[03:59] In last night's meeting, we passed over the testing of gthumb vs. f-spot. For the current images gthumb+gphoto2 requires an additional 15.5MB of pre-compressed disk space, and f-spot requires an additional 141MB of pre-compressed disk space. [04:00] 141MB!? [04:00] I'm strongly tempted to use gthumb based just on that (although it sacrifices functionality) [04:00] 141MB. directhex is working to make mono dependency chains smaller, but it doesn't appear to have made intrepid. [04:00] Yes, it was a statement of suprise, not questioning [04:01] Also, we already have a lot of the dependencies for gthumb, for various reasons. -mobile has f-spot, so users with the larger hardware get more features. [04:02] Despite the increasing processing and storage in a modern MID, I'm just not sure we need anything too fancy, as people will want to do real photo processing on some other device anyway. [04:03] The complete list of new pacakges pulled by gthumb + gphoto2 is : gphoto2, gthumb, gthumb-data, libcdk5, libgphoto2-2, libgphoto2-port0, libltdl7 [04:04] We could probably drop gphoto2: that's 928k on-disk for the CLI front-end (I may be in the minority of people who uses shells on MIDs) [04:08] * persia tests gthumb [04:24] gthumb seems to look OK. [04:25] At least I can preview pictures, and flip through them to display them to others. [04:25] * StevenK kicks moblin-applets [04:25] It depends on icons under /usr/share/icons/gnome/32x32/apps/*.png === matt_c is now known as RACCOON === RACCOON is now known as matt_c [06:59] crevette: I'm giving up on bluez-gnome 0.28 again. The changelog looks great. Your packaging works great. The communications and audio portions show no regressions from 0.27. It solves the appearances bugs. I can't verify that it works with any bluetooth input devices. Mind you, I can't get 0.27 to work either, but I'm nervous about any update without knowing it doesn't break something. [08:23] heya [08:23] * StevenK waves to lool [08:24] persia: The uninstallability is still here with the meta kernel in place, and ogra is hit by this as well on i386 images [08:25] lool: Hrm? Which uninstallability? The one from me breaking the seeds, the one from grub missing, or the one from the kernel not being installed? [08:26] I know it's kernel / grub interaction, but don't have the specific error [08:27] lool: OK. It would help to know which error you mean if you say you're seeing it on i386 images. Each has a rather different cause, and only the one about me breaking the seeds (now fixed) should have affected i386. There may also be other issues (I know of two different grub issues that were reported at least once, but remain unverified). [08:30] 23:16 < ogra> according to my users the install works flawless ... but they end up with error 15 after reboot [08:32] That's a different image. I can't get ogra's image to work well for a number of reasons. It looks beautiful if you boot off USB. It doesn't work in KVM. It doesn't boot on either my Core 2 Duo or my Atom Z250. It doesn't install for me, even on the hardware on which it boots, and I get different errors each time. [09:15] Anyone with an Aigo MID about? What's the battery life like? [09:29] amitk: Topic wifi: what was the idea with ath5k fixing for Q1U? [09:29] Or was the plan to get restricted modules? [09:30] lool: since I am a bit pressed for time here, I am considering doing an LRM for LPIA. I would've liked to study why ath5k doesn't work but that can be left for later. [09:30] k [09:31] lool: from beta freeze point of view would it be better to have an LRM package in before the freeze? Or do you want aufs first? [09:32] ...aufs fixed first [09:34] amitk: We are in beta freeze already :-( [09:34] amitk: I think we really want both for intrepid; order perhaps doesn't matter too much [09:34] I'd say the regression potential is higher for aufs, so it would be best to start testing it soonest [09:35] LRM's madwifi is just boring stuff, but we want wifi :) [09:35] I don't see much risk with LRM though [09:35] ack... working on aufs now [09:35] Cool [09:40] crevette: I think I missed you earlier. I'm just not having any success with getting bluetooth input to work :/ If I don't use bluez-gnome, it doesn't work. If I do use bluez-gnome, it doesn't work more, but the not working is the same for 0.25 and 0.28. As I can't test, I'm not comfortable pushing the upload, unless someone can confirm it doesn't break a working setup. [09:40] okay [09:41] can you get in touch with ubuntu dev who can test it ? [09:42] hello persia by the wau [09:42] way [09:42] I'm not sure who has the hardware. I don't think it matters if it's an Ubuntu dev. I've just not heard of anyone who is known to have bluetooth input working, who could be asked to test the update and make sure it doesn't break. [09:42] Oh yeah. Good morning :) [09:42] perhaps a message on bluetooth mailing would do the trick [09:42] ? [09:43] I thought slytherin sent one a couple weeks ago. You could send another. [09:49] * lool