[00:00] chrisccoulson: What are you doing up :) Isn't it like 2am Sunday or something? [00:00] * RAOF pulls from bzr. [00:02] heh, it's not quite that late for me just yet ;) [00:02] it's midnight here now [00:03] i'm all messed up tonight because of daylight saving [00:06] * bryceh waves [00:07] bryceh: Good morning. [00:07] / evening. [00:11] At least unity gets respawned after crashing now. [00:13] hey RAOF [00:14] RAOF, by chance have you poked through fedora's patches to see if they have anything interesting we're missing? (guessing there isn't anything major) [00:16] bryceh: I haven't, but I'd be willing to go digging today5. [00:17] RAOF, great, yeah [00:23] RAOF, also dunno if you saw the discussion but sarvatt noticed friday that ia32-libs has an old snapshot of mesa in it [00:23] and he's pinpointed that at least some of the sandybridge gpu lockups on amd64 are caused by that [00:23] bryceh: That's probably because yokozar hasn't refreshed it yet. [00:23] so now I'm wondering what other reported bugs might be due to it [00:23] Hm. Hurray for Oneiric and the obliteration of ia32-libs! [00:23] indeed [00:36] Oh, man. If that ends up being the reason why this always crashes… [00:37] potentially I could imagine same issue is at heart of some of the screen corruption issues [00:46] Protip: C++ does not implicitly initialise member variables to 0. If you're going to use null as a guard value before calling into a pointer, initialising it to NULL in the constructor makes it less likely to crash by calling off to a random memory address :/ [00:47] yep [00:47] C doesn't do that either afaik? [00:47] No, it does not. [00:47] Well, except for static variables. [01:05] RAOF, aha it's due to missing support for >4096x4096 render targets on arrandale [01:05] Sarvatt, ^^ [01:06] Who'se rendering to >4096x4096 targets? (Apart from my LCD's now broken EDID suggesting it's that big) [01:06] RAOF, https://bugs.freedesktop.org/show_bug.cgi?id=35682 [01:06] Freedesktop bug 35682 in Drivers/DRI/i965 "[Arrandale] support for >4096x4096 render targets" [Major,New] [01:08] Modes: 3840x1080… Why don't I have one of those? :) [01:13] that screen must be quite large, then twice the 1080p res in width but usual height [01:37] robert_ancell, TheMuso, RAOF, jasoncwarner, happy Monday [01:37] rickspencer3, Happy Monday to you too! Welcome to the future! [01:37] aah [01:37] I'm in the airport waiting to fly to London, so I am really in the land of no time [01:38] robert_ancell, how's natty looking for you? [01:40] rickspencer3, on and off unstable, I haven't tried the Beta release yet. I'm going to do a fresh install when we get the cd image [01:40] hmmm [01:40] robert_ancell, it's really stable for me, and even the dash is working really well [01:40] rickspencer3, are you going to do an update of you spiderweb (?) diagram [01:40] robert_ancell, yeah [01:41] it's actually kinda time consuming, and tbh, I didn't really realize anyone looked at it ;) [01:41] but that's kinda why I'm asking, except for accessibility, Natty is working well for me on both 'puteres [01:41] I get alt-tab crashers, and have had general problems. I'm using radeon, which sounds like it has the most problems. The dash has always been slow for me, and I don't really find it useful [01:41] I love the panel and the fullscreening though [01:42] robert_ancell, so, now I can do Meta key, type a few letters, hit enter [01:42] apps launch really fast this way [01:42] you're not having a similar experience, it sounds like [01:43] rickspencer3, yeah, but I'm not sure if that's useful for standard users. I feel like it's really hard to work out what applications are available [01:43] robert_ancell, interesting [01:43] because there are no categories? [01:43] yes [01:43] someone should make a categories place [01:43] I think someone did, I'm sure I saw one on OMG [01:44] interesting [01:44] but yeah, I guess you only get a list of all your apps [01:44] interestingly, I had the same experience when first using OSX, I wonder if they consider it to be a bad thing [01:45] Also, the three types of find apps on the dash seems very confusing. I'd really only expect one... [01:45] huh [01:45] 3 types of apps? [01:46] I get "Find Media Apps", "Find Internet Apps" and "Find More Apps" [01:46] oh, right [01:46] well, you can choose "applications" from the lanucher [01:46] and you get mfu, installed, available for install [01:46] And I just think, Uh, I just want some apps, I don't care where they come from! [01:47] aha! [01:47] you can filter by type! [01:47] it's hard to find [01:48] if you click on "find more app" [01:48] in the upper right you can filter by category [01:49] robert_ancell, we should get your alt-tab crashers fixed [01:49] rickspencer3, I haven't restarted since dist-upgrading, It may have gone :) [01:49] that would be nice [01:49] It tends to toggle between upgrades [01:50] well, there shouldn't be any new features, so there shouldn't be any more toggling for the rest of the cycle === asac_ is now known as asac [05:58] sb levelclear -level clientcrap,crap,joins,parts,quits,nicks,clientnotice [07:38] good morning everyone [07:44] heya chrisccoulson :-) [07:45] hi bryceh, how are you? [07:45] chrisccoulson, good, just facebooking about this weekend's woodworking project [07:47] nice! it seems like you had a productive weekend :) [07:50] chrisccoulson, yep and had an old college friend and his family over, had fun and our kids had lots of time to play and wear each other out [07:50] Good morning [07:53] RAOF: unity crashes about 3 or 4 times at start with various stack traces (reporting my latest one right now), after that it's pretty stable though [07:55] pitti: Well, I hunted down the cause of why unity would crash when opening the dash or right-clicking on any of the launchers and proposed a merge. Unity hasn't crashed since. [07:55] I'll shout at DBO until the trivially correct fix is merged :) [07:56] pitti, hey, has the beta CD been done? [07:56] hey robert_ancell [07:56] TheMuso: Incidentally, you're running out of time to endorse me for core-dev; the next DMB meeting is 19:00UTC (or, as I calculate it, 6am tomorrow ☺) [07:57] :) [07:57] robert_ancell: it's due on Thursday, we don't have test candidates yet [07:57] RAOF, \o/ [07:57] pitti, so, should I upload GNOME releases (e.g. GTK3) or wait until after the CD? [07:57] RAOF: nice! my crash of the day is bug 