[00:53] Dear libtool: WTF. === duflu_ is now known as duflu [06:38] Good morning [06:39] hey pitti [06:39] storms are hitting! http://www.bom.gov.au/products/IDR021.loop.shtml [06:45] darkxst: Where are you? [06:45] TheMuso, A little north of Melbourne [06:45] darkxst: Ah yeah, nasty. [06:45] darkxst: We are supposed to get storms here in Sydney too I think at some point. [06:46] And hey there pitti. [06:46] TheMuso, we haven't had any rain at all in about a month, its probably useful! [06:46] hey TheMuso, how are you? [06:46] morning darkxst [06:46] Oh yeah, as long as its not too heavy. [06:46] pitti: Well thanks, yourself? [06:46] TheMuso: quite well, thanks [06:47] pitti: I've actually got something interesting to discuss re systemd, and its pkg-config file. For various reasons, I've been working to build pulseaudio with different prefixes/install locations, which gave me errors. This lead me to trying distcheck with puseaudio. I found that systemd's pkg-config file has the prefix hard-coded in the various service/unit directory variables. I see no bugs upstream/anywhere about this. Is this known? [06:48] pitti: i.e systemduserunitdir=/usr/lib/systemd/user -- I would expect that variable to use ${prefix} rather than hard-code /usr... [06:49] pitti: I was actually trying to come up with a patch to fix it, but my autofoo is lacking it seems... [06:49] TheMuso, maybe Lennarty did it on purpuse so you don't change the prefix ;) [06:50] darkxst: But that makes no sense. If you look at other .pc files, when referring to the include dir etc, they all refer to prefix with a variable. [06:50] darkxst: And it is likely to break the distcheck of packages that use systemd in some way, i.e pulseaudio. [06:50] darkxst: Actually, it does break distcheck. [06:50] There is no unlikelyness about it. :) [06:51] TheMuso, I wasn't actually being serious there! [06:51] darkxst: Yeah saw the smily, but words on teh screen alone are not enough to convey such things, so yeah makes sense in context. :) [06:52] TheMuso, hardcoded paths are in the pc.in file? [06:53] darkxst: No, it uses substitution to fill them in. [06:53] But that has me wondering actually... [06:53] * TheMuso checks something. [06:54] Hrm ok, as I expected. Never mind, thought the prefix variable was part of the .pc.in files I checked that look sane. [07:01] TheMuso, why are you building pulseaudio with different prefixes? if you just want to isolate from your system? jhbuild can do that [07:04] darkxst: Reasons, I cannot really say. [07:06] darkxst: But even so, distcheck is designed to make sure a package is buildable with custom prefixes as well as using a different DESTDIR. [07:06] just saying, jhbuild has quite some hacks, to make things work with messed up mixed prefixes ;) [07:06] darkxst: Nice to know, but jhbuild is overcomplicated for what I am doing. [07:06] TheMuso, some things take vars from the actual installed .pc files [07:08] darkxst: I'm sure jhbuild is wonderful and all, but more than what I need, and, still we come back to pulseaudio distcheck failing. [07:13] TheMuso, maybe a bug in pulseaudo configure.ac [07:14] darkxst: It could certainly be worked around in pusleaudio's configure.ac, but IMO that is rather hacky, and not the correct solution. [07:14] good morning [07:15] Morning didrocks. [07:15] hey TheMuso [07:15] TheMuso, this is a fruitless conversation since you can't say what you are doing ;) sure you can work it out! [07:15] hey didrocks [07:16] good evening darkxst ;) [07:16] didrocks, stormy evening! first rain in over a month! [07:17] darkxst: waow, is it planned to stay over the week-end or just doing the night? [07:17] during* [07:17] darkxst: Sure I can, but I am still of the thought that working things out *properly* means fixing systemd's pkg-config. [07:17] Anyway, EOD for me. [07:17] Tonight and sat [07:18] TheMuso, pkg-config does what it is told [07:18] We had a lovely late storm that roled into my area of Sydney last night, just as I went to sleep, was lovely. [07:18] darkxst: Yeah I know that. Anyway, I can work around it for now. [07:21] didrocks, https://code.launchpad.net/~noskcaj/ubuntu/vivid/appstream-glib/mir-fixes/+merge/249562 