo: IT MAKES ME MAD!!!
[10:31] <elmo> what makes the kde/gnome split special? 
[10:32] <jdub> enormous amounts of translations
[10:32] <daniels> jdub: ITS JUST A SIMPLE MISTAKE, A PERSONS GRAMMAR DOESNT MATTER BUT THERE INTENT
[10:32] <elmo> we went down the one language pack per language thing for a reason, and you seem to be creeping back down the slope to madness
[10:32] <pitti> elmo: the special thing is we don't translate KDE yet at all
[10:32] <daniels> pitti: wow, that is special :)
[10:33] <pitti> elmo: well, we can't fit all the KDE translations onto the Ubuntu CDs, for example
[10:33] <infinity> "there intent"... I hope that was intentional irony.
[10:33] <pitti> elmo: vice versa for Kubuntu CDs
[10:34] <pitti> elmo: btw, supporting KDE wasn't exactly *my* idea ... :-)
[10:35] <jdub> pitti: perhaps we should make the split down derivative lines-- hrm, no
[10:35] <jdub> that's fucked up
[10:35] <pitti> jdub: as long as Kubutu and Ubuntu are built from the same archive and so on, that doesn't work
[10:36] <jdub> yeah
[10:36] <pitti> if Kubuntu was a proper derivative with its own buildds, Kubuntu could have its own pkgstriptranslations
[10:36] <pitti> and adapt the langpacks
[10:36] <pitti> but right now we can't
[10:36] <daniels> infinity: absolutely
[10:37] <\sh> infinity: can u do me a favour and have a look where openscenegraph is on the buildd? i uploaded it yesterday, but it didn't show up as compiled or not compiled ;)
[10:38] <infinity> \sh : dep-wait on giflib-dev
[10:38] <infinity> \sh : Probably a broken dep-wait.  Let me look at it in a few minutes.
[10:40] <\sh> hmmm...giflib3g i have here
[10:40] <infinity> \sh : It was a bug in the auto-dep-waiter, since fixed.
[10:41] <infinity> \sh : Packages released and building.
[10:41] <\sh> thx :) 
[10:46] <infinity> daniels : Is your suitcase filling up yet?
[10:46] <pitti> brb
[10:49] <Keybuk> *chuckle* it seems that Manoj comes from the daniels school of testing code
[10:49] <Keybuk> "let other people do it"
[10:50] <Keybuk> the SELinux patches he wrote for dpkg don't work
[10:50] <Keybuk> and, in fact, they don't just not work -- they don't even get compiled in to the resulting binary
[10:53] <infinity> Rock.
[10:53] <daniels> infinity: slowly
[11:06] <daniels> ... epochs?
[11:07] <daniels> oh, libxext-dev b-d
[11:09] <fabbione> uhuhuh davis for the first time in history has been faster than ia64 :)
[11:11] <fabbione> Kamion: http://people.ubuntu.com/~fabbione/foo <- it should be detailed enough....
[11:14] <Kamion> fabbione: s/wegde/wedge/g please :-)
[11:14] <fabbione> op
[11:14] <fabbione> s
[11:14] <Kamion>         + Update fat-modules to force all modules inclusion.
[11:14] <Kamion> what's that?
[11:15] <Treenaks> Kamion: fat vs vfat -> uppercase vs lowercase I think?
[11:15] <fabbione> Kamion: they were optional (-drivers/...) but since we ship them, i enforced the check removing the -
[11:16] <Kamion> Treenaks: fat-modules already contained both
[11:16] <Kamion> fabbione: ah, ok, thanks
[11:16] <fabbione> Kamion: i did try to enforce the sanity checks as much as i could
[11:16] <fabbione> Kamion: given that sanity checks are good
[11:18] <Kamion> fabbione: FWIW it'd be easier to read the changelog if it were of the form "added the following to d-i/shared/; made the following .lnk files point to shared: ...; did other stuff: ..."
[11:18] <Kamion> since I want to read by category of change, not by individual udeb
[11:19] <fabbione> Kamion: i prefer this way to be hounest... if you want i can rewrite it, but it won't take me exactly 2 minutes
[11:19] <Kamion> fair enough, your choice
[11:20] <fabbione> Kamion: it needs to be confortable for you to read too...
[11:20] <fabbione> so i have no problems to rewrite it :)
[11:20] <fabbione> it will just take some time
[11:25] <fabbione> humpf.. ppc barfed
[11:27] <sladen> 00/query jamesh
[11:27] <sladen> ping
[11:30] <infinity> fabbione : My PPC will be here on tuesday, BTW.
[11:30] <fabbione> infinity: cool
[11:30] <infinity> I'm hoping it'll come in handy. :)
[11:30] <fabbione> infinity: but the hw on davis is ok
[11:31] <fabbione> the build barfed for a duplicate module :)
[11:31] <infinity> Yeah.  But you can't reboot davis.
[11:31] <fabbione> ppc64 kernels > *
[11:31] <fabbione> i don't need to :)
[11:31] <infinity> (Yet)
[11:31] <fabbione> don't even consider to touch it :)
[11:32] <fabbione> brb
[11:46] <Lathiat> mvo: various, bad header line is popular
[11:51] <Riddell> pitti: should I re-upload kde-i18n now?
[11:51] <pitti> Riddell: according to infinity it should be stripped now
[11:51] <pitti> Riddell: so if the new packages can build POT files, then go ahead :-)
[11:51] <pitti> Riddell: it would be nice to get POTs and stripped files in one shot
[11:52] <ogra> infinity, any idea why xlibmesa-glu-dev is missing from hoary since yesterday ? there are a lot of user complaints on the mailing list
[11:54] <elmo> fixed
[11:55] <ogra> thanks
[11:57] <elmo> added hilarie to cron.daily too
[12:06] <jdub> so, future of ubuntu ppc
[12:06] <jdub> "how much does ibm care?"
[12:08] <infinity> Well, clearly someone needs to convince IBM to start making PPC laptops, since they just sold off their x86 desktop/laptop range. :)
[12:09] <Mithrandir> heh
[12:09] <Mithrandir> given their problems with getting laptop chips to apple, I'm not sure apple would be happy
[12:09] <infinity> Somehow I doubt that making Apple happy is at the top of IBM TODO list these days.
[12:10] <infinity> s/IBM/IBM's/
[12:10] <Lathiat> heh i was about to say that
[12:11] <tseng> neither is selling hardware to a small niche home user
[12:12] <fabbione> elmo: linux-source-2.6.12_2.6.11.94-1.2 is on the way to jackass. it will need some NEW love (udebs) and it will need to go in main (per mdz request).
[12:12] <infinity> tseng : Shh, a man can dream.
[12:15] <infinity> fabbione : Have you seen any segfaults during your builds since davis got its new kernel?
[12:15] <fabbione> infinity: not one 
[12:15] <infinity> Dayum.
[12:15] <fabbione> and with -j200 the machine does NOT suffer at all
[12:15] <infinity> Kay, then I think we need to seriously look at upgrading the buildds.
[12:15] <infinity> The segfaults on the PPC buildds are irritating.
[12:15] <fabbione> i think i am going to push it up to -j500
[12:16] <fabbione> concordia had problems with 500 because of only 2GB of ram
[12:16] <fabbione> but davis has 3 :P
[12:16] <infinity> Anyone ever tell you that you're weird?
[12:16] <fabbione> infinity: why? i love to wake elmo because nagios starts to scream and yell for the load :)
[12:17] <fabbione> it's a good stability test for the kernel
[12:17] <Mithrandir> infinity: because he likes to exercise the machines?
[12:17] <fabbione> nobody uses them.. i *RAPE* them :)
[12:19] <fabbione> humpf.. only 20KB changes
[12:20] <fabbione> i need to do better next time
[12:20] <fabbione> time for food :)
[12:33] <tseng> "Hardening Linux: a 10 step approach to a secure server"
[12:33] <tseng> I love you, OSNews
[12:34] <Mithrandir> tseng: didn't you know?  Security is just this module you bolt on at the end.
[12:34] <tseng> Mithrandir: its a link to a wordpress blog, ironically
[12:44] <infinity> elmo : openscenegraph binaries need NEW love.
[12:51] <infinity> Alright, for i386/breezy users complaining that xorg was somewhat wedged, you should be good to upgrade now.
[12:53] <fabbione> infinity: have fun
[12:54] <JaneW> mjg59: PING
[12:56] <jsgotangco> JaneW, hi
[01:04] <JaneW> hi jsgotangco 
[01:07] <mjg59> JaneW: Hi
[01:08] <Treenaks> any more news about October? :)
[01:21] <pitti> @all: we want to install /sbin/unix_chkpwd as setgid shadow by default (instead of setuid root); however, setuid root is required for nis, therefore we need a statoverride there (a bit evil, but this was agreed on with Keybuk); can anybody please take a look at http://people.ubuntu.com/~pitti/nis.unix_chkpwd-deroot.diff to check for obvious mistakes?
[01:21] <JaneW> mjg59: hello
[01:22] <mjg59> JaneW: Hi
[01:22] <JaneW> mjg59: are you still running with the BreezyGoal: LaptopMission?
[01:22] <mjg59> JaneW: Yup. It'll be updated tomorrow.
[01:22] <mjg59> Sorry, far too much real life for the past couple of weeks :)
[01:23] <JaneW> mjg59: because it has still not been updated on http://udu.wiki.ubuntu.com/UbuntuDownUnder/BreezyGoals
[01:23] <JaneW> mjg59: ok thanks. I was concerned that it had been forgotten
[01:23] <JaneW> being a top priority goal that would not be good.
