**** BEGIN LOGGING AT Fri Nov 23 02:59:57 2007
Nov 23 04:11:01 <cdbot2> * * OM Bug 1022 has been created by mail(AT)mmontour.net
Nov 23 04:11:02 <cdbot2> * * splashimage.pl always generates 32bpp images
Nov 23 04:11:03 <cdbot2> * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=1022
Nov 23 05:07:48 <xzcvczx> nnpiggy: hey you aruond?
Nov 23 05:49:31 <CM> Good morning
Nov 23 05:50:11 <fgau> MOIN
Nov 23 05:51:01 <mbuf> CM, good morning
Nov 23 06:14:26 <Hopscotch> good morning
Nov 23 07:19:07 <edistar> hi
Nov 23 07:25:02 <mbuf> edistar, hi
Nov 23 07:25:18 <ewon> happy thanksgiving, etc.
Nov 23 07:26:18 <edistar> is thanksgiving today?
Nov 23 07:26:40 <Aria> Ayuh.
Nov 23 07:26:45 <ewon> yuo
Nov 23 07:26:46 <Aria> If Today is a Thursday anyway.
Nov 23 07:26:51 <ewon> *yup
Nov 23 07:27:04 <ewon> jusy back from stuffing my face with turkey
Nov 23 07:27:54 <edistar> it's friday here..
Nov 23 07:27:55 <edistar> ;)
Nov 23 08:22:09 <CM> OSM is starting to get known.. Hope it picks up here in Sweden too. (UK centric article) http://www.regdeveloper.co.uk/2007/11/22/blognation_os_maps/
Nov 23 08:48:47 <gordonsyme> I had to apply the following patch to build/tmp/staging/arm-angstrom-linux-gnueabi/include/X11/extensions/render.h, is this a known issue or should I file a bug? (http://pastebin.ca/794602)
Nov 23 08:49:38 <balrog-kun> yes, i believe something like this in bugzilla already
Nov 23 08:50:57 <balrog-kun> gordonsyme: but there was no patch actually
Nov 23 08:51:05 <balrog-kun> !ombug 1021
Nov 23 08:51:07 <cdbot2> * * Bug 1021, Status: VERIFIED (FIXED), Created: Unknown
Nov 23 08:51:08 <cdbot2> * * yoyo(AT)ghost.pl: xserver-kdrive
Nov 23 08:51:09 <cdbot2> * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=1021
Nov 23 08:51:18 <balrog-kun> gordonsyme: you could mention the patch there
Nov 23 08:51:45 <gordonsyme> balrog-kun: well it's not exactly a complex patch, just adds an include :)
Nov 23 08:52:20 <balrog-kun> gordonsyme: yea, but it's better than the current resolution i think :)
Nov 23 08:52:27 <gordonsyme> balrog-kun: on reading the bug it seems that patching isn't the right fix anyway...
Nov 23 08:53:07 <balrog-kun> the bug is that it doesn't update cleanly AIUI
Nov 23 08:53:29 <balrog-kun> the comments to #1021 basically say let's leave it like it is
Nov 23 08:53:56 <balrog-kun> (i don't think that's the right thing to do)
Nov 23 08:54:54 * balrog-kun is off
Nov 23 09:30:55 <CM> ~hail raster 
Nov 23 09:30:56 * apt bows down to raster and chants, "I'M NOT WORTHY!!"
Nov 23 09:31:45 * raster creeps into the gutter so he cannot be bowed down to because he is wallowing about in such low places
Nov 23 09:32:04 <raster> CM: actually it has only bee a bit of rain and drizzle
Nov 23 09:32:05 <raster> :)
Nov 23 09:32:15 <CM> Hehe, good to hear :)
Nov 23 09:32:28 <CM> Been working on anything fun lately? 
Nov 23 09:33:59 <raster> hmmmm
Nov 23 09:34:01 <raster> not entirely
Nov 23 09:35:47 <raster> unfortunately
Nov 23 09:36:38 <CoreDump> =)
Nov 23 09:36:50 <CVirus> raster: you got the GTA02 .. right ?
Nov 23 09:37:04 <CoreDump> It these "not-so-fun" things that show how fun much fun the "funny" things are
Nov 23 09:38:28 <raster> CoreDump: hahahahahahahah
Nov 23 09:38:32 <raster> CVirus:  yes - of course.
Nov 23 09:39:02 <CVirus> raster: so what do you think of it so far ?
Nov 23 09:39:15 <CVirus> raster: do we really need another revision ?
Nov 23 09:39:26 <CoreDump> my grammar and spelling are as always  impeccable *shudder*
Nov 23 09:39:26 <raster> ummmmm
Nov 23 09:39:39 <raster> i cańt much say
Nov 23 09:39:44 <raster> video is barly alive
Nov 23 09:39:55 <raster> internal builds are painful in supporting the gta-02
Nov 23 09:39:56 <CoreDump> and the word-order...truly a masterpiece
Nov 23 09:40:04 * CoreDump goes and hides
Nov 23 09:40:07 <raster> as the new kernels that fix usb 1000ms ping issues break gta-01 support
Nov 23 09:40:21 <raster> x driver support is barely up
Nov 23 09:40:24 <CVirus> I see
Nov 23 09:40:31 <raster> its too early to say
Nov 23 09:40:45 <raster> and do u think we need a gta02-v5 u mean?
Nov 23 09:41:27 <CVirus> yes
Nov 23 09:41:39 <raster> don't know yet
Nov 23 09:41:51 <raster> i kind fo ignore hw reveisions
Nov 23 09:41:59 <raster> as we still are working on wireless drivers
Nov 23 09:42:00 <raster> and video
Nov 23 09:42:04 <raster> and other thigns need work
Nov 23 09:42:08 <raster> so irrespective of hw revision
Nov 23 09:42:15 <raster> software is definitely behind
Nov 23 09:42:25 <CVirus> too bad
Nov 23 09:42:42 <CM> CVirus: v5 will be needed for that power issue
Nov 23 09:42:50 <CVirus> CM: yeah I think so
Nov 23 09:42:56 <CVirus> bug 957 iirc
Nov 23 09:43:02 <CM> *nod*
Nov 23 09:43:53 <ScaredyCat> oe,org dead ?
Nov 23 09:44:36 <CM> works for me
Nov 23 09:44:55 <raster> no idea
Nov 23 09:45:01 <raster> power consumption i am not sure about
Nov 23 09:45:22 <ScaredyCat> mtn: connecting to monotone.openembedded.org
Nov 23 09:45:23 <ScaredyCat> mtn: network error: failed to connect: Connection refused
Nov 23 09:45:23 <raster> we need to be able to suspend and resume cleanly to even hope to have good lifetime (Standby time)
Nov 23 09:45:36 <raster> in addtion while suspended you dont want to leak lots
Nov 23 09:45:44 <raster> of current and want to b able to suspend/shut down as
Nov 23 09:45:47 <raster> much hw as you can
Nov 23 09:45:48 <CM> ScaredyCat: Ah, didn't try monotone
Nov 23 09:47:10 <CM> raster: Of course. The power thing we talked about is in cooment 1 here: http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=957
Nov 23 09:48:23 <ScaredyCat> seems to be alive again
Nov 23 09:48:33 <CM> ScaredyCat: Was just about to say that :)
Nov 23 09:58:52 <mbuf> hrw, can your friend who wanted an IDE, try this? http://free-opensource.qvantel.net/mediawiki/index.php/Preity-plugin
Nov 23 10:06:07 <raster> CM: ummm. doesn't mean much to me actually :(
Nov 23 10:06:22 <raster> looks like we can;'t fetch battery status properly
Nov 23 10:06:33 <raster> without using 2 i2c lines
Nov 23 10:08:29 <ruxpin> what's the status of the bluetooth driver?
Nov 23 10:09:05 <madwoota> he got some porcelein veneers ... he's now a sparkly white tooth driver
Nov 23 10:09:13 <ruxpin> on gta01..
Nov 23 10:09:50 <raster> ruxpin: no idea. as best i know its a standard bluetooth chipset wired up by usb
Nov 23 10:09:55 <raster> so it should "just work"
Nov 23 10:10:15 <raster> but i havent payed much attention to it frankly
Nov 23 10:11:46 <jeddy3> ruxpin, bluetooth works fine on GTA01, if that was what you meant
Nov 23 10:17:58 <CM> ruxpin: It uses the current bluez libs
Nov 23 10:20:01 <Stephmw> mornin
Nov 23 10:20:37 <hrw> mbuf: will give him info
Nov 23 10:21:32 <mbuf> hrw, thanks, also have a detailed user guide there (with screenshots)
Nov 23 10:22:22 <CIA-14> 03thomas * r3482 10/trunk/src/target/OM-2007.2/applications/openmoko-dialer2/ (4 files in 2 dirs): 
Nov 23 10:22:22 <CIA-14> * src/phone-kit/moko-dialer.c: (moko_dialer_rejected):
Nov 23 10:22:22 <CIA-14> * src/phone-kit/moko-talking.c: (moko_talking_class_init),
Nov 23 10:22:22 <CIA-14> (window_delete_event_cb), (on_pad_user_input), (moko_talking_init):
Nov 23 10:22:22 <CIA-14> * src/phone-kit/moko-talking.h:
Nov 23 10:22:23 <CIA-14> Add DTMF pad while in call
Nov 23 10:28:41 <CIA-14> 03olv * r3483 10/trunk/src/target/xserver/: Move the development of Xglamo under trunk/
Nov 23 10:34:40 <raster> olv: yay! work!
Nov 23 10:34:40 <raster> :)
Nov 23 10:44:12 <CIA-14> 03laforge * r3484 10/trunk/src/target/u-boot/patches/ (12 files): 
Nov 23 10:44:12 <CIA-14> sync with mainline git rev 9a337ddc154a10a26f117fd147b009abcdeba75a
Nov 23 10:44:12 <CIA-14> (only makefile changes)
Nov 23 10:44:28 <CM> Oh.. Nice too
Nov 23 10:45:52 <CIA-14> 03olv * r3485 10/trunk/src/target/xserver/patches/ (7 files): (log message trimmed)
Nov 23 10:45:52 <CIA-14> Split smedia-glamo.patch into 4 patches:
Nov 23 10:45:52 <CIA-14> xglamo-build.patch
Nov 23 10:45:52 <CIA-14>  integrate Xglamo into the build system
Nov 23 10:45:52 <CIA-14> xglamo.patch
Nov 23 10:45:52 <CIA-14>  new Xfbdev-based server, Xglamo
Nov 23 10:45:54 <CIA-14> xglamo-kaa.patch
Nov 23 10:49:05 <flexd> wow, alot of action today o_0
Nov 23 10:54:19 <CIA-14> 03thomas * r3486 10/trunk/src/target/OM-2007.2/applications/openmoko-dialer2/ (ChangeLog src/phone-kit/moko-dialer.c): 
Nov 23 10:54:19 <CIA-14> * src/phone-kit/moko-dialer.c: (on_pin_requested): Send registration
Nov 23 10:54:19 <CIA-14> request one second after sending pin
Nov 23 10:55:21 <CM> flexd: Fo sho ;)
Nov 23 10:55:45 <ScaredyCat> yoink!
Nov 23 11:06:49 * ScaredyCat wonders ....... uclibc ..
Nov 23 11:10:17 * CVirus is dying for a GTA02
Nov 23 11:13:26 * mbuf already died; in ghost mode, hunting FIC for a GTA02
Nov 23 11:14:26 <CVirus> I lost my Nokia 6600 and I'm in need of a mobile phone as soon as possible
Nov 23 11:15:43 <borg_> http://pastebin.ca/794707
Nov 23 11:15:46 <borg_> is this c99?
Nov 23 11:15:49 <borg_> what does this do?
Nov 23 11:17:33 <CIA-14> 03olv * r3487 10/trunk/src/target/xserver/patches/ (xglamo-kaa.patch xglamo-xv.patch xglamo.patch): regenerate patches with --no-timestamps flag
Nov 23 11:21:44 <borg_> ah ok
Nov 23 11:21:46 <borg_> i got it
Nov 23 11:21:48 <borg_> never seen this :)
Nov 23 11:23:05 <tuukkah> mjr, are you having success with using gta01 as your main phone?
Nov 23 11:23:36 <CIA-14> 03olv * r3488 10/trunk/src/target/xserver/patches/xglamo-kaa.patch: some commands take very long time to finish
Nov 23 11:24:40 <mudyc> tuukkah: do you use it regularly?
Nov 23 11:25:34 <tuukkah> mudyc, as a gprs modem and music player only
Nov 23 11:26:47 <mudyc> is there any problems with the gsm side?
Nov 23 11:27:31 <tuukkah> what do you mean?
Nov 23 11:27:59 <mudyc> like why you don't use it as a gsm phone?