744063, perhaps the trace sounds familiar [07:58] robert_ancell: gtk3 is fine, as it's not on the CDs [07:59] pitti, and leaf applications like gcalctool? [08:00] robert_ancell: we might just keep them in the queue until Thursday [08:00] pitti: Nah, that's not the crash I fixed. I've seen it, though. [08:00] robert_ancell: but you can upload them [08:01] pitti, and ok to upload thing like glib-networking/libsoup as well? [08:01] * robert_ancell pushes his luck [08:03] RAOF: ooo right, keeps slipping my mind. Will do it this evening. [08:04] It's ok if you miss it; I've got a bunch of other endorsements. [08:10] Hrm. Is it just me, or does pressing control + alt + shift maximize windows for others as well? Using compiz. [08:11] Nope, not for me. [08:15] TheMuso: not here [08:15] Hrm ok, I'll have to reset my compiz settings and see if that helps. [08:17] good morning [08:21] bonjour didrocks [08:21] Guten Morgen pitti [08:22] 5~/c [08:28] Morning all! [08:30] hey Sweetshark [08:30] Sweetshark: wie gehts? [08:32] pitti: Alles gut, und selbst? [08:32] Sweetshark: prima, danke [08:33] Sweetshark: we'll need to fix some transitional packages in openoffice.org-dictionaries [08:33] Sweetshark: http://people.canonical.com/~ubuntu-archive/testing/natty_probs.html, scroll all the way down [08:34] Sweetshark: e. g. openoffice.org-hyphenation-ro depends: hyphen-ro, but hyphen-ro Conflicts: openoffice.org-hyphenation-ro [08:34] Sweetshark: we need versioned replaces/conflicts here [08:34] Sweetshark: i. e. Conflicts:/Replaces: openoffice.org-hyphenation-ro (<< 1:3.3.0-2ubuntu1) [08:34] Sweetshark: hyphen-ro should conflict/replace the older packages which still actually had data/files in it, but it shouldn't conflict to the transitional package [08:35] Sweetshark: do you have time to fix this, or shall I look into this? [08:35] *grumpf* I had no idea that blogging causes that much comment spam, what a mess [08:35] Sweetshark: I guess this affects Debian as well [08:35] mvo: akismet FTW [08:35] mvo: it's really good [08:35] (and yes, spammers suck [08:36] great, I have a look [08:36] mvo: it's a wordpress plugin [08:36] I use wordpress.com directly, i.e. not my own instance [08:37] mvo: hm, AFAIR they have plugins as well [08:37] aha, cool. it appears the comment section of the post is actually sane, its just that I have a gazillion mails about comment spam in my inbox :) [08:40] micahg, chrisccoulson: hm, why do we still build abrowser-branding and firefox-branding? [08:40] oh, transitionals [08:40] mvo: there are currently a lot of transitional packages which want to go to universe; I take it we should seed them and keep them in main until the next LTS, right? [08:41] for OOo? yeah, that would be good [08:41] mvo: in general [08:42] mvo: ok, I'll walk through http://people.canonical.com/~ubuntu-archive/component-mismatches.txt and seed the transitional bits [08:46] thanks! [09:02] hi pitti [09:02] yeah, we need to keep those transitional packages until after the next LTS, to ensure users who used to have abrowser end up with firefox [09:02] chrisccoulson: right, noticed a second after saying.. :) [09:07] hmmm, i think a session restart is needed already [09:07] after compiz crashed about 4 times in half a minute, the bfb and launcher no longer works [09:14] pitti: k [09:27] morning [09:29] hey rodrigo_ [09:29] hi pitti [09:38] hello there [09:38] pitti, did you restart the retracers? [09:38] seb128: yes, and fixed the chroots [09:38] ok thanks [09:38] I was just looking at them because it seemed they didn't retrace for a while but they seem to be running [09:39] pitti, how are you? ;-) [09:39] seb128: I'm quite fine, thanks! currently sorting out natty archive problems [09:39] seb128: how about you? [09:39] pitti, I'm fine thanks, currently catching up with the weekend emails [09:39] ;-) [09:40] seb128: spent some hours last night with my mother's ipod [09:40] seb128: this currently works rather poorly :/ [09:40] indeed [09:40] I don't see it in RB, and nautilus keeps hanging [09:40] what generation, type of ipod is that? [09:40] the afc:// mounts work [09:40] seb128: ipod touch, how do I tell the generation? [09:41] no easy way, the format of what they write behind changed, but easier is "when did you buy it" [09:41] seb128: she got it from her company for christmas [09:41] ok, so likely the most recent one [09:42] "Model: PC544FD" [09:42] libgpod didn't get any update this cycke and doesn't seem actively maintained [09:42] libimobiledevice etc are also not getting lot of work [09:42] I would not be surprised if they didn't get what is needed for recent ones [09:42] seb128: I get a crash of the udev prober, I'll look at that one at least [09:43] ok [09:43] it can't identify the ipod as such because of that, this might also be the reason why rb doesn't see it [09:44] could be indeed [09:56] seb128: is libgpod obsolete in favor of imobiledev? [09:56] pitti, no, libgpod is used for the ipod who don't use the afc protocol [09:57] which is all the "old" ones, classic, nanos, shuffle [09:57] seb128: ah, so this shouldn't even be necessary [09:57] (for rb) [09:57] right, I think it's not for yours [09:58] hm, millions of "Syncdaemon service died" [09:59] and indeed this thing keeps restarting [10:01] seb128: fun, after I removed ubuntuone packages, it suddenly works now [10:01] urg, weird indeed [10:03] E: ID_MEDIA_PLAYER=apple_video-ipod [10:03] E: USBMUX_SUPPORTED=1 [10:03] good 'nuff [10:12] pitti, is the retracer dead again? [10:12] fta2: it was over the weekend, now catching up [10:13] ok === ogra_ac is now known as ogra_ [10:36] seb128: what's the current situation with cairo? can I help you with this (sorting out the ldconfig stuff?) [10:37] pitti, I've the debdiff from the other day but I need to sit down and think about what to do with the dev [10:37] pitti, especially with the .pc paths etc [10:37] seb128: it can't be a differently named .pc? [10:37] it's easy to drop a lib to use at runtime but I was not sure what to do for builds [10:38] pitti, what do you mean? so far I started with a "install the .so in a different dir and the -gl.pc in the dev" [10:38] but I'm not sure it will work [10:38] seb128: ah, -gl.pc [10:38] since applications might use cairo.pc and cairo-gl.pc [10:38] seb128: that would require changing the wayland packge to use cairo-gl.pc instead of cairo.pc, right? [10:38] but cairo.gl being the