[12:46] <Mithrandir> tko: who would be a good person to talk with about hildon and translations?
[02:18] <Mithrandir> osso-gwcompat should be in now.
[02:18] <Mithrandir> agoliveira: do you have any idea how much work it'd be to move to the official bluez interfaces?
[02:21] <agoliveira> Mithrandir: I'ts hard to say as I don't have experience on this but for the looks I would say that won't easy. Bluetooth is quite a messy stuff.
[02:21] <Mithrandir> hmm
[02:22] <agoliveira> Mithrandir: and there is some things there, related to proprietary part I guess, that won't make much sense to me but, as I said, I don't have experience with bluez.
[02:23] <Mithrandir> me neither. :-)
[02:25] <agoliveira> Mithrandir:This is something needed, no doubt, but we should check with Nokia to see if they have something already in place, maybe even Intel has something but they didn't say anything about it also.
[02:26] <Mithrandir> nokia seemed to want to transition, but nothing firm yet.
[02:27] <agoliveira> Mithrandir:BTW, after the small "anouncement" I did yesterday, I was contacted by Bob Sinclair (Intel) asking me if he could already use the packages, if he could help, etc. I didn't say anything yet but I believe he can access LP if he wants grab the sources right?
[02:28] <Mithrandir> yes, the branches are publically available.
[02:29] <agoliveira> Fine, I'll tell him that he can jump in from there then.
[02:30] <Mithrandir> as for helping out, I would prefer if people started by contributing by either sending patches to the mailing list or branching off into their own space on launchpad before we give them commit access
[02:30] <Mithrandir> I'm happy being easy on the requirements, but just giving it out completely freely is something I would like to avoid.
[02:31] <agoliveira> Understood.
[02:32] <agoliveira> And it's Bob Spencer. Bob Sinclair is a musician :)
[02:33] <Mithrandir> oh, Bob Spencer is a member of the team, I thought
[02:36] <agoliveira> Yes, he is already.
[02:40] <daniels> Mithrandir: so, you guys doing anything interesting with x?
[02:40] <Mithrandir> daniels: not yet at least.
[02:41] <Mithrandir> we're so far basically just working on getting hildon into ubuntu.
[02:41] <Mithrandir> fixing all kinds of minor and major stuff, like.. completely wrong debian/copyrights and such
[02:42] <daniels> Mithrandir: no surprise there.
[02:43] <daniels> the package quality is ... variable.
[02:43] <Mithrandir> yes, some bits are quite good, some not so much.
[02:43] <Mithrandir> you don't happe to know who'd be the right person to chat with about l10n infrastructure?
[02:47] <daniels> i have no idea, tbh
[02:47] <daniels> i'd probably have to ask the same people you would
[02:48] <Mithrandir> ok, I'll just mail maemo-developers and see what comes of it.
[02:48] <daniels> they don't lurk on m-d; you'd have to talk to nokians directly
[02:49] <daniels> they're sort of part of the corporate machinery
[02:50] <Mithrandir> hm, ok.
[02:50] <Mithrandir> we're open, but not really, and not always, and not all bits.
[02:51] <agoliveira> Mithrandir: got it now? ;)
[02:51] <agoliveira> That's *the* single major complain the maemo developpers have.
[02:52] <Mithrandir> the openness, or lack thereof?
[02:55] <agoliveira> the lack of it.
[02:56] <Mithrandir> in that respect, we have a much better job
[02:56] <daniels> not a discussion i'm really keen to get into, but yeah, talk to your nokia contacts to find the l10n people.
[02:56] <agoliveira> indeed
[02:59] <Mithrandir> s/better/easier/, sorry.
[03:05] <tko_> Mithrandir, I know something about l10n..
[03:06] <tko_> Mithrandir, as I mentioned yesterday, our roadmap is something along the lines of 1) migrate to gnome.org, 2) make stable release, 3) fix po
[03:06] <Mithrandir> ok
[03:06] <Mithrandir> what's blocking those and how can I/we help?
[03:06] <Mithrandir> we're basically trying to answer the question whether we can use malone to translate hildon for gutsy or if that's not feasible.
