13:30:42 <cmurf> #startmeeting Workstation WG (2020-06-02) 13:30:42 <zodbot> Meeting started Tue Jun 2 13:30:42 2020 UTC. 13:30:42 <zodbot> This meeting is logged and archived in a public location. 13:30:42 <zodbot> The chair is cmurf. Information about MeetBot at http://wiki.debian.org/MeetBot. 13:30:42 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic. 13:30:42 <zodbot> The meeting name has been set to 'workstation_wg_(2020-06-02)' 13:30:44 <cmurf> #meetingname workstation 13:30:44 <zodbot> The meeting name has been set to 'workstation' 13:30:46 <cmurf> #chair cmurf 13:30:46 <zodbot> Current chairs: cmurf 13:30:48 <cmurf> #topic Rollcall 13:30:50 <King_InuYasha> .hello ngompa 13:30:50 <zodbot> King_InuYasha: ngompa 'Neal Gompa' <ngompa13@gmail.com> 13:30:59 <King_InuYasha> let's see if my connectivity holds up... 13:31:36 <King_InuYasha> (I don't tend to have reliable IRC connectivity) 13:33:06 <cmurf> well this could be short :D 13:35:12 <cmurf> is it the meeting location or is it the agenda or is it IRC connectivity? 13:35:23 <petersen> Hello 13:35:26 <mclasen> err 13:35:29 <tpopela> Didn't know it's on IRC :D we were on Bluejeans :D 13:35:32 <mclasen> we were all on bluejeans 13:35:43 <cmurf> haha 13:35:43 <mclasen> aday is on pto today 13:35:53 <mcatanzaro> .hello catanzaro 13:35:54 <zodbot> mcatanzaro: catanzaro 'Michael Catanzaro' <mcatanza@redhat.com> 13:36:00 <mclasen> .hello mclasen 13:36:01 <zodbot> mclasen: mclasen 'Matthias Clasen' <mclasen@redhat.com> 13:36:02 <petersen> Despite reading your agenda mail 10min ago I still ended up in BJ lol 13:36:06 <cmurf> ok nice 13:36:09 <tpopela> .hello tpopela 13:36:10 <zodbot> tpopela: tpopela 'None' <tpopela@redhat.com> 13:36:13 <petersen> .hello2 13:36:14 <zodbot> petersen: petersen 'Jens Petersen' <petersen@redhat.com> 13:36:34 <mcatanzaro> We have langdon coming too, half of didn't read the meeting agenda and wound up in BlueJeans 13:36:52 <mcatanzaro> aday sends his regrets today, but I guess that's OK if we have meetbot to take minutes ;) 13:36:58 <cmurf> old new habits die hard 13:37:05 <petersen> yeah haha 13:37:15 <cmurf> we'll see what happens next week, do we all end up here? 13:37:28 <petersen> hopefully not ;-P 13:37:30 <cmurf> i haven't seen kalev in a while, is he OK? 13:37:46 <mclasen> I haven't seen him today 13:38:10 <tpopela> cmurf: mclasen he replied to some email where I'm CC'ed few minutes ago.. 13:38:11 <mclasen> I exchanged emails with him last week. he was responding 13:38:35 <cmurf> #topic Approve 26 May minutes 13:38:37 <cmurf> #link https://meetbot.fedoraproject.org/teams/workstation/workstation.2020-05-26-23.53.log.html 13:38:48 <cmurf> objections? 13:39:11 <cmurf> #agreed No objections - approved 13:39:18 <cmurf> #topic Announcements 13:39:19 <otaylor> .hello otaylor 13:39:22 <otaylor> (sorry to be late - rebooted to try and fix sound, then realized it was irc today) 13:39:22 <zodbot> otaylor: otaylor 'Owen Taylor' <otaylor@redhat.com> 13:39:50 <cmurf> I don't think I have anything qualifying as an announcment, got a couple things for followup that will be brief. 13:41:20 <cmurf> oh there's maybe an idea for updating the bootloader on Silverblue and CoreOS, https://github.com/coreos/fedora-coreos-tracker/issues/510 13:41:25 <langdon> .hello2 13:41:26 <zodbot> langdon: langdon 'Langdon White' <langdon@redhat.com> 13:41:29 <langdon> sorry.. my bouncer fell over and didn't want to come back 13:41:47 <cmurf> bootupd 13:42:15 <cmurf> ok no other announcements 13:42:18 <feborges> .hello2 13:42:19 <zodbot> feborges: feborges 'Felipe Borges' <feborges@redhat.com> 13:42:27 <cmurf> #topic Followups 13:42:29 <cmurf> #info Supporting hibernation in Workstation ed., draft 2 13:42:31 <cmurf> #link https://pagure.io/fedora-workstation/blob/master/f/hibernationstatus.md 13:42:33 <cmurf> quick followup 13:42:43 <cmurf> Done and published. Got a bump already over the weekend to clarify the nature of the hibernation as loophole around Secure Boot risk. 13:42:57 <cmurf> Yesterday found a new hibernation bug, and I've taken it directly upstream. 