[10:13] <CIA-19> ubiquity: cjwatson * r2154 ubiquity/ (debian/changelog ubiquity/components/partman_commit.py):
[10:13] <CIA-19> ubiquity: * Fix question_type vs. self.question_type crash in partman_commit
[10:13] <CIA-19> ubiquity:  (LP: #125539).
[11:25] <CIA-19> ubiquity: cjwatson * r2155 ubiquity/ (debian/changelog partman/commit.d/01unmount_busy):
[11:25] <CIA-19> ubiquity: * Unmangle entries from /etc/mtab before trying to unmount them (also
[11:25] <CIA-19> ubiquity:  LP: #125539).
[11:32] <xivulon> cjwatson, evand, are you in for a beer later on today?
[01:38] <xivulon> cjwatson, evand are you tied up later on today?
[05:06] <cjwatson> xivulon: I don't think it's going to work out for me this evening, unfortunately - last day is always mad
[05:06] <cjwatson> sorry :(
[05:10] <xivulon> cjwatson, np will do next time you are in london
[05:11] <xivulon> evand, are you in?
[05:44] <CIA-19> ubiquity: evand * r2156 ubiquity/ (4 files in 3 dirs):
[05:44] <CIA-19> ubiquity: * Don't preseed seen to false, or stay in the gtk event loop on the
[05:44] <CIA-19> ubiquity:  summary page, when using --automatic.
[06:04] <evand> xivulon: I'm afraid I cannot make it either.  They're throwing some sort of wine and cheese party in the office and then I've got to go present shopping before I leave.
[06:04] <evand> I'll be back here at the next sprint though
[06:26] <xivulon> have fun, and a nice trip home
[09:39] <superm1_> cjwatson, is there an eta when 1.5.5 will be released as a package?  I've got probably one more thing i wanted to get into the mythbuntu frontend yet, and wanted to see how much time i've got to do it
[11:01] <cjwatson_> superm1_: err, 1.5.5 was released on Wednesday; see the changelog
[11:01] <superm1_> oh, haven't bzr merged since my last commit
[11:01] <superm1_> :)
[11:01] <superm1_> okay
[11:01] <cjwatson_> superm1_: there is no reason why you should be bound by any particular version, though
[11:02] <superm1_> cjwatson, well I was hoping to be able to source the archives rather than rely on a mythbuntu mirror when generating the next alpha ISO
[11:02] <cjwatson> we are kind of planning to do further uploads, you know ;-)
[11:03] <superm1_> haha, well i wasn't sure on how often they will be generated - so for future reference then, what's the frequency on this sort of thing?
[11:09] <cjwatson> depends on what's there to uploads - look back through the changelog for a sampling
[11:10] <cjwatson> can be anything from days to a small number of weeks, typically
[11:10] <superm1_> alright, that's kinda what i was assuming
[11:12] <cjwatson> historical average since we took it over from Guadalinex (so all espresso or ubiquity uploads) is once every 2.9 days, which is a touch disturbing
[11:12] <superm1_> i had another question regarding casper, i wanted to get some input on.  I've toyed with the idea of generating an overlay filesystem similar to how you can use a flash drive to save changes to a live boot.  I was considering adding that as an append option to the boot menu - one with fglrx and one with nvidia, so that if someone really wanted to use a disk as a full live frontend, they could boot right into the proprietary drivers.
[11:12] <superm1_> What would be your take on us doing something like that?
[11:13] <superm1_> yes that is wow.
[11:13] <cjwatson> (you might reasonably expect development to slow down over time though)
[11:13] <cjwatson> hmm, it would require some care to make that work with ubiquity, but it should be possible
[11:14] <cjwatson> TBH I really think we should just fix the crappy way the proprietary drivers work with libGL
[11:14] <superm1_> i'm imagining the hook that generates the xorg.conf in casper may need additional work too, to choose them
[11:14] <superm1_> is there a spec out there to do such things regarding libGL?
[11:14] <cjwatson> just calls dpkg-reconfigure xserver-xorg or the moral equivalent
[11:14] <cjwatson> no
[11:14] <cjwatson> (not to my knowledge; not my area)
[11:15] <superm1_> perhaps i'll consider drafting one up for the dev summit in boston this year, but i'll have to investigate how much is involved
[11:16] <cjwatson> the problem is that both require overlay libGL implementations and they conflict 'cos they need to go in the same place in the filesystem
[11:17] <superm1_> perhaps in the place that they would need to be, replace a dynamic loader
[11:17] <superm1_> to load the correct libGL implementation from another area on the FS
[11:18] <superm1_> and then you would be able to store all three implementation in their own area
[11:19] <cjwatson> it's not trivial and requires much thought, anyway
[11:20] <cjwatson> anyway, time to go to the hotel; offline until Monday
[11:20] <cjwatson> have fun :)
[11:20] <superm1_> okay cu :)