**** BEGIN LOGGING AT Sun Sep 02 02:59:57 2007 Sep 02 03:05:32 well i'll be Sep 02 03:05:36 with the other psu Sep 02 03:05:41 we get spining fans Sep 02 03:05:47 and beeps Sep 02 03:05:59 what on earth happened to the working psu... Sep 02 03:06:21 and now if i stick in my memory Sep 02 03:06:27 it seems to vaguely power on Sep 02 03:06:54 Alright! Sep 02 03:11:10 when i want update openmoko i have message "Killed" what is it? Sep 02 03:12:14 what does "update openmoko" mean precisely? Sep 02 03:13:19 root@fic-gta01:~$ ipkg update Sep 02 03:13:19 Killed Sep 02 03:14:29 hha! Sep 02 03:14:37 and with my "unsuppoted" 4-4-4 mem in... Sep 02 03:14:39 it boots Sep 02 03:14:40 nicely Sep 02 03:14:43 to bios Sep 02 03:14:56 even tries to netboot (since i unplugged the hdd) Sep 02 03:15:02 bizarre Sep 02 03:15:17 how on earth did the psu go bums up between moving from box a to box b Sep 02 03:15:47 raster: something got loose? Sep 02 03:15:53 maybe Sep 02 03:16:15 raster: if it works when back on the original box I'll have to come up with a better explanation... Sep 02 03:16:22 Or it's been tits up, but the new MB is just more sensitive. Sep 02 03:16:29 raster: like "the new box used too much power" for instance Sep 02 03:16:41 nup Sep 02 03:16:49 psu that worsk = 250w Sep 02 03:16:53 psu that doest - 400w Sep 02 03:16:55 err Sep 02 03:16:56 PSUs and printers are two of the most problematic things on computing... Sep 02 03:16:59 350 for working psu Sep 02 03:17:29 the "tits up" mb i actually bought only 1 months ago or so Sep 02 03:17:33 its almost brand new Sep 02 03:17:43 so this is truly bizarre Sep 02 03:17:47 Huh. Sep 02 03:18:32 unless it was the xfx gfx card Sep 02 03:18:45 it has this bizrre connector that wants 2 normal power connectors Sep 02 03:18:48 so i plugged in 2 Sep 02 03:20:39 raster: if it's the so-called "PCI express" 6-pin connector, it's not that bizarre Sep 02 03:20:54 hmm Sep 02 03:20:57 raster: it's quite common on higher-powered gfx boards Sep 02 03:21:05 my other nv cars only need the normal "hdd" connector Sep 02 03:21:19 raster: and it does need 2 separate leads from the power supply, since just one cannot carry that much current Sep 02 03:21:25 http://i2.neoseeker.com/neo_image/155084/article/8600_roundup/xfx%208600gts%20bundled%20hardware_thumb.jpg Sep 02 03:21:31 that connector on the right there Sep 02 03:21:54 the one with 2 white bits and the black Sep 02 03:21:58 the black goes into the card Sep 02 03:22:12 the 2 white bits are your usual hdd/cd etc. power connectors (whatver they are calld) Sep 02 03:22:13 raster: yeah, it's not that exotic... some PSUs can even plug directly, without that adapter Sep 02 03:22:29 hmm Sep 02 03:22:35 well then i idnt fuck that up Sep 02 03:22:37 raster: the white ones are commonly called "molex" connectors, and are the single most annoying power connector on a PC Sep 02 03:22:52 i guss somehow i bummed out and my powersupply went dud on me Sep 02 03:22:58 aah yes Sep 02 03:22:59 molex Sep 02 03:23:00 tats it Sep 02 03:23:06 and ys Sep 02 03:23:12 they nver fit quite right Sep 02 03:23:17 also theyre bulky Sep 02 03:23:33 * cesarb hopes SATA finishes taking off quickly, since it has its own power connector which is used in most places one of these "molex" annoyances are Sep 02 03:23:47 cesarb: amen Sep 02 03:23:54 aye Sep 02 03:24:02 thought the sats connector is way too big for its own good Sep 02 03:24:07 (the power one) Sep 02 03:24:08 raster: I've heard some people can manage to get these "molex" connectors INVERTED, which can easily burn a PSU and things connected to it Sep 02 03:24:24 raster: ah, you know WHY is the SATA power connector so large? Sep 02 03:24:32 nup Sep 02 03:24:36 raster: because it has lots of pins... and why does it have lots of pins? Sep 02 03:24:40 because the normal conecotr was size X Sep 02 03:24:47 raster: because each pin can only carry a limited amount of current Sep 02 03:24:56 and they had Y left from the old pata connector and wanted to fill the space? Sep 02 03:25:14 hmm Sep 02 03:25:22 just so they can make thinner ribbon cables? Sep 02 03:25:32 why not just have fatter pins? Sep 02 03:25:43 (it also has a couple of extra hotplugging-specific pins, staggered connection, a pin for delayed startup, and other niceties) Sep 02 03:25:55 in fact Sep 02 03:26:04 why didnt they just combine sata dat aand power Sep 02 03:26:06 raster: because they are not pins like the ones in the old "molex" connector... they are card-edge connectors Sep 02 03:26:08 so u have 1 plug Sep 02 03:26:09 thats it Sep 02 03:26:12 nice and simple Sep 02 03:26:15 aqla usb Sep 02 03:26:35 raster: they can be combined, and are on a hotplug enclosure Sep 02 03:26:45 raster: the connector design was made to be able to be combined Sep 02 03:27:16 raster: but for a normal PC it's simpler to have the 7-pin data one go to the MB and the 15-pin (IIRC) one to go to the PSU Sep 02 03:27:31 raster: without the extra power distribution complexities of a server Sep 02 03:28:14 Yeah. I wish the connector was narrower and combined too. There's not that much point in it being so wide. Sep 02 03:28:24 mmmmm Sep 02 03:28:32 still doesnt make sense Sep 02 03:28:33 That said, eSATA not having power bugs me more. Internally, I just use backplanes. Sep 02 03:28:44 usb manages 500ma as well as data Sep 02 03:28:51 Aria: if it weren't that wide, it would be fatter. you would end up with what you started with (the "molex" ones) Sep 02 03:28:59 Aria: indeed Sep 02 03:29:04 that too sucketh Sep 02 03:29:08 cesarb: could still make it hotpluggable. Sep 02 03:29:08 raster: 500mA isn't enough for a HD Sep 02 03:29:15 cesarb: i know Sep 02 03:29:18 Yeah. Need more like 3A to cover your bases. Sep 02 03:29:22 Still, that seems solvable. Sep 02 03:29:22 but i am sure u could provide enough power Sep 02 03:29:28 just make a fat round plug Sep 02 03:29:45 doesnt have to be skinny Sep 02 03:29:58 its easier to handle too with fingers Sep 02 03:30:04 raster: well, IIRC eSATA is a hack Sep 02 03:30:13 eSata is a hack. Sad. Sep 02 03:30:13 cesarb: a nice hack tho Sep 02 03:30:14 raster: people were using the SATA connector outside the box Sep 02 03:30:18 But it works damn well. Sep 02 03:30:19 raster: which it wasn't designed to Sep 02 03:30:25 i can throw my hd's into an eclosure with its own poer Sep 02 03:30:39 and not need to funnel thigns down a puny 480mb usb cable Sep 02 03:30:40 :) Sep 02 03:30:43 raster: so they simply got it, added a shield, and said "now it's allowed" (IIRC it's also more resistent) Sep 02 03:30:58 Yeah. And the connector doesn't fall out so easily. Sep 02 03:31:01 hehe Sep 02 03:31:07 It's physically well designed now. Sep 02 03:31:14 user driven development Sep 02 03:31:16 But sad that it doesn't carry 2-3A. Sep 02 03:31:30 yeah Sep 02 03:31:32 I'm just happy that SATA drivers support hotplug so well. Sep 02 03:31:39 'cause it saves me a ton of downtime. Sep 02 03:31:40 i am tired of the 8billionand1 power plugs Sep 02 03:31:43 i want 1 Sep 02 03:31:45 Me too. Sep 02 03:31:53 then the pc can power everythgn plugged into it Sep 02 03:31:59 Exactly! Sep 02 03:32:08 no more 23 boxy ac->dc adaptors floating around under a desk Sep 02 03:32:13 I wouldn't mind 18AWG nearly as much as the brick hell I have. Sep 02 03:32:16 those damn dc adaptors suck Sep 02 03:32:25 i am hanging for the day we have dc plugs on walls Sep 02 03:32:32 raster: people have been using USB for that... sadly, only 500mA is a handicap Sep 02 03:32:38 either that or everyone can power off usb for everything Sep 02 03:32:38 :) Sep 02 03:32:38 My workbench at the office, with a huge industrial strip, can still only support 3 machines. Sep 02 03:32:44 Because of the sheer number of AC adapters Sep 02 03:32:53 * * OM Bug 788 has been created by mwester(AT)dls.net Sep 02 03:32:54 * * Starting or stopping gsmd completely locks up the Neo Sep 02 03:32:55 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=788 Sep 02 03:32:58 cesarb: yeah - i know Sep 02 03:33:00 raster: firewire has more power... but IIRC the voltage is not fixed, it has an unpowered variant, etc Sep 02 03:33:02 500ma is a bit of a downer Sep 02 03:33:15 Yeah, Firewire's patents and variation made it lose out. Sep 02 03:33:16 maybe we need usb3 Sep 02 03:33:28 raster: 500mA is too few for most devices who want to use it, and too much for most devices who want to provide it Sep 02 03:33:29 providing 100ma, 500ma 1, 3 or 5a of power Sep 02 03:33:36 (negotiated up from 100ma) Sep 02 03:33:48 USB3 would be awesome. .oO(USB packets over SATA PHY, powered to >= 2A ) Sep 02 03:33:58 yup Sep 02 03:34:01 that'd be nice Sep 02 03:34:13 raster: ah, did you see the new "battery charger specification"? it's only for chargers, but it does allow for autodetection of more than 500mA over a normal USB cable... Sep 02 03:34:29 its eitherthat or wireless power AND data Sep 02 03:34:44 raster: another cool option would be ethernet with power-over-ethernet... (already several watts, being increased to more) Sep 02 03:34:48 cesarb: no - i didnt... cool nanas! Sep 02 03:34:48 Nikola Tesla would love to tackle that, I bet. Sep 02 03:34:58 lets hope laptops and desktops provide it too... soon Sep 02 03:35:05 Yeah, really. Sep 02 03:35:09 and yes Sep 02 03:35:11 poe is nice too Sep 02 03:35:23 all of it nice Sep 02 03:35:34 but the closest to a standard thats out there that works ins usb Sep 02 03:35:38 with powr + data Sep 02 03:35:44 Yep. Sep 02 03:35:53 shame its a little weap on boht fronts Sep 02 03:35:59 480mbit is fine for toys Sep 02 03:36:04 not enough for serious things Sep 02 03:36:07 like high end hdd's Sep 02 03:36:17 Yeah. It's just underspecced in every way. Sep 02 03:36:30 and power-wise too Sep 02 03:36:30 Good enough to keep other standards from competing, bad enough to annoy Sep 02 03:36:36 yup Sep 02 03:36:39 raster: you know what would really be cool? PCI express cable... running (hotpluggable) PCI express to outside the main box Sep 02 03:36:45 well - here's to usb3 Sep 02 03:36:45 :) Sep 02 03:36:47 raster: the standard for that seems to already exist... Sep 02 03:37:00 cesarb: yup - but again.. POWER Sep 02 03:37:02 * cesarb notes there are way too many different standards Sep 02 03:37:09 need power delivered down the same cable Sep 02 03:37:13 The standards that last are the ones that layer, most often. Sep 02 03:37:20 Keep the USB packet format, just change the PHY Sep 02 03:37:29 i'm amazed that the hdmi camp havent decided to throw poer down the cable yet Sep 02 03:37:37 they have stuck in video and audio Sep 02 03:38:01 raster: probably because the cable's wires/connectors are too thin for any decent amount of power Sep 02 03:38:12 :) Sep 02 03:38:15 i know Sep 02 03:38:22 just musing Sep 02 03:38:32 if they upped the connector sizes/cable thickness Sep 02 03:38:37 then it'd be possible Sep 02 03:38:39 raster: but I do think it has power (at least DVI IIRC does, and since they are compatible...), for the I2C chip on the monitor/TV... a very small amount of power, but power Sep 02 03:38:58 oh Sep 02 03:39:00 interesting Sep 02 03:39:15 so u can negotite edid info while the display is entirely unplugged Sep 02 03:39:41 raster: it came from VGA, which also has it Sep 02 03:39:47 i idnt know that Sep 02 03:39:51 well fancy that eh Sep 02 03:40:11 raster: there are lots of these kinds of interesting side channels Sep 02 03:40:21 raster: for instance, memory boards Sep 02 03:40:42 raster: they have a very small chip for information on the memory (supported speeds, etc) Sep 02 03:40:52 that makes sense Sep 02 03:40:58 raster: the chip is (again) I2C, and completely separate from the actual memory Sep 02 03:41:01 which is why i expected my "over-spcceD" ram to still work Sep 02 03:41:05 just get cloked down a bit Sep 02 03:41:12 like most cpus do these days Sep 02 03:41:13 etc. Sep 02 03:41:24 Yeah. It's under-specced that fails less gracefully. Sep 02 03:41:27 raster: other than being on the same PCB, the identification chip and the memory chips have no connection at all Sep 02 03:41:59 well Sep 02 03:42:02 time to get a new psu Sep 02 03:42:14 i want to stop living off my laptop as my workstation Sep 02 03:42:23 very unhappy batteries Sep 02 03:43:02 I know the feeling. Sep 02 03:43:21 then i need to build my raid array.... so i finally can stop having slepless nights over losing my data... Sep 02 03:43:23 hehe @ neo un-bricking by mail on community list Sep 02 03:46:43 Psi: that would be me ;-) Sep 02 03:47:45 :) Sep 02 03:48:26 I'll make sure to get the advanced kit for GTA02, so there's one in Sydney Sep 02 03:49:05 dotmef: i'll be in sydney in a few months Sep 02 03:49:09 semi-permanently Sep 02 03:49:15 i have the jtag kit Sep 02 03:49:20 so why do people flash uboot anyway? ive not needed to so far Sep 02 03:49:35 maybe i will bring it regularly to slug metings for people who briked... :) Sep 02 03:49:48 cannot find JavaScriptGlue_svn.webkit.org_.repository.webkit.trunk.