| duflu | Does anyone know if there is a particular date for 16.10 EOL in "July 2017"? | 04:53 |
|---|---|---|
| jbicha | duflu: https://motd.ubuntu.com/ :) | 04:54 |
| jbicha | I'd expect an ubuntu-announce email for that | 04:55 |
| duflu | jbicha, thanks. If verified then this needs updating: https://wiki.ubuntu.com/Releases and maybe https://wiki.ubuntu.com/YakketyYak/ReleaseNotes | 04:56 |
| jbicha | https://bazaar.launchpad.net/~ubuntu-motd/ubuntu-motd/trunk/revision/15 | 04:57 |
| duflu | Yeah, thanks. Although I just remembered there's no point directing users to the Releases page till after moves into EOL | 04:59 |
| duflu | RAOF: Do you have any experience or knowledge as to why rendering in Gnome Shell is higher CPU (for both shell and app) than Weston? | 05:04 |
| duflu | Or is that just uninteresting -- Weston is much simpler and obviously more optimized | 05:05 |
| RAOF | duflu: dunno. | 05:15 |
| RAOF | But Shell has more levels of abstraction. | 05:16 |
| duflu | Doesn't matter... Some profiling should provide solid answers as to where the extra CPU is going | 05:16 |
| RAOF | And an X server connection (still) | 05:16 |
| RAOF | Unless that particular wart is done | 05:17 |
| RAOF | Yeah, profile 😀 | 05:17 |
| duflu | I realised I was asking the wrong question. It's not interesting that Weston is more efficient than Gnome Shell. It's interesting that hardware acceleration claims to be working but actually results in higher CPU than software decoding | 05:17 |
| RAOF | SW decoding for what? | 05:18 |
| RAOF | vainfo may be of interest | 05:19 |
| duflu | RAOF: vainfo reports everything is OK. Seems I have just shell/rendering problems | 05:20 |
| didrocks | good morning | 06:12 |
| jibel | morning | 06:50 |
| seb128 | good morning desktopers | 07:16 |
| duflu | Morning seb128 | 07:18 |
| seb128 | hey duflu! how are you? had a good w.e and start of week? | 07:18 |
| duflu | seb128, yeah going OK. How was your weekend? | 07:19 |
| jamesh | hi seb128 | 07:19 |
| seb128 | quite uneventfull, we had some rainy days and I mostly relaxed and did nothing | 07:19 |
| seb128 | hey jamesh, how are things down there? ;-) | 07:20 |
| duflu | Sounds nice actually | 07:20 |
| jamesh | wet | 07:20 |
| jamesh | (not right this minute though) | 07:20 |
| jamesh | So, I'm wondering if there is any reason snapd couldn't support polkit authorisation for operations where you don't need to authenticate to the store | 07:21 |
| seb128 | jamesh, do you have specific actions in mind? like installing a local snap or removing one? | 07:25 |
| jamesh | seb128: basically anything that you'd need to use "sudo snapd command" for currently if you don't have a store macaroon | 07:26 |
| jamesh | so installing and removing snaps | 07:26 |
| jamesh | if we had that, gnome-software could delay asking for store auth until the user tried to pay for a snap | 07:27 |
| seb128 | jamesh, that's worth raising for discussion on the forum if you think that would be a better solution | 07:27 |
| seb128 | jamesh, I just saw that Laney added you to the n-m portal trello card, you are looking to that one? | 07:30 |
| jamesh | seb128: that's what I'm looking at next, yeah. | 07:31 |
| seb128 | cool | 07:31 |
| oSoMoN | good morning desktoppers | 07:38 |
| seb128 | jamesh, oh, I see that you commented on the post from robert about the polkit thing, good | 07:38 |
| seb128 | lut oSoMoN, how are you? had a good w.e? | 07:39 |
| oSoMoN | salut seb128, yeah I had a good week-end, with the right balance of getting some rest and activities. How about you? | 07:42 |
| seb128 | oSoMoN, w.e was good, didn't do much but that was good :-) | 07:43 |
| oSoMoN | not doing much is good indeed | 07:45 |
| seb128 | oSoMoN, so the libreoffice build issue is a buildd/kernel problem? | 07:56 |
| oSoMoN | seb128, I think so, although I launched another build over the week-end, which ran with an updated kernel that presumably has a fix, but it failed in the same way | 07:57 |
| oSoMoN | maybe the kernel fix is incomplete | 07:58 |
| seb128 | :-( | 07:58 |
| seb128 | we should perhaps talk to the kernel team about that | 07:58 |
| oSoMoN | yes, I’ll talk to them this morning | 07:58 |
| seb128 | thanks | 07:59 |
| willcooke | morning all | 08:05 |
| duflu | Morning willcooke | 08:05 |
| willcooke | ar'noon | 08:06 |
| oSoMoN | good morning willcooke | 08:06 |
| seb128 | duflu, that pulseaudio xenial SRU segfault, do we have any concrete data telling us that it's not important, or is that comment based on the number of reports? I mean it's an issue if users get in a state their experience with the system is impacted, like audio cut or they need to restart their session to get it working again, less so if it's just the service segfaulting and restarting | 08:07 |
