[01:15] <SpamapS> micahg: hey, what ever happened to xulrunner? I kind of lost track of mongo and javascript in general...
[01:16] <SpamapS> micahg: err, I meant libmozjs, not xulrunner
[01:17] <SpamapS> thinking and typing should usually be in sync
[01:17]  * SpamapS makes a note
[01:25] <micahg> SpamapS: so, libmozjs is libmozjs185 in ubuntu, I've meant to merge mongodb, will probably do it over vacation and patch it to use the system mozjs
[01:25] <micahg> they ship their own copy now of mozjs
[01:31] <SpamapS> micahg: yeah, thats the answer to all the worlds' problems now.. embed your own version
[01:31] <micahg> yep, that's what mozilla tells people
[01:32]  * micahg wishes Debian would ship mozjs instead of building it from xulrunner, but oh well
[01:39] <mbiebl> micahg: who is maintaining mozjs 1.8.5? What about security updates etc?
[01:39] <micahg> mbiebl: well, if there's a security update, I'll push it out, but we haven't seen any updates yet
[01:40] <mbiebl> I mean, who maintains mozjs upstream?
[01:41] <micahg> wes garland seemed to be doing a lot with it, idk if he's officially maintaining it or not
[01:42] <mbiebl> micahg: we've been discussing that some time ago on #debian-gnome, and there was the question about how security updates are handled for the mozjs branch
[01:42] <mbiebl> afaik it's based off FF4
[01:42] <micahg> was wondering myself recently
[01:43] <mbiebl> and you can't build FF against mozjs, so we'd have two (different) copies of libmozjs
[01:43] <micahg> well, debian still builds mozjs out of xulrunner
[01:43] <mbiebl> so we'd need to discuss that with the security team and the mozilla maintainers
[01:43] <mbiebl> micahg: yeah
[01:44] <micahg> well, Ubuntu is ok with Firefox having its own copies of things since it gets updated every 6 weeks or faster
[01:44] <mbiebl> well, actually it's build from src:iceweasel
[01:44] <mbiebl> not sure if that is what you meant
[01:45] <micahg> mbiebl: well, isn't xulrunner also built from src:iceweasel now?
[01:45] <mbiebl> seems to be
[08:19] <dholbach> good morning
[10:48] <dr3mro> hello .. I have an Idea for precise+1 : That is to rewrite all current Apps that in python like software center and others in Vala and create an ubuntu development environment that will ease the new developers to create projects for ubuntu .. windows got all the apps because it's easier to develop and maintain projects than in linux ..
[10:48] <dr3mro> we need to rewrite compiz for low memory usage  too so plain ubuntu install shouldn't use more than 300MB as windows 8 airms for 256 Mb only we should be prepaired
[10:48] <dr3mro> else we could write unity for mutter if it's difficult to make compiz rewrite as for now unity is great but memory hug and perormance is uncomparable for gnome shell
[10:50] <dr3mro> I am thinking of Glade + gedit + debugger + compiler + UI in Gtk+-3.0 for managing launchpad repository .. test the compilation before upload .. a form for proposing you app in ubuntu/debian  PPA  .. Documentations ..
[10:51] <dr3mro> ubuntu should adopt vala as the default programing language and invest in Vala as it's the future of ubuntu
[10:52] <who_me> what exactly is wrong with python ?
[10:52] <dr3mro> and OS with an integrated development with good documentations for both the OS and the Dev language is what ubuntu needs
[10:53] <who_me> python doesn't have good documentation ? :)
[10:53] <dr3mro> who_me, python is great but it is slower than vala and memory hug
[10:53] <dr3mro> who_me, no I meant vala has no good doumentations
[10:55] <dr3mro> who_me, imagine an integrated dev environment with glade + text editor + debugger + vala documentations + ubuntu documentations about all the infrastructures of ubuntu like how to message the kernel dbus and all that vage stuff for beginners ..
