/srv/irclogs.ubuntu.com/2015/03/12/#ubuntu-desktop.txt

=== duflu_ is now known as duflu
=== l3on_ is now known as l3on
=== mchro- is now known as mchro
pittiGood morning06:04
didrocksgood morning06:21
didrockshappyaron: hey, I see your patch on presage, wasn't there supposed to be one on fcitx-qt5 for the build-dep (and extra-cmake-modules?)06:32
didrockshappyaron: ah good, seeing it now (for some reasons, didn't spot it in -changes)06:45
didrockshappyaron: however, you are pulling gnome-icon-theme-symbolic/gnome-icon-theme (through  fcitx-config-gtk, mind fixing that one?06:46
didrocksthe rest looks mostly good from germinate-output06:46
pittibonjour didrocks06:47
didrockshey pitti!06:47
mvo_didrocks: there was talk that the qr-creatorimplements a daemon for the handling of the chroots, but I was not following that (cf the discussion with seb128 yesterday)06:53
didrocksmvo_: ah, so we will never *ever *ever* know he was executing the old code which was printing the new code on error (not a biggie, a reboot worked for him) :p06:54
mvo_didrocks: *cough* yeah06:56
didrocksmvo_: surely a north-french-german thing :p06:56
mvo_:P06:56
mlankhorstmorning07:28
didrockshey mlankhorst07:30
didrockspitti: hum, for bug #1428486, seems there are a lot of NEED_*, should I only take care of STATD in your opinion? With the strategy I wrote on my last comment?07:43
ubot5bug 1428486 in nfs-utils (Ubuntu) "Only start rpc.statd if $NEED_STATD" [Low,Triaged] https://launchpad.net/bugs/142848607:43
robert_ancellCan anyone help me try and work out how the ubuntu-desktop depends on the ttf-indic-fonts-core package? It's not in the seed afaict and reverse-depends isn't showing any package that should be bringing it in.07:46
pittididrocks: "lot of NEED_*"? there is only NEED_STATD and NEED_GSSD, right?07:48
pittididrocks: and the startup of gssd is done automatically based on whether you use kerberos, so this is just a dead setting07:48
pittididrocks: i. e. if it was empty, it's determined automatically07:48
didrocksrobert_ancell: hey, it's in the desktop-common seed (I guess pulled by the usb)07:49
didrockspitti: some jobs are using NEED_IDMAPD07:50
robert_ancelldidrocks, ah, is "usb" the standard target (i.e. the CD replacement)07:50
pittihm, that's not even in the default file07:50
didrocksdebian/nfs-common.init used it07:51
didrockspitti: it executes rpc.idmapd07:52
didrocksno idea what that is07:52
didrockspitti: so, as it's a complex case, I propose to be more conservative than for others: I don't ship .override files for upstart, keeping the variables for them (only sedding "this isn't used under systemd" in the conffiles)07:53
didrocksdo the one time transition for systemd07:53
didrocks+ the [Install] patch for enabling/disabling via sytemctl07:53
didrockssounds good to you?07:53
pittididrocks: (followed up)07:54
pittiphone, b4rb07:54
didrocksok, will do the STATD part already07:55
pittididrocks: still feeling a bit uneasy about that, but I guess if there's a use case for controlling these manually, so be it08:06
pittididrocks: I actually like the simple nfs-{client,server} split which you can enable/disable, and the rest is automatic08:07
didrockspitti: hum, it seems you used git to format your patched in nfs-utils, but if I branch git://git.debian.org/kernel/nfs-utils.git I only saw debian branches, no ubuntu?08:07
pittididrocks: no, I did some of them against the upstream git and just copied them there08:07
pittithe ubuntu upload was just a simple apt-get / quilt / dput thing, no VCS08:08
didrockspitti: agreed that should be automatic, but maybe let's just get on parity until LTS and revisit once we removed the need for all those configs08:08
didrocksok08:08
pittididrocks: well, once you introduce an [Install] it's more work to get it back to automatic08:08
pittididrocks: anyway, I do see why you would want to disable statd (if you only use NFSv4)08:08
pittididrocks: but I'd leave gssd on automatic until we at least get a comment from Steve08:09
didrockspitti: +1, doing that then08:09
happyarondidrocks: I can weaken the dependency on configuration tool, to get them (including fcitx-config-gtk) out for the moment08:10
didrockshappyaron: yeah, please do a suggests08:10