| seb128 | hey willcooke, did you have a good w.e and recovered from that London week? | 08:08 |
| duflu | seb128, yes the links I provided are concrete. And the bug exists in zesty and artful already without people actively complaining | 08:08 |
| willcooke | morning seb128 - still tired :) | 08:09 |
| duflu | seb128, I mean nobody noticed it in usage. Only errors.ubuntu.com noticed it | 08:09 |
| duflu | There are other crashes orders of magnitude more common and people generally don't seem to report those either. | 08:09 |
| seb128 | duflu, it's difficult to say, all users don't report bugs and maybe some "my sound stop working when I do <...>" reports are due to it but we just not have a way to make that link | 08:10 |
| seb128 | duflu, I mean you are probably right, but it's always tricky to rule out a real issue for sure | 08:11 |
| seb128 | duflu, it might be good to at least report the segfault upstream in any case | 08:11 |
| duflu | seb128, across all distros the crash you refer to has happened 383 times, out of 30000+ pulseaudio crashes | 08:12 |
| duflu | It's tiny | 08:12 |
| duflu | (in the past year) | 08:12 |
| duflu | It's roughly 1% of all pulseaudio crashes | 08:13 |
| seb128 | yeah, at the same time it has not been in xenial which is the distro serie which has the most users, it already picked 12 from the SRU | 08:13 |
| seb128 | but yeah, it's probably still going to be on the low side of the scale | 08:13 |
| duflu | seb128, OK then, consider just xenial: It's 12 out of almost 40000 xenial pulseaudio crashes | 08:14 |
| seb128 | well, you count reports on a longer time to a version shipped to more users | 08:15 |
| seb128 | but yeah, I'm not arguing with you that it's not low | 08:15 |
| seb128 | anyway, let's see what the SRU team says | 08:16 |
| seb128 | they tend to not like regression, even small/non important ones | 08:16 |
| duflu | seb128, If you just narrow it down to 1:8.0-0ubuntu3.3 then there are still other crashes much more common than the one you're talking about | 08:16 |
| seb128 | right | 08:18 |
| seb128 | duflu, what about reporting upstream, do you think that might be useful? | 08:18 |
| seb128 | or do you think they are not going to be able to do much about it without an actual description of the problem or of how to trigger it? | 08:18 |
| duflu | seb128, perhaps. But we haven't paid enough attention to the dozens/hundreds of more common crashes. I would focus on those first | 08:18 |
| duflu | dozens in crashes. hundreds in user-reported | 08:19 |
| seb128 | right, except that it would probably help the SRU team to feel more confortable about the update | 08:19 |
| seb128 | they tend to not be comfortable with any regression, even small ones | 08:19 |
| duflu | seb128, in fact I've probably received more than 12 comments from users pointing out the new version fixed their systems :) | 08:20 |
| seb128 | which is understandable, you never know if it's not going to impact some production system even if it's a corner case and only impacting specific setups | 08:20 |
| duflu | seb128, yeah as I mentioned I totally understand that. But it's a net positive benefit to keep. I understand the usual policy would actually prefer zero regressions, but that's a net negative benefit | 08:22 |
| seb128 | I think opening a bug upstream and then arguing that we did what we think it reasonable should be something the SRU team can listen to | 08:23 |
| seb128 | I can forward the bug if you prefer | 08:23 |
| duflu | seb128, yeah that's easy. I just feel like I'm wasting time doing so though, since it's not actually as big a deal as other bugs we ignore | 08:23 |
| duflu | I'll look at it now... | 08:24 |
| seb128 | thanks | 08:24 |
| seb128 | I think it's going to waste less time than arguing with the SRU team would cost us if we don't | 08:25 |
| duflu | Yeah I know. There just isn't any team arguing that "you're ignoring 99% of the real problem" | 08:25 |
| duflu | seb128, seems I found upstream reports/fixes, will link them | 08:31 |
| seb128 | oh, great | 08:31 |
| seb128 | easy fix that might be included in the SRU if the SRU team asks us to add that regression fix? | 08:31 |
| duflu | seb128, well the fix doesn't exist in artful or zesty yet | 08:33 |
| oSoMoN | chrisccoulson, have you had a chance to start validating the chromium-browser update in https://launchpad.net/~canonical-chromium-builds/+archive/ubuntu/stage/+packages ? | 08:33 |
