[12:00] Kamion: warty has bigger warts than that :-) [12:01] Hi, is there some info somewhere on the wiki about how to build custom distribs based on ubuntu ? [12:01] lucas_: i think thats a goal of hoary [12:01] mdz: oh yeah [12:02] tseng: arg, not before hoary ? I would have liked to build a warty-derivate ... [12:02] lucas_: you can certainly derive from warty if you would like [12:03] there is no step-by-step howto [12:04] mdz: who could give me an overview of how to build a derivate that only adds a dozen of packages ? === Kamion wonders vaguely if such a person exists yet. :-) [12:05] lucas_: it would be very nearly the same process as building a Debian derivative [12:05] if I knew how to build a debian derivate, I wouldn't be there asking :) [12:05] the archives of the debian-custom mailing list may have useful information [12:06] I think there's a "debian-cdd" howto (or some such name) [12:06] pants off dudes [12:06] zi[ [12:06] *p === Kamion squirts a water pistol at jdub [12:07] ok, thanks for the pointer [12:07] ow! morning! cold! [12:07] I think I am learning about this "cold" [12:08] it was only ~16C yesterday [12:08] wuss [12:08] I had to wear sleeves! === mdz gathers firewood frantically [12:09] rock, the dude who ITPed ifolder has passed it over :) === Mitario [~michiel@sikkes.xs4all.nl] has joined #ubuntu-devel [12:14] lo everyone [12:15] jdub: Make ifolder non-sucky [12:15] tberman is helping with that [12:15] hey Mitario [12:15] or use epittance instead [12:16] it performs a different function === jdub will look at it more after someone like alexl gets his mitts on it [12:18] 1 === rastaman [~sgtshatta@24-193-25-90.nyc.rr.com] has joined #ubuntu-devel [12:35] i think beagle cvs is broken atm, make in the top dir does nothing. [12:35] so much for packaging === doko [doko@dsl-084-057-024-165.arcor-ip.net] has joined #ubuntu-devel [12:58] Night [01:05] Kamion: which d-i bit creates /sbin/unconfigured.sh? === elmo_ [~james@83.216.141.215] has joined #ubuntu-devel [01:21] mdz: which d-i bit copies all .deb from the cdrom to the HD ? [01:21] archive-copier [01:21] I thought it was archive-copier, but I read its code [01:21] from the code and apt-cdrom's manpage, it only adds the cdrom entry in /etc/apt/sources.list [01:21] was apt-cdrom hacked to change that ? [01:21] bob2: i see the syntax error in tomboy now [01:21] elmo: I read the apt-cdrom source code and still can't find a place where it copies the .deb to the HD. [01:21] hornbeck: finishing up libdbus-cil pkg now [01:21] hornbeck: care to test on a clean box? [01:24] lucas_: apt-cdrom does not do any copying. archive-copier does [01:25] oh, got it [01:25] I was only reading the "prebaseconfig" script [01:25] postinst makes more sense :) [01:26] tseng: yes I would like to try it [01:26] is it in your repos? [01:26] one moment [01:26] and not on the box you already installed dbus from cvs on [01:26] since we wont easily know if the dep is filled by the package, or something you installed already [01:27] eg, do we need anything more than the dll's [01:27] tseng: I do not think so, I could be wrong though [01:28] its rare but it has been known to happen :) [01:28] hm i guess it needs the .pc also [01:28] oops === lamont [~lamont@mix.mmjgroup.com] has joined #ubuntu-devel [01:31] daniels: ping [01:35] mdz: which bit of d-i installs the packages in ubuntu-desktop but not in base after the reboot ? [01:36] lucas_: base-config [01:36] oh ok [01:38] hornbeck: libdbus-cil is in my repo, with all the +cvs dbus stuff [01:39] hornbeck: i took it from thom so im not sure if it is new enough for beagle.. dated 10/7 [01:39] tseng: I will check in alittle while [01:39] I am doing school work right now [01:40] any clue why blam resets to defaults everytime it launches? [01:42] ok, I now have a much clearer view of how d-i works. I'll write a nice howto if I succeed in building an Ubuntu-derivate =) thanks for answering my questions ; good night. [01:52] oh hornbeck [01:52] could you send me that tarball after all, my cvs checkout seems b0rk === T-Bone is now known as T-None [01:53] when you have a chance that is === T-None [varenet@d190.dhcp212-198-139.noos.fr] has left #ubuntu-devel ["Client] [02:02] <__daniel> sleep tight [02:12] yeah tseng [02:13] can I get a mail for you [02:13] brandon@smarterits.com [02:14] on its way [02:17] evolution is crapping out on sending attachments [02:17] will try to send from other account [02:23] tseng: now it is on its way [02:25] hornbeck: er dude [02:25] i meant beagle [02:25] did i say dbus? [02:26] haha, yeah [02:26] ok I will send beagle [02:26] thanks === rastaman [~sgtshatta@24-193-25-90.nyc.rr.com] has left #ubuntu-devel ["Leaving"] [02:27] tseng: now that one is on its way [02:28] great [02:28] tseng: you should shoot for a gecko-sharp-0.6 also :-) [02:28] perhaps [02:28] nice === srbaker_ [~srbaker@blk-224-143-227.eastlink.ca] has joined #ubuntu-devel === sabdfl [~mark@host217-37-231-28.in-addr.btopenworld.com] has left #ubuntu-devel [] [02:39] okay, after a full day of ubuntu on my laptop, i'm even more impressed than before. great work, guys [02:46] thanks [02:53] " [02:53] i've done with install snd_via82xx /bin/true" [02:53] ^ eeeeeek! === nmf [~nmf@adsl-b4-66-44.telepac.pt] has joined #ubuntu-devel [03:06] jdub/mdz: I think rc2 == release. thoughts? === sgtshatta [~sgtshatta@24-193-25-90.nyc.rr.com] has joined #ubuntu-devel === jamesh [~james@203-59-217-232.dyn.iinet.net.au] has joined #ubuntu-devel [03:27] kinda neat how when you move around the website you are logged in one page than not another, good way to keep people from being able to do anything [03:33] hmmm === hazmat [~hazmat@c-24-15-10-12.client.comcast.net] has joined #ubuntu-devel [03:49] hornbeck: you can use https://site-edit.ubuntulinux.org/ until that's fixed [03:53] mdz: thanks [03:55] man that is way to complicated [03:57] anyone know how i can find out when to expect ubuntu cds? i'm planning our LUG meeting, and i want to know which one my "introduction to ubuntu" talk (with cds) will take place [04:00] hornbeck: it's a bug [04:01] lamont: I think we have no choice :-) [04:01] srbaker_: http://lists.ubuntu.com/archives/ubuntu-users/2004-October/007832.html === mdz gleefully closes an FTBFS-on-m68k bug [04:20] aw, daniels beat me to it [04:48] mdz: true, I think there are bugs that would be nice to have fixed, but I think they mostly come back to hwdetection diffs and minifo :-( === srbaker_ [~srbaker@blk-224-143-227.eastlink.ca] has joined #ubuntu-devel [04:51] mdz: quick draw, yo === hazmat [~hazmat@c-24-15-10-12.client.comcast.net] has joined #ubuntu-devel === srbaker_ [~srbaker@blk-224-143-227.eastlink.ca] has joined #ubuntu-devel [05:02] morning guys [05:03] morning papa fabbione [05:03] good morning [05:03] night here :-) [05:06] hey kid [05:08] daniels: we will probably have a kitchen the first week you will be here :-))) [05:08] awesome! [05:08] and hopefully the bigger office during the second one [05:08] your timezone shift is bad, dude [05:08] i know [05:10] it's not like i am having fun waking up at 5 am every day [05:22] ;) [05:30] amu: you are not supposed to be awake either ;) === srbaker_ [~srbaker@blk-224-143-227.eastlink.ca] has joined #ubuntu-devel [05:37] fabbione: correct, i lie straight on a slopematte in the southseas, 2 pretty girls make breakfast for me, wow i dream stilD[D[D[D[Dl === mdz_ [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu-devel [05:53] amu: eheheh === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has left #ubuntu-devel ["Client] === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu-devel === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu-devel === mdz [~mdz@69-167-148-207.vnnyca.adelphia.net] has joined #ubuntu-devel === hazmat [~hazmat@c-24-15-10-12.client.comcast.net] has joined #ubuntu-devel === SuperLag [~colbyirc@CPE-69-76-188-71.kc.rr.com] has left #ubuntu-devel ["Leaving"] === Keybuk [scott@descent.netsplit.com] has joined #ubuntu-devel [07:26] morning fabbione, did you get the debhelper things working? === ..