[00:02] <LaserJock> anybody know who wrote the debcheck script for qa.ubuntuwire.com?
[00:04] <RoAkSoAx> norsetto, so i just copy the last changes showed in the hardy records (the last one) and paste it in the intrepid records... and since i've mande any patch or anything else... that would be it?
[00:05] <norsetto> RoAkSoAx: not just the last, ALL ubuntu changes (there might be previous ones)
[00:07] <RoAkSoAx> norsetto, including the records of the previous merge? http://pastebin.ubuntu.com/9865/
[00:08] <norsetto> RoAkSoAx: if its a change you are carrying over, yes
[00:10] <norsetto> gotta go now, g'night all
[00:29] <alteregoa> he-man
[00:29] <alteregoa> MUD? multi user dungeon?
[00:29] <jdong> err... wrong channel :)
[00:29] <jdong> this isn't a MUD :)
[00:29] <jdong> or maybe it is
[00:30] <alteregoa> fnord...
[00:30] <jdong> You have walked into a room full of nerdy coders. Press (a) to merge packages from debian. Press (b) to exit
[00:30] <alteregoa> press alt-f4 to continue
[00:30] <jdong> that key combination would free so many of us from our Ubuntu addiction.
[00:30] <pochu> what about syncs? :)
[00:31] <jdong> pochu: syncs are like merging without a payload....
[00:31]  * jdong reads that back for innuendos...
[00:31] <pochu> heh, I never thought about syncs that way
[00:33] <crimsun> alt+f4 seems only to display workrave
[00:33] <crimsun> oh well, workrave time :-)
[00:34] <alteregoa> yeah,
[00:34] <alteregoa> i have to fix those fidonet stuff now
[00:40] <alteregoa> would be nice to see synchronet for ubuntu
[00:40] <alteregoa> in the universe
[02:38] <sven_> so where's the link for the rules in this place?  i don't wanna start asking things and get kicked for being off topic.
[04:00] <coppro> where's a good guide to packaging libraries?
[04:40] <cheatr> Could someone give me a hand? I'm trying to recover my revu password. I entered my email address in the box, and left the password blank. I then clicked on the recover link. I entered "gpg -d <<EOT ; echo" in a terminal. I then pasted the encrypted message that was generated after I clicked the recover link. I then typed "EOT" followed by return. I entered the correct password, but nothing gets displayed. Any ideas?
[04:43] <wgrant> cheatr: It tells you it has decrypted it?
[04:43] <wgrant> cheatr: What's the email address associated with your REVU account?
[04:43] <cheatr> wgrant: mrcheatr[at]gmail[dot]com
[04:44] <cheatr> wgrant: And it doesn't say anything about it being decrypted. It also doesn't complain about an incorrect password. All it does is show info about my gpg key
[04:45] <wgrant> cheatr: The encryption was correct; you have no password set.
[04:45] <wgrant> I'll set one.
[04:46] <cheatr> wgrant: Thanks
[04:47] <wgrant> cheatr: Try again.
[04:47] <cheatr> wgrant: Yeah, it worked. Thanks
[04:48] <cheatr> wgrant: Is there any way to change the password?
[04:48] <wgrant> cheatr: Not at this point.
[04:48] <cheatr> wgrant: Ok, I'm glad firefox can remember my password
[05:12] <LaserJock> wgrant: ping
[05:17] <wgrant> LaserJock: Hi.
[06:29] <tbielawa> so there's a ticket in LP asking us to sync dx from debian unstable.
[06:29] <tbielawa> which is marked to do in MoM
[06:30] <tbielawa> Whats the proper way to respond to a ticket like this (#226088)
[07:16] <white> is there a problem with cowbuilder and simple-patchsys from cdbs?
[07:16] <white> i tried to rebuild two packages in cowbuilder (which build fine in the normal chroot) and both tell me that patches can't be applied, although it all works with patch
[07:16] <Hobbsee> heya white!
[07:16] <white> hi :)
[07:18]  * wgrant uses sbuild instead.
[08:06] <RAOF> Anyone want to play "Upload to Debian"? :)
[08:07] <jdong> RAOF: doesn't sound like a very fun game
[08:07] <RAOF> Sure it is!  I provide some .dsc, and you put your name on them!  How could there be _more_ fun?
[08:08] <Hobbsee> RAOF: if someone else does the work, and hands the .dsc files.
[11:26] <bobbo> nxvl: ping
[12:26] <tbielawa> anyone have an opinion on wether it's more proper to ask policy questions here or on the mailing lists?
[12:31] <highvoltage> my guess would be that the mailing list is best, probably just because more people would receive your question
[12:31] <highvoltage> although, if you want a quick answer, it usually doesn't hurt asking on IRC
[12:32] <highvoltage> you can always follow to the mailing lists after asking on IRC :)
[12:33] <tbielawa> highvoltage: good policy you have there, I'll try throw it out in here
[12:34] <jsgotangco> hey highvoltage
[12:36] <tbielawa> on second thought, I'm not sure what I mean to ask... I'm just thrown off on this part in the PackagingGuide/Basic about dealing with source that has existing debian/ dirs
[12:38] <tbielawa> I think I figured it out
[12:39] <tbielawa> https://wiki.ubuntu.com/PackagingGuide/Basic#head-66bb429f7da4cd243a5977c776c661d5e7603e69
[12:39] <highvoltage> hey jsgotangco
[12:39] <tbielawa> is that last bullet implying that you should ask for a clean source from upstream and have THEM provide a diff.gz from the source with debian/ included?
[12:40] <x1250> tbielawa: nope, it refers to another packager
[12:42] <tbielawa> I'm not following this part then. How do you deal with debianized sources?
[13:34] <Arby> If pbuilder tells me 'pbuilder: debootstrap failed -> Aborting with an error' how do I find out what the error was?
[13:35] <Hobbsee> look further up
[13:36] <Arby> Hobbsee: it says Failure trying to run: chroot /var/cache/pbuilder/build/16928/. dpkg --force-depends --install var/cache/apt/archives/libc6_2.7-10ubuntu3_i386.deb
[13:36] <Arby> but no suggestion why it failed
[13:36] <tbielawa> Arby:  i've been getting that same error!
[13:36] <Hobbsee> probably because libc6 is not installable.
[13:36] <Arby> ah, it's not just me then
[13:36] <Hobbsee> chroots probably can't be created yet.