| duflu | So technically they should come first | 08:33 |
| seb128 | right | 08:35 |
| duflu | seb128: Ha. Yes, that crash can only occur on a bluetooth audio system that's working. Hence didn't occur before :) | 08:37 |
| seb128 | that sort of verify the fix :p | 08:37 |
| seb128 | "see, bluetooth work good enough now that we hit bugs in the blutooth code" | 08:38 |
| duflu | Yes, weirdly, that crash verifies the improved behaviour | 08:38 |
| davidcalle | Laney: hi (and hi everyone else too :)), I'm posting your "welcome" blog post on insights, would you have an alternative title to make it more descriptive? | 09:21 |
| oSoMoN | davidcalle, La_ney is off this week | 09:22 |
| oSoMoN | hi, btw :) | 09:22 |
| davidcalle | Oh! | 09:22 |
| seb128 | what o_SoMoN said | 09:22 |
| seb128 | hey davidcalle :-) | 09:22 |
| davidcalle | willcooke (willcook_e), any suggestion? I would go for "Switching from Unity to Gnome-Shell: first challenges" | 09:24 |
| willcooke | davidcalle, wfm :) | 09:24 |
| davidcalle | Yay :) | 09:24 |
| davidcalle | Hey seb128 and oSoMoN ;) | 09:25 |
| willcooke | davidcalle, there is another laney blog post already, which I will add to co-schedule today | 09:25 |
| davidcalle | +1, thanks! | 09:26 |
| andyrock | good morning! | 09:30 |
| seb128 | hey andyrock, how are you? had a good w.e? | 09:30 |
| andyrock | hey seb128 I went to the beach | 09:31 |
| andyrock | what about you? | 09:31 |
| seb128 | ah, nice | 09:31 |
| seb128 | it was grey cold and rainy here | 09:31 |
| oSoMoN | hey andyrock | 09:31 |
| seb128 | I didn't do much, went to have a look at some shops to find a table to put outside in summer and otherwise just stayed home and didn't do much | 09:33 |
| Trevinho | hey guys | 09:33 |
| seb128 | good morning Maaarrccooo | 09:33 |
| seb128 | Trevinho, how are you? had a good w.e? | 09:34 |
| Trevinho | seb128: I'm good... Yeah, i had a good one... You saw the pictures, isn't it? :-) | 09:34 |
| Trevinho | seb128: how about yourself? | 09:34 |
| seb128 | Trevinho, yeah, I saw that, seems you always have crazy weekends :-) | 09:35 |
| seb128 | Trevinho, w.e was good as I was writing on the line just before your hello :-) | 09:35 |
| Trevinho | seb128: ahah, recently it happens... But just unplanned things that I join at the last moment | 09:36 |
| seb128 | well, seems you have fun in any case | 09:36 |
| seb128 | way to go! | 09:36 |
| seb128 | hum | 09:37 |
| seb128 | does anyone know why the daily iso are not getting promoted to current? | 09:38 |
| seb128 | I always forgot what those tests are and where are the logs | 09:38 |
| seb128 | jibel, hey, you probably know the details ^? | 09:39 |
| jibel | seb128, automated promotion is blocked due to bug 1700557. I'll promote it manually | 09:40 |
| ubot5 | bug 1700557 in linux (Ubuntu) "VM doesn't boot after installation until an input event is received" [High,Incomplete] https://launchpad.net/bugs/1700557 | 09:40 |
| seb128 | jibel, was that the issue you were looking at when you had that packages list diff where none seemed like a candidate for the issue? did you manage to find one that would create the issue on updating? | 09:41 |
| seb128 | jibel, thanks | 09:41 |
| jibel | seb128, yeah but the package list is unrelated | 09:41 |
| jibel | seb128, it's a really weird issue | 09:42 |
| jibel | seb128, it seems to depend on the order of installation of the packages | 09:42 |
| seb128 | weird | 09:42 |
| jibel | seb128, and it happens very early in the boot process. the boot order is involved, it's like the system is blocked on the lack of entropy | 09:43 |
| seb128 | going to be a fun one to debug :-/ | 09:43 |
| duflu | OK, I think Tuesday I'll get back into Pulse. But only because GStreamer make me want to violently break things | 09:44 |
| duflu | Fix one bug and you just hit a bunch more, in a mountain of plugins | 09:44 |
| seb128 | :-( | 09:45 |
| duflu | seb128: It's fine. Another day. I just had to give it a few days analysis to assess the situation | 09:49 |
| seb128 | in any case it sounds like you should think about calling it a day and get dinner at this time! | 09:50 |
| duflu | seb128, indeed. I was waiting for a clean context switch opportunity | 09:55 |
| duflu | Although separating all the decode bugs from the display bugs, I might be able to actually close some | 09:57 |
| oSoMoN | seb128, while I work to resolve the i386 LO build failures, an updated amd64 snap (5.3.4, strict confinement) is available for testing at https://code.launchpad.net/~osomon/+snap/libreoffice/+build/52056 | 09:57 |
