[00:01] the new theme looks pretty cool [00:01] * jdong backports it [00:02] Because running Hardy is pointless if you can just backport the lot? [00:02] StevenK: not nearly enough is backported to draw that comparison :) [00:04] jdong: Heh. [00:28] <_MMA_> jdong: I wouldnt backport it yet as it still needs some work. [00:29] _MMA_: only personally for now [00:29] I just wanted it on my gutsy machine [00:29] call it greed [00:29] <_MMA_> Sure. [00:29] * StevenK is content to wait. [00:36] and what is about my bug report? [00:36] well be fix it? === jelmer_ is now known as jelmer [01:17] Should a package remove obsolete config files when it get's upgraded? They seem to stay in /etc and get listed as "obsolete" in "dpkg -s". [01:18] oops.. I've meant to ask in #ubuntu-motu.. === greeneggsnospam is now known as jsgotangco === kitterma is now known as ScottK2 [03:30] * Hobbsee waves [03:30] can some admin ban the spammers at http://brainstorm.ubuntu.com/search?ordering=new [03:30] they are screwing up the site [03:31] Heya Hobbsee! [03:31] hey RAOF! [03:31] Heya Hobbsee [03:31] * Hobbsee deals with gio. [03:31] * TheMuso waves to RAOF and Hobbsee. [03:31] With a hammer? [03:32] And a howdie to you too, TheMuso. [03:32] Hmmm. Spammers messing up a site where users come up with work they think we should do... Bug or feature? [03:32] Heh. [03:33] I was thinking a bit about that. [03:33] It should have been tied to Launchpad User IDs. [03:33] IMO [03:33] Oddly enough Ubuntu is the distro where I have to agree to play nice and Debian is the distro where I have to agree to care about the users. [03:37] hi TheMuso [03:37] TheMuso: I was advised that it will be soon. [03:37] RAOF: no, not really. trouble is, i have conflicting dates on when my rego was *actually* cancelled. [03:38] so they want a fax, etc, etc, etc. nyah nyah nyah. [03:39] Hobbsee: Aaaah. gio, not gio :) [03:39] RAOF: which were you thinking of? [03:39] GIO, the new gnome IO subsystem. [03:39] * Fujitsu initially thought of the GVFS-related GIO. [03:39] That one. [03:39] Since we are, after all, in #ubuntu-devel :) [03:40] Not #cars-are-expensive [03:40] Haha [03:40] RAOF: oh. [03:40] :P [03:40] no, this is #Hobbsee-hates-making-phone-calls-that-are-overdue [04:06] blink? [04:06] Netsplit? [04:07] oh [04:07] was about to say - i shouldnt' be timing out from there [04:08] hurrah. done my 5 today. [04:08] * mpt applauds [04:08] should have got the spangly new code to do with it, though [04:13] ScottK: nuked one of yours (can't reproduce) [04:16] Hmmm. Looks for bugmail. [04:17] ScottK: get-build-deps on !bash [04:17] Ah [04:17] I think he fixed that one. [04:18] It's OK. I filed 10 bugs on ubuntu-dev-tools in one day. Even if one falls out, I'm still well over 5. [04:18] heh [04:18] heh [04:18] I thought it was fix 5, not file 5 [04:19] if you file 5, you have to fix 10. [04:19] Heh [04:19] If people would stop uploading without thinking, I'd be fixing, not filing. [04:19] bah. bzr curled up it's toes and died. [04:19] Oh? [04:19] bzr: ERROR: Cannot lock LockDir(lp--1217675668:///~5-a-day/5-a-day-data/main/.bzr/branchlock): Transport operation not possible: readonly transport [04:19] bzr failed with error code 768 [04:19] Delete the checkout and re-checkout, I think [04:19] Hobbsee: That looks like you're trying to push to an http branch? [04:20] My favorite ubuntu-dev-tools bug was that the AUTHORS file was executable. [04:20] RAOF: no idea. i'm just using whta's in haryd [04:20] ScottK2: What did it do when you executed it? :-P [04:20] It wouldn't have done anything much, but I didn't execute it. [04:20] It's just symptomatic of a lack of care. [04:20] How does that happen? [04:21] I often chmod +x random files. Surely you do too? :? [04:22] I'm on a different tack now anyway. I'm going to see how well I survive without ever logging into LP. [04:22] I should learn the mail-based LP interface, yeah. [04:23] Well I'm seriously looking at Lenny right now. I've pretty well had it up to my eyeballs with LP for a while. [04:24] oh, meh. it doesn't create ~/.5-a-day-data, so dies [04:24] and then still dies [04:25] ScottK2: so be producive, and find ways of diong ubuntu without LP [04:26] Hobbsee: I'm planning on it. I've been looking into actually teaching bughelper to report Ubuntu bugs instead of just whining to a mailing list. [04:26] that'd be handy [04:27] I think all the relevant bits of code exist, it's just a matter of pulling them together. [04:30] The other thing is that Lenny with KDE 3.5.9 may be a better desktop for my than the Ibex. [04:31] I'm still waiting for a reassuring answer to my comment on kubuntu-devel ML about us being over a year away from a stable/reliable kde4pim. [04:31] * ScottK2 goes back to filing klamav bugs with upstream. [04:33] Wow, over a year? How does that happen? [04:33] It's like this... [04:33] KDE4.0 - No kdepim for kde4. [04:33] KDE4.1 - Initial kdepim port to qt4 [04:33] kubuntu will carry kde 3.5.9 [04:34] KDE4.2 - Initial integration with the new Akdondi data engine back end [04:34] KDE4.3 - I figure this is the first shot at a stable/reliable one. [04:34] Releases happen every 6 months. [04:35] Hobbsee: True, but how much attention is it going to get? [04:35] Unless the Debian KDE people pull a miracle out of their hats Lenny will release with 3.5.9 and KDE4 will stay in experimental. [04:36] but don't we get whatever work Debian does? [04:37] Probably. Assuming someone does the merges. [04:37] I'm not saying which way I'm going. I've not decided. It's just that I'm looking into it. [04:39] speaking of KDE, any kubuntu folks want to rebuild kdebase-workspace for libxklavier12? [04:39] ScottK2: how much is gnome going to get? [04:39] What is from my perspective declining usability in our main tool set and developers that don't care is my main concern. [04:39] slangasek: i;ll do it [04:39] Hobbsee: I've no idea. [04:40] Hobbsee: cheers; fyi, there's a dep and a build-dep that each need changed [04:41] come on, pull faster tahn 700 kbps. [04:41] er, kBps [04:41] The straw that broke the camel's back today was finding out after a long disucssion about a bug that it was known before the last LP release and neither documented nor fixed. [04:42] that happesn for a lot of bugs. [04:42] just watch a few that get marked for critical cherry picking, after sitting there for 3 months. i find that sadly amusing. [04:42] You would think if being able to include attachments in new email bug reports was being announced as a major feature it would actually work (or at least have it's limitations documented) [04:43] not really. i'm afraid i've seen a little too much of launchpad to expect such things. [04:43] especially after seeing ppa release without a delete function, for months. [04:43] Everyone over there claims it's getting better, but I don't see it. [04:44] it is. it's getting more code, more features, and more bugs. [04:44] I'm also seeing more and more replies along the lines of "we know it's not ideal, but that's what sabdfl said he wanted. nothing to do about it." [04:45] * Hobbsee suspects a lot of companies operate around "it's because the boss said so" [04:45] is a ubuntu qa password different than lp or the wiki? [04:45] Yes. [04:45] fun [04:45] that being said, with all their time, and all their features, and all their new people, i would have hoped for better QA - which, at least for some parts, hasn't happened. [04:45] Hobbsee: But very few FOSS projects do. [04:45] ScottK2: and launchpad is not FOSS. [04:46] Exactly. [04:46] So I'm not going to play free QA for them any more. [04:47] i'm actually not sure how much their "free QA", as it were, is helping them, seeing as they appear to get important bugs filed, then either ignore, or not understand the importance to people in the real world (or class them as not important) [04:47] and yet do very little with them [04:48] Hobbsee: actually, im not certain a delete feature is a good idea. technically the gpl requires you to provide the source code to people for some number of months after