15:04:42 <rdieter> #startmeeting kde-sig
15:04:42 <zodbot> Meeting started Tue Mar  1 15:04:42 2016 UTC.  The chair is rdieter. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:04:42 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
15:04:42 <zodbot> The meeting name has been set to 'kde-sig'
15:04:44 <rdieter> #meetingname kde-sig
15:04:44 <zodbot> The meeting name has been set to 'kde-sig'
15:04:47 <heliocastro> here
15:04:49 <tosky> hi
15:04:51 <than> present
15:04:56 <rdieter> #topic roll call
15:05:01 <dvratil> hola
15:05:15 <rdieter> #info rdieter heliocastro tosky than dvratil present
15:05:19 <rdieter> #chair heliocastro tosky than dvratil
15:05:19 <zodbot> Current chairs: dvratil heliocastro rdieter than tosky
15:07:00 <rdieter> #topic agenda
15:07:01 <rdieter> hi all, what to discuss today?
15:07:40 <rdieter> qt-5.6 status (and related blocker bug(s))
15:07:59 <dvratil> Plasma 5 for Centos - need a volunteer to look into it
15:08:25 <Kevin_Kofler> QupZilla by default, now that it is all packaged and working?
15:08:30 <rdieter> well, baby steps, need to work on kf5 first (for epel I assume)
15:08:40 <Kevin_Kofler> (i.e., instead of Firefox on the KDE Spin)
15:09:10 <rdieter> that makes 3 topics, anything else?
15:09:55 <rdieter> #info Kevin_Kofler present
15:10:00 <Kevin_Kofler> I would also like to formally complain (mainly for the meeting log) about the toxic behavior of heliocastro, accusing me of toxic behavior in a meeting I wasn't even attending.
15:10:19 <rdieter> ok, if you insist, that makes 4
15:10:23 <Kevin_Kofler> (so no chance to defend myself from the accusation)
15:10:27 <rdieter> #topic qt-5.6 status
15:10:50 <rdieter> So, qt-5.6.0-rc landed, but we hit a new, fairly big bug wrt plasma
15:11:06 <rdieter> .bug 1311635
15:11:07 <zodbot> rdieter: Bug 1311635 Unable to create io slave: klauncher said: Error loading /usr/lib64/qt5/plugins/kf5/kio/desktop.so - https://bugzilla.redhat.com/1311635
15:11:13 <rdieter> heliocastro: ^^ that's the one I mentioned
15:11:36 <heliocastro> rdieter: The strange part is that is not in all achines
15:11:40 <heliocastro> *machines
15:11:56 <heliocastro> And i have one with nvidia, prop, and other with intel chipsets
15:12:00 <rdieter> heliocastro: oh?  it's on all mine (though I have only 2)
15:12:01 <heliocastro> all f23\
15:12:24 <rdieter> heliocastro: it's fine on at least one of yours?
15:12:41 <heliocastro> Both is running
15:12:51 <heliocastro> And the nvidia one, even with three screens
15:12:58 <rdieter> interesting, both using plasma ?
15:13:11 <heliocastro> I experienced a single crash on composite konsole when unplugged one screen
15:13:14 <heliocastro> But all plasma
15:13:25 <heliocastro> I just don't have any plasmoids beyonf the taskbar
15:13:44 <rdieter> mine's a default/vanilla plasma config, fwiw
15:14:04 <rdieter> desktop containment fails to load due to the desktop.so plugin not loading
15:14:26 <rdieter> anyway, I'll make a call onlist for more testing/feedback
15:14:39 <rdieter> anyone else successfully using qt-5.6.0-rc/plasma5 ?
15:14:46 <heliocastro> rdieter: Humm, all my machines are using folder view
15:14:48 <heliocastro> not desktop
15:14:59 <rdieter> like I said, mine's whatever is the default
15:15:14 <rdieter> heliocastro: and, I reproduce the error with dolphin, trying to use pretty much any kio slave
15:16:28 <Kevin_Kofler> Have you tried whether a rebuild of the offending kioslave fixes it?
15:16:35 <rdieter> no
15:16:41 <rdieter> wait, yes I did
15:17:05 <rdieter> I rebuilt plasma-workspace (which owns desktop.so)
15:17:17 <rdieter> like I said, other kio's failed too, the same way
15:17:57 <rdieter> I think I even tried rebuilding kf5-kio
15:18:03 * jgrulich is here
15:18:08 <rdieter> #info jgrulich present
15:18:09 <rdieter> hi
15:18:32 <jgrulich> hi
15:18:58 <rdieter> anyway, any debugging assistance would be appreciated
15:19:56 <rdieter> I'd tried settings QT_DEBUG_PLUGINS env, but it didn't give anything too useful, I may have to enable more debugging
15:21:36 <rdieter> anyway, any help would be appreciated, please comment in the bug
15:21:49 <rdieter> moving on...
