=== jtisme is now known as jtholmes === bjf is now known as bjf-afk [08:38] Good morning [08:41] Guten Tag Her Pitti ;) [08:42] pitti: did you have a good flight? [08:42] hey didrocks [08:42] didrocks: yes, uneventful; slept 3 hours, which is pretty good for my standards in a plane :) [08:42] didrocks: how about you? [08:43] pitti: good, thanks. Only slept 2 hours :( We arrived one hour in advance, but wait for 30 minutes before the plane can get a gate. [08:44] still a little bit tired, but less than expected :) [08:44] * pitti gets some breakfast [08:45] enjoy your breakfast [09:24] hello everyone [09:24] hey chrisccoulson! how are you? [09:25] pitti - good thanks. how are you? did you have a good week? [09:25] yes, it was great; still feeling a bit weak (jetlag, lack of sleep in plane, etc.), but that will be fine soon :) [09:26] hey chrisccoulson! [09:26] when did you arrive back? [09:26] hey didrocks [09:26] chrisccoulson: yesterday around noon [09:27] yeah, i think i would probably be quite tired too ;) [09:27] hello chrisccoulson pitti [09:28] hey seb128 [09:28] chrisccoulson, how are you? [09:28] hello seb128 :) [09:28] chrisccoulson, the baby arrived? [09:28] I had no time for IRC during UDS [09:28] hey didrocks [09:28] seb128 - yeah, i'm ok. we have a baby now :) [09:28] good! [09:28] bonjour seb128 [09:28] everybody doing fine? [09:28] she arrived on tuesday [09:28] lut pitti [09:28] chrisccoulson: congratulations! So I guess you have a reason to be tired, too? [09:29] * pitti hugs chrisccoulson [09:29] * pitti hugs seb128, too [09:29] yeah, they're doing ok. had to stay in hospital a couple of days, but they got back home on friday [09:29] * seb128 hugs chrisccoulson pitti [09:29] seb128: I tried my 16 hours fasting -- seems to have worked ok [09:29] * seb128 hugs didrocks [09:29] * chrisccoulson hugs pitti and seb128 [09:29] * didrocks hugs pitti, seb128 and chrisccoulson :) [09:29] pitti, when did you sleep this night? [09:29] * seb128 did midnight to 10am [09:29] usually after a trip from US to home it takes me like 4 hours to find some sleep [09:29] still a bit tire [09:29] tired [09:30] yesterday I went to bed at 22:30, and fell asleep after 10 mintues; I woke up at 8:30 [09:30] excellent [09:30] well, easy to find sleep [09:30] * chrisccoulson hugs didrocks [09:30] 21h15 - 6h45 for me :) [09:30] just don't sleep in the plane [09:30] seb128: but I only slept 3 hours in the plane, so I had good reason to be tired [09:30] didrocks: well [09:30] after skipping a night you usually crash easily [09:31] pitti, yeah, same here [09:31] I didn't sleep much this time in the plane [09:31] I finished my book instead! [09:31] seb128: were you so interested in the movies? :) [09:31] ah ;) [09:32] no, but I was interested in my book and in being tired enough to sleep later [09:33] seb128: I read the vala tutorial and did some first exercises [09:34] vala is pretty nice, although vastly underdocumented [09:34] I'll fiddle a bit more to get my d-bus example working and then post it to the vala wiki [09:34] ah good [11:01] Where could I find the algorithm that maps a Desktop Menu Specification category into an Ubuntu "Applications" menu category? [11:02] e.g. that takes an application with "Category: Network" and puts it in "Internet" === asac_ is now known as asac === sabdfl1 is now known as sabdfl [11:21] mpt - i'm not sure if it's what you are looking for, but the menu structure defined in /etc/xdg/menus has all of that information (about which categories appear in each sub-menu) [11:21] specifically, applications.menu for the default menu [11:22] chrisccoulson, that's exactly what I was looking for, thanks [11:23] you're welcome [11:54] aloha [12:02] anyoe seen similar issues using gwibber 2.0 ? https://bugs.launchpad.net/gwibber/+bug/487064 [12:02] Launchpad bug 487064 in gwibber "Message pane repaints frequently and resets the scroll status, losing reading position" [Undecided,New] [12:28] morning [12:35] huats: aloha [12:35] hey czajkowski [12:36] are you ok, not suffering from the jetlatg ? [12:36] huats: I'm not too bad thanks, rested yesterday and went