[topic/#ubuntu-devel:mdz] : Ubuntu development -- general discussion and support on #ubuntu | Happy Hoary Trail! | BE THE SIGNAL | Warty release is DONE | Hoary kickoff meeting Monday, 2004-10-25 1600UTC in #ubuntu-meeting === SuperLag [~colbyirc@CPE-69-76-188-71.kc.rr.com] has joined #ubuntu-devel [07:41] xmlconfig.c:912: error: `program_invocation_short_name' undeclared (first use in this function) [07:41] and errno.h is included [07:41] what am I missing here? [07:42] #define __USE_GNU ? [07:43] Keybuk: er, _GNU_SOURCE? [07:44] Redhat found a vulnerability in fileutils (ls and mkdir), that could allow a remote attacker to execute arbitrary code with root privileges. [07:44] ^ ouch [07:45] though highly odd *how* you'd do that without making ls or mkdir setuid root [07:46] Received: from 2ens11.uta.edu (2ens11.uta.edu [129.107.2.122] ) by [07:46] menubar.gnome.org (Postfix) with ESMTP id 951303B0D8A for [07:46] ; Sun, 24 Oct 2004 18:00:13 -0400 (EDT) === Keybuk decides that mail is highly suspicious [07:46] Keybuk: also, Red Hat know how to spell their company name [07:46] also the instructions are to untar, build and install a random tar file [07:47] RH ship security updates as RPM [07:47] interestingly, they ship 'fileutils-patch.bin' in that tarfile [07:47] which is an RPM [07:48] ah yes, bit red notice on RH.com [07:48] daniels, Keybuk: thanks... [07:48] i wonder more why it wasn't set automatically [07:49] fabbione: you have to set it to say you're happy that your code is no longer C99/POSIX/blah and you're happy about that [07:49] also, _BSD_SOURCE, _POSIX_SOURCE, and _XOPEN_SOURCE [07:50] the latter two accept numbers as to your level of posix/x/openness [07:53] hmm [07:53] the only lib complaining about it is libgl [07:53] i wonder if others are affected too [07:54] Keybuk: this is dri (black magic wodoo) stuff [07:54] very difficult to say [07:55] ehm [07:55] daniels: what do you think about confining the -D_GNU_SOURCE to libgl ? [07:55] you know how I worked out which flags most modular libs used? [07:55] buildd.d.o [07:56] if you can do per-file cflags in imake, even better [07:56] maybe even just a #define _GNU_SOURCE rather than -D [07:56] daniels: yes i can do that [08:06] daniels: btw.. all these little changes are just because of -DUseInstalled [08:06] a real pain in the butt [08:07] and it needs to be done at Imake level [08:09] but i can still confine it to that specific dir [08:09] there is no need to force it on the entire lib [08:17] fabbione: building mesa/glx and the server separately is a huge pain in the arse [08:17] right now you can't do it and keep dri [08:18] that's why my monolithic tree still contains lib/GL and extras/Mesa [08:18] (also, Xfont is really bizzaro, but that's another story) [08:18] daniels: yes i know they circular build-dep [08:18] i already have the xserver source tree as dependency [08:19] hm [08:19] right now libgl is the worst crap i have seen in terms of Imakefile [08:20] i'm glad you guys can bond over X horrors [08:20] fabbione: isn't it great? [08:20] jdub: sort of like old war stories [08:20] jdub: all this stuff is coming out because we are splitting the tree [08:20] YAY! [08:20] 'and then there was the time I realised half of the XKB code was probably exploitable if someone tried hard enough' [08:20] daniels: yes, but that doesn't scare me too much [08:22] jdub: as i already told Mark, the amount of work is much higher than what i planned in the beginning [08:22] jdub: unfortunatly i realized it later [08:22] but we will do our best [08:24] daniels: ''But I just typed "2*&)$}{;')(*!" at the keyboard..'' [08:26] daniels: so how did you happen upon the Boeing glider story? === vorlon had never heard that the mechanics ran out of gas on the way there, though, that's hilarious. :) [08:27] yeah was reading that, is a good telling of it :o) [08:27] I first heard about it reading some racetrack trivia stuff [08:27] vorlon: chris blizzard linked to it on his blog [08:27] ah. :) [08:28] there's several examples of races being interrupted by planes doing emergency landings (lots of race tracks are old airfields), but that's the only one which was a commerical liner === vorlon had heard about it because his dad is a glider pilot; the story was occasionally brought up as an "and that's why all commercial airline pilots should have to learn to fly gliders", when in reality it's a "and that's why airlines should ground planes that are suffering computer failures" ;) [08:30] although, knowing your pilot can glide the plane in if there's a problem is kinda nice [08:31] depends on the plane really :) [08:32] sure, though they showed that the glide ratio on those things really is lousy, and there aren't all that many airline pilots who do know how to fly gliders. :) [08:32] dude, if it's a 747-400, i still want to know the guy can glide it [08:32] indeed; the natural state for an out-of-fuel jetliner is "crater" [08:34] Silverstone is pretty fantastic for that; it's still an in-use airfield as well as a F1 Grand Prix race track -- on race day a fleet of helicopters arrives and lands in the middle of the track as all the rich kids get around the parking problems [08:35] Keybuk: how was the race yesterday? [08:35] i don't have a TV yet :) [08:35] fabbione: got a bit wet, but no major casualties [08:35] who won? [08:35] Montoya [08:35] argh [08:35] Kimi 2, Barichello 3 [08:35] schummy? [08:36] somewhere a little way back, he toasted his car in the wall during practice so had to take an engine change [08:36] montoya! [08:36] fabbione: you seen the rule changes for 2005 yet? [08:36] Keybuk: nope.. i only read the one you posted here [08:36] about no tyres change during the race [08:37] that is going to be sooooo boring [08:37] front wing height and rear wing size reduced; so will be interesting whether we'll see 1980-era wings-fixed-to-noses again [08:37] i hope there is at least the exception in case of rain/dry weather [08:37] Keybuk: you'd land a plane on alicia silverstone? bastardo! [08:37] no tyre changes during race (except for environmental conditions) [08:37] in fact, same tyres have to last for both Qualy *and* the race [08:37] amen [08:38] we will see more cars out with falt tyres than anything else [08:38] which is interesting that they didn't forbid refuelling as well; because the pit boys won't have much to do now [08:38] engines have to last for two race weekends, not just one === hazmat [~hazmat@c-24-15-10-12.client.comcast.net] has joined #ubuntu-devel [08:39] so if one engine blows up [08:39] 9/10 teams want to eliminate tyre testing entirely, so effectively use standardised tyres for all teams [08:39] you are out for 2 races [08:39] fabbione: no, just have to take a 10-position penalty [08:40] bah that sucks [08:40] 9/10 teams want to reduce testing to just 10 days throughout the entire season [08:40] it's getting to complicate [08:40] Qualy is going to be Saturday afternoon for a first low-fuel run in previous race finishing order [08:40] then on Sunday morning, they'll run again with race fuel in reverse order of previous finishing [08:40] Position will be based on an aggregate of both times [08:41] why don't they just use /dev/urandom? [08:41] it will save them tons of headackes === sivang [~dannyh@80.179.93.130.forward.012.net.il] has joined #ubuntu-devel [08:42] heh, I still think they should go back to the *original* Qualy format, but with the addition that you have to run laps in 15 minute windows (to prevent everyone going out with 10 minutes left on the clock) [08:43] morning all [08:43] next year's going to be kinda interesting [08:43] Keybuk: i think they should go back to the original. that's it.. same as it was at Prost and Senna time [08:43] hi sivang [08:43] Minardi are probably safe, they have their own engines they can run [08:43] Jordan are probably safe, they are rumoured to be running Toyota engines next year [08:43] fabbione : morning, how are you? :) [08:44] Jaguar are still in the shit though, no buyer yet [08:44] sivang: fine thank and you? [08:44] Keybuk: too bad... [08:44] fabbione: will be amusing, because Ferrari will have to run a third car if they drop out [08:44] i wonder when they will allow more than 2 cars x team [08:45] Keybuk: uh why? [08:45] it's not a matter of "allow", it'll be a requirement for the top teams [08:45] F1 rules state a minimum of 20 cars must run [08:45] ahhh [08:45] the third car won't be a points-scorer [08:45] so if you happen to be in the third Ferrari, you don't count [08:45] and what's the point of having a third car than? [08:45] just to fill up the grid? [08:45] fabbione : not bad at all, I see it's race car discussion day? :) [08:45] thaytan: you shrink you [08:45] it counts for the driver, not the teeam [08:45] bah. sorry [08:46] sivang: well yeah.. i don't have a tv yet in the new house [08:46] and Keybuk is updating me [08:46] though I think it'd be sweet if they let Jordan and Minardi have the third cars, and be points-scorers :) [08:46] Keybuk: understandable... will the team decide which cars will bring points to the team or they will pick the 2 highest score for each race? [08:47] 2 highest runners, I believe [08:47] fabbione : don't major sports channels broadcast on the internet as well? [08:47] though Brundle was being cynical and suggesting that a Ferrari 1, 2, 3 is the last thing the sport needs next year :o) [08:47] sivang: no, F1 isn't net-broadcast [08:48] Keybuk : this is an european exlusive sport channel I suppose? (I know Eurosport, ESPN for the states etc) === bigbrother0074 [~bigbrothe@cdm-24-250-145-26.bcst.cox-internet.com] has joined #ubuntu-devel [08:48] sivang: F1 = FIA Formula 1 [08:49] Keybuk : Ah so that's the famous formula 1 races , those are better watched from within the audience ;) [08:49] Keybuk: ahahah that would own the F1.. Ferrari 1 2 3 :P [08:49] actually, I tend to disagree there ... they are fantastic to attend; but you get a better race if you watch at home [08:50] yeah [08:50] you don't see anything at the race [08:50] it gets too messy [08:50] it's way too much to go here [08:50] prices start at about $au450 or so [08:50] fabbione: not really likely though, Ferrari are notably mid-field at the moment -- all the other teams have caught up; even Schumy is notably off-pace [08:51] it's almost as if he's bored, he's made silly mistakes in the last several races [08:53] Keybuk: yeah well... he is demotivated after winning the championship [08:54] which is odd for Michael, he's normally even more motivated afterwards and just driving for fun [08:55] he is the "oldest man in F1" now ... though I give him a few more years yet, I expect we're starting to see the peak of his career [08:55] yeah i agree [08:56] still waiting to see who'll be in the 2nd Williams [08:56] and, if they have to, 3rd Ferrari and BAR [08:56] (I guess the latter will be Anthony Davidson) [08:57] probably Fittipaldi? [08:58] isn't he the test driver for Ferrari? [08:58] no idea [08:59] yay Webber [09:08] make[1] : Leaving directory `/usr/src/xorg/xorg-lib-gl-6.8.1/build-tree/xc/lib/GL' [09:08] touch stampdir/build [09:08] YES [09:08] finally [09:09] daniels: isn't about time to sign my gpg keys? [09:09] fabbione: last time i asked daniels about signing your key, he said he wasn't sure that you really were 'Fabio, The Most Beautiful Man In The World' === vorlon vouches for fabbione's beauty [09:12] jdub: you didn't even want to exchange id's with me :( [09:12] vorlon: eheh [09:13] fabbione: we've already exchanged, haven't we? at the original uk meeting [09:13] jdub: only for one of my key [09:13] i am searching sigs for my new shiny 4096RSA key and the canonical key ;) [09:14] gpg --list-sigs fabbione@fabbione.net |grep Jeff [09:14] $ [09:15] so that means that even if we exachanged... you didn't sign [09:15] ;) [09:15] i didn't ;) [09:15] you suck :P === pitti [~martin@195.227.105.180] has joined #ubuntu-devel [09:17] Morning [09:19] heh [09:20] yeah, i should probably sign keys from oxford ... and akademy ... and ols ... and lca ... [09:20] heh, i should probably sign keys from lca ... 2001 [09:21] aj: and debconf4 :-) [09:21] or was it 3? [09:21] the one in Olso.. [09:21] olso [09:22] OSLO [09:22] ok [09:23] 3 was oslo, 4 was brazil === Gmail [~Google@gnu-debian.user] has joined #ubuntu-devel === KeyserSoze [unbound@pound.ifndef.com] has joined #ubuntu-devel [10:01] daniels: is there actually any reason why we build libgl1.2 when 1.4 is available in debian? [10:01] usr/lib/libGL.so.1.4.500 libs/mesag3,libs/mesag3-glide2,libs/mesag3+ggi === carlos [~carlos@69.Red-80-33-181.pooles.rima-tde.net] has joined #ubuntu-devel [10:04] fabbione: yeah -- mesa built independently from X is incapable of direct rendering [10:04] only software [10:05] working on using debian's mesa on a branch was what got me kicked out of the XSF the first time :P [10:05] in how many hr is the meeting? [10:05] i want to getready [10:06] but i think its in 8hrs [10:07] daniels: ok... and do you happen to remember why there is a /usr/lib/libGL link to /usr/X11R6/lib/libGL ? [10:07] probably because people suck and hardcoded the path === fabbione is extremely tempted to kill it === enrico_ [~enrico@81-174-12-206.f5.ngi.it] has joined #ubuntu-devel [10:08] fabbione: just like the rest of /usr/X11R6 :) === Ferry [~Ferry@cust.15.118.adsl.cistron.nl] has joined #ubuntu-devel [10:10] daniels: eh the libgl1 stuff is complex [10:11] hm? [10:11] because libgl1 is a virtual package provided by several other packages [10:11] there is all a rationale behind it [10:11] it's in debian/changelog [10:11] yeah [10:11] that also means that i cannot call the package libgl1 [10:11] oh, right [10:11] and that seriously SUCKS HARD [10:11] so you're saying there's a /usr/lib/libGL->/usr/X11R6/lib/libgl link? [10:12] daniels: there is a link because that link is handled by different packages as far as i can see [10:12] so yes.. it needs to stay as symlink [10:12] /usr/lib/X11/libGL.so [10:12] hey, wait ... === fabbione is seriously disappointed by this GL mess [10:17] mmm [10:17] they're trying to fix that upstream, so mesa will be capable of direct rendering when built separately [10:17] daniels: i think i will keep the same name schema we have in Debian now.. even if i think it sucks [10:18] xlibmesa-* has nothing to do with libgl [10:18] but that's the easiest i think [10:19] i don't really like xlibmesa [10:19] i would personally prefer libgl1-xorg [10:20] or xorg-libgl1 [10:20] that is slightly more coherent with all the other packages [10:20] like 171 xorg-source-* [10:20] ;) [10:21] Package: xorg-libgl1 [10:21] Package: xorg-libgl-dev [10:21] Package: xorg-libgl-dri [10:21] ehm [10:22] libgl1-dri [10:22] mmm, but libgl1-* is far more compliant with standard naming scheme's [10:22] x's is total bong [10:22] (the xlib* mess being a bad hangover) [10:22] s/scheme's/schemes/ [10:23] also, libgl1-dri is bad because we're not actually dri [10:23] that's dri.sf.net [10:23] so i propose libgl1-xorg, but your call === ph_ [~ph@pD9E6BC38.dip.t-dialin.net] has joined #ubuntu-devel [10:25] let's keep it as xorg-libgl* [10:26] we can easily change it later [10:26] fabbione: hm [10:26] there is the same mess with xorg-libosmesa [10:26] fabbione: as i said, your call [10:26] daniels: as i said.. it's a detail right now [10:26] but in my packages, it's libgl1-xorg, libosmesa4-xorg, et al [10:26] yeah [10:26] we can change later ;) [10:26] we can fight it out in denmark :) [10:27] daniels: the ring is ready.. time for the cage :P [10:27] heh === seb128 [~seb128@ANancy-151-1-8-176.