[10:55] <who_me> you mean "hog" probably" and it's not like the linux kernel is written in python, just some utilities. I'd rather have those be complete, easy to debug etc. then redisgn something for the new programing language that seems to be the "rage" of the day :)
[10:56] <who_me> redesign*
[10:56] <dr3mro> who_me, I love the pythonic way in development but after playing with vala for a few days and see the performance difference and low memory usage I am now a vala +1
[10:57] <brendand> i'd say things are more likely to move towards Qt
[10:57] <who_me> that's what I feel too :)
[10:58] <dr3mro> I think ubuntu should focus on the IDE thing and Document offline so new programmers don't have to google for things that very obvious if they are a developers for windows
[10:59] <dr3mro> brendand, Qt project is great but less free than Gtk and thats why its supported by nokia ..but ubuntu is Gtk ?
[10:59] <who_me> umm, developers for windows are part of the "Visual C, C++, Basic and etc" toolset and mindset, it's quite hard to recreate that on any other platfrom
[11:00] <who_me> it's up to them to use something like the QT SDK and try to target multiple platforms
[11:00] <dr3mro> who_me, what about making something for ubuntu like that ? it will make ubuntu tempting for new developers for linux
[11:01] <brendand> to be honest, it's naive to think the problem is around having the 'wrong' tools
[11:01] <brendand> anyway, what's the say the app development community is weak on Linux/Ubuntu?
[11:01] <brendand> games development, yes, but that's a whole different matter
[11:01] <brendand> and some commercial software
[11:02] <brendand> still a nice dev environment would be, well, nice :)
[11:02] <brendand> anway, this should all be discussed on #ubuntu-app-devel
[11:02] <dr3mro> My idea is a Documentation for both Ubuntu services like DBUS and other stuff with Vala and Gtk + and Gobject and ... + debugger + text editor + glade + a frontend for launchpad and an easy way to propose your app for ubuntu /debian repos ?
[11:02] <who_me> dr3mro: problem is like this, try to find a book about programming about C++ and I bet that 99% will make you use MS Visual studio... nothing on using some other IDEs
[11:02] <brendand> why Vala though?
[11:05] <dr3mro> brendand, vala performance and memory usage very smilar to C and easier to learn and write code in few hours like python and it has syntax similar to C# / Java so developers will find it very familiar and it has  a fork that uses the same compiler but syntax is pythonic  it's called `Genie`  ??
[11:09] <who_me> dr3mro: here's an idea, try to implement some of those ideas yourself, get enough followers and maybe you guys will get noticed.  A spin of Ubuntu with Vala based utilities...
[11:10] <dr3mro> who_me, pinguy OS did that already :)
[11:10] <dr3mro> who_me, gnome 3.0 is written mostly in VALA
[11:11] <who_me> gnome 3 is rubbish, not the best example of usage IMO :P
[11:11] <TiMiDo> can someone tell me if my wiki is ok https://wiki.ubuntu.com/aaronfarias
[11:11] <dr3mro> who_me, you mean gnome shell :) ..
[11:11] <TiMiDo> I'm going for membership ;)
[11:11] <who_me> no, I mean gnome 3 with mutter and extensions as a whole
[11:13] <dr3mro> who_me, I find it faster on my laptop that gnome 2 and stylish .. themes are in CSS and the problem is that it's immature and need a year or two before it matures
[11:13] <who_me> they did a rush release, with most of the functionality from the previous version left out and now we are supposed to "beta test" while they slowly fix things by version 3.6 :)
[11:13] <dr3mro> who_me, agree .. but it works not as gnome 2.x but not that bad
[11:14] <dr3mro> who_me, it's better in quality than kde 4.0 when it was released
[11:14] <who_me> that is why I keep my trusty LTS install and try to help kubuntu and kde iron out bugs :/
[11:15] <dr3mro> who_me, looks like you are preparing to go the KDE way soon ..
[11:15] <who_me> dr3mro: yeah, and what did the gnome guys learn from a poorly managed kde 4.0 release ? nothing.
[11:15] <who_me> they did the exact same thing
[11:16] <dr3mro> who_me, they delayed to release a year and more and you can't iron things out if it's now widely tested
[11:18] <who_me> dr3mro: KDE offers a tablet like interface, but they don't force it on me. They give me the classic desktop interface that I've known for a long time and can easily discover. Gnome 3 + shell just chnaged everything and it like "hey, this is how your desktop will look now. like it or else"
[11:19] <who_me> I tend to get moody when stuff is forced down my throat
[11:20] <dr3mro> who_me, you are right ..