__20070902.tar.gz Sep 02 03:49:49 Psi: i needed to to fix a charging problem Sep 02 03:49:57 didnt negotiate to 500ma properl Sep 02 03:50:25 i wonder what made you have that problem and me not ? Sep 02 03:50:47 my neo was pre gta-01 Sep 02 03:50:49 raster: awesome. I'll have three months of no-fear messing with my uboot Sep 02 03:50:50 ah Sep 02 03:50:53 Anyone know what the Stat application is on http://scap.linuxtogo.org/ ? Sep 02 03:51:01 :-P Sep 02 03:51:39 dotmef: still beware :) Sep 02 03:51:53 ok Sep 02 03:51:57 time to get soem pants on Sep 02 03:52:22 raster: nah - it's ok. my uboot erasing days are done... I hope :-D Sep 02 03:53:23 I flashed the 2007.2 uboot (even before I flashed the rootfs) to fix the bug about the rootfs DFU not erasing completely the rootfs Sep 02 03:57:21 dotmef: thx for the nice email Sep 02 03:58:33 rwhitby: no probs. :-) Sep 02 04:09:37 * dotmef is attempting to learn Ruby Sep 02 04:39:16 woo hoo, irssi on the neo Sep 02 04:44:03 wish I had bluetooth so I could go wireless Sep 02 05:04:30 gcb77_NEO: _sweet_ Sep 02 05:11:08 * gcb77_ is trying to run freeciv on the NEO for some reason Sep 02 05:24:27 This thing (freeciv) is actually playable! Sep 02 05:33:59 moin Sep 02 05:36:18 blindcoder:moinsen Sep 02 05:37:29 gcb77_, the stat application is one that I wrote, I posted on the app mailing list, but the message seems not to be arrived yet. Sep 02 05:38:04 daltona: that thing looks really really cool, I can't wait to try it out Sep 02 05:38:21 I've been suspecting my NEO had a heartbeat :) Sep 02 05:39:12 gcb77_, I think you should find it in the openmoko-apps archives, later I'll setup a gforge project. Sep 02 05:50:01 I posted a civclient screenshot on scap Sep 02 05:52:19 daltona: did you load the ipk package somewhere? Sep 02 05:53:28 gcb77, No I did not, as I don't know how to create one Sep 02 05:54:02 but it's not that hard to build... I con give you instructions Sep 02 05:54:24 I'm still looking for the source :( Sep 02 05:55:10 gcb77, http://lists.openmoko.org/pipermail/openmoko-apps/attachments/20070901/1a6c059d/mokostat-0.1.tar-0001.bin Sep 02 05:55:21 rename it to tar.gz Sep 02 05:57:57 awesome, I'm building it Sep 02 05:59:41 In order to build it, start a devshell from the build environment, and then extract the file, and you can ./configure, make and scp the binary to the neo Sep 02 05:59:42 If somebody has a tutorial explaining how to create an ipk, I would be interested BTW... Sep 02 06:02:00 gcb77, let me know if and how that works Sep 02 06:02:42 In the version I posted, the graph are smoother as I did a rolling average instead of samples Sep 02 06:08:04 daltona: have you looked through http://wiki.openmoko.org/wiki/Application_Development_Crash_Course Sep 02 06:16:03 gcb77, no I didn't, let me read that... Sep 02 06:24:52 gcb77, I read the course, very interesting not seem to show how to create the ipk though Sep 02 06:25:17 did you make a .bb file? Sep 02 06:26:22 not yet, but from the tutorial, it seems that the bb recipie will directly copy the binary on the fs without going through the ipk step Sep 02 06:26:31 I may be wrong... Sep 02 06:39:37 gcb77, did you built it ? Sep 02 06:41:21 daltona: lol, no. I'm trying to figure out how to make it into an ipkg Sep 02 06:41:48 I had that working with my own app before, and I can't remember how I did it Sep 02 06:42:39 gcb77, if you find out, do you mind mailing me how to proceed ? Sep 02 06:45:14 daltona: you read through the "Your First Application"? Sep 02 06:46:07 on the same link as before, halfway down the page Sep 02 06:47:28 yes I did Sep 02 06:48:14 just saw that Sep 02 06:48:31 let me try to create the bb file and see if I can make a ipk Sep 02 08:00:16 damnit, my build failed again >_< Sep 02 08:00:17 df -h Sep 02 08:00:19 whoops Sep 02 08:00:23 anyways, that's why, ran out of space :P Sep 02 08:00:42 rtyler: heh. I might beat you yet ;-P Sep 02 08:00:59 I haven't been paying attention to that terminal >_< Sep 02 08:01:14 major time delay, who knows how long it's been sitting there, failed >_< Sep 02 08:02:22 * dotmef thinks about hacking together an sms alert system through his Neo Sep 02 08:02:37 SMS works now? :P Sep 02 08:03:01 dunno - waiting for my build :-D Sep 02 08:03:27 is it possible to do it through libgsm-tool? Sep 02 08:03:43 Has anyone been able to build libgsmd lately? Sep 02 08:04:29 CM: it depends what do you mean 'lately' Sep 02 08:04:44 zdanek: About 5 minutes ago ;) Sep 02 08:04:47 i have build from repo from last week Sep 02 08:04:48 :) Sep 02 08:04:52 nope Sep 02 08:05:02 I have dialer working in my build Sep 02 08:05:23 but I didn't try to dial from libgsm-tool Sep 02 08:05:44 but it should work because it asked my for my pin before registering to net Sep 02 08:05:55 I din't try sms sending Sep 02 08:05:57 I have the opposite, dialing with libgsmd-tool worked, but never the dialer Sep 02 08:06:25 dialer works from time to time Sep 02 08:06:58 and it asks me for pin at first boot of X Sep 02 08:07:29 if it does then dialing works perfectly Sep 02 08:07:39 and incoming calls too\ Sep 02 08:08:21 maybe you'd like to try our build Sep 02 08:08:34 www.openmoko.org.pl/builds Sep 02 08:08:39 :-) Sep 02 08:10:12 it's now recommended to flash uboot Sep 02 08:10:22 with this builds Sep 02 08:11:08 I have a recent u-boot and I won't flash unnecessarily Sep 02 08:11:13 wtf! Sep 02 08:11:24 ScaredyCat: Good morning to you too :D Sep 02 08:11:38 the build doesn't build all dependancies! Sep 02 08:12:24 i'v downloaded recommended uboot for this build Sep 02 08:12:37 to builds dir Sep 02 08:12:42 CM: Ok Sep 02 08:12:58 but many did not so we recommend for it Sep 02 08:14:04 http://pastebin.ca/678716 Sep 02 08:14:16 I have to make breakfast for my son, so bye for now :) Sep 02 08:19:00 ok, so midori satisfies the libsexy... Sep 02 08:19:14 Has anyone managed to compile 07.02 today? My builds are all failing when it comes to webkit. Anyone passed that point today? Sep 02 08:20:05 i just need gnome-vfs and libxpm4 Sep 02 08:20:10 joao: I haven't gotten that far yet, but I'll tell you if I get past it Sep 02 08:20:17 I'm passed it... Sep 02 08:20:29 but I started my build about 6 years ago Sep 02 08:21:05 ScaredyCat: ok mine is from 45 minutes ago Sep 02 08:21:49 I wanted to test the new gsmd patch from rwhitby. He said gsm attach was now working flowlessly Sep 02 08:24:37 ok , just updated, rebuilding Sep 02 08:34:13 joao: a couple of us have had trouble with building webkit Sep 02 08:35:09 joao: are you getting errors with undefined reference to blah blah blah? Sep 02 08:35:29 dotmef: I thought as much. I have been looking at the logs for the channel yesterday. Sep 02 08:35:49 dotmaf: no i'm getting lots of 404 and also a few path issues Sep 02 08:36:25 joao: is it saying "can't find directory"? Sep 02 08:38:59 Sorry had to swap computers to get the build log ... Sep 02 08:39:18 np :) Sep 02 08:39:41 dotmef: I'm getting a few undefined references in libicui18n.so Sep 02 08:40:19 joao: that's the one. cesarb's recommendation was to delete the build/tmp directory and rebuild Sep 02 08:41:37 dotmef: ok will try that. I'll probably be building for the next eight hours after that though :-) Sep 02 08:41:40 joao: cesarb said "the error is because of a new gcc version on openembedded" Sep 02 08:42:09 joao: yeah - I dunno if there's a better way, and my build hasn't finished yet Sep 02 08:42:21 joao: so I can't tell you for sure that it works Sep 02 08:42:50 rtyler: you there? Sep 02 08:43:04 dotmef: new compiler wrong symbols... sounds logical at least... before resorting to those rm /buid/tmp I'm doing a quick make clean for webkit. Nt much hope but then if it works it saves me a few hours Sep 02 08:44:09 joao: good luck. if my build completes anytime soon, I'll let you know whether it was a success or not Sep 02 08:44:44 dotmef: ok cool Sep 02 08:46:33 how much space does the full build consume? Sep 02 08:47:14 Jiten_: It used to be 12Gb before, but it might grow even more Sep 02 08:49:37 Jiten_: Adding INHERIT += "rm_work" to your local.conf should make it just 2gb or so Sep 02 08:50:14 I can dial the phone by hand but the applet crash right after enable the anantna ... anyoe know about this bug? Sep 02 08:50:27 exit Sep 02 08:50:33 doh! Sep 02 08:51:07 I was just wondering. I cut&pasted the wrong command from the wiki instructions for MokoMakeFile and left it for a while. When I came back and interrupted it, I had a 20GB folder :) Sep 02 08:51:51 * Jiten_ was just trying to look at the qemu version to see what the 2007.2 version looks like. Sep 02 08:53:30 mine is 12 GB now (built one image, skipped the locales) Sep 02 08:53:39 2007.2 Sep 02 08:54:54 Where is this gui: http://scap.linuxtogo.org/files/b5336f7bbf5efd063a08dcf9e083ce04.png Sep 02 08:55:31 kind of makes you wonder what all that is for when the end result has to fit on 64MB flash :) Sep 02 08:55:54 all that unnecessary OE shit -_- Sep 02 08:56:02 dotmef: you rang? Sep 02 08:56:20 rtyler: yeah - just wondering how your build is going? Sep 02 08:56:46 ignore the ? Sep 02 08:56:48 just restarted it after I nuked another whole tmp directory I had lying around >_< Sep 02 08:57:33 ah weak :( Sep 02 08:57:35 btw, do you think they'll manage to get the mass market version out in a month? Sep 02 08:57:40 Jiten: no Sep 02 08:58:43 I'd settle for a developer version with GTA02 hardware :) Sep 02 08:59:23 I'd settle for it for cheap ._> Sep 02 08:59:53 well, yeah, if it's developer version, it should be cheaper :) Sep 02 09:02:37 CM: that's the settingsgui Sep 02 09:03:38 * ScaredyCat slaps CM Sep 02 09:03:40 http://wiki.openmoko.org/wiki/SettingsGUI Sep 02 09:13:06 ls Sep 02 09:13:21 joao: wrong terminal ;) Sep 02 09:13:34 rtyler :-) Sep 02 09:14:25 * dotmef yawns at his task count of 1160 out of 3350 Sep 02 09:15:03 hm, the qemu plays a sound on a mouseclick, but doesn't respond much otherwise... Sep 02 09:15:34 1078 out of a fucking lot Sep 02 09:15:44 lol Sep 02 09:16:36 cheer up guys, only another 7 hours to wait ;) Sep 02 09:17:07 * dotmef throws a clock at ScaredyCat Sep 02 09:17:20 * ScaredyCat watches it Sep 02 09:18:45 * Kero wonders what is wrong with x+y coordinates... Sep 02 09:18:58 perhaps calibration f*cked up with my multiple clicks... Sep 02 09:19:49 anyone experience matchbox-panel being crash ? Sep 02 09:19:58 (and the thing could break off the call it is trying to make, having no SIM in qemu...) Sep 02 09:21:14 AvengerMoJo: when you play with the gsm menu ? Sep 02 09:21:35 time for food Sep 02 09:21:40 Kero: I think there's a calibration file in /etc/ Sep 02 09:22:08 if you remove it (or rename it) then i think you can re-run the calibration Sep 02 09:22:30 There is also a cal program Sep 02 09:22:34 xtscal? Sep 02 09:22:38 Xtscal? Sep 02 09:23:47 probably. I'll just start hacking what I want to hack and create a new image. Sep 02 09:23:53 yes Sep 02 09:24:40 the important thing for _later_ is that I can see whether an image works, before uploading to Neo (thus overwriting a potentially useful install). Sep 02 09:24:40 run ts_calibrate from the shell. It will create a pointercal file in /etc Sep 02 09:25:20 ScaredyCat, 7 hours to what? Sep 02 09:25:22 Kero: where's you sense of adventure ;) Sep 02 09:25:34 jeddy3: another rebuild Sep 02 09:25:41 ScaredyCat: yes :) Sep 02 09:25:55 AvengerMoJo: yes, it's broken Sep 02 09:25:56 ScaredyCat: I'm searching the bugzilla now Sep 02 09:26:09 ScaredyCat: I seem to able to dial by hand Sep 02 09:26:09 ScaredyCat: adventure? I haven't even tried to put a SIM in my Neo... Sep 02 09:26:20 but the match-box-panel crash :) Sep 02 09:26:36 yes, don't use the panel Sep 02 09:26:45 ssh in and restart the gsmd Sep 02 09:27:28 ScaredyCat: thanks :) Sep 02 09:28:24 morning Sep 02 09:29:29 just ;) Sep 02 09:29:38 ScaredyCat: is this bug in the bugzilla? Sep 02 09:29:52 not sure, it is a known bug though Sep 02 09:30:29 ScaredyCat: I'll try to file that then :) ...... Sep 02 09:31:21 ScaredyCat, ah! Sep 02 09:31:26 ScaredyCat: however starting the matchbox-panel directly seem not working right .... do you know how to start matchbox-panel are being called? I can only restart the X completely? Sep 02 09:31:47 iirc you restart x Sep 02 09:34:52 ScaredyCat: How do I start the settings gui? Sep 02 09:35:04 no idea Sep 02 09:35:10 ScaredyCat: look like the panel is working if I started the gsmd frst Sep 02 09:35:11 s it not on the menus? Sep 02 09:35:12 ScaredyCat: look like the panel is working if I started the gsmd first Sep 02 09:35:12 Is even in 2007.2? Sep 02 09:35:20 no you have to add it Sep 02 09:35:34 AvengerMoJo: yes, I think that might be the hack :) Sep 02 09:35:42 hum Sep 02 09:35:58 many be the panel can check the gsmd status first Sep 02 09:36:14 and call activiate it if it is not Sep 02 09:36:34 rwhitby has done some patches, so the fixes should appear in newer images Sep 02 09:39:15 CM: it needs a .desktop file Sep 02 09:39:17 python SettingsGUI-x.x/SettingsGUI.py Sep 02 09:39:30 ScaredyCat: Ah, danke Sep 02 09:45:14 anyone know where can I look for this kind of error explain? +CME ERROR: 32' Sep 02 09:48:51 http://216.239.59.104/search?q=cache:Iken6f3B_xwJ:net4uonline.com/gsm-modem-cms-cme-errors.pdf+CME+error&hl=en&ct=clnk&cd=1&gl=uk Sep 02 09:49:28 32 is network not allowed, emergency calls only Sep 02 09:49:34 I see Sep 02 09:49:38 SpeedEvil: thanks Sep 02 09:50:00 can't get to your ip :) maybe because I'm in side china hehe Sep 02 09:51:16 That's a google cache IP Sep 02 09:51:20 it may be private only Sep 02 09:51:50 anyway, google 'cme error' 32 Sep 02 09:52:18 thanks Sep 02 09:53:23 (script) wiki RecentChanges: [[Package_management]] [[OpenMoko_Local_Groups:_Sydney]] Sep 02 09:55:05 so where is this image coming from? http://wiki.openmoko.org/wiki/Dialer Sep 02 09:55:21 is that out of date? or I'm using something else ? :) Sep 02 09:56:44 I'm curious ... Sep 02 09:56:49 AvengerMoJo: look on the history tag on the page Sep 02 09:57:34 SpeedEvil: thanks Sep 02 09:57:54 * * OM Bug 789 has been created by fabchevalier(AT)free.fr Sep 02 09:57:55 * * Occasional phone crash on X startup Sep 02 09:57:56 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=789 Sep 02 09:58:35 why would the build fail on not being able to build gnome-vfs .. but wok if i bitbake gnome-vfs Sep 02 09:59:17 ! Sep 02 10:06:46 ScaredyCat: I have like that too on all svn builds Sep 02 10:07:06 can't find libxpm4 either now Sep 02 10:07:15 any idea what will build it ? Sep 02 10:09:59 hello, how can I rebuild only the kernel with mokomkafile? Sep 02 10:10:26 make rebuild-package-linux does not do the work Sep 02 10:11:40 (cd build ; source ../setup-env ; bitbake -c rebuild virtual/kernel) Sep 02 10:12:24 or make rebuild-package-linux-openmoko Sep 02 10:12:42 rwhitby: thx Sep 02 10:12:47 or even make rebuild-package-virtual/kernel Sep 02 10:13:05 did someone remove libxpm4 from oe ? Sep 02 10:16:08 * ScaredyCat sighs ... Sep 02 10:16:26 ~blame CM for something Sep 02 10:16:27 * apt blames mtoups and perhaps CM for something and all the evil in the world Sep 02 10:29:06 pft, managed to kill contacts, which took down the GUI and now the Neo reboots. Sep 02 10:29:20 there, ehm, are some stability issues here... Sep 02 10:37:11 <_diego__> Kero: you are right... it happens quite often to me too Sep 02 10:45:20 now