libcairo2-dev one it will not point to the right dir [10:39] oh, there is already a cairo-gl.pc [10:40] seb128: woudl that actually matter? for apps which don't use cairo-gl, they can use libcairo2 as well? [10:41] pitti, well the wayland configure.ac check for "cairo" and "cairo-gl" [10:41] pitti, I'm concerned that builds will not work if they try to resolve symbols from the cairo-gl build [10:41] since the cairo.pc will point to /usr/lib [10:41] not /usr/lib/cairo-gl [10:42] and the .so is a symlink, it will not be changed by the ld hack [10:43] I was pondering now do a libcairo2-gl-dev with c,r,p libcairo2-dev and ship all the files rather [10:43] seb128: I think in this case libcairo2-gl-dev should ship all .pcs of libcairo-dev, and point to /usr/lib/cairo-gl/ [10:43] that combined with using --libdir=/usr/lib/cairo-gl [10:43] wayland has versioned build-depends, so we'll need some diversions there [10:43] (instead of just a provides) [10:44] hm, no, we could just change the build dep, we'll need that anyway [10:44] well [10:44] so a mere conflicts: should suffice? [10:44] right, just conflicts,replace then [10:44] then update wayland to use that one [10:44] *nod* [10:45] it just starts filling a stack of hacks and that is would be much easier to just dump what we have in a ppa for natty [11:01] vuntz, ping [11:15] rodrigo_, not sure but he might be travelling to gnome asia or already there [11:16] seb128, ah, right [11:16] rodrigo_, you better context ping him so he can reply when he's around [11:16] seb128, I'll send him a mail then, easier to reach each other via mail [11:16] ok [11:16] dpm, hi [11:16] dpm, you indicator-datetime label being cut, does it happen under unity 3d? [11:19] heya seb128, on unity-3d it's fine, the problem is only on the classic desktop (no effects) session [11:20] dpm, oh ok, that might explain why most people don't get it [11:20] dpm, thanks [11:22] yeah, I'd run unity, but I cannot do it on my main pc due to bug 726496 [11:22] Launchpad bug 726496 in nvidia-graphics-drivers "Cannot use Unity or Classic desktop with effects after the latest nvidia+xorg update" [Undecided,Incomplete] https://launchpad.net/bugs/726496 [11:24] Sweetshark: just for coordination, are you working on the oo.o transitional conflicts, or want me to? [11:25] dpm, no worry, it's just that the bug description didn't make it clear it was specific to 2d [11:39] pitti: I had a look at it, but havent really gotten to the root cause. [11:42] pitti: hmm, openoffice.org-hyphenation-af depends on hyphen-af 3.3.0-7ubuntu1, but we only have 3.3.0-2ubuntu1 [11:43] Sweetshark: hm, not here [11:43] Package: openoffice.org-hyphenation-af [11:43] Depends: hyphen-af [11:43] i. e. it's unversioned [11:43] (as it should be) === MacSlow is now known as MacSlow|lunch [12:07] rodrigo_: pong [12:07] hey vuntz, already in gnome.asia? [12:11] vuntz, just wanted to ping you about https://bugzilla.gnome.org/show_bug.cgi?id=634988 [12:11] Gnome bug 634988 in general "my computer is too fast" [Normal,New] [12:12] vuntz, I've tried several things, even adding a sleep to gsd to force the problem in my not so fast computer, but I never able to replicate [12:12] vuntz, so, any suggestion as to where to try next? [12:14] I could check with desrt [12:14] since I'm with him [12:14] but he's sleeping on the floor right now [12:15] desrt: ping me when you wake up :-) [12:16] :D [12:29] pitti: I really dont get it. I have hyphen-af installed, but cant install openoffice.org-hyphenation-af because "Depends: hyphen-af but is not going to be installed" [12:35] pitti: also natty_probs.html claims openoffice.org to be not installable, but it does here quite fine (and only depends on libreoffice anyway). [12:41] vuntz: my left hand is asleep [12:41] ow [12:42] vuntz: ping! [12:43] hey desrt [12:43] seb128: what's up? [12:43] desrt, how are you? [12:43] my left hand is asleep [12:43] desrt, vuntz, just wanted to ping you about https://bugzilla.gnome.org/show_bug.cgi?id=634988 [12:43] Gnome bug 634988 in general "my computer is too fast" [Normal,New] [12:43] it's vuntz's fault [12:43] desrt, I think they wanted to talk to you about that one [12:43] ah. coool. [12:44] desrt, yes, it is, we just ping you so that you force him to fix it :-) [12:44] new computers are not getting slower so extra users run into it ;-) [12:44] desrt, vuntz wanted to do a quick test on your machine [12:44] it's my computer at home [12:44] and i have only my laptop with me [12:44] you should have brought it here [12:45] it was a bit large for my carry-on [12:45] and i don't trust the bag-throwers [12:45] rodrigo_: so yeah, no good suggestion from me :/ [12:46] vuntz, ok, I'll keep trying other stuff [12:48] rodrigo_: does gdm wait for gnome-session to exit before starting the user session? [12:50] vuntz, seems so [12:50] vuntz, also, I've added a sleep(10) to g-s-d to try to force the issue, but g-session seems to wait for it to finish [12:52] if you think about it, just about any way of gdm telling its g-s-d to quit is going to be asynchronous with respect to the X server [12:52] even kill -9... [12:52] because X doesn't necessarily notice right away that the process is dead [12:52] there are basically two things we can do: [12:53] 1) have some way to kill g-s-d such that we get a notice that X has definitely recognised that the XSettings provider has been unregistered [12:53] or [12:53] 2) we take the code how it is now and if the registration of the Xsettings manager fails then we pause for a short time and try again [12:53] #1 is more correct, but #2 has the advantage that it doesn't introduce additional serialisation where we don't have it right now (thus slowing down login) [12:54] desrt, I'm writing a patch to test #2 [12:54] #2 is definitely easiest :) [12:54] yeah [12:55] in my testing, 1 second of extra sleep is always enough to completely get rid of the bug [12:55] but probably less is required [12:56] so sleep for 100ms is probably good if you try for 20 times or something [12:56] desrt, I guess we can sleep for 1/2 a second, and do 2 or 3 tries [12:56] yes, also [12:58] hey what's the best way to fix a string in a patch?. Could someone comment on bug 743928 and explain the submitter how to do it? He usually creates branches to upstream projects to fix them, but I think in this case he might have not done it because he might not have known how this works for fixes