**** BEGIN LOGGING AT Sat Oct 13 02:59:56 2007 Oct 13 03:26:01 * * OM Bug 930 has been created by mwester(AT)dls.net Oct 13 03:26:02 * * kexec fails on the neo; requires existing kexec ATAG patch Oct 13 03:26:03 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=930 Oct 13 03:30:20 * dotmef starts the long road to a full compile Oct 13 05:01:01 re Oct 13 05:07:00 * * OM Bug 931 has been created by mail(AT)mmontour.net Oct 13 05:07:01 * * matchbox-panel-2 segfault in pango_font_get_coverage() Oct 13 05:07:02 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=931 Oct 13 05:44:07 'allo roh Oct 13 06:35:57 hi all, i wanted to recompile the kernel for my neo to enable fbcon rotation, but I've never cross compiled a kernel before, and when i go to make the kernel, it tries to build an x86 kernel... Where can i look to find the settings I need to set to build the kernel? Oct 13 07:13:18 good morning Oct 13 07:15:55 Hopscotch, good morning Oct 13 07:29:29 Hello Everybody Oct 13 07:29:54 Does uboot serial console support most of the commands? Oct 13 07:30:16 Like formatting, extracting tar archives, etc? Oct 13 08:34:43 Hey Guys, anybody there who knows about porting OpemMoko to hardware other than the Neo? Oct 13 08:36:04 Fabian2_de well.. the openmoko-stuff is the least to 'port' the linux beneath is is most of the work Oct 13 08:37:33 roh: well, yeah, I know... I tried to compile an openmoko-image for h5000 using MokoMakefile, but the image is way too big for my device (h5400). Oct 13 08:46:10 what's h5000? Oct 13 08:50:16 it's the name of the machine.conf. (well, the type of the machine) Oct 13 08:54:27 I know I can't ask anything about gta02, but... :P Oct 13 08:55:01 Will it be sold before december without the stable openmoko? :P Oct 13 08:55:29 no Oct 13 08:55:48 if in december it will be sold, likely without stable openmoko Oct 13 08:56:05 ... ok :P Oct 13 08:56:23 so it is imossible to get my hands on a gta02 before december :'( Oct 13 08:56:49 maybe if you break into openmoko office and steal one Oct 13 08:57:00 well.. even if it does not make a working device ;) Oct 13 08:57:04 other than that, i dont think you will get one Oct 13 08:57:23 i have a gta02v2 here.. but thats still not a done device (half the subsys have known bugs) Oct 13 08:58:38 if battery live anywhere near useable now? Oct 13 08:58:54 on gta01 Oct 13 08:59:04 well.. there is no software which reensembles complete powermanagement yet.. so no Oct 13 08:59:45 but any data how long it might survive once suspend actually works? Oct 13 08:59:58 nope. Oct 13 09:01:30 there also a lot to play with by clocking it down and using only some subsys. Oct 13 09:04:01 in my book, it must do at least 3days with light usage Oct 13 09:04:01 any less than that is basically useless for a phone Oct 13 09:06:59 pfeh, my first mobile phone was flat by lunchtime if unplugged in the morning after being on change all night. Oct 13 09:07:02 *on charge Oct 13 09:07:09 ... so bi-weekly is that twice a week or once every two weeks (fortnightly) ? Oct 13 09:07:39 fortnightly Oct 13 09:07:44 ScaredyCat: think it depends on who's saying it - regardless of what the correct usage is Oct 13 09:07:57 twice a week would be semi weekly Oct 13 09:08:08 no it wouldn't Oct 13 09:08:38 if it's fortnightly then fine, I was just checking... Oct 13 09:08:43 * cjb_ie protests weakly Oct 13 09:08:59 every half week gives twice a week in my book Oct 13 09:09:13 bi-weekly would generally mean twice a week, hence the question about what was meant... it wasn';t a critism Oct 13 09:09:40 http://en.wikipedia.org/wiki/Biweekly WP disagrees Oct 13 09:09:47 Biweekly, sometimes referred to as fortnightly, is a reference to an event that occurs once every two weeks. Oct 13 09:09:49 wp is often full of shit Oct 13 09:10:21 you only have to have seen the news in the uk to know that Oct 13 09:10:27 semi-weekly would be almost weekly Oct 13 09:10:28 http://www.google.com/search?hl=en&q=define%3Abiweekly&btnG=Search Oct 13 09:10:31 (this past week) Oct 13 09:11:06 i'd read semi-weekly as "half as often as weekly" Oct 13 09:11:17 "Semiweekly: Issued twice a week. Compare Biweekly." Oct 13 09:11:22 http://www.usd.edu/library/instruction/glossary.shtml Oct 13 09:11:41 semi annually doesnt mean every two years, either