[11:49] <ogra_> wow, britney has a fast day (or my timing was very lucky)
[11:52] <xnox> ogra_: you probably do not have autopkgtests! =( otherwise it now takes longer ;-)
[12:02] <cjwatson> slangasek,bdmurray: working on the copy/phased-updates thing now.  I'm a bit rusty ...
[13:02] <psivaa> cjwatson: just curious if there will be saucy desktop images arriving later today
[13:06] <ogra_> depends when libqt5webkit5 enters main
[13:18] <cjwatson> psivaa: blocked on bug 1192567, I believe currently taken by doko
[13:18] <ubot2> Launchpad bug 1192567 in qtwebkit-opensource-src (Ubuntu) "[MIR] qt5webkit " [Undecided,New] https://launchpad.net/bugs/1192567
[13:19] <doko> yes, will look at this today
[13:19] <psivaa> cjwatson: doko: ack, thanks
[15:00] <slangasek> per discussion on #ubuntu-devel, cleaning up daily-preinstalled/* on nusakan... as it only contains stale nexus7 desktop images
[15:00] <ogra_> yeah
[15:28] <stgraber> cjwatson: hey, do you happen to know what I'd need to do to hook into live-build's compression code? Basically I want to have it spit out both the current .tar.gz and a new .tar.xz (with everything shifted into a system/ directory in there) at the end of an ubuntu-touch build
[15:33] <cjwatson> stgraber: One or the other is easy, but to do both, I think it'd be simplest to just recompress it in livecd-rootfs/live-build/auto/build
[15:37] <stgraber> cjwatson: ok
[15:49] <stgraber> cjwatson: ok, got it working here, though looking at how slow it's on my laptop, I'm not sure we want that to run on armhf. I think I'll instead just write a python tool that does the unpack, move stuff to a sub-dir, repack on nusakan.
[15:49] <cjwatson> ok
[15:50] <stgraber> once we get rid of the existing .tar.gz rootfs, we could even grab an uncompressed copy from the livefs builder which I suspect would cut build time by a few minutes (looking at how slow compression is on pandas...)
[15:50] <stgraber> (or we'll just have the shiny new builders and won't have to care about this anymore... not sure how fast those calxeda nodes are at doing gzip/xz)
[21:55] <cjwatson> infinity: So I just commented out auto-sync for Debian import freeze, and was going to send mail about it, but I just noticed that raring's schedule had DIF in week 17 and saucy's schedule has it in week 8.  This seems a weirdly enormous discrepancy.  I thought I mentioned this in the UDS discussion - didn't we agree to move it?
[21:56] <infinity> cjwatson: Hrm, I didn't have moving it in my notes.
[21:56] <ScottK> IIRC it was supposed to be close to FF.
[21:56] <ScottK> (not from UDS, but generally from the raring discussion)
[21:57] <infinity> So, Raring had it at alpha2.  I'm not wildly picky about exactly when and where it lands, so long as it's not after FF. :P
[21:57] <infinity> And, honestly, we have people who like to play the role of the autosync bot after DIF anyway (a practice I'd love to stamp out)...
[21:58] <cjwatson> I know day-of is a crappy time to be asking about it, but it does seem incredibly weirdly early to me this time round.
[21:58] <cjwatson> Two weeks earlier than quantal, even.
[21:58] <infinity> cjwatson: I don't think anyone will mind if it slips later.  Maybe we can hit some mild concensus among people in the channel right now.
[21:58] <cjwatson> I agree it clearly ought to be before FF.
[21:59] <cjwatson> Raring had it coincident with Alpha 2.  That seemed OK to me.
[22:00] <infinity> cjwatson: I's be comfy with either A2freeze or FF.
[22:13] <doko> now you build everything, and everything is fine, and then somebody uploads a new texinfo :-/
[22:13] <ScottK> cjwatson: Sounds good to me.
[22:40] <ScottK> KDE 4.10.4 is all accepted for raring now, so the wall of accepts should be over.