[00:01] and the code is not really trivial [00:02] so, page_size -> 0, and keep page_increment? [00:03] I would do that yes [00:46] seb128: bug 267892 if you would like to sponsor tomorrow [00:46] Launchpad bug 267892 in gnome-system-tools "[time-admin] unable to manually adjust the clock past 13:49:49" [High,Triaged] https://launchpad.net/bugs/267892 [00:46] james_w: right, tomorrow, I was just about to go to bed ;-) [00:47] thanks [00:47] thank you [00:47] and sleep well :-) [00:48] thanks, you too [03:20] hey need help loading ubuntu to my dell [08:07] hi === BugMaN1 is now known as BugMaN [09:21] morning [09:25] lut huats [09:25] hey mvo [09:27] hello seb128 and mvo [09:36] hey seb128, huats [09:50] mvo: bug #268636 [09:50] Launchpad bug 268636 in gnome-control-center "gnome-keyboard-properties crashed with SIGSEGV in g_cclosure_marshal_VOID__VOID()" [Medium,Confirmed] https://launchpad.net/bugs/268636 [09:51] mvo: is that fixed in the new version you uploaded yesterday or is that something else that you changed? [09:51] gicmo: hey hey [09:51] hey [09:51] yo mvo [09:51] long time no see [09:52] gicmo: waouh, alex just commented on a gvfs bug ;-) [09:53] gicmo: gnome bug #551337 [09:53] Gnome bug 551337 in general "gedit crashes browsing fileselector ssh locations" [Critical,Unconfirmed] http://bugzilla.gnome.org/show_bug.cgi?id=551337 [09:54] seb128: maybe he is slowly getting back ... [09:54] would be so awesome [09:54] gicmo: looks like, he wrote that he's going to work on a fix for this one [09:55] wohooot [09:55] ;-) [09:55] gicmo: did the dbus trash crash thing go somewhere yesterday? [09:55] gicmo: I'll mark the bug as a blocker I think otherwise GNOME 2.24 will be crashland [10:04] yeah [10:09] man, flash is so broken here [10:09] -here [11:07] seb128: neither swfdec nor adobe's thing work [11:13] seb128: I still get the cups error [11:13] grrr [11:13] which one? [11:14] E [11/Sep/2008:12:12:56 +0200] CUPS-Add-Modify-Printer: Unauthorized [11:14] wtf [11:15] gicmo: try asking on #ubuntu-devel maybe [11:21] gicmo: could you update http://bugzilla.gnome.org/show_bug.cgi?id=547568 to reflect what the issue is? [11:21] Gnome bug 547568 in trash backend "gvfsd-trash crashed with SIGSEGV in g_main_context_dispatch()" [Critical,Unconfirmed] [11:22] gicmo: and comment on the bug [11:23] sounds like a plan, although I am not quite sure, further investigation needed [11:24] gicmo: code freeze is on monday, the bug is milestoned for 2.24, just write your though and maybe alex or mclasen or davidz can have a look too [11:25] yep [11:25] I told it davidz already === mvo__ is now known as mvo [11:27] ok [11:27] seb128: adding the printer worked through the web interface [11:28] weird [11:30] it asked me for u/p though [11:58] seb128: do you know more about the new gconf system mechanism? that was added very recently in gconf. it allows system settings via dbus/policykit. I played with it and have code that uses it, but when I try it on a new user it seems to be not used [11:59] mvo: no I didn't look at it, do you know where is writes the changes? [11:59] i.e. /etc/gconf/gconf.xml.system is created and looks ok [11:59] but its not read [11:59] mvo: did you unset the gconf user key if there is one? [11:59] and the permissions are funny (0600) [11:59] eh, 0700 for the dir I mean [12:00] mvo: ah, this dir is not in /etc/gconf/2/path [12:01] mvo: the debian/default.path needs to be updated, good catch [12:02] hum [12:03] not sure that's required now though [12:03] defaults/gconf-defaults.c lists it explicitly [12:04] hm [12:06] seb128: it seems to be needed here, I added it into the path that now it is fine, without that, it does not work (permissions had to be corrected too) [12:12] meh, both firefox and epiphany do not honor the http_proxy env? [12:34] seb128: I have a close look at the gconf stuff now and see what I can do [12:43] seb128: if you don't mind I upload a new gconf that adds /etc/gconf/gconf.xml.system and that also adds it to the default gconf2 path? [12:55] mvo: What is it for? [12:55] ah sorry read backlog [12:56] mvo: We forked the gconf path settings to add our own Debian/Ubuntu pathnames; I guess that the upstream default path now lists this system path and we forgot to add it to the Debian/Ubuntu fork of the config [12:56] mvo: So yeah definitely a good idea to merge these changes in our fork [13:21] lool: ok, thanks [13:26] mvo: feel free to do any change you want [13:27] lool: apparently they didn't but that might be a