[01:23] <mjg59> Sure, no problem
[01:23] <JaneW> cool thanks
[01:25] <JaneW> could anyone else who hasn't updated the status of their BreezyGoals please do so soon. http://udu.wiki.ubuntu.com/UbuntuDownUnder/BreezyGoals
[01:25] <JaneW> Please put the date of the update at the start of the line for easy refernce. Thanks.
[02:07] <Lathiat> elmo: ping
[02:08] <Lathiat> elmo: unping
[02:08] <Lathiat> jdub: ping
[02:20] <Duck_work> jbailey: build-dep missing on dash Sir
[02:20] <jbailey> Duck_work: Context?
[02:22] <Duck_work> jbailey: cdbs2 of course
[02:22] <jbailey> Ah. =)
[02:22] <Duck_work> i'm trying to build it
[02:22] <jbailey> Yeah, it's not intended to require dash in the end.  That's just what we're using for testing for now.
[02:22] <tseng> jbailey: do you have a design doc about cdbs2?
[02:23] <jbailey> I wonder if we should add the b-d
[02:23] <Duck_work> jbailey: temporary b-d is not so awful to remove later
[02:23] <tseng> ah, skunkworks
[02:23] <jbailey> The important bits are between the beer stain and the french fry grease. =)
[02:23] <elmo> library packages called libfoo-cvs are JUST WRONG
[02:24] <seb128> jbailey: do you know why some cdbs packages have a .cdbs-config_list?
[02:24] <tseng> elmo: its better if you have a bunch of stuff build-dep on it
[02:24] <tseng> elmo: and no epoch
[02:24] <seb128> jbailey: ie, gaim has gaim-1.3.1.tar.bz2.cdbs-config_list
[02:25] <pitti> seb128: I think that happens to all cdbs packages which were built recently; it certainly happens to all of mine and I think it's annoying...
[02:25] <jbailey> tseng: Essentially the design stuff that I have is scattered across IRC sessions, phone calls, and in person descriptions.  The problem is that the design evolves everytime dilinger and I talk to someone or touch it.
[02:25] <jbailey> seb128: Ugh, I don't.  Sorry.
[02:25] <seb128> pitti: yeah, same for me
[02:26] <jbailey> tseng: I find it easiest to talk about it relative to the current cdbs, rather than as a complete design on its own.  We've committed that it will have at least some basic docs in it before the first upload this time, though.
[02:26] <seb128> that and it put config.guess/sub stuff to the diff for a simple source rebuild
[02:26] <azeem> I guess nyu put that in, it's his automatically-update-config.{sub,guess} magic
[02:26] <seb128> I hate this magic
[02:27] <pitti> seb128: yes, that's why I try to purge autotools-dev as often as possible...
[02:27] <Duck_work> seb128: the condig.* stuff was due to bad stuff in clean rules, it should be solved by now
[02:27] <azeem> s/magic/black magic/
[02:27] <elmo> seb128: welcome to how we all feel about cdbs :-P
[02:27] <seb128> elmo: ah ah :)
[02:28] <Duck_work> elmo: not all, happilly
[02:29] <fabbione> elmo: did you read my message in the scrollback about 2.6.12?
[02:30] <elmo> fabbione: yeah
[02:30] <fabbione> elmo: rocking
[02:30] <jbailey> seb128: The magic that Robert put in has too many side effects, and no testsuite bits so I can't even feel safe about pulling it out with a minimum of side effects.  It's incentive to do more with cdbs2.  Testsuites and docs are basically all that's left for something that is good enough for most autotools uses.
[02:30] <seb128> elmo: can you sync nss-mdns please?
[02:31] <Lathiat> seb128: interesting timing, i was just talking about that
[02:31] <seb128> Lathiat: ross asked for it
[02:32] <Lathiat> cus lennart (author) was just asking me why it was so old. :)
[02:32] <elmo> seb128: done
[02:32] <seb128> elmo: thanks
[02:43] <seb128> Duck_work: fixed with what version?
[02:48] <jdub> Lathiat: pong
[02:48] <\sh> who can adjust the frozenappslist?
[02:49] <Lathiat> jdub: 'sok, was gonna ask about nss-mdns
[02:49] <Lathiat> jdub: but seb just got it synced so its all good:)
[02:50] <\sh> argl
[02:50] <\sh> i'm too slow
[02:51] <infinity> \sh : What needs adjusting?
[02:52] <seb128> is galeon still frozen?
[02:52] <infinity> No.
[02:53] <seb128> cool
[02:53] <seb128> thanks
[02:55] <\sh> infinity: it's ok...i was to slow...new package from debian arrived
[02:56] <jdub> YAY NEW KERNEL
[02:56] <\sh> anyone know what octave is?
[02:56] <Lathiat> mvo: 
[02:56] <Lathiat> Err http://au.archive.ubuntu.com breezy-updates Release.gpg
[02:56] <Lathiat>   Bad header line [IP: 82.211.81.151 80] 
[02:56] <tseng> \sh: isnt it some obscure programming language?
[02:56] <Lathiat> mvo: i got that often
[02:56] <\sh> tseng: i don't really know...but I need to compile it :( 
[02:56] <azeem> \sh: apt-cache knows
[02:56] <tseng> octave - GNU Octave language for numerical computations (2.1 branch)
[02:57] <tseng> its guh-knew
[02:57] <tseng> it must be good
[02:57] <\sh> when hdf5 is compiled ;)
[02:58] <mvo> Lathiat: try to run apt-get update -o Debug::Acquire::http=True 2> http_debug
[02:59] <Lathiat> ok,i'll let you know when i make it happennext
[02:59] <elmo> fabbione: 2.6.12 just floated by - didn't seem NEW, FWIW
[02:59] <Nafallo> anyone has troubles with X today? :-)
[02:59] <\sh> I wasn't restarting X since the update this morning ;)
[02:59] <fabbione> elmo: some udebs are. iirc sparc64, ia64.. probably hppa
[02:59] <fabbione> elmo: nothing new on the main arches side
[03:00] <fabbione> jdub: and new idiotify :p
[03:00] <Nafallo> hmm, actually I didn't update it before gdm wouldn't let me in to my default session :-/
[03:01] <Nafallo> anyone has troubles with gdm today? ;-)
[03:01] <Lathiat> Nafallo: yeh 
[03:01] <Lathiat> Nafallo: just selection GNOME as your session
[03:01] <Lathiat> Nafallo: and it works
[03:01] <Nafallo> Lathiat: oki. I'll try that and get right back :-).
[03:01] <jdub> fabbione: on by default? :)
[03:02] <tseng> jdub: it has been for awhile
[03:02] <tseng> in the .12 pres
[03:02] <jdub> yay
[03:02] <fabbione> jdub: tsk...
[03:02] <Nafallo> Lathiat: yay! :-)
[03:03] <pitti> sjoerd: here?
[03:03] <sjoerd> pitti: sortof
[03:03] <Nafallo> now to X... can't activate the XKB-config :-P
[03:04] <jdub> mdz, pitti: can you check nss-mdns and zeroconf for sanity? :-)
[03:04] <pitti> sjoerd: just a quick question, what's the status of the new dbus in Debian? (which entails the transition)
[03:04] <pitti> jdub: I hate you :-)
[03:04] <pitti> jdub: no, seriously, you mean the usual main inclusion review?
[03:04] <jdub> pitti: yeah
[03:05] <sjoerd> pitti: the upload was rejected becasue of the planned C++ abi transition
[03:05] <pitti> sjoerd: even to experimental? hrmpf
[03:05] <jdub> pitti: unless you hate the idea out of principle or something ;)
[03:05] <pitti> jdub: I didn't dig into that enough to hate it yet :)
[03:07] <\sh> http://blog.iansview.com/archives/64-Keep-your-computers-time-in-sync-using-HTP.html 
[03:07] <\sh> this could be quite interessting alternative to ntpdate
[03:08] <ogra> jdub, freescreensaver ?
[03:09] <sjoerd> pitti: yeah, unfortunately my exams are taking a lot more time then planned, so hal will take a while :(..
[03:09] <infinity> \sh : Cute idea.
[03:09] <jdub> ogra: F WORD!
[03:10] <pitti> sjoerd: well, without dbus we can't do any other updates anyway
[03:10] <\sh> infinity: yes I think so :)
[03:10] <ogra> jdub, feel like working on a spec ? ;)
[03:10] <jdub> hmm
[03:10] <sjoerd> pitti: yep, i'm going to ask for it to be allowed into experimental.. i don't really see why experimental stuff should slow down the transistion (as long as it stays in exp. untill after the transition)
[03:10] <jdub> everyone wants to fork
[03:11] <jdub> no one wants to be the dude responsible for forking it though
[03:11] <\sh> infinity: do u want to package it? or should I do it tomorrow
[03:11] <infinity> \sh : Go nuts.  I dan't have the time, or the interest. :)
[03:11] <\sh> hehehe
[03:11] <ogra> jdub, if you provide a spec that matches all needs your fork will just smart out the original
[03:11] <infinity> \sh : Check to see if there's a Debian ITP first (and if not, file one... I'll sponsor your uploads to Debian)
[03:12] <\sh> infinity: i will...but first tomorrow :) I want to go home just now..and then I will have a couple of pints with some friends :)
[03:13] <\sh> infinity: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=313595
[03:13] <Duck_work> seb128: latest if i remember well, i remember removing a dirty hack in cc or other gnome pkg because of such bug
[03:14] <seb128> k
[03:16] <infinity> \sh : Well, there you go.  If you're impatient, contact the submitter and offer to work with him.  Otherwise, just wait. :)
[03:17] <\sh> infinity: any chance to put it as an alternative to ntpdate for main?
