[03:40] Attention is drawn to LP Bug #1215182 [03:40] Launchpad bug 1215182 in xfce4-session (Ubuntu) "xfce 4 session manager does not save session" [Undecided,Incomplete] https://launchpad.net/bugs/1215182 [04:08] knome, ochosi: ---v [04:08] Before I forget, so some thoughts for the meeting tomorrow. Even if multimonitor support and bypass land this week (which look like they're landing), you still have many failing cards, and not just falling back to X.org, ones that either have corrupt screen ranging from login to completely unusable. Apart from that, there seems to be quite a lot of people still having mouse lag, and of course it still hasn't landed with virtual machine ... [04:08] ... support. Another thought is if we decide to use some Xfce 4.11 applications, we'd be running two experimental package sets which can make it more fun to figure out where something may be breaking. As for the tearing issue, I personally think it'd be nice to have compton in the repos, and ship a config file but don't actually ship compton itself (Not in Saucy). Any questions? [05:55] I'm not going to make the meeting, but i think we should run either mir or 4.11 not both. I'm talking with debian about 4.11 currently. [07:15] Unit193: sounds very sensible to me [07:18] that's my take too - but even if we didn't have both, given what I get with mir I'd not be using it currently [07:18] which is my position for the meeting this afternoon as I'll not be making it [07:22] mhm [07:22] thing is, we're going for xfce4.11 packages to get better multi-monitor support [07:23] crapping that up with a worse general experience kinda makes that move senseless [07:23] yep [07:23] (although the 4.11 packages aren't in yet and mr_pouit hasn't really given us his opinion on the matter) [07:23] (so it's far from sure we'll get them) [07:24] yep [09:39] {Sure, if 4.11 still doesn't land that doesn't change any stability problems, only possible future ones.) [09:46] Unit193, compton being? [09:47] noting -1 from Unit193 and elfy. [09:47] knome: Compositor, seems to be a nice one. [09:47] Unit193, mir compositor? :P [09:48] Heh. [09:48] Thought ochosi tried to get you to use it once, but could be wrong. [09:48] * knome sneezes while has some muesli in the mouth [09:48] oop. [09:48] okay, so non-related [09:48] elfy, what's your opinion on waiting for new things to land? [09:49] it's a replacement for xfwm4's built-in compositor [09:49] oh wait [09:49] he said already [09:49] right [09:49] the main advantage of it being that it uses opengl (optionally) [09:49] i'll promise i'll test it if we decide not to take mir. [09:49] so you don't get any window tearing anymore when watching videos [09:49] so it's kinda mir-unrelated, we can use or not use it with or without mir [09:50] it seemed stable enough, but i assume the performance varies across hardware (as is usually the case, because it's the driver's fault) [09:50] yeah [09:50] xfwm4's compositor uses XRender afaik, which is why there is window-tearing and stuff [09:50] so basically all you have to do after compton is installed is switch off the xfwm4 compositor and start compton [09:51] aha [09:51] it also has some effects builtin, but i wouldn't recommend any of them [09:51] at least the transparency-blur (which might be useful) made my system slow [09:51] does it have at least the same features than xfce's? [09:51] bleh, that's a bugger [09:51] That's why I think in this case it might make sense to ship config but not the compositor itself. Also http://www.mail-archive.com/mir-devel@lists.ubuntu.com/msg00048.html indicates they are following true to what they said. [09:52] it's kind of stupid to ship something that has a feature that's interesting but damn slow [09:52] knome: yeah, it has more features than xfce's compositor [09:52] and xfce's compositor doesn't do blur at all [09:52] so in the config, that would simply be disabled [09:52] same with fade-outs on window close or minimize [09:53] it might also just be that those make the system _feel_ slower... [09:53] "in the config"? [09:53] does that mean it doesn't have a GUI config tool? [09:53] so yeah, there are no regressions over xfwm4's compositor, but potential gains [09:53] because what i meant was: [09:53] if