[13:36] <tbielawa> I was making a sid environment when I was having problem
[13:37] <Arby> oh, I've seen people in this channel say they did it already
[13:37] <Hobbsee> tbielawa: the fact that your sid wants to install ubuntu versions of important packages is a bigger problem.
[13:37] <Hobbsee> Arby: i suspect they upgraded theirs from hardy.
[13:37] <Hobbsee> were you doing a create?
[13:37] <tbielawa> Hobbsee: I had changed my mirror to make sure it was grabbing from debian ;)
[13:37] <Arby> Hobbsee: any estimate when it is likely to be possible?
[13:37] <Hobbsee> Arby: 'when it's done'
[13:37] <Hobbsee> tbielawa: so, why did it pick an ubuntu version?
[13:37] <Arby> :)
[13:38] <tbielawa> Hobbsee:  It was failing grabbing the libc6 from debian
[13:38] <Hobbsee> oh, so that libc6 is also broken.  right
[13:38] <tbielawa> W: Failure trying to run: chroot /var/cache/pbuilder/build/15482/. dpkg --force-depends --install var/cache/apt/archives/libc6_2.7-10_amd64.deb
[13:41] <Hobbsee> tbielawa: that telsl you what failed to install - not why it did
[13:42] <tbielawa> Hobbsee: I know :( I wish it had some more verbose output
[13:42] <tbielawa> I had hoped that running it with the -log flag would accomplish that, instead it just replicated stdout to my .laslog
[13:43] <tbielawa> *.lastlog
[13:47] <tbielawa> Arby: What was your target dist? I had success making hardy environments
[13:47] <Arby> tbielawa: intrepid
[13:48] <tbielawa> DId you start right off pointing at intrepid or try and update to it?
[13:49] <Arby> started off pointing to it
[13:49] <tbielawa> I'm running it again to be sure, but IIRC I had luck with pbulder --update intrepid; you may try that
[13:49] <tbielawa> once i had a hardy working, of course.
[13:50] <Arby> tbielawa: thanks, I'll try that
[13:50] <tbielawa> Arby: sure thing :)
[13:56] <tbielawa> the command to run would have been: pbuilder --update --distribution intrepid --override-config
[13:56] <tbielawa> though I may have hallucinated it working, i got unmet dependencies on libpam-modules, e3fsprogs, util-linux....
[14:01]  * Hobbsee wonders why you expected it to work
[14:05]  * tbielawa likes to dream
[15:02] <cheatr> Would someone here be willing to look at my package on revu? I doubt it is perfect, but I need some feedback.
[15:09] <RainCT> cheatr: perhaps you'd have more luck if you say the URL :)
[15:10] <cheatr> RainCT: http://revu.ubuntuwire.com/details.py?package=handbrake
[15:11] <RainCT> cheatr: http://revu.ubuntuwire.com/revu1-incoming/handbrake-0805040510/lintian  here you have to things to fix
[15:11] <RainCT> cheatr: the changelog should close a bug (Syntax for that is: (LP: #xxxxx) )
[15:12] <RainCT> cheatr: debian/control has no section
[15:13] <RainCT> cheatr: copyright is empty in debian/copyright. and replace "Author(s):" with "Authors:"
[15:13] <cheatr> RainCT: I searched the entire website, and they do not mention a copyright anywhere, any suggestions on what I should do?
[15:14] <RainCT> cheatr: what is that HandBrakeCLI thing?
[15:14] <cheatr> RainCT: It is a video converter. It can convert video files or rip and convert a DVD.
[15:15] <RainCT> cheatr: yes but what sort of file is it?
[15:15]  * RainCT 's connection is to slow to download that now
[15:15] <cheatr> HandBrakeCLI: ELF 32-bit LSB executable, Intel 80386, version 1 (SYSV), for GNU/Linux 2.4.1, dynamically linked (uses shared libs), not stripped
[15:16] <cheatr> RainCT: This is their linux version of the file. They also offer an SVN version. Should I use that instead?
[15:17] <RainCT> cheatr: yes
[15:17] <RainCT> cheatr: you can't just put a binary in a debian source package, you've to use the sources and compile them there
[15:18] <cheatr> RainCT: Thanks for the feedback. This should keep me busy for a while. I'll probably be back later after I make all the changes you suggested.
[15:20] <RainCT> cheatr: no problem, good luck
[15:21] <cheatr> RainCT: One last question. The source code includes file for Macs and Windows. I know we are not meant to modify the source code, but is it really necessary to distribute these files?
[15:22] <coppro> where's a good guide to library packaging?
[15:22] <RainCT> cheatr: yes, unless there is *a lot* of non-linux stuff, in which case I'd consider repackaging the tarball.
[15:24] <RainCT> coppro: also, there's a tarball on the website (http://handbrake.fr/rotation.php?file=HandBrake-0.9.2.tar.gz). use that instead of getting the source from the SVN
[15:25] <coppro> wrong username :)
[15:25] <albert23> coppro: http://www.netfort.gr.jp/~dancer/column/libpkg-guide/libpkg-guide.html
[15:25] <RainCT> coppro: sorry
[15:25] <coppro> that one doesn't explain the basics.
[15:25] <RainCT> coppro: this might also help https://wiki.ubuntu.com/MOTU/School/LibraryPackaging
[15:25] <coppro> thanks
[15:25] <RainCT> cheatr: ^^
[15:25] <cheatr> RainCT: Yeah, I saw that, and that's what I'm using now. The unpacked .tar.gz is about 5.5 mb. ~3.5 of those are from a macosx folder in it. Should I repack it or leave it as is?
[15:26] <RainCT> cheatr: then it might be worth repackaging it. but better ask for a second opinion
[15:27] <RainCT> cheatr: and if you repack it create a get-orig-source rule in debian/rules in order to automatically repack later versions
[15:30] <cheatr> RainCT: I'm not that familiar with the rules file at this point. So I think that I will first try packaging it as is. Once I figure out how to do that correctly, I can try to repackage it to remove the macosx stuff.
[15:34] <coppro> what do I do for an upstream without a soname?
[15:49] <cheatr> Are there anymore sections than the ones listed here: http://www.debian.org/doc/debian-policy/ch-archive.html#s-subsections? I know that Synaptic shows many more sections.
[15:52] <RainCT> cheatr: afaik, no. Synaptic shows each 3 or 4 times (for each component).