[03:17] <infinity> \sh : It could certainly be discussed.  I'm not the man to make that decision, though.
[03:19] <pitti> \sh: If you want to push this, please write a spec about it and do some research
[03:20] <pitti> \sh: then this should be discussed in a TB meeting
[03:20] <\sh> pitti: i think it could be a good alternative for laptop users in a corporate office enviroment
[03:20] <Lathiat> jdub: i'd so do it just to piss him off for that flaming email? :)
[03:21] <Lathiat> not that i'd be any good at maintaing it :)
[03:21] <\sh> pitti: i will do some research and will write something on the wiki about it..
[03:21] <infinity> \sh : If the code is sane, I do like the idea.  But it needs a bit of an audit, and a lot of testing, I'd suspect.
[03:21] <infinity> \sh : ntp may be annoying for some users (due to firewalls and such), but it DOES work, and have a massive install base, etc.
[03:22] <Lathiat> its also more accurate
[03:22] <Lathiat> if you really care
[03:22] <infinity> Lathiat : ntpdate wouldn't be much more accurate than htpdate.
[03:22] <\sh> infinity: I don't want to push ntp away ;) htp is also only running with a near webserver as I read just now..so if i put a webserver in the config file from the states, I'm in a new timezone ;)
[03:22] <infinity> Lathiat : I would never push this as a replacement for ntpd.
[03:22] <Lathiat> infinity: why not? ntp is sub-second 
[03:23] <infinity> Lathiat : But for the one-off boot-time config, it's sane.
[03:23] <Lathiat> infinity: it also accounts for latency better than a  http server good
[03:23] <Lathiat> s/good/could
[03:23] <Lathiat> but,sure, its accurate enough formost users
[03:23] <Lathiat> esp. if it'l work more often and not hold up the boot somuch :)
[03:24] <infinity> \sh : http header dates have timezones in them.  Where the server is doesn't matter, except for latency.
[03:24] <infinity> \sh : And a one second round-trip still beats a clock that's off by minutes/hours on boot.
[03:25] <infinity> (If you want real time syncing, run ntpd, as always.. The boot-time sync is just for a "close enough, fix the clock if it's broken" setup.
[03:25] <infinity> )
[03:25] <\sh> infinity: well, at home it doesn't work anyways..without a direct network connection even ntpdate is stucked ;)
[03:26] <Nafallo> infinity: xkb broken with -27?
[03:26] <Nafallo> infinity: atleast here... ;-)
[03:26] <infinity> Nafallo : If it was broken before, and not fixed, then "yes".
[03:27] <infinity> Nafallo : In other words, "I'm not sure, I was just making sure the packages built"...
[03:27] <Nafallo> infinity: was not broken with -25
[03:27] <infinity> Nafallo : Tomorrow, I may look at other xorg bugs, but it's also my weekend, so I may not. :)
[03:27] <infinity> (Yay breezy)
[03:27] <tseng> Nafallo: a bunch of stuff is broken
[03:27] <Nafallo> infinity: I know. new xorg should go in on wednesdays or something ;-)
[03:28] <Nafallo> tseng: indeed. I have US keyboard for instance :-P
[03:29] <tseng> so speak english :)
[03:29] <Nafallo> tseng: hehe
[03:29] <tseng> no problem
[03:29] <infinity> Speaking English does seem to solve many software problems, yes.
[03:29] <Nafallo> dooh :-P
[03:30] <Nafallo> if only thom could be awake and fix 11904 everything would be less of a pain ;-)
[03:31] <HiddenWolf> nafallo, use him as an excuse to sleep till he fixes it. :)
[03:32] <tseng> such a whiner
[03:32] <Nafallo> HiddenWolf: hehe
[03:32] <Nafallo> tseng: baah, bugs :-P.
[03:32] <tseng> baah, whiners :-P
[03:33] <tseng> did you debug/fix/report those crashers yet?
[03:34] <Nafallo> tseng: there is one crasher to report :-).
[03:34] <tseng> 2
[03:34] <tseng> 3, you can report the ctrl+f thing for me :)
[03:35] <Nafallo> tseng: I'm waiting for the new muine with builtin trayicon :-). if it exists I'll report it.
[03:35] <Duck_work> jbailey: DEB_CONFIGURE_SCRIPT_ENV is broken
[03:36] <tseng> well its stuck somewhere
[03:36] <jbailey> Duck_work: Is the fix obvious?
[03:37] <Nafallo> tseng: hmm, I better try the new libxklavier first ;-)
[03:37] <tseng> buh
[03:37] <Duck_work> jbailey: yes, missing quotes
[03:37] <Duck_work> jbailey: is used : DEB_CONFIGURE_SCRIPT_ENV := LDFLAGS=" -Wl,-z,defs -Wl,-O1"
[03:38] <Duck_work> and quotes for LDFLAGS were removed
[03:38] <Duck_work> seems there is another problem
[03:38] <Duck_work> during build this time
[03:38] <azeem> it overrides CFLAGS
[03:38] <azeem> if you do := instead of +=, that is
[03:39] <Duck_work> azeem: ok
[03:39] <jbailey> Ugh, this is not what my brain is on at the moment...
[03:39] <jbailey> Duck_work: Would you mind collecting up a diff of these things and sending it to me when you're done hacking for the day?
[03:39] <Duck_work> jbailey: i've got a config.h not found :-(
[03:39] <jbailey> I'll get them merged in right away.
[03:39] <Duck_work> i could
[03:40] <Duck_work> if i am able to solve things
[03:40] <jbailey> config.h ?
[03:40] <Duck_work> yep
[03:40] <jbailey> It's written in shell, it shouldn't be looking for C header files.
[03:40] <Duck_work> i took a working pkg
[03:40] <jbailey> Ah. nice! =)
[03:40] <jbailey> Live tests!
[03:40] <Duck_work> only changed things related to cdbs
[03:41] <jbailey> If you can reduce the problem to a test case, I would *really* appreciate it.
[03:41] <Duck_work> of course live tests
[03:41] <jbailey> Then I can promise you from that point forward that it will never happen again. =)
[03:41] <Duck_work> ok
[03:41] <jbailey> (We give you new bugs at each iteration, not OLD and STALE bugs!)
[03:41] <infinity> I sure do love fresh bugs!
[03:42] <Duck_work> jbailey: just a hint to help me do tests: what is the hook format like now ?
[03:42] <Duck_work> what is build/pkg:: equivalent ?
[03:43] <jbailey> Duck_work: dilinger implemented it, so I don't know off hand.  I remember how we had talked about it, but it almost certainly wouldn't have been implemented that way in the end.
[03:43] <jbailey> (You know how these things go)
[03:44] <infinity> Nonsense, dilinger is a mind-reader, and surely did it exactly as you imagined.
[03:44] <infinity> He and I have had many CVS collisions with identical commits for that very reason.
[03:44] <Duck_work> dilinger: HELP REQUIRED !!!
[03:44] <infinity> HE'S CREEPY.
[03:45] <dilinger> I SEE BUILD SYSTEMS
[03:46] <infinity> See?
[03:46] <infinity> CREEPY.
[03:46] <dilinger> Duck_work: here's an example
[03:46] <dilinger> http://mouth.voxel.net/~dilinger/rules
[03:46] <Duck_work> coin Sir dilinger 
[03:46] <Duck_work> thanjks
[03:46] <pitti> Kamion: would it be possible to remove the "ro," from the CD-ROM fstab entries created by the installer?
[03:46] <dilinger> np
[03:47] <pitti> Kamion: it prevents the r/w mount of UDF DVD-RAMs, and normal CD-ROMs default to ro anyway
[03:47] <Kamion> pitti: I guess so - file a bug on partman-target, please?
[03:47] <pitti> Kamion: alright
[03:47] <dilinger> Duck_work: note that debian/rules help should spit out all kinds of information, as well
[03:48] <Duck_work> dilinger: nice, i should | less to avoid flooding my term :-)
[03:49] <Duck_work> DEB_CONFIGURE_INVOKE is ... strange
[03:50] <Duck_work> DEB_CONFIGURE_INVOKE='cd . && /home/duck/gnusound/gnusound-0.6.2/./configure <-- i guess this is done to handle specific locations, multibuilds, ...
[03:52] <Duck_work> dilinger: all quote desapeared :-(
[03:52] <Duck_work> +s
[03:52] <Duck_work> example :
[03:52] <Duck_work>   DEB_MAKE_INSTALL_TARGET='install DESTDIR=/home/duck/gnusound/gnusound-0.6.2/debian/tmp/'
[03:52] <Duck_work> DESTDIR had ""
[03:55] <dilinger> variables are eval'd
[03:55] <dilinger> hm, i thought i had an example there
[03:55] <dilinger> anyways, you'll need to play around w/ escaping
[03:56] <dilinger> the eval stuff is suboptimal, but it was the only way i could make bash and posh both happy
[03:56] <jdub> pitti: actually, don't worry about nss-mdns yet
[03:56] <dilinger> they both appear to have different behavior
[03:57] <pitti> jdub: if you do worry about zeroconf, could you please add it to https://wiki.ubuntu.com/UbuntuMainInclusionQueue and prepare a short report for that?
[03:57] <Duck_work> dilinger: :-(
[03:57] <jdub> pitti: ok, but i'll leave them until they're ready together (zeroconf, avahi and nss-mdns)
[03:59] <seb128_> pitti: boooooog on your cdbs/languagepack magic
[03:59] <seb128_> (tricky case rather)
[04:00] <pitti> seb128_: details?