Oct 13 09:12:28 I've never met or heard anyone who uses those.. Oct 13 09:12:45 maybe it's just another americaism Oct 13 09:12:52 americanism Oct 13 09:12:56 ie made up Oct 13 09:13:20 ScaredyCat: nah, i've heard it this side of the water too. but mostly as the subject of "so which one means which?" discussions Oct 13 09:13:22 semi annual is quite often heard Oct 13 09:13:50 (especially in context of half year financial reports) Oct 13 09:14:04 I've never heard that... most normal poople would say every six months... Oct 13 09:14:36 "twice weekly" and "fortnightly" are nearly impossible to misinterpret, so those are what i use. Oct 13 09:14:55 i've yet to meet "normal" financial analysts Oct 13 09:15:05 cjb_ie: meaning twice a week? Oct 13 09:15:07 none of them are normal Oct 13 09:15:30 guaqua: twice weekly meaning twice a week.. fortnightly meaning once a fortnight Oct 13 09:15:41 cjb_ie: i.e. once in two weeks? Oct 13 09:15:54 fortnightly == once every 2 weeks Oct 13 09:15:56 well.. dont think about the wording too long.. i think michael and i are glad it we get that done semi-punctual :) Oct 13 09:15:59 a fortnight being two weeks (contraction of fourteen nights) Oct 13 09:16:46 fortnight is somewhat hard to understand for most non-native speakers as the concept of fortnight isn't in every language (english being the only language i know) Oct 13 09:16:52 although apparently bi-fortnightly would be every week Oct 13 09:17:10 in my book, that would be monthly Oct 13 09:17:16 :) Oct 13 09:17:16 that is monthly Oct 13 09:17:20 bi-weekly is fortnightly Oct 13 09:17:32 semi forthnightly would be weekly Oct 13 09:17:33 guaqua: french has quinze-jours which literally translates as 15 days but means two weeks. Oct 13 09:17:40 hmm.. isn't there a specialpage in mediawiki which list the highest ranking contributors? Oct 13 09:17:57 cjb_ie: okay :) Oct 13 09:18:18 the concept doesnt exist in german, but it's simple enough to wrap your head around in my book Oct 13 09:18:25 * ScaredyCat thinks buz is semi-bi.... weekly Oct 13 09:18:32 LOL Oct 13 09:18:48 buz: pfft, germans can't even agree what the days of the week are called ;-) Oct 13 09:18:50 * buz has semi clue of english Oct 13 09:18:56 wtf? Oct 13 09:18:57 finnish, swedish and german...being my 3 other languages don't have the concept of fortnight Oct 13 09:19:24 hey, don't worry about it the Dutch aren't even sure if they're German or not... Oct 13 09:19:40 ScaredyCat: oh i think you'll find they're sure alright :) Oct 13 09:19:55 sure enough to punch you if you suggest they are Oct 13 09:19:59 really.... "I come from German blood".... Oct 13 09:20:18 standard response if a german tourist asks for directions is "drei tagen geradeaus" Oct 13 09:21:08 or "oi ! where's my grandad's bike" Oct 13 09:21:32 seriously need to get over that... Oct 13 09:21:34 or where's my granny :( Oct 13 09:21:50 ffs slaughter of millions and all you care about is a friggin bike Oct 13 09:22:00 anyway... Oct 13 09:22:04 enought of that... Oct 13 09:22:11 it's saturday... Oct 13 09:22:15 caturday! Oct 13 09:22:17 and I haven;t had breakfast Oct 13 09:22:27 yaaaaay for Caturday! Oct 13 09:22:41 * ScaredyCat checks ichcb Oct 13 09:26:21 .oO(ICanHaveCaturdayBrunch?) Oct 13 09:27:19 *fsck* .. mediawiki hates me Oct 13 09:27:47 http://www.thecheezburgerfactory.com//completestore/128340998580937500ADSLcatisconn.jpg Oct 13 09:30:01 12,4 http://st-pitch.miniville.fr/  Oct 13 09:30:03 URL http://st-pitch.miniville.fr/ Oct 13 09:31:11 play Oct 13 09:31:22 okay, i mean Oct 13 09:31:29 someone should correct the topic Oct 13 09:32:06 thank you Oct 13 09:32:14 kids... Oct 13 09:32:42 roh: no no no! it's ICanHasCaturdayBrunch! Oct 13 09:32:58 *sigh* Oct 13 09:33:27 * roh had breakfast already... after a night bar and hours of work ;) Oct 13 09:33:34 the topic is fine Oct 13 09:33:39 the change didn;t happen Oct 13 09:34:05 you are all in my power... Oct 13 09:34:14 send all your neo to ScaredyCat Oct 13 09:51:41 could somebody do me a favour and try to build ttf-liberation? Oct 13 09:52:58 aye Oct 13 09:54:49 fine here Oct 13 09:56:00 what probs are you getting? Oct 13 10:01:58 ScaredyCat: it does not fetch Oct 13 10:02:16 ScaredyCat: http://www.pastebin.ca/raw/735099 Oct 13 10:04:08 hm.. why does it try