[00:04] Trevinho: were you going to reorder the gnome-shell patch series or should I do that? [00:05] there is a gnome-3-28 branch in upstream gnome now (HEAD points to 3.28.1) [00:05] jbicha: you mean for ubuntu or debian? [00:08] reordering for Ubuntu. the gnome-3-28 branch is in gnome git [00:43] jbicha: reordering how, sorry, I don't understand what's wrong [00:48] Trevinho: it's not wrong, I just thought it would be helpful to sort the patches in debian/patches/series so it was easier to see there which patches are already applied in git master [00:49] I thought we discussed that earlier. That's why I hadn't uploaded to bionic yet. https://irclogs.ubuntu.com/2018/04/17/%23ubuntu-desktop.html#t18:23 [01:41] jbicha: ah, ok... how you want them to be separated? Want me commenting them? [01:41] It's already defined in Origin: tag though [02:34] Afternoon thumper [02:34] Evening Trevinho [02:34] And hi jbicha [02:34] hey duflu [02:34] o/ duflu [02:34] and Trevinho [02:34] hey thumper [02:34] :) [02:49] duflu: woudl be the case to include also https://gitlab.gnome.org/GNOME/mutter/commit/2d80fd02e76bbe17dc52072299dda92ab88c99c0 ? [02:50] Trevinho, that would certainly fix our #1 crasher bug 1748450 [02:50] bug 1748450 in gnome-shell (Ubuntu Bionic) "gnome-shell crashed with SIGTRAP in _g_log_abort() from g_log_default_handler() from default_log_handler(message="Connection to xwayland lost") from g_logv() from g_log() from " [High,Triaged] https://launchpad.net/bugs/1748450 [02:50] duflu: well, or at least would make us understand what's leading it [02:50] I'm including it let's see [02:51] Trevinho, it doesn't help to solve the problem, but it will silence the huge amount of crash reports. I think it's a good small step, yes [02:52] Although some people will report bugs without debug info then... saying "my shell just vanished" [02:59] We still have the problem of Xwayland failing and no crash reports for it. [03:00] that's a thing that whoopsie should handle though [03:00] But silencing those crash dialogs in 18.04 would be excellent [03:00] Trevinho, no Xwayland fails without crashing, and triggers a crash in gnome-shell only. It's an Xwayland issue that never leaves a core/crash file [03:02] duflu, jbicha: https://salsa.debian.org/gnome-team/mutter/merge_requests/2/diffs [03:02] GNOME bug (Merge request) 2 in mutter "debian/patches: handle xwayland crashes better" (comments: 0) [Opened] [03:02] * Trevinho loves salsa [03:02] can't wait to have the same way of working with patches in ubuntu too... Well, I'm doing it already but still need to handle with bzr eventually. But still better way [04:33] Trevinho, I feel conflicted about letting gnome-shell just exit. It will result in some people logging bug reports that we can't diagnose, but those will be orders of magnitude fewer than what we see now [04:35] So short term for 18.04, it's a good idea. Slightly beyond that, it's going to keep hurting us if nobody fixes Xwayland [06:47] seb128, hi [06:48] seb128, to answer your question from yesterday evening, it's attached xwayland of the gdm user [06:48] +to [07:05] duflu: https://gitlab.gnome.org/GNOME/gnome-shell/merge_requests/73 I missed this in the g-s patches, you wanted to include it? [07:05] GNOME bug (Merge request) 73 in gnome-shell "js/ui: Choose some actors to cache on the GPU" (comments: 1) [Opened] [07:06] Trevinho, yes, if possible. I am answering your question on it now [07:06] duflu: is there an ubuntu bug for that? [07:07] Lemme check [07:07] Trevinho, yes the merge request links to two upstream bugs, and one of them links to LP too [07:08] duflu: ok [07:14] good morning [07:24] good morning desktopers [07:24] hey duflu Trevinho jibel [07:24] lut didrocks [07:25] hey seb128 [07:26] Hello seb128 [07:27] jibel, weird that it doesn't catch an error :/ [07:27] hey seb128, Trevinho, duflu [07:28] hi didrocks [07:28] Hi