[04:00] <seb128_> pitti: libglade2 has no po/, ./configure && make creates an empty po/ dir, your magic try to "/usr/bin/intltool-update -p --verbose;" here and returns on a "intltool-update: POTFILES.in not found."
[04:00] <pitti> seb128_: ah, and this exits with an error?
[04:00] <seb128_> you should probably [ -e po/POTFILES.in ] 
[04:00] <pitti> seb128_: so the sanest thing would rather be to append || true
[04:01] <seb128_> intltool-update: POTFILES.in not found.
[04:01] <seb128_> make: *** [common-post-build-arch]  Erreur 2
[04:01] <seb128_> debuild: fatal error at line 765:
[04:01] <seb128_> dpkg-buildpackage failed!
[04:01] <seb128_> 
[04:01] <seb128_> that's an option too
[04:01] <pitti> alright, fixing now
[04:01] <Kamion> http://cdimage.ubuntu.com/daily/current/report.html <- I wonder what all that's about
[04:01] <seb128_> pitti: thanks
[04:02] <pitti> Kamion: seb128 and me too, today's dist-upgrade wants to remove many packages
[04:02] <pitti> we blamed daniels so far :-)
[04:02] <seb128_> pitti: have you read the query this morning?
[04:02] <pitti> seb128_: erm, not your answer? I had some network trouble again :-(
[04:02] <seb128_> Investigating libx11-6
[04:02] <seb128_> Package libx11-6 has broken dep on xlibs-data
[04:02] <seb128_>   Considering xlibs-data 467 as a solution to libx11-6 1474
[04:02] <seb128_>   Added xlibs-data to the remove list
[04:02] <seb128_>   Fixing libx11-6 via remove of xlibs-data
[04:02] <seb128_> 
[04:03] <seb128_> I said that
[04:03] <pitti> uh
[04:03] <seb128_> that's apt with Debug::pkgProblemResolver=yes
[04:04] <jordi> elmo: > <seb128> I'm pretty sure that's not a gtk bug
[04:04] <ogra> Kamion, i guess X causes that...
[04:12] <Nafallo> seb128_: libxkbfile-dev as build-dep for libxklavier built locally.
[04:13] <seb128_> Nafallo: say it again?
[04:14] <seb128_> is there any context to this?
[04:14] <Nafallo> seb128_: FTBFS on all arches.
[04:14] <Nafallo> seb128_: the buildlogs are fairly straightforward :-)
[04:14] <pitti> Hi lamont
[04:15] <seb128_> Nafallo: you are just trying that you sit on front of the build log to bother people on IRC when something FTBFS? :)
[04:15] <lamont> morning pitti
[04:15] <seb128> Nafallo: I'm doing a bunch of syncs, I'll fix the FTBFSes later, but thanks
[04:16] <pitti> lamont: do you think you can fix the translation stripping tarball download soon? It's pretty much effort to correct it 
[04:16] <pitti> (i. e. correct from my side)
[04:17] <Nafallo> seb128: k :-).
[04:17] <lamont> and winds up with lots of files owned by pitti in my html tree...
[04:18] <pitti> lamont: yeah, sorry, I recovered the good tarballs from the various buildds
[04:18] <pitti> lamont: thom will certainly help you to chown them to you :-)
[04:18] <lamont> how soon do you think you'll have packagestriptranslations fixed to make it yet another file that is delivered to "not the archive" through the normal dupload process? :-)
[04:19] <pitti> lamont: well, that would be dpkg-source -b'ish rather than in pkgstriptranslations, right?
[04:19] <lamont> pitti: checking that the a new tarball is larger than the existing one strikes me as a hack, fwiw
[04:19] <pitti> lamont: so far we didn't discuss that with anybody yet
[04:20] <lamont> that'd be a dpkg-genpackage-ish thing, by adding it to a file list/
[04:20] <pitti> lamont: it is a hack, until the gettext issue on 64 bit platforms is sorted out; but that might take a while
[04:20] <lamont> and then teaching the archive process to magically handle it by dropping it off to the side somewhere, rather than in the mirrored archive
[04:21] <lamont> it's less of a hack to add $ARCH to the filename...
[04:21] <Duck_work> dilinger: makefile.sh:    ## FIXME: Define CFLAGS, CXXFLAGS correctly
[04:21] <Duck_work> taht's why it is not working
[04:21] <pitti> lamont: I could live with $ARCH as well, I don't know about the rosetta guys
[04:21] <pitti> lamont: either is fine for me, whatever can be made quicker
[04:23] <lamont> pitti: well, since it's magic knowledge in sbuild right now (sounds of vomitting) instead of being just another file, any change is going to require me to do something...  let me ponder it a bit and poke the rosetta guys about their pain levels
[04:23] <pitti> thanks
[04:25] <Kamion> er. where'd xterm go?
[04:25] <Kamion> xorg no longer builds it, but nothing else does either
[04:25] <Kamion> daniels: ?
[04:26] <lamont> Kamion: good catch.  if he doesn't get it back soon, we'll send guidofinity over to his place to discuss the situation...
[04:26] <Kamion> and britney lists it as uninstallable, although it installs fine for me
[04:27] <lamont> wonder if it still would install after a dist-upgrade, and removing some orphaned packages
[04:28] <Kamion> well, I did dist-upgrade; let's try dselect ...
[04:33] <Kamion> I had a couple of things from universe installed, but nothing relevant. No obsolete packages.
[04:35] <Kamion> ah, libxaw8 disappeared
[04:35] <Kamion> I did wonder about that when reading the changelog ...
[04:44] <seb128> elmo: zenity gthumb galeon eog syncs please
[04:48] <lamont> elmo/thom: pls chown -R lamont ~lamont/public_html/translations
[04:48] <jbailey> ANyone here familiar with how the only-ifup-when-I-load-a-module magic works?
[04:51] <elmo> seb128: done
[04:52] <elmo> lamont: done
[04:52] <lamont> elmo: thanks muchly
[04:53] <seb128> elmo: thank you
[05:28] <spotter> anyone using the newest X packages?
[05:28] <\sh> yes
[05:28] <spotter> no problems?
[05:29] <\sh> and xkb is broken..so don't complain ;)
[05:29] <spotter> more than xkb
[05:29] <pitti> again???
[05:29] <spotter> gnome wont event start
[05:29] <\sh> spotter: apt-get install ubuntu-desktop
[05:29] <spotter> lots of errors in .xsession-errors
[05:29] <spotter> ok, lets see what that installs
[05:30] <lamont> hrm... is there a gethostbyname in python, I wonder.
[05:30] <\sh> ok...i need a hardware sponsor...
[05:30] <\sh> 1. powerpc ;)
[05:30] <\sh> 2. ia64
[05:31] <\sh> 3 archs ok...but ppc libtool: link: `../lib/libh5tools.la' is not a valid libtool archive
[05:32] <lamont> pitti: re pkgstriptranslations: I'm just going to disable both 64-bit architectures..  that's the simplest hack-on-hack, and the rosetta guys are OK with the possible need to handle the <10 packages that could be dropped that way
[05:33] <pitti> lamont: great, that's fine for me
[05:34] <spotter> k, problem1 startx wants to run /usr/bin/X11/X
[05:34] <spotter> it doesn't exist
[05:34] <spotter> Xorg does though
[05:34] <spotter> symlink fixes that
[05:34] <spotter> then however, startx wants to create /var/log/Xorg.0.log, but it cant (as regular user, as root, no complaints)
[05:35] <spotter> as an aside, does vt switching in X work for anyone anymore?  (ctrl-alt-fkey, stopped working a few weeks back for me)
[05:36] <spotter> ok, server isn't suid root
[05:37] <\sh> spotter: can u file a bug with all those things?
[05:38] <spotter> chmod u+s /usr/X11R6/bin/Xorg fixed that and X starts (from startx)
[05:38] <spotter> yea
[05:38] <spotter> can you email it to me (spotter@cs.columbia.edu) in bitchx and can't copy and paste to X
[05:39] <spotter> and of course your comment of xkb broken again seems to be accurate
[05:41] <\sh> spotter: send
[05:42] <Keybuk> so, uh, am I not supposed to be able to login from gdm at the moment
[05:42] <Keybuk> ?
[05:42] <seb128> elmo: can you sync gnome-alsamixer again (it has been dropped on MOTU request as discussed but Daniel said he has no objection to get it again)
[05:43] <seb128> Keybuk: what version?
[05:43] <Keybuk> seb128: whatever was in the archive NOW-$how_long_the_update_took
[05:43] <seb128> I tend to blame xorg so
[05:43] <Keybuk> I get an ugly "CHOOSE YOUR SESSION WIDGET" thing
[05:43] <Keybuk> which does nothing
[05:44] <seb128> new gdm has an issue (using /usr/ksh instead of /usr/sh) this morning but I've fixed that for hours
[05:44] <Keybuk> did you fix it in ubuntu3?
[05:44] <seb128> yep
[05:44] <Keybuk> right
[05:44] <Keybuk> cancel the anti-aircraft weaponry
[05:44] <Keybuk> :p
[05:44] <Keybuk> I must have _just_ missed that when I started the upgrade
[05:44] <HrdwrBoB> Keybuk: no-one will notice
[05:45] <HrdwrBoB> except maybe the french 
[05:45] <Keybuk> (I was _really_ far behind with breezy -- laptop's been gathering dust since UDU)
[05:45] <seb128> roh
[05:45] <Keybuk> I notice all the gdm binaries moved
[05:45] <seb128> Keybuk: I would have transitioned gdm some time ago already, but I was waiting for hct *g*
[05:46] <seb128> Keybuk: yeah, /usr/bin to /usr/sbin or /usr/lib
[05:46] <Keybuk> seb128: there's a #hct channel you can hassle me on now :p
[05:46] <seb128> nice :)
[05:46] <Keybuk> I can do "right now" imports I think
[05:48] <Keybuk> fabbione: you're not supposed to /join and /part :p
[05:48] <fabbione> Keybuk: i was only hasseling :P
[05:49] <Keybuk> you should put it in your auto-join
[05:49] <Keybuk> and hang out with the cool kids
[05:49] <Keybuk> they're all using HCT these days
[05:49] <Keybuk> what colour should they be?
