[01:33] <qengho> Hi hi hi!
[04:52] <qengho> Hmm, I want my GPG agent to forget passwords at suspend/hibernate. I can HUP it. Should I put that in pm-utils' config dir, or a new systemd service that "sleep" target wants? This last bit sounds suspicious to me.
[04:53]  * qengho silently mourns systemd eating everything.
[05:50] <pitti> Good morning
[05:50] <qengho> Hi!
[06:49] <seb128> good morning desktopers
[06:49] <desrt> saluton, karaj!
[06:49] <desrt> hey seb, pitti, qengho :D
[06:49] <seb128> oh, desrt is back?
[06:49] <seb128> hey desrt ;-) had good holidays?
[06:49] <desrt> yup.  awesome.
[06:50] <desrt> and now i have an awesome inbox... for some definition of "awesome"
[06:50] <qengho> desrt, seb128: heyheyhey!
[06:51] <seb128> the post holidays deal-with-emails day ;-)
[06:52] <desrt> ....on the topic of email
[06:52]  * desrt guesses she needs to figure out gmail at some point
[06:52] <seb128> if you use it?
[06:52] <desrt> i just got google apps... i'm kinda sick of fastmail
[06:53] <flocculant> desrt: what you need to figure out is mark everthing as read and hope important things get resent :p
[06:53] <desrt> and they sorta forced my hand by disabling their secondary authentication systems last monday, with two days advance notice.  i use the secondary system as my primary, with my primary password written only on a piece of paper at home.  sigh.
[06:53] <desrt> so i lost email until i go home to toronto (in half a month)
[06:53] <desrt> so for now i sent it all to NSA^Wgoogle instead
[06:54] <qengho> desrt: fastmail talks some U2F now?
[06:54] <desrt> u2f would be cool
[06:55]  * qengho thinks he read that.
[06:55] <desrt> i'm sure their new system is better than the old one.  what's totally not cool about it is that they disabled the old one with two days of notice.
[06:55]  * qengho nods.
[06:55] <qengho> Two months is barely acceptable.
[06:57] <didrocks> good morning!
[06:57] <qengho> The Didier. Hi.
[06:58] <didrocks> hey qengho ;)
[06:59] <qengho> desrt: Yep, I did read that. Fastmail talks U2F now. Not that you can trust them any more. But, a small consolation.
[07:00] <pitti> hey desrt, how are you?
[07:00] <pitti> bonjour seb128
[07:01] <desrt> awesome. looking forward to start working again
[07:01] <desrt> ...after i dig myself out from email
[07:02] <qengho> desrt: You're throwing away an excuse to declare email bankruptcy! :)
[07:03] <desrt> hmmmmm
[07:03] <desrt> i'm fairly sure there's some important stuff in here ;)
[07:06] <pitti> I just don't get along with gmail's web UI; it's still bearable for a couple of emails, but way too clumsy for the volume of work email that we get
[07:07] <qengho> It's annoyingly close but-not-exactly Mutt's keybindings.
[07:17] <larsu> desrt: morning. subscribe to their "blog" (you get email notifications) - they've been talking about this for a while
[07:18] <larsu> also, they said they won't remove the old system until Aug 31
[07:18] <desrt> well, it stopped working on monday, as mentioned in the email that they sent to me
[07:18] <desrt> i've already decided to let it go, in any case
[07:18] <desrt> i'm living almost 100% in the G universe in other aspects now, and this is sort of the last thing i need to do
[07:19] <desrt> and fastmail's phone app is so so so bad
[07:19] <desrt> which is pretty much the thing that is really killing it for me
[07:20] <larsu> ya, this is why I went ios, their email app is amazing
[07:21] <larsu> and works with open protocols ;)
[07:22] <qengho> FWIW, I connected Gmail app to fastmail. It was pretty nice.
[07:23] <seb128> salut pitti
[07:23] <seb128> hey larsu
[07:23] <seb128> wie gehts?
[07:23] <larsu> hey seb128!
[07:23] <larsu> good. married life is SOOOO different
[07:23] <didrocks> oh, a larsu here!