didrocks [07:32] duflu: I've added that in https://code.launchpad.net/~3v1n0/gnome-shell/bionic-patches-picks-reorder/+merge/343479 [07:34] Trevinho, thanks. Are we getting 3.28.1 too? [07:34] duflu: that's based on 3.28.1 [07:34] OK [07:51] willcooke, hey, sadly i need to skip our morning mtg today [07:52] morning koza, no worries. I dont have anythng anyway. duflu you want to meet anyway? [07:53] hey willcooke [07:53] hi didrocks, how goes? [07:53] (he asks him knowingly) [07:54] willcooke, Nope. [07:54] hey willcooke koza, nothing from me either on the bluetooth front [07:54] I'll do an update post-release [07:55] willcooke: lalala :) [07:55] It might be a single bugfix, or it might be 5.49 [07:55] * didrocks does good use of his time, grepping for copyrights and filing up debian/copyright (which will be out of date in few weeks, like all packages in distro :p) [07:55] good morning desktoppers [07:55] hey oSoMoN [07:55] salut didrocks [08:01] yo [08:02] hey Laney [08:02] hey Laney [08:05] Hi Laney [08:06] willcooke, Sorry, it's an awkward time of day. Either like today, the question is asked when I am afk doing pre-meeting prep, or like other weeks we decide to cancel only minutes before you log on [08:08] * didrocks uploads to a ppa for test build in a clean env [08:13] hey Laney oSoMoN [08:13] hey oSoMoN didrocks duflu seb128!!! [08:13] salut seb128 [08:16] seb128: willcooke: ok, vendored ubuntu-report uploaded and waiting now in UNAPPROVED. (I still did a little extra to use distro deps for the golang source bindings which stays in universe). More info on bug #1764961 and https://bugs.launchpad.net/ubuntu/+source/ubuntu-report/+bug/1759540/comments/10 (commented on the MIR) [08:16] Ubuntu bug 1759540 in ubuntu-report (Ubuntu) "[MIR] ubuntu-report: send telemetry data to ubuntu server" [Undecided,Incomplete] [08:16] bug 1764961 in ubuntu-report (Ubuntu) "Use vendor dependencies" [Undecided,New] https://launchpad.net/bugs/1764961 [08:16] \m/ [08:18] didrocks, great, thanks! [08:19] yw :) [08:25] didrocks, "Note thus that "golang-github-ubuntu-ubuntu-report-dev" should still be kept in main." [08:25] didrocks, you meant "universe" there? [08:26] oupsss, where? [08:26] I wrote that 3 times ;) [08:27] ok, not the ones in the package and comments [08:27] ah, in the MIR comment [08:28] commented, thanks seb128! [08:28] np! [08:47] Oh, I did get credited in the gstreamer-vaapi git log even though they didn't use my original patch. That's nice. [08:47] * duflu relearns from his own code [08:47] :/ [08:51] seb128: gnome-initial-setup wasn't promoted? Should I? (as ubuntu-report is still in main) [08:51] at least, even if the MIR team doesn't get back to us on the final changes for u-r, we'll have g-i-s tested in tomorrow's iso [09:07] didrocks, wfm [09:07] thx [09:08] ok, doing it [09:08] yw ;) [10:40] andyrock: those xdg autostart files - don't we need to stop the path units when the user logs out? [10:48] or otherwise they could be started in default.target and the corresponding units made to cleanly exit if they can't run? [11:03] * duflu wanders into the night [11:13] Laney: mmm how? [11:14] Laney: I mean how to stop when the user logs out? [11:14] I don't know :( [11:14] mmm let me take a look [11:14] we get that with graphical-session.target, but I'm not sure how to do it without that [11:25] apart from something watching the XDG session or gnome-session [11:30] Trevinho: you're not still around now, right? [11:32] tseliot: It appears that Ubuntu flavours (MATE, Budgie, Kubuntu, Xubuntu have been tested) all boot to a black screen on hybrid graphics laptops with nvidia 390 drivers installed on 18.04. [11:33] This doesn't affect Ubuntu proper with GNOME. [11:33] Wimpress: is the nvidia-prime package installed? [11:33] What information do you required to investigate this? [11:33] On Ubuntu MATE yes. [11:34] I have an OEM with 