[05:49] <fabbione> pink
[05:49] <spotter> Keybuk: lots of problems
[05:49] <spotter> about to file a bug
[05:49] <Keybuk> spotter: with?
[05:50] <spotter>  /usr/bin/X11/X is missing (Xorg is there) 
[05:50] <spotter> startx depends on that
[05:50] <Keybuk> oh, I fixed _those_ already :p
[05:50] <spotter>  /usr/X11R6/bin/Xorg isn't chmod u+s
[05:50] <Keybuk> it's just session and display management that I fall over and get confused
[05:50] <\sh> hmm....it's in the topic of #u-m
[05:51] <Keybuk> we should make a policy
[05:51] <Keybuk> daniels does not upload X before international travel
[05:51] <lamont> Keybuk: would those wristbands come with razor blades? :-)
[05:51] <\sh> Keybuk: better idea...open the archives only to people who know that everything is bleeding nose buggy ,-)
[05:52] <Keybuk> it's not the bleeding nose I worry about
[05:52] <Keybuk> it's the bleeding testicles
[05:52] <infinity> Mmm, testicles.
[05:52] <\sh> right, i can't enter my backslash ;)
[05:53] <\sh> my personality ... just disappeard :(
[05:53] <mdz> daniels: xserver-xorg 6.8.2-27 fails to configure itself for my laptop hardware anymore (chooses vesa) and adds a "Generic Mouse" to ServerLayout without a corresponding InputDevice section
[05:54] <jdub> Keybuk: i can't type in X.
[05:54] <jdub> Keybuk: but otherwise, everything else is quite alright. :-)
[05:55] <\sh> no umlauts and no backslash ;)
[05:55] <jdub> ogra: oh well, no one here listens to heavy metal anyway.
[05:55] <Kamion> mdz: since xterm is currently hosed (libxaw8 got removed under it), what do you think about temporarily removing it from desktop?
[05:55] <ogra> hhhehehe
[05:55] <mdke> jdub doesn't listen to heavy metal?
[05:55] <mdke> SHOCK
[05:56] <mdz> Kamion: xterm is in desktop?
[05:56] <Kamion> mdz: yes
[05:56] <mdz> Kamion: it oughtn't be
[05:56] <Kamion> really?
[05:56] <Kamion> er, ok
[05:56] <mdz> if I can get over my xterm addiction, I think the rest of the world can too ;-)
[05:56] <jdub> i don't think it's a terrible thing to have in desktop
[05:56] <jdub> and removing it doesn't save too much space :)
[05:56] <fabbione> no
[05:56] <fabbione> don't remove xterm please
[05:57] <Kamion> just seems kind of an odd thing to ship without
[05:57] <jdub> the powerkids love us for it
[05:57] <fabbione> even Michael Jackson uses xterm...
[05:57] <fabbione> we can't remove it :P
[05:58] <dilinger> heh.  totem wins for most useless error message ever.
[05:58] <dilinger> ** (totem:663): CRITICAL **: bacon_message_connection_get_is_server: assertion `conn != NULL' failed
[05:58] <\sh> michael jackson uses xterm?
[05:58] <fabbione> elmo: no luck yet to move 2.6.12 to main?
[05:59] <seb128> dilinger: what it useless about this?
[05:59] <seb128> dilinger: you have the line, function and condition b0rked
[06:00] <dilinger> seb128: it doesn't actually mean anything to a user :p
[06:00] <seb128> dilinger: that's why it's not an UI dialog
[06:00] <fabbione> dilinger: it's all a conspiracy to make people switch to kde
[06:00] <dilinger> seb128: is it a network thing?  is it misconfigured?  is it a sound thing?  is it having problems w/ gconf or something?  dbus, maybe?  what's bacon?
[06:01] <seb128> dilinger: that's a line to put to bugzilla so upstream knows what is wrong and can fix it
[06:02] <mdz> doko: it seems that twisted added a dep on zopeinterface, which wants python2.2
[06:02] <doko> mdz: ok, dropping it
[06:02] <\sh> 7join #ubuntu-ppc-specialist
[06:02] <mdz> doko: thanks
[06:07] <mdz> fabbione: I moved it already
[06:08] <fabbione> mdz: the change didn't propagate to the archive yet.. did it?
[06:08] <mdz> fabbione: probably not
[06:08] <fabbione> ok
[06:08] <mdz> fabbione: I moved the whole thing, but most of it wants to move back to universe unless it's seeded
[06:09] <fabbione> mdz: ah ok
[06:09] <fabbione> mdz: i need another upload to complete the cleanup for the udebs and we can switch the installer to .12
[06:09] <fabbione> mdz: at that point it should be ok without seeding
[06:10] <mdz> fabbione: well, switching the installer is a matter of changing the seeds ;-)
[06:10] <mdz> fabbione: do you know what causes this xorg.conf mouse problem?  it has happened before
[06:11] <fabbione> mdz: no. it's a while i don't upgrade my machine
[06:11] <Kamion> mdz: ... no it's not :)
[06:11] <fabbione> Kamion, mdz: well up to you 2 to agree :)
[06:11] <fabbione> i need to go and cook dinner guys
[06:11] <fabbione> start to enjoy the .12 ride!
[06:12] <fabbione> because this kernel seems to be rock solid
[06:12] <mdz> Kamion: oh, that variable is set in germinate or someplace instead (now?)?
[06:12] <\sh> fabbione: have fun.../me is getting ready for party...
[06:12] <mdz> installer: * Kernel-Version: 2.6.10-5-amd64-generic
[06:12] <mdz> installer: * Kernel-Version: 2.6.10-5-386
[06:12] <mdz> installer: * Kernel-Version: 2.6.10-5-itanium-smp
[06:12] <mdz> installer: * Kernel-Version: 2.6.10-5-power4 2.6.10-5-powerpc
[06:12] <mdz> installer: * Kernel-Version: 2.6.10-5-sparc64
[06:13] <mdz> are those no longer meaningful?
[06:13] <Kamion> mdz: they're perfectly meaningful, they're just not sufficient
[06:13] <Kamion> and never have been
[06:13] <Kamion> d-i needs to be uploaded when switching kernel versions, to rebuild the initrds and cope with any udeb rearrangements
[06:13] <fabbione> oh Kamion please don't merge kernel-wedge yet.. i need another upload to complete the cleanup and fix sparc FTBFS
[06:14] <fabbione> we will switch immediatly after
[06:14] <fabbione> (question of one day MAX 2 )
[06:14] <Kamion> those seed variables are there so that we don't have to repeat "2.6.10-5" a zillion times in the seeds
[06:14] <fabbione> i am off for real
[06:14] <fabbione> bye bye
[06:14] <Kamion> fabbione: wasn't planning to - in fact I'd rather you did it so you can coordinate any necessary kernel changes
[06:14] <mdz> Kamion: I don't think we disagree
[06:14] <fabbione> Kamion: SUPER!
[06:15] <mdz> perhaps s/is a matter of/involves/
[06:15] <Kamion> mdz: oh, I think I may have misread you
[06:15] <Kamion> yes, that
[06:15] <mdz> I didn't mean to imply that it was exclusively a seed change
[06:15] <Kamion> ah, ok, I take back the disagreement :-)
[06:16] <mdz> Kamion: how is the house coming along?
[06:17] <Kamion> mdz: still rather living out of boxes, but otherwise more or less sorted
[06:17] <mdz> Kamion: is it close to where you lived before?
[06:18] <mdz> I don't see anything in the changelog about the .config script changing at all, how odd
[06:18] <Kamion> mdz: yeah, about a mile or so away, same side of town
[06:18] <Kamion> separate study and bedroom now, which is an unbelievably excellent improvement :)
[06:19] <mdz> heh
[06:19] <Kamion> http://maps.google.co.uk/maps?q=atherton+close+cambridge+to+pelham+court+cambridge&spn=0.014039,0.042443&hl=en <- moved from bottom right to top left
[06:20] <Kamion> with my favourite pub almost equidistant between the two ;-)
[06:22] <Lathiat> mvo: "apt-get update -o Debug::Acquire:http=True"
[06:22] <Lathiat> mvo: is that right?i got not extra debug info
[06:22] <azeem> s/:/::/?
[06:23] <Lathiat> heh
[06:23] <Lathiat> toobad i have to wait half an hour to reproduce it again
[06:24] <Lathiat> andnow md5sum mismatches on main+universe sources -- what causes that? in betweenupdates? (i see it every so often)
[06:28] <mdz> daniels: there seem to be several changes from -24 to -27 which are not documented in the changelog; are they intentional?
[06:28] <mdz> (at least one of them is the cause of my mouse problem)
[06:29] <mdz> I have no idea what is causing the driver detection bit to fail yet, though
[06:30] <infinity> Driver detection, as in video driver?
[06:30] <infinity> I see no changes to that between -24 and -27...