Nov 23 11:28:28 <borg_> because there is no working gui für sms :)
Nov 23 11:28:49 <tuukkah> we don't have proper power management so the phone needs to be charged every four hours or so
Nov 23 11:28:52 <borg_> i am not the sending-a-log-of-sms guy
Nov 23 11:28:56 <tuukkah> and gsmd is very flaky
Nov 23 11:28:59 <borg_> but i get some sms from people
Nov 23 11:29:06 <borg_> and i have to be able to read them :)
Nov 23 11:30:16 <CIA-14> 03olv * r3489 10/developers/olv/xserver/: 
Nov 23 11:30:16 <CIA-14> Deprecated.
Nov 23 11:30:16 <CIA-14> Development has moved to
Nov 23 11:30:16 <CIA-14> http://svn.openmoko.org/trunk/src/target/xserver/
Nov 23 11:33:55 <ScaredyCat> anyone got an opinion on the pocketsurfer2 ?
Nov 23 11:35:12 <mudyc> tuukkah: we have designed a new gsmd from scratch based on glib. it would be a interesting to see if om is interested in for the design
Nov 23 11:36:49 <robster> mudyc: is the source code for that available anywhwere?
Nov 23 11:37:18 <rwhitby> mudyc: who is "we"
Nov 23 11:37:43 <mudyc> rwhitby: the company i work in
Nov 23 11:38:02 <rwhitby> mudyc: are you able to say which company that is?
Nov 23 11:38:19 <mudyc> robster: currently no, i may provide a snapshot for review
Nov 23 11:38:38 <mudyc> rwhitby: sure, ixonos
Nov 23 11:40:20 <ScaredyCat> is it goin gto be gpl'd ?
Nov 23 11:41:38 <mudyc> ScaredyCat: what licence would you prefer?
Nov 23 11:41:55 <ScaredyCat> well, this is an opensouce project :)
Nov 23 11:42:07 <ScaredyCat> so you could dul licence it...
Nov 23 11:42:10 <ScaredyCat> dual 
Nov 23 11:42:28 <CVirus> An Introduction to Bluetooth Programming http://people.csail.mit.edu/albert/bluez-intro/
Nov 23 11:42:31 <tuukkah> mudyc, very cool option, i think many people have been hoping for a gsmd replacement to arrive :-)
Nov 23 11:42:34 <ScaredyCat> so your company could allow people to buy it for closed devices...
Nov 23 11:42:58 <ScaredyCat> but the open version could be used in projects like om
Nov 23 11:43:53 <mudyc> ScaredyCat: i would prefer gpl 2 or greater. is that ok?
Nov 23 11:44:09 <ScaredyCat> well, to me that sounds good...
Nov 23 11:44:18 <ScaredyCat> but I don;t get to decide it :)
Nov 23 11:45:05 <ScaredyCat> is your company opensouce frinedly then ? :)
Nov 23 11:45:10 <ScaredyCat> friendly
Nov 23 11:46:45 <tuukkah> gsmd is currently GPLv2 and libgsmd LGPLv2.1
Nov 23 11:48:19 <tuukkah> http://svn.openmoko.org/trunk/src/target/gsm/
Nov 23 11:48:22 <ScaredyCat> gpl v2 is probably good enough - imho of course
Nov 23 11:49:17 <tuukkah> at least some of the source files include "or later"
Nov 23 11:49:51 <mjr> "or later" is more practical
Nov 23 11:50:22 <ScaredyCat> yeah, so long as someone doesn;t force it to 3
Nov 23 12:10:03 <mickeyl> morning florian 
Nov 23 12:10:19 <mickeyl> florian: i loved our meeting. we should do this more often
Nov 23 12:12:26 <yoyo> hi !
Nov 23 12:12:56 <florian> hi mickeyl 
Nov 23 12:13:17 <florian> mickeyl: yes indeed, that was really really nice. 
Nov 23 12:13:24 <florian> hi yoyo 
Nov 23 12:13:39 <yoyo> :)
Nov 23 12:40:07 <torpor> yo
Nov 23 12:40:16 <torpor> mickeyl: what meeting!
Nov 23 12:40:23 <torpor> florian: what meeting!
Nov 23 12:44:46 <hrw> yoyo: hej
Nov 23 12:46:09 <yoyo> czesc
Nov 23 12:48:02 <yoyo> dose some work on the lock screen (resume problem)?
Nov 23 12:49:08 <hrw> yoyo: power button should resume
Nov 23 12:50:15 <yoyo> hrw: yes it resume bat. after screen back light on and flush main screen on screen it freeze
Nov 23 12:51:15 <hrw> yoyo: checked bugtracker?
Nov 23 12:51:33 <yoyo> hrw: quiet about it
Nov 23 12:54:24 <CIA-14> 03thomas * r3490 10/trunk/src/target/OM-2007.2/applications/openmoko-dialer2/ (ChangeLog src/phone-kit/moko-talking.c): 
Nov 23 12:54:24 <CIA-14> * src/phone-kit/moko-talking.c: (moko_talking_set_clip),
Nov 23 12:54:24 <CIA-14> (incoming_timeout), (moko_talking_incoming_call),
Nov 23 12:54:24 <CIA-14> (outgoing_timeout), (moko_talking_outgoing_call),
Nov 23 12:54:26 <CIA-14> (on_pad_user_input), (moko_talking_init):
Nov 23 12:54:28 <CIA-14> Reset "talking" UI on new call
Nov 23 13:05:27 <CIA-14> 03thomas * r3491 10/trunk/src/target/OM-2007.2/artwork/icons/openmoko-standard/ (4 files in 2 dirs): * Add images for stock_person
Nov 23 13:09:29 <edistar> hi
Nov 23 13:12:56 <torpor> 404: NOTE: fetch http://downloads.openmoko.org/sources/openmoko2_svn.o-hand.com_.repos.dates.branches_661_.tar.gz
Nov 23 13:34:23 <CM> torpor: Unofficial oe/om meeting in Berlin earlier this week I think. :)
Nov 23 13:37:36 * ScaredyCat wonders what happens if you actually look like 'stock_person'
Nov 23 13:38:03 <CM> stock = log in swedish :D
Nov 23 13:38:33 <ScaredyCat> lart = lart in any language
Nov 23 13:38:35 <ScaredyCat> ~lart CM
Nov 23 13:38:35 * apt slaps CM upside and over the head with one freakishly huge killer whale named hugh
Nov 23 13:38:46 <zash> hahaha
Nov 23 13:38:49 <CM> ~bop ScaredyCat 
Nov 23 13:38:52 <zash> god noe
Nov 23 13:38:53 <zash> one
Nov 23 13:38:56 <CM> Pff.. 
Nov 23 13:39:00 <torpor> CM: damnit, why mention it in the channel if we're not allowed to know!  tease!
Nov 23 13:39:02 <CM> ~poke ScaredyCat 
Nov 23 13:39:04 * apt cuts down a small tree, sneaks up behind ScaredyCat, pokes ScaredyCat repeatedly, hilarity ensues.
Nov 23 13:39:17 <ScaredyCat> ~fish
Nov 23 13:39:18 <apt> somebody said fish was FISHFISHFISH! DO THE FISH DANCE! "Give a man a fish and you'll feed him a day. Teach him how to fish and he'll feed himself for the rest of his life." This is so appropriate, instead of asking us to tell you exactly what to do, why not read some docs, then come back and ask specific questions which aren't covered?, or ...
Nov 23 13:39:19 <ScaredyCat> ~fish cm
Nov 23 13:39:20 * apt slaps cm around with a large trout
Nov 23 13:39:26 <ScaredyCat> wtf
Nov 23 13:39:31 <CM> ~commands
Nov 23 13:39:49 <ScaredyCat> ~none found
Nov 23 13:39:56 <CM> I like the emulate ones
Nov 23 13:39:59 <ScaredyCat> ~blame CM for that one
Nov 23 13:40:00 * apt blames trivex` and perhaps CM for that one and all the evil in the world
Nov 23 13:40:07 <zash> ... slap a man in the face and he'll never come to you begging for food again
Nov 23 13:40:08 <CM> ~emulate stevie
Nov 23 13:40:16 <zash> (with a fish)
Nov 23 13:40:18 <CM> Äh.. 
Nov 23 13:40:34 <ScaredyCat> 'Ä'h ?
Nov 23 13:40:38 <CM> ~emulate stewie
Nov 23 13:40:39 <apt> No Sprinkles. For every sprinkle I find, I shall kill you.
Nov 23 13:40:43 <CM> ~emulate stewie
Nov 23 13:40:47 <ScaredyCat> lol
Nov 23 13:40:47 <zash> &Auml;h!
Nov 23 13:40:54 <ScaredyCat> stewie, not steve!
Nov 23 13:41:09 <CM> ~emulate koen
Nov 23 13:41:10 <apt> the release notes have hints
Nov 23 13:41:19 <CM> ~emulate Crofton
Nov 23 13:41:20 * apt don't be so elitist!
Nov 23 13:41:33 <ScaredyCat> shouldn;t that be "whatever you do is wrong, I'm right" for Keon
Nov 23 13:41:35 <zash> ~rape CM
Nov 23 13:41:36 * apt takes CM behind the WallMart and makes a few grunts and screams
Nov 23 13:41:47 <ScaredyCat> oe
Nov 23 13:41:48 <ScaredyCat> eo
Nov 23 13:41:48 <CM> Eeek!
Nov 23 13:41:50 <ScaredyCat> eieio
Nov 23 13:42:01 <CM> I'
Nov 23 13:42:12 <CM> I'd spell that Wal-Mart..
Nov 23 13:42:25 <ScaredyCat> with a lart lart here and a lart lart there, here a lsrt there a lart everywhere a lart lart
Nov 23 13:42:28 <zash> Wal-Mört
Nov 23 13:42:37 <CM> hehe
Nov 23 13:42:43 <ScaredyCat> ~emulate CM
Nov 23 13:42:55 <ScaredyCat> emulate that!
Nov 23 13:43:05 <CM> ~emulate XorA
Nov 23 13:43:12 <CM> No luck..
Nov 23 13:43:36 <CM> ~hate APL
Nov 23 13:43:43 <CM> Pff.. Not even that?
Nov 23 13:43:52 <CM> apt: You suck!
Nov 23 13:43:53 <apt> Sorry!
Nov 23 13:43:57 <ScaredyCat> naa it did it right... emulating XorA results in doing sod all ;P
Nov 23 13:44:10 <CM> ScaredyCat: Hehe :P
Nov 23 13:44:13 * ScaredyCat giggles
Nov 23 13:44:24 <ScaredyCat> ~poke XorA
Nov 23 13:44:25 * apt cuts down a small tree, sneaks up behind XorA, pokes XorA repeatedly, hilarity ensues.
Nov 23 13:44:28 <CM> That's almost unfairl..
Nov 23 13:44:30 <Sup3rkiddo> ~rape apt
Nov 23 13:44:31 * apt takes apt behind the WallMart and makes a few grunts and screams
Nov 23 13:44:35 <CM> Hehe
Nov 23 13:45:03 <Sup3rkiddo> ...but why walmart
Nov 23 13:45:07 <mickeyl> ~emulate mickeyl
Nov 23 13:45:12 <mickeyl> hmm
Nov 23 13:45:13 <CM> ~blame ScaredyCat for starting this silliness
Nov 23 13:45:14 * apt blames rob_w|laptop and perhaps ScaredyCat for starting this silliness and the end of the world as we know it
Nov 23 13:45:22 <ScaredyCat> pppfff
Nov 23 13:45:38 <ScaredyCat> ~emulate emulation
Nov 23 13:45:39 * apt performs an exercise in hopeless futility.
Nov 23 13:45:40 <mudyc> ScaredyCat, robster, rwhitby: ok, a snapshot of unfinished concept: http://www.ixonos.com/opensource/gsmd-snapshot-2007-11-23-for-design-discussion.tar.gz
Nov 23 13:45:47 <rob_w|laptop> ???
Nov 23 13:46:01 <ScaredyCat> cool mudyc
Nov 23 13:46:08 <ScaredyCat> lol @ rob_w|laptop
Nov 23 13:47:03 <CM> mudyc: Sure that link works?
Nov 23 13:48:33 <ScaredyCat> worked here
Nov 23 13:48:36 <mudyc> CM: yes (just tried from outher host)
Nov 23 13:48:49 <ScaredyCat> perhaps its only for special people....
Nov 23 13:49:22 <CM> mudyc: Ok. The link you pasted above gives me: The requested URL /opensource/gsmd-snapshot-2007-11-23-for-design-discussion was not found on this server.
Nov 23 13:50:50 <mudyc> CM: try to append the missing .tar.gz
Nov 23 13:51:17 <CM> Heh, thanks..
Nov 23 13:52:41 <robster> mudyc: looks nice.
Nov 23 13:53:52 <CM> mudyc: Cool
Nov 23 13:54:00 <mickeyl> whew!