should clean up the appstream-glib MIR [07:22] darkxst: oh DSO link errors… insane we still have this nowdays, put my mind back more than 5 years ago :) [07:22] didrocks, I don't think Fedora builds with as-needed [07:22] even still [07:23] s/builds/links/ [07:23] I hit them all the time in jhbuild [07:23] darkxst: oh really? I thought they turned on the gold linker… but probably not when seeing that :) [07:23] darkxst: need sponsoring? [07:23] looking good to me [07:23] didrocks, yes, I don't have powers for appstream-glib [07:24] darkxst: running a sbuild pass first to ensure the tests are working and then, sponsoring [07:28] thanks didrocks [07:28] yw Noskcaj :) [07:36] Noskcaj: darkxst: built successfully, sponsored and MIR approved [07:36] * didrocks closes tabs [07:49] didrocks, Since you approved the MIR, would you have time to s/appdata-tools/appstream-util in all the r-deps? The same binary is in both [07:49] It seems to be too trivial of a patch for it to be worth my time as someone without upload rights [07:50] Noskcaj: quite busy today, (and off Monday). If it can wait Tuesday (and you remind me about it), sure can do [07:50] cool [07:50] Noskcaj, some naughty people can do copies from ppa's into archive [07:50] yeah, that works as well :) [07:51] * didrocks likes darkxst_'s idea [07:51] ok, i'll look at making a ppa tomorrow if no one steps up to do the work [07:51] Noskcaj: keep me posted on Tuesday, I'll do the package copy if nobody did it before then [07:51] ty [07:51] yw === tiheum_ is now known as tiheum === darkxst_ is now known as darkxst [08:50] morning [08:50] hey willcooke [08:57] hey willcooke [08:59] anyone else having multi-monitor pain on vivid on Intel today? http://imgur.com/80LcmgV [09:00] popey, on you x220? So Intel gfx? [09:00] popey: urgh no :/ x220, intel gfx here [09:00] popey: -14 kernel? [09:01] 3.18.0-13.14 here [09:03] yes [09:03] bug 1421575 [09:03] bug 1421575 in xserver-xorg-video-intel (Ubuntu) "Desktop corruption when changing monitor config" [Undecided,New] https://launchpad.net/bugs/1421575 [09:03] oh hang on, I'm on some dodgy 3.19 kernel [09:03] * popey reboots to sanity [09:04] popey: what are you installing on your poor system? :) [09:04] installed 3.19 upstream to diagnose some other issue (gpu lockup) [09:04] but this is way worse [09:04] willcooke: postponed the arduino stuff [09:04] as discussed yesterday [09:04] thanks didrocks [09:04] yw [09:05] I didnt get time to explore Eclipse yesterday [09:05] maybe over the weekend [09:05] willcooke: no worry, we'll backport that to the ppa at worst :) [09:05] popey: trying to compete with attente_ on kernel bisecting? [09:06] bah, still busted on 3.18 [09:06] but busted differently [09:07] popey: -13.14? -13.13 did have some issues… [09:08] dunno, i can't ssh into it now, disk light solid, it's not happy [09:08] waow [09:08] just dist-upgraded (quite a lot since yesterday) [09:08] rebooting [09:08] and see if I blame you *personally* [09:08] *brrring* *brrrring* Hello, IT [09:09] networking is also busted [09:09] seems to want to have wifi and wired on at once and is getting upset with this situation [09:10] I had that yesterday on my U8 box, I think it's a different NetMan issue [09:10] popey: sorry, everything's working well here [09:13] 3.18.0-13-generic is what I am on [09:13] Laney, https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1420404/comments/3 btw, summary of the u-s-s translations issue [09:13] Launchpad bug 1420404 in ubuntu-system-settings (Ubuntu) "Name entry in desktop file not localized" [High,In progress] [09:13] seb128, Laney is on swap today [09:14] willcooke, I know, but he reads backlog when back usually, that can be monday [09:14] kk [09:14] just making sure [09:15] willcooke, it's just a fyi on a bug I spent my day on yesterday [09:15] Hello Laney from the past [09:15] he helped me a bit [09:15] :-) [09:17] didrocks: https://www.youtube.com/watch?v=W3ZQ9xW0PhQ :( [09:18] popey: private video [09:18] fixed, ta [09:19] popey: waow, if we wanted to do it on purpose, we wouldn't be able I guess :) [09:19] I guess it's a question for mlankhorst [09:21] didrocks, just like I hit 