2 laptop models that are affected. [11:34] So we can gather useful information. [11:34] What do you need. [11:34] Wimpress: I'm going to need /var/log/gpu-manager.log, /var/log/Xorg.0.log, dmesg output [11:34] OK. [11:35] I'll raise a new bug a gather some debug. [11:36] Wimpress: thanks [11:45] tseliot: Let me introduce mwilson-e [11:46] He runs Entroware, an OEM that sells Ubuntu laptops. [11:47] mwilson-e: Is going to collect the debug you've requested and attach to this bug https://bugs.launchpad.net/ubuntu/+source/nvidia-graphics-drivers-390/+bug/1764005 [11:47] Ubuntu bug 1764005 in nvidia-prime (Ubuntu) "Black-screen on boot with nvidia 390 for Budgie, MATE, Xubuntu, Kubuntu" [Undecided,Confirmed] [11:47] Wimpress: ok [11:48] tseliot: Mike has two laptop models to test against so if you need more info he can provide it. [11:48] I'll do the same for the XPS 15 9550. [11:48] Wimpress: that's good. I'll wait for mwilson-e to attach the logs for now [11:49] Nice to meet you. Working on gathering all the logs now. [11:49] Sure, will be a little while since he is going to be thorough :-) [11:56] Laney: stopping the services on logout does not seems easy [11:57] seb128: ^^^ [11:57] Laney: seb128 I would drop the upload of update-notifier for the moment, I'll try to find a way to show livepatch notification without this [11:57] maybe it doesn't matter that much??? [11:58] what's the impact - ugly failures? [12:01] andyrock: sorry for not thinking of that before [12:01] Laney: so all the other services are not effected because there is a GFileMonitor fallback [12:02] we can do the same [12:02] there's a service running all the time? [12:02] yeah [12:03] update-notifier [12:03] oh right [12:03] well then that could watch logind and stop the path units too [12:04] but just doing the file monitor directly might be simpler [12:06] andyrock: ok, I'll reject, ping me for a re-review if you want [12:06] Laney: thx [12:11] andyrock, k [12:28] tseliot: See https://pad.lv/1764005 - I've added logs and details I what I experience on the XPS 15 9550. [12:28] Launchpad bug 1764005 in nvidia-prime (Ubuntu) "Black-screen on boot with nvidia 390 for Budgie, MATE, Xubuntu, Kubuntu" [Undecided,Confirmed] [13:06] gedit fans - is the right click menu using a strange font for you? [13:06] looks like its monospace condensed or something [13:06] willcooke: mine is on artful [13:06] mdeslaur, is funny? [13:07] yeah, [13:07] willcooke: yeah, like you said, it's a small monospace [13:07] thanks mdeslaur [13:07] Trevinho, any ideas on that one? [13:07] Can't belive I haven't noticed it before [13:16] willcooke: That content menu font issue is an upstream bug. [13:17] Affects all gtk3 editors with context menus. [13:17] ah ha! thanks Wimpress [13:17] Wimpress, this doesn't change my mind about gedit though, before you start on that one ;) [13:17] tseliot: You'll find more debug from mwilson-e in https://pad.lv/1764005 [13:17] Launchpad bug 1764005 in nvidia-prime (Ubuntu) "Black-screen on boot with nvidia 390 for Budgie, MATE, Xubuntu, Kubuntu" [Undecided,Confirmed] [13:29] Wimpress: that's a lot of logs, thanks! [13:29] Most welcome :-) [13:30] tseliot: I've also just discovered `ubuntu-drivers autoinstall` fails. [13:30] I'll raise a bug for that too. [13:31] Wimpress: that would be very welcome. Thanks [13:31] tseliot: If you need any others, just let me know. [13:35] seb128: I guess we need to add dh-migrations as a dep in software-properties [13:35] ? [13:38] otherwise we can just drop the migration script [13:39] mwilson-e: maybe the output of "journalctl --since=today" would help, as I don't see anything suspicious in the logs [13:40] tseliot: I'll get that now, any particular distro? [13:41] mwilson-e: just one that fails would be fine. I'm looking for clues [13:42] andyrock: man dh_migrations ;) [13:42] it tells you what to do in term of deps, files