13:43:12 <cmurf> I guess I like pain? 13:43:46 <petersen> hehe 13:43:46 <cmurf> Discussion on devel@ is going pretty smoothly, and light. 13:44:01 <cmurf> That's all I've got at the moment. Any questions on that? 13:44:33 <tpopela> cmurf: nothing from my side. Good work! 13:45:20 <cmurf> I can totally see why users pull their hair out on it. Not their fault! 13:45:25 <cmurf> #info enable swap-on-ZRAM by default 13:45:27 <cmurf> #link https://pagure.io/fedora-workstation/issue/127 13:45:35 <cmurf> let's see... 13:45:43 <cmurf> change proposal preview is being circulated to various affected groups: anaconda, editions: cloud server iot coreos 13:45:55 <cmurf> some tweaks are needed and then i'll set it to ready for wrangler status. No sadness from cloud or server folks about hibernation. 13:46:26 <cmurf> that's all i've got on that 13:46:38 <cmurf> any other followups/status reports needing attention? 13:47:21 <cmurf> alrighty 13:47:22 <otaylor> I can give a quick status on assembling a whitelist for flathub 13:47:29 <cmurf> oh yeah go ahead 13:47:50 <otaylor> I started working from the flathub statistics about what is downloaded most frequently 13:48:24 <otaylor> Looking in particular at *from scratch* downloads, not deltas, as a proxy for "fresh install" - since otherwise, you see the apps that people update most frequently at the top of the stats 13:49:03 <langdon> sorry.. i need more coffee... what is the context for the whitelist for flathub? 13:49:12 <langdon> my memory is failing me 13:49:16 <otaylor> In quick summary, what people download most are "extra-data wrappers around proprietary apps" 13:49:23 <mcatanzaro> #topic Whitelist for flathub 13:50:00 <otaylor> langdon: we want to add flathub, but not *all of flathub* 13:50:09 <cmurf> #chair mcatanzaro 13:50:09 <zodbot> Current chairs: cmurf mcatanzaro 13:50:34 <cmurf> #topic Whitelist for flathub 13:50:35 <otaylor> Not all of flathub because a) potential patent issues b) fedora legal wants the ability to remove individual apps if concerns arise 13:51:06 <langdon> oh .. this guy? https://pagure.io/fedora-workstation/issue/108 13:51:27 <petersen> yes 13:51:35 <langdon> gotcha.. thanks 13:51:46 <cmurf> #link https://pagure.io/fedora-workstation/issue/108 13:52:21 <otaylor> Of the top 50 apps, 15 are proprietary - mostly wrappers (1 with redistribution permission), 10 are things we already have fedora flatpaks for, 6 have patent issues 13:52:39 <cmurf> interesting 13:52:40 * petersen ponders voice-to-chat :) 13:52:52 <otaylor> 14 are in fedora, but Flatpaks 13:53:05 <langdon> is the list somewhere? 13:53:09 <otaylor> and 7 are unpackaged OSS software 13:53:40 <petersen> So only 7? 13:53:43 <langdon> any consistency to the "frameworks"? ie.. are they all electron apps? (all == the 7) 13:54:19 <otaylor> https://fishsoup.net/misc/flathub-popular-may-downloads.txt 13:54:29 <otaylor> that's my notes 13:54:49 <feborges> yay, Boxes is in the list :) 13:55:14 <otaylor> langdon: "planner" is not - it's a Elementary app in vala, others are, I think 13:56:01 <otaylor> (I tend to doubt playonlinux is packageable leaving aside the electron part) 13:56:15 <otaylor> Well, pycharm clearly isn't electron either 13:56:37 <langdon> postman ranking so highly seems weird 13:56:47 <tpopela> Telegram is a Qt app.. 13:57:03 <King_InuYasha> planner is packaged in Fedora 13:57:06 <King_InuYasha> as elementary-planner 13:57:27 <sub_pop> Having a first-rate VSCode experience would be _amazing_. The Flatpak is frustrating to work with. 13:57:28 <kwizart> telegram relies on ffmpeg (unless one want a disabled version) 13:57:32 <otaylor> I think one main question I have is "In Fedora, No Flatpak" - most of these aren't wouldn't be hard to make Fedora Flatpaks of 13:57:45 <petersen> Do those ".........................." dots mean something? 13:57:57 <langdon> ms packages code as an rpm.. with a repo.. 13:58:16 <otaylor> petersen: the ..... dots indicate the cut-off lines if we wanted to whitelist 20 apps 13:58:24 <petersen> Ah I see 13:58:45 <mclasen> otaylor: if you say "main question"... what is the question ? 