=== qengho is now known as CardinalFang === CardinalFang is now known as qengho === ara is now known as Guest22932 [08:03] Ok, I think I've worked out a solution to building VM images./c [08:03] whoops, wrong window. [08:07] Good morning [08:12] hey pitti! [08:13] hey Laney, how are you? [08:13] doing good - sunny days in London! [08:13] how's the other side of the river? [08:14] Laney: equally sunny :) [08:14] seb128: bonjour ! où est l'autre français ? :-) [08:15] pitti, salut ! il dort encore le feineant ! [08:15] * seb128 summons didrocks [08:16] hey pitti! [08:16] bonjour didrocks ! ça va ? [08:17] pitti: ça peut aller, et toi ? [08:18] didrocks: j'ai dormi mieux, mais encore trop tard [08:18] à cause de jibel, c'est ça ? [08:19] ces français… [08:19] didrocks: non, nous avons chambres pour une personne [08:19] il t'a trainé dans les bars, j'imagine :p [08:29] all: today's pad for notes: [08:29] http://pad.ubuntu.com/MQ01t50gLb [08:32] bregma, desrt, didrocks, FJKong, happyaron, Laney, seb128, Sweet5hark, attente, ^^ [08:32] happyaron, FJKong - yesterday's notes: http://pad.ubuntu.com/N4aJtsz2V7 [08:34] willcooke, thanks [08:34] got it [08:50] popey, pingaling - are you free for a HO? [08:51] willcooke: ya [08:52] popey, https://plus.google.com/hangouts/_/canonical.com/will?authuser=0 [08:52] popey: http://pad.ubuntu.com/MQ01t50gLb [08:52] for context [08:53] one mo [09:26] Trevinho: any idea why compiz doesn't draw shadows for windows that have an argb visual? [09:27] larsu: it's by design [09:27] davmor2, btw, that nautilus trash issue is bug #1445595, seems to be an unity issue (calling the dbus method manually doesn't lead to the extra win) [09:27] bug 1445595 in unity (Ubuntu) "Empty Trash Results in File Windows Opening" [Low,Confirmed] https://launchpad.net/bugs/1445595 [09:27] larsu: it never did because in general it's up to them to draw these [09:27] Trevinho: hi. how are you. [09:27] larsu: or things like cairo-clock or other widgets might get double shadows [09:28] seb128: oh nice thanks for the update [09:28] Trevinho: you're making larsu very happy right now [09:28] hi desrt allright, you? [09:28] davmor2, yw [09:28] Trevinho: great :D [09:28] thanks for asking [09:28] Trevinho: "awesome" [09:28] * happyaron wonders whether we'll put input methods into snaps? [09:28] you guys are enjoying london? [09:28] happyaron: add it to the etherpad [09:28] desrt: ok [09:29] happyaron: we had ideas of snaps for fonts and themes, but input methods didn't come up, i think [09:29] Trevinho: your workaround for not supporting frame extents was to draw the shadows for csd windows server side, right? [09:29] Trevinho: but those windows have argb visuals for rounded corners [09:29] desrt: added [09:31] * Trevinho larsu: it used to be at trusty times, or well, it was possible to assume scenarios [09:31] to get csd and shadows [09:32] I don't understand [09:32] but, things have changed, so if there's an argb visual we can just end up checking if it has gtk extents... And maybe add shadows anyway [09:32] it won't have frame extents [09:33] we disable them on unity now so that we don't draw shadows client side [09:33] that's the patch you proposed.. [09:33] larsu: ah, ok... and with that they still have an argb visual? [09:33] Trevinho: not upstream, but I want to change that [09:34] Trevinho: otherwise we won't get rounded corners [09:34] larsu: well, there multiple things acting in this case.. [09:35] larsu: for example, chrome does rounded corners by adding multiple subwindows (so the window is not actually a rectangle), and in this case we still can't add shadows (well we could, but not in the "best way")... [09:36] Trevinho: we don't draw shadows for chrome? [09:36] larsu: the other way is the one you want to do, but it's quite hard to us to recognize wether the