to patches in packages [12:58] Launchpad bug 743928 in banshee "Ubuntu patches add untranslated strings" [Undecided,New] https://launchpad.net/bugs/743928 [12:59] rodrigo_: there may be a 3rd option [12:59] rodrigo_: depends if we can implement some sort of --replace functionality for XSettings [13:00] i'm not sure how the spec works or if it would allow it [13:00] desrt, hmm, I'll have a look [13:02] rodrigo_: i'm looking now [13:02] rodrigo_, while you are on g-s-d do you know why it doesn't respawn when it crashes? and why some applications drops their theme use but don't get it back when g-d-s is restarted? [13:02] desrt, ^ [13:02] like xchat or nautilus stay without theme until restarted [13:02] no. [13:02] seb128, if it doesn't respawn, I guess it's because we don't mark it as respawned for gnome-session [13:03] Sweetshark: (re from lunch) [13:03] as for the apps not getting back the theme, no idea really, they should really get it afaik [13:03] Sweetshark: right, that's the very problem we need to fix :) [13:03] Sweetshark: hyphen-af must not have an unversioned conflicts/replaces to openoffice.org-hyphenation-af; it needs to be (<< 1:3.3.0-7ubuntu1) [13:04] Sweetshark: uninstallable openoffice.org is due to main packages depending on universe ones; I'm currently sorting that out [13:04] fta2: since Friday, I have chromium going made with a lot of javascript site, like google reader [13:05] Sweetshark: do you know whether openoffice.org-hyphenation-lt works? (from ispell-lt source); it currently depends on OO.o, should the package be removed/updated/works as it is? [13:05] rodrigo_: the specification doesn't provide an explicit mechanism for replacing the existing settings manager, but from my reading it should be possible to do XSetSelectionOwner to blindly steal it [13:05] rodrigo_, the g-s-d .desktop has X-GNOME-AutoRestart=true [13:05] Sweetshark: ah, see http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=617569 [13:05] Debian bug 617569 in openoffice.org-hyphenation-lt "please rename to hyphen-lt" [Wishlist,Open] [13:05] rodrigo_, so I guess g-s-d session registration is broken [13:06] rodrigo_: maybe it's not such a good idea, though. the check for an existing settings daemon is clearly legitimate -- not good to go blindly stomping [13:06] desrt, yes, it will conflict with another g-s-d process already running [13:07] seb128, hmm, is that only on natty? [13:07] rodrigo_, no [13:08] rodrigo_, it's nothing new, g-s-d crashes bug we usually get are described are GNOME dropping his theme [13:09] or it restarts but the xsettings are not updated [13:11] seb128, I guess it might be the same issue as for desrt's bug [13:11] yeah [13:11] Sweetshark: I'll fix ispell-lt to recommend libo | oo.o, this will stop pulling in the oo.o transitional metapacakge and fix the installability report [13:11] so let me finish this (hopefully) fix, and we can test in both situations [13:11] although I'm having lunch 1st, bbiab [13:12] it could be that it thinks that a settings manager is still running when respawned [13:12] seb128, yes [13:12] rodrigo_, enjoy! [13:12] thanks :-) [13:12] rodrigo_: so an idea [13:12] rodrigo_: you could check the _NET_WM_PID property of the existing selection owner [13:12] and if it is set, use it to 'check on' the process with this PID for 'signs of life' [13:13] ie: not a zombie or something [13:13] yes, will try that also [13:13] but later, I'm out for lunch now, bbl :-) [13:24] hey mterry [13:25] seb128, hello! [13:25] mterry, how are you? [13:25] seb128, good, tracking down a bug in gtk+ :) [13:25] oh? which one? [13:26] wouldn't the correct response have been "there are no gtk bugs"? [13:26] pitti, ;-) [13:26] seb128, a bug that I found via indicator-datetime. I haven't reported it yet, working on the smaller test case. Note if you use the indicator-datetime prefs that if you are selecting a location in the locations dialog, it won't work if your mouse is hovering over the completion popup window [13:27] seb128, I have what seems to be a patch, but not sure if it's the best of all possible patches. Will talk to upstrea [13:27] mterry, quite a corner case ;-) [13:28] mterry, if you get borred or want new challenges feel free to take one of those [13:28] https://bugs.launchpad.net/ubuntu/+source/indicator-appmenu/+bug/741488 [13:28] Launchpad bug 741488 in indicator-appmenu "unity-panel-service crashed with SIGSEGV in props_cb()" [Medium,New] [13:28] seb128, maybe, but it's making that interaction feel very unrobust, because it seems to work sometimes, not others [13:28] bug #722375 [13:28] Launchpad bug 722375 in wxwidgets2.8 "High CPU usage of wxWidget apps caused by dbusmenu" [High,Triaged] https://launchpad.net/bugs/722375 [13:28] seb128, ok [13:28] the second one should be fun :p [13:34] pitti: ah, now I understand the hyphen-* openoffice.org-hyphenation-* issue. [13:47] pitti: Should I fix the stuff in openoffice.org-dictionaries? [13:47] Sweetshark: if you have time for it, please [13:47] Sweetshark: i. e. make the conflicts/replaces: versioned [13:50] pitti: should i bump dict to debian rev 3 too? [13:51] Sweetshark: "dict"? [13:52] pitti: openoffice.org-dictionaries [13:53] debian is at 3.3.0~rc10-3, we are at 2ubuntu1 [13:53] Sweetshark: http://packages.qa.debian.org/o/openoffice.org-dictionaries/news/20110311T174946Z.html sounds wrong, though [13:53] Sweetshark: the conflicts shouldn't be removed, but versioned [13:54] otherwise upgrades from lucid (or from Debian squeeze) will fail [13:54] Sweetshark: so if you upgrade to that version, you'll need to add them back :) [13:54] so, I merge from -3 and replace renes changes with versioned conflicts? [13:55] Sweetshark: the C/R: versioning should really be done in sid, and then merged over, but it would be nice to fix this today so that we can get this right for upgraders to beta-1 [13:55] Sweetshark: I don't think it's worth merging now [13:55] it's harder to fix -3 than to fix -2 [13:55] Sweetshark: versioned replaces, too, pleas [13:56] k [13:56] * Sweetshark still has a checkout [14:02] pitti: there arent any versioned or unversioned conflicts from hyphen-* against openoffice.org-hyphenantion-* should I add a versioned conflict? [14:03] Package: hyphen-fr [14:03] Conflicts: openoffice.org-hyphenation-fr [14:03] Replaces: openoffice.org-hyphenation-fr [14:03] Sweetshark: ^ [14:10] doh. I had the debian branch, not the