distribution, upon request [04:48] Yes. Even more encouragement to quit. [04:49] pwnguin: ah yes - now i'm waiting for that bug to hit critical status, but it doesn't mention that it's violating the gpl, and kiko has already said "oh please...", so i'm guessing it will take a few more months, and a few more people going "uh...." [04:49] pwnguin: feel free to comment about the violation on the bug. [04:49] Hobbsee: I'm not sure that's a very fair assessment [04:49] pwnguin: but it's nice to actually have control over the binaries that are ther.e [04:49] there's certainly a lot of improvements that need to be made [04:49] but there are a lot of bugs getting fixed [04:50] and we certainly don't do much better with our bugs [04:50] LaserJock: how do you disagree? i'll freely admit, iv'e seen more improvements in some areas than others, and i suspect i'm seeing some of the worse ends of LP, as most outside people don't see them [04:50] yeah well. i have various things to say about that too. [04:50] https://bugs.edge.launchpad.net/ubuntu/+bug/196859 <-- please notify admin asap, site is under spam attack [04:50] Launchpad bug 196859 in ubuntu "Brainstorm is susceptible to spam" [Undecided,Invalid] [04:50] LP has a lot of functionality and is quite usable, not perfect, but it gets the job done [04:51] ionstorm: they're not awake. [04:51] LaserJock: But it's getting worse not better. [04:51] ScottK2: I wouldn't say that [04:51] ionstorm: oh, that admin [04:52] there are some regressions here and there for sure [04:52] Hobbsee, we need to wake them [04:52] no doubt [04:52] view http://brainstorm.ubuntu.com/search?ordering=new [04:52] Personally I think the web site U/I peaked just before the beta and it's been getting more complex and less usable since. [04:52] ionstorm: good luck in finding people who work for canonical at this time of day [04:52] but they've also made progress [04:52] they are massively spam attacking them [04:52] StevenK: can you call someone? [04:52] well canonical needs to be on irc [04:52] LaserJock: Sure. It's some good and some bad. IMO the balance is to the bad. [04:53] stgraber: ping? [04:53] ionstorm: someone needs to not sleep, yeah [04:53] hrm. i wonder if i can delete thru here [04:54] no, i can only edit [04:54] how useful [04:56] yes! it's even worse than LP, in that you can't clear a dupe! [04:57] ionstorm: you'll need someone on https://edge.launchpad.net/~ubuntu-qa-website-devel [04:57] * mpt works for Canonical [04:58] mpt: But you live on a sane side of the world at the moment. [04:58] mpt: oh, i forgot to specify "who could do something about it" [04:58] So you're not someone that we're looking for, as they're logically going to be in the wrong timezone when you need them. [04:58] My side of the world is *always* sane :-P [04:58] unless you'r eoffering calling [04:59] Call someone about what? [04:59] Fujitsu: you mean everyone doesn't operate on UK time? [04:59] mpt: recent backscroll is your friend :) [04:59] Hobbsee: What? Impossible! [04:59] and it relaly helps if i actually increment the changelog [04:59] Oh. [04:59] oh, hee hee [05:00] * ScottK2 finds it help if I upload to Hardy instead of Gutsy (last night anyway). [05:00] hah [05:00] Oh that bastard. [05:00] yeah, that too. i got that one right htough [05:00] Erm, sorry. [05:00] But, the spammer rejecting the bug. [05:00] WTF. [05:00] heh [05:00] Bug was invalid against Ubuntu anyway. [05:01] I can't access the bug report, I get an offline message [05:01] Hm, edge looks dead. [05:01] So do I. [05:01] Try !ege. [05:01] *!edge [05:01] It worked for me a couple of minutes ago :( [05:01] !edge must work, as it redirects me... [05:01] ouch [05:02] This is an example of why I don't think edge should be a separate hostname [05:02] Doesn't the code update run around now? [05:02] It's back now. [05:02] That must have been it. [05:02] BTW, I do find mpt is an exception to my broad generalizations I was making a few minutes ago. [05:03] Well, I confirmed the bug report, but there's not much more I can do at this time of day [05:05] We have yet to learn the lesson of planning for spam before releasing something publicly [05:05] Mhm. [05:05] Hopefully this will be an object lesson [05:05] Hopefully. [05:05] it's too bad it didn't even make it like more than 1 day [05:05] Having some kind of coverage of multiple timezones would be good too. [05:05] Having admins sitting on one side of the world is ++ungood. [05:06] Fujitsu: It's not that. It's the combination of sitting on one side of the world AND letting them sleep that's the problem. [05:06] LaserJock: Slashdot and Digg do that to websites. [05:06] ScottK2: Good point. Canonical needs to sponsor caffeine IVs. [05:06] mpt: i live in hope. [05:06] Hobbsee, I live in a settlement that is actually called Hope [05:07] mpt: it does actually have admins in most timezones, so it is getting better. [05:07] mpt: nevertheless, with a GUI for delete, and a gui for 'mark as spam', that's not overly helpful [05:07] Gaaaah. [05:07] * Fujitsu stabs repeatedly. [05:08] He rejected it. [05:08] * Fujitsu kills. [05:08] hah [05:08] now now Fujitsu. you can't kill people, otherwise people inside and otuside of canonical will whinge about you being heavy handed. [05:08] Apparently this is after "months of planning and development" what was come up with (reading planet). [05:08] ScottK, Launchpad has its problems, but if you say that the pre-1.0 UI was better than the post-1.0 UI, people may have difficulty taking your other issues seriously [05:09] Even Launchpad has beta test phases, whereas this came from nowhere! [05:09] mpt: That's how I find it, but I'm old and I like text. [05:09] The old U/I also worked via my smartphone. New one doesn't. [05:09] * Hobbsee shudders [05:09] I also clicked on the wrong thing a lot less before. [05:09] ScottK2: you ran launchpad over your *smartphone*???? [05:10] I agree the newer one is prettier. [05:10] Hobbsee: I have an all you can use data plan. [05:10] you really are a masochist. how long did the pageloads take? [05:10] Long time. [05:10] i'll bet [05:10] There aren't any LP admins on this side of the world either, are there? [05:10] They take a lot longer now and I can't change stuff, so I quit doing it. [05:10] Yay for status wars. [05:11] Fujitsu: nope [05:11] Fujitsu: You could try to get his account cancelled, but you'll have to warn him, take him in front of some council somewhere, warn him again, and then maybe. [05:11] ScottK2: And that requires an admin. [05:12] He's just done it for the third time. [05:12] Fujitsu: By the time you get the other stuff done, the admin will be awake. [05:12] ScottK2: no you don't. [05:12] ScottK2: for actual spam, they'll nuke it. sometimes they'll try an email first. [05:13] * ScottK2 confesses to engaging in hyperbole and sarcasm. [05:13] really?? [05:13] ;-) [05:14] The part about thinkin LP is getting less and less usable was dead serious though. [05:15] We need a system to vote users out, damnit! [05:16] "Survivor: Ubuntu Edition" [05:17] s/Ubuntu/Launchpad/ [05:20] ScottK, does your phone browser try to do Launchpad's multi-column CSS layout? [05:20] mpt: Yes. [05:20] I think [05:20] Let me try it. [05:21] non-Opera/iPhone browsers are often execrably bad at CSS [05:21] It's a Palm Treo 600, so it's not precisely obscure. [05:21] Wouldn't suprise me. [05:21] but a phone that ignores CSS altogether should be much happier with post-1.0 Launchpad than with the -ridden pre-1.0 Launchpad. [05:21] Their included mail client qualifies as excerable. [05:22] The thing that got me was the switch to the current approach for the status/importance/etc section. The old one worked. [05:23] hurrah. finally made all the phonecalls for the day. [05:23] ScottK2: This is what the mail interface is for. [05:24] Fujitsu: This is true. There is an alternative, but it's still a regression from my perspective. [05:25] ScottK, if you can send me a screenshot, that might help a lot [05:25] mpt: I'll see if I can figure out how to get a decent one. [05:27] Actually it's gotten