[11:21] <who_me> so while G3 is like, look dudes, your workstation just became a tablet PC
[11:22] <who_me> KDE gives me the option to use my machine for what it actually is, be it tablet or workstation
[11:23] <who_me> I'm actualy old enough to remember how Gnome actually came to be, as an answer to the way Qt was licensed back in the days
[11:23] <who_me> KDE 2.x was beautifull in comparison to Gnome back then
[11:26] <who_me> http://www.visionfutur.com/img/histoire/gnome1-1.jpg  versus  http://www.kde.org/screenshots/images/large/kde2final_4.jpg
[11:29] <mbiebl> micahg: urgh, the ubuntu mozjs package has different symbols files for each architecture
[11:37] <chrisccoulson> mbiebl, that's because the name mangling is different on each architecture :/
[11:38] <mbiebl> chrisccoulson: hasn't KDE the same problem?
[11:38] <chrisccoulson> mbiebl, yeah, i think so
[11:39] <mbiebl> I think they developed a solution for that
[11:39] <chrisccoulson> i'm sure i asked someone how they solved this problem, and this was the result
[11:39] <chrisccoulson> but that was a while ago
[11:39] <mbiebl> iirc MoDaX worked on that
[11:42] <mbiebl> chrisccoulson: http://anonscm.debian.org/gitweb/?p=pkg-kde/kde-sc/kde4libs.git;a=tree;f=debian;hb=HEAD
[11:43] <mbiebl> seems they managed to have a single symbols file
[11:44] <chrisccoulson> interesting
[11:44] <chrisccoulson> i'll try and figure out how :)
[11:45] <doko> Riddell, ScottK: kdeplasma-addons and gwenview are the two real ARM build failures (double/float casts)
[11:59] <mcclurmc> hi all. does ubuntu allow multiple distributions to be specified in the debian/changelog?
[11:59] <mcclurmc> can I specify both oneiric and precise? or do I need to build two separate debs?
[12:01] <geser> mcclurmc: you need two source packages, one for each release
[12:01] <mcclurmc> okay, thanks geser
[12:27] <Riddell> mbiebl, chrisccoulson: http://pkg-kde.alioth.debian.org/symbolfiles.html is how we have one symbols file for all arches
[12:27] <Riddell> doko: ok
[12:40] <mbiebl> Riddell: where does pkgkde-getbuildlogs get the build logs from?
[12:41] <mbiebl> does that mean I need to upload -1 with broken symbols files, the run pkgkde-getbuildlogs and upload fixes symbols files in -2?
[13:11] <Riddell> mbiebl: ignore pkgkde-getbuildlogs it's debian only
[13:12] <Riddell> you can use locally compiled build logs or in buildds but only if there's only 1 library
[13:13] <Riddell> I usuae my on local build logs
[13:47] <doko> Riddell, libmygpo-qt needs a MIR (amarok)
[13:47] <doko> and, can amarok be built without OpenGL for armel/armhf?
[13:48] <doko> https://launchpadlibrarian.net/88144488/buildlog_ubuntu-precise-armhf.amarok_2%3A2.5.0-0ubuntu1_FAILEDTOBUILD.txt.gz
[14:07] <ScottK> lamont: Could you give ross a kick in the shins so we have some hope of powerpc catching up this year.
[14:09] <Riddell> doko: yeah I know, on my TODO
[14:09] <Riddell> (libmygpo-qt, amarok needing no GL on ARM is new to me but I can look at it too)
[14:33] <sabdfl> Daviey, ping
[14:35] <nigelb> He's supposed to be vacationing.
[14:40] <doko> Laney, we should introduce dedicated buildds for haskell :-p
[14:47]  * iulian nods.
[14:52] <doko> iulian, btw, are all the rebuilds needed, or could these a bit better organized to avoid some of the rebuilds?