5 screws today screwing down architraves ;( certainly can't do that if you try! [09:21] just put some duct tape over it [09:22] ogra_, duct tape will fix it ;) [09:22] darkxst: it's really "Friday 13th" it seems for some of you :) [09:23] :) [09:23] popey, mlankhorst is off today. [09:23] didrocks, doubt that is related to my achitrave problems! [09:24] balls, it is fine in a guest session [09:24] so something is knackered in my session [09:24] do we have a "reset compiz back to defaults" button anymore? [09:25] popey: you think it's compiz? do you have gnome-shell or any other DE? [09:25] no [09:25] I'm pure as the driven snow. [09:26] :) [09:26] popey, try removing the monitors.xml file [09:26] done that [09:26] ah [09:26] * popey copies monitors.xml from a good guest session in [09:26] \o/ [09:26] heh [09:26] popey, just delete it, u-s-d will recreate a good one then [09:26] Not touching that again today. [09:26] nope, it didn [09:27] it created a _worse_ one [09:27] that's weird… maybe it's the gsd config [09:27] let me find it [09:27] http://paste.ubuntu.com/10201762/ is my good one [09:27] popey: do yo uhave the "bad" one? [09:28] no, but I can probably make a bad one by breathing near the display applet [09:28] lemme try [09:28] yeah, would be interesting [09:28] popey, if its creating broken configs, probably xrandr bug [09:28] k [09:28] darkxst: well, it does create a "good one" in his guest session [09:28] so puzzling… [09:29] didrocks: http://paste.ubuntu.com/10201816/ is the bad one [09:30] identical! [09:30] of course [09:30] but as my user if I touch the monitor config, even if I let it revert back, it still makes the display all messed up and flickery [09:31] popey: yeah, let's see with mlankhorst I guess, he will probably ask you to dump your xrandr configs [09:32] kk [09:33] popey, that looks like a driver bug [09:33] it must be something in my session [09:33] (unless xrandr is stuffing up the timings on your modes) [09:51] seb128, do you think I should be able to use gsettings set com.canonical.Unity8 usage-mode Staged to get U8 desktop in to tablet mode? [09:53] willcooke, that should work yes [09:53] in which case.... [09:53] sad face [09:53] should I be able to do it from an SSH session? [09:53] like, [09:54] do I need to be in the same session or something? [09:55] willcooke: did you get any error? (I think you are not in the same session bus) [09:55] no errir [09:55] when I tried from ssh [09:55] when I try from U8 terminal [09:55] I get a seg fault [09:56] ? [09:56] willcooke: if you then gsettings get com.canonical.Unity8 usage-mode, did you get the same value in return? [09:56] $ echo $? [09:56] 0 [09:57] willcooke: it should print the value of usage-mode key (so "Staged") [09:57] nada [09:57] hey mlankhorst [09:58] mlankhorst, I know its your day off and all, but there's some weirdness for you to take a look at on Monday from popey up there ^^ [09:58] define weirdness :P [09:58] mlankhorst, http://imgur.com/80LcmgV [09:59] mlankhorst, https://www.youtube.com/watch?v=W3ZQ9xW0PhQ [09:59] (not Rick Astley)3 [10:00] looks more like a compiz bug :P [10:00] seen the same before [10:00] yeah, that was my feeling too [10:01] willcooke, ssh, you might need to define the dbus env [10:02] seb128, if I get env from U8 terminal, grep for bus, and then copy that in to my SSH session, that might work? [10:02] willcooke: you need DBUS_SESSION_BUS_ADDRESS at least :) [10:02] willcooke, just look to /proc/`pidof unity8`/environ [10:03] easier to cat that, so you can select/paste [10:03] rather than having to type it over [10:03] seb128, coolio, thanks [10:03] yw === vrruiz_ is now known as rvr [10:16] seb128, worked \o/ [10:16] willcooke, great! [10:18] weird that it wasn't working in your session though [10:18] (from the U8 terminal) [10:22] seb128: nice fix! [10:23] Laney, hey, thanks! it was "interesting" to debug [10:26] Laney, btw did you go anywhere with the about panel/storage/qt5.4? Mirv wants to land qt5.4 early next week, we need to fix the panel to at least load, even if the disk computation is a bit off in a first time [10:27] seb128: no sorry not yet, but could make it just use '/' or something [10:27] Laney, right [10:28] Laney, well, Mirv said he would have a look today, since they want to land qt and you are off work, just wanted to make sure he doesn't dup work [10:29] oh well let me push the small bit I did [10:30] Mirv: lp:~laney/ubuntu-system-settings/storageinfo-5.4 [10:34] Laney, thanks [10:35] Laney, https://bugs.launchpad.net/ubuntu/+source/ubuntu-system-settings/+bug/1420404 btw for easy review karma if you want to ack it ;-) [10:35] Launchpad bug 1420404 in ubuntu-system-settings (Ubuntu RTM) "Name entry in desktop file not localized" [High,In progress] [10:35] shrug [10:35] why is CI grumpy on settings? [10:36] seems like it fails to start or something [10:36] seb128: there are been a new CI Train rollout, maybe linked? [10:36] didrocks, not sure [10:38] seb128: I wonder if the uitk shouldn't have a fix here too [10:38] Laney, yeah, I plan to open at least 2 bugs [10:39] Laney, one to have the documentation mention that it sets the textdomain to that value [10:39] Laney, and maybe one to say that is should set the domain only if it hasn't been set manually before [10:39] (thanks didrocks for suggesting that one) [10:39] Laney, is that what you were thinking about? [10:40] indeed [10:40] yeah, you will clearly not be the only ones to be trapped by this, even if documented… [10:40] well, to be fair it's rather a special case [10:41] but yeah, likely going to bite others [10:41] you can do i18n.domain = ... from your qml [10:41] and it's fine for the UI [10:41] I'm unsure that enthousiast would be happy to spend a day on something similar :) [10:42] the fact that g_dgettext() decide on its strategy on the first call and that this calls happen between the init and the domain change is a bit special [10:42] but yeah, we are probably not going to be the only one to use glib [10:42] yeah ;) [10:42] any quick workarounds for these wierd network manager issues? I've already stopped ofono [10:43] willcooke: I didn't get it again since Brussels here btw… [10:43] ohh [10:43] last time I did restart the network manager service [10:43] ah crap, I had a nm bug just now [10:43] C+Ped some nmcli output to file one but I forgot and lost it :( [10:43] I have wired networking and wifi - and I can't get to the internet most of the time [10:43] vivid? [10:44] if I unplug the ethernet it works again# [10:44] yeah, on my u8 vivid machine [10:44] ah, not the same, only wifi here [10:44] ya, try sudo restart network-manager [10:44] tried that, no different [10:44] I'm running a ping to 8.8.8.8 [10:44] and it'll suddenly just work for about 10 seconds, and then stops again [10:45] disabled wifi - looks better now [10:45] could be some loop on my internal network [10:46] but I doubt it, since everything else works [10:49] Laney: seb128: ah, excellent! I did http://pastebin.ubuntu.com/10202733/ as the bandaid solution to the rest of the about page if there wouldn't be a real fix [10:50] (in the ppa) [10:53] so if that branch helps it a bit further of not disabling the storage portion (even if wrong numbers), that would be better [11:41] mlankhorst, could we, if we wanted, interpret touch events as left-clicks in Xmir? === alan_g is now known as alan_g|lunch [13:03] seb128, Trying to pair that kbd... "Please enter the following pin...." but no pin is shown [13:04] willcooke, what device? [13:04] seb128, N7 running devel prop [13:05] willcooke, yeah, bluez is buggy on that device [13:05] https://bugs.launchpad.net/ubuntu/+source/bluez/+bug/1421598 [13:05] Launchpad bug 1421598 in bluez (Ubuntu) "DisplayPasskey() "entered" number wrong on some devices" [Undecided,New] [13:05] willcooke, https://code.launchpad.net/~seb128/ubuntu-system-settings/bluetooth-key-hint-workaround/+merge/249622 as a workaround [13:06] seb128, nice one! [13:06] seb128, will that land in devel proposed at some point, or do I need to *shudder* compile code [13:06] oh [13:06] Mirv, k, would be better to not disable storage indeed, are you looking at a better fix? [13:06] it's qml [13:06] I can just copy it? [13:06] right [13:06] it's one line to change [13:07] so yeah, you can even vi edit it [13:07] need to set your device to rw first though [13:11] seb128: I'm EOD after this hangout, but maybe Laney knows what's the next step that should be done to his MP now that the button + free space reporting is possible to get back. I guess implementing/copying QML class that used to be StorageInfo? FYI, here's the current diff I have http://pastebin.ubuntu.com/10204250/ [13:11] Mirv, thanks, I guess we can look at doing better on monday [13:16] seb128, can you give me the path to DisplayPasskeyDialog.qml? [13:16] seb128, ignore [13:16] willcooke, dpkg -S DisplayPasskeyDialog.qml [13:17] oh, neat trick, thanks [13:17] yw! [13:19] yay! works [13:19] thanks seb128 [13:19] willcooke, great ;-) === MacSlow is now known as MacSlow|lunch [13:44] seb128: yes, this solution at least fulfills the minimum level, and the main problem is probably thinking what's wise to do, the missing functionality itself is probably easy enough to replicate [13:44] Mirv, right === alan_g|lunch is now known as alan_g [13:46] I'd like some advice please. [13:46] I'm the Ubuntu MATE lead. [13:47] I've got a couple of files ubuntu-mate-settings that get installed to /etc/skel, just as other flavours have done. Such as Xubuntu. [13:47] Just want to get some feedback on that practice? [13:47] flexiondotorg_: skel is evil. stop doing that :) [13:47] Because I'm being asked by my sponsor to check with the desktop team. [13:48] desrt, Well, for I am using it for there is not other option. [13:48] flexiondotorg_: is there not a way that you could make changes to the affected software to have the correct settings by default? [13:48] flexiondotorg_: what are you doing with it? [13:49] desrt, Provide a default configuration for Tilda and qt4-config. [13:49] Much as Xubuntu currently do. [13:49] because, almost certainly, it's not impacting all of the people that it ought to be, and is improperly impacting some of the people who it shouldn't be [13:49] Well, it is only relevant for system installs or new user additions. [13:50] what can't qt by configured via more normal means? [13:50] Normal means? [13:50] like xsettings, or so? [13:50] ie: why are the needs of mate different here than the needs of gnome or unity? [13:53] Well, I guess my needs are the same as Xubuntu. [13:54] sure... but that's not the question... in what way are they different than gnome/unity? [13:54] The default Qt4 look and feel doesn't fit and therefore requires a little tweak. [13:54] it could be that xubuntu is also doing it wrong [13:54] desrt, Possibly. But I have tried to use the existing flavours as a guide. [13:54] what's the tweak? is it a new theme, basically? [13:55] desrt, Basicaly use GTk+ theme and Ubuntu font. [13:55] and this isn't already the default under unity or gnome? [13:55] maybe the fix to your problem is to get the defaults properly changed... [13:56] anything else that you do really needs to be determined at runtime [13:56] or at the absolute worst permissible case, via something like update-alternatives [13:56] So, how does Ubuntu (or rather Unity) set the Qt4 look? [13:56] determined-by-packages-that-happened-to-be-installed-when-the-user's-homedir-was-created is just bonkers [13:57] flexiondotorg_: i'd guess xsettings [13:57] desrt, Which package? [13:57] {gnome,unity}-settings-daemon [13:57] If I have a reference I can follow. [13:57] so probably mate-settings-daemon to you? [13:57] I'll take a peek. [13:57] desrt, Indeed. [13:57] i'm not sure if this is true or not... [13:58] i'm wildly guessing at this point [13:58] * desrt is not a qt guy [13:58] i just know that it's surely possible to do a better approach than your skel thing, and the person who flagged it as suspicious was absolutely right to have done so [13:59] even a dpkg-diversion on the relevant system config file would be a better option... (but still a bad one) [14:01] desrt, There is not system level configuration AKAIK. [14:02] there always is :) [14:02] sometimes they just put it inside the .c files.... [14:02] desrt, Which I can't influence. [14:03] but i'd be surprised to hear about something that could be changed via a dotfile and not by a similar file in /etc [14:03] flexiondotorg_: why not? [14:03] this is #ubuntu-desktop [14:03] this