**** BEGIN LOGGING AT Sat Oct 18 02:59:57 2008
Oct 18 07:25:31 <quickdev> Hey
Oct 18 07:41:41 <sicu> hey
Oct 18 07:46:05 <sicu> quickdev: in SHR, do i need openmoko-agpsui to get a fix in tangoGPS ?
Oct 18 07:50:31 <sicu> quickdev: i flashed the latest SHR image (17oct) which had tangoGPS already ... is this supposed to work as it is, or do i need to install other bits?
Oct 18 07:54:16 <quickdev> sicu, it should work without agpsui
Oct 18 07:54:48 <sicu> hmmm ... no such luck :/
Oct 18 07:55:15 <quickdev> but maybe there's a problem - will test it later
Oct 18 07:56:10 <sicu> is there a command needed to activate/start gps, or is this done automatically?
Oct 18 08:02:45 <sicu> well, i'll leave it outside in the cold for a while with an ext antenna
Oct 18 08:09:09 <quickdev> sicu, it *should* be done automatically
Oct 18 08:30:32 <sicu> quickdev: ok, thanks ... i've left it outside running agpsui, so should get a fix soon i hope
Oct 18 08:31:29 <quickdev> should be there in less than 2 minutes with an external antenna
Oct 18 08:32:31 <sicu> ah, crap
Oct 18 08:33:04 <sicu> well, i just bought the ext antenna ... perhaps it's broken :/
Oct 18 08:38:59 <sicu> hmmm, i just gave it a hot reset and suddenly ttff was 4 seconds, and now it's found 8 sats .... so it must've been freezing ;p
Oct 18 08:43:52 <sicu> quickdev: when changing power settings (dimm and suspend times), is there a way to force the changes without having to restart?
Oct 18 08:45:11 <quickdev> does setting the dimm or suspend times work at all atm?
Oct 18 08:45:12 <sicu> quickdev: as it is now if i change time before suspend from say '30 sec' to 'Never', it will still keep suspending until i reboot :/
Oct 18 08:45:32 <quickdev> sicu, could you add a ticket?
Oct 18 08:45:35 <sicu> oh yeh ... they do for me
Oct 18 08:45:42 <sicu> quickdev ... will do
Oct 18 10:40:00 <quickdev> Hey
Oct 18 10:40:28 <Hire> hi quickdev
Oct 18 10:49:05 <Ainulindale> sicu: The settings aren't used
Oct 18 10:53:58 * raster is disturbed
Oct 18 10:54:15 <raster>  1256 root      20   0 19736  12m 5092 S  1.0 10.3   0:31.27 python             
Oct 18 10:54:22 <raster> frameworkd is fat
Oct 18 10:54:35 <raster> what on earth is is nuking 7m of ram for?
Oct 18 10:55:26 <mickey|zzZZzz> a GSM server, a gps server, a resource manager, a settings daemon, an events daemon
Oct 18 10:55:28 <mickey|zzZZzz> all in one
Oct 18 10:55:31 <mickey|zzZZzz> :D
Oct 18 10:56:04 <Hire> a pim deamon
Oct 18 10:56:14 <Hire> a future network deamon
Oct 18 10:56:58 <raster> mickeyl: thats almost double e's memory footprint
Oct 18 10:56:59 <raster> :)
Oct 18 10:57:07 <raster> and it's doing a lot more memory heavy stuff
Oct 18 10:57:07 <raster> :)
Oct 18 10:57:51 <raster> all of those things you shoudl be able to get away with in 1-2m
Oct 18 10:57:56 <raster> but its at about 7
Oct 18 10:58:03 <raster> (excluding shared pages)
Oct 18 10:58:18 <raster> methinks u have leaks
Oct 18 10:58:29 <Ainulindale> How dare you ! :-)
Oct 18 10:58:42 <raster> or... the code is utter junk (whihc i seriously doubt)
Oct 18 10:58:52 <raster> or... python is not being nice to you on the memory-front
Oct 18 10:58:56 <raster> not one tiny bit
Oct 18 10:58:57 <raster> :(
Oct 18 10:59:49 <mickeyl> dunno, have you watched it over a longer term?
Oct 18 10:59:51 <raster> the only reason i bring this up is...
Oct 18 11:00:03 <raster> this prettu much is going to put a major spanner int he works of getitng fso sanely onto my treo
Oct 18 11:00:05 <raster> :(
Oct 18 11:00:11 <raster> mickeyl:  no - this is just after boot
Oct 18 11:00:31 <raster> but its noew been about 10 mins
Oct 18 11:00:35 <raster> so its had time to gc
Oct 18 11:00:51 <raster> the main thing is thats 7m of nopn-pagable pages of ram
Oct 18 11:00:55 <raster> cant page them back to disk
Oct 18 11:01:03 <raster> if heavily fragmented
Oct 18 11:01:07 <raster> :(
Oct 18 11:01:53 <raster> do you do any memory profiling for fwd?
Oct 18 11:02:27 <mickeyl> not at all
Oct 18 11:02:34 <mickeyl> we're still not in optimizing mode
Oct 18 11:02:46 <mickeyl> after 0.9 we can do all that 
Oct 18 11:02:53 <mickeyl> but i still need to get the promised functionality done
Oct 18 11:02:56 <raster> ok
Oct 18 11:03:04 <raster> just that i was astounded at the footprint
Oct 18 11:03:13 <raster> and it's tickled my "something is wrong here" bone
Oct 18 11:03:17 <mickeyl> hehe
Oct 18 11:03:20 <mickeyl> call python
Oct 18 11:03:24 <mickeyl> and check when it's idle
Oct 18 11:03:27 <mickeyl> should already have some megs
Oct 18 11:03:28 <raster> so you may find some major oversight/bug/design problem if u do a bit of profiling now
Oct 18 11:03:32 <mickeyl> i think it's the base footprint
Oct 18 11:03:37 <raster> its idle
Oct 18 11:03:42 <raster> my sim card has nothnig on it
Oct 18 11:03:47 <raster> in fact doesnt even work
Oct 18 11:03:49 <raster> no contacts
Oct 18 11:03:53 <raster> no calls made or sms's
Oct 18 11:04:00 <raster> so it shoudl be pretyt much an empty fwd
Oct 18 11:04:15 <mickeyl> try python
Oct 18 11:04:21 <raster> ?
Oct 18 11:04:21 <mickeyl> just call the command line
Oct 18 11:04:27 <mickeyl> and then check the footprint
Oct 18 11:04:35 <mickeyl> should already have some megs
Oct 18 11:04:47 <mickeyl> so i don't think framework is doing too bad
Oct 18 11:04:53 <mickeyl> but anyways, will profile at some point of time
Oct 18 11:05:01 <raster>  1382 root      20   0  5164 2868 1700 S  0.0  2.3   0:00.27 python             
Oct 18 11:05:13 <raster> base footprint to interactive cmdline is 2.8-1.7m (shared)
Oct 18 11:05:14 <raster> so about 1m
Oct 18 11:05:19 <raster> before doing anything
Oct 18 11:05:28 <raster> cool
Oct 18 11:05:37 <raster> not sure how u profile memory usage in python...
Oct 18 11:06:51 <mickeyl> there's some cool things you can do with the garbage collector
Oct 18 11:07:15 <raster> maybe the gc isnt being set to be aggressive enough?
Oct 18 11:07:38 <mickeyl> could be
Oct 18 11:07:47 <mickeyl> i think one should try individual subsystems first
Oct 18 11:07:49 <raster> thats just why i mentioend it
Oct 18 11:07:52 <raster> might be worht a look
Oct 18 11:08:04 <DocScrutinizer2> we are looking into MOKO-fw to see what can be done for echo issue
Oct 18 11:08:25 <raster> but a idle, just from boot with no data or stuff to actualyl do fwd should be much less than 7m of non-pagable ram :)
Oct 18 11:10:05 <mickeyl> don't forget it's a VM
Oct 18 11:10:06 <mickeyl> after all
Oct 18 11:10:25 <raster> sure
Oct 18 11:10:29 <raster> that comes with overhead
Oct 18 11:10:33 <raster> agreed
Oct 18 11:10:42 <raster> but i'm amazed at its overhead
Oct 18 11:11:13 <raster> i originally thoguht "oh. thats so simple and lightweight as a back-end service - it can stay in python. not speed or memory critical"
Oct 18 11:11:19 <raster> but now i am thinking otherwise
Oct 18 11:11:21 <raster> :(
Oct 18 11:11:37 <mickeyl> well
Oct 18 11:11:57 <mickeyl> if that couple of megs is disturbing you, you have to write code for a couple of years
Oct 18 11:12:06 <mickeyl> before you are up to par w/ the same functionalit
Oct 18 11:12:06 <mickeyl> y
Oct 18 11:12:09 <raster> it is disturbing me
Oct 18 11:12:42 <raster> i'm not convinced
Oct 18 11:13:02 <mickeyl> *shrug*
Oct 18 11:13:04 <Ainulindale> Considering the fact that the amount of memory is going up and up on each and every device
Oct 18 11:13:08 <Ainulindale> I seriously think that's not a matter
Oct 18 11:13:18 <Ainulindale> You're just jealous because e can't have that memory ! :-)
Oct 18 11:13:24 <raster> no
Oct 18 11:13:28 <raster> i am worried about e';s usage
Oct 18 11:13:33 <raster> but i actually know where e's usage goes
Oct 18 11:13:38 <raster> i can detail it for you if u want
Oct 18 11:13:39 <raster> :)
Oct 18 11:13:48 <Ainulindale> Someday, yes, I'm interested :-)
Oct 18 11:13:54 <Ainulindale> I tried e a long long time ago
Oct 18 11:14:01 <Ainulindale> And didn't for a while, I should retry
Oct 18 11:14:15 <raster> i regularly punt it thru memoroif
Oct 18 11:14:16 <quickdev> Ainulindale, trying to setup that serial forwarder atm ;)
Oct 18 11:14:23 <raster> and anylise where the most # of allocations go
Oct 18 11:14:26 <raster> and the biggest amoutn of ram
Oct 18 11:14:30 <raster> and biggest chunks
Oct 18 11:14:32 <raster> who allocs them
Oct 18 11:14:33 <raster> why
Oct 18 11:14:35 <raster> and whats there
Oct 18 11:14:42 <Ainulindale> raster: I do that from times to times also
Oct 18 11:14:46 <raster> sometimes i have been surprised
Oct 18 11:14:48 <Ainulindale> valgrind/dmalloc
Oct 18 11:14:54 <raster> i found i had alloced about 10,000 strings
Oct 18 11:15:05 <raster> a few thousand were ""
Oct 18 11:15:10 <raster> and there were lots of repeats
Oct 18 11:15:14 <raster> ie each object making its own copy
Oct 18 11:15:17 <Ainulindale> Hhe :-)
Oct 18 11:15:19 <raster> of a file path
Oct 18 11:15:26 <mickeyl> you will only get such a quality of optimizing if you do all yourself
Oct 18 11:15:31 <mickeyl> so you have to make a decision
Oct 18 11:15:32 <raster> i quickly qhipoped up a "read only strign token sharere"
Oct 18 11:15:33 <mickeyl> reinvent everything
Oct 18 11:15:39 <raster> (basically just a simple string hash with refcounts)
Oct 18 11:15:45 <mickeyl> or just live with some things
Oct 18 11:16:01 <raster> mickeyl:  the problem is - on 32m - its just not going to fly :(
Oct 18 11:16:12 <raster> so fso will be for big hefty devices only
Oct 18 11:16:13 <raster> :(
Oct 18 11:16:20 <mickeyl> un true
Oct 18 11:16:27 <mickeyl> first, we did _NOT_ do _ANY optimization
Oct 18 11:16:39 <mickeyl> second, the plan is to integrate Vala plugins for some things
Oct 18 11:16:47 <raster> imho u should optimise as u go
Oct 18 11:16:50 <Hire> yes, raster, you need to wait after 0.9 is release
Oct 18 11:16:53 <raster> dont "wait until the end"
Oct 18 11:16:55 <mickeyl> nope
Oct 18 11:16:58 <mickeyl> that's not my way
Oct 18 11:17:00 <mickeyl> sorry
Oct 18 11:17:02 <Ainulindale> raster: this kind of thing introduces more problems than it solves
Oct 18 11:17:06 <raster> as u invariably uncover design faults
Oct 18 11:17:17 <Ainulindale> (Or that was my experience with optimization during a project)
Oct 18 11:17:29 <raster> Ainulindale: not my experience
Oct 18 11:17:33 <quickdev> "premature optimization is the root of all evil", hehe..
