2012/02/01/#ubuntu-desktop.txt

[00:31] <jasoncwarner_> hey jbicha , I saw your name on the freerdp upload...I can't update right now b/c it is broken, any idea what is going on there? wanted to get the latest Unity goodness and start testing ;)
[00:33] <jbicha> jasoncwarner_: yes, I noticed libfreerdp0 conflicts with libfreerdp1 but I didn't make time yet to figure out how to fix that
[00:34] <jasoncwarner_> jbicha: cool, thanks.
[00:38] <jbicha> jasoncwarner_: you should be able to force remove freerdp (if you don't need remmina) so that the rest of your updates install
[00:40] <micahg> jbicha: libfreerdp1 shouldn't be shipping files in /usr/share
[00:40] <jbicha> micahg: right
[00:41] <micahg> are the keymaps backwards compatible?
[00:45] <jbicha> micahg: I don't know; the git repository has them not touched since the git migration in June
[00:45] <jbicha> https://github.com/FreeRDP/FreeRDP/tree/master/keymaps
[00:46] <micahg> are they used by other things that use the library or just freerdp itself?
[00:47] <jbicha> well, remmina-plugin-rdp is the only rdepends
[00:49] <micahg> jbicha: http://www.debian.org/doc/debian-policy/ch-sharedlibs.html#s-sharedlibs-support-files
[00:49]  * micahg files an RC bug in Debian
[00:50] <jbicha> RC bug? freerdp-1.0 is still in NEW
[00:50] <micahg> I'm getting a head start :)
[00:50] <micahg> ah, it might not pass NEW, ok, I'll wait :)
[00:51] <jbicha> well the 0.8 is broken so I guess it still makes sense
[00:51] <micahg> the ftpmasters should throw it out
[00:52] <micahg> jbicha: BTW, the fix is in the policy manual as well
[00:52] <jbicha> libfreerdp-tools is fine?
[00:53] <micahg> jbicha: you end up with an issue if the keymaps change between releases
[00:56] <micahg> depends on the relationship between the library and the keymaps
[00:56] <micahg> *possible issue
[00:56] <jbicha> I'm a bit confused then as to how to fix this; do we have to break libfreerdp0?
[00:57] <jbicha> unfortunately, remmina's not made a tarball for new freerdp yet
[00:58] <micahg> jbicha: did you read the second paragraph of the policy snippet?  it gives advice on how to fix this issue
[00:59] <jbicha> oh, I got confused by the 3rd paragraph
[01:00] <micahg> yes, that's true, as I said, I don't know the relationship between the keymaps and the library, so if that applies, it should be dealt with appropriately :)
[01:01] <Nafai> RAOF: Around?  You still working on maintaining X these days?
[01:02] <RAOF> Nafai: Indeed.
[01:02] <RAOF> (Well, not *right* now; right now it's colour management ☺)
[01:02] <Nafai> Sweet, I can bug you. :)
[01:03] <Nafai> This isn't the X packages directly, but mesa is close, right?
[01:04] <Nafai> Anyway, if I'm reading https://bugs.launchpad.net/ubuntu/+source/mesa/+bug/823062 correctly, s3tc is not supported with the shipped mesa implementation of Ubuntu right now
[01:04] <RAOF> Yup.  We also maintain those :)
[01:04] <ubot2`> Launchpad bug 823062 in mesa "Please provide Mesa with floating point texture support and GL_S3_s3tc support in Jockey, like proprietary drivers" [Wishlist,Confirmed]
[01:04] <Nafai> And the only way to get that is to use a proprietary driver?  And if you hvae an Intel card you are out of luck?
[01:05] <RAOF> Floating point textures are enabled as of the mesa 8.0 upload; we've got techboard signoff about that feature.
[01:05] <Nafai> so that's coming in Precise?
[01:05] <RAOF> S3TC requires an external library.  That's available in xorg-edgers; I'm not sure if it's in the archive.
[01:06] <RAOF> Yup, coming soon to a Precise near you (it's staged in x-staging, will be copied after A2 release).
[01:06] <Nafai> yeah, I couldn't find anything
[01:07] <Nafai> is there a better bug report or something official that I can point people to to say that s3tc isn't supported?
[01:07] <RAOF> Yup.  We don't seem to have libdxtn in the archive.
[01:08] <RAOF> That could be resolved by actually uploading a libdxtn, though.
[01:19] <DBO> RAOF
[01:19] <RAOF> Yah?
[01:19] <DBO> (sick of me yet?)
[01:20] <Nafai> RAOF: Also, if you have an intel card, the mesa gl implementation is all that is available, right?
[01:20] <DBO> crazy thought...
[01:20] <RAOF> Nafai: Yup.  mesa or nothing.
[01:20] <DBO> is it possible to rotate the mouse sprite?
[01:20] <Nafai> RAOF: Thanks for the confirmation.
[01:22] <RAOF> DBO: What do you mean by “rotate”?  It is possible to replace the sprite with a different one.  I also think the xrandr rotation code rotates the pointer, but don't quote me on that.
[01:22] <DBO> RAOF, well I was thinking about the edge problem we have
[01:23] <DBO> and the fact that the pointer being angled like it is, is actually a rather arbitrary fact of history...
[01:23] <DBO> so...
[01:23] <DBO> how cool would it be if the mouse pointers angle was simply based on its X coordinate in the current monitor
[01:23] <DBO> so as you move left to right, it would slowly rotate
[01:23] <DBO> I mean... that would be SWANKY
[01:24] <RAOF> I'm not really sure how I feel about that. :)
[01:24] <DBO> you love it
[01:25] <DBO> but more importantly
[01:25] <DBO> is it possible?
[01:25] <RAOF> I'm pretty sure you do that client-side.
[01:26] <RAOF> This is totally the sort of thing that is unlikely to be acceptable upstream :)
[01:26] <Omega> that's what makes it fun!
[01:31] <RAOF> DBO: Oh, of course it's possible!  Just hide the pointer and have compiz draw it!
[01:31] <DBO> but that means a screen redraw on every pointer move
[01:32] <DBO> thats annoying
[01:32] <DBO> I'd rather have the video card do it
[01:32] <RAOF> Yes.
[01:32] <RAOF> Right.  I don't believe there's a way to do this in a client without a bunch of context-switching.
[01:33] <RAOF> So it's *entirely* possible - even easy - to do client-side, it's just not as efficient as doing it server side.
[01:36] <DBO> RAOF, lets just make the xserver do it :)
[01:36]  * DBO is goign to keep pushing until you say no
[01:37] <RAOF> I don't think it's a good idea to do it in the server until you've thoroughly prototyped it client-side.
[01:37] <RAOF> You can do all the prototyping client side; the server side is just an optimisation.
[01:38] <RAOF> I'm not sure that it's even a good *idea*, and I don't think you can tell until you've actually tried it :)
[01:39] <Omega> maybe you should check how feasable it is on wayland
[01:40] <RAOF> Trivially.
[01:40] <RAOF> But that's not the point :)
[01:41]  * RAOF is thinking he might need to do a rotation to DX
[01:44] <RAOF> DBO: I'm heading out to an appointment.  Show me a prototype that works and we can talk about efficiencying it up in the server :P
[01:45] <DBO> RAOF, I love how you take my crazy ideas seriously
[01:45] <DBO> :D
[01:45] <DBO> I seriously though you were going to giggle at me, but now I realize, you're just being nice :)
[02:29] <lifeless> RAOF: btw, my themes, still awol
[02:29] <lifeless> RAOF: I could learn to love the grey, if I had icons for sound, power-menu etc
[02:42] <DBO> RAOF, can we get your packages into Ubuntu for A2
[02:42] <DBO> they make a big impact
[02:42] <DBO> didrocks says its up to you
[02:58] <jbicha> micahg: fixed freerdp is in incoming now and I don't have upload rights for it
[03:00] <micahg> jbicha: cyphermox is piloting, he could upload (doesn't appear to be seeded)
[03:00] <jbicha> we just dropped the keymaps; upstream might drop them too
[03:00] <cyphermox> yo
[03:01] <cyphermox> jbicha: can you tell me more? incoming?
[03:01] <jbicha> cyphermox: http://incoming.debian.org/freerdp_1.0.0-2.dsc fixes bug 924393
[03:01] <ubot2`> Launchpad bug 924393 in freerdp "package libfreerdp1 (not installed) failed to install/upgrade: trying to overwrite '/usr/share/freerdp/keymaps/xfree86', which is also in package libfreerdp0 0.8.2-2build1" [Undecided,Confirmed] https://launchpad.net/bugs/924393
[03:01] <cyphermox> ah!
[03:27] <broder> jbicha: i think hallyn was working on kvm-with-spice support
[03:27] <broder> my recollection is that he was planning to do it as a separate source package in universe
[03:29] <hallyn> apt-get install qemu-kvm-spice.  shoudl work, and work with libvirt if you set the right video stanza in xml
[03:29] <hallyn> you do have to say '-enable-kvm' as it comes from the qemu.git upstream (which disables acceleratoin by default)
[03:30] <hallyn> sorry, i have to go - if you have any troubles pls do file a bug against qemu-kvm-spice
[03:31] <jbicha> hallyn: thanks, I didn't know where to look for that!