is where changes like this happen all the time... [14:04] I'm trying to see if Ubuntu have already changed the Qt4 config. [14:30] desrt, While I continue looking for if/how Qt4 is being tweak by Unity/Gnome could you take a peek at the following please? [14:30] https://bugs.launchpad.net/ubuntu/+source/mate-control-center/+bug/1351890 [14:30] Launchpad bug 1351890 in mate-control-center (Ubuntu) "Changing external screen resolution with dual monitors" [Medium,Confirmed] [14:30] See the patch we've prepare in #9 [14:30] How can we get this merged and released? [14:32] flexiondotorg_: you're talking to the wrong guy, unfortunately [14:32] desrt, Do you know who I should approach? [14:33] oh. this is a gtk patch? [14:33] i'd talk to Laney about that [14:33] desrt, gtk patch yes, from 2010, which apparently didn't get applied to gtk2 [14:33] desrt, would be nice to have applied upstream as well? [14:33] * desrt thought it was a mate issue [14:33] seb128: i doubt anyone would get very much objection from the gtk maintainers about that [14:34] desrt, https://bugzilla.gnome.org/show_bug.cgi?id=635380 to apply to gtk2 apparently [14:34] Gnome bug 635380 in Backend: X11 "gdk_event_apply_filters is unsafe against changes in filter list" [Normal,Resolved: fixed] [14:34] flexiondotorg_, I guess you can ask on the upstream bug if they can apply to gtk2, or maybe open a new bug about it on bugzilla [14:34] er... https://git.gnome.org/browse/gtk%2B/commit/?id=323df2b2800383832ed3c2e43626f2c6821c33ec ? [14:35] oh. you mean it didn't get backported [14:35] desrt, Correct. [14:37] non-trivial backport there... [14:37] The patch is the TK2 backport. [14:37] considering it modifies a file that doesn't even exist in gtk2 [14:39] you want to ping in #gtk+ on irc.gnome.org [14:39] they'll almost certainly be very happy to accept that patch [14:40] not really sure who monsta-mint is ... :) [14:40] desrt, monsta is a MATE dev. [14:40] desrt, Like myself. [14:45] willcooke: maybe [14:45] why? :P [14:50] didrocks: lovely! [14:50] didrocks: I get nice fsck progress (with the dummy) now *hug* [14:50] didrocks: but neither in qemu nor on my real laptop Control-C works -- does that part need a plymouth fix? [14:52] well, something for the next upload [15:03] didrocks, congrats on getting some of that landing ;-) === MacSlow|lunch is now known as MacSlow [15:31] mlankhorst, just playing with it on my touch screen laptop - I think it'd be a nice tweak if it's something that's easy to do [15:39] pitti: excellent! Control-C on the real laptop should work, the thing is that you have to press for ~1s [15:39] pitti: that was less noticeable with raw C [15:39] seb128: *finally*, phew! :) [15:39] didrocks: ah! I suppose I didn't do that [15:39] didrocks, :-) [15:39] * didrocks back from cycling [15:40] didrocks, ready for the tour de france? ;-) [15:40] seb128: ahah, seeing how many cramps I have the day after cycling for 30kms… I would say… no :p [15:40] and 15kms/h is like 4 times less than tour de france ;) [15:40] didrocks, Laney, btw I opened https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1421672 and https://bugs.launchpad.net/ubuntu/+source/ubuntu-ui-toolkit/+bug/1421661 [15:40] Launchpad bug 1421672 in ubuntu-ui-toolkit (Ubuntu) "Overwrite the gettext to MainView.applicationName if already set" [Undecided,New] [15:40] Launchpad bug 1421661 in ubuntu-ui-toolkit (Ubuntu) "The applicationName documentation doesn't state that it define the gettext domain" [Undecided,New] [15:41] seb128: excellent! let's see how it goes (keep us posted, I'm interested ;)) [15:41] just as a fyi [15:41] didrocks, sure [16:15] beurk beurk [16:22] switching VTs seems to kill the audio for me [16:25] attente_, i thought thats expected since you lose your polkit seat [16:36] ogra_: has it always been like that? i seem to remember it being able to recover as soon as i log into the vt [16:36] i was complainin aboout it years ago ... that was pitti's explanation i got :) [16:37] not sure if that behavior had changed later and is now broken again [16:37] it could well be [16:37] attente_, desrt trying to run gedit on a device: http://paste.ubuntu.com/10206929/ [16:38]