Oct 18 11:17:35 <Ainulindale> I see that :-)
Oct 18 11:17:38 <raster> yes
Oct 18 11:17:43 <raster> u can go quote all u like
Oct 18 11:17:48 <raster> iv'e found otherwise
Oct 18 11:17:57 <raster> over the years
Oct 18 11:17:57 <Ainulindale> raster: Though I think e is kind of more settled regarding its functionnalities set than frameworkd is
Oct 18 11:18:08 <raster> dont sit and try and squeeze every cycle out of something
Oct 18 11:18:20 <raster> but at least look for the major junk and nasties
Oct 18 11:18:23 <raster> and nuke them early
Oct 18 11:18:46 <raster> Ainulindale:  hmm... u might be surprised how much of e's internals shuffle about and what it does chnages
Oct 18 11:18:52 <raster> iv'e had a hard time keepign it in check.
Oct 18 11:19:14 <Ainulindale> :-)
Oct 18 11:19:40 <Ainulindale> raster: anyway, your approach isn't bad, it's just different. I'm myself more like mickeyl in the way to do things
Oct 18 11:19:45 <raster> every day someone comes and asks for somethng not there and not intended
Oct 18 11:19:50 <raster> but sometiems it slips in because its easy"
Oct 18 11:19:53 <raster> and make a user happy
Oct 18 11:19:57 <quickdev> mickeyl, sometimes this appears: http://rafb.net/p/oH2wAs24.html - is that a problem you mentioned yesterday?
Oct 18 11:20:06 <Ainulindale> I found out that if you can't have a definitive set of functionnalities, you won't have a good project vision, with all the problems you found out while developing
Oct 18 11:20:14 <raster> i do agree that spending all your dev cycles trying to tweak the crap out of everyting is a waste
Oct 18 11:20:16 <Ainulindale> But that's only for me :-)
Oct 18 11:20:28 <raster> but early on - as you go, just keeping the house 'mostly clean" is good
Oct 18 11:20:34 <Ainulindale> raster: indeed
Oct 18 11:20:41 <raster> as i said
Oct 18 11:20:54 <raster> sometimes your profiling (for mem or speed) bring sup much nasty design errors
Oct 18 11:20:55 <raster> eg
Oct 18 11:21:08 <raster> u decided u'd just duplicate the data when passing it aroudn
Oct 18 11:21:26 <raster> but reality shows its duplicated more than u ever thought
Oct 18 11:21:33 <mickeyl> quickdev: yeah
Oct 18 11:21:36 <raster> and so u need to come up with some complex zero-copy sharing mechanism
Oct 18 11:21:38 <raster> (for example)
Oct 18 11:21:38 <admiral0> hello raster
Oct 18 11:21:48 <raster> and sometimes that directly impacts the design
Oct 18 11:21:57 <raster> as u just cant do things anymore
Oct 18 11:22:02 <raster> or need to do them differently
Oct 18 11:22:05 <raster> admiral0:  yo
Oct 18 11:22:07 <raster> anyway
Oct 18 11:22:28 <DocScrutinizer2> raster: as soon as you *know* where crap is, you may go ahead dealing with it later
Oct 18 11:22:29 <raster> i was just wondering if there was any knowledge of why its so heavy when it essentially should be empty and not doing anything
Oct 18 11:22:45 <raster> DocScrutinizer2:  thats what i asked
Oct 18 11:22:46 <raster> eg
Oct 18 11:22:57 <DocScrutinizer2> i know
Oct 18 11:23:37 <raster> if it was "yeah - we know. we put all gsm messages in a queue in ram and run thruy that quque and rebuild the state machine from it so as to get it right - we shoudl throw them out and only keep state data in the state machine, but thats riscky as we may over-optimise"
Oct 18 11:23:45 <raster> if there was an answer.. that'd be cool :)
Oct 18 11:24:07 <DocScrutinizer2> ack
Oct 18 11:24:10 <raster> "we create an empty 5mb in-ram pim database just for speed/testing.. so don't worry"
Oct 18 11:24:11 <raster> :)
Oct 18 11:26:06 <quickdev> mickeyl, do you know why that read write errors occur?
Oct 18 11:30:57 <mickeyl> quickdev: no
Oct 18 11:31:07 <mickeyl> did you get it to work one time?
Oct 18 11:31:10 <quickdev> ok, will try to figure out
Oct 18 11:31:21 <mickeyl> i just seen it happening once you close a pty and then try to reopen another one
Oct 18 11:31:26 <quickdev> mickeyl, it's partly working...very unstable....but some dbus calls are answered
Oct 18 11:31:30 <mickeyl> right
Oct 18 11:31:36 <mickeyl> i'd be really glad if you find something
Oct 18 11:31:44 <quickdev> and I have to restart gsm0710muxd each time...as you said
Oct 18 11:32:56 <mickeyl> yep
Oct 18 11:38:59 <sicu> Ainulindale: well, i'm 100% sure that the 'Blank Time' settings work
Oct 18 11:40:33 <sicu> just try it out ... the settings apply straight away, no need to reboot
Oct 18 11:41:15 <sicu> but 'Suspend After Blank' requires a reboot before it takes affect
Oct 18 11:42:40 <sicu> anyways, i created a ticket for it
Oct 18 11:47:37 <Ainulindale> sicu|zz: the illume settings you mean ?
Oct 18 11:47:43 <Ainulindale> They never worked for me here
Oct 18 11:48:05 <Ainulindale> raster: are the illume settings doing anything with frameworkd ?
Oct 18 11:48:36 <raster> Ainulindale:  none
Oct 18 11:48:39 <raster> nada
Oct 18 11:49:26 <Ainulindale> raster: Oh by the way, a little question
Oct 18 11:49:51 <Ainulindale> Does it seem possible to you to build a plugin/gadget for illume to manage the blank/lock/suspend settings ?
Oct 18 11:50:04 <Ainulindale> i.e. tie the illume settings with events in frameworkd
Oct 18 11:50:14 <Ainulindale> and tie frameworkd signals to illume lock abilities 
Oct 18 11:50:53 <raster> anything is possible in e with modules
Oct 18 11:50:55 <raster> its like saying
Oct 18 11:51:09 <raster> "is it possible to write a process in linux that displays something on the screen?"
Oct 18 11:51:10 <raster> :)
Oct 18 11:51:24 <Ainulindale> raster: Then could you point me to some resources on how to do that ? :-)
Oct 18 11:52:13 <raster> there's 60 moduels that ship with e17
Oct 18 11:52:15 <raster> for examples
Oct 18 11:52:17 <raster> some simple
Oct 18 11:52:18 <raster> some complex
Oct 18 11:52:37 <raster> theres a whole E-MODULES-EXTRa svn dir in e17 svn with more modules...
