[03:00] hi [03:00] if anyone is still awake ;) [03:00] /etc/X11/Xsession.d/25ume-config-common_startup [03:01] used to export the the GTK2_RC_FILES environment variables for Gtk+2 widget themes here as follows: [03:01] export GTK2_RC_FILES=/usr/share/themes/proview/gtk-2.0/gtkrc:/usr/share/themes/proview/gtk-2.0/gtkrc.maemo_af_desktop [03:02] oops ...even export GTK2_RC_FILES=/usr/share/themes/mobilebasic/gtk-2.0/gtkrc:/usr/share/themes/mobilebasic/gtk-2.0/gtkrc.maemo_af_desktop [03:02] it no longer does this and i wondered why [03:02] ian_brasil: On Intrepid? [03:02] no hardy [03:03] The theme is set in gconf [03:03] (For both Hardy and Intrepid [03:04] so you do not override in /usr/share/ume-config-common/ume-gui-start in matchbox /usr/bin/matchbox-window-manager \ [03:04] -theme /usr/share/themes/mobilebasic/matchbox/theme.xml \ [03:05] StevenK: ok thx..i will look at gconf [03:05] ian_brasil: It's set in the moblin-applets postinst, for some reason. [03:06] yikes [13:54] hi there [13:54] does someone plan to upgrade bluez-gnome to 0.28 ?If not I can try to ask a Feature Exception [14:14] crevette: Some of us have thought about it, and we're planning to look at the code again tomorrow. Is there something specific about 0.28 that you need? [14:14] no appart to stick to the latest bluez-gnome [14:15] I think it should be pretty dsafe as new changes happens in bluez 4 now [14:15] I preparing a pbuilder env to build it [14:15] crevette: I've built it. Builds fine. [14:15] okay [14:16] do you have a deb to test it ? [14:16] If you want a local copy, I'd recommend basing of the debian packaging, as that applies without any editing. If you use the Ubuntu packaging, you'll need to revert some patches (as the Ubuntu upstream tarball is older) [14:18] ah I didn't look to debian package :/ [14:18] davmor2: ping [14:18] cgregan: pong [14:18] How's your afternoon going? [14:19] Quite at the moment why (think Chris is sounding ominous) [14:19] muahahahah [14:19] yeah, he has this looong list of tasks for you :P [14:20] see how I knew ;) [14:20] heh [14:20] * ogra was joking :) [14:20] davmor2: I was cleaning up the UME case area and the installation area needed updating for Intrepid [14:20] I deleted all the MIC/Xephyr stuff [14:21] But the KVM area needs to be altered so it refers to the correct repos [14:21] Since you put that section together...could you get the info to update it? [14:21] Not too long of a list :-) [14:22] Yes no probs sticking with KVM this round then :) [14:22] Yes...the current images already support KNM boot [14:22] KVM [14:22] So we might even be able to shorten the process [14:23] Like..install KVM...point to latest image from cdimage and go [14:23] I'll have a play with it this afternoon and see what I come up with then :) [14:24] Sweet! Thanks [14:24] Next! now serving.....lool: ping [14:25] :-) [14:27] * davmor2 lool: don't answer it's a trick ;) [14:28] hey...no fair davmor2! :-) [14:28] I have lool's home phone number..so he will have to answer eventually! ;-) [14:30] I also have a man inside his house I bribed with soft cuddly things.....he is still in diapers, but I'm sure he'd back me up. [14:30] :D [14:43] cgregan: Ah I don't see the pings in the form "blah lool: ping"; try "lool: ping" to get my attention ;) [14:43] cgregan: How may I help? === ogra_ is now known as ogra [14:47] lool: foolish mortal [14:47] Yup, that's me [14:47] you've fallen for his trick [14:47] Either that or the guy with the diapers would have taken over my office [14:49] :) [14:58] cgregan: just a quick double check the latest images are the ones in mid correct? [14:59] davmor2: correct [14:59] * persia updates the /topic [14:59] np's [14:59] Oh, I already did that :) [14:59] lool: Do we have a list of what apps are being ripped out or replaced in Intrepid? I do not want to get overzealous with my wiki updates [15:00] cgregan: We didn't build such a list; I think we might rip down more apps down the road to the release [15:00] cgregan: We mostly care about functionality such as chat, web browsing, email [15:01] lool: And they are moving to the mozilla stack? [15:02] cgregan: hmm why is that? [15:02] Well...we had that discussion of