I'm guessing some script causes the reboot explicitely (timeout, cause GUI won't restart), because I simply can't believe the entire system would be that unstable Sep 02 10:45:37 no, it's that unstable Sep 02 10:46:03 which driver is that? can't be the kernel itself Sep 02 10:46:33 <_diego__> to be honest, i see openmoko applications crash, but not reboot Sep 02 10:46:49 kernel is fine. X and WM and apps need work. Sep 02 10:47:12 then why did my Neo reboot? Sep 02 10:47:35 oh, it rebooted spontaneously? Sep 02 10:47:39 <_diego__> atm the only problem i see now, is applet crashing and not be able to restart them Sep 02 10:47:47 not just cause X died? Sep 02 10:48:02 wanted to restart the GUI; I'm assuming X died. Sep 02 10:48:37 <_diego__> can u connect via ssh to the cell? Sep 02 10:48:42 that's what it said, at least :) it never got to playing the sound, instead rebooted Sep 02 10:49:14 so that can be graphics driver problem, or audio driver, or who knows what. Sep 02 10:49:43 hello, again, I would like do somethin like cd moko/build/tmp/work/fic-gta01-angstrom-linux-gnueabi/linux-openmoko-2.6.22.5-moko11-r2/linux-2.6.22.5 and make CC=..something.. , so I can access to normal linux kernel targets (vmlinux, modules, menuconfig ...) is it that possible? Sep 02 10:51:42 * rwhitby comes home from fathers day stuff Sep 02 10:52:51 in Sep? we have that in Jun... Sep 02 10:52:57 we too Sep 02 10:53:12 or maybe rwhitby had a long journey :) Sep 02 10:53:27 the greeting card companies need to spread the profits throughout the year ... Sep 02 10:54:24 it's too cold in june in the southern hemisphere to enjoy a barbie on fathers day Sep 02 10:56:41 openmoko: 03mickey * r2888 10/trunk/src/target/OM-2007.2/artwork/themes/ (75 files in 17 dirs): add openmoko-standard-2-qvga as proposed by Philipp Zabel (pH5) Sep 02 10:56:55 * rwhitby needs to look at how to get the date on the today screen to show the date in the non-american way. Sep 02 10:57:44 showing date as 09/02/07 on an international project is simply unacceptable Sep 02 10:58:07 rwhitby: Are you from Australia? Sep 02 10:58:16 yep Sep 02 10:58:53 * * OM Bug 790 has been created by phil.openmoko(AT)ashbysoft.com Sep 02 10:58:54 * * system clock not reset on resume Sep 02 10:58:55 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=790 Sep 02 10:58:58 "barbie" gave it away ;-) Sep 02 10:59:02 openmoko: 03mickey * r2889 10/trunk/src/target/OM-2007.2/artwork/themes/openmoko-standard-2-qvga/gtk-2.0/ (gtkprogressbar openmoko-calculator): openmoko-standard-theme-2-qvga: sync a bit w/ vga theme Sep 02 10:59:19 rwhitby: i'll probably rip it out Sep 02 10:59:28 rwhitby: and make it more textual Sep 02 10:59:38 6th September 2007 Sep 02 10:59:39 or what Sep 02 10:59:41 not Sep 02 11:00:03 please keep it. my phone must show the date on the home screen :-) Sep 02 11:00:32 yes, date Sep 02 11:00:34 but in textual form Sep 02 11:00:41 Either DD MMM YYYY or YYYY-MM-DD will do. the latter being the ISO standard :-) Sep 02 11:01:05 full textual is fine too of course :-) Sep 02 11:02:01 would be nice to see the gsm carrier on that screen too Sep 02 11:02:14 doesn the locale have info how to display dates? Sep 02 11:02:18 openmoko: 03mickey * r2890 10/trunk/src/target/OM-2007.2/artwork/themes/openmoko-standard-2-qvga/gtk-2.0/ (gtkpaned gtkrc): openmoko-theme-standard-2-qvga: sync gtkpaned w/ vga theme Sep 02 11:02:19 +'t Sep 02 11:03:16 mickey|mdd: line 29 of today-pin-summary.c I think Sep 02 11:03:37 rwhitby: thanks. Sep 02 11:04:00 rwhitby: where is Aus are you from? Im in Melb Sep 02 11:04:05 Adl Sep 02 11:07:53 mickey|mdd: s/%x/%A, %d %B %Y/ Sep 02 11:08:11 rwhitby: is that textual form? Sep 02 11:08:15 yep Sep 02 11:08:22 great. I'll commit that then Sep 02 11:08:24 with weekday name too Sep 02 11:08:33 Friday, 1 October 2007 Sep 02 11:08:35 ah, might be too long horizontally Sep 02 11:08:38 but let's try Sep 02 11:08:57 wouldn't think so, looking at it Sep 02 11:09:40 hmm Sep 02 11:09:44 what's that %I ? Sep 02 11:09:59 that's the hour, 12 hr format Sep 02 11:10:06 ah Sep 02 11:10:08 we don't need that Sep 02 11:10:10 it's in the panel Sep 02 11:10:13 http://www.cppreference.com/stddate/strftime.html Sep 02 11:11:00 * ScaredyCat prefers 24 hour fmt Sep 02 11:11:02 * rwhitby doesn't bother with bug trackers, he just harrasses those with commit access directly :-) Sep 02 11:11:09 hehe Sep 02 11:11:12 that's the spirit Sep 02 11:11:18 although today is OpenEZX hacking day Sep 02 11:11:21 and i want to concentrate on that Sep 02 11:11:33 24h clock++ Sep 02 11:11:46 I knew that mickey|mdd wouldn't let an americanism stay there too long when pointed out ;-) Sep 02 11:12:18 ScaredyCat, CM: the one in the panel is 24h Sep 02 11:12:31 quite right too! Sep 02 11:12:37 so if mickey|mdd takes the time out of the today-pim-summary, then everyone is happy Sep 02 11:13:14 yes, don't need it twice I guess Sep 02 11:13:34 he could replace it with the gsm carrier currently reg'd :D Sep 02 11:14:05 I prefer not to give them advertising space on my phone's screen. Sep 02 11:14:07 let me test that and then commit Sep 02 11:14:09 indeed, the openmoko bit Sep 02 11:14:18 could be the carrier name.. Sep 02 11:14:32 rwhitby: I like to see which one I'm using... Sep 02 11:15:55 mickey|mdd: how do you personally test a change like that? it's in SVN, but you can't rebuild the bitbake package until you've checked in the svn, so where do you make the change and build? Sep 02 11:16:30 rwhitby: i often use bitbake devshell Sep 02 11:16:38 ~praise RP Sep 02 11:16:39 All hail RP! Sep 02 11:16:55 but that doesn't break the OE->SVN dependency of all the openmoko packages Sep 02 11:17:19 so how do you get around that? do you make the change twice? (once in tmp/work/.. and then again in svn?) Sep 02 11:18:14 yes Sep 02 11:18:17 that's what i'm doing Sep 02 11:18:23 first testing as quilt patch Sep 02 11:18:28 then later applying the patch to svn Sep 02 11:18:49 there's gotta be a better way ... Sep 02 11:18:54 sure Sep 02 11:18:58 using an external toolchain Sep 02 11:19:01 which we'll have soon Sep 02 11:21:42 rwhitby: We should make it so the workdir is a valid svn checkout and there you could just commit directly from the workdir... Sep 02 11:22:10 yeah, that would be good. Sep 02 11:22:47 need some way to massage svn urls to change anon access to developer access. or I guess you just svn switch each time you want to commit something. Sep 02 11:31:29 Guys, Im trying to run an executable that used to work but since upgrading to 2007.2 i get a 'Not Found" error Sep 02 11:32:37 executable abi format has changed Sep 02 11:32:58 2007.1 was oabi, 20072. is eabi. the two are not compatible. Sep 02 11:35:44 oh ok. Not exactly sure what that means but I shall look it up Sep 02 11:36:48 is there a picture viewer for openmoko? Sep 02 11:40:36 hi Sep 02 11:44:26 dcordes: that'll be a software question, answers to those are usually.. not yet, feel free to chip in :) Sep 02 11:48:24 I just tried gpe-gallery Sep 02 11:49:12 working great with moko ui. also kind of finger usable (except when it comes to opening a folder to view: gpe file locator is too small) Sep 02 11:49:45 dcordes: gpe-gallery is the current picture viewer for openmoko ;-) Sep 02 11:52:04 regarding the eabi change. Unless i had the source and could recompile the binary then there is no hope correct? Sep 02 11:52:34 correct, unless you run an oabi chroot. If you're talking about gllin, then someone is working on an oabi chroot for that. Sep 02 11:53:08 is there a way to get those scrollbars bigger? Sep 02 11:53:43 (script) wiki RecentChanges: [[User:Phlash]] Sep 02 11:54:14 lol, you already answered my next question. I guess I'll wait patiently :-) Sep 02 11:54:58 I was saving you from having to disclose your possession of something which you shouldn't have. Sep 02 11:55:18 ty Sep 02 11:55:38 hahaha Sep 02 11:57:29 http://www.youtube.com/watch?v=9TOXAzh7vA4 Sep 02 12:00:35 So what's going on with gllin and the licensing problems ? Sep 02 12:02:04 I to am interested to know. Sep 02 12:02:26 hi Sep 02 12:02:42 i'm trying to emulate OpenMoko on qemu Sep 02 12:02:50 Federico2: nice :) Sep 02 12:02:51 i'm using Fedora 7 Sep 02 12:03:29 i get this error Sep 02 12:03:30 WARNING: "gcc" looks like gcc 4.x Sep 02 12:03:30 Looking for gcc 3.x Sep 02 12:03:30 gcc 3.x not found! Sep 02 12:03:30 QEMU is known to have problems when compiled with gcc 4.x Sep 02 12:03:30 It is recommended that you use gcc 3.x to build QEMU Sep 02 12:03:32 To use this compiler anyway, configure with --disable-gcc-check Sep 02 12:03:34 make: *** [setup-qemu] Fehler 1 Sep 02 12:03:38 how can i solve this? Sep 02 12:03:38 elmarb: See http://wiki.openmoko.org/wiki/FAQ (and also SH1, SH1_FAQ, Neo1973, Developer_preview and Neo1973_Hardware#GTA01Bv4 pages) (Hopefully these links answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Sep 02 12:05:43 * ScaredyCat watches it via neuros Sep 02 12:06:07 elmarb: yum install compat-gcc-34 Sep 02 12:06:20 aloril: i already read the FAQ but i did not finf anything Sep 02 12:07:31 rmoravcik: okay thanks I'm installing ;-) Sep 02 12:08:44 elmarb: make sure you update the wiki page which contains the instructions you were following too Sep 02 12:08:59 where are the openmoko videos in a playable video format and not for the flash banner ad player that i boycott? Sep 02 12:09:31 http://people.openmoko.org/ninjutsu/ Sep 02 12:10:36 thanks a lot, that will be much better and brighter Sep 02 12:19:57 i would like to have the neo for a more powerful gp2x, n800 with built in modem and has a tomtom navigator - I need a d-pad and 2 - 4 buttons on the hardware side at least - maybe there's another irc channel for end users Sep 02 12:26:26 is the buildhost down again? Sep 02 13:02:54 appears to be dead borg_ Sep 02 13:03:23 damn Sep 02 13:04:00 right, ssh'd to qemu image. Sep 02 13:04:58 Q, I edited /openmoko/trunk/src/target/OM-2007.2/panel-plugins/openmoko-panel-battery/src/openmoko-panel-battery.c Sep 02 13:05:19 how do I make an image, or package that uses the changes? Sep 02 13:05:38 the `make` command on this thing has apparently been ran somewhere else... Sep 02 13:09:08 Kero: I'm not sure, but try: cd build; source ../setup-env; bitbake -c rebuild openmoko-panel-battery Sep 02 13:13:17 Hey. I had to travel for about a week so I left my neo unplugged from my comp. when I came back it woudnt turn on so I figured that it had lost its charged, so I have had it connected to the comp for 1 day now but it still wont turn on. any ideas? Sep 02 13:14:51 calamous, http://wiki.openmoko.org/wiki/Neo1973_Battery_Charger#from_deep_discharged_battery Sep 02 13:16:54 CM: that takes a fresh SVN checkout, but not via the location I edited. Sep 02 13:18:42 jeddy3, Ty, trying it now. Let you know in an hour ;p Sep 02 13:19:37 calamous, :) Sep 02 13:20:43 but, it does give me the configure command, so I can continue with my sources now, I think. Sep 02 13:22:12 openmoko: 03mickey * r2891 10/trunk/src/target/OM-2007.2/applications/openmoko-today2/ (ChangeLog src/today-pim-summary.c): openmoko-today2: Switch to textual display of date, remove time (it's in panel anyways). Thanks Rod Whitby. Sep 02 13:23:41 rwhitby: here we are Sep 02 13:24:58 hell, no, that configure line is only a tiny piece of the puzzle. Sep 02 13:26:08 Kero: Maybe you should use quilt and patch your changes into the build? (Not that I know how..) Sep 02 13:26:17 'quilt' ? Sep 02 13:26:50 * Kero is going to look at the wiki for some page "how to change things in your home-baked image" Sep 02 13:28:41 Anyone know anything about any GSM modem update to make at&t 3g sims working? Sep 02 13:29:11 Cause I can't find anything new on bugzilla #666 or the wiki page Carriers/ATT Sep 02 13:38:40 slowly, I drag some info from the mailthread 'Adding Applications To OpenMoko The Right Way' and wiki pages referenced from there. Sep 02 13:41:00 Good morning> I'd like to remove qemu from OE so it will compile on an x86_box with gcc-4.x. How can I remove it? Sep 02 13:41:53 (The RAID card on the i386 machin is failing, and I can't get through a build.) Sep 02 13:51:14 DukeOfURL what hinders you to install gcc 3.4 additionally? Sep 02 13:51:46 qemu is needed for 2 things: as emulator (not built by oe, but by mokomakefile) and for locales Sep 02 13:54:03 (script) wiki RecentChanges: [[Business:Mobile_photo_frame]] [[Application_Development_Crash_Course]] [[Vertical_applications]] [[Business_Development]] [[Main_Page]] [[User_talk:Thseiler]] [[User:Thseiler]] Sep 02 14:01:20 * Kero finds it scary that his qemu image actually has variations in GSM signal strength... Sep 02 14:11:28 * Kero restarts xserver and... has an 'upgraded' battery applet! Sep 02 14:17:38 Kero: these vatiations in qemu are on purpose.. else you would never see if the applet is alive or not Sep 02 14:20:49 eh variations Sep 02 14:22:37 makes sense. scary nevertheless. Sep 02 14:23:02 wiki edit time, I guess. Sep 02 14:24:04 and time to dig up my old build-ipkg scripts and tailor them for single-package building (so I have those ./configure cmd and various PATHs documented for myself) Sep 02 14:24:49 then onto building bigger fish, see if I can compile Ruby that way. Sep 02 14:31:36 jeddy3, My neo turn on again. Thanks! Sep 02 14:32:09 calamous, good to hear! Sep 02 15:01:56 where are the program launch entries from "+" in today saved? Sep 02 15:03:28 dcordes: IIRC they are *.desktop files, search for them Sep 02 15:04:32 k Sep 02 15:05:12 Anyone know how to upgrade the GSM modem's firmware? Sep 02 15:06:15 I thought it would not be possible/allowed to access gsm rom? Sep 02 15:06:33 I guess using the jtag it's possible to flash it Sep 02 15:06:44 but the firmware is proprietary and closed Sep 02 15:07:31 thomasg, do you know if I can ever get my neo's firmware upgraded so I can use at&t sims? Sep 02 15:08:31 thomasg, or will I need to get a neo that already has a working firmware? Sep 02 15:08:34 well, the problem is, that AT&T only uses the 850 MHz band. and afaik the neo has no approval from the FCC for this band Sep 02 15:09:19 I think (but thats only my opinion!) that you can