2008/05/29/#ubuntu-mobile.txt

[03:02] <StevenK> persia_: Your package for Update Manager is just the normal update-manager?
[03:03] <persia_> StevenK: There ought be three binary packages.  update-manager-hildon, update-manager-core, and update-manager.  For ubuntu-mobile, update-manager-hildon is interesting.
[03:03] <StevenK> persia_: Well, the thiny veiled question is, where is the source package hiding, and is it in the PPA?
[03:03] <StevenK> thinly
[03:04] <persia_> My intrepid branch (from which that derives: backporting just wasn't happening) isn't 100% merged with mvo's, but it's close.
[03:04] <persia_> update-manager is the name of the source package.
[03:05] <StevenK> Does that mean the update-manager in Hardy has what we need, or we need to upload a changed version to the PPA?
[03:05] <persia_> The update-manager in the PPA has what we need, and the standard update-manager in intrepid is expected to have what we need when mvo finishes merging.
[03:06] <StevenK> Ah, okay.
[03:07] <persia_> I uploaded update-manager 1:0.90.1~ume804 this morning after giving up on the backport and not hearing anyone say there was much of a use case for upgrades to hardy using it.  It should be fine for upgrades from hardy, but doesn't have all the hints for gutsy (or older) -> hardy.
[03:07] <persia_> If you want to play, http://launchpadlibrarian.net/14777823/update-manager-hildon_0.90.1%7Eume804_all.deb is the package to use.
[03:08] <StevenK> Ah. I was actually looking to seed it.
[03:08] <persia_> I thought I already did that.
[03:09] <StevenK> There's no ubuntu-meta upload ... :-)
[03:09] <persia_> Ah.  Right.  It's in the seed, but I didn't run germinate.  I really ought not do these things at that hour in the evening.
[03:09] <StevenK> Heh
[03:10] <StevenK> persia_: It isn't just running germinate, it's a little bit more complicated, but not much.
[03:11] <persia_> StevenK: Right.  I updated ubuntustudio-meta for gutsy I think, but haven't touched them since.
[03:11] <StevenK> Oh, that should be the same method.
[03:13] <persia_> StevenK: I'm away from my keys for about 9 hours.  Would you happen to have time to run it (and I'll be sure to remember for next time)?
[03:14] <StevenK> persia_: Perparing to do so now
[03:14] <persia_> StevenK: Thank you.
[03:14] <StevenK> Just trying to figure out if there was something else I needed to seed.
[05:24] <cl0s-home> fixed my theme problem by the way.. the tutorial for setting it up has a startup script that points to a theme that doesnt exist..?
[07:04] <dholbach> good morning
[07:21] <bpsew> good morning
[08:41] <davmor2> Yay rc's up :)
[08:59] <stgraber> davmor2: is it better to use rsync for ume images ? (as they are compressed with bz2)
[09:00] <davmor2> stgraber: no Idea I just download them
[09:02] <davmor2> stgraber: I was going to modify the dl-iso script but for one Image I couldn't see the point.  Besides which it enables me to rename them to include the date of the image.
[09:30] <lool> stgraber: I use rsync; it's a bit faster when you update an image rather than downloading it from scratch
[09:33] <davmor2> lool: I found it more useful to keep more than 1 copy the new one wasn't always guaranteed to work :)
[09:33] <lool> davmor2: One thing doesn't prevent the other :)
[09:34] <davmor2> lool: rsync over writes what is there though doesn't it ?
[09:34] <lool> persia: Thanks for the update to the seed -- will you update ubuntu-meta as well?
[09:35] <lool> davmor2: Sure, but you can cp the file before rsync
[09:42] <lool> StevenK: The icons in dates-hildon are missing; smagoun says this is probably due to sdk-default-icons which I think we tried to remove to see what would break; if I recall correctly its licensing wasn't clear enough for main promotion
[09:42] <lool> StevenK: I think I'll readd it to the ppa and to the seed
[09:43] <StevenK> It's non-free
[09:43] <StevenK> You can't modify it
[09:43] <StevenK> And I doubt it it's due to sdk-default-icons, dates-hildon isn't from Nokia and sdk-default-icons is.