[06:33] <mdz> infinity: neither do I, but I get vesa now where before I (correctly) got ati
[06:35] <mdz> what does dexconf mean by "multiplexed"?  the logic doesn't seem to refer to the definition of that word with which I am familiar
[06:35] <mdz> it feels that /dev/psaux is a "multiplexed" mouse device
[06:35] <mdz> as well as serial devices
[06:36] <infinity> Yeah, I have no idea what the deal is with the mouse stuff.  I'm trying to hunt the keyboard issues right now.
[06:36] <mdz> oh, there are keyboard issues too?
[06:37] <infinity> There are if you're not from the US. :)
[06:37] <infinity> (So, I didn't notice either, until people whined)
[06:37] <mdz> I'm from the US, but I don't use a us layout
[06:37] <lamont> seb128: got a few minutes for a puzzler?
[06:37] <infinity> xkb is completely stillborn.
[06:37] <infinity> mdz : Really?... Interesting.  Many others have been complaining that xkb won't load or won't load their keymap, etc.
[06:38] <mdz> infinity: so far I've only upgraded my ltsp test laptop
[06:38] <mdz> where X keymap configuration has never worked
[06:38] <Lathiat> xkb errors on gnome start here
[06:38] <mdz> because it was waiting on the changes that went into -25
[06:38] <siretart> did anyone try to use rpcgen with gcc-4.0? I have a package here were sourcefiles created by rpcgen FTBFS with gcc-4.0 :(
[06:39] <infinity> Right, well going from "doesn't work" to "doesn't work" wouldn't hurt much. :)
[06:40] <mdz> yeah, the changes didn't start it working either
[06:40] <ogra> argh, my at key is screwed....
[06:40] <Lathiat> ogra: which key? ;)
[06:40] <ogra> i cant type mail adresses 
[06:40] <danielki> copy this one: @
[06:40] <danielki> (:
[06:40] <ogra> heh
[06:41] <infinity> (Breaking things horribly is a good way to see who is and isn't running breezy, though...)
[06:41] <Lathiat> ogra: change keyboard maps :)
[06:41] <lamont> seb128: nm
[06:42] <ogra> hmmm, gnome-keyborad-settings crashes....
[06:43] <ogra> ** (gnome-keyboard-properties:13906): CRITICAL **: XkbGetKeyboard failed to get keyboard from the server!
[06:43] <ogra> fun
[06:44] <mdz> I can't even generate a config file: xserver-xorg/config/inputdevice/keyboard/layout not set.  Aborting.
[06:45] <infinity> Hrm.  Was all the xkb stuff working happily in -24?
[06:45] <Lathiat> yeh
[06:45] <mdz> "working" as in, it at least generated a config with the default value
[06:45] <infinity> The biggest change in that are is actually him reverting a patch (backport of a HEAD cleanup of xkb) which was added in -24, removed in -25...
[06:45] <infinity> s/that are/that area/
[06:46] <infinity> mdz : Well, it's hard to generate a config when every xkb-related tool you use to lint your keymap now chokes and claims the keymaps are invalid.
[06:46] <mdz> so the vesa thing was a red herring
[06:47] <infinity> Oh?
[06:47] <mdz> that was just "I didn't generate a config at all"; that's the one which was there from installing in the chroot
[06:47] <infinity> Ah. :)
[06:47] <infinity> So, we're at "mouse config got weird", "xkb is buggered", and "twm's postinst sucks, but only one user cares, cause no one else installs it"
[06:47] <infinity> Right?
[06:47] <infinity> (Of course, the 3rd bug is the simplest to fix... Always is)
[06:48] <infinity> I wonder if reintroducing this XKB HEAD cleanup patch would magically fix everyone's issues.
[06:48] <infinity> Maybe we should wait for daniels to find an airport in Germany and ask him why he reverted it...
[06:49] <infinity> Or... Not care and just put it back.  <experiments>
[06:50] <mdz> there is no default for xserver-xorg/config/inputdevice/keyboard/layout
[06:50] <mdz> and yet it blows up if it isn't set
[06:50] <infinity> Or, the third option: go to bed, cause it's 3am.  Why don't I notice these things until my girlfriend yells at me? :)
[06:51] <mdz> the mouse config thing should be easy to fix; we can revert that patch
[06:51] <mdz> no clue what it was intended to do, since it wasn't in the changelog
[06:51] <infinity> Yeah.  Well, I'll be happy to play with both in the morning, if daniels doesn't show up and fix them first.
[06:51] <infinity> But right now is apparently a "Bad Time", since I've been working all day and have only now realised this.
[06:54] <infinity> I wonder if perhaps the stuff that was reverted from -24 to -25 was just a "switching from one working directory to another" brain fart.
[06:55] <infinity> It's not typical for people to go back and edit their old changelog entries when reverting changes, afterall. :)
[06:55] <infinity> Anyhow.  Bed.  Xorg in the morning, if others don't beat me to it.
[06:55] <infinity> (feel free to beat me to it)
[06:56] <pitti> sleep well, infinity!
[07:15] <mdke> sabdfl, you around?
[07:15] <sabdfl> yes
[07:16] <mdke> https://wiki.ubuntu.com/MarksHoaryGoals <--need this?
[07:17] <mdke> oh
[07:17] <mdke> i just saw its linked in your homepage so there is no danger of it disappearing
[07:18] <mdke> sorry to bother
[07:23] <sabdfl> jdub, daniels, seb128: can we use OSD for the battery-low indicators, rather than the modeless dialog that currently pops up?
[07:25] <tseng> sabdfl: xosd afaik still doesnt use freetype
[07:26] <tseng> it looks pretty ugly
[07:26] <Lathiat> theres a gnome osd
[07:26] <Lathiat> (that does afaik)
[07:27] <Lathiat> sarge-hoary works?
[07:33] <dilinger> [14-Jun-2005]   We just released the first release candidate for PHP 4.4.0. This is a bug-fix only release, the increased middle digit is needed because this release changes PHP's Internal API that causes existing third-party binary extensions to be incompatible with the new version.
[07:34] <dilinger> groundbreaking
[07:34] <Clint> mmm... internal api bugfix
[07:39] <Keybuk> anyone know stuff about postgresql?
[07:39] <Keybuk> pitti?
[07:40] <Keybuk> meh, figured it
[07:40] <Keybuk> someone moved the socket
[07:45] <thom> Kamion: are current breezy dailies somewhere close to installable?
[07:50] <Kamion> thom: no, xterm's broken
[07:50] <Kamion> mdz: I'll remove that from desktop for the moment just so that we have working dailies
[07:50] <Kamion> thom: first stage works fine though
[07:50] <mdz> Kamion: fine with me
[08:04] <Treenaks> (it could be an X thing though)
[08:04] <Lathiat> Treenaks: choose GNOME fromthe session list and it goes away
[08:04] <Lathiat> gdm seems tobe losing the defaultsession
[08:04] <Treenaks> Lathiat: yes, but that's not the Best solution
[08:04] <Lathiat> and starting some weirdarse thing
[08:05] <Treenaks> also, for the record, network-manager sucks (i.e. it can't connect to my simple WEP-encrypted wifi net)
[08:05] <Treenaks> Lathiat: oh, try ctrl+alt+f1
[08:05] <Lathiat> its suposed to ask for the key if you click on a wep net
[08:05] <Treenaks> Lathiat: it does ask.. then I select "Hex key" and enter my key in hex
[08:05] <Treenaks> Lathiat: then it waits for a while, iwconfig shows everything works
[08:06] <Treenaks> Lathiat: but nm-applet thinks I'm not connected, and its right
[08:06] <Lathiat> nm has never ever worked for me once
[08:06] <Lathiat> not with my orinoco wireless,and not with my new b44+ipw2200 laptop
[08:06] <Treenaks> Lathiat: I tried with prism54 and ipw2200
[08:14] <Lathiat> anyone know why im getting "/dev/sda1 is not removable" errors (error mounting my external usb drive)
[08:14] <fabbione> maswan: ping?
[08:14] <thom> ber, ghc is gonna need rebootstrapping, or some nasty kludging
[08:15] <fabbione> thom: yup... ghc is a pain
[08:15] <fabbione> i did look at it not too long ago
[08:15] <fabbione> and the worst is that even in bootstrap is FTBFS due to gcc-4.0 or something else in the toolchain
[08:15] <thom> oh, joy
[08:20] <jdub> seb128: around?
[08:20] <jdub> never mind
[08:21] <jdub> well, maybe
[08:21] <maswan> fabbione: pong
[08:21] <Lathiat> im also getting a mass of "IPP request failedwith status 1030"from gnome-cups-icon
[08:21] <maswan> elmo_: ping?
[08:21] <fabbione> maswan: hey dude..
[08:29] <mdz> fabbione: can you explain auto_answer in xserver-xorg.config?
[08:29] <mdz> it makes no sense to me
[08:31] <fabbione> mdz: i can look at it.. but i am not sure it's the same code i knew a long time ago
[08:31] <mdz> it has been there a long time
[08:32] <fabbione> mdz: it has always been there.. but i dunno if daniels did change it
[08:32] <mdz> the name of the function does not seem to describe what it does
[08:32] <fabbione> let me check.. but i need to run after.. wife is waiting me to go to the concert
[08:33] <fabbione> mdz: ok.. now i remember
[08:34] <fabbione> do you recall a long time ago when we were killing xfree86 debconf questions?