a lot worse than the last time I tried it. On a package page I just get the 'table' of releases. I don't even get the tabs to go to the bugs page. [05:28] ScottK2: Oh, people use the tabs? [05:28] Fujitsu: I don't usually. I usually type in the urls on my computers, but on the phone it's a bit harder. [05:28] Ah, true. [05:29] * lamont uses the tabs [05:29] much easier that remembering the magic hidden URLs for stuff [05:29] Fujitsu: Alternative answer is: No. That's why clicking on the package name in the status section of a bug now takes to you the package page instead of giving you the chance to change the affected package. [05:29] so I don't even bother trying to remember even the simple ones [05:29] And it can't be changed. [05:29] * Hobbsee has aliases [05:30] * ScottK2 doesn't remember them either. The browser knows where I've been before. [05:30] Hobbsee: I have an alias for getting to a source package (which still has /distros in it, iirc), and maybe an lpbug alias [05:30] lamont: pity the url's aren't always shown on the gui [05:30] yeah [05:31] I just type the urls by hand mostly [05:31] * ScottK2 needs to get to bed. Good night all. [05:31] * Fujitsu does the same as LaserJock. [05:31] Night ScottK2. [05:32] * StevenK has an alias for source package, person and bug [06:34] hrm. how do i get firefox-2 to work? [06:38] loading that, i appear to get firefox3, as usual [06:39] is there a separate .desktop? [06:39] i was using the terminal [06:39] ie, sarah@saturn:~% /usr/bin/firefox-2 5:39PM [06:48] Good morning [06:49] Morning pitti [06:50] jdstrand: nut bzr> thanks [06:54] morning pitti [06:54] Good morning [06:54] hey Hobbsee! [06:54] moin StevenK [06:55] bon giorno warp10 [06:55] morgen pitti! [06:55] hiya pitti [07:22] slangasek, according to http://people.ubuntu.com/~ubuntu-archive/cd-build-logs/mythbuntu/hardy/daily-20080229.log the ISO generated, but it looks like some error with the md5sum template that may be out of my control? [07:25] oh nvm, just took a sec to show up on cdimages.ubuntu.com [07:26] ok :) === hunger_t is now known as hunger === \sh_away is now known as \sh [08:10] superm1: ping? [08:10] (nothing urgent) [08:12] pitti: you remember we do that thing in ubuntu-docs where translations that are less than 40% don't get included? I'm just working on updating translations in the gutsy package; can I drop that a bit now that iso space is no longer a consideration? [08:15] hi [08:18] hello seb128 [08:18] hi kagou [08:19] hey dholbach [08:19] good morning [08:19] hi seb128 [08:19] morning seb128 [08:20] hey dholbach [08:20] dholbach: your shiny is borken :( [08:20] hey Hobbsee [08:20] hi kagou, hi Hobbsee [08:20] shiny? [08:20] dholbach: http://pastebin.ca/923001 [08:20] morning dholbach [08:20] hi mdke [08:21] Hobbsee: seems we need to transition it from "all commit to one branch" to "everybody commits to their own branch" [08:21] dholbach: i'm just doing some updates for the gutsy package of ubuntu-docs, with updated translations. Is it appropriate for me to remove the packaging guide from that, or do we just do that in hardy and going forward? [08:21] mdke: it'll make it still quite ugly to read, and CDs are still full, as always [08:21] that'll make statistics a bit hairy, but it'll fix the issues [08:21] dholbach: what's the problem with all committing to one branch? [08:21] mdke: I'd just do it in hardy and keep the changes to gutsy as small as possible [08:21] pitti: are CDs still being regenerated for gutsy? [08:21] also, it doesnt' seem to create the ~/.5-a-day-data [08:21] dholbach: fine, thanks [08:22] Hobbsee: it should run bzr checkout [08:22] at least it did for a lot of other people :) [08:22] dholbach: this is a stock install of it. no changes. [08:23] can you move .5-a-day-data away and try to run add-5-a-day 194631 190050 150577 140612 178903 196786 again? [08:23] and paste the output? [08:26] dholbach: http://rafb.net/p/6xkuJL96.html [08:27] readonly transport? [08:27] do you have the right SSH key in LP? [08:27] yes [08:27] * Hobbsee only has one ssh key. [08:27] hrmhrmhrm [08:28] Have you run "bzr launchpad-login hobbsee"? [08:28] RAOF: nope [08:28] * RAOF presumes that's your lp id. [08:28] dholbach: i get promped for a ssh p/w, and put that in... [08:29] RAOF: same problem after doing so [08:29] There goes my guess :) [08:31] Hobbsee: where's the 'hobbsee' file? [08:31] is it in ~/.5-a-day-data/? [08:31] pitti: steve gave his ack to the dbus update ;-) [08:31] seb128: oh? didn't see it by mail [08:32] sarah@saturn:~/.5-a-day-data% ls -la hobbsee 7:32PM [08:32] -rw-r--r-- 1 sarah sarah 204 2008-02-29 19:28 hobbsee [08:32] dholbach: so, yes [08:32] pitti: he added a comment on the bug some hours ago [08:32] Hobbsee: but the branch seems to be in ~/.5-a-day-data/main ? [08:32] seb128: ah, indeed! /me uploads [08:33] pitti: danke [08:34] ah no, nevermind [08:34] Hobbsee: what happens if you run cd ~/.5-a-day-data; bzr update; bzr commit -m "updated log for 'hobbsee'" ? [08:35] if that makes it work,we should look into splitting up the branch quickly :) [08:35] dholbach: same problem on that commit [08:35] dholbach: you mean bzr doesn't handle big things? [08:36] Hobbsee: no, if lots of people hammer the same branch, you can always get out of sync in a few seconds :) [08:36] dholbach: right, but this has happened multiple times today? :) [08:37] morning pitti, seb128, dholbach and Hobbsee [08:37] hi glatzor [08:37] hi glatzor [08:38] morning glatzor! [08:41] hi glatzor [08:47] Hobbsee: can you pastebin bzr info -v somewhere? [08:48] dholbach: http://rafb.net/p/LKucRO74.html [08:53] Hobbsee: dholbach: as a workaround, hobbsee should be able to "bzr switch sftp://hobbsee@bazaar.launchpad.net/%7E5-a-day/5-a-day-data/main/" [08:54] Hobbsee: https://bugs.launchpad.net/launchpad-bazaar/+bug/196913 [08:54] Launchpad bug 196913 in launchpad-bazaar "Cannot lock LockDir(lp--1218658708:///~5-a-day/5-a-day-data/main/.bzr/branchlock): Transport operation not possible: readonly transport" [Undecided,New] [08:55] hey mvo [08:55] hey dholbach! [08:55] hello mvo! [08:55] mvo: back to germany? [08:56] hey mvo, welcome back [08:56] pitti: yeah, thanks :) [08:56] seb128: yes, arrived last night (later, flight was slightly delayed) [08:56] * seb128 hugs mvo [08:56] dholbach: btw, suggestion: could update-signature do a 'bzr update ~/.5-a-day-data/'? [08:57] dholbach: so that I can call it on all my machines [08:57] * mvo hugs seb128 dholbach pitti [08:57] dholbach: could you namespace update-signate btw? 5aday-update-signature or something? [08:58] pitti: bug 196914 [08:58] Launchpad bug 196914 in five-a-day " dholbach: btw, suggestion: could update-signature do a 'bzr update ~/.5-a-day-data/'?" [Undecided,New] https://launchpad.net/bugs/196914 [08:59] dholbach: wow, do you have an irc2lpbug script? :) [08:59] seb128: bug 196915 [08:59] Launchpad bug 196915 in five-a-day " dholbach: could you namespace update-signate btw? 5aday-update-signature or something?" [Undecided,New] https://launchpad.net/bugs/196915 [08:59] * seb128 hugs dholbach [09:00] pitti: no, I'm just quick today [09:00] no time until end of next week to hack on 5-a-day stuff though [09:00] dholbach: that's like inbox handling, do it now and switch to something else? ;-) [09:00] so if somebody else wants to get started on those bugs.... :) [09:00] seb128: ah, I know why I didn't get bug mail for the dbus FF request, ubuntu-release wasn't sub'ed [09:01] pitti: I did! [09:01] seb128: ah, so slangasek unsub'ed ubuntu-release after ack'ing apparently [09:01] likely [09:01] * dholbach -> dogwalk [09:01] that's annoying you don't get a bug in such cases [09:01] a mail rather [09:01] that's the same when somebody reassign a bug to an another package [09:02] you have mails about the bug [09:02] but nothing suggesting it has been moved on somebody's else plate [09:17] * pitti hugs seb128 [09:21] seb128: I just put my wife's new audio CD into the CD-ROM drive [09:21] seb128: RB popped up, offered me a button "copy into music collection" [09:21] and did all the