[09:44] <davmor2> lool: ah yes true
[09:44] <lool> I grabbed sdk-default-icons; it's CC 2.5 and it only allows sharing and "remixing"
[09:44] <lool> So yes, it's probably non-free
[09:45] <lool> Hmm I don't see it in https://stage.maemo.org/svn/maemo/projects/haf/trunk; I wonder where we got it
[09:47] <lool> dates-hildon is using icon names such as qgn_list_messagin_editor, qgn_list_gene_fldr_opn, etc.
[09:48]  * StevenK whimpers.
[09:48] <StevenK> Dun wanna to add sdk-default-icons back.
[09:49] <lool> modest has embedded copies of the icons
[09:49] <lool> https://garage.maemo.org/plugins/scmsvn/viewcvs.php/trunk/src/gtk2/pixmaps/qgn_list_gene_fldr_opn.png?root=modest&view=markup&rev=377&pathrev=377
[09:50] <suihkulokki> Oh, there should now be permission to license those (and the themes) with CC SA 3.0
[09:50] <suihkulokki> I need to go kick someone ->
[09:50] <lool> suihkulokki: Cool!
[10:01] <lool> suihkulokki: I see it in version 2.0-4.5 in sardine
[10:01] <lool> And we had branched from -4.1
[10:02] <lool> Hmm still 2.5 in this version
[10:02] <lool> suihkulokki: Are such packages kept in a separate SVN, or no SCM at all?
[10:04] <suihkulokki> lool: I have no idea if there is any SCM for the icons. themes are in haf svn.
[10:05] <lool> Ok; thanks
[10:54] <speilicke> Hi guys, can somebody point to a nightly build (or at least something recent) to try out on x86 without much hassel?
=== dholbach_ is now known as dholbach
[11:21] <davmor2> Guys why does the moblin-media keep saying Application is disabled while in docking mode?
[11:23] <lool> suihkulokki: Do you recall the name of the gobby document we had on the ARM port?
[11:23] <lool> Or does someone have logs of #uds-mobile?
[11:32] <davmor2> https://bugs.launchpad.net/ubuntu-mobile/+bug/234230 this is still present in RC does anyone know what is happening with it?
[11:41] <lool> davmor2: I do'nt reproduce on my Q1U; could it be some xephyrish issue?
[11:44] <davmor2> lool: It was cgregan who reported is not functioning on the JAX10 which is menlow rather than mcasslin I think.  So I think it is more likely to be menlow than xephyr but I'll get cgregan to confirm on the jax10 again
[11:44] <lool> davmor2: I'm upgrading my Q1U, so perhaps it's only for new installs
[11:46] <davmor2> lool: pass.  I've had it since 20080526 or 27 and confirmed cgregan report but he has had the issue longer on the jax10 device
[11:46] <davmor2> I'll add a screenshot of it to the report anyway 
[11:47] <lool> StevenK: Hmm if we release on snapshots.ubuntu.com, we will need to have a /stable or something entry in sources.list, defaults.cfg points to an explicit snapshot version, but you override this when building; do you want to use /current for this?
[11:52] <davmor2> Guys why is seahorse launcher not visible?  I'd of thought that gpg keys would of been very important.
[13:32] <teletommy007> i wonder if ume is usable only with a keyboard?
[13:47] <persia> teletommy007: It's not ideal.  Most applications are optimised for touchscreen use.
[13:48] <teletommy007> persia: i think about using UME in a car pc. there i only have a wheel using to emulate keystrokes
[13:48] <teletommy007> i read about intels / windrivers engagement in moblin bringing it to the car.
[13:48] <zumbi_> !hi
[13:49] <persia> teletommy007: Depending on which applications you expect to use, that might work, but it likely needs a different default interface.
[13:50] <zumbi_> is ubuntu mobile already working?
[13:50] <teletommy007> persia: what are available interfaces
[13:52] <persia> zumbi_: Somewhat.  It's not really ready for general use without modification at this point, but definitely works for a number of people who use it daily.
[13:55] <zumbi_> i do not understand very well the infrastructure, what platforms does it support? McCaslin and Menlow ?
[13:55] <zumbi_> does it support different architectures ?
[13:56] <zumbi_> what finally what moblin is, is it like the lowest layer?
[13:57] <ian_brasil> lool: is there some wiki page about the plans for an arm port?
[13:57] <persia> teletommy007: No limit, really.  The default is touchscreen optimised tought.  You'd nneed to find something else that works for you.
[13:57] <zumbi_> is it Debian based? like Ubuntu desktop 
[13:58] <persia> zumbi_: It's a collection of Ubuntu packages, and optimised for McCaslin & Menlow handhelds.
