**** BEGIN LOGGING AT Sat Aug 01 02:59:58 2009 Aug 01 06:14:26 DocScrutinizer: hehe. Aug 01 06:14:53 DocScrutinizer: bitbake already has a method to make a dependency graph one can visualise with graphvis Aug 01 06:22:05 PaulFertser: hallo..... X e17 runs in SHR-gta01 ....but without t-screen .....http://pastebin.org/5691 Aug 01 06:22:13 PaulFertser:good morning... Aug 01 06:22:20 PaulFertser: :-) Aug 01 06:22:20 schasch: morning Aug 01 06:22:34 schasch: pastebin your dmesg output Aug 01 06:22:40 schasch: and ls /dev/input Aug 01 06:24:27 PaulFertser: http://pastebin.org/5693 Aug 01 06:25:17 http://pastebin.org/5695 Aug 01 06:25:41 schasch: /dev/input/ Aug 01 06:25:47 Ok Aug 01 06:26:14 So both mice and touchscreen0 are there... Aug 01 06:28:13 schasch: what about dmesg? Aug 01 06:29:22 PaulFertser: under the first in http://pastebin.org/5693 Aug 01 06:29:44 Ah :D Aug 01 06:29:47 Just woke up Aug 01 06:30:11 PaulFertser:OK... :-) Aug 01 06:31:59 schasch: cat /sys/class/input/input1/name Aug 01 06:32:55 PaulFertser: s3c2410 TouchScreen Aug 01 06:33:16 schasch: ok, if you do cat /dev/input/event1 and then touch TS do you see any output? Aug 01 06:34:04 PaulFertser: YES Aug 01 06:34:48 schasch: then stop X completely. Start only "strace Xfbdev" Aug 01 06:36:17 PaulFertser:...done... Aug 01 06:36:49 PaulFertser:see an x .... no action on TS-actions... Aug 01 06:37:29 PaulFertser:mouse x not moving... Aug 01 06:37:38 schasch: strace output is what is interesting here Aug 01 06:41:37 PaulFertser. http://pastebin.org/5698 Aug 01 06:45:32 PaulFertser: added a echo "2:2345:respawn:/sbin/getty 38400 SAC0" >> ${IMAGE_ROOTFS}/etc/inittab ....hope it is no problem... Aug 01 06:45:56 PaulFertser:...or causing th problem... Aug 01 06:47:09 schasch: should be ok Aug 01 06:49:21 schasch: your strace output doesn't say anything about mice or TS. Aug 01 06:50:00 PaulFertser:hmm.... Aug 01 06:55:07 PaulFertser:and this one http://pastebin.org/5700 Aug 01 06:56:36 schasch: looks the same (and truncated btw Aug 01 06:57:37 PaulFertser, schasch: build from scratch went bad too :( Aug 01 06:59:59 mrmoku:good morning! Aug 01 07:00:26 mrmoku: do you mean the TS-bug? Aug 01 07:00:44 schasch, morning... the whole image Aug 01 07:00:46 evas stuff missing Aug 01 07:00:53 frameworkd-config-shr missing Aug 01 07:00:57 mrmoku`: i had a (very) quick look at bitbake and it seems that RDEPENDS and RRECOMMENDS should be the same... I still don't get the idea. Probably you should ask OE folks. Aug 01 07:01:07 fsousaged missing Aug 01 07:02:38 mrmoku: this patch fro PaulFertser helped http://pastebin.org/5701 building an image with X-e17 Aug 01 07:03:13 mrmoku: without TS :-( Aug 01 07:04:38 schasch, does frameworkd start? Aug 01 07:06:15 mrmoku:root@om-gta01 ~ $ ps aux | grep work -> root 2042 0.0 0.4 3328 532 pts/0 S+ 08:47 0:00 grep work ...NO? Aug 01 07:07:12 schasch, looks like no Aug 01 07:07:38 mdbus -s shows what is listening on the d-bus Aug 01 07:07:54 anyway... got to bring my mum her computer back now... Aug 01 07:07:56 bbl Aug 01 07:08:51 mrmoku|away: http://pastebin.org/5703 Aug 01 07:21:32 schasch: i can't understand why without strace it writes some messages about TS you mention and with strace it doesn't. Aug 01 07:24:07 PaulFertser: schasch: i can't understand why without strace it writes some messages about ....you mean the cat /dev/input/event1 -watching? Aug 01 07:24:37 schasch: no, i mean Couldn't find pointer driver /dev/input/mice, [tslib/TslibEnable] failed to open /dev/input/touchscreen0 Aug 01 07:27:25 PaulFertser: I do a strace Xfbdev &> /tmp/x_strace3.log Aug 01 07:28:54 schasch: if you start Xfbdev without strace do you see those messages? Aug 01 07:34:05 PaulFertser. http://pastebin.org/5705 Aug 01 07:35:15 PaulFertser: NEW http://pastebin.org/5706 Aug 01 07:36:48 schasch: do you have tscal? Aug 01 07:38:14 PaulFertser: in rootfs? or what do you mean? Aug 01 07:39:16 schasch: try to "export TSLIB_TSDEVICE=/dev/input/event1" Aug 01 07:39:20 schasch: and then Xfbdev Aug 01 07:41:23 schasch: if you don't have this file yet, do "echo -67 36365 -2733100 -48253 -310 45219816 65536 > /etc/pointercal" Aug 01 07:42:22 PaulFertser: export..... http://pastebin.org/5707 Aug 01 07:43:56 schasch: try to create /etc/pointercal and start x the usual method (nodm) Aug 01 07:47:20 PaulFertser. NO TS : more /etc/pointercal and ps aux >>> http://pastebin.org/5709 Aug 01 07:49:18 schasch: so, looks like specifically tslib is not working, for whatever reason... Probably some missing dependency too, but no idea which one. Aug 01 07:50:18 schasch: install tslib-conf package Aug 01 08:03:06 PaulFertser: ipk copied by scp ....opkg...ERROR.... http://pastebin.org/5710 Aug 01 08:08:47 schasch: well, install the config file somehow.. Aug 01 08:13:40 PaulFertser:you're GREAT opkg-cl install -force-depends -force-reinstall http://build.shr-project.org/shr-unstable/ipk/armv4t/tslib-conf_1.0-r20_armv4t.ipk TS is working..... Aug 01 08:15:06 PaulFertser:thanks! is it possible to update my build-host too? *.bb? Aug 01 08:15:46 schasch: sorry, gtg Aug 01 08:15:56 schasch: i think mrmoku should ask OE folks about RRECOMMENDS Aug 01 08:16:14 schasch: probably some bitbake change altered semantics... Though it doesn't look like that. Aug 01 08:18:30 PaulFertser:OK...i will ask mrmoku... Aug 01 08:18:39 schasch: and yes, i _am_ great. Aug 01 08:18:41 :D Aug 01 08:19:07 :-)...can't say anything else... :-) Aug 01 08:20:00 PaulFertser:isn't it possible to FXI my own build by installing apk/ipk manually... Aug 01 08:20:10 PaulFertser:...and my own apps too? Aug 01 08:20:27 s/FXI/FIX Aug 01 08:48:32 <\marco> hi to all Aug 01 08:48:55 <\marco> I'm looking at the version of my gsm firmware Aug 01 08:49:03 <\marco> and... it's old :D Aug 01 08:49:17 <\marco> Moko9-beta1 Aug 01 08:49:48 <\marco> now.. the wiki suggest to use uSD-card method Aug 01 08:50:12 <\marco> but " for flashing MOKO11 to GTA02 Freerunner only" Aug 01 08:50:45 <\marco> oh... Aug 01 08:50:48 <\marco> sorry.. Aug 01 08:50:57 <\marco> for flashing Moko11 Aug 01 08:51:01 <\marco> lol Aug 01 08:51:19 <\marco> delete all I've said :P Aug 01 08:51:28 xD Aug 01 09:02:47 I *really* *really* wonder how moko9beta can be found on so many devices, though I *clearly* stated it's deprecated to use this version Aug 01 09:03:09 <\marco> :D Aug 01 09:30:28 DocScrutinizer: people like living on the edge ? :p Aug 01 09:30:40 maybe they need something tangible to blaim for all the FR problems :) Aug 01 09:31:38 hmm, it seems it's either resellers or our own fab that used moko9 Aug 01 09:32:47 <\marco> upgraded to moko11 :) Aug 01 09:32:49 *brag* Zorkman has 10k+ cells on openbmap, I think I'm expecting my certificate real soon from onen :) Aug 01 09:33:47 <\marco> it's very simple with uSD-method.. congratulation with the creator :) Aug 01 09:34:03 <\marco> *with=to Aug 01 09:41:14 \marco: (11) good boy ;-) Aug 01 09:41:34 <\marco> hehehe :) Aug 