rest [09:22] pitti: yes, it's the only way I see to manage our queue of FFes [09:22] pitti: nice ;-) [09:22] (unsubbing u-release, that is) [09:22] this is *sooo* mindbogglingly well integrated!! [09:22] slangasek: right, I understand [09:23] seb128: I noticed that g-v-m also started sound-juicer; I think it should stop doing that then? [09:23] (and do we still need s-j at all? [09:23] Keybuk: ^ WDYT? [09:23] pitti: I though we dropped this tab, from gvm? [09:24] seb128: not for playing audio CDs, that's still s-j [09:24] pitti: oh, I though you used the fedora patch I pointed to you some time ago which masks the tab [09:24] oh, did you? I just disabled the gconf keys [09:25] ah, I remember [09:25] we need to do that for upgrades, too [09:25] the fedora patch takes care of upgrades [09:25] seb128: shouldn't that happen upstream, too, with gvfs'/nautilus' recent progress? [09:25] pitti: http://bugzilla.gnome.org/show_bug.cgi?id=509823 [09:25] Gnome bug 509823 in general "remove autorun/automount options" [Normal,Unconfirmed] [09:25] seb128: IIRC it just completely disables the usage of them, right [09:25] pitti: sounds like it doesn't need my opinion ;) [09:26] pitti: right, which is what we want [09:26] Keybuk: having s-j still does [09:26] seb128: what do you think? [09:26] I think we should keep it, it has its users === cjwatson_ is now known as cjwatson [09:26] the UI is clean and it's small [09:26] rhythmbox can't play auto cds, right? [09:26] it does in hardy now [09:27] oh, interesting [09:27] what do you call "auto cds"? [09:27] Keybuk: WFM [09:27] when started with a drive parameter it switches to the audio CD and start playing [09:29] seb128: ok, I agree we could keep it in main, it doesn't hurt; but for new installs, too? [09:29] Keybuk: oh, "WFM" under the assumption that s/auto/audio/ :) [09:29] pitti: it's part of the upstream desktop, not sure [09:30] err, I mean audio cds [09:31] seb128: which do you think we should use by default for audio cds? [09:32] I'm not decided [09:32] Pici: yay, dbus 1.1.20 :) [09:33] sound-juicer has clean interface and is easy to use [09:33] rhythmbox is nice but it might be too much to just play a CD [09:33] what do other people think? [09:33] we can still switch to rhythmbox by default now [09:33] and see feedback we get until beta [09:33] * pitti ♥ RB, but I'm not claiming to be a representative user [09:34] and then decide on whether we use it or sound-juicer [09:34] that's a good idea [09:35] since it integrates better with your existing music collection [09:35] I think *if* you are using RB, then you want it for CD ripping, too [09:35] but if you *don't* use/like RB, then it's too heavy [09:35] (like users prefering banshee, or so) [09:37] but then you'd want to use banshee instead of sj most probably (i still prefer sj for ripping though ;) ) [09:38] hey carlos [09:38] seb128: hi [09:38] seb128: so, I'll apply that g-v-m patch soon [09:38] pitti: which takes us back to http://bugzilla.gnome.org/show_bug.cgi?id=510614 [09:38] Gnome bug 510614 in general "Allow more user friendly selection of applications to use" [Enhancement,Unconfirmed] [09:38] pitti: thanks [09:38] seb128: so the application for new devcies is now selected in nautilus? [09:38] pitti: yes [09:39] pitti: the media tab in the preferences dialog [09:39] brb [09:40] slomo: IMHO that's solved much better now [09:41] pitti: how? [09:41] pitti: the extension to the .desktop files? [09:41] slomo: look at nautilus -> preferences -> media [09:41] slomo: there's a dropdown with all available apps now [09:42] pitti: how is this list generated? [09:42] slomo: not entirely sure, TBH; seb might know? === geser_ is now known as geser [10:07] RB by default> sound-juicer doesn't do last.fm, does it? my wife uses RB for that [10:11] cjwatson: the question was about CD playing (when you insert an audio CD in the drive) [10:12] but it might make sense to start the music player so people can switch to listen something else easily [10:21] seb128: I was talking about CD playing [10:21] seb128: last.fm users like to notify last.fm when they play a CD so that it gets more information about their music preferences [10:21] ah, right [10:23] that's not on by default in RB, but it's a fairly easy to find checkbox [10:26] yes, the plugin is not activated by default and you need to configure an account but that's easy enough [10:39] I dislike the fact that configuration for plugins is hidden inside the plugins [10:40] does RB do musicbrainz? [10:40] when ripping [10:41] apparently yes [10:44] pitti: is dropping the xinit.d dbus-launch thing still required? was for gnome-keyring, right? [10:44] ArneGoetje: did you get in touch with thep yet (thai fonts)? [10:44] slomo: right; maybe the newer version fixes it [10:49] tjaalton: what's musicbrainz? [10:50] mvo: hm, there's another update-manager upload in -proposed [10:51] mvo: shouldn't we get the current one into -updates first? otherwise verification get even more hairy and dragged [10:51] mvo: (bug 172609) [10:52] Keybuk: cddb/freedb on steroids [10:52] ie. much better === dholbach_ is now known as dholbach [11:00] pitti: Did you get my remarks on the langpack stuff? [11:00] pitti: BTW, new dbus uploaded; only change is a CVE id [11:00] lool: ah, yay; I added the CVE to the Ubuntu changelog [11:01] lool: langpacks> sorry, seems I missed them [11:01] pitti: Concerning the langpack: we have to do the scripting and all to have hildon langpackable in all cases, but the planned savings (something like 10 MB per language so max 30 MB) are not highly interesting [11:02] So the part with new templates and copying translations into multiple langpacks are not too important for us [11:02] lool: 10 MB per language? that sounds really much [11:02] And hence seeds support [11:02] pitti: yes, its not easy as it needs a hppa or ia64 box to test the 0.81.2 [11:02] pitti: gnome and main langpacks are currently pulled [11:02] mvo: lamont can't test this? [11:02] s/main/common [11:02] mvo: also, I'm actually more interested in getting a normal amd64/i386 upgrade test [11:03] mvo: i. e. to check for regressions [11:03] lool: so we just need the categorization for mobile and create l-p-mobile-XX, depending on l-p-gnome? [11:04] pitti: What we _need_ is only hook support, like for firefox, openoffice.org etc. [11:04] pitti: for regressions? right, I think I can ask bdmurray for this [11:04] The rest is all optional to whether we want mobile langpacks, but it's a lot of work for little savings === thekorn_ is now known as thekorn [11:25] hum... is cdimages.u.c slow for somebody else too? [11:25] (7k/s) [11:30] would it be a stupid question to ask why my keycodes for my multimedia keys appear not to work? === cprov-out is now known as cprov [11:46] hi, if anyone has a moment, this is very simple fix https://bugs.launchpad.net/ubuntu/+source/firefox-3.0/+bug/196953 , Polish users would appreciate this fix :] [11:46] Launchpad bug 196953 in firefox-3.0 "Add Polish translation to firefox.desktop (diff included)" [Undecided,New] [11:46] *this is a [13:17] Do we have anything in the distribution right now that can parse GtkBuilder files? [13:17] Our glade-3 doesn't seem enthusiastic, and gazpacho fails on startup [13:18] feed it some red cordial. [13:21] Hm. Yeah, need gazpacho 0.7.2... [13:21] hey mjg59 [13:22] mjg59: did you read my comments about the dbus-send call and the backlight issue yesterday? [13:22] seb128: Yeah. No clue what's going on there. As I said, dell-addon-backlight is an entirely separate chunk of code [13:22] It's run as a separate process === allee_ is now known as allee [13:22] k [13:23] I'll try if that's happening using the git version [13:23] and open a bug upstream if that's the case === Pricey is now known as PriceChild [13:25] lool: Any chance you could update gazpacho? We seem to have been on 0.7.1 since April :) [13:28] mjg59: *stab in the dark* - where would i start looking for why my multimedia keys have stopped showing up in xev, but used to all work fine? [13:28] all but one seem to come up in the