[13:58] <zumbi_> who are the main developers?
[13:59] <persia> zumbi_: Participants in this channel
[14:04] <cl0s> teletommy007: thats what im trying to do.. use it for my car..
[14:07] <davmor2> NTP install fails 
[14:07] <zumbi_> persia: anyway, sorry for shooting so many questions at a time
[14:08] <zumbi_> just trying to find out information, i'm related to embedded debian project and we are having a meeting in september
[14:09] <zumbi_> i was wondering if maybe some developer could come to our worksession
[14:09] <zumbi_> i have heard about loic minier, anderson lizardo, tollef, .. are those the right people working in these project?
[14:10] <Hobbsee> tollef no longer does
[14:10] <Hobbsee> loic, yes.
[14:13] <persia> zumbi_: Probably best to ask on the ubuntu-mobile@lists.ubuntu.com mailing list.  The approved list of developers is available from https://launchpad.net/~ubuntu-mobile/+members, although lots of other Ubuntu developers also contribute (same software), and other people help in various ways.
[14:15] <teletommy007> persia: I don't know what exactly i have to search for :-)
[14:16] <persia> teletommy007: I can't really help you there (as I know little about CarPCs, and that's not really the regular Ubuntu Mobile target).  You might collaborate with cl0s who is doing something similar.  You might also try installing in a VM, and trying different things.
[14:17] <teletommy007> persi: ok, thank you!
[14:17] <cl0s> teletommy007: what are you trying to do?
[14:18] <teletommy007> I thought about usin ume for my car pc that has no touchscreen. only keyboard.
[14:19] <cl0s> oo.. just curious but why no touch screen? would seem the best for a car..
[14:20] <cl0s> shouldn't have a problem using a keyboard though, but then you'd need a mouse..
[14:20] <teletommy007> cl0s: i tested ume in the xephyr windows yesterday and wondered i it is usable without mouse
[14:20] <cl0s> oo i c.. were you able to use maybe the tabs atleast?
[14:20] <cl0s> thats as far as i am right now also.. 
[14:21] <teletommy007> there are this fancy control in the car. BMW tells it a wheel. i will use it as control
[14:21] <cl0s> ahhh
[14:22] <teletommy007> i will use this wheel to emulate keystrokes to linux
[14:23] <teletommy007> like "left" "right" "enter"
[14:23] <cl0s> ok i see
[14:23] <cl0s> is it picking up any of your keystrokes in xephyr
[14:24] <cl0s> the hardest part to me is getting that wheel to talk to the computer... unless thats what its meant for...
[14:25] <teletommy007> this part is solved. i have connection to the can bus receiving the right messages
[14:26] <teletommy007> the xephyr window doesn't pick anything
[14:29] <teletommy007> cl0s: are key strokes not supported or something?
[14:30] <cl0s> i dont have it installed here at work yet but ill play with it in a little while, maybe start by looking at the program that picks up hotkeys in gnome and start by trying to install and use that..
[14:31] <cl0s> i just started looking at everything like a day or two ago, wanted to build a carpc from scratch also..
[14:32] <teletommy007> exactly the same with me :-)
[14:34] <cl0s> you have some hardware already to start with/
[14:35] <teletommy007> i have a normal little pc
[14:35] <teletommy007> with mobile board / processor
[14:35] <teletommy007> i just have to change the power supply to DCDC
[14:36] <cl0s> nice.. 
[14:36] <cl0s> i was just looking at some mini-itx boards and stuff like that
[14:36] <teletommy007> i use the mb899 from ibase but there shoult be other better boards available
[14:37] <cl0s> regular small lcd and turning that into a touch screen,... didnt want to spend any money till i had the software part mostly done, and plus i dont have to have money to do that part.. hah
[14:37] <teletommy007> i read that intel plans a automotive infotainment system using the moblin system based on their atom processor
[14:39] <cl0s> depending on how open the hardware is i might look at gettin something like that.. but thats kind of why i want to build from scratch
[14:45] <cl0s> http://moblin.org/community/ivi/
[14:46] <cl0s> dont know if you seen that but what you were just talking about.. hadnt seen this page before..