15:21:54 <rdieter> #topic kf5/plasma5 for centos
15:22:07 <rdieter> anyone able/willing to work on this?
15:22:20 <heliocastro> I want to, but no, no time
15:22:28 <dvratil> same here :)
15:22:28 <heliocastro> I'm focusing in make the ci works
15:22:39 <rdieter> first step would probably be to kf5 (copr initially, eventually in epel-7)
15:22:47 <dvratil> we have a fairly new KF5 there
15:22:49 <dvratil> 5.19 I think
15:22:57 <dvratil> so it's just about getting Plasma 5 built
15:23:06 <rdieter> there's a copr already?  where?
15:23:26 <dvratil> http://copr.fedoraproject.org/coprs/dvratil/epel-playground
15:23:46 <dvratil> it should be fairly simple to get the stuff work with the new distgit feature in Copr
15:24:39 <rdieter> dvratil: wow, it 'just worked' without any changes ?
15:24:51 <dvratil> it was fairly simple, yes
15:24:55 <rdieter> I expected at least a few, nice
15:25:07 <dvratil> mostly I just needed minor fixes in spec files
15:25:23 <dvratil> but since we build for F22, getting it work for EPEL was rather easy
15:25:39 <dvratil> Plasma might be more tricky, I don't know yet
15:25:48 <heliocastro> rdieter: https://heliocastro.fedorapeople.org/Screenshot_20160301_162148.png
15:26:00 <heliocastro> Current packages, dolphin with fish
15:26:03 <dvratil> but as I said, we can use specfiles diurectly from Fedora distgit, we just need someone to do the initial setup and write the tooling probably
15:26:35 <dvratil> and eventually to do the mass-import of KF5 into EPEL
15:26:36 <rdieter> dvratil: neat, I didn't know that was possible
15:26:43 <heliocastro> dvratil: We can read/get from distgit
15:26:50 <heliocastro> But not write directly
15:26:56 <dvratil> we don't need to write
15:27:01 <heliocastro> Yep
15:27:08 <dvratil> we can build from f22 branch of Fedora distgit
15:27:28 * heliocastro is basing the same idea on use distgit
15:27:36 <dvratil> there are even some hooks for automatic build triggers, but I haven't investigated
15:28:24 <dvratil> ok, looks like build triggers only work with github
15:28:46 <rdieter> heck, if kf5 works without changes, any objection to requesting epel-7 kf5 branches asap ?
15:29:09 <dvratil> it *builds*. I haven't tested if they actually work
15:29:14 <rdieter> heh
15:29:23 <rdieter> SHIPIT
15:29:33 <dvratil> and someone needs to review build logs if we have all the optional deps
15:29:40 <rdieter> <nod>
15:30:15 <rdieter> dvratil, heliocastro: either of you willing to post onlist about progress so far, maybe asking for help too?
15:30:44 <heliocastro> Yes, for both cases, epel + ci ?
15:31:02 <dvratil> regarding the EPEL, I can write something after work
15:31:07 <rdieter> sounds like testing in copr may be a reasonable prereq before pushing anything to epel
15:31:28 <rdieter> but I'm ok with doing so asap too, esp if that means more testing/feedback
15:32:06 <rdieter> side-issue: *may* be worth waiting until qt-5.6.0 lands in epel before pushing kf5/plasma5 there too
15:32:16 <dvratil> sounds reasonable
15:33:03 <rdieter> I mispoke , only kf5 for epel, plasma5 can't go there
15:33:16 <rdieter> plasma5 must be copr only
15:33:32 <dvratil> yep, that we discussed previously - setting up a kde-sig/plasma-5 copr for EPEL
15:33:48 <rdieter> ok, sounds like we have a plan, anything else?
15:33:55 <dvratil> nfm
15:34:12 <heliocastro> Done
15:34:22 <rdieter> moving on...
15:34:32 <rdieter> #topic Qupzilla for f24
15:34:37 <rdieter> Kevin_Kofler: go ahead
15:35:28 <Kevin_Kofler> Yeah, so…
15:35:48 <Kevin_Kofler> QtWebEngine 5.6 RC is now in Rawhide and F24 Branched.
15:36:32 <Kevin_Kofler> (GStreamer support is not there and I'm not sure it'll land for F24, it also depends on gstreamer1-plugins-bad-free packaging, among other things. But unencumbered codecs work with the current packaging.)