[07:23] <didrocks> hey pitti
[07:24] <seb128> larsu, shouldn't you be larsk btw? ;-)
[07:24] <larsu> seb128: yeah...............
[07:25] <larsu> I think I'll keep larsu for nostalgia (and tons of people know me as that)
[07:25] <pitti> hey larsu, congratulations! *hug*
[07:25] <larsu> thank you pitti!
[07:25] <larsu> how are you?
[07:25] <seb128> desrt, you don't have somebody in Toronto who has your keys and could go see your password and give it to you over phone or something?
[07:25] <desrt> nope
[07:25] <seb128> :-/
[07:25] <pitti> larsu: quite fine indeed; we've been in Berlin over the weekend for a wedding, now looking forward to our summer vacations from Wednesday on
[07:25] <desrt> almost gave them to a friend.... but in the end, i only gave her the key to my mailbox
[07:25] <pitti> still two days to break stuff :)
[07:25] <desrt> in any case, i can still receive all of my new emails
[07:26] <seb128> yeah
[07:26] <larsu> pitti: oh nice, enjoy!
[07:26] <larsu> seb128: how are you?
[07:26] <didrocks> pitti: git push; exit 0 :-)
[07:27] <pitti> well, I broke things on Friday (systemd 231) and Sunday (systemd session start) already, so actually it's two days to sort out the fallout :)
[07:28] <Trevinho> good morning folks
[07:28] <seb128> larsu, I'm good thanks! had a nice & busy w.e, slightly tired this morning but I'm working on it with coffee ;-)
[07:28] <pitti> seb128: ah, where did you go?
[07:28] <seb128> hey Trevinho, how was your long w.e?
[07:28] <desrt> pitti: thanks for dealing the death blow to systemd-shim as an upstream project
[07:28] <desrt> i appreciate that i officially never have to think about it again :)
[07:29] <pitti> desrt: R.I.P. shim!
[07:29] <larsu> oh wow - congrats :)
[07:29] <larsu> it only took a couple of years
[07:29] <pitti> desrt: btw, prodding the Devuan guys wasn't very successful - https://twitter.com/DevuanOrg/status/758630419185209344
[07:29] <pitti> I think someone grossly misunderstood things
[07:30] <desrt> lol
[07:30] <desrt> k thx
[07:30] <desrt> it was me, and this was all quite intentional from the start, k thx :)
[07:31] <pitti> role of acquisition #285 -- No good deed ever goes unpunished :)
[07:31] <seb128> pitti, nowhere fancy, we just had busy days, we went to look for new bathrooms floor tiles on saturday, visited several shops etc, then had dinner with friends and played board games, yesterday was busy with house cleaning and some painting and small fixes
[07:31]  * Trevinho switched laptop
[07:32] <pitti> Trevinho: oh, new toys?
[07:32] <Trevinho> hey seb128, it was very nice... I've discovered a part of Tuscany I didn't know.... Awesome sea :-)
[07:32] <seb128> great!
[07:32] <seb128> what laptop did you get?
[07:32] <Trevinho> pitti: no, I mean... I got two thinkpads this year... So I switched between the t460s (mobile toy) and the t460p (docked toy :-))
[07:33] <Trevinho> so... hot-switch it was :-)
[07:34] <seb128> thinkpads... you should follow andyrock's lead, he got an actually nice laptop ;-)
[07:34] <desrt> pitti: somehow i have trouble seeing an angry tweet directed at me from a bunch of malcontents as "punishment"
[07:34]  * duflu wishes a thinkpad as lovable as the X220 existed
[07:34] <Trevinho> seb128: I spent less to get two machines... with that red thing in between my keys which... is something I can't live without now
[07:35] <desrt> duflu: it does exist.  it's called Thinkpad X220.