| oSoMoN | and anyone else interested in giving it a try, feedback welcome ^ | 09:57 |
| duflu | Trevinho, the new battery icons are now design approved. Do we have to wait on some other review for visual things? | 09:58 |
| Trevinho | duflu: I guess, no... I'll check that shortly | 09:59 |
| duflu | Kay, night | 09:59 |
| jibel | seb128, build 20170703 have been promoted to current | 10:05 |
| seb128 | jibel, thanks! | 10:06 |
| seb128 | oSoMoN, great, I had that on my list for today | 10:07 |
| ricotz | oSoMoN, hi, jfyi, more logs regarding the i386 ftbfs -- https://launchpad.net/~ricotz/+archive/ubuntu/ppa/+packages | 10:49 |
| oSoMoN | ricotz, yeah, I’m getting the same failures in another PPA and when building an i386 snap too | 11:41 |
| jbicha | didrocks: hi, could you promote the xwayland binary to main to unblock gnome-session from -proposed? | 11:53 |
| didrocks | jbicha: sure, let me look at the bug report about xwayland dep first | 12:00 |
| didrocks | why 3.24.1-0ubuntu7 and 3.24.1-0ubuntu8 have the same changelog? | 12:01 |
| didrocks | amd64 failed for -0ubuntu7. Normally, we do upload -0ubuntu8 with the additional fix + -v<olderversion> to include both | 12:02 |
| didrocks | (xwayland promoted) | 12:03 |
| jbicha | didrocks: they aren't the same changelog | 12:03 |
| jbicha | thanks | 12:03 |
| seb128 | jbicha, why was the arch any/all change needed? | 12:19 |
| jbicha | dpkg-gencontrol: error: the Depends field contains an arch-specific dependency but the package is architecture all | 12:21 |
| jbicha | in Debian, it's gnome-session-bin that depends on xwayland but that's not really right (at least for Ubuntu flavors) | 12:22 |
| seb128 | does it depends on it for technical reasons? | 12:24 |
| seb128 | like what happens if xwayland is not installed? | 12:24 |
| didrocks | no wayland session available | 12:24 |
| didrocks | in gdm it doesn't list it | 12:25 |
| didrocks | in lightdm, it does list it but doesn't start | 12:25 |
| seb128 | I guess there is no issue worth discussing there and I will just stop asking questions and let people deal with work :p | 12:27 |
| seb128 | thanks didrocks, jbicha | 12:27 |
| jbicha | np | 12:27 |
| seb128 | the first question was because I was unsure if an arch all package could depends on "binary" which is arch specific if the arch is not coded | 12:27 |
| seb128 | like libraries or built code is not fine | 12:27 |
| seb128 | but if it's just depending on "xwayland" for example I was curious if that was allowed or not | 12:28 |
| seb128 | that might have to do with the [linux] arch use though | 12:28 |
| jibel | didrocks, there are 2 issues with the migration script after an upgrade from a default installation: 1. amazon is gone. 2. system settings is kept to unity-control-center which is correct but should probably be migrated to gnome-control-center | 12:28 |
| didrocks | jibel: amazon is gone? you had amazon in the old launcher but it didn't transition to the new desktop version from Lane_y? | 12:29 |
| jibel | didrocks,yes | 12:29 |
| didrocks | jibel: on the second one, good idea, we should transition from u-c-c to g-c-c | 12:30 |
| jbicha | seb128: oh, that's probably it then; I just copied the [linux-any] from Debian, but I guess Ubuntu could drop it | 12:30 |
| seb128 | jbicha, I don't think it matters much either way, I was mostly curious | 12:30 |
| seb128 | it's not a case we hit often | 12:30 |
| didrocks | jibel: let me have a look, I did the amazon thingy on purpose and tested it, probably another issue after it then | 12:30 |
| didrocks | thanks for testing :) | 12:30 |
| jibel | didrocks, right, that's why i checked amazon :) | 12:30 |
| seb128 | didrocks, does amazon requires to pull in a new package? | 12:30 |
| jbicha | seb128: thanks for thinking about it more, I didn't understand why I hadn't seen that kind of error more often | 12:31 |
| didrocks | seb128: the package is pulled normally, no? | 12:31 |
| jibel | seb128, what's the name of the package? | 12:31 |
| seb128 | didrocks, I didn't try upgrades or looked at what pulled it in | 12:31 |
| didrocks | jibel: did you try an upgrade? | 12:31 |
| didrocks | or did you try to have the unity session | 12:31 |
| didrocks | make some tweaks | 12:31 |
| jibel | didrocks, it's an upgrade from 17.04 | 12:31 |
| didrocks | and transition to GNOME? | 12:31 |
| didrocks | ahah | 12:31 |
| didrocks | maybe seb128 is right, and Lane_y didn't make it pulled on upgrade | 12:31 |
| didrocks | let me look | 12:32 |
| jbicha | ubuntu-desktop Recommends: ubuntu-web-launchers | 12:33 |
| seb128 | jibel, ^ I guess jbicha is saying it's this binary | 12:34 |