happyaronit's not needed for out-of-box experience, but input method is very related to user's tastes, people may end up installing it themselves08:10
happyaronok08:10
didrocksthanks :)08:11
=== dbarth__ is now known as dbarth
seb128good morning desktopers08:38
didrockssalut seb12808:38
seb128lut didrocks, en forme ?08:39
didrocksรงa va, et toi ?08:39
larsuochosi: cool!08:48
larsumorning desktoppers08:48
seb128hey larsu08:48
willcookemorning08:52
didrockshey willcooke, larsu08:54
larsumorning!08:55
larsuhow are you all doing?08:55
ogra_pitti, seems there is still some systemd fallout in desktop-next images (they fail)08:55
seb128hey willcooke08:56
seb128ogra_, fail to build? or boot? or...?08:56
ogra_build08:56
ogra_https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/vivid/ubuntu-desktop-next08:56
ogra_i wonder why i only got mail about it today08:56
ogra_seems to happen since three days already08:56
seb128 systemd-sysv : Conflicts: upstart but 1.13.2-0ubuntu9 is to be installed08:57
seb128:/08:57
ogra_right08:57
pittiogra_: hm,  could it be that these also want to install ubuntu-standard?08:57
ogra_yeah, i suspect so08:57
pittiubuntu-standard wants systemd-sysv, but the touch seeds want upstart08:57
ogra_thats a tricky one08:57
pittiso I guess we'll need to see if u-d-n works with systemd, then we can just drop the upstart seed in the touch seeds08:57
ogra_only desktop pleae ;)08:58
pitti(yes, sure :) )08:58
pittiogra_: did the fakechroot debootstrap work now?08:58
seb128pitti, we use upstart for user sessions08:58
ogra_pitti, yeah, works wonderfully08:58
pittiogra_: \o/08:58
pittiseb128: sure, we do that everywhere (that's upstart-bin)08:58
seb128pitti, desktop-next should works fine under systemd (the only issue I had this cycle was the cgmanager not start making the dash app start buggy, but that was resolved)08:59
seb128not starting*08:59
pittiah, great09:00
Laneymorning09:10
seb128hey Laney09:11
* Laney is getting annoyed with no network on every resume09:11
willcookepopey, ping09:16
willcookepopey, free for a quick chat?09:16
Laneyok, updated the bug again09:17
Laneyhey seb128!09:17
popeywillcooke: sure09:21
popeygot 9 mins before next call09:21
=== cking_ is now known as cking
pittiogra_, seb128: touch seeds pushed, https://launchpad.net/ubuntu/+source/ubuntu-touch-meta/1.213 updated09:37
seb128pitti, danke09:38
* ogra_ crosses fingers09:38
pittiogra_: do you have a retry button on https://launchpad.net/~ubuntu-cdimage/+livefs/ubuntu/vivid/ubuntu-desktop-next ?09:38
pitti(once it's published)09:38
pittiand nice, my world model still had these as cronjobs and shell scripts and logs on people.u.c.; I didn't see that page before09:38
ogra_i'll just fire a build on nusakan09:39
Laneythose builds are triggered by the rest of cdimage09:39
Laneyyou can't really retry them there09:39
seb128I didn't know about that page either09:40
seb128ogra_, pitti, please let me know when there is an image build, I can test it09:42
seb128rsalveti, hey, I'm not sure that the bluez change to have a timeout by default on the discoverable state play nicely with the desktop, we have a manual switch in the indicator/settiings there and the status keeps reverting by itself now09:44
Laneybah, now I have a fcitx indicator10:00
pittitoday's dist-upgrade installed a gazillion fcitx packages, was that really intended?10:01
pittidon't we already have ibus for that?10:01
LaneyI think the plan is to have it installed on-demand by language-selector10:04
Laneyseems like too much is being installed currently10:05
seb128pitti, yeah, it shouldn't pull it by default, we need to look at fixing that10:05
seb128those landing aimed at adding better support for fcitx for those who use it10:05
seb128which is mostly China users10:05
seb128we don't want to change the default or pull both framework for everyone10:05
LaneyI guess a bug report would be welcome10:06
Laneyah crap, now I'm trying to remove "shouldn't" from my bug text10:08
* Laney stares at Lars Paul Thomas10:08
* Laney just decides to use the damn word :p10:09
Laneyseb128: do we have a way of tracking bugs for this change?10:11
Laneytag or assign someone?10:11
larsuLaney: ts ts ts10:13
Laneyha10:13
Laneyit's "I shouldn't see this"10:13