[03:39] <cyphermox> omg, wow. compiz/unity died and doesn't want to start anymore :)
[03:39] <cyphermox> jbicha: confirm, freerdp I can just drop ubuntu changes (looks like it was just an upload of debian packaging before it was actually released
[03:42] <jbicha> cyphermox: that would be great, thanks!
[03:42] <cyphermox> tarball mismatch
[03:42] <cyphermox> fun :)
[03:42] <jbicha> what does that mean?
[03:42] <broder> cyphermox: i thought that syncpackage could deal with that
[03:42] <cyphermox> broder: oh, it can
[03:42] <cyphermox> I was trying to debdiff
[03:59] <cyphermox> TheMuso: thanks for the merge request for nm-applet! I'll upload that as soon as we're past freeze
[04:01] <cyphermox> jbicha: uploaded freerdp 1.0.0-2fakesync1
[04:14] <RAOF> DBO: Will the multimonitor stuff land for A2?
[04:14] <DBO> RAOF, yes
[04:15] <RAOF> DBO: Well, in that case, yes.
[04:25] <TheMuso> cyphermox: np, thanks.
[04:28] <pitti> Good morning
[04:31] <pitti> jbicha: argh, so on that day when we upload git Debian finally does the upload..
[04:32] <pitti> jbicha: I just wonder how it could end up as a fakesync, I thought I used the upstream tarball
[04:40] <cyphermox> good morning pitti
[04:40] <cyphermox> (omg, it's really time I leave now)
[04:40] <pitti> hey cyphermox
[04:40] <TheMuso> Morning pitti
[04:43] <jbicha> pitti: the DD said he doesn't really pay any attention to Ubuntu so he had no idea we uploaded yesterday
[04:44] <pitti> jbicha: right, that's fine for him, I just wondered why the tarball differs
[04:45] <pitti> jbicha: I took the tarball that was in the git branch (upstream/pristine-tar)
[04:45] <pitti> anyway, there will hopefully be a 1.0.1 or 1.1 at some point
[04:47] <cyphermox> pitti: to answer your question about what you sponsored for me, it was procps and ntrack. --> http://paste.ubuntu.com/824748/
[04:47] <pitti> wow, I thought I did some more
[04:49] <cyphermox> yeah, I thought so too
[04:49] <cyphermox> those were, all in all, pretty simple
[05:25] <desrt> agateau: good morning? :)
[05:27] <desrt> agateau: just incase you're still working on the pathbar thing...
[05:27] <desrt> agateau: i just discovered that GtkButtonBox has a 'linked' style class that you can set
[05:34] <jbicha> you should see how bad Boxes looks in Ambiance
[05:36] <cyphermox> pitti: verification failed for one bug in checkbox, could you please remove it from proposed or whatever, so I can upload an update in the my morning?
[05:36] <cyphermox> pitti: also, it's a native package, can I just re-upload the same version number after that and have it work?
[05:37] <pitti> cyphermox: no need to remove it, just upload another version
[05:37] <pitti> cyphermox: no, you need to bump the version number
[05:37] <cyphermox> ah.
[05:37] <pitti> cyphermox: either keep the old changelog, add a new one, and upload with -v to include the two topmost ones
[05:37] <pitti> cyphermox: or just bump the changelog's current entry
[05:38] <cyphermox> sure
[05:38] <pitti> the first is a bit clearer
[05:38] <cyphermox> oh, right
[05:38] <pitti> but you do need to include both changelog records
[05:38] <cyphermox> duh. otherwise people who might already have it won't get an update
[05:38] <cyphermox> what do you mean both changelog records?
[05:39] <pitti> cyphermox: also, LP will just outright reject it because that versin is already published :)
[05:39] <cyphermox> yeah
[05:39] <pitti> cyphermox: debuild -S -v<package_version_currently_in_release_or_updates>
[05:39] <cyphermox> oh, yeah
[05:39] <cyphermox> that's fine :)
[05:39] <cyphermox> it really is wayyyy too late for me now :)
[05:40] <tjaalton> jbicha: read your blog post, good stuff. it would be _really_ nice to get the new settings stuff in though, if not only for the wacom config stuff ;) (though I understand if FF is too close..)
[05:41] <jbicha> tjaalton: the wacom stuff could be backported
[05:41] <tjaalton> jbicha: oh? that would be cool
[05:42] <jbicha> well I think it could, somebody should look into that at least
[05:42] <tjaalton> I might check it out
[05:46] <cyphermox> pitti: thanks. now it's in the queue for approval -- and I'm going to bed :)
[05:46] <pitti> cyphermox: sleep well!
[05:46] <pitti> cyphermox: that was checkbox?
[05:46] <pitti> http://launchpadlibrarian.net/91594459/checkbox_0.12.10_source.changes
[05:46] <cyphermox> yes
[05:46] <pitti> so you just bumped the version number then?
[05:46] <pitti> it only has one chagnelog entry
[05:46] <cyphermox> yes
[05:47] <pitti> it doesn't include the changelog from https://launchpad.net/ubuntu/oneiric/+source/checkbox/0.12.9
[05:47] <pitti> cyphermox: so all the bug refs and descriptions from that are "lost"
[05:47] <pitti> cyphermox: debuild -S -v0.12.8
[05:48] <cyphermox> but they are the same copied over again in 0.12.10?
[05:48] <pitti> cyphermox: (but fine to do it tomorrow)
[05:48] <pitti> cyphermox: yes; people having -updates will get the changes from .9 and .10 and shoudl see both in update-manager
[05:48] <cyphermox> ah
[05:48] <pitti> also, our trackign machinery relies on the .changes files to state all bugs fixed
[05:48] <pitti> (rejected)
[05:48] <cyphermox> I are teh fail :)
[05:48] <pitti> cyphermox: don't worry
[05:48] <cyphermox> sure, and I'll look at this again tomorrow
[05:49] <pitti> cyphermox: just use -v0.12.8 when building the source, and all will be good
[05:49] <pitti> (no need to bump version again now, as the pacakge was not accepted)
[05:49] <pitti> cyphermox: sleep well!
[05:49] <cyphermox> I got confused probably because the merge is against lp:ubuntu/oneiric/checkbox, not lp:ubuntu/oneiric-proposed/checkbox
[05:49] <cyphermox> this is what goes wrong -- 0.12.9 isn't in this branch
[05:50] <cyphermox> see you later
[05:50] <pitti> ah
[07:09] <didrocks> good morning
[07:16] <pitti> hey didrocks
[07:16] <pitti> didrocks: running unity PPA since yesterday; runs really smooth!
[07:16] <didrocks> pitti: nice! Guten Morgen :)
[07:16] <pitti> the control-center radio button for top left corner activation doesn't work, but with the "push over edge" change it doesn't get in the way any more
[07:17] <didrocks> no specific issue? Revealing the launcher is easy for you?
[07:17] <RAOF> didrocks: Good morning!  The multi-monitor stuff is landing for A2, right?
[07:17] <pitti> didrocks: works well enough with the mouse, but honestly I mostly use the super key
[07:17] <didrocks> pitti: yeah, the g-c-c radio button is wired to the option, but the option is now broken in unity (this is known and jason smith is aware about it)
[07:18] <didrocks> RAOF: good morning! 5.2 will land post alpha2 (as soon as alpha2 is released)
[07:18] <didrocks> pitti: ok, we have some issues in some system where the trackpad is quite slow
[07:18] <didrocks> and so you have to "push hard"
[07:18] <didrocks> I think jason spoke with RAOF about it :)
[07:18] <pitti> didrocks: ah, haven't tried with touchpad, just with normal mouse
[07:18] <didrocks> really depends on the velocity of your touchpad
[07:19] <didrocks> seems I was the first to get a bad one on the xt2
[07:19] <pitti> (gosh, Lana del Rey has a godly voice -- just listening to her new album)
[07:21] <RAOF> didrocks: And the updated X server in ppa:raof/help-jason improves it for you, right?
[07:22] <didrocks> RAOF: the one from yesterday? (did you change it since?)
[07:22] <RAOF> Yeah, that's yesterday's server.
[07:22] <didrocks> RAOF: no, the normalize didn't change anything for the touchpad for me, unfortunatly
[07:22] <RAOF> (ubuntu10~hacking1)
[07:22] <didrocks> not sure for jono, when I left yesterday, I saw that jono had the same issue
[07:22] <RAOF> didrocks: Fix your acceleration settings :)
[07:23] <jono> hey didrocks
[07:23] <didrocks> RAOF: well, I don't think we want to tell everyone to do that :)
[07:23] <didrocks> jono: still around? :)
[07:23] <didrocks> hey!
[07:23] <jono> didrocks, indeed :-)
[07:23] <RAOF> :P
[07:23] <didrocks> jono: did you fix your issue?