didn't know about /+junk [09:49] Cool stuff [09:49] Handy, at least. [10:49] * ogra hopes persia made a list of these various issues [10:49] i'm only aware of one which seems to be that grub doesnt install anything [10:49] OK. #1: doesn't boot in KVM [10:50] #2: doesn't boot on D4 [10:50] D4 ? [10:50] #3 doesn't book on Y7 [10:50] Y7 ? [10:50] Sharp D4. [10:50] ah [10:50] Panasonic Y7. [10:50] doesnt boot as to which point [10:50] * persia couldn't find a good large-format desktop from Sharp [10:50] Doesn't boot as in BIOS doesn't recognise the image as a bootable device. [10:51] ah [10:51] Boots fine on the SR. [10:51] Looks great. [10:51] Can't install. [10:51] yeah, i read things like that on the blog posts on umpcportal [10:51] You *really* want to change the preseed. About the only thing you want to preseed is the automatic login. [10:51] though i wouldnt know why, its a vfat usb image with bootsecor [10:52] Yeah, that's confusing me. Does it work in virtualbox? [10:52] i didnt try [10:52] and i actually cant since i use ose [10:52] I especially don't know why the D4 can't boot off it: I can boot off the -mid dailies just fine. [10:52] (no usb support) [10:52] No disk image support? [10:52] Claim it's an IDE drive when you boot. [10:52] disk image, yes, usb, no [10:53] hmm [10:53] That's what I do with KVM. Works like a treat for other images. === asac_ is now known as asac [10:53] urgh [10:54] newest VB asks for registration ? [10:54] and hangs after that [10:54] ogra: suns great isn't it :) [10:55] * persia likes Sun a lot, but the internal communication paths need some work [10:55] davmor2: Have you tried the -mobile image? Does it work on your hardware? [10:56] bah, and you still cant alt-tab with it [10:56] they should rather fix that than adding reg forms [10:56] Anyway, back to issues: [10:59] The install failed to copy some stuff. I'm redownloading to see if that's my download vs. the manifest files. [11:00] The install also replicated http://people.ubuntu.com/~stevenk/IMG_2090.JPG which makes me very happy. I've planned a liveCD install tomorrow, and if I can replicate there as well, then it's not me. [11:05] very likely caused by the missing task [11:05] that should be fixed in my next build [11:06] OK. Are you populating the manifest files also? The filenames are hardcoded in the ubiquity source, so cannot be changed. [11:06] btw whatever you removed from the seeds didnt make cjwatson happy ... that needed germinate changes as well [11:06] Well, I removed a *lot*. One thing was restored. The failure was in my testing. [11:06] i'm using livecd-rootfs so they should be populated [11:07] but since the task was missing very likely with the wrong content [11:07] Right. That would make sense. [11:08] Yay. redownload finished. Now for reburn, and another install attempt. [11:19] ogra, persia: what are the currently open bugs on installation issues with the mid and mobile images? [11:19] Any other issue which is not filed as a bug yet? [11:20] I want to milestone all of these to make sure all get fixed for intrepid, and I'd like to raise them at the release meeting later today [11:20] i dont have any bugs yet [11:20] will set them up during the day while i do installation testing [11:20] lool: Good point. We'll try to get the installation stuff filed as bugs tonight : at least the tasks, daily building, etc. [11:20] ogra: How many different issues are we speaking about right now for your image? one? [11:21] lool, i guess only one [11:21] might be two [11:21] persia, ogra: please file them asap, meeting is pretty soon ;) [11:21] The fact that the image itself is unrecognised by KVM and several HW devices is likely related to handling, rather than to the code. [11:21] query on MID should it stop at the boot prompt at the beginning? I'm guessing not. [11:21] * persia runs an install test. [11:21] lool, i wont manage before the meeting, simply because i havent seen them yet [11:21] davmor2: It does. Whether it should or not is open for debate :) [11:22] ogra: Do an install :) [11:22] * ogra added a timeout to mobile [11:22] davmor2: No, it shouldn't, but it makes things easier to test :) [11:22] davmor2: But yeah, we should fix that [11:22] persia, i will but i doubt it finishes in time and i need a backup of my Q1 first [11:22] StevenK: I think we want an isolinux.cfg timeout as well [11:23] ogra: What do you have on your Q1 that dpkg --get-selections can't handle? [11:23] ++ [11:23] StevenK: Yeah: having a little menu would be cool too, like the LiveCD has. [11:23] persia, personl data, some movies (ripped DVDs) i dont have backed up etc [11:23] ogra: When do you think we could get dailies on cdimage for mobile? [11:23] ogra: Ah. Makes sense. [11:24] lool, i hope to get that done over the weekend, i need to understand StevenK's setup on antimony which i dont yet [11:24] ogra: Have I said yet how much I like Mobile? It's *really* a pleasure to use. [11:24] :D [11:24] my blog has another 6 comments pending .... i suddenly have that every morning now [11:25] and another .fdi for the touchscreen :) [11:25] synaptic fails to open due to admin issues and locks up the system when you hit close on the error message [11:25] damned, i should have done all that a month ago ... i didnt expect such a success [11:26] lool, my prob with cdimage is that it really scares me, you can so easily mess the whole system up if you make a mistake [11:27] ogra: No? There's been demand for Ubuntu Mobile since at least half-way through the hardy cycle. [11:28] persia, still i havent thought it would kick off a landslied of feedback ... i somewhat underestimated the amount of UMPC users [11:29] ogra: Oh. My apologies if I didn't make that sufficiently clear previously. There's *lots* of us, and no good software to install unless one puts it together by hand. [11:29] (until now) [11:29] yeah, apparently [11:30] anyway, no usb support in vbox, the .img file cant be read as cdrom either [11:30] ogra: If you have any doubts, do the cdimage changes but run then through StevenK and cjwatson [11:30] Mind you, -mobile still isn't best for all my hardware, but after today, I'm expecting -mid to be a good install for me. I just have to figure out how to preserve the Vista partition, because I'm not supposed to delete it. [11:30] lool, i will anyway [11:30] *them [11:31] persia, btw, what about your touchscreen, please tell me if i need to include another touchscreen driver for you [11:31] (though the calibration tool wont work for it) [11:32] Which? [11:32] ogra: One thing I didn't understand is what the image you blogged on was based of? I thought it was the mobile seed, but the umpcportal review and you said it was using standard panel and launcher [11:33] the one that didnt work with evtouch [11:33] Oh, right. On the SR, I need to sort out the stuff you pasted me last night. I'll do that when I have a bit of free time (likely tomorrow). For the D4, it's *definitely* evtouch, with the two devices and everything. I'll fiddle about with that, and see if I can't get it calibrated as well. [11:33] But I do see maximus and netbook-launcher in the seed, so I'm confused [11:33] lool, they are in the seed but not used, i will drop them before the next build [11:33] lool, the launcher is simply to broken, maximus doesnt work with compi [11:33] z [11:34] The image is definitely running gnome-menus, but in an implementation that works well at this DPI. [11:34] ogra: Hmm why did you decide to move away from them? I kind of expected we would get them in an intrepid image [11:34] And I think the users of these as well [11:34] ogra: Yeah, but why use compiz?!? [11:34] lool, i dont see that i have the time to solve all issues and upstream doesnt either apparently [11:34] And concerning the launcher, what else apart of fonts? [11:34] lool: The image is *very* nice, and works well. I like it better than the remix. [11:34] all sizes and positions are hardcoded [11:35] ogra: Hmm ok, will we revisit this for jaunty? [11:35] to get it working with a panelsize of 48px alone i have to rewrite and patch a lot of it [11:35] we can [11:35] lets have a spec and BOF bout it [11:36] i'm not convinced by the apps [11:36] and as the reviews show you dont really need the launcher [11:36] I actually prefer not having the launcher. With the large-font menu, it's a *lot* easier to handle running multiple apps. [11:36] the go home applet is essentially only a rewrite of the show desktop applet [11:37] I understand your choice, but perhaps it should have been announced more clearly; like, I'd like to have some pointer with the information we just exchanged if someone asks me why the UNR apps weren't part of the images after all [11:37] i perfer to go with the existing one here [11:37] (even when talking with OEM people :-) [11:37] the window picker applet has all sizes hardcoded to 16px which isnt usable with a fingertip [11:37] and maximus *only* works with metacity [11:37] Well that's not too grave [11:37] The last one [11:38] i really perfer to use compiz where we can [11:38] lool: Let's ask this the other way: what do we gain by using the netbook remix applications? [11:38] it impressed people a lot [11:38] my aim was to be as close to gnome upstream as we can [11:39] ogra: The thing is right now you have to chose between compiz and any clutter or pigment app, such as the launcher or elisa for instance [11:39] adding four new apps that need a *lot* of fixes and changes that nearly end up in a rewrite isnt maintainable imho [11:39] lool: Then don't use those. [11:39] persia: The apps are popular and widely used by people who only heard about them from a blog post and went the route if installing manually [11:39] persia: Uh thanks for constructive discussion [11:39] i want gnome upstream love so we only have to care for seeds and the default settings [11:40] my main target was to keep the maintenance level on our side as low as possible with the biggest outcome possible [11:40] ogra: Well the choice