[07:35] <duflu> Huh. Cult of the Trackpoint
[07:35] <desrt> and you can own one for a truly excellent price
[07:35] <Trevinho> duflu: that x220 seems to be a chimera... everyone wants that. I don't see why :-P
[07:35] <duflu> desrt: Hello to you and your wit. I meant new :)
[07:35] <desrt> duflu: good morning to you too :)
[07:36] <seb128> Trevinho, yeah, the nibble is much nicer than a touchpad
[07:36] <seb128> it's still over me how people can like/use touchpads
[07:37] <duflu> A nibble is four bits. AFAIK another name for Trackpoint is nipple :)
[07:37] <seb128> it's so much less efficient and in the way of you hands which always lead to random things you don't want
[07:37] <Trevinho> Yeah... Especially when moving... I mean, I was always travelling with a mouse before. Now I can be free.
[07:37] <seb128> duflu, lol, thanks ;-)
[07:37] <Trevinho> duflu: and... For someone is also something else.... reddit docet :-)
[07:38]  * duflu realizes only now his desktop "ThinkPad" USB keyboard has one too
[07:38] <Trevinho> duflu: is that a full keyboard?
[07:38] <Trevinho> I've the compact bluetooth one...
[07:39] <duflu> Trevinho: ;)  https://support.lenovo.com/au/en/documents/migr-73183
[07:39] <duflu> It's the keyboard of an X220 in an external chassis
[07:39] <Trevinho> ah, ok... compact one then... I mean with no KP
=== davmor2_ is now known as davmor2
[08:02] <Laney> mooooooooooooorninggggggggg
[08:03] <pitti> hey Laney, wie gehts?
[08:03] <pitti> systemctl --user start laney.service
[08:05] <Laney> Process: 10256 ExecStartPre=/bin/is-awake-yet (code=exited, status=1/FAILURE)
[08:05] <davmor2> Morning all
[08:06] <Laney> it's davmor2
[08:06] <Laney> RUN
[08:06] <Trevinho> Laney, pitti: LOL :-)
[08:07] <willcooke> beep
[08:07] <davmor2> Laney: I was gonna be nice and not find any issues with 16.10 or 14.04.5 but now you've made me think I need to take a closer look so all the incoming bugs willcooke can blame you for ;)
[08:07] <Laney> pitti: good thanks! filled the pond at the allotment yesterday, hoping to attract some frogs/toads to keep the slugs down
[08:08] <Laney> how are you?
[08:08] <pitti> Laney: and hopefully not too many mosquitos? :)
[08:08] <Laney> oh davmor2 did I show you this cake I baked just for you?
[08:08] <pitti> Laney: I didn't know that frogs drive out snails
[08:08] <Laney> pitti: we're already at mosquito bankruptcy :P
[08:08] <Laney> so the others say
[08:08] <pitti> Laney: quite fine, thanks! We went to Berlin over the weekend for a wedding, it was nice (but lots of train travel again)
[08:09] <Laney> awesome
[08:09] <Laney> it's all about the Berlin weddings this month
[08:09] <davmor2> Laney: hahahaha
[08:09] <Laney> woah - last month, August already
[08:10] <davmor2> Laney: I know I need to book some holiday at some point too
[08:10] <seb128> hey Laney willcooke davmor2
[08:10] <davmor2> morning seb128
[08:11] <Laney> hey seb128
[08:11] <Laney> how's it going?
[08:11] <davmor2> hey willcooke you ping me on friday but I had the day off do you know what it was about save me digging through logs
[08:11] <seb128> good! though a bit tired
[08:12] <Laney> late night?
[08:12] <Trevinho> Laney: alpha freeze is over  now, right?
[08:12] <seb128> not really, but lot of house cleaning & small work (painting, etc) yesterday
[08:13] <seb128> apparently that was more physical work that I though :p
[08:13] <willcooke> happyaron, good afternoon!  Pls could you update your Trello board?  I added to NM Git tree issues.
[08:13] <willcooke> davmor2, <processing>
[08:14] <Laney> Trevinho: yes, why?