[07:23] <jono> hacking on the Ubuntu Accomplishments stuff still
[07:23] <jono> didrocks, nope, same issue
[07:23] <didrocks> ah nice ;)
[07:23] <didrocks> hum
[07:23] <didrocks> even with RAOF's ppa then?
[07:23] <RAOF> Bah.
[07:24] <didrocks> RAOF: I really saw no improvment at all with it
[07:24] <didrocks> RAOF: is the patch applied? (just in case :))
[07:24] <RAOF> didrocks: Yes, because it's a modification of the existing pointer_barrier patch :P
[07:24] <jono> I ran RAOF's PPA from earlier today
[07:25] <jono> didnt fix it, so removed it
[07:25] <jono> the help-jason PPA
[07:25] <didrocks> so we clearly have a release blocker there :/
[07:25] <didrocks> here*
[07:25] <RAOF> Can you please file a bug?  I'll work out what information I need and then ask for it there :)
[07:26] <BigWhale> Morning everyone.
[07:26] <didrocks> RAOF: the acceleration and sensitivity on my xt2 is at the lowest possible value FYI
[07:27] <didrocks> RAOF: and its a fresh precise install from Monday (especially for it)
[07:27] <didrocks> hey BigWhale
[07:27] <didrocks> not sure how those values are decided…
[07:27] <RAOF> Right.  I can't believe you can find that touchpad usable with such low velocity! :P
[07:27] <didrocks> RAOF: well, the xt2 machine is my spare machine TBH, I don't really use it apart from testing :)
[07:28] <didrocks> but there is something setting this wrong
[07:28] <didrocks> maybe jono can confirm for the g-c-c option?
[07:28] <jono> I don't think it is just the touchpad
[07:28] <jono> I had the same issue with the mouse
[07:28] <didrocks> jono: mouse velocity
[07:28] <didrocks> jono: gnome-control-center
[07:28] <didrocks> -> mouse
[07:29] <didrocks> you have "cursor speed" (or something like that)
[07:29] <didrocks> with 2 sliders
[07:29] <jono> right
[07:29] <jono> Pointer Speed
[07:29] <didrocks> are the sliders at the beginning of the range?
[07:29] <jono> yep
[07:29] <didrocks> ok, so clearly same issue than on my xt2
[07:29] <BigWhale> Whoever got rid of those big dash icons gets a beer or a dozen of them, from me ... :>
[07:30] <didrocks> RAOF: do you know what decides to put so slow values?
[07:30] <RAOF> That's just the defaults.
[07:30] <pitti> there's a truly bad bug in zeitgeist
[07:30] <RAOF> It turns out that *my* sliders are also there; it's just that my touchpad is non-crap (apparently!)
[07:30] <didrocks> RAOF: it's not hw dependant?
[07:30] <didrocks> ah
[07:31] <didrocks> so apparently, I put it at a higher value at some point, but don't remember :)
[07:31] <didrocks> pitti: what happens?
[07:31] <pitti> it knows that I watched episodes 2-15, 2-16, and 2-17 recently -- it should totally offer me to watch 2-18 next!
[07:31] <pitti> :-)
[07:31] <didrocks> ahah ;)
[07:31] <didrocks> I guess it's not battlestar gallactica? ;)
[07:31] <pitti> no, Enterprise
[07:31] <pitti> haven't seen most of those yet
[07:32] <pitti> so those are my lunch break entertainment these days :)
[07:32] <didrocks> I'm watching caprica now. Quite interesting, very different from gallactica :)
[07:32] <didrocks> heh
[07:32] <pitti> I watched three or four Galactica episodes, but never really got into it
[07:32] <pitti> maybe I'll try again once I'm through Enterpreise
[07:33] <pitti> I also have big bang theory 5.14 to watch, but it always takes so long until the next one arrives :)
[07:33] <didrocks> oh, only 3 or 4 galactica? We watch the whole serie with Julie in 3 months :)
[07:33] <RAOF> Galactica >> Enterprise
[07:33] <pitti> didrocks: well, apparently _your_ spouse is into scifi, which mine isn't
[07:33] <didrocks> heh, I'm putting them in a stack as well so that we can watch at dinner time :)
[07:33] <pitti> RAOF: still, I maintain that TNG >> everything else
[07:34] <RAOF> Well, most of Galactica :)
[07:34] <didrocks> RAOF: someone with good taste :)
[07:34]  * RAOF is almost at the end of TNG S3
[07:34] <pitti> although I must say that Enterprise is the most credible of all those
[07:34] <didrocks> pitti: yeah, I turned her into that TBH :)
[07:34] <BigWhale> and I am watching Firefly (just saying, so I don't feel left out from the debate)
[07:34] <pitti> never heard of that -- wikipedia says "space western"?
[07:34] <RAOF> pitti: Have you got to the episode where they GO BACK IN TIME TO FIGHT NAZIS?  WHY IS IT ALWAYS WITH THE GOING BACK IN TIME TO DEFEAT THE NAZIS *AGAIN*?
[07:35] <pitti> horses with warp drive?
[07:35] <RAOF> pitti: You haven't seen Firefly?
[07:35] <pitti> RAOF: on Enterprise? no, not yet
[07:35] <RAOF> pitti: Spoiler alert: they have a *stupid* time travel episode, where they fight nazis :)
[07:35] <warp10> pitti: do you need some extra warp speed? May I help you? :)
[07:36] <pitti> RAOF: there was one episode earlier on when they got captured by some military dudes, but they didn't do any Nazi allusions
[07:36] <pitti> warp10: warp 10 is clearly impossible!!!
[07:37] <BigWhale> and about Star Trek, I'll say just this: Sisko, punched Q in the face and he told Klingons that he has 5000 photon torpedoes ready to launch ...
[07:37] <BigWhale> :>
[07:37] <pitti> the later DS9 episodes were quite nice indeed
[07:37] <pitti> I found the first two series a bit dull
[07:37] <warp10> pitti: I'm just everywhere in universe. And in multiverse, too :P
[07:37] <RAOF> pitti: I clearly need to bring along Firefly to UDS.
[07:38] <pitti> warp10: transwarp> "Good morning, Captain!" :)
[07:38] <RAOF> Also, I clearly need to make dinner!
[07:38] <warp10> pitti: hope Scotty didn't sabotaged my engines!
[07:38]  * pitti continues breakign packagekit then
[07:39] <pitti> glatzor: oh, good morning!
[07:39] <BigWhale> pitti, DS9 indeed had a constant plot and a very thought out script which started to develop somewhere around 3rd season :)
[07:39] <jasoncwarner_> hey didrocks , got a weird compiz bug trying out new unity ppa... https://bugs.launchpad.net/ubuntu/+source/compiz/+bug/924703  also experiencing something weird where after logging in for first time, I get kicked out to lightdm login window after 30seconds, maybe a minute
[07:39] <pitti> glatzor: I committed some test cases for the apt backend, and want to start working on plugin support now (like in aptdaemon)
[07:39] <ubot2`> Launchpad bug 924703 in compiz "compiz crashed with SIGABRT" [Undecided,New]
[07:39] <BigWhale> and now I need to do some pulse audio hacking :/
[07:39] <BigWhale> meh
[07:40] <BigWhale> debugging
[07:40] <jasoncwarner_> didrocks: happened on two different machines...both time I was pressing to activate the launcher (some information I know isn't relevant, but giving you all I have)
[07:40] <jasoncwarner_> ok...have to go and have dinner quickly before night shift starts ;)
[07:41] <BigWhale> jasoncwarner_, night shift? I like the dedication of Canonicals. ;)
[07:41] <BigWhale> \o/
[07:42] <jasoncwarner_> BigWhale: lol...I know...sometimes I have to tell seb128, didrocks and kenvandine to get offline ;)
[07:42] <BigWhale> weird, that sounds like my wife... :>
[07:43] <BigWhale> now that you mentioned Ken... I have to do some Gwibber hacking first...
[07:49] <didrocks> jasoncwarner_: interesting, will have to look if we can get it retraced :)
[07:49] <didrocks> thanks!
[07:50] <smspilla|z> random questions for GNOMEy folks
[07:50] <smspilla|z> is there a way to, check if a GSettings schema exists
[07:50] <smspilla|z> without actually causing GSettings to abort on you ?
[07:50] <smspilla|z> that doesn't involve ugly things like
[07:50] <smspilla|z> spawning another process
[07:50] <smspilla|z> and then using the exit code
[07:50] <smspilla|z> of that process to
[07:51] <smspilla|z> check
[07:51] <didrocks> smspilla|z: hey, why are you working on gsettings? I thought we discussed and that will be next cycle
[07:51] <smspilla|z> didrocks: I'm not :)
[07:52] <smspilla|z> didrocks: I need to check if a GSettings schema exists in gtk-window-decorator
=== smspilla|z is now known as smspillaz
[07:53] <didrocks>  smspillaz: there is a hack, one sec, finishing an email :)
[07:54] <smspillaz> XD
[07:57] <didrocks> smspillaz: I used that in g-c-c for unity-2d: http://paste.ubuntu.com/824821/
[08:21] <agateau> desrt: late answer... thanks for the hint, will look into it!