to bed last night early. back at work today [12:36] ok [12:37] huats: congrats on loco council :) === MacSlow is now known as MacSlow|lunch [12:38] czajkowski, hum... I think I can congrat you too ! [12:38] :D [12:39] :) true === MacSlow|lunch is now known as MacSlow [14:17] salut seb128 [14:17] lut rickspencer3 [14:17] hey tedg [14:17] good morning pitti, ccheney` asac etc... [14:17] Good morning seb128 [14:18] rickspencer3: hey [14:18] rickspencer3, http://picasaweb.google.com/sdelcroix/FSpot#5407299873207857650 [14:18] hey asac [14:18] rickspencer3, upstream f-spot already did the "edit in viewer" [14:18] * asac hugs seb128 [14:19] he pinged me today about what we need and did it during lunch time apparently [14:19] seb128, seriously? [14:19] said it was trivial to do since the code was already there [14:19] that is awesome [14:19] it just needed to make it available in view mode [14:19] FOSS ftw! [14:19] ;-) [14:22] hey rickspencer3! [14:22] hi pitti [14:22] here we go again ;) [14:23] seb128: that sounds like the wrong way? I thought we want the button in eog to "edit in f-spot"? [14:23] pitti, maybe, but we need editability in the view mode anyway [14:24] pitti, f-spot need to be able to edit... [14:24] currently you have to load images into the library to edit them [14:24] seb128: ah, I understand [14:24] then we can add the eog button [14:24] so it's unrelated to that [14:24] not really [14:24] the "add an eog button" is easy [14:24] but if you don't have anything to call from there... [14:24] pitti, well, only in that there would be no point in putting in an "edit" button if f-spot couldn't edit ;) [14:25] MacSlow, seb128: btw, any chance we could make lucid's notify-osd a little less hideously looking by removing the lines? [14:27] pitti, +1 [14:27] pitti, i think he will want those turned on for a little while [14:28] kenvandif, he wants urgency to be on screen [14:28] yeah [14:28] the layout lines are probably of no use [14:28] oh... [14:28] good point [14:28] perhaps those are both enabled by the same debug setting [14:28] kenvandif, hey btw, did you see the f-spot comment? === kenvandif is now known as kenvandine [14:28] it's not that my design-untrained eye will ever report a bug against off-by-one pixels or so [14:29] seb128, awesome [14:29] kenvandine, right, which is why we ask now if we could have different level or option or turn layout lines [14:29] seb128, i guess he didn't close the bug though (f-spot), i didn't get mail about it [14:29] MacSlow, I agree with pitti, whatever way is easier for you but having a way to have only the colored bar would be nicer [14:29] pitti: hi. what is planned work-item wise this cycle? e.g. what improvements to tools etc. [14:30] kenvandine, no, he might still be tweaking [14:30] * kenvandine figured it should be trivial [14:30] DBO has a patch which lets you browse all photos in the current directory, even if you didn't specifically open them [14:30] he's also asking for what other changes we need [14:31] or want to get done [14:31] he = sde [14:31] oh ... just found a mail about subscribing to work-items.py etc. [14:31] he would welcome design team though about edit mode too [14:31] we want him to merge that patch, i will make sure we add it to the bug [14:31] ie how to conciliate browsing and editing [14:31] seb128, who is workong on it? [14:31] using sidepane for both [14:31] or having a grid similar to f-spot library [14:32] seb128, rickspencer3 also prefers having the browse sidebar on the bottom like eog does it [14:32] basically rickspencer3's request was to make it look just like eog :) [14:32] asac: I might add support for using bugs as WIs, but it's not there yet [14:32] asac: thus so far we have whiteboard or wiki page [14:32] asac: I'd rather ask it the other way round: is there something that your team needs/wants urgently? [14:33] pitti: ok thanks .... where is that workitems.py thing for the subscription lool asked for? [14:33] kenvandine, right, I'd also like a nicer looking button bar, like in EOG [14:33] asac: it's running on my server, and I set the MAILTO to you now [14:33] asac: i. e. you'll be spammed with "this BP has no work items defined", or "invalid WI status" [14:33] howver, I feel that the editing ability is the requirement, the rest are nice to have [14:34] asac: http://piware.de/bzr/bin/workitems.py [14:34] pitti: nothing needed atm. maybe some doc what the current syntax allows in whiteboard (e.g. i know there was ways to assign work items to individual membmers != assignee) [14:34] pitti: ok great. let me bookmark it [14:34] asac: doc> right, I need to set that up [14:34] asac: the "assignee" one was mere convention right now ("[asac] do foo") [14:35] asac: once I add a per-assignee report, I'll parse that [14:35] pitti: did the mobile team run that script on their own or was the graph generated for them by you too? [14:35] asac: all on my machine so far [14:36] ok http://piware.de/workitems/mobile/lucid/report.html [14:36] i assume i need to ping you when we are done with work items so you can reset the trend line starting point? [14:36] pitti: ? [14:37] asac: exactly [14:37] thx [14:37] asac: i. e. I can either trash the DB for a clean start, or just adjust the starting # [14:37] right. thx a lot. great service ;) [14:39] pitti: one thing i wonder if we can better support it: e.g. we want to do targetting for alpha-2 etc ... is there a way we can make burndowns for those milestones somehow? do you plan to do something like that for desktop team? [14:41] asac: right, that's in fact on my TODO list [14:41] since we all have these three subcycles [14:43] good ... so we could use general "spec" milestones, but i guess this needs to be done on a per-work-item base [14:45] asac: for now I only plan to parse spec-level milestones [14:46] asac: as soon as we have support for using bugs, we can of course use WI-level granularity [14:47] kk [14:48] pitti, how come apport is outdated in ubuntu? ;-) [14:48] * seb128 looks at the versions.html list today [14:48] seb128: ah, I did a new upstream release last Friday to test lp-project-upload [14:48] so it's in fact true :) [14:49] hehe [14:53] seb128, serialorder here, sorry for all those sync request bugs. I thought since they showed up on piware.de they had failed to autosync for some reason. [14:53] mannyv, hey [14:53] mannyv, no problem, but no, autosync are run manually so not every day [14:53] and we sync from testing by default this cycle [14:54] but usually if a version in ubuntu is coming directly from debian no need to bother [14:54] is anyone processing the -archive queue? [14:54] Laney, I've started doing autosync [14:54] will have a look to sync and new after that [14:55] hmmm [14:55] I expect there is quite some lag after uds [14:55] yes it's huge [14:55] prepare yourself ;) [14:55] according to slashdot their is a large group of people who are technical enough to depend on the gimp, but not technical enough to install new software on their desktop [14:55] I will no clean everything [14:55] but since half of archive admin are on holidays this week I will try to do some [14:56] does autosync also do new packages? [14:56] or is that a separate run [14:56] to be honest I really care about autosyncs catchup and desktop syncs and newing [14:56] I will probably let other things to other people [14:57] there is a different command to run for those [14:57] but I will run it too [14:57] alright [15:10] mvo: hi ;) ... did you recover from travel? [15:10] mvo: what determines which package gets into software center? is that whatever has a .desktop file? [15:11] and ... is there a way to tweak this? [15:14] asac, I guess that's .desktop but there was a spec at uds about listing other things there [15:15] hey, people talk here again [15:15] oh, right, UDS over :) [15:15] hey Amaranth [15:15] Amaranth: nono, go back to sleep [15:15] Amaranth, how are you? [15:23] oops, was reading scrollback [15:23] seb128: good, did you have a good trip back from texas? [15:23] yes thanks [15:24] I managed to read in the plane and sleep a bit, correspondance was easy to get no need to run, everything went smoothly [15:24] lucky [15:24] and I managed to sleep at normal hours yesterday [15:24] so all good [15:24] dang, was hoping I could get you guys on US time :) [15:26] Amaranth, so, we need to speed compiz this cycle any plan or hint for that? [15:26] Amaranth, is your version not using the wrapper good to go in lucid? [15:27] seb128: well I showed it to upstream and they noticed a small memory leak in the blacklist checking but I've fixed that now [15:27] that plus splitting up the packaging more will likely be all the speedup we get [15:28] do you have any clue about how much speed up that will be? [15:28] Unless we want to do something crazy like backport the module loading setup from the 0.9 branch (master) [15:28] Amaranth: splitting up packaging> into plugins we enable by default, and an "extra-plugins" kind of pacakge? [15:29] Then we will not have to read XML on start [15:29] Amaranth: "crazy" describes the challenge we face well [15:29] Amaranth: we have to cut down desktop startup time by 2/3.. [15:29] pitti: Well, it would completely change the plugin interface ABI [15:30] well, if we don't get it to run quick enough no compiz by default anymore in lucid [15:30] Amaranth: that alone doesn't scare me at all two weeks after opening a release :) [15:30] compositing in general will slow down login time [15:31] also it seems kind of silly that we're fighting for a 10 second boot+login then in 10.10 running gnome-shell will change that to 20 seconds [15:31] we never said we would take gnome-shell without any consideration [15:31] we will ponder pro and con before any change [15:32] asac: yes and yes [15:32] asac: mostly recovered anyway [15:32] asac: what do you need in s-c ? [15:32] in that case bring on the crazy [15:32] compositing makes things slower but there is no reason to be that slow [15:33] I wish we could get upstream releases of 0.9.x faster though :/ [15:33] do you have any clue how much difference the new loader will do? [15:33] 0.9 is not on the table for lucid [15:33] I know, thus the :( [15:33] err, :/ [15:33] hehe [15:34] If 0.9 had come out before karmic release I think it would have been feasible even if karmic didn't actually end up with an 0.9 release [15:34] anyway [15:34] what would have a been feasible? [15:34] using 0.9/0.10 in lucid [15:35] mvo: just saw a mail by bdrung about eclipse on devel-discuss (and tha the wrong package is included in sc) [15:35] mvo: maybe you can follow up directly there? [15:35] not that it matters now, we'll likely see 0.9 in time to stick it in lucid+1 [15:35] Amaranth, it's not in the table even if 0.9 was available next month [15:35] we will not change the compiz codebase in a lts when we have no compiz maintainer [15:35] But I have no idea how much time not loading the XML will save, benchmarking compiz is tricky [15:35] robert_ancell will be on oem shift this cycle [15:36] not that you don't do a rocking job there [15:36] I'd likely just have to do it and see what bootchart says [15:36] but better to be on the safe side [15:36] what do you suggest doing then? [15:37] It has to make some kind of difference though, it freaking loads XML using XPath [15:37] 1- use you ppa version with wrapper changes [15:37] 2- don't install the things we don't use [15:37] 3- look at the new loader? [15:37] the changes to drop the wrapper should go into lucid ASAP [15:37] ok [15:37] Then once we have an idea on the default settings we can split up the packaging [15:37] are those ready for upload? [15:37] in which case I can have a look to that today or tomorrow [15:37] let me build again with the new patch and push [15:38] thanks [15:38] we have an idea of the default settings we want [15:38] we can do the binary change [15:38] we can tweak later if required [15:39] although splitting the packaging up is not going to make much of a difference [15:39] and will make almost no difference at all if we change the plugin loader to not load XML [15:40] what is taking time if that's not the loader? [15:41] we want to split those anyway so people installing cssm don't get too much crack [15:41] the shell script and redirecting existing windows [15:41] hey... [15:41] hi paul [15:42] * paul wonders if there are any #gwibber