[08:15] <Trevinho> Laney: there was that silo to land
[08:15] <andyrock> morning all
[08:15] <Trevinho> hi andyrock
[08:15] <happyaron> willcooke: ok, doing
[08:16] <willcooke> davmor2, The "Turn off secure boot" installer option.  But s_eb128 answered, so I think you're in the clear
[08:16] <willcooke> thanks happyaron
[08:16] <willcooke> hey andyrock
[08:17] <davmor2> willcooke: ah cool thanks
[08:19] <willcooke> anyone else getting lag on Freenode?  I think it's my internet connection though
[08:20] <Laney> fast enough here ;-)
[08:20]  * willcooke checks ISP status pages
[08:21] <Laney> 01/08 09:20:07 [ctcp(willcooke)] PING
[08:21] <Laney> 01/08 09:20:07 CTCP PING reply from willcooke:
[08:21] <willcooke> so no lag there then
[08:22] <Trevinho> ah, Laney actually it seems that things have been released... not merged though, sorry
[08:23] <Laney> Trevinho: yeah that got published last week
[08:23] <Trevinho> Laney: sorry... I didn't see the mails because of bad filters but yeah...
[08:24] <Trevinho> so the fact that it wasn't merged worried me
[08:24] <Laney> np
[08:30] <willcooke> urgh.  When you think you've reached inbox zero, then you scroll the window up...
[08:39] <duflu> Or in Thunderbird you have reached inbox zero and the summary says some non-zero number
[08:39] <willcooke> ha!
[09:49] <Trevinho> what happened to yakkety? Compiz fails to build there...
[09:50] <Trevinho> The following packages have unmet dependencies:
[09:50] <Trevinho>  sbuild-build-depends-compiz-dummy : Depends: libgtk-3-dev but it is not going to be installed
[09:50] <Trevinho>                                      Depends: libmetacity-dev (>= 1:3.16.0) but it is not going to be installed
[09:50] <Trevinho>                                      Depends: libwnck-3-dev but it is not going to be installed
[09:51] <davmor2> Trevinho: upstart user session replaced with systemd but I would assume that would have no effect on the issue you see
=== pesari_ is now known as pesari
[10:05] <Trevinho> davmor2: no, I don't think so
[10:11] <seb128> Trevinho, could be the new gtk in yakkety-proposed? Laney might know about known installability issue
[10:11] <Laney> yes
[10:11] <Laney> just wait a bit
[10:19] <pitti> apport has failed yesterday and today too on gtk3 and openjdk uninstallability
[10:20] <Laney> it's just because of the icon theme
[10:21] <Laney> ...which I'm now copying
[10:21] <Laney> also there's a pending MIR, but nothing here cares about that
[10:23] <Trevinho> ok
[10:23] <Laney> copy-package to the archive is scary
[10:23]  * Laney closes eyes and presses enter
[10:24] <davmor2> Booooooooooom Did I hear the sound of a distant explosion from Laney's direction
[10:26] <pitti> tedg: I like all the green checkmarks in https://launchpad.net/~ci-train-ppa-service/+archive/ubuntu/landing-022/+packages !
[10:26] <pitti> tedg: do you know what these two mean?
[10:26] <pitti> Destination version missing from changelog (yakkety/indicator-display, yakkety/indicator-session, yakkety/indicator-sound).
[10:26] <pitti> Ready to build (vivid/indicator-location, xenial/indicator-location, yakkety/indicator-location).
[10:28] <seb128> pitti, the version missing usually means that the archive has a version that is not in the changelog
[10:28] <seb128> pitti, that usually means somebody did a manual upload/dput and didn't get it commited to the vcs
[10:30] <pitti> ah, thanks; and the "ready to build"?