Laneyprobably not a violation of the rule anyway10:14
larsuinteresting edge case10:14
Laneyremoved one case from the bug title at least10:16
Laneybug #143125710:16
ubot5bug 1431257 in fcitx (Ubuntu) "fcitx indicator shown when fcitx isn't in use" [Undecided,New] https://launchpad.net/bugs/143125710:16
larsuLaney: I think mpt would still consider that as a wrong use of 'should'10:17
larsu"showing the indicator even though I only have one input source configured is not useful and visually distracting"10:18
Laneyhaha10:20
LaneyI think "this shouldn't be shown because x" and "I don't want to see this because x" is quite a fine distinction10:20
larsuindeed10:21
* larsu stops being pedantic and gets back to work10:21
Laney:p10:21
mptLooks fine to me ๐Ÿ‘Œ10:28
seb128Laney, "this changes" being fcitx? not that I know, just assign to happyaron or attente_ I guess10:42
didrockspitti: ok, I think most of the raw edges around statd are now gone, as always with conffiles, I prefer a double checking, mind giving a look if your eyes bleed on something? http://people.canonical.com/~didrocks/tmp/nfs_statd.debdiff10:42
didrocksseems trivial looking at the diff, but there were quite some corner cases :p10:42
didrocks(tried different updates/install patterns, under systemd)10:43
Laneyseb128: okay!10:43
didrocksseb128: oh FYI, we discussed this morning and the dep on gnome-icon-theme will be removed10:43
didrocks(spotted that from germinate)10:44
seb128didrocks, oh, thanks10:44
didrocksI guess fcitx-blah-config will degrade the dep to suggests10:44
seb128or change it to adwaita instead?10:45
didrockshappyaron: any opinion on this? ^ I know you invoked user's taste, but that could be a better option?10:46
pittididrocks: can you please add a bug ref to the changelog?10:46
Laneydid we get g-i-t demoted?10:46
pittididrocks: and "let the admin enabl*e* (not *ing*)10:47
didrockspitti: yeah, this was already done after the debdiff, ok on enabl* :)10:47
Laneyoh right, it's in main for metacity10:48
seb128Laney, it's out of the desktop iso at least10:48
Laneyya10:48
Laneywe don't get alerted via component-mismatches though10:49
seb128right10:49
Laneyhttp://people.canonical.com/~ubuntu-archive/component-mismatches.svg10:49
Laneywhat's all that font stuff?10:49
Laneywhat even is the usb seed?10:49
seb128robert_ancell's doing10:50
seb128I didn't follow closely but those were sync blacklisted for years10:50
Laneyah10:50
seb128apparently Colin did that to get the autosync going in some cycles and they got ignored since10:50
Laneya lot of font packages got renamed in debian10:50
seb128right10:50
Laneyprobably the new one is in universe or something10:50
seb128Robert cleaned up the mess10:50
Laneymakes sense10:51
Laneyhe probably just needs to chase promoting stuff10:51
seb128seems like some users in India complained about the fonts being outdated10:51
seb128right10:51
willcooke@ Indian fonts - robert_ancell was chasing some people on that10:51
meetingologywillcooke: Error: "Indian" is not a valid command.10:51
seb128lol10:51
Laneytell that to the british empire10:52
willcookelol10:52
willcookeoh, I see seb128 already said that Robert was sorting it10:52
seb128willcooke, well, anyway robert_ancell found me and we unblocked the syncs, so should be mostly good10:52
willcooke:) thx seb12810:52
seb128yw10:53
willcookeback to reviews :(10:53
LaneyI will ignore this for a week ;-)10:53
willcookeYou know, staring at a blank page for an hour isn't helping10:53
willcookeI don't know what else to try10:53
willcookeWait...10:53
willcookeTea10:53
Laneyis this writing the reviews?10:53
willcookeyeah10:53
seb128still have some of those to do :-/10:54
Laneyhttp://www.lipsum.com/10:54
willcookehahahah10:54
seb128willcooke, http://mschmitt23.com/performance/ as well10:56
willcookeniiiiiice!10:56
seb128:-)10:56
Laneyhaha10:57
seb128why staring a blank page when you can google for random ideas to get the work done for you :p10:57
LaneyPerformance review for Sebastien Bacher:10:57
LaneyPhysically fit, very rarely sick, and able to meet the demands of a physically demanding job.10:57
willcookekeep it coming10:57
seb128lol10:58
Laneythe only box I ticked was "Hygiene" to get that ...11:03
Laneyretrying desktop-next btw11:07