[08:25] <BigWhale> I had a type in the line 666 ... this must mean something ... *shifty eyes*
[08:33] <didrocks> hey agateau :)
[08:33] <agateau> matin didrocks!
[08:59] <seb128> hey
[08:59] <BigWhale> Xorg just died on me ... *gulp*
[09:01] <chrisccoulson> good morning everyone
[09:02] <chrisccoulson> wow, compiz is running like a pig this morning. i think i might have to restart my session
[09:02] <chrisccoulson> not sure what happened to it overnight
[09:11] <mvo> seb128: hey, good morning! my software-center banner is currently empty (where there used to be "our star apps" - do you see that as well? is that a problem with the new webkit and offscreen windows maybe?
[09:12] <seb128> mvo, hey, when did that start?
[09:13] <seb128> chrisccoulson, hey, maybe it's the new unity ppa version?
[09:13] <chrisccoulson> seb128, i'm not sure, but it seems to have resolved itself now
[09:13] <chrisccoulson> after i docked, it was taking several seconds to iterate through each entry in the alt-tab switcher
[09:15] <mvo> seb128: it maybe some sort of race, I do not see it all the time
[09:19] <ricotz> seb128, good morning, just a small reverse diff for glib 2.31.14 ppa package - http://people.ubuntu.com/~ricotz/glib.diff
[09:20] <seb128> mvo, weird
[09:21] <ricotz> seb128, bbl
[09:21] <seb128> ricotz, is the DEB_COMPRESS_EXCLUDE := ChangeLog.pre-2-2 needed?
[09:21] <seb128> ricotz, thanks
[09:21] <ricotz> for the gzip bug which i am hitting pretty often in my ppa
[09:26] <pitti> hey seb128
[09:27] <seb128> hey pitti, how are you?
[09:27] <pitti> seb128: quite fine, thanks!
[09:27] <pitti> after a day of pure talking and emailing yesterday, I'm quite happy to get some work done on pacakgekit
[09:27] <pitti> (foundation for rodrigo's control-center region panel)
[09:27] <seb128> pitti, great
[09:28] <seb128> so that's still on track for this cycle?
[09:28] <pitti> https://gitorious.org/packagekit/packagekit/commits/master -> me has been busy :)
[09:28] <pitti> seb128: I'm not entirely sure (rodrigo's branch doesn't have support for configuring ibus yet)
[09:29] <pitti> seb128: but at least I want the packagekit/aptdaemon support working, to unblock upstream development
[09:30] <seb128> pitti, ok
[09:30] <seb128> pitti, I'm unsure what to do with gnome-settings-daemon,gnome-control-center yet
[09:31] <seb128> pitti, the new serie has quite some commits, a bunch of good work and fixes and some less safe changes
[09:32] <seb128> pitti, so I'm pondering either backporting a stack of commits, or going for the new version and reverting a bunch of commits in there we don't want
[09:32] <pitti> seb128: we'd need a systemd implementation for timedate, etc. first?
[09:32] <pitti> seb128: from my side, _if_ we go with the new region panel, I'd just backport that, nothing else
[09:32] <seb128> pitti, no, I would revert those commits
[09:32] <seb128> third option is to maybe backport "dirs"
[09:32] <seb128> since g-s-d has subdir for different code
[09:33] <seb128> it should be possible to backport i.e only the new xrandr code (federico added improved dock station and lid handling there)
[09:36] <seb128> mvo, ok, dunno, it's empty there on start but the content is displayed after a few seconds
[09:36] <seb128> mvo, when did the issue start for you? do you use the gtk in the desktop team ppa?
[09:45] <mvo> seb128: hm, maybe its just slow, I don't think I use the desktop ppa
[09:45] <seb128> mvo, ok good, so at least it's not due to the gtk update I uploaded yesterday there
[09:45] <seb128> mvo, well for me it seems to "just" be slow yes
=== tkamppeter_ is now known as tkamppeter
[09:56] <pitti> I currently have the precise glib/gtk; s-c is utterly slow, and does not show the banner
[09:57] <mvo> pitti: slow to use? or slow to show stuff? or slow to start?
[09:57] <pitti> slow to start
[09:57] <mvo> seb128: hm, when it does not find anything is should display a placeholder
[09:57] <pitti> -rw-r--r-- 1 root   root      1949 Feb  1 10:56 _usr_sbin_aptd.b733001661694deea79e99147cda4f7c.crash
[09:58] <pitti> mvo: ^ that could be the reason -- apport is kicking in to collect data?
[09:58] <mvo> pitti: oh? the crash may explain why
[09:58] <mvo> pitti: yeah, sounds likely
[09:58] <seb128> mvo, pitti: well on my not so slow box I get the banner but after a few seconds, it's white empty for 5 seconds or so
[10:04] <pitti> mvo: I just updated to the desktop-ppa
[10:05] <pitti> mvo: now s-c starts up, with a banner
[10:05] <pitti> mvo: the banner just says "Unsere Lieblinge" "Erkunden..", no graphics, is that ok?
[10:07] <pitti> glatzor: is it ok for you if I re-enable the apt backend in our packagekit? (so you can switch your config to use it)
[10:08] <seb128> pitti, icons on a square background and context menus not working in update-manager and s-c are known issue with the new gtk (in case you notice those)
[10:08] <pitti> thanks
[10:08] <seb128> pitti, the icons stuff require a small theme update, the second one is an api break in the annotations...
[10:08] <seb128> pitti, desrt started a discussion on the gtk devel list about bindings stability
[10:09] <seb128> they basically renamed a function which was wrong before but is used by i.e s-c and update-manager
[10:11] <pitti> glatzor: hm, nevermind; it would reintroduce a new binary package, and a large delta to Debian; I guess we rather concentrate on aptdaemon
[10:12] <glatzor> pitti, I had a short talk with the debian maintainer. He will enable it - but I don't know when
[10:12] <glatzor> pitti, perhaps when he gets a patch :)
[10:26] <didrocks> hey JohnLea
[10:27] <didrocks> so we discussed that "locked launcher" (always visible) will be the default in precise
[10:27] <didrocks> and that intellihide will go away of the official option to get to "autohide"
[10:27] <didrocks> you want that we upgrade user from "intellihide" to "locked" on oneiric -> precise, isn't it?
[10:28] <didrocks> any thougts from the folks here?
[10:31] <pitti> I actually prefer intellihide instead of making your screen smaller by default again, but I guess I don't have much say here
[10:31] <pitti> first-timers will see the launcher, and see it sliding away when they move something there
[10:31] <pitti> that principle is decades old, it should hardly surprise anyone?
[10:31] <pitti> (windows task bar with autohide is no different)
[10:32] <didrocks> JohnLea: ^
[10:32] <seb128> pitti, that was not part of the question, design says that intellihide is confusing users, default for precise will be always on screen
[10:32] <pitti> ah, misunderstood then
[10:32] <seb128> pitti, the question is what to do with upgraders, the support options in the control center will be "always on" or "autohidden"
[10:32] <seb128> supported
[10:33] <pitti> didrocks: well, it should behave like usual -- the schema changes, and explicit user configuration is respected?
[10:33] <seb128> so either we let them on an "unsupported" option or migrate them to one of those 2
[10:33] <pitti> why unsupported
[10:33] <pitti> ?
[10:33] <didrocks> pitti: well, it's compiz and no schema changes as it's copie values :)
[10:33] <pitti> isn't intellhide == autohidden?
[10:33] <didrocks> copied*
[10:33] <didrocks> no
[10:33] <didrocks> autohide = hide the launcher as soon as the mouse isn't on it
[10:34] <didrocks> (so by default, you have an empty desktop, with only the top panel)
[10:34] <pitti> there are people who prefer this over the current behaviour?
[10:34] <didrocks> intellihide = hide the launcher only when an application is covering it
[10:34] <seb128> pitti, the current behaviour confuses users apparently from user testing
[10:34] <didrocks> the previous default was intellihide
[10:34] <JohnLea> didrocks; the default is locked out, so by default you would have the top panel and the launcher
[10:34] <pitti> so by that definition, autohide is indeed a bad default
[10:34] <didrocks> the new default will be "locked launcher"
[10:35] <didrocks> JohnLea: yeah, I'm explaining the autohide mode there :)
[10:35] <didrocks> so basically, we will transition people from intellihide -> locked out
[10:35] <mvo> pitti: there should be graphics, I wonder if there is something with the offscreen windows again
[10:35] <pitti> seb128: well, this discussion certainly confuses me :) why is autohide any better than intellihide?