01 09:41:51 and thabks for the congrats above :D Aug 01 09:51:00 \marco: you think the wikipage is ok (gsm-flashing)? Aug 01 09:53:10 <\marco> DocScrutinizer: yes, I do, the wikipage is clear and simple :) absolutely no problem :) Aug 01 09:53:23 fine. thanks Aug 01 09:53:47 <\marco> ^^ you're welcome Aug 01 10:17:48 DocScrutinizer: I also agreed, it's a very foolproof way and instructions are clear Aug 01 10:18:54 do you think it's commonly known to be recommended to update gsm-fw? Aug 01 10:19:22 DocScrutinizer: *I* think so Aug 01 10:19:33 isn't it also stated in the "getting started" on the wiki? Aug 01 10:19:35 (i'll check) Aug 01 10:20:01 dunno. it should, I'd say Aug 01 10:20:15 hmmm it isn't... Aug 01 10:20:30 still has a lot of 2007.2 stuff also... Aug 01 10:21:21 yup... annoying Aug 01 10:21:23 it's mentionned here: http://wiki.openmoko.org/wiki/Freerunner_Hardware_Issues#Some_SIMs_Don.27t_Work but that is a step too far for a new user Aug 01 10:21:33 I think it should be stated in the getting started... Aug 01 10:21:52 me too Aug 01 10:24:58 PaulFertser: reading buzzword "dos1" -- how can PMU AD data be read out? E.g. result of magic-R test Aug 01 10:26:29 PaulFertser: maybe a task for our kernel guy to supply a few missing sysfs nodes to access that? ;) Aug 01 10:32:07 * DocScrutinizer pondering about rationale behind rejecting AT@POFF from kernel space, but OTOH (apparently) not even exposing *any* means to read out actual AD-value in userspace. Seems insane Aug 01 10:33:23 PaulFertser: maybe we can find some pmu_reg-dump sysfs node ? Aug 01 10:35:03 ... which allows to write new register values as well. otherwise it's probably still difficult to use PMU-ADC from userspace Aug 01 10:49:23 BATS!! No dump related to PCF50633?!!!? Aug 01 10:50:33 PaulFertser: maybe go direct I2C-bus IO? Aug 01 10:51:32 nasty, annoying, and stupid... :-/ Aug 01 11:01:33 Ainulindale, ping... still headache? ;) Aug 01 11:15:41 Qt4.5 is in oe.dev Aug 01 11:28:04 wow :-P Aug 01 11:28:49 nah, honestly Qt isn't that bad. Greeeeedy but rather convenient to code with it Aug 01 11:29:12 (my 2€cent ;) Aug 01 11:30:03 DocScrutinizer: I'm writing qt apps at work now. Some things are made there like pain in the ass Aug 01 11:30:33 i'm currently writing UI in javascript Aug 01 11:30:37 THAT is a pain in the ass Aug 01 11:31:22 mrmoku: there are new commits in oe.dev regarding policykit and packagekit Aug 01 11:32:44 mrmoku: will we have qt in shr repo? both dev and libs Aug 01 11:34:06 Q-Master: after merging with oe.dev we'll have qt 4.5.2 Aug 01 11:34:38 dos1: sounds good. some qt apps are flawless Aug 01 11:37:06 mrmoku: please cherry-pick a1adb7f8b667756434bfe5082b28fb90ad4205a6 Aug 01 11:41:04 dos1, nah Aug 01 11:41:54 Q-Master: depends on what you do Aug 01 11:41:58 some things really rock with qt Aug 01 11:42:25 bumbl: complicated UIs in qt are pain in the ass Aug 01 11:42:52 Q-Master: what about Qt Designer? Aug 01 11:43:19 dos1: what about it? it suxx. 8) Aug 01 11:43:41 but e17 lacks even this Aug 01 11:44:21 there is some edje-creator, but i don't know how it works Aug 01 11:44:27 anyway, bbl Aug 01 11:45:40 edje-creator is much stranger than qt-designer Aug 01 11:46:04 DocScrutinizer, /me hates qt :) Aug 01 11:46:34 though that's probably because, that's how I was raised. Aug 01 11:48:04 TAsn: some qt apps roxx Aug 01 11:48:35 k3b for instance is better than all the alternatives Aug 01 11:48:38 I know. Aug 01 11:48:47 but still, gtk is nice :) Aug 01 11:48:49 I like gtk. Aug 01 11:49:05 anyhow, shower. Aug 01 11:49:05 :) Aug 01 11:51:07 to me qt-designer seems superior to E* editing plain text edj files ;-) Aug 01 11:52:00 but nevermind my mumbling, I neither used one or the other extensively Aug 01 11:53:37 TAsn: PSI is a great jabber client, much superior than others Aug 01 11:53:42 also the signals/slots concept feels rather natural to me, and there's a couple of other nice data/obj-types Aug 01 11:53:51 DocScrutinizer: yep. editing edjs suxx Aug 01 11:54:02 baseballs ;-P Aug 01 11:54:25 DocScrutinizer: even when you don't understand what exactly does some text mean Aug 01 11:54:27 8) Aug 01 12:01:50 hmm inkscape crashes quite often Aug 01 12:05:36 bumbl, on the moko? Aug 01 12:11:37 * dos1 don't like GTK+ :P Aug 01 12:12:15 but 2007.2 GTK apps > 2008.8 ported to X11 Qtopia apps :P Aug 01 12:12:34 but generally Qt > GTK+, as KDE > GNOME ;D Aug 01 12:14:57 mrmoku: nah for the logo competition Aug 01 12:18:35 dos1, you logic has flaws Aug 01 12:18:38 for instance: Aug 01 12:18:40 GNOME > KDE Aug 01 12:18:50 not the other way around. Aug 01 12:18:59 furthermore, gtk+ > Qt. Aug 01 12:19:10 of course KDE is better than GNOME :P Aug 01 12:19:12 although those two statements are not logically bound. Aug 01 12:19:15 gnome sucks :P Aug 01 12:19:41 dos1, to be honest, I stopped using KDE a long time ago Aug 01 12:19:51 i'm using KDE 4.2 now Aug 01 12:20:04 I stopped using kde and switched to gnome 8 years ago, give or take Aug 01 12:20:13 and then gave kde another go 3 years ago Aug 01 12:20:17 (though not for a long period) Aug 01 12:20:24 and decided that's not for me Aug 01 12:20:24 \ Aug 01 12:20:33 well, gnome is not for me ;) Aug 01 12:20:40 though I have used fluxbox, xfce and e17 Aug 01 12:20:46 and they all are great ;) Aug 01 12:20:55 dos1, I haven't gave kde4 a try yet Aug 01 12:21:05 i have used kde3, gnome, xfce and kde4 Aug 01 12:21:18 and for me kde4 > kde3 > xfce > ...long pause... > gnome Aug 01 12:21:20 dos1, flux is awesome on laptops Aug 01 12:21:20 :P Aug 01 12:21:31 dos1, really? but xfce sucks! Aug 01 12:21:51 TAsn, nah... xfce4 is just fine :) Aug 01 12:21:58 kde is better Aug 01 12:22:11 mrmoku, it's better than the previous, that's for sure Aug 01 12:22:13 but still... Aug 01 12:22:25 xfce4 is quite good Aug 01 12:22:30 but i prefer KDE :P Aug 01 12:22:36 kde roxxx Aug 01 12:22:55 but it has flaws on powerpc Aug 01 12:22:56 8( Aug 01 12:23:11 mrmoku, I started liking xfce a lot better what they moved to thunar Aug 01 12:23:21 though still, not as good as flux/gnome/kde/e17 Aug 01 12:23:43 kde = great apps, great theming, great configuration options Aug 01 12:23:54 the good thing is... we have a choice :D Aug 01 12:24:00 windoof users don't ;) Aug 01 12:24:02 mrmoku: ++ :) Aug 01 12:24:12 mrmoku, sure they do Aug 01 12:24:22 they can install vista themes on their xp Aug 01 12:24:29 :) Aug 01 12:25:35 mrmoku: any prefered colors for the logo? Aug 01 12:25:58 bumbl, although you didn't ask me Aug 01 12:26:06 I'd like to see "live and fun" colors :)( Aug 01 12:26:19 TAsn: any prefered colours? Aug 01 12:26:53 :) Aug 01 12:26:57 "live and fun" Aug 01 12:27:14 live and fun = black and white? Aug 01 12:27:19 I'm trying to think about actual coloring Aug 01 12:27:38 blue! Aug 01 12:27:50 as on http://dos.jogger.pl/ :D Aug 01 12:27:51 bumbl, no :) Aug 01 12:27:57 dos1, also thought about light blue Aug 01 12:28:01 maybe dark purple Aug 01 12:28:32 