ubuntu one === MacSlow|lunch is now known as MacSlow [14:24] hey all [14:24] hey rickspencer3 [14:27] howdy rickspencer3 [14:28] hey guys [14:28] just got to London! [14:28] rickspencer3: you have time again! [14:29] hi desrt [14:29] hey [14:29] desrt, I brought my codethink shirt! [14:29] nice [14:29] :) [14:29] be sure to wear it a lot around mark [14:31] rickspencer3: ok, beta release sprint or so? [14:31] ok, pitti tedg one of you broke nautilus [14:32] not sure who yet though [14:32] bug #740765 [14:32] Launchpad bug 740765 in nautilus "nautilus crashed with SIGABRT in __kernel_vsyscall()" [Medium,New] https://launchpad.net/bugs/740765 [14:32] desrt, will do! [14:32] NAUTILUS_IS_VIEW (view) assert in nautilus_window_disconnect_content_view [14:33] those started on the 23 on which day you both did patch it [14:33] pitti, well, a few things I wanted to get done, easier face to face [14:33] dinner time =) [14:34] seb128: doesn't ring a bell here, but feel free to assign to me [14:34] seb128, Yeah, I'd be surprised if my code did that as well... but let the blame game begin! [14:34] :-) [14:35] tedg, well, your patch created 2 other bugs at least in nautilus, I assigned those to you [14:35] tedg, hey btw ;-) [14:35] tedg, the desktop context menu lacks items now and some action group warning are printed as well [14:36] pitti: proposed patch: https://pastebin.canonical.com/45281/ [14:37] pitti: The only thing missing seems to be openoffice.org-thesaurus-de, I guess that needs a fix in openthesaurus. [14:37] seb128, Ah, okay. I'll probably be spending my whole day reviewing all the mterry branches though :-) [14:37] Sweetshark: looks fine, thanks! feel free to change -7ubuntu1 to just -7, so that it's applicable to Debian as well [14:37] ;-) [14:37] :) [14:38] tedg, there is also a bug about indicator-appmenu making wxwidget application eat cpu [14:38] if mterry doesn't beat you to it while you review his previous fixes ;-) [14:38] Hmm, okay. That's interesting. [14:38] * mterry files a bunch of junk branches to slow tedg down [14:38] NOOOOOO!!!! [14:40] pitti: fixed to be debiancompatible. Should I debuild and put it in a ppa? [14:43] Sweetshark: or just put on chinstrap, will sponsor from there [14:43] k [14:47] pitti, ok, sorry but I think you win the nautilus crash bug the very first bug was opened on the revision before ted's change [14:47] seb128: heh, k :) [14:48] pitti, it seems it would be due to 17_disable_locked_automounting.patch [14:48] pitti, assigned to you anyway [14:48] ok, will look [14:49] pitti, it "seems" from duplicate description to have something to do with copies or devices handling [14:49] like quite some people got it while copying files to or from devices [14:56] mterry, I was going to look at this now https://bugs.launchpad.net/ubuntu/+source/audacity/+bug/731451 [14:56] Launchpad bug 731451 in audacity "audacity not working in Ubuntu 11.04 (natty)" [High,Confirmed] [14:56] mterry, just want to make sure you are not taking this on also [14:58] ronoc, awesome! No I wasn't [15:03] Sweetshark: mythes-en-us has a conflict against openoffice.org-thesaurus-en-us, but that package is actually availlble as a transitional package. oculd you please add a (<< version-before-transitional) ? [15:03] to the conflict? [15:10] mterry, mpt: do you know if there is an open bug about the indicator-datetime add location entry behaviour being confusing? [15:10] mterry, mpt: like some users do type a name, "ie: london" and press enter [15:10] mterry, grand [15:10] without waiting for the list to make suggestions or notice that it does [15:11] seb128, I reported half a dozen last week [15:11] I think bug #729791 is that issue [15:11] seb128, I think that was one of them [15:11] Launchpad bug 729791 in indicator-datetime "Cannot add multiple cities" [Low,New] https://launchpad.net/bugs/729791 [15:11] mpt, ok, bug #740945 I guess [15:11] Launchpad bug 740945 in indicator-datetime "Possible to add blank or non-existent locations" [Undecided,Confirmed] https://launchpad.net/bugs/740945 [15:13] seb128, yeah, that one (740945) I was treating as the issue you described [15:13] seb128, mpt, I guess the desired result is to blank out the entry when they do that? [15:13] or just ignore the press is probably better [15:14] mterry, how about completing to the first suggestion if there are any, or if there aren't, ignore the keypress entirely and keep editing. [15:14] Would that work? [15:14] mpt, I think so [15:15] well the thing is that it's not obvious you will get suggestions [15:15] I can type london and enter before getting any there [15:15] seb128, in that case the keypress would do nothing, and then a fraction of a second later the menu would appear [15:15] and typing "London" enter should add the London location no? [15:15] mvo: already in the works for all openoffice.org-dictionaries and openthesaurus packages [15:16] seb128, London Ontario? :-) [15:16] mpt, no, east london , eastern cape, south africa :p [15:16] (that's the first in the list) [15:16] mpt, but right, you have a point [15:17] hmm [15:17] mterry, this might be pie in the sky, but is it possible to sort locations by probability or population or anything like that? [15:17] Sweetshark: nice [15:19] mvo: thank pitti, he nudged me to do it ;) [15:19] mpt, we don't have that data right now. I have a branch filed to put the locations that actually share a prefix at the top [15:20] mpt, that would be a good future wishlist item to file though [15:23] ok [15:24] mterry, bug #744261 [15:24] Launchpad bug 744261 in unity-place-applications "unity-applications-daemon crashed with SIGABRT in raise() (dup-of: 738884)" [Medium,New] https://launchpad.net/bugs/744261 [15:24] Launchpad bug 738884 in unity-place-applications "unity-applications-daemon crash in unity_package_searcher_get_random_apps()" [High,Fix released] https://launchpad.net/bugs/738884 [15:24] mterry, when did you restart your session? [15:24] mterry, it's supposed to be fixed since tuesday, did you restart since? [15:25] seb128, yup, this morning. kamstrup was asking in #ayatana too === bcurtiswx_ is now known as bcurtiswx [15:47] hey, any idea why pulseaudio just quits randomly? [15:48] I've launched it with --log-level=debug but I can see nothing out of place [15:52] Sweetshark: when do you plan a new libreoffice upload? there is still this file overwrite problem that I would like to get