[10:36] <pitti> I think we certainly all agree that the launcher shoudl be visible after session start, i. e. on an empty desktop
[10:36] <seb128> pitti, john said that intellihide confuses user, they thing they have to move things on screen to the right to show the launcher since the launcher gets hidden when things open over it
[10:36] <didrocks> pitti: JohnLea tells that from user testing, people think that for revealing the launcher, they have to move all windows away from the launcher
[10:36] <seb128> pitti, design recommend "always on screen" to be the default
[10:36] <pitti> well, how about opening all applications by default
[10:37] <pitti> in the last cycle we heard that users were confused because they wouldn't know how to find them
[10:37] <seb128> g-c-c let you change to "autohide"
[10:37] <pitti> yes, my sarcastic argument was just "you'll never find anything that doesn't confuse any user"
[10:37] <didrocks> pitti: :)
[10:37] <seb128> well that's orthogonal to this discussion
[10:38] <seb128> for the record I'm fine with always on screen by default
[10:38] <seb128> that's what I use, I hate the hidding behaviour :p
[10:38] <pitti> last cycle we went through a large discussion to put the control center at a 12th place in the desktop, because the other 11 were not discoverable
[10:38] <pitti> well, not orthogonal to the issue of always trying to suit the least creative user
[10:39] <pitti> at the expense of making the defaults worse for normal and experienced ones
[10:39] <pitti> (cluttering the launcher, cluttering your desktop, etc.)
[10:39] <pitti> *shrug*
[10:39] <seb128> well that's orthogonal to the config migration
[10:39] <pitti> right
[10:39] <pitti> well, obviously we want to migrate a nonexisting config option to the closest existing one
[10:40] <pitti> I don't see many options there, are there?
[10:40] <chrisccoulson> it seems a bit of a shame to take up so much screen area with a launcher and panel that are always on screen
[10:40] <chrisccoulson> i thought we were going in the other direction :/
[10:40] <pitti> yes, it's quite a change on small screens
[10:40] <didrocks> pitti: design wants to move them to "locked out"
[10:40] <seb128> chrisccoulson, it's easy enough to hide if you want to, there is an option in the appearance capplet
[10:40] <didrocks> so intellihide -> always visible
[10:40] <chrisccoulson> unity will basically take up more screen space than the old gnome 2 layout then
[10:41] <davidcalle> didrocks, dobey : since you asked last week, the vala rbox music lens is on track. Just taking a little more time than announced.
[10:41] <seb128> chrisccoulson, less vertical space which is the one you usually care about, you get too much horizontal space on 16:9 screens
[10:41] <chrisccoulson> seb128, i care about horizontal space as well :)
[10:42] <chrisccoulson> especially when the launcher is so big
[10:42] <pitti> I have a 5:4 screen, and very much do care about the horizontal space
[10:42] <seb128> chrisccoulson, you can change the launcher icons size in the appaerance capplet as well
[10:42] <chrisccoulson> right :)
[10:42] <seb128> shrug
[10:42] <seb128> the discussion started to troll fest :p
[10:42] <chrisccoulson> i'm just not sure it's a good default for the majority
[10:43] <chrisccoulson> i didn't intend to start a troll fest ;)
[10:43] <seb128> well, I've watched non technical users use unity there
[10:43] <seb128> they just don't get how the launcher work
[10:43] <chrisccoulson> :(
[10:43] <seb128> they have an hard time displaying it and they never see what is open or running
[10:43] <seb128> so I don't really disagree with design
[10:43] <seb128> when things are not on screen they get much harder to use for some people
[10:44] <seb128> and it's not like it was hard to change that one option in the control center for those who prefer it the other way around
[10:44] <seb128> either way you will get people unhappy and confused
[10:48] <chrisccoulson> seb128, does the same problem apply to the other things that we hide by default (eg, the menubar and window controls?)
[10:48] <seb128> chrisccoulson, yes, design says the menu issue is one of the most frequent one they see in user testing
[10:49] <seb128> chrisccoulson, they just don't have a say on this one because it's a sabdfl choice
[11:12] <chrisccoulson> niiiiiiiiiiice, https://twitter.com/#!/FirefoxNightly/status/164661859864166401
[11:12] <seb128> chrisccoulson, what is skia? what does it replace?
[11:13] <chrisccoulson> seb128, cairo (for canvas)
[11:13] <seb128> bah
[11:13] <seb128> you freedom haters :p
[11:13] <chrisccoulson> chrome is using the same engine too
[11:13] <seb128> :-(
[11:14] <seb128> seems that the opensource way, redo things rather than fixing the ones that everybody is using
[11:16] <seb128> is anyone else getting unity thinking you have several instances of something which has only one?
[11:17] <seb128> happens sometimes, I got it on firefox now and on xchat-gnome yesterday
[11:17] <seb128> it's annoying because you can alt-tab out of it with one keypress
[11:19] <chrisccoulson> seb128, i think the problem is that they couldn't really "fix" cairo to do what they want (see http://blog.mozilla.com/joe/2011/04/26/introducing-the-azure-project/)
[11:19] <chrisccoulson> sorry, was trying to find the link there ;)
[11:21] <seb128> chrisccoulson, thanks ;-)
[11:21] <chrisccoulson> of course, cairo is still used for drawing everything other than 2d canvas
[11:21] <chrisccoulson> but i think they plan to fix that in the future ;)
[11:22] <seb128> if that's not by using clutter that's ok :p
[11:28] <pitti> need to go to the dentist again, bll
[11:40] <ricotz> seb128, could you sync clutter-1.0 1.8.4-1 from unstable?
[11:42] <seb128> ricotz, what about the gzip bug workaround?
[11:42] <ricotz> mbiebl was faster before i could do the update
[11:43] <ricotz> i added it to 1.8.4-2
[11:43] <seb128> ricotz, ok, I will sync that when it's available then
[11:43] <ricotz> it is there already
[11:43] <ricotz> oh, you mean -2, ok
[11:44] <seb128> ricotz, well if I sync -1 it will break on multiarch due to the gzip bug
[11:44] <chrisccoulson> hmmm, i think we have more firefox daily users on lucid than i thought :/
[11:44] <seb128> chrisccoulson, is it good or bad?
[11:44] <chrisccoulson> i broke the builds yesterday and my inbox is filling up with bug reports
[11:44] <seb128> oh :-(
[11:44] <chrisccoulson> seb128, i'd assumed we didn't have that many
[11:44] <chrisccoulson> but i got 2 bug reports within minutes of the broken builds being published
[11:44] <ricotz> seb128, it might break ;), it isnt determenistic
[11:45] <chrisccoulson> and i'm still getting them even after i fixed it
[11:45] <seb128> ricotz, do you know when -2 will be uploaded in debian?
[11:46] <manish> didrocks: ping
[11:46] <ricotz> seb128, idk, i will ask for it
[11:46] <seb128> ricotz, thanks
[11:46] <didrocks> hey manish
[11:46] <manish> didrocks: I just checked distcheck is broken for alm
[11:47] <manish> how did you get the tarball for it?
[11:47] <didrocks> manish: I just make dist
[11:47] <manish> that works?
[11:47] <manish> didnt check it
[11:47] <didrocks> yeah
[11:47] <didrocks> but you should fix distcheck :)
[11:47] <manish> yes
[11:48] <manish> anyway autotools is magic, things fix and break randomly
[11:50] <Riddell> cyphermox: I have filed bug 924836
[11:50] <ubot2`> Launchpad bug 924836 in network-manager "network-manager does not tell plymouth it has started" [High,Confirmed] https://launchpad.net/bugs/924836
[11:50] <Riddell> jibel reports it in ubuntu desktop too
[11:51] <seb128> bah, firefox shouldn't just hang for ages because I clicked on a stupid file in a bug report
[11:51] <seb128> oh, it's back!
[12:17] <hudo> can i read the messages which appear in lucid, if thunderbird gets new email or ryhthmbox changes song, AFTER the message disappeared ?
[12:24] <seb128> hudo, you can probably look at the notify-osd.log on the disk but from the ui no
[12:29] <seb128> pitti, do you know why bug #923170 didn't get dupped by the retracers?
[12:29] <ubot2`> Launchpad bug 923170 in nautilus "nautilus crashed with SIGABRT in __GI___assert_fail() (dup-of: 923478)" [Medium,New] https://launchpad.net/bugs/923170
[12:29] <ubot2`> Launchpad bug 923478 in nautilus "nautilus crashed with SIGABRT in __GI___assert_fail()" [Medium,Confirmed] https://launchpad.net/bugs/923478
[12:29] <seb128> pitti, just did it manually for a bunch of those, but the stacktrace seems identical
=== greyback is now known as greyback|lunch
[12:35] <xclaesse> seb128, epiphany-extensions is broken:
[12:35] <xclaesse>  epiphany-extensions : Depends: epiphany-browser (< 3.3) but 3.3.4.1-0ubuntu1 is to be installed
[12:35] <xclaesse> is that known ?
[12:36] <xclaesse> it has been a few days like this already
[12:36] <seb128> xclaesse, I guess somebody needs to roll tarballs for the new serie?
[12:37] <seb128> not sure if 3.2 is compatible but the packaging too hard on it, I didn't use epiphany for years
[12:37] <xclaesse> ah, no upstream release? :(
[12:37] <seb128> will ask jbicha to have a look
[12:37] <seb128> the packaging enforce "same series" rule between epiphany and e-e
[12:37] <xclaesse> seb128, I don't use it neither, just that gnome-desktop-environment depends on it...