there is a GUI tool to configure [09:53] exactly, it has a config file that lives in .config/compton.conf (iirc) [09:54] aha [09:54] then i don't think it's a good thing [09:54] i mean it shouldn't be too hard to write a config-dialog for compton tbh [09:54] for me, it's a regression in itself that there is no GUI tool to manage that [09:54] sure [09:54] if that's the only blocker [09:54] i'm not questioning that [09:54] well my other point [09:54] if there is a GUI [09:54] and even if feature X would be disabled by default [09:54] but as Unit193 said (and i agree with that), shipping such a config-file by default might still be a good idea to make it easier for users to test it [09:55] if enabling feature X would make the system performance really poor, that sucks [09:55] most definitely [09:55] +1 for that, feel free to do that for 13.10 [09:56] yeah, for users to enable a feature, they'd have to dig in the config-file, so that's almost regression-safe imo :) [09:56] yes, but then again i also think it sucks that we don't have a GUI ;) [09:56] who digs around in config files and flips switches has to live with the consequences [09:56] yeah, i agree [09:56] someone should contact the compton dev and ask him whether there are plans for that [09:56] Unit193: ? ^ [09:57] i agree with poking config files, but i don't want to make them do that [09:57] sorry folks, i gotta run now [09:57] np [09:57] hope i can make it to the meeting though [09:57] have fun ochosi [09:57] well, if you call plumbing fun... [09:57] if you want, and if you're certain, tell your vote now and we'll take into account [09:57] Mir/Xmir: -1 [09:57] okay [09:58] (i don't want to turn our user-base into a herd of guinea-pigs) [09:58] (if anyone asks why -1 :)) [09:58] sure [09:58] i'll point them to the logs [09:58] cause then we can also start shipping xfce in random git snapshots [09:58] bbl [09:59] I have yet to install it, I know it exists and seems to make the most sense. [09:59] Unit193, once you do, can you check performance-wise if some of the features have a big effect [10:06] Well, I wasn't planning on it, but if need be I could. There was a forum post linked in #xubuntu geared towards Xfce (and LXDE) that could be used. [10:07] well if you do it... [10:10] by no means do feel like you're obliged [10:11] i just think you've did great with other similar things [10:11] and i'm sure it could benefit xubuntu [10:11] but if you don't feel like doing it, it's ok [10:14] Heh, I just said that I didn't plan on it, hardware that I typically use isn't the best so chances of being slower are high. I'd be up for it though. (Fastest computer, which I don't use often or little slower one.) [10:15] but isn't that a good benchmark anyway? [10:15] do you use the xfce compositor on those? [10:16] Yep, use it on all of them (slow one too. :P ) [10:17] yeah, then it's a fair comparison [12:53] just for everyone's information before today's meeting: http://www.phoronix.com/scan.php?page=news_item&px=MTQ0MjA (security issue with XMir) [12:54] security issue heh [12:56] phoronix too. [12:56] * smartboyhw will object (although not a Xubuntu person) [12:57] I mean, Mir. [12:57] you can object, but only team members can vote. [12:57] knome, yep [12:57] knome, ochosi https://code.launchpad.net/~robertcarr/mir/client-focus-notifications [12:58] The Mir guys gave me that link to the phonorix bug fix [12:58] *phoronix [12:59] smartboyhw, quite a big change and as it is so crucial, we have given it extra attention and reviews [12:59] tvoss|lunch, :) [13:04] knome: I will watch the meeting as well. Whatever xubuntu does is of interest to studio [13:05] OvenWerks, sure, everybody is free to join the meeting (and speak, only voting on certain issues is limited to team members) :) [13:06] I think Studio will run Xorg till after 14.04 [13:06] * cub is running Ubuntu Studio on xmir as we speak ... [13:07] OvenWerks, that sounds like a valid strategy. [13:07] or at least, one of the valid ones... [13:08] bbl -> [13:20] knome: my opinion is how long do we wait - how long is a piece of string ... [13:22] elfy: I think in this case, even if all the things get fixed that are "about to be fixed", xMir is not ready. [13:24] my understanding is that