Nov 23 13:54:05 <CM> mudyc: You should talk some with Writchie|turkey 
Nov 23 13:54:12 <Sup3rkiddo> mudyc: the above one uses dbus?
Nov 23 13:54:17 <fatal-> Sup3rkiddo: yes
Nov 23 13:54:20 * Sup3rkiddo takes a look
Nov 23 13:55:26 <CM> There are even nice docs :)
Nov 23 13:55:27 <fatal-> it has not been split into a gsm library and a separate dbus interface in terms of binaries, but they exist as separate files anyhow so it's more or less a formality
Nov 23 13:56:05 <ScaredyCat> wow! there's more docs there than ..... well ... you know ... ;)
Nov 23 13:56:14 <CM> ScaredyCat: :P
Nov 23 13:56:28 <Sup3rkiddo> fatal-: if theres a plan, i am in for a python interface, which should be trivial anyway
Nov 23 13:57:15 <CM> Sup3rkiddo: The test python example is using dbus
Nov 23 13:57:37 <ScaredyCat> ~lart Sup3rkiddo for not reading the readme
Nov 23 13:57:37 * apt breaks out the Hoover and sucks up Sup3rkiddo for not reading the readme
Nov 23 13:58:30 <Sup3rkiddo> oh..dang...i am outta work then :/
Nov 23 13:58:37 <Sup3rkiddo> ~lart Sup3rkiddo 
Nov 23 13:58:37 * apt says "boot to the head" and knocks Sup3rkiddo over
Nov 23 13:59:36 <balrog-kun> ewww glib :p
Nov 23 14:00:09 <Sup3rkiddo> ...man, look at the friggin docs in there, i love it
Nov 23 14:00:46 <Sup3rkiddo> hallelujah, i found *.doc in the doc folder....
Nov 23 14:02:04 <CM> No pdf in the tech-spec folder, but the filenames of the eps files sound promising.. :P
Nov 23 14:03:01 <mickeyl> "As a result of limited memory resources the C++
Nov 23 14:03:01 <mickeyl> support was dropped"
Nov 23 14:03:03 <mickeyl> that's very sad
Nov 23 14:03:07 <mickeyl> can you provide more info?
Nov 23 14:03:22 <mickeyl> i seriously doubt that C++ has a real-life memory impact
Nov 23 14:03:30 <mudyc> CM: uh, sorry about that. i'll fix that on next week
Nov 23 14:03:57 <fatal-> mickey1: it's not just memory usage; the whole thing was pain to use in multiple ways
Nov 23 14:04:32 <mickeyl> i need to go now, but i'm very interested about your experiences
Nov 23 14:04:36 <mickeyl> since i'm a strong C++ supporter
Nov 23 14:04:37 <CM> mudyc: Np. I'll run the makefile once I get home :)
Nov 23 14:04:39 <mickeyl> please lets talk 
Nov 23 14:04:40 <mickeyl> l8er
Nov 23 14:05:24 <balrog-kun> i also like C++ but i don't think separate C++ bindings for gsmd would be of much use
Nov 23 14:06:06 <fatal-> I've become less addicted to C++ recently, I've noticed
Nov 23 14:06:25 <balrog-kun> and the favourite part of writing C++ applications for many C++ fans is actually making the object oriented wrappers for C stuff
Nov 23 14:06:28 <mudyc> note -- we used sqlite3 for sms storage and there's dependency to very light db wrapper library which was not provided with the source snapshot
Nov 23 14:08:04 <tuukkah> fatal-, are you also part of the gsmd2 effort?
Nov 23 14:08:16 * ScaredyCat thinks sqlite is a good place to stor all sorts of stuff
Nov 23 14:09:36 <tuukkah> mudyc, could you easily provide the stuff in doc/ as pdf somewhere?
Nov 23 14:10:04 * balrog-kun thinks that if gsmd was to be officially replaced with a new gsmd it should support the current APIs so that all of existing OM stack doesn't have to be changed at this point
Nov 23 14:10:26 <balrog-kun> currently i think the messages are stored outside gsmd, in a place called journal or something
Nov 23 14:10:27 <tuukkah> balrog-kun, is there much that uses the current APIs?
Nov 23 14:10:44 <ScaredyCat> it's not that much of a change is it.... ?
Nov 23 14:10:52 <ScaredyCat> s/of/to/
Nov 23 14:10:52 <apt> ScaredyCat meant: it's not that much to a change is it.... ?
Nov 23 14:11:02 <ScaredyCat> s/of a/to/
Nov 23 14:11:02 <apt> ScaredyCat meant: it's not that much to change is it.... ?
Nov 23 14:11:23 <ScaredyCat> s/ it's not that much to change is it/cheese/
Nov 23 14:11:31 <balrog-kun> tuukkah: other than the kernel and u-boot, i think the dialer/today stuff is all openmoko has (which is ready to use)
Nov 23 14:11:32 <ScaredyCat> pffaft
Nov 23 14:11:45 <balrog-kun> so it's pretty much all of existing user-space openmoko
Nov 23 14:11:57 <tuukkah> balrog-kun, but how much is that dependent of gsmd directly?
Nov 23 14:12:13 <tuukkah> as there's libmokogsm or what was the name
Nov 23 14:12:19 <balrog-kun> tuukkah: theoretically there's libmokogsmd which abstracts gsmd
Nov 23 14:12:36 <Sup3rkiddo> but want that supposed to replaced by phonekit?
Nov 23 14:12:51 <balrog-kun> tuukkah: but the place where messages are stored woul be change not only in gsmd api
Nov 23 14:13:20 <tuukkah> balrog-kun, as long as the gsmd2 api allows us to extract the messages, we can store them wherever =)
Nov 23 14:13:40 <balrog-kun> i haven't looked at the proposed new gsmd but from what i hear it would not only replace gsmd but also libmokogsmd, phonekit, journal and the rest of the design
Nov 23 14:14:22 <tuukkah> i think we better look at the api, docs, and code before speculating :-)
Nov 23 14:14:30 <balrog-kun> true
Nov 23 14:17:23 <mudyc> tuukkah: http://ip.fi/~mudyc/projects/ixonos/tech-spec-gsm-daemon.pdf
Nov 23 14:18:10 <CM> ~praise mudyc 
Nov 23 14:18:11 <apt> All hail mudyc!
Nov 23 14:18:27 <tuukkah> mudyc, what about the doc :-)
Nov 23 14:18:42 <tuukkah> perhaps i can do the export
Nov 23 14:18:47 <fatal-> balrog-kun: it can replace all of those, if wanted, but that's just one option
Nov 23 14:20:03 <fatal-> we do define our own storage for sms'es, we have the d-bus interface etc but on the bottom, the gsm functionality is it's own part of the whole
Nov 23 14:20:15 <fatal-> and defineable as a single .so if so wanted
Nov 23 14:20:44 <balrog-kun> fatal-: nice
Nov 23 14:21:22 <mudyc> tuukkah: i think it has been created with openoffice but stored in doc format :(
Nov 23 14:23:48 <tuukkah> if anyone prefers: http://tuukka.iki.fi/tmp/gsmd2-req-spec-gsm.pdf
Nov 23 14:26:22 * balrog-kun likes the design (maybe except glib) :p
Nov 23 14:26:34 <balrog-kun> i think it would be considered as the #1 option for openmoko if the code was available at the beginning of 2007
Nov 23 14:27:07 <tuukkah> mudyc, are you sure you don't get in trouble for publishing documents with "company confidential" labels ?-)
Nov 23 14:27:36 <fatal-> tuukkah: well, they probably should be cleared away but they are there per default anyhow
Nov 23 14:28:12 <tuukkah> fatal-, i notice you're an "other stakeholder" :-)
Nov 23 14:29:00 <fatal-> one of them, yes.
Nov 23 14:36:04 <edistar> why are there no updates about gta-02 :(
Nov 23 14:39:34 <CM> edistar: Probably because they're stil working on rev5
Nov 23 14:40:47 <CIA-14> 03thomas * r3492 10/trunk/src/target/OM-2007.2/applications/openmoko-dialer2/ (4 files in 2 dirs): 
Nov 23 14:40:47 <CIA-14> * src/phone-kit/moko-dialer.c: (moko_dialer_dial),
Nov 23 14:40:47 <CIA-14> (moko_dialer_init):
Nov 23 14:40:47 <CIA-14> Fix direction field of log entry when dialing calls (outgoing)
Nov 23 14:40:47 <CIA-14> * src/phone-kit/moko-talking.c:
Nov 23 14:40:47 <CIA-14> * src/phone-kit/moko-talking.h:
Nov 23 14:40:49 <CIA-14> Remove moko-journal requirement
Nov 23 14:50:33 <tuukkah> so with gsmd2, multiplexing needs to be elsewhere?
Nov 23 14:51:31 <CM> tuukkah: Isn't Writchie|turkey working on moving the multiplexing to the kernel?
Nov 23 14:51:59 <tuukkah> that's something i haven't been able to follow
Nov 23 14:53:16 <tuukkah> i'd expect the gsmd part to have some requirements to support that configuration too
Nov 23 14:53:47 <CM> It was balrog-kun and mickey_away who discussed with Writchie, I just saw it in my backlog
Nov 23 14:54:20 <tuukkah> yeah, i think i saw the discussion but i don't have the full picture
Nov 23 14:54:29 <CM> Me neither :)
Nov 23 14:54:46 * CM doesn't have the full picture of anything
Nov 23 14:54:51 <fatal-> currently atleast the situation is such that the multiplexing needs to be done outside
Nov 23 14:55:20 <fatal-> we have done some experiments with multiplexing (well, not me actually) but there have been lots of problems with the h/w we are using
Nov 23 14:55:31 <fatal-> trying to resolve that
Nov 23 14:55:47 <tuukkah> you can always use the neo :-)
Nov 23 14:55:51 <balrog-kun> there are also problems with GTA01 h/w + multiplexing :)
Nov 23 14:56:17 <fatal-> I guess we should try neo, I have one on my table. otoh, I have not had much success with dialing with that particular device so far
Nov 23 14:56:34 <CM> torpor: Something for you maybe? http://gizmodo.com/gadgets/synthesizer/korg-kaossilator-brings-groovy-music-to-the-masses-verdict-must-get-one-325837.php
Nov 23 14:56:51 <CM> Damn, it was cut off
Nov 23 14:56:56 <CM> http://gizmodo.com/gadgets/synthesizer/korg-kaossilator-brings-groovy-music-to-the-masses-verdict-must-get-one-325837.php
Nov 23 14:57:04 <tuukkah> fatal-, dialing is something i'm able to do :-)
Nov 23 14:57:14 * CM appologises for the spam
Nov 23 14:58:08 <CM> torpor: Sorry: http://tinyurl.com/yqya8n
Nov 23 14:58:10 <tuukkah> CM, your setup may display the lines as cut off but they're not
Nov 23 14:58:29 * CM hides in the corner in shame
Nov 23 14:59:29 <tuukkah> CM, perhaps there's no problem with your setup and you can read the complete lines from your logs =)
Nov 23 15:00:09 <CM> I think i can, but I guess irssi doesn't chop up single words longer than 80 chars
Nov 23 15:00:30 <CM> Or it's because I have the same screen attached from more than one comp
Nov 23 15:12:35 <torpor> CM: yeah i know about those devices (worked on something similar myself once) but right now the big ticket item for that sort of thing is in fact the Nintendo DS Lite! :)
Nov 23 15:13:03 <torpor> cos ther's apps just like this for NDS ..
Nov 23 15:19:20 <tuukkah> fatal-, out of curiosity, do your specs and implementation currently provide for AT command cancelling?
Nov 23 15:19:25 <mmp> hello, anyone noticed that web archive of community mailing list is not being updated?
Nov 23 15:19:37 <mbuf> mmp, yes
Nov 23 15:19:50 <mbuf> mmp, quite recently
Nov 23 15:20:00 <Vegar> I haven't received any emails from the list either
Nov 23 15:20:15 <mmp> mbuf: ok, so it isn't just buggy transparent proxy somewhere on way between me and lists.openmoko.org :)
Nov 23 15:20:24 <CM> torpor: Ok, cool :)
Nov 23 15:20:36 <CIA-14> 03thomas * r3493 10/trunk/src/target/OM-2007.2/artwork/themes/openmoko-standard-2/gtk-2.0/gtkcombobox: * Fix GtkComboBoxEntry entry fields
Nov 23 15:21:00 <mbuf> mmp, i couldn't find my recent posts in the archives
Nov 23 15:21:37 <PBeck> hi
Nov 23 15:23:09 <Vegar> mbuf: do you receive emails from the list?
Nov 23 15:24:31 <mbuf> Vegar, yes
Nov 23 15:24:45 <Vegar> meh
Nov 23 15:24:56 <Vegar> I haven't gotten any in quite a while
Nov 23 15:25:37 <ScaredyCat> you should try being married.... you don;t get any, ever.