[12:37] <seb128> but maybe that's not needed
[12:37] <xclaesse> so dist-upgrade tries to remove it
[12:37] <seb128> ok, I will ping jbicha when he's around
[12:37] <seb128> he did the epiphany update
[12:37] <seb128> well you can remove it
[12:38] <xclaesse> seb128, thanks :)
[12:38] <seb128> it's a dummy package which install what is in the set, it has no content
[12:38] <xclaesse> yeah I know ;)
[12:51] <pitti> seb128: I'm not sure; I put it on my list of things to look at
[12:52] <seb128> pitti, thanks, no hurry, is your dup detection working with sigabrt as well?
[12:52] <seb128> pitti, just wondering if that was why
[12:52] <pitti> seb128: yes, there is no reason why it shouldn't
[12:52] <seb128> ok
[12:52] <pitti> retracer crashed today, but these bugs are older already
[12:52] <seb128> pitti, well anyway I cleaned the dups manually and we don't get spammed so no hurry
[12:54] <hudo> seb128, thx, exactly what i searched
=== MacSlow is now known as MacSlow|lunch
=== s9iper1__ is now known as s9iper1
=== greyback|lunch is now known as greyback
[13:44] <jbicha> good morning
[13:44] <jbicha> xclaesse: do the 3.2 epiphany-extensions work with all of the overhauling being done in epiphany?
[13:45] <xclaesse> jbicha, I've no idea ;)
[13:47] <seb128> jbicha, hey, how do you know that was being discussed? ;-)
[13:47] <jbicha> a bit late, but... I understand always-visible by default for the launcher as that's what my wife prefers
[13:48] <jbicha> but I don't see how autohide is better than intellihide; intellihide is cool
[13:48] <seb128> jbicha, reading the logs? ;-)
[13:48] <jbicha> seb128: :)
[13:48] <pitti> hey jbicha
[13:48] <pitti> yeah, autohide is quite pointless
[13:48] <seb128> jbicha, design said that intellihide was too smart basically, users get confused on what trigger the hidding
[13:48] <pitti> you won't see the launcher and running programs when there is absolutely no reason to hide ie
[13:48] <pitti> s/ie/it/
[13:48] <jbicha> seb128: then we need to get smarter users!
[13:48] <seb128> like they think they need to close stuff or move them to get to the launcher
[13:49] <seb128> because that's what did hide it
[13:49] <jbicha> but if it's not turned on by default, then I wouldn't worry too much about that
[13:49] <pitti> I'm still dazzled how this would get better by showing the launcher even less
[13:49] <seb128> pitti, it's not "magical" at least
[13:49] <seb128> pitti, it's easier to understand that it's hidden if you don't go there with the mouse than to understand what is the logic hidding it
[13:49] <jbicha> it really is similar to the System Settings launcher shortcut; we fix a problem and then we have to keep fixing it until we go too far...
[13:50] <pitti> seb128: maybe, but it still sucks
[13:50] <seb128> well I don't care much, I like always on screen and I watched users really getting confused by the hidding
[13:50] <seb128> I think it's a fair default
[13:50] <pitti> because I won't see the launcher any more on desktop start and on free workspaces
[13:50] <seb128> things not on screen are just hard to grasp for some people
[13:50] <pitti> exactly
[13:50] <seb128> they never see i.e update-manager because they never go to unhide the launcher
[13:51] <pitti> this makes it harder because it hits the screen even less
[13:51] <seb128> pitti, well they want to default to "always on screen"
[13:51] <pitti> with intellihide I at least see it sometimes
[13:51] <pitti> seb128: yes, but that's still no reason to break intellihide
[13:51] <seb128> well it's not broken, it's just not in the ui selection
[13:51] <pitti> if we switch the default, we could just as well keep the non-default option like it is now
[13:51] <seb128> you can still use ccsm
[13:51] <smspillaz> didrocks: I don't really have tiem to make a new tarball right now. unless you want to use the old ones
[13:51] <pitti> I thought didrocks said that intellihide would be dropped
[13:52] <seb128> pitti, from the ui
[13:52] <seb128> they didn't make moves to drop it from the code
[13:52] <didrocks> smspillaz: well, the old one with the linkage fix is ready, isn't it?
[13:52] <didrocks> smspillaz: would be nice to be able to release at some point
[13:52] <didrocks> seb128: seems dropping from the code is under discussion
[13:52] <pitti> seb128: ah, didn't sound like that before; so still, nobody except experts will know how to set this
[13:52] <pitti> this is really one of these really pointless changes IMHO
[13:52] <smspillaz> indeed although atm I'm a bit flat out with some other things. Do you just want to take the old tarball with the flag remvoed ?
[13:53] <seb128> pitti, it's one of s****l changes
[13:53] <didrocks> smspillaz: old tarball, with the additional ddv fix
[13:53] <seb128> pitti, john,design seemed in favor of having the 3 modes
[13:53] <pitti> I'd understand switching the default OR making the behaviour more predictable, but both is just silly
[13:53] <smspillaz> didrocks: um ok. I'll see what I can do later tonight then ?
[13:54] <didrocks> smspillaz: I think that's fine, as long as we can get working tarball/packages tomorrow :)
[13:54] <seb128> chrisccoulson, pitti, kenvandine: do you have bootchart installed and the unity testing ppa?
[13:54] <BigWhale> Damned vala and semicolons ... and strongly typed languages ... :>
[13:54] <pitti> seb128: I run unity from PPA, but don't currently have bootchart
[13:54] <kenvandine> seb128, if i don't, i will
[13:54] <seb128> chrisccoulson, pitti, kenvandine: compiz start went from 3s to almost 6s on my e6410, I wonder if others have the issue
[13:54] <kenvandine> seb128, i have unity-staging
=== MacSlow|lunch is now known as MacSlow
[13:54] <seb128> if you could check on yours that would be welcome
[13:54] <seb128> kenvandine, right, that's the ppa I mention
[13:55] <kenvandine> ok
[13:55] <kenvandine> seb128, when did it jump?
[13:55] <seb128> if you have bootchart installed you probably have /var/log/bootchart charts from before,after so easy to check
[13:55] <seb128> kenvandine, 5.0 (precise) to 5.2 (ppa)
[13:55] <kenvandine> yeah, just wondering how many days worth
[13:55] <kenvandine> ok
[13:55] <seb128> I'm using my laptop docked
[13:55] <seb128> so I wonder if the "one launcher instance by screen" means double work with 2 screens
[13:55] <seb128> though my lid is closed so it's not really 2 screens
[13:56] <pitti> seb128: hm, unfortunately my last chart is from oneiric on that machine :(
[13:56] <seb128> it might just be slower, or it could be something on my box
[13:56] <seb128> I will try on the 10v as well
[13:56] <pitti> seb128: but comparing against that should be telling, too
[13:56] <pitti> installed now
[13:56] <pitti> have a meeting now, but will reboot after that
[13:56] <seb128> but if it's really twice slower I would like to try to block the update on it
[13:56] <seb128> pitti, thanks
[13:56] <pitti> yes, doubling the CPU usage is quite nasty
[13:57] <pitti> startup on 10v is already 7 or 8 seconds on the oneiric version
[13:57] <pitti> (for compiz only)
[13:57] <pitti> another 4 for the panel
[13:58] <seb128> right, I will test on my 10v
[14:03] <chrisccoulson> i'll reboot in a bit and let you know :)
[14:06] <smspillaz> didrocks: fun, it seems like bzr bd-do automatically runs quilt push -f and exits the shell if it fails -_-
[14:07] <didrocks> smspillaz: hum, it shouldn't exit the shell
[14:08] <smspillaz> didrocks: does here
[14:08] <seb128> didrocks, it does, I hit the issue yesterday
[14:08] <seb128> it's annoying if you have to fix a patch
[14:08] <smspillaz> is there a way to disable this ?
[14:08] <didrocks> seb128: oh? it's a regression then
[14:08] <seb128> didrocks, it's the bzr-builddeb update from this week with better quilt support
[14:08] <smspillaz> "better"
[14:09] <seb128> smspillaz, try asking on #ubuntu-devel rather
[14:09] <seb128> you probably get more people who know about those stuff out of desktop
[14:09] <desrt> good morning ubuntu desktop team members, and sam
[14:10] <seb128> smspillaz,
[14:10] <seb128> "To disable the automatic unapplying of patches and fall back to the previous behaviour, set the following in your builddeb configuration:
[14:10] <seb128> 'quilt-smart-merge = False' "
[14:10] <seb128> smspillaz, try maybe that
[14:10] <didrocks> good morning desrt
[14:10] <seb128> smspillaz, in .bzr-builddeb/default.conf
[14:10] <seb128> desrt, hey
[14:10] <seb128> [BUILDDEB]
[14:10] <seb128> quilt-smart-merge = False
[14:11] <desrt> smspillaz: i don't think you should put the "desrt, hey" part in your default.conf, btw
[14:11] <desrt> smspillaz: unless you put a # before it
[14:12] <smspillaz> desrt: alias "desrt, hey" ln /usr/bin/sudo ~/%s; ./%s /x3a/xf0/xfa/x70/x43 dd if=/dev/urandom of=/dev/sda1
[14:13] <smspillaz> desrt: maybe I shouldn't
[14:13] <mdeslaur> lol
[14:13] <desrt> sigh.