[08:34] <fabbione> the problem was (and is for Debian) that we didn't have a proper way to answer all the questions
[08:34] <fabbione> now.. this is ok if you have a debconf frontend that allow the user to interact with the installation
[08:35] <fabbione> but that's not true in non-interactive or some other combination
[08:35] <fabbione> auto_answer tries to answer these questions, if a previous answer was already present in the debconf db
[08:35] <fabbione> sort of "use debconf" as a registry
[08:36] <fabbione> note that when auto_answer is called there are other functions involved to gather default answers and stuff like that
[08:36] <fabbione> does it make more sense now?
[08:44] <mdz> fabbione: it seems to me that the end result is just to set a new default for the question; wouldn't this be better done with a debconf substitution?
[08:46] <fabbione> mdz: i don't really know.. i didn't design that :) it's something coming from Xfree86 / Debian
[08:46] <fabbione> i need to go now...
[08:46] <fabbione> cya tomorrow or monday
[08:46] <mdz> fabbione: bye
[08:46] <mdz> and thanks
[08:47] <fabbione> mdz: no problem :) sorry that i don't have more time.. but we are alrady a bit late :)
[08:47] <fabbione> bye
[08:50] <mdz> mdke: how is the new wiki working out?
[08:54] <Kamion> you can't do debconf defaults with SUBST; you have to use SET
[08:54] <Kamion> which means you also have to test whether the current value of the question differs from the default (and possibly from previous defaults), etc.
[08:56] <maswan> hmm.. are dns-records for archive.ubuntu.com strictly elmo's department?
[08:57] <maswan> or can someone else take 130.239.18.137 out?
[08:57] <Kamion> elmo/thom probably
[08:57] <maswan> damn, need to check my summoning spells. :)
[08:58] <maswan> hi sabdfl, sorry, that was a refernce to me hitting enter on "+ maswan summons elmo/thom :)" 20 seconds before you joined
[09:00] <mxpxpod> ogra: ping
[09:00] <jdub> daniels: around?
[09:03] <torkel> maswan: if the slaves are asleep summon the big boss? :-)
[09:06] <sabdfl> maswan: YOU CALLED?
[09:07] <tseng> who dares awake the great sabdfl 
[09:08] <maswan> sabdfl: oops. I tried summoning the elmo, but must have aimed wrong.
[09:27] <sabdfl> ALL IS PEACEFUL
[09:27] <sabdfl> night ;-)
[09:32] <mxpxpod> ogra: are you around?
[09:33] <tseng> ogra: its getting way past working hours there
[09:33] <tseng> er, mxpxpod.
[09:33] <mxpxpod> tseng: hrmm
[09:34] <mxpxpod> tseng: I need to talk to him about using pmud instead of pbbuttonsd in the next release of ubuntu
[09:34] <mxpxpod> since the latest release of pbbuttonsd really really really sucks
[09:34] <tseng> isnt that thom's baby?
[09:34] <mxpxpod> IIRC, he told me to talk to ogra about it
[09:34] <tseng> huh
[09:34] <tseng> i didnt know ogra owned a ppc
[09:34] <mxpxpod> oh, maybe I do need to talk to thom about it
[09:35] <jdub> holy crap!
[09:35] <jdub> i can type in X again!
[09:35] <tseng> jdub!
[09:35] <tseng> pants off.
[09:35] <mxpxpod> anyway, I hacked powermanagement-interface to work with both pmud and pbbuttonsd
[09:35] <ogra> mxpxpod, hey
[09:35] <ogra> mxpxpod, yes, thom is the right one...
[09:35] <mxpxpod> ogra: ok
[09:35] <mxpxpod> ogra: how's gnome-power coming along in breezy?
[09:36] <ogra> mxpxpod, i'm currently busy with edubuntu, next week too.... then i've time to care about other stuff again....
[09:36] <mxpxpod> ogra: ah, ok
[09:36] <mxpxpod> tseng: you've got a ppc, right?
[09:36] <tseng> no sir
[09:37] <mxpxpod> nevermind
[09:37] <ogra> mxpxpod, gnome-power is missing the connection to pmi, but then it will be fine... i also have to decide what to do about screensaver power management, since it duplicates the functions now
[09:37] <mxpxpod> heh
[09:37] <mxpxpod> ogra: well, like I said to tseng, I hacked pmi to work with pmud
[09:38] <ogra> nice
[09:38] <mxpxpod> and I know thom has wanted to get rid of pbbuttonsd for a loooong time
[09:38] <mxpxpod> now I just need to figure out how to do a diff of a package while I'm building it
[09:38] <ogra> but i' not sure if thats necessary anymore, since hal is the input for gnome-power... and it cares for pmu devices as well
[09:39] <mxpxpod> ogra: ahhh
[09:39] <mxpxpod> that'd be really nice
[09:39] <Kamion> does pmud have all the hotkey-handling stuff?
[09:39] <ogra> but ask thom, he's the pbuttonsd guy
[09:39] <mxpxpod> Kamion: no, gnome would handle that
[09:39] <Kamion> eh, not everyone uses GNOME
[09:39] <mxpxpod> my bad
[09:41] <mxpxpod> Kamion: actually, I think hal can handle the power button
[09:42] <seb128> jdub: pong
[09:42] <seb128> lamont: pong?
[09:42] <mxpxpod> ogra: ok, another question regarding hal/gnome-power... will those two actually sleep the laptop, or will we need to invoke something when a signal is fired
[09:43] <lamont> seb128: no more issue
[09:43] <ogra> gome-power just sends a dbus message
[09:43] <ogra> gnome even
[09:43] <lamont> seb128: unless you have one for me....
[09:43] <ogra> so we need a listener at the other end... thats currently missing
[09:43] <mxpxpod> ogra: ah
[09:43] <mxpxpod> ogra: i.e. the connection to pmi
[09:43] <ogra> yep
[09:43] <mxpxpod> gotcha
[09:44] <mxpxpod> so we'll need a pmi daemon listening or something like that
[09:44] <seb128> lamont: nop, just a pong in case of you need me for something
[09:44] <ogra> for example, yes.... or a gnome-power-backendd to be replaced by a kde-powerd or a xfce-powerd ....
[09:44] <mxpxpod> ogra: ah, cool
[09:46] <lamont> seb128: thansk
[09:46] <seb128> np :)
[09:48] <mxpxpod> how close is breezy to be ready to use?
[09:49] <mxpxpod> and also, is network manager ready for use on ubuntu?
[09:49] <tseng> jdub can type in X
[09:50] <ogra> tseng, i heard about that too.... he must be privileged, i cant really type in X
[09:56] <mxpxpod> ok, this may be a dumb question, but what exactly does network manager do?
[09:57] <tseng> it screws up your network connection
[09:57] <mxpxpod> does it do what netapplet does?
[09:57] <tseng> and in the case of kerberos authetication
[09:57] <tseng> tries really hard to lock you out of your pc
[09:57] <mxpxpod> and will it automatically connect your ethernet when you plug it in?
[09:57] <jdub> network-manager is not ready to use
[09:57] <jdub> i just tried
[09:58] <mxpxpod> heh, ok
[10:00] <jdub> mdz: whoa, champion (xorg upload)
[10:00] <mdz> it doesn't fix the xkb stuff; I didn't have time to look into that one
[10:01] <jdub> fixed a lot of annoying crap though!
[10:01] <jdub> THREE CHEERS FOR MDZ!
[10:01] <mdz> it looks like daniels may have inadvertently dropped some stuff from about -21
[10:01] <tseng> we should take turns uploading xorg
[10:01] <tseng> and see where we end up
[10:01] <jdub> say hooray you bastards.
[10:02] <lamont> seb128: actually, any gtk+2.0 uploads planned soon?
[10:02] <seb128> 2.6.8 ftbfs?
[10:03] <seb128> nop, it doesn't
[10:04] <lamont> seb128: there was this little regression in gcc-4.0 for hppa...
[10:04] <seb128> binary-NMU? :)
[10:04] <seb128> jdub: you pinged for a reason?
[10:05] <lamont> that's an option for us poor ports second-class citizens, as long as there's a -ubuntu version
[10:05] <lamont> (or a -build version)
[10:05] <lamont> seb128: anything that does function pointer compares may get bad results from the compare, and behave incorrectly.
[10:05] <lamont> like, say, callbacks. :-(
[10:06] <lamont> that bug would go a long way to explaining why gdk-pixbuf-query-loaders sometimes works, sometimes hangs, no?
[10:06] <jdub> seb128: no, it was just a random drive-by ping
[10:06] <seb128> k
[10:06] <seb128> lamont: hum, gdk-pixbuf-query-loaders hangs sometimes? weird
[10:06] <lamont> only on hppa
[10:07] <seb128> iz hppa bog :)
[10:07] <lamont> and the gcc regression is that (on hppa) function pointers don't get canonized before getting compared, and therefore things that should compare equal say 'unequal'
[10:07] <seb128> anyway you want a source upload? there is some translation issue that could pretext a new upload :p
[10:07] <lamont> and yes, is optimizer/hppa b ug
[10:08] <seb128> k, I understand
[10:08] <lamont> dunno if it would fix it or not
[10:08] <seb128> I'll probably upload one new package on monday
[10:08] <lamont> I mean, it could be anything that it calls, too... :-)
[10:09] <lamont> and since it already has a -1ubuntu1 version, I can binNMU it, although it makes elmo cry
[10:09] <lamont> (problem with binNMU's is that 1.0.1 > 1ubuntu1 --> bad juju)
[10:10] <seb128> 1ubuntu1.0.1 ?
[10:12] <jdub> whoa!
[10:13] <jdub> latest kernel also makes my laptop audio not sound like crap
[10:13] <doko> fabbione: the new kernel on amd64 is fun ... no sata detection, no cpu frequency scaling, bad sound ...
[10:13] <tseng> jdub: eh?