https? Oct 13 10:04:58 SRC_URI = "https://www.redhat.com/f/fonts/liberation-fonts-ttf-3.tar.gz" Oct 13 10:05:29 someone needs a slap Oct 13 10:05:41 heh.. change that to http or do 'wget https://www.redhat.com/f/fonts/' on a console and accept the cert Oct 13 10:06:40 there's no good reason for using https.. Oct 13 10:07:00 unless the site owner is retareded and only has https setup.... Oct 13 10:07:07 but that's redhat... Oct 13 10:07:18 https is http secure something? Oct 13 10:07:19 so it's probably whoever made the .bb file Oct 13 10:07:22 with certs? Oct 13 10:07:38 yes.. https is what you use for stuff like online transactions Oct 13 10:07:54 there's no need for it when downloading fontd.. Oct 13 10:08:07 -d+s Oct 13 10:08:14 and when I download that file manually with wget and accept the certificate wget does recall it when it's called to download from that url? Oct 13 10:08:52 wget does not ask for it. it suggests using --no-check-certificate Oct 13 10:09:02 once you have downloaded it with wget, put it in your sources dir, then creat the .lock file (same filename+ .lock on the end) Oct 13 10:09:25 I'd rather correct the recipe and file a bug Oct 13 10:09:35 then it wont try and download it again Oct 13 10:09:58 the bug you fix will get overwritten unless you push the fix up Oct 13 10:10:18 NOTE: build 200710131209: completed Oct 13 10:10:26 with http instead of https in the recipe Oct 13 10:10:45 if you're using mokomakefile I'm pretty sure it'll revert on an update Oct 13 10:11:12 no I'm using normal oe setup and I'm build for akita Oct 13 10:11:57 ahh ok Oct 13 10:11:59 it will ask me if I want it 'corrected' or not on update Oct 13 10:12:26 at least that's what it always did when I did a manual change to the metadata and updated it afterwards Oct 13 10:13:55 ScaredyCat: thank you very much for the advise Oct 13 10:14:08 between, are there any news on gta02 release :D ? Oct 13 10:14:18 no problem... Oct 13 10:14:20 hehe Oct 13 10:14:21 dcordes best would be if you file a bug against that package in the oe bugtracker Oct 13 10:14:26 last I heard December... Oct 13 10:14:30 but... Oct 13 10:14:49 and also the gps chip has been decided on, which is good... Oct 13 10:14:49 roh: yes of course Oct 13 10:15:31 ScaredyCat: I have to make some santa buy me one then Oct 13 10:17:53 yay @ topic Oct 13 10:18:07 some more gta02 info for addicts like me :) Oct 13 10:19:31 hm.. i need some list of people who want to help reorganizing the wiki and are fit in all the wiki-isms Oct 13 10:25:53 reorganising the wiki? what's up with it? Oct 13 10:26:51 .. besides every time you move (copy paste) somehting it removes the original author/contributor Oct 13 10:32:48 -160 rx sensitivity... any good? Oct 13 10:37:16 roh, some of that stuff is in there for the simple reason that the search function is shit Oct 13 10:41:20 and errm,why did you delete the Canberra local users entry... Oct 13 10:55:11 ScaredyCat well.. we want to clean it up and split out docs/specs from talk and howtos Oct 13 10:55:37 and also integrate automated documentation as well as to be written one Oct 13 10:59:42 ScaredyCat i only deleted empty and unused pages Oct 13 11:06:47 * cjb_ie wonders how feasible it would be to linuxify something like a treo 680 Oct 13 11:07:30 works ;] Oct 13 11:21:52 hey I want to test openmoko with an emulator Oct 13 11:22:16 debian sid 64bit here, what is the easiest way to do that? Oct 13 11:33:22 blinx, sid 64 bit? which arch did you install? amd64? Oct 13 11:34:00 amd64, right Oct 13 11:34:07 I see http://wiki.openmoko.org/wiki/Using_QEMU_with_MokoMakefile#Debian_4.0_and_Ubuntu_7.04 Oct 13 11:34:17 blinx, if you used 32bit I would have said to download mokomakefile and run make qemu Oct 13 11:34:56 :) Oct 13 11:35:00 blinx, but I think you may have to change the ARCH in build/conf/local.conf to i686 or something like that Oct 13 11:35:01 I will try it anyway Oct 13 11:37:16 no news concerning the release date of the neo? Oct 13 11:41:11 blinx, nope but atleast we know what gps chipset we have Oct 13 11:42:11 speaking of said gps chip, wonder if RMS will endorse GTA-02 now, as in the previous communication the gllin stuff was (unsurprisingly) the biggest hurdle with that Oct 13 11:45:28 well.. i think 'antares4' is what helps people... ATR0635 