Nov 23 15:25:40 <mbuf> Vegar, openmoko-devowner, -apps, -devel
Nov 23 15:26:49 <Vegar> I think I'm subscribing to -community, -devowner and -commitlog
Nov 23 15:27:08 <Vegar> I've only received commitlog mails this week
Nov 23 15:28:28 <CM> Vegar: Enjoy the silence! ;)
Nov 23 15:28:39 <Vegar> trust me, I have
Nov 23 15:28:55 <CM> ScaredyCat: Hehe
Nov 23 15:55:30 <mwester> ScaredyCat:  That's why women get married in the first place.
Nov 23 16:15:27 <wibbit> Are there ANY woman in this channel? :)
Nov 23 16:20:03 <CM> wibbit: Not sure right now, but there used to be at least two :)
Nov 23 16:20:20 <whm> shocking.
Nov 23 16:20:40 <CM> I'm assuming they keep quiet about it not to be harassed by a bunch of greeks ;)
Nov 23 16:20:49 <wibbit> I don't blame them
Nov 23 16:20:54 <CM> Sad it's still that way
Nov 23 16:21:01 <wibbit> quite so
Nov 23 16:21:11 <mmontour> http://www.xkcd.com/322/
Nov 23 16:22:38 <CM> mmontour: Hehe :)
Nov 23 16:23:23 <edistar> mmontour: funny ;)
Nov 23 16:25:39 * CM finally heads home after a long week
Nov 23 17:10:30 <ScaredyCat> of doing nothing....
Nov 23 17:10:32 <ScaredyCat> ;)
Nov 23 17:16:59 <CM> ScaredyCat: :P
Nov 23 17:17:11 <CM> ScaredyCat: Not far from the truth though... :D
Nov 23 17:18:00 <dinamich> hey, what up with mailing lists? I have not received any emails for couple of days.
Nov 23 17:18:11 <edistar> me neither
Nov 23 17:27:54 <mwester> The cone of silence has descended up us all.
Nov 23 17:28:50 <tuukkah> busy reading the gsmd2 docs =)
Nov 23 17:29:06 * CM is on his way to a party
Nov 23 17:34:11 * Writchie passed mwester an ice cream cone
Nov 23 17:42:36 <tuukkah> does this say anything: "The first target hardware platform [for the software] is ARM9."
Nov 23 17:43:44 <Writchie> tuukkah: virtually all target smartphones are ARM 9 or better
Nov 23 17:44:39 <CIA-14> 03thomas * r3494 10/trunk/src/target/OM-2007.2/applications/openmoko-today2/ (ChangeLog src/today-main.c src/today-task-manager.c): 
Nov 23 17:44:39 <CIA-14> * src/today-main.c: (main): Initialise OpenMoko stock items
Nov 23 17:44:39 <CIA-14> * src/today-task-manager.c: (today_task_manager_page_create): Use
Nov 23 17:44:39 <CIA-14> gtk_tool_button_new_from_stock when loading stock items
Nov 23 17:45:32 <tuukkah> Writchie, ok. so the neo processor is ARM 9 too?
Nov 23 17:45:57 <Writchie> yes - this is also minimum for android 
Nov 23 17:46:19 <Tieku> can plans to put android onto the neo?
Nov 23 17:46:22 <Writchie> arm7 lacks mmu so not suitable for a standard kernel
Nov 23 17:46:23 <Tieku> any plans*
Nov 23 17:46:33 <tuukkah> i read how it's not possible to run android on the neo
Nov 23 17:46:48 <tuukkah> since the current binaries are ARMv5 and the neo is ARMv4?
Nov 23 17:47:35 <Writchie> source for android has not been made available - if and when it is it will likely be ported in due course
Nov 23 17:47:57 <Tieku> Writchie: cheers
Nov 23 17:48:31 <edistar> why would you want android?
Nov 23 17:49:46 <Writchie> edistar: to run android applications 
Nov 23 17:50:08 <tuukkah> porting android to openmoko would make sense then
Nov 23 17:50:15 <edistar> hm, ok
Nov 23 17:50:41 <Writchie> it remains to be seen, its many months away and lots to do before then.
Nov 23 17:53:36 <Writchie> if you think phones are broken now, wait till you see android phones ;-)
Nov 23 17:53:42 <edistar> heh
Nov 23 17:54:08 <tuukkah> i would give google the benefit of doubt
Nov 23 17:54:34 <Writchie> google has no control over the phone 
Nov 23 17:55:10 <Writchie> you will get what the phone manufacturer provides and if that's broken it will likely remain broken as it does today. 
Nov 23 17:56:07 <tuukkah> google may do their best in the collaboration with the manufacturers to get a result that doesn't destroy their brand
Nov 23 17:56:49 <edistar> tuukkah: I think they will test the devices.. and not officially support it when it's too broken
Nov 23 17:58:44 <Writchie> it still apears to me that the only open platform is and will remain the neo. That is not likely to change until what runs on that open platform is worth running. OM is proving whether the concept of an open platform phone is viable and the jury is still out on that, IMNSHO.
Nov 23 17:58:57 <edistar> exactly..
Nov 23 18:00:18 <Writchie> I will say, however, that based on the qtopia, the concept does appear to be technically valid. 
Nov 23 18:01:01 <tuukkah> i don't want to contest neo being the only open phone
Nov 23 18:01:34 <Writchie> tell me another and I'll buy it ;)
Nov 23 18:01:36 <tuukkah> on the software side, it's less becoming less clear
Nov 23 18:01:55 <tuukkah> indeed =)
Nov 23 18:02:34 <tuukkah> should be easy for openmoko to cash on the open hardware
Nov 23 18:03:01 <edistar> if a lot of people like open hardware..
Nov 23 18:03:13 <tuukkah> or a lot of companies
Nov 23 18:03:55 <edistar> yes
Nov 23 18:03:57 <tuukkah> Writchie, what does qtopia validate?
Nov 23 18:04:53 <Writchie> a workable (sort of) phone on an open platform - what is running is completly in control of user not phone mfg or network. 
Nov 23 18:05:44 <Writchie> in my mind neo is the IBM-PC of phones and the OM Linux is the bios. 
Nov 23 18:06:19 <edistar> om is the bios? hm
Nov 23 18:06:40 <Writchie> right now we have p-code, cpm, and msdos equivalents to extend the metaphor
Nov 23 18:06:41 <dottedmag> OM Linux is the Linux :)
Nov 23 18:07:22 <tuukkah> openbios at that
Nov 23 18:07:24 <Writchie> the role played by the OM Linux is compatibility with the platform. The platform has moved up to FB and above
Nov 23 18:08:17 <CIA-14> 03chris * r3495 10/trunk/src/target/OM-2007.2/applications/openmoko-dialer2/ (6 files in 2 dirs): (log message trimmed)
Nov 23 18:08:17 <CIA-14> * src/phone-kit/Makefile.am:
Nov 23 18:08:17 <CIA-14> * src/phone-kit/moko-dialer-mcc-dc.h:
Nov 23 18:08:17 <CIA-14> Add a header with mappings from MCC to dialing codes
Nov 23 18:08:17 <CIA-14> * src/phone-kit/moko-dialer-dbus.xml:
Nov 23 18:08:19 <CIA-14> * src/phone-kit/moko-dialer.c: (on_network_registered),
Nov 23 18:08:21 <CIA-14> (moko_dialer_finalize), (net_msghandler), (pb_msghandler),
Nov 23 18:08:51 <Writchie> linux (with drivers), + dbus + phone middleware create the new base platform on which you can run gtk, kde, direct fb, java, whatever.
Nov 23 18:09:31 <Writchie> the difference with neo (which is critical) is that is all open and can be fixed (provided the hardware is not broken). 
Nov 23 18:09:55 <Writchie> and eventually someone will may hardware that is not broken. 
Nov 23 18:10:11 <Writchie> on most phones it is the software that is broken - not the hardware
Nov 23 18:15:58 <CIA-14> 03chris * r3496 10/trunk/src/target/OM-2007.2/applications/openmoko-messages2/ (ChangeLog src/sms-compose.c): 
Nov 23 18:15:58 <CIA-14> * src/sms-compose.c: (send_clicked_cb):
Nov 23 18:15:58 <CIA-14> Don't try to get selected contact to send a message, unnecessary
Nov 23 18:20:23 <tuukkah> well said: <Writchie> and eventually someone will may hardware that is not broken.
Nov 23 18:38:17 <mudyc> tuukkah: about AT-command cancelation. i don't know about that there have been plan to add timeout for commands which do not respond. also hang up should be prioritized or use muxed  tty to get faster to the modem hw
Nov 23 18:48:29 <tuukkah> mudyc, the more i learn about AT commands the worse they seem to provide what'd be needed. it's a bit tricky, but ATD sometimes needs to be cancelled when it doesn't return - otherwise you can't cancel a call!
Nov 23 18:49:36 <summatusmentis> is there a spot I can download the sounds that will come w/ openmoko?
Nov 23 18:50:02 <flexd> summatusmentis: i guess you can just download any version of the source, the sounds should be there (?)
Nov 23 18:50:17 <flexd> I'm just guessing, i have no clue sorry :/
Nov 23 18:50:30 <summatusmentis> flexd: any idea if there's a web interface for the source, so I could just get the sounds?
Nov 23 18:50:36 <mudyc> tuukkah: uh. which AT command cancels the previous command? (i have only read the basics of the AT set)
Nov 23 18:50:55 <flexd> summatusmentis: there's a svn or something around here somewhere, i'm not sure
Nov 23 18:51:20 <Writchie> mudyc: anything send while a command is pending should cancel the command
Nov 23 18:51:44 <tuukkah> mudyc, for more see http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=906 and http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=928
Nov 23 18:51:50 <mwester> summatusmentis:  The current audio is in the svn repo
Nov 23 18:52:23 <summatusmentis> mwester: what's the url for that?
Nov 23 18:53:05 <mwester> already downloaded in your build directory, if you use the MokoMakefile:   openmoko/trunk/src/target/OM-2007.2/artwork/sounds 
Nov 23 18:53:25 <summatusmentis> is there an online interface? I don't have openmoko downloaded
Nov 23 18:53:57 <mudyc> Writchie: is this a hw specific or general case of the 07.07?
Nov 23 18:53:57 <mwester> http://svn.openmoko.org/trunk/src/target/OM-2007.2/artwork/sounds
Nov 23 18:54:12 <summatusmentis> thanks
Nov 23 18:54:31 <Writchie> mudyc: general case, pre 07.07 
Nov 23 18:54:47 <Writchie> the CR at the end of the command starts the command processing. 
Nov 23 18:54:54 <balrog-kun> lone "AT" is normally used to interrupt a command because it does nothing
Nov 23 18:54:59 <balrog-kun> (that's what gsmd does too)
Nov 23 18:55:15 <Writchie> balrog-kun: a single CR will do the same
Nov 23 18:55:26 <balrog-kun> Writchie: i see
Nov 23 18:55:48 <Writchie> there can be a very long time, many seconds for example in the case of network registration
Nov 23 18:56:25 <Writchie> unsolicited commands can be presented if enabled as well as intermediate responsed e.g. call progress
Nov 23 18:57:20 <Writchie> command processing mode ends and data mode is entered on data when the CR after a CONNECT is sent by the modem.
Nov 23 18:58:26 <Writchie> the most reliable way to operate is to dedicated a mux channel to unsolicited commands and keep them off the other channels. 
Nov 23 18:58:41 <Writchie> and dedicate a channel for hangups. 
Nov 23 18:59:04 <Writchie> and don't forget that a second call can arrive during a call so you really need to handle call reference values
Nov 23 18:59:14 <tuukkah> an issue with the current gsmd is that there is no handle per command so the cancel command can cancel a later command if the intended one completes unexpectedly
Nov 23 19:00:31 <Writchie> i'm not sure its even possible to have a free-for-all from multiple apps sharing a single AT command channel.
Nov 23 19:00:42 <mwester> tuukkah:  That's one of the major flaws of the existing gsmd that makes a rewrite the most practical approach to moving forward.
Nov 23 19:00:49 <Writchie> amen
Nov 23 19:01:03 <edistar> heh
Nov 23 19:01:24 <tuukkah> now that we have the gsmd2 developer, we should give these requirements to him so he can take care of them :-)
Nov 23 19:01:35 <Writchie> what is gsmd2?
Nov 23 19:01:37 <mudyc> Writchie: did you see the link(s) I gave earlier today?
Nov 23 19:01:52 <Writchie> yes
Nov 23 19:02:00 <tuukkah> Writchie, that's gsmd2
Nov 23 19:02:02 <Writchie> is this stuff for public consumption?