mistake, the patch comes from fedora and they might have their own path too and didn't update the upstream one [13:27] Yay for fedora [13:27] thanks seb128 [14:00] vuntz: hello [14:07] seb128: allo, wie geht's? [14:07] ja, ja, genau [14:08] * vuntz is willing to speak german again [14:10] vuntz: guten tag [14:10] vuntz: http://bugzilla.gnome.org/show_bug.cgi?id=551818 [14:10] Gnome bug 551818 in clock "wrong adjustement limit in the clock applet" [Normal,Unconfirmed] [14:11] hi [14:11] does anyone know where can i find the code of the "Log Off" applet? [14:12] seb128: did you know? [14:12] nxvl: what applet? what ubuntu version? [14:12] seb128: the Quit applet [14:13] nxvl: that's not an applet, what do you want to know exactly? [14:13] seb128: feel free to commit :-) [14:13] seb128: the applet where you click and have Restart, Shut Down and stuff [14:13] nxvl: or rather the applet is a button calling a dialog [14:13] nxvl: still not a clear question no [14:14] ok, let me take a screenshot, une sec [14:15] nxvl: do you want to speak about the applet or the dialog? [14:15] nxvl: and what ubuntu version are you using because that changed in intrepid [14:15] nxvl: the screenshot is not really useful there [14:16] * tedg hides :) [14:16] hey tedg, why do you hide? ;-) [14:16] tedg: did you change it? [14:16] I'm guessing he's meaning the FUSA applet. [14:16] ah [14:17] Though, I'm not sure. [14:17] tedg: did you fix the crasher I pointed the other day btw? [14:17] seb128: http://people.ubuntu.com/~nxvl/Quit_Button.png [14:17] speaking of the quit dialog (from the menu), is it supposed to count down 60, 50, 40, 30, 29, 28, 27, 26... ? [14:17] I think so, but I need to get some feedback on PPA versions before I can make an "official" release and try a new live CD. [14:17] seb128: ok, better i tell you what i'm specifically looking for [14:18] andreasn: yes, talk to vuntz about it if you don't like it, I think the idea was than changing every second is somewhat stressing, I find it confusing too and though the counter was not working [14:18] seb128: i want to write and automates stress test for a machine, and what i want is it to reboot several times in a row [14:18] tedg: intrepid is an unstable version it's there to get user feedback, no need to use ppa before uploading to intrepid usually [14:19] seb128: Well, for those of us who aren't core-dev the PPA is much much faster ;) [14:19] seb128: and i din't want to use 'sudo reboot' since it will ask for my password and i don't want to write it all the time, so i was looking for the code of that dialog to check what is it doing [14:19] seb128: I don't have a strong opinion really, but my first reaction was that it was broken too. [14:20] tedg: right but don't hesitate to subscribe the sponsor team so those get actual testing, ppa will give you testing nowhere near intrepid [14:20] seb128: but yes, let me get mpt's review this morning and I think it's ready. [14:20] nxvl: again what ubuntu version are you using, that changed in intrepid [14:21] seb128: oh, sorry, hardy [14:21] the session dialog uses the gdmflexiserver interface [14:22] you can also look at what update-notifier is doing for the "reboot now" dialogs, mvo can probably point you to the source for that, it's an easy gnome-session interface [14:22] seb128: right i forgot about update-notifier [14:22] seb128: thank you! [14:22] * nxvl HUGS seb128 [14:22] you're welcome [14:57] ... gnome-keyring ... [14:58] mvo: what was the sponsoring trouble with bug 56792 again? [14:58] Launchpad bug 56792 in apt "String consisting entirely of variables has no translation hint" [Low,Fix committed] https://launchpad.net/bugs/56792 [14:58] otherwise the desktop-wise sponsoring queue looks good, good job everyone [14:59] seb128, is the current choice between asking for keyring access every time you launch a program that uses it, vs. never asking at all? [15:00] mpt: it's between asking the first time you run the application and not at all [15:00] first time in the session rather [15:00] That's unfortunate [15:01] The use case is preventing an unauthorized program from accessing your passwords and doing naughty things with them [15:01] pitti: that one is fine I think, I just commited it and did not upload because the change is really tiny and I wanted to check if I have other pending apt stuff first [15:02] If a program has exactly the same file size and hash as it did last time you logged in, it shouldn't need to ask again [15:02] mpt: right, the things is "do users read those dialog or just click allow", knowing that programs asking for a password to do not nice things will probably use a name which will make it looks like a regular desktop application to users [15:03] mpt: in fact I'm not sure about "in the session" now, because we obviously don't get this dialog at every login for network-manager, evolution, etc [15:04] mpt: alright, that's the first time it's used, not only for the session [15:05] which is not too bad I guess [15:05] right [15:05] This seems like it was half-copied from the Mac, which has different problems and different available solutions [15:06] mvo: right; please unsub u-m-s then [15:07] On the Mac it's conceivable that an attacker's Web site could download an application that pretends to be (for example) Safari, which has access to your passwords, so the alert is supposed to protect you in that case [15:07] mpt, seb128: still, it shouldn't ask at all about standard Ubuntu applications IMHO; then the dialog becomes much more useful, since it's something users haven't seen before (if a malicious program wants to access it) [15:08] In Ubuntu that won't happen, because anything you download won't be executable in the first place [15:08] pitti: done, sorry, I was not aware that I can do that [15:08] seb128, mpt: can we reasonably determine which program accesses the keyring? if so, could we suppress the confirmation for programs in /usr/bin and so on? [15:08] mvo: no problem at all, thank you! [15:09] pitti, exactly [15:09] pitti: that's out of the scope of this configure option now and join the upstream bug discussion about improving the behaviour [15:09] so I guess we want to keep the current behaviour for now [15:09] seb128: ok, I will comment on the upstream bug [15:09] and work with upstream on making those dialog showed only for non known applications [15:09] pitti: thanks [15:10] * pitti turns this into an action item [15:10] seb128: WDYT about the approach? [15:10] pitti: looks good to me too [15:12] great [15:41] hum [15:41] anybody looking for an update to do? ;-) [15:41] Ampelbein, pochu: ^? ;-) [15:41] seb128: memememe! [15:41] ;-) [15:42] Ampelbein: http://ftp.gnome.org/pub/GNOME/sources/conduit/0.3/conduit-0.3.14.tar.gz [15:42] k [15:42] thanks [15:42] ooh, conduit [15:42] * pochu is reading the desktop team meeting backlog [15:42] Ampelbein: make it use python-webkitgtk if possible ;-) [15:43] ok, i'll see. [15:52] Ampelbein: http://www.conduit-project.org/wiki/0.3.14 [15:53] Ampelbein: upstream has asked for webkit/gio love for this release. [15:53] Ampelbein: so if you could enable all that that would be awesome [15:53] jcastro: i'll try my best. thanks for the wiki-page hint [15:54] Ampelbein: I'm in contact with upstream if you have questions/issues, just poke me if you need help. [15:54] ok, thanks. [15:56] Ampelbein: if it all magically works the first time around let me know so I can tell upstream we're all set. :D [16:16] anyone here with intel gfx and problems with the screensaver hanging the machine? [16:17] I don't use the screensaver [16:18] I can try though, is that happening every time you use it? [16:18] pretty much, you need to set the screen to blank from the power-save settings [16:18] so screensaver 1min, blank 2min and see what happenss [16:18] -s [16:21] gfx? [16:21] graphics [16:21] it hangs at least with i965 [16:22] I have i965 [16:22] and I found the commit that is the culprit [16:22] ah, so no need to test? ;) [16:22] so once you have confirmed that it hangs, setting vblank_mode=0 in drirc makes it work again [16:22] well, I'd like to know that the workaround works for others too [16:23] the real fix is to update the kernel drm module [16:23] I don't know how to do those things... [16:24] no need to, just try if it hangs, and if yes, copy http://users.tkk.fi/~tjaalton/dpkg/drirc to /etc [16:24] I'll try drm git next [16:24] hmm, if I mount a digital camera and try to copy some photos with nautilus it doesn't work, and even files in my desktop dissapear (until I kill nautilus)... [16:24] tjaalton: ok, let me try that [16:25] pochu: cool, thanks [16:26] I guess if it hangs killing gnome-screensaver will be good enough? I'm copying some things over ssh and I wouldn't like to lost the connection :) [16:30] tjaalton: it didn't hang [16:31] blank screen 1 min, then power saving [16:31] (screen to sleep, I mean) [16:32] ok, no worries [16:32] ssh would work, but killing g-s or X wouldn't help [16:34] jcastro: ok, for conduit i added the