app want to just round a corner or have a full transparent block inside [09:36] larsu: not with its own decorations [09:37] Trevinho: that's what the opaque region hint is for, no? [09:37] larsu: I have a fix for that scenario, but I was worried of possible regressions for other cases [09:37] what's the fic? [09:37] larsu: yaeh, but we don't support it yet [09:37] larsu: nor most of the widgets I know of [09:38] Trevinho: widgets? Do you mean toolkits? [09:38] so, there might be some scenarios were we might draw unwanted shadows [09:39] larsu: no sorry... I used that word in the "osx/windows term".. .I mean things like cairo-clock or similar things [09:39] "screenlets"?! [09:40] oh! [09:40] people still use those? [09:40] those should just set an empty opaque region so that we don't draw a shadow, right? [09:41] yeah.... Unfortunately... As soon as I changed some shadowing rules I got tons of bugs for very-very edge cases I didn't think of (or at least, that I thought were not important anymore) [09:41] or just don't draw one if *no* opaque region is set for argb windows, but honor the region if it is set [09:41] yeah [09:41] csd are becoming a non-edge case [09:41] yep [09:44] mdeslaur: hey! we're working on that gsettings confinemnet dameon. Where do we get the list of keys an app is allowed to access? [09:51] jdstrand, ^ fyi, we start working on that === MacSlow is now known as MacSlow|lunch === pstolowski is now known as pstolowski|lunch [11:36] pitti: hey... got a second to chat? [11:36] i'm wondering about this 'appname.developer' naming scheme [11:36] for snaps... [11:36] it's making our lives very difficult [11:37] "our" [11:37] mostly larsu and seb -- i don't have to deal with this part of it =) [11:53] larsu: I don't really know, perhaps tyhicks does [11:54] desrt: I'm the wrong Martin to bark against; I think you want Martin Albisetti [11:55] mdeslaur: thanks [11:55] arf arf [11:55] pitti: seb told me to bark your way :) [11:56] seb128:? [11:56] desrt, ? [11:56] * pitti <- not a snappy designer/developer, I just occasionally do some consultancy when mvo is being nice to me :) === MacSlow|lunch is now known as MacSlow [11:56] larsu: ? [11:56] pitti: sorry, I was the one to recommend you [11:56] pitti, just for the record I didn't and don't see why I would have [11:56] desrt, stop lying please [11:56] * pitti hugs seb128 and desrt [11:57] seb is angry at me now :( [11:57] * seb128 hugs pitti [11:57] no hugs [11:57] yes -- indeed it was larsu... i confused them [11:57] :-) [11:57] too many people with beards these days [12:00] pitti, I can't imagine mvo being anything other than nice, so I guess that consultancy is full time === chrisccoulson_ is now known as chrisccoulson === greyback__ is now known as greyback [12:01] * ogra_ thinks mvo needs to seriously work on his image [12:01] :) [12:03] ogra_: https://s-media-cache-ak0.pinimg.com/236x/ab/b3/57/abb3576d2fc35561d8c6472556d042ce.jpg [12:04] lol [12:04] * ogra_ hugs mvo [12:15] mvo: :) [12:17] seb128, are you sprinting this week? === pstolowski|lunch is now known as pstolowski [13:08] chrisccoulson, hey, yes, next to hyde park ... are you in London? [13:12] seb128, that sounds like fun :) [13:14] seb128, I'm not in london. I don't think I've been there since the last desktop sprint I attended [13:15] chrisccoulson, you should come on thursday for the release/party at the pub in the evening ;-) [13:20] larsu, seb128: I would start with just the things we need. iirc, locales, grid units and vibrate. add others when needed [13:20] tyhicks: ^ [13:21] jdstrand, tyhicks, hey, I think the question from larsu was rather how/where do we specify what keys can be read and what's the mechanism to get that list to the new service [13:23] jdstrand: this is for desktop. We need a lot more than that [13:23] the desktop