15:37:01 <rdieter> \o/
15:37:11 <Kevin_Kofler> (with the bundled remnants of FFmpeg, with all encumbered codecs removed)
15:37:48 <Kevin_Kofler> There is also a matching QupZilla 1.9.99 snapshot that should also just work. Last I tested the stuff, it all worked, at least. :-)
15:38:07 <Kevin_Kofler> Other people have also confirmed it working well, being fast, etc.
15:38:23 <rdieter> <nod>, worked fairly well in my own limited testing.  good work
15:38:32 <Kevin_Kofler> So I am now proposing again that the F24 KDE Spin should drop Firefox and include QupZilla as its default browser instead.
15:38:56 <Kevin_Kofler> (When I last proposed it, it was shot down due to not being packaged yet. Now it is.)
15:38:57 <rdieter> I thought we'd already decided to wait a release?
15:39:05 <Kevin_Kofler> See above.
15:39:11 <tosky> no, it was not the reason why I voted no
15:39:32 <Kevin_Kofler> What would waiting a release buy us?
15:39:35 <tosky> the reason still stand: too early, first release not yet released, happy to re-evaluate in F25
15:39:35 <rdieter> though, I would agree to shipping it on the spin by default
15:39:52 <tosky> fine with having it on the spin
15:39:53 <Kevin_Kofler> You didn't wait a release to switch to Firefox during Final Freeze (!!!).
15:39:57 <rdieter> as a tech-preview, backup browser (possibly replacing konqueror)
15:40:03 <tosky> Firefox was not a new software
15:40:20 <rdieter> as mentioned, ff was not new and was already well-tested, well-understood
15:40:24 <Kevin_Kofler> "First release not yet released" is not true, QupZilla has been around for years. :-)
15:40:37 <tosky> it's a rewrite with a new engine
15:40:39 <rdieter> webengine-based qupzilla, *is* quite new
15:40:43 <Kevin_Kofler> It's a port, not a rewrite.
15:40:56 <tosky> same complexity
15:41:54 <Kevin_Kofler> If you insist on shipping both, the Qt browser with Plasma integration should be the default and Firefox the emergency fallback, not the other way round.
15:41:59 <rdieter> Any objections to including qupzilla on kde-sig spin (as a seconary/tertiary browser)?
15:42:03 <lupinix> hi all
15:42:25 <tosky> no objections to including qupzilla on the spin
15:42:32 <Kevin_Kofler> But shipping both means yet another HTML engine on the spin.
15:42:35 <Kevin_Kofler> There are already too many.
15:42:40 <Kevin_Kofler> Gecko needs to go away.
15:43:02 <dvratil> +1 for including QupZilla as a secondary browser (and re-evaluating in subsequent releases)
15:43:03 <rdieter> Kevin_Kofler: other kde/plasma apps will eventually start depending on qtwebengine, its only a matter of time
15:43:21 <Kevin_Kofler> That's exactly why QtWebEngine is not the one that should go away. :-)
15:43:30 <rdieter> Kevin_Kofler: baby steps
15:43:38 <Kevin_Kofler> The KDE Spin is a KDE Spin, not a Mozilla Spin.
15:43:52 <Kevin_Kofler> And Firefox will soon no longer comply to the Free Software Definition.
15:43:58 <rdieter> Kevin_Kofler: you're repeating yourself again, we've been over that many many times
15:44:06 <Kevin_Kofler> See the signed extension fiasco.
15:44:21 <rdieter> and hint: each time you do it, the weight of the argument lessens (to me)
15:45:12 <rdieter> ok, so followup proposal: consider removing konqueror from the spin (since qupzilla is essentially filling the role of backup browser)
15:45:38 <Kevin_Kofler> Konqueror should be the backup for QupZilla (instead of Firefox).
15:46:11 <rdieter> Kevin_Kofler: if you insist, we can take a vote, and then will you promise to stop bringing it up?
15:46:12 <Kevin_Kofler> Then if we're confident enough that QupZilla works for everyone, drop Konqueror too.
15:46:40 <Kevin_Kofler> Your problem is that QupZilla is too new? Well, Konqueror sure isn't. ^^
15:47:12 <tosky> Konqueror lacks seriously a Frameworks version
15:47:20 <tosky> sadly
15:47:40 <Kevin_Kofler> Yes, so Konqueror should be dropped in F25. F24 should have it as a fallback for QupZilla.
15:48:19 <rdieter> ok, let's get this over with
15:48:20 <Kevin_Kofler> And Firefox was a one-time mistake in F23 that should never have happened. (F23 should have shipped with Konqueror, it would just have been one more release to bridge.)