[03:07] <daniels> hopefully we can move off logical strings soon, too
[03:07] <tko_> well, migration is currently blocking on accounts, then on svn repositories, then on us penetrating the firewall for commit access.. stable release is just blocking on fixing all bugs :)
[03:08] <Mithrandir> daniels: logical strings, as in msgid "sfil_li_folder_root"
[03:08] <Mithrandir> I presume
[03:08] <tko_> and we can't really touch the l10n during this development cycle
[03:08] <daniels> Mithrandir: right
[03:08] <Mithrandir> ok.
[03:09] <tko_> Mithrandir, if malone could display the en_GB string instead of the logical id, it should be simple to use
[03:09] <Mithrandir> gutsy is October, so not feasible for then.
[03:09] <Mithrandir> sorry, not malone, but rather rosetta
[03:09] <tko_> I was wondering about that...
[03:09] <Mithrandir> I'm making a mess of all words today.
[03:09] <Mithrandir> I can ask the rosetta people if that's feasible.
[03:09] <tko_> but figured you must know better :-P
[03:10] <daniels> tko_: well, you have irc now, which is an improvement, so surely svn can't be too far off ;)
[03:11] <Mithrandir> svn-over-irc.
[03:11] <Mithrandir> :-P
[03:11] <tko_> daniels, well, I'm abusing jabber irc transport, which is a bad argument as I really shouldn't be running jabber either :-P
[03:11] <tko_> svn over dns
[03:12] <Mithrandir> ip-over-dns is working quite well, but I suspect you'll get angry employers that way.
[03:13] <daniels> Mithrandir: there are ways to bust tunnels out
[03:39] <Mithrandir> it sounds like rosetta should be able to handle this way of doing translations, so we can use it for gutsy.
[04:07] <lool> Hmm some bzr branches are in an old format and bzr complains about it
[04:07] <lool> Working tree format 3 is deprecated and a better format is available.
[04:07] <lool> e.g. hildon-thumbnail
[04:07] <Mithrandir> just run bzr upgrade and you should be fine
[04:07] <Mithrandir> ideally, hildon-thumbnail shouldn't ship the .bzr directory in the tar.gz at all
[04:07] <lool> Mithrandir: I mean on bazaar.launchpad.net
[04:08] <lool> At least when I bzr co all branches, only some display the "obsolete format" error
[04:08] <Mithrandir> hm
[04:08] <lool> Do you want a list?
[04:08] <Mithrandir> weird.
[04:08] <Mithrandir> that's just your working tree.
[04:08] <Mithrandir> hildon-thumbnail is that way?
[04:09] <lool> I doubt it's local, otherwise I would get the warning on all checkouts
[04:09] <lool> BTW, I'm bootstrapping a pkg-maemo on alioth
[04:09] <lool> And wrote a script to checkout all bzr branches of the ubuntu-mobile team as "/ubuntu"
[04:11] <Mithrandir> bzr co sftp://bazaar.launchpad.net/~ubuntu-mobile/hildon-thumbnail/ubuntu/ hildon-thumbnail ; cd hildon-thumbnail ; bzr status doesn't make it whine here
[04:11] <lool> http://paste.debian.net/29679
[04:11] <lool> Mithrandir: It's with a bzr tarball from upstream installed in my home
[04:12] <lool> Mithrandir: Ah, it might have been interrupted in a previous run
[04:12] <haary> What browser will Ubuntu mobile use?
[04:12] <Mithrandir> haary: custom gecko/xulrunner-based, most likely.
[04:12] <lool> Mithrandir: Right; I wiped everything, and now it's fixed; sorry for the noise
[04:12] <Mithrandir> lool: ah, ok.  Good. :-)
[04:12] <Mithrandir> you had me confused for a bit
[04:13] <lool> Yeah, I was confused too
[04:14] <haary> Oh, I thought it would be something based on Gtk Webcore, since gecko is pretty bloated
[04:30] <lool> Is there some kind of mailing list for tarball releases in maemo?