Nov 23 19:02:12 <tuukkah> it is, gpl licensed
Nov 23 19:03:07 <Writchie> it's a good start but there is a lot more dbus definition required and many more use cases. 
Nov 23 19:03:37 <tuukkah> indeed. how to collect the specs?
Nov 23 19:03:47 <mudyc> Writchie: ok, it provides a start point to discussion
Nov 23 19:04:00 <Writchie> mudyc: absolutely, I encourage you to continue
Nov 23 19:04:48 <Writchie> it does, however, required very thorough understanding of gsm specs to design this properly
Nov 23 19:05:01 <Writchie> unformately this is tons of stuff. 
Nov 23 19:05:14 <mudyc> Writchie: could you provide more of the use cases and features. we are willing to provide the daemon for public good
Nov 23 19:05:54 <Writchie> I don't think AT commands belong at all to higher layers except for data modem and fax applications. 
Nov 23 19:06:23 <Writchie> the higher layers IMNSHO should be pure dbus abstrations. 
Nov 23 19:06:50 <mudyc> i agree with that
Nov 23 19:07:25 <Writchie> the lower layers of fully defined by the gsm specs - what we need now is a rather complete definition and abstruction of the upper layers. 
Nov 23 19:07:45 <Writchie> then a gsmd daemon can be designed to implemented to connect the two. 
Nov 23 19:08:09 <anrp> ooh, more layers!
Nov 23 19:08:22 <tuukkah> less
Nov 23 19:08:43 <anrp> D:?
Nov 23 19:08:47 <tuukkah> no AT command multiplexing layer
Nov 23 19:08:48 * anrp adds nothing to this conversation ..
Nov 23 19:08:49 <Writchie> applications should deal with one d-bus layer can this already pickup lots of bindings 
Nov 23 19:09:04 <anrp> ahh
Nov 23 19:09:32 <anrp> fsm-based multi-to-one serial muxing is .. problematic, at best
Nov 23 19:11:21 <Writchie> the neo is both a phone and a terminal which can lead to several issues.
Nov 23 19:12:07 <Writchie> the AP/BP interfaces today are usually multiplexed
Nov 23 19:12:13 * mwester agrees with the d-bus concept, but adds "hacking" to the list of exceptions.
Nov 23 19:12:25 <anrp> ?
Nov 23 19:12:58 <mwester> Experimental stuff before it has gotten far enough along for d-bus -- location awareness, for example.
Nov 23 19:13:00 <Writchie> mwester: it may be necessary to have some "state following" 
Nov 23 19:13:27 <mudyc> Writchie: so, you think that the dbus interface should have fax in additional to modemcontrolling, voicecall and sms?
Nov 23 19:13:35 <Writchie> no
Nov 23 19:13:49 <Writchie> i think d-bus can be used to set up channels
Nov 23 19:14:04 <wirrwarr> huhu
Nov 23 19:14:21 <Writchie> you can use existing fax application on neo or connect through BT
Nov 23 19:15:12 <tuukkah> Writchie, 07.10 channels, you mean?
Nov 23 19:15:17 <Writchie> the fax application just sees a port like bt
Nov 23 19:15:25 <Writchie> yes 07.10 channels 
Nov 23 19:15:34 <tuukkah> ok, just as i thought
Nov 23 19:15:46 <mudyc> Writchie: still confusing: do you mean with channel a unique dbus path /application/1234pid or channel aquired from mux (07.10)?
Nov 23 19:15:49 <Writchie> you can even export a full mux if you want over another channel or BT
Nov 23 19:16:46 <tuukkah> so i was thinking that's something important to add to the use cases and the d-bus interface at an early phase
Nov 23 19:16:48 <Writchie> mudyc: for applications running on the neo they would normally just use d-bus except for legacy apps like data/fax which would just use ports
Nov 23 19:17:07 <tuukkah> pppd would use ports too, right?
Nov 23 19:17:31 <tuukkah> or is that the "data" part already :-)
Nov 23 19:17:33 <Writchie> yes, pppd would be just be used directly on the channel. 
Nov 23 19:18:25 <Writchie> pppd line discplines connects in the kernel and the "app" is just basically for setup and this would be unchanged. 
Nov 23 19:18:45 <Writchie> but you could also export the channel for BT for use with your laptop. 
Nov 23 19:19:03 <mudyc> Writchie: as you seem to understand the issues quite well: why is multiple channels needed (except for fax and ppp)
Nov 23 19:19:56 <mudyc> or is the gsmd just used to open those channels
Nov 23 19:20:19 <Writchie> multiple channels allows you to have an active pppd session for example, and handle a fax or data call at the same time, and manage you sms messages on the SIM card, and do other fiddling with the sim card, etc. all at the same time. 
Nov 23 19:21:09 <Writchie> the major architectural issues is that some chipset allow anything of each mux port and some dedicate certain DLCI's for specific purposes. 
Nov 23 19:21:35 <Writchie> so in the general case you may have some set of allowed functions and each particular DLCI
Nov 23 19:21:46 <Writchie> s/and/on/
Nov 23 19:21:47 <apt> Writchie meant: so in the general case you may have some set of allowed functions on each particular DLCI
Nov 23 19:22:48 <Writchie> BT is another issue and a good look at how the BlueZ folks handle this (especially with dBus) is instructive. 
Nov 23 19:24:25 <tuukkah> BT wrt what?
Nov 23 19:24:55 <tuukkah> something special needed of gsmd for dund?
Nov 23 19:24:57 <Writchie> BT services, for example FAX
Nov 23 19:25:42 <mudyc> Writchie: this sounds as every interface should have it's own command queue and if the hw does not support enough channels then just use fewer
Nov 23 19:26:16 <Writchie> basically yes. 
Nov 23 19:26:43 <Writchie> you still have some contention issues.
Nov 23 19:27:43 <Writchie> d-bus already handles lots of issues - we just need to handle what goes on outside of d-bus. 
Nov 23 19:28:16 <Writchie> this may require some state following/filtering/ on channels exported, for example, through BT. 
Nov 23 19:28:50 <tuukkah> right
Nov 23 19:29:24 <Writchie> There are different views on whether the neo should do all that a phone does. If you want it be a phone, you should do what
Nov 23 19:29:27 <Writchie> phones do. 
Nov 23 19:29:42 <Writchie> That being said, lots of stuff is becoming rather obsolete. 
Nov 23 19:31:15 <tuukkah> something like should you be able to use gammu or just some higher-level syncin solution
Nov 23 19:31:28 <Writchie> yes
Nov 23 19:31:41 <tuukkah> or dund vs. pan
Nov 23 19:31:47 <Writchie> yes
Nov 23 19:32:09 <Writchie> or even a full blown 7.10 mux over BT to your desktop if that's what you want
Nov 23 19:32:40 <tuukkah> good to keep in mind if there are some hard trade-offs to make
Nov 23 19:33:30 <mudyc> i will add these as a feature requests for us
Nov 23 19:34:21 <Writchie> it's probably best to think of AT command processor ports in terms of capabilities, with each port having a set of app capabilities any of which may be possibly busy at any given time. 
Nov 23 19:34:56 <mudyc> btw. if om finds our solution practical enough what scm do you prefer?
Nov 23 19:35:32 <Writchie> mudyc: i am not OM - just another member of community. 
Nov 23 19:36:26 <Writchie> it seems to me, however, that dbus capabilitiy is free to use for any purpose. 
Nov 23 19:36:41 <mudyc> the question was not directed to you as a person :)
Nov 23 19:37:52 <tuukkah> mudyc, perhaps ixonos should speak with openmoko
Nov 23 19:38:12 <balrog-kun> mudyc: if that happens om will probably mirror the code on their svn
Nov 23 19:38:39 <balrog-kun> or at least have a gta01 patchset in the om local svn
Nov 23 19:38:51 <balrog-kun> based on some stable release of gsmd2
Nov 23 19:39:01 <mudyc> balrog-kun: yep. sounds good 
Nov 23 19:39:06 <tuukkah> perhaps git or svn then
Nov 23 19:52:27 <Writchie> mudyc: are you planning to require glib?
Nov 23 19:53:34 <roh> what are you guys engineering?
Nov 23 19:54:07 <mudyc> Writchie: it provides a plenty of stuff, e.g. lexical parser, queue, io_channels, dbus-glib etc. if you can point me a better one then we can drop glib
Nov 23 19:55:08 <roh> glib is nice.. too sad its so fat due to compatibility with rotten libcs like glibc,bsd libc solaris libc and all that incompatible shit with different behaviours ;)
Nov 23 19:55:24 <Writchie> it possible to do d-bus without it and its possible to use embedded glib in order to do the d-bus stuff in a glib like manner. 
Nov 23 19:55:28 <roh> but its cool for platform-independance (glib)
Nov 23 19:56:13 <Writchie> roh: how big on the neo?
Nov 23 19:56:25 <roh> hm.. are you trying to re-do gsmd in a sane way?
Nov 23 19:56:39 <Writchie> yes
Nov 23 19:56:55 <roh> Writchie dunny by hand. but atleast a quarter is only due to compatibility with all that different unix-flavours
Nov 23 19:57:19 <roh> Writchie nice. have you had a look at the phonekit proposal and the source of it?
Nov 23 19:57:42 <Writchie> i looked at it a few months ago and it seemed to be dead in the water
Nov 23 19:57:58 <roh> well.. there is code and the dialer also uses it already afaik
Nov 23 19:59:02 <Writchie> one nice thing about a d-bus approach is that it is usable by any upper layer, whether gtk, kde, java, or roll your own. 
Nov 23 19:59:08 <balrog-kun> roh: yes, the sms app also
Nov 23 19:59:28 <balrog-kun> roh: gsmd2 doesn't have to replace phonekit though
Nov 23 20:00:04 <borg_> so this thing in openmoko-dialer2/src/phone-kit/ is some sort of gsmd replacement?
Nov 23 20:00:17 <ewon> I'd be a big fan of the phone functionality working like the xbox360 dashboard
Nov 23 20:00:19 <roh> thats good. what i see that it needs replacement is gsmd, the rest is fine afaik
Nov 23 20:00:21 <balrog-kun> borg_: not at all
Nov 23 20:00:22 <Writchie> the mobile terminal layer (as I think of it) should be devoid of UI completely
Nov 23 20:00:24 <roh> http://svn.openmoko.org/trunk/src/target/OM-2007.2/applications/openmoko-dialer2/src/phone-kit/
Nov 23 20:00:31 <ewon> e.g. completely independant of the user process
Nov 23 20:00:34 <borg_> i am totally confused about this things :(
Nov 23 20:01:32 <borg_> i looked at gsmd and libgsmd, they dont do dbus at all, right?
Nov 23 20:01:50 <borg_> phone-kit uses libgsmd and does and it seems to do something with dbus
Nov 23 20:01:55 <borg_> right?
Nov 23 20:02:38 <balrog-kun> rihgt
Nov 23 20:02:52 <balrog-kun> phonekit also skips the libmokogsm layer
Nov 23 20:03:12 <borg_> ok
Nov 23 20:03:24 <borg_> but phonekit is intended to work with openmoko-dialer2
Nov 23 20:03:45 <borg_> why does it do dbus, shouldnd gsmd do it?
Nov 23 20:05:10 <balrog-kun> borg_: the whole thing is just siplit into more processes this way, they together do a bigger task
Nov 23 20:05:19 <borg_> i wanted to write some small popup application that shows up when a sms arrives
Nov 23 20:05:32 <balrog-kun> splitting tasks into smaller tasks is always discutible and is always based on some arbitrary choice
Nov 23 20:05:36 <borg_> but i couldnt figure out if i already can use dbus for this :(
Nov 23 20:05:58 <roh> borg_that would be a dbus-listener only with some nice ui in the end
Nov 23 20:06:16 <borg_> roh: yes, but i searched in gmsd and libgsmd for dbus
Nov 23 20:06:20 <borg_> there is nothing :)
Nov 23 20:06:52 <roh> borg_ the whole field of modem-gsmd-libgsmd-phonekit-dialer is currently in flux and hipefully soon in a much better shape.
Nov 23 20:07:18 <borg_> it is totally confused which app is doing what thing
Nov 23 20:07:26 <borg_> one app for one task is good
Nov 23 20:07:32 <borg_> but there is no clear splitting
Nov 23 20:07:36 <borg_> at least i cant see it
Nov 23 20:07:57 <Writchie> roh: the kernel mux will be ready soon and this will permit a saner approach. 
Nov 23 20:08:05 <roh> Writchie ack.
Nov 23 20:08:09 <borg_> i really would like to help, but i cant figure out where :)
Nov 23 20:08:11 <balrog-kun> i think you are supposed to use phonekit for this (through dbus) in the final version but now the design is just shaping up and changing
Nov 23 20:08:21 <balrog-kun> borg_: ^
Nov 23 20:08:22 <roh> Writchie who does that? does he have a device?