and debian/rules ;) [13:43] (waow 2012… 6 years ago) [13:45] seb128, I just got a gnome-shell crash with gdb attached to xwayland. What do you need now? [13:47] tseliot: https://pad.lv/1765053 [13:47] Launchpad bug 1765053 in ubuntu-drivers-common (Ubuntu) "ubuntu-drivers autoinstall fails to install nvidia 390 drivers" [Undecided,New] [13:47] tseliot: Just posted the requested log to the bug report. [13:50] Wimpress, mwilson-e: thanks [13:58] andyrock, indeed, I'm fixing that [13:58] seb128: thx! [13:58] jibel, what is gdb telling you, did it stop? [13:59] andyrock, np, sorry for not spotting that before upload [13:59] seb128, nothing, it just stopped. I dumped a core of the process just in case [13:59] jibel, can you make a photo? ;) [13:59] like it stopped giving you back a prompt where you can type "bt"? [14:00] yes [14:00] which I did https://paste.ubuntu.com/p/wHWtZsKwts/ [14:00] not really exciting trace [14:01] hum [14:01] did it indicate a sigsegv before? [14:01] the bt looks like it's active/waiting for events [14:01] not that it hit a bug [14:01] no, really nothing. It just stopped and gave me a prompt [14:02] weird [14:02] and if you do "c" it resumes? [14:02] yes [14:02] I'll keep it attached and try again [14:02] and you got the gnome-shell issue as well? [14:03] yes, g-s crashed and apport generated a crash file [14:04] it's uploading [14:10] k, let's see what it gives us === pstolowski is now known as pstolowski|lunch [14:37] Trevinho: Please can you take a look at this - https://bugs.launchpad.net/ubuntu/+source/indicator-sound/+bug/1737834 [14:37] Ubuntu bug 1737834 in indicator-sound (Ubuntu) "[FTBFS] indicator-sound fails to build, test failures" [High,Confirmed] [14:38] Is that merge proposal OK with you? [14:39] tseliot: Not sure if it helps you but if I install Ubuntu MATE BETA 2, not the daily, and install the nvidia driver it works fine. But once the `ubuntu-driver-common` package is upgraded the problem then presents itself. [14:40] Wimpress, Trevinho, disabling tests doesn't seem the right way out no [14:40] mwilson-e: what version of ubuntu-drivers-common does Ubuntu MATE BETA 2 ship with? [14:40] seb128: OK, thanks. [14:40] np [14:40] tseliot: It ships with 1:0.5 [14:41] the upgrade to 1:0.5.1 kills it [14:42] mwilson-e: do things work if you boot the system with the "nogpumanager" parameter? [14:45] tseliot: Just tested on two machines, the nogpumanager param does not work. [14:47] mwilson-e: remove /usr/share/X11/xorg.conf.d/11-nvidia-prime.conf first, then boot with "nogpumanager" [14:52] tseliot: That works, but in the nvidia-settings application many of the options are gone. only PRIME profiles, Application Profiles, and nvidia-settings Configuration tabs are displayed [14:52] kenvandine, when you have a minute: https://code.launchpad.net/~osomon/evince/snap-allow-printing/+merge/343521 [14:54] seb128, kenvandine: thanks for the endorsement! one to go… [14:55] oSoMoN, happy too! [14:55] mwilson-e: that was just to find what was causing the problem. Can you post your /var/log/Xorg.0.log please? [14:55] oSoMoN, i just reviewed it, i think we need avahi-observe too [14:57] kenvandine, ack, adding it [14:58] tseliot: added it to the bug report. [14:59] kenvandine, updated [15:01] kenvandine, we should request auto-connection of the cups-control plug once this is merged and published in the store, I'm happy to create the request on the forum if you let me know when to do it [15:01] oSoMoN, yw, should be easy to get the 3rd one :) [15:01] oSoMoN, yeah [15:02] oSoMoN, merged and the snap is building now [15:04] mwilson-e: thanks [15:05] tseliot: No problem, if you need anything else, just ask. [15:05] hum [15:06] is the trello board having background textures on most of the card now for some reason? or is it me/my browser being buggy? [15:06] seb128, i saw the same too [15:06] it looks very 1990s :) === pstolowski|lunch is now known as pstolowski [15:07] that's one of the powerups [15:07] kenvandine, it looks like it's proportional to the time they haven't been updated maybe [15:07] card aging [15:07] lol [15:07] someone must have put that on [15:07] you're right [15:08] sounds like a trick from seb128 to keep us on our toes [15:08] lol [15:08] Trevinho did it [15:08] he activated "Card Aging" [15:09] Trevinho, please don't do such changes to the board without discussing them here first [15:11] next step: make the cards to remind us to update them :) [15:11] :-p [15:11] :p [15:11] which remind me, some people still didn't update their inprogress card, I should do another follow up [15:15] Hi jbicha, did you see my docs idea about the screen keyboard? [15:16] GunnarHj, hey, I saw your comment but I don't really know what different it makes to set that IM variable to ibus or not, I agree it's late to change it though so probably best as a SRU [15:16] kenvandine [15:17] kenvandine, jamesh, jibel, you are basically the ones who still have in-progress cards in the trello board that didn't get updated [15:17] could you please review your cards? [15:17] seb128: We can do that as an SRU, but it still doesn't help on Xorg AFAIU. [15:19] seb128: So the very latest idea is to inform the users in the desktop guide about what they need to do to make it work. [15:20] well, or to fix it for xorg in a SRU as well [15:20] seb128: Sure, if that is at all possible. I didn't get that impression from the comments on the upstream issue. [15:24] bugs are always possible to fix, might not be trivial though :/ [15:24] the cycle has been too busy for me to look at OSK issues but we should make sure it works good enough by .1 [15:25] seb128: Ok, so then I'll wait with that docs text. Plan A: Fix it, also on Xorg as an SRU and plan B: inform the users with a docs SRU. [15:26] GunnarHj, what are the main issues in your opinion? (I didn't follow much, capital letter and special chars can't be inputed?) [15:27] seb128: That's it. Bad enough IMO. [15:27] yeah... [15:28] andyrock, Laney, since the update-notified upload has been rejected I uncommited the change and push --overwrite the vcs without those to not confuse the next uploader, unsure if that's the right thing to do in those case (it makes the status of the merge mp wrong/confusing) [15:29] seb128: thanks [15:29] I was already working on an alternate fix [15:29] let me "rebase" [15:31] I usually try to not push the tag until it's accepted [15:31] that is easier with git than bzr, but still can be done [15:31] biggest risk is forgetting [15:31] oSoMoN, evince has built, i've tested it [15:31] oSoMoN, in stable now [15:32] Laney, right... [15:32] kenvandine, excellent, thanks, I'll write the auto-connection request [15:32] oSoMoN, cool [15:32] anyway, thx, probably not that much chance somebody did pull in the meantime [15:32] apart from me [15:32] :D [15:33] right [15:33] haha [15:33] poor andyrock === sergiusens_ is now known as sergiusens [15:40] kenvandine, https://forum.snapcraft.io/t/auto-connecting-the-cups-control-interface-for-the-evince-snap/5033 [15:45] GunnarHj: I'm not really a fan of the docs idea. I agree we need to understand more what the purpose of GTK_IM_MODULE is [15:46] GunnarHj: on GNOME/Ubuntu on Wayland, it's my understanding that ibus is the only supported input method so if we could skip that variable when run in those sessions I think the regression potential is limited [15:47] we'll need to open a Debian bug because Debian has the same problem (ibus there recommends im-config but I think ibus isn't installed by default there) [15:48] jbicha, but it means =ibus or empty leads to different components to be used? [15:48] so it's a codepath/feature/behaviour change [15:48] and we didn't test much the other one, so seems a risky change that late, especially without someone in