13:00:12 <mcatanzaro> #startmeeting Workstation WG
13:00:12 <zodbot> Meeting started Mon Apr 10 13:00:12 2017 UTC.  The chair is mcatanzaro. Information about MeetBot at http://wiki.debian.org/MeetBot.
13:00:12 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
13:00:12 <zodbot> The meeting name has been set to 'workstation_wg'
13:00:21 <mcatanzaro> #meetingname Workstation WG
13:00:21 <zodbot> The meeting name has been set to 'workstation_wg'
13:00:39 <mcatanzaro> #meetingname workstation
13:00:39 <zodbot> The meeting name has been set to 'workstation'
13:00:44 <mcatanzaro> #topic Roll call
13:00:49 <mcatanzaro> .hello catanzaro
13:00:50 <zodbot> mcatanzaro: catanzaro 'Michael Catanzaro' <mcatanzaro@gnome.org>
13:00:58 <rdieter> .hello
13:01:00 <zodbot> rdieter: (hello <an alias, 1 argument>) -- Alias for "hellomynameis $1".
13:01:05 <rdieter> ha
13:01:07 <rdieter> .hello rdieter
13:01:08 <zodbot> rdieter: rdieter 'Rex Dieter' <rdieter@math.unl.edu>
13:01:18 <juhp> .hello petersen
13:01:19 <zodbot> juhp: petersen 'Jens Petersen' <petersen@redhat.com>
13:02:25 <mclasen> .hello mclasen
13:02:26 <zodbot> mclasen: mclasen 'Matthias Clasen' <mclasen@redhat.com>
13:03:39 <mcatanzaro> Need one more...
13:04:04 <mcatanzaro> So stickster is obviously not here today. Also otaylor is not going to be here.
13:04:05 * mclasen goes to hunt cschalle
13:04:14 <mcatanzaro> Hopefully someone else will show so that we have quorum.
13:04:38 <mclasen> don't see him on irc :-(
13:04:57 <mclasen> who else is missing - kalev ?
13:06:08 <kalev> .hello kalev
13:06:09 <zodbot> kalev: kalev 'Kalev Lember' <klember@redhat.com>
13:07:00 <mcatanzaro> OK that's a quorum
13:07:25 <mcatanzaro> #chair mcatanzaro rdieter juhp mclasen kalev
13:07:25 <zodbot> Current chairs: juhp kalev mcatanzaro mclasen rdieter
13:07:49 <mclasen> yay
13:07:49 <mcatanzaro> Note that we'll be skipping the Atomic Workstation part of the agenda until Owen returns
13:08:06 <mcatanzaro> #topic F26 Marketing talking points
13:08:34 <mcatanzaro> So we might have to skip this too, unless anybody knows what URL stickster means by "Point out URL and requesting edits (info)"
13:08:46 <mcatanzaro> I guess it's a wiki page that we're requested to fill out.
13:08:57 <mclasen> anybody have a link ?
13:09:35 <mcatanzaro> Well we can brainstorm a few marketing points even without a link. What are the big new features in GNOME this cycle? Besides Recipes. ;)
13:09:52 <mclasen> sure
13:10:00 <mcatanzaro> Not much action happened in GNOME Shell. The notification/calendar area is a bit nicer, but I think that's it.
13:10:05 <mclasen> recipes is not a big one, and not a fedora talking point
13:10:54 <mclasen> the gnome release notes have plenty of things to talk about
13:11:38 <x3mboy> .hello x3mboy
13:11:39 <zodbot> x3mboy: x3mboy 'Eduard Lucena' <eduardlucena@gmail.com>
13:12:09 <x3mboy> Do you mean the F26 Talking Points linK????
13:12:36 <x3mboy> https://fedoraproject.org/wiki/Fedora_26_talking_points
13:13:28 <mclasen> night light! thats the headline feature
13:13:29 <x3mboy> And specific for Workstation: https://fedoraproject.org/wiki/Fedora_26_talking_points#Fedora_Workstation
13:13:38 <mcatanzaro> Yup, that would be it
13:13:45 <mcatanzaro> #link https://fedoraproject.org/wiki/Fedora_26_talking_points#Fedora_Workstation
13:13:47 <mcatanzaro> Thanks
13:14:03 <x3mboy> There are already some point highlighted in workstation
13:14:08 <mcatanzaro> Yeah, there are already several talking points listed, and they look sane to me.
13:14:15 <x3mboy> Cool!