MIR does not equal xMIR either, so the same deal for 14.04, another new SW block. That is why my thought is: after 14.04 [13:28] knome: now I've read the whole scrollback my vote is still the -1 thanks [13:28] hi OvenWerks, would chat but I get 30 minutes for lunch :) [13:30] elfy: no prob [14:36] meeting is in 15 minutes right? [14:37] GridCube, I thought it's 24:P [14:37] Your time is a bit too fast. [14:37] ? [14:38] its 22 [14:38] https://wiki.ubuntu.com/Xubuntu/Meetings [14:38] maybe Grid lives in on of theose countries that has the time off-set by 15 or 30 minutes :) [14:38] :/ [14:38] ooo [14:38] the minutes [14:38] right [14:38] :P [14:38] my clock might be wrong [14:41] Maybe time moves differently in a quiet library? [14:43] oh, I know what it is...he was looking at the clock on his computer running Xubuntu, and, as everybody knows, Xubuntu doesn't have ntp installed/configured by default :) [14:48] :P [14:49] i had my computer runing xubuntu in a vm :( but they took it to repair in May because its video card got broken, in May [14:50] GridCube, now it's 10 minutes to meeting:P [14:53] hallo. [14:53] * smartboyhw waves at knome [14:53] \o/ [14:55] so i need to be as quick as possible today (though no hard deadlines) [14:55] who's here for the meeting? [14:55] o/ [14:55] o/ [14:56] (as a guest, alright) [14:56] tvoss_ is here it seems:) [14:56] smartboyhw, yup, o/ [14:57] * OvenWerks is an observer [14:57] * smartboyhw too [14:59] #startmeeting Xubuntu community meeting [14:59] Meeting started Thu Aug 22 14:59:22 2013 UTC. The chair is knome. Information about MeetBot at http://wiki.ubuntu.com/meetingology. [14:59] Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired [14:59] #topic Items carried on [14:59] #subtopic Open action items from previous meeting [14:59] #action skellat to prepare blog article discussing updating & upgrading for users and why it is okay to do so [14:59] ACTION: skellat to prepare blog article discussing updating & upgrading for users and why it is okay to do so [14:59] #topic Team updates [14:59] Please use #info and #action [14:59] people, go ahead [15:00] :) [15:00] #info unfortunately, due to various obstacles, the docs SRU is not going to make it to .3, but we will upload it to -updates ASAP (and eventually, it'll be in .4) [15:01] #info Desktop of the week will have a list ready for later today of the first few images proposed for the project, we hope to have them posted an anounced for the first days of the next week [15:01] #action knome will prepare the website side later this week [15:01] ACTION: knome will prepare the website side later this week [15:02] #undo [15:02] Removing item from minutes: [15:02] #action knome will prepare the website side for desktop of the week -gallery later this week [15:02] ACTION: knome will prepare the website side for desktop of the week -gallery later this week [15:02] anything else? [15:02] not from me in this topics [15:02] looking at a busier week next week, because we're one week to FF :) [15:03] indeed [15:03] #topic Announcements [15:03] none from me at this time. other team members? [15:03] oh, i do [15:04] #info knome has received money from the magazine. the project now about 250 euros to spend. [15:04] #undo [15:04] Removing item from minutes: [15:04] #info knome has received money from the linuxidentity magazine articles. the project has now about 250 euros to spend. [15:04] better :P [15:04] \o/ [15:05] #topic New and emerging items [15:05] #subtopic Mir decisions [15:05] is this time for opinions? [15:05] soon [15:05] is there something people want to discuss? [15:06] knome, tvoss and I can give an update [15:06] what's the status updates for Mir, is there any new ground-breaking tests? [15:06] o/ [15:06] olli, sure, that would be good :) [15:06] we were heading for a Call for testing for today around Multimonitor and composite bypass [15:07] sad reality is that we don't have the PPA ready for testing today [15:07] while we hope it's ready by tomorrow, we might even see further delays [15:07] I understand that this negatively impacts your decision making process [15:08] the ultimate plan is to land these features sufficiently tested by FF, i.e. 8/29 [15:08] so we don't have a lot of time to