[00:01] i'll set up a PPA next week,there we can prepare newer packages... currently i'm still a bit busy getting the images built in the datacenter though [00:26] rhp: You can certainly apt-get install on a live image. Be warned that you're limited in the size of our changes, because the overlay to the unionfs lives in tmpfs, so if you want a lot of changes, you'll want to do them to the image itself. [00:37] persia, do you mind if we have a catch up call sometime in the next couple of hours instead of tomorrow? I'll be in Oregon and it might be easier for me tonight. === robr__ is now known as robr [07:43] persia: clear enough. First things first, though. Without wifi I will not get very far. Any idea's why my nic is recognized as wired? In the 'regular' ubuntu releases I never had that problem. [07:45] rhp: It's a different kernel. I'm not sure why it reports it that way for you. I'm finding some oddnesses with network interfaces as well. [07:47] Hm... [07:47] Since I broke my laptop on de mccaslin image last night, I figure I might as well install 8.10 on it. That opens the possibility to create my own image. [07:48] I'll be doing that tonight to see if I get any further. [07:50] rhp: Good luck. Did the mccaslin image not work? [07:52] As soon as I noticed it started screwing up my partition table I shut down the laptop in hopes of being able to restore it (but no)... after that I didn't try again. Might as well though, since it is broken anyway. [07:54] Yeah. It's broken now, but it ought have the mccaslin environment installed if you try again. Sorry for that oddity: it's one of the big things that was changed for intrepid. [07:55] And rightly so... [07:55] Anyway, it gives me something to do tonight :) [08:13] * lool waves [08:15] hey lool. [08:31] lool: Do you know much about pygtk? It seems to be FTBFS on several arches, including lpia, which is blocking a ubiquity build. I've tried the last few versions uploaded, and they all seem to FTBFS. [08:31] (at least for lpia) [08:35] it built on i386 [08:35] It's just the Xvfb breakage again [08:35] Is it? It looked like API issues to me. [08:35] * persia looks again [08:36] I looked at the amd64 build log [08:36] It seems to die in the testsuite [08:36] After some randr warnings [08:36] Ah, yeah, you're right. The things I thought were the cause are just warnings. [08:36] I don't think we die if the ts fails, but it seems make gets confused [08:38] Anyway, I bet seb128 is already looking into this [08:39] OK. It's beta-critical for -mid, as lpia can't be installed until ubiquity is built which can't happen until pygtk has the same version for -dev and lib. [08:41] And indeed he is, judging by -devel :) [08:54] Hmmm. Changing the policy of scrolling gives us 5 icons across not 4 [08:56] Let's do that, as it ought look nicer. [08:56] I was planning on [08:57] I currently have three. Does that match you? [08:57] I think maybe not. [08:57] I'm on a Q1, mind [08:57] That's 1024x600 isn't it? [08:57] I'm getting three across by found down at 1024x600 with vesa-on-psb [08:58] s/found/four/ [08:58] Yeah, 1024x600. [08:58] I'm getting 5 across [08:58] Is the text a little more squashed? Does the same trick work for thunar as well? [08:59] heya [08:59] No, the text looks okay [09:00] In Thunar I'm getting 4x4.7 or so, but having 5 there as well would be nifty. [09:00] Is the scroll policy thing kourou-specific, or a setting that ought apply more widely? [09:03] Kourou specific [09:03] It's a GTK-ism I wasn't calling [09:04] persia: If you want to see, open kourou.py and add 'scrolled_win.set_policy(gtk.POLICY_NEVER, gtk.POLICY_AUTOMATIC)' below the 'scrolled_win = gtk.ScrolledWindow()' line [09:04] I presume I ought restart kourou after? [09:05] Right [09:07] I still get 3x4 for Home, but no scrollbar. [09:08] I get 3x4.2 for All, with a scrollbar [09:08] There ought to be one? [09:08] (For Home) ? [09:10] There isn't one: no need to scroll. [09:11] There was one before I changed the policy. [09:12] * persia is running today's image, so may be a few hours out of date, and has no working networking, despite a variety of devices that are intended to provide a network [09:12] persia: That's the behavior I was intending [09:13] Then it works, but I'm not seeing more icons. I wonder if there's something funny