people here [15:42] Amaranth, ok, so let's get this wrapper changer as a first step [15:42] alright [15:42] paul, talk to kenvandine [15:42] should I name the packages compiz-plugins-universe or something? [15:42] seb128: cheers :) [15:42] Amaranth, yes [15:42] Can't use compiz-plugins-extra because then we'd have compiz-fusion-plugins-extra-extra :) [15:42] universe sounds good [15:43] we already have extras, etc [15:43] right [15:43] Ideally we wouldn't use anything from extras but alas [15:43] what do we need from there? [15:43] kenvandine: hey, just wondering if you know whether bug 487064 will be a trivial fix or not? (i.e. should I go back to 1.4) [15:43] Launchpad bug 487064 in gwibber "Message pane repaints frequently and resets the scroll status, losing reading position" [Undecided,New] https://launchpad.net/bugs/487064 [15:44] seb128: hmm, I actually can't think of anything [15:44] let me check what we load by default again [15:46] paul, not sure [15:46] i don't think that is happening here [15:46] pitti, nice mail wrt work items, could I ask you to share it with other tech leads to help them get started? [15:47] paul, can you run "gwibber-daemon -d" and gwibber in terminals to catch output? [15:47] i wonder if it is failing to render something and looping or something crazy [15:48] kenvandine: you can reproduce this by scrolling somewhere and hitting reload [15:48] paul, also which gwibber theme are you using? [15:48] asac, well on reload yes it does do that [15:48] which it probably shouldn't [15:48] right. so i would expect every refresh to do that oo ;) [15:49] but that is different, he says it happens even when it is set to refresh less often [15:49] i dont think that reload is technically much different from refresh [15:49] hmm [15:49] it isn't [15:49] no different [15:49] ok ... so probably when daemon gets update (which is independent from refresh) [15:49] seb128: hehe, we only use plugins-extra for extrawm [15:49] it triggers a UI refresh [15:49] kenvandine: it's the default [15:49] kenvandine: i can try that, yeah [15:49] but if he is set to fetch updates every 15m [15:50] kenvandine: isnt that fetch update setting the UI trigger time? [15:50] seb128: and we don't even set any keybindings in extrawm so it doesn't get used at all [15:50] i guess the daemon threads still keep on going as usual? [15:50] asac, we should make sure... but it should be the interval the daemon is set to use [15:51] seb128: so we can kick that one out to universe entirely [15:51] i scrolled mine down a little and left it sitting here and it hasn't moved [15:51] but i am sure it will when it updates next [15:51] Amaranth, excellent [15:51] asac, paul's bug is that it does that every 10s or so [15:55] seb128: https://code.edge.launchpad.net/~amaranth/compiz/no_wrapper is the changes to get rid of the shell script [15:55] kenvandine: well, it's not doing it now.. maybe it takes time or some other actions to manifest. [15:55] (though, it has done across several restarts before, so i'm confident it's reproducable) [15:55] Amaranth, ok, thanks, will have a look to that in a bit [15:55] I don't know if mvo wants to make a karmic branch from the main compiz branch and I was hoping someone would look it over first so I pushed it there instead of the main branch [15:56] lets wait 15 min till it does its next fetch of new messages [15:56] mvo, ^ want to do that or should I have a look to those changes? [15:57] kenvandine: yes, my problem is it will update the UI pane every 10s or so - no new messages [15:57] paul: aloha [15:58] czajkowski: [ALOLRA protocol changes to state m1_waiting] [15:58] paul: I'm safe from ye lot on freenode :) [15:58] czajkowski: ooh, i see you're iLolra on this server ;) [15:58] paul, could it be images loading? [15:58] kenvandine: it could [15:59] so the images loading push everything down as they come in [15:59] which is very annoying [15:59] Oh, I forgot I also refreshed all the patches so it looks like a lot of changes [15:59] but once they are all downloaded it should say put [15:59] The only real changes are to the packaging stuff and the 060 