pull the PPA together _and_ have enough headway until FF [15:09] there has been some FUD around the VT security bug [15:10] that is currently being worked on at https://code.launchpad.net/~robertcarr/mir/client-focus-notifications [15:10] and a mandatory target for 13.10 [15:10] internal testing has shown improvements for the bypass feature [15:10] we are about to publish numbers via the long promised & outstanding dashboard [15:11] tvoss_, anything I missed (sorry, am in a real life meeting while I type;) [15:11] olli, nope, all good :) for racarr's branch: it has taken quite a while to review it as it impacts such an important area and is security relevant [15:13] Riddell, o/ [15:13] anybody has questions to olli, tvoss_? [15:14] thanks for the updates [15:15] yup [15:15] tvoss_, when is the ETA of the fix branch? [15:15] I mean, landing [15:16] i think we should vote about waiting or not waiting first [15:17] everybody agree or do we have better ideas? [15:17] maybe talk about it before vote? :) [15:17] people are so quiet... [15:17] hehe [15:17] we need opinions before voting [15:17] sorry reading work emails :P [15:17] smartboyhw, so eta for landing was yesterday :) but one reviewer found small issues [15:17] to take a reasonable desition [15:17] im spectating only [15:17] tvoss_, ah [15:18] so here's the thing, we're still seeing a fair number of failures, where the failover doesn't work and folks end up with blank screens, slow mouse pointers and screen corruption on some cards [15:18] smartboyhw, small issues as in: syntactic mostly. racarr is online in ~1 hour and should tackle those, so there are no major blockers [15:18] even if the multi-monitor stuff landed today, I'm not happy [15:18] ...which is why i personally see a little point in postponing the decision [15:18] mmhm, i had better results on 0.0.8 than 0.0.9 [15:19] 0.0.9 is most certainly better than 0.0.8. I agree with GridCube [15:19] postponing until today was already quite late for us given limitations of our team (we have a pretty small testing team in general, smaller development for fixing bugs) [15:19] smartboyhw, that's not what he said.. [15:19] smartboyhw: he said the opposite [15:19] i had more lag on .9 than in .8 [15:19] knome, pleia2 oh:( [15:19] * smartboyhw says the direct oppostie of GridCube then [15:19] :P [15:20] in any case there are too many result in the fail area [15:20] I don't know XUbuntu well, but is there an anticipated large difference between XUbuntu's DE and Ubuntu's DE [15:20] But would still be possible to add xmir as of today with apt-get install unity-system-compositor? [15:20] from my point of view, while there certainly has been improvements, xmir hasn't been proving to generally improve for everybody [15:20] olli, yes [15:20] cub, yes [15:20] i.e. shouldn't Ubuntu & XUbuntu see feature/bug parity [15:20] olli, partly, but there is a lot that is different [15:21] So the question is more "Should mir be the default already on installation?"? [15:21] I haven't had a good experience of xmir yet [15:21] knome, I think olli is referring to the underlying graphics stack [15:21] cub, ...no :) [15:21] cub: right, that's the question [15:21] knome: oh? [15:21] oh? [15:21] cub: the question would be if it will be shipped default [15:21] that ^ [15:21] it will be in the repos regardless, is I think what cub is saying [15:22] then it should be "for" installation :) [15:22] tvoss_, olli: yes, sure [15:22] exactly. But the people who would like to test it could just add it themselves. Everybody happy. [15:22] we could change the ubiquity installer to ask if xmir or x.org? [15:22] but the reason why we are here that the Canonical decision to indlude mir or not is a different from the Xubuntu decision to include (X)Mir [15:23] Only if it includes an expailation [15:23] not only because the other uses Mir and the other XMir [15:23] regarding QA resources & test coverage... would it help if we (upstream) would contribute to your testing [15:23] GridCube, not happy with that either. [15:24] knome, I rather think an explanation on the website or the release notes might help. [15:24] GridCube: that would mean shipping xmir & x.org packages on the cd, and removing the one not needed at the end of