[14:54] <cjwatson> doko: the current builds are mostly syncs of packages not previously in Ubuntu, not rebuilds
[14:55] <doko> cjwatson, yes, known. just fearing the follow-up of rebuilds
[14:56] <cjwatson> they aren't that bad except for the cases where they sit and hang for a day :-)
[14:56] <cjwatson> http://people.canonical.com/~ubuntu-archive/transitions/ghc.html is looking not particularly awful right now
[14:57] <doko> lets see this page after this round of uploads
[14:57] <zyga> hi
[14:57] <zyga> who is the best person to talk about u1db?
[14:57] <cjwatson> doko: syncs of new packages aren't going to make it worse
[14:58] <cjwatson> zyga: I'd start with aquarius
[14:58] <zyga> thanks
[14:58] <zyga> aww, away
[14:58]  * iulian will keep his eyes open.
[14:58] <cjwatson> lp/~sil
[14:58] <iulian> There will be a few rebuilds to deal with.
[15:34] <james_w> zyga, #u1db
[15:52] <mdeslaur> doko: I'll merge virt-manager (you TIL)
[15:53] <doko> mdeslaur, anything you want =)
[15:53] <mdeslaur> :)
[16:03] <zyga> james_w: thanks
[16:08] <roadmr> hey! is 32-bit Ubuntu installing the pae kernel by default now? is this a recent change or has it been like that for a while?
[16:08] <micahg> roadmr: recent
[16:09] <roadmr> micahg: thanks! I'm triaging a bug from someone installing on a VirtualBox (PAE is disabled by default) - so without that particular bit of information, his report was "ubuntu fails to boot"
[16:09] <micahg> and`: I think the extensions compatible by default landed in 10, not 9
[16:18] <and`> micahg: do we have a firefox 10 already?
[16:18] <micahg> and`: in firefox-next PPA soonish
[16:19] <and`> micahg: ah, I found this out right after installing firefox 9 on Debian sid
[16:20] <and`> micahg: and I got the confirmation that ffox asks for informations to addons.mozilla.org if the network is active
[16:20] <and`> micahg: so that you can enable the addon even if max_version is < than the installed ffox package
[16:21] <micahg> well, in 9 I thought that only works if you have the extension that overrides compatibility installed, unless glandium backported a patch
[16:23] <and`> micahg: I tested it out today, I upgraded iceweasel to version 9, all my addons got disabled, I switched them back to enabled on the addons interface and restarted iceweasel.
[16:23] <and`> micahg: result, all my addons are back and working again without touching install.rdf
[16:24] <and`> micahg: and I don't think glandium backported a patch for that according to what he told me on d-devel before.
[16:24] <and`> I couldn't find an official statement upstream so I don't know :)
[16:33] <Laney> doko: yeah sorry. feel free to score stuff down if you want.
[16:38] <micahg> could i please get an archive admin to copy thunderbird 3.1.16+build2+nobinonly-0ubuntu0.10.04.1 to ubuntu/primary lucid from ubuntu-security-proposed?
[16:40] <micahg> doko: would you be able to do this? ^^
[16:41] <micahg> oops partial copied, that needs to go to lucid-security
[16:53]  * micahg wonders if it's too early for slangasek
[17:05] <micahg> Riddell: could you copy the above for me?
[17:07] <Riddell> micahg: let's see
[17:09] <Riddell> micahg: from ubuntu-security-proposed ppa lucid to ubuntu-security lucid?
[17:09] <Riddell> does it have the necessary QA and security team approval for that?
[17:10] <micahg> Riddell: thunderbird from ubuntu-security-proposed ppa lucid to ubuntu lucid-security, I am the security team and QA approval in this case :)
[17:10]  * micahg already copied maverick and natty, but lucid times out
[17:12] <Riddell> micahg: copy binaries too?
[17:12] <micahg> Riddell: please
[17:14] <Riddell> micahg:  copy-package.py -p ubuntu-security-proposed -s lucid --to-suite=lucid-security -b thunderbird ?
[17:15] <roadmr> SRU question this time: an package in -proposed had two verification failed bugs. What's the best way to handle that? remove those two fixes from the SRU and request another merge?