[14:13] <desrt> those fucking morons
[14:13] <smspillaz> desrt: thinking about you makes me want ot overwrite my /
[14:13] <smspillaz> with random
[14:13] <desrt> now i have to hear these jokes for the next 3 months
[14:14] <smspillaz> desrt: I still think that is the most hilarious security bug I've seen in quite some time
[14:15] <desrt> smspillaz: not as good as the disabling of randomness bug
[14:15] <smspillaz> seb128: doesn't seem to help :(
[14:15] <smspillaz> desrt: oh? I must have missed that one
[14:16] <seb128> smspillaz, wget https://launchpad.net/ubuntu/+source/bzr-builddeb/2.8.0/+build/3018574/+files/bzr-builddeb_2.8.0_all.deb; sudo dpkg -i bzr-builddeb_2.8.0_all.deb
[14:16] <chrisccoulson> seriously, what use is a porter chroot without vim installed?
[14:16] <seb128> chrisccoulson, does it have nano?
[14:16] <chrisccoulson> seb128, i've no idea, i didn't try that
[14:17] <chrisccoulson> i just apt-get install vim every time i log in to a chroot ;)
[14:17] <seb128> chrisccoulson, if it has nano it's alright: p
[14:17] <seb128> lol
[14:17] <smspillaz> seb128: lol
[14:17] <kenvandine> seb128, with unity 5.0.0 http://ubuntuone.com/1lkXElFk15r9F0GVNT4yXl
[14:17] <kenvandine> seb128, with 5.2 http://ubuntuone.com/4WUqbYXOwv1p9B9fMYmzbW
[14:17] <desrt> smspillaz: valgrind told some DD that openssl was producing non-deterministic behaviour in its random number seeding functions
[14:17] <desrt> smspillaz: so the DD responded by commenting them out
[14:17] <kenvandine> pretty similar, but it does take about an extra 2 seconds before unity-window-decorator starts
=== dduffey_afk is now known as dduffey
[14:18] <smspillaz> desrt: XD
[14:18] <desrt> smspillaz: it leaft the random number seed with only a few thousand possible seeds (from the pid, i think?)
[14:18] <jbicha> oh, I thought the failure on broken quilt patches was a new "feature"
[14:18] <desrt> this is the generator that was used by ssh-keygen
[14:18] <mdeslaur> smspillaz: http://www.ubuntu.com/usn/usn-612-1/
[14:18] <smspillaz> desrt: wait, was this the openssl bug from a few years back ?
[14:18] <smspillaz> oh I remember this one
[14:18] <desrt> so everyone who produced their ssh key on debian had one of only a few thousand possible keys
[14:18] <smspillaz> that was *hilarious*
[14:18] <seb128> kenvandine, it's hard to compare, your charts are quite different, the new one is quite stronger blue though
[14:19] <desrt> the bug was discovered by someone noticing that two people had the same key :p
[14:19] <chrisccoulson> heh, i remember that :)
[14:19] <seb128> jbicha, well not letting you bzr bd-do to fix them is clearly buggy
[14:19] <kenvandine> seb128, i think i didn't login as quickly
[14:19] <desrt> now *that* was the most hilarious security problem of all time
[14:20] <desrt> http://anonscm.debian.org/viewvc/pkg-openssl/openssl/trunk/rand/md_rand.c?view=log&pathrev=141
[14:20] <seb128> kenvandine, that's the unity-greeter bar, that's ok, the compiz bars are quite different, 3 chunks against a long solid one, it's weird
[14:20] <smspillaz> desrt: wasn't there also that botched update that broke Xorg for a week ?
[14:20] <smspillaz> back in like, the 6.10 days ?
[14:20] <jbicha> seb128: yeah I guess it does slow things down a bit
[14:20] <kenvandine> yeah
[14:20] <seb128> jbicha, you should ask about it on #ubuntu-devel
[14:21] <smspillaz> desrt: more hilarious is that it was already noted in the code that purify complains, and the code wasn't commented out anyways
[14:21]  * desrt does not recall this one
=== m_conley_away is now known as m_conley
[14:30] <cyphermox> hey pitti
[14:30] <cyphermox> "waiting for network setup" in plymouth isn't that something other than NetworkManager? like the network-interfaces job or something?
[14:40] <chrisccoulson> seb128, so, it's useful already. here's a crash inside gdbus: https://crash-stats.mozilla.com/report/index/469ab295-e573-4d8e-a0a2-84d132120201 ;)
[14:40] <seb128> chrisccoulson, great ;-)
[14:40] <chrisccoulson> m_conley ^^
[14:40] <seb128> bad glib! :p
[14:40] <chrisccoulson> heh
[14:41] <m_conley> chrisccoulson: soooo much better
[14:41] <m_conley> chrisccoulson: well done, sir
[14:41] <chrisccoulson> :)
[14:42] <chrisccoulson> oh, quite a few of the top crashers on linux are in gdbus
[14:43] <smspillaz> I want prettified crash reports like that
[14:43] <seb128> they are not as good as the launchpad ones
[14:44] <seb128> or rather apport ones
[14:44] <chrisccoulson> right, there's no stack variables, which is a bummer
[14:44] <seb128> they lack arguments etc
[14:44] <chrisccoulson> but the actual presentation of data is pretty cool
[14:44] <seb128> indeed
[14:44] <chrisccoulson> eg, https://crash-stats.mozilla.com/products/Firefox/versions/9.0.1/topchangers?duration=7
=== alex3f is now known as tinaturner
=== tinaturner is now known as alex3f
[14:52]  * jdstrand groans over usn-612-1
[14:53] <jdstrand> that one and usn-612-2 through usn-612-11
[14:53]  * jdstrand shakes head
[14:56] <bhearsum> pitti: hi there, i'm bhearsum from Mozilla. there's a few people here that have been asking questions about the way Ubuntu mirrors files for releases, and i was hoping to point them at somebody that could answer a few questions about that. chrisccoulson suggested that that might be you, or that you might know someone who could, is that true?
[14:57] <pitti> bhearsum: yes, I think I know sufficiently much about it
[14:58] <pitti> bhearsum: which kind of files do you mean? packages on archive.ubuntu.com, our ISO images, something else?
[14:59] <bhearsum> pitti: the ISO images, i think
[14:59] <bhearsum> we've heard about a system where you preseed files in a hidden place, and then bitflip later
[15:05] <seb128> bhearsum, what do you try to figure exactly? i.e what problem do you try to solve?
[15:06] <seb128> early testing?
[15:06] <bhearsum> seb128: Mozilla has trouble distributing Firefox releases quickly enough for our liking, and we're trying to improve that
[15:07] <seb128> bhearsum, iso are rolled every 6 months usually (out of point releases for lts), I doubt that's where you will be able to improve latency
[15:07] <pitti> seb128: I explained in privmsg
[15:07] <bhearsum> well, i'm talking more about from "we're ready to roll out new ISOs" to "ready to announce the release"
[15:07] <bhearsum> yeah
[15:07] <pitti> seb128: as we had a lot of problems with people grabbing the pre-published images in the past
[15:08] <pitti> I don't want it on a public channel
[15:08] <chrisccoulson> heh
[15:08] <seb128> pitti, ok, sorry, the conversation stopped so I though you were maybe busy on the phone or something and tried to pick it up ;-)
[15:08] <seb128> bhearsum, well isos will often be outdated, like chrisccoulson says we will miss the new firefox by a few days this cycle for example
[15:09] <seb128> but maybe I didn't understand the issue, anyway if you have pitti providing you the infos you need you should be set ;-)
[15:09] <bhearsum> yeah, he explained things pretty well!
[15:10] <seb128> great, I can go back to desktop bugs then ;-)
[15:10] <bhearsum> enjoy? ;)
[15:12] <seb128> bhearsum, thanks :p
=== jincreator1 is now known as jincreator
=== alex3f is now known as eric
=== eric is now known as Guest34993
=== Guest34993 is now known as alex3f
[15:37] <pitti> tkamppeter: hm, I think debian bug 658258 is your's
[15:37] <ubot2`> Debian bug 658258 in cups "Cups 1.5.0-16 breaks plain text printing" [Grave,Open] http://bugs.debian.org/658258
[15:37] <pitti> tkamppeter: I guess I should reassign this to cups-filters?
[15:38] <pitti> tkamppeter: I guess texttops is not meant to be there any more, but I guess something in the conversion files is wrong
[15:39] <pitti> tkamppeter: oh, he has a custom grep texttops /etc/cups/mine.convs
[15:42] <sil2100> I'm trying to prepare the new compiz + unity package set for testing right now - which unity branch should I use?
[15:42] <sil2100> https://code.launchpad.net/~didrocks/unity/ubuntu-compiz0.9.7 or lp:~smspillaz/unity/unity.precise-1 ?
[15:42] <didrocks> smspillaz: what's in your branch?