13:58:53 <langdon> sub_pop: i generally think flatpak needs to solve the "add on" issue.. how do you make extensions work that really on natively installed things.. e.g. python ext using pylint (although i think that example is solved) 13:59:35 <tpopela> otaylor: it will depend on whether they are as RPMs in Fedora.. Otherwise you would have to find maintainers, .. 13:59:38 <otaylor> mclasen: Do we want to whitelist these from flathub, or do we want to encourage/wait-for them to be Flatpak'ed in Fedora - of course that's dependent on a different question 14:00:09 <otaylor> Do we want to whitelist apps that have Fedora Flatpaks, or concentrate on things that are additions 14:00:25 <tpopela> otaylor: I would prefer the additions.. 14:00:43 <otaylor> tpopela: " it will depend on whether they are as RPMs in Fedora" - don't quite follow 14:01:41 <cmurf> re: in Fedora, no Flatpak not being hard to make Flatpaks; how hard? what's needed? 14:03:00 <otaylor> cmurf: how hard it depends on the app - it's somewhere between 10 minutes and weeks and weeks (weeks and weeks being libreoffice mostly :-) 14:03:11 <cmurf> oh man 14:03:14 <cmurf> that's a spread 14:04:14 <otaylor> cmurf: https://docs.fedoraproject.org/en-US/flatpak/tutorial/ is the documentation, https://fedoraproject.org/wiki/Flatpak:Easy categorizes some apps, though it's way out of date at this point 14:04:20 <cmurf> i'm not sure how to go about answering the question 14:04:33 <tpopela> otaylor: I mean all the dependencies and so on.. You really can't make a Fedora Flatpak from anything that's not available as an RPM in Fedora no? 14:04:39 <mclasen> otaylor: I think my vote would be to focus the whitelist on additions, and not wait for extra packaging effort 14:04:40 <King_InuYasha> you cannot 14:04:52 <King_InuYasha> tpopela: ^ 14:04:58 <otaylor> tpopela: yes, My question was about things that *are* availbale as rpms in fedora 14:05:03 <otaylor> mypaint, say 14:05:54 <langdon> this might be off topic.. but are there any "converters" from say appimages or snaps to flatpaks? like build srcs->build srcs 14:07:16 <tpopela> otaylor: ah, ok.. at that point I would prefer to try to flatpak them in Fedora.. Sorry I didn't understood it.. 14:08:07 <mclasen> langdon: there are a few apps on flathub that have been converted from snaps, I believe. I haven't looked in detail if that could be described as "a converter" 14:08:49 <langdon> mclasen: you mean by hand? yeah.. i was looking for something that would get you 90% of the way there.. 14:09:16 <mclasen> I saw the link go by in #flatpak, but that was a month ago... 14:10:53 <mclasen> regarding telegram, why would ffmpeg affect it ? other than sharing videos 14:10:55 <cmurf> i think i'm also +1 to focusing on the whitelist and wouldn't wait for them being flatpak'd in Fedora 14:11:13 <tpopela> mclasen: let me ask jgrulich, that's maintaining it.. 14:11:30 <cmurf> but i'm not sure if there's a way to encourage them being flatpak'd or if there's a distinct advantange to them being flatpak'd in Fedora rather than just whitelisted from flathub 14:12:23 <mclasen> for some people "its all fedora" is considered an advantage 14:12:30 <langdon> mclasen: +1 14:12:33 <cmurf> for sure 14:12:40 <otaylor> cmurf: well, "if there's a distinct advantange to them being flatpak'd in Fedora rather than just whitelisted from flathub" can be a can of worms to open 14:12:40 * mclasen not in that camp 14:13:03 <feborges> we have some interns in the brno office working on converting rpms to flatpaks, so we could point them to work on these specific conversions 14:13:13 <cmurf> i wasn't looking for a can of worms :D but what I meant by distinct is, one that makes it worth waiting for it to happen 14:13:31 <otaylor> cmurf: some people might trust Fedora packaged stuff more, Fedora packaged stuff *with considerable handwaving* is more likely to reliably get security updates 14:13:47 <langdon> i don't think it is worth waiting for.. but i do think we should be encouraging as many as possible to be coming from fedora 14:14:04 <langdon> is the fedora flatpak repo available to other distros? 