dos.jogger.pl is my only web design which i like :D Aug 01 12:29:58 dos1, nah, too blue :) Aug 01 12:30:03 ;D Aug 01 12:30:10 bumbl, maybe red, and a touch of yellow/green Aug 01 12:30:28 (not necessarily together) Aug 01 12:30:36 bumbl, rainbow then :P Aug 01 12:31:23 rainbow? Aug 01 12:31:58 then politics from right side won't use it! Aug 01 12:31:59 :D Aug 01 12:39:21 dos1, that would be fine for me... even preferred Aug 01 12:39:30 ;) Aug 01 12:39:35 as the evil side should never use superiour tools :P Aug 01 12:50:00 dos1: here they would Aug 01 12:50:42 rainbow = red green yellow orange blue (we don't have something yellow but two rightwing parties (orange and blue) Aug 01 12:54:49 * mrmoku sends the german free democratic liberals to austria and is happy that bumbl then has the full rainbow and he got rid of them ;) Aug 01 12:56:14 ;D Aug 01 12:57:09 mrmoku: we have them here too but with ~2% they are too small for the rainbow Aug 01 12:57:15 * mwester adds all the liberals in the US, and wishes austria well in taking care of their high-maintenance, especially the ones from Hollywood. Aug 01 12:57:34 bumbl, ORANGE! Aug 01 12:57:36 ROCKS! :) Aug 01 12:57:46 TAsn: they are a rightwing party here ;) Aug 01 12:59:23 Of course, Austria may have to do a little re-education; it seems our liberal President thinks Austrians speak "Austrian" as a language. Sigh. Aug 01 13:00:26 mwester: really? Aug 01 13:00:32 bumbl, I'm not involved in politics, so I really don't care :) Aug 01 13:00:32 that's progress Aug 01 13:00:50 he does not think we speak "Australian" Aug 01 13:00:51 ;) Aug 01 13:01:13 yaaaAAAAAAAAAAAAAWWWNN Aug 01 13:01:18 sorry Aug 01 13:02:11 bumbl, lol. Aug 01 13:02:21 30°, actually too high for me to move Aug 01 13:02:40 bumbl, it's easy to confuse you with australia because of the kangaroos... Aug 01 13:03:11 DocScrutinizer, here it's 35° (I think) Aug 01 13:03:12 dos1: you prepared a nice little package yet? Aug 01 13:03:31 I just came back from surfing, the water was 30.5° Aug 01 13:03:40 so believe me' it's hotter here :) Aug 01 13:04:00 TAsn: pheew, but yeah, you're close to the sea, that's usually more comfy at high temp Aug 01 13:04:26 dunno why Aug 01 13:05:43 ~weather telaviv Aug 01 13:05:45 I can't find station code "TELAVIV" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:05:55 bah Aug 01 13:06:37 interesting Aug 01 13:06:41 ~weather MUC Aug 01 13:06:43 I can't find station code "MUC" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:06:50 ~weather MUNIC Aug 01 13:06:52 I can't find station code "MUNIC" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:06:53 +h Aug 01 13:06:54 ;) Aug 01 13:07:00 ~weather mue Aug 01 13:07:01 I can't find station code "MUE" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:07:14 ~weather muenchen Aug 01 13:07:16 I can't find station code "MUENCHEN" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:07:25 ~weather munich Aug 01 13:07:28 I can't find station code "MUNICH" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:07:37 fsckU Aug 01 13:07:42 ~weather 20100 Aug 01 13:07:44 I can't find station code "20100" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:08:44 DocScrutinizer, :) Aug 01 13:09:01 ~weather jerusalem Aug 01 13:09:03 I can't find station code "JERUSALEM" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:09:06 ;\ Aug 01 13:09:41 ~weather LLSD Aug 01 13:09:43 Tel Aviv / Sde-Dov Airport, Israel; (LLSD) 32-06-53N 034-46-56E; last updated: 2009.08.01 1250 UTC; Dew Point: 73 F (23 C); Heat index: 93.9 F (34.4 C); Pressure (altimeter): 29.71 in. Hg (1006 hPa); Relative Humidity: 62%; Temperature: 87 F (31 C); Visibility: greater than 7 mile(s); Wind: from the WNW (300 degrees) at 13 MPH (11 KT) Aug 01 13:09:48 :) Aug 01 13:10:08 nah, not that hot ;) Aug 01 13:10:16 ~weather 10870 Aug 01 13:10:17 I can't find station code "10870" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:10:40 ~weather EDDN Aug 01 13:10:42 Nuernberg, Germany; (EDDN) 49-30N 011-03E 318M; last updated: 2009.08.01 1250 UTC; Dew Point: 53 F (12 C); Pressure (altimeter): 29.94 in. Hg (1014 hPa); Relative Humidity: 41%; Temperature: 78 F (26 C); Visibility: greater than 7 mile(s); Wind: Variable at 2 MPH (2 KT) Aug 01 13:11:12 ~wheater LLJR Aug 01 13:11:39 hmm it does not do anything Aug 01 13:11:51 ~weather LLJR Aug 01 13:11:53 bumbl: bzzbot and apt still ignore you ;-0 Aug 01 13:11:53 I can't find station code "LLJR" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:12:20 mwester: i want auto correction ;) Aug 01 13:12:57 ~weather LOWL Aug 01 13:12:59 Linz / Hoersching-Flughafen, Austria; (LOWL) 48-14N 014-11E 313M; last updated: 2009.08.01 1250 UTC; Dew Point: 60 F (16 C); Pressure (altimeter): 29.97 in. Hg (1015 hPa); Relative Humidity: 54%; Sky conditions: mostly clear; Temperature: 78 F (26 C); Visibility: greater than 7 mile(s); Wind: from the E (100 degrees) at 8 MPH (7 KT) Aug 01 13:13:43 mrmoku, not indeed Aug 01 13:13:47 felt hoter :) Aug 01 13:14:05 yeah, heat is very subjective :) Aug 01 13:14:09 hotter* Aug 01 13:14:26 ~weather 27347 Aug 01 13:14:28 I can't find station code "27347" (see http://www.nws.noaa.gov/oso/site.shtml or http://www.nws.noaa.gov/tg/siteloc.shtml for ICAO locations codes). Aug 01 13:14:29 TAsn, maybe because of some hot girls? Aug 01 13:15:04 mrmoku, especially when you wear a tight lycra shirt... Aug 01 13:15:16 mrmoku, no ;( only tight lycra Aug 01 13:15:44 neoprene even better ;-P Aug 01 13:15:57 or latex Aug 01 13:23:03 PaulFertser, g Aug 01 13:23:08 oops Aug 01 13:23:41 PaulFertser, what about the wifi_ifupdown.patch? Aug 01 13:23:59 (http://docs.openmoko.org/trac/ticket/2277) Aug 01 13:28:44 BBQ for the rescue :-)) Aug 01 13:29:13 paul is unresponsive ;) Aug 01 13:29:32 maybe BBQ as well ? Aug 01 13:37:40 mickeyl: how to query why auto-suspend is disabled (can't even be enabled in shr-settings)? [hint: ls-resources tool request] Aug 01 13:43:36 lsar for ls-allocated-resources, analog to lsof Aug 01 13:45:26 another mandatory building block for FSO, like mdbus and fsoraw Aug 01 13:48:41 dos1, can you use ObjectPath instead of string in opimd? Aug 01 14:11:35 what do linux guys use as PIM on thier desktop pc? (like outlook on windows?) Aug 01 14:13:54 F4t: are there any linux guys here? ;) Aug 01 14:14:15 F4t: (just trying to point out that it is not linux specific) Aug 01 14:16:12 ok, so whats the prefered software that does that? Aug 01 14:17:11 F4t: i use gnus for email Aug 01 14:17:13 Better yet, in a corporate working environment, what is used for internal email communication, meetings, and stuff like that? Aug 01 14:22:38 playya_: yes, i want to Aug 01 14:24:03 lindi-, does it have meeting scheduling? Aug 01 14:24:47 F4t, in the US at least, there is very little Linux PIM used in the corporate environment. Aug 01 14:25:09 F4t: nope, it's MUA Aug 01 14:25:34 In fact, the Linux "die-hards" I know have mostly been forced to run a Windows