14:00:11 <stickster> #startmeeting Workstation WG
14:00:11 <zodbot> Meeting started Wed Oct 14 14:00:11 2015 UTC.  The chair is stickster. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:00:11 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
14:00:25 * mcatanzaro is present
14:00:30 <stickster> #meetingname workstation
14:00:30 <zodbot> The meeting name has been set to 'workstation'
14:00:34 <stickster> #topic Roll call
14:00:49 <stickster> .hello pfrields
14:00:50 <zodbot> stickster: pfrields 'Paul W. Frields' <stickster@gmail.com>
14:01:05 <stickster> Hey guys, literally as I was typing the meetingname command, power went out here.
14:01:22 <stickster> My router, wifi, and laptop should be good for the time being
14:01:38 <stickster> Oh look, power's back
14:02:37 <stickster> kalev: mclasen__: juhp: rdieter_work: around?
14:02:44 <kalev> I am!
14:03:21 * mclasen__ too, but slightly distracted
14:03:27 <stickster> #chair kalev mcatanzaro juhp rdieter_work kalev mclasen__
14:03:27 <zodbot> Current chairs: juhp kalev mcatanzaro mclasen__ rdieter_work stickster
14:04:24 <rdieter> stickster: hola
14:04:27 <stickster> o/ :-)
14:04:34 <stickster> #chair rdieter
14:04:34 <zodbot> Current chairs: juhp kalev mcatanzaro mclasen__ rdieter rdieter_work stickster
14:04:42 <cschalle> yo
14:04:46 <stickster> #chair cschalle
14:04:46 <zodbot> Current chairs: cschalle juhp kalev mcatanzaro mclasen__ rdieter rdieter_work stickster
14:04:52 * stickster waits 30 more sec'
14:05:56 <stickster> #topic Release notes SVG issue
14:05:58 <stickster> #link https://bugzilla.redhat.com/show_bug.cgi?id=1135114
14:06:31 <stickster> I haven't had a chance to reproduce this issue, couldn't see it on any of my systems but it seems it happens only in odd-sized laptops or VMs. Anyone know more?
14:07:02 * mclasen__ is of the opinion that the real fix for this would be to stop showing documentation like apps
14:07:52 <stickster> Yeah, I'm of the same... release notes seem of dubious help *on the box* after installation
14:08:06 <stickster> as opposed to going to a live site that can be updated continuously
14:08:14 <stickster> but... that's OT for now.
14:08:30 <mcatanzaro> That "app" is not installed by default anymore (for the reason mclasen__ mentioned)
14:08:41 <sgallagh> stickster: I'd prefer to see it as a hybrid thing; an offline-cached web service
14:09:01 <stickster> mcatanzaro: So this is not marked as a blocker for GA then?
14:09:38 <mcatanzaro> stickster: Nope, the bug is even closed (someone says it is still a problem and requested to reopen it, but there's no blocker proposal that I see)
14:09:45 * stickster answers his own dumb question, the bug is still marked CLOSED
14:09:59 <stickster> randomuser: ^ FYI
14:10:21 <stickster> If it's not installed by default, I doubt it would trigger release criteria.
14:10:23 <mcatanzaro> The only blocker I see that's relevant to us is https://bugzilla.redhat.com/show_bug.cgi?id=1269581
14:10:36 <stickster> mcatanzaro: We're getting to that, hold your horses :-D
14:11:04 <mclasen__> assigned to me, no less
14:11:06 <stickster> OK, I guess I would propose we talk to the Docs folks about a better solution, but nothing to do here. +/-1?
14:11:12 <mcatanzaro> (Actually that one is only a proposed blocker to....)
14:11:12 <randomuser> the release notes was not rebuilt since that bug was addressed because of tooling issues, until.. yesterday
14:11:15 <mcatanzaro> *too
14:11:34 <randomuser> also, wasn't it decided not to ship the release notes RPM with workstation?
14:11:47 <stickster> Correct, afaik randomuser
14:12:20 <mcatanzaro> TBH there was some user claiming it is still installed by default, but I cannot reproduce that. It's definitely not installed on my F22 computer or my F23 VM and I didn't remove it.
14:12:28 <randomuser> I wouldn't consider a gnome-specific icon rendering improvement a blocker, then
14:13:09 <mcatanzaro> +1 do nothing
14:13:19 * kalev agrees.