TB [15:02] figured there might be a move to Firefox as well [15:02] * davmor2 download the latest image [15:02] cgregan: The current images use midbrowser, TB, and pidgin. [15:03] Firefox was there, but it's a beast on smaller hardware, and not very UI-efficient. [15:03] (so we pulled it today) [15:03] persia: Ok....was wondering [15:04] cgregan: Understandably: it's been an effort to get anything even vaguely inspectable, so most of the changes have only been visible on the seeds, which isn't very visible. [15:04] Please try an image: if there's an app you want added or removed, it's worth discussion. [15:05] persia: Will do [15:05] thanks all [15:08] i downloaded the current image last night and i can get to the UI in kvm but there X freezes..is this known issue or shall i file a bug? [15:09] freezes ? [15:09] how ? [15:09] ian_brasil: It's actually not frozen, it's just not accepting any input. [15:09] there is a known bug with X starting before hal [15:09] right [15:09] There's a race condition with starting X. Switch to a different VT, and run `sudo stop session; sudo start session` to get it working. [15:10] This ought be sorted in the next image. [15:10] persia: cool, thx [15:12] images are created daily right? [15:17] ian_brasil: There is an automated process to create them daily, and sometimes additional images are created if there is some reason it's worth while. [15:18] There are two images for 20080912 for example, because 20080912 had a build failure, so 20080912.1 was created to test some other things. [15:21] persia: ok cool..there seems to be lots of new goodies to play with in the image anyway :) [15:22] ian_brasil: You mean that it works? [15:22] Also, which image do you have? 20080911, 20080912, or 20080912.2 ? === njpatel is now known as njpatel_at_docto [15:30] i mean that i can see the UI with what looks like lots of new stuff to play but it is not working no..i have 20080912.2 [15:30] if this is current on http://cdimage.ubuntu.com/ubuntu-mid/intrepid/current/ === njpatel_at_docto is now known as njpatel === asac_ is now known as asac [16:46] using ogras script i get this error [16:46] cp: writing `/tmp/image/casper/filesystem.squashfs': No space left on device [16:46] How much extra stuff did you add? [16:47] i just did a dist-upgrade [16:47] There's only about 20G free, minus the preseed files, the kernel, and other miscellaneous bits. [16:47] Hrm. That oughtn't make it that much bigger :( [16:48] Maybe do an `apt-get clean` afterwards? It may be that your /var/cache/apt/archives directory contains a bunch of files. [16:49] ok [16:50] If that doesn't work, you'll need a bigger VFAT in which to put it. This can be done, but it's a little trickier. [16:52] persia, 20G ? [16:52] you mean s/G/M [16:52] Erm. 20M ! [16:52] we should make that 100M [16:52] * persia probably needs to be a little more diurnal [16:53] I was thinking we might hardcode at 510M so we can know we fit on 512M sticks, and then try to have as much space left over as possible. [16:53] where are we at not ? [16:53] *now [16:53] 480M ? [16:54] 20080912.1 is 570M, but it's too big. [16:54] woah [16:54] Yeah. Needs more trimming. [16:54] my last downloaded one has 419M [16:54] 2008-08-28 though [16:55] I suspect that doesn't have the installer. Note that we're currently shipping fvwm1, so there's plenty of stuff we neither want nor use to drop. [16:56] fvwm1 o_O [16:56] who seelcted that ? [16:56] Also, it's worth considering if we really want to ship *all* of openoffice.org [16:56] I've not tracked it down yet: it's not seeded, but comes from the package relations. It's surely a bug. [16:56] ah, i thought it was the WM we use [16:57] what happened to trebuet ..the open office loader? [16:57] seems not to be in the archive [16:57] ian_brasil: Currently we're shipping both treb and oo.o. [16:58] ian_brasil: From what I understand, this isn't entirely intentional, but just a side effect of some other changes that happened in intrepid that also affect us. [17:00] ogra: If you have time, take a look at http://cdimage.ubuntu.com/ubuntu-mid/intrepid/20080912.1/ubuntu-mid.manifest to see if anything else strikes you as odd. [17:00] I'm sure you'll