the team who understands the difference [15:49] I agree with not rushing it in now while we don't fully understand what's going on and Debian hasn't made the suggested change either [15:50] jbicha, seb128: My gut feeling is that if IBus works on GNOME without that var being set, it probably would work for us too. But still... [15:50] interestingly, there is a proposed simple patch to get the screen keyboard working in the Xorg session with GNOME 3.26 (17.10) but it may be a lot more complicated with current GNOME [15:51] seb128: the fix is almost ready. it's quite small [15:51] just giving a final test [15:52] mwilson-e: I'm going to need the output of these two commands too, please: uname -r , dkms status [15:53] andyrock, great! [15:56] tseliot: 4.15.0-15-generic [15:56] tseliot: nvidia, 390.48, 4.15.0-15-generic, x86_64: installed [15:57] mwilson-e: the weird thing is that, when you disabled gpu-manager, the X log showed an actual error, and the driver wasn't really working [15:58] jbicha: I don't think either that Debian installs any IM stuff by default. So in their case, with GNOME on Wayland, the screen keyboard works until they install IBus which pulls im-config? [15:58] seb128: can I drop the UIFe status from the bug? [15:59] tseliot: Do you want me to run those commands with gpumanager disabled? [16:00] GunnarHj: I'd have to try again. My initial testing was broken because the screen keyboard didn't work how I expected (I filed bugs) [16:00] mwilson-e: it wouldn't change the output of those commands [16:00] seb128: you skipped a version in software-properties, did you want to fix that? [16:01] jbicha: Ok. [16:01] also it's not control.in :p [16:03] tseliot: I see, are there any other logs that I can provide? [16:04] seb128: https://code.launchpad.net/~azzar1/update-notifier/fix-1761841-v2/+merge/343540 [16:04] mwilson-e: I might need you to test packages from a PPA. Not sure if today or tomorrow [16:04] oSoMoN, thx [16:05] tseliot: no problem [16:06] oSoMoN, i'm going to request it for gedit as well [16:06] * Laney just got the meaning of azzar1 for the first time [16:07] azzarONE [16:07] :D [16:07] mind is blown [16:07] oSoMoN, with a bit of copy and paste from your request :) [16:07] they usually call me azzarONE in USA [16:07] at least in the airports [16:08] where does the rock come from? [16:09] you can smell what Andrea is cooking [16:10] andyrock, thanks, I'm getting ready for sport so I might not get to it before later this evening, but it's on my list [16:10] Laney, ups, can do but ^ [16:11] if somebody wants to fix it for me please do, otherwise I can do when I'm back in the evening [16:11] andyrock, UIFe drop is fine I believe yes [16:11] ok [16:11] jbicha: from a mistake :D [16:11] I was young and wanted to copy didrocks [16:11] but I forgot the final 's' [16:12] :D [16:12] andyrock: this monitor_files function is :( [16:12] did_rocks is a model for the new generation! [16:12] Laney: you mean the existing one or my changes? [16:12] the one that was already there [16:12] everyone wants to be like didrocks [16:13] I agree that update-notifiers needs some love [16:13] next-cycle-love [16:14] there's some weird indirection between the file monitor cb and a timer [16:14] and it leaks all over the place [16:24] \o/ [16:36] andyrock: might want to add some error checking around the atoi thing, you get a 0 outputted if the file contains weird contents [16:36] (minor) [16:47] mwilson-e: can you add this PPA, and when they packages build, upgrade the xserver, please? https://launchpad.net/~albertomilone/+archive/ubuntu/xtesting [17:00] tseliot: Just tested that package and it does not fix the issue. [17:01] mwilson-e: did you reboot? Did you boot without "nogpumanager"? [17:02] tseliot: yes I did reboot, this was also done on a fresh install with only the nvidia drivers installed. [17:03] mwilson-e: ok, I'll install one