installation (similar is done with unneeded file-system utilities et al) [15:24] GridCube, if XMir is ready and stable enough to be chosen from the installer, then it should probably be default. if not, then it shouldn't be installable at all [15:24] olli: so we've put a lot of effort into testing xmir specifically, we really need to get back on track for more general Xubuntu testing [15:24] ofcourse it would help, but our users regard xubuntu as a pretty stable release, they use us on many places where they want to have a simple-stable system, not providing that would scare many of our users [15:24] knome: good point [15:24] (installable from the CD) [15:25] we could make sure a wiki page is maintained for folks who do want to enable it (I don't love installer option either) [15:25] At this moment, just at this moment, xmir its not stable enough for xubuntu, that is my opinion [15:25] pleia2, you can put it in release notes [15:25] its pretty good though [15:25] Or website [15:26] so... [15:27] vote time [15:27] do people think we should bother to vote about waiting or not waiting [15:27] * smartboyhw doesn't [15:27] * olli does [15:27] vote [15:27] i think we should do that. [15:27] It's one week BEFORE ff now [15:27] sorry, that was selfish ;) [15:27] * tvoss_ does, too :) [15:27] olli: hehe :) [15:27] olli, np, that's expected ;) [15:27] #vote Should we wait for improvements landing to XMir later (but before the FF)? (+1 for wait, -1 for not wait) [15:27] Please vote on: Should we wait for improvements landing to XMir later (but before the FF)? (+1 for wait, -1 for not wait) [15:27] Public votes can be registered by saying +1, +0 or -1 in channel, (private votes don't work yet, but when they do it will be by messaging the channel followed by +1/-1/+0 to me) [15:27] #voters bluesabre elfy GridCube jjfrv8 knome micahg mr_pouit ochosi pleia2 skellat Unit193 [15:27] Current voters: GridCube Unit193 bluesabre elfy jjfrv8 knome micahg mr_pouit ochosi pleia2 skellat [15:27] -0 [15:28] bah, no -0? [15:28] +0 [15:28] +0 received from knome [15:28] -1 [15:28] -1 received from pleia2 [15:28] -1 [15:28] -1 received from GridCube [15:28] -1 [15:28] -1 received from jjfrv8 [15:28] Well... [15:29] there was one -1 from elfy and one -1 from Unit193. [15:29] (covering this questions as well) [15:29] -6... [15:29] people who want to check that the votes are valid can check the logs from this channel earlier today. [15:30] let's wait for a few more minutes. [15:31] -1 [15:31] -1 received from ochosi [15:31] * smartboyhw thinks -7 is a majority already... [15:31] smartboyhw, please. [15:31] #endvote [15:31] Voting ended on: Should we wait for improvements landing to XMir later (but before the FF)? (+1 for wait, -1 for not wait) [15:31] Votes for:0 Votes against:4 Abstentions:1 [15:31] Motion denied [15:31] #vote Include (+1) or not include (-1) Mir in Xubuntu 13.10? [15:31] Please vote on: Include (+1) or not include (-1) Mir in Xubuntu 13.10? [15:31] Public votes can be registered by saying +1, +0 or -1 in channel, (private votes don't work yet, but when they do it will be by messaging the channel followed by +1/-1/+0 to me) [15:31] #voters bluesabre elfy GridCube jjfrv8 knome micahg mr_pouit ochosi pleia2 skellat Unit193 [15:31] Current voters: GridCube Unit193 bluesabre elfy jjfrv8 knome micahg mr_pouit ochosi pleia2 skellat [15:31] -1 [15:31] -1 received from pleia2 [15:32] -1 [15:32] -1 received from knome [15:32] -1 [15:32] -1 received from GridCube [15:32] -1 [15:32] -1 received from ochosi [15:32] -1 [15:32] -1 received from jjfrv8 [15:32] there was one -1 from elfy and one -1 from Unit193 to this as well. [15:33] again, you can check the logs [15:33] #endvote [15:33] Voting ended on: Include (+1) or not include (-1) Mir in Xubuntu 13.10? [15:33] Votes for:0 Votes against:5 Abstentions:0 [15:33] Motion denied [15:33] tvoss_, olli: thanks [15:33] efforts are appreciated :) [15:34] its not ready just yet [15:34] so, while this is not the resolution we were hoping for (but we understand & appreciate your efforts) I really wanted to say thanks to the team for giving it a fair chance [15:34] indeed! it's been fun working with the Mir team, and helping folks get the hands in testing at Ubuntu Hours