system as a Virtual Machine on their Linux laptops. Aug 01 14:26:02 fortunately i'm not linux die-hard but free software user : Aug 01 14:26:52 fortunately I'm a pragmatist, and don't get too worked up about having to run Windows-based non-free software as part of one's employment. Aug 01 14:26:54 lindi-: mwester: F4t: howdy!! Aug 01 14:27:31 Salutations, Mr. DocScrutinizer -- are you roasting over there? (I seem to recall that it's unusually warm there) Aug 01 14:27:48 you bet Aug 01 14:27:48 DocScrutinizer, hi Aug 01 14:28:32 Meanwhile, my area (Chicago) has just set a record for the coldest month of July. Aug 01 14:28:42 ~weather EDDN Aug 01 14:28:45 Nuernberg, Germany; (EDDN) 49-30N 011-03E 318M; last updated: 2009.08.01 1420 UTC; Dew Point: 55 F (13 C); Pressure (altimeter): 29.91 in. Hg (1013 hPa); Relative Humidity: 44%; Temperature: 78 F (26 C); Visibility: greater than 7 mile(s); Wind: from the ESE (110 degrees) at 3 MPH (3 KT) Aug 01 14:29:29 so what you are telling me is that there is no collabartion/working environment for linux based corporation? (as in meeting scheduling, contacts, mails, document publishing.. etc...) Aug 01 14:29:40 that's subburbs (aka airport). Downtown worse Aug 01 14:37:57 "Kolab" seems like it.. Aug 01 14:40:52 F4t: yup Aug 01 14:41:23 though a little hard to run it on E I guess ;-) Aug 01 14:41:48 would prolly need a little bit of love until works on FR Aug 01 14:42:12 aah, wait. you said linux, not FR Aug 01 14:44:01 Im just thinking in terms of a Sync app for the moko... one that would work with outlook contacts meetings and shit, and was trying to figure out whats the linu equivalanet... Aug 01 14:44:10 equivalent*? Aug 01 14:44:30 mwester, which bitbake branch to use, now that its in git? Aug 01 14:46:48 DocScrutinizer, FR would have its own calendar/contacts and apps... im just looking into Syncing with a desktop pc Aug 01 14:48:00 F4t, might help: http://www.osalt.com/exchange Aug 01 14:49:49 furthermore, I wonder how do iCal and vCard fit as I think they are pretty widespread... (and free?) Aug 01 14:49:50 F4t: why outlook? Aug 01 14:49:58 vcards roxx Aug 01 14:50:40 * Q-Master has a script which imports all contacts to not so working PIM from vcard file Aug 01 14:50:51 Q-Master, he said outlook-equivalent Aug 01 14:50:58 btw F4t I use evolution for that Aug 01 14:51:02 (see libebook) Aug 01 14:51:05 vcard and evolution Aug 01 14:51:06 on my box Aug 01 14:52:30 Novell Evolution?! Aug 01 14:52:41 F4t, huh? Aug 01 14:52:49 gnome's evolution ;) Aug 01 14:53:07 http://en.wikipedia.org/wiki/Evolution_%28software%29 Aug 01 14:53:41 novell suxx Aug 01 14:53:45 yup Aug 01 14:55:58 so yes Aug 01 14:56:04 novell evolution Aug 01 14:58:45 ximian Aug 01 14:58:53 novell suxx 8) Aug 01 14:59:08 I wonder if we should write our own human interface guidelines... Aug 01 14:59:08 ;\ Aug 01 14:59:29 as we already had a share of mishaps with our GUI's consistency. Aug 01 14:59:35 freesmartphone.org: 03seba.dos1 07framework * r6dd6a3e550c9 10/framework/subsystems/opimd/ (pimd_calls.py pimd_contacts.py pimd_messages.py): opimd: Queries: implement (Contact|Message|Call)Added signal Aug 01 15:00:02 dos1, you are crazy! :) you actually code 24/7, how do you even find the time?! Aug 01 15:00:20 * TAsn wishes he had more free time. Aug 01 15:00:27 hehehe ;D Aug 01 15:01:19 TAsn: (styleguide) full ack Aug 01 15:01:33 dos1 knows why ;-) Aug 01 15:01:42 dos1, why? Aug 01 15:01:46 DocScrutinizer, cool ;) Aug 01 15:02:14 I'm now trying to figure out if I should write it, or maybe someone that actually likes/knows GUIs. Aug 01 15:02:18 Ainulindale, up for the job? Aug 01 15:02:19 ;) Aug 01 15:02:37 dos1: btw, I'm really happy with button placement right now Aug 01 15:02:39 we have many samples to learn from Aug 01 15:02:50 apple's, gnome's kde's ... Aug 01 15:03:34 TAsn: stick to apple for template, and to kde for actual decisions Aug 01 15:04:16 DocScrutinizer, what about gnome? :) Aug 01 15:04:27 then, after you're done with it, nasty Doc will nag it apart ;) Aug 01 15:04:45 bah, I hate gnome. don't ask why Aug 01 15:05:13 DocScrutinizer, what about Vista's guidelines? don't you think vista's UI is just perfect (j/k, troll) Aug 01 15:05:29 :X Aug 01 15:05:45 don't get me started ;) Aug 01 15:06:54 have to run now, buy a bull, err beef Aug 01 15:06:57 BBq Aug 01 15:06:57 apple + kde looks fine Aug 01 15:07:00 cya Aug 01 15:07:12 dos1, :| Aug 01 15:07:19 apple is evil :) Aug 01 15:07:28 but I guess ppl like their UI Aug 01 15:07:38 TAsn: i know apple is evil Aug 01 15:07:49 TAsn: apple knows how to write styleguides Aug 01 15:07:54 that's the point Aug 01 15:07:59 exactly Aug 01 15:08:05 DocScrutinizer, aye, that's what matters. Aug 01 15:09:04 I particularly hate apple GUI Aug 01 15:09:41 but I adore it's consistency, due to good styleguide Aug 01 15:10:03 DocScrutinizer: i can't agree more Aug 01 15:10:05 ;) Aug 01 15:10:21 both about hating gui, and adoring consistency Aug 01 15:15:49 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * rd562347894fc 10framework/framework/subsystems/opimd/ (pimd_calls.py pimd_contacts.py pimd_messages.py): Merge branch 'master' of git@git.freesmartphone.org:framework into dos/opimd-tracking Aug 01 15:20:42 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r6c8ebc1ffa48 10framework/framework/subsystems/opimd/pimd_generic.py: opimd: GenericDomain: implement EntryAdded signal in queries Aug 01 15:39:49 damn, mayday! autosuspend sticky on off. How to fix (except reboot)? Aug 01 15:40:21 TAsn: ^ Aug 01 15:40:33 no idea. Aug 01 15:40:36 :) Aug 01 15:40:55 * DocScrutinizer curses dbus and fsousaged Aug 01 15:41:03 alphaone: ^ Aug 01 15:44:15 nevermind, have to run now Aug 01 15:44:17 bbl Aug 01 15:55:06 mrmoku: what's the idea of libframeworkd-phonegui? I'm trying to understand how it's working Aug 01 15:56:52 Q-Master, the idea is to be able to have different libframeworkd-phonegui-xyz frontends Aug 01 15:58:27 ok Aug 01 15:58:42 i'll ask you a bit late more precisely Aug 01 16:02:39 raster_, do you accidently know something about the nüvifone? Aug 01 16:23:40 DocScrutinizer: plug usb in and out Aug 01 17:00:50 freesmartphone.org: 03mickey 07fso-term * r4eafbb127b92 10/ (.gitignore configure.ac): check for vala version; use AC_GNU_SOURCE Aug 01 17:12:27 mickeyl, ping Aug 01 17:12:56 mickeyl, before comitting shr stuff to oe.dev I have to sort out our build problems Aug 01 17:13:23 mickeyl, somehow a whole bunch of RRECOMMENDS does not get installed into the image Aug 01 19:17:21 dos1(away): (plug usb) I did, to no success Aug 01 20:10:31 i have great news! Aug 01 20:10:41 i broke opimd ;D Aug 01 20:11:59 dos1, thats not what I would call news - old story instead ;) Aug 01 20:12:32 and I'm sure the even greater news is you will commit a fix and want me to build it, right? :P Aug 01 20:12:33 trying to do GenericDomain and having Contacts domain inheriting it ended with not working Contacts domain Aug 