dbus stuff [13:29] Hobbsee: Erm. No clue. X keymap change? [13:29] No, that shouldn't affect it [13:29] mjg59: don't think so [13:29] Are you using evdev? [13:30] mjg59: how do i check? [13:30] Xorg.0.log [13:30] nope [13:31] lool: Eh. 0.7.2 fails in the same way for me [13:31] AttributeError: GazpachoObjectBuilder instance has no attribute '_version' [13:32] mjg59: any idea about who i should bug about where to start looking? [13:33] Hobbsee: I genuinely have no idea what could be causing that. If they're showing up in dbus, then the only thing I can think of is that it's an X problem [13:33] Because they're clearly being generated by the kernel [13:33] pitti: would should we do with bug #196740? [13:33] Launchpad bug 196740 in gnome-utils "There are no debugging symbols for gnome-utils available" [Wishlist,Invalid] https://launchpad.net/bugs/196740 [13:34] mjg59: any idea why one of them isn't showing up in dbus then? [13:34] Which one? [13:34] next track [13:34] previous track does [13:35] but next track is broken. [13:35] Is it getting mapped? [13:35] how do i check? [13:35] See what it generates on the console with showkeys [13:35] Hobbsee: are they acpi event producing keys? [13:36] I ask only because my blue thinkpad button has stopped sending events to xev, but acpid sees it [13:36] n [13:36] g [13:36] mjg59: [13:36] mjg59: I never uploaded or used gazpacho *cough* ;) [13:36] g [13:36] ah [13:36] . [13:36] a [13:36] o [13:36] lool: packages.qa seems to have you in uploaders [13:36] y [13:36] mjg59: I'm just an uploader as a result of pkg-gnome scripts [13:36] . [13:36] s [13:36] lool: Ah [13:36] i [13:36] g [13:36] h [13:36] Hobbsee: Erm. [13:36] * lool slaps Hobbsee [13:36] a [13:36] r [13:36] g [13:36] h [13:36] 1 [13:37] 1 [13:37] 1 [13:37] a [13:37] Hobbsee: are you ok? [13:37] n [13:37] o [13:37] someone pleas stop Hobbsee [13:37] Seems over [13:37] o.o [13:37] Keyboard bugs are fun [13:37] mjg59: But I can still have a look [13:37] seb128: no idea other than 'wait for new upload', I'm afraid [13:37] lool: How on earth do you do any gtkbuilder work? We don't seem to have anything in the archive that actually works [13:37] hm, libsvg wants to go to universe? that's surprising [13:38] mjg59: I ... don't lalala [13:38] Hobbsee: /msg me when it's fixed and i'll unmute you [13:38] mjg59: You work on the old glade file and convert it to gtkbuilder haha [13:38] (just in case +q - no offence intended) [13:38] lool: Fail. [13:38] mjg59: work on a glade source and convert it to gtkui [13:38] gazpacho is supposed to do this, but is broken [13:39] does GtkBuilder fix the fact that your custom window cannot be derived from GtkWindow but has to be an instance of that class? [13:39] seb128: Sound server isn't started anymore with the new gnome-session [13:39] I'll try again [13:41] lool, isnt that started by an initscript ? [13:41] heya people [13:41] lool: do you have the gnome-settings-daemon update uploaded yesterday? [13:41] seb128: Ah I might not [13:41] well, that was interesting. [13:42] lool: you need it, it's the one starting the sound server now [13:42] * lool tries that now [13:42] 13:41 -!- mode/#ubuntu-devel [-b %*!n=hobbsee@ubuntu/member/hobbsee] by Keybuk, Hobbsee [13:42] mjg59, Keybuk, pitti, lool, etc: my apologies. keyboard was on crack. [13:42] don't believe I've ever seen that happen before [13:42] seb128: I didn't; will pull it now [13:42] Hobbsee: no apology necessary [13:42] Hobbsee: itz gtk bug [13:42] it appears that it was sending the enter keypress every few miliseconds or so :P [13:43] Hobbsee: e [13:43] v [13:43] just kidding ;) [13:43] cjwatson: yours reports -b too, how weird [13:43] (the mode change was -q) [13:43] cjwatson: chanserv sometimes gets confused. i've actually found a bug in it, too [13:43] Keybuk: yes, but it has a % which means a quiet. [13:44] ahh [13:44] I get plenty of 404s from http://archive.ubuntu.com hmmm === Ubulette_ is now known as Ubulette [13:45] Keybuk: it reported the +q as +b too [13:45] (Same here) [13:45] cjwatson: that's normal. wish it didn't though, it keeps confusing users. [13:45] +q == +b % [13:46] Bah my squid is on crack again [13:51] Riddell: kdepim-kde4 Depends: kdebase-runtime-bin, which is NBS [13:53] dholbach, ping (query) [13:53] emgent: he's dog walking [13:53] emgent: I saw the query but was too busy up until now and was just about to take my dog for a walk - sorry [13:54] emgent: please be patient and wait for an email on the list, I did not forget about it [13:54] Hobbsee, lol :) [13:54] keescook, ping [13:54] ok thanks dholbach :) [13:57] calc: oops, you broke it. [13:57] E: /var/cache/apt/archives/openoffice.org-hyphenation-en-us_2.3.1-1_all.deb: trying to overwrite `/usr/share/myspell/dicts/hyph_en_US.dic', which is also in package openoffice.org-hyphenation [13:59] superm1: ubuntustudio-audio depends on rosegarden4, which is NBS; can you please fix the dependency? [13:59] superm1: (oh, NBS == not built from source, thus it'll go away by the final release) [14:02] <_MMA_> pitti: superm1 does Mythbuntu. Ill get one of our guys to fix it. [14:03] I thought we fixed that a while back? Does it need a merge? [14:04] <_MMA_> persia: I thought so. Though, Im a little fuzzy as to what changed. [14:10] seb128: hey ! I'm packaging a gnome app yet, and I wondered if there is a metapackage for gnome dev files like kdelibs4-dev [14:10] seb128: atm I have something like 20 -dev packages..... I'm sure there is a way to do better no ? [14:11] tonio: gnome-devel [14:12] anjuta (>= 1.2.0), bluefish, [14:12] seb128: are you *sure* that's the one he wants? [14:13] gnome-core-devel looks more appropriate [14:13] jdong: dunno, he took a example I don't know about [14:13] I'm not using KDE [14:14] seb128: I believe he just wanted a metapackage for common build-deps associated with gnome [14:14] rather than listing like 30 packages for the simplest of GNOME apps [14:14] jdong: feel free to reply to him, he asked on a public chan [14:14] tonio: ^^ see if gnome-core-devel fits the bill [14:14] seb128: I wasn't entirely sure of the answer myself [14:14] I've other things to do than discussing if my reply is what he wants ;-) [14:16] seb128: thanks, jdong, same [14:16] * Hobbsee looks for a large brick [14:16] * jdong hands Hobbsee a macbook pro [14:16] hrm. do you want that back, in working order? [14:17] Hobbsee: it worked to begin with? [14:17] one of my buds bought one to work with Linux. An interesting idea. [14:17] yeah, good point [14:17] of course 3 hours later he opened it up and swapped out the broadcom wifi card [14:17] I'm quite sure that's not how Apple intended it to be used ;-) [14:18] the macbook air is _sweet_ [14:18] now it runs Linux great but doesn't run OS X at all [14:18] lifeless: for values of sweet approaching sour [14:22] mjg59: bitter at the kernel team much? :) [14:29] * Hobbsee uses jdong's macbook, and knocks herself out. night all. [14:38] thom: oh, you don't like [14:39] ? === asac_ is now known as asac === dholbach_ is now known as dholbach [15:01] * soren kicks /usr/share/initramfs-tools/scripts/local-premount/resume [15:04] Failed to fetch http://archive.ubuntu.com/ubuntu/pool/main/libg/libgweather/libgweather0_2.21.92-0ubuntu1_i386.deb 404 Not Found [IP: 91.189.88.31 80] [15:04] one of the archive.ubuntu.com servers seems to be 404'ing like mad [15:05] lifeless: no, played with one, unimpressed === DktrKranz2 is now known as DktrKranz [15:08] seb128: Even with latest updates, logon sound is still borkne for me [15:08] pulseaudio is started, but I don't hear the logon sound [15:10] lool: ok, same for me, I'll try without the debian gsd startup patch and with the svn change version [15:13] lool, did you see my email wrg to ekiga? [15:14] superm1: Yes [15:14] superm1: I'm not particular tightened to ekiga, but will have a look [15:15] lool: doesn't work better [15:15] lool, okay well you had just done the last merge from what it looked, so rtg said to speak with you. Thanks [15:17] superm1: Yeah, that's what I guessed [15:22] Keybuk: It seems there's