is currently unconfined. are you talking about unity8 desktop? [13:23] jdstrand: ah sorry yes. This is futurey work [13:24] jdstrand: desrt is busy writing that gsettings dbus service [13:24] jdstrand, yes, trying to make that service properly so it covers our futur convergence needs [13:26] let's wait til tyhicks gets here. https://blueprints.launchpad.net/ubuntu/+spec/security-r-app-gsettings dictates the implementation ("Option 3 is the one we've selected") but I imagine we would start with it maintaining its own whitelist and we would add apparmor policy at a later date [13:26] since we're pretty much booked atm [13:28] jdstrand: you mean for the changes needed in apparmor? (are any needed?) [13:29] can we maybe store arbitrary metadata with a profile? [13:29] larsu: the spec says that you would express the list of whitelisted keys in apparmor, and gsettings would use the libapparmor api to query if it is allowed for that profile [13:30] larsu: I'm saying we won't do that at first and instead whitelist via some other method [13:31] so just toss something in /etc ? [13:31] jdstrand: what could this other method be? Something in the package? [13:31] sure [13:31] /etc/gsettings-access-thing.d/appname with a list of keys? [13:32] s@/etc@/usr/share@ sorry :) [13:32] /etc, hardcoded, whatever makes sense for the implementation [13:32] you're making larsu cry :) [13:33] maybe we are talking about different things [13:33] we just want a way to add some randomish unstructured data to apparmor profiles that we can query [13:33] key/value style [13:33] where key is whatever and the value is a list of dconf paths [13:33] I'm not saying that the first iteration will define different whitelisted keys per app. I'm saying, list whitelisted keys that are safe for all apps. then later, we can add per app [13:34] no -- we need to do app settings from the start [13:34] in fact, this is the thing we want to do first [13:34] well, we don't have resources for the apparmor side atm [13:34] we're not asking you to do anything [13:34] we're only asking for information [13:35] ie: is it possible to add random key/value style metadata to apparmor profiles for an app? [13:36] not at the moment. apparmor doesn't function as a database. that is something along the lines of where we would want to go-- where a trusted helper (in this case, gsettings) can query the libapparmor api [13:36] this is how dbus currently works [13:36] right. okay. that's what we needed to know. [13:36] but, we don't have a generalized way of doing this atm [13:37] could we abuse something like listing read/write access to files under some imaginary path prefix? [13:37] like filesystem read access to /dconf/org/gtk/* for example [13:37] just as a temporary shim.... [13:38] that would work if the file query interface was implemented, which it is not yet (planned, but not short term) [13:38] okay. so same problem :) [13:38] yes [13:40] what some other trusted helpers have done is use the libapparmor api to see what label an app is confined under, and then either make decisions based on that and/or integrate with something like trust-store [13:40] do labels have to exist? [13:40] everything has a label. if it is unconfined, it has the "unconfined" label [13:41] oh. got it [13:41] label is like unique identifier [13:41] is that the same ss the 'policy_group' thing in .apparmor files? [13:41] well, not necessarily unique [13:41] larsu: no [13:41] this is the APP_ID [13:41] hm? [13:42] dotty has a "networking" policy_group... that's not its app id I assume? [13:42] https://wiki.ubuntu.com/AppStore/Interfaces/ApplicationId [13:42] larsu: it is not [13:42] larsu: the APP_ID is composed of the package name, the appname and the version [13:43] jdstrand: do we have a way of translating that to the current app id we use in