fixed (I'm getting the latest source currently) [15:55] mterry, https://wiki.ubuntu.com/TimeAndDate?action=diff&rev2=28&rev1=27 [15:55] Does that look right? [15:56] mpt, I think so. And if the user just focuses away, stopping the edit with, say "london" in the entry, should we just cancel the edit and ignore it? [15:57] oh, poo [15:57] good catch [15:58] mterry, if we blanked it out, that would be a little harsh if you misclicked while trying to click an item in the menu [15:59] didrocks, hey, what do you mean? [15:59] mpt, we could select the first answer when it comes in later? but if there are no answers, blank it? In the former case, they may end up getting something they didn't expect though [16:00] seb128, WRT bug 741848 what do you think about backporting https://bugzilla.gnome.org/show_bug.cgi?id=622658 [16:00] Launchpad bug 741848 in indicator-appmenu "Indicator-appmenu needs localization" [Low,Confirmed] https://launchpad.net/bugs/741848 [16:00] didrocks, (wrt chromium) [16:00] fta: I have a lot of javascript getting stuck those days in chromium (no issue in firefox) [16:00] Gnome bug 622658 in gtk "GTK_STOCK_FILE Stock Label" [Trivial,Resolved: fixed] [16:00] mterry, yeah. So how about we leave it as plain text, but discard it if you close the window without finishing it? [16:00] didrocks, hm, try with a fresh profile (--temp-profile) [16:00] tedg, oh right, I knew we discussed it previous cycle but I forgot it was a gtk thing [16:00] fta: ok, will try [16:01] mpt, seems like the discard would also be an unexpected action [16:02] seb128, Okay, I'll move it to a GTK task and link it then. [16:02] mterry, so the lack of time next to it wouldn't be enough to show that it's not an understood location? maybe so [16:02] tedg, bug #657961 in fact then [16:02] Launchpad bug 657961 in ubuntu-translations "stub menu not translated" [Medium,Fix committed] https://launchpad.net/bugs/657961 [16:02] mpt, what about little red squigglies like a spellchecker or something [16:02] mterry, or just color the whole thing red. Is there a theme-safe error color? [16:02] tedg, but well yes let's reassign the current one, I will backport that after beta [16:02] mpt, color only solutions are bad for a11y [16:03] tedg, thanks [16:03] mpt, Not yet. Symbolic icons will give us that :-) [16:03] gnarr [16:03] seb128, Oh, I marked it a dup? Is that okay, or did you want something else? [16:03] tedg, that's ok thanks [16:03] mterry, how about a little error icon inside the trailing end of the field? [16:03] mvo: I have no concrete plan for the next natty upload yet. [16:04] mpt, and only show it in this situation? (when you focused away) I suppose we could also do that for the "enter" case instead of the previous idea? [16:05] mterry, I think highlighting the first result for the Enter case is probably a good idea regardless [16:05] so yes, error icon only for the focus-away case [16:06] mpt, OK, I can't see why this all wouldn't be technically possible, but I'll let you know if I find any issues [16:20] mterry, https://wiki.ubuntu.com/TimeAndDate?action=diff&rev2=29&rev1=27 [16:22] mpt, you could have multiple invalid rows if user edits existing locations. but your edit reads like you expect just one [16:22] oh, you're good === bjf is now known as bjf[afk] [16:28] mterry, https://wiki.ubuntu.com/TimeAndDate?action=diff&rev2=30&rev1=29 [16:32] didrocks, so, does it work there? [16:32] fta: didn't have time to test with this, will do that later [16:32] ok [16:33] didrocks, btw, http://people.ubuntu.com/~fta/acroread.ogv [16:33] and http://people.ubuntu.com/~fta/dash.ogv [16:33] known? [16:33] mpt, looks good. Looking into implementing it though, GTK does not give us access to the popup window at all. So I'm not sure I'm going to be able to force one to be always highlighted or even tell whether we should ignore an enter press or not. I'm looking further, but it's discouraging so far. [16:34] Could always reimplement the work that GTK does a bit, but that would be a larger task [16:34] mterry, ok, let me know if it's impractical, and we'll find a plan B [16:34] Thanks for being so thorough in reviewing it, you've saved us a lot of time :-) [16:35] fta: first one, I would first check in the appmenugtk side [16:36] fta: second one is known [16:37] mterry, tedg: ^ do you have any clue about the acroread video? [16:38] seb128, Uhm, they must build their labels somehow funny. [16:39] didrocks, ok, thanks. I just filed bug 744361 for the 2nd one then [16:39] Launchpad bug 744361 in appmenu-gtk "acroread appmenu unreadable (no text)" [Undecided,New] https://launchpad.net/bugs/744361 === 20QAADT57 is now known as mterry [16:39] fta: thanks :) [16:40] seb128, I just disconnected for a second, but looks like I missed something? what was that about the acroread? [16:40] mterry, bug #744361 [16:40] fta, seems similar to bug #664647 [16:40] Launchpad bug 664647 in appmenu-gtk "acroread menu captions not transfered" [Medium,Triaged] https://launchpad.net/bugs/664647 [16:40] seb128, yikes [16:41] seems we should just add acroread to the ignore list [16:41] right, looks the same [17:04] mpt, I think I found a super hacky way to get access to the gtk locations popup, so we can probs rest easy [17:05] mterry, brilliant [17:15] chrisccoulson, hi [17:15] chrisccoulson, how busy are you nowadays? [17:15] pitti: is openoffice.org-dev current NBS? [17:16] hi seb128 [17:16] seb128 - fairly, but i'm sure i can fit in some desktop work ;) [17:16] chrisccoulson, what do you work on? still natty work? xulrunner cleaning? [17:17] pitti: I ask because its currently causing grief in the auto-upgrade tester and I wonder if we can get rid of it [17:18] seb128 - yeah, mostly. also, i'm working a bit on firefox translations too [17:18] chrisccoulson, can you take on some libdbusmenu issues? [17:18] sure, no problem [17:18] chrisccoulson, ok, you won bug #729187 [17:18] Launchpad bug 729187 in libdbusmenu "xchat - separators are shown as 'Label Empty'" [Low,Confirmed] https://launchpad.net/bugs/729187 [17:19] thanks [17:19] chrisccoulson, thank you ;-) [17:19] tedg, ^ [17:19] tedg, I'm bouncing a few issues chrisccoulson's way [17:19] Ah, cool. Thanks chrisccoulson! [17:19] :) [17:27] annoying tmux crashes are annoying [17:27] pitti: there are some new changes files for openthesaurus and openoffice.org-dictionaries on chinstrap [17:32] mvo, could you add bug #713984 