[10:10] <persia> So, I'm working on a new ubuntu-mid-default-settings package that will replace ume-config-* for intrepid.
[10:11] <persia> Basically, I'm presuming that xinput hotplug should be doing the right thing for intrepid, so we really only need the startup and shutdown stuff.
[10:11] <persia> Some of the environment creation stuff will become the responsibility of the installer.
[10:12] <persia> Also, looking at the moblin-applets package, it seems to do very policy-incompliant with gconf: if anyone can sort that out to work for arbitratily named users, it would be great.
[10:12] <lool> persia: The gconf defaults mechanism in dh_gconf is not user specific, that would be a good option
[10:13] <persia> lool: Sounds good.  Is that something you have time to chase?
[10:13] <lool> Depends for when it's due  :-)
[10:13] <lool> Another thing I noticed is that moblin-applets is using the /desktop/gnome namespace
[10:14] <lool> There might be a good reason for that such as relying on the real GNOME applets to change the keys
[10:17] <persia> I wouldn't think that changes to the postinst to be policy compliant would violate FeatureFreeze, so we ought be safe there.
[10:17] <persia> That said, I'd only extremely limited understanding of why it's doing what it's doing.
[10:26] <lool> persia: I poked dev@moblin.org on this stuff; I fear we're putting the finger in "cleaning up GConf key usage" across moblin
[10:26] <lool> persia: What I fear is that we approach feature freeze and hildon related changes to Ubuntu main packages end up being considered feature additions
[10:27] <persia> That makes sense.  I don't think we can clean everything for intrepid, but I don't think anyone else uses moblin-applets, so it ought be OK.
[10:28] <persia> In other words, as long as the chances of breaking anything for anyone else are nil, and we're dealing with something so incredibly messy in one of our packages, I think it will make our lives easier to fix it.
[10:33] <lool> persia: I'll try to remember to move these to dh_gconf
[10:33] <lool> persia: Did you rename ume-config-common already?  Are you aware it's an upstream module?
[10:36] <persia> I wasn't renaming, so much as dropping it entirely in favour of something else.  I wasn't aware it was an upstream module.
[10:37] <persia> My specific rationale is that I don't think that we get any useful benefits from hardcoding each config: it essentially means that we can't create images that work properly on arbitrary hardware,
[10:38] <lool> persia: Anyway, I agree it should die; I just mention it FYI
[10:39] <persia> lool: OK.  I was afraid you had some reason to keep it.
[10:44] <ogra> you guys want global gconf changes ? why dont you do them in /usr/share/gconf/defaults/ ?
[10:45] <ogra> (thats why my gnome MID desktop will do at least, i think we use it stadardized in the main distro as well)
[10:55] <persia> ogra: I thought you were working on ubuntu-mobile, not ubuntu-mid.  Am I mistaken?
[10:56] <ogra> no, i still get the wording wrong :)
[10:57] <ogra> though what i understood from davidm was that he wants me to do http://people.ubuntu.com/~ogra/ume/
[10:57] <ogra> which is definately closer to -mid atm
[10:58] <persia> Hrm.  I thought we were using hildon again for -mid, and using that for -mobile.  We definitely need to block out and document the featureset better next cycle.
[10:58] <ogra> definately
[10:59] <ogra> but next cycle we wont be busy with merges and PPA moving for 2/3 of the development time
[10:59] <ogra> and we will be able to write proper specs in advance
[11:00] <persia> Well, there's that.  Also, next cycle we'll get to start at the beginning, rather than having our release be well after UDS.
[11:00] <ogra> right
[11:16] <lool> ogra: That's where we intend to have them
[11:16] <persia> OK. I lose.  I wanted to make the system not depend on the username, but without depending on the username, we can't autologin.
[11:16] <persia> That said, I'd like to change the username from "ume" to "ubuntu".  Any objections?
[11:16] <lool> ogra: dh_gconf is the best way to put them in /usr/share/gconf/defaults
[11:18] <ogra> ah, i usually just put the file with keys/values in there and run update-gconf-defauls :)
[11:18] <ogra> i bet thats what dh_gconf does in postinst then 
[11:18] <ogra> persia, since ubuntu is what we use on the liveCD that sounds fine
[11:18] <lool> ogra: Putting the files and running update-gconf-defaults is what dh_gconf does!
[11:19] <lool> But then you must have missed its introduction as both tools were written at the same time
[11:19] <ogra> heh
[11:19] <ogra> i didnt do that from packages since ages :)
[11:20] <ogra> (its about a year ago that i touched edubuntu-artwork last, which does these settings for edubuntu ... in the classmate the installer puts the file in place and calls the command)
[11:54]  * ogra-Q1 hates daily disconnects
[11:54] <ogra-Q1> wow, cellwriter really makes a good kbd
[11:56] <persia> ogra-Q1: screenshot?
[11:56] <davmor2> ogra-Q1: isn't it cellwriter that is predictive or is that something else?
[11:56]  * ogra-Q1 is in "eat your own dogfood" mode today
[11:58] <ogra-Q1> davmor2, cw is actually writing recognition... but comes with an internationalized kbd as well
[11:58] <davmor2> ogra-Q1: okay cool :)
[11:59] <ogra-Q1> the o ly drawback is that it sts on top with metacity
[12:04] <ogra-Q1> persia,  http://people.ubuntu.com/~ogra/cellwriter.png
[12:05] <persia> ogra-Q1: Looking very nice indeed :)
[12:06] <ogra-Q1> with  a bitmore trainig and better screen calibration i could rellally type like with any other kbd
[12:06] <ogra-Q1> i got relly fast within 10 mins of usage
[12:54] <lool> Meeting in 5 mns in #ubuntu-meeting
[12:59] <lool> Meeting starting soon in #ubuntu-meeting
[13:04] <emgent> lool: ok thanks
[22:56] <nxvl> good afternoon