doesnt help anybody ;) Oct 13 11:48:32 roh, I assume there aren't any news on GPS status for GTA01, aren't they? Oct 13 11:48:43 not yet. Oct 13 11:50:08 but mickey is back from portugal ;) Oct 13 11:52:11 roh, :) Oct 13 12:11:39 * ScaredyCat returns - bloated Oct 13 12:13:38 have you seen http://sourceforge.net/projects/osgps/ before? Oct 13 12:15:10 is that the one that's supposed to be able to read navtec/garmin maps but they don;t tell you how to do Oct 13 12:15:35 nope.. thats a soft-gps solution as far as i could tell Oct 13 12:16:00 roh: is it possible to measure some distances with the actual moko? Oct 13 12:16:42 * cb22 sees that the new gps chip supports WAAS and DGPS among others Oct 13 12:16:58 roh: errm no it's not a 'soft' gps Oct 13 12:17:21 I mean something like this: hit a button -> start point ... go to your destination and hit the button again -> it shows the air line or the walked way between the points... Oct 13 12:17:43 http://home.earthlink.net/~cwkelley/ Oct 13 12:20:01 ScaredyCat well.. not that soft as the gl, but seemingly much more low-level than parsing completely correlated nmea Oct 13 12:20:42 oh, yeah.. the guy claims not to know much about gps, but looks like he knows a hell of a lot to me Oct 13 12:21:05 blinx that could be implemented easily - but you'd have to use a bluetooth-gps-receiver Oct 13 12:24:07 kristian-m: why? Oct 13 12:24:17 kristian-m: other apps can use gps, too Oct 13 12:24:29 why I would need a bluetooth receiver? Oct 13 12:24:33 +gps Oct 13 12:25:00 blinx, because GTA01 needs a closed source driver for its GPS and this driver cannot be distributed atm Oct 13 12:25:05 hey kristian-m Oct 13 12:28:26 hey Ghiottone :) Oct 13 12:29:20 blinx yea but thats with some biary driver that can't be distributed with openmoko Oct 13 12:29:43 Ghiottone ah, eh right Oct 13 12:29:47 Ghiottone well.. the problem is not distribution anymore, its just not working on eabi (eabi build broken) Oct 13 12:30:49 roh: Problem is not distribution anymore... does that mean it can be distributed now? Oct 13 12:32:06 abraxa_ if delivered with the right licence text... i'll see what i can find out. Oct 13 12:32:15 roh, is it too much to ask them to recompile it for the eabi Oct 13 12:32:27 roh: Awesome! =) Oct 13 12:32:39 Ghiottone well.. they did and the binary is not working Oct 13 12:32:55 roh that's strange Oct 13 12:32:59 Hi, is umts/3g support expected for the neo1973 ? Oct 13 12:33:21 KenSentMe not in the near future, no Oct 13 12:35:01 kristian-m, hmm, that's sad. I'm following the Openmoko project for a while and am very interested in buying the neo1973 when my current phone stops running, but i can't miss umts anymore Oct 13 12:35:32 kristian-m, is it hard to implement, or is there some other reason for this? Oct 13 12:36:24 KenSentMe: It would require totally different GSM hardware - with as open specs as possible Oct 13 12:36:52 roh, pity. is there a way to help on that? and does it mean that we could have an abi executable? Oct 13 12:36:52 abraxa_, and that isn't available yet? Oct 13 12:37:27 KenSentMe: It's more a matter of "do we really want to redesign this crucial part of the hardware when we are already way behind schedule?" Oct 13 12:37:48 Ghiottone i'll try to push that forward. Oct 13 12:38:16 roh, thanks! Oct 13 12:38:18 abraxa_, i understand. But it's a technology that will only be used more and more in the future. It could be that umts support is planned in a future release of the neo Oct 13 12:39:19 KenSentMe: roh would be the one to give a more definite answer on that but I am 100% positive that OpenMoko, Inc. can't miss out on that market in the long run - so UMTS/3G versions will follow eventually Oct 13 12:39:30 kristian-m: why is there the gps chip implemented with only binary drivers???? Oct 13 12:40:18 well.. 3G will get used as soon as there is a spin which makes it possible. which means: no there is no hardware design put together, but its clearly on the 'want to have in new devices'-list Oct 13 12:40:29 blinx: Because the Hammerhead (in GTA01) is a mere receiver and does very little in hardware - the real work is done in software. You can imagine the company doesn't want that knowledge open to competitors Oct 13 12:40:54 roh, thanks for the clarification Oct 13 12:40:55 just not not gta01 and gta02. 