14:13:47 <mclasen__> thats the kind of vote I can get behind, +1
14:13:50 <stickster> ha
14:13:50 <cschalle> +1
14:13:54 <kalev> +1
14:14:17 <randomuser> thanks for bringing it to my attention, though, I will ensure the issue goes away
14:14:53 <stickster> #agreed Doesn't need any immediate action, non-blocker. But if there are issues with release notes presentation, talk with Docs folks about alternate solutions.
14:14:59 <stickster> #topic Blockers for F23 GA
14:15:03 <stickster> #link https://qa.fedoraproject.org/blockerbugs/milestone/23/final/buglist
14:15:28 <stickster> #info Auto login delay -- https://bugzilla.redhat.com/show_bug.cgi?id=1269581
14:15:50 <stickster> Lucky you, mclasen__ -- want to summarize this one and what you're doing to resolve?
14:16:16 <mclasen__> err
14:16:37 <mcatanzaro> I suspect mclasen__ is not currently doing to resolve :)
14:16:47 <rdieter> while annoying, I wouldn't necessarily consider login/shutdown delays a blocker
14:16:51 <mclasen__> it looks like owen started to dive into it
14:16:52 <stickster> ha, maybe not? Should we ping halfline for this? :-)
14:16:54 <kalev> ohh, I think the question is phrazed like that is because it's assigned to mclasen :)
14:17:03 <stickster> kalev: correct
14:17:08 <mclasen__> but no owen atm
14:17:16 <mclasen__> cschalle: did he say he was travelling ?
14:17:34 <mcatanzaro> The problem is cosimo ported gnome-keyring to gdbus, and now $PROBLEMS; this isn't the only one, I think there are huge delays with git clone or git pull or something as well.
14:17:50 <cschalle> mclasen__, yes
14:18:02 <cschalle> mclasen__, he is on the road for a week
14:18:03 <mcatanzaro> $PROBLEMS go away if we revert to gnome-keyring 3.16 but then that breaks GPG, which is very broken in F22 and fixed in new gnome-keyring
14:18:11 <mclasen__> I'll see if I can get some time from cosimoc
14:18:25 <stickster> rdieter: Right, this is just a *proposed* blocker right now
14:18:32 <mcatanzaro> I would argue the git problem is more serious, at least to developers.
14:18:47 <mcatanzaro> And that one isn't a proposed blocker (nor would it be accepted since it's easily fixed post-release)
14:19:14 <stickster> *nod, let's stick to blockers in this discussion since that's what is going to decide go/no-go tomorrow
14:19:30 * stickster ruthlessly enforces agenda... what a jerk
14:19:43 <kalev> https://git.gnome.org/browse/gnome-keyring/commit/?id=3cf744f67939dc23c2cc8715cda999a7ec13f1b6 might fix the git pull issue
14:19:44 <mcatanzaro> The issue is relevant since both bugs are caused by the same patchset :)
14:20:04 * mclasen__ can't promise anything atm, but will work that bug
14:20:10 <kalev> I could try backporting it to the fedora package and asking someone to test if it makes anything better wrt login
14:20:20 <stickster> mclasen__: OK -- the go/no-go meeting AIUI is tomorrow early afternoon EDT. Whether this bug affects it is going to depend on whether it's ruled a real blocker.
14:20:31 <kalev> stickster: I think you are one week off? isn't it next week?
14:20:47 <mcatanzaro> Final release is two weeks from yesterday
14:20:51 * kalev nods.
14:21:01 <mcatanzaro> Usually go-nogo is the week before, not two weeks before
14:21:11 <kalev> yes, we just entered the freeze, definitely not going to ship yet :)
14:21:13 <mclasen__> kalev: that commit at least sounds relevant
14:21:24 <mcatanzaro> I might add that it is rather crazy we are still on schedule so late in the cycle ;)
14:21:32 <mclasen__> if you could produce test packages, that would be fantaastic
14:21:37 <stickster> mcatanzaro: JEEZ SHUT UP SOMEONE WILL HEAR YOU
14:21:38 <rdieter> great mcatanzaro just jinxed it
14:21:38 <kalev> sure, will do
14:21:38 <juhp> mcatanzaro, hehe
14:21:44 <stickster> mcatanzaro: :-D
14:21:53 <mclasen__> it sucks too, because we'll miss 3.18.1 if we stay on schedule...
14:21:53 <stickster> you'd better be knocking on wood right nos!