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel === __daniel [~daniel@td9091b68.pool.terralink.de] has joined #ubuntu-devel === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has joined #ubuntu-devel [10:41] <__daniel> hai [10:46] 1,320 changesets for warty === sabdfl [~mark@host217-37-231-28.in-addr.btopenworld.com] has joined #ubuntu-devel === daniels [daniel@fooishbar.org] has joined #ubuntu-devel [11:01] daniels: can you gice me the output of objdump --all-headers libGL.so.1.2 | grep TEXTREL on your libGL ? [11:01] s/gice/give [11:01] and on libosmesa4 [11:01] hey sabdfl [11:02] fabbione: there should be no TEXTREL -- see my make libGL PIC-compliant patch [11:03] daniels: that's what i was searching for [11:03] i have no TEXTREL here [11:03] but yeah, I wrote a patch to make GL PIC-compliant [11:03] anyway, dinnertime [11:04] yeah it didn't apply clean.. i will check it again [11:04] i remember the first time i didn't do it deeply [11:05] jakub jelinek did a better one, might be applied now [11:05] that will make gl *significantly* *quicker* [11:05] anyway, dinner [11:05] hiya fabbione [11:06] mdz: d-i doesn't create /sbin/unconfigured.sh; that's a boot-floppies thing [11:07] mdz: instead, /etc/inittab is just set to run base-config until base-config rewrites it not to do that === cc [~byte@c211-28-119-102.eburwd3.vic.optusnet.com.au] has joined #ubuntu-devel === fabbione starts to wonder if do we really have to support GL [11:35] fabbione: that's the unfortunate reality [11:39] daniels: i rediffed your patch and now libGL is PIC compliant [11:39] cool [11:39] but libosmesa4 isn't [11:39] yeah, that probably still needs doing [11:39] and probably your version isn't linked properly either [11:40] the Imake is missing a REQUIREDLIBS = MathLibrary [11:40] i start to eat Imakefiles for breakfast [11:41] yeah, that's a patch from Gentoo I never got to integrating [11:41] their BTS has a few patches to practically eliminate all weak references [11:41] i was working on debrix at that stage, and couldn't be arsed dealing with Branden to get it into the Debian tree [11:43] the debian patch applie on xfree86 but not x.org [11:44] which debian patch? [11:44] ok fixed [11:44] the 062_ [11:44] is the same you geve to me [11:44] the libGL PIC one? [11:44] gave to me [11:44] yes [11:44] right [11:44] that one doesn't apply to x.org [11:44] and i rediffed [11:44] hmm [11:44] now it is ok [11:44] also [11:44] rad [11:44] libosmesa was ranting about TEXTREL because it was not properly linked with MathLibrary [11:45] so that is fixed too [11:45] cool [11:45] hmm no hold on [11:45] i used the wrong the script for the PIC === fabbione sighs [11:46] at least it is linked properly [11:48] http://dev.gentoo.org/~spyderous/xfree/patchsets/4.3.0/patch-2.1.17/0192_all_4.3.0-missing-lib-sharedreqs.patch [11:49] that's a small part of the patch [11:49] http://dev.gentoo.org/~spyderous/xorg-x11/patchsets/6.8.0/patch/0130_all_4.2.1-fix-shared-libXau-link.v2.patch [11:49] you'll want that if you want a shared Xau lib (I want a shared Xau lib) [11:50] we already have a share libXau [11:51] the last patch is only partially true [11:51] s/true/correct [11:51] anyway [11:51] the pic problem is still there [11:51] and it's not a link issue [11:51] these are easy to fix [11:52] yeah [11:52] gl pic is a bitch [11:52] and when you do get it right, you lose arseloads of performance [11:52] GL is PIC now [11:52] we only miss libosmesa4 [11:52] mmm [12:01] yo, keybuk! [12:02] aj: hey [12:02] l33t [12:02] is there a generic way to find code that makes a lib non-pic? [12:02] hrm, now i should find that bugnumber [12:03] fabbione: not easily [12:03] grepping the resulting assembler *shrug* [12:04] Keybuk: Bug 62529 -- what're the chances of getting something done about it? [12:04] Keybuk: ok.... [12:04] well... almost === mvo_ [~egon@suprimo-131.ping.de] has joined #ubuntu-devel [12:05] aj: reasonably high, I think that's one I read over the weekend [12:05] yeah, you did [12:05] or someone pretending to be you set it to wishlist, anyway [12:05] I favour the "does the -revision contain .0.x" trick [12:06] it's one lamont and elmo keep poking me about :) so expect it to be one of the first bugs fixed once sarge has its translation update release [12:07] goodo [12:07] i'll see if i can get kamion to poke you about it too [12:23] Keybuk: (consider yourself poked) === daniels pokes Keybuk, for the hell of it. [12:24] (all the cool kids are doing it) === Keybuk pokes daniels back [12:25] Keybuk: might be worth going over queue/done to check whether any sourceful uploads in the last have been versioned .0.x [12:25] oh, .0.x doesn't work, binNMUs of sourceful NMUs are .1.x etc. [12:25] Kamion: there are a few upstream examples (usually bpb) but seeing as policy says .0.x in the revision is a bin-nmu, I think anyone who does it is silly [12:25] doesn't work on its own, anyway [12:26] Kamion: elaborate? [12:26] MU 1.0-2, sourceful NMU 1.0-2.1, binNMU 1.0-2.1.1 [12:27] why 1.1 ? [12:27] because -2.0.1 would be less than -2.1 and -2.1.0.1 would be excessively long? [12:27] Keybuk: rebuild of 2.1 [12:27] hence, 2.1.1 [12:27] I think the idea is third level of Debian revision = recompile [12:28] dunno, just know it's historical fact :) [12:28] I get worried about not making it *look* magic [12:28] .0.x is pretty rare [12:29] .1.x is common for general usage [12:29] I'm just observing what people have historically done [12:29] also, katie supports .. [12:30] re_bin_only_nmu_of_mu = re.compile("\.\d+\.\d+$"); [12:30] re_bin_only_nmu_of_nmu = re.compile("\.\d+$"); [12:31] all I'm saying is it'd be worth checking queue/done, since people do all sorts of weird shit and it might be worth choosing something with maximal weird-shit deflection potential :-) [12:34] heh :) [12:34] the software design metric of the future! [12:39] Keybuk: feel free to make up your own magic, like .recN, or .0.N$ or something different [12:39] Keybuk: we can always just REJECT anything that doesn't match the magic === Keybuk hunts for a spare character on the keyboard [12:40] new characters would be bad, though; cf ~... [12:40] at least ~ actually makes sense [12:40] cf. patch, diff -ru foo-1.0~ foo-1.0 [12:41] yeah, but it's a nuisance trying to get it to actually work [12:42] hrm, ".." could be cute foo_1.0-1..1 [12:42] eww! that so gross! [12:42] [12:43] eww, only [0-9A-z+.] to play with [12:44] .0+1? [12:44] + tends to get used for what ~ is intended for [12:44] 1.0-1+but.really+0.99.1 [12:45] * Not changed: geda-gschem ... (ugh, that's a horrible package name) [12:45] ajt@newraff:~/queue/done$ find | grep -- '-.*\.0+1_' | wc -l [12:45] 0 [12:46] aj: that's certainly a useful idea then [12:46] foo_1.0-1.0+1 would work [12:46] *shrug* i'm happy if you break things though, REJECTing is easy :) [12:46] anything that actually gets implemented gets my vote [12:46] are there any in queue/done that end in "+\d*" ? [12:46] upstream included [12:47] ./2002/11/28/openoffice.org-debian-files_1.0.1-6+1_i386.changes [12:47] ./2002/03/05/cfdisk-utf8_2.11n-5+1_powerpc.changes [12:47] ./2002/04/07/alsa-modules-2.4.18-i386_0.9+0beta12+3+2.4.18+4+1_i386.changes [12:47] ./mutt_1.5.6-20040907+1_powerpc.changes [12:48] ./vim_6.3-025+1_arm.changes [12:48] mantis will be used by ubuntu after bugzilla? [12:48] ./2002/03/04/tetex-bin_1.0.7+20011202-5_i386.changes [12:49] *nods* I'll bounce off lamont when he wakes up and see if he has any opinions [12:49] "rc+\d" doesn't match anything, so foo_1.0-1.rc+1 would work too [12:49] "rc" ? [12:49] +rc1 matches release candidates otoh [12:49] ReCompile [12:51] yeah I read "rc" as "Release Candidate" [12:51] rb+1 for rebuild, maybe [12:52] +rb\d works too [12:54] I like the +\d+ form ... foo_1.0-1.1+rebuild3 ... kinda reads as "and 3 rebuilds" [12:57] 'rebuild' is relatively short so it could be spelt out, IMHO [12:57] doesn't make people like me wonder what it means, as e.g. with 'ds' [01:01] it doesn't really matter too much, as it's not important for anyone but buildd freaks :) [01:01] aj: is +b\d used? === sparkes [~sparkes@host81-154-216-173.range81-154.btcentralplus.com] has joined #ubuntu-devel [01:03] no +b\d [01:04] cool, that's reasonably short === seb128 [~seb128@ANancy-151-1-8-118.w83-194.abo.wanadoo.fr] has joined #ubuntu-devel [01:05] hi seb128 === sid77 [~sid77@host98-44.pool8020.interbusiness.it] has joined #ubuntu-devel [01:13] hi all === sid77 is away: Far from here (close to you) [01:30] daniels: ping [01:31] pong [01:32] fabbione: 2676 is yours [01:33] daniels: i was talking with Overfiend right now [01:33] he wrote that section of policy about */X11 [01:33] and he said that we can go as we like [01:33] awesome [01:33] that means no X11R6 [01:33] right [01:34] remember you're the maintainer, ultimately -- /usr and /usr/X11R6 is your call [01:34] you know my opinion by now, I'm sure :) [01:35] daniels: yes i know your opinion [01:36] right now i want to get libgl out of my way [01:36] but the problem is still that TEXTREL [01:36] and i can't find the piece of code generating it [01:36] osmesa.c includes 3 tons of crap [01:39] crap [01:39] yeah === SuperLag [~colbyirc@CPE-69-76-188-71.kc.rr.com] has joined #ubuntu-devel [01:43] * \file imports.h [01:43] it seems to be the one at fault [01:44] ./extras/Mesa/src/mesa/main/glheader.h [01:44] ./extras/Mesa/src/mesa/main/imports.h === sid77 is back (gone 00:21:16) [01:45] sid77: please turn off public away [01:51] moo [01:51] Keybuk: what version number? [01:55] lamont: we were discussing bin-nmus [01:56] the idea that dpkg could recognise a certain form of version number, and generate the changes so it's missing the bit at the end in the Source: header [01:58] daniels: i think i found a solution to the TEXTREL [01:58] fabbione: wassat? [01:58] but it's going to impact performance. there is no other way around [01:58] check imports.g [01:58] ehm .h [01:58] all our PIC changes have killed performance [01:58] i'm not sitting at a computer with an unpacked tree right now [01:58] it has several specific asm definition [01:58] right [01:59] and also a generic portable C one [01:59] so it's question of do in such a way that it will compile the generic one [01:59] always [01:59] hm [02:00] personally I'd be shooting to make the ASM PIC-compliant [02:00] Keybuk: .0.N is kinda defacto, you know. === Gmail [~Google@gnu-debian.user] has joined #ubuntu-devel [02:05] lamont: except Kamion was saying it wasn't, and that when a source-NMU has been done, you do .N; and it's not safe when applied to upstream [02:06] daniels: well.. if you know enough about asm.. patches are welcome :-) [02:06] and no real standard exists for native packages. === gruberman [~gruberman@h9n2fls35o294.telia.com] has joined #ubuntu-devel [02:07] lamont: so we were discussing going with something else entirely ... I personally like +b1, +b2 etc. [02:07] I think that's why I ended up liking aj's DEB_RECOMPILE suggestion after thinking about it, but some convention that nobody's used yet would work too [02:07] Keybuk: you know some smartarse will use it for beta versions though - why abbreviate? [02:07] the code I've seen basically says "if there are 3 more components in the debian version number, and there is source excluding either the last 1 or 2 components (which must be numeric), then that's it...' [02:08] +binNMU [02:09] of course, anything that is not '.x.y' will require changes in katie and sbuild, to name just a few. [02:09] fabbione: look at the way I did libGL :P === Gmail [~Google@gnu-debian.user] has left #ubuntu-devel ["Leaving"] [02:11] Keybuk: the goal here is for dpkg to know that it's building an NMU? [02:11] er, binNMU? [02:12] more that the archive can know reliably, I suspect [02:13] note also that a binNMU of a package in ubuntu is newer than the ubuntu version... (1-1.0.1 > 1.1ubuntu1) [02:13] lamont: er, is it really? [02:13] elmo: ah, right. [02:13] daniels: yep. [02:13] i would've thought that 1ubuntu1 > 1.0 [02:14] dpkg --compare-versions 1.0.1 gt 1ubuntu1 && echo y [02:14] y [02:14] daniels: thanks for playing, though. [02:14] :-) [02:14] heh [02:14] daniels: "." > "u" [02:15] lamont: (ITYM, 'kthxbye' :P) [02:15] fortunately, the only binNMU in warty has a newer version in sid [02:15] elmo: lol [02:15] sablevm, for those with scorecard [02:15] s [02:15] lamont: yeah, basically it's so dpkg-genchanges can go "ah, a recompile, I'll stick 'Source: foo (1.0-1)' in then" [02:16] elmo: the ubuntuforums folk are find of 'k thnx. [02:16] ' [02:18] bah [02:18] that asm isolation killed my ccache [02:20] Keybuk: then it's probably just a matter of getting all the principals in one place an picking a format,eh? [02:20] yup [02:20] personally, I don't really care, so long as the next sourceful upload has a higher version [02:21] for extra credit, pick something < 'u' :-) [02:22] that's kinda tricky [02:22] you'd actually have to pick a new character for that, and put it < 'a' [02:22] yeah - that's why it's _extra_ credit [02:22] all punctuation is > 'z' right now [02:22] or choose a-t to be the character. :-) === lamont ducks [02:23] see, if you used "x.ubuntu.y" instead of "xubuntuy" you could use '+', + < '.' ;o) === jdub is still wondering wtf derivatives are going to do [02:24] Keybuk: yeah, the issue was that we wanted 1ubuntu1 < 1.1 [02:25] jdub: 1ubuntu1myderivative1 [02:25] heh [02:25] foo_1.0-1ubuntu5skole3guade2_i386.deb :p === lamont must run [02:25] back in about 90 minutes or so, actually closer to 2 hours now that I think about it, [02:26] 326, 326, 350, 22 <-- need-update, up-to-date, ubuntu-modified, ubuntu-specific [02:26] s/22/20/ [02:27] Keybuk: truthfully, I don't really mind what it is, so much as I want there to be an official standard for it. [02:27] and buyin from all concerned, of course.. [02:28] bbl [02:28] elmo: I have 438 ubuntu-modified [02:32] keybuk: I'm only doing main [02:32] right now [02:32] ahh ok [02:32] Kamion: that so doesn't scale :) [02:33] I did main+universe I think [02:33] hmm, I'd kind of forgotten about universe, meh [02:33] elmo: ubuntu-specific soon to be += ~80 [02:36] elmo: you're doing the sync of the need-update ones, right? [02:37] Keybuk: getting ready to, yeah [02:37] cool [02:37] and mailing folks about stuff that would like to be updated, but are ubuntu modified [02:37] I really wish abiword wasn't first on ubuntu-modified :o) [02:38] it'd help if there were proper hoary seeds [02:39] most of my rejects seem to be autofuck updates so far [02:39] which is encouraging [02:40] oh, and po/* hell [02:40] I swear, the guys who invented ".po" sat down and designed a file format deliberately intended to be totally unpatchable [02:40] it's a feature (TM) === fabbione kicks libgl straight in the core [02:59] thom: did apache2 steal dpkg-style sorting for filenames ?! [02:59] while we're speaking of httpds, what's up with the indexing on auckland? [03:00] * kdemultimedia trying 3.2.2-1ubuntu2 + 3.3.0-1 [03:00] 1 out of 1 hunk FAILED -- saving rejects to file debian/changelog.rej [03:00] 1 out of 4 hunks FAILED -- saving rejects to file debian/control.rej [03:01] lamont is teh suck [03:01] Keybuk: hrm? [03:02] thom: I just noticed that the files are sorted correctly [03:02] rather than 1 10 11 2 [03:02] ah, heh [03:03] doubt it's dpkg stealage :-) [03:03] it's not quite right actually, but is close :p [03:03] [ ] aiksaurus_1.0.1+cvs.2004.03.15+dev-0.12-0ubuntu1.patch 25-Oct-2004 07:02 91M [03:03] [ ] aiksaurus_1.0.1+cvs.2004.03.15-1ubuntu1.patch 25-Oct-2004 07:02 9.8K [03:03] those two should be the other way around :p [03:03] http://sourcefrog.net/projects/natsort/ [03:03] so how do i distribute the op love? === HauntedUnix [~hauntedun@HauntedUnix.student.supporter.pdpc] has joined #ubuntu-devel === Hawkwind [~SoS@Hawkwind.linuxfordummies] has joined #Ubuntu-Devel [03:04] /m chanserv access #ubuntu add nickname 20 === gruberman [~gruberman@h9n2fls35o294.telia.com] has joined #ubuntu-devel === Hawkwind [~SoS@Hawkwind.linuxfordummies] has left #Ubuntu-Devel [] [03:05] aha [03:05] done for daniels and Keybuk [03:06] and thom [03:06] anyone else? [03:06] thanks dude [03:06] i think bob2 wants it, too [03:06] done for lamont [03:06] bah add me too [03:06] he was mentioning it a while ago, and is #debian [03:06] just in case [03:06] Hello all. [03:06] done === HauntedUnix [~hauntedun@HauntedUnix.student.supporter.pdpc] has left #ubuntu-devel [] [03:07] why not :) [03:07] done :) [03:08] thanks :) [03:09] * opencv trying 0.9.5-4ubuntu1 + 0.9.5-10 [03:09] 1 out of 1 hunk FAILED -- saving rejects to file debian/changelog.rej [03:09] 2 out of 2 hunks FAILED -- saving rejects to file debian/control.rej [03:09] o/~ oh, lamont === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has left #ubuntu-devel ["Leaving"] === tuo2 [~foo@adsl-36-114.swiftdsl.com.au] has joined #ubuntu-devel [03:15] daniels: getting an wireless AP for the sprint :-) [03:15] good thing, too === fabbione gives up on Mesa [03:16] this house is now pretty reliant on my wrt54g :\ [03:16] i have tried everything i could to kill that TEXTREL [03:16] it's gonna be 11Mb === gruberman [~gruberman@h9n2fls35o294.telia.com] has joined #ubuntu-devel [03:32] rookery scott% grep -c "to file po/" failed.txt [03:32] 1277 [03:32] ugh [03:33] in fact [03:33] rookery scott% egrep -c "to file (debian/)?po/" failed.txt [03:33] 1987 [03:33] rookery scott% egrep -c "to file debian/(changelog|control)" failed.txt [03:33] 582 === pitti [~martin@195.227.105.180] has joined #ubuntu-devel === Mitario [~michiel@62.58.176.206] has joined #ubuntu-devel [03:55] lo everyone === mvo_ [~egon@suprimo-131.ping.de] has joined #ubuntu-devel === plovs [~plovs@62.84.21.44] has joined #ubuntu-devel === nmf [~nmf@213.30.75.8] has joined #ubuntu-devel === T-Bone [~varenet@shirka.esiee.fr] has joined #ubuntu-devel [04:21] mako: ping === plovs [~plovs@62.84.21.44] has joined #ubuntu-devel [04:26] Keybuk: how bad are the conflicts, I wonder. === sap [~setup@212.13.209.105] has joined #ubuntu-devel [04:27] I'm not able to make the Hoary meeting [04:27] Is there anyone else who would like to raise laptop stuff? [04:29] Hi all [04:29] I have a question I already made on the devel ml but got no (usefull) response, so... [04:29] mjg59: can you dump your "stuff" that you want to raise on a wiki page? [04:29] Are there any plans to make packages like tomcat available? [04:30] Guys, does anybody have a PCMCIA card which provides an IDE interface? === lamont does somewhere [04:30] pitti: Yeah [04:30] well, some/most CF cards do that [04:30] pitti: Well, PCMCIA->CF adapter === sivang [~dannyh@80.179.93.130.forward.012.net.il] has joined #ubuntu-devel [04:30] Presents as IDE [04:30] and you can get "dumb" CF -> PCMCIA adapters [04:31] dammit :) [04:31] what mjg said [04:31] mjg59: quite happy to, yeah [04:31] mjg59: I'm currently debugging #2265, and I need people with such cards to execute two commands [04:31] the problem is how to reliably tell PCMCIA and "normal" IDE adapters apart; hal's current method is broken [04:32] Oh, and I have a PCMCIA CD drive, too [04:32] I don't have a laptop with PCMCIA running Ubuntu, though :) [04:32] I need the output of 'cat /var/lib/pcmcia/stab' and 'cardctl ident' [04:33] mjg59: I don't think that Ubuntu matters, Debian or another Linux is probably okay, too [04:33] mjg59: but it should be kernel 2.6, though [04:33] mjg59: what sort of laptop stuff do you want me to represent? [04:33] hal upstream wants to collect the output of these commands to find a better way [04:34] lamont: can you please send me the output of these commands for your card as well? [04:34] nmf: we will soon develop a policy for adding user contributed packages to our Universe [04:35] nmf: by now you can try to get packages from Debian unstable, or if unstable does not have it, from www.apt-get.org [04:35] nmf: (that is, most unstable packages shoudl already be in our universe) [04:36] daniels: We need to figure out what sort of suspend work to concentrate on - StD is easier but less satisfying, StR is going to need more effort but /ought/ to be possible. We'd need lots of hardware testing, though. [04:36] mjg59: *craploads* [04:37] pitti: I have no /var/lib/pcmcia/stab === vuntz [~vuntz@fennas.vuntz.net] has joined #ubuntu-devel [04:37] (Or, indeed, a /var/lib/pcmcia) [04:37] mjg59: hmm, and cardctl ident? [04:37] Socket 0: [04:37] product info: SAMSUNG, Rev 1.18.4, [04:37] manfid: 0x00ce, 0x0000 [04:37] function: 4 (fixed disk) [04:38] mjg59: thanks! [04:38] pitti: yeahh, I know I can get them from debian, I'm using them now [04:38] nmf: the package is most likely not in universe because we don't have Java at the moment; but we will find a solution for this [04:38] pitti: the problem is packages like tomcat depend on not-free packages (sun JDK) to build [04:39] yes, although we should try and work out a way to build them with gcj and see if they work [04:39] this needs a test suite, and tomcat didn't have one last i looked [04:39] cocoon would be phat to have [04:39] pitti: ok, if there's anyway I can help to speed up the process, let me know === rburton [~ross@84.12.22.159] has joined #ubuntu-devel [04:40] thom: (nor last I looked, short of 'it does stuff and works' or 'it crashes') [04:40] rburton: hey dude [04:40] daniels: aye [04:40] hi daniels [04:40] afternoon ross [04:40] thom: I think most packages in debian that can work with free JVMs/JDKs already have the correct dependencies [04:40] thom: afternoon [04:40] hey ross [04:41] hey fabbione [04:41] good afternoon all [04:41] mvo_, synaptic --set-selections isn't doing the right thing for me [04:42] i clearly tell it "gaim\tremove" and it doesn't do it === mvo_ is having a look [04:42] aaah [04:42] ignore me === rburton fixes stupid bug [04:43] rburton: gaim delete ... === mvo_ blushes [04:43] hey thom, any idea how to solve this one.. dbus-1 starts networkmanager before hald [04:44] nmf: wrt tomcat: [04:44] multiverse/web/tomcat4_4.1.30-6: Dep-Wait by buildd+mcmurdo [optional:uncompiled] [04:44] Dependencies: j2sdk1.4 [04:45] so warty only has source. :-( [04:45] pitti: booting now [04:45] rburton: it's either "delete" or "uninstall" [04:46] mvo_, aah. that would stop the run i was about to do. whats the difference? [04:46] pitti: 4MB pcmcia card, and CF-adapter, fwiw [04:46] lamont: does /var/lib/pcmcia exist on your box? [04:46] rburton: there is none. it's just for conenience [04:46] thom: should /etc/dbus/event.d have numbered scripts perhaps? [04:46] mvo_, k [04:46] rburton: I can add "purge" if you need it [04:47] mvo_, could be useful, but i may just use remove [04:47] pitti: no [04:47] uninstall even === lamont_r [~lamont@rover3.mmjgroup.com] has joined #ubuntu-devel [04:47] lamont: hm; /var/cache/pcmcia maybe? [04:48] lamont: find /var -name stab ? [04:48] /var/run/stab... === lamont upgrades [04:48] wohoo [04:49] mjg59: does /var/run/stab exist? Does it look as if it has sth to do with PCMCIA? [04:49] anyone want to tell me why apt wants to upgrade gnome-cpufreq-applet over and over... [04:49] mvo_, is therea way to force the terminal window to close when it is finished? [04:49] Preparing to replace gnome-cpufreq-applet 0.1.3-1 (using .../gnome-cpufreq-applet_0.1.3-1_i386.deb) ... [04:49] le huh? [04:49] lamont: can you please /msg me this file and the output of 'cardctl ident'? [04:50] rburton: it's a bug if it does not close with "--non-interactive". I'll fix this [04:50] lamont: maybe version +0.0001 and dpkg rounds to two digits? :-) [04:50] pitti: Nope [04:50] pitti: (As in, it doesn't exist) [04:50] mvo_, but app-install just removed and installed gaim for me. synaptic's cli stuff is useful, thanks [04:50] mjg59: hmm, thanks anyway [04:51] Hmm. That's odd - I plug the card in, hal-device-manager briefly shows two partitions, and then they go away again === mvo_ would love to try rburtons app-install [04:51] pitti: Oh, hang on - it's suddenly appeared [04:51] Socket 0: ATA/IDE Fixed Disk [04:51] 0 ide ide-cs 0 hde 33 0 [04:52] lamont: that's the problem I was talking about, it depends on Sun JDK so it never gets built. [04:52] nmf: right [04:52] Anyone knows how contrib packages build is handled on debian? [04:53] tseng: yes [04:53] nmf: It's not autobuilt [04:53] nmf: buildd's try, if