3g also means zillions of licence-daemons more and the added cost through this Oct 13 12:41:24 abraxa_: hrm Oct 13 12:41:35 meaning even if someone could build the hardware for the same price as with 2g, 3g costs quite a bit only in royalties per device Oct 13 12:41:38 abraxa_: how the other gps based apps handle that? Oct 13 12:41:43 moin Oct 13 12:42:18 hola hhf423 Oct 13 12:42:34 hi Ghiottone Oct 13 12:43:06 blinx: "other GPS apps"? I would assume other GPS-aware applications just handle NMEA data - not raw data coming from a proprietary GPS chip. Note that GTA02 has a different chip now which (afaik) does everything in hardware, so no driver should be needed Oct 13 12:43:24 abraxa_ libgpsd, not raw nmea Oct 13 12:43:41 kristian-m, are there some packages that still needs to be done? I have stopped working on the NEO for one whole week cause of my new laptop Oct 13 12:43:53 roh: Still better than HH, yes? Oct 13 12:44:19 abraxa_ meaning: for the app its totally uninterresting which hardware its on. its only talking to gpsd (which gets fed nmea or ubx from serial or nmea from socket) Oct 13 12:44:24 KenSentMe Ghiottone if you've got time it'd be nice to have a package of the currnet pyPenNotes Oct 13 12:44:56 roh: Well yeah, was just wanting to clarify that the HH needs a binary driver because of its hardware design Oct 13 12:45:24 abraxa_ it does not need to be binary because of the design, but because of the businesscase Oct 13 12:45:26 ;) Oct 13 12:45:26 hello, anybody working her on the qemu emulator ? Oct 13 12:45:35 s/her/here Oct 13 12:45:47 roh: Oh? Hmm... interesting. Oct 13 12:51:36 kristian-m, version 3b? Oct 13 12:53:16 kristian-m, mmm what did you say you wanted as icon? Oct 13 12:58:18 Ghiottone gtk-directory Oct 13 13:04:43 hmm, the top menu still does only show up on every second boot of the neo Oct 13 13:06:11 Are there any other phones with 3g that will run openmoko when it's stable? Oct 13 13:06:43 KenSentMe: when its stable there will be loads of phones Oct 13 13:06:55 when its stable Oct 13 13:08:44 Is gta02 considered stable? Oct 13 13:09:08 gta02 is hardware, not openmoko which is software Oct 13 13:09:50 yeah, but the gta02 is released in december for mass market and runs openmoko. So is the software used in that release considered stable? Oct 13 13:09:50 well, is gta02 considered stable hardware? Oct 13 13:10:16 gta02 is stable hardware. gta01 is also stable hardware Oct 13 13:10:27 okay, good Oct 13 13:10:46 I doubt that there will be stable software in december, but it is not impossible Oct 13 13:12:19 in any case, I expect to have gta02 in my hands early january. maybe some lucky bastards will have it in late december (like if you are one of the first 100 orders and live next door to from where they ship :-) Oct 13 13:12:27 KenSentMe: "gta02 is released in december for mass market" - says who? Oct 13 13:13:10 abraxa_: that has been said by a number of people in the past months Oct 13 13:13:32 abraxa_, says here: http://wiki.openmoko.org/wiki/Neo1973 Oct 13 13:13:32 "past months" Oct 13 13:14:27 said by mickey on sept 20th Oct 13 13:14:43 qualified with something like "i hope" Oct 13 13:18:34 is there still hope that GPS will be made to work in gta01? Oct 13 13:19:22 of course Oct 13 13:19:34 great Oct 13 13:19:37 hey mickeyl Oct 13 13:19:40 wb Oct 13 13:20:56 mickeyl: can you maybe expand on the basis for your hope? Oct 13 13:20:56 hey Oct 13 13:21:02 well Oct 13 13:21:11 a) there's the oabi-gllin which works Oct 13 13:21:22 b) there's the eabi-gllin which doesn't work Oct 13 13:21:35 we may just distribute oabi-gllin with a chroot or other hacks Oct 13 13:21:42 or find a way to make eabi-gllin work Oct 13 13:22:01 there will be a way to make it work, it's just a question of how ugly and when Oct 13 13:22:57 I see Oct 13 13:25:50 mickeyl thats great news Oct 13 13:26:06 is it? Oct 13 13:26:09 is there any news on gsmd development? now that I have a new eplus sim card wich works in my neo I be me more interested to use it as a phone than before :-) Oct 13 13:26:22 i'm surprised that you think we abandoned that topic. Oct 13 13:26:43 we are committed to get the phone to work as advertised Oct 13 13:26:44 with all the focus on gta02 Oct 13 13:27:07 well Oct 13 