01 20:12:47 mrmoku: no, broken opimd isn't commited yet Aug 01 20:12:49 ;) Aug 01 20:12:53 ahh :) Aug 01 20:12:58 but that's really general breakage :D Aug 01 20:17:03 mrmoku: but the worst thing is it doesn't work, and it doesn't print any error :x Aug 01 20:19:38 dos1: tzztzztzz Aug 01 20:24:16 DocScrutinizer: buzzzzzz Aug 01 20:35:00 ;] Aug 01 20:46:47 buy Aug 01 20:59:28 DocScrutinizer: i'm sorry, was out for the whole day. Aug 01 21:00:06 PaulFertser: no reason to feel sorry. Hope you enjoyed it Aug 01 21:00:30 DocScrutinizer: there's no easy way to read the result of adc conversion as it should be initiated first. As to the external charger, i think /sys/class/power_suply/ac/present is enough. Aug 01 21:02:04 PaulFertser: dos1: for purposes of resetting hostmode I dunno if it's detected in hostmode anyway. IF yes, then yes, sufficient Aug 01 21:02:25 DocScrutinizer: hehe, no BBQ, just a bit of fooling around with my friend (including car (his) repair) discussing nonsense questions, watching stupid cartoons, talking about modern military tanks etc. Aug 01 21:03:00 DocScrutinizer: hm, i need to check that. Seems like i should really start to keep a todo list. :) Aug 01 21:03:26 mrmoku: hey, so you haven't asked OE folks yet? Aug 01 21:03:50 PaulFertser: I'm glad to hear you take a day off from IT every now and then ;-) Aug 01 21:03:50 mrmoku: and didn't dare to read bitbake sources too? ;) Aug 01 21:04:10 PaulFertser, I don't know exactly *what* to ask in first place Aug 01 21:04:18 and I checked bitbake changelog Aug 01 21:04:22 tried different version Aug 01 21:04:23 mrmoku: i did too Aug 01 21:04:35 PaulFertser, btw. problem is *not* om-gta01 specific Aug 01 21:04:40 mrmoku: i'd ask for the semantics of RRECOMMENDS first. Aug 01 21:04:40 same happens for gta02 too Aug 01 21:04:53 mrmoku: not a surprise for me. Aug 01 21:05:09 PaulFertser, I'm studying bitbake -D output right now Aug 01 21:06:09 DocScrutinizer: doing AT@POFF from kernel space is hard enough to be impractical. One needs to write a new "line discipline" iirc and some other tricky things. Aug 01 21:07:10 PaulFertser is correct. I examined that early on, it is not an easy thing (although not impossible either) Aug 01 21:07:51 PaulFertser: anyway if fsckng power driver does a magic-R AD probe, it MUST expose raw result to sysfs Aug 01 21:07:57 DocScrutinizer: werner's "pmu" does exactly that -- direct communication with PMU. I think it can be used to query ADC but it will result in warnings in kernel log as for the driver it'll be like an unexpected "adc ready" interrupt. Aug 01 21:09:02 PaulFertser: ack. driver needs augmentation Aug 01 21:09:24 to provide a sysfs node for AD-result Aug 01 21:10:06 bbiab Aug 01 21:10:16 DocScrutinizer: are you sure it will be useful anyway? Aug 01 21:10:30 YES ;-))) Aug 01 21:10:33 :D Aug 01 21:10:42 DocScrutinizer: btw, just tried nokia BL-6C Aug 01 21:10:47 1150 mAh Aug 01 21:10:52 ooh Aug 01 21:11:04 Fits perfectly, i guess it's exactly the same form-factor as gta02 battery. Aug 01 21:11:15 (BL-5C being thiner) Aug 01 21:11:56 anyway, have to run... later Aug 01 21:11:57 DocScrutinizer: i'm currently preparing a battery faq, will ask for you comments prior to releasing. Aug 01 21:12:01 See you :D Aug 01 21:13:36 mwester: (sending @POFF) i think it'd require a userspace binary to do special ioctl to connect a serial port to the driver, so it'd depend on distro to work anyway. Aug 01 21:14:09 And proper distro will no to turn off modem by itself. Aug 01 21:16:21 IMO, the lack of a power switch for GSM on the gta01 makes "airplane mode" on that hardware too fragile. It's a fatal flaw, IMO. Aug 01 21:16:48 Pull the battery, it's the only way to be sure that hardware does not have the GSM operating. :( Aug 01 21:17:18 i managed Contacts domain to work again :) Contact class now inherits from GenericEntry, now it's time for ContactsDomain to inherit from GenericDomain :x Aug 01 21:17:19 So therefore it's absolutely pointless to worry about the AT@POFF sequence; too unreliable no matter how you do it. Aug 01 21:17:23 mwester: so what's use of "airplain mode"? Aug 01 21:18:25 larsc: ping Aug 01 21:21:01 PaulFertser: give the one using it the warm and fuzzy feeling of compliance :) Aug 01 21:21:50 PaulFertser, Airplane Mode is required by law in certain countries, and by safety rules in certain companies. It's called "airplane mode" because in NA one cannot use devices that transmit during take-off and landing, and certain devices cannot be used in-flight. Whether that is Aug 01 21:21:55 wpwrak: i can't say lack of a feature like that can be considered fatal for a developer device (and gta01 undoubtly is). Aug 01 21:22:31 technically sound is another matter, but it is necessary to keep oneself out of jail. A more technically-sound case is blast zones, and certain companies that cannot tolerate the GSM interference in certain areas of a lab or plant. Aug 01 21:22:50 mwester: i've heard that common law often leads to many meaningful rules. Aug 01 21:23:20 (developer device) Politics. Don't go there. We all know that the gta0x were not marketed as "developer devices" to begin with. But I do agree that in terms of quality, they are engineering prototype devices. Aug 01 21:23:29 ok, going to sleep now Aug 01 21:23:32 mwester: gta01 was always marketed as a dev device Aug 01 21:23:37 iirc Aug 01 21:23:59 mwester: we actually have a very good use case: the freerunner that made it into space. that one i believe even without going into politics :) Aug 01 21:24:07 PaulFertser, I had a screen shot of the OM.com web site that seemed to imply different. But I'm not going to argue the point. Aug 01 21:24:28 mwester: gta01? Aug 01 21:24:36 wpwrak: for that specific usecase i'd desoldered the inductor thus cutting power to the modem really reliably. Aug 01 21:24:39 mwester: does it laso include the supposed quand band ? :) Aug 01 21:24:54 mwester: i always heared gta02 will be mass market device. and that was *before* gta02 release Aug 01 21:24:56 PaulFertser: yeah :) Aug 01 21:25:01 mwester: i might be wrong on gta01 but i never had that impression. I trust you. Aug 01 21:25:04 on gta02 release and after it was "for developers" :P Aug 01 21:25:04 wpwrak, the quad-band issue is the reason I saved that web-site image in the first place ;) Aug 01 21:25:18 DocScrutinizer: talking about gta01 here Aug 01 21:25:22 dos1: ^^ Aug 01 21:25:44 dos1: as gta02 has a hardware gsm modem power switch. Aug 01 21:26:06 Yep, the gta02 did it right. Aug 01 21:26:17 mwester: though technically even gta02 has RF part powered all the time. Aug 01 21:26:39 mwester: ask wpwrak ;) Aug 01 21:26:41 It does, but the rules for "airplane mode" are suitably vague, I have been told. Aug 01 21:27:29 good night Aug 01 21:27:29 The rules doesn't make sense. You can get arrested for actually talking on phone but i think not for failing to activate mythic "airplane mode". Aug 01 21:27:38 (in