| jibel | it's installed | 12:34 |
| jbicha | jibel: when you say it's gone, you just mean the side-launcher shortcut is gone (GNOME calls their side dock the "dash"), not the app itself right? | 12:34 |
| jibel | after an upgrade | 12:34 |
| jibel | jbicha, it is not in the favorite apps after an upgrade | 12:34 |
| jbicha | the .desktop may have been renamed? | 12:34 |
| jibel | but it was before | 12:34 |
| didrocks | jibel: seb128: I think I know what happens | 12:35 |
| didrocks | you upgrade, the old package is removed | 12:35 |
| jibel | it's com.canonical.launcher.amazon.desktop | 12:35 |
| didrocks | so no more .desktop corresponding to old name | 12:35 |
| seb128 | unity removes it | 12:35 |
| jibel | same than in didrocks' script | 12:35 |
| didrocks | unity removes it | 12:35 |
| seb128 | :-) | 12:35 |
| didrocks | and migration… without it | 12:35 |
| jibel | ah | 12:35 |
| didrocks | :p | 12:35 |
| jibel | didrocks, do you want bug reports so you can come back to it later? | 12:36 |
| didrocks | argh @laney to rename the .desktop | 12:36 |
| didrocks | jibel: please file it, yeah | 12:36 |
| jibel | okay | 12:36 |
| didrocks | at least so that we can double check | 12:36 |
| seb128 | do we rename for a technical reason? | 12:36 |
| didrocks | I guess it was in the "let's follow the GNOME scheme" | 12:36 |
| seb128 | and GNOME issues :p | 12:37 |
| didrocks | so, symlinks? | 12:37 |
| andyrock | mpt: hey! please can you let me know when the design is ready, otherwise i'm blocked | 12:37 |
| seb128 | guess so, same hack as for others | 12:37 |
| seb128 | compat | 12:37 |
| didrocks | yep | 12:37 |
| didrocks | then, once all G-S, we won't have that issue | 12:37 |
| didrocks | as it doesn't check for .desktop existence | 12:37 |
| seb128 | right | 12:37 |
| jbicha | I don't like those compat symlinks, can't we just use the old name for the .desktop? | 12:37 |
| seb128 | it means gnome-shell lists inexistant/not working entries? | 12:38 |
| didrocks | in the gsettings key, it does, right | 12:38 |
| seb128 | but filter them away from the rendering then? | 12:38 |
| didrocks | yep | 12:38 |
| didrocks | no "?" | 12:38 |
| seb128 | I see | 12:38 |
| jbicha | seb128: that's useful for Ubiquity in particular | 12:39 |
| didrocks | I guess let's add the compat symlink and we can remove post-LTS | 12:39 |
| seb128 | yeah | 12:39 |
| seb128 | jbicha, what's the issue with ubiquity? | 12:39 |
| jbicha | we want it to be a "favorite", at least in the live image | 12:40 |
| mpt | andyrock, understood, sorry for the blockage. I’m working on it today. | 12:40 |
| andyrock | np! I've other stuff to work on :D | 12:40 |
| didrocks | well, unity had that, and checking for the .desktop made sense | 12:40 |
| seb128 | jbicha, right, well it was like that under unity | 12:40 |
| jbicha | although maybe we just hack that in the live CD for gnome-shell actually | 12:40 |
| seb128 | I guess they are pro and co for each way | 12:40 |
| jbicha | it's useful if we don't install an email client by default ;) | 12:41 |
| seb128 | the current way means your config doesn't match what you see, and that the shell does extra work checking for items to discart at every start | 12:41 |
| didrocks | yeah, I guess none has the perfect solution | 12:41 |
| seb128 | but that's trivial work so no big deal | 12:41 |
| seb128 | jbicha, I fail to see why? | 12:41 |
| seb128 | what do email clients have to do with that?! | 12:42 |
| jbicha | When you install thunderbird, it will show as a favorite. When you uninstall it, Thunderbird won't show up as a favorite | 12:43 |
| jibel | didrocks, bug 1702089 and bug 1702091 | 12:47 |
| ubot5 | bug 1702089 in gnome-session (Ubuntu) "Amazon removed from the favorite apps after an upgrade from 17.04 to artful" [Undecided,New] https://launchpad.net/bugs/1702089 | 12:47 |
| ubot5 | bug 1702091 in gnome-session (Ubuntu) "upgrade to artful: Migrate unity-control-center to gnome-control-center " [Undecided,New] https://launchpad.net/bugs/1702091 | 12:47 |
| didrocks | jibel: thanks a lot! I'll reassign to the correct component | 12:48 |
| seb128 | jbicha, that sounds like the issue Didier reported, under Unity installing something from gnome-software would add the something icon to the launcher/favorites | 12:49 |
| seb128 | which is currently missing under GNOME | 12:49 |
| didrocks | yep, I don't see why some apps would be different | 12:49 |
| didrocks | cleaning up makes sense, the only issue is those use case: transitions to new names without having the possibility to delay it | 12:50 |