unity, dbus, gsettings, indicators, etc.? [13:43] so something like: com.ubuntu.developer.rancell.dotty_dotty_1.0 [13:44] larsu: the APP_ID is the common theme in the whole process [13:44] larsu: the launcher determines the APP_ID, then launches the program in such a way that unity has it and that it runs under the apparmor profile [13:45] the APP_ID is the profile name for store apps [13:46] for non-store/unconfined, I don't know what unity sees, but apparmor will use "unconfined" [13:46] therefore you can do something like: [13:46] profile = libapparmor api call [13:46] if profile == "unconfined": [13:46] give it what it asked for [13:46] else: [13:46] do something else [13:47] unconfined apps shouldn't need to use this service [13:47] and the "do something else" is the interesting part that we're asking about here…. [13:48] this sounds like it would be a good topic for UOS [13:49] yeah sure [13:49] tyhicks isn't here now, and I'd like him to be, and I'm a bit strapped for time atm with snappy work for the next couple of days. can we do this at UOS and if not then, perhaps a hangout friday/next week? [13:50] ok, let's do the uos thing [13:59] awesome, thanks [14:37] Is there a way to remove dash icon from launcher (ubuntu 14.04)? [14:58] tiheum, hey, how are you? we are talking about bug #1388235 and the equivalent desktop ... some time ago you said you said you would look at using touch indicator icons on desktop or updating the desktop on, do you still plan to work on that? [14:58] bug 1388235 in Ubuntu UX "[indicators] Battery icon turns red at 30%" [Medium,Triaged] https://launchpad.net/bugs/1388235 [15:04] seb128: Hi, I am fine and you? In the last few weeks I worked to expandthe Suru icon theme to categories of icons that are more specific to the desktop (files, folders, etc). There's still lots of things to do on this side. According to JohnLea_, the plan is to use Suru icons for the next desktop release but only for Unity 8 (to push users to test it). [15:05] of course, the theme could be used with Unity 7 as well but it won't be the default [15:05] tiheum, would the indicator icons fit in the unity7 themes? [15:06] seb128: no, unity 7 theme needs an update in order to support the new battery levels [15:06] tiheum, would that be lot of work? [15:07] seb128: just one hour or two [15:07] so maybe we can get that done next cycle? ;-) [15:07] seb128: sure :) [15:07] great! [15:08] tiheum, thanks for the update ;-) [15:08] seb128: np [15:11] kenvandine, waits boooo [15:12] well... wait_select :) [15:12] those are often better [15:12] yeah, just looked at the change, I though you added sleeps ;-) [15:12] seb128, i asked for a qa review of that [15:12] oh.... no way :) [15:13] kenvandine, changes look ok to me, I can approve the mp if you want, would still be good to get a qa review [15:14] thx [15:14] an approval from you would be good [15:14] but still want to wait for qa before landing it [15:16] kenvandine, do you know why the "self.assertThat(self.page, NotEquals(None))" that you removed was there? [15:17] if the page isn't there, the other assert will fail [15:17] and elopio complained that we had too many asserts per test [15:17] which i don't completely agree with [15:17] but this one seemed un-necessary [15:18] right, agreed, thanks [15:18] np [15:18] comment approved [15:31] meeting!! [15:31] o/ [15:31] o\ [15:31] seb128, didrocks, larsu, Laney, attente, qengho, happyaron, FJKong, Sweet5hark, willcooke: roll call! [15:32] hi desrt [15:32] hey [15:32] desrt, you master of the bot?! [15:32] everyone ready with their notes? [15:32] hey [15:32] * Sweet5hark is late ... [15:32] #startmeeting [15:32] Meeting started Tue Apr 21 15:32:15 2015 UTC. The chair is desrt. Information about MeetBot at http://wiki.ubuntu.com/meetingology. [15:32] Available commands: action commands idea info link nick [15:32] ..