[04:32] <tko_> lool, we don't do tarball releases... :-] 
[04:32] <tko_> lool, we've requested one for hildon on gnome.org but it'll take time
[04:33] <lool> I'll use http://repository.maemo.org/pool/sardine/main/source/libo/libosso/ as download URL for libosso then
[04:33] <lool> I can't help but smile as I cross Johan Bilien in the ChangeLog mouarf
[04:33] <tko_> mmm?
[04:42] <lool> tko_: We were in the same school and even association
[04:42] <lool> via.ecp.fr
[04:43] <lool> http://via.ecp.fr/via/historique/index.html
[04:45] <tko_> ah
[04:54] <lool> Hmm the maemo version numbers conflict with the Debian ones
[04:54] <lool> I wonder whether Debian packages should use -1ubuntu3debian1 now :)
[05:27] <lool> mdz: Hmm, I think you uploaded mce-dev, but I can't find anything in the ubuntu branch of "mce-dummy"; is this on purpose?
[05:28] <mdz> lool: I just synced it unmodified from the maemo repository, since it didn't require any changes (it's trivial)
[05:28] <lool> Ok, thanks
[05:29] <mdz> I've added it to TODO by way of documentation
[06:37] <bobux> agoliveira, what time zone are you in?  Is it the same as east coast US?
[06:43] <agoliveira> bobux: I don't know, what's US timezone? :) Seriously, I'm at gmt -3.
[06:43] <bobux> I'm pdt, 9:38am now
[06:43] <bobux> you just back from lunch, so 12:38pm?
[06:43] <agoliveira> 13:43 actually.
[06:43] <bobux> I was thinking we were the same but you are working at my 4am
[06:44] <bobux> good to know. thx
[06:44] <bobux> I saw your comment above on LP access and am doing that today
[06:44] <agoliveira> In Brazil it's mostly GMT -3 but some parts are -2 and even -1 IIRC.
[06:44] <agoliveira> Nice to know :)
[06:53] <lool> I cleaned mce-dummy a little; would someone be so kind to have a look at the changes, tell me whether they are not too intrusive, and perhaps merge some in the ubuntu branch?  bzr.debian.org/bzr/pkg-maemo/mce-dummy/debian/
[06:54] <daniels> a diff might be nice, being somewhat easier to review than a bzr branch.
[06:55] <tko> daniels: you wouldn't mind a git branch though, would you? :)
[06:57] <daniels> tko: it would be easier than bzr (given that git lets you clone new repositories with 'old' versions; i couldn't clone postr with edgy's bzr, which sucks horribly), but would still definitely prefer a diff.
[06:57] <lool> http://people.dooz.org/~lool/debian/mce-dummy-diff.patch
[06:58] <tko> diff is easier, true.. shouldn't the bzr web thingamagic be able to produce a diff between arbitrary revisions?
[07:02] <lool> Who should I talk to to get such cleanups in the upstream SVN?
[07:02] <lool> Individual maintainers?
[07:03] <tko> lool: bug in bugzilla and individual maintainers I'd say, at least until we get the hildon mailing list running
[07:04] <tko> lool: IIRC we don't have debhelper 5 yet so that part might become annoying to us if we can't apply the whole patch
[07:05] <lool> tko: Oh yeah; but not bumping debhelper is going to be a pain for all packages, especially those providing -dbg packages :-/
[07:05] <lool> tko: Is there a chance to get an updated debhelper
[07:05] <daniels> yeah, debhelper 5 is an enormous pain, and not solved anytime soon.
[07:05] <lool> aha
[07:05] <daniels> lool: no, and you're still stuck with dpkg 1.13.19 in the devkits.
[07:05] <tko> and as you may have noticed we're not too good at the finer details of packaging either so chances of us tweaking the necessary bits to fit for debhelper 4 are bitty slim
[07:05] <lool> I wonder whether we should refrain from updating debian/compat in this case
[07:07] <tko> daniels: fer was saying something that sdk guys were quick to agree on getting debhelper 5 as it's apparently needed for python bindings
[07:07] <daniels> lool: thanks for the diff.  looks fine, modulo dh5.