Oct 18 11:52:42 <raster> moutnains fo examples
Oct 18 11:52:43 <raster> :)
Oct 18 11:52:43 <Ainulindale> Hmm now that I think about it I could use the gsm gadget
Oct 18 11:53:08 <raster> gsm gadget is part of illume (the module)
Oct 18 11:53:13 <Ainulindale> I would like to build a sound settings plugin for the top shelf
Oct 18 11:53:18 <raster> a module can provide as manu or as few gagdets as it wants
Oct 18 11:53:19 <raster> :)
Oct 18 11:53:20 <Ainulindale> And to build an invisible gadget for suspend
Oct 18 11:53:35 <raster> dotn need an invisible gagdte
Oct 18 11:53:37 <raster> just a module
Oct 18 11:53:40 <Ainulindale> Ok then :-)
Oct 18 11:53:43 <raster> think of it this way
Oct 18 11:53:47 <raster> moduels are like processes in linux
Oct 18 11:53:51 <raster> they dont HAVE to have a window
Oct 18 11:53:54 <raster> or connect to x
Oct 18 11:53:56 <raster> or they can
Oct 18 11:53:59 <raster> depends what they want to do
Oct 18 11:54:00 <raster> :)
Oct 18 11:54:13 <raster> eg
Oct 18 11:54:18 <raster> the dropshadows in e.. is done by a module
Oct 18 11:54:26 <raster> the alt-tab box that pops up is a module
Oct 18 11:54:31 <raster> alt+escapoe exec box - module
Oct 18 11:54:46 <raster> got moduels that do pengiuns runing aroudn your desktop
Oct 18 11:54:54 <raster> or flame and rain and clouds
Oct 18 11:54:58 <raster> ones thst do gadgets in shelves
Oct 18 11:55:00 <raster> or on the desktop itself
Oct 18 11:55:09 <raster> a module even manages N other gadgets on the desktop
Oct 18 11:55:33 <raster> moduels are not sandboxed into just doing gadgets
Oct 18 11:55:34 <raster> :)
Oct 18 11:56:28 <sicu|zz> something weird is going on then ... because if i select illumes 1 sec to dimm .... it takes one second, if i select 30 secs to dimm, it takes 30 secs
Oct 18 11:57:00 <sicu|zz> all your settings are belong to illume
Oct 18 11:57:38 <raster> illume sets x's screensvaer
Oct 18 11:57:39 <raster> thats all
Oct 18 11:58:39 <sicu|zz> so that's what's happening =]
Oct 18 11:58:52 <raster> just lets x handle it from there
Oct 18 11:59:02 <raster> and when x issues an event like "screensaver/blank just kicked in"
Oct 18 11:59:13 <raster> illume m,ay also issue a "please suspend now" request
Oct 18 11:59:18 <raster> or anything it likes
Oct 18 12:02:31 <DocScrutinizer2> all your base are belong to us
Oct 18 12:02:55 <raster> e and illume are pretty good at taking over the world
Oct 18 12:10:20 <quickdev> mickeyl, 2008.10.18 14:02:13 ogsmd    DEBUG    <CallChannel via /dev/pts/12>: got 19 bytes from: '\r\n+CMS ERROR: 310\r\n'
Oct 18 12:10:20 <quickdev> 2008.10.18 14:02:13 ogsmd    DEBUG    <CallChannel via /dev/pts/12>: COMPLETED 'AT+CSMS=1' => ['+CMS ERROR: 310']
Oct 18 12:10:35 <quickdev> that appears in my logs...although I have a SIM card installed
Oct 18 12:10:46 <quickdev> experienced that before?
Oct 18 12:11:12 <mickeyl> that's ok
Oct 18 12:11:30 <mickeyl> +CSMS=1 will return 310 if your SIM card is not unlocked yet
Oct 18 12:11:35 <mickeyl> 310 is kind of wrong there
Oct 18 12:11:39 <mickeyl> but hey... that's GSM
Oct 18 12:11:41 <mickeyl> :)
Oct 18 12:11:59 <mickeyl> everyone follows the standard only as far as they could understand it
Oct 18 12:12:19 <quickdev> ok
Oct 18 12:12:28 <mickeyl> i should move CSMS=1 to the chunk of commands only sent once your SIM is unlocked
Oct 18 12:12:32 <mickeyl> but it's just cosmetic
Oct 18 12:12:42 <mickeyl> since it's the default anyawys
Oct 18 12:16:44 * raster is happy the at commandset gsm pain is being done somewhere far away from him
Oct 18 12:16:46 <raster> :)
Oct 18 12:17:25 <mickeyl> :D
Oct 18 12:17:28 <mickeyl> be glad
Oct 18 12:17:33 <raster> yeah
Oct 18 12:17:33 <mickeyl> it's a dark corner
Oct 18 12:17:40 <raster> dont get me wrong abotu complaining on the memory thing
Oct 18 12:17:47 <raster> i do hope that when u get to it
Oct 18 12:17:50 <raster> u'll profile it nicely
Oct 18 12:18:07 <raster> it just was far beyond what i dreamed it would use at this stage
Oct 18 12:18:18 <mickeyl> i will profile that for sure
Oct 18 12:18:32 <mickeyl> i just think it's mainly python itself and allocating a ton of dbus things
Oct 18 12:18:43 <mickeyl> but anyways, i will scrutinize it
Oct 18 12:18:48 <mickeyl> as well as run time perf.
Oct 18 12:19:09 <mickeyl> my personal goal is OpenEZX
Oct 18 12:19:11 <mickeyl> 48Megs :)
Oct 18 12:19:12 <raster> that could be it
Oct 18 12:19:16 <mickeyl> so that's pretty close to your 32
Oct 18 12:19:23 <raster> with al lthe introspection and niceness it adds to doing dbus
Oct 18 12:19:27 <raster> it adds overhesad too
Oct 18 12:19:29 <mickeyl> yep
Oct 18 12:19:41 <raster> aaah
Oct 18 12:19:42 <raster> 48
Oct 18 12:19:46 <raster> ok
Oct 18 12:19:50 <raster> well keep in mind the 32
Oct 18 12:19:51 <mickeyl> i promised to make it work w/ 48
Oct 18 12:19:55 <raster> in some ways i like 32m
Oct 18 12:19:57 <mickeyl> so I'm bound to that
Oct 18 12:20:04 <raster> as android claims that as their minimum platform req
Oct 18 12:20:13 <mickeyl> ah, didn't know that
Oct 18 12:20:15 <raster> so... i think it'd behove us to equal it at least
Oct 18 12:20:18 <raster> yeah
Oct 18 12:20:25 <raster> just the
Oct 18 12:20:32 <raster> "well if they can... damnit! we can!"
Oct 18 12:21:03 <raster> fyi
Oct 18 12:21:06 <quickdev> alphaone, do you remember the dbus timeout patch?
Oct 18 12:21:08 <raster> on a 32m platform
Oct 18 12:21:13 <raster> after kernel and goop u have 29m
Oct 18 12:21:22 <raster> with x and base illume u have about 17m left
Oct 18 12:21:53 <raster> (Thats just dbus daemon e + all its stuff, a suprios waker daemon i wrote for the alarm and some scripts and minoir daemons around)
Oct 18 12:22:07 <raster> thats why 7 m of non-swappable pages was a worry
Oct 18 12:22:13 <raster> as it east through 7/17 left
Oct 18 12:22:19 <raster> not including disk cache/mapped files
Oct 18 12:22:24 <raster> so just a heads up
Oct 18 12:22:25 <raster> :)
Oct 18 12:22:39 <mickeyl> oki
Oct 18 12:23:12 <mickeyl> quickdev: what's with that?
Oct 18 12:23:35 <quickdev> mickeyl, I need a greater timeout for dbus on my host system
Oct 18 12:23:54 <quickdev> If there isn't another failure
Oct 18 12:23:55 <quickdev> ** ERROR **: Unknown error: Did not receive a reply. Possible causes include: the remote application did not send a reply, the message bus security policy blocked the reply, the reply timeout expired, or the network connection was broken. dbus -2
Oct 18 12:24:13 <mickeyl> client or server timeout?
Oct 18 12:24:23 <mickeyl> try first w/ giving a larger client timeout
Oct 18 12:24:30 <quickdev> client timeout
Oct 18 12:24:31 <mickeyl> and check the logs (the server might have been died)
Oct 18 12:25:31 <mickeyl> default client timeout you can configure with --dbus-default-reply-timeout
Oct 18 12:26:00 <mickeyl> although we're using 60 seconds atm.
Oct 18 12:26:04 <mickeyl> that should really be ok
Oct 18 12:26:09 <mickeyl> which operation takes longer for you?
Oct 18 12:27:01 <quickdev> mickeyl, http://rafb.net/p/Cu5ore84.html
Oct 18 12:28:06 <CIA-57> freesmartphone.org: 03mickey 07framework * r5c89c8b8495c 10/framework/subsystems/ousaged/generic.py: ousaged: improve error specs, clean up, bump version
Oct 18 12:28:06 <CIA-57> freesmartphone.org: 03mickey 07framework * r5b2e4fdf7b30 10/framework/subsystems/ousaged/generic.py: ousage: fix potential errors detected by pychecker
Oct 18 12:28:32 <CIA-57> freesmartphone.org: 03mickey 07framework * r1f9002235a49 10/framework/subsystems/ogsmd/modems/abstract/channel.py: ogsmd: send +CSMS=1 only after the SIM has been unlocked
Oct 18 12:29:08 <mickeyl> quickdev: sounds like something else is broken. SetAntennaPower should not take 60 seconds, never.
Oct 18 12:30:05 <quickdev> it's not set antenna power....that callback is called....I'm just seeing: The set antenna power callback calls sim_auth_get_status()...that may be the evil
Oct 18 12:40:18 <quickdev> mickeyl, it's definteley set antenna power
Oct 18 12:40:57 <mickeyl> if that's on your host, you may have a default delay of 25 seconds
Oct 18 12:41:05 <mickeyl> try increasing it
Oct 18 12:41:39 <quickdev> yes, it's on my host
Oct 18 12:51:52 <quickdev> mickeyl, in order to configure dbus with a default retry timeout of 25 seconds I have to recompile it, right
Oct 18 12:51:54 <quickdev> ?
Oct 18 12:52:42 <mickeyl> unfortunately yes
Oct 18 12:52:44 <mickeyl> ~lart dbs
Oct 18 12:52:48 <mickeyl> dbus, that is
Oct 18 12:52:58 <mickeyl> s/retry/response/ bt.
Oct 18 12:53:00 <mickeyl> w.
Oct 18 14:50:26 <Kensan> raster: oe is broken again for me :(
Oct 18 14:50:58 <Kensan> raster: need to clean up some stuff before I can get some real work done *hmpf*
Oct 18 14:52:33 <raster> ugh
Oct 18 14:58:20 <lindi-> mickey|bbl: btw, is there some way to restart frameworkd without having to also restart all apps that talk to it over dbus?