of metacity isn't exactly a GNOME one ;-) but that's unrelated [11:40] err compiz [11:40] adding four apps that need a fulltime person to maintain them isnt really in that scope [11:41] Sure, again, that I understand [11:41] i didnt make that coice [11:41] what choice? [11:41] it is our default ubuntu desktop ... it uses compiz if it detects composite [11:41] if it doesnt it uses metacity [11:42] i didnt pick *anything* [11:42] lool: Essentially, I don't see any benefit to the netbook apps. The netbook remix was popular in part because there are *no* other choices. [11:42] its all desktop team work i'm using here [11:42] the only thing i did was set the panel layout and applet layout as well as some theme changes [11:43] ogra: Ok, so you inherited compiz, but I don't think it's in argument in including clutter apps that the default is just to run compiz [11:43] nah [11:43] the maintenance overhead is the reason [11:43] It's ok to inherit that, but the default of inheriting should be put in question if that prevents us from running apps we care about [11:44] lool: Why do we care about the netbook remix apps? [11:44] after using it for two weeks i must admit that i dont care that much about the launcher [11:44] ogra: grub is being installed: but no kernel is being installed. [11:45] i'm fine maintaining the packages in the archive and nagging upstream aboout fixes (or even contributing them) but i *dont* fell the apps are ready for production [11:45] persia: The UNR apps were tailored specifically for netbooks, are nice and usable; they are already popular and desired [11:45] ogra: Ok; the bugs in the intrepid version really sound like it's not the time to use them [11:45] right [11:46] lool: I guess. I'd rather continue with the extension of ograsac desktop as continually discussed since June. [11:47] Which has what upstream? [11:47] GNOME. [11:48] Erf [11:48] No really, it's just a theme tweak, and some panel positioning. Take a look. [11:48] Oh, and maximise-by-default. [11:49] lool: [11:49] This project aims to develop prototypes to make it possible to use a plain ubuntu-desktop base on mobile platforms by just changing small parts of the UI theming and minimal toolkit patches instead of rewriting the world from scratch like other projetcs do ... the aim is to be able to use the existing awesome desktop integration and translation that is in ubuntu already and make the apps behave right on small screens with and without fingertip navi [11:49] gation (suitable for moble devices as well as netbooks). [11:49] thats the text of https://launchpad.net/ograsac-desktop [11:49] From which the ubuntu-mobile project grew. [11:49] Exactly the way I understood the state of ograsac-desktop: prototyping [11:50] well [11:51] just look at the feedback (and listen to what the people say that tried UNR) [11:51] My point is not that the UI experiments aren't interesting [11:52] At some point, we need to move to an upstream/packaging model [11:52] We're not enough people to do upstream development [11:52] We are already there. There's not a significant volume of code that differentiates Ubuntu Mobile from Ubuntu Desktop. [11:52] There's no launcher, and it doesn't have the menu changing hacks. [11:52] I want us to build cool netbook images out of the best existing free software technologies [11:53] well ... if there is no other viable upstream, what choices do we have? [11:53] asac: Which part are you commenting on? :) [11:53] "we cannot do upstream development" :) [11:54] is there a viable alternative? [11:54] ogra: Just to make sure: is there anything other than ubuntu-mobile-default-settings that makes this special? [11:55] the seed [11:55] I'm tempted to respond from a Canonical pov; from an Ubuntu pov I'd simply say that we need to do our best in app selection and configuration of existing components [11:55] and meta [11:55] lool: How is that different from what was done? [11:55] But avoid writing new software [11:55] There is no new software in Ubuntu Mobile. [11:56] well, there is stuff from universe like xournal ... [11:56] This is in response to usage of a) ograsac desktop and b) UNR apps [11:57] lool: We used the idea of ograsac desktop: none of the ograsac desktop applications. [11:57] Yes, and that's fine for now [11:57] there are no ograsac applications :) [11:57] ogra: Well, there's the launcher :p [11:57] (and yes, it doesn't work, and yes, that doesn't matter) [11:57] which in a similar form is now in mid [11:57] persia: It's like you think your statement contradicts my pov [11:58] I do think that the choice of using GNOME and Gtk+ components for mobile for now makes sense, and customizing its look [11:58] lool: It seems to me that you have an attachment to the applications in netbook remix. I don't understand the attachment, as I find the launcher to actively interfere with the way that I use such a device. [11:58] But as you were saying, you are pushing for ograsac-desktop "since June" [11:59] persia: I have an attachment to