[15:53] <cheatr> RainCT: Yeah, but for example, Synaptic lists a Multimedia section. There is no Multimedia section listed on the page I linked to
[15:55] <RainCT> cheatr: well, it changes the names to be user friendlier
[15:57] <cheatr> RainCT: Thanks again. And do you have any ideas on what I should put for a copyright if the site and original .tar.gz don't mention one anywhere? I have the authors and license, I just don't know who has the copyright
[16:00] <RainCT> cheatr: check in the source files
[16:00] <RainCT> cheatr: for example with:  egrep -i Copyright * */* */*/* */*/*/* */*/*/*/*
[16:00] <StevenK> Or find . -type f | xargs grep -i copyright  ?
[16:01] <StevenK> grep -Hi, actually
[16:01] <laga> RainCT: fgrep -r might be easier than all those wildcards ;)
[16:03] <RainCT> laga: oh, hadn't noticed the -r option. thanks!
[16:04] <tbielawa> are new packages still being put into <= hardy
[16:04] <cheatr> RainCT: None of those commands provide anything
[16:04] <cheatr> tbielawa: No, but you can put them into Intrepid.
[16:04] <RainCT> tbielawa: no, they get into Intrepid now
[16:05] <tbielawa> thanks guys
[16:05] <tbielawa> Is anyone on that can sync the REVU keys?
[16:05]  * RainCT 
[16:05] <tbielawa> :-)
[16:06] <tbielawa> lp/~tbielawa
[16:07] <RainCT> tbielawa: done ;)
[16:08] <tbielawa> RainCT: thank ypu
[16:08] <RainCT> np
[18:53] <tbielawa> so I send an upload to REVU and I spotted something wrong with it. using --force to try and reupload failed, is there another way to push this to revu without bumping the -ubuntu# ?
[18:54] <nixternal> tbielawa: remove the .upload file
[18:55] <tbielawa> nixternal: that seems almost too simple, thanks
[18:56] <nixternal> no prob
[18:56] <tbielawa> it worked :) and revu updates every half hour?
[18:57] <nixternal> is it that long? I never paid attention to that part really
[18:57] <tbielawa> oh
[18:57] <tbielawa> I thought I saw that in conversation once
[19:16]  * tbielawa has been at his console too long. Typing 'exit' into the google search bar does not close your browser
[19:18] <LaserJock> I think it updates every 5-10 min
[19:18] <LaserJock> REVU that is
[19:21] <sebner> LaserJock: I'll mail the GNUStep maintainer group tomorrow when I have my new mail adress ;)
[19:21] <LaserJock> lol
[19:22] <LaserJock> that's assuming your address works tomorrow
[19:22] <sebner> LaserJock: hmm I hope so :)
[19:22] <LaserJock> mine has been messed up for at least 2 weeks now
[19:23] <sebner> LaserJock: buh, maybe because you are not that annoying like me ^^
[19:23] <LaserJock> probably :-)
[19:23] <sebner> ARGH
[19:23] <sebner> afflux: damn ping
[19:23] <afflux> sebner: damn reping :P
[19:24] <sebner> afflux: did I already say that I hate new bugsquad members?
[19:24] <afflux> I think so
[19:24] <sebner> ^^
[19:24] <sebner> afflux: is the new doc ready so they can stop setting my syncs to "confirmed" ?
[19:26] <sebner> jono again =)
[19:26] <LaserJock> what doc are people looking at that says to set them to "confirmed"?
[19:26] <sebner> LaserJock: could it be that the MC mailinglist is somehow ****** up?
[19:26] <LaserJock> sebner: why?
[19:27] <sebner> LaserJock: latest entries are from 2 May
[19:27] <afflux> bdmurray, ScottK and persia (or was it pochu? lost my logs, can't check anymore) had a talk about that on friday. bdmurray seemed to feel uncomfortable with using LP as a workflow tracker at all
[19:27] <LaserJock> sebner: is that a problem?
[19:27] <sebner> LaserJock: Well, I can't follow my application progress :P
[19:28] <LaserJock> sebner: I'm asking, do you know if there has been an email since the 2nd?
[19:28] <LaserJock> that's only 2 days ago
[19:29] <sebner> LaserJock: well, for example norsetto wrote a comment on the 2nd of may and it seems that it arrived today
[19:29] <LaserJock> well
[19:29] <LaserJock> it may have been moderated
[19:29] <sebner> LaserJock: hmm the other mails arrived pretty quickly but nvm
[19:30] <LaserJock> if the people who sent them are subscribed then it goes through right away
[19:31] <sebner> LaserJock: kk, it's just that the 48 hours timeframe for the MC vote on me is now over and I'm so damn excited :)
[19:31] <LaserJock> I see
[19:32] <afflux> sebner: which bug is it this time?
[19:32] <sebner> afflux: the same xD  bug 225669
[19:32] <LaserJock> afflux: I'm sure bdmurray doesn't like using LP as a workflow tracker, but on the other hand it's not really up to him either
[19:33] <sebner> afflux: by Alex Mayorga Adame
[19:33] <afflux> LaserJock: probably. I'm not really sure how to handle it here since some "workflow bugs" are not too easy identifiable by new triagers.
[19:34] <LaserJock> afflux: IMO, then they shouldn't be triaging
[19:34] <LaserJock> triagers and developers need to be working close enough that triagers know what a process bug is and developers can keep and eye on new triagers
[19:35] <afflux> LaserJock: feel free to adjust the HowToTriage page, or tell bdmurray/pedro to do it.
[19:35] <LaserJock> there are obvious things like merges and syncs
[19:35] <LaserJock> which we put in the bug title
[19:35] <LaserJock> that should be a no-brainer to document
[19:36] <afflux> yes, but then we got SRU and MIR which are workflow-things that often (okay, not too often for MIR) integrate with usual bugs
[19:36] <LaserJock> sure
[19:36] <afflux> not sure what other things we have, I'm not a motu.
[19:36] <afflux> (not even hopeful ;))
[19:37] <LaserJock> MIR should be easy to spot
[19:37] <LaserJock> and SRUs will have stable release bug tasks
[19:37] <LaserJock> so they shouldn't be all the difficult
[19:39] <proppy> Hi, regarding prevu, is prevu http://archive.ubuntu.com/ubuntu/pool/universe/x/xmonad/xmonad_0.7-1.dsc the way to test xmonad backport build from intrepid to hardy ?