of those distros, and I'll see if I can reproduce the problem here [17:03] tseliot: Just so you know, I tested this on Ubuntu MATE daily. [17:04] mwilson-e: ok, I'll go with that then, thanks === pstolowski is now known as pstolowski|eod [17:52] seb128: eh, see... the aging is what you want in fact :-D [17:52] I was trying some power-ups and I forgot to disable sorry [18:09] jbicha: about not adding that patch, both me and duflu are fine with that... Upstream wasn't keen, but in the new PRs there's no sign of that [18:16] Trevinho: I guess you're getting close to being on Robert's timezone :) [18:39] Laney: can you open a bug about that? [18:39] if you got spare time [18:39] I guess not [18:40] night all [18:58] good night all === mwilson-1 is now known as mwilson-e [21:04] andyrock, how did you make https://gitlab.gnome.org/azzaronea/gnome-initial-setup ? I didn't think we could do that until GNOME/gnome-initial-setup existed. [21:04] I'm trying to work out how we push to the Community/Ubuntu/gnome-initial-setup so we can all commit [21:05] robert_ancell: I just created a project called gnome-initial-setup [21:05] it's not a fork of the main one [21:05] I mean it's a fork but it has not been created with the fork button [21:07] ah [21:24] robert_ancell: the very rough concensus was that we should use a LP ubuntu-desktop git repo for that next month [21:25] there are 3 logical places for hosting and we have advocates for all 3 🙁 [21:26] jbicha, for packaging or working on branches? [21:26] it's the same thing here isn't it? [21:27] not necessarily... The branches should be close to upstream as that's where we're trying to get them. The packaging doesn't go upstream, so it doesn't matter so much where it lives. [21:28] ok I see, maybe it makes sense for those to be separate commits and it wouldn't hurt to have that in GNOME gitlab [21:29] it just makes things a bit complex as we need 3 different remotes to handle gnome-initial-setup then 😐 [21:30] I can deal with that workflow; it's just going to be a bit of a learning curve for some on our team I suppose [21:33] but I guess we almost do that with gnome-software already except 1 of the 3 is bzr [21:44] robert_ancell: hey could you please publish https://bileto.ubuntu.com/#/ticket/3223 ? [21:55] Trevinho, done [21:55] robert_ancell: thanks [22:18] robert_ancell: if you've some time also https://code.launchpad.net/~3v1n0/gnome-shell/bionic-patches-add-gpu-cached-texture/+merge/343565 would need merging [22:18] (and sponsoring) [22:19] Trevinho, not sure if I'll find time to look at that today and I'm on holiday from tomorrow [22:19] robert_ancell: ah ok, no worries... In case I'll ask other dev :) [22:21] robert_ancell: sorry to second guess you, but I'm surprised you removed the gnome-getting-started-docs popup part at the end of gnome-initial-setup [22:21] jbicha, design asked to remove it [22:21] ok [22:34] robert_ancell: on the Ready to go screen (software) panel in your proposed ubuntu4, I see back and forward arrows but they aren't active here [22:34] it was working in the ubuntu3 version [22:39] jbicha, what do you mean "aren't active"? [22:45] they are greyed out: I only see one page worth of snap apps on that panel [22:51] jbicha, oh, right. Yes, there's more shown on a page, so they all fit into one page. I haven't asked design what we should do, we might as well hide them I guess. [22:54] jbicha, I updated the git branch to disable them when there's one page [22:58] andyrock, I moved the branch to https://gitlab.gnome.org/Community/Ubuntu/gnome-initial-setup so feel free to merge in your GOA changes when you're happy with them. [23:17] robert_ancell: hey, another nitpick. In the Google Doc it says "approximate location you chose" but in Github, it says "approximate location you choose" (note final word) [23:17] I like the "chose" tense better, what do you think? [23:18]