[14:48] <teletommy007> yes, that's it
[14:48] <teletommy007> but i think it is too new. they don't have much information
[14:48] <cl0s> yea :(
=== Zic_ is now known as Zic
[15:54] <lool> FYI people http://paste.ubuntu.com/15550/
[15:54] <lool> (Just at the time of the IRC meeting, shouldn't affect us)
[16:50] <emgent> heya
[16:57] <lool> suihkulokki: Please check http://people.ubuntu.com/~lool/ARM-EL-port 
[16:57] <lool> suihkulokki: Does this look like the latest version of the doc?
[16:59] <davidm> Looks like mootbot is with us again
[16:59] <suihkulokki> lool: looks sane. Thanks for digging it up.
[16:59] <suihkulokki> Now I have something to point people at
[17:00] <davidm> #startmeeting
[17:00] <lool> suihkulokki: I'll move it to the wiki
[17:00]  * lool kicks MootBot 
[17:01] <davidm> Hmm, mootbot is stuck
[17:01] <GrueMaster> Just an FYI:  I'm d/l'ng the RC now, if anyone asks how it looks.
[17:01] <davidm> #startmeeting
[17:01] <davidm> Yep, stuck, oh well
[17:02] <davidm> OK starting the meeting.
[17:02] <davidm> I have no prior action items
[17:02] <davidm> And only one current item
[17:02] <davidm> Topic: Tobin (GrueMaster): segfault when upgrading sudo package in Beta 5 with apt-get update;apt-get upgrade. Pastebin at [WWW] http://moblin.pastebin.org/39480
[17:03] <davidm> GrueMaster, you have the floor
[17:03] <GrueMaster> I'm not sure how important it is, or if I should file a bug yet.
[17:03] <GrueMaster> It may be a conflict with running apt-get in a sudo environment.
[17:04] <lool> GrueMaster: That's fixed with the -17 kernel
[17:04] <lool> There's an open bug about that
[17:04] <lool> It was an unionfs issue with hardlinks
[17:05] <GrueMaster> ok, then.  Guess it's moot.
[17:05] <davidm> Sounds like.
[17:05] <lool> https://bugs.launchpad.net/bugs/224754
[17:05] <lool> GrueMaster: The new kernel should be in the current RC, if it's -17 it should be fixed
[17:05] <davidm> OK, I have no other things on the agenda
[17:06] <davidm> We have with the assistance of Intel (sorting a license issue) released the first RC image.
[17:06] <GrueMaster> lool:  I'm still downloading the RC.  Will pass on my test info to Don sometime later today.
[17:06] <lool> You can test by updating a device to the new kernel
[17:07] <davidm> Are there any other open issues to discuss?
[17:07] <cgregan> Is there a listing of the bugs that are listed as fix commit but are not in RC1?
[17:08] <cgregan> So we do not duplicate work
[17:08] <davidm> cgregan, as best I know there is at least one, the treb bug (adilson fixed it yesterday) 
[17:09] <davidm> lool, did you fix something also that did not make the image?
[17:09] <cgregan> Can an email go out to mobile group with a list?
[17:10] <davidm> cgregan, sure, that can happen
[17:10] <cgregan> thanks
[17:10] <lool> davidm: I fixed things in MIC which were tracked as bug reports
[17:10] <lool> And I also fixed some missing MD5sums in some packages which were not tracked in bug reports
[17:10] <davidm> lool, thanks, that would not show up in the image anyway.
[17:11] <davidm> cgregan, to the best of my info then just the treb fix, but I'll ask StevenK for sure when he wakes up again.
[17:11] <davidm> I'll get an email out now and then later if need be with an update.
[17:11] <cgregan> davidm: ok
[17:11] <lool> cgregan: I can send you a list of changed packages in the ppa since the last snapshot
[17:12] <davidm> OK, any other opens?
[17:13] <davidm> OK about to close the meeting, going once.........................................................
[17:15] <davidm> OK about to close the meeting, going twice...........................
[17:16] <davidm> #endmeeting
[17:16] <davidm> Again no mootbot, bummer....
[17:16] <davidm> Thanks everyone.
[17:17] <lool> cgregan: What I know I sent to the list and yourself
[17:18] <bspencer> lool:  any outstanding issues with how we are handling the PPA?  complaints, suggestions, etc.?
[17:18] <cgregan> lool: thanks
[17:18] <lool> bspencer: I'm afraid I noticed that you broke the source packages for many of your last uploads
[17:18] <bspencer> what does that mean?