[19:39] <afflux> I'm not the one who is able to decide for any of those things, but IMHO we end up having a rule like "don't touch bugs containing A, B and C, and maybe D, also check for subscribers E and F and release-tasks"
[19:39] <proppy> (if launched from a prevu-init'ed hardy)
[19:39] <jdong> proppy: prevu lp:xmonad/intrepid
[19:40] <jdong> proppy: or if you add intrepid deb-src lines to sources.list prevu xmonad works too (it uses apt-get source)
[19:40] <proppy> jdong: ho I thought those were for package hosted on launchpad
[19:40] <jdong> proppy: no for anything with an Ubuntu source package
[19:40] <proppy> jdong: I mean those whose source are hosted on launchpad with bzr and stuff)
[19:40] <proppy> jdong: nice
[19:40] <jdong> proppy: it scrapes from ubuntu/release/+source/pkgname :D
[19:40] <jdong> muahaha
[19:41] <proppy> jdong: using beautifulsoup ? :)
[19:41] <LaserJock> afflux: that sounds reasonable :-)
[19:41] <jdong> proppy: using wget and regex :D
[19:42] <afflux> LaserJock: you mean having or not having the rule?
[19:42] <jdong> (read: it's a hack till LP has a proper interface)
[19:42] <LaserJock> afflux: have a rule
[19:42] <LaserJock> afflux: a list of bugs to not worry about
[19:42] <proppy> jdong: yep I don't know if they got +rdf url for package
[19:42] <jdong> proppy: I uploaded a new prevu to Intrepid that supports intrepid chroots, and also now defaults to LP urls
[19:43] <jeromeg> jdong: i backported the whole empathy/telepathy suite to my ppa for Gutsy, if you want to have a look
[19:43] <afflux> LaserJock: having one sounds entirely reasonable to me too, but we should keep it easy. Like adding a [MOTU] to the summary (or something like that).
[19:43] <jeromeg> jdong: it seems to work pretty well
[19:43] <jdong> jeromeg: ok, I'm extremely busy today, I'll tkae a look when I have some time
[19:43] <jeromeg> jdong: no problem
[19:43] <LaserJock> afflux: well, it's not just MOTU
[19:43] <afflux> yes
[19:44] <proppy> jdong: I'm not intrepid enabled yet :)
[19:44] <afflux> was an example ;)
[19:45] <LaserJock> afflux: I'm more in favor of triager education and oversight
[19:45] <LaserJock> I've had a number of "my" bug messed around with by triagers
[19:46] <LaserJock> it veries from low-level irritation to hampering progress
[19:46] <LaserJock> *varies
[19:46] <proppy> jdong: thanks for the help, prevu creating base tarball
[19:46] <jdong> proppy: no probs :)
[19:46] <jdong> proppy: the only reason I mentioned it is because I also use prevu as my pbuilder replacement for test-building to Intrepid
[19:47] <afflux> LaserJock: IIRC we badly need more bug triagers to cope with new bugs. Education is very much needed, but as I said above, identifying the workflow bugs seems only to be easy when you've been involved in them earlier.
[19:48] <proppy> jdong: nice, seems it is still using pbuilder behind the scene, but with a nicer user interface :)
[19:48] <afflux> LaserJock: anyway, again: I'm just a simple bug triager ;) Going to search some food now
[19:48] <LaserJock> afflux: we badly need more triagers who can properly triage
[19:49] <LaserJock> afflux: but yes, it's not always very straightforward to triage
[19:51] <proppy> jdong: should I attach the build log to the bug report ?
[19:51] <highvoltage> yo LaserJock
[19:51] <proppy> jdong: or just notice that it was successfull
[19:52] <i4x> what is 'packaging from scratch'? I want to build my first package with some files (2 scripts, 1 file.c, 1 README and 1 html page).. is 'packaging from scratch' what I need to do? what is the easiest way to build a package?
[19:52] <LaserJock> highvoltage: hiya
[19:53] <LaserJock> i4x: well, generally "packaging from scratch" mean created a new package that hasn't existed before
[19:53] <LaserJock> i4x: as opposed to updating or fixing an existing package
[19:53] <i4x> yes, that's what I want!
[19:54] <i4x> create a new one!
[19:54] <LaserJock> i4x: it can also mean packaging without any sort of helper system (debhelper and cdbs are common ones)
[19:54] <LaserJock> which you don't want to do
[19:54] <i4x> LaserJock: so, it is easier to build with cdbs, right?
[19:55] <LaserJock> well
[19:55] <LaserJock> you might want to go with a debhelper-based package to start with
[19:56] <LaserJock> cdbs can be very easy, but can quickly fail if you need non-standard stuff
[19:56] <LaserJock> i4x: have you read https://wiki.ubuntu.com/PackagingGuide at all?
[19:57] <i4x> I'm reading it!
[19:57] <LaserJock> great
[19:58] <i4x> but it's too much theory!.. and my technical English is not that great!!
[19:58] <LaserJock> ah, sorry
[19:58] <i4x> that's why I've asked..
[19:58] <LaserJock> there's a lot of packaging theory
[19:59] <LaserJock> https://wiki.ubuntu.com/PackagingGuide/HandsOn#head-c8ef3f625afd4722778766bac36a65cefbb6f26c
[20:00] <LaserJock> that might help some
[20:00] <i4x> I've already tried a way!.. and I was pretty close, but there was an error that made me search more and discover the PackagingGuide !!
[20:01] <i4x> but there are some ways and I can't decide what is the better and easier!.....
[20:01] <i4x> I'll take a look!..
[20:02] <LaserJock> the Packaging from Scratch section
[20:02] <LaserJock> it has a lot of theory but it's important stuff to know
[20:03] <i4x> lol, that's where I was! :)
[20:03] <i4x> thx!!
[20:04] <LaserJock> if you have specific questions you can ask them here
[20:04] <i4x> that was a general question that helped some!!
[20:05] <i4x> but I'll do that!! thx again!
[20:23] <ScottK> sebner: My suggested rule was ubuntu-bugsquad/qa not touch bugs that have ubuntu-universe-sponsors, ubuntu-main-sponsors, ubuntu-archive, ubuntu-release, motu-release, ubuntu-sru, motu-sru, or ubuntu-mir subscribed.
[20:24] <sebner> ScottK: sounds good to me
[20:24] <LaserJock> ScottK: or bugs that are assigned?
[20:24] <LaserJock> that's another thing that get's people I think
[20:25] <ScottK> Dunno.  That's true, but bugsquad also uses assignment to keep track of who is triaging.
[20:25] <ScottK> I think you'd need to kill of that practice first (which I would support).