Nov 23 20:08:24 <borg_> ok
Nov 23 20:08:48 <Writchie> i am finishing up one written from scratch - yes I have two devices. 
Nov 23 20:09:14 <Writchie> roh: there are a couple of others too - i think. 
Nov 23 20:09:23 <roh> nice. you know the ts07.10 has differen 'dialects' on ezx and neo devices, right?
Nov 23 20:09:40 <balrog-kun> Writchie knows that :)
Nov 23 20:09:45 <mudyc> Writchie: how does the kernel driver tell the device names to the user space?
Nov 23 20:10:03 <mudyc> that is, the mux channels
Nov 23 20:10:20 <Writchie> the mux is a line discpline which is attached to any tty that is the aggregate link for the mux. 
Nov 23 20:10:33 * roh fears that he needs to buy a razr2 v8/9 thingie when we do not build a clamshell soon ;)
Nov 23 20:10:38 <Writchie> the module exports N ttys which are the individual channels for any mux
Nov 23 20:10:52 <roh> mut /me then also needs some nice softwarez
Nov 23 20:11:00 <CIA-14> 03mickey * r3497 10/trunk/src/target/gsm/src/ (5 files in 2 dirs): gsmd: add missing license headers
Nov 23 20:11:03 <Writchie> these have names like muxport1, muxport2, etc. 
Nov 23 20:11:31 <Writchie> an ioctl on the aggregate link binds channels to a mux
Nov 23 20:11:41 <Writchie> you can have as many mux's as you like. 
Nov 23 20:12:49 <balrog-kun> in the ezx implementation and other implementations th dlc's were called /dev/mux0, /dev/mux1 and so on
Nov 23 20:13:02 <balrog-kun> since it became kind of standard it might be good to use the same names
Nov 23 20:13:17 <mudyc> Writchie: however, there's hw limitation of how many channels can be muxed
Nov 23 20:13:21 <Writchie> they can be anything 
Nov 23 20:13:53 <Writchie> mux1 is confusing because its a port not a mux - a mux is normally association with its aggregate link
Nov 23 20:14:35 <Writchie> mudyc: these is a limit of 62 channels per mux by the protocol any particular mux may have a lower limit. 
Nov 23 20:14:51 <balrog-kun> right, but it is what existsing programs will try to use
Nov 23 20:15:17 <Writchie> symlink ;)
Nov 23 20:15:37 <Writchie> its confusing enuf without poor choices for naming things. 
Nov 23 20:16:10 <roh> well.. having more than 2 is luxus ;) but 2 are needed (e.g. for ppd on one and controlling the modem, unsolicited messages on the other)
Nov 23 20:16:12 <Writchie> muxport or muxtty makes it clear that its the port on the mux. 
Nov 23 20:16:15 <balrog-kun> Writchie: do you add a framework like the harald's gsmmux for generic muxing or is there ts0710 code only?
Nov 23 20:16:17 <roh> s/ppd/pppd/
Nov 23 20:16:19 <apt> roh meant: well.. having more than 2 is luxus ;) but 2 are needed (e.g. for pppd on one and controlling the modem, unsolicited messages on the other)
Nov 23 20:17:30 <Writchie> it's a ts0710 mux at the moment, and "advanced" option UIH frames only. 
Nov 23 20:17:40 <Writchie> but it can be extended to other protocols 
Nov 23 20:17:42 <balrog-kun> in my version i added a Kconfig entry for general serial muxing and then the list of available muxers (counting 1) below it
Nov 23 20:17:50 <Writchie> including motorola's basterdized 710
Nov 23 20:18:01 <balrog-kun> i believe that's what laf0rge intended it too look
Nov 23 20:18:57 <roh> well.. if you want, just drop laforge a mail.
Nov 23 20:19:20 <Writchie> the mux protocol is entirely in the ldisc part, so it could be easily separated. 
Nov 23 20:19:20 <roh> he did not stop hacking, just wideing his scope *ducks*
Nov 23 20:20:13 * balrog-kun thinks it should be
Nov 23 20:22:18 <roh> Writchie did i get it right that you want to put it in kernelspace, like this graphics suggests: http://people.openmoko.org/sean/specifications/images/gsm_interface.png ?
Nov 23 20:23:01 <Writchie> yes- i am not in agreement with laforge that it belongs in the kernel
Nov 23 20:23:11 <roh> (ignore the changes to the userspace for now, that graphic is old)
Nov 23 20:23:17 <roh> nice
Nov 23 20:23:24 <Writchie> not inherintly, but because of the pecuiliarities of the linux tty subsystem
Nov 23 20:24:18 <CIA-14> 03mickey * r3498 10/trunk/src/target/gsm/src/util/ (event.c libgsmd-tool.c shell.c): gsmd: fix typos and indentation
Nov 23 20:24:39 <roh> please explain, serial and line-discriplines are not my expertise (but i want to learn)
Nov 23 20:29:38 <Writchie> roh: the tty subsystem has a concept of a line discipline which is attached to a serial port and acts as a kind of filter. 
Nov 23 20:30:41 <Writchie> the default N_TTY discpline implements the basic default. 
Nov 23 20:31:16 <Writchie> there is an interface from the tty layer to line discipline and from line discpline to tty driver. 
Nov 23 20:32:30 <Writchie> N_PPP is a line discpline that implements ppp. 
Nov 23 20:33:12 <Writchie> for the mux, we creat a bunch of virtual terminals and essentially connect them from the trunk. 
Nov 23 20:33:46 <zash> 22kr
Nov 23 20:33:54 <Writchie> so the line discpline operates on the trunk, e.g. as the serial port to the gsm modem or to a pty simulating one
Nov 23 20:34:03 <zash> damn the lag of gprs
Nov 23 20:34:13 <roh> i see.. so connecting one like shown for ppp/gprs is just like a different filter setting
Nov 23 20:34:15 <Writchie> and the mux ports behave like pretty ordinary tty's 
Nov 23 20:34:31 <zash> and remote irssi + screen
Nov 23 20:34:53 <Writchie> yes, you just open the virtual muxport and change the line discipline to ppp and voila! 
Nov 23 20:34:56 <roh> because having a real gprs0 interface that speaks ip on the kernel would be really cool
Nov 23 20:35:00 <Writchie> or open one as a modem
Nov 23 20:35:14 <roh> fax will be funny
Nov 23 20:35:28 <Writchie> roh: yes - but do you have to be careful of routing and the "connection cost" issue. 
Nov 23 20:35:31 <balrog-kun> roh: it would be named ppp0
Nov 23 20:36:21 <Writchie> the routing issue will have to be addressed because in some case the user will want to route certain things over gprs always and
Nov 23 20:36:30 <Writchie> other things through usb, wifi, etc.
Nov 23 20:36:47 <roh> Writchie yes.. but hey.. we still have mr iptables around for asking tricky questions ;)
Nov 23 20:36:59 <Writchie> yes
Nov 23 20:37:35 <roh> i would like policy rounting or similar stuff for that. could work like a crazy routing/personal firewall style exception handlink  thing
Nov 23 20:38:09 <Writchie> roh: there are lot's of policy issues to consider in a complete phone stack. 
Nov 23 20:38:35 <roh> asking for new policy/case by case when something does not fit the stored default (new accesspoint for example , need to set 'just use, its flat or 'only use when i tell so'
Nov 23 20:38:48 * balrog-kun still doesn't know what Writchie finds bad in the muxing on kernel side due to linux tty peculiarities
Nov 23 20:39:19 <balrog-kun> what you described so far is hwo ldisc's work on any OS
Nov 23 20:39:33 <roh> Writchie users should be able to select by app and traffic class. apps should somehow tag if its neccessary for operation or just an extra to get data and if its bulk or just some bits
Nov 23 20:39:41 <Writchie> some OS have arbitrary stacking of drivers
Nov 23 20:40:33 <roh> my ssh client should be able to use any connection, refreshing rss feed-images i only would like on flat high bandwith connections, if none avail ask..  and so on
Nov 23 20:40:43 <Writchie> the linux tty stack grew ad-hoc from the looks of it, ld was originally in the tty_io.c IIRC
Nov 23 20:40:45 <emdete> anyone around here who got gps to work? getting a location from the gps chip? i never got one and don't know why
Nov 23 20:41:11 <roh> emdete you know about gllin?
Nov 23 20:41:20 <emdete> roh: yes, i use that
Nov 23 20:41:35 <balrog-kun> Writchie: yes, although i find that irrelevant
Nov 23 20:41:51 <emdete> i read nmea from it via a python module but the location is still empty
Nov 23 20:42:14 <roh> weird. ive not used it myself, but ive seen it working
Nov 23 20:42:17 <mmontour> emdete: Are you outside with a good view of the sky?
Nov 23 20:42:24 <Writchie> balrog-kun: the gsm stack is really in user space 
Nov 23 20:42:31 <Writchie> conceptually
Nov 23 20:42:41 <anrp> a clash of ideas
Nov 23 20:42:51 <emdete> mmontour: i have a window in the roof here, a garmin 301 is laying there also with lots of satilites
Nov 23 20:43:30 <balrog-kun> Writchie: definitely it can be done in userspace, and it's believed that whatever can be done without too much overhead in userspace belongs there (not only on Linux), but the main argument was precisely this, the overhead
Nov 23 20:43:41 <emdete> i get $GPGSA,A,1 and $GPRMC,,V which means there is no location found
Nov 23 20:44:13 <Writchie> balrog-kun: yes, that's why changed my opinion of where it belongs. 
Nov 23 20:44:28 <balrog-kun> Writchie: ah okay
Nov 23 20:45:36 <roh> when you like crazy 3g stuff.. see this http://www.wireless3g4free.com/
Nov 23 20:45:41 <mudyc> Writchie: so you are implementing it as a module in kernel space 
Nov 23 20:45:52 <Writchie> mudyc: yes
Nov 23 20:46:05 <anrp> has the overhead been measured?
Nov 23 20:47:41 <Writchie> there is very little overhead with the kernel approach because the data is just being moved over from one line discpline on the aggregate link to the other line discpline on the virtual ports.
Nov 23 20:48:02 <balrog-kun> anrp: i don't know, but i trust laforge's prediction on this evenm if it's a complete guess :p
Nov 23 20:48:59 <Writchie> i think laforge agrees there is little difference at todays speeds - its more for 3g and HSDPA in future
Nov 23 20:49:47 <balrog-kun> one thing that is now sure for me is that multiplexing is separate from gsmd
Nov 23 20:50:06 <balrog-kun> ts0710 is used by several non-modem devices that use serial ports in some form
Nov 23 20:50:21 <emdete> do i have to do some `echo 1 > /sys/.../power`?
Nov 23 20:50:22 <balrog-kun> there are usb cameras that use ts0710 for instance
Nov 23 20:50:23 <Writchie> balrog-kun: except that gsmd (or equiv) needs to manage state
Nov 23 20:51:19 <balrog-kun> Writchie: if you mean issuing the AT+CMUX, yes, although AT+CMUX is not part of ts0170 document
Nov 23 20:51:44 <anrp> usb cameras .. ??
Nov 23 20:51:48 <anrp> holy mismatch, batman
Nov 23 20:52:11 <Writchie> if all of the mux ports are used by gsmd there is no problem. When some ports are used outside of gsmd, what is going on in those
Nov 23 20:52:21 <Writchie> ports can effect the mobile terminal as a whole
Nov 23 20:52:27 <mmontour> emdete: I don't have gllin, but I think it does the power-up itself. The power control is in /sys/devices/platform/s3c2410-i2c/i2c-adapter/i2c-0/0-0008/gta01-pm-gps.0/
Nov 23 20:53:05 <balrog-kun> Writchie: what i'm saying is that TS0710 doesn't have to be used with any "mobile terminal" at all, it's a generic multiplexing protocol
Nov 23 20:54:31 <Writchie> balrog-kun: ack - it was designed for gsm but is also used by BT and other stuff
Nov 23 20:55:12 <mudyc> Writchie: I was thinking between the lines of that gsmd could provide a pseudo terminal to "external" applications, i.e. delegate data
Nov 23 20:56:31 <balrog-kun> mudyc: that was the original idea on gsmd-devel list i think
Nov 23 20:57:25 <mudyc> balrog-kun: are there other variants (i may have missed these discussions)
Nov 23 20:58:20 <balrog-kun> mudyc: afaik nope, this idea was first discussed and it was concluded that kernel should do that instead
Nov 23 20:58:47 <balrog-kun> i had missed this discussion and partially implemented the muxing in gsmd with pseudo terminals and the code is still in bugzilla
Nov 23 20:59:20 <CVirus> long time no community updates
Nov 23 20:59:20 <Writchie> balrog-kun: are you the gsm "ports" guy
Nov 23 20:59:43 <Writchie> s/gsm/gsmd/
Nov 23 20:59:44 <apt> Writchie meant: balrog-kun: are you the gsmd "ports" guy
Nov 23 21:00:07 <balrog-kun> Writchie: yes, i sent this patch, i think it's still good :p
Nov 23 21:00:37 <balrog-kun> i mean unrelated to multiplexing 
Nov 23 21:00:52 <Writchie> the idea looked good to me, but the whole of gsmd has looked ugly to me from the begining.