Oct 18 15:01:53 <mickey|bbl> yes and no
Oct 18 15:02:17 <mickey|bbl> you can always watch for name owner changes to get the dbus connection "repaired" on demand
Oct 18 15:02:26 <mickey|bbl> most bindings make this very simple
Oct 18 15:02:58 <mickey|bbl> you need to watch out for the resource system though, e..g when the frameworkd reappears you need to reclaim resources before continuing to use them
Oct 18 15:03:05 <mickey|bbl> (RequestResource("foo"))
Oct 18 15:07:55 <DocScrutinizer> obviously. There might have been state changes inside the restarted dbus service, trigered by previous messages. These will get lost when service is restarted, so association to the resource/service has to be handled/renewed somehow
Oct 18 15:21:45 <mickey|bbl> yep
Oct 18 15:21:55 <mickey|bbl> all dbus servers and clients should be restartable though
Oct 18 15:22:09 <mickey|bbl> should be the goal, at least
Oct 18 15:33:49 <lindi-> mickey|bbl: any idea if zhone does it currently?
Oct 18 15:37:27 <mickey|bbl> pretty sure it does not
Oct 18 16:48:21 <sicu> Ainulindale, quickdev: i keep bumping into errors when trying to access Messages and Contacts
Oct 18 16:48:41 <quickdev> sicu, which errors?
Oct 18 16:49:36 <sicu> sometimes it works, but a lot of the time it just crashes
Oct 18 16:50:10 <sicu_> quickdev: http://pastebin.com/m66ffdd3f
Oct 18 16:51:33 <quickdev> sicu, that's a problem with ophonekitd, because ophonekitd enables the GSM resource and the other apps rely on the GSM resource...to get sim informations for example
Oct 18 16:51:45 <quickdev> sicu, cat /tmp/ophonekitd.log
Oct 18 16:54:49 <sicu_> quickdev: http://pastebin.com/d6decc950
Oct 18 16:56:45 <quickdev> sicu, http://shr.bearstech.com/trac/ticket/40
Oct 18 16:56:49 <quickdev> thanks for bug reporting
Oct 18 16:56:52 <quickdev> just added it
Oct 18 16:57:05 <quickdev> it's because your received an incoming message
Oct 18 16:57:11 <quickdev> restart and it should work again
Oct 18 16:57:20 <quickdev> or: /etc/X11/Xsession.d/80ophonekitd
Oct 18 16:57:50 <quickdev> kill ophonekitd before
Oct 18 16:57:54 <quickdev> if it's still running
Oct 18 16:58:09 <sicu> quickdev: yeh, it does stay like that until i restart ... or do something about it
Oct 18 16:58:27 <sicu> ok, will do ... thanks
Oct 18 17:05:02 <quickdev> mickey|dinner, that setup is great :)
Oct 18 17:05:49 <sicu> quickdev: snappier development now i take it =]
Oct 18 17:06:09 <quickdev> yeah, it's _much_ better now
Oct 18 17:06:26 <quickdev> but the next week I'm in munich, so no development then ;)
Oct 18 17:06:45 <Hire> :\
Oct 18 17:06:57 * Hire now is sad
Oct 18 17:07:20 <sicu> :/
Oct 18 17:07:49 <quickdev> but I'll try to fix some bugs tomorrow ;)
Oct 18 17:08:06 * Hire is less sad 
Oct 18 17:09:20 <sicu> btw, i did /etc/X11/Xsession.d/80ophonekitd but still get error when trying to access either contacts or messages :/
Oct 18 17:09:54 <sicu> how do i kill ophonekitd? it's not listed when i do top
Oct 18 17:12:51 <sicu> i should get more used to using linux
Oct 18 17:15:26 <AJuOnLiNE> does emulation in qemu work ? i followed the instructions for ubuntu/makefile isee a very old qtopi build running.
Oct 18 17:15:31 <AJuOnLiNE> any way i can run the latest ones?
Oct 18 17:16:59 <AJuOnLiNE> Sup3rkiddo: how you tried running openmoko on qemu?
Oct 18 17:18:03 <AJuOnLiNE> Sup3rkiddo: i mean. how did you develop beofre when you didnt have the FR :P
Oct 18 17:18:47 <Sup3rkiddo> AJuOnLiNE, i used the MokoMakefile to build images.....it has targets for running qemu images as well, but not sure it if people use it now
Oct 18 17:20:01 <AJuOnLiNE> Sup3rkiddo: i have the qtopia build running. shows 4 icons. i suppose you saw the screenshots earlier. when i click. nothing works.
Oct 18 17:20:23 <Sup3rkiddo> AJuOnLiNE, yeah i did, i suppose thats way old...
Oct 18 17:20:57 <Sup3rkiddo> AJuOnLiNE, but i am pretty sure, many test their code on their desktops or laptop first unless if its really low level stuff
Oct 18 17:21:16 <Sup3rkiddo> (atleast that makes debugging easy for me
Oct 18 17:23:00 <AJuOnLiNE> Sup3rkiddo: ok. shall try that way. once i freeze on the language to program in. i am in research phase at the moment.. but certainly i need to speed up things now :)
Oct 18 17:23:28 <Kensan> Sup3rkiddo: mickey|dinner was looking for you
Oct 18 17:23:56 <Sup3rkiddo> AJuOnLiNE, right... then write a recipe and get it built using the toolchain
Oct 18 17:24:05 <Sup3rkiddo> mickey|dinner, ping
Oct 18 17:24:26 <Sup3rkiddo> Kensan, thanks...
Oct 18 17:25:30 <Sup3rkiddo> AJuOnLiNE, the bitbake recipe that is ;), then its a matter of scp'ing into the fr and opkg install
Oct 18 17:26:31 <AJuOnLiNE> yes i get that part :)
Oct 18 17:26:58 <Sup3rkiddo> hehe..... ergo eatito, ergo sum
Oct 18 17:27:25 <quickdev> sicu, ps aux | grep ophonekitd
Oct 18 17:27:45 <quickdev> then kill -9 $PID....replace $PID with the process number(first column of ps aux I think)
Oct 18 17:28:27 <AJuOnLiNE> Sup3rkiddo: english/hindi :P
Oct 18 17:33:06 <Sup3rkiddo> AJuOnLiNE, latin, i think :D
Oct 18 18:22:12 <AJuOnLiNE> is there a way i can change the boot image that the qemu build loads? and try the newer images?
Oct 18 18:27:15 <Hire> argh
Oct 18 18:27:24 <Hire> i want raster on my facebook|
Oct 18 18:29:28 <quickdev> Ainulindale?
Oct 18 19:36:51 <nikolajsheller> where do I report errors in the toolchain?
Oct 18 20:53:08 <nikolajsheller> where can I browse the code for om2008.x?
Oct 18 20:57:11 <wurp2|working> nikolajsheller: Which part?
Oct 18 20:57:34 <wurp2|working> You have to dig around in the bitbake recipes for where the various bits are checked out from
Oct 18 20:57:40 <wurp2|working> or downloaded from, or whatever
Oct 18 20:58:56 <nikolajsheller> well, Mainly I was interested in experimenting with qt4
Oct 18 20:59:46 <nikolajsheller> I got the toolchain to compile an example gtk app, and was able to run it.
Oct 18 21:00:23 <nikolajsheller> So now I wanted to look at a sample Qt app for openmoko
Oct 18 21:01:28 <nikolajsheller> additionaly I think I found a small bug in the toolchain
Oct 18 21:01:44 <nikolajsheller> where do I report this?
Oct 18 22:00:55 <wurp2|working> I'm sorry, I don't know.  I would look at the history on the toolchain page on the wiki to see who created the page
Oct 18 22:01:04 <wurp2|working> Or post your bug to the Talk panel
Oct 18 22:01:17 <wurp2|working> I also don't know where to find the Qt code :-)
Oct 18 22:04:03 <quickdev> Hey
Oct 18 22:19:52 <nikolajsheller> wurp2|working: ok thanks for you reply. I found some code on the qtextended.org page
Oct 18 22:20:39 <nikolajsheller> I am a bit mystified though that it is not straightforward to find the openmoko 2008.x code....
Oct 18 22:27:56 <Kensan> nikolajsheller: git.openmoko.org has most of it except what's actually upstream
Oct 18 22:29:39 <nikolajsheller> Kensan: ok. Thanks
Oct 18 22:30:11 <nikolajsheller> Is getting a project to compile using Qt described anywhere?
Oct 18 22:38:46 <Kensan> nikolajsheller: hm I don't know much about qt... Maybe you'll find something in the wiki or at the qtextended website.
Oct 18 22:40:05 <nikolajsheller> I assume that qt is the way to for for om2008.x ?
Oct 18 22:40:48 <Kensan> nikolajsheller: om2008.x is basically putting the qt apps ontop of X/FSO etc.
Oct 18 22:42:01 <nikolajsheller> Kensan: I'm not sure I follow. Can you elaborate?
Oct 18 22:42:40 <Ainulindale> quickdev ?
Oct 18 22:42:48 <quickdev> Ainulindale !
Oct 18 22:42:50 <Kensan> nikolajsheller: http://wiki.openmoko.org/wiki/Distributions#Om_2008.8_.28ASU.2C_April.2FAugust_Software_Update.29
Oct 18 22:42:59 <Ainulindale> quickdev: What were you willing to tell me ?
Oct 18 22:43:10 <Kensan> nikolajsheller: Om 2008.8 has been started to integrate the Qtopia stack - on X11 - with a new set of graphically pleasing applications based on the Enlightenment Foundation Libraries.
Oct 18 22:43:14 <quickdev> I don't know anymore
Oct 18 22:44:15 <nikolajsheller> Kensan: I interpret that as meaning future applications should use Qt. Is that correct?
Oct 18 22:44:19 <quickdev> Ainulindale, do you have some minutes?
Oct 18 22:44:54 <Kensan> nikolajsheller: well it is X using enlightenment libs.
Oct 18 22:45:15 <Ainulindale> Quickly then I'm not alone
Oct 18 22:45:39 <nikolajsheller> Kensan: I'm not sure I follow the implications of that.
Oct 18 22:45:57 <Kensan> nikolajsheller: I assume you are playing with the idea to write an application for openmoko, correct?