[20:25] <afflux> ScottK: that's obsolete
[20:25] <ScottK> They don't do that anymore?
[20:25] <sebner> ScottK: btw, I have courier debdiff but I have stress at school and I want to ask Mortel 1 or 2 question so you have to wait a little bit, ok?
[20:25] <afflux> ScottK: some "old" triagers do that
[20:26] <afflux> ScottK: but the wiki says subscribing now.
[20:26] <ScottK> sebner: I haven't looked at it, so no rush.
[20:26] <ScottK> afflux: OK.  Good then.  That's progress.
[20:26] <ScottK> LaserJock: Then yes, I'd say or assigned.
[20:26] <sebner> ScottK: kk :) I should be a contributros member so, at least I think so =)
[20:27] <afflux> Assignments are a general problem though. Some people (mostly non-bugsquad people) just press "assign myself" because it sounds fancy.
[20:27] <afflux> (at least I think that's the reason)
[20:27] <ScottK> Or the bug filer assigns MOTU because they think that'll get it fixed faster.
[20:27] <afflux> right
[20:28] <afflux> maybe assigning should be restricted on LPs side to be only available for the bug superviser
[20:28] <sebner> afflux: I personally assign me if I'm working on a merge or something similar though it's also "In Process" then
[20:28] <LaserJock> well, *somebody* should be monitoring triager's activites
[20:28] <afflux> sebner: that's the (imho only) correct way to do it.
[20:28] <ScottK> OTOH, if the bug reporter incorrectly assigning the bug to someone caused a delay in triaging (as in "bonehead, when you do that, the triagers leave it alone") it might prove a deterrent.
[20:29] <ScottK> LaserJock: For packages I'm subscribed to, I try to do that.
[20:29] <sebner> There are always pros and cons. hard to find a good solution
[20:32] <Artemis_Fowl> @LaserJock: hey
[20:33] <mushroomsven> anyone out there?
[20:37] <bobbo> norsetto: you around?
[20:37] <norsetto> bobbo: yes sir
[20:37] <sebner> huhu no0tic
[20:37] <bobbo> norsetto: have i done Bug #226088 right so far?
[20:37] <sebner> huhu norsetto
[20:37] <sebner> argh
[20:38] <bobbo> norsetto: the comments make me wonder if i messed something up?
[20:38] <norsetto> bobbo: no, the guy is a new bug triager and is confused
[20:38] <i4x> LaserJock, dpkg lets us build packages, doesn't it? but it is more difficult, right?
[20:38] <bobbo> norsetto: ah thanks :)
[20:39] <norsetto> bobbo: btw,there is a script you can use to automate sync requests
[20:39] <norsetto> bobbo: requestsync
[20:39] <bobbo> norsetto: yeah jpatrick told me that about 10 seconds after i filed the bug
[20:40] <sebner> norsetto: btw, your comment showed up at the ML today O_o
[20:41] <tbielawa_irl> my bad
[20:41] <norsetto> sebner: ah, took  its time
[20:41] <sebner> wired, yes
[20:41] <norsetto> sebner: if you look for a fun merge, there is flightgear
[20:41] <emgent> heya
[20:42] <norsetto> hi emgent
[20:42] <emgent> norsetto: :)
[20:42] <sebner> norsetto: I should start crying about merges ^^. Well first I have school stuff todo, then courier. hmm ok you convinced me. I add flightgear on my todo list =)
[20:43] <ScottK> norsetto: Can we try to get flightgear fixed in Debian now and maybe sync later?
[20:43] <tbielawa_irl> norsetto: what should i have done on that dx sync bug?
[20:43] <norsetto> tbielawa_irl: nothing, if it is subscribed to u-u-s its fine
[20:44] <norsetto> tbielawa_irl: also, if it is a request from a dev and it is subscribed to the archive its fine
[20:44] <sebner> Ok, ping me when the sync is ready and I'll file a bug then =). school stuff is crying :(  afk
[20:45] <mushroomsven> hello folks.  quick question:  exactly how experienced do you have to be in writing software to become a successful MOTU?
[20:46] <tbielawa_irl> norsetto: thanks for clarifying that. i'll check the activity log next time to see if its been subscribed
[20:46] <i4x>  in "single binary, multiple binary, library, kernel module or cdbs", what is the difference between single and multiple binary?
[20:47] <norsetto> tbielawa_irl: you can also check directly through the subscribers tab; anyway thanks!
[20:47] <highvoltage> multiple binary allows you to build multiple packages from one source package, iic
[20:47] <tbielawa_irl> norsetto: well look at that! didn't even see that before
[20:48] <highvoltage> norsetto: ooh, you're online. I looked for you earlier to tell you that I love your sense of humour :)
[20:48] <norsetto> tbielawa_irl: lp can be VERY confusing :-)
[20:48] <norsetto> highvoltage: what sense of humour!? I'm serious
[20:48] <tbielawa_irl> norsetto: ya, that's becoming more apparent as time passes. especially policy on setting and subscribing blablabla
[20:49] <norsetto> scottk: I think we opened a bug on that already which has never been acted upon
[20:49] <i4x> highvoltage, but are they independent from each other?
[20:49] <highvoltage> norsetto: Subject: "Application for "Ubuntu Universe Contributors" (Stefan Ebner)" ;)
[20:49] <ScottK> norsetto: OK.  Just trying to make sure we use time early in the cycle to get Debian and Ubuntu in sync where we can.
[20:50] <tbielawa_irl> maybe in an effort to better understand policy I'll write some mail client that'll connect to your account that gets your LP & mailing list mail and apply rules for filtering.. alert you if something shouldn't be touched. etc..
[20:50] <highvoltage> norsetto: the final .deb packages? if you choose multiple binary, then yes
[20:50] <norsetto> highvoltage: yes, I'm serious (oh well ....)
[20:50] <tbielawa_irl> grab relevant data from the mail and give you quick links to items like package.ubuntu homepage, related bugs... dunno..
[20:51] <tbielawa_irl> what would you guys like to see for managing mail from LP and the mailing lists?
[21:02] <sebner> norsetto: however. I claim it. just to let you know :)
[21:03] <norsetto> sebner: scottk is right btw, I checked and the guy to whom I asked to report this has never reported it
[21:04] <sebner> ScottK: ok a merge then. I'll prepare a debdiff at weekend. Damn. Sometimes debian folks don't like us :\
[21:05] <sebner> norsetto: ah for you
[21:05] <sebner> ScottK: sry
[21:06] <ScottK> sebner: Spend the time reporting the bug and providing the patch to Debian.