get a firmware upgrade for the gsm-module that allows to use this band - as soon the neo has the approval for the states Sep 02 15:09:43 thomasg, the neo dosent have FCC approval in the US yet? Sep 02 15:09:56 it has, but only for 900/1800/1900 mhz Sep 02 15:10:23 but I don't have further informations, too. Sep 02 15:10:23 I think the + icons are in /usr/share/applications Sep 02 15:10:49 Ok, so even though its quadband, since they don't have 850mhz permission its disabled in the firmware? Sep 02 15:10:55 it's a pitty that the US goverment don't like international standards - it makes the things much more difficult Sep 02 15:11:24 I know, I've seen specturm charts and recomendations and we don't follow any :( Sep 02 15:11:32 calamous, I know that the chip is quadband. I also know that there is no 850 mhz approval. but I don't know if/how the activation wil be possible Sep 02 15:11:53 It's possible that activating that band is a simple AT command Sep 02 15:12:00 cesarb, might be Sep 02 15:12:06 thomasg, alright. Where do you live by the way? Sep 02 15:12:10 europe Sep 02 15:12:14 On the Razr, you can chose between 850/1900, 900/1800, or quadband Sep 02 15:12:18 Europe is a big place, :) Sep 02 15:12:26 germany :) Sep 02 15:12:37 So, it wouldn't surprise me if the Calypso also had a way to chose the band Sep 02 15:12:59 cesarb, too bad all the info on the calypso is closed Sep 02 15:13:12 cesarb, well, it surely has, because it also uses one band at the same time and can chose between 3 or 4. Sep 02 15:13:13 (and it also wouldn't surprise me if that setting was persistent between powering on/off of the modem) Sep 02 15:13:40 calamous: even with the closed info, someone has already found engineering mode commands... Sep 02 15:14:03 thomasg: well, it autodetects. the trick would be to change the list of bands it will autodetect in Sep 02 15:14:06 cesarb, wow, thats great. do you know if they are posted in public? Sep 02 15:14:23 calamous: no, someone found a gsm modem with similar commands ;-) Sep 02 15:14:26 there is a band command but its disabled Sep 02 15:15:07 Writchie|: is it a standard 07.07 command? if it is, the real command might be a vendor one, and the 07.07 command simply not implemented Sep 02 15:15:21 it's not a standard command Sep 02 15:15:25 Anyone know what "GSM baseband download mode" is? Sep 02 15:15:33 Writchie is an at% command? Sep 02 15:15:50 wait one Sep 02 15:16:07 at% are the propriarty commands right? Sep 02 15:16:12 * cesarb is interested in GSM850 because one of the carriers here is either CDMA or GSM850 Sep 02 15:16:31 (and one of my friends who is interested on the neo uses that carrier...) Sep 02 15:16:40 * calamous Is also intersted in the 850 band Sep 02 15:17:23 * cesarb has already tested, the neo can see 3 carriers, the razr all four... chosing 900/1800 you can see the same three as the neo, chosing 850/1900 you can see the remaining one Sep 02 15:18:54 AT%BAND=20 is the command to set GSM850/1900 Sep 02 15:19:13 AT%BAND=31 sets full quadband Sep 02 15:19:42 these are probably disabled in FIC's configurations Sep 02 15:20:18 Writchie|: do you have a link to the document these came from? Sep 02 15:20:23 good to know; before, somebody here said (probably not knowing better) that it should be fully automatic and you can't switch bands manually Sep 02 15:20:25 Someone Disabled GSM850 accidentally or intentionally and is stonewalling any information about it. Sep 02 15:20:29 Wow, I just looked at the GSM freq ranges, Why does the US have to be different?!? Sep 02 15:20:44 calamous: because the GSM bands were already in use there Sep 02 15:21:34 Perhaps it would be best if everyone threw away their radios and the entrie spectrum was compeletly reallocated. (but that would be very expensive, and unrealastic) Sep 02 15:22:26 Writchie|: I think the problem is that nobody who has already tried 850 filed a bug report on bugzilla Sep 02 15:22:36 Writchie|: bug reports on bugzilla are harder to forget or ignore Sep 02 15:22:51 cesarb: http://www.enfora.com/index.cgi?CONTENT_ID=577&User:LANGUAGE=en Sep 02 15:22:59 * cesarb will try to get a 850 SIM to test when gsmd stops being annoying Sep 02 15:23:19 these guys have modules, some of which are caylpso based. Sep 02 15:24:04 cesarb, at&t prepaid sim Sep 02 15:24:48 Writchie|: you do know there are TWO different firmware for calypso, right? Sep 02 15:24:52 the engineering mode 2 commands for the Enabler IIE work on the neo as documented up to 13 Sep 02 15:24:58 Writchie|: just take a look at gsmd's vendor_* files Sep 02 15:25:19 root@fic-gta01:~$ ./ruby -ve 'puts "Hello, world!"' Sep 02 15:25:19 ruby 1.8.6 (2007-03-13 patchlevel 0) [arm-linux-eabi] Sep 02 15:25:19 Hello, world! Sep 02 15:25:45 the IIE is as edge version and the neo is not. Sep 02 15:26:09 re Sep 02 15:26:19 cesarb: what makes you think there are "two" different calypso firmware bases? Sep 02 15:26:39 Writchie|: the vendor_* files on gsmd Sep 02 15:27:11 gtg, lunch Sep 02 15:27:21 cesarb: AFAIK these related to the HTC which uses a qualcomm chipsek IIRC Sep 02 15:29:43 Here is my output STR=`at%band=?' Sep 02 15:29:43 RSTR=`%BAND: (0-1),(1-31)' Sep 02 15:30:14 so 1-31 means manual-GSM/EGSM/DCS/PCS/850 Sep 02 15:30:35 and 0-1 means automatic-GSM900 Sep 02 15:31:30 It means it is in automatic mode - it's supposed to select any band Sep 02 15:31:55 Is there any way I can force it on the 850 band? Sep 02 15:32:28 re Sep 02 15:32:47 As I said before, mannually setting the band appears to be disabled. Sep 02 15:33:13 * hhf423 has issues with irc via pidgin via N770 Sep 02 15:33:31 ScaredyCat, where are those images you mentioned supposed to appear? Sep 02 15:33:42 Damn Sep 02 15:34:36 At the present time, the GTA01BV4 is not legally registered in US to transmit on GSM850 - at least according to the publicly filed documents. Sep 02 15:35:16 could the chip even do 850? Sep 02 15:35:19 i don't mind this at all at this stage of developement - what I mind is FIC/opemmoko.com being completely silend Sep 02 15:35:47 while people waste valuable time trying to make works things that FIC knows full well is disabled. Sep 02 15:36:05 Wow, that makes me angry Sep 02 15:36:15 where can I find an overview of applications used in openmoko currently? (like text editor, image viewer and so on) Sep 02 15:36:16 the chipset is quadband Sep 02 15:36:40 Bug #666 talks about the at&t sims and there has been no activity on it for a month Sep 02 15:36:55 however, antennae, impedeances, etc. all have to be proper for quadband. Sep 02 15:37:18 "Upgrading the modem's firmware is technically possible but no proper software is currently legally available to users outside OpenMoko staff." Sep 02 15:37:18 Writchie: one thing I do appreciate about the whole OpenMoko information availability is that it's a hell of a lot more than is typically available publicly... or even to operators at this stage of development Sep 02 15:37:23 #666 problem and its timing are not unusal Sep 02 15:37:43 Spehpmw: so do I Sep 02 15:37:51 in 2007.2 i have a gtk+ dialog which asks for password and the keyboard vanishes at that point...anyone know how to bring back the keyboard? Sep 02 15:39:21 * calamous cries over NDAs Sep 02 15:39:41 zedstar, maybe it toggles the keyboard, try entering the password dialog with keyboard turned off Sep 02 15:39:54 calamous: NDAs are part of what made me quit my last job Sep 02 15:39:57 hit the entry w the stylus? Sep 02 15:40:03 calamous: I'm getting sick of them Sep 02 15:40:09 Stephmw: It is actually FIC who is loosing out because there are a fair number of people here that could help them quite a bit with low level things if there were a bit more forthcoming. Sep 02 15:40:19 Yeah Sep 02 15:40:25 NDA is not the problem. Sep 02 15:40:43 kero: how u mean? Sep 02 15:40:43 zedstar, they might have changed it previous behaviour was putting up double keyboards in a keyboard toggling input, still not right...but Sep 02 15:41:28 Stpehmw, I feel that the whole copyright and patent system needs a reform. Sep 02 15:41:35 i think cos the prompt hides characters (password prompt) it disables keyboard and dunno how to get it back if i click in the field for normal entry it appears Sep 02 15:42:03 But, lets take a guess, do you think we will hear from FIC about the firmware / 850 band first, or ATI will opensoure their graphics driver Sep 02 15:42:30 calamous: patents and copyrights are a separate issue to NDAs Sep 02 15:43:06 ah ok solved it...can bring the other keyboard up from input manager and works Sep 02 15:43:09 Yes, but I was always under the impression that patents existed to protect inventions so you could make the information public to spread knowledge Sep 02 15:43:39 calamous: yes, it's a conditional monopoly.... but that's still less desirable than trade secrets Sep 02 15:43:49 calamous: Patents exist you could make money from inventions Sep 02 15:43:55 calamous: and with encrypted ROMs and stuff, trade secrets are much more attractive Sep 02 15:44:18 Stephmw, But with a trade secret, if someone figures it out they can publish it right? Sep 02 15:44:27 The only protection is that it is "secret" Sep 02 15:44:28 jeddy3: my images? Sep 02 15:44:45 that 2 keyboard thing is confusing Sep 02 15:44:48 calamous: hence NDA... whoever figures it out has to be damn sure they can prove they did so legally Sep 02 15:44:51 ScaredyCat, you mentioned something about 7h left this morning Sep 02 15:45:03 patents are a license to sue and be sued - little more Sep 02 15:45:14 calamous: with the litigious name of the game these days, it's not worth going public Sep 02 15:45:57 ahh well, they're done :) Sep 02 15:46:13 http://buildhost.automated.it/OM2007.2/ Sep 02 15:46:39 Stephmw, this is off the subject of computing, but I think this example demonstrates the best abuse of the patent system. Have you heard of the drug either nexium or Prilosec? Sep 02 15:47:17 http://automated.it/buildhost/OM2007.2/?M=D :) better view Sep 02 15:47:23 calamous: no, can't say I have Sep 02 15:47:53 scaredycat-openmoko? Sep 02 15:48:07 there are 2 Sep 02 15:48:07 ah, scummvm etc... Sep 02 15:48:26 the standard Openmoko- image,and an experimaental one of mine Sep 02 15:48:27 They are "heart burn" drugs. In chemistry an enantiomer is just a mirror image, so your right hand is an enantiomer of the left, a mirror image. Sep 02 15:48:39 * ScaredyCat points at the readme text at the bottom of the page Sep 02 15:49:03 Those two drugs are just a mirror image of eachother, and when the patent ran out on the first drug, they got it on the second Sep 02 15:49:09 whats the difference between these, and the ones on buildhost? Sep 02 15:49:17 http://automated.it/buildhost/OM2007.2/OpenMoko-rootfs-glibc-ipk-P1-August-Snapshot-20070902-fic-gta01.rootfs.jffs2 Sep 02 15:49:26 is probably a good one for you. Sep 02 15:49:45 jeddy3:depends when thay get built :) Sep 02 15:49:51 ScaredyCat, ah :) Sep 02 15:50:15 and just grab latest (1/9) uimage Sep 02 15:50:24 ya Sep 02 15:50:37 http://automated.it/buildhost/OM2007.2/uImage-2.6.22.5-moko11-r2-fic-gta01.bin Sep 02 15:51:41 modules are just kernel modules ment for extracting into filesystem, but the openmoko-devel image contains these modules also right? Sep 02 15:52:08 ya, however openmoko-devel-image does not contain all modules Sep 02 15:52:14 module-foo.tar.bz2 contains all Sep 02 15:52:15 mickey|mdd, will you be at the next CCC Congress in berlin? Sep 02 15:52:35 mickey|mdd, ah, how so? Sep 02 15:52:46 you shouldn't need that, unless you want to add stuff like ext2 for example Sep 02 15:52:54 thomasg: i'm not sure. it always conflicts a bit with my birthday (26th) and silvester ;) Sep 02 15:53:04 hehe Sep 02 15:53:10 jeddy3: *all* modules is big Sep 02 15:53:51 that would be 44mb image + 24 odd mb modules (compressed) Sep 02 15:53:52 ScaredyCat, nod, but the needed (sd-card, etc...) are in openmoko-devel-image? Sep 02 15:53:57 yes Sep 02 15:54:03 right, thanks Sep 02 15:54:03 (script) wiki RecentChanges: [[Student_phone]] [[Business_phone]] [[Vertical_applications]] [[Talk:Wish_List]] [[User:Jcwang]] [[Wish_List]] [[Mobile_photo_frame]] [[Category:Business]] [[Business_Development]] [[Main_Page]] Sep 02 15:54:32 everything you need should be in there... it's only if you're trying 'other' stuff :) Sep 02 15:57:33 ScaredyCat: striped modules has ~2Mb compressed Sep 02 15:57:59 all ? Sep 02 15:58:17 ScaredyCat: yes Sep 02 15:58:24 ScaredyCat: does dialing work from boot on your builds, or is the stop/start trick still needed? Sep 02 16:00:32 jadams: the Openmoko- builds are just standard builds. Sep 02 16:00:56 the *scaredycat* builds are.. errm slightly different from standard Sep 02 16:02:18 scummfromtehgetgo! Sep 02 16:02:29 what's myresolv? Sep 02 16:03:07 also, here's a big gripe I've always had about these default file listings...why can't I set the number of characters it shows me in the filename? sigh... Sep 02 16:03:26 I always use firefox fullscreen on one desktop, and things like that make me bring it down so I can have my status bar to see the filename Sep 02 16:03:29 jadams: myresolv has been removed.. :) Sep 02 16:03:55 I think I'm going to attempt to bt network this sucker through my n770 today Sep 02 16:04:19 first, I'll flash this guy with your latest scaredycat image...then I'll flash the 770 with os2007 hacker edition Sep 02 16:04:30 then just a bit of ifconfig, eh? Sep 02 16:05:01 :) Sep 02 16:05:12 good luck... Sep 02 16:05:17 because I want the same :D Sep 02 16:05:35 I'll let you know... Sep 02 16:36:47 I'm wondering why is the build "interleaved"... I see it compile webkit-gtk, compile several other packages, and hours later come back to webkit-gtk for a "do_package" Sep 02 16:37:16 Wait, pcmciautils? This is getting ridiculous Sep 02 16:40:24 task-based granularity Sep 02 16:40:24 some tasks depend on other tasks Sep 02 16:41:23 ScaredyCat: how hard is it to get edje-viewer built? Tried? Sep 02 16:41:28 and if those are spaced out enough you get better parallelization Sep 02 16:41:44 not tried tbh Sep 02 16:42:02 tbh? Sep 02 16:42:28 to be honest Sep 02 16:42:39 Also, I've *got* to get my GPRS routing my