13:27:20 you have to stay in business to earn money for the software team Oct 13 13:27:21 heh Oct 13 13:27:36 so we can fix the issues for all models Oct 13 13:28:55 I do not get that. who has to stay in which business? Oct 13 13:29:05 the guys who pay my salary Oct 13 13:29:46 and gta02 might be the first device which actually brings a bit of money for FIC Oct 13 13:29:51 which is FIC, no? Or OpenMoko, a 100% daugther of FIC Oct 13 13:31:34 if FIC does not have the breath to invest for three to five years then they should probably stay out of this. But it was my impression that they have. Oct 13 13:32:00 and gta01 was not sold at a loss, was it? just not big profit, right? Oct 13 13:33:49 i don't know the numbers. i just know that every company works in parallel on future models. it is not a serial process. the focus of the hardware people is on gta02, yes, of course, gta01 is done. it's not the focus of the software people. Oct 13 13:34:06 [as far as I and the colleagues i know about are concerned] Oct 13 13:34:13 i know not everything what happens in .tw Oct 13 13:34:59 although i expect to be updated in exactly one week :) Oct 13 13:35:17 *g* Oct 13 13:35:19 so, about software, any news on gsmd development? Oct 13 13:35:40 not from me. i'm still catching up from vacation Oct 13 13:36:01 as far as i can see there are patches going in Oct 13 13:36:04 hhf423 the gsm modem stays the same for gta02. so it needs to work nonetheless. there will be work happening Oct 13 13:36:08 incremental improvements Oct 13 13:36:20 but not yet the revamp which I suppose Oct 13 13:36:31 when do you think we might see that? next week? Oct 13 13:36:45 (the patches, not the revamp) Oct 13 13:36:54 there are constantly patches going in Oct 13 13:36:58 read the commit list Oct 13 13:37:03 see OE Oct 13 13:37:04 etc. Oct 13 13:37:18 well.. sofa for me.. bbl Oct 13 13:37:21 nah, there was nothing happening last week Oct 13 13:40:29 where do you see anything about gsmd or related? Oct 13 13:41:03 oh, 762 was closed. great Oct 13 13:42:24 NOTE: fetch https://stage.maemo.org/svn/maemo/projects/haf/trunk/fontconfig/device_symbols.h Oct 13 13:42:28 ERROR: Certificate verification error for stage.maemo.org: unable to get local issuer certificate Oct 13 13:42:31 To connect to stage.maemo.org insecurely, use `--no-check-certificate'. Oct 13 13:42:54 (when running mokomakefile / make openmoko-devel-image) Oct 13 13:58:38 urghs Oct 13 13:58:42 * mickeyl forgot his tea water Oct 13 13:58:44 *sigh* Oct 13 13:59:00 it's been cooling down for 60 minutes now Oct 13 13:59:57 separate water heater for the win :) Oct 13 14:00:17 put the cup on a neo, take out the battery, and plug it into the usb port Oct 13 14:00:18 or an alarm clock... Oct 13 14:03:46 i have tried to run open moko on qemu but it is unbelievably slow. is this a problem with the emulation or the is it just so slow? Oct 13 14:04:24 both Oct 13 14:04:47 mickeyl: hey how goes? Oct 13 14:05:24 Dheath: what kind of hardware you running it on? Oct 13 14:05:38 normal pc (x86) Oct 13 14:05:46 need more info? Oct 13 14:05:53 like, how fast Oct 13 14:06:08 more than 5 minutes to start Oct 13 14:06:12 mickeyl: so, there's some very cool plans for hal - basically hald will be going away, replaced with some file/db that has all teh device info Oct 13 14:06:18 Dheath: no, how fast is the PC Oct 13 14:06:22 push a button wait 30 sec Oct 13 14:06:32 dheath hm Oct 13 14:06:37 ... Oct 13 14:06:51 hello Oct 13 14:07:00 damn i have to check the exact number but it definiteli isnt that slow Oct 13 14:07:01 does anyone have a link to a working gsmhandset.state file? Oct 13 14:07:14 mickeyl: udev will grow some rules additions so particular system activatable daemons can be started for particular devices Oct 13 14:07:34 mickeyl: still not quite figured what happens with fdi's though ;) Oct 13 14:09:38 robster: hey rob. so so. the stresslessness i got from two weeks in portugal were eaten up pretty instantaneously due to OEDEM and my forthcoming Taipei journey... Oct 13 14:09:47 robster: (udev) interesting. Oct 13 14:10:06 OEDEM was stress? Oct 13 14:10:37 for me yes. arrived in FRA 02:00, took the ICE to Berlin @ 06:00. had to moderate two days long, etc. Oct 13 14:11:27 ah, right. thats too little sleep probably. I had