[21:07] <sebner> ScottK: I thought that's useless?
[21:08] <no0tic> sebner?
[21:08] <ScottK> sebner: Why do you think that?
[21:08] <sebner> no0tic: sry. damn tab ;)
[21:08] <no0tic> :)
[21:08] <sebner> ScottK: because norsetto already reported the changes back?
[21:08] <sebner> ScottK: ah sry
[21:08] <ScottK> sebner: No. norsetto asked someone to do that and they didn't.
[21:09] <sebner> ScottK: Checked it now. So no merge and waiting for a sync since I'll report the changes back. I'm fine with that
[21:13] <ScottK> sebner: Yes.  Then if it's not fixed in a month or two we go ahead and merge.
[21:14] <sebner> ScottK: great. /me is writing his todo list and going back to school work
[21:22] <RainCT> bobbo: please add new changelog entries from Debian to your sync request (bug 226088)
[21:23] <bobbo> RainCT: ok :)
[21:24] <RainCT> bobbo, norsetto: and status should be New, or have the rules changed again?
[21:25] <nhaines> Greetings, everyone!  :)  I was wondering if anyon could sync the REVU uploaders keyring?
[21:25] <RainCT> nhaines: sure, one moment
[21:26] <norsetto> rainct: for what I remember status should be new, and will be confirmed by the sponsor if acceptable
[21:26] <RainCT> norsetto: yes. Why have you set it back to Confirmed then? :)
[21:27] <i4x> anyone has time to explain me how to build a package?
[21:28] <nhaines> i4x There have been two really excellent Ubuntu Open Week sessions on packaging.  That's how I learned.
[21:28] <bobbo> RainCT: debian changelog up for bug #226088
[21:28] <norsetto> rainct: because it was set to in-progress
[21:29] <i4x> i'm trying my best to understand the wiki PackagingGuide, but now I'm starting to confuse what seems to be easy!!
[21:33] <i4x> nhaines, could you explain some quick ideas?
[21:36] <RainCT> bobbo: sync request ack'd. (btw, it seems like Debian hasn't added the Homepage field, you may want to ask the Maintainer about this)
[21:37] <RainCT> nhaines: (keyring synced)
[21:37] <bobbo> RainCT: thanks, i'll email the maintainer about it
[21:37] <nhaines> RainCT: mucho thanks!  :)
[21:38] <RainCT> nhaines: you're welcome :)
[21:38] <nhaines> i4x: it's sort of tricky.  :)  Well, you might want to try and check out https://wiki.ubuntu.com/MeetingLogs/openweekhardy/PackagingA
[21:38] <RainCT> bobbo: np, thank *you* :)
[21:38] <bobbo> RainCT: what happens now? Does it need another ACK?
[21:39] <sebner> RainCT: I should also start asking you for acking my syncs :P
[21:39] <i4x> nhaines: I just need to build a package that installs some files (2 scripts, 1 file.c, 1 README.txt and a web-page.html) to their folders!! it seems easy!! and I've almost there doing it with 'dpkg-deb --build debian' !! but then I got an error and latter found wikis' PackagingGuide while searching!! now I'm wondering if there's another way!!
[21:39] <nhaines> i4x: Daniel Holbach went through with a bunch of others and described step by step how to package a program, including how to set everything up.
[21:39] <RainCT> bobbo: no, now just wait for an archive admin to do the sync.
[21:40] <bobbo> RainCT: nice one :)
[21:40] <nhaines> It's not terribly difficult, but maybe seeing the process through will help.  That's the only way I could figure it out!
[21:40] <i4x> thx!!
[21:40] <i4x> nhaines, I'll take a look at it!!
[21:43] <Arby> I'm trying to merge k3b, the report file has 'C* debian/k3b.files' I can find k3b.files.DEBIAN, shouldn't there also be a k3b.files.UBUNTU?
[21:43] <Arby> because I don't have one
[21:43] <Arby> so I don't understand what the conflict is.
[21:46] <crimsun> Arby: because the current k3b Ubuntu source package lacks a debian/k3b.files altogether.
[21:47] <Arby> ok, so adding something new counts as a conflict, I hadn't realised that
[21:48] <Arby> do I just rename it to k3b.files then?
[21:48] <crimsun> if that's the appropriate action, yes.
[21:48] <crimsun> (MoM/DaD is only a starting point)
[21:49] <Arby> well it's my best guess based on https://wiki.ubuntu.com/UbuntuDevelopment/Merging
[21:49] <Arby> from what I've read if I have foo.DEBIAN and foo.UBUNTU I should chose one or the other or combine them
[21:49] <Arby> and I seem to have a choice of 1 here
[21:50] <Arby> that's why I'm confused
[21:52] <Arby> the file in question contains a list of directory paths, I think it describes where various components should be placed.
[21:53] <Arby> do we stick to where debian recommends to place things?
[21:54] <crimsun> yes, unless there's a really good reason to do otherwise
[21:54] <crimsun> there's a debian/KUBUNTU-DEBIAN-DIFFERENCES file worth reading.
[21:55] <Arby> crimsun: I'm just reading it now
[21:56] <Arby> although it doesn't all make sense
[21:57] <crimsun> I've not merged k3b (that I recall); try asking the previous merger
[22:08] <nhaines> RainCT: Now that I've uploaded my package to REVU, I just patiently wait, correct?  :)
[22:11] <RainCT> nhaines: yes. I guess it won't take much (a few weeks perhaps) until people start looking at REVU again, then you can ask here on Mondays (REVU Day :)) for reviews
[22:11] <nhaines> Okay, that sounds good!  :)  Thanks.  This is my first package, so I'm excited.  Hopefully if I do well I can pick up another package or two.
[22:13] <bobbo> RainCT: is tomorrow a REVU day or is it too early in the cycle?
[22:13] <RainCT> bobbo: I haven't seen any announcement so I don't think so
[22:13] <bobbo> RainCT: ok, thanks
[22:14] <RainCT> and if it was a REVU day it would already have started; it's already Monday in some places in the world (not here though) :)
[22:14] <nxvl> bobbo: pong
[22:15] <bobbo> hey nxvl is it OK if i do the apt-torrent merge? You were the last uploader so i thought i'd check
[22:17] <Kopfgeldjaeger> n8
[22:17] <nxvl> bobbo: yes
[22:17] <nxvl> bobbo: go ahead!