internet. I have unlimited internet and I've used it for like 15 seconds total Sep 02 16:42:43 aha Sep 02 16:42:46 my acronym-foo is weak Sep 02 16:43:19 :x Sep 02 16:43:21 :P Sep 02 16:43:31 HOLY CRAP THOSE LOOK JUST LIKE FACES Sep 02 16:44:12 the :x is Miffy the cat? (-: Sep 02 16:44:46 jadams: just building edje-viewer Sep 02 16:46:43 0.3svn.R563 here Sep 02 16:46:55 sorry, wrong channel Sep 02 16:48:01 ScaredyCat: awesome. By the time I get everything downloaded and flashed between the phone and the 770 it'll be gravyu Sep 02 16:48:04 -u Sep 02 16:48:41 warning, do not put gravy on your neo or 770. Sep 02 16:49:34 what, they don't taste better then? Sep 02 16:55:42 only wiht cheese Sep 02 16:56:04 http://scap.linuxtogo.org/index.php?page=1 <--- what is this "satscan"? Sep 02 16:56:11 http://scap.linuxtogo.org/files/488c9fee2eb8eb488ffc45e2003184f3.png Sep 02 16:57:27 borg_, in the mailinglist Sep 02 16:57:39 scap.linuxtogo.org is fun Sep 02 16:57:49 pretty cool things there nowadays Sep 02 16:57:54 yes :) Sep 02 16:59:05 satscan sounds like the kind of thing the sphyrna guys would do... Sep 02 16:59:06 who did this battery monitor, btw? - i guess i don't have to add this to settingsgui anymore then... :-) Sep 02 16:59:20 borg_, but can't seem to find post now... Sep 02 16:59:30 :P Sep 02 16:59:34 cesarb, yes it's from the sphyrna project Sep 02 17:00:34 borg_, here http://lists.openmoko.org/pipermail/device-owners/2007-August/000408.html Sep 02 17:00:54 thx :) Sep 02 17:02:38 now it's my turn :)...what is this "treemodelIF..." http://scap.linuxtogo.org/files/55f8e257ff776196576b0e604383fe2a.png Sep 02 17:04:33 * cesarb thinks this build is a lot like that silly duracell bunny Sep 02 17:04:38 It keeps going and going and going... Sep 02 17:05:03 jeddy3 thats from Lars Hallberg see http://lists.openmoko.org/pipermail/community/2007-September/009925.html Sep 02 17:05:21 kristian-m, thanks Sep 02 17:09:15 hmm Sep 02 17:10:14 i'm trying to install scummvm from http://buildhost.automated.it/scummvm/ but ipkg just spits out "Terminated" - even with -V 3 Sep 02 17:14:35 is anyone here from Madrid? Sep 02 17:36:41 jadams: it's done Sep 02 17:36:59 untested, but it's available from my repo Sep 02 17:39:23 ScaredyCat: I'll test it as soon as this latest image flash is done (halfway through, scaredycat image) Sep 02 17:39:34 man nothing's better than fresh squeezed grapefruit juice Sep 02 17:45:10 how is the pin managed, btw? Should a dialog pop up or is it to be stored in a config-file? Sep 02 17:46:09 riot, as far as is know it is not allowed to store the pin somewhere Sep 02 17:47:05 mdt: well, the phone never asks me for one, but watching the logs, i can see that upon gsm-powerup the modem asks for a pin.. Sep 02 17:47:23 that may be a gui problem Sep 02 17:48:10 i don't know if openmoko follows gsm specs in any way (like emergency calls and other base requirements for mobile phones) Sep 02 17:48:42 I'd not wanna test that... Sep 02 17:50:32 shit.. i bitterly need a new HDD.. no space left for any build-environment.. and i don't wanna build notepad-apps (i blatantly stole one from the pygtk manual btw, works great, even i18n), when i cannot even make/receive calls properly Sep 02 17:53:33 is u-boot missing from OM2007.2 ???? Sep 02 17:54:04 * SpeedEvil passes riot 6 DVD-rs. Sep 02 17:54:39 ;) Sep 02 17:55:01 erm, so I get "root@(none)" for my prompt, and hostname foo doesn't seem to make it root@foo...any ideas? Sep 02 17:55:45 jadams: tried restarting the shell after changing the hostname? Sep 02 17:56:18 cesarb: just now rebooting, so it'll do that for me I suppose :) Sep 02 17:57:24 jadams: if it doesn't set the hostname back (at least on what I recall from Debian's initscripts, hostname is set on boot from /etc/hostname) Sep 02 17:59:02 openmoko: 03mickey * r2892 10/trunk/src/target/OM-2007.2/libraries/libmokopanelui2/ (5 files in 2 dirs): libmokopanelui2: use MBPanelScalingImage as icon Sep 02 18:04:06 wow, updates to lots of stuff coming down with ipkg update && ipkg upgrade, hurrah Sep 02 18:04:51 cesarb: etc/hostname is a directory... Sep 02 18:06:02 what are the odds something like an xmms always-on-top would work in matchbox? Sep 02 18:06:19 jadams: /etc/hostname should be a file. Sep 02 18:06:37 riot: alright, then something awkward is going down on my fs Sep 02 18:06:50 ruby runs, feed http://chmeee.dyndns.org/om/ruby/feed provided (and mail to ML sent) Sep 02 18:07:15 either this, or some tool b0rked your hostname-file. should be a simple text-file containing just your hostname (without domain et al) Sep 02 18:07:19 * Kero thinks he earned a beer :) Sep 02 18:07:51 Kero: awesome awesome Sep 02 18:07:56 Kero: where do you live? Sep 02 18:08:15 jadams: hm... do you have a post-2007.2 uboot? Sep 02 18:08:18 this was on my list, but work keeps getting in the way Sep 02 18:08:25 cesarb: I don't Sep 02 18:08:31 jadams: and if you don't, did you run "nand erase rootfs" before flashing the rootfs? Sep 02 18:08:33 I don't guess... Sep 02 18:08:36 I didn't Sep 02 18:08:47 jadams: netherlands Sep 02 18:09:18 jadams: if you don't have a post-2007.2 uboot (the one from sean's snapshot announcement is great), and don't run "nand erase rootfs" before flashing the rootfs, you will get jffs2 filesystem corruption Sep 02 18:09:39 jadams: either a recent uboot or running "nand erase rootfs" before flashing the rootfs avoids it Sep 02 18:09:59 cesarb: alright, thanks. I'll flash the uboot shortly, for now I have to play with ruby on my corrupted fs :) Sep 02 18:10:10 and check out edje-viewer that ScaredyCat built Sep 02 18:10:16 jadams: so, you probably will have to reflash the rootfs Sep 02 18:10:29 openmoko: 03mickey * r2893 10/trunk/src/target/OM-2007.2/panel-plugins/ (4 files in 4 dirs): Sep 02 18:10:29 openmoko: openmoko-panel-usb: catch up with libmokopanelui2 changes Sep 02 18:10:29 openmoko: openmoko-panel-gps: catch up with libmokopanelui2 changes Sep 02 18:10:35 jadams: so you can do gtk bindings for ruby now, ok? :) Sep 02 18:11:20 jadams: if you don't have a debug board nearby (either yours or someone else's), you should only flash uboot if you are sure you know what you are doing (and have a UPS or a battery-backed computer!) Sep 02 18:11:37 jadams: (same goes for playing with the uboot environment, in fact) Sep 02 18:12:31 cesarb: my debug board's at the office, I'll wait til tuesday or wednesday to flash uboot to avoid going in over the holiday Sep 02 18:21:22 meh, corrupt fs caused problems...couldn't launch edje-viewer, but only because terminal was screwed up Sep 02 18:21:26 sih, starting over Sep 02 18:22:07 jadams: it's best to do an "erase nand rootfs" and reflash the rootfs Sep 02 18:22:23 jadams: so you won't waste time tripping over a broken rootfs Sep 02 18:23:09 A question to all who use ipkg feeds: which networking do you use? cdc_ether? bluetooth pan? GPRS? Sep 02 18:23:13 cesarb: I'm doing that now Sep 02 18:23:31 I use usb gw (cdc_ether, right?) Sep 02 18:23:32 s/erase nand rootfs/nand erase rootfs/ Sep 02 18:23:38 yeah, yeah Sep 02 18:23:42 installing cu atm Sep 02 18:24:05 * cesarb would like to know why Ctrl-C'ing gsmd crashes the whole phone Sep 02 18:26:54 * riot would like to know, what component is responsible for pin-entry. The gsm-applet doesn't have anything like that in its code. Sep 02 18:27:07 cu? :/ Sep 02 18:27:34 how long does nand erase rootfs take? Sep 02 18:27:47 from memory, a few secs only Sep 02 18:27:59 yeah, it finished Sep 02 18:27:59 hi riot :-) Sep 02 18:28:09 riot its the dialer, i guess Sep 02 18:28:13 SpeedEvil: so, theres no ui-stuff to really setup the gsm? Sep 02 18:28:15 do I need to do anything special to leave the uboot env? no such commands as quit or exit... Sep 02 18:28:32 kristian-m: hi. Are you sure?? Didn't see any pin-entry. And pin is being asked before i fire up any dialer. Sep 02 18:28:49 jadams: it makes no sense to quit or exit from uboot, since it's the operating system Sep 02 18:28:53 heh, hangup works fine Sep 02 18:28:55 ahh Sep 02 18:28:56 jadams: you can power-off, reboot or boot Sep 02 18:29:01 hnf. the dialer's readme is just wonderful.. Sep 02 18:29:04 I just meant quit my terminal session Sep 02 18:29:15 but it just hung up on me, so easy enough Sep 02 18:29:26 jadams: heh, you just got hit by the timeout then ;-) Sep 02 18:29:37 riot the dialer is at leaseed used to enter the pin Sep 02 18:30:11 cesarb: see http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=788 Sep 02 18:30:18 kristian-m: oh yes, i see, there's a source-file containing such code.. Sep 02 18:31:05 hmmm, this shouldn't have anything to do with the dialer, imho. Sep 02 18:31:33 riot: ditto Sep 02 18:32:21 oh, but it actually can send the pin... nice thing is, it displays the pin, not some password-character.. Sep 02 18:32:24 evil, evil. Sep 02 18:32:55 riot: pin requrest for bluetooth and gsm PIN1 and PIN2 etc should be handled the same way Sep 02 18:33:03 and i think they eventually will Sep 02 18:33:52 hmm. I'd rather like to store the pin and use some other means of "authentication/authorization" like a unix-password mechanism ;) Sep 02 18:34:28 Writchie: well, generally, there should be a nice dialog to enter such data.. esp. since bt-pins are likely to be different quite often Sep 02 18:34:33 mwester: thanks, submitted my accidental finding to it Sep 02 18:37:05 what exactly does AT+COPS=0? Is it normal, that it gets retransmitted every few seconds? Sep 02 18:37:16 riot: so will GSM pins in the case where more features are implemented Sep 02 18:37:38 has work on an SMS send/recv app started yet ? Sep 02 18:37:40 riot: you have an old dialer, right? Sep 02 18:37:52 riot: rwhitby (IIRC) fixed that this week Sep 02 18:38:06 cesarb: the stuff is upgraded since about 4 hours. Sep 02 18:38:10 riot: it now will only hang after sending a single AT+COPS=0 instead of looping Sep 02 18:38:50 cesarb: hmm. so COPS=0 is a request for registration? Sep 02 18:39:14 riot: well, it's what is sent when you ask libgsmd-tool to register, so I guess it is Sep 02 18:39:29 at+cops=0 is a request to logon to network 0 Sep 02 18:39:32 riot: the +CFUN=1 (just above it) is sent when you tell it to power on the antenna Sep 02 18:39:35 in the lsit of at+cops=? Sep 02 18:39:47 SpeedEvil: thanks, was too lazy to look it up in the spec ;-) Sep 02 18:40:03 cesarb: yes, that works, except gsmd sometimes seems to fail in tracking the state. Sep 02 18:40:09 its a request to automatically register default network Sep 02 18:41:13 !ombug 766 Sep 02 18:41:14 * * Bug 766, Status: NEW, Created: Unknown Sep 02 18:41:15 * * cesarb(AT)cesarb.net: gsmd misbehaves on back-to-back AT+CFUN=1/AT+COPS=0 Sep 02 18:41:16 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=766 Sep 02 18:41:18 !ombug 754 Sep 02 18:41:18 * * Bug 754, Status: RESOLVED (DUPLICATE), Created: Unknown Sep 02 18:41:19 * * yoyo(AT)ghost.pl: uicmoc4-native_4.3.1.bb Sep 02 18:41:21 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=754 Sep 02 18:41:23 oops Sep 02 18:41:27 !ombug 574 Sep 02 18:41:28 * * Bug 574, Status: NEW, Created: Unknown Sep 02 18:41:29 * * mickey(AT)vanille-media.de: timing issues / buffer parsing issues in gsmd Sep 02 18:41:30 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=574 Sep 02 18:41:44 riot: see these two bugs Sep 02 18:45:19 cesarb: hmm. They all come further than i do. Sep 02 18:45:38 and now the matchboxpanel died upon deactivation of the gsm. Sep 02 18:46:42 riot: GSM on the neo supports 16 various passwords - at least with the SIM i am using Sep 02 18:46:44 riot: you can try the following: Sep 02 18:46:50 riot: stop gsmd Sep 02 18:46:54 riot: start it again Sep 02 18:47:00 jadams: update you image Sep 02 18:47:02 riot: on libgsmd-tool -m shell, type "O" Sep 02 18:47:06 riot: wait a few seconds Sep 02 18:47:11 riot: type "r" Sep 02 18:47:22 ScaredyCat: which image? I nand erased rootfs and just reinstalled the scaredycat openmoko-devel rootfs image Sep 02 18:47:30 riot: that should be enough for it to register, now restart the GUI (restart X completely) Sep 02 18:47:41 download the scaredycat image again Sep 02 18:47:51 rebooting it right now, then i have to do my post-install networking stuff to change its default IP to 192.168.1.202 so it doesn't conflict with the real network here on .0 Sep 02 18:47:53 rootfs Sep 02 18:47:58 ScaredyCat: aha, k Sep 02 18:48:16 cesarb: ok, i'm trying that Sep 02 18:48:25 brb fewd!!! Sep 02 18:49:05 damn, this thing loves hanging when I exchange SIMs Sep 02 18:49:13 * cesarb got a GSM850 one, should be fun Sep 02 18:49:37 Sigh. Sep 02 18:50:00 Just got my neo back from a trip, logging GSM EM=2,2 data - to see if I can do ranging Sep 02 18:50:07 And the GPS died for some reason. Sep 02 18:50:28 cesarb: yep, it seems registered now. Sep 02 18:50:37 cesarb, how to restart X? Sep 02 18:50:45 does the signal-strength-monitoring work? Sep 02 18:50:49 jeddy3: some script on /etc/init.d Sep 02 18:50:54 are you sure the network isi gsm 850 only Sep 02 18:50:56 cesarb, ag Sep 02 18:50:58 cesarb, ah* Sep 02 18:50:59 jeddy3: call it passing "reload" as the argument Sep 02 18:51:06 Writchie: yes Sep 02 18:51:11 Writchie: as sure as google can tell Sep 02 18:51:41 riot: yes - it seems it may also give crude ranges to the tower. Sep 02 18:51:53 yup, now its updated Sep 02 18:51:55 Writchie: the carrier is Vivo (the CDMA carrier), every place I've looked says they use GSM850 Sep 02 18:52:29 what about the audio-profiles? i had a call just some secs ago, but no mic et al. Is it still necessary to switch the alsa-settings by hand? Sep 02 18:52:30 CDMA carrier? how is that going to work with GSM Sep 02 18:53:04 Writchie: they also have GSM now Sep 02 18:53:17 riot if you get python-pygtk running you could try settingsgui Sep 02 18:53:24 Writchie: perhaps they got tired of being the only odd one (all the others in the whole country are already GSM) Sep 02 18:53:59 What do you get from AT+COPS=? Sep 02 18:54:16 kristian-m: i have pygtk runnin Sep 02 18:54:41 Writchie: with my Claro SIM, I got the other three carriers, but not this one Sep 02 18:55:01 kristian-m: you should hang around in #neo1973-germany, too