|.. [15:32] Laney not taking part in the roll call again... [15:32] will said there was none. [15:32] seb128: you're up first [15:32] he's trolling us all [15:32] desrt took over [15:32] don't rise to the bait [15:32] qengho: trolling in London... [15:32] * sprinting in London [15:32] [15:32] so now that we are all here: Who will bring me a bottle of Cola? [15:33] seb128: sounds very interesting. [15:33] * in London [15:33] Sweet5hark: your future self [15:33] over [15:33] #topic didrocks === meetingology changed the topic of #ubuntu-desktop to: didrocks [15:33] didrocks: hey. what's up this week? [15:33] desrt: nothing good :p [15:33] . [15:33] didrocks: :( [15:33] larsu: my present self! [15:33] anything we can help with? [15:34] okay. thanks didrocks. hope things look up. [15:34] #topic larsu === meetingology changed the topic of #ubuntu-desktop to: larsu [15:34] meh [15:34] (meh) [15:35] [15:35] #topic Laney === meetingology changed the topic of #ubuntu-desktop to: Laney [15:35] show me some love, please [15:35] everyone else seems pretty depressed [15:35] I did a glib patch [15:35] waiting for review [15:35] [15:35] cool. i'll try to get around to that. [15:35] ሴ [15:36] Laney's patch is about fixing trash on the live cd, fwiw [15:36] thanks Laney! [15:36] #topic attente === meetingology changed the topic of #ubuntu-desktop to: attente [15:36] attente: your turn [15:37] attente: don't let us down [15:37] 🐧 [15:38] continued work on service for switching between input method frameworks [15:38] thanks attente [15:38] #topic qengho === meetingology changed the topic of #ubuntu-desktop to: qengho [15:38] qengho: hey. want to go? :) [15:39] 💩 [15:39] we can come back to qengho later [15:39] #topic happyaron === meetingology changed the topic of #ubuntu-desktop to: happyaron [15:39] happyaron: hey. what's going on at that end of the table? [15:39] [15:39] 在伦敦开会 [15:39] [15:40] Hah [15:40] the above are real unicode characters [15:40] didrocks, seb128, didrocks, desrt, etc: hey guys, want to meet for dinner later this evening? [15:40] done [15:40] pitti: we're going rock climbing [15:40] desrt: all of you? [15:40] but maybe after? [15:40] (or come) [15:40] pitti: all except didrocks :/ [15:40] so maybe tomorrow [15:41] want to come climbing? [15:41] desrt: I suck at that [15:41] also, -ETOOMUCHTODO here [15:41] pitti: me too, but it will be fun anyway [15:42] we have team dinner tomorrow [15:42] but you are welcome [15:42] pitti: depends, if we can meet half-way, why not :) [15:42] 以逸待勞以逸待劳 [15:43] okay then [15:43] moving on... [15:43] #topic FJKong === meetingology changed the topic of #ubuntu-desktop to: FJKong [15:43] FJKong: hey. what's up? :D [15:43] didrocks: yeah, we can meet at some tube station [15:43] looking into scope developing in U7, but no one maintain it... oh no, horrible [15:43] U7 API has a lot of changes it,there is no Lens at all in the new API, but only Scopes. [15:44] well [15:44] that was a very nice status update [15:44] i am happy about your honest participation in this process [15:44] good show [15:44] #topic Sweet5hark === meetingology changed the topic of #ubuntu-desktop to: Sweet5hark [15:44] Sweet5hark: i know you've been waiting... [15:44] pitti, you should come to our team dinner :-) [15:44] wat? [15:45] as stated: 以逸待勞／以逸待劳 [15:45] EOF [15:45] i agree. that sounds like a great plan. [15:45] okay. thanks! [15:45] #topic willcooke === meetingology changed the topic of #ubuntu-desktop to: willcooke [15:45] willcooke: any updates for the week? [15:45] seb128: sounds good too (we are five people here at the release sprint) [15:45] I saw a cat [15:45] [15:45] nice! cats are awesome! [15:46] It was a pretty good week [15:46] #topic other_business === meetingology changed the topic of #ubuntu-desktop