the UI experiment as well and it has the advantage of having a real upstream; this upstream might not be up to the level where we're confident using it in official images, especially not in intrepid, but there are upstream development resources there [11:59] And user interest [12:00] OK. I can understand that. I'm not opposed to revisiting whether upstream has code in condition for deployment for jaunty. I just don't think the current images deserve criticism for application selection. [12:01] I expressed criticism in the choice of compiz which is only an inheritance and goes in the way of choics which could have involved clutter of pigment [12:01] *choics [12:01] Grah, my e key has some stuff below it [12:02] Of course you don't use "e" that much [12:02] OK. Maybe I read it wrong. [12:03] I also expressed that I would have liked more explicit dropping of the UNR apps recently [12:03] (Cause I want such choices to be documented, not because I disagree with them) [12:04] lool, the thing is that ubuntu-mobile was never really documented or specced [12:04] OK. That makes more sense :) Actually, that's true for many of the choices we make: we tend to just grab what works, rather than actually mention more than arbitrary griping here. [12:04] i will change that for jaunty, theer will definately be a spec and BOFs [12:04] but that wasnt the case up to the first build [12:04] ogra: That's right, it was simply vaguely targetted at netbooks with the same constraints as for other ubuntu flavours [12:05] right [12:05] I think choice of UI some days before beta freeze qualifies as needing some announcement, even if the seed is such loosely specified [12:05] Anyway [12:05] btw, did you try it ? [12:06] No, I want to [12:06] Which is also why I'm pushing you to publish dailies :) [12:06] yeah [12:06] i wanted dailies on monday but the amount of cdimage changes somewhat pushed me back, sorry, thats definately my bad [12:06] It's not that I wouldn't trust downloading it from people.u.c, but the way it's published makes me think of a prototype as well, which it's not meant to be [12:07] yeah [12:07] agreed [12:07] That's ok, I really want that you publish your work with intrepid [12:07] it wont be or look much different if built from cdimage though, i use all the tools we use there and all changes i needed in the tools are committed [12:08] Perfect [12:08] ogra: But dailies means I'll install it twice a week, and that the kernel actually gets installed :) [12:08] (or maybe more often, depending) [12:08] persia, i doubt the kernel will be magically fixed by building it on cdimage :) [12:08] i'm sure its not that easy [12:09] ogra: No, but I made some changes earlier in the week that might fix that: I don't know if they do. [12:09] ogra: Anyway: when you have dailies, I'll be testing and fiddling bits. Even better if you have dalies that work in KVM. [12:10] well, i dont see why cdimage would fix that magically either [12:10] the syslinux version is the same in both environments [12:11] anyway, i need to go on wit testing, lost way to much time for this discussion now ... [12:11] cdimage doesn't fix it. cdimage gives me fresh images to fiddle with whilst you sleep. [12:11] ah [12:13] persia, since you are the only one yet seeing the install failure on mobile, can you file a bug i can follow up to ? [12:13] ogra: Anyway, I'm just testing the complete boot after a fixed install. I have a feeling about how this needs to be, but not yet confirmed. [12:13] we urgently need something with a description for the release meeting [12:13] Sure, although I'm going to try to fix it first. [12:14] Yeah, before the release meeting, you'll have a bug. I have a couple to file against -mid as well. [12:14] I guess file against the meta packages? I'm not sure where else the image bugs go, unless we understand which component needs tweaking. [12:16] lool: ok. i think i didnt get the topic right then ;) [12:16] (sorry had connectivity issues) [12:20] persia: Does the install work, though? [12:21] StevenK: I think so. Just patching it up manually, and will reboot. The kernel is missing. [12:23] persia: Is that due to the kernel not being in a repo on the image? [12:23] lool: And isolinux.cfg?! We don't use that [12:23] persia, meta is fine for now [12:23] StevenK, he means syslinux.cfg [12:23] I know that :-) [12:24] :) [12:24] Patches welcome :-P [12:24] ogra: With a manually installed kernel, -mobile works great. I'll see if I can sort out why it's not being installed, and let you know. [12:25] persia, but please file a bug ... release meeting is after team call and we will be blocked if loic cant present a bug number [12:25] Although the set of available bluetooth services for -mobile looks very bad. [12:25] ogra: I will. [12:25] merci :) [12:26] * persia is doing a -mid install with manual hinting to track down the issues there too, and make sure that the list is complete for the meeting. [12:27] StevenK: Actually, if everything is built, could you generate a new -mid image? I think what I have, and what I will have are sufficiently different to be interesting. [12:28] persia: Such as the new meta? [12:29] And having a terminal emulator :) [12:31] ogra: Do you use bluetooth for anything? [12:32] no, i just notice the icon in the panel [12:32] my phone is supposed o be capable ... i could try to use the Q1 as handsfree device [12:33] ogra: OK. By default, it only works for audio. Not for transfer to the phone or input. I suspect you may need a couple extra packages. [12:33] (or maybe this is an issue for -desktop) [12:35] persia: Just checking, looks like everything has. [12:35] StevenK: Excellent. I'd like to see how it looks, so we can make sure to feed lool any bugs he wants to raise. [12:38] OK. Word back from the Xubuntu folk is that there is a good reason for thunar to recommend xfce4-panel, but it's safe to miss. Does anyone happen to know if blacklisting a recommendation in the seeds works? [12:38] Nope. [12:38] Beg Colin. [12:39] Oh well. [12:51] OK. Outstanding installation issues. Please chime in if anyone has experienced anything not on this list (as I'm filing bugs): no grub on -mid, no kernel on -mid, no kernel on -mobile, preseeding issues on -mid, preseeding issues on -mobile, no support for 480 vertical on -mid. [12:54] no support for 480 vertical on -mid ?? [12:55] ogra: Nope. There's ways around it, but it's a bug. I don't expect to fix it for intrepid. [12:55] ah, ubuquity [12:56] i was thinking generally :) [12:56] Yep. It's a wishlist bug at best. [12:56] The solution is to create glade files for a new front-end, but that defeated me (and I should have been concentrating on function rather than appearance at the beginning) [12:57] StevenK: Sorry, meant syslinux.cfg; syslinux and isolinux are the same upstream software suite [12:58] lool: I know. I'm teasing. :_) [12:58] StevenK: Damn Australian humor! [12:58] Muahaha [12:58] StevenK: You had too much Champagne at the party [12:58] I had no champagne [12:58] I wonder whether you can get Australian white wine with CO²? [12:59] * lool had some tasty Australian wine a couple of times [12:59] persia, oooh, see the ML [12:59] https://lists.ubuntu.com/archives/ubuntu-mobile/2008-September/002094.html [12:59] its a menu.list issue it seems [13:00] he doesnt say he needed to install a kernel [13:00] lool: why do you have make sparkling wine so scientific :-/ [13:00] amitk: I was searching for the word sparkling; I had Kohlensäure in mind and couldn't think of the English word anymore :) [13:01] heh [13:01] * lool was thinking of Sekt, Sprudel, this kind of things [13:01] lol [13:02] Arf, the menu.lst issue is almost the same as the MIC one [13:02] lool: just buy a soda stream ;) === davmor2 is now known as davmor2_Lunch [13:14] Oh, vlc 0.9.2, cool === Andrade is now known as Guest90321 === Andrade[1] is now known as Andrade [13:42] ogra: You have a bug for the install issue? === davmor2_Lunch is now known as davmor2 [13:59] lool, persia said he would put it up [13:59] persia, ? [13:59] I don't have the number in front of me. Read my mail to the list :) [14:00] * persia is filing other bugs. [14:28] lool: Bug list: bug #182004 bug #274752 bug #274753 bug #274781 bug #274785 bug #274786 I have local branches for 274781 and 274785, and will make them available to the installer team after another test round. [14:28] Launchpad bug 182004 in ubiquity "partitioner fails if partially preseeded due to seen flag madness" [Medium,In progress] https://launchpad.net/bugs/182004 [14:28] Launchpad bug 274752 in mobile-meta "No kernel installed for ubuntu-mid flavour" [Undecided,New] https://launchpad.net/bugs/274752 [14:28] Launchpad bug 274753 in mobile-meta "No kernel installed with ubuntu-mobile" [Undecided,New] https://launchpad.net/bugs/274753 [14:28] Launchpad bug 274781 in grub-installer "Please add support for the lpia architecture" [Critical,In progress] https://launchpad.net/bugs/274781 [14:28] Launchpad bug 274785 in ubiquity "Ubiquity should select grub as the boot loader for lpia" [Medium,In progress] https://launchpad.net/bugs/274785 === njpatel is now known as njpatel_away [14:29] lool: bug #274789 is specifically *NOT* targeted for intrepid. [14:29] Launchpad bug 274789 in ubiquity "ubiquity should have a smaller interface to work on 640x480 screens" [Wishlist,Confirmed] https://launchpad.net/bugs/274789 [14:30] persia: thanks [14:46] * ogra is shocked by DHL [14:46] my n800 just arrived [14:47] and instead of ringing at the bell they just put it in front of the door on teh ground [14:47] ... [14:48] ogra: At least you got it :D [14:48] indeed [14:48] :) [14:48] i'm in love [14:49] Het it's good but it's not that good ;) [14:49] yeh, the 810 is better [14:49] but i got it cheap [14:49] and i wanted recent arm HW [14:50] ogra: I got mine free :) the 810 is more up-to-date arm though isn't it? [14:50] indeed [14:50] but the n800 totally suffices my needs atm [14:50] ogra: if you want really up to date you may wish to