Sep 02 18:55:11 Writchie: since it didn't register yet, somehow AT+COPS=? doesn't work (I also noticed that without a SIM, it only works after you register to a network) Sep 02 18:55:42 * jeddy3 just made his first phone call _with voice_ with his neo Sep 02 18:55:48 it should see all of the networks after you AT+CFUN=1 Sep 02 18:55:49 aah...it's the little things =) Sep 02 18:56:01 Writchie: before registering? Sep 02 18:56:15 yes - other wise you don't know which network to register? Sep 02 18:56:25 if you're doing manual Sep 02 18:56:30 like when roaming Sep 02 18:57:11 Writchie: no~pe... even powering off, powering on again (which sends CFUN=1), it doesn't quite answer to COPS=? Sep 02 18:57:11 it should show whether each network is either your current registered network, available network, or not-allowed network Sep 02 18:57:12 lol! my soundcard is gone, according to alsa. Sep 02 18:57:29 Writchie: COPS=? only works for me after a successful +COPS=0 Sep 02 18:57:44 no modules loaded.. wtf Sep 02 18:57:52 i'm afraid i don't trust anything other than CU to the chipset at the moment. Sep 02 18:58:10 Writchie: cu to the chipset hangs for me, so I have no other option than to trust gsmd Sep 02 18:58:15 do you see more than one network after you register Sep 02 18:59:01 how can i adjust the mouse of my openmoko ( qemu ) Sep 02 18:59:02 Writchie: with my SIM, yes (the three older GSM networks, including mine); with the Vivo SIM (or without a SIM), +COPS=0 doesn't work Sep 02 18:59:29 Writchie, i am working on an "AT Test Suite" application; have a simple prototype; have registered the project at projects.openmoko.org, but, haven't got confirmation Sep 02 18:59:53 Writchie, it has a GUI, and uses libgsmd as backend; if anyone can approve it fast, I can upload the code package; Sep 02 19:00:01 cesarb: Are you running the tty stuff in a parallel session after you start cu ? Sep 02 19:00:09 hey guys :( Sep 02 19:00:38 BasiC_: no idea - what's the problem? Sep 02 19:01:11 what is SettingsGUI on http://scap.linuxtogo.org/ ? Sep 02 19:01:15 what is the response from AT+COPS=0 Sep 02 19:01:26 the "mouse" ( touchscreen ) on openmoko is not where my mouse is pointing Sep 02 19:01:36 mbuf why don't you use an other repository and port it to the wiki? - any screenshots yet? I'm curious Sep 02 19:02:07 jadams http://wiki.openmoko.org/wiki/SettingsGUI *plug again* ;-) Sep 02 19:02:11 cesarb: when you run cu are you using --nostop Sep 02 19:02:14 Stephmw: :(? Sep 02 19:02:58 Writchie: I saw no --nostop on the wiki (where it explains how to use gsm manually) Sep 02 19:03:01 kristian-m: cool. Also, I've got to find out if Xournal is in a repos somewhere, but waiting on an upgrade to finish right now Sep 02 19:03:14 (wtf, this samsung I took the SIM from _loses the time_ if you remove the battery) Sep 02 19:04:21 i cant use mouse @ qemu :( Sep 02 19:04:27 cesarb: it may not be necessary depending on what you are doing but the modem uses hardware flow control and xon xoff must not be used. It may not matter for small data but cu should work. Sep 02 19:04:59 kristian-m, http://shakthimaan.com/downloads/openmoko/images/at-test-suite-window.png - a prototype Sep 02 19:05:00 Writchie: I cannot even type "AT" on cu Sep 02 19:05:07 i have one cu session, for example, that has been up and running through acm port for 3 days Sep 02 19:05:10 Writchie: it simply doesn't answer, or even echo Sep 02 19:05:27 how does one install packages to the SD card? Sep 02 19:05:42 does anyone know if gprs uses the same ppp configuration for every provider? - any differences on pap/chap authentification or the like? Sep 02 19:05:44 is gsmd off Sep 02 19:06:02 Writchie: when I tried, yes Sep 02 19:06:03 kristian-m: its different for each provider Sep 02 19:06:14 * cesarb turns back on the neo after removing the Vivo SIM, which isn't his Sep 02 19:06:21 mbuf thanks - wow thats nice :-) Sep 02 19:06:22 cesarb: I have to run stty -F /dev/ttySAC0 crtscts to get my cu to work. Sep 02 19:06:46 frma71__: stty -F /dev/ttySAC0 -a shows crtscts is already enabled Sep 02 19:06:49 Writchie I was suspecting that... Sep 02 19:07:01 most common is zeros for ip and no username and no password - your mileage may vary Sep 02 19:07:02 :-/ Sep 02 19:07:08 (just wait for this slow thing to boot, and I'll test... Sep 02 19:07:12 APN needs set. Sep 02 19:07:24 APN must be what you provider requires Sep 02 19:07:28 kristian-m: try to search Palm's site for your provider's configuration Sep 02 19:07:43 t-mobile is t-mobile.general.uk Sep 02 19:07:48 (in Uk/) Sep 02 19:07:51 kristian-m: the opera mini pages also have a good list, with APN Sep 02 19:08:31 sony-ericsson has a site that will send an sms to phone to program the specific settings for your carrier. Eventually i hope to grab these messages and translate them to the neo Sep 02 19:08:31 cesarb I've got a working configuration, just want to find out, what options I have to offer in the program Sep 02 19:08:33 i request OM team to put the burden of GForge projects to the community, so we can share their work/burden Sep 02 19:08:50 cesarb thanks, i'll have a look at it Sep 02 19:10:03 ok, I'm booted and killed -9 gsmd Sep 02 19:11:18 root@fic-gta01:~$ cu -l /dev/ttySAC0 Sep 02 19:11:18 Connected. Sep 02 19:11:23 It hangs there Sep 02 19:11:44 ScaredyCat: so the terminal is what was messed up in the other scaredycat image, or no? And how do I know that I downloaded the right one? (seemsto have the same name)... Sep 02 19:12:28 * cesarb does an "echo 0 > /sys/bus/platform/devices/gta01-pm-gsm.0/power_on" foolowed by an "echo 1 > /sys/bus/platform/devices/gta01-pm-gsm.0/power_on", and the phone HANGS Sep 02 19:12:37 I've done that several times in the past! Sep 02 19:12:59 clearly, something WTF is going on... Sep 02 19:13:25 you need more than that - baud rate etc. wait one Sep 02 19:14:00 Writchie: http://wiki.openmoko.org/wiki/Manually_using_GSM says nothing about baud rate, and even then didn't gsmd already set most things up for me? Sep 02 19:14:59 how are new modules updated? My debian's manpage to update-modules says "does nothing - is obsolete" Sep 02 19:15:18 i have three modules-dirs under /lib/modules only one (matching the kernel) containing anything Sep 02 19:15:26 but no modules get loaded Sep 02 19:16:02 modprobe rather tries old module-paths (and fails..) Sep 02 19:17:46 Cool, someone on bug 788 started debugging the crashes with the debug board Sep 02 19:18:34 !om bug 788 Sep 02 19:18:51 !ombug 788 Sep 02 19:18:52 * * Bug 788, Status: NEW, Created: Unknown Sep 02 19:18:53 * * mwester(AT)dls.net: Starting or stopping gsmd completely locks up the Neo Sep 02 19:18:54 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=788 Sep 02 19:18:57 ah Sep 02 19:19:16 oh,funny. ipkg upgrades on the kernel seemingly didn't work Sep 02 19:19:32 riot: the kernel is on a "raw" separate partition, so it's not surprising Sep 02 19:19:41 riot: you have to use uboot to upgrade the kernel Sep 02 19:19:50 Yeah. You have to update the kernel partition with dfu-util (or I suppose internally flash it, but I've never seen instruction for that) Sep 02 19:19:57 why is it upgraded by ipkg then? Sep 02 19:20:07 i was already wondering. Sep 02 19:20:10 There's an internal copy (with modules) too. Sep 02 19:20:50 just wasting space? Sep 02 19:21:16 ScaredyCat: also, fyi edge-viewer doesn't seem to work...at least when I run it ssh'ed in. I'll try directly from the phone, but the error seemed not to be display-related... Sep 02 19:21:24 err, edje-viewer Sep 02 19:21:45 cesarb: i'm running still on an om2007.1 - i use cu -l /dev/ttySAC0 -h -s 115200 Sep 02 19:21:58 the -h sets half duplex so you can see what you type Sep 02 19:24:00 Writchie: cool, will try it later Sep 02 19:24:11 * cesarb now playing with libgsmd's passthrough Sep 02 19:24:17 cesarb: After you issue an AT+COPS=? you have to wait for the network scan to complete (about 10 seconds). If you type anything it will abort the scan and give you an OK Sep 02 19:26:32 Connected. Sep 02 19:26:32 AT^M Sep 02 19:26:40 It doesn't like my enter key :( Sep 02 19:26:51 that what you should get Sep 02 19:26:58 the control M is carriage return Sep 02 19:27:08 the modem should respond with an OK Sep 02 19:27:10 Writchie: it doesn't even go to the next line Sep 02 19:27:55 are you running a recent OM2007.2 Sep 02 19:28:04 Writchie: yes, today's Sep 02 19:28:27 well, i saw something about applications trying to automatically restart gsmd Sep 02 19:28:30 Writchie: (a few hours ago, in fact... things change every swearword minute on this random build) Sep 02 19:29:02 Writchie: they don't even notice it went away, in fact they never even seem to have connected to it in the first place Sep 02 19:29:23 you did turn the modem on? Sep 02 19:30:43 oh no.. kernel-modules installs _everything_ bad idea Sep 02 19:30:57 riot: especially the usbgadget Sep 02 19:31:43 Writchie: gsmd's init script turned it on Sep 02 19:31:47 Writchie: no, _everything_.. scsi e.g. Sep 02 19:31:49 Writchie: turning it off hangs the kernel... Sep 02 19:32:04 Writchie: (well, turning it back on is what hangs the kernel) Sep 02 19:32:24 i'll try it here Sep 02 19:35:59 cesarb: Mine works perfectly Sep 02 19:36:20 when i power back up, i get AT-Command INterpreter ready Sep 02 19:36:51 Writchie: older kernel? Sep 02 19:37:22 yes. Sep 02 19:37:26 Writchie: oh, btw, earlier you were saying about ATAND... but your syntax was wrong Sep 02 19:37:34 Writchie: it seems it's a kernel bug on the newer kernel Sep 02 19:37:45 Writchie: (the hang, that is) Sep 02 19:37:48 which AT Sep 02 19:38:07 Writchie: but even with the older kernel, I neve managed to make cu work Sep 02 19:38:18 Writchie: oh, btw, earlier you were saying about AT%BAND... but your syntax was wrong Sep 02 19:38:33 Writchie: it's not AT%BAND=x Sep 02 19:38:39 Writchie: it's AT%BAND=1,x Sep 02 19:41:30 cesarb: yes Sep 02 19:42:47 Writchie: with the correct syntax, I get "conditional info element error" Sep 02 19:42:54 Writchie: (+CME ERROR: 100) Sep 02 19:43:16 cesarb: i think my error was a little different but it was a CME ERROR Sep 02 19:43:31 Writchie: I get the CME ERROR, the text I looked up in the manual Sep 02 19:43:36 CME ERROR 21 Sep 02 19:43:43 at%band? Sep 02 19:43:49 * cesarb will try with a SIM Sep 02 19:44:04 Writchie: 21 is "call rejected", makes no sense Sep 02 19:44:50 ah, wrong table Sep 02 19:44:56 100 is "unknown" (wtf?) Sep 02 19:45:07 21 is "invalid index" (???) Sep 02 19:45:17 cesarb: without registration my AT+COPS=? returns (after 10 seconds or so) ERROR Sep 02 19:45:36 After I register, it returns the two networks in range Sep 02 19:47:05 SpeedEvil: AT%BAND sets which bands to operate on - or automatic. Sep 02 19:47:12 it's clearly implemented Sep 02 19:47:21 but it doesn't accept anything other than automatic Sep 02 19:47:50 openmoko: 03mickey * r2894 10/trunk/src/target/OM-2007.2/panel-plugins/openmoko-panel-gsm/ (ChangeLog src/openmoko-panel-gsm.c): openmoko-panel-gsm: catch up with libmokopanelui2 Sep 02 19:49:07 does anybody have u-boot images in their OM2007.2 build. Mine have disappeared Sep 02 19:50:41 Writchie: mine doesn't have any either Sep 02 19:50:52 Writchie: probably isn't being built by default anymore Sep 02 19:51:03 Writchie: (and, IIRC, the package name changed, just to be more fun) Sep 02 19:51:23 that's just great - now we need to build two overnight builds Sep 02 19:53:37 Writchie: it should be in the feed autobuilders use, shouldn't it? Sep 02 19:54:02 Writchie: it probably was being built before by accident (an accidental dependency on uboot somewhere) Sep 02 19:54:04 i have uboot from 31st aug Sep 02 19:55:31 well u-boot is very hardware specific and is couple by the PM to the specific kernel - it needs to be in the builds. Sep 02 19:55:49 otherwise why bother with all this bb stuff Sep 02 19:56:48 i am geeting openocd and duf-util Sep 02 19:56:56 s/geeting/getting/ Sep 02 19:56:57 Writchie meant: i am getting openocd and duf-util Sep 02 19:57:08 s/duf/dfu/ Sep 02 19:57:09 Writchie meant: i am geeting openocd and dfu-util Sep 02 19:57:37 openmoko: 03mickey * r2895 10/trunk/src/target/OM-2007.2/panel-plugins/openmoko-panel-bt/ (ChangeLog src/openmoko-panel-bt.c): openmoko-panel-bt: catch up with libmokopanelui2, reindent Sep 02 19:57:40 is there a stamp for it? Sep 02 19:58:01 holy crap that was an important commit... Sep 02 19:58:02 it appears that both om2007.1 and om2007.2 are still alive Sep 02 19:58:30 Writchie: did you try "ATAND=0"? Sep 02 19:58:33 urgh Sep 02 19:58:35 Writchie: did you try "AT%BAND=0"? Sep 02 19:58:42 yes Sep 02 19:58:53 Writchie: it is giving me an error... Sep 02 19:59:05 Writchie: which is strange, since it is already in mode 0 Sep 02 19:59:11 (+CME ERROR: 100) Sep 02 19:59:38 Writchie: I agree with you that it looks a bit like a heavy-handed "early exit" from the band-setting function... Sep 02 20:00:55 cesarb: I get error also when setting the band Sep 02 20:01:13 the set function looks disabled - possibly in the wrong place. Sep 02 20:01:40 it should not report that the command options are present and then not permit you do set them Sep 02 20:05:08 openmoko: 03mickey * r2896 10/trunk/src/target/OM-2007.2/panel-plugins/openmoko-panel-bt/src/openmoko-panel-bt.c: openmoko-panel-bt: refactor a bit Sep 02 20:07:29 Hi guys and gals Sep 02 20:08:29 hello everyone, i cant wait untill GTA02 gets out. Dev`s are doing a teriffic good job! Sep 02 20:08:53 Any of you have any clue of the progress of gta 2.0? I've been scouring the net but i can't seem to find anything consice? Anything i've missed? Sep 02 20:09:11 october, morten_ ? Sep 02 20:11:08 October. quite early, thinking of how delayed gta 1 was. Sep 02 20:12:39 true.. Sep 02 20:14:33 Sorry to change the topic but: I've made a software keyboard to the moko. How do I simulate keypresses? in linux (openmoko) I've done all the gui, layout things, i only need to hook the keyboard Sep 02 20:16:03 http://geekouzouille.antiville.fr/ Sep 02 20:16:05 http://geekouzouille.antiville.fr/ Sep 02 20:17:49 morten_: maybe sources of some other virtual keyboards will help a