following build-depends: python-webkitgtk-dev (>= 1.0), python-gobject-dev (>= 2.15.3), libgnomevfscommon-dev (>= 0.99.7.1) [16:35] jcastro: and their respective packages in depends. [16:36] but somehow the ppa-build-machine seems to be broken. it hangs. [16:37] ok, awesome [16:38] jcastro: i attached the files to bug 269011 . could you check? [16:38] Ampelbein: where? [16:38] Launchpad bug 269011 in conduit "Please upgrade to 0.3.14" [Wishlist,In progress] https://launchpad.net/bugs/269011 [16:38] Ampelbein: looking [16:38] last lines in build-log: [16:38] make[6]: Leaving directory `/build/buildd/conduit-0.3.14/conduit/modules/GoogleModule/gdata' [16:38] make[5]: Leaving directory `/build/buildd/conduit-0.3.14/conduit/modules/GoogleModule/gdata' [16:39] since 15 minutes or so. [16:39] https://edge.launchpad.net/+builds/lemon [16:39] maybe it tries to access online datas === asac_ is now known as asac [16:41] Any final interface design requests from anyone before the User Interface Freeze? :-) [16:45] seb128: and with pbuilder on my local machine i get: pbuilder-satisfydepends-dummy: Depends: python-vobject which is a virtual package (same for python-pygoocanvas, python-webkitgtk-dev and libgnomevfscommon-dev) [16:45] Ampelbein: libgnomevfscommon0? that package doesn't exist in intrepid [16:46] libgvfscommon. [16:46] sorry [16:46] Ampelbein: that's the one you added to the depends [16:46] same about libgnomevfscommon-dev [16:46] and those are GNOME 1 things [16:46] that's wrong [16:46] why did you add those? [16:47] thought they would be needed since conduit-wiki said: We require the latest releases of both python-gobject, and Gobject/GIO/Gvfs [16:47] gvfs != gnomevfs [16:48] new gobject,gio,gvfs means libglib 2.16 [16:48] ie, new python-gobject [16:48] ah, ok. [16:48] by doing a diff between configure.in versions there is no build-depends change required [16:49] you just need to add python-gobject (>= 2.15.3) to the depends [16:49] and python-webkitgtk [16:49] ok. [16:50] so no, build-depends for those needed? [16:50] no [16:50] there is no compilation at build time for python [16:51] so there is no build-depends required usually if the configure doesn't look for those [16:52] ok, thanks again for the info. reuploaded. [16:53] Ampelbein: "+ - added Depend on python-gobject-dev", that's python-gobject [16:53] the control is right that's just a changelog typo error [16:55] just to make sure: if its not indicated by configure.in changes there is no need for adding build-depend? only in depends if (like in this case) a software changes runtime-requirements [16:55] seb128: reuploaded [17:11] Ampelbein: looks good now, thanks [17:12] seb128: but still does not build in pbuilder. but the version in the repos doesn't build either so i guess its a problem with pbuilder. [17:13] right [17:13] Ampelbein: I'm wondering if the pywebkitgtk patch is correct though [17:14] Ampelbein: could you change the debian/rules to install conduit.real rather than patching the wrapper? [17:15] seb128: ok [17:28] seb128: now after looking and searching i must ask: how? i have no idea... [17:31] Ampelbein: edit debian/rules [17:32] seb128: ok. that one is clear. i added cp -f /usr/bin/conduit.real /usr/bin/conduit to the binary-indep: build install - section. [17:32] Ampelbein: after $(MAKE) install DESTDIR=$(CURDIR)/debian/tmp [17:32] ah, pl [17:33] ah, your line is wrong, it tries to modify the system one [17:33] you need to cp $(CURDIR)/debian/tmp/usr/bin/conduit [17:35] ok, so cp -f $(CURDIR)/debian/tmp/usr/bin/conduit.real $(CURDIR)/debian/tmp/usr/bin/conduit - should there be a "backup" of the wrapper? [17:36] no need to backup it no [17:39] I'm away for a bit, bbl [17:39] reuploaded [18:11] gicmo: what do I have to do (if any) after I invoked a org.freedesktop.DBUs.GLib.Async method via dbus_g_proxy_call() ? I understand I need to read the reply somewhere (even if I don't care about it)? [18:24] mvo: you can use _call_no_reply if you want to ignore the return [18:46] Ampelbein: still around? [18:46] walters: thanks, I will try that. my code currently dbus_g_proxy_call() but that seems to work only for the first call, subsequent ones do not seem to work [18:46] I will try _no_reply() [18:50] Ampelbein: an another small change, would be nice to move rather than copy the binary so there is no a duplicate copy installed [18:50] anyway I've to run again, bbl [19:31] seb128: i'm back again. reuploaded now with mv. [19:43] mpt: still here? === pochu_ is now known as