[07:07] <lool> The thing is, debhelper 4 is out of my memories; I don't know what expectations I have which would be broken in v4 mode
[07:07] <tko> :)
[07:07] <daniels> tko: ah, nice.  i argued with them about that and dpkg for a while, got nowhere.
[07:07] <lool> For example, I don't recall whether it's ok to wrap build-deps in DH 4
[07:07] <tko> daniels: sdk != si
[07:07] <daniels> lool: no --print-missing (or whatever), --dbg-package semantics reversed (i.e. use --dbg-package=libfoo1 to get libfoo1-dbg)
[07:07] <daniels> tko: yes, both teams. :)
[07:08] <lool> daniels: Yeah, that kind of stuff
[07:08] <lool> Hmm daniels, you said it's unlikely to be very soon, and tko you mentionned is a requirement for python
[07:09] <tko> I'm still sort of wondering if we could get canonical maintain our build environment and tools and such :)
[07:09] <lool> Until when can you get debhelper 5, and how to help?
[07:09] <daniels> tko: if only
[07:09] <lool> tko: hehe
[07:09] <daniels> lool: we don't know, and there's nothing you can do to help, sadly
[07:09] <daniels> dinnertime, bbl
[07:09] <tko> lool: I meant currently the python packaging requires debhelper 5, we need to either get it, or make it work with v4
[07:10] <lool> tko: I don't know whether it's for sale, but I suppose it could help Nokia save money on that part of the maintenance
[07:11] <lool> tko: Python stuff has changed a lot recently; I am not sure what you mean with "python packaging", but restraining your python packages to the old mode of dh_python sounds awful
[07:12] <tko> I've gone through the pain of downgrading some packaging, so yes, I know
[07:12] <tko> but still, in cases, if you want something to happen this year...
[07:12] <lool> tko: Is it discussed internally in Nokia?
[07:13] <lool> Or are there people one can talk to, to explain that upgrading debhelper should be self thanks to its concept of compat modes?
[07:14] <tko> lool: not actively that I know of (would require some higher level business decision) individuals complain occasionally but eventually resign
[07:15] <tko> on debhelper specifically I believe daniels has most recent experience
[07:17] <lool> Hmm where's mce-dev in SVN?
[07:18] <lool> I see mce-dummy Provides mce-dev, but it's not the good mce-dev it seems
[07:34] <tko> lool: it probably isn't in svn (not all teams use stage)
[07:53] <mdz> lool: I don't think I found it in svn
[07:53] <mdz> I only noticed it by accident
[07:54] <mdz> it's essentially a stripped-down version of the proprietary mce package
[07:57] <agoliveira> Indeed. mce-dev is not there, just mce-dummy because of the dependency IIRC.
[07:58] <mdz> agoliveira: no, there are three MCE packages
[07:58] <mdz> mce (the proprietary one), mce-dummy (in svn) and mce-dev (in the archive)
[07:58] <tko> IIRC mce-dummy is historical because it was needed to satisfy mce dependency, but it was later superseded by the real thing
[07:58] <mdz> I pulled in the latter
[07:59] <agoliveira> I meant in the stage svn.
[07:59] <agoliveira> Weird, I don't recall to have other one.
[08:50] <agoliveira> not at all.
[09:24] <happycube> -mtune=pentium seems wrong... the A100 series are pentium-m which is k6
[09:24] <happycube> *P6/ppro
[09:35] <tko> hmm, I just realized I'll be in SF at the same time as apple is organizing their conference. I wonder if I could somehow sneak that into my visit and get nokia pay for it :-P
[10:31] <tfheen> tko: debhelper 4 is acceptable to us if that makes it easier for you guys to merge our changes.
[11:40] <happycube> ubuntu mid can work on something trackpad-based like the asus eee right?
[11:40] <happycube> that looks like a *very cheap* laptopized MID
[11:41] <happycube> also there needs to be provision for not wearing holes in the SSD's that all this gear uses ;)
[11:42] <tko> tfheen: at the moment it would, but fer was trying to get debhelper updated. hard to say how/when that'll go