end up with similar things for -mobile. [17:00] once i have images i'll do a huge cleanup [17:00] Can you not construct images still? I know they aren't automated, but what's blocking local creation? [17:01] though i want to stick as close o -desktop as i can with -mobile [17:01] to much other stuff on my todo ... i'll probably build one on the weekend [17:01] In that case, I think you'll want to create a CD image, and use evand's tool to generate USB images. [17:01] i need to prepare the patches for steven as well [17:02] Should be minor: it's just another flavour, and not arch-specific or anything. [17:02] right [17:02] but i dont want a CD image [17:03] Well, livecd-rootfs doesn't generate a CD image: it only generates a squashfs. Wrapping that into a VFAT image is fairly easy (as your scriot demonstrates). [17:03] right [17:03] but i want to use the same tool [17:03] I suspect it's a 5 line change to livecd-rootfs to allow generation of -mobile images, after which you just need a wrapper. [17:03] right [17:04] wrapper -> one if or case statement to stevens build script [17:04] to use a different size for the vfat and the other names [17:05] Actually, if you're planning to update that script, I'd recommend just appending, or doing a loop, rather than using if or case. [17:05] Unless you think you'll have a significant use case for a different frequency of image generation. [17:05] i dont [17:05] but i will need a bigger vfat [17:06] It's currently calculated as squashfs size + constant, so you get that for free as long as nobody decides to take my thoughts about 512M sticks seriously. [17:07] i'd like to make 800M ones [17:07] so people can esily modify [17:07] i dont expect my squashfs to be bigger than the liveCD one though [17:08] I'm tempted by your idea of just increasing the buffer to ~100M or so, rather than using hardcoded sizes. [17:08] if that works ill enhance my image editor to a synaptics wrapper ;) [17:08] Probably good for both images (although downloading 100M of empty space will annoy some people) [17:08] so you can just call it with --gui and have synaptics spawned inside the squashfs [17:09] Are you really expecting a lot of people to need image modifications? Why not just install it? [17:10] well, its only me toying around ... but synaptic is inside the image anyway, so why not have a point and click option [17:10] its tirvial :) [17:12] I guess, but I think you're building the best buggy-whip ever with that sort of thing. Cool, but ultimately not that useful. [17:12] I agree. 100M of empty space is excessive. [17:12] 30M *maybe*, but not 100M [17:12] Essentially, the only use case I see for modified images is for 1) more aggressive preseeding for factory installs or 2) adding a couple extra bits to get networking post-install. [17:13] rebranding is another one [17:13] there are at least three hackish solutions to modify the live iso in a gui way [17:13] ogra: One feature I'd like to see in the script is an apt-get clean: it allows users to apt-get update rather than rsyncing a new image. [17:13] if possible i'd like to prevent that [17:16] yeah, apt-get clean should be added ... [17:16] feel free to change in the bzr branch :) [17:18] People keep encouraging me to do stuff instead of pointing stuff out ... [17:18] :) [17:19] And the wiki too? Is that up-to-date? Have you set it to automatcially pull from bzr head? [17:19] Where's the code in bzr? [17:19] https://code.launchpad.net/~ubuntu-mobile-dev/ubuntu-mobile/mobile-scripts [17:20] image-tools subdir [17:20] can the wiki pull from bzr ? [17:20] * ogra didnt know [17:20] i wanted to simlply point to LP to the branch [17:21] I believe moin can include an arbitrary URL [17:21] though i think lool wanted to have the scripts packaged at some point [17:22] Might as well toss that one in ubuntu-dev-tools, no? [17:23] i think he meant to have a ubuntu-mobile-scripts package once we have a collection in that branch [17:23] or ubuntu-mobile-dev-scripts [17:23] Do you think we'll have enough of them? [17:24] stevens build wrapper [17:24] loic has some casper modifying scripts [17:24] i might add some to mine as well [17:25] actually i'd like to have various frontend