[05:57] <RAOF> Wooo!
[05:57] <RAOF> An MP that's less than 1000 lines!
[05:58] <RAOF> I CAN DO IT!
[07:32] <duflu> anpok! Fixed Mesa yet? :)
[07:34] <anpok> yes
[07:35] <anpok> patches have not yet been integrated..
[07:36] <anpok> there is still a failure during unity8-dash context creation -- which is suspicious since unity8 and indicators start properly
[07:37] <anpok> duflu: i guess the right answer is 'never'
[07:38] <duflu> anpok: Aww. You should branch it then. Branching solves everything
[07:39] <anpok> i thought rewriting does that?
[07:39] <anpok> or was it adding another layer?
[07:40] <RAOF> Indirection is what you're thinking of, yeah. :)
[07:42] <anpok> possibly, as long as you make sure that those indirections are redundant.. it is always odd if similarities are indicated by sharing implementatioons
[07:44] <anpok> we have to get to a point where think of software components as a genetic algorithm that test runs multiple specimen at the same time
[07:48] <anpok> .. ok something different
[07:49] <anpok> RAOF: i guess i was to late yesterday sending the patch - just couldnt send it without anothe test run on my laptop
[07:49] <RAOF> anpok: Yeah, I think it came in overnight.
[07:50] <RAOF> And I got distracted with duflu and his ridiculous “needs to be reviewable” criteria for branches :)
[08:18] <duflu> Hmm, why is Mir so much slower to build than it was a few months ago?
[08:18] <duflu> What did we change?
[08:18] <duflu> Might need to script up some builds next weekend
[11:18] <alan_g> alf__: libmiromnibus has "all" the mir code (and no test code). Are you sure that "libmirtesting" is a better name? Or do we need to think of a third?
[11:45] <alf__> alan_g: Well, I had to look up what "omnibus" is, and that is not a good sign. Perhaps, though, "omnibus" is a well known term for native speakers. I think it would be good to consider a few more alternatives.
[11:47] <alf__> alan_g: of course, this is an internal library, so getting a perfect name is not critical (but it would be nice if we could find one)
[11:47] <alan_g> alf__: I don't consider it obscure, but I don't use it everyday either.
[11:49] <alan_g> It might not stay internal. Downstream have expressed interest in having our test doubles available for their tests - and many of these rely on internal interfaces. Depending on how that progresses it may have a wider consumption.
[11:53] <alf__> alan_g: just brainstorming (not necessarily good names) -> libmirall, libmirinternals, libmirfull
[11:54] <alan_g> alf__: ack
[11:55] <alan_g> libmir-all-naked
[11:55] <alan_g> (That should get a phoronix article)
[11:55] <alf__> :)
[11:58] <alan_g> kdub: is "omnibus" an obscure word?
[12:10] <kdub> alan_g|lunch, eh, I know what it means, but dont use it often
[12:52] <alan_g> kdub: alf__ and I were discussing whether "libmiromnibus" is a good name and what else we might call it.
[12:54] <kdub> its sort of an obscurely-used library, so I'm okay with the name