[17:19] <lool> bspencer: For example check hildon-desktop, you moved from .dsc listing .diff.gz + upstream tarball to just a native tarball
[17:19] <lool> There are many broekn source packages which are painful to fix  :-/
[17:20] <lool> The reason it's hard to fix is because we need to use a higher upstream version number than the one you used, but there isn't necessarily a new upstream version to pakage
[17:20] <lool> *package
[17:20] <lool> bspencer: Anyway, this is all moot
[17:20] <bspencer> ok, before you explain the moot point, explain the hildon-desktop pkg issue
[17:20] <lool> bspencer: What will happen in the next days is that we will declare UME released and the ppa should not be uploaded to anymore except for very important fixes and security updates
=== cprov is now known as cprov-out
[17:21] <bspencer> ok.
[17:21] <bspencer> all I did was add a single patch file and "debuild... -S key<>  "
[17:21] <lool> bspencer: Source packages for upstream software need to use version numbers with Debian revisions
[17:21] <bspencer> and the dch number was ubuntu+1 ... ?
[17:21] <lool> bspencer: When you did this, did you get a warning like "This is a native blah (y/n)?"
[17:22] <bspencer> don't recall
[17:22] <lool> bspencer: So whenever you get that warning, you did the wrong thing :)
[17:22] <bspencer> sad that after all this time I still don't know what I did wrong nor how to learn to do it right
[17:22] <lool> bspencer: Before uploading, check what's in your .changes file or .dsc file and that you're about to upload: only a .tar.gz + .dsc => bad, .dsc + .diff.gz + .tar.gz => good
[17:23] <bspencer> how to build the .diff.gz?
[17:23] <bspencer> is there a step before/after "debuild " ?
[17:24] <lool> bspencer: The typical cause for this mistake is when you 1) extract the package into e.g. libfoo-1.0 2) run debuild without the .orig.tar.gz for libfoo (libfoo_1.0.orig.tar.gz) in ../
[17:24] <bspencer> my steps are:   1) dget -x <url>, 2) add patch file, 3) debuild.
[17:25] <bspencer> 2.5) dch -i  
[17:25] <lool> bspencer: The .diff.gz is created when creating the source package (the .dsc); if there's an upstream tarball for the current version of the package, dpkg-source with build a patch (.diff.gz) between your tree and the upstream tarball (good); if the tarball is missing or not named properly, dpkg-source will think this is a "native upload" and create a .Tar.gz of your tree (bad)
[17:25] <lool> bspencer: Your steps shouldn't have caused this
[17:26] <lool> bspencer: But perhaps the previous upload was already broken
[17:26] <bspencer> ok.  I assumed that dget -x  ... would have handled creating the needed upstream tarball
[17:26] <lool> bspencer: Let me check who broke the hildon-desktop package
[17:26] <bspencer> I didn't check it though.   I'll be more careful
[17:26] <lool> bspencer: hildon-desktop is special in that it's in bzr
[17:26] <lool> https://launchpad.net/~ubuntu-mobile/+archive?field.name_filter=hildon-desktop&field.status_filter=any
[17:27] <lool> If you expand Show files  hildon-desktop - 1:2.0.11-1~svn15367-0ubuntu1~804um1 
[17:27] <lool> =>     *  hildon-desktop_2.0.11-1~svn15367-0ubuntu1~804um1.diff.gz  (61.4 KiB)
[17:27] <lool> (good)
[17:27] <lool> Show files  hildon-desktop - 1:2.0.11-1~svn15367-0ubuntu2 (bfiller)
[17:28] <lool> => no .Diff.gz, bfiller broke it
[17:28] <bfiller> lool: huh?
[17:28] <bspencer> bfiller: is such a slacker
[17:29] <bspencer> oy, hi bfiller!
[17:29] <bspencer> :)
[17:29] <lool> bspencer: But let us look at mobile-basic-flash
[17:29] <bspencer> uh oh
[17:29] <lool> Show files  mobile-basic-flash - 0.44-0ubuntu1  (inuka) => good
[17:29] <lool> Show files  mobile-basic-flash - 0.44-0ubuntu2  (bspencer) => bad
[17:29] <bspencer> checking
[17:30] <bspencer> lool:  why do we need a .diff.gz for mobile-basic-flash?
[17:30] <lool> bspencer: Upstream is moblin.org; downstreams could be gaston, Debian, Ubuntu, Fedora etc.