[22:17] <bobbo> nxvl: thanks :)
[22:17] <nxvl> bobbo: if you need some help just ask for it
[22:17] <nxvl> :D
[22:18] <bobbo> nxvl: It looks fairly straightforward but thanks a lot :)
[22:21] <RainCT> good night
[22:22] <nhaines> Good night, RainCT.
[22:41] <greg-g> ok, packaging question, it might be a typo/mistake in the guide I'm not sure
[22:41] <greg-g> looking at https://wiki.ubuntu.com/PackagingGuide/Complete
[22:41] <greg-g> at the "dh_make -e your.maintainer@address -f ../hello-2.1.1.tar.gz" step
[22:41] <greg-g> ask you see the dir you are in is hello-debhelper-2.1.1, and thus my Package Name from dh_make is hello-debhelper
[22:41] <greg-g> however, in the guide it is just "hello"
[22:41] <greg-g> is this my mistake or the guide's?
[22:42] <greg-g> (I'm not sure if "thus" is the correct word in that sentence, since I am just assuming that is where dh_make got the name from)
[22:45] <Arby> trying to build a package after a merge by running debuild -S fails with 'debian/rules:63: *** target file `clean' has both : and :: entries. Stop.'
[22:46] <Arby> does that mean the target file is called clean or the target of the section named clean in debian/rules?
[22:46] <superm1> Amaranth, ping
[22:46] <Amaranth> superm1: pong
[22:46] <superm1> Amaranth, wanted to get an idea about how you felt about this before getting the SRU together on it: https://bugs.launchpad.net/bugs/160264
[22:47] <Amaranth> oh yeah, that one
[22:47] <Amaranth> probably should get that into our bzr repo and into intrepid before doing an SRU
[22:47] <Amaranth> talk to mvo tomorrow
[22:47] <superm1> well hence why i'm talking to you before going above your head :)
[22:48] <Amaranth> i'm not core-dev so mvo ends up doing most of the packaging work as all the pieces have to be coordinated
[22:48] <norsetto> greg-g: your tarball name is hello-2.1.1.tar.gz and your source tree top directory is hello-debhelper-2.1.1
[22:48] <superm1> ah, i thought you were
[22:48] <superm1> okay
[22:49] <greg-g> norsetto: correct...
[22:49] <norsetto> greg-g: so, where is the problem?
[22:49] <greg-g> the discrepency between what I see and what the guide shows
[22:49] <norsetto> greg-g: what do you see?
[22:50] <greg-g> Package Name: hello-debhelper
[22:50] <norsetto> greg-g: the package name is from debian/control
[22:51] <greg-g> I thought the debian dir was created by dh_make (this is my first time obviously)
[22:52] <norsetto> greg-g: yes, thats correct
[22:52] <greg-g> so, I guess my question still stands, in the example in the guide, should the Package Name as reported by dh_make be "hello-debhelper" or "hello"
[22:53] <norsetto> greg-g: what is filled by dh-make in debian/control and debian/changelog ?
[22:53] <RAOF> Arby: It sounds like the merge is broken; is there both a clean: and a clean:: target in debian/rules?
[22:54] <greg-g> when I just did it, hello-debhelper
[22:54] <RoAkSoAx> norsetto, hello there!! have a question... should i filter all .po files from a debdiff?
[22:54] <norsetto> greg-g: ok , so the package name will be hello-debhelper
[22:54] <norsetto> RoAkSoAx: yes
[22:54] <RAOF> Arby: Just because there aren't any conflicts, doesn't mean that the merge is correct :)
[22:54] <RoAkSoAx> norsetto, thanks :D
[22:55] <greg-g> norsetto: then should the guide be changed to reflect that?
[22:55] <norsetto> greg-g: let me check
[22:55] <i4x> people.. telling that the copyright is GPL, I won't get any errors from any GPG, right?
[22:55] <greg-g> norsetto: k, thanks
[22:55] <Arby> RAOF: no there is only clean: in /debian rules
[22:55] <Arby> or debian/rules even
[22:56] <i4x> what is GPG?
[22:56] <RAOF> Arby: Want to pastebin debian/rules?
[22:57] <Arby> RAOF: http://paste.ubuntu.com/10098/
[22:58] <RAOF> Arby: Heh.  Merging gone haywire :)
[22:58] <Arby> so it would seem
[22:58] <RAOF> Arby: That rules file is totally broken; it looks like you've merged a CDBS and debhelper rules file together.  Hilarity ensues.
[22:59] <jdong> RAOF: ROFL nice!
[22:59] <jdong> RAOF: good reason to double-check automated merge output :D
[22:59]  * jdong glares at git merge....
[22:59] <Arby> that's what I'm trying to do :(
[22:59] <norsetto> greg-g: just change hello to hello-debhelper in the sample debian/control and debian/changelog, thx
[23:00] <i4x> anyone knows the difference between GPG and GPL??
[23:00] <jdong> Arby: hang in there man, we'll figure it out :)
[23:00] <Arby> RAOF: the untouched version is http://paste.ubuntu.com/10099/
[23:00] <greg-g> norsetto: ok, but now I'm confused :)
[23:00] <jdong> i4x: apples and oranges
[23:00] <RAOF> Arby: Right.  So, you can't merge the rules file; you'll need to pick one (probably the Debian one, unless there's a good reason why we diverge).
[23:00] <jdong> i4x: GPG is a cryptographic system for encrypting/signing stuff. GPL is an open source license.
[23:00] <greg-g> norsetto: oh, you mean on the wiki or my build?
[23:00] <norsetto> greg-g: on the wiki
[23:01] <greg-g> norsetto: gotcha, not confused anymore :)
[23:01] <jdong> i4x: the "whatis" tool at the command line is pretty cool for deciphering 3-letter jargon like this
[23:01] <Arby> RAOF: there is only one rules file, so I removed the Ubuntu section (one line), and left the debian section
[23:02] <greg-g> norsetto: thanks for the help/confirmation on that.  I know it was a very picky question.
[23:02] <norsetto> greg-g: I guess also in the example showing the output of dh_make
[23:02] <RAOF> Arby: In this case the merged file is useless to us; you'll need to manually look at Debian's rules and Ubuntu's rules, figure out why they diverge, and pick one.
[23:02] <RAOF> Arby: Talking to the last person who touched k3b is likely to help in this, as is the changelog(s)
[23:02] <norsetto> greg-g: where it says "Package Name    : hello"
[23:03] <crimsun> sigh, we really should have waited until 8.10 to have PA use hw:*
[23:03] <i4x> jdong: thx..