to your natty list? [17:32] Launchpad bug 713984 in sessioninstaller "session-installer crashed with TypeError in _render_warning(): could not convert argument to correct param type" [Undecided,New] https://launchpad.net/bugs/713984 [17:33] mvo, it seems to happen every time someone tries to use file-roller on a format which needs something to be installed [17:34] Sweetshark: thanks, uploaded! will shepherd it through ASAP [17:35] mvo, https://code.launchpad.net/~gilir/sessioninstaller/defer-migration/+merge/46346 has a trivial typo fix waiting review for a while it seems if you want an easy review === bjf[afk] is now known as bjf [18:03] seb128: thanks! [18:05] mvo, you're welcome ;-) === alecu is now known as alecu-lunch [18:06] chrisccoulson, ok, I've added 2 dbusmenu_client_send_about_to_show() crashers with different stacktrace to your list as well since you started investigating one of those before [18:06] chrisccoulson, will be enough for today ;-) [18:06] or for this week mayb [18:06] e [18:06] seb128 - cool, thanks. i look forward to it ;) [18:07] seb128: added a bugtask [18:07] tedg, mterry: bug #743895 [18:07] Launchpad bug 743895 in libdbusmenu "indicator-applet-appmenu crashed with SIGABRT in __kernel_vsyscall()" [Medium,New] https://launchpad.net/bugs/743895 [18:07] those "0 <= index && index < 24" asserts, are they supposed to be fixed by https://code.edge.launchpad.net/~mterry/dbusmenu/741726/+merge/54882 [18:07] ? [18:08] mvo, thanks! [18:08] brb [18:09] seb128, looks like those bugs are probably the same as my merge [18:09] pitti, can we still get uploads in today? [18:09] Yeah, seems the same to me as well. [18:10] thank you [18:10] mterry, Have you looked at bug 733050 at all? The reporter says it's in rev 125, which was when you added the cancellable, but I can't see anything that'd cause an issue with nautilus. [18:10] Launchpad bug 733050 in appmenu-gtk "appmenu-gtk 0.1.96-0ubuntu1 prevents desktop global menu (desktop appmenu in unity panel) from appearing" [Medium,Confirmed] https://launchpad.net/bugs/733050 [18:11] tedg, seems to be a race since if you restart nautilus it works [18:11] seb128: just upload, and then poke folks in #u-release [18:11] ok [18:11] seb128: we built a first set of images for smoke-testing, but don't expect them to be the final ones [18:12] kenvandine, mterry: can one of you backport https://code.edge.launchpad.net/~mterry/dbusmenu/741726/+merge/54882 to natty? [18:12] tedg, yeah, I haven't seen that bug, but I'm not sure why my cancellable would do that [18:12] seems it would filter out noise from other components [18:12] seb128, sure [18:12] so would be nice to get it in [18:12] seb128, for beta 1? [18:12] it seems to trigger sibabrt [18:12] kenvandine, yes, it's a small diff and it leads to random crash [18:12] will do [18:12] got unity to crash on it twice in my guest session today [18:13] mterry, K, I'll assume it's not 125 and look around some more. Blame the reporter ;-) [18:13] kenvandine, thanks [18:13] tedg, well... the reporter seems technical enough to claim it with authority. It's possible they actually compiled the two versions and tested [18:14] mterry, Yeah, it seems so. I'll have to bisect it a bit. [18:14] mterry, It's a good place to start. [18:15] Maybe he meant 125 worked... [18:15] didrocks: given how close we are to b1, I take it that the remaining a11y stuff won't get done in time, so should I postpone them and we'll reconsider them for o? [18:15] didrocks: (TheMuso said that we should better go with classic for some a11y profiles anyway) [18:15] pitti: yeah, seems that's the wisest and safest thing to do at this point [18:16] ok time for sport and dinner, I will be back later [18:20] ronoc: do you think you can do the rb->banshee upgrade testing this week in time for beta-1? (at that time a lot of people will upgrade) [18:21] pitti, sure I'm pretty happy with it now, Banshee are about to release 1.9.6 into natty, once that lands [18:21] I'll give it a run through [18:21] ronoc: ok, sounds like postponing to b2 then [18:23] pitti, the only thing really I had to test was that the player behaves properly with the menu, there was a task for jcastro to check if the tracklist importer was working as expected [18:24] pitti, i guess this is pretty important for task labelled as such [18:24] didrocks: "get some plugins to be linked statically" (for compiz) -> is there still work needed there, or is it what it is now? [18:25] pitti: it should, but it depends on Sam's work and realistically, it won't be there [18:26] didrocks: and it's an optimization, not functionality, right? [18:26] pitti: yeah [18:26] [smspillaz] implement improved borders & shadows (phase 2): INPROGRESS [18:26] hm, isn't that done now? [18:26] don't know what's the phase 2 :/ [18:27] guys anyone know who is the maintainer for clementine ? [18:27] ronoc: QA's going to write a test for the importer, and upstream hasn't had any bugs filed on it in a while, so I think it's good to go. [18:28] didrocks: setting to done for now; they changed not too long ago, and now is UIF and all that anyway [18:28] [smspillaz] polish dialog and MT functionality: INPROGRESS [18:28] that's also open from alpha-3, dropping [18:28] pitti: smspillaz won't have the time to do that [18:28] jcastro, good stff [18:28] stuff even [18:29] didrocks: right, postponing to o [18:29] jcastro, do you know who packages clementine for ubuntu? [18:29] pitti: thanks :) [18:30] ronoc: -- Artur Rona [18:30] jcastro, ah thanks [18:37] hey kamstrup, I'm now generating docs weekly from the giraffe output and have the links on the d.u.c. server http://developer.ubuntu.com/api/ ;-) did you have any luck talking to the vala guys re: that bug whereby the generated gir files were not correct (they had too many entries of the same type under a record, IIRC)? [18:38] Sweetshark: yay! http://people.canonical.com/~ubuntu-archive/testing/natty_probs.html [18:38] Sweetshark: all oo.o stuff gone (except transitional on arm), thanks [18:38] kamstrup, I'm basically using a workaround for the unity docs [18:40] good night everyone, Taekwondo o'clock [18:41] nite pitti [18:43] shouldn't bug 743941 be re-assigned to libdbusmenu-gtk or something? [18:43] Launchpad bug 743941 in indicator-weather "indicator-weather crashed with SIGABRT in raise()" [Medium,Confirmed] https://launchpad.net/bugs/743941 [18:50] pitti, fyi, apport-retracer