about my resolution [09:13] * persia reads the logs [09:15] Ah. Vesa decided to use 800x600 for some reason. That's the source of my problem. [09:18] 800x600 still fitting 12 icons for Home makes me happy [09:19] Yeah. It's a nice bonus. Personally, I'd like full resolution, but I'll settle for it not being broken :) [09:35] any plans for ubuntu mobile on arm eabi yet? [09:39] samjam: It depends on Ubuntu generally. Currently Ubuntu doesn't have any builds for armel. [09:40] thanks [09:40] samjam: Unless the practice changes, you can probably expect to find a recompilation of intrepid for arm at mojo.handhelds.org sometime post-release, and you can see if the Ubuntu Mobile build works, but I don't think it's an explicit target for the mojo folk. [09:40] I didn't know that, thanks! [09:42] samjam: No problem. I've several arm devices myself :) [09:42] I see the Hasty ARMv5EL, I may have to try it. I'm using the HTC Universal, do you have one? [09:43] samjam: Nope. I've Zaurii and kuroboxen. [09:45] (not that kuroboxen are "mobile" in the sense that they work whilst being moved, without heavy modification, but they are armel) [09:55] Argh [09:55] ? [09:55] * StevenK is getting defeated by GTK [10:01] iconview.window.set_back_pixmap(...) gives AttributeError: NoneType object has no attribute [10:01] So iconview.window == None, but why [10:02] persia: I think I've uploaded a fixed pygtk [10:03] lool: I see that, and am very excited. I have a list of 10 FTBFS that I'd like solved pre-beta, and that's probably the most critical. Thank you. [12:05] persia: Dunno who will give back the pygtk FTBFSes [12:05] ubiquity is still in the "Failed to build" state [12:09] lool: There's another issue with ubiquity as well, as it also fails on i386. I expect it will be reviewed in a few hours. [12:33] My. Repeatedly using a USB stick only as a target for dd results in some interesting partition maps. [12:57] persia: synaptic accepted [12:58] Hrm? pygtk rdeps? [13:04] anyone with an opinion on bug 275299 here ? [13:04] Launchpad bug 275299 in ubuntu-mobile-default-settings "ubuntu-mobile-default-settings: better descriptions describing target MIDs" [Undecided,New] https://launchpad.net/bugs/275299 [13:05] * ogra changed the description for -mobile to: [13:05] Description: The Ubuntu mobile system [13:05] This package depends on all of the packages in the Ubuntu mobile system [13:05] Ubuntu Mobile is an alternative to ubuntu-desktop, intended for use on [13:05] Ultra Mobile PCs (UMPCs) [13:05] the description in the bug would make sense for -mid though i think [13:06] No. That's -mobile. [13:06] persia: Err xf86-synaptic that is [13:06] * lool coffee & [13:06] A "MID" is a special Intel word for the class of devices once known as "PDA", but with more RAM, processor, and local storage. [13:06] The description in the bug does sound like mid-default-settings [13:07] lool: Yeah. That makes more sense :) I've been playing with it all day: it's a lot easier to work on a device when the pointer can be controlled. [13:08] I don't like listing storage there. The only device I have with < 10G storage is 4 years old. [13:09] And recommended memory? No. [13:09] It should list the form-factor [13:09] Right, and clearly differentiate the two. As much as anything, it's about DPI: "How big is a finger?" [13:09] right, list touchscreen dependency and screen size [13:09] -mobile works fine without a touchscreen. [13:10] but it is designed with focus on touchscreens [13:10] * persia tried it on the Kohjinsha, and ignored the miscalibrated touchscreen, and was very happy with the result [13:10] that will be more noticeable with jaunty though [13:10] where i.e. gksu will popup cellwriter automatically [13:10] and things like that [13:10] Please don't make it not work with a touchpad, trackpoint, or mouse. I *really* like having the simple, clean interface, but don't want to get fingerprints all over my screen. [13:11] Oh, I'm fine with that, as long as it's done through the accessibility interfaces, and can be disabled by users with touchpads (e.g. Eee users). [13:11] persia: Use your nose? [13:11] or your big toe :) [13:11] Heh [13:12] though we'd have to lower the dpi for that i guess :P [13:12] Bah. *all* the lpia devices I own have perfectly functional synaptics touchpads. I don't