| seb128 | renaming .desktop is just wrong and creating issues | 12:51 |
| seb128 | it's not the only one | 12:51 |
| seb128 | we talked the other days about mimetype associations as well | 12:51 |
| didrocks | yeah, it's complex and not well-supported for sure | 12:51 |
| seb128 | oh well, it's done now | 12:51 |
| seb128 | thanks GNOME | 12:51 |
| didrocks | (TBH, that one isn't really GNOME's fault :p) | 12:52 |
| seb128 | right | 12:52 |
| * didrocks tests quickly the symlink | 12:52 | |
| seb128 | it was just L_aney being overzealous and copying them | 12:52 |
| jbicha | didrocks: I'm not sure using a symlink like that is a good idea | 13:32 |
| didrocks | jbicha: it's what we did for quite some desktop file AFAIK | 13:33 |
| didrocks | and what we discussed here ^ | 13:33 |
| didrocks | the only issue is duplication in the dash/app search, but I guess it's the pain of this transition | 13:35 |
| jbicha | didrocks: that's not how we did it for other apps | 13:35 |
| jbicha | and yes, duplication in the Activities Overview is a bug | 13:35 |
| didrocks | any suggestions thus? | 13:35 |
| jbicha | if you want to see how we did it for other apps, look at the gedit packaging for instance | 13:36 |
| jbicha | my #1 recommendation is that since we control the .desktop filename for this app is to just go back to the old name | 13:36 |
| didrocks | could have been more explicit before the upload btw | 13:37 |
| didrocks | rather than venting afterwards :p | 13:37 |
| didrocks | I don't see a OnlyShowIn in the .desktop | 13:37 |
| jbicha | GNOME changed .desktop names primarily to help with Flatpak integration or something but I don't think anyone wants to Flatpak this | 13:37 |
| jbicha | it adds a NoDisplay=true | 13:37 |
| didrocks | right | 13:37 |
| didrocks | doesn't help the unity case though | 13:38 |
| didrocks | OR | 13:38 |
| didrocks | we need another transition file to keep unity | 13:38 |
| jbicha | the compat thing we use for gedit and others is broken too but different | 13:38 |
| didrocks | which isn't ideal | 13:38 |
| didrocks | so you trade a symlink for 2 transitions | 13:38 |
| jbicha | I'm sorry you felt I was venting | 13:38 |
| jbicha | I'm not sure I could have given good feedback before I knew what you were doing ;) | 13:39 |
| didrocks | 14:37:09 didrocks | so, symlinks? | 13:39 |
| didrocks | a symlink is a symlink :p | 13:39 |
| didrocks | 14:37:18 seb128 | guess so, same hack as for others | 13:39 |
| jbicha | yes, sorry | 13:40 |
| didrocks | any reason the NoDisplay=true is on the new desktop name? | 13:40 |
| didrocks | I guess just to avoid having that migration script running | 13:40 |
| jbicha | that's one of the issues I have with the compat hack! ;) | 13:40 |
| didrocks | so, it means: migration script in unity session + additional desktop instead of symlink with NoDisplay=true | 13:41 |
| didrocks | could do | 13:41 |
| didrocks | just a little bit more involved | 13:41 |
| jbicha | ok, but why? | 13:41 |
| didrocks | well, if we are going to migrate everything at some point, let's not add another one to migrate | 13:42 |
| jbicha | why do we need to use a different name for the .desktop in artful than we used in zesty? | 13:42 |
| didrocks | that was a question for L_aney I gues | 13:42 |
| didrocks | guess* | 13:42 |
| didrocks | he did that name change | 13:42 |
| didrocks | without considering backward compat apparently | 13:42 |
| didrocks | seb128: thoughts? we could do that (double migration) or just go back to older name ^ | 13:43 |
| didrocks | I'm fine with either | 13:43 |
| seb128 | there is no hurry, maybe wait next week for Iain to be back so we can check if he had a reason for the rename or if he feels strongly about it? | 13:44 |
| didrocks | then, if one day we migrate and handle desktop name properly, we migrate everything in batch (including mimetypes) | 13:44 |
| didrocks | ok | 13:44 |
| jbicha | didrocks: there's more discussion about the compat issues at LP: #1662296 | 13:44 |
| ubot5 | Launchpad bug 1662296 in unity (Ubuntu) "Remap renamed .desktop files" [Medium,Triaged] https://launchpad.net/bugs/1662296 | 13:44 |
| didrocks | so, we keep the symlink for now? | 13:44 |
| jbicha | I'm fine with waiting until he gets back | 13:44 |
| didrocks | jbicha: yeah, I have that bug opened in a tab | 13:44 |
| jbicha | it was too complicated for me to do everything La_ney suggested, all I did was point out some issues I had with the current implementation :( | 13:45 |
| didrocks | I guess at some point we'll just have to be able to do that in a performant way. Implementations need to be decided | 13:46 |