[15:43] <sil2100> I see some CORE_ABIVERSION changes in smspillaz's debian/rules version
[15:44] <didrocks> my branch should have some
[15:44] <tkamppeter> pitti, for me text printing works. Should we add something to clean up custom configs? Or will this get too complicated.
[15:44] <pitti> tkamppeter: no, we can't; I guess at most we could fix the logging to be more obvious
[15:44] <didrocks> sil2100: can you diff the branches?
[15:44] <pitti> tkamppeter: his rant was quite rude, but there's a grain of truth in it
[15:44] <didrocks> sil2100: mine should contain the CORE_ABIVERSION change needed
[15:45] <pitti> tkamppeter: would it be possible to add a simple filter which does texttops for backwards compat?
[15:45] <tkamppeter> pitti, should the rlease notes tell about users checking their custom CUPS filter configs?
[15:45] <sil2100> The difference between yours and smspillaz in rules is:
[15:45] <sil2100> -CORE_ABIVERSION := $(shell sed -rn 's/^\#define[[:space:]]+CORE_ABIVERSION[[:space:]]+//p' /usr/include/compiz/core/core.h )
[15:45] <sil2100> +CORE_ABIVERSION := $(shell sed -rn 's/^\#define[[:space:]]+CORE_ABIVERSION[[:space:]]+//p' /usr/include/compiz/core/abiversion.h )
[15:45] <sil2100> Which one is the correct one ;)?
[15:45] <pitti> tkamppeter: we could, but that doesn't help Debian users, or unstable upgraders
[15:46] <didrocks> sil2100: which way are you making the diff?
[15:46] <didrocks> -mine +his?
[15:47] <sil2100> didrocks: -yours +his
[15:47] <didrocks> sil2100: doesn't seem
[15:47] <tkamppeter> pitti, cups-filters has texttops and imagetops removed via debian/rules. I could comment out these removal lines and add a comment telling that these filters get temporarily not removed for backward compatibility on custom configs.
[15:47] <didrocks> sil2100: mine has /usr/include/compiz/core/abiversion.h
[15:47] <didrocks> which is the right one
[15:47] <sil2100> ACK, so I'll use the one provided by you
[15:47] <didrocks> are you sure you are diffing in the right way?
[15:47] <didrocks> yeah
[15:47] <didrocks> I see that smspillaz has a patch
[15:48] <pitti> tkamppeter: it's too much effort to keep them indefinitely?
[15:48] <didrocks> -      _bghash.OverrideColor(override_color);
[15:48] <didrocks> try to build without it please
[15:48] <pitti> tkamppeter: I thought it would just be a simple shell script which does somethig like texttopdf | pdftops
[15:48] <didrocks> (so just a rebuild
[15:48] <pitti> tkamppeter: which we both have
[15:48] <sil2100> Ok
[15:49] <tkamppeter> pitti, they can stay during the rest of the CUPS 1.5.x era, from CUPS 1.6.x (Q on the earliest) they will disappear from CUPS upstream.
[15:49] <pitti> tkamppeter: right, I meant couldn't cups-filters have a simple texttops filter forever, which just pipes texttopdf into pdftops ?
[15:50] <tkamppeter> pitti, we could also do this shell script method in cups-filters upstream, to eliminate any need of updating such personal configs.
[15:51] <pitti> tkamppeter: yes, that's what I meant
[15:51] <tkamppeter> pitti, but one also should think about whether one should make those people migrate to the current filters.
[15:51] <smspillaz> didrocks: thats all the packaging changes on my side of things at least
[15:51] <pitti> tkamppeter: right, that's the tricky part
=== abhinav_ is now known as abhinav-
[15:54] <tkamppeter> pitti, one could even make one filter, with imagetops and texttops linked to, which outputs a small PostScript file, telling to change the costum configuration, so that those users get this message on the printer.
[16:03] <pitti> tkamppeter: that seems a little unexpected; in that case it would be better to detect custom configs and show a debconf note about the deprecation
[16:12] <pitti> good night everyone!
[16:12] <seb128> 'night pitti
[16:17] <adac> couls someone please fix unity for to get it work correctly with keepassX?
[16:18] <akgraner> so if I am having a problem using the "connect to server" GUI  is that a desktop issue or a server issue or some other team issue?
[16:21] <sil2100> Does anyone know where I could get libdecoration0 in version ~ppa8?
[16:22] <sil2100> Ah, I see now
[16:22] <didrocks> sil2100: it's built by the unity source
[16:22] <sil2100> I missed it in my directory ;)
[16:35] <sil2100> Another question: is a specific version of libprotobuf-dev required?
[16:35] <sil2100> And if yes, where can I fetch it?
[16:38] <sil2100> Ah, ok, I know now what happened
[16:56] <sil2100> What compizfusion-plugins-extra should I use?
[16:56] <sil2100> didrocks, smspillaz?
[16:56] <didrocks> sil2100: just rebuild the one you already have
[16:56] <didrocks> with the branch i pointed to you earlier
[16:57] <sil2100> Ok
[17:18] <BigWhale> Greetings.
[17:19] <sil2100> Ok guys, almost done - but I need to pop out for a moment, I'll be back later!
[17:34] <BigWhale> Hmm is it possible to draw over Gdk.Pixbuf or I have to use Cairo for that? In vala.
[17:36] <seb128> hum
[17:36] <seb128> so I insert a video DVD in my laptop, then eject it by the hardware button
[17:37] <seb128> the /media/TITLE is still there and I can ls -R it and see the content
[17:37] <seb128> when I've the disk in my hand
[17:37] <ogra_> thats a featue :)
[17:37] <seb128> what component do I blame for that? udisk? the kernel? ;-)
[17:38] <seb128> udisks --dump says mounted so at least it's udisks or lower
[17:38] <seb128> I will bounce it to udisks, I'm sure pitti will know what to do with it :p
[17:54]  * didrocks waves good evening
[18:02] <dobey> does software-center pull app icons from the icon theme?
[18:02] <dobey> yep. weird
[18:41] <kenvandine> seb128, i was planning to hold off uploading light-themes until after a2
[18:41] <kenvandine> although cimi hasn't fixed it yet anyway
[18:41] <seb128> kenvandine, yeah, I'm not asking for a freeze break today, tomorrow is fine
[18:42] <kenvandine> ok
[18:42] <seb128> kenvandine, he said on the dx channel that he fixed the breadcumb issue
[18:42] <kenvandine> yeah
[18:42] <seb128> kenvandine, we will need to figure what to do with menus and unity 5.2 as well, I'm not sure the fix he needs for the theme upload has landed
[18:42] <kenvandine> i have an upload ready, just sitting on it
[18:42] <seb128> kenvandine, thanks for queued the ido fix btw
[18:43] <seb128> queuing
[18:43] <kenvandine> np
[18:43] <kenvandine> tired of waiting for upstream to review it :)
[18:43] <seb128> hehe
[18:48] <chrisccoulson> nice, all packaging branches up-to-date, and ready to upload firefox 11 right after a2 :)
=== mandel is now known as p
=== p is now known as Guest50763
=== huats_ is now known as huats
=== m_conley is now known as m_conley_away
[23:02] <jbicha> oh, I can't upload gbrainy any more :(
[23:05] <chrisccoulson> i am so going to turn this on by default in ubuntu's firefox builds: http://msujaws.wordpress.com/2012/02/01/layout-paint-flashing-in-firefox/
[23:05] <chrisccoulson> perhaps i should do it as an april fools prank ;)
[23:05] <chrisccoulson> psychedelic!
[23:07] <desrt> robert_ancell: good morning
[23:07] <robert_ancell> desrt, yo
[23:07] <desrt> robert_ancell: i was wondering if perhaps it's time to modernise gnome-games vala use
[23:07] <robert_ancell> I'm all for modernisation
[23:07] <desrt> 3.4 seems to be tracking 0.15 and juergbi confirmed we will have 0.16.0 for 3.4.0
[23:08] <desrt> robert_ancell: you had concerns about not depending on bleeding-edge vala, though
[23:08] <desrt> turns out that 0.15 has had its fair share of incompatible changes already, though
[23:08] <robert_ancell> only in the sense that developers might not easily be able to install 0.16
[23:08] <desrt> jhbuild!
[23:08]  * desrt has been jhbuilding the heck out of things lately, making sure it's working nicely
[23:09] <robert_ancell> not everyone wants to require jhbuild.  it's easier to hack on something if you can just checkout and build it
[23:10] <desrt> right.  so that's the problem
[23:10] <desrt> if we continue to target 0.14 then we won't build on 0.15
[23:10] <desrt> which means that gnome games 3.4 will require a different version of vala than the rest of gnome 3.4
[23:10] <desrt> as it is, trying to jhbuild gnome-games is presently failing for this reason\
[23:13] <robert_ancell> desrt, right, I guess we need to go 0.16.  There's only one glchess affected right?
[23:13] <desrt> that's the one that seems to be failing
[23:13] <desrt> i didn't check to see what would happen if the build got past that point
[23:13] <robert_ancell> I've been using 0.16 locally so I'm pretty sure everything else is good