a race in the resume code in our initramfs.. Do you have a few minutes? [15:23] soren: sure [15:23] Keybuk: For reasons that will go unmentioned, I had to nuke my swap lv at some point and have now recreated it. This has reordered my lvm layout so that swap_1's /dev/disk/by-uuid link gets created a bit too late for the resume stuff to be able to see it. [15:23] Adding a tiny sleep or a call to udevsettle will fix it, but I'm sure you have a better idea. :) [15:24] Well, udevadm settle, but ykwim. [15:25] I could make a wait_for_resume_device loop akin to the one that waits for the root, but I'm not too hot on that either. [15:26] resume is run from local-premount, no? [15:26] Right. [15:26] so it's run after the wait-for-root loop? [15:27] Yes. [15:27] so shouldn't the swap device be there too? [15:27] I can see what you're getting at.. [15:27] Precisely. [15:27] but it's not. [15:27] The thing is.. [15:27] hmm [15:27] it's on the same disk? [15:27] Yup. [15:28] Due to the fact that I removed the lv and then created it again, it's now discovered as the last lv rather than one of the very first ones. [15:28] Well... I assume that's the reason. [15:28] It worked just fine before then, and now, if I break=bottom, and manually do the resume /dev/disk/by-uiid/blahblahbl, it resumes just finne. [15:28] fine. [15:29] Hobbsee | Riddell: can one of you two give a hint about the FFe for new package keurocalc? (bug #196123) thanks [15:29] Launchpad bug 196123 in ubuntu "[FFe] keurocalc port to kde4" [Undecided,Incomplete] https://launchpad.net/bugs/196123 [15:29] I put a few "ls -l /dev/disk/by-uuid/" into the resume script with a "sleep 1" in between. In the first one, it shows all the lv's up until and including the root, in the last one, it also has my swap. [15:29] what does $RESUME contain when you do break=bottom ? [15:29] Oh, that stuff is all fine. [15:30] yeah, simple race then [15:30] The path reference is just fine. [15:30] to fix it properly [15:30] I'd move the $RESUME -> $resume code from the resume script into init itself [15:31] and then adjust the loop inside local to check for $ROOT and $resume existing [15:31] Well.. [15:31] That would work. [15:31] Sort of. [15:31] The problem is that it shouldn't fail if the resume device never shows up. [15:32] seb128: g-v-m with that patch uploaded; I had to fix the second upstream patch, took a little longer [15:32] thom: I thought it was better than my d430 - lighter, less volume [15:33] Keybuk: I'm thinking about adding a second loop that times out after something as low as 5 seconds. [15:33] pitti: ok, thanks [15:33] I *hate* never fail if never shows up problems [15:33] pitti: do you think we should revert the dual battery hal change? [15:33] Keybuk: ?!? [15:34] soren: racing $ROOT and $RESUME [15:34] Keybuk: I can't parse that sentence. [15:34] in an ideal world, you'd just mount the root filesystem on the event for it being available [15:34] seb128: back to double batteries? [15:34] and you'd resume from swap on the event for it being available [15:34] seb128: AFAICS we should rather change it to only consider /proc? [15:34] Keybuk: Right. [15:34] but you can't do that, because you have to wait for both to determine which you need to do [15:34] Keybuk: Exactly. [15:34] (see also: why we're not taking advantage of Upstart yet) [15:35] it's worse, since depending on circumstance, one or the other may not show up at all [15:35] Keybuk: One could make the argument that hibernation and subsequent resuming from hibernation is somethat that will only reasonable take place on systems that don't need to wait several minutes for the resume partition to show up. [15:35] seb128: I don't know whether it's the right solution (i. e. whether the kernel puts wrong stuff in /sys or hal just reads it wrongly), but since batteries in /proc are known to work, we should maybe just go with that in hardy [15:35] if you have $ROOT, how long do you wait before assuming $RESUME isn't going to show up [15:35] pitti: dunno if you read it yesterday but crimsun_ got some git changes fixing the battery notifications [15:35] you can always guarantee someone will file a bug, because their $RESUME shows up one second after whatever you pick [15:36] seb128: no, I didn't reat id [15:36] Keybuk: I realise. [15:36] Keybuk: Just as is the case for the current loop that waits for the root. [15:36] carlos: any idea about this translation regression? bug 196106 [15:36] Launchpad bug 196106 in language-pack-kde-de "context menu entry "Paste File" [and other dialogs] not translated into German (anymore)" [Undecided,New] https://launchpad.net/bugs/196106 [15:37] you'd end up with [15:37] Keybuk: could we put a marker in the root filesystem to say that you should wait for resume? [15:37] on filesystem-available $ROOT and (filesystem-available $RESUME or 5s after filesystem-available $ROOT) [15:37] type things [15:37] hmm, I suppose it would have to be on the device such that you didn't need to mount the filesystem to get at it [15:38] cjwatson: doesn't protect you from corrupt swap partitions :-( [15:38] or partial hibernation, etc. [15:38] You can't partially hibernate [15:38] mjg59: what happens if the power goes out while it's writing to swap? [15:38] Oh, sorry, I see what you mean [15:39] or (and I have seen this) some idiot using md as their swap partition [15:40] cjwatson: of course, we could always use swap files on the root partition ;) [15:40] :-) [15:40] * soren twitches [15:40] I think we have to support swap partitions regardless of the default [15:40] or mandate that the resume partition has to be on the same physical disk as the root [15:40] That is the case here. [15:40] which presumably still doesn't guarantee that they show up simultaneously [15:41] just quite close together [15:41] Yet, it still races (and loses). [15:41] not when md, lvm, devmapper are involved - no [15:41] even without any of that, it can't create both devices at the same time [15:41] Er... lvm is involved. [15:41] And it races. [15:42] cjwatson: I mean that it's easy to force waiting for the rest of the partitions on the same disk to all exist [15:42] ah [15:42] but it's hard when you're actually waiting for a resultant block device [15:42] udev may have seen the partitions, but it now has to go and run lvm, mdadm, etc. [15:42] Well, the block device is there. [15:42] and those things use watershed [15:42] udev just hasn't quite gotten around to creating the uuid symlink. === greeneggsnospam is now known as jsgotangco [15:43] soren: honestly, the best quick fix solution is just to copy the busy-loop out of local into resume [15:43] and reduce it from 180 to 5 [15:44] Keybuk: That works. [15:47] Keybuk: Ok, I'll work something to that effect out when I get off the phone again. Thanks. === j_ack__ is now known as j_ack [16:00] excuse me, does anyone know where can I find the _Unlock button for translation in gnome-system-tools admin applets? It's not in g-s-t, it's not in policykit-gnome... [16:04] forget it, my bad [16:10] doko_: please commit your oem-config changes to bzr, preferably as a branch from the revision at which 1.28 was released which you then merge with evand's un-uploaded changes [16:11] cjwatson: ok, same for ubiquity? [16:11] you uploaded ubiquity? eek [16:11] yes, absolutely [16:12] down to one build dependency on python-xml ... [16:12] erm, but ubiquity is awkward because you aren't in the ubuntu-installer team [16:12] please publish a branch that we can merge [16:14] doko_: also please don't use ubuntu1 versioning for packages maintained in Ubuntu as native packages [16:14] oh, actually, you didn't, it was just the tarball ... [16:14] yep, noticed, but forgot to rename [16:15] doko_: localechooser should be done in bzr too, if you were planning to do that [16:15] seb128: I just updated https://wiki.ubuntu.com/StableReleaseUpdates [16:16] seb128: the procedure is *much* easier to read now IMHO; comments? [16:18] doko_: if you haven't uploaded ubiquity yet, please don't, as evand is preparing an upload too [16:18] * Keybuk tries