see why I shouldn't use them. [13:12] I see a new standard. Toe-driven interfaces [13:12] yay [13:12] the new toe bluetooth keyboard ... at the size of a 37" Tv [13:13] handsfree computing :) [13:13] * persia can toe-type on 15" keyboard. [13:13] ogra: 37 inches? That sounds more like a feet-driven interface [13:14] And you get shifted keys by stomping [13:14] well, your toe or your heel :) [13:14] I like that. velocity sensitivity for alphanumeric keyboards. [13:14] persia, do we have a prper fix for the error 15 bug btw ? [13:14] hmm, and do we actually have a bug number ? [13:14] ogra: Yes. That's sorted. [13:14] ah [13:14] Yes, and I closed it. [13:14] k, no number needed [13:15] I actually opened *two* bugs, one for each flavour, but they turned out to be the same problem, so I dup'd them, and StevenK did the necessary adjustment to fix them. If you build a mobile image somewhere sane, it will work. [13:15] good [13:16] ogra: You need to put the kernel under casper [13:16] StevenK, yeah, i saw that, but wasnt aware thats enough for the installed system [13:17] Yeah, it's a hardcoding in ubiquity. [13:17] ah, k [13:18] * persia fails again at using syslinux, and decides to wait for usb-creator to work [13:19] ?? its trivil to use ... you just need mtools installed and a syslinux.cfg in the root of the target fs [13:20] Yeah. I did that. The result causes a loud beeping noise. [13:22] My procedure was fdisk /dev/sdb; mkfs.vfat /dev/sdb1; mount /dev/sdb1 /tmp/mnt; cp -pRa /mnt/intrepid-desktop-i386/* /tmp/mnt; cp /mnt/ubuntu-mid/syslinux.cfg /tmp/mnt; umount /tmp/mnt; syslinux /dev/sdb1 [13:23] ah [13:23] dont partition :) [13:23] syslinux doesnt work on partitions [13:23] Just mkfs.vfat /dev/sdb ? [13:24] yeah [13:24] * persia tries again, but knows there is supposed to be an easy tool, and wants it to work [13:40] ogra: Thanks. I'm booting the desktop now. [13:40] * persia may do the same to sort out not being able to boot the ubuntu-mobile image [13:57] urgh [13:58] does midbrowser work for anyone with the latest upgrades ? [13:58] asac, ^^^ [13:58] hi all, [13:58] i have been trying to build mobile-basic-flash to use xulrunner-1.9 [13:59] i'v seen all the posts from asac and bob spenser that it's broken [13:59] but those posts are from April time frame I was wondering if someone has better clue how to solve the issue? [14:00] basically what i'm seeing is a black window in hildon UI which is the home.html page; but afterwards nothing; as if the javascript: calls are not being executed. [14:00] rabeeh2000: I don't think anyone was able to come up with a good solution. For intrepid, m-b-f has mostly been replaced with kourou. [14:00] any tip would help. thanks [14:00] ogra: what happens? [14:01] Mind you, kourou doesn't support flash or xul, so it may not meet your needs. [14:01] asac, well, it coredumps pretty quietly if i run it from a termnal [14:01] We tried to move m-b-f to xulrunner-1.9 for Hardy, and reverted it [14:01] ogra: midbrowser hasn't worked for me in about a week. [14:01] persia: midbrowser works for me using xulrunner-1.9 (both x86 and ARM versions) [14:01] well, i just tried it for the first time after installing ubuntu-mobile on the Q1 and upgrading [14:01] (but I've been fiddling with increasingly messy environments, so this may not be indicative) [14:02] its a fresh intrepid install fully up to date [14:02] rabeeh2000: In the Ubuntu Mobile or Ubuntu MID intrepid development snapshots? [14:02] persia: ubuntu mobile from hardy [14:03] ogra: just "core" ... or any other output? [14:03] rabeeh2000: Yeah. It works fine there. [14:04] ogra@ogra-Q1:~$ midbrowser [14:04] pure virtual method called [14:04] terminate called without an active exception [14:04] Aborted (core dumped) [14:04] ogra@ogra-Q1:~$ [14:07] ogra: Following your instructions gets me to an initramfs prompt, because casper can't find the right disk (it's not /dev/sda, and it's not /dev/scd0, but it's /dev/sdb). Any suggestions? [14:07] hmm, note really (phonecall now) [14:10] Oh well. Thanks. [14:25] how can i launch kourou? [14:25] i just built it for hardy; how do i integrate it with hildon? [14:47] stevenk: sorry for buging you on this; but can you please provide a hint how to launch kourou? is it ready at all in demo level? [14:47] rabeeh2000: You need to edit /etc/hildon-desktop/home.conf and home-layout.conf, replacing what's there with the path to kourou.desktop [15:16] persia, i dont get where you get the passwd/auto-login from ... its not documented anywhere [15:17] It's in user-setup-apply in user-setup. ubiquity processes it and flips the checkmark. d-i is used as a prefix so it doesn't get included in the debconf DB on the installed system. [15:19] It ends up doing a bunch of ugly sedery, but if you don't look at the implementation, it works pertty good. [15:25] ogra: Hmm. Looking at gdm, it doesn't seem to have much in the way of debconf questions. Sorry. [15:25] Maybe you could add one for jaunty? [15:27] wel, you said there is a checkbox somewhere for it [15:27] No, I said there was a checkbox for automatic login. [15:27] ah, user-setip ... [15:27] i should read the backlog first :) [15:27] I was looking for something for timed login. [15:27] (or read my mind in this case, as I lost track of context) [15:28] it does timed login [15:28] $log $chroot $ROOT sed -i \ [15:28] -e "s/^AutomaticLoginEnable=.*\$/AutomaticLoginEnable=true/" \ [15:28] -e "s/^AutomaticLogin=.*\$/AutomaticLogin=$USER/" \ [15:28] -e "s/^TimedLoginEnable=.*\$/TimedLoginEnable=true/" \ [15:28] -e "s/^TimedLogin=.*\$/TimedLogin=$USER/" \ [15:28] -e "s/^TimedLoginDelay=.*\$/TimedLoginDelay=10/" \ [15:28] "$GDMCONF" [15:28] great [15:28] no problem then [15:28] Oh, cool. preseed it then. [15:29] you should just have pointed me there and i wuld have kept quiet ;) [15:30] I don't want you to be quiet. It's generally too quiet here. On the other hand, asking me from where something comes may be more useful than asking me about something. [15:32] yeah, 'ill try to ask more intelligently in the future [15:33] i see user-setup even has proper handling for gdm-cdd.conf now :) [15:33] Cool! I wasn't sure if it handled that, but if it does, you should be completely sorted. [15:40] geez [15:40] * ogra just closed all apps since the last upgrade and wasnt aware the default wallpaper had changed [15:41] heh. [15:52] stevenk: tried changing home.conf and home-layout.conf, and also rebuilt hildon-desktop from interpid http://packages.ubuntu.com/intrepid/hildon-desktop [15:53] stevenk: unfortunately this doesn't seem to be good enough; since m-b-f is still being launched [15:53] uninstall it [16:48] * persia has now successfully installed Ubuntu Desktop on a telephone. For those watching at home: this is not what you want. === asac_ is now known as asac [17:04] stevenk / ogra - so it works for me now; only few inputs - [17:05] 1. got a python backtrace on this - bgcolour = gtk.gdk.Color("#000000") --> unclear why python doesn't understand gtk.gdk.Color? gtk version? [17:05] 2. unclear why kourou needs - /usr/share/pdk/pixmaps/image-creator-32x32.xpm to run ? [17:06] 3. if using standard X-windows with a mouse then double click on any icon doesn't work. [17:07] and finally; thanks for saving the day :) [17:07] rabeeh2000: #3 confuses me, as I have had just as much success with mice, touchpads, remote screen access (vnc), virtual mice, and touchscreens to launch applications. [17:07] For #1, you're running this in intrepid, or backported to hardy? [17:08] * ogra guesses 2) is just a hardcoded path in the .desktop file (which is a big nono but wouldnt surprise me with MIC) [17:08] persia: for #1 it's a backport for hardy [17:09] persia: for #3, i'll check it right now with Xephyr [17:09] gtk.gdk.Color is there since ages [17:09] shouldnt be the issue [17:09] probably you need to import gdk separately in hardy [17:11] ogra: for #1; the error message - TypeError: an integer is required [17:11] ogra: i'll try doing that too. thanks for the tip [17:11] well, i think you need gtk.gdk.color_parse() [17:12] not sure though, i avent seen the kouru code but have a nearly identical app here === davmor2 is now known as davmor2_away [17:12] where i use the abov eline instead [17:13] rabeeh2000: Might be a Xephyr thing, but it does work for me for intrepid for real HW and KVM. === njpatel is now known as njpatel_away === davmor2_away is now known as davmor2 === robr__ is now known as robr === mterry is now known as mterry_afk === mterry_afk is now known as mterry [22:37] ogra: does accessibility work with the theme on ubuntu-mobile.img or didn't you try that