[00:16] rhp: something like dd bs-1024 if= of=/dev/sdb [00:16] where /dev/sdb is whatever your usb mounts as [00:17] rhp: dd bs=1024 if= of=/dev/sdb === ALo is now known as ALoGeNo [10:09] rhp: These were the older MIC images; the new images are live images, and shouldn't overwrite your disk === Celtiorus is now known as Celtiore [12:16] lool: cool [12:17] so now they run from the usb stick? [12:18] like a livecd (they are built using the same tools now) [12:18] rhp: Indeed. [12:18] ian_brasil: as far as I can tell, there is no difference between dd and cat > /dev/sdb, but for some reason that does not result in a bootable usb scick. [12:18] nice. [12:19] dd does a binary copy by blocks, cat cats a stream from a to b [12:19] rhp: It's copying bytes vs. blocks. [12:19] * persia stops typing, being inevitably too late and therefore redundant [12:19] ogra-maemo: what is the difference between those? [12:19] I've put images on disks before with cat, without problems.. [12:20] using cat the bootsector wont work, it needs to have a certain blocksize [12:20] Well, rather, it might work, but only if you get really lucky. [12:21] Hrm. I appear to have melted my USB stick :( [12:23] i managed to wear one out recently... the one i used for install tests on the cmpc since the beginning :) [12:23] but it survived 1.5y of nearly daily install tests [12:24] I don't think I wore it out. I just let someone else touch it. My mini-A to full-B cable needed replacement, and I foolishly used my test stick to make sure the replacement cable worked at the shop. [12:26] * ogra-maemo cant belive nobody ported any dvb tools to maemo yet [12:26] (thats my usecase for such an adapter) [12:29] aside from dd-ing it directly to /dev/sdb, would it also be possible to create two partitions on the stick, one for the img file and the other for data files? [12:31] rhp: Yes, but there's no guarantee the result would boot. What I do is to add the couple files I need via loopmount, and they appear on the live system under /cdrom. [12:31] not with the current setup... caspper (the tool used for booting the image) has a persistent mode though, you cold use a second usb key for that [12:31] *casper [12:32] or a partition on the local hdd... look at wiki.ubuntu.com for persistent livecd [12:33] ok, clear enough. putting it on /dev/sdb1 was the first mistake then... [12:33] yeah [12:33] Maybe some info on the dd-incantation would be usefull for the wiki/howto... [12:34] If things work nicely, I can just copy the image to the harddrive? [12:34] Or is there an installer for that purpose? [12:35] http://ograblog.wordpress.com/2008/09/23/announcing-the-first-ubuntu-mobile-image/ [12:35] we uszally give the necessary commandline in the announcements [12:39] and i'm working on a gui tool,but its not ready yet https://launchpad.net/usb-imagewriter [12:41] welll, its functional, but the code isnt production ready [12:42] ok, I was looking through http://wiki.ubuntu.com/MobileAndEmbedded and couldn't find anything about it. [12:42] You're blob is a more authorative source of information? [12:43] blob -> blog... [12:43] no,its my blog :) [12:43] Those screenshots look great, by the way... [12:44] we should put something into the howto or faq section about dd i guess [12:44] (on the wiki) [12:45] i guess at some point we need a mobile team wiki for such announcements [12:46] the image announced there isnt yet built in the datacenter, once that happens i'll do an authoritative annou cement [12:47] (some time nexr week) [12:47] * ogra-maemo notices how different his typos are on an onscreen kbd :) [12:59] Turns out that old stylus 3400 does not support booting of the usb-stick directly. There is grub on the hd, though. I guess it would be easy to boot the usb-stick from grub? [13:00] rhp: You'll need grub > 0.97, but it ought work. [13:02] Also, you need the right hints available. Seems Fujitsu may be annoying this way: if it doesn't work, you might need to do something like http://64.124.13.3/hacks/USB_Boot_using_GRUB.html#mozTocId210830 [13:02] Alternately, you might want a CD image. [13:02] (but i don't know how to convert a USB image to a CD image) [13:05] Why would I want a CD image? There is no CD player in the device... [13:05] Oh. Then you don't. [13:05] (unless you have an attachable bootable CD drive, but you may not) [13:06] On another system I also boot from the installed hd to run the root-fs on a usb-disk. [13:06] So I'm pretty sure I can so something similar here. [13:06] You ought be able to do the same thing. [13:07] The initrd.img probably contains usb modules already? [13:13] It ought, or at least USB stuff works for me. [13:31] ok, it is booting the kernel now, but after mentioning the squashfs info, it seems to stall... [13:34] ... or I just need to learn to have a little patience... I booted just fine! [13:37] At first glance, it looks great. [13:37] Unfortunately, the X input driver for my device is not included, so that'll be the first thing to fix. Trying that in a few hours. [13:40] rhp: Which driver? [13:57] ogra: You're just matching on "IDEACO", right? Not the model