Oct 18 22:46:05 <quickdev> it isn't urgent, let's talk about it at a different time....I won't be online from monday to friday, did I tell you this?
Oct 18 22:46:32 <Ainulindale> Yep
Oct 18 22:46:38 <Kensan> nikolajsheller: the whole diistribution situation is rather messy I am afraid :(
Oct 18 22:46:40 <Ainulindale> And I won't be here tomorrow morning
Oct 18 22:46:41 <nikolajsheller> Kensan: yes, I recently got the sample application to compile and run, for GTK
Oct 18 22:46:42 <quickdev> btw..I got my setup working
Oct 18 22:46:46 <Ainulindale> And probably not the whole next week too
Oct 18 22:47:03 <quickdev> ok
Oct 18 22:47:09 <Kensan> nikolajsheller: so there are many possibilities to write applications: qt, gtk, enlightenment etc
Oct 18 22:47:30 <nikolajsheller> Kensan: And my understaning is that Qt is the way openmoko is heading
Oct 18 22:48:13 <Kensan> nikolajsheller: it is safe to assume that as a "telephony backend" FSO will be used (see freesmartphone.org). It's a dbus interface which lets your app talk to the device and certain system-level services.
Oct 18 22:48:47 <Kensan> nikolajsheller: well openmoko does not have a clear direction imho.
Oct 18 22:48:57 <nikolajsheller> Kensan: which toolkit do you think would be the the preferable one?
Oct 18 22:49:42 <Kensan> nikolajsheller: there's a push from the community to "salvage" the gtk applications from 2007.? and port them to use FSO as a backend.
Oct 18 22:50:04 <quickdev> Kensan, that goal isn't followed anymore
Oct 18 22:50:09 <Kensan> nikolajsheller: rasterman is developing an easy to use widget toolkit called elementary based on the enlightenment libs
Oct 18 22:50:10 <quickdev> afaik
Oct 18 22:50:24 <Kensan> quickdev: oh ok. please correct me where I am wrong.
Oct 18 22:50:33 <Kensan> quickdev: which way is SHR going then?
Oct 18 22:50:35 <quickdev> a toolkit called elementary? never heard of..
Oct 18 22:50:57 <Kensan> quickdev: have you seen rasters illume-image with the alarm clock? that's done with elementary
Oct 18 22:51:10 <quickdev> The Stable Hybrid Release (SHR) is intended to be a community driven distribution composed of the FSO and some basic applications, that can be configured to use several diffrent graphical toolkits, for example GTK or EFL. SHR is based on the FSO build. 
Oct 18 22:52:02 <Kensan> quickdev: http://www.rasterman.com/files/illume-buttons.png
Oct 18 22:52:06 <nikolajsheller> quickdev: ELT ?= Enlighment?
Oct 18 22:52:21 <nikolajsheller> nikolajsheller: quickdev: ELF ?= Enlighment?
Oct 18 22:52:37 <quickdev> EFL = Enlightenment foundation libraries
Oct 18 22:52:41 <Kensan> nikolajsheller: yes for all practical purposes ELF = Enlightenment
Oct 18 22:52:51 <Kensan> nikolajsheller: you can ask lpotter about QT :)
Oct 18 22:53:31 <quickdev> "elementary" toolkit..ah yeah, E has a builtin toolkit, but that should not be used for the apps afaik
Oct 18 22:53:56 <nikolajsheller> Kensan: I would perfer to use whichever toolkit will be supported and backed for openmoko
Oct 18 22:54:06 <Kensan> quickdev: elementary is specifically written by raster for phones/embedded devices etc. He's been contracted by Swisscom to do this.
Oct 18 22:54:26 <Kensan> nikolajsheller: the backend is FSO that's rather fixed.
Oct 18 22:54:50 <nikolajsheller> Kensan: But the UI is up for grbs?
Oct 18 22:54:55 <nikolajsheller> grabs
Oct 18 22:56:05 <Kensan> nikolajsheller: well quite frankly I don't know which way om2008.x is headed since OM is now focusing on stability etc. Mayb other people can give you better advice
Oct 18 22:56:12 <nikolajsheller> Kensan, quickdev: What is you impression of the community direction? What is being used?
Oct 18 22:56:29 <quickdev> Kensan, you're right...it's an own toolkit - haven't read that before :)
Oct 18 22:57:08 <Kensan> nikolajsheller: this is only from my own observation so this might be totaly wrong: lot of people use python and enlightenment
Oct 18 22:57:16 <quickdev> nikolajsheller, Ainulindale and I are trying to write some basic apps with the EFL and libframeworkd-glib as the base....I don't know what's the community direction is ;)
Oct 18 22:57:19 <Kensan> nikolajsheller: there's many existing gtk assp.
Oct 18 22:57:21 <nikolajsheller> quickdev: elementary is its own toolkit?
Oct 18 22:57:56 <Kensan> nikolajsheller: I don't know how to classify it :)
Oct 18 22:58:06 <Kensan> nikolajsheller: but it looks rather simple to use.
Oct 18 22:58:37 <Kensan> nikolajsheller: raster will write dialer, sms and some other apps in it once he has finished elementary so he will put out an image with all that + fso.
Oct 18 22:58:46 <quickdev> nikolajsheller, http://www.rasterman.com/
Oct 18 22:59:08 <Kensan> yeah, check out his latest blog post
Oct 18 22:59:38 <Kensan> nikolajsheller: you will find the code in the svn if you are interested
Oct 18 22:59:52 <Kensan> nikolajsheller: http://svn.enlightenment.org/svn/e/trunk/TMP/st/
Oct 18 23:00:31 <nikolajsheller> Kensan, quickdev: is libframeworkd-glib in the repository? 
Oct 18 23:00:47 <quickdev> git.freesmartphone.org
Oct 18 23:01:09 <Kensan> quickdev: what kind of apps are you guys writing?
Oct 18 23:01:41 <quickdev> messages, contacts, dialer, settings...stuff like that
Oct 18 23:02:00 <nikolajsheller> quickdev: PIM?
Oct 18 23:02:05 <Kensan> quickdev: cool! where do you guys keep the code? ;)
Oct 18 23:02:23 <quickdev> PIM will be integrated in frameworkd and we'll use it, yes
Oct 18 23:02:39 <nikolajsheller> quickdev: PIN synchronization?
Oct 18 23:02:44 <quickdev> Kensan, atm it's still in Ainulindale's svn, svn://daria.forty-two.fr/shr/trunk/
Oct 18 23:02:48 <nikolajsheller> PIM synchonization?
Oct 18 23:02:52 <quickdev> later, yes
Oct 18 23:04:47 <nikolajsheller> quickdev: that was basically what I wanted to start working on. Has bothered me for a couple of years now that there i no decent way of synching my Nokia phones with Evolution
Oct 18 23:05:15 <nikolajsheller> quickdev: Would by nice if openmoko could
Oct 18 23:05:31 <Kensan> nikolajsheller: maybe you can join quickdev and Ainulindale :)
Oct 18 23:05:31 <quickdev> nikolajsheller, join us..
Oct 18 23:07:12 <nikolajsheller> but I am a bit mystified why no decent usabilitiy platform is in place... diversity is good, but there are some basics that need to be in place for the phone to be usable...
Oct 18 23:08:22 <quickdev> there are no real basics..
Oct 18 23:08:40 <Kensan> nikolajsheller: yes, it's all very painfull and OM has just started to realise that...
Oct 18 23:08:55 <nikolajsheller> quickdev: is your work in any of the distributions?
Oct 18 23:09:02 <quickdev> nikolajsheller, SHR
Oct 18 23:09:52 <nikolajsheller> How many developers are actually working on the openmoko?
Oct 18 23:10:36 <quickdev> mickey|dinner, please have a look: http://rafb.net/p/0RVs0x53.html
Oct 18 23:10:58 <quickdev> nikolajsheller, unknown...openmoko pays some...and there are very few community devs
Oct 18 23:12:45 <nikolajsheller> quickdev, kensan: You are very active on this channel. Are you hobbyists or do you work professionaly with this?
Oct 18 23:12:51 <Hire> nikolajsheller: join us... we had cookie
Oct 18 23:13:05 <quickdev> nikolajsheller, hobby
Oct 18 23:13:07 <Hire> cookies
Oct 18 23:13:35 <Kensan> nikolajsheller: I follow this project in my spare time
Oct 18 23:13:45 <Kensan> nikolajsheller: I've been sick all week so I had some time on my hands :)
Oct 18 23:13:48 <nikolajsheller> Hire: :)
Oct 18 23:14:08 <Hire> seriously
Oct 18 23:14:18 <Kensan> nikolajsheller: but don't be fooled, there are people getting paid to develop for the Openmoko plattform.
Oct 18 23:14:30 <Hire> if you want to contribute into shr... perfect :)
Oct 18 23:14:32 <Hire> yes
Oct 18 23:14:51 <Hire> but openmoko inc pays devs for fso
Oct 18 23:14:54 <Kensan> nikolajsheller: what's your interest in openmoko if I might ask?
Oct 18 23:14:56 <Hire> kernel hacker
Oct 18 23:14:59 <Hire> x hacker
Oct 18 23:15:18 <nikolajsheller> The project is really a great idea, and I want to contribute. But I would like some kind of idea where it would do the most benefit
Oct 18 23:15:21 <Hire> and current, sadly, stack ( qtopia + efl )
Oct 18 23:15:30 <Kensan> Hire: there's also other companies than OM working with the plattform
Oct 18 23:15:38 <Hire> as swisscom :)
Oct 18 23:16:12 <Kensan> nikolajsheller: well writing userspace apps such as quickdev and Ainulindale is a great way to contribute since that's what's currently missing.