scripts using the current one for different modifications [17:25] I think the build wrapper belongs in livecd-rootfs along with the CD build wrapper, although given the timing, I don't mind it not being there for intrepid. [17:25] well, not for intrepid anyway [17:25] for now we have the branch and should go on collecting them there [17:25] I think the casper-modifying scripts belong to the set of lore devoted to modifying liveCDs (and live images), rather than being mobile specfic. [17:25] in jaunty we should review the branch and do a cleanup [17:26] Well, I'm not going to complain about revision control and public source :) Still, I don't think it ought be collected as a package. [17:26] lets see [17:26] nothing we need to decide now :) [17:27] http://paste.ubuntu.com/46294/ look right to you? [17:28] i would put it into a separate chroot call and line [17:28] for better readability [17:29] oh [17:29] heh [17:29] yes, looks good to me [17:29] * ogra saw a - where none is [17:29] i should stop for the day [17:29] susie is also starting to look unpleased ... [17:47] I am trying to turn accessibility on in gconf but it is not working [17:47] root@lawrence:/# gconftool-2 --type bool --set /desktop/gnome/accessibility/enable true [17:47] root@lawrence:/# gconftool --get /desktop/gnome/interface/accessibility [17:47] false [17:47] any ideas? [17:47] ian_brasil, do you try that in the chroot ? [17:47] that wont work ... [17:48] gconftool expects a running gconfd [17:48] i am inside the chroot [17:48] if you want a system setting use a file in /usr/share/gconf/defaults/ [17:49] echo '/desktop/gnome/accessibility/enable "true"' >/usr/share/gconf/defaults/65-a11y [17:50] then run update-gconf-defaults in the chroot [17:50] ahyway, really out now ... susie moans ... [17:52] ian_brasil: You'll really need to do that in the booted system, rather than in the image. [17:52] persia: why is that? [17:53] ian_brasil: Because you're not running the daemons in the image, so you don't have the right environment for the tools. I suppose you could start and stop the daemons manually, or manually adjust the gconf settings DB, but neither of those are at all easy. [18:16] persia: the apt-get clean worked and i have a new image..the apt-get dist-upgrade has also solved the X problem [18:16] ian_brasil: Excellent news. I'd tested that locally, but hadn't gone back since it was uploaded. [18:16] What HW are you using? [18:17] i am using kvm [18:18] Oh well. Usability testing only then. I've yet to hear of anyone testing the new images on HW other than the Samsung, and was curious how the new Xinput stuff worked for special keys and the like. [18:19] I know the X team has been focusing on getting all the special function keys working for laptops, but I've not heard of much effort to set the right keymappings for MID-type function keys (not that there are so many MIDs available: mostly UMPCs and netbooks) [18:22] ha ha..clicking on the terminal brings up time and date! [18:22] For me it was launching Thunderbird. There's something funny with the terminal. [18:23] One workaround is to use ctrl-alt-1 and then type "sendkey ctrl-alt-f2" to get a different VT. [18:23] (or maybe it's ctrl-alt-2 on the host keyboard : I forget exactly, as I usually use -monitor stdio) [19:57] persia: what is -monitor stdio ? [21:47] Will Ubuntu mobile work for HTC P4350 smartphone? [21:50] kadakas: from the FAQ link at the top [21:50] Ubuntu does not currently provide ARM binaries for any flavour. Although the Ubuntu MID flavour currently has some optimisations for the "lpia" architecture, one of the goals of the MobileTeam is to develop architecture-neutral solutions. Some people and groups have compiled portions of Ubuntu for ARM processors with a fair degree of success, but there is no firm expectation that any Ubuntu flavour will work on an ARM-based devi [21:50] ce at this time. [21:51] thank you [21:51] ill return this subject in a year :p [22:11] has anyone tried running this on the eee pc? [22:16] chuck, see #eeepc too [22:34] bspencer_ works at intel?! [22:35] chuck is that you? the /real/ chuck? [22:35] yes, i'm chuck norris [22:35] * bspencer_ quickly runs "Who is"