consider a beagle board [14:50] i'll definately subscribe to the maemo dev program for the next HW release [14:51] i was to late for the 810 [14:52] http://beagleboard.org/ [14:52] yeah, i know it [14:52] i would love to develop a thin client variant in ltsp for it it should make the best thin client ever [14:53] I love the fact that the chip it runs has a separate MM core [14:53] its quite limited in ram though [14:53] you wont run a gnome desktop on it that way [14:56] ogra: you could always get a developer version of these http://store.neurostechnology.com/neuros-osd20-developer-p-55.html === njpatel_away is now known as njpatel [15:40] * ogra hugs persia thanks for getting all these guys a working install :) [15:41] Well, I wanted a working install, so I had to figure it out. [15:41] hmm [15:42] * ogra woders whats up with g-p-m [15:42] it doesnt show me that the battery is charging in the applet [15:43] moun -t proc proc /proc should help with the proc errors btw [15:43] Yeah, but it doesn't matter. Just ignoring them works. [15:43] yeah [15:58] OK. Release meeting. Let's see how much pain we get for the bugs. [15:59] persia, about bug 274786, mobile expects the user to be called ubuntu by default, i have no onscreen kbd in gdm yet [15:59] Launchpad bug 274786 in mobile-meta "ubuntu-mobile install should be less agressively preseeded" [Low,New] https://launchpad.net/bugs/274786 [15:59] so its set to autologin with that username [15:59] i agree about the locaion though [15:59] ogra: You can preseed the "log in automatically" value, and then gdm will autologin with the user's selected username. [15:59] oh, i didnt know that [16:00] i'll try that as soon as i have the first cdimage build then [16:00] Yep. That's what the checkbox is for. Should actually make it easier for you. [16:00] phew another 5 blog comments within the last 2h [16:00] What's the URL again? [16:01] http://ograblog.wordpress.com/ [16:02] i didnt approve all of them yet, i want to have the time to comment on them if i do [16:02] Right. I just like to read them :) [16:03] :) [16:04] * ogra needs to go shopping soon, i promise susie since three days that we do that ... will be afk for 1-3h [16:04] No, that blog entry is a placeholder until we get a volunteer to be a forums moderator. [16:04] OK. [16:04] not yet though [16:05] hmm, steve didnt notice that i moved, i guess i need to clear that up with him (i used to live about 50km away from him, now its more like 350) [16:15] while true; me = me + 1; sleep 31556926; done [16:15] ogra, amitk: do we have a bug for madwifi or ath5k on Q1U not working for intrepid? [16:15] lool: For i386 or amd64 or both? [16:15] s/amd64/lpia/ [16:16] lpia [16:16] i386 has madwifi in lrm [16:16] Who suggested we should blacklist poulsbo? [16:16] Me. [16:17] I have poulsbo hardware. X only works if I uninstall xserver-xorg-video-psb [16:17] (yes, this is counterintuitive) [16:18] persia: May I ask you to file a bug to remind us to remove it from xorg.conf generation? [16:18] amitk: Is there a bug for lrm on Q1U? [16:18] err lrm/lpia to support Q1U [16:18] lool: Hrm? Where does that belong? [16:18] no [16:19] lool: I think it's fixable, just not quickly. I'd rather keep the package as a placeholder for an SRU. [16:19] Or do you mean fiddling the .fdi files to not prefer it? [16:19] persia: Is any software explicitely pulling the package? [16:19] If it's not part of video-all, then we probably don't care [16:19] * persia checks [16:20] It was on the image anyway [16:20] persia: The mid dailies? [16:20] Yeah. It's part of -all [16:21] Maybe it shouldn't be part of -all, and when we fix it, it can get restored? [16:21] Ok; so we should drop it from there if it's broken, ok with you? [16:21] Yeah, that's what I have in mind as well [16:21] lool, bug 274832 [16:21] Launchpad bug 274832 in mobile-meta "no linux-restricted-modules for lpia flavour" [Undecided,New] https://launchpad.net/bugs/274832 [16:21] That's fine. I was just tired last night and annoyed that my hardware required fussing. [16:21] * persia files a bug [16:21] ogra: Thanks [16:21] nominated for intrepid as well [16:21] persia: thanks! [16:21] ogra: perfect [16:21] * lool appreciates the help with the bugs :) [16:22] sorry for slacking, i wanted to file that weeks ago [16:23] lool, do i need to put up a bug for ubuntu-mobile vs. cdimage.u.c ? [16:23] (woudl that help ?) [16:23] ogra: You can [16:23] ogra: You mean the DC-built images (cdimage) dailies based on mobile/i386 seed? [16:24] we could also have a bug for ath5k not working on Q1U [16:24] lool: bug #274833 [16:24] Launchpad bug 274833 in xorg "Please disclude xserver-xorg-video-psb from xserver-xorg-video-all" [Undecided,New] https://launchpad.net/bugs/274833 [16:24] yes, i mean for the inclusion of the ubuntu-mobile images in the build scripts [16:24] That does need a bug. [16:25] * persia forgot it when filing earlier: apologies [16:25] persia: Thanks [16:25] ogra: I think it's nice to track it as a bug, but if you