=== asac_ is now known as asac [08:28] ok [10:03] ya nvidia is fixed [10:03] yay even [10:05] asac, people keep opening bugs against prism, while there's nothing prism could do about that, they want desktop apps features from webapps [10:11] fta: thats how it works ;) ... set wontfix or invalid [10:15] firefox-3.1 and 3.2 help>about is broken, XML error [10:29] ok thats strange, a package updated but i didnt have it installed :( [10:55] sometime this week i should have seamonkey 2.0 done. i doubt today since its monday and ive been off for a while [11:14] jdstrand: we have https://bugzilla.mozilla.org/show_bug.cgi?id=471715 in the pipeline, but for 1.8 branches its not yet really QAed [11:14] Mozilla bug 471715 in Libraries "Add cert to nssckbi to override rogue md5-collision CA cert" [Enhancement,Resolved: fixed] [11:14] not sure what to do with that for this update round [11:14] jdstrand: are there any server apps using nss? [11:16] jdstrand: so my plan atm is to get the new nss (3.12latest) into hardy, intrepid and jaunty [11:16] dapper and gutsy will come when we know what to do [15:17] how can setup thunderbird to see only unread emails [17:47] fta: are you here? [18:57] piratenaapje, hi [18:57] piratenaapje, no, it didn't work, i fixed a few other things but it's still not enough [18:57] fta: But it configures? [18:57] the whole python hack is far too messy [18:57] no, configure worked, like for you [18:58] the build fails afterwards [18:58] Oh :s [18:59] i guess i have to figure out another way to use our system python. [18:59] What's the problem now then? Still the python build code? [18:59] using --python=xxx in xul-configure is apparently a bad idea [19:00] no, by default, it expects a full copy of python in the moz objdir. if you don't provide it, you end up with just the python modules from both xul and o-k but no usable python [19:01] not sure if I can help with that [19:01] they should create a something we can ship in /usr/lib/python-support or something [19:02] too bad i know almost nothing about python packaging [19:02] don't worry, it's a though project [19:02] much tougher than usual [19:03] Well, I don't know much about python packaging either :s, probably less than you even [19:03] i committed you stuff so asac could have a look, maybe he'll have a better idea [19:04] don't worry, whatever happens, you'll be credited for your work :) [19:04] Ah, yay :p [19:04] youR [19:04] I'll take a look later, I can try at least [19:04] sure [19:05] asac, ^^ any progress with o-k? [19:05] asac, maybe you can corner some python packaging magician in your sprint :) [19:06] hmm, he's probably drinking beers in some dark place [19:06] Alright, no rush :p [19:34] fta: You didn't commit your changes? [19:35] hm, i thought i did, maybe not all [19:36] fta: bzr says my tree is up to date, but I didn't update since saturday [19:37] fta: oops sorry, bzr is a bit confusing, update doesn't actually update :p [19:37] you need rev6 [19:37] bzr pull [19:38] update is different [19:38] you usually just need bzr init/push/pull/commit/diff/status [19:38] fta: I'm used to svn :p [19:39] yep, each vcs has its own tricks [19:42] no more snow in paris :( [19:45] piratenaapje, another big problem with o-k is that they ship patches everywhere so when you have to patch those files, it becomes really ugly [19:48] fta: We're supposed to patch the patches? [19:49] not if we can avoid it [19:49] we should really discuss with upstream [19:49] but i don't have much time to jump into yet another big project [19:50] fta: And I'm really not experienced enough for this one :p [19:51] maybe we can drag another motu or core-dev into this [19:51] (i secretly hoped asac was the one) [19:51] ;) [19:51] fta: Hopefully :) [19:52] ok, dinner time. cu later [19:52] Enjoy your meal [20:11] asac: questions about libnotify [20:12] or fta, I guess [20:30] mconnor, yesN [20:30] ? [20:31] fta: libnotify doesn't let you set an action for onclick? [20:34] mconnor, i don't know the details. i know the cannonical devs are working on notifications and actions are hot debate. same for click through. asac should know more, he's in a sprint the whole week, so he probably has fresh news [21:13] fta: ok, awesome, thanks [21:17] oh my! http://www.journaldugeek.com/2009/01/14/reparer-un-ecran-plasma/ [21:48] mconnor, in case you missed it, the initial rework idea for notifications started from there: http://www.markshuttleworth.com/archives/253 [21:53] I think that the actions thing is stupid [21:53] the rationalization there is half-assed [21:54] because by saying "hey, X" you're telling someone information that they may want to act on [21:55] otherwise why are you even telling me anything? [21:55] fta: where can I weigh in on this misguided decision process? [21:56] hm.. UDS would have been a good place. [21:56] I didn't know [21:57] i'm looking on lp if there's a place with fresh info on this [21:57] thanks [21:58] because as it stands, I'm mulling backing out libnotify support, because it's a usability regression [21:59] and then asac and I will have to get drunk at FOSDEM and fight to the death [21:59] or something [21:59] mpt seems to be at the center of this topic [22:02] <[reed]> ah, good ol' mpt ;) [22:02] <[reed]> he's ex-Mozillian [22:08] can't find much :( [22:08] http://www.ubuntu.com/testing/jaunty/alpha3#New%20style%20for%20notifications%20and%20notification%20preferences [22:11] some info about the new Desktop Experience Team: http://www.kdedevelopers.org/node/3805 [22:11] http://arstechnica.com/open-source/news/2008/12/canonical-linux-notification-system.ars [22:11] no real spec, sorry [22:16] fta: gotta love the transparency [22:25] i thought some parts of the mozilla tree where twisted and ugly, but it's in fact very well polished compared to openkomodo :P [22:25] -where+were [22:27] fta: rule 106 of openkomodo-configure.patch should be removed, it's output I used for debugging [22:29] fta: rule 97-114 can all be removed actually :p [22:29] piratenaapje, yes, i know but i'm not satisfied at all by the last two patches in the series (mine included). so i didn't even try to clean them. i just push them to let asac start from where we left. [22:30] i'm quite sure there's a far better way to do this [22:30] fta: ah alright [22:30] fta: yeah it's quite hacky what I've done :p [22:31] not your fault, the design is flawed to start with [22:31] fta: I don't see how the configure would've worked in the first place [22:32] fta: The first error you were getting had nothing to do with the siloed python version [22:33] i also fixed a lot of stuff that could not possibly have worked for anyone. I wonder how they manage to even build that thing upstream [22:34] it's using a build system in perl, in python, with an autoconf and m4 macros, with Cons scripts, makefiles, shell scripts, etc.. [22:34] plus patches [22:35] plus tons of foreign binaries [22:35] fta: Yeah, easily the ugliest build system I've seen so far :p [22:37] maybe we could just bypass this whole python "black" build system and build each piece ourself [22:39] fta: Sounds like alot of work :p [22:40] maybe