[17:30] <bspencer> if there are no patches
[17:30] <lool> .diff.gz is the packaging and patches for Ubuntu
[17:31] <bspencer> but I took all the patches and pushed them upstream
[17:31] <lool> bspencer: It's the patch between upstream tarball and Ubuntu source package, it should carry debian/
[17:31] <bspencer> oh
[17:31] <lool> It represents the delta between upstream and Ubuntu
[17:31] <bspencer> so all PPA pkgs should have .diff.gz
[17:31] <lool> Yes
[17:31] <lool> To be honest, .diff.gz has some bad characteristics which motivated the use of .tar.gz native packages in the gutsy cycle for mobile
[17:32] <bspencer> ok.  I updated 5-6 pkgs.  Doubtful that any had a .diff.gz
[17:32]  * agoliveira waves quickly to everyone.
[17:32] <bspencer> agoliveira: are you in a parade?
[17:32] <lool> (You can't ship binary files in .diff.gz, so if you e.g. add an image, you need to encode it as text before putting in the .Diff.gz)
[17:32] <bspencer> interesting
[17:33] <agoliveira> bspencer: Yep and ready for your cheering and demostrations of sheer admiration :)
[17:33] <bspencer> all hail agoliveira !! 
[17:33]  * agoliveira waves back
[17:34] <davidm> agoliveira, happy day
[17:34] <agoliveira> davidm: Hi! Busy day actually...
[17:35] <lool> bspencer: We will certainly announce it, but as we're about to release, it would be nice if you could pass around that ppa uploads to hardy should be done carefully if at all starting from now and for the next 18 months
[17:35] <bspencer> lool: I think we can restrict any hardy PPAs to ubuntu requests
[17:36] <bspencer> we will continue to release on moblin.org and you can pull things if you think they are absolutely needed.
[17:36] <bspencer> the one thing that lingers is mobile-basic-flash --> xul 1.9
[17:36] <lool> bspencer: I don't quite now what we will be pulling; certainly security updates will need to be prepared and uploaded to ppa, I don't know which stability fixes will go in
[17:37] <lool> bspencer: This is a terribly missing transition indeed
[17:40] <ToddBrandt> is the weekly ubuntu meeting already over?
[17:41] <lool> Yes
[17:41] <lool> bspencer: By the way, did you talk to asac on the topic?
[17:41] <bspencer> I've sent a few emails, yes.
[17:41] <lool> bspencer: He explained to me where his research brought him, and I can explain what I understood of it if it can help
[17:41] <bspencer> ok
[17:41] <lool> bspencer: How far are you now?
[17:41] <davidm> ToddBrandt, it was very short.  No opens, and no new opens to speak of.
[17:41] <ToddBrandt> lool: any ARs for me?
[17:42] <ToddBrandt> davidm: ahh, thx
[17:42] <bspencer> everything builds, seems ready to go... except that javascript calls from C  don't come through
[17:42] <lool> ToddBrandt: I don't think so; however I would be happy to give you some work :)
[17:42] <davidm> ToddBrandt, no there was one for us and it's been done.
[17:42] <bspencer> the same code that worked with gecko 1.8 doesn't seem to fail, but the Javascript methods don't get called
[17:42] <lool> ToddBrandt: You handle moblin-applets IIRC?  One of our archive admins made comments on a couple of security issues in moblin-applets source code
[17:43] <lool> It's in the main promotion request bug for moblin-applets
[17:43] <bspencer> this means that no apps get shown on the home screen.  It loads without error but is just black.
[17:43] <ToddBrandt> lool: oh, right, I remember a mail on that
[17:43] <ToddBrandt> what's the URL?
[17:43] <lool> bspencer: So the reason it's not called is because the javascript receives blank values for all data becasue the data seems to come from an unsafe source
[17:44] <lool> bspencer: One way to fix this would be to use XUL at the toplevel rather than HTML (and it would make sense anyway)
[17:44] <bspencer> "make sense" perhaps, but is a bigger change. 
[17:44] <bspencer> is there a smaller change that doesn't make as much sense?
[17:45] <lool> bspencer: I don't think it's that big a change
[17:45] <bspencer> Having written a couple of simple xul apps, it is tedious debugging
[17:45] <bspencer> with no clear reason why the seemingly perfect xul page doesn't work
[17:45] <bspencer> and when you figure it out, it is very simple
[17:45] <lool> bspencer: For any other solution, I think asac waits for input from upstream as it's a very sensitive code path and we don't want to endanger the security of e.g. midbrowser...