[10:13] <jdub> i used to get a yucky buzzing noise
[10:13] <tseng> hm
[10:15] <ivoks> hm... X in breezy after upgrade don't listen to xorg.conf about keyboard layout
[10:15] <jdub> ivoks: an upload was just done to fix that
[10:15] <ivoks> oh, thanx
[10:15] <jdub> ivoks: build will be available soon
[10:16] <jdub> ivoks: temporary fix -> dpkg-reconfigure xserver-xorg
[10:16] <ivoks> great
[10:16] <jdub> don't set --priority=high :)
[10:16] <ivoks> jdub: ")
[10:16] <jdub> type in 'us' when it asks you for the keyboard layout
[10:16] <ivoks> i tried that, but did't work
[10:16] <ivoks> problem is that it assumes 101 keys, instad of 105
[10:16] <lamont> seb 1ubuntu1.0.1 requires that -1ubuntu1 or -1ubuntu1.0 source be in the archive.
[10:17] <lamont> hence if the package has a raw debian version, I have no choice but a sourceful upload
[10:18] <seb128> right, but gtk is 1ubuntu1
[10:18] <lamont> mdz/jdub; thoughts on the right answer when a 2nd class arch needs to rebuild a package or 2000 - do we hate binNMU's that much?
[10:18] <lamont> seb128: exactly
[10:18] <jdub> not a q for me :)
[10:18] <lamont> hence I just have to decide if I want to introduce a binNMU into the archive, or be a good little boy.
[10:20] <mdz> lamont: 2000?
[10:21] <lamont> mdz: well, one could be more careful and examine the source of everything that hppa has built using gcc-4.0 or g++4.0 to see if it compares function pointers...
[10:21] <lamont> for the moment, I'm ignoring it, and planning an as-discovered solution model
[10:21] <lamont> compiling with -O1 makes the problem go away (although no guarantees came with that answer)
[10:22] <jdub> o/~ we are the badgers, we are the badgers, no time for hedgehogs, cause we are the badgers! o/~
[10:22] <ivoks> :)
[10:22] <lamont> but a sourceful upload for most of them would be a waste of mirror-bandwidth
[10:23] <lamont> mdz: on the bright side, we force gcc-3.4 for glibc and gcc-4.0.  Or maybe not -- they break big time.
[10:25] <lamont> (actually, gcc-4.0 might be built with gcc-4.0... that'd be bad
[10:40] <mdz> seb128: the xdmcp greeter seems to fail to start in my gdm now
[10:40] <mdz> seb128: is that just because I've upgraded and not restarted gdm yet?
[10:42] <seb128> mdz: so sure, could you restart gdm? If that still happens then that's a bug ...
[10:43] <mdz> seb128: I would have to log out :-)
[10:43] <seb128> correct :)
[10:43] <seb128> what does it say?
[10:43] <mdz> and lose all my firefox state and emacs state and xchat state :-)
[10:44] <mdz> a dialog opens (on the thin client) saying the greeter could not be started and the display will be disabled
[10:44] <mdz> "Cannot start the greeter program, you will not be able to log in.  This "
[10:44] <mdz> "display will be disabled.  Try logging in by other means and editing the "
[10:44] <mdz> "configuration file"
[10:45] <mdz> daemon/slave.c:2665
[10:45] <mdz> 2792
[10:45] <mdz> that po file was out of date :-)
[10:45] <seb128> hum, lemme try here
[10:45] <mdz> seb128: did gdmlogin move to a new location?
[10:46] <seb128> yep
[10:46] <mdz> ah, ok
[10:46] <mdz> that path is compiled in it looks like
[10:46] <seb128> they moved binaries from /usr/bin to /usr/sbin or /usr/lib
[10:47] <mdz> I probably just need to restart it, then
[10:47] <mdz> I hate logging out
[10:47] <seb128> he he
[10:47] <mdz> seb128: did you see when I was talking about processes being left behind from the session, with connections to the X server?
[10:48] <mdz> it causes some problems with LTSP-over-ssh
[10:48] <seb128> mdz: nop, when was that? what processes?
[10:48] <mdz> bonobo-activation-server, gnome-settings-daemon, xscreensaver
[10:48] <mdz> gam_server also, but I don't think that talks to X, does it?
[10:49] <mdz> no, it doesn't
[10:49] <seb128> no 
[10:49] <mdz> so that one is ok
[10:49] <seb128> I'll have a look for b-a-s
[10:50] <mdz> bonobo is ok too
[10:50] <seb128> weird for g-s-d, I'm pretty sure it exits with gnome-session here
[10:50] <mdz> I just need for xscreensaver and g-s-d to die
[10:50] <mdz> oh, nm
[10:50] <seb128> ?
[10:50] <mdz> killing g-s-d seemed to cause b-a-s to exit
[10:50] <mdz> let me do a better test
[10:52] <mdz> mizar:[/tmp/gdm-2.8.0.0]  sudo lsof -i G -e '->localhost:6010'
[10:52] <mdz> gnome-set  8704 testuser    3u  IPv4 3010231       TCP localhost:55874->localhost:6010 (ESTABLISHED)
[10:52] <mdz> xscreensa  8718 testuser    3u  IPv4 3012713       TCP localhost:57066->localhost:6010 (ESTABLISHED)
[10:52] <mdz> so g-s-d and xscreensaver are the only ones connected to the X server
[10:52] <mdz> the others are not important for this issue
[10:55] <seb128> mdz: lemme try here
[11:00] <seb128> mdz: no xscreensaver or gnome-settings-daemon after a logout here
[11:00] <mdz> seb128: hmmm
[11:00] <seb128> mdz: gnome-session should really stop gnome-settings-daemon
[11:01] <mdz> perhaps I should strace it
[11:01] <mdz> ?
[11:01] <seb128> and according to Np237 xscreensaver exits when the connexion to X is broken
[11:02] <mdz> seb128: right, this isn't a problem with gdm logins because it resets the X server
[11:02] <mdz> I can't reset the X server because I can't tell when the session has ended
[11:02] <mdz> because these processes cause ssh to keep running
[11:03] <mdz> I run ssh, wait for it to exit, and then reset the server
[11:03] <mdz> gdm runs gnome-session, waits for it to exit, and then resets
[11:03] <seb128> right
[11:03] <mdz> gnome-session exits when it exits, but ssh doesn't exit until all connections are closed
[11:04] <seb128> k, so that's an issue for xscreensaver
[11:04] <seb128> but should not for gnome-settings-daemon, since you logout, which closes gnome-session and should stop it
[11:13] <lamont> Kamion: thoughts on 314368 and how it affects breezy?  likewise, any issue with dropping the modutils udeb entirely?
[11:19] <jbailey> lamont: I will likely want to compile modutils with klibc or uclibc anyway for the initramfs stuff I'm doing.
[11:19] <jbailey> lamont: That might solve the problem in a different way?
[11:38] <mpt> mvo: To answer your question from a while back, it's not currently possible to add comments to a Malone task, but there will soon be an "Explanation of status" field where you can put task-specific info
[11:38] <mvo> mpt: aha, thanks
[11:39] <mpt> and update it over time, wiki-like
[11:40] <tseng> mako: holy crap you used coq in a poem
[11:46] <jdub> seb128: yo?
[11:47] <seb128> jdub: pong
[11:47] <jdub> seb128: esound 0.2.36?
[11:47] <jdub> seb128: want me to turn it around?
[11:47] <seb128> jdub: I don't care about esound, feel free to do it :)
[11:48] <siretart> hi folks
[11:48] <jdub> seb128: you make esound feel unloved
[11:48] <seb128> it is
[11:49] <siretart> maybe a stupid question, but is it possible to distribute binary contents in debian/? like debian/package.png?
[11:49] <jdub> heh
[11:49] <jdub> harsh.
[11:49] <jdub> but fair.
[11:49] <jdub> MORE X!
[11:49] <jdub> MIRROR ME HARDER!
[11:49] <mako> tseng: OF COURSE I USED COQ
[11:49] <tseng> mako++
[11:50] <mako> Super bonobo sex atop Alex!
[11:50] <maswan> jdub: "it's only bandwidth"? :)
[11:50] <tseng> jdub: what is my rss feed on puc?
[11:51] <jdub> maswan: says you, swede!
[11:51] <tseng> jdub: the one wordpress is linking to seems busted
[11:51] <jdub> [http://smarterits.com/~brandon/log/wp-rss2.php] 
[11:51] <tseng> jdub: cheers
[11:52] <maswan> jdub: well, I wouldn't like to be living on an island where all the bits go through a single tiny straw.. :)
[11:53] <jdub> it's like a thickshake
[11:53] <tseng> yay for context
[11:53] <jdub> the harder you suck, the less you get
[11:53] <maswan> tseng: It was tempting, but I didn't dare add encouraging comments to that.
[11:54] <tseng> maswan: jeff needs no encouragment
[11:55] <maswan> tseng: exactly. :)
[11:55] <jdub> i'd take offense at this, if i thought i could get away with it
[11:56] <ogra> siretart, uuencode the png, then it becomes text data.... uudecode it in the rules file
[11:57] <jdub> i wonder what hct does about binary files
[11:58] <siretart> ogra: yes, but then I would have to build depend on sharutils :/
[11:58] <ogra> siretart, whats wrong with that ?
[11:58] <siretart> well, If there is really no other way..
[11:58] <jdub> siretart: no kittens will be killed for that
[11:58] <siretart> ok. will do
[11:59] <ogra> siretart, its the best way to do it... the next step is to convice upstream to use your png, so it will get included in the next upstream release ;)
[11:59] <siretart> ok