and such [15:34] rather than making uneducated decisions, this is worth a lot to us [15:34] i guess ppl will continue testing [15:34] hope mir goes well for ubuntu :) [15:34] tvoss_, olli: i'm sure people can test mir with xubuntu in the following days as well, to see if things have been improving and bugs fixed :) [15:34] olli: :) and thank you for your presence it was greatly appreciated [15:35] yeah, and thanks or all your testing efforts :) [15:35] I hope that by 14.04 the vote will be in the different direction ;) [15:35] tvoss_, olli: if you don't mind, we won't mind you being on the channel available for possible questions and discussion in the future [15:35] +1 [15:35] knome, I will try to idle around [15:35] knome, I will be around [15:35] but you guys know, that #ubuntu-mir is just around the /j corner ;) [15:35] tvoss_, olli: and finally, thanks again for working with us and really taking us into account! [15:36] anybody else with closing thoughts? :) [15:37] I think that what we all think its that is great to have the developers here, trying to help us to make the best desition [15:37] definitely [15:37] and that we hope that if things keep this way in the future we can rethink our position for future releases [15:38] #subtopic Proposal for more structured handling of Xubuntu bugs [15:38] skellat is not around, so passing [15:38] #subtopic Inclusion of Xfce 4.11 components in Xubuntu 13.10? [15:39] since developers aren't around, we can't make final decisions... however, [15:40] ochosi, would you like to list components that we should consider, since you have been overlooking the xfce development lately? [15:40] well the most important part of xfce4.11 in my personal opinion is the improved multi-monitor support [15:41] that is part of xfce4-settings [15:41] bluesabre, mr_pouit and me were working on that [15:41] xfdesktop also bears quite some improvements [15:41] i assume that would bring the much discussed display-dialog in as well [15:41] yes, xfce4-settings == display dialog (amongst others) [15:42] and i think some of the keyboard-shortcut issues (primary vs. ctrl) were solved recently [15:42] for the display dialog we should at least (if we don't include it by default) point ppl to the 4.12 dev ppa [15:42] what's in with xfdesktop more specifically? [15:43] a revamped settings-dialog, nicer workspaces support [15:43] #info xfce4-settings to bring in improved multi-monitor support along with a nice display dialog [15:43] the settings-dialog doesn't just show a fixed list, but the content of a folder [15:44] i've also been involved in that a bit [15:44] #info xfdekstop to bring improvements in settings dialog and better workspace support [15:44] you can also set different wallpapers per workspace [15:44] which is a feature many wanted for a long time [15:45] what about the panel indicators patch? [15:45] i know that's not exactly 4.11, but is something that relates to that [15:45] i've been sitting behind a really slow mobile internet connection, so i wasn't able to put the pieces together unfortunately [15:46] okay, but would you think that would be something we wanted to consider? [15:46] "that" being a patch that allowed running gtk3 indicators in a gtk2 panel [15:46] well there are two options: [15:46] which would mean we wouldn't need to maintain the gtk2 indicators, and we'd get back some lost features [15:46] 1) use patched gtk2 panel with gtk3 indicators [15:47] 2) use gtk3 panel with gtk2 xfce plugins and gtk3 indicators [15:47] tbh i've only tested the latter [15:47] that was stable though, i even have theming for it in a separate branch of greybird [15:48] i've been trying to get a hold of andrzejr, the author of the patches and maintainer of the indicator-plugin for a few days now [15:48] do you think you would have some time and a decent enough connection to work on that before the FF? [15:48] #info Several options to enable GTK3 indicators for the panel [15:48] i'll have a bad connection until the end of august, then i'll be moving to italy, so possibly no regular connection at all for the first weeks [15:49] at least not in a way that i can promise [15:49] okay [15:49] let's postpone that for 14.04 then [15:49] yeah, if no-one else can test it [15:49]