14:14:09 <cmurf> ok something sorta missing in flatpaks before you have them, is, recency 14:14:23 <otaylor> langdon: note that once we whitelist something, we can't unwhitelist it, since people would then get stuck on whatever we had when we removed the whitelist 14:14:27 <mclasen> langdon: you can install it anywhere, yes 14:14:38 <langdon> mclasen: not flathub.. to be clear 14:14:48 <otaylor> langdon: yes (does require flatpak-1.6) 14:14:48 <cmurf> i see that information fairly clearly with an RPM, using dnf info X, but i don't offhand get any indication how up to date a flatpak is kept unless it's installed already 14:14:51 <cmurf> and not by date 14:15:18 <langdon> otaylor: i think the "stuck on" is fine.. people reinstall all the time.. 14:15:20 <mclasen> otaylor: but that (removing things from the whitelist) is exactly the capability that legal asks from us, no ? 14:15:40 <otaylor> mclasen: well, if there was a legal fire, "stuck on" would be OK, at least for now 14:16:02 <mclasen> so this is just to get fedora onto dry land... users can sink or swm 14:16:07 <mclasen> swim 14:16:16 <mclasen> works for me 14:16:17 <otaylor> mclasen: I do think the whitelist really should have an equivalent to the flatpak eol functionality 14:16:30 <cmurf> i agree 14:16:30 <langdon> and... wouldn;t it operate like rpms.. who ever had latest would be the version you get on update? so if we took it off the whitelist.. the fedora rev would take over on next update 14:16:41 <otaylor> mclasen: But I don't think that would be appropriate if the app just got flatpak'ed in fedora 14:17:09 <otaylor> langdon: flatpaks don't make any assumption that there is a world-spanning set of comparable version numbers :-) 14:17:48 <langdon> oh... is "xyz" from flathub not the same as "xyz" from fedora? intrinsically? 14:17:48 <otaylor> langdon: an app is not-up-to-date, if the current version in the repo is different than what you have installed (unless you've "pinned" what you have installed) 14:18:16 <langdon> yeah.. thats what i was afraid of.. 14:18:43 <cmurf> yeah they can be completely different 14:18:45 <otaylor> langdon: "Version" in Flatpak terms is just a free-form text field 14:19:55 <otaylor> OK, I think we've devolved into random discussion :-) 14:20:05 <King_InuYasha> no surprise there 14:20:09 <langdon> whitelist++ :) 14:20:12 <King_InuYasha> easy to do on IRC :) 14:20:33 <langdon> i did say i was potentially going off topic.. :) 14:20:36 <cmurf> otaylor: could you update #108 with some of this info? 14:20:57 <otaylor> My next step is to turn my notes and questions into more formal form, and send the notes to the desktop list, and put the questions on the ticket 14:21:19 <otaylor> cmurf: yes, will do 14:21:31 <cmurf> #action otaylor to update #108 ticket and send notes to desktop@ 14:21:52 <cmurf> ok so keeping to our preferred 50 minute aimpoint... 14:21:56 <cmurf> #topic Open Floor 14:22:23 <cmurf> on item i forgot, i am working on more speakers for this cycle 14:22:33 <cmurf> if anyone has suggestions let me know 14:22:35 <petersen> great 14:23:13 <mclasen> cmurf: coreos folks ? various topics come to mind 14:24:05 <cmurf> sure - throw out 1 or 2 topics? 14:24:42 <mclasen> installer (ignition vs anaconda), toolbox 14:25:08 <cmurf> ok good 14:25:36 <cmurf> add suggestions to #145 or let me know on IRC by pm if you prefer 14:25:39 <King_InuYasha> fb btrfs folks to talk about their experiences? 14:26:02 <King_InuYasha> since people seem to be weird about btrfs in the WG, maybe that would help allay their fears 14:26:23 <cmurf> ok 14:26:27 <cmurf> anything else? 14:26:50 <mclasen> weird ? 14:26:53 * mclasen lets that slide 14:27:16 <cmurf> thanks for showing up everyone 14:27:30 <cmurf> next week back to the blue :D 14:27:35 <cmurf> #info adjourned 14:27:37 <cmurf> #endmeeting