to survive the week on 3-5h sleep/night and that was also a bit too little. Oct 13 14:11:32 robtaylor: do you have a link? Oct 13 14:11:35 *nod* Oct 13 14:11:51 traveling like that takes a huge toll Oct 13 14:11:57 its amazing though, how refreshed I was after 5h sleep when I had only 4 and 3 the nights before Oct 13 14:12:15 heh, yeah, it's all relative, but usually i need 8h Oct 13 14:12:21 same here Oct 13 14:12:31 mickeyl: yeah, thats always the problem with holidays - you need to come back ;) Oct 13 14:12:54 pH5: Nope, this was just what we discussed at the gnome boston summit Oct 13 14:13:20 robster: i'm seriously thinking about taking 6 months off Oct 13 14:13:23 the question is when Oct 13 14:13:41 mickeyl: (hey, get the right rob ;)) Oct 13 14:13:45 oops Oct 13 14:13:47 sorry ;) Oct 13 14:14:06 robtaylor: ok, cool. looking forward to more info about that :) Oct 13 14:14:23 mickeyl: yeah, there's nothing like a good long break.. Lets hope you get some time after GTA02 release :) Oct 13 14:14:28 mickeyl: you can 6 months off when openmoko is stable :-) Oct 13 14:14:57 hhf423: heh Oct 13 14:15:00 hhf423: slave driver Oct 13 14:15:08 want me bound for the next 2 years? :D Oct 13 14:15:35 robtaylor: probably not. but i'll aim for Q1 and Q2 2009 Oct 13 14:15:54 that sounds reasonable Oct 13 14:16:50 with a small tour through ireland and scotland Oct 13 14:17:12 hopefully doing some gigs in clubs Oct 13 14:17:16 so is there a working alsa .state file? I couldn't fine anything on the wiki that actually worked without feedback for me Oct 13 14:17:18 * mickeyl dreamingm Oct 13 14:17:31 ifoox: query alphaone when he's awaken. he should have one iirc Oct 13 14:18:17 mickeyl: thanks! Oct 13 14:19:30 mickeyl: Would you mind granting me commit access to the artwork dir so I can fix/update things there, too? Oct 13 14:20:24 cjb_ie: 2.00 GHz Oct 13 14:21:58 abraxa_: granted. please try again Oct 13 14:22:16 mickeyl: Thanks :) Oct 13 14:22:20 np Oct 13 14:22:28 yay. all my paperwork done for today Oct 13 14:22:37 sofa-time with the new c't now Oct 13 14:22:40 cu Oct 13 14:24:49 openmoko: 03abraxa * r3156 10/trunk/src/target/OM-2007.2/artwork/icons/openmoko-standard/ (12 files in 3 dirs): Swap gtk-media-play-*.png's LTR and RTL versions Oct 13 14:24:55 Worked :) Oct 13 14:25:23 so, when can we expect compiz-fusion on the neo ^-^ Oct 13 14:27:09 cb22: Does it speak OpenGL ES? Oct 13 14:27:26 i wish Oct 13 14:27:39 although how dissimilar are the two? Oct 13 14:29:01 infact, i dont really care about most things... Oct 13 14:29:04 i just want the cube Oct 13 14:29:11 powered by acceleromters Oct 13 14:29:21 so when i put the phone landscape, it goes into cube mode Oct 13 14:29:30 and then tilt shaking the phone moves the cube Oct 13 14:33:08 I'm sure it'll be done as soon as the glamo driver(s) are ready Oct 13 14:34:34 cb22: you know, the neo is a portable device with limited power supply Oct 13 14:40:33 LaF0rge, do you have a few minutes ? Oct 13 14:41:19 Vegar: Yes but showing off what the Neo can do is always a good thing ;) Oct 13 14:45:02 hehe Oct 13 14:45:04 daltona, just ask your question Oct 13 14:47:49 I just wanted to ask when the patch for turning off the Audio amp on suspend and shut down would be added in the kernel Oct 13 14:48:52 and I also created a patch for returning negative current measurements in order to get the current drawn by the moko when discharging. Oct 13 14:49:01 * pH5 wonders why the amp is not turned off as soon as the last audio stream stops playing. Oct 13 14:49:26 It's been a long time that they've been in bugzilla but seemed to be forgotten Oct 13 14:49:33 It would be nice if the patch for avoiding kernel lockups due to flow-control issues with GSM would find its way into the kernel... Oct 13 14:49:35 pH5: because you haven't written the code. Oct 13 14:49:52 pH5: it's unfortuantely generically not that simple. Oct 13 14:50:11 pH5: consider that the CPU can in fact be asleep and still have the GSM modem outputing to the amp. Oct 13 14:50:14 I've been using a kernel with the flow-control bug fixed, and the audio amp power-off fix for quite some time, no problems. Oct 13 14:50:21 Routed through the WM chip Oct 13 14:50:35 SpeedEvil: why, the whole asoc infrastructure is based around the principle of