14:21:54 <mcatanzaro> oops ;)
14:21:55 <stickster> er, now
14:22:08 <stickster> mclasen__: Schedules are the enemy of good!
14:22:12 <jwb> i have a blocker bug right now that isn't going to be fixed.  we're going to have to delay a few weeks
14:22:16 <kalev> so, about 3.18.1: I talked to adamw a few weeks ago and mentioned that we have a schedule conflict
14:22:23 * stickster glares at jwb
14:22:24 <kalev> and he seemed open to freeze exceptioning it in
14:22:33 <kalev> and I also mentioned it at a blocker review meeting
14:22:39 <mcatanzaro> kalev__, mclasen: I wonder if that commit might fix the autologin issue as well as the git problem.
14:22:40 <jwb> odd number releases are development releases anyway, right?
14:22:45 <kalev> sgallagh who usually object wasn't there :)
14:22:53 <kalev> and others seemed open to freeze exceptioning it in too
14:22:57 <stickster> :-\
14:23:04 <mcatanzaro> jwb: 3.17 and 3.19 are development versions, 3.18.1 is a bugfix release.
14:23:09 <mclasen__> might be good to get a keyring 3.18.1 release in any case
14:23:11 <jwb> i meant f23
14:23:18 <mcatanzaro> jwb: It's going to be a massive zero-day update if it doesn't get a freeze exception.
14:23:18 <jwb> and i was being a wiseass
14:23:21 * mclasen__ goes to look for volunteers
14:23:23 <mcatanzaro> jwb: ;)
14:23:24 <sgallagh> kalev: in a meeting. What's the question?
14:23:33 <kalev> nothing, just made a joke :)
14:24:38 <mclasen__> if the keyring commit fixes the delays, that might be out ticket to get 3.18.1 :-)
14:24:46 <mclasen__> 'all or nothing'
14:24:57 <stickster> From a QA standpoint that does not sound awesome
14:25:04 <mclasen__> true
14:25:11 <mclasen__> lets get back to seriousness and to the agenda
14:25:15 <mcatanzaro> stickster: I think he's joking ;)
14:25:23 <stickster> mcatanzaro: partly at least ;-)
14:25:32 <stickster> Anyway, https://bugzilla.redhat.com/show_bug.cgi?id=1269581#c14 seems to be of interest here
14:25:44 <kalev> stickster: yes ... I did try to get Fedora to ship one week later when FESCo was putting together the schedule, but it didn't seem feasible to push out the schedule one week back then
14:25:54 <stickster> mattdm: Is the go/no-go tomorrow for GA, or next week?
14:26:04 <kalev> neither option is good from QA standpoint; freeze exceptioning it in isn't great, and doing a massive 0 day update isn't great either
14:26:09 <kalev> two bad choices :(
14:26:16 <stickster> kalev: yeah, lesser of two weevils here
14:26:52 <mcatanzaro> kalev: I got a texlive update the other day that took multiple hours to apply. I think 3.18.1 would not be that bad, but still bad.
14:27:02 <stickster> OK, back to the issue at hand guys
14:27:05 <mclasen__> we could delay it a bit and do an actual qa'ed f23.1 megaupdate...
14:27:13 <kalev> for f24 / gnome 3.20, gnome hasn't put together a schedule yet but fedora's is out yet
14:27:22 <kalev> I'll try my luck on gnome side this time to make sure schedules sync up
14:27:44 <mclasen__> right, we should get the schedule written, thanks for the reminder
14:27:50 <stickster> mclasen__: So you said you'll talk to cosimoc -- any chance of that today, in case my guess of the go/no-go schedule is right?
14:27:50 <juhp> mcatanzaro, yeah texlive...
14:28:19 <kalev> stickster: it cannot be right, since we don't have an RC yet
14:28:31 <stickster> kalev: Ah right. You pointed that out before, sorry.
14:28:33 <mclasen__> stickster: I've left him some messages (still a bit early on the west coast)
14:29:08 <stickster> mclasen__: OK, hopefully our window's still good. I think we're good to say you're looking into it, we'll certainly get bonked by QA if this goes to blocker for real.
14:29:18 <stickster> Shall we mark this #agreed and move on?