13:14:34 <x3mboy> Thanks for discuss this, it's really important for us in the marketing team
13:14:55 <mcatanzaro> The main problem I see is "Wacom Graphics Improvements" which is a GNOME 3.24 feature that we had backported to F25, so that's not new for us. I'll remove that.
13:15:17 <mcatanzaro> Can anybody can think of anything else to add?
13:17:49 <mcatanzaro> OK well let's move on.
13:17:59 <mcatanzaro> #topic Alpha release: any feedback to consider?
13:19:02 <mcatanzaro> So... I'm not aware of any feedback to consider.
13:19:16 <mcatanzaro> Unless anybody else has received feedback, I guess we can move on to the next point.
13:19:25 <mcatanzaro> Might be a short meeting today. ;)
13:20:38 <mcatanzaro> #topic Plymouth/keyboard layout issue
13:20:44 <mcatanzaro> #link https://bugzilla.redhat.com/show_bug.cgi?id=1405539
13:21:02 <mcatanzaro> Agenda says: "This is a non-trivial fix.  Although now it's not a blocker, has there been any developer discussion of how to address?"
13:22:14 <mcatanzaro> Discussion in the bug amounts to "this is a non-trivial fix, let's not make it a blocker because it's too hard." Which is fair, but it'd be good to see if someone was tasked with addressing this in the long term. mclasen, it looks like that's an action item for you, anything to report?
13:22:32 <mcatanzaro> I know you had halfline look at it already.
13:24:00 * mcatanzaro is beginning to feel alone :)
13:25:37 <mclasen> sorry, distracted
13:25:43 * juhp is still here
13:25:49 * x3mboy still here
13:26:07 * kalev is here too, but doesn't have anything to add.
13:26:13 <mclasen> the keyboard thing is fundamentally a 'needs a better architecture' type problem
13:26:41 <mclasen> you can only put bandaids on top of bandaids so often
13:27:20 <juhp> at least it is a Prioritized bug now
13:27:21 <mclasen> and since all attempts at getting a better archicture in place have petered out, I don't have much hope here
13:28:39 <juhp> in terms of console IO the only workaround I can think of is falling back to text console
13:29:03 <rdieter> given adamw's comment about ascii, it doesn't seem to be *that* bad of a problem
13:30:21 <mcatanzaro> #link https://bugzilla.redhat.com/show_bug.cgi?id=1405539#c29
13:30:29 <mcatanzaro> Ray describes the situation pretty well there
13:30:51 <mcatanzaro> I don't think anybody is expecting miracles here, but it would be good for this to stay on the radar for a long-term fix, mclasen.
13:31:02 <mcatanzaro> OK let's move on, I don't think we're going to accomplish anything more on this today.
13:31:07 <rdieter> agreed
13:31:21 <mclasen> I don't see a long-term fix either, but anyway, moving on
13:31:28 <mcatanzaro> This was actually the last item on stickster's agenda, but I'll pile on a bit more we can move through quickly.
13:31:48 <mclasen> we had one about the workstation spec
13:32:02 <mcatanzaro> mclasen: Well halfline has alluded to a couple different things that can be done... I think he would have to be part of the process. Or another plymouth expert, but I'm not sure if there are other plymouth experts left.
13:32:07 <mcatanzaro> mclasen: Ah yeah, let's do that.
13:32:17 <mcatanzaro> #topic Workstation technical spec review
13:32:49 <mcatanzaro> #link https://fedoraproject.org/wiki/Workstation/Technical_Specification
13:33:39 <mcatanzaro> So the first bug I see here is it says "firefox will be used as the web browser" but maybe we should not discuss that today. :D
13:34:09 <mclasen> I think the main correction we should do here is to move away from a fixed package list in a wiki page - that is bound to be perennially outdated
13:34:09 <mcatanzaro> What mclasen and I wanted to discuss is the package list. There is a big list specifying the exact package set for Workstation, which probably has not been accurate since F21.
13:34:13 <mcatanzaro> Yes.
13:34:30 <mcatanzaro> I think we can simply remove this list. It's not really needed. Any objections?
13:35:04 * kalev agrees.
13:36:22 <mcatanzaro> #agreed Package list to be removed from technical specification.
13:36:47 <mcatanzaro> #topic F26 blocker: goa-daemon not stopped on logout, and gnome-keyring unusable on next log in
13:36:59 * mcatanzaro moving through topics quickly
13:37:04 <juhp> Sounds okay to me...