[17:45] <bspencer> true
[17:46] <bspencer> so convert home.html to home.xul
[17:46] <lool> bspencer: You can have a toplevel xul and mostly HTML below, you don't need to rewrite the whole UI in xul
[17:46] <lool> the toplevel xul container will help from a security context PoV IIUC
[17:46] <bspencer> I will play with it and see if I can get that to work
[17:46] <lool> ToddBrandt: https://bugs.launchpad.net/ubuntu/+source/moblin-applets/+bug/219087
[17:47] <lool> bspencer: At least that's the only way forward I know about for xul 1.9
[17:47] <lool> bspencer: Oh and did people from your team organize for the next 18 months?
[17:48] <lool> bspencer: We will setup a security contact here; we'd be happy to use a security contact in case we need to discuss moblin security holes too, but we also need a channel for security and/or stable updates from moblin
[17:49] <bspencer> organize:  we have had a bit of restructuring to prepare for sustaining and future development
[17:49] <lool> e.g. if you discover a security hole in moblin 1.0 software, we'd like to know about it, we need a patch, preferably against released version of the software; we need a separate channel to ensure proper urgency and discretion is used
[17:49] <bspencer> Mauri can help clarify the details
[17:49] <bspencer> security:  sure, who is organizing this ?
[17:50] <bspencer> asac: hello.  got a sec?
[17:50] <asac> bspencer: yes
[17:50] <lool> bspencer: Who is organizing on which side?
[17:50] <asac> 10 min to be precise :)
[17:50] <lool> bspencer: On the UME side, Canonical people are finalizing this and will probably announce it
[17:50] <bspencer> lool: who is organizing what the separate channel is for security messages and delivering that message to community
[17:51] <bspencer> asac: lool was explaining that for security reasons the C-->Javascript calls are failing
[17:51] <bspencer> and that we probably need to change our top-level HTML to XUL
[17:51] <asac> bspencer: thats my current story
[17:51] <asac> i have to verify that with upstream still
[17:52] <asac> bspencer: we have a javascript issue in epiphany where arguments we pass to a function from c -> javascript are nulled out
[17:52] <bspencer> I'll try to play with it.  It is probably not a big change, but from past experience, getting xul to work was tedious twiddling to get it just right.
[17:52] <bspencer> great news -- someone else has the same issue
[17:52] <asac> we could track that down to place where there is a check for a XUL window
[17:52] <asac> bspencer: i am not sure, but i have the feeling that its the same issue
[17:52] <bspencer> it might be.  If I put "alert ('got here') "  in my javascript functions, they never get called
[17:52] <lool> asac: http://people.dooz.org/~lool/ubuntu-mobile-xul-1.9.txt
[17:53] <asac> bspencer: in firefox running javascript:xxx() in location bar still works, so it cant be something generally broken
[17:53] <lool> (log of this chan before you joined starting with xul 1.9 discusison)
[17:53] <asac> bspencer: does the home screen work when you open it in firefox?
[17:54] <bspencer> hm... how would the C code run?
[17:54] <bspencer> it isn't packaged in an extension format
[17:55] <asac> bspencer: well ... i wonder if we could run your javascript functions through location bar. the C code obviously wont work.
[17:55] <asac> just to see that its really something due to the embedding api
[17:55] <bspencer> I see.  I haven't tried that.   This is something that I didn't change between 1.8 and 1.9
[17:56] <asac> ok. anyway: my kicker is that we keep 1.8 for now, but take care that nothing else is using it. in that way we can transition to xul 1.9 when the security support is dropped for 1.8 without risking regressions outside the mobile-basic-flash
[17:57] <asac> obviously we wont get the memory saving we hoped for for this release
[17:57] <asac> but with thorough QA we can put this transition into an update
[17:57] <asac> would that help?
[17:57] <bspencer> ok
[17:57] <bspencer> sounds like a safe approach
[17:58] <asac> i just wonder how significant this release is? I guess that manufacturers will use the final release + updates to build their images, right?
[17:58] <asac> or will they always use the unmodified release for the next years?
[17:59] <asac> lool: bspencer: ok, so if we go for it, i'd suggest that someone reviews if there are other rdepends on xul 1.8. if there are none, then this transition should not hold back release IMO.