practice) Aug 01 21:27:40 Frankly, I think the rules as presented to the public are nothing but bullcrap. For example, no GPS can be used on an airplane. WhAT??! Aug 01 21:28:02 PaulFertser: (rf powered) yes, sigh ... Aug 01 21:28:16 PaulFertser, in practice, no -- nobody checks, and they lack the training and instruments to find a running radio reciever anyway. Aug 01 21:29:01 mwester: exactly because there's no sense to do it. If there was a real threat they wouldn't allow electronic equipment on board at all (including luggage). Aug 01 21:29:28 So on recent flights, the rules explained by the crew before departure are this: "If it has an on-off switch, it must be off". Which, of course, means nothing... Aug 01 21:29:34 PaulFertser: pong Aug 01 21:30:01 PaulFertser, or they would simply issue lead-lined bags for passengers who insist on taking electronics. Aug 01 21:30:08 Security theatre. TSA. Those are almost synonyms now. Aug 01 21:31:09 larsc: i wonder what to do with all my patches i needed to get gta01 driver working... No feedback from upstream, no feedback from OM devs, no feedback from users. I guess general public would like to see it work but they lack kernel compiling skills/desire. Aug 01 21:31:18 The problem is that they are underpaid, undertrained, and take themselves way too seriously. A very bad combination, as they will take out their frustrations if you give them a chance. It's best not to push them, I have found from experience. (Once. Bad.) Aug 01 21:32:14 larsc: for me problem is that i still have a working gta02 battery so i can't say i did that just scratching my personal itch. Aug 01 21:33:17 mwester: so i can't say that lack of airplane mode is an issue all in all. If you asked, you can type something in the terminal claiming that it "rfkills" everything. Aug 01 21:33:47 PaulFertser: I guess as long as nobody complains merge them. Aug 01 21:34:16 Yes, but on the off-chance that someone at the gate wants to see it "demonstrated", frankly I would not risk the gta0x guis (any distro) -- I would just yank the battery and show the guy that it's off. Aug 01 21:34:18 larsc: ok, will do tomorrow then. Aug 01 21:34:43 larsc: thanks Aug 01 21:35:40 PaulFertser, the _real_ reason I was so disturbed about the GTA01 and the lack of a GSM power switch is not Airplane mode, it is that powering off the phone can leave it often in a state where the GSM remains running, killing your battery. This is especially bad iff you turned off the phone to preserve your battery (while hiking, or such for example). Aug 01 21:35:58 mwester: ack Aug 01 21:36:06 mwester: that really sucks, i agree. Aug 01 21:36:28 The principle of least surprise would suggest that the GSM is powered off when the phone is power off, and the gta01 made accomplishing that goal very difficult. The gta02 fixes that, so I'm happy in that regard. :) Aug 01 21:37:34 PaulFertser: silence means assent ;-) Aug 01 21:37:42 quick rant: Aug 01 21:37:52 FUCKING TUXBRAIN STILL HAVEN'T PAID MY MONEY!!! Aug 01 21:37:54 mwester: is the modem waking up related to the console mux ? Aug 01 21:37:56 wpwrak: or that nobody cares/has spare time. Aug 01 21:37:58 FFS!~ Aug 01 21:38:00 * TAsn is now calm again. Aug 01 21:38:17 PaulFertser: which is the same, isn't it ? ;-) Aug 01 21:38:20 (though deep inside I'm still pissed, this is *NOT* normal) Aug 01 21:38:43 wpwrak: and since larsc is the most active in merging with upstream atm i decided to ask him explicitly to not make his work harder. Aug 01 21:39:11 TAsn: did you already try public embarrassment ? (theirs :) Aug 01 21:39:24 wpwrak, no I haven't. Aug 01 21:39:37 I don't want to harm them Aug 01 21:39:44 as they are good ppl in general Aug 01 21:40:17 (contributing money to open source software) Aug 01 21:40:18 wpwrak: (mux) not directly, but I did think about plumbing that together at one time. It's just too hard, too many layering violations. Aug 01 21:40:20 TAsn: well, if they're not responding, a bit of prodding can't be all wrong Aug 01 21:40:29 but still, this is fucking pissing me off. Aug 01 21:40:32 mwester: yeah :-( Aug 01 21:40:42 wpwrak, I can reach them once in every couple of weeks Aug 01 21:40:54 though unfortunately this once was 2 weeks ago Aug 01 21:41:04 which means I have 2 more weeks to wait till a response :) Aug 01 21:41:44 mwester, PaulFertser, in practice, no -- nobody checks, and they lack the training and instruments to find a running radio reciever anyway. Aug 01 21:41:48 wpwrak: in fact i'm a bit concerned about lack of leadership in lowlevel OM stuff atm. In this situation given we have 2 almost equally good solutions we can discuss them to no end not knowing how to decide. Lack of leadership also can contribute to spending time of community members in some other ineffective ways (since noone knows what's more urgent/needed and what can wait). Aug 01 21:41:52 TAsn: maybe they're just busy selling millions of freerunners :) Aug 01 21:42:00 is it even possible to find a running radio receiver? Aug 01 21:42:12 TAsn, ask any HAM radio operator. Aug 01 21:42:17 Yes, they can find it. Aug 01 21:42:18 wpwrak, I hope that's why ;) Aug 01 21:42:26 HOW?! Aug 01 21:42:29 mwester: receivers? Aug 01 21:42:35 Listen for the local oscillator. Aug 01 21:42:46 PaulFertser: ah, didn't know there were conflicting solutions. what are they ? Aug 01 21:42:49 oscillator? Aug 01 21:43:26 ok, this means what I thought this means (oscillator) Aug 01 21:43:41 mwester, you mean "listening" to vibrating crystals? Aug 01 21:43:43 PaulFertser: I just created a branch in the kernel repo based on .31-rc5 to play a bit around with topgit. Feel free to do the same, so we can see how it works with multiple people doing commits. Aug 01 21:43:45 All (practical) radio recievers use at least one local oscillator to generate an RF signal near the desired signal - the two are mixed, and the difference signal is filtered out. Aug 01 21:43:52 wpwrak: we can use topgit or do not use it e.g. Well, now nobody said anything against using topgit but if somebody says it'll evolve into a useless discussion without anybody in charge. Aug 01 21:44:22 The local oscillator "leaks" slightly, and if you can tune for it, you can not only find the radio, you can find what it was tuned to listen for. This is very useful for finding bugs in rooms, for exampole. Aug 01 21:44:40 larsc: i like topgit idea. No problem for me to learn a new tool. Aug 01 21:44:55 mwester, sweet. Aug 01 21:45:06 Indeed. Aug 01 21:45:08 mwester: can local oscillator be heavily shielded? Aug 01 21:45:11 Yes. Aug 01 21:45:25 PaulFertser: i haven't used topgit myself. stefan seems to be very happy with it and has used it for a while. Aug 01 21:45:33 mwester, actually, very sweet. :) man this channel is great, I learn something new on a daily basis! Aug 01 21:45:43 wpwrak: i'm not talking about topgit, i'll be happy contributing either way. Aug 01 21:45:47 PaulFertser, seems probable. Aug 01 21:45:50 