15:48:36 <Kevin_Kofler> The arguments for Firefox are still as irrational as ever.
15:48:50 * rdieter giggles at the irony of that statement
15:49:02 <rdieter> (sorry)
15:50:01 <rdieter> Kevin_Kofler: can you make a formal proposal that we can vote on please?
15:50:22 <Kevin_Kofler> Irony? Are you trying to imply desktop integration, look&feel consistency, freedom (e.g., no signed extension nonsense) are irrational? Do you really think that???
15:50:42 <rdieter> i think it's irrational to repeat yourself , and expect a different outcome
15:51:11 <rdieter> now, do you want a vote or not?
15:51:18 <rdieter> I want to get it over with, so we can move on
15:51:24 <Kevin_Kofler> Proposal: QupZilla 2/1.9.99 shall replace Firefox as the default browser on the Fedora 24 KDE Spin. Konqueror shall remain the fallback for now, dropping it will be considered for Fedora 25.
15:51:32 <Kevin_Kofler> Rationale: as above.
15:51:50 <tosky> two combined votes? Uhm
15:51:58 <tosky> ok, -1 then
15:52:12 <rdieter> well, the primary issue is the first sentence
15:52:22 <rdieter> -1
15:52:26 <lupinix> +1
15:52:28 <Kevin_Kofler> If your issue is the second sentence, then let me offer the alternative:
15:52:32 <jgrulich> -1
15:52:42 <Kevin_Kofler> Proposal B: QupZilla 2/1.9.99 shall replace Firefox as the default browser on the Fedora 24 KDE Spin. Konqueror shall be dropped immediately.
15:52:51 <Kevin_Kofler> (the first one is A)
15:53:03 <rdieter> Kevin_Kofler: pick one
15:53:04 <heliocastro> -1
15:53:10 <jgrulich> A: -1, B: -1
15:53:11 <rdieter> A implies to me that firefox is being removed
15:53:26 <Kevin_Kofler> That's the plan for B too.
15:53:26 <heliocastro> A; -1 B: Abstain
15:53:27 <rdieter> B implies to me that firefix is kept as secondary browser
15:53:43 <tosky> A: -1, B: -1
15:53:46 <rdieter> Kevin_Kofler: oh, so no secondary browser at all?
15:53:54 <Kevin_Kofler> Proposal B (clarified): QupZilla 2/1.9.99 shall replace Firefox (which shall be dropped) as the default browser on the Fedora 24 KDE Spin. Konqueror shall be dropped immediately.
15:54:29 <Kevin_Kofler> Yes, Firefox needs to go away, it's a waste of space, and a landmine for privacy and freedom.
15:54:45 <Kevin_Kofler> And we had 22 releases without a fallback browser.
15:54:51 <rdieter> ok, neither proposal passes as-is, I'll assume you're +1'ing it
15:55:01 <lupinix> from my point anothe reason for dropping ff: quite bad integration in plasma 5
15:55:02 <Kevin_Kofler> I'm not in the voting members anymore.
15:55:09 <rdieter> Kevin_Kofler: feel free to take issue onlist and ask kde-sig'ers to vote
15:55:46 <Kevin_Kofler> I see only lupinix for it, so…
15:56:09 <Kevin_Kofler> Thanks for reminding me again why I left the SIG!
15:56:17 <rdieter> ok, quickly, my proposal was just:  remove konqueror from f24 spin
15:56:27 <Kevin_Kofler> <SARCASM>It also really shows how much my work on packaging QtWebEngine is appreciated.</SARCASM>
15:56:32 <rdieter> (mostly because we now already have 2 browsers)
15:56:58 <rdieter> Kevin_Kofler: I'm sorry if it's a surprise, but we're following the plan we agreed on in prior meetings
15:57:04 <Kevin_Kofler> So you want Firefox/QupZilla?
15:57:18 <Kevin_Kofler> Whereas I proposed QupZilla/Konqueror or just QupZilla.
15:57:25 <rdieter> yes, implies keeping firefox default, qupzilla as backup/tech-preview
15:57:51 <Kevin_Kofler> rdieter: I'm annoyed by the ever-changing excuses for sticking to Firefox.
15:57:54 <rdieter> shipping 3 browsers seems a bit much
15:58:10 <Kevin_Kofler> It's true that tosky had already brought up the "wait one release" idea, but you jumped on that bandwagon now.
15:58:15 <rdieter> Kevin_Kofler: again, I'm sorry if you feel that way, *my* reasoning has always been consistent
15:58:23 <Kevin_Kofler> You said it needs to be packaged before being considered. Now it is packaged.