[10:31] <seb128> unsure about that one, usually it means the silo is ready to build but it has built in this case so I don't know
[10:31] <seb128> or maybe there were changes to some of the merge requests and it needs a rebuild to pick those
[10:31] <seb128> which would make sense because e.g https://launchpadlibrarian.net/274526838/indicator-display_0.1+16.04.20160504-0ubuntu1_0.1+16.10.20160722-0ubuntu1.diff.gz
[10:32] <seb128> that revert a bugfix upload from yakkety
[10:32] <Laney> don't see indicator-location in the ppa
[10:32] <pitti> no, indicator-location... ^ that
[10:32] <seb128> oh, sorry I misread
[10:32] <seb128> so I guess somebody added that to the landing list
[10:32] <pitti> perhaps tedg took it out to land it separately
[10:32] <seb128> but didn't trigger a build for it since
[10:32] <seb128> or added it to the silo
[10:32] <seb128> but didn't do the build step yet
[10:33] <Laney> weird
[10:33] <Laney> it looks like it's in the build log though
[10:33] <Laney> https://requests.ci-train.ubuntu.com/log/1710/build/9/info/
[10:34] <Laney> probably more efficient to ask a train person :)
[10:37] <seb128> there are some error if you toggle debug in that log
[10:37] <seb128> 2016-07-29 22:24:12,911 INFO Diffed xenial/indicator-location.
[10:37] <seb128> 2016-07-29 22:24:12,911 DEBUG
[10:37] <seb128> yakkety/indicator-location: Diff failed: No DSC found, was it ever built?
[10:38] <Laney> it doesn't say that's an error
[10:39] <Laney> ho hum!
[10:39] <seb128> the DSC_URL is empty for it earlier in the log
[10:41] <seb128> looks like it could be a citrain bug
[10:41] <seb128> so yeah, better to ask on #ubuntu-ci-eng (or let ted sort it out)
[11:15] <Laney> Trevinho: should be fixed now, please retry
[11:38] <Trevinho> Laney: thanks
[11:39] <Laney> no problemo homie
[13:33] <tedg> pitti: seb128: We also need to put unity7 back in there, but I was waiting for their silo to land before rebuilding.
[13:33] <tedg> Since we'd have to once it landed anyway.
[13:34] <seb128> hey tedg, how are you?
[13:34] <pitti> tedg: hey, good morning
[13:34] <seb128> their silo landed on friday I think
[13:34] <Trevinho> tedg: there were few changes to do, but yeah..
[13:34] <tedg> Ah, cool. I didn't see it on Friday.
[13:34] <tedg> I'll merge trunk and get that back up.
[13:35] <Trevinho> tedg: merging trunk should be enough now
[13:38] <pitti> tedg: I tested the current landing PPA on yakkety, and with it I only have the session indicator; presumably because the inidiator-common bits landed but not the corresponding unity7 bits?
[13:39] <pitti> and I'm *really* missing my xeyes!
[13:41] <Trevinho> pitti: I guess so... is ups that requires such indicators...
[13:42] <Trevinho> pitti:  in the past it was the other way around... ups was saying "indicators, come here!" and they were started...
[13:42] <Trevinho> now they're in "Wants"...
[13:42] <pitti> they actually do seem to run in upstart, I just wonder why I can't see htem
[13:43] <pitti> but I didn't really check much, deep in debugging netplan stuff today I'm afraid
[13:43] <Trevinho> Not sure I prefer this way though... I mean, indicators won't probably change, but having unity not having the list of indicators it needs  was nicer
[13:43] <pitti> why would it?
[13:43] <pitti> we can make ubuntu-session.target Wants=indicator.target, and hook all indicators in there
[13:44] <pitti> or have unity.service Wants=indicators.target, not sure what is more appropriate conceptually
[13:44] <Trevinho> so we don't have to maintain the unity-panel-service.service (and lockscreen one), but instead it's up to indicators to come up when the service calls them
[13:44] <pitti> but unity certainly should not individually enumerate all indicators?
[13:44] <Trevinho> I agree
[13:44] <Trevinho> tedg: ^ ?
[13:45] <pitti> we should keep teh same conceptual structure as with the upstart jobs for now
[13:45] <pitti> I mean, it's fine to change it of course, but preferrably not couple that structural change with the upstart->systemd transitino
[13:46] <tedg> Well, no. We want unity-panel-service to queue a different set of indicators than for instnace unity8.
[13:46] <tedg> So there can't be an "all indicators" target.
[13:47] <pitti> so how is that done with upstart jobs?
[13:47] <tedg> We only have the shared target for the after.
[13:47] <pitti> I suppose an indicator-foo.conf has a "start on" condition for unity 7, 8, or both?