[17:17] <micahg> Riddell: not sure on the syntax, I just have an old guide on the Archive Admin page, but the from and to look right
[17:17] <ScottK> That or fix the fix so it works.
[17:19] <micahg> Riddell: according to that, it should be copy-package.py -b --ppa=ubuntu-security-proposed -s lucid --to-suite=lucid-security -e 3.1.16+build2+nobinonly-0ubuntu0.10.04.1 thunderbird
[17:19] <roadmr> ScottK: for one of the bugs the fix is in the devel version, I just messed up while adding it to the SRU
[17:20] <roadmr> ScottK: but for the other one, the bug is still present in development, so we'd have to fix it there and then push that to the SRU again
[17:20] <ScottK> Right, so I'd drop that one and resubmit, I guess.
[17:20] <Riddell> micahg: done
[17:21] <roadmr> ScottK: ok, I'll do that then. Thanks!
[17:21] <micahg> Riddell: thank you much
[17:21] <micahg> slangasek: unping
[17:34] <slangasek> micahg: not too early, just to on-the-phoney, sorry :)
[17:36] <doko> ScottK, gwenview, see bug #907735
[17:36] <ScottK> doko: Already fixed.
[17:37] <doko> ScottK, please close it after verifying =)
[17:37] <ScottK> OK
[17:46] <doko> make[1]: Entering directory `/build/buildd/drush-4.5'
[17:46] <doko> sh: 1: wget: not found
[17:46] <doko> sh: 1: curl: not found
[17:46] <doko> people should hide things better ...
[17:46] <doko> gah, should be make[1]: Entering directory `/build/buildd/drush-4.5'
[17:46] <doko>  /build/buildd/drush-4.5/drush.php > /dev/null
[17:46] <doko> sh: 1: wget: not found
[17:46] <doko> sh: 1: curl: not found
[17:47] <cjwatson> I thought I fixed that
[17:47] <cjwatson> oh, no, there was a Debian bug about it
[17:48] <cjwatson> grr, it was working and then the Debian maintainer broke it again
[17:48] <cjwatson> missing build-dependency on php-console-table I believe
[18:09] <adam_g> is there any way to have a lp PPA build satisfy its Build Depends from within the PPA?
[18:13] <mdeslaur> adam_g: that should already be the case AFAIK, just make sure your build depends are finished building before uploading the other ones
[18:21] <adam_g> mdeslaur: ok, ill try again. thanks
[18:27] <ScottK> Nice.
[18:28] <ScottK> lamont: Ross fell over and can't get up again.  Is something up or just unlucky today?
[18:52] <doko> Riddell, ScottK: I think we need to improve the kde situation for build-deps/breaks
[18:52] <doko> e.g. kdenetwork fails with:
[18:53] <ScottK> doko: Soyuz just needs to implement the same BD unistallable condition Debian has and it'll be fine.
[18:53] <doko> The following packages have unmet dependencies:
[18:53] <doko>  kde-sc-dev-latest : Breaks: kde-workspace-dev (< 4:4.7.90) but 4:4.7.3a-0ubuntu4.1 is to be installed
[18:53] <doko>                      Breaks: kdepimlibs5-dev (< 4:4.7.90) but 4:4.7.3-0ubuntu2 is to be installed
[18:53] <doko>                      Breaks: libkonq5-dev (< 4:4.7.90) but 4:4.7.3-0ubuntu2 is to be installed
[18:53] <doko> E: Unable to correct problems, you have held broken packages.
[18:53] <doko> apt-get failed.
[18:53] <doko> and doesn't automatically retry
[18:53] <ScottK> Yep.  On Debian that would get a BD uninstallable state, not a failed build and would eventually work.
[18:54] <doko> if you know that, don't use it
[18:54] <doko> just adjust the version for the b-d's
[18:54] <doko> ScottK, is there a bug about this?
[18:54] <ScottK> So you want us to fork the KDE packages from Debian?
[18:54] <ScottK> I don't think that's going to happen.
[18:55] <doko> ScottK, no, contribute to it. these adjusted versions don't hurt debian
[18:55] <ScottK> No, the versions are correct.