Oct 18 23:16:15 <Kensan> imho
Oct 18 23:16:15 <Kensan> Hire: yes :)
Oct 18 23:16:20 <Hire> however, nikolajsheller. the future of openmoko is FSO + paroli + tichy
Oct 18 23:16:22 <Hire> but
Oct 18 23:16:23 <Kensan> Hire: but also others I believe
Oct 18 23:16:23 <quickdev> nikolajsheller, because om develops the base (frameworkd)...we can concencrate on the ui...and that's in my eyes the most beneficial work atm
Oct 18 23:16:28 <nikolajsheller> Kensan: I just received my FR. And I would like to learn a bit about embedded linux development
Oct 18 23:16:40 <quickdev> Hire, no really ;)
Oct 18 23:16:49 <Hire> why not?
Oct 18 23:17:20 <Kensan> nikolajsheller: great, for that its a good device. Just don't get scared off by the community. It can be very negative at times. but there's also great people doing cool things.
Oct 18 23:18:14 <nikolajsheller> And having been fed up with closed phones for too many years, I think this is an area that must open up
Oct 18 23:18:22 <Kensan> jebus, I am just stumbling into one issue after the other without actually being able to do what I wanted to *grrr*
Oct 18 23:19:15 <quickdev> Hire, I think that paroli / tichy could do it on the short term...but they'll be replaced in the long term ;)
Oct 18 23:20:17 <nikolajsheller> quickdev: paroli / tichy? are they developers?
Oct 18 23:20:24 <Hire> uhm, probably yes. i think it will be replaced with a frontend based on opengl es when them release the GTA04
Oct 18 23:20:27 <Hire> nono
Oct 18 23:21:39 <nikolajsheller> nikolajsheller: heh :) UI frameworks?
Oct 18 23:21:47 <Hire> yes
Oct 18 23:21:55 <quickdev> python plugin manager..
Oct 18 23:22:04 <Hire> tichy
Oct 18 23:22:09 <Kensan> Hire: GTA04? *shakes head*
Oct 18 23:22:28 <Hire> paroli is a suite of apps as contacts, messages and dialer
Oct 18 23:23:02 <Kensan> why "dream" of hardware that might or might not come in one or two years when there's the FR that has so much potential?
Oct 18 23:23:16 <Kensan> just my 2 cents
Oct 18 23:23:46 <nikolajsheller> FR has a lot of potential.
Oct 18 23:24:30 <Hire> you are right kensan
Oct 18 23:24:53 <Hire> so I think we need to give more love to fso and shr :)
Oct 18 23:27:18 <Kensan> *grrr* I am getting nowhere today/night... rather frustrating day :(
Oct 18 23:27:29 <Hire> nikolajsheller, join shr :o
Oct 18 23:27:56 <Hire> more hackers + more code + more bugs + more fun
Oct 18 23:28:08 <nikolajsheller> Hire: does SHR build on FSO and...?
Oct 18 23:28:11 <Ainulindale> nikolajsheller: wait a minute for more information on SHR :-)
Oct 18 23:28:26 <Hire> and a suite of apps :)
Oct 18 23:28:33 <nikolajsheller> eek. Too many TLA's
Oct 18 23:28:34 <Hire> as dialer3, contacts3 and messages3
Oct 18 23:28:37 <Kensan> somebody should explain nikolajsheller how it all comes together :)
Oct 18 23:28:52 <Hire> talk it with Ainulindale
Oct 18 23:28:53 <Kensan> because
Oct 18 23:28:55 * Kensan has no clue
Oct 18 23:29:08 * Hire has no money
Oct 18 23:29:43 <Ainulindale> Well then nikolajsheller 
Oct 18 23:29:48 <Ainulindale> SHR is built upon FSO as basis
Oct 18 23:29:52 <nikolajsheller> :)
Oct 18 23:30:01 <Ainulindale> We first inteded to port old 2007.2 apps on a frameworkd basis
Oct 18 23:30:07 <Ainulindale> Hence I first tried to developed a wrapper
Oct 18 23:30:14 <Ainulindale> it was a no brainer as it's really different
Oct 18 23:30:21 <Ainulindale> That's why I came with libframeworkd-glib/ophonekitd
Oct 18 23:30:36 <Ainulindale> First being a C library to call frameworkd dbus methods
Oct 18 23:30:43 <Ainulindale> (thus not bothering about dbus in the underlying apps)
Oct 18 23:30:52 <Ainulindale> Second being a daemon to do the normal telephony things
Oct 18 23:31:00 <Ainulindale> (wait for calls and display a window, etc)
Oct 18 23:31:17 <nikolajsheller> ok...
Oct 18 23:31:25 <Ainulindale> As of now, you can use any kind of library implementing the libframeworkd-phonegui headers for ophonekitd
Oct 18 23:31:42 <Ainulindale> As a matter of fact there's no developer of the GTK library still in activity
Oct 18 23:32:02 <Ainulindale> And I'm myself allergic to any kind of UIs
Oct 18 23:32:11 <nikolajsheller> :)
Oct 18 23:32:23 <Ainulindale> But I'd enjoy any help on that
Oct 18 23:32:41 <Ainulindale> I can even offer some help if forced to under threat with the GTK stuff
Oct 18 23:33:02 <nikolajsheller> hehe :)
Oct 18 23:33:16 <Ainulindale> There's also ton of stuff we want to do
Oct 18 23:33:22 <Ainulindale> libframeworkd-phonegui-qt is one of them
Oct 18 23:33:27 <Ainulindale> Some modules for illume
Oct 18 23:33:34 <Ainulindale> Some basic packing of the image
Oct 18 23:33:45 <Ainulindale> (having a nice set of stable packages to include)
Oct 18 23:33:47 <Ainulindale> All in all
Oct 18 23:33:52 <Ainulindale> We don't intend to redevelop everything
Oct 18 23:33:58 <Ainulindale> The main idea was to use stable stuff
Oct 18 23:34:01 * Kensan wonders what Ainulindale is waiting for...
Oct 18 23:34:04 <Ainulindale> As a matter of fact there's no UI for frameworkd
Oct 18 23:34:11 <Ainulindale> Kensan: in terms of what ?
Oct 18 23:34:15 <Kensan> hehe
Oct 18 23:34:41 <Kensan> becuase you are listing all the things you want to do but still you have time to chat on irc *hehe*
Oct 18 23:34:48 <Kensan> silly joke
Oct 18 23:35:10 <Ainulindale> I'll accept any kind of comments of this explicit kind when you'll have commited as much as me on SHR :-p
Oct 18 23:35:23 * Kensan failed to be funny and shall not disrupt nikolajshellers audience
Oct 18 23:35:26 <Ainulindale> Meanwhile, subscribe, get to work, and move your ass :->
Oct 18 23:35:29 <nikolajsheller> Is there anywhere where all this is summarized? 
Oct 18 23:35:37 <Ainulindale> nikolajsheller: We started to do that on trac
Oct 18 23:35:41 <Ainulindale> shr.bearstech.com/trac
Oct 18 23:35:54 <Ainulindale> Feel free to put there anything you want in a community oriented distribution
Oct 18 23:36:00 <Ainulindale> Because *this* is the main goal of SHR
Oct 18 23:36:04 <Ainulindale> Being a community driven distro
Oct 18 23:36:21 <Ainulindale> Let OM focus on basic stuff, and let ourselves build what we want in the end
Oct 18 23:36:53 <Ainulindale> It was wurp2|working's idea, and it suits me as I don't like not being able to choose
Oct 18 23:37:07 <Ainulindale> And we will be able to choose as soon as we will have a basic and stable set of functionnalities
Oct 18 23:37:46 <nikolajsheller> ok. How many people are working on this?
Oct 18 23:37:57 <Ainulindale> I truly don't know
Oct 18 23:38:07 <Ainulindale> There was many people since July
Oct 18 23:38:14 <Ainulindale> I don't know who is still active
Oct 18 23:38:33 <Ainulindale> I can name three of them as of now, quickdev, wurp2|working (who doesn't have time to develop) and me
Oct 18 23:38:44 <Ainulindale> There's a bunch of other people eager to test and give ideas
Oct 18 23:38:52 <Ainulindale> And people I don't have any news of
Oct 18 23:39:09 <Ainulindale> Problem is, and will be :
Oct 18 23:39:28 <Ainulindale> We need developers, ready to develop quickly a simple and working stuff to put together a working and usable distro
Oct 18 23:39:34 <Ainulindale> To then, focus on delivering a complete thing
Oct 18 23:39:37 <Ainulindale> First basic, then complete
Oct 18 23:40:31 <nikolajsheller> is there a wishlist/needed features on the trac?
Oct 18 23:40:45 <nikolajsheller> 12 tickets...
Oct 18 23:40:50 <Ainulindale> 12 ? No
Oct 18 23:41:00 <Ainulindale> We're at the #40 right now
Oct 18 23:41:10 <nikolajsheller> ok
Oct 18 23:41:10 <Ainulindale> It may be that you don't have the rights to see them and that would be my fault
Oct 18 23:41:21 <Hire> nono
Oct 18 23:41:43 <Ainulindale> nikolajsheller: My first wish as a user is a sound profile module for illume as a shelf gadget
Oct 18 23:41:58 <Ainulindale> i.e. silent/default
Oct 18 23:42:05 <Hire> http://shr.bearstech.com/trac/report/1
Oct 18 23:42:10 <Hire> 23 tickets open
Oct 18 23:42:38 <Kensan> Ainulindale: I believe once you guys have something to show you will attract more devs etc
Oct 18 23:42:38 <Ainulindale> nikolajsheller: my second wish is PIM but that is tied to frameworkd integration
Oct 18 23:42:46 <nikolajsheller> Wow, I have some reading up do do on all this information
Oct 18 23:42:46 <Ainulindale> Kensan: there's already something to show
Oct 18 23:42:56 <Ainulindale> But I feel totally unable to release until this is stable
Oct 18 23:43:06 <Kensan> Ainulindale: show in the sense, that there's some buzz in the community about your project.
Oct 18 23:43:08 <Ainulindale> I won't do the same mistake Om did on AU
Oct 18 23:43:10 <Ainulindale> +S
Oct 18 23:43:29 <nikolajsheller> Ainulindale: which was?
Oct 18 23:43:37 <Ainulindale> Kensan: If there's some feedback and fixed tickets I agree
Oct 18 23:43:38 <Ainulindale> But not for now
Oct 18 23:43:46 <Ainulindale> nikolajsheller: they released to soon, IMHO
Oct 18 23:43:49 <Ainulindale> +o
Oct 18 23:43:49 <Hire> them release asu too early :)
Oct 18 23:44:24 <Kensan> ASU was a total failure imho because it had all the concepts wrong.