completely messed up my install :P i had a bunch of -dbg and -dbgsym already, it installed zillions of -dbgsym (why so many?) last time i retraced a crash file with -g (as root), dropped some of my -dbg (why??), leaving me with an incomplete install requiring -f === warp10 is now known as aquila1 [20:01] Sweetshark: any idea why soffice.bin starts randomly? === alecu-lunch is now known as alecu === bdrung_ is now known as bdrung [20:43] tremolux, are there any known bugs about not being able to click on anything in software-center except for the sidebar? [20:43] mterry: hmm, I don't think there are [20:43] mterry: what'd you do?? [20:44] * tremolux looks accusingly at mterry [20:44] tremolux, it probably doesn't like deja-dup being installed :) [20:44] tremolux, accents stopped input processing [20:44] mterry: wha?? [20:44] tremolux, just joking about all the other deja-dup related bugs [20:44] mterry: yeahhh [20:45] mterry: so you can select items in the sidebar, and does that change the view? === aquila1 is now known as warp10 [20:45] tremolux, it's an invisible input window in the middle of the screen [20:45] tremolux, only with software-center [20:46] tremolux, if I move the window around, I can click on parts of it [20:46] mterry: ah! listen, I had that happen for about a day! [20:46] tremolux, yeah, it does change the view. But then I can only click on the view contents if I move the window to be out from under the invisible input window [20:46] mterry: and, it's not just software-center, it's everything [20:46] tremolux, but here, it's just software-center. If I switch to another window, it's gone [20:46] mterry: yeah, a big empty "dead space" square in the middle of the screen [20:47] mterry: oh really? when I saw it, it affected my editor, firefox, thunderbird...hrm hrm [20:47] mterry: I assume you are fully up to date? [20:48] tremolux, it's even outside the window. Like I can't click on the titlebar and drag it if it'ss under the invisible window [20:48] tremolux, yeah [20:48] mterry: right [20:48] mterry: try switching to another workspace and see if an app there has the problem [20:48] tremolux, nope, all clear [20:48] mterry: tho... I dunno [20:49] tremolux, so do you want this on software-center (only software affected, and persists even after close/restart of s-c) or compiz with note about software-center-uniqueness? [20:50] mterry: good question, I guess I really don't think it's a problem in software-center specifically [20:51] mterry: when it happened to me, I checked for such a bug in compiz and other places, but found nothing [20:51] mterry: and it cleared on my next update, so I thought it was either fixed or...sunspots :/ [20:52] mterry: but as I said, for me, it affected all apps [20:52] mterry: and it did persist across reboots [20:53] mterry: so you pull another app in front of software-center, and the dead zone isn't there? bring s-c back, and it is there, correct? [20:54] mterry: fwiw, I haven't seen any reports of this in incoming s-c bugs yet [20:55] tremolux, right [20:55] tremolux, OK, will file against compiz [20:55] ricotz, you're probably good to make your gjs build use the libmozjs package now if you like [20:55] mterry: k, thanks, and I will watch for any reports of this [20:55] (i think you were asking a few days ago) [20:56] nothing is going to change much now, although we might add a pkg-config file [20:58] mterry: oh, and plz tell me the bug so I can follow it (or subscribe me if you want to) [21:01] chrisccoulson, i see, i think i wait for the pkg-config file first, gjs should be changed first then which is in the natty repo, and then gnome-shell can be patched to use it [21:02] chrisccoulson, this spidermonkey version is the current firefox4 one, right? [21:04] ricotz, yes, with some small changes [21:06] ricotz, the tarball release of mozjs has this change in which firefox doesn't have: https://bugzilla.mozilla.org/show_bug.cgi?id=630209 [21:06] Mozilla bug 630209 in JavaScript Engine "Unsafe usage of JS_Compile*/JS_ExecuteScript in shell and tests" [Normal,New] [21:06] (which is an API break) [21:06] but it's a small one [21:06] (JSScript is not public anymore) [21:07] tremolux, now s-c is unresponsive and grayed out... how do I get a trace from it in this state? [21:07] i guess that walters already knows about that though [21:07] tremolux, if i force-kill it, will that cause apport to prompt me? [21:07] chrisccoulson, ok [21:07] mterry: I'm not sure it will [21:08] i didn't actually [21:08] tremolux, is there a trick I can do? [21:08] mterry: I'm usually running from the terminal, so I can interrupt it and see where it's hung up [21:09] mterry: I don't know of one actually when it's launched normally :( [21:09] mterry: unless it writes something out to ~/.cache/software-center/software-center.log .. [21:10] mterry: nah, I wouldn't think it would [21:10] walters, is this used by g-s? [21:10] oh well [21:11] i haven't actually tried gjs against the split release [21:11] walters, do you use JS_CompileScript/JS_DestroyScript? [21:11] mterry: if you know how you got that hang, you could start again from the terminal, but my guess is it will be hard to repro [21:12] chrisccoulson: yeah, but only for /usr/bin/gjs [21:12] i will look at the bug [21:13] tremolux, yup. :-/ subscribed you to bug 744533 [21:13] Launchpad bug 744533 in compiz "Invisible input window over software-center" [Undecided,New] https://launchpad.net/bugs/744533 [21:13] walters, alright [21:15] mterry: thx! so, you mention "dreaded invisible input" window, so, is that the symptom of that? it's not over the full screen, just an area in the center? [21:15] tremolux, correct. it's roughly 650x500 pixels [21:16] mterry: ah, so that's what I had then, sounds like the dimentions I saw also [21:16] mterry: so is that bug fixed now? [21:16] (dimensions, heh) [21:17] tremolux, which one? the "dreaded" one? If it was a while ago, then yes. That one tended to start at (0, 0) -- i.e. the top left corner -- though. [21:17] mterry: ah, ok, yeah, so what I saw was what you are seeing, it seems [21:17] tremolux, the invisible input window has come up a few times. But I thought we'd seen the last of it a while ago [21:18] mterry: right [21:19] mterry: btw, I found that I couldn't mouse/trackpad scroll either if I had the cursor sitting over the "dead area" [21:20] makes sense, yeah [21:20] those come in as button presses