[23:03] <Arby> hmm, that sounds like a big job best postponed until tomorrow
[23:03] <Arby> it's getting late here
[23:03] <greg-g> norsetto: yep will do
[23:06] <superm1> jdong, so what did you have to say about the handbrake build system?
[23:08] <Amaranth> crimsun: why is that?
[23:08] <RoAkSoAx> norsetto, coudl you check this please? https://bugs.launchpad.net/ubuntu/+source/blam/+bug/226670
[23:08] <crimsun> Amaranth: oh, the whole ALSA regression bits
[23:09] <Amaranth> crimsun: ah, right
[23:09] <crimsun> we've just broken an entire class of audio/video apps that don't have native PA plugins
[23:09] <Amaranth> i thought you meant make it run on dmix, i know nothing about how sound works :)
[23:09] <crimsun> well, yes, that's how it _should_ have been done for 8.04.
[23:10] <crimsun> not to mention that a lot of ice17xx-based cards don't work by default in GNOME
[23:10] <wgrant> Might that be appropriate for 8.04.1 as long as there is a lot of testing?
[23:10] <crimsun> wgrant: I've asked for a lot of testing
[23:10] <wgrant> Where?
[23:10] <crimsun> wgrant: 1928888
[23:10] <crimsun> 192888*
[23:10] <wgrant> -8?
[23:10] <Amaranth> bug 192888
[23:11] <norsetto> greg-g: we can also just modify the dh_make command to force the hello name (dh_make -p hello etc. etc.)
[23:11] <crimsun> note that the proper fix for that bug is really to enable nspluginwrapper for i386 in hardy-proposed
[23:12] <greg-g> norsetto: should we do that instead, since it appears there are other references later in the guide to the package as only "hello" ?
[23:12] <wgrant> crimsun: Won't that only stop Firefox going down with it?
[23:12] <norsetto> greg-g: I think it would be easier
[23:12] <crimsun> wgrant: yes, which is the concern of that bug report
[23:12]  * greg-g nods
[23:13] <wgrant> I would have thought the solution would be to stop Flash crashing. But I guess that's not possible.
[23:13] <crimsun> TheMuso_: which HDA codec (and revision) is on the machine(s) that have the caching issue?
[23:14] <norsetto> RoAkSoAx: I really haven't got the time now, did you talk with asac as I suggested?
[23:15] <RoAkSoAx> norsetto, havent find him online asa
[23:16] <superm1> crimsun, this may be a silly question; has Adobe been made aware of the issue that you know of?
[23:16] <crimsun> superm1: yes, both privately and publicly
[23:16] <norsetto> RoAkSoAx: please do try to talk with him, he is THE mozilla guy
[23:16] <superm1> crimsun, and have they indicated that a resolution is at least in progress?
[23:16] <superm1> crimsun, perhaps with some sort of ETA?
[23:17] <norsetto> g'night all
[23:17] <crimsun> superm1: they've pretty much ignored the issue.
[23:17] <crimsun> SEP, IOW.
[23:17] <crimsun> "not our code, not our problem"
[23:17] <RoAkSoAx> norsetto, ok i'll try to be aware if he comes online, thanks :D
[23:17] <superm1> crimsun, but it has been proven to be a bug in their code as my understanding though?
[23:18] <sebner> gn8 folks
[23:18] <superm1> with it's 'interactions' with "other" people's code
[23:19] <crimsun> superm1: a bug, yes.  The _only_ bug?  No.
[23:21] <superm1> given this release being an LTS, I would hope that there is an OSV/ISV interaction going on from Canonical in terms of prioritizing it, and that just not being publicized, but i'd be reluctant to fully believe that since it still doesn't live in partner, multiverse, or restricted
[23:21] <superm1> (mearly that installer living in multiverse)
[23:22] <TheMuso_> crimsun: Caching issue?
[23:23] <crimsun> TheMuso_: PulseAudio from bzr
[23:23] <crimsun> TheMuso_: Sorry, meaning the sample cache issue
[23:23] <TheMuso_> crimsun: Well one machine I tested with doesn't have hda. Its got standard ac97/intel8x0 module.
[23:23] <crimsun> I've tested six different HDA codecs this weekend running around retail outlets; none of them have the issue so far
[23:24] <TheMuso_> Let me test on another machine right away.
[23:24] <crimsun> TheMuso_: ok, Sigmatel AC'97?
[23:24] <TheMuso_> crimsun: No intel.
[23:24] <TheMuso_> Let me check to be sure.
[23:28] <TheMuso> crimsun: Its Intel ICH4.
[23:28] <TheMuso> Going to test on a machine with an emu10k1 card now.
[23:28] <crimsun> TheMuso: that's what's in /proc/asound/card*/*codec*/* ?
[23:28] <TheMuso> hang on I'll have a look.
[23:35] <TheMuso> crimsun: http://www.pastebin.ca/1007502
[23:36] <crimsun> cool, a ThinkPad?
[23:37] <crimsun> I think I can poke the people at AD
[23:38] <crimsun> TheMuso: ok, I have a testing point now.  I'll try again tonight to reproduce it.
[23:38] <crimsun> TheMuso: thanks.
[23:38] <TheMuso> crimsun: np.
[23:39] <TheMuso> yeah a thinkpad.
[23:51] <TheMuso> crimsun: Same behavior on an emu10k1 based card.
[23:52] <crimsun> TheMuso: ltrace would be useful at this point
[23:52] <crimsun> not that I expect anything eye-opening, but...
[23:52] <TheMuso> crimsun: So should I run gnome-sound-properties through that for example?
[23:53] <TheMuso> Since I can trigger the message/error with that.
[23:54] <crimsun> TheMuso: and through esdplay (in esound-clients), please.
[23:54] <crimsun> sorry, s/through/via/
[23:54] <TheMuso> crimsun: Does that do sample caching?
[23:54] <crimsun> well, libesd* is common
[23:55] <crimsun> gah, lag.  No, it doesn't AFAICT.
[23:56] <crimsun> (basically, see if it's PA that's the culprit, and if so, why using additional libasound features effectively neuters caching)
[23:56] <crimsun> (it doesn't make any sense that it's only reproducible on certain cards, since the PA bit is identical)
[23:57] <TheMuso> crimsun: http://www.pastebin.ca/1007521 for gnome-sound-properties
[23:59]  * TheMuso gathers output for esdctl.