Oct 18 23:44:31 <Ainulindale> I can't agree
Oct 18 23:44:39 <Ainulindale> Concepts can't be wrong
Oct 18 23:44:43 <Ainulindale> They are not yours
Oct 18 23:44:46 <Ainulindale> (Or mine for that matter)
Oct 18 23:44:50 <Ainulindale> But they can't be wrong
Oct 18 23:44:54 <Kensan> ?
Oct 18 23:45:00 <Kensan> I think we are miscommunicating here
Oct 18 23:45:04 <Ainulindale> It may be
Oct 18 23:45:08 <Ainulindale> I'm french, tired
Oct 18 23:45:13 <Ainulindale> I drank some wine earlier
Oct 18 23:45:13 <Kensan> I am swiss, tired
Oct 18 23:45:15 <Kensan> :)
Oct 18 23:45:17 <Kensan> me too
Oct 18 23:45:19 <Kensan> lol
Oct 18 23:45:23 <Ainulindale> And I have a nice bottle of beer next to me :-p
Oct 18 23:45:43 <Ainulindale> Kensan: well, what were you saying then ?
Oct 18 23:45:45 <quickdev> I assume german beer ;)
Oct 18 23:46:07 <Kensan> I am a bit agravated since I wanted to get something done instead of bumping into other things that need to get fixed first
Oct 18 23:46:21 <nikolajsheller> I had a danish beer earlier
Oct 18 23:47:06 <Kensan> The move to do asu, was imho wrong because during that time they should have done what they announced in the latest weekly engineering news.
Oct 18 23:47:16 <Kensan> stability, get the core features rock solid
Oct 18 23:47:53 <Ainulindale> quickdev: german beer is crap
Oct 18 23:47:58 <Ainulindale> quickdev: belgian is the way to go
Oct 18 23:48:29 <quickdev> Ainulindale, look out what you're saying :)
Oct 18 23:48:41 <Kensan> quickdev: where are you from?
Oct 18 23:48:46 <quickdev> germany :)
Oct 18 23:48:49 <quickdev> nrw
Oct 18 23:48:50 <Ainulindale> I would be happy to drink any quality beer from germany
Oct 18 23:48:55 <Ainulindale> Or switzerland for that matter
Oct 18 23:48:55 <Kensan> quickdev: ah ic :)
Oct 18 23:49:00 <Ainulindale> (samischlaus is just awful)
Oct 18 23:49:10 <Kensan> Ainulindale: hahaha Samichlaus
Oct 18 23:49:15 <nikolajsheller> mmmm hoegaarden
Oct 18 23:49:24 <Kensan> Ainulindale: Samichlaus means Santa Clause ;)
Oct 18 23:49:41 <Kensan> Ainulindale: we have some good local brew but were not famous for the beer
Oct 18 23:50:59 <Ainulindale> quickdev: http://www.calaquendi.org/om/etiquettes.jpg
Oct 18 23:51:12 <Ainulindale> Kensan: I know it means santa claus
Oct 18 23:51:24 <Ainulindale> it's also a triple fermentation beer from switzerland
Oct 18 23:51:38 <Ainulindale> Which has roughly the taste of tar
Oct 18 23:51:51 <raster> ewww
Oct 18 23:52:30 <Ainulindale> http://brouette-de-bieres.com/beers/kind/BRU-Brown/EGGSAMBIE-Samichlaus-Bier
Oct 18 23:52:36 <Kensan> raster: oh hey there
Oct 18 23:52:45 <Kensan> raster: I am pulling my hair out here, I am getting nowhere.
Oct 18 23:52:59 * raster hands Kensan scissors instead
Oct 18 23:53:05 <raster> they work better than pulling
Oct 18 23:53:09 <raster> :)
Oct 18 23:53:17 <Kensan> raster: latest oe kernel is borked
Oct 18 23:53:41 <Kensan> raster: bootchart can't write to /etc/...
Oct 18 23:53:48 <raster> oooh
Oct 18 23:53:55 <Kensan> raster: building and flashing for the =)(??`? time
Oct 18 23:54:05 <nikolajsheller> quickdev, Kensan, Ainulindale, Hire: Thank you for you input. I will do some reading and get back to you with a lot of quiestions :)
Oct 18 23:54:15 <raster> ahhahahaha
Oct 18 23:54:17 <Ainulindale> nikolajsheller: I'm really interested in any help you can give
Oct 18 23:54:29 <raster> i've done that often enough
Oct 18 23:54:31 <Ainulindale> quickdev: by the way
Oct 18 23:54:34 <raster> do i understand :)
Oct 18 23:54:36 <raster> pain
Oct 18 23:54:37 <Ainulindale> would you be able to fix some issues by tomorrow ?
Oct 18 23:54:38 <Kensan> nikolajsheller: stick to quickdev, Ainulindale and Hire because they know what they talk about - unlike me *heh*
Oct 18 23:55:05 <raster> i'm not coding today
Oct 18 23:55:07 <Ainulindale> Kensan: one way to settle that is to come and give us help :-p
Oct 18 23:55:10 <raster> today... artwork day
Oct 18 23:55:11 <quickdev> Ainulindale, atm I'm fixing some things...
Oct 18 23:55:15 <quickdev> messages app
Oct 18 23:55:22 <quickdev> tomorrow I won't have much time
Oct 18 23:55:24 <quickdev> almost no time
Oct 18 23:55:29 <Kensan> raster: damned I am rather disapointed. wanted to get mdev in but I have acomplished nothing *grrr*
Oct 18 23:55:39 <Kensan> Ainulindale: I have my own couple projects but thanks for the offer :)
Oct 18 23:55:41 <nikolajsheller> hehe :) I'm sure you have valuable insights
Oct 18 23:55:43 <Ainulindale> quickdev: as long as the basic things are fixd :-)
Oct 18 23:55:53 <Ainulindale> quickdev: oh and
Oct 18 23:55:55 <Kensan> raster: artwork? Capoera dancing?
Oct 18 23:55:57 <Ainulindale> We should figure out this list calls problem
Oct 18 23:56:03 <Ainulindale> I didn't have a look recently
Oct 18 23:56:15 <quickdev> is there a problem?
Oct 18 23:56:20 <Ainulindale> Yep bad marshalling
Oct 18 23:56:25 <raster> rebuild-reflash-boot-rebuild-reflahs-boot is a slowway to do this
Oct 18 23:56:27 <raster> it sucks
Oct 18 23:56:34 <quickdev> ah, I remember
Oct 18 23:56:40 <raster> but for lower level things like kernel and boot stuff (speed) the only way
Oct 18 23:56:48 <raster> thats why i prefer to do all my dev in xephyr on x8
Oct 18 23:56:50 <raster> err
Oct 18 23:56:50 <raster> x86
Oct 18 23:56:57 <raster> it is sane
Oct 18 23:56:59 <raster> fast
Oct 18 23:57:09 <Kensan> raster: Somehow I am not able to catch all log messages... logread does not spew out everything.
Oct 18 23:57:18 <raster> imho that is the BIGGEST plus of a system like om
Oct 18 23:57:19 <raster> or oe
Oct 18 23:57:27 <raster> u CAN just install a few packages on your desktop
Oct 18 23:57:31 <raster> no need to emulate the whole os
Oct 18 23:57:36 <Kensan> raster: well I intermittently had the grande idea to build qemu... another waste of time :(
Oct 18 23:57:54 <raster> just small key things like dpi, x nested server and the wm env
Oct 18 23:58:02 <raster> the rest u can do withotu ever seeing a freerunner
Oct 18 23:58:03 <raster> :)
Oct 18 23:58:19 <Kensan> raster: ah that's why you have me do that stuff
Oct 18 23:58:23 <Ainulindale> quickdev: anyway fixing messages is the way to go now (that and having a way to look faster for a contact :-p)
Oct 18 23:58:24 <raster> hehehehe
Oct 18 23:58:28 <raster> of course!
Oct 18 23:58:34 <raster> delegate the pain to others!
Oct 18 23:58:35 <raster> :)
Oct 18 23:58:43 <Kensan> raster: touché
Oct 18 23:58:49 <Ainulindale> raster: could you tell me where I can find the sources for the illume gsm shelf gadget ? (FSO)
Oct 18 23:58:56 <raster> often u can even convince them that its fun
Oct 18 23:58:58 <raster> and they want to do it
Oct 18 23:59:02 <raster> and its "cool" to hack the kernel
Oct 18 23:59:03 <raster> :)
Oct 18 23:59:15 <raster> Ainulindale:  e_mod_gad_gsm.c
Oct 18 23:59:17 <raster> in illume
Oct 18 23:59:33 <raster> as for artowrk
Oct 18 23:59:34 <Ainulindale> Are modules shared objects ?
Oct 18 23:59:37 <raster> doing theme work for e17
Oct 18 23:59:43 <Kensan> raster: well it's fun if I see progress. Simply solving issues that pop up while trying to start working on the actual issue is not so much fun
Oct 18 23:59:46 <raster> doing a "about this theme" design
Oct 18 23:59:55 <raster> as the theme provides its own about box design and artowkr
Oct 19 00:00:09 <Kensan> raster: ah ic
Oct 19 00:00:28 <Kensan> raster: btw, if you are artworking, you could stomp out a uboot splash screeny
Oct 19 00:00:28 <raster> Ainulindale:  yes. all e moduels are shared objects. e finds them, dlopen()'s them and calls specific init/shutdown calls
Oct 19 00:00:36 <raster> the gsm gagdte is just part of illume
Oct 19 00:00:47 <Ainulindale> Ok then :-)
Oct 19 00:00:53 <Kensan> raster: well kernel is fun I agree :)
Oct 19 00:00:57 <Ainulindale> Now, find the sources of illume...
Oct 19 00:00:59 <raster> Kensan: ooh i stomped one out for myself...