turning off whatever is not needed. Oct 13 14:51:19 daltona, mwester, laforge is currently updating the patch queue, he will probably look at your patches then Oct 13 14:51:22 mwester, I also used my amp disabling patch, but when I upgrade the kernel, I have to re-patch it each time I think it is time to include it Oct 13 14:51:32 Shoragan, ok thanks Oct 13 14:52:00 do you have the relevant bug ids handy? Oct 13 14:52:00 pH5: yes - but determining which is needed the right way involves getting quite complex. Oct 13 14:52:20 pH5: It's more than 'if sound isn't playing from /dev/dsp, turn the amp off'. Oct 13 14:52:29 pH5: that's a starting point of course. Oct 13 14:52:54 Shoragan, let me search Oct 13 14:53:16 http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=788 <==== gsmd patch Oct 13 14:53:31 http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=930 <===== kexec patch Oct 13 14:54:47 for negative current repport : #193, and Amp switch off: #781 Oct 13 14:55:24 kristian-m, nice work with pyPenNotes Oct 13 14:56:15 morning Oct 13 14:56:45 * mwester looks at the patch for #193, and notes that it is remarkable in its simplicity :) Oct 13 14:57:22 * daltona spent almost one week in developing and testing it :-) Oct 13 14:57:45 daltona: very useful fix, that one -- I'll be adding that to my kernel. Thanks! Oct 13 14:57:47 Ghiottone thanks - well there are still some bugs with background color on restore - and loading the strokes file takes way to long - but it really can be used now :-) Oct 13 14:58:40 mwester, yeah, it would allow to do some calculus for getting remaining battery life actually Oct 13 14:59:24 the measurements does not allways seems accurate though, but doing a kind of average seems ok Oct 13 15:00:00 Is this an ADC oversampling patch for battery current? Oct 13 15:00:23 SpeedEvil, no it's not Oct 13 15:00:40 Sigh. I need to get a build environment up and running so I can test my kernel patches. Oct 13 15:00:49 SpeedEvil, it just allows to repport negative currents when battery is not charging Oct 13 15:00:54 can i compile games like supertux for openmoko? Oct 13 15:01:19 daltona: the hardware can measure discharge current? Oct 13 15:01:27 SpeedEvil, yes Oct 13 15:01:40 It was just software that was screwing it? Oct 13 15:01:41 Nice. Oct 13 15:02:12 SpeedEvil, voltage is measured at the two sides of a resistor so if v1>v2 we are charging and if v2 >v1 we are discharging Oct 13 15:02:33 If the resistor is connected in the correct place Oct 13 15:02:48 SpeedEvil, yes, unsigned integers were used, making the negative values to be screwed Oct 13 15:03:04 I thought the battery tap for the main system was isolated from the battery charging circuit, so didn't bother checking Oct 13 15:03:15 Was just assuming it was noise. Oct 13 15:03:17 SpeedEvil, I think so, but I did not checked the schematics as I don't have them Oct 13 15:03:23 Do you get sane discharge numbers? Oct 13 15:03:57 I hopefully should have a build environment up in the next couple of days, so I can test my battery life patch. Oct 13 15:03:59 SpeedEvil, yes but as you said some kind of oversampling + averaging in the kernel would be a good feature to implement Oct 13 15:04:16 SpeedEvil, what have you implemented ? Oct 13 15:04:19 (dramatically improve battery life prediction, including battery aging. Oct 13 15:04:54 SpeedEvil, by sampling voltage more frequently ? Oct 13 15:05:13 No. By using a much, much better battery model. Oct 13 15:05:34 SpeedEvil, cool, I'd like to see it ! Oct 13 15:05:53 The existing one is a joke - it thinks that it's linear from 4.2V to 2.8V when in fact 3.5V or so is 10% Oct 13 15:06:21 Discharge current should also help with that too. Oct 13 15:06:32 SpeedEvil, yes I guess Oct 13 15:06:48 We need someone with 02 hardware to test the batcur reading to see if it's so horribly noisy Oct 13 15:06:53 SpeedEvil, and you've implemented that in the kernel then ? Oct 13 15:06:58 daltona: yes Oct 13 15:07:18 SpeedEvil, in apm driver ? Oct 13 15:07:31 Yes. Oct 13 15:08:12 * mwester looks for SpeedEvil's patch... Oct 13 15:08:19 It's not out there. :) Oct 13 15:08:41 It's currently on a drive that's disconnected while I move data around prior to getting a build system up Oct 13 15:11:57 if the iphone is BSD powered, couldn't i