Laneyoh right, it's got to wait for Task: to update11:12
pitti ubuntu-touch | 1.213 | vivid/universe | amd64, armhf, i38611:19
pittiogra_: ^ can you press ze button, please?11:19
Laneypitti: I just tasked about that11:21
Laneytalked11:21
Laney*and* tasked(!)11:21
pittihah, thanks Laney11:21
Laneyyou have to wait, I think an extra cycle, for Task: to change too11:22
ogra_pitti, building11:28
Laney...11:28
ogra_meh11:28
=== MacSlow is now known as MacSlow|lunch
attente_Laney, seb128, is doing that dist-upgrade supposed to pull in fcitx-bin as a recommends?12:36
Laneyyou tell me12:36
attente_i thought it shouldn't, but i don't understand why it gets installed on upgrade12:37
Laneyrecommends will be installed, yes12:38
LaneyI have no idea what's right in this situation though12:38
larsuLaney: styling the osd stuff for totem isn't trivial12:51
LaneyI looked at the rules in adwaita12:51
Laneyseemed confusing12:51
larsuwe still use .png borders for buttons for example12:51
larsuI can make pure css ones, but they'll look very different12:51
* larsu needs to do visual design yet again :( 12:52
Laneyfor which buttons?12:52
larsubuttons in a osd12:52
larsulike play/pre/net in totem12:53
larsu*prev, *next12:53
Laneyah these ones12:53
Laneyis that what adwaita does?12:53
larsuno, they can just switch some colors around since they're draing everything in css already12:54
* larsu should not eat and type, sorry12:54
Laneyya that's what I meant12:55
Laneycssing them12:55
larsubut actually they change quite a bit about the appearance of those buttons12:55
larsuso I guess we can do that as well?12:55
* Laney isn't particularly attached12:57
Laneywould be a user interface freeze break, but I think it'll be okay12:57
larsuya indeed12:57
* Laney tries to tab complete GunnarHj and fails12:57
larsusorry I didn't get to it earlier12:57
Laneymeh, shit happens12:57
larsuwe could postpone it, but it does look pretty ugly now12:58
Laneyindeed12:58
Laneyfixing is better imho12:58
Laneyalso I noticed that if you right click on a video in the bookmarks page (the first screen) you get a huge pixellated check mark12:59
Laneyguess that's a scaled up png too12:59
happyarondidrocks seb128 I'll see if I can change it to adwaita first13:00
larsuLaney: indeed13:00
larsuI really don't want to get into fixing those until we have a new theme13:00
larsubecause it involves redrawing those in css13:01
larsuor making them into proper icons in the theme (which are scalable)13:01
* larsu wonders if we have svg assets for those13:01
Laneycan't we use a โœ“?13:02
LaneyI guess this is the hard bit?13:02
larsuno, gtkcheckbox doesn't draw text13:02
larsuah, adwaita goes the icon-theme route13:02
larsuinteresting13:02
Laneyanyway13:03
Laneyone thing at a time :P13:03
=== czajkows1i is now known as czajkowski
=== MacSlow|lunch is now known as MacSlow
willcookeFinally catching up with the meeting notes from Tuesday.  This Android notifications thing sounds cool14:18
willcookelarsu, my 2 cents >  Yes, that's something we should have.  ^14:19
seb128that's initially why didrocks started looking at the bluez update no?14:20
nessitapitti, hey there! quick question: I need to keep using juju on my vivid installation, and juju does not work with upstart yet. I was suggested by sinzui to install upstart and retry, but seems like sudo apt-get install upstart complains but then shows upstart as installed: (first 50 lines of https://pastebin.canonical.com/127463/)14:21
nessitaany suggestion how to have upstart up and working again on vivid?14:21
willcookeseb128, ah, yes  - thats right, needed some new profile support14:21
happyarondidrocks, seb128, switching to adwaita-icon-theme works14:22
rsalvetiseb128: yeah, I did raise that to cyphermox before doing the upload, but he also agreed to make it have a timeout even on desktop14:22
rsalvetiseb128: and I also saw that bluez5 would have the same change14:22
rsalvetiso if we want that to be like before, we need to change indicator-bluetooth to force timeout as 014:23
didrockshappyaron: nice!14:23
rsalvetione additional dbus call, but while it's a different behavior now, I actually find it the correct one14:23