14:29:42 <kalev> sure
14:29:44 <rdieter> yes
14:29:57 <juhp> +1
14:29:58 <stickster> #action mclasen__ consult with cosimoc on keyring issue; notwithstanding the 3.18.1 update this should probably be fixed (one way or the other)
14:30:10 <kalev> also, if anything else fails, it should be possible to revert the gdbus port and keep the other gpg related fixes
14:30:11 <mcatanzaro> +1 everyone except mclasen__ does nothing ;)
14:30:16 <stickster> haha
14:30:20 <stickster> Sorry Matthias :-(
14:30:25 <kalev> hey, I am doing a test build!
14:30:35 <stickster> I'm running this awesome meeting :-)
14:30:41 <mcatanzaro> OK :)
14:31:02 <stickster> #topic F23 Final release announcement
14:31:04 <stickster> #link https://fedoraproject.org/wiki/F23_Final_release_announcement
14:31:19 <stickster> OK, this is pretty simple, thankfully -- we need to put together text for the F23 Final release announcement.
14:31:22 <mcatanzaro> Copy the text from the beta announcement?
14:31:41 <stickster> It should be a bit more rah-rah than our abbreviated Alpha/Beta material
14:31:51 <stickster> But mcatanzaro, you're right that we want to carry the topics in, for sure
14:32:37 <stickster> Each release in the cycle typically gets a more detailed announcement, just as each one gets more publicity/notice
14:33:07 <stickster> mcatanzaro: Do you want to take a whack at this?  I would be happy to take second whack, in that case :-)
14:33:55 <mcatanzaro> stickster: I'm going to try to dodge this one, sorry. :( This week and next are (unusually) super busy for me, it's just bad timing.
14:34:13 <stickster> mcatanzaro: hey, it's no problem. In that case, I volunteer for first whack :-)
14:35:02 <stickster> I'd like to ask all the WG members to stay tuned to the list, and once I have the draft out there, please review it for technical accuracy, and also please add other material you think is missed.
14:35:36 <stickster> #action stickster Write up F23 Final release announcement draft section for Workstation and notify list for review by Friday night EDT
14:35:51 <juhp> ok thanks
14:36:11 <stickster> #topic All other business, open floor
14:36:59 <stickster> Here's one... release note on the Chrome user agent. I completely forgot this, and it was assigned to me
14:37:18 <stickster> randomuser: ^ if you're still around, is there any way to get in some note about this topic, such as for a zero-day relnotes update?
14:38:16 * stickster self.lash(tool='wet_noodle')
14:38:41 <stickster> Oh, there's something on the list now
14:38:56 <stickster> #action stickster follow up with randomuser on docs@ list
14:39:01 <stickster> Anyone else?
14:39:42 <randomuser> stickster, yeah, it can be done; you have commits there, I can do it for you if you'd like
14:40:04 <stickster> randomuser: I'll get with you elsewhere to recall the details :-) I'm a bit rusty
14:40:11 <randomuser> ack
14:41:08 <mclasen__> I've started to do updates to https://fedoraproject.org/wiki/Workstation/Tasklist
14:41:21 <mclasen__> mainly shuffling lots of unfinished stuff of the f23 section
14:41:24 <mclasen__> needsd more work
14:41:46 <stickster> mclasen__: That will be helpful for release announcement too, since I usually look through that list to see if we missed anything major in the marketing blurbs
14:41:47 <mclasen__> I think cschalle may want to drop in a set of new goals for f24 at some point
14:42:28 <stickster> mclasen__: cschalle: It would be cool for us to have that on the list too
14:42:28 <cschalle> yeah, basically just 'codifying' some of the development efforts we have underway
14:43:25 <mclasen__> if you just want to make your draft public at some point, I'll be happy to work that into the tasklist
14:43:46 <cschalle> will do, just need to verify some of the items first
14:43:58 <stickster> #info cschalle bringing F24 goals to the group shortly, mclasen will add to running Tasklist on the wiki
14:44:15 <mclasen__> strip out the secret stuff :-)
14:45:02 * stickster notes there may even be an opportunity for some Cloud WG tie-in coming up
14:45:31 <mclasen__> another open floor item: maybe some of what we've discussed at the gnome summit would be of interest to the wg members who weren't there
14:45:45 <mclasen__> I did post a blog update on sunday, and there's some notes on the wiki
14:45:56 <juhp> aha
14:46:01 <mclasen__> https://blogs.gnome.org/mclasen/2015/10/11/boston-gnome-summit-update/
14:46:22 <mclasen__> https://wiki.gnome.org/Events/Summit/2015#Notes_from_Saturday
14:46:23 <mcatanzaro> I read your blog. Sorry I wasn't there; it was just bad timing for me this year.