[18:56] <ScottK> Once kde-workspace is built on armel, I'll retry it.  It'll be fine.
[18:56] <doko> ScottK, it's not about the retrying
[18:56] <doko> ScottK, is there a bug report open about this?
[18:56] <ScottK> Yes.
[18:57] <doko> which one?
[18:57] <ScottK> Let me see if I can find it.
[18:57] <ScottK> doko: Bug 527245
[18:58] <ScottK> AIUI, 'Low' in LP bug terms means they aren't going to do it.
[18:59] <doko> thanks for the pointer
[18:59] <micahg> it means it's not scheduled, opportunistic devs can fix I think
[18:59] <ScottK> Could be.
[19:00] <ScottK> doko: No problem.
[21:25] <adrian_berg> If you tinker with the kernel, you are then not able to upgrade to future ubuntu releases, correct?
[21:26] <adrian_berg> My boot time is pathetic, and this is a new laptop (4gb ram, dual core 2ghz processors), shouldn't be having these problems, so my only guess is that there's too much crap being loaded in the kernel
[21:31] <RAOF> adrian_berg: Have you done any measurements?  I don't think kernel load time (except, perhaps, initramfs load time) is going to be a large part of your boot time.
[21:43] <SpamapS> http://reports.qa.ubuntu.com/reports/boot-speed/
[21:43] <SpamapS> adrian_berg: note the results there
[21:44] <SpamapS> kernel is quite tiny
[21:44] <SpamapS> plumbing is the pain point
[21:44] <SpamapS> udev, mounting, fsck'ing, etc
[21:44] <ScottK> Funny how when we forget to care about boot time it goes up.
[21:45] <SpamapS> its just like my waistline
[21:45] <SpamapS> the minute the diet is over, I gain 10lbs back. :)
[21:46] <SpamapS> I would be interested in the breakdown of what constitutes plumbing
[21:47] <ScottK> I guess you could run bootchart yourself.
[21:48] <SpamapS> no the bootcharts are linked there
[21:49] <SpamapS> looks like a lot of it is ureadahead
[21:49] <infinity> ScottK: That bug mostly misses the point (as does BD-Un in Debian) 99% of the time.  You can almost always express a BD-U state as a dep-wait, it just requires drilling down.
[21:49] <ScottK> infinity: Implementation detail.
[21:49] <infinity> ScottK: And I may well fix *that* in lp-buildd, rather than implementing another build state.
[21:49] <ScottK> That's fine.
[21:50] <ScottK> I don't care what you call it.
[21:50] <RAOF> SpamapS: Yeah, but time speant in ureadahead is (or should be) directly offset by time not spent at other points of the boot process.
[21:50] <ScottK> (and I agree, BD uninstallable is a sub catagory of depwait)
[21:51] <SpamapS> RAOF: the lucid ones hit 93MB/s w/ ureadahead
[21:51] <SpamapS> RAOF: and the more recent ones are down around 86MB/s
[21:51] <RAOF> :(
[21:51] <SpamapS> is it just as simple as the drivers being more careful and underperforming?
[21:52] <SpamapS> The numbers work out quite well % wise
[21:53]  * SpamapS decides to blame firmware
[21:57] <infinity> SpamapS: Blame Apple.  That's my default fallback for everything now.
[21:59] <psusi> SpamapS: you gotta pack the data with e2defrag ;)
[22:06] <zyga> hi, where can I find some canonical/ubuntu media packs
[22:06] <zyga> stuff like the ubuntu logo
[22:08] <jpds> zyga: http://design.canonical.com/brand/
[22:08] <zyga> jpds: thanks
[23:02] <hallyn> @pilot in
[23:37] <hallyn> jinkeys!  suddenly no aufs module?  must have happened with today's update?   huh...
[23:40] <RAOF> hallyn: That's been the case for some time; overlayfs (mostly) replaces it.
[23:42] <hallyn> i expected that one day, but figured schroot/sbuild would quietly dtrt when that happened :)
[23:50] <hallyn> here's hoping that sed -i 's/aufs/overlayfs/' * in /etc/schroot/chroot.d does the trick
[23:51] <hallyn> \o/