pochu [20:07] jcastro: ping? conduit 0.3.14 is now in my ppa in case you want to test. [20:25] Ampelbein: that would be great, link? [20:25] https://edge.launchpad.net/~andreas-moog/+archive [20:25] erm: http://launchpadlibrarian.net/17540789/conduit_0.3.14%7Eppa-0ubuntu1_all.deb [20:25] wrong copy ;-) [20:32] Ampelbein: everything seems to be working [20:37] jcastro: not really. just tried a sync f-spot <-> flickr and got that upon refreshing: http://paste.ubuntuusers.de/391911/ [20:37] I'm doing a picassa sync right now [20:41] jcastro: was my mistake. i forgot to enable dbus writing in f-spot. [20:42] now it works [21:32] hi all. When former Windows users see a menu item called "Add/Remove..." they immediately think it is only for removing software, not adding, since the Windows "Add/Remove" control panel is useful only for removing and useless for adding. [21:32] How can we fix this problem? [21:40] unfo: there is a tooltip help which explicitly states that you can install and remove applications from there. [21:41] Ampelbein: i, for one, never read tooltips. [21:41] Do any of you? [21:41] unfo: when i don't know what an option does, sure. [21:42] Ampelbein: also, have you ever seen the Windows "Add/Remove"? [21:42] it is useless for adding. [21:42] I think most former Windows users think they know what "Add/Remove" does. They think it removes and doesn't really add. So they figure they have no reason to check the tooltip. [21:44] i don't think so. add/remove is self-explaining. [21:55] Ampelbein: really: have you ever seen the Windows "Add/Remove"? [21:55] :) [21:55] sure i have. [21:55] from what i know its possible to add optional components to already installed software from there. [21:56] m$ office for example. [21:56] Ampelbein: yes. But not to add whole new apps. [21:56] unfo: we must not follow bad style. [21:57] Ampelbein: correct. That is why gnome-app-install can add new apps. [21:57] unfo: what do you want the menu to look like? "Add, YES YOU CAN ACTUALLY INSTALL SOFTWARE FROM HERE/Remove applications?" [21:57] how about Get/Remove? [21:58] or, how about showing new users a "Ubuntu Tour"? the first item in the tour could be a screenshot of gnome-app-install and a paragraph of text about what it does. [21:58] or a Tip of the Day? [21:58] or rename the menu item Download/Install/Remove? [21:58] anyone? other ideas? [22:02] Ampelbein: what do you think? [22:04] i'm fine with add/remove... get sounds awkward. the ubuntu-tour would be a good idea, if its not started by default after installation. but i still don't think many users will falsely assume they can only remove software from there. [22:05] Ampelbein: you should hang out in #ubuntu more often. Many many newbies go to a website, click "download" just like they would in windows, then try to extract and compile the tarball themselves. [22:06] Has anyone else here also ever noticed that newbies often do that? [22:06] unfo: that wouldn't change even with the button relabeled. [22:06] unfo: I suggest you report a bug about it with some alternatives. Otherwise your suggestion will get lost [22:07] pochu: i should list all the alternatives in one bug? [22:07] or should i make it a mailing list post instead? [22:08] Ampelbein: i hear your point. Maybe when newbies run "configure" or "make" for the first time it should warn them to try apt-get? [22:08] s/it/Ubuntu [22:09] unfo: there is "add" in the name, if that doesn't suggest you can add... [22:10] seb128: did you read the entire scrollback ever since I joined this channel 40 minutes ago? :) [22:11] unfo: no it's too much backlog ;-) [22:12] unfo: I quickly read over it, microsoft doing stupid thing is not a reason to break ubuntu too [22:12] seb128: please read it :) do you now see why "add" does not suggest to Windows users that they can add? [22:12] oops, mid-air enter collision. s/.