patch [15:59] kenvandine: its not a push down - its a reset back to the top [15:59] ok, so it isn't that [15:59] i.e. the scroll slider gets reset to beginning [16:00] kenvandine: ok, its not the images loading - i just scrolled down to a point where there were still images to be loaded and they got painted without any scroll-state reset [16:02] kenvandine: you got back ok then. tired from a long week ? [16:02] yeah :) [16:02] and you? [16:02] paul, what version of webkit? [16:02] seb128: I planed to look at it, but maybe not today [16:02] paul, actually can you post the webkit version to the bug? [16:02] yeah 6am yesterday. Was slithgly wrecked, back at work today and no wanting to be here :) 1 hr left though [16:02] Amaranth: I can branch karmic if there is a need for this, not sure, there is some pending stuff, so its probably a good idea [16:02] mvo, ok thanks [16:02] kenvandine: sure [16:02] czajkowski, i was anxious to get to work... lots of stuff to do now that uds is over :) [16:03] kenvandine: I've all my gobby docs downloaded and glanced at them last night and see my name against a good few tasks ;) [16:03] hehe [16:03] good :) [16:03] for lucid I've defiantely got some clear goals and tasks to get through for community [16:05] kenvandine: 1.1.10 (webkit) [16:05] ok, we have 1.1.15.2-1 in ubuntu [16:06] i'll talk to ryan and see if he has ideas [16:07] kenvandine: i may update to F12 at some stage, which has 1.1.15 [16:07] though, that might not be till around christmas.. [16:08] yeah, but i would like to see if that is the problem [16:09] paul: see amazing what hapens when you log and dont just keep whinging on twitter and annoying me :) [16:10] :) [16:10] kenvandine: ok, i might have an F12 test system i can try (dont use it as dsktop at mo) [16:10] czajkowski: indeed [16:10] paul, that would actually be great [16:10] really help narrow the scope [16:10] paul: bugs = fixes :) [16:10] that pane is rendered with webkit, so seems most like it would be the culprit [16:11] kenvandine: ok, it's happening (but large interval) - nothing in the debug output for either daemon or app [16:11] kenvandine: interesting [16:12] kenvandine: it handles the scroll state too then? [16:12] wait, i might be assuming that [16:13] paul, yes... most likely [16:13] that could be javascript [16:17] kenvandine: ok, seems i need to leave it a while / use it a while [16:22] kenvandine: fwiw, it's not leaving 15 minutes between updates [16:22] its more like 5 [16:25] * Ng files bug #487165 and wonders if other people get that without transparent windows and think everything has somehow crashed [16:25] Launchpad bug 487165 in compiz "screensaver unlock dialog under other windows" [Undecided,New] https://launchpad.net/bugs/487165 [16:34] mvo, I found "man apturl" [16:38] mpt: great, I'm not sure if it covers it fully, but it should be good enough to give a overview [16:39] mvo, it gives examples of multiple packages (foo,bar), sections (foo?section=universe), and PPAs [16:40] mvo, in future it would be nice to have a syntax for departments, and a syntax for canned searches [16:40] and apt.ubuntu.com equivalents for all of the above [16:41] mpt: makes sense, could you just mail me about it, I think we can add this trivially [16:42] mpt: the only one I found is "minver=1.0" to ensure the package has the right version [16:42] * mvo is away for a bit for lunch [16:49] mvo, still on us time? ;-) [16:49] mvo, https://wiki.ubuntu.com/SoftwareCenter?action=diff&rev2=240&rev1=238 [17:08] mpt: thanks [17:09] seb128: yes, at least my brain [17:09] mvo, adding those other URLs is v4-type stuff, no need to worry about it now :-) [17:09] * mpt -> home [17:15] rickspencer3: can do [17:33] kenvandine: bingo - i think i know whats causing the refresh - searches [17:34] kenvandine: they seem to get scheduled very frequently at times [17:34] (with the general message pane open - not a search specific pane even) [17:35] kenvandine: i had a run of 3 updates just there, circa 40s, 7s and 13s apart. [17:38] humm [17:38] paul, so you have a search stream open right, but you are in the messages pane? [17:39]