Nov 23 21:01:35 <Writchie> it does take time and lots of discussion to get this part right. 
Nov 23 21:01:45 <balrog-kun> Writchie: this was an attempt to partially clean it up
Nov 23 21:02:17 <Writchie> balrog-kun: ack
Nov 23 21:02:41 <balrog-kun> one of the things i don't like in gsmd is that the mask given to select() is at the same time used as the kindof semaphore which tells you whether a command is executing or not
Nov 23 21:03:04 <balrog-kun> in the current svn it still does it and it's bound to cause problems
Nov 23 21:03:52 * balrog-kun will bbl
Nov 23 21:25:48 <anders_> | /home/andersg/moko/build/tmp/work/armv4t-angstrom-linux-gnueabi/gcc-cross-4.1.2-r11/gcc-4.1.2/libgfortran/mk-kinds-h.sh: Unknown type
Nov 23 21:26:03 <anders_> ..
Nov 23 21:26:03 <anders_> NOTE: package gcc-cross-4.1.2-r11: task do_compile: failed
Nov 23 21:26:08 <anders_> Known issue?
Nov 23 21:32:08 <abraxa_> anders_: Did you check OE bugzilla?
Nov 23 21:34:54 <anders_> abraxa_: Couldn't find it.
Nov 23 21:35:10 <abraxa_> anders_: In that case - file a new one, please :)
Nov 23 21:35:59 <anders_> abraxa_: Not sure if I have broken my environment.
Nov 23 21:40:38 <mwester> anders_:  do you have gfortran installed on your host?
Nov 23 21:41:13 <mwester> if not, try installing that.
Nov 23 21:44:46 * Writchie contemplates writing the mux in Fortran - hmmm - WRITE(6,101) flag, IFLAG, IADDRESS, ICONTROL
Nov 23 21:51:12 <starox_> good evening. I'm trying to make openmoko work with xoo, but dbus stop me. When I launch x-window-manager, I got error message like that :  GConf-WARNING **: Client failed to connect to the D-BUS daemon
Nov 23 21:52:03 <starox_> I made users (avahi,messagebus) for dbus, and I'm sure the daemon is running
Nov 23 21:53:13 <fatal-> what was the kernel side MUX stuff status? does it work at all or not?
Nov 23 21:53:30 <zash> whats up with the muxing?
Nov 23 21:55:22 <Writchie> Writchie: i am working on a new kernel mux implentation from scratch and it should be released for testing soon.
Nov 23 21:55:39 <Writchie> soon = shorter than an openmoko soon ;)
Nov 23 21:56:21 <fatal-> it'd be interesting to test would it work also with our telit stuff. it has proven to be not so co-operative
Nov 23 21:57:23 * balrog-kun wonders if something is less cooperative than the neo1973 modem in this regard
Nov 23 21:57:35 <Dark_Apostrophe> Will the GTA02 have a camera?
Nov 23 21:57:49 <zash> no
Nov 23 21:57:52 <SpeedEvil> Dark_Apostrophe: no
Nov 23 21:57:57 <balrog-kun> Dark_Apostrophe: see topic
Nov 23 21:58:23 <Dark_Apostrophe> Damn.
Nov 23 21:58:44 <Dark_Apostrophe> Are there any other open phones (with a sane price, so the greenphone doesn't count) that will have a camera?
Nov 23 21:59:03 <Writchie> fatal-: telit mux is BASIC option - TI chipset is ADVANCED option. 
Nov 23 21:59:42 <Writchie> i am planning to add BASIC option but it will be after it is working on the neo. 
Nov 23 22:00:09 <Writchie> I eventually plan to add full support for all of TS.07.10. 
Nov 23 22:00:37 <fatal-> Writchie: hmm. perhaps I should finally to bother to read the spec myself too :)
Nov 23 22:00:46 <Writchie> but mototola support will be have to be patched in by somebody else since they don't follow the standard. 
Nov 23 22:02:31 <emdete> mmontour: i got gps to work - want to know how?
Nov 23 22:03:38 <Writchie> balrog-kun: what problems have you had with calypso?
Nov 23 22:04:15 <fatal-> Writchie: do you have a link for the 07.10 spec nearby?
Nov 23 22:04:24 <mmontour> emdete: sure - what was it?
Nov 23 22:04:31 <balrog-kun> Writchie: randomly stopping responding basically, or adding a random 0x00 in a middle of a frame
Nov 23 22:04:37 <emdete> mmontour: i had to switch on gsm... :/
Nov 23 22:04:50 <mmontour> ???
Nov 23 22:04:52 <Writchie> fatal-: http://webapp.etsi.org/key/queryform.asp
Nov 23 22:05:07 <emdete> mmontour: strange is i did it with my gprs script (i deinstalled gsmd). but: when gllin starts pppd stopps
Nov 23 22:05:08 <Writchie> balrog-kun: in mux mode?
Nov 23 22:05:24 <balrog-kun> Writchie: btw, i have the code that detects the supported "option" and can do either basic or advanced frames out of the data you give it, i wonder if someone can reuse it
Nov 23 22:05:28 <balrog-kun> Writchie: yes
Nov 23 22:05:55 <balrog-kun> as well as decoding of frames and retrying on timeouts etc
Nov 23 22:06:01 <torpor> anyone know a solution to the xserver-kdrive issue?
Nov 23 22:06:11 <emdete> mmontour: does that mean 'assisted'-gps?
Nov 23 22:06:23 <balrog-kun> torpor: the build error?
Nov 23 22:06:29 <torpor> yup
Nov 23 22:06:33 <balrog-kun> torpor: yes, there's a solution in bugzilla
Nov 23 22:06:38 <torpor> ok
Nov 23 22:06:54 * balrog-kun tries something..
Nov 23 22:06:57 <balrog-kun> !ombug 1021
Nov 23 22:06:58 <cdbot2> * * Bug 1021, Status: VERIFIED (FIXED), Created: Unknown
Nov 23 22:06:59 <mmontour> I guess it's trying to download LTO data or something. There should be a command-line flag to tell gllin to run standalone, though.
Nov 23 22:06:59 <cdbot2> * * yoyo(AT)ghost.pl: xserver-kdrive
Nov 23 22:07:01 <cdbot2> * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=1021
Nov 23 22:07:25 <borg_> NOTE: package neod-0.1.0+svnr3421-r5: task do_unpack: started
Nov 23 22:07:25 <borg_> NOTE: Unpacking /data/openmoko2/sources/neod_svn.openmoko.org_.trunk.src.target.OM-2007.2.daemons_3421_.tar.gz to /data/openmoko2/build/tmp/work/neo1973-angstrom-linux-gnueabi/neod-0.1.0+svnr3421-r5/
Nov 23 22:07:29 <borg_> mh
Nov 23 22:07:48 <borg_> why does it use revision 3421 when 3498 is the newest
Nov 23 22:07:49 <borg_> ?
Nov 23 22:08:27 <CM> Because of locked scr_revs
Nov 23 22:08:37 <torpor> balrog-kun: thanks..
Nov 23 22:08:46 <borg_> CM: if i want to write a patch for it
Nov 23 22:08:50 <borg_> how do i do it?
Nov 23 22:09:13 <borg_> i modified the latest revision
Nov 23 22:09:15 <borg_> made a patch
Nov 23 22:09:28 <borg_> copied it to openembedded/packages/openmoko2/neod
Nov 23 22:09:35 <fatal-> Writchie: ok, thx
Nov 23 22:09:35 <borg_> and made a rebuild
Nov 23 22:09:45 <borg_> but then it doesnt work because of the wrong version
Nov 23 22:09:53 <CM> borg_: Not sure, maybe edit openmoko/trunk/oe/conf/distro/include/preferred-openmoko-versions.inc to get the latest?
Nov 23 22:09:54 <borg_> how is the workflow to do such a thing?
Nov 23 22:09:58 <borg_> ah ok
Nov 23 22:10:01 <CM> I'm not really familiar with it either..
Nov 23 22:14:37 <emdete> mmontour: it's not using gprs, it seems to use gsm somehow...
Nov 23 22:16:30 <torpor> i wish i could use gprs
Nov 23 22:20:13 <emdete> torpor: where is your problem?
Nov 23 22:21:52 <CIA-14> 03andrew * r3499 10/trunk/src/host/qemu-neo1973/openmoko/env: 
Nov 23 22:21:52 <CIA-14> Ensure every SIGALRM causes a cpu_loop_exit().
Nov 23 22:21:52 <CIA-14> Take back the forcing of clocksource, the dynticks should now be functional.
Nov 23 22:25:58 <CIA-14> 03andrew * r3500 10/trunk/src/host/qemu-neo1973/ (cpu-all.h cpu-exec.c exec.c vl.c): Changes missing from previous commit due to svn vs. git mutual hate.
Nov 23 22:27:01 <torpor> emdete: with kdrive-xserve, but i've done the details described in the bugzilla entry and it seems to have moved on ..
Nov 23 22:27:12 <mmontour> emdete: I guess you could 'strace -x -s9999' gllin to see what it's doing. I've never seen it try to open /dev/ttySAC0 (GSM serial port), and I'm not sure how else it would talk to the GSM
Nov 23 22:27:28 <anders_> mwester: Installing gfortran on host seemed to do the trick..
Nov 23 22:28:46 <mmontour> lol - a gfortran dependency is somehow appropriate for the "neo1973" phone.
Nov 23 22:30:04 <anders_> Finally an explaination for "1973"
Nov 23 22:32:09 <zash> the 70's called, they want their fortresses back .. or something ;)
Nov 23 22:32:09 <torpor> n000
Nov 23 22:35:13 <Ghiottone> hi all
Nov 23 22:37:15 <emdete> torpor: kdrive & gprs? how that?
Nov 23 22:38:19 <emdete> mmontour: yes, that seams resonable, but i got tcsetattr errors with pppd when gllin startet to show position. i could not reproduce that again... will investigate further
Nov 23 22:44:12 <emdete> mmontour: there are many option to gllin that seem be related to gsm (like -gsm_cell, -msisdn, -nai (Network Access Identifier), -freq FRQ_PLAN_GSM_R26B13_TRK)
Nov 23 22:47:04 <abraxa_> emdete: I thought the hammerhead is an AGPS chip - would make sense that it tries talking to the GSM under certain conditions
Nov 23 22:47:27 <PBeck> hm have anybody screenshots from the openmoko-messages2 application?
Nov 23 22:47:32 <PBeck> *had
Nov 23 22:48:00 <emdete> abraxa_: :D i still have not understand Agps...
Nov 23 22:48:21 <CM> PBeck: Haven't seen any yet, but I'm sure there'll be a blog post about it soon :)
Nov 23 22:48:37 <emdete> abraxa_: but how does it talk to gsm? is there another chanel in hw?
Nov 23 22:48:52 <abraxa_> No clue
Nov 23 22:49:01 <PBeck> CM: is the application relativ new?
Nov 23 22:49:16 <PBeck> +e
Nov 23 22:50:01 <SpeedEvil> abraxa_: no - it soedn't
Nov 23 22:50:11 <SpeedEvil> abraxa_: that's basically in software
Nov 23 22:50:34 <SpeedEvil> emdete: http://en.wikipedia.org/wiki/A-GPS
Nov 23 22:51:00 <abraxa_> SpeedEvil: ...meaning gllin, which might have AGPS features built in that we just don't know of
Nov 23 22:51:03 <CM> PBeck: Well, been a lot of check ins to it today, last one 4.5h ago
Nov 23 22:51:25 <PBeck> ah yes, cool :D
Nov 23 22:51:28 <SpeedEvil> abraxa_: some of the switches visible with -h say something about it AIUI
Nov 23 22:51:40 <PBeck> CM: is this application for sms and email?
Nov 23 22:51:54 <abraxa_> SpeedEvil: Hmm, maybe emdete can check them out then
Nov 23 22:51:56 <CM> I don't know, but I'm guessing mainly sms right now
Nov 23 22:52:06 <PBeck> ok
Nov 23 22:52:39 <PBeck> so i musst go now ;) Bye. CM thanks for your informations
Nov 23 22:52:45 <CM> Bye :)
Nov 23 22:53:59 <hhf423> ssh phone
Nov 23 22:54:39 <mwester> ssh: phone: Name or service not known
Nov 23 22:54:41 <hhf423> hmm
Nov 23 22:54:46 <hhf423> heh
Nov 23 22:54:56 <borg_> :)
Nov 23 22:55:03 <hhf423> the nov21 snapshot does not ask for pin
Nov 23 22:55:05 <Vegar> mwester: you should've asked for the password
Nov 23 22:55:21 <mwester> Vegar:  Aaggh!  A golden opportunity wasted!