[13:47] <tedg> pitti: We send a signal and everyone that is installed runs. Which is a problem. For instance on my machine right now I have two bluetooth indicators. gnome-bluetooth and indicator-bluetooth.
[13:48] <pitti> I see
[13:48] <tedg> There's a similar problem with indicator-network and nm-applet. U7 uses nm-applet and U8 uses indicator-network.
[13:49] <tedg> Trevinho: In general, everything *is* backwards in systemd vs. Upstart. Upstart is event based while systemd is state based.
[13:49] <pitti> so an indicator-foo package should add its .service to unity7.service.wants/ and/or unity8.service.wants/ ?
[13:50] <pitti> right; in upstart you say "start" and let the "event avalanche" go lose; in systemd you say "this is the final state I want to have", and it starts transitive dependencies
[13:50] <tedg> pitti: We could go that route, I went the route where unity-panel-service lists the ones it wants. And then U8 could do the same. That's where the decision is made organizationally.
[13:50] <pitti> but this can (mostly) be mapped to each other
[13:50] <Trevinho> tedg: yeah... I know that... But there are targets. And we could maybe make targets differently for u7/8 sharing what can be shared
[13:51] <pitti> tedg: ok; indicator-foo package adding itself to wants.d/ is more decentralized and easier for new packages, with an explicit Wants= list in unity7 you have more control
[13:51] <tedg> I don't think in this case decentralized is better, you'd want to add a package recommends/requires as well. They should go together.
[13:52] <Trevinho> well, I think that's a different story honestly
[13:52] <tedg> If someone wants to hack it, they can always add it to the wants directory.
[13:52] <pitti> so for the "core" indicators a "forward" Wants= list, and auxilliary ones like indicator-multiload etc. could still use an indicators.target.wants.d/
[13:52] <Trevinho> well those kind of indicators are app-indicators...
[13:53] <Trevinho> So it doesn't apply to that
[13:53] <tedg> Yeah, multiload won't work on U8.
[13:53] <pitti> oh, ok
[13:53] <pitti> well, you guys tell me what you want and I can then hopefully tell you how to model that :)
[13:53] <seb128> static list sounds wrong
[13:54] <Trevinho> I personally would prefer avoid keeping a list of indicators in unity7...
[13:54] <tedg> seb128: You should tell debian you have issues with their control files :-)
[13:54] <Trevinho> Although the debian/* bits could have those, I'd prefer avoiding touching th service files any time
[13:55] <tedg> It seems to me that the service file should map to the control file. If it's "wants" there, it's "recommends".
[13:55] <tedg> We don't have to list it in the service file per se.
[13:55] <tedg> It could be done with a links file
[13:55] <tedg> The state in systemd would be the same.
[13:57] <seb128> I don't understand
[13:57] <seb128> the panel would load third party system indicators if they are installed in the system location no?
[13:57] <tedg> Have a links file that symlinks all of the indicator service files into the wants directory.
[13:58] <tedg> They would have to add a symlink in the wants directory, then systemd would start them at the right time.
[13:58] <seb128> k
[13:58] <seb128> but then you have 2 systems for our indicators and for others
[13:58] <seb128> that seems confusing
[13:59] <tedg> Two systems? Two places to add symlinks depending on which desktop you're interested in.
[14:00] <tedg> It allows for indicators to be in one and not the other.
[14:01] <pitti> right now I don't see how indicator upstart jobs would not start in unity8; but I guess that's what you are trying to fix here
[14:01] <seb128> sorry I though you guys were saying that the current mp has an coded list of indicators to start in unity7 code
[14:01] <pitti> oh, /usr/share/upstart/sessions/indicator-application.conf has a quirk for that
[14:01] <seb128> pitti, easy ... right
[14:01] <pitti> so right now an indicator package decides for itself where it wants to run
[14:01] <seb128> indeed
[14:02] <Trevinho> seb128: it has that
[14:02] <seb128> which makes sense imho
[14:02] <Trevinho> seb128: not in the code, but in the .service
[14:02] <pitti> the corresponding model would be to ship a unity7.service.wants.d/ link but not a unity8.service.wants.d/
[14:02] <tedg> pitti: Yes, it is a change, to make it so that people can easily have both installed.