15:58:37 <tosky> the packaging of QtWebEngine *is* appreciated; the delay of QupZilla has nothing to do with the level of packaging
15:58:46 <tosky> it's not just "it's packaged, it's ready"
15:59:32 <rdieter> any votes/comments on removing konqueror?  (we're running short on time)
15:59:43 <rdieter> I'm torn, but tentative +1
16:00:00 <Kevin_Kofler> And this is also one of the reasons I was so strongly against Firefox in F23, I knew it'd be a fight to get the Spin off that drug. And this proves my point back then.
16:00:04 <jgrulich> +1 for removing konqueror in favor of qupzilla
16:00:13 <tosky> I share the same for removing konqueror, I abstain on that
16:00:41 <dvratil> +1, same as jgrulich
16:01:08 <rdieter> than ?
16:01:51 <rdieter> pretty sure we don't have quorum, I'll take it onlist after
16:02:38 <rdieter> moving on...
16:02:45 <rdieter> #topic toxic behavior
16:03:06 <rdieter> Kevin_Kofler: your topic, go ahead
16:03:29 <Kevin_Kofler> Basically, I already said what I had to say in the introduction: I consider it toxic behavior to accuse me of toxic behavior in absentia, when I am not there to defend myself.
16:03:40 <Kevin_Kofler> (which is exactly what heliocastro did last week).
16:04:03 <Kevin_Kofler> heliocastro: That's what I had to say.
16:04:07 <heliocastro> Sorry, i really commented, and that's why i requested to we implement rules for everyone
16:04:11 <heliocastro> As rdieter took
16:04:17 <rdieter> duly noted
16:04:28 <heliocastro> Not only you, neither me, everyone
16:04:34 <heliocastro> Is going too far.
16:04:50 <heliocastro> I made a mistake few days ago, you direct call me for expulsion
16:04:51 <rdieter> last week we decided to take a harder stance on code of conduct expecatations
16:05:22 <Kevin_Kofler> As for the actual alleged "toxic behavior" at stake, I think we've been through this on the chan enough for the whole week not to rehash it again.
16:05:45 <rdieter> ok, so matter is (essentially) closed ?
16:05:45 <heliocastro> So now we have proper policies, reinforced
16:05:48 <heliocastro> That's ok for me
16:06:27 <rdieter> worksforme too
16:06:29 <rdieter> #topic open discussion
16:06:41 <rdieter> anything else to mention today before we close the meeting?
16:06:52 <dvratil> just got an email about Plasma 5.5.5 tarballs being available for packagers, will get to building tonight
16:06:59 <rdieter> woot
16:07:09 <rdieter> dvratil++
16:07:15 <heliocastro> go go go
16:07:23 <dvratil> Plasma 5.6 beta tarballs will be available tomorrow, I'd like to get that to Copr too, but can't promise when
16:07:41 <heliocastro> Quick question, which git tag should be more appropriate for the packages
16:08:04 <dvratil> upstream tag?
16:08:10 <heliocastro> Meaning 0.<release>.<gitrshort>
16:08:36 <heliocastro> So, the next qt packages will be 5.6.1-0.1.adasda
16:08:46 <heliocastro> Or there is a better approach ?
16:08:53 <rdieter> heliocastro: oh, if they are snapshots, then yes, that's the way to go
16:09:02 <dvratil> I like the explicit "beta", "alpha", "rcX" more
16:09:13 <dvratil> git tag revision does not tell you much
16:09:24 <heliocastro> dvratil: This is for nightly builds
16:09:28 <heliocastro> not the official released
16:09:32 <dvratil> ah!
16:09:36 <rdieter> heliocastro: depends, if there are real 5.6.1-beta git tags to use, then use the tag
16:09:38 <heliocastro> Will be same for kf5 packages
16:09:44 <Kevin_Kofler> There is a clear policy to follow for Release tags for prereleases, which you are expected to know. One minute, let me dig up the link.
16:09:44 <dvratil> sorry, in that case +1
16:09:56 <heliocastro> rdieter: The released ones will keep the beta/rc tags
16:10:08 <rdieter> in the absense of tags, then use snapshot scheme
16:10:09 <heliocastro> The ci ones will be the reference
16:10:21 <Kevin_Kofler> https://fedoraproject.org/wiki/Packaging:NamingGuidelines#Non-Numeric_Version_in_Release
16:10:34 * rdieter has to run to appointment soon, we're over time
16:10:40 <heliocastro> kk
16:10:47 <heliocastro> There's a git tag recomendation already
16:10:51 <heliocastro> So, see you guys
16:10:59 <rdieter> thanks everyone!
16:11:01 <rdieter> #endmeeting