seb128rsalveti, right, or drop that UI bit and make the desktop do like other systems, just enable it when the UI is open14:23
didrockswillcooke: yeah, we need the new profiles to do something with it without installing any app phone-side14:23
rsalvetinot having a timeout is kind of annoying, as it will help reducing your battery for sure, specially in laptops14:23
rsalvetiseb128: right, yeah14:23
ogra_Laney, pitti, assuming the task update has happened, i started a new build a few mins ago ... which failed again :/14:38
ogra_same error14:38
happyarondidrocks: I'm adding extra-cmake-modules to the MIR, can I assign it to you for a review? it's not urgent but ECM will be used in future fcitx releases14:52
didrockshappyaron: sure, please assign14:53
happyaronthanks14:53
Laneyogra_: ok, maybe I'm wrong about how often it updates, check with $launchpad folks15:05
Laneyit's probably better to check the field instead of assuming :P15:05
ogra_Laney, it should run every time the publisher runs15:05
ogra_afaik15:06
=== alan_g is now known as alan_g|EOD
Sweet5harklarsu, desrt: how is gtk3 on MIR? just asking because LibreOffice on native gtk3 seems to be making great strides: http://caolanm.blogspot.de/2015/03/gtk3-vclplug.html ;)19:07
willcookeThis is very good news :)19:10
mlankhorstSweet5hark: you run it in XMIR!19:11
Sweet5harkwillcooke: yes, indeed.19:11
Sweet5harkmlankhorst: thats boring, willcooke runs LibreOffice in XMir every second day ... ;)19:13
mlankhorsthaha19:13
* Sweet5hark is greedy and wants LibreOffice on native MIR ...19:13
mlankhorstyou can also become a Xmir maintainer and add windowed mode to Xmir :D19:13
* Sweet5hark hides19:14
willcooke:D19:14
Sweet5harkmlankhorst: you cant fool me, when we moved LibreOffice bugzilla to our own, I saw (and cleaned) the quips file, including all the dirty laundry of Xorg hackers ...19:16
Sweet5harkincluding the self referencing "[alanc] https://bugs.freedesktop.org/quips.cgi?action=show are the most  interesting bits of this channel. not the useful ones though"19:17
* willcooke -> EOD19:36
mlankhorstSweet5hark: best one is still th topic in dri-devel...19:57
mlankhorst <ajax> nothing involved with X should ever be unable to find a bar19:58
sarnoldhah set in 2012 :)19:58
Sweet5harkmlankhorst: yeah, thats also in the quips ..19:58
mlankhorstits older than 201220:00
mlankhorstbut its double, since all graphic cards have bars too\20:00
seb128robert_ancell, hey20:06
robert_ancellseb128, howdy20:07
seb128robert_ancell, how are you?20:07
robert_ancellgood20:07
seb128great20:07
seb128doko was looking for you20:07
seb128you put http://people.canonical.com/~ubuntu-archive/component-mismatches.svg on grumpy mode it seems20:07
robert_ancellah, MIR...20:08
seb128yeah20:08
seb128is that something you are looking at?20:09
* robert_ancell reverts20:09
robert_ancellI am now20:09
seb128thanks20:09
seb128revert what?20:09
robert_ancellseb128, the seed change20:09
seb128if those are just renames it's easy20:09
robert_ancellseb128, can you approve MIRs? They're just metapacakges20:09
seb128but dunno if that's the case20:09
seb128no I can't20:10
seb128I'm not in the MIR team20:10
seb128doko mterry or didrocks can20:10
seb128not sure any of them is around though20:10
robert_ancellmterry should be...20:10
seb128right20:11
seb128but they don't seem to be20:11
seb128e.g fonts-kalapi is a normal font package it seems20:11
seb128(just picked a random one)20:11
seb128or you mean font-indic?20:12
seb128did it pull in the same fonts from some other binaries?20:12
robert_ancellyeah, fonts-indic20:12
seb128why did you seed it?20:13
seb128does it replace something else?20:13
seb128how much does it pull in?20:13
robert_ancellyeah, perhaps we need a subset. For libreoffice it was pulling in the old top package, so we should probably MIR them all though20:13
* robert_ancell plays dependency following20:14
Helphello23:49
=== Help is now known as Guest88739
Guest88739I have a Dell latitude e6400 and my ubuntu wireless will not work23:49
Guest88739will anyone help23:50
Guest88739this is stupid23:51
Guest88739nobody is helping23:51
Guest88739I have a Dell latitude e6400 and my ubuntu wireless will not work23:51
Guest88739will anyone help23:51

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!