09:52:44 #startmeeting Workstation WG (2020-03-31) 09:52:44 Meeting started Tue Apr 7 09:52:44 2020 UTC. 09:52:44 This meeting is logged and archived in a public location. 09:52:44 The chair is aday. Information about MeetBot at http://wiki.debian.org/MeetBot. 09:52:44 Useful Commands: #action #agreed #halp #info #idea #link #topic. 09:52:44 The meeting name has been set to 'workstation_wg_(2020-03-31)' 09:52:45 #meetingname workstation 09:52:45 The meeting name has been set to 'workstation' 09:52:45 #chair cmurf 09:52:45 09:52:45 #topic Rollcall 09:52:45 Current chairs: aday cmurf 09:52:45 09:52:47 #info present: tpopela, neal, aday, cmurf, fborges, kalev, mcatanzaro, langdon, mclasen, owen 09:52:49 #info regrets: 09:52:53 #info missing: 09:52:55 09:52:57 #topic Approve minutes for 24 March 09:52:59 #link https://meetbot.fedoraproject.org/fedora-meeting-2/2020-03-31/workstation.2020-03-31-13.07.html 09:53:02 It says 17 March, but it was the 24th. 09:53:04 #agreed Deferred until next week, to give people time to review them 09:53:06 #topic Announcements 09:53:08 #info Final freeze for F32 is scheduled for 7 April 09:53:10 #topic Guidelines for preinstalled and non-removable apps 09:53:12 #link https://pagure.io/fedora-workstation/issue/125 09:53:14 Allan: guidelines for the "default" app set are in progress, but there's a question that needs to be answered first: currently non-removable apps are non-removable in Software only - they can still be removed from the command line. Does the working group agree with that, or do we want to make them non-removable everywhere? Is that even technically possible? 09:53:19 Neal: non-removable apps should be non-removable everywhere. 09:53:23 Matthias: apps should always be removable - if they're not removable, they're not an app. 09:53:25 Discussion ensues - what is an app? Is it something with a launcher? Something that's a flatpak? 09:53:27 Owen: the various aspects of the definition of an app (has a launcher, can be removed, shows a window, is a flatpak, etc) won't always line up. 09:53:30 Michael: apps should not choose for themselves whether they are non-removable. i.e. gnome-software should stop respecting in appstream metadata. It's nuts that any random app can decide to make itself nonremovable. That tag should be deprecated and ignored. 09:53:34 #action Allan to incorporate this discussion into the draft guidelines and bring them back to the WG for discussion. 09:53:37 #topic Default monospace font is extremely small 09:53:39 #link https://pagure.io/fedora-workstation/issue/134 09:53:41 Kalev: this issue was originally raised by Fedora QA. They are always testing clean installs - and so have to use the default settings. We should increase the font size. 09:53:44 Various discussion about the appropriateness of making the change downstream... Should the WG diverge from upstream? Should the WG independently make UX changes? 09:53:47 Allan: is it important to fix this for F32, considering where we're at the end of the cycle? 09:53:49 Kalev: suggests that we go with what's in the Freedesktop runtime (use the system monospace - typically DejaVu Sans 11pt - instead of Source Sans 10pt). 09:53:54 Kalev: we have 6 days until final freeze. 09:53:56 Neal: we should fix it now. 09:53:58 Michael: we shouldn't diverge from upstream if we don't have to. Also we shouldn't flip back and forth between settings, and the font was changed for F32. It would be better to fix this in F33. 09:54:01 Matthias and Owen agree. Owen does think that we should revisit the size. 09:54:03 Michael: can we change the monospace font at the fontconfig level? 09:54:05 #agreed Fix this in F33. 09:54:07 #topic Review Taiga Kanban 09:54:09 #link https://teams.fedoraproject.org/project/workstation-wg-1/kanban 09:54:11 #info The nano task is up for grabs if anyone wants it 09:54:13 #topic Encryption of system data (excludes user) 09:54:15 #link https://pagure.io/fedora-workstation/issue/136 09:54:17 #info We have an encryption subgroup, but it seems a bit stalled. Should the WG take this on directly? 09:54:20 Neal: we have systemd-homed now, which changes things. We can start evaluating that after F32. 09:54:24 Chris: homed does solve some problems. It supports a few different options that we can consider. 09:54:26 Allan: there's some overlap with system migration which is something that might be worked on soon. 09:54:29 Chris: yubikeys are interesting and relevant. 09:54:31 Michael: we need integrity protection if we are going to rely on user directory encryption. 09:54:33 #agreed The WG will take ownership of improving the encryption story, and will invite domain experts to speak to us. 09:54:36 #endmeeting