the non-free stuff is there, then maybe it works. [04:53] mjg59: thanks [04:53] otherwise, some random d-d uploads the binaries [04:53] just spoke to (one of) the maintainers about it [04:53] wonderful. [04:53] networkmanager, or dbus? [04:53] daniels: dbus-1 [04:54] mjg59: hppa has 88 contrib/ entries in w-b... contrib is autobuilt [04:54] just not with great success... [04:54] Oh, it is? Ha. [04:54] thom: the other maintainer sucks [04:54] non-free, OTOH, is definitely not in w-b [04:57] heh. [05:03] hrm... SMC card seems to "just work". Linksys card not so fortunate. [05:04] new linksys cards tend to be broadcom chips [05:04] while smc goes with prismX [05:04] netgear card happy too [05:05] tseng: WPC11 ver 4 802-11B wireless [05:05] not exactly "new" :-) [05:05] hmm thats an 11b even [05:06] survey says prism2, *should* work [05:06] yeah - I need to actually get my AP working, you see... [05:10] tseng: btw, url? === Mitario [~michiel@sikkes.xs4all.nl] has joined #ubuntu-devel [05:14] http://www.kismetwireless.net/cards.shtml === Kamion fixes up wiki/InstallFromKnoppixHowto lots [05:16] still recommends base-config in a chroot, which is kinda suboptimal, but it'll do for now [05:16] hello everyone! [05:16] hi Mitario [05:17] mvo_, seen my updates on upgrade-notifier? [05:18] Mitario: yes. I checked in a small fix also :) [05:19] cool :) [05:19] is the upgrade-center available for download somewhere? === Mitario svn ups [05:19] mvo_, umm, well I did the rewrite in python [05:19] with python-apt [05:19] I wonder if it is worth the efford to put the build-in window back [05:19] but I have not yet made a package or buildsystem for it [05:19] true [05:19] IMO it isn't nescesairy [05:20] ok [05:20] does anyone here know a neet buildsystem for python apps? [05:23] mvo_, i can send you the source I have now though [05:23] Mitario: yes, please do :) === Mitario makes a tarball [05:24] oh, I was a bit stuck on setting selections in synaptic though, but i'll ask that later [05:26] mvo_, see http://luon.net/~michiels/ubuntu/update-manager.tar.gz [05:28] Mitario: looks very nice! [05:29] :) [05:30] i best refactor app-install and make a tarball for people to poke at too [05:30] app-install was that neet app you showed us some days ago? [05:30] aye [05:30] nice :) [05:31] it just removed and installed gaim [05:31] (via synaptic) [05:31] cool :) [05:31] oh, can you show me your synaptic --set-selection method? === mvo_ is very happy about all that progress === Mitario played with it a bit, couldn't get it to work [05:31] Mitario, okay to /query? [05:32] sure [05:35] mvo_, anyways, seen the --with-package-manager option? [05:39] Mitario: yes! [05:39] ok, i did that so distributions/maintainers can use the package manager of choice [05:40] so for ubuntu (or at least my packages :) I would use --with-pkg-manager=/usr/bin/update-manager or something [05:42] oh, of the 326, only 248 actually have newer versions [05:45] elmo: don't count xfree86. we are not going to merge it from sid [05:45] elmo: we will kill it as soon as we have x.org [05:47] elmo: and many can be just re-synced. [05:51] T-Bone: hey there.. give me a couple minutes [05:51] mako: ok sure! just wanted to make sure you were alive ;) [05:53] fabbione: http://bugs.gentoo.org/show_bug.cgi?id=49038 [05:54] daniels: i didn't build Xx86rush yet [05:54] i can test it on the fly [05:55] i was just searching for a simple lib to trash before the meeting :-) [05:55] daniels: if it doesn't contain *gl* anywhere in the name it's ok :) [05:56] heh [05:57] fabbione: anyway, that bug and the attached fd.o one are pretty much the canonical fixes [05:57] look like they're fixed upstream anyway [05:57] oh yeah, dlloader works out of the box [05:58] if you want to go for a triple-whammy we can KILL ELFLOADER with EXTREME PREJUDICE [05:58] dlloader rocks [05:58] it sure does [05:58] even better than dlloader is the loader in debrix [05:59] daniels@nanasawa:~/x/debrix/debrix/hw/xorg/loader% wc -l *.[ch] | tail -1 [05:59] 897 total [05:59] daniels@nanasawa:~/x/xorg/xc/programs/Xserver/hw/xfree86/loader% wc -l *.[ch] | tail -1 [05:59] 13417 total === x4m [~max@215.164-136-217.adsl.skynet.be] has joined #ubuntu-devel [06:02] daniels: one thing at a time [06:03] let's get X.org out of my harddisks first [06:03] heh [06:04] oh daniels, do you know anything about tv out on ati? [06:04] daniels: Xx86rush compiles perfectly here [06:04] they are on crack [06:05] fabbione: merged into xorg, then [06:05] daniels: could be [06:05] tseng: 'don't' [06:05] tseng: how recent? [06:06] i have an radeon mobility, few months old and a radeon aiw 9200 === inklingx [~inklingx@u212-239-167-206.adsl.pi.be] has joined #ubuntu-devel [06:06] should be fine with the mobility, likely sol with the 9200 [06:06] just wondering if xorg will do anything for me, ati-drivers are the suck [06:06] for the mobility, google for atitvout === robtaylor [~robtaylor@217.204.121.82] has joined #ubuntu-devel === srbaker_ [~srbaker@blk-224-143-227.eastlink.ca] has joined #ubuntu-devel === __daniel [~daniel@td9091b68.pool.terralink.de] has joined #ubuntu-devel [06:37] <__daniel> hai [06:57] heya [06:59] <__daniel> brb === maskie [~maskie@196-30-111-250.uudial.uunet.co.za] has joined #ubuntu-devel === pasc [pasc@gandalf.redellipse.net] has joined #ubuntu-devel [07:05] Kamion/Thom: please rename warty-rc2-live-i386.iso to warty-release-live-i386.iso. Ditto for .torrent === __daniel [~daniel@td9091b68.pool.terralink.de] has joined #ubuntu-devel === sjoerd_ [~sjoerd@fire.ipv6.luon.net] has joined #ubuntu-devel [07:12] kamion/thom/whoever: and please let me know, when it's done so I can upload it to the CD pressers [07:12] not me, this is all kamio [07:13] thom: you need to do torrents :) [07:13] I'll look, may not be done until the meeting's over [07:15] thom: isn't it great how much time tab completion saves you? :P [07:16] lamont: (mdz's confirmed?) [07:16] elmo: confirm the md5sum with lamont [07:16] his latest bits are the ones which should go out === lamont md5sums [07:19] dac84a3abf5a1a104d768d569a62579e warty-rc2-live-i386.iso [07:19] dac84a3abf5a1a104d768d569a62579e warty-live-i386-20041022-04.iso [07:19] that should match rookery [07:19] should match releases.u.c === ctalkep [~ctalkep@212.21.138.21] has joined #ubuntu-devel [07:23] hi guys [07:23] is daf here? [07:24] ctalkep: he's on the channel [07:24] well... [07:25] ctalkep: hi [07:27] daf, , hi [07:27] daf, been looking for you for several days [07:28] I'm not always around on weekends [07:28] daf, wanted to contact you about the translation project, since you are in charge [07:29] i read at the list that the installer is about to be the most important part to be translated, so is that where we need to start? [07:33] I recommend you choose something you would personally like to see better translated [07:34] also, it's a good idea to take into account what experience you have when you start translating [07:34] some programs are easier to translate than others [07:34] and you have an advantage if you're translating a piece of software you're familiar with [07:35] i see [07:36] daf, so there is no structured schedule ? [07:37] no [07:37] it's a free-for-all [07:37] daf, sorry, got to leave for a while, be right back, wanted to ask you and on the web site translation [07:37] ok, see you later === lucas_ [~lucas@ca-grenoble-1-139.w80-8.abo.wanadoo.fr] has joined #ubuntu-devel [08:07] hi [08:08] thom: updated live CDs, please check torrents [08:08] Kamion: 'k [08:13] how are you making your live cds? [08:13] manually, or fai, or other automation? [08:13] manual, aiui [08:15] well [08:15] morphix-manual [08:16] lamont/kamion: torrents torrentified [08:16] aj: morphix-mmaker [08:17] invoked manually atm [08:19] lamont: I assume you or somebody will figure out who's sending the release announcement [08:19] canonical URL is http://releases.ubuntu.com/warty/warty-release-live-i386.iso BTW [08:21] Kamion: yes, once the meeting is