or anything? [13:58] i'm matching the model because i apply defaults [13:58] ideaco in a different device needs a new fdi [13:59] Hrm. OK. [14:00] I guess I'll run the calibration tool as soon as I can boot the Mobile image then. [14:01] i'm trying to generalize but i somewhat want it usable on a basic level without any calibration already [14:01] its a tricky situation [14:14] Oh, I understand. It's probably best to identify as many models as possible, and then try to set sensible defaults for all of them, with calibration for unsupported devices. [14:15] At least now I actually have an evtouch device, which means I can produce something useful for you. [14:19] yay [14:20] well, i think the best thing for sensible defaults would be to ship a database and apply them in /etc/evtouch/config from the postinst [14:20] That would make sense. [14:20] but that might be complicated [14:21] yu need to find out the hal device data etc [14:21] Hrm. Maybe the postinst isn't best then: imagine how to make it work in a live environment. [15:29] persia: I believe it is called the fpit driver. [15:30] I was included by default in the xubuntu 7.10 release (as far as I can recall), so it should not be difficult to get it included in ubuntu-mid too. [15:30] Is that the radio pen? [15:30] radio pen? [15:31] No, not likely difficult at all, it's just knowing which package supplied it, and making sure it works in -mid. [15:31] I do not know what that means, but it is a passive pen. I can operate it with my fingers too. [15:31] The package was actually called -fpit, so that's a good clue :) [15:31] My Stylistic 1000 had a radio-based pen, but that was ages ago, so I've no idea if newer tablets use the same tech. [15:31] ah, ok [15:32] Try apt-get install xserver-xorg-input-fpit and see if that works. [15:33] (It may not, but I'm currently having issues checking things) === asac_ is now known as asac [17:19] persia: can I apt-get install on the usb image as it is running? [17:53] ogra: Did you commit po/de/LC_MESSAGES/usb-imagewriter.mo in usb-imagewriter on purpose?!? [18:16] indeed [18:17] anything wrong about that ? [18:41] ogra: Isn't it a generated file? [18:41] there are scripts in the tree [18:41] Anyway, not my business [18:41] i needed a .po for rosetta [18:42] Yeah, but why commit the .mo? [18:42] Anyway, you do as you please [18:44] oh,the .mo... i was only testing the translatability to make sure i didnt miss strings [19:09] ok, I'm up to the point where everything boots fine. However I have no wifi-access (it puts my wifi nic in the wired list...), and the touch screen doesnot work (but I cannot apt-get, since I have no network...) [19:09] Any ideas how to get wifi working? [19:24] Hi, there! The HELP on Ubuntu MID is needed, looking for a team members or proes able to assist, thanks [19:37] ANYONE ALIVE HERE!!!!!!!!!!?????????? [19:41] biz_den6140; no, but I am pretty sure that is not the best way to get help. [19:41] Maybe it would help to state your problem... [19:41] Oh. I'm really sorry [19:41] OK [19:42] Here it goes [19:42] Does MID have locals in other languages? [19:44] biz_den6140: i'm still working on getting it running, then I'll bother running it in dutch :) [19:46] I meant does it has the same way of localization as Main Ubuntu has: like downloading it while installing or something? [20:02] biz_den6140: I think the pro's have gone to bed, it seems pretty quiet now... [20:08] rhp: oops, ok time for me to go too, thanks anyway, I'll try to get help later on [20:14] okido [20:19] hello [20:20] crevette, Hi, there! === davidm_ is now known as davidm [21:59] <[doctor]> hello, can anybody list locales in ubuntu mid? [22:22] ogra-maemo: I tried following your script to update an image, but it seems that my system does not recognize squashfs as a filesystemtype. I tried unsquash instead, but that complains about a too low version. [22:22] What are the minimal system-requirements for running the script? It seems 7.10 is too outdated. === robr__ is now known as robr [22:37] it was developed on 8.10 [22:38] ah, that's... recent. [22:39] unfortunately I broke my laptop today trying to see whether the mccaslin image would have working wifi... didn't remember it would overwrite my harddrive. [22:39] So maybe I should just install 8.10 on that. [22:43] ok, see you all tomorrow. [23:08] anyone know if there is a driver available for the Q1 web cam? [23:48] ian_brasil, uvcvideo would be the right one, but still cant handle it [23:52] ogra-maemo: yes, i saw that it is not uvc standard..incidentally I am not sure you saw this last week but i was playing around learning pygtk and wrote some really simple frontend to your evtouch calibration tool http://paste.ubuntu.com/50238/ ...maybe it is a start (if you did not already write some UI that is) [23:53] cool ! [23:55] i threw together a glade file for the extended options today, but the calibration is still the old script with some zenity/xdialog crap for user intteraction [23:56] lets merge that in jaunty, intrepid is sadly in beta and ui freeze, i'll only be able to get .fdi files in to avoid hw regressions