to remember if you can match string literals in the C preprocessor #if [16:18] pitti: much nicer indeed [16:19] pitti: the procedure description is clear now [16:19] cjwatson: sorry, already done. or remove if from the queue? [16:19] too late :-( [16:19] uh oh [16:19] the publisher is already munching on it [16:19] Keybuk: what do you want to achieve? [16:19] I imagine I should cancel this then [16:19] evand: you'll have to [16:19] sistpoty|work: #if SBINDIR == "/usr/sbin" [16:20] done [16:20] evand: I looked for you on the distro channel, but didn't find you there. forgot to look here [16:20] until doko gives us a branch to merge so that you can punt this to 1.7.14 [16:20] there were vcs-bzr headers in both packages, I believe [16:21] doko_, sorry, I'm somewhat in flux today and don't have xchat set up for Canonical IRC [16:21] * seb128 hugs pitti, good work, I think it's great now ;-) === \sh is now known as \sh_away [16:21] * pitti hugs seb128 [16:24] evand: patch is here, http://people.ubuntu.com/~doko/tmp/ubiquity.diff my connection to the data center seems to be dead slow [16:24] doko_: choose-mirror needs to go to bzr too (and again, there was a small unuploaded change; branch from r580) [16:24] cjwatson: I'll take care of that [16:24] thanks [16:25] doko_, thanks [16:25] Keybuk: seems like you can't: http://gcc.gnu.org/onlinedocs/cpp/If.html#If [16:26] sistpoty|work: yeah, reading C99 doesn't help either [16:26] it doesn't really go into detail about what you can put there [16:26] other than saying it has to be a constant expression [16:26] only characters, but you cannot index a variable as well, so :( [16:27] (apart from integers of course) [16:28] Keybuk: if it uses autoconf, you could just set a definition to s.th. if sbindir is /usr/bin? [16:28] well, *yes* :p [16:28] heh [16:29] Keybuk: C99 6.4.5 para 6 "It is unspecified whether these arrays are distinct provided their elements have the appropriate values" which I think means you can't rely on pointer equality [16:29] I wanted value equality though :) [16:30] well the preprocessor will only give you pointer equality [16:30] err, not even that actually, hmm [16:30] the preprocessor just gives me an error [16:30] which pretty much answers the question no matter what the standard says ;) [16:30] yeah, C99 says you can't do that [16:31] 6.10.1 para 1 "shall be an integer constant expression"; 6.6 para 6 lists the operands that may be in an integer constant expression none of which include string literals [16:31] *nods* [16:31] agree [16:31] * Keybuk returns C99 to its job of being a bookmark in Stevens APUE [16:33] (I'm not the only one who uses books as bookmarks for other books, right? :p) [16:33] Keybuk: folding them closed around each other? [16:34] lifeless: yeah, or leaving one open with the other on top of it to keep my place [16:34] yah, I fo the dormer [16:36] http://www.amazon.co.uk/dp/0470845732/ -- best buy ever [16:46] bzr: ERROR: bzrlib.errors.UnlockableTransport: Cannot lock: transport is read only: [16:47] * doko_ grumbles [16:47] pitti: You around by any chance? (about clanbomber) [16:49] pitti: I feel so rejected ;-) [16:50] LaserJock, what, did he not approve the MIR for squeak ? :0 [16:50] hah [16:51] heh [17:02] sistpoty|work: the problem is that in autoconf, sbindir isn't fully evaluated yet [17:11] doko_, I've merged your changes in and uploaded a new ubiquity that reverted them. It looks like you're already fixing this in localechooser and choose-mirror, though, so they'll go back in as soon as we release another ubiquity. [17:12] assuming that the new localechooser and choose-mirror are in the archive at that point [17:12] evand: localechooser is [17:12] but not choose-mirror [17:12] ok [17:13] Keybuk: oh... can't it work with some m4 that's getting evaluated once configure is called (assuming that you know autoconf much than I do *g*)? [17:13] sistpoty|work: yes, but it's icky :p [17:13] heh [17:13] doko_: yeah, if you ever find yourself touching anything inside d-i/source/ in ubiquity or oem-config, please read d-i/README which tells you not to ;-) === alleeHol is now known as allee [17:20] Pumpernickle: do you mean to have so many clones? [17:21] Keybuk: he's host reminds me of a network troll I've had problems with.. [17:21] jpatrick: the nick is known to me [17:22] so it's likely just network problems on his end and an unattended reconnecting IRC client [17:22] (giving him the benefit of the doubt - and in no way looking at the fact the numbers joined out of order ) [17:23] forwarding him to ##fix_your_connection would be best [17:23] Hobbsee: for whatever reason it was biting me before the update, so i requested the update hoping it might fix it, since it didn't i'm going to have to update the packaging as well :-\ [17:23] jpatrick: how do I do that? [17:23] Keybuk: /mode +b *!*@CPE0014d13c957e-CM0012c9a9a6dc.cpe.net.cable.rogers.com!##fix_your_connection [17:24] Keybuk: and remove the last ban [17:25] jpatrick: it won't let me add that as a ban [17:25] Keybuk: you could /op me a sec [17:25] MODE +b redirection channel: :That channel doesn't exist [17:26] Keybuk: whops, forgot: /mode #ubuntu-devel +b *!*@CPE0014d13c957e-CM0012c9a9a6dc.cpe.net.cable.rogers.com!##fix_your_connection [17:26] it's the fix_your_connection thing it's complaining about [17:27] hmm, the channel does exist and Pumper* is in there.. [17:27] no idea then [17:27] silly network [17:29] Keybuk: hmm, curious guy [17:29] jpatrick: hmm? [17:29] /mode #ubuntu-devel +b *!*@CPE0014d13c957e-CM0012c9a9a6dc.cpe.net.cable.rogers.com [17:30] I've always wondered whether those still work if the host is cloaked :) [17:30] Keybuk: it does [17:30] bored now, anyway [17:30] he's clearly either got a major problem, or trying to cause one === doko_ is now known as doko [17:54] cjwatson: Hi Colin. Is there a list for oem-config discussion/development? [17:55] coastGNU: ubuntu-installer@ will do [17:57] cjwatson: Ok, thanks. Next question, do you plan to handle persistent-net and persistent-cdrom via oem-config or is it up to the oem to place needed hook files for this? [17:59] To #ubuntu-devel: "Thank you for ubuntu!" -- from http://brainstorm.ubuntu.com/idea/1514 [17:59] coastGNU: hmm, good question, I hadn't particularly thought of that [18:00] coastGNU: could you file a bug for that? comes under the general category of reconfiguration for specific hardware [18:01] cjwatson: I'm writing an Article for freeX magazine and there are quite a lot question marks which came up in my mind over the last weks. [18:01] cjwatson: s/wek/week/ [18:01] heno: :) [18:02] coastGNU: oem-config development is basically whenever people ask for stuff :) [18:11] cjwatson: Hhhm, if you would ak me oem-config is the most promising tool to fix bug #1, isn't it === \sh_away is now known as \sh [18:27] coastGNU: I'm not sure I'd go that far :-), but it's certainly useful [18:55] Hi. Do you know how can I use internet inside PPA chroot? My package (maven) requires download pom and jar files from internet in order to build maven itself. [18:56] you can't; you need to put any necessary files in the source package itself [18:56] (this is intentional) [18:56] ok, i'll try again but... there is a limit on the size of .orig.tar.gz ? [18:57] terrex: no, aside from the usual PPA size limits [18:57] ok, thanks! [19:06] will http://brainstorm.ubuntu.com/ influence your work, developers ? [19:09] nicolah: too late for Hardy, but AFAIK it'll be taken into account for the UDS to come (Intrepid Ibex) [19:09] great [19:10] thanks for the answer stgraber I won't bother you anymore [19:19] slangasek: Any chance you could apply some love to bug 196868? If I have to redo the kernel merge, kittens will die. [19:19] Launchpad bug 196868 in kvm "[ffe] Upgrade kvm from vers. 60 to 62" [Undecided,New] https://launchpad.net/bugs/196868 [19:20] That's meant as a threat, not a prophecy. I have a bag full of kittens and a sharp knife, and I'm ready to go. [19:21] Will someone *please* think of the kittens? [19:22] hahaha [19:25] <_MMA_> For