over, I'll work with mdz/jdub/whoever and make it happen === sivang is now known as sivang_away === sivang_away is now known as sivang [09:17] daf, you still here? [09:19] yep [09:22] daf, so, do i take the files for translation from the debian repository? [09:23] that depends on what you want to translate [09:25] daf, i was hoping there was someone out there to tell me what should be done,:), but since i'm on my own i don't generally care [09:27] daf, and i am so impressed with ubuntu, that i wanted to begin with it's translation [09:29] well, it's really up to you [09:29] find something you like using which needs work and work on it [09:29] have you already managed to get Ubuntu working in your language? [09:31] haven't started yet [09:32] wanted to first get a grip of the situation [09:33] i guess i'll just start with that [09:33] then what about the web site translation? [09:34] there are a lot of people here working with linux, yet few of them are comfortable with english [09:34] web site translation is not possible at the moment [09:34] i think it would bea great advantage, since there are only few online sources of information on linux/unix now [09:34] we're working on that [09:34] i see [09:35] I'll make an announcement as soon as it's ready [09:35] running Ubuntu in your language would be a good start [09:35] i will begin with that then === mvo__ [~egon@suprimo-131.ping.de] has joined #ubuntu-devel === gro [~gro@u212-239-167-206.adsl.pi.be] has joined #ubuntu-devel === nmf [~nmf@adsl-b4-170-137.telepac.pt] has joined #ubuntu-devel === ferdo [~ferdo@212.21.138.61] has joined #ubuntu-devel === ferdo [~ferdo@212.21.138.61] has joined #ubuntu-devel === dany [~dany@212.21.138.61] has joined #ubuntu-devel [10:18] to beshe na purvata vuzmojna opcia be === T-Bone [~varenet@T-Bone.developer.debian] has joined #ubuntu-devel === x4m [~max@215.164-136-217.adsl.skynet.be] has joined #ubuntu-devel [10:49] So we can now upload our stuff into Hoary? Or shall we wait for the sync first? === jbailey [~jbailey@dragonfly.fundserv.com] has joined #ubuntu-devel === T-Bone is now known as T-Dinner [11:11] is there anyway to find out what was said in the meeting today? [11:11] like a log or something [11:13] pitti: if you have packages to upload which do not interefere with the sync, I think it is fine. elmo can confirm [11:13] hornbeck: there will be a transcript and summary posted to the list [11:13] hornbeck: if you want something now, I can send you a copy of my scrollback [11:13] mdz: I already synced some packages (hal and gvm), so it would actually ease the merge [11:14] hornbeck: I have a copy here [11:14] hmm, I'm wonderwing if it'd be usefull for ubuntu/me if i'd sign up for wannebe maintainer :) [11:14] hornbeck: http://www.piware.de/ubuntu-meeting-20041025.txt [11:15] mdz: but if there is already an automated merging process, I can defer uploading [11:16] pitti: should be fine, send mail to Keybuk to notify him that you have done it [11:16] mdz: okay, I will CC elmo as well [11:16] eh, I think we should try and sync where possible? [11:17] rather than do uploads? [11:17] pitti, mdz: thanks I am reading pitti's copy [11:17] elmo: I packaged a completely new hal version which just arrived at experimental [11:17] elmo: I just don't know what is easier: let you finish the automatic merging and upload afterwards, or upload immediately [11:18] mdz: a bounty for a python port of yelp? [11:18] elmo: but we changed so many things in hal and gvm that even manual merge was a PITA; I think it isn't possible automatically [11:19] pitti: what am I saying is, if ubuntu-version will == debian-version, I think we should sync [11:19] elmo: not really [11:19] elmo: I took the new Debian version, and redid the Ubuntu modifications as clean patches [11:20] elmo: same with g-v-m [11:20] elmo: I will need to upload anyway, but I want to do it at a time when it does not interfere with merging [11:22] then go ahead and upload [11:22] tho the cron jobs are disabled right now [11:22] only 800 or so more packages to sync in universe [11:22] elmo: okay, if now is a good time === User496 [~User@207.102.22.2] has joined #ubuntu-devel === __daniel [~daniel@td9091b7a.pool.terralink.de] has joined #ubuntu-devel === User911 [~User@207.102.22.2] has joined #ubuntu-devel [11:26] mdz: I would like to do the Ubuntu in a Nutshell, if there are no other takers [11:29] hornbeck: that'll be a more-than-one-man project, probably involving some canonical contracts too [11:30] jdub: I would really like to work on it if I can [11:30] I really want to put something out that is not just internet based [11:35] well I am off to work again [11:42] hornbeck: i like the idea, who is the publisher of the "in a nutshell" series? [11:42] oreilly and associates [11:43] tseng: thanks [11:43] sabdfl oreilly [11:43] we could say [11:43] "ubuntu in a clamshell" :D [11:43] "ubuntu in a akorn" [11:43] hornbeck: go ahead and approach them if you like, with my support [11:43] sabdfl: I will try :-) [11:44] If I go about this, I don't think I will get much real docs done [11:44] there are several other publishers with an eye towards open source as well [11:44] but it will be a major doc in and of itself [11:44] if that is the route you are going [11:44] well it would be nice to get a Ubuntu Book out there [11:44] I like books [11:44] newriders, no starch press === hornbeck works in a library [11:45] no starch is through orielly is it not? [11:45] there is some sort of partnership i believe [11:45] yeah that is what I thought [11:45] but no starch is at least externally its own company [11:46] sabdfl: the nutshell series is more just facts [11:46] I could approach no starch about doing a "Ubuntu book" [11:46] or even Oreilly [11:46] http://www.nostarch.com/about.htm [11:47] oreilly is a distributer for no starch [11:47] what do you guys think? === tseng cares less about who prints it than the content [11:47] night [11:47] bye pitti [11:47] well, lets discuss later I have to be at work in 10 minutes [11:47] ok [11:47] content = #1 [11:48] sabdfl, tseng: I think I would want to work with all the dev's on this [11:48] to make it very good [11:48] yes. [11:48] but discussion later [11:48] bye [11:48] hornbeck: to get it done for hoary it may be better to start with a tighter format === rjb [~arjaybe@207.102.22.2] has joined #ubuntu-devel [11:49] so, I can upload merges to hoary now? === Kamion cracks his knuckles [11:49] sabdfl: I will put together a outline and mail to you [11:50] yeah, if you want [11:50] they won't be built/mirrored out for a bit tho [11:51] hornbeck: i'm already a bottleneck, can you figure this out within the doc team? ping me on irc if you need any specific commitment [11:51] sabdfl: I will work something out and let you guys know [11:51] elmo: if we can avoid mirroring for a while, it might be worthwhile === hornbeck is off to work [11:51] maybe publish hoary somewhere where it can break [11:52] especially if we are going to be uploading new gcc etc [11:52] once the toolchain is in, we can rebuild [11:52] or is that toenail-smoking? [11:52] sounds like a plan to me [11:53] or maybe just publish source only [11:55] err, I thought we discussed this? [11:55] +in the meeting [11:56] this is why, I'd held off on hoary so long, because people keep telling me to do entirely conflicting things with hoary :-/ [11:57] if we were going to update the toolchain, we should have done that _before_ importing new versions of everything [11:57] I can't not mirror hoary, without essentially forking the archive infrastructure. I can do that, if you want but I need to know now. [11:58] The only other thing we could do is restrict access to the Packages/Sources files but that certainly won't allow us to en-masse rebuild [11:58] or option c) we just put hoary out there, which is what I thought was the plan, and what I had started on doing [11:59] by "update the toolchain" you mean switch to gcc3.4/4.0 by default? [11:59] that sounds like bong ...