*// [22:12] seb128: i redone the changes to conduit and uploaded the updated dsc and diff.gz. jcastro already looked into it and it works for him. bug 269011 [22:12] Launchpad bug 269011 in conduit "Please upgrade to 0.3.14" [Wishlist,Confirmed] https://launchpad.net/bugs/269011 [22:12] yes, because microsoft and dumb and users switch and don't want to think and don't read documentation [22:12] Ampelbein: ok, looking [22:13] seb128: but the fact that users don't read documentation *is* a fact of life. IMO we must deal with it [22:13] s/must/should [22:14] unfo: their don't read title either? [22:14] unfo: then what makes you think the users would read a guidancetour or a warning when running make/configure? they would just think: "damn linux, does not let me install software" [22:14] seb128: what is title? [22:14] unfo: not a lot we can do for users who refuse to think and read [22:14] unfo: add and remove [22:14] there is add written there [22:14] seb128: oh. [22:15] i see what you meant now [22:15] seb128: what do you think of my several suggestions at I made at 20:58 local time? [22:16] Ampelbein: that is true, i hate the windows welcome.exe pointing me to the tour, i always skip it [22:16] and check "Do not show me the Welcome to Windows dialog again" [22:16] unfo: we decided again the wizard or tour some time ago I think, and renaming ... suggest a better option which is not any longer [22:16] Get/Remove [22:16] unfo: the current title is already long in some locale and that makes the menu looks ugly because it gets streched due to the longer title [22:17] add or get [22:17] not a real different if you ask me [22:17] hmmmm [22:17] seb128: maybe "get" implies it will download and install? [22:18] to me it implies it'll download, not install [22:19] IMO it's worth it for the menu to be stretched and ugly in some locales if it prevents even 1% of newbies from trying to compile their new software from source. [22:19] seb128: you don't agree? [22:19] :) [22:20] no [22:20] seb128: i don't know what i can do then. [22:21] I don't think it's worth making the menu ugly in some languages only because some users don't want to read a tooltip or the documentation and are not curious enough to click on an icon [22:21] I think it is. [22:21] any other votes? :) [22:21] well, if users prefer to build tarball than spend 15 seconds to read a tooltip good for them [22:22] we will not lower quality only to please those users who prefer to jump on tarballs rather than read [22:22] seb128: when you install new GUI apps, do you read the docs? [22:22] no, but when I install a system I look in the menus [22:22] we have a system, administration, synaptic package manager [22:22] and you look at the tooltips for things that you understand? [22:22] and an add and remove entry in applications [22:22] :) [22:23] well, when I search to install something I do look at what menu entry allow me to do that [22:23] I suspect most newbies have no clue what a "package manager" is [22:23] and I do read tooltip for the ones where the title is not clear enough to me yes [22:24] seb128: but do you look at the tooltips for things that you think you already understand? [22:24] :) [22:24] maybe not, but I don't find something I open the documentation before trying to build things on a command line [22:24] +when [22:25] @21:22 "lower quality" > plus, when users try to build tarballs, it doesn't just waste time for those users, it also wastes time for helpers in #ubuntu. [22:25] I don't think users are that stupid [22:25] seb128: so ask in #ubuntu [22:25] you are to be a pretty advanced user to start building tarballs [22:25] I am pretty sure they will confirm what I am saying. [22:25] seb128: no, just read "INSTALL" then blindly follow [22:25] it will fail at "looking for C compiler" [22:25] IRC is a pretty focussed minority which is technical enough to join IRC [22:26] seb128: true. [22:26] that's people who think they are clever enough to not read the documentation [22:26] but who don't understand "add" in a menu item [22:26] not a lot we can do about those [22:26] so how do we collect usability data from people who are not technical enough to go on IRC or forums or mailing lists? [22:26] we do usuability studies [22:27] have we done that? are there results online? /me googles [22:27] ie, take an user who never used ubuntu but has some vague clue what a computer is [22:27] sit him or her in front of an ubuntu desktop [22:27] tell him to install an image viewer [22:27] and look at what he or she does [22:27] repeat: have we done that? are there results online? :) [22:27] no, we can't do that for every menu item [22:28] that cost lot of money [22:28] ah I just found a usability test [22:28] http://contentconsumer.com/2008/04/27/is-ubuntu-useable-enough-for-my-girlfriend/ [22:28] that's one user comment [22:28] I would not call that usuability test [22:28] she was sent to adobe.com then she downloaded a tarball. [22:28] seb128: i stand corrected. [22:29] seb128: i found an experiment: http://contentconsumer.com/2008/04/27/is-ubuntu-useable-enough-for-my-girlfriend/ [22:29] that's something the firefox guys work on [22:29]