do it straight away you can skip it :-P [16:25] i wont be able to [16:26] ogra: Perhaps best to have this on the release team's radar's bug list [16:26] I will mention it for sure though [16:26] need to do some real life stuff soon ... i'll attack it tonight [16:26] bug 274838 [16:26] Launchpad bug 274838 in mobile-meta "ubuntu-mobile images need to be built ob cdimage.ubuntu.com machines" [Undecided,New] https://launchpad.net/bugs/274838 [16:26] also nominated [16:27] meh, typoed it [16:27] nominated? I thought we were supposed to have them milestoned by the release manager for the flavour. [16:27] well, i want it to be on the intrepid radar in any case [16:28] At UDS I distinctly remember slangasek saying he didn't use nominated bugs as a search criteron. [16:28] ogra: thanks [16:29] persia, well, better than nothing [16:29] ogra: Indeed. [16:29] i cant set milestones it seems for whatever reason [16:30] If they are on the agenda for the meeting, the rest hardly matters. [16:30] You aren't a driver. [16:30] right [16:30] well, i was able to for hardy [16:30] but that changed apparently for intrepid [16:31] it used to work for all of ubuntu-core-dev [16:32] Right. It got restricted. [16:39] * lool seems to be able to [17:13] lool: no to both === robr_ is now known as robr === njpatel is now known as njpatel_away [18:33] * persia just remembered there is no icon for the installer [18:34] lool: Are you still about? There's a couple ways we could add an icon for the installer, but they're all a bit messy, and I'd like to run them by you. [18:35] persia: I'm all ears [18:35] Well, the basic issue is that ubiquity installs a .desktop file on the home directory Desktop. [18:35] Since -mid doesn't actually show the desktop, this isn't visible. [18:35] Also, we want to run in --automatic mode, so we don't want to use that .desktop file. [18:36] (which makes the preceeding not so bad). [18:36] So, here are some possible solutions: [18:36] 1) add a ubuntu-mid-installer-wedge binary package to ubuntu-mid-default-settings that contains the launcher we want, and gets deleted post-install. [18:37] 2) hack kourou to pull the ubiquity desktop file iff ubiquity is installed (we lose --automatic) [18:37] 3) install the installer launcher into the unionfs during the image boot (but not have it be in the squashfs) [18:37] There are probably others, but those are the first that come to mind. [18:38] I think we want a .desktop file which would show up like the others [18:38] This should be shipped by ubiquity, just like the Desktop thing [18:39] We don't need a new package for that, and I think it's ok if it shows up in the menus for everybody; we should just pick the proper place [18:39] Except we only want it for -mid. Everyone else doesn't want to be in --automatic mode. [18:39] You might want to discuss the specifics with cjwatson [18:39] Ah, didn't see the automatic mode let me think a sec [18:39] is there another way to set automatic mode? such as preseeding [18:40] We can have a special flavour of ubiquity that does it, but that seems like more overkill than ubuntu-mid-installer-wedge [18:40] that does what? [18:40] that includes our special .desktop entry. [18:40] I was proposing to have it in the main ubiquity, for everybody [18:41] Yeah. They don't want that. [18:41] You checked with cjwatson already? [18:42] No need. running in --automatic would significantly change the standard install behaviour. [18:43] My idea was to just have ubiquity in the .Desktop file and set automatic mode with preseeding [18:43] 19:39 < lool> is there another way to set automatic mode? such as preseeding [18:44] I'll double check, but I'm fairly sure --automatic can't be preseeded. [18:44] Could it be changed easily? [18:44] s/changed/implemented [18:45] which? [18:45] automatic mode's preseeding [18:45] I mean preseeding of automatic_mode=true|fakse [18:45] *false [18:48] No. There might be a way to hack it with environment variables. [18:48] Doesn't help [18:48] persia: Ok; so we could create the .desktop file in casper below the home dir [18:48] Which would be somewhat 3) [18:49] and hack kourou to look there? [18:49] No, should be xdg [18:49] Like .local/share blah [18:49] But I don't have the specifics :) [18:49] Oh, right. We might be able to do that. You like 3) better than 1) then? [18:49] I think 1) is cleaner, but it does require NEW. [18:49] Yeah, package is overkill [18:50] There are plenty some such hacks in casper; I prefer not piling too many, but it's ok to add a small one like this [18:50] I think I'll ask cjwatson whether he has plans for preseeding automatic mode though [18:50] If you like, but given the architecture of ubiquity, I'm not sure how that would work. [18:51] Generally, the preseeding works against d-i, rather than against ubiquity, and ubiquity just uses it. In --automatic mode, ubiquity doesn't show interfaces where it knows the answer, and in regular mode, it prepopulates the preseed answers. === robr__ is now known as robr