Nov 23 22:55:24 <hhf423> and I can not reboot from GUI, just shutdown. what a brilliant improvement
Nov 23 22:55:46 <hhf423> ah, oh yeah, my Neo password
Nov 23 22:55:47 <mwester> You like that, huh?  I noticed that as well.
Nov 23 22:55:52 <emdete> SpeedEvil: thnx, yes, i read that. but i am not shure what that all means. which layer is used to get the data? internet/gprs? something on gsm layer? what happens if that is not possible? will satelite data be used?
Nov 23 22:56:09 <hhf423> mwester it shucks
Nov 23 22:56:26 <mwester> And that after the argument with the GSM OM folks about how the neo should never be turned off, but if it gets screwed up, it should be restarted.... How ironic ;)
Nov 23 22:56:33 <SpeedEvil> emdete: in this case - at the moment, it's not actually 'A'
Nov 23 22:56:51 <SpeedEvil> emdete: it's just GPS only
Nov 23 22:57:00 <balrog-kun> we don't need no stinking reboot!
Nov 23 22:57:08 <hhf423> mwester I agree
Nov 23 22:57:25 <hhf423> but its the Unix way!
Nov 23 22:57:31 <hhf423> if it does not work: reboot!
Nov 23 22:57:45 <balrog-kun> is it?
Nov 23 22:57:53 <hhf423> thats why Sun got a foothold in the Unix world: because Sun's boot fast!
Nov 23 22:58:05 <mwester> No - my point was that if people got frustrated enough, they would power off the phone and use another.  OM thinks they will just reboot the phone.
Nov 23 22:58:09 <hhf423> if you wanted to see a Sun at its best: boot it!
Nov 23 22:58:22 <hhf423> (phun intended)
Nov 23 22:59:13 <hhf423> well I know from previous images that the asking for PIN sometimes comes after a few reboots...
Nov 23 22:59:56 <hhf423> how do I query again if GSM is up via command line?
Nov 23 22:59:56 <mwester> And after all that arguing with OM, we STILL do not have the code to power off the GSM modem *anywhere* in the device.  Sigh.
Nov 23 23:00:18 <SpeedEvil> mwester: nothing sends at@poff?
Nov 23 23:00:28 <mwester> Nope.
Nov 23 23:00:34 <SpeedEvil> ...
Nov 23 23:00:36 <hhf423> and we're still waiting for Google to provide an Android build :-)
Nov 23 23:01:20 <anrp> who would want to power off the modem? ;)
Nov 23 23:01:51 <emdete> SpeedEvil: if it /would/ show a position :/
Nov 23 23:01:56 <balrog-kun> right, you could run SETI@home on it when you're not using it
Nov 23 23:01:57 <mwester> SpeedEvil:  Somebody added code to gsmd so that it could power off the modem (which I believe is insufficient), but there's still nowhere that it's actually invoked.  So when you power off your phone, the GSM modem is still happily running, and sucking your battery dry. 
Nov 23 23:02:09 <balrog-kun> on the modem that is
Nov 23 23:02:43 <mwester> anrp:  It's a nice feature to keep the battery from getting sucked dry if *everything* powers off when you shutdown your phone ;)
Nov 23 23:03:48 <hhf423> init 0
Nov 23 23:03:50 <hhf423> oops
Nov 23 23:04:00 * mwester powers off
Nov 23 23:04:00 <anrp> init: command not found
Nov 23 23:04:55 <hhf423> mwester so thats the reason for the battery issue? pathetic
Nov 23 23:05:40 <hhf423> but in any case, the media player now works all the time
Nov 23 23:05:54 <hhf423> of course it sometimes skips a split second
Nov 23 23:06:11 <Vegar> some audio thing was also causing the battery issue, hhf423
Nov 23 23:06:18 <mwester> hhf423:  That's not the sole reason.  The audio device that was also sucking power has been fixed in the kernel, and there's still the whole power management issue when it's running.
Nov 23 23:06:37 <hhf423> audio? how bizarre
Nov 23 23:07:04 <hhf423> there was some audio chip still counting sheep or something? 
Nov 23 23:07:15 <hhf423> busy phone :-)
Nov 23 23:07:16 <mwester> So with recent kernels, and a shutdown script that sends AT@POFF to the GSM modem, I've reliably been able to shut down the phone and have it actually draw no current.  :)
Nov 23 23:07:38 <hhf423> yay
Nov 23 23:08:02 <hhf423> so how long do you think will take this to get into the official builds?
Nov 23 23:08:06 <Writchie> mwester: did you actually measure it at the battery, - it should still draw tens of microamps AFAIK
Nov 23 23:08:10 <mwester> hhf423:  It turns out that there are two devices AFAIK that are "upstream" from the power control chip -- the GSM modem, and the audio device.  They need to be turned off separately from the phone itself, or they continue to draw power.
Nov 23 23:08:41 <hhf423> mwester amazing
Nov 23 23:09:09 <mwester> Writchie:  Yes, it draws some small amount - I didn't change the scale to microamps - but I expect it needs some power to be able to sense the USB insertion (for charge) and of course the power button itself.
Nov 23 23:09:37 <mwester> hhf423:  That might actually be a feature, I guess. 
Nov 23 23:10:27 <Writchie> mwester: have you got anywhere with suspend/resume?
Nov 23 23:10:43 <mwester> hhf423:  How long it takes will depend on many factors, the most significant of which is "openmoko time", which seems to only be loosely related to our dimensions timescales.  :D
Nov 23 23:10:47 <hhf423> ah, qtopia asks for PIN
Nov 23 23:10:59 <mwester> Writchie:  I haven't looked at suspend/resume yet (been busy on a SlugOS release!)
Nov 23 23:49:14 <mjr> bah, it appears Asus screwed up with the Eep: http://cliffhacks.blogspot.com/2007/11/asus-eeepc-first-impressions-and-gpl.html
Nov 24 00:00:35 <nbd> mjr: i can understand why asus is not distributing the madwifi sources
Nov 24 00:00:55 <nbd> that would be a tricky thing to do
Nov 24 00:04:46 <Vegar> mjr: disappointing
Nov 24 00:05:17 <nbd> mjr: iirc asus is using atheros' own version of madwifi, which is a tangled mess of proprietary and free code
Nov 24 00:05:23 <nbd> it isn't just a patch against the madwifi.org version
Nov 24 00:05:52 <egn> on the openmoko platform, is it possible to have access to the "press x key" to do something functions like voicemail/whatever, or is that handled in the carriers services?
Nov 24 00:07:00 <mjr> Whaever, my interest in the Eep just died. Maybe if they change the wifi or the rev-eng driver gets up to speed _and_ supporting that platform's particular kinks (yeah not holding breath), _and_ they get that source business cleared out, esp. wrt. acpi
Nov 24 00:07:33 <Vegar> egn: press/hold a key to check voicemail or press a key to redirect a call to voicemail?
Nov 24 00:07:38 <mmontour> egn: voicemail should work. With my carrier you just have to dial a special phone number. I'm not sure about some of the other special features
Nov 24 00:07:42 <nbd> mjr: reverse engineering work for that particular chip is in progress
Nov 24 00:08:08 <egn> Vegar: mmontour: i mean like say someone calls me, can i program something on the openmoko side to say "press 5 to do this.. press 1 to do this"
Nov 24 00:08:11 <mjr> nbd, ack.
Nov 24 00:08:23 * mjr is waiting for an XO for a mini-lappie-toy anyway
Nov 24 00:08:25 <Vegar> egn: ah, PBX-like
Nov 24 00:08:58 <mjr> and yeah, voicemail afaik generally is just your own special phone number, at least here
Nov 24 00:09:01 <mmontour> mjr: me too
Nov 24 00:09:05 <egn> Vegar: yeah i believe so
Nov 24 00:09:08 <Vegar> egn: it might be possible with software like asterisk, but I'm not sure how well it would work
Nov 24 00:09:25 <mjr> egn, yeah, it's possible
Nov 24 00:09:33 <mjr> or should be anyway
Nov 24 00:11:10 <egn> mjr: Vegar: hm alright thanks
Nov 24 00:11:41 <egn> any idea if it would it be implimented as like an openmoko function or something?
Nov 24 00:11:44 <egn> it would be*
Nov 24 00:12:26 <mjr> you basically just route the audio to the SoC, play prerecorded messages to the GSM chip, detect the DTMF frequencies associated with the buttons...
Nov 24 00:12:45 <mjr> I'm not an OM guy but I'd guess that'd be "low priority"
Nov 24 00:12:54 <abraxa_> lol mjr
Nov 24 00:13:23 <egn> hmm alrighty.. sounds kinda scary
Nov 24 00:13:46 <abraxa_> egn: Not really
Nov 24 00:14:18 <egn> abraxa_: okay, well as long as its possible
Nov 24 00:14:31 <abraxa_> Asterisk would indeed be one way to do it, just dialer integration might be... tricky
Nov 24 00:14:42 <egn> hmm okay
Nov 24 00:15:25 <egn> basically i'm looking to make "dont bother me" away messages, ie. "hey, i'm sleeping, if you really need to reach me, press 5, if not, leave a message"
Nov 24 00:15:30 <egn> anything close to that already?
Nov 24 00:16:05 <abraxa_> egn: A "do not disturb" mode seems quite doable to me, though not high priority indeed
Nov 24 00:16:18 <abraxa_> It could just be one of the phone profiles 
Nov 24 00:16:21 <egn> abraxa_: okay sweet
Nov 24 00:16:53 <egn> thanks for the help, may be on once i get a neo...
Nov 24 00:17:33 <mmontour> egn: A quicker hack might be to look for 3 or more calls from the same caller-ID within a short time period (with the ringer off by default)
Nov 24 00:17:58 <egn> mmountour: ah thats not a bad idea
Nov 24 00:18:09 <egn> and if three come through, enable ringer
Nov 24 00:18:35 <Writchie> the feature used to be called "reject a jerk"
Nov 24 00:19:03 <Writchie> on some networks you can redirect an incoming call to another number on the fly
Nov 24 00:19:15 <Writchie> redirect = divert
Nov 24 00:20:38 <Writchie> there are some calls i'd like to just divert to the FCC
Nov 24 00:20:55 <Writchie> but alas t-mobile doesn't support this feature on their US network
Nov 24 00:22:14 <webjames> hi, did openmoko just go down?
Nov 24 00:22:29 <webjames> as in wiki.openmoko.org openmoko.com
Nov 24 00:22:40 <webjames> buildhost.openmoko.org etc..
Nov 24 00:22:57 <Vegar> nope
Nov 24 00:23:02 <Writchie> not from here
Nov 24 00:23:11 <webjames> really? hmm..
Nov 24 00:23:34 <webjames> i'm getting nothing from them, but the rest of the internet works fine.
Nov 24 00:24:02 <Vegar> could it be a dns issue?
Nov 24 00:24:25 <webjames> no idea, are you saying it's my end?
Nov 24 00:25:06 <webjames> hmm ebay.co.uk doesn't work either, but gmail.com does
Nov 24 00:25:57 <Vegar> webjames: must be
Nov 24 00:26:10 <webjames> how can i sort this?
Nov 24 00:26:54 <Vegar> check if the domain names in question are resolving
Nov 24 00:29:10 <Writchie> webjames: ping wiki.openmoko.org  then traceroute wiki.openmoko.org
Nov 24 00:30:39 <webjames> seems to be working now. weird
Nov 24 00:30:48 <webjames> thanks Vegar and Writchie 
Nov 24 02:13:39 <b0ef> ehlo
Nov 24 02:14:02 <b0ef> I just read that the gpsd for the neo requires proprietary software; is this still true and is someone working on a free driver?
Nov 24 02:18:56 <SpeedEvil> yes for GTA01
Nov 24 02:19:16 <SpeedEvil> I and some other people were, however that's stopped largely as GTA02 has different GPS chip in
Nov 24 02:22:34 <b0ef> SpeedEvil: oh, excellent;) 
Nov 24 02:22:41 <b0ef> I only care about being able to make calls without proprietary software, though 
Nov 24 02:23:06 <b0ef> I could live without GPS if it was only that that used proprietary software
Nov 24 02:24:59 <SpeedEvil> a chip which outputs NMEA, so requires no non-free software to use.
Nov 24 02:25:16 <SpeedEvil> In a way it's a pity - the original one is more flexible.
Nov 24 02:25:18 <SpeedEvil> but...
Nov 24 02:26:53 <b0ef> what format do you get from the new chip, then?
Nov 24 02:27:55 <SpeedEvil> NMEA
Nov 24 02:28:44 <b0ef> ah
**** ENDING LOGGING AT Sat Nov 24 02:59:57 2007