bit, e.g. gok, or something from handhelds.org ... Sep 02 20:17:53 How do i suppress the splash-screen? Sep 02 20:18:48 mmp, Yeah, i was actually thinking of the keyboard that's already on the neo? maybe that has waht i need? Sep 02 20:19:13 morten_: maybe, I don't personally have neo, nor have seen the codebase:) Sep 02 20:19:30 mmp, same here, that's my problem Sep 02 20:20:19 riot: command-line option, but it's easier to rename rcS.d/S00psplash to something not starting is "S"... Sep 02 20:20:40 If someone could find me the codebase of the input widget i could very likely use it, the problem is that i have NO whatsoever idea of where to get it Sep 02 20:20:45 i've just flashed a new image, which just doesn't boot completely Sep 02 20:20:57 the progress bar fills only the first half :( Sep 02 20:22:06 oh, it boots! Second try always goes better Sep 02 20:26:30 riot: congratulation Sep 02 20:28:22 hey peeps Sep 02 20:28:48 is there anyone alive Sep 02 20:30:18 dcordes: and hangs upon the last pixel. at least i can ssh into it now Sep 02 20:33:15 hello ? Sep 02 20:33:16 awesomo4000: See http://wiki.openmoko.org/wiki/FAQ (and also SH1, SH1_FAQ, Neo1973, Developer_preview and Neo1973_Hardware#GTA01Bv4 pages) (Hopefully these links answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Sep 02 20:33:53 anyone doing development with debian 4.0 ? Sep 02 20:34:25 Cool, the engineering mode together with a table I found on wikipedia tells me my carries uses GSM1800 (I can see channel 885 nearby) Sep 02 20:34:31 s/carries/carrier/ Sep 02 20:34:31 cesarb meant: Cool, the engineering mode together with a table I found on wikipedia tells me my carrier uses GSM1800 (I can see channel 885 nearby) Sep 02 20:35:09 cesarb: that expains why it works Sep 02 20:35:22 can anyone see this message ? Sep 02 20:35:26 IIRC different parts of brazil use european band Sep 02 20:35:29 awesomo4000: no, we can't Sep 02 20:35:33 :) Sep 02 20:35:50 triband phone is different 3 bands depending on where the phone is marketed. Sep 02 20:36:27 i'm setting up a debian install under parallels on mac and i have a neo1973 .. just wanted to make sure its the same as everyone elses Sep 02 20:36:49 is the buildhost for openmoko releases on debian 3.1 ? Sep 02 20:37:39 riot: my zaurus-angstrom-moko is doing great :D Sep 02 20:38:07 will i have trouble with debian 4.0 trying to build for my neo1973 ? Sep 02 20:38:17 no Sep 02 20:38:50 Writchie: the only other option would be the 850 band, which both my carrier has (it uses TDMA/AMPS on that band) and the other carrier has (also AMPS, now GSM on that band... perhaps also CDMA on that band, might be on another band...) Sep 02 20:38:56 awesomo4000, yes/no/maybe/wth Sep 02 20:38:56 Hi all. What gstreamer modules should I install to use mediaplayer? Sep 02 20:39:12 Writchie: (the other two carriers have always been GSM, probably on 1800 or 1900) Sep 02 20:39:23 sarge (3.1) should be no trouble though right ? Sep 02 20:42:17 I believe that all 4 GSM bands are in use in brazil Sep 02 20:43:18 Writchie: no, AFAIK 900 isn't Sep 02 20:43:48 * cesarb looks again Sep 02 20:43:52 Writchie: yeah, all four are used :P Sep 02 20:44:21 Writchie: see http://www.teleco.com.br/Bandac.asp and http://www.teleco.com.br/Bandac_ext.asp for the gory (and nearly undecipherable) details Sep 02 20:44:55 are there any compact flash cards that are usable with gsmd? Sep 02 20:44:57 Writchie: ah, it's _1900_ which isn't used. I knew there was one not used ;-) Sep 02 20:45:15 1900 i think is reserved for UMTS Sep 02 20:45:36 Writchie: we don't have UMTS yet AFAIK... Sep 02 20:46:17 no but its planned Sep 02 20:46:32 Writchie: http://www.teleco.com.br/3g_brasil.asp has the details... does that look like it's using 1900 or 2100? Sep 02 20:46:34 reserving the whole band for UMTS should provide better service Sep 02 20:47:21 Writchie: well, the whole band is being split in four parts, so we can have four carriers... that if I'm reading that page right Sep 02 20:47:22 looks like two bands Sep 02 20:48:19 meh, the GSM frequency spectrum here in Brazil is a crazy patchwork mess Sep 02 20:48:40 well at least you have spectrum, unlike the US Sep 02 20:49:11 we have AMPS, TDMA, CDMA, and now UMTS, the frequency bands are split on a lot of subbands, and each region on Brazil has a different allocation of bands to the carriers Sep 02 20:49:39 and GSM Sep 02 20:49:40 And also, almost none of the carriers has coverage to the whole country (IIRC that's TIM's main selling point; it does have coverage in all the states) Sep 02 20:49:52 Writchie: and GSM... 5 technologies total Sep 02 20:50:34 But only people with old handsets still use AMPS and TDMA, and IIRC both are going away (IIRC AMPS coverage will not be required anymore either this year or next year...) Sep 02 20:50:34 we actually AFIK have the same, except that I think AMPS is finally dead Sep 02 20:51:01 AMPS and TDMA is cloning city, which is why I think it's great we are moving away from them Sep 02 20:51:03 but WiMAX adds another incompatible mess Sep 02 20:51:23 http://www.teleco.com.br/comentario/com188.asp <- cool band allocation graph Sep 02 20:52:34 IMHO the only succuessful policy is requiring all operators to comply with a standard and monopolistic practices like the tie-in of handsets to networks. Sep 02 20:53:09 There is no real competition if the operators use incompatible technologies which is what they have maneuvered for themselves in the US Sep 02 20:54:14 re Sep 02 20:54:20 Writchie: well, around here so far we only have two winning technologies (CDMA and GSM), and the CDMA carrier seems to have noticed they painted themselves into a corner and are moving to GSM Sep 02 20:54:59 Writchie: one of the carriers (Oi IIRC) started for some random reason providing unlocked phones, and made a big marketing campaign about it Sep 02 20:55:13 Writchie: the ad is on every single bus stop you look Sep 02 20:55:38 Writchie: TIM IIRC always had a policy where you can unlock the phone after some time or payment... Sep 02 20:56:16 The worst one is the CDMA one (Vivo), which is the only one who sells "defeatured" phones (things like locking bluetooth) Sep 02 20:56:19 with GSM it hard to keep the phone unlocked from an educated population and it usually backfires. Sep 02 20:56:52 But the non-GSM handsets just aren't going to work because the networks are incompatible Sep 02 20:57:09 the US carriers de-featured their GSM phones as well Sep 02 20:57:18 they don't to support anything Sep 02 21:00:35 cesarb: did you get the EM 2 data only after you registered? Sep 02 21:01:43 Writchie: no, in fact I forgot to register and even then I received... but with the SIM, I don't recall if I tried without the SIM Sep 02 21:02:07 * cesarb powers off and tries, hoping the SIM holders will not wear out too soon Sep 02 21:02:19 I'm very confused why the AT+COPS=? doesn't work without registration Sep 02 21:02:27 it's necessary for emergency calling Sep 02 21:02:44 could be an interaction with the SIM Sep 02 21:03:22 Writchie: I powered off then on ("o" and "O"), but did not do the "r"... now let's see if AT+COPS=? work (still with the SIM) Sep 02 21:03:47 Writchie: it auto registered, AND gsmd crashed :( Sep 02 21:04:05 which is why i don't waste much time with gsmd Sep 02 21:04:24 Writchie: you know I didn't manage to make cu behave :( Sep 02 21:04:46 i'll have to try it on a later kernel Sep 02 21:05:28 I have two devices, and i'm keeping one on the original OM2007.1 kernel because that is known working up to a stable point with the modem. Sep 02 21:06:00 the latest kernel i am quite happy with, as bt now works mostly Sep 02 21:06:04 the other i've loadded with various OM2007.2 but they regress Sep 02 21:06:13 hhf423: which one is that? Sep 02 21:06:40 see my latest entry in p1 testing Sep 02 21:06:59 i'm not very happy if I don't have a source code base that matches the binary Sep 02 21:07:18 how can you debug anything if you don't know what source is in the build Sep 02 21:07:18 oh, doesnt? Sep 02 21:08:14 sure, but then i do not debug, i only test ;-) Sep 02 21:09:15 forward progress is going to remain difficult until there is a matchup of code base and binaries Sep 02 21:09:36 it still looks to me like trying to change tires while the car is rolling down the road Sep 02 21:09:57 and i sooo want gsmd to work Sep 02 21:10:23 Writchie: now booting without the SIM, let's see what EM says Sep 02 21:10:50 hhf423: i am too, but I fear it is already hacked to unusability Sep 02 21:11:21 probably Sep 02 21:11:23 openmoko: 03mickey * r2897 10/trunk/src/target/OM-2007.2/panel-plugins/openmoko-panel-battery/ (11 files in 3 dirs): openmoko-panel-battery: Precalculate images to make scaling easier. Catch up with libmokopanelui2 Sep 02 21:11:41 it's not realistic to support multiple phones with different architectures when you don't even have a defined API Sep 02 21:12:03 yes, unlikely Sep 02 21:12:36 i hope to be able to fix some these shortcomings in the next week - if I can get to a reasonably stable code base. Sep 02 21:13:13 BT is still using glib? Sep 02 21:13:40 Writchie: AT%EM works without the SIM, and seems to be seeing only the SIM's operator (probably a cached value) Sep 02 21:14:01 bt is bluez, do not know what that uses Sep 02 21:14:32 last I looked it uses glib but there was talk of moving to native C interface to dbus Sep 02 21:15:16 agsm phone without sim will still be able to make emergency calls Sep 02 21:16:10 Writchie: AT+COPS=? without the SIM says +CME ERROR: 13 Sep 02 21:16:21 Writchie: which the manual says is "SIM failure" Sep 02 21:16:33 i think its a bug Sep 02 21:16:53 i need to try to find some of my old sim cards Sep 02 21:17:08 there must be a way to implement the typical "Select Network" Sep 02 21:17:10 Slight problem. Sep 02 21:17:21 That AT%EM docs site is now passworded Sep 02 21:17:22 Writchie: well, it does return a signal strength (AT+CSQ) Sep 02 21:17:31 take care, old sim cards could be 5v Sep 02 21:17:31 SpeedEvil: what you say??? Sep 02 21:17:50 SpeedEvil: I was afraid that would happen if I opened by big mouth. Sep 02 21:18:09 And I was looking to see if the signal strength is in the %EM2,3 response Sep 02 21:18:13 Writchie: it's possible that it's simply latching to the strongest signal, which happens to be the same from my carrier (the antenna is like directly in front of my window) Sep 02 21:18:45 cesarb: It's a scan function that has to take place on all bands. Sep 02 21:18:47 SpeedEvil: fourth line Sep 02 21:19:06 SpeedEvil: two lines above the cell ID Sep 02 21:19:24 Ah - k. Sep 02 21:19:50 SpeedEvil: rxlev_nc ("Receive Field Strength") Sep 02 21:20:02 SpeedEvil: my experiments show the highest one is the one which will be chosen Sep 02 21:20:04 found it (locally) Sep 02 21:20:15 I tried today to do a nice graph. Sep 02 21:20:31 However, it diddn't work - as I lost GPS. Sep 02 21:20:54 I was trying to correlate them. Sep 02 21:21:37 SpeedEvil: btw, AT%EM=? shows it also accepts another value for the first field... which isn't in the documentation Sep 02 21:21:52 SpeedEvil: might be risky (setting strange parameters), but I wonder what's in there Sep 02 21:22:34 * Writchie tucks away downloaded pdf from enfora for safekeeping. Sep 02 21:23:45 I just wish I could get cell names. Sep 02 21:23:48 * cesarb does the same Sep 02 21:23:55 SpeedEvil: cell names? Sep 02 21:23:57 But as I understand they are only from cell broadcasts, ... Sep 02 21:24:01 SpeedEvil: ah Sep 02 21:24:09 * cesarb doesn't have that around here Sep 02 21:24:10 Textual equivalents of the cell-number Sep 02 21:24:35 SpeedEvil: http://www.enfora.com/index.cgi?CONTENT_ID=283&User:LANGUAGE=en <- is still working for me Sep 02 21:24:43 SpeedEvil: isn't the cell id and LAC enough? Sep 02 21:25:26 To an extent. Sep 02 21:25:35 SpeedEvil: BTW, cell IDs and LAC on at least my carrier have an interesting structure (which I found out some years ago) Sep 02 21:25:39 However, the cell name may give hints as to the actual location. Sep 02 21:26:14 SpeedEvil: the cell ID has the format "side" (1 digit) "cell" (2 digits) "area" (2 digits) (names are mine) Sep 02 21:26:32 in decimal? Sep 02 21:26:34 SpeedEvil: and the area id has "area" (?? digits) and "area code" (2 digits) Sep 02 21:26:38 SpeedEvil: yep, decimal Sep 02 21:26:53 SpeedEvil: since a mast will never have more than 6 sides (0-5), it all fits nicely Sep 02 21:27:12 SpeedEvil: I wrote a silly J2ME app for my razr which shows the cell ID Sep 02 21:27:39 SpeedEvil: it was quite interesting once when the bus did a curve all the way around a mast Sep 02 21:27:48 SpeedEvil: the first digit changing, the other ones not changing Sep 02 21:28:28 SpeedEvil: also, what I'm calling the cell ID "area" doesn't change as often as what I'm calling the "cell", since it seems to indicate a group of nearby (or logically grouped) cells Sep 02 21:28:55 SpeedEvil: the Rio-Niterói bridge, for instance, has at least 3 cells, with cell IDs 1xx, 2xx, 3xx Sep 02 21:29:00 I assume cell IDs repeat nationally. Sep 02 21:29:09 SpeedEvil: take a look at your data and see if it has a similar structure Sep 02 21:29:29 I've only got some 88 sites logged. Sep 02 21:29:45 I want to integrate it with GPS Sep 02 21:30:23 damn Sep 02 21:30:46 cesarb: each cell should normally be associated with 3 towers. Sep 02 21:30:53 * * OM Bug 791 has been created by joseph(AT)pfeifferfamily.net Sep 02 21:30:54 * * webkit-gtk won't link Sep 02 21:30:55 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=791 Sep 02 21:31:05 I had been noticing my razr's battery compartment was having difficulty closing since some months... Sep 02 21:31:14 just noticed the cause Sep 02 21:31:21 THE BATTERY GOT FAT Sep 02 21:31:24 if you drive around the tower you should move into three different cells Sep 02 21:31:32 i.e. it's bulging Sep 02 21:31:39 cesarb: beware Sep 02 21:32:04 cesarb, Are you telling us your battery is expanding? Sep 02 21:32:16 take the battery to where you got it and make them give you a new for free. Sep 02 21:32:21 Indeed. Sep 02 21:32:42 cell topology should be a whole field of mathematics Sep 02 21:33:02 it is ;) Sep 02 21:33:18 