| didrocks | not a fan with a .d/ checking for more | 13:46 |
| seb128 | jbicha, what do you think about building a gnome-control-center-faces from g-c-c? | 14:20 |
| rbasak | My system seems to have switched to GNOME smoothly. Good job! | 14:33 |
| rbasak | One minor issue: I had to remove xserver-xorg-input-synaptics manually. With it installed (as it was by default from Unity previously) the mouse was a little broken. | 14:33 |
| rbasak | Known issue? | 14:34 |
| jbicha | rbasak: yes, LP: #1686081 there was talk about fixing Unity to work with libinput and backporting that to xenial | 14:35 |
| ubot5 | Launchpad bug 1686081 in xorg (Ubuntu) "If -synaptics is installed, GNOME Mouse & Touchpad Settings doesn't work" [High,Confirmed] https://launchpad.net/bugs/1686081 | 14:35 |
| seb128 | jbicha, fixing unity is another issue of having synaptic not removed on upgrade | 14:38 |
| rbasak | jbicha: thanks, though I'd s/18.04/17.10/ in most of that, since it'll happen to all users upgrading from 17.04 to Artful as I understand it? | 14:38 |
| seb128 | jbicha, we could remove synaptic and make the unity config dialog useless | 14:38 |
| rbasak | (sort of a higher level bug I mean - not that it happens, but that it is triggered on the specific default upgrade case) | 14:38 |
| seb128 | rbasak, it's something we know about and will resolve before 17.10 one way or another | 14:39 |
| rbasak | Thanks :) | 14:39 |
| seb128 | yw | 14:39 |
| seb128 | good to know that things mostly worked for you though :-) | 14:39 |
| seb128 | did you have a chance to use a bit GNOME yet? how does it feel? | 14:40 |
| jbicha | GNOME (or at least the former gnome-control-center maintainer) wishes that distros would just remove synaptics completely | 14:40 |
| rbasak | It seems straightforward enough | 14:40 |
| jbicha | not sure how much of a problem that is for other desktops | 14:40 |
| rbasak | (never used recent GNOME) | 14:40 |
| seb128 | things you like or find weird? (always good to get some feedback from unity users upgrading) | 14:40 |
| tjaalton | jbicha: I think most have migrated to libinput by now | 14:40 |
| tjaalton | if not all | 14:41 |
| oSoMoN | seb128, if you don't object to it I’m going to push that LO 5.3.4 snap to the store (beta channel) and write about it to get some community testing. It'll be amd64 and armhf only for now, until the i386 build issue is resolved | 14:41 |
| rbasak | The Mouse & Touchpad settings appears with a shortened height until it gets refreshed somehow. Papercut level. | 14:41 |
| seb128 | jbicha, typical GNOME :-) | 14:41 |
| tjaalton | seb128: well, libinput upstream did port the other desktop capplets tho | 14:41 |
| tjaalton | just not unity, for some reason ;) | 14:42 |
| seb128 | oSoMoN, sure, go ahead, and thanks for the reminder I'm going to test your new snap now (well, time to install it etc) | 14:42 |
| rbasak | Going from "Test Your Settings" back to the actual settings themselves is confusing. I press the back button expecting it to take me there and instead I get the settings menu. | 14:42 |
| jbicha | rbasak: there's an overhaul of g-c-c scheduled for 3.26 so that height bug may be obsolete then | 14:42 |
| oSoMoN | seb128, cheers | 14:42 |
| rbasak | I'm being told about an XPS 13 9360 update, but nothing seems to happen when I click Update. | 14:43 |
| tjaalton | rbasak: if it's a bios update then something might happen when you reboot | 14:44 |
| rbasak | tjaalton: I've rebooted since; nothing happened that I saw. The update still shows as available (AIUI). | 14:44 |
| tjaalton | ah | 14:44 |
| rbasak | But even if it did, the UI should tell me I need to reboot :) | 14:45 |
| tjaalton | right | 14:46 |
| rbasak | I was a bit surprised to find that the power button suspends; I was expecting a menu. I can get used to that though :) | 14:47 |
| seb128 | you are not the first to say that | 14:48 |
| rbasak | Just did some Googling to figure out that I have dynamic workspaces. I can get used to that now that I understand that they're there :) | 14:50 |
| rbasak | That's everything from the last half hour or so :) | 14:50 |
| * rbasak resumes regular work | 14:51 | |
| seb128 | thanks for the feedback | 14:52 |
| jbicha | rbasak: you can make them static in Tweak Tool ( sudo apt install gnome-tweak-tool ) | 14:52 |
| seb128 | I wonder if the power button behaviour is worth arguing upstream for | 14:52 |
| rbasak | jbicha: I found that in my Googling, but I prefer to be minimal in what I tweak, so I don't want to install no tweak tool :) | 14:53 |
| jbicha | hmm, ok | 14:55 |