13:37:07 <mcatanzaro> #link https://bugzilla.redhat.com/show_bug.cgi?id=1340203
13:37:32 <mcatanzaro> This is an accepted F26 blocker that's seen no progress. It's a regression we introduced in F24, so it's been broken for a year already.
13:38:27 <mcatanzaro> The problem is that gnome-keyring and gnome-online-accounts break when you log out of your desktop and log back in, so you have to restart your computer to use them again. I've trained myself to always restart instead of logging out, which is pretty bad.-
13:38:47 <mcatanzaro> E.g. I cannot read any email in Evolution without restarting, because g-o-a doesn't work, because gnome-keyring doesn't work.
13:39:07 <mcatanzaro> So I think we should not leave this one until the last minute, because we do not have any active developers working on gnome-keyring anymore, and it's a fairly tricky component to modify.
13:40:45 <mcatanzaro> mclasen, is it OK if I ask you to ask someone to look into it, so we don't hit June and discover this is still unsolved? It's sort of fallen between the cracks of maintainer responsibility; nobody is caring for gnome-keyring right now.
13:41:02 * mclasen clicks on the link
13:42:36 <rdieter> too bad logind 'KillUserProcesses=yes' is still not a (default) thing
13:43:08 <mclasen> I'll shop the bug around, sure
13:43:19 <mcatanzaro> OK, that's all I was looking for today. Thanks.
13:43:44 <mcatanzaro> Only one point left on my improvised agenda, before open floor
13:43:53 <mcatanzaro> #topic GNOME Software bandwidth issues
13:44:27 <mcatanzaro> Users have been complaining about this for ages, but it came up on reddit recently which brought it to mind. There are basically two problems.
13:45:27 <mcatanzaro> (a) GNOME Software is smart enough to not download updates and metadata in the background on a metered connection, but NetworkManager is not always smart enough to accurately report which connections are metered. There's no user interface in System Settings for marking a connection as metered if NetworkManager guesses wrong.
13:46:32 <mcatanzaro> (b) PackageKit and dnf have separate metadata caches, and the metadata is huge. So instead of downloading a huge amount of data in the background without user intervention just once, we do it twice for no reason.
13:46:48 <juhp> hmmm
13:47:26 <mcatanzaro> kalev, are these points that you have time to look into? I know you have experience with both control-center and PackageKit. ;)
13:47:51 <juhp> Is there an RFE bug?
13:48:05 <mcatanzaro> juhp: I'm sure there are somewhere, but I don't have a link.
13:48:12 <juhp> okay
13:48:35 <mclasen> I've brought it up with the designers, and Allan agrees that we need something there
13:48:39 <mcatanzaro> It's really two separate issues with the same effect: can't use Fedora on metered connection
13:48:56 <juhp> It would be nice to fix (b) anyway
13:49:28 <kalev> I can look into the second issue, but the first one needs to go through design first
13:49:44 <mclasen> perspective-wise, I think packagekit will be less relevant in the future, as we are working on rpm-ostree now
13:49:48 <mcatanzaro> I imagine (a) should be just a matter of adding a checkbox to the settings dialog somewhere, so I'm not sure it needs a ton of design input, but of course it never hurts to ask for a mockup.
13:50:23 <mcatanzaro> Perspective-wise, even if PackageKit is less relevant in five years, I suspect we're going to be using it for a long time yet. :)
13:50:26 <kalev> I've suggested a few times to do a preferences dialog for gnome-software where one could turn off background downloads, but mostly it ends up with people suggesting it should be in system settings instead and then it never goes anywhere
13:50:44 * kalev is still happy to do a preferences dialog for gnome-software that makes this configurable.
13:51:42 <mcatanzaro> kalev: I think that's a valid option too. I don't think Workstation WG needs to decide itself where the preference should go -- control-center or gnome-software -- we just want to make sure the issue is on somebody's radar. Thanks for volunteering to take (b). mclasen, can you or kalev follow up with designers about (a) as well?
13:53:31 <mclasen> thats what I discussed with them already, but sure
13:56:18 <mcatanzaro> OK then.
13:56:22 <mcatanzaro> #topic Open floor
13:57:26 <mcatanzaro> Mark Shuttleworth says Ubuntu is going to ship GNOME "the way the GNOME designers intend it to be shipped." We are going to have some serious competition pretty soon!
13:58:44 <juhp> more competition and collaboration are good :-)
14:00:06 <mcatanzaro> OK, thanks for attending everyone!
14:00:08 <mcatanzaro> #endmeeting