But it makes costs for consumer gear very very high to sheild it to the point where it cannot be easily detected inside a typical room or airplane. Aug 01 21:46:08 mwester, though if so, then bugs can have sheilds as well. Aug 01 21:46:11 shields* Aug 01 21:46:13 Yes. Aug 01 21:46:13 PaulFertser: so ... what's the conflict then ? Aug 01 21:46:27 or are they too small to do it properly? Aug 01 21:46:35 But shielding bugs makes them very expensive, and adds weight and size -- which cause their own problems, TAsn Aug 01 21:46:56 (spread the actual ant from the osc) Aug 01 21:47:04 wpwrak: i'm afraid lack of leadership is harmful. Aug 01 21:47:07 mwester, as I just said :) cool. Aug 01 21:48:45 PaulFertser: well, as long as you can work things out by consensus, that should be fine as well Aug 01 21:49:23 wpwrak: for that a procedure needs to be established. Aug 01 21:49:29 PaulFertser: at least i didn't have the impression that issues ended up being discussed in endless circles. Aug 01 21:50:02 wpwrak: i know consensus mechanism is used in "activist" activities. But they usually have an established procudure for that. Aug 01 21:50:55 PaulFertser: the informal procedure is to just post what you're going to do if you're about to do something that might upset others Aug 01 21:50:56 wpwrak: it was a hypothetical situation. But now we e.g. decide to employ topgit. Who actually makes the decision? Who has voting rights? etc. Aug 01 21:51:07 PaulFertser: then give them time to disagree. if they don't, go ahead Aug 01 21:51:49 PaulFertser: if not, work out what they disagree with. often enough, it's either something off-topic (i.e., if they didn't understand what you're trying to do) or a real flaw. Aug 01 21:51:59 wpwrak: sounds reasonable. Aug 01 21:52:05 PaulFertser: the flaw can of course be in their reasoning :) Aug 01 21:53:23 PaulFertser: (topgit) i don't know it well enough to give advise on that. i think any solution that limits the number of concurrent branches to items that are actively being worked on is acceptable. Aug 01 21:53:47 anyhow gn. Aug 01 21:53:52 mwester, thanks for the info. Aug 01 21:53:59 wpwrak: (topgit) it's not about number of branches, it's about how they're worked with if i understood that correctly. Aug 01 21:54:30 PaulFertser: also, rebases are best avoided. now, i don't know how many situations there are where you can't avoid them sensibly, but as long as it's understood that rebases are not to be undertaken lightly, i think that's okay Aug 01 21:55:13 wpwrak: (consensus) it can work very well if there's a core team of experienced cool devs, but can be disasterous for less experienced/bright. Aug 01 21:55:37 mwester: gta02 actually *is* quadband. Just RF-rx sensitivity is kinda poor for the secret 4th band. but we operated 850 devices on 900 here for months and only heard complaints about poor coverage. (power for GSM on 02)bzzzzz, wrong. It's only TX-amp that's constantly powered. But Amp doesn't have any oscillator. baseband is hardswitched. Aug 01 21:55:44 PaulFertser: yes, topgit basically makes it easier to handle a lot of branches. so there needs to be a guideline for when to branch and when not. so my criterion would be to branch only for something that will be worked on. this keeps branches few and short-lived. Aug 01 21:56:52 PaulFertser: hence the peer review. the kernel reorg proposal did get a critical review Aug 01 21:57:01 mwester: (know what the RX was tuned to) only if you know the IntermediateFreq Aug 01 21:57:25 wpwrak: but no decision yet made. Aug 01 21:57:54 DocScrutinizer, true. but that's a known set, and a small set, for most commercial consumer devices. Certainly that's a challenge for custom-designed devices, though. Aug 01 21:58:08 ack Aug 01 21:58:15 just mentioning Aug 01 21:58:19 :) Aug 01 21:58:26 PaulFertser: yes .. would be good if lars could post an update on the plan, taking into account the feedback Aug 01 21:58:40 455k, 10.7M Aug 01 21:58:47 I bet you could figure out the LO freq for the GTA02, just from looking at the specs. Aug 01 21:58:47 ... a few more Aug 01 21:59:13 wpwrak: i think he will after having more experienced with tg. Aug 01 22:00:47 PaulFertser: fair enough. so everything is moving smoothly, even without dictator :) Aug 01 22:02:38 wpwrak: imho it's too slow to be called smoothly :-/ Aug 01 22:03:21 PaulFertser: wouldn't you want the decision to be based on more than a few minutes of running topgit ? :) Aug 01 22:04:33 wtf topgit Aug 01 22:04:48 wpwrak: or another issue: for merging with upstream we should employ upstream solutions as much as possible. larsc is working on that, using mainline gpio leds driver, gpio keyboard driver etc. I think there's quite a possibility for regressions on this way. For me it's perfectly tolerable. But i know others might disagree. Distro maintainers would want something that work good. Another problem since we are so low on manpower. Aug 01 22:04:58 wpwrak: no i wouldn't. Aug 01 22:06:28 PaulFertser: disagree here. strong favour for working kernel - fsck upstream Aug 01 22:06:38 wpwrak: i think that for an experienced git developer it's obvious that both topgit way and some other sane way of using git are roughly on par so whatever he chooses will work. So if he sees that it takes too long to decide he'll choose what he feels like. Aug 01 22:06:57 PaulFertser: distribution maintainers are free to use .29 if want to use something 'tested' Aug 01 22:07:58 as far as i can see, the only thing which topgit gets us are automatic merges. Aug 01 22:08:08 DocScrutinizer: (topgit) roughly speaking a git "addon" to maintain topic branches with an aim to make decent patch series for upstream submission. Aug 01 22:08:53 k,thnx Aug 01 22:09:09 PaulFertser: upstream is the only way to survive long-term, so any cleanup needed for that is good Aug 01 22:09:44 PaulFertser: if this causes regressions, so be it. i think distros aren't in a great hurry to track things anyway. Aug 01 22:09:54 wpwrak: DocScrutinizer already disagrees. I'm not sure mwester will like that either. Aug 01 22:10:30 wpwrak: some people will say that kernel devs "abandoned" the popular distros and community. Aug 01 22:10:51 * mwester feels he's lost the right to an opinion, since he has not had time to contribute for a long while and it doesn't look like he'll be able to spend a lot of time for a while anyway. Aug 01 22:11:11 PaulFertser: hmm, so they prefer an unmaintained kernel ? Aug 01 22:11:40 WHO *is* "upstream" anyway in the end??? isn't it US who *use* the kernel and add patches to fix whatever "upstream" messed up? Aug 01 22:12:01 DocScrutinizer: upstream is linux.org Aug 01 22:12:01 PaulFertser: because people will move away from maintaining the openmoko kernel. the only way to keep it alive long-term is if it's merged into linus' kernel. Aug 01 22:12:27 ITYM kernel.org Aug 01 22:13:12 wpwrak: don't tell me obvious things. :D But people will complain anyway. :-/ Aug 01 