shame. Laughing at the thought of dying kittens. ;) === afflux_ is now known as afflux [19:30] soren: I doesn't negoshurait wif terrists [19:30] soren: yeah, I'll be looking at that one today, but not for a few hours yet [19:33] ZOMG! i haz bin lolcatzetzed. [19:34] hi I was wondering if the kernel module firmware_class has been removed from the gutsy kernel or it has been compiled into it? [19:36] Does anyone know if Hardy will use cx18 beta driver for ivtv? [19:39] slangasek: http://outflux.net/kittens.png [19:41] keescook: aieee [19:43] <\sh> lol [19:49] keescook: that guy looks like my old maths teacher... [19:50] <\sh> jpatrick: that guy is himself ... the master in security ,-) [19:51] \sh: Er.. No. It's slangasek. Not kees. [19:51] that pictures is not of me. :) [19:51] <\sh> what? [19:51] s/s [19:51] s/s/z/ [19:51] <\sh> s/security/release managing/ ,-) [19:51] <\sh> sorry ;) but the website is yours... [19:52] heh, yawp [19:54] <\sh> but this smile on the face...I guess he hacked something before ;) [19:55] tjaalton: is there going to be any sort of final resolution of bug #133192? it's been release-noted for quite a few alphas in a row now... [19:55] Launchpad bug 133192 in xserver-xorg-video-ati "Blank screen or distorted image because of wrong default AGPMode value" [High,Confirmed] https://launchpad.net/bugs/133192 [19:55] <\sh> I had such a smile yesterday, when I stresstestbroke our servers [20:00] <\sh> so time to leave for today... === \sh is now known as \sh_away [20:06] slangasek: oops, sorry.. it's fixed now [20:06] 6.8.0-1 reverted the behaviour [20:10] slangasek: I don't remember the gory details, but there is no silver bullet to the problem, there will be setups where the only resort is to play with BIOS settings [20:12] so the current status is that the behavior matches the behavior from previous releases? [20:12] correct [20:12] ok, that's probably good enough to take it out of the release notes then [20:12] yep [20:13] we should make a MASTER bug about this.. [21:02] doko: I'd appreciate it if you wouldn't remove the details from the "Automatic update" changelog message [21:03] they are useful for tracking down bugs later on [21:03] cjwatson: err, I didn't ... [21:03] doko: did you not run 'debian/rules update'? [21:04] yes [21:04] [ Matthias Klose ] [21:04] * Automatic update of included source packages. [21:04] that was included [21:04] something must have gone wrong if it didn't include any details [21:04] I would like to handle the next upload [21:04] ok [21:04] a little worried something's out of sync somewhere subtle [21:04] but thanks for the merge/upload [22:27] cjwatson: hello. cdimage.ubuntu.com is now building mythbuntu alternate disks. do i have to talk to you if i want to add preseeds? [22:27] how does https://translations.launchpad.net read translations from projects? [22:28] Anyone having troube with X.org taking 90+ % of CPU usage with Hardy Alpha 5? [22:31] EnderTheThird: No, at least not with an NVIDIA driver and Compiz. Are you using VESA or something along those lines? [22:32] mbt: no. intel [22:33] EnderTheThird: Strange. Sounds like a bug to me. [22:33] Dell GX260. No powerhouse by any stretch of the imagination, but it should at least be able to do 2D. (no Compiz) [22:34] yeah, it's unbearably slow. [22:35] Well, maybe try switching to the VESA driver and see if that speeds things up. At least that would likely pinpoint the cause as being the Intel driver. [22:37] I'll give it a whirl. brb [22:39] much better with vesa. It's still not as responsive as I'd think it should be, but it's usable. [22:39] Yeah, the vesa driver is very not optimized. But, I think that probably just gave you enough information to report the problem [22:40] moving windows is awful, heh [22:40] lol yeah, though you can tweak some settings in gconf to do wire-frames and the like [22:41] Gutsy performance wasn't too great either. Weird thing is that I remember it running better with 6.06 and 6.10, even with Compiz/Beryl. [22:44] nevermind. it's not just with the Intel driver [22:44] It seems to be jumping around with Vesa too. [22:44] It will. [22:44] It's doing most everything with the host CPU instead of with the GPU when using the VESA driver. [22:45] But CPU use shouldn't stay consistently high unless you've got a lot of screen updates. [22:46] I'm going to try downgrading to 6.06 after the d/l completes. I remember it being much faster on 6.06. We'll see if it's just my crappy memory, heh. [22:48] Thanks for the help. [22:49] No problem. [22:49] And also for being the only one active in this channel, ha [22:50] lol I idle a lot too [22:52] Yeah, but 233 idlers, that's just ridiculous [22:53] should be a idlerpg competition in here [22:53] My god man, you just might be onto something! [22:53] lol [22:53] there is a #idlerpg [22:53] That is terrifying. [22:55] User: Joe_Berelum Total time idled: 1501 days, 18:47:39 [22:56] Wow. [22:56] wouldn't your freenode registration expire? [22:56] because he hasn't identified to nickserv for 120 days or whatever? [22:57] anyone else find it funny that talking about being idle is bringing idle people out of their idle slumber? [22:58] I was never in idle slumber :) [22:58] or bringing them back from the toilet... ;) [23:06] mvo: how can I do an upgrade from dapper to hardy on a server ? [23:06] mvo: is there a do-release-upgrade command ? [23:07] mathiaz: yes - see https://help.ubuntu.com/community/HardyUpgrades#head-e7f287c730b93116f89de7ea7e05efbe95fa6dd1 [23:07] * sistpoty still believes that apt-get dist-upgrade should just work for people not installing 3rd party stuff [23:07] mvo: ok - thanks. [23:08] sistpoty: for dapper->hardy? it will not, you will have to upgrade apt and dpkg first if you do it manually, the problem is the new "Breaks" field [23:09] mvo: than that's wrong and should be fixed imho *g*... but I must admit that I upgrade my gf's laptop with dist-upgrade, and it was just too smooth to be real ;) [23:09] not with dist-upgrade, but with the updater python thingy you wrote [23:10] sistpoty: from dapper? [23:10] mvo: from feisty to gutsy [23:10] sistpoty: nice, good to hear :) [23:11] sistpoty: the fix for this would be to not use breaks until the next lts got released in order to maintain compatiblity. that is a bit drastic IMO - or to backport it to dapper, but that is quite a change as well [23:11] the first fix wouldn't work surely, because you'd just have the same problem with lts+2, no? [23:13] but I can see sistpoty's point, I would have thought a lot of server admins will try dist-upgrade from dapper to hardy [23:14] mvo: right, I guess even having s.th. getting upgraded first (e.g. libc) pre-depend on the new apt wouldn't work, due to the circular nature [23:15] mdke: I agree that this is not a ideal situation - but I can not see what we can do (except for backporting) - the solution we have is the release-upgrader that does some additional checks and shouldn't get in the way of any admin in server mode [23:15] mvo: yeah. The only thing we can do is try and get the message to as many people as possible that dist-upgrade won't work [23:15] sistpoty: yeah, even if we could make it work somehow it would be rather messy [23:16] that's true, and I guess it would open a can of worms trying to fix it [23:16] mdke: we discussed issuing a update to dappers apt that will give a big warning on dist-upgrade (if any of the upgraded-to packages are hardy) [23:16] sistpoty: *nod* [23:17] mvo: sounds like a plan [23:17] yeah [23:19] mathiaz: please let me know how well the upgrade worked for you - in case of problems, please send me the logs in /var/log/dist-ugprade === Ubulette_ is now known as Ubulette [23:56] Is there any way in pbuilder to set make -j NUM_CPUs? [23:58] jdong: Add a hook that sets MAKEFLAGS before calling dpkg-buildpackage? [23:58] persia: is that the proper way to do it (i.e. what the buildds would do)? [23:59] jdong: I'm not sure. Maybe setting parallel=n in DEB_BUILD_OPTIONS? From past traffic in debian-devel, it seems to generate more FTBFS issues.