14:46:38 <juhp> mclasen__, thanks
14:47:12 <mclasen__> in terms of concrete f24 ties, two things might be worth calling out:
14:47:33 <stickster> The stuff coming with Builder looks super-interesting
14:47:42 <mclasen__> by f24 gnome-builder will probably be complete enough to consider including it on the workstation
14:48:23 <kalev> I think it would fit way better than devassistant that we have right now
14:48:41 <mclasen__> we've looked at remaining wayland gaps - one thing that I'm planning to have jonas work on for f24 is remoting
14:49:19 <mcatanzaro> mclasen__: What about accessibility?
14:49:46 <mclasen__> hasn't been discussed at the summit, but clearly also gaps there
14:50:26 <mclasen__> https://wiki.gnome.org/Events/Summit/2015#Wayland_in_general is the list of annoyances that I got from summit participants
14:50:59 <mclasen__> I'd love to get more on there, feel free to add or send me your 'pet issues'
14:51:10 * stickster should probably switch over to Wayland :-\
14:51:56 <kalev> stickster: if you are willing to file bugs as you find them, definitely :)
14:52:07 <stickster> #info mclasen__ gave a good summary of GNOME summit/hackfest details... see full log for the scoop
14:52:22 <mcatanzaro> "no xsettings interface" <-- seems like a feature not a bug?
14:52:34 <mclasen__> bit of both
14:52:35 <mcatanzaro> "Window placement (always at 0,0 ?)" <-- should be entirely controlled by mutter and not by clients?
14:52:55 <stickster> mcatanzaro: depends if that means an influx of complaints about e.g. font antialiasing?
14:53:14 <mclasen__> on the fonts front, I have good news too
14:53:23 <stickster> I think xsettings is where people currently tweak... but not sure if we're talking about same thing
14:53:27 <kalev> how does the "no xsettings interface" hook into the font discussion from a few days ago?
14:53:34 <stickster> oooh! /me listens...
14:53:38 <mclasen__> we have gotten quite a bit of fixes for cantarell from a new contributor who's experienced in font design
14:53:39 <kalev> is it gtk's default that gets used for hinting on wayland?
14:53:43 <mcatanzaro> stickster: Well people tweak GTK+ settings, but that's implemented via xsettings isn't it? (Not sure?)
14:54:04 <mclasen__> so the 0.17.x cantrarell releases that are in f23/24 should be much better in some details
14:54:34 <mclasen__> I have no illusions that it will stop people from complaining about font rendering
14:54:42 <mclasen__> but still, worth pointing out
14:54:44 <kalev> also, the new contributor is strongly convinced that we should be changing the hinting default on Fedora
14:54:54 <mcatanzaro> kalev: To what?
14:55:10 <kalev> hold on, let me write his own writeup, otherwise I might summarize it wrongly
14:55:15 <mcatanzaro> Any public discussion?
14:55:26 <kalev> https://bugzilla.gnome.org/show_bug.cgi?id=756204
14:56:28 <kalev> and also, http://lists.freedesktop.org/archives/fontconfig/2015-October/005562.html
14:56:35 <mclasen__> I'll state that I disagree somewhat with his ideas
14:57:48 <kalev> for what it's worth, I am not totally convinced that hintslight is actually a better default, I only posted that patch because he seemed to know what he was talking about :)
14:58:46 <stickster> Interesting.
14:58:52 <mcatanzaro> FWIW hintslight would be convenient for WebKit since it would make web fonts look more like system fonts
14:59:09 <mcatanzaro> But that is a loooong topic
14:59:20 <juhp> mmm
14:59:22 <kalev> I tried hintslight locally here and it seemed to make some things better and some things worse
14:59:32 <stickster> #info some discussion of proposal of hintslight as default, mainly happening upstream in GNOME
14:59:38 <mcatanzaro> I think in general it would be worse, but who am I to question the font experts....
14:59:39 * stickster notes we are at top of hour.
14:59:53 <mclasen__> kalev: not too surprising - that is _always_ the result when you make a font configuration change
15:00:07 <stickster> So... I'm going to close this meeting, we can take more font discussion to #fedora-workstation
15:00:16 <stickster> Thanks for coming, everyone!
15:00:18 <stickster> #endmeeting