| jbicha | most GNOME distros include it by default (Fedora doesn't but RHEL does!) | 14:56 |
| seb128 | lol | 14:57 |
| seb128 | seems reverse :-) | 14:57 |
| jbicha | Fedora tries to follow the GNOME definitions of what should be in GNOME | 14:58 |
| jbicha | RHEL cares more about pleasing their customers ;) | 14:58 |
| seb128 | I though they would care also about looking polished | 14:59 |
| jbicha | ugh, I don't think GNOME Classic's theme is very polished personally | 14:59 |
| seb128 | "here is a toolbox you might need to unscrew you because our settings don't let you configure what you need" is not the best message to send to your users | 15:00 |
| jbicha | because GNOME developers don't use GNOME Classic so Classic is basically in maintenance mode for years | 15:00 |
| seb128 | you have a point there | 15:00 |
| andyrock | seb128: do we have a list of extension that we are going to include in 17.10? | 15:28 |
| seb128 | andyrock, we didn't decide on that yet, the ones we asked about in the survey are listed on https://insights.ubuntu.com/2017/06/12/ubuntu-desktop-gnome-extensions-poll-results/ | 15:35 |
| Beret | rbasak, you're going to have install the tweak-tool at least in the short term | 15:46 |
| Beret | until the default extensions are installed and our settings are tweaked the way we want them | 15:46 |
| Beret | at least if you liked a majority of the defaults in unity | 15:46 |
| gQuigs | is the tweak tool the only way to manage the extensions? | 15:49 |
| Beret | I don't use that to manage extensions | 15:49 |
| Beret | I just use a web browser for that | 15:49 |
| Beret | the tweak tool let's you mess with some settings (like date format) that are not found by default in configuration | 15:50 |
| rbasak | Beret: I'm going to try and ride the defaults as they change. Let's see how I manage :) | 15:50 |
| Trevinho | anyone know what changed in artful so that now I need to explicitly build-dep on dbus-x11 in u7? | 16:03 |
| Trevinho | it wasn't needed last friday or yesterday... I think | 16:03 |
| willcooke | I was talking to flexiondotorg about shipping tweak tool by default last week actually. My initial thought was "no way", but there is some really useful stuff in there. IMO it's analogous to ccsm (perhaps less dangerous) so I'd still say don't ship by default, and we can change the settings we need without it. | 16:08 |
| jbicha | some of that useful stuff is stuff that we had in unity-control-center | 16:29 |
| jbicha | Trevinho: see https://lists.debian.org/debian-devel/2016/08/msg00554.html | 16:34 |
| jbicha | Trevinho: unity recommends unity-session which depends on gnome-session-bin which depends on dbus-x11 though | 16:36 |
| Trevinho | jbicha: yeah, but we use it for running the tests... | 16:37 |
| jbicha | https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=835887 | 16:37 |
| ubot5 | Debian bug 835887 in src:gnome-session "gnome-session: don't run dbus-launch if XDG_RUNTIME_DIR/bus is available" [Normal,Open] | 16:37 |
| Trevinho | jbicha: I've adeded it to build deps anyway | 16:37 |
| Trevinho | not directly though | 16:37 |
| Trevinho | I guess gio does it | 16:38 |
| jbicha | so I don't know if GNOME or Unity actually runs without dbus-x11 installed since it's kind of hard to run things without gnome-sesion-bin | 16:38 |
| Trevinho | I've not checked | 16:38 |
| Trevinho | gnome-shell runs withouth for sure, at least... without having it running | 16:38 |
| jbicha | the test probably needs to be fixed to not require dbus-x11? | 16:39 |
| jbicha | the only other thing in main that depends on dbus-x11 is ubiquity but I was hesitant to push there because it's hard for me to tell if ubiquity works fine without it | 16:40 |
| oSoMoN | seb128, willcooke: https://forum.snapcraft.io/t/call-for-testing-libreoffice-5-3-4-snap/1205 | 16:57 |
| oSoMoN | (will now re-post to g+) | 16:58 |
| willcooke | oSoMoN, great! | 16:58 |
| oSoMoN | https://plus.google.com/100486649727917304321/posts/g2zWPFx75DW | 17:01 |
| * willcooke shares | 17:03 | |
| willcooke | sharing is caring, etc. | 17:03 |
| willcooke | ;) | 17:03 |
| oSoMoN | :) | 17:07 |
| Trevinho | seb128: once this has been approved (if you care) by britney please publish it https://bileto.ubuntu.com/#/ticket/2843 :-) | 17:26 |
| willcooke | night all | 18:10 |
| oSoMoN | nighty night | 18:14 |
| === JanC_ is now known as JanC | ||
| seb128 | Trevinho, k, done now | 20:22 |
| Trevinho | seb128: thanks | 21:27 |
| === carlo_ is now known as clobrano__ | ||
| === Guest4347 is now known as RAOF | ||
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!