Oct 19 00:01:08 <raster> already
Oct 19 00:01:09 <raster> :)
Oct 19 00:01:17 <Kensan> raster: especially when stuff makes it upstream :)
Oct 19 00:01:20 <Kensan> raster: where?
Oct 19 00:01:24 <raster> Ainulindale: in e svn.. like everything else on the planet :)
Oct 19 00:01:37 <Ainulindale> I know, I was just voicing out loud my search :-)
Oct 19 00:01:40 <Kensan> raster: don't tell me it's in svn
Oct 19 00:01:41 <raster> Kensan: ummm... just here atm
Oct 19 00:01:47 <Kensan> phew
Oct 19 00:01:50 <Kensan> hehe
Oct 19 00:01:54 <Kensan> because I hadn't looked there
Oct 19 00:01:56 <Kensan> heh
Oct 19 00:01:58 <raster> u'll have to modify uboot as its bigger than the default read size
Oct 19 00:02:03 <raster> but it does fit in the splash partition
Oct 19 00:02:15 <raster> u just need to tell uboot to read the whole splash part
Oct 19 00:02:18 <raster> not some of it
Oct 19 00:02:37 <Kensan> let me get my fr back on track first.
Oct 19 00:02:49 <raster> sure
Oct 19 00:02:57 <quickdev> Ainulindale, you'd like to code an illume plugin? what for?
Oct 19 00:03:00 <Kensan> then I'll fiddle around with uboot again
Oct 19 00:03:04 <Ainulindale> Heh, too bad, the GSM gadget doesn't use lfg
Oct 19 00:03:09 <Ainulindale> quickdev: sound profiles
Oct 19 00:03:32 <raster> Kensan:  http://www.rasterman.com/files/sp.gz
Oct 19 00:03:35 <raster> :)
Oct 19 00:03:42 <quickdev> Ainulindale, we have to think about where to integrate "Missed calls" for example...
Oct 19 00:03:44 <Kensan> heh
Oct 19 00:03:45 <raster> u'll have to flash on manually yourself
Oct 19 00:03:49 <Ainulindale> quickdev: we won't
Oct 19 00:03:52 <Kensan> right
Oct 19 00:03:56 <Ainulindale> quickdev: that has to be done in opimd
Oct 19 00:04:01 <raster> i was wanting to make a package actually that does it from userspace
Oct 19 00:04:13 <raster> either that or add it to the gta02 flash
Oct 19 00:04:20 <raster> but i'd ALSO need to write something to moify the uboot env
Oct 19 00:04:34 <raster> and decided that was going to take a bit more than a 5 minute hack to make sure its safe
Oct 19 00:04:56 <raster> Ainulindale:  lfg?
Oct 19 00:04:58 <Kensan> raster: yeah true
Oct 19 00:05:04 <Ainulindale> raster: libframeworkd-glib
Oct 19 00:07:03 <Hire> *yawn*
Oct 19 00:07:13 <Hire> 2.07am here
Oct 19 00:07:28 <Ainulindale> Same here
Oct 19 00:08:08 <Hire> it's the bed time
Oct 19 00:08:15 <quickdev> yeah, for me too
Oct 19 00:08:21 <quickdev> in a few minutes :)
Oct 19 00:10:57 <raster> Ainulindale: not possible. glib mainloop conflicts with ecore mainloop
Oct 19 00:11:09 <raster> u cant mix mainloops (easily)
Oct 19 00:11:12 <raster> u can
Oct 19 00:11:17 <raster> but u'd need a separate glib thread
Oct 19 00:11:29 <raster> and the pipes to pass messages back/forth from ecore to glib loops
Oct 19 00:11:37 <raster> just do raw dbus
Oct 19 00:12:17 <Hire> the bed calls me
Oct 19 00:12:21 <quickdev> raster, that's exactly what I'm doing
Oct 19 00:12:36 <raster> raw dbus isn't too bad
Oct 19 00:12:37 <Hire> so if I going to read some lovecraft's novell :)
Oct 19 00:12:39 <raster> just footwork
Oct 19 00:12:50 <raster> if u want u can wrap it in a convenience lib
Oct 19 00:12:54 <raster> for framework
Oct 19 00:13:05 <Hire> good night and good hacking :)
Oct 19 00:13:12 <raster> but i'm pretty rabid on keeping everything async
Oct 19 00:13:17 <raster> i HATE calls that do stuf flike
Oct 19 00:13:34 <raster> value = dbus_something_get_value();
Oct 19 00:13:44 <raster> when i know that call need to set up a dbus request
Oct 19 00:13:47 <raster> send it of
Oct 19 00:13:51 <raster> then sit and block waiting for a reply
Oct 19 00:13:57 <raster> then when ti gets the reply, return
Oct 19 00:14:12 <raster> i despise having stuff like that in my code
Oct 19 00:14:29 <raster> so it always loosk complex with setting up marshalling callbacks to handle the reply async in a callback
Oct 19 00:15:04 <Ainulindale> raster: lfg is async
Oct 19 00:15:18 <Ainulindale> And lfg doesn't do dbus connection
Oct 19 00:15:26 <Ainulindale> It is explicitely done by the developer
Oct 19 00:16:13 <raster> hmm
Oct 19 00:16:18 <raster> then why is glib involved?
Oct 19 00:16:24 <raster> if all it does is sit on dbus?
Oct 19 00:16:36 <raster> (unless thats for datatypes) ?
Oct 19 00:17:29 <Kensan> since when is hw-ecc active in moko kernels?
Oct 19 00:18:47 <Ainulindale> raster: because I used dbus-glib :-)
Oct 19 00:20:59 <raster> Ainulindale: :)
Oct 19 00:21:09 <raster> well sorry to make life hard but not using glib over here
Oct 19 00:21:34 <Ainulindale> Though now that I see there's e methods it would be interesting to do some stuff to be able to use them in lfg
Oct 19 00:21:34 <quickdev> yeah, qt people would bother
Oct 19 00:21:55 <Ainulindale> quickdev: then we should manage this thread issue in lfg
Oct 19 00:22:02 * Kensan swears in swiss-german so nobody understands him
Oct 19 00:22:05 <Ainulindale> When I tell you I'm far away from this UI issues :-)
Oct 19 00:22:30 <Ainulindale> by the way raster 
Oct 19 00:22:37 <Ainulindale> signal_unmarhsall <=
Oct 19 00:22:43 <Ainulindale> Like I say
Oct 19 00:22:50 <Ainulindale> dislexics have more fnu
Oct 19 00:23:47 <raster> yeah
Oct 19 00:23:50 <raster> i have a few of those
Oct 19 00:23:50 <raster> :)
Oct 19 00:23:58 <quickdev> Ainulindale, I'll commit some changes tomorrow...going to sleep now
Oct 19 00:25:04 <quickdev> see you tomorrow
Oct 19 00:26:35 <Kensan> arghhh
Oct 19 00:26:46 <Kensan> enough is enough
Oct 19 00:27:31 <Ainulindale> raster: I think I'll adapt lfg to handle "plugins" for different dbus environment
Oct 19 00:27:36 <Ainulindale> i.e. redefinition of connect and so on
Oct 19 00:27:50 <Ainulindale> That way I'll do a e plugin, it should be interesting
Oct 19 00:28:11 <Kensan> ok I need to do something else/ go to sleep
Oct 19 00:28:21 * Kensan is seriously pissed-off
Oct 19 00:29:30 <raster> or
Oct 19 00:29:42 <raster> make it setup and datayupe agnostic
Oct 19 00:29:47 <raster> so not really plugin
Oct 19 00:29:51 <raster> but u just set up callbacks
Oct 19 00:30:07 <raster> to handle the "native loop/setup etc. etc."
Oct 19 00:30:14 <Kensan> well I am off
Oct 19 00:30:41 <Ainulindale> raster: that's what I was thinking about
Oct 19 00:30:45 <Kensan> raster: with the current image you get kernel warnings on boot, since most modules are compiled in but are still being loaded on boot
Oct 19 00:30:52 <Ainulindale> I don't care about glib types, I just used them out of convenience at the beginning
Oct 19 00:31:17 <Kensan> good night and better look with what you guys are doing
Oct 19 00:34:55 <raster> Kensan:  i havent paid much attention :)
Oct 19 00:42:35 <raster> Ainulindale: they are convenient
Oct 19 00:42:42 <raster> e has its own datatypes too
Oct 19 00:42:52 <raster> we are shufflign that around right now though
Oct 19 01:18:47 <sicu> Ainulindale: sigh, ticket #40 is definately a showstopper :/
Oct 19 01:21:21 <sicu> seems to happen all the time now :/
Oct 19 01:22:28 <Ainulindale> That will be corrected by tomorrow
Oct 19 01:22:55 <Ainulindale> sicu: to resolve the problem relaunch ophonekitd
Oct 19 01:23:00 <Ainulindale> (in between)
Oct 19 01:29:28 <sicu> like this:? /etc/X11/Xsession.d/80ophonekitd
Oct 19 01:29:50 <sicu> Ainulindale: cause i have tried that and it doesnt work
Oct 19 01:32:13 <sicu> Ainulindale: quickdev told me to kill ophonekitd first by doing:
Oct 19 01:32:15 <sicu> [20081018-192725] <quickdev> sicu, ps aux | grep ophonekitd
Oct 19 01:32:15 <sicu> [20081018-192745] <quickdev> then kill -9 $PID....replace $PID with the process number(first column of ps aux I think)
Oct 19 01:33:22 <sicu> but don't think it was running as the only process that got listed was the ps one
Oct 19 01:50:46 <Ainulindale> ps auxw | grep ophonekitd
Oct 19 01:50:52 <Ainulindale> if it is then killall -9 ophonekitd
Oct 19 01:51:03 <Ainulindale> I can ensure you it is working I do that :-)
Oct 19 01:51:14 <Ainulindale> But if it isn't
Oct 19 01:51:22 <Ainulindale> And it should not be there as it quits after the sms screen
Oct 19 01:51:33 <Ainulindale> Type DISPLAY=":0" ophonekitd&
Oct 19 01:51:36 <Ainulindale> Should be enough
**** ENDING LOGGING AT Sun Oct 19 02:59:56 2008