22:13:12 DocScrutinizer: don't forget that upstream also merges new drivers, protocols, file systems, etc. Aug 01 22:13:33 linus' kernel NEVER will *generate* GPIO routines that override the GTA02-specific ones. how could it? Aug 01 22:14:11 DocScrutinizer: you can freeze your kernel, but then you miss out on all the good stuff that's happening there. at some point in time, where will be userspace-visible changes, so eventually you also have to start freezing applications, etc. it's the road to hell. Aug 01 22:14:13 it's OM kernel where those routines are developed Aug 01 22:14:14 They often change things that break the APIs to the older machine-specific drivers. Aug 01 22:14:43 DocScrutinizer: the trick is that if some driver is upstream it will be changed "automatically" if API is changed by the one changing that API. Aug 01 22:14:46 If a machine does not have a maintainer that is active, and keeps up with the API changes, then that machine will bit-rot. Aug 01 22:14:46 wpwrak: offtopic, not my point Aug 01 22:14:50 DocScrutinizer: gpios don't only exist on openmoko ... Aug 01 22:15:08 DocScrutinizer: that is ideal. Sometimes it happens like mwester described. Aug 01 22:15:19 mwester: true. but the pressure is a lot lower if you're in mainline. Aug 01 22:16:00 mwester: and very often, people who break your stuff will even fix it for you :) Aug 01 22:16:25 wpwrak: (gpio) but they work in a unique way on this particular hw Aug 01 22:16:27 wpwrak: in fact for pcf50633 driver it's already happened. Aug 01 22:16:29 mwester: of course, if you're not in mainline, they won't even know they broke something ... Aug 01 22:16:52 PaulFertser: breakage, update, or just the merge ? Aug 01 22:17:52 DocScrutinizer: there are some special cases, yes. but by and large, there shouldn't be an unresolvable conflict. btw, which exception are you talking about ? gpb ? Aug 01 22:19:36 wpwrak: guys fixed stuff for us: http://git.kernel.org/?p=linux/kernel/git/sameo/mfd-2.6.git;a=history;f=drivers/mfd/pcf50633-core.c;h=8d3c38bf9714857fbe97507e52498a39cec122c2;hb=HEAD Aug 01 22:19:49 bah, the whole hw is unique, from schematics to gpio pullups to pcftpzee variant and our special way to handle that crap. why should I care about any "upstream" who thinks he knows better than *me* (OM EE) Aug 01 22:19:58 wpwrak: gpb is a good keyword. i'm wondering why it's shadowed in software? Aug 01 22:20:35 s/pcftpzee/pcf50633/ Aug 01 22:20:35 DocScrutinizer meant: bah, the whole hw is unique, from schematics to gpio pullups to pcf50633 variant and our special way to handle that crap. why should I care about any "upstream" who thinks he knows better than *me* (OM EE) Aug 01 22:20:44 larsc: the led outputs went to transistors without base resistors on some devices. so whatever you write, you always read back a 0. Aug 01 22:21:18 DocScrutinizer: because you want to use the kernel they maintain ? Aug 01 22:21:36 not sure about that Aug 01 22:22:32 DocScrutinizer: hardware vendors that provide BSPs say the same and we *hate* them for that ;) Aug 01 22:22:56 PaulFertser: (fixes) nice ! Aug 01 22:24:57 well, my point is: for *all* SW that's accessing any hw (commonly known hw drivers) *I* am upstream for all I understand the word Aug 01 22:25:13 wpwrak: random devices, or devices of a specific hardware revision? Aug 01 22:25:34 larsc: fortunately the latter ;-) up to gta02v5 Aug 01 22:26:27 wpwrak: including v5? Aug 01 22:26:38 larsc: unfortunately, yes Aug 01 22:26:46 wpwrak: thats bad :/ Aug 01 22:26:48 DocScrutinizer: you can be a maintainer but you're not upstream Aug 01 22:26:53 larsc: there's no publically sold A5 Aug 01 22:27:03 not? Aug 01 22:27:16 DocScrutinizer: unless i'm very confused, the first run was a5 Aug 01 22:27:17 ok Aug 01 22:27:41 in any case, there are devices with that problem in the hands of customers Aug 01 22:27:42 for the MP A5 that's been fixed for power-dual-LED, but not for AUXled Aug 01 22:27:57 DocScrutinizer: yup. so you still need the hack Aug 01 22:28:10 wpwrak: the first sold was A5 with a6 hw backports Aug 01 22:29:28 DocScrutinizer: yeah, but no complete fix -> the problem does exist in the field Aug 01 22:29:33 wpwrak: (hack) ack. So I should prepare to se some "upstream" "fixing" this and thus introducing a regression eventually? I really dislike Aug 01 22:30:28 for AUX GPIO *we* are "upstream" Aug 01 22:30:41 DocScrutinizer: well, talk to upstream. explain the problem, show your solution. if they have a better idea that solves the problem, use it. Aug 01 22:31:14 DocScrutinizer: upstream flows from linux's tree. you can push stuff "up", but unless you're linus, you're not upstream Aug 01 22:31:27 wpwrak: I'm rarely talking to myself ;-P Aug 01 22:31:42 DocScrutinizer: don't confuse upstread/downstream with who has the technical authority on an issue Aug 01 22:32:15 well, ok. So nevermind. Aug 01 22:32:25 prolly I got that wrong Aug 01 22:33:16 DocScrutinizer: hm, this states BL-5C is 1050mAh http://www.nokia.co.uk/find-products/accessories/all-accessories/power/batteries/nokia-battery-bl-5c Aug 01 22:33:42 This is BL-6C, 1050mAh http://www.nokia.co.uk/find-products/accessories/all-accessories/power/batteries/nokia-battery-bl-6c Aug 01 22:34:10 1150 Aug 01 22:34:28 I can't believe it's only 100mAh difference. Aug 01 22:35:01 Probably that's some modern modification of BL-5C that has 1050. Aug 01 22:35:13 yup Aug 01 22:35:50 Anyway 6C is also comptabile Aug 01 22:36:11 second source boosts capacity upto a claimed 2800 or sth Aug 01 22:36:51 In that form factor? Wow. Aug 01 22:37:13 I'd not use a LiIon battery from a vendor i don't trust. Aug 01 22:40:43 another screenie from that cool qt iface....http://scap.linuxtogo.org/ Aug 01 22:41:13 yup. those are either fakes, or separator is so thin and weak you actually owned a pyrotechnical device rather than a cell Aug 01 22:49:04 how do you mean fake? Aug 01 22:49:25 why would anyone put a fake cap un scap.linuxtogo? Aug 01 22:51:02 ok, i did not see the other caps ;) Aug 01 22:51:25 by fake I meant chinese rogue sellers putting a label on their bat that states 2800mA, while actual cell has maybe 1800 Aug 01 22:51:58 oh, i thought you were talking about the screen captures on scap.linuxtogo.org ? Aug 01 22:52:40 nope, talking about a web research I did some 18months ago Aug 01 22:53:00 its not fake, its some new UI... qt based or something, dont remmber the name of it though Aug 01 22:54:13 looks nice though Aug 01 22:55:32 http://scap.linuxtogo.org/index.php?page=3 Aug 01 22:56:15 yes, i saw those too, nice isn't it? Aug 01 22:56:22 yep Aug 01 22:57:24 is qt considered non free or bad ? Aug 01 22:57:29 by the SHR dev team? Aug 01 23:11:05 not really afaik Aug 01 23:17:34 anyone tried the qt creator IDE yet? Aug 01 23:18:39 pieterc, main development in shr doesn't connected with UI at all. Aug 01 23:19:22 oh, okay Aug 01 23:19:28 I'm sorry for my ignorance :) **** ENDING LOGGING AT Sun Aug 02 02:59:57 2009