to: other_business [15:46] anything else? [15:46] thanks [15:46] I saw two different types of birds as well [15:46] oh, sorry for breaking your meeting [15:46] that is quite some other business indeed [15:46] pitti, maybe the few of us who go climbing can meet you for beer tonight after the climbing? [15:46] pitti, still having the same mobile number? [15:46] pitti: you didn't, this is some kind of long troll by desrt ... [15:47] seb128: yes, hasn't changed in many years [15:47] it takes patience.... [15:47] pitti, k, good, let's see how the evening goes then [15:47] okay. thanks everyone! [15:47] #endmeeting === meetingology changed the topic of #ubuntu-desktop to: Welcome to the Desktop Team | https://wiki.ubuntu.com/DesktopTeam | For support please join #ubuntu | Issues with GTK 3.14? Check here: http://pad.ubuntu.com/gtk-update-v [15:47] Meeting ended Tue Apr 21 15:47:21 2015 UTC. [15:47] Minutes: http://ubottu.com/meetingology/logs/ubuntu-desktop/2015/ubuntu-desktop.2015-04-21-15.32.moin.txt [15:47] thanks! [15:47] thanks! [15:47] seb128: looks we can come tomorrow, too [15:49] pitti, yeah, well we can do that 2 nights it's not that often that we see each others nowadays [15:49] seb128: right; when do you guys get climbing? [15:49] s/get/go/ [15:49] 6pm [15:49] Laney said it should take 1.5-2h [15:49] + traval [15:49] travel [15:49] wow, my arms are usually a wreck after 1 h :) [15:50] so maybe 8:30-9 pm [15:50] yeah, I'm not even sure I'm going yet [15:50] I never tried climbing ;-) [15:50] anyway, dinner around 8 seems fine to me, we can come your way [15:50] those guys don't joke though [15:50] great [15:50] let's keep in touch [15:50] just tell us a tube station and a time? [15:54] pitti, would around Convent Garden at ~20:30 works? [15:54] seb128: sure [15:56] great [15:57] pitti: should we met at Convent Garden a little bit earlier? [15:57] have a beer and so on [15:58] didrocks: if we manage to get out of here early enough, sure :) [15:59] hum, seems you won't then :p [15:59] ok, I'll see if I meet you guys there [16:00] didrocks: yesterday we went out around 19:30, plus tube and stuff, so 20:00 sounds possible [16:01] pitti: ok, let's forget about it, if I was going there, I would go walking, so I'll stay here it seems [16:01] will see you tomorrow! [16:01] didrocks: hmkay [16:02] * Sweet5hark is afraid of heights: climbing -> scary. jumping out of a flying plane -> no problem. [16:03] or maybe, Im just afraid of exercise ... [16:04] Sweet5hark: My wifes first flight she was more scared looking out of the window when the plane was on the tarmac than she was when the thing was in the air :D [16:20] * ogra_ curses [16:20] evolution got completely unusable for me on trusty :( [16:20] doesnt survive 10min without turning the window grey today [16:20] but it didn't change? [16:21] no, and i didnt upgrade or anything [16:22] i have like 80 mails that i have marked read for about 20 times since this morning though ... every time i restart the app they come back ... very annoying [16:35] * ogra_ sighs ... again ... [16:36] can you get a backtrace and/or file an upstream bug? [16:36] yeah, i guess i have to [16:38] sorry for your bug :( [16:40] heh, not your fault [16:40] my general desktop experience is quite awful since i switched my desktop from precise to trusty [16:40] use vivid! [16:40] * larsu hides [16:40] this is only one piece of the picture [16:41] larsu, i do, on my laptop, but my desktop runs critical scripts and stuff ... even rebooting takes me ~1h to get everything up in order [16:41] oh wow [16:42] (and every time i do that i swear to myself to write a master script to do all that :P ) [16:59] * qengho is afk for driver's-license appointment. [17:00] * ogra_ hopes you get a well open licensed driver then ... not some binary crap [17:00] ;)