[17:59] <lool> asac: I came to the same conclusion with davidm 
[18:00] <lool> asac: Except that if a xul 1.9 home screen comes up, we would probably be happy to include it as a stable update
[18:00] <asac> lool: ok. for 1.8 security support, i will keep updating it in a timely-fashion
[18:00] <lool> asac: Ok; I didn't expect that, cool then
[18:00] <asac> not as instantly and coordinated with upstream as the 1.9 branch, but i try to be good at it
[18:00] <bspencer> asac: is mobile-basic-flash the last lingering 1.8 user?
[18:00] <asac> same goes for the firefox-2 in universe, which we are updating until it reaches EOL
[18:01] <lool> On my Q1, I only see mobile-basic-flash pulling libxul0d
[18:01] <asac> bspencer: i think so, but someone should review it to be sure
[18:02] <asac> lool: bspencer: midbrowser: merge is ready from what i can tell. now waiting for jimmy to integrate two more patches and bake a packaging release OR for my new ssh keys becoming enabled on moblin again (so that i coudl bake the release).
[18:02] <bspencer> asac: oy -- you've sent new keys and they aren't activated?
[18:02] <bspencer> I'll get that resolved first thing
[18:03] <bspencer> cwong1: 
[18:04] <asac> so either lets get RC1 into UME final release (PPA) or let it go through the normal -proposed -updates process we are going for ffox RC1. 
[18:04] <asac> lool: you decide what you want (guessing that we have the bits by tomorrow morning this time)
[18:04] <asac> lool: b5 in final release and RC1 update in a week? or RC1 in ppa tomorrow?
[18:06] <lool> asac: sounds risky, please delay post UME release
[18:07] <asac> ok that makes sense.
[18:07] <asac> ill think about cherry-picking one tiny bug fix tomorrow
[18:07] <asac> (broken homepage if proxy set=
[18:07] <asac> )
[18:08] <bspencer> lool: there is a weekly meeting (taking place right now) where Canonical meets with Intel to review any open issues / requests.  Tony usually attends but is on vacation.  Currently the list is empty.  (fyi)
[18:08] <lool> bspencer: This is for the customer builds
[18:08] <bspencer> yep, exactly.  
[18:08] <lool> (which is why I don't attend and tony does)
[18:09] <asac> ok thanks ... anything else? otherwise Ill drop out for a while again :)
[18:10] <bspencer> asac: nothing else from me. thx.
[18:15] <sportman> hello
[18:21] <sportman> hey
[18:21] <sportman> is there a version out of this for x86?
[18:34] <lool> sportman: We don't have a generic build yet, but you can install ubuntu-mobile from a regular install on a device, or install the generic kernel from an UME install
=== cprov-out is now known as cprov
[19:15] <davmor2> does any one know of an ebook location so I can try the e-book viewer properly?
[19:20] <GrueMaster> davmor2:  I have access to some through the University of Phoenix (I think).  You also might try amazon.com or Barnes and Noble.
[19:20] <davmor2> GrueMaster: ta
[19:36] <lool> davmor2: project gutenberg has ebooks I think
[19:37] <davmor2> lool: ta. I thought I had found one but they were all pdf :(
[19:40] <sportman> lool so could i install this on a regular gnome ubuntu desktop system?
[19:41] <lool> sportman: I wouldn't recommend installing it on a system you use for anything else
[19:41] <lool> sportman: You should install it in a rather isolated use case: chroot, virtual machine, etc.
[19:42] <sportman> i have a test system lool
[19:43] <sportman> i reformat it about once a week
[19:43] <sportman> i really dont mind
[19:43] <sportman> :)
[19:44] <lool> In this case, you can try installing UME on it directly; perhaps the easiest way is to do a minimal install and add the ubuntu-mobile package, not quite sure
[19:44] <sportman> cool
[19:44] <sportman> alright
[19:44] <sportman> brb
[19:49] <davmor2> Found some I had a thought to check out the fbreader website they link to a few :)
[19:49] <GrueMaster> cool
[19:57] <cl0s> i don
=== asac_ is now known as asac
[23:08] <lool> persia: I started update-manager, pressed "ok" (no password) and the "starting update manager..." dialog remained on the screen forever
[23:08] <lool> Second run didn't display anything more
[23:08] <lool> xsession-errors didn't show anything useful
[23:10] <GrueMaster> davidm:ping  RC1 missing libvorbisfile3, should be included for frozen-bubble sound to work.
[23:11] <davidm> GrueMaster, please file a bug on UME for this.   Thanks
[23:12] <GrueMaster> already did.  #228407
=== robr_ is now known as robr