I wouldn't be surprised Sep 02 21:33:19 you can have cells in other cells, non-radial cells, handoffs within a cell to migrate a receiver, etc Sep 02 21:33:30 overlays Sep 02 21:33:36 riot: you got my querry? Sep 02 21:33:45 last I remember, agent swarms were used as part of the weighting Sep 02 21:34:08 morten_: yeah... it's been for some time, since I had been having difficulty closing the compartment... Sep 02 21:34:14 morten_: but only now I noticed Sep 02 21:34:32 Writchie: well, it came with my cell phone :P Sep 02 21:34:56 so maybe moto is next to recall tens of millions of li-ion batteries Sep 02 21:35:00 * SpeedEvil has difficulty closing the battery compartment on his neo too, and it bulges. Sep 02 21:35:09 cesarb, Ok, but it's not expanding as-we-speak, because then you should bury it in your garden as fast as you can spell moghadishu Sep 02 21:35:10 (that'd be the internal wires) Sep 02 21:35:20 SpeedEvil: yeah but you've been abusing the thing Sep 02 21:35:31 It's the current shunt wiring actually. Sep 02 21:35:41 and your 14 day warrantee is up Sep 02 21:36:56 harold has added some nice things to u-boot, espeicaly for low level gsmd work Sep 02 21:37:04 s/harold/harald/ Sep 02 21:37:04 Writchie meant: harald has added some nice things to u-boot, espeicaly for low level gsmd work Sep 02 21:37:22 What sort of stuff? Sep 02 21:37:36 access to different serial ports. Sep 02 21:37:49 see his blog on openmoko planet Sep 02 21:37:54 ah Sep 02 21:41:28 root@fic-gta01:~$ mono Hello.exe Sep 02 21:41:28 Hello Moko! Sep 02 21:41:29 root@fic-gta01:~$ Sep 02 21:41:31 * rtyler dances Sep 02 21:41:46 hi all Sep 02 21:41:50 yay Sep 02 21:42:06 mickey|mdd: it was easy, cliff brake did all the hard work with the OE recipe Sep 02 21:42:12 rtyler: infidel! Sep 02 21:42:32 what's happening? Sep 02 21:42:34 Stephmw: would you prefer: "java Hello.class" >_< Sep 02 21:42:44 rtyler: tis already working, gui-level :) Sep 02 21:42:59 rtyler: Swing's pretty much there iirc Sep 02 21:43:26 forget I said anything, I'd rather not get into a "OMFG JAVA VERZUS CSHARPZ" Sep 02 21:43:44 * rtyler goes back to fixing the recipe to build the .NET 2.0 profile Sep 02 21:44:24 did someone manage to port java to the neo? Sep 02 21:44:31 * rtyler hopes not Sep 02 21:44:56 there's already work in OE for this ;) Sep 02 21:45:04 There is a special organisation doing it. Sep 02 21:45:08 rtyler: you're just a sourgrape :) Sep 02 21:45:09 Java Hut. Sep 02 21:45:31 SpeedEvil: Java in under thirty minutes delivered or it's free! :P Sep 02 21:49:17 is there a way to deploy just some ipk packages in a separate directory? and then let bitbake create an additional Packages.gz for them? Sep 02 21:53:00 you would need to change their PACKAGE_ARCH then Sep 02 21:54:27 (script) wiki RecentChanges: [[Buying_Interest_List]] [[Phase_1_Software_Testing]] [[Wish_List]] [[User:MTH]] Sep 02 21:56:41 mmmm should a simple redefinition in the bb recipe be enough? Sep 02 21:57:02 I would like to use it to create a repository of my own packages and then put it online Sep 02 21:57:07 could somebody give me the URI from ipkg repo source list? Sep 02 21:58:18 Ghiottone: changing package arch is insane then Sep 02 21:58:24 just add a custom src/gz Sep 02 21:58:31 bbiab, rebooting Sep 02 22:06:43 morning Sep 02 22:06:47 ahoy Sep 02 22:06:59 heya Sep 02 22:14:05 dcordes, something like src/gz updates-all http://buildhost.openmoko.org/OM2007.2/tmp/deploy/glibc/ipk/all ? Sep 02 22:15:29 exactly thanks Sep 02 22:17:14 Now I have only two options... I can use the Neo as my primary phone (sooner than I wanted to), or I can borrow my parent's phone Sep 02 22:17:32 I'm not using the razr anymore until I get a new battery Sep 02 22:18:16 dcordes, that is only one of the several files. you have to change all to armv4t and fic-gta01 also Sep 02 22:18:52 dcordes, for updates. for snapshot the lines are like src/gz snapshot-all http://buildhost.openmoko.org/snapshots/2007.08/ipk/all Sep 02 22:27:39 mickeyl, what do you mean by just add a custom src/gz? I want to create the repository so that then I can use a custom src/gz for ipkg update Sep 02 22:28:14 yeah, well why not uploading stuff to somewhere Sep 02 22:28:18 creating a packages file Sep 02 22:28:26 then adding this as a seperate feed source Sep 02 22:29:07 hrm, I already don't like bitbake >_> Sep 02 22:29:09 wow - attempt to install windows drivers for u-boot serial = instant black screen of death - game over - no more live testing for me with USB neo. Sep 02 22:29:31 I can't find a place in the world where the actual "make" might be executed, let alone configure params Sep 02 22:29:49 Writchie: ouch. can it work in vmware? Sep 02 22:30:02 something's causing this to package the monolite version which (sucks and) is too minimal Sep 02 22:30:05 i think vmware is funky with usb Sep 02 22:30:15 i'll have to dedicate a box to USB testing. Sep 02 22:30:37 i thougt something simple like serial port ACM would just work - silly me. Sep 02 22:30:54 probably eventually Sep 02 22:31:22 it will have to - like or not this stuff will have to play nice with xp and vista to go mainstream Sep 02 22:31:43 Writchie: isn't uboot hacker-only? Sep 02 22:31:44 SpeedEvil: did you ever get a chance to measure things with the audio code/amp off? Sep 02 22:31:55 Writchie: what have to play with windows is linux usb, not uboot... Sep 02 22:32:00 the LM???? chip - yes Sep 02 22:32:07 SpeedEvil: i'm getting not much over 24hrs battery life, i'm assuming it's the audio to blame Sep 02 22:32:13 SpeedEvil: yeah, how did you power it off? Sep 02 22:32:16 it measures about 0 - when turned off, with that set in alsamixer on 'off' Sep 02 22:32:21 cw666: set the mixer "Amp Mode" to "Off" Sep 02 22:32:23 'amp-mode' in alsamixer Sep 02 22:32:26 turn it to off Sep 02 22:32:34 Also send the modem at@poff Sep 02 22:32:40 to turn that off. Sep 02 22:32:46 no, i need modem on :-) Sep 02 22:32:51 im trying to use it as a phone Sep 02 22:32:53 cw666: there's a patch somewhere which automatically turns it off on suspend or shutdown Sep 02 22:32:58 cesarb: Ideally you would not need to touch u-boot Sep 02 22:33:15 carrar: ah ok, any idea where or what the subject might be? Sep 02 22:33:26 cesarb: but sometimes the u-boot environment might have to be fiddled. Sep 02 22:33:45 SpeedEvil: basically i get little more than a day as a phone, im aiming for a lot more and wondering what else i should be checking Sep 02 22:33:52 Writchie: isn't fiddling with the uboot environment, again, hacker-only? Sep 02 22:34:13 cesarb: think - how does the user unbrick the device Sep 02 22:34:34 i'm thinking an SD card based install Sep 02 22:34:39 Writchie: the user should not be able to brick it Sep 02 22:34:49 Writchie: unless the user is already using hacker-only things Sep 02 22:35:12 some end-users will smash uboot, there is probably no way to avoid it Sep 02 22:35:18 Writchie: upgrades should be made from within linux (ipk plus dd to the kernel partition) Sep 02 22:35:27 cesarb: if you have to upgrade u-boot and power fails you may brick it. Sep 02 22:35:55 mickeyl, cesarb: either of you know any good bitbake tutorials, or maybe documents explaining wtf is actually happening in do_compile when the .bb doesn't explicitly set it up? :P Sep 02 22:36:27 rtyler: check the inherits (base.bbclass is implicitly inherited) Sep 02 22:36:34 rtyler: (some other inherits also come from elsewhere) Sep 02 22:36:37 if the linux system is borked, you can use it upgrade - you need a baseline installer of some kind compatible with the u-boot in the device. Sep 02 22:37:35 cesarb: ideally a small initrd and not dependant on the current flash. Sep 02 22:38:03 rtyler: read base.bbclass Sep 02 22:38:05 except for u-boot Sep 02 22:39:01 mickeyl, yeah! but how do I create a Packages.gz? Sep 02 22:39:20 Ghiottone: check what bitbake package-index does Sep 02 22:40:01 hrm, 2007.2 has some killer UI-stability problems, I can't use it for more than 5 minutes before Today crashes and I see the splash screen again Sep 02 22:40:10 mickeyl: ZOMG ZOMG PLZ FIX KTHX :p Sep 02 22:40:12 * rtyler ducks Sep 02 22:40:53 rtyler: sounds like your hardware is broken Sep 02 22:41:00 it works very stable here Sep 02 22:41:04 you upgraded uboot? Sep 02 22:41:23 no, just uImage and rootfs, should I upgrade uboot as well? Sep 02 22:41:27 well Sep 02 22:41:28 * rtyler hasnt done that yet Sep 02 22:41:33 doesn't anyone read my release notes? Sep 02 22:41:37 mickey: if I actually use the GUI, it ends up crashing quite easily Sep 02 22:41:51 mickeyl: and I have the 2007.2 uboot Sep 02 22:41:54 can't say anything except that it's really stable here Sep 02 22:41:57 mickeyl: I usually only catch things on the mailing lists and the planet Sep 02 22:42:17 but since most people are building their own images, i have a hard time reproducing problems Sep 02 22:42:55 mickeyl: the crashing GUI is not the most important problem, however... it's the crashing gsmd which is very annoying Sep 02 22:43:19 mickeyl, what release notes? ;) Sep 02 22:43:29 mickeyl: that, and the fact that it never works without init.d hacking (because the dialer thinks it's fun to send AT+CFUN=1 and AT+COPS=0 back-to-back) Sep 02 22:43:38 mickeyl: there doesn't appear to any u-boot built in the om2007.2 build Sep 02 22:43:48 well Sep 02 22:43:53 really, no one Sep 02 22:44:03 is supposed to use om2007.2 stuff Sep 02 22:44:04 mickeyl, you mean that I should just call rootfs_ipk_do_rootfs after setting IMAGE_ROOTFS to my own path? Sep 02 22:44:05 that is not even tested Sep 02 22:44:10 please stick to the snapshots for now Sep 02 22:44:23 Ghiottone: uhm... or use ipkg-make-index directly Sep 02 22:44:26 from ipkg-utils-native Sep 02 22:44:29 the august 29th image (self built) with updated u-boot works good for me and very stable - even gsmd doesn't work per guy, but thats not new for me :) Sep 02 22:44:35 %s/guy/gui Sep 02 22:45:21 mickeyl, oh that is not even mentioned in package-index.bb or its included class. thanks! Sep 02 22:45:26 :) Sep 02 22:45:30 * mwester scratches his head, and wonders how the community can contribute even in testing, if they are on a binary-incompatible version several weeks old... Sep 02 22:45:32 mickeyl, but switching to i686 made me running into a bunch of new problems :) Sep 02 22:45:33 thomasg: when you say "updated u-boot" do you mean a u-boot built from the OM2007.2 build or some a u-boot from somewhere else. Sep 02 22:45:51 mwester: binary incompatible? Sep 02 22:46:04 Writchie, as I use 2007.2 for a month now and I said self-built, that's what I meant :) Sep 02 22:46:07 .2 changed to eabi, no? Sep 02 22:46:20 mwester: yes. and OM 2007.08 is the first official snapshot w/ EABI Sep 02 22:47:06 anyways, i need to wake up in 4 hours. need to sleep Sep 02 22:47:07 g'nihgt Sep 02 22:47:13 * rtyler waves to mickeyl Sep 02 22:47:35 hrm, does MokoMakefile not build u-boot by default/ Sep 02 22:47:47 rtyler: apparently, no Sep 02 22:48:01 d'oh Sep 02 22:48:05 rtyler: and I don't know the package name for it either (I think I saw something with its name changing) Sep 02 22:49:26 rtyler: if does for me Sep 02 22:49:29 or did the other day Sep 02 22:51:18 nite all Sep 02 22:52:37 u-boot-mkimage-openmoko-native maybe? Sep 02 22:52:37 I will make mokomakefile build u-boot later today Sep 02 22:52:53 rwhitby: cool, will you ping me when you do? :D Sep 02 22:53:02 rwhitby: thank you Sep 02 22:53:06 you'll see the commit message here Sep 02 22:53:30 or subscribe to http://cia.vc/stats/project/openmoko/.rss Sep 02 22:53:37 rwhitby: note harald has just committed a bunch of changes to u-boot Sep 02 22:53:51 * rwhitby goes to work Sep 02 22:53:52 i'm not sure how well tested they are Sep 02 22:55:24 aw man, we totally gotta do something about that startup sound :P Sep 02 22:57:10 rwhitby: what was the solution wrt to compiler blowing up? Sep 02 22:57:19 rwhitby: patch out the change that was to blame? Sep 02 23:49:54 cw666: hrw revert koen's bad gcc patches Sep 02 23:54:27 (script) wiki RecentChanges: [[Manually_using_Bluetooth]] [[OpenMoko_Local_Groups:_Berlin]] [[OpenMoko_Local_Groups:_Perth]] [[OpenMoko_Local_Groups]] Sep 03 00:45:52 ello all Sep 03 01:06:22 quiet room tonight Sep 03 01:06:55 BryceLeo: everybody must've gone to sleep early Sep 03 01:07:49 cesarb: Either that or it's the middle of the day and they're working. Sep 03 01:09:58 And it's a holiday in the US Sep 03 01:19:19 pshhh Sep 03 01:19:24 this is not a real holiday Sep 03 01:19:38 * rwhitby notes the new international date style in the Today screen :-) (but wonders about the dot after the day number) Sep 03 01:19:38 it's labor day, aka yet another lazy american excuse to skip work and go drinking Sep 03 01:20:08 Is that a European thing, to put a '.' between the day number and the month name? Sep 03 01:20:22 i do that if i don't use a space Sep 03 01:20:34 there is a dot, then a spacae Sep 03 01:20:40 yeah that's a european thing Sep 03 01:20:42 hmm... that seems odd Sep 03 01:21:19 "Monday, 03. September 2007" is what it says. Sep 03 01:21:22 cesarb: who still does work? :-P Sep 03 01:21:28 I was expecting "Monday, 3 September 2007" Sep 03 01:22:03 oh right, i thought you meant 03.09.2007 Sep 03 01:22:29 BryceLeo: isn't labor day May 1st? Sep 03 01:22:34 Yeah, "Monday, 03. September" is just odd. Sep 03 01:22:53 cesarb: ... is it? heh now i'm not sure! Sep 03 01:23:03 It's an improvement, anyway. At least it's not ambiguous now. Sep 03 01:23:05 BryceLeo: at least here labor day is May 1st Sep 03 01:23:29 Labor day in the US is at the end of the summer Sep 03 01:23:36 (always a Monday) Sep 03 01:23:42 May 1 here is called May DAt Sep 03 01:23:45 er.. May Day. Sep 03 01:23:49 And nobody really celebrates it Sep 03 01:23:55 nope labor day is this coming monday Sep 03 01:24:06 cesarb: where's here for you? Sep 03 01:24:57 BryceLeo: Brazil Sep 03 01:25:14 It's the 4th Monday in October here Sep 03 01:25:50 cesarb: how is it down there? Sep 03 01:25:50 http://pt.wikipedia.org/wiki/Dia_do_trabalho Sep 03 01:43:31 * gcb77 finally filled up the NEO to max... time to reflash **** ENDING LOGGING AT Mon Sep 03 02:59:56 2007