[14:02] <pitti> ... in indicator-application
[14:03] <Trevinho> seb128: see https://code.launchpad.net/~ted/unity/systemd-unit./+merge/300624 the ups service file
[14:03] <seb128> right
[14:03] <seb128> don't do that!
[14:11]  * Laney plays the bisecting tune
[14:21]  * tedg apparently fell off freenode, if someone replied, he missed it.
[14:22] <seb128> wb tedg, no much reply no but you didn't ask a question either did you?
[14:22] <seb128> "<tedg> pitti: Yes, it is a change, to make it so that people can easily have both installed." was what we got before you left the channel
[14:23] <tedg> seb128: I just answered, no question: "seb128: The service job lists the ones it wants by default. Others can be added with symlinks in the wants directory."
[14:23] <seb128> sorry if I created confusion, I didn't follow closely the start of the discussion
[14:24] <seb128> but I think it makes more sense having the indicator packages to opt in for the desktop they wants by using symlinks in .wants dirs
[14:24] <seb128> rather than having the panel service listing a defined static set
[14:24] <seb128> I understand others can still do the .wants
[14:25] <seb128> but you end up having different mechanisms in use depending of what indicator you look at then no?
[14:26] <tedg> No, it's the same mechanism, we're just saying that unity7 should have a list of the ones it wants by default and unity8 should have a list of the ones it wants by default.
[14:26] <tedg> So if u7 decides to drop nm-applet to go for indicator-network, that's a change in the u7 package, not an indicator-network and network-manager change.
[14:26] <seb128> well, then e.g indicator-session doesn't have a .wants symlink for unity7 in its binary deb?
[14:27] <seb128> so you end up having some debs including a symlinks and others not
[14:27] <seb128> I mean there is no symlink for indicator-session
[14:27] <tedg> Eh, kinda. I doubt there are many indicators that would have symlinks in them. Reality is that there aren't many non-core indicators other than app indicators.
[14:28] <tedg> I can't think of any, but there probably is...
[14:28] <seb128> I can't either, but nothing blocks a company to have a custom one on their installs
[14:28] <seb128> I agree it's not a big deal in practice
[14:29] <tedg> Correct, indicator-session doesn't encode who wants it. U7 and U8 encode that they want it.
[14:29] <tedg> For instance, if we think of Xubuntu, they'd select indicators and make their own list.
[14:29] <seb128> it feels like giving the decision power to the desktop maker and not the user
[14:30] <seb128> also kylin and ubuntu might want a different set
[14:30] <seb128> like they might not like our keyboard indicator
[14:31] <seb128> then how do they opt out?
[14:31] <seb128> I guess they need an override or a symlink in another dir?
[14:32] <shookees> Hey all
[14:32] <tedg> seb128: Not sure how kylin customizes, new package?
[14:33] <seb128> tedg, that was a random invented example, but yeah they have a -settings with schemas overrides and such
[14:34] <tedg> seb128: Sure, more asking because I really didn't know :-)
[14:35] <tedg> seb128: I guess I don't see the set of indicators as a user config thing. They can tweek it with overrides and the such, and they can remove packages, but I dont' think it's a normal thing for a user to adjust.
[14:36] <seb128> well it is but in practice we have "show this indicator" settings in u-c-c
[14:36] <jbicha> Laney: could you rebuild the GTK 3.20 stuff in main from http://qa.ubuntuwire.org/ftbfs/ I think I got all from universe
[14:36] <seb128> which is bit stupid because it means we have the service active for users who don't use it
[14:37] <Laney> jbicha: later on, or sooner if you give me a list :)
[14:38] <seb128> jbicha, if you are the one who retried gnome-builder no point doing that it fails with the new vala, but you might want to merge 3.20 it should be trivial
[14:38] <tedg> seb128: Yes. Miss the gsettings bridge for Upstart, even though we never got it into production, was a cool idea.
[14:39] <jbicha> clutter-gtk gnome-sudoku gtkmm3.0 gtksourceview3 notification-daemon
[14:39] <seb128> indeed
[14:40] <Laney> done
[14:40] <Laney> merci
[14:41] <jbicha> thanks, yeah I syncd gnome-builder but it still needs a few more 3.20 stuff to build so...later this week
[14:41] <seb128> tedg, anyway, as said I think either way is fine, so I should probably just shut up and let Trevinho tell you what he wants since he actually maintain unity and is the one giving you an ack or not ;-)
[14:42] <seb128> jbicha, oh, the libpeas loader split went to debian? great
[14:44] <jbicha> seb128: yes, Laney did most of the work, I just nudged it a bit when it got stuck
[14:44] <seb128> nice
[14:45] <tedg> pitti: So I tried moving the "wants" to a links file and it's complaining that the directory doesn't exist. Is there a way around that?
[14:45] <Trevinho> Laney: I've a new SRU ready for xenial, when you have time please consider publishing that (https://requests.ci-train.ubuntu.com/#/ticket/1737)
[14:45] <pitti> tedg: dh_links automatically creates links; how does your .links file look like and where/how does it fail?
[14:46] <pitti> tedg: note that you need to specify absolute target paths in the .links file; dh_links will make them relative (that's a bit eww..)
[14:46] <Laney> Trevinho: k, in a bit, poking inside the greeter atm
[14:46] <Trevinho> Laney: no worries, take your time.... And there will be a yakkety landing coming in a bit too.
[14:47] <tedg> pitti:  https://www.irccloud.com/pastebin/NgcB4kph/
[14:47] <Trevinho> actually is there...
[14:48] <pitti> tedg: can you show the line from the .links file here?
[14:48] <tedg> pitti: http://paste.ubuntu.com/21764853/
[14:50] <pitti> tedg: I think the second component can't be a dir, it needs to be a file; i. e. append foo.service to each line
[14:50] <pitti> see man dh_link
[14:50] <jbicha> Laney: could you retry the s390x builds from the list too?
[14:50] <Laney> I retried all arches for the things you gave me
[14:50] <pitti> tedg: or create them in the upstream build (which is conceptually better, but not that relevant if the packages only exist in ubuntu)
[14:51] <Laney> unless ubuntu-build doesn't know about them
[14:51] <Laney> there
[14:53] <tedg> pitti: Ah cool, I was trying to take into Trevinho's concern that he'd like it to be in the packaging.
[14:54] <jbicha> Laney: it looks like you need an unreleased ubuntu-dev-tools for s390x https://bazaar.launchpad.net/~ubuntu-dev/ubuntu-dev-tools/trunk/changes
[14:54] <Laney> yeah I got it
[15:28] <seb128> jbicha, you might want to look at bug #1607663?
[15:28] <ubot5> bug 1607663 in vino (Ubuntu) "vino-server does not autostart on Ubuntu-GNOME" [Undecided,New] https://launchpad.net/bugs/1607663
[15:31] <jbicha> thanks
[17:43] <attente> Laney: is pidgin supposed to show up in the results when doing 'appstreamcli search pidgin'
[18:13] <ricotz> TheMuso, hi, could you plan to cherry-pick this with the next pulseaudio sru for xenial? https://cgit.freedesktop.org/pulseaudio/pulseaudio/commit/?id=81d3eb84672726e61d46dcb429a73422f43b691c
=== jhenke_ is now known as jhenke
[19:24] <seb128> k, enough for today
[19:25] <seb128> have a nice evening desktopers
[19:25] <seb128> see you tomorrow
[19:25] <willcooke> cya seb128
[19:25] <seb128> night willcooke!
[20:03] <willcooke> morning robert_ancell
[20:03] <robert_ancell> willcooke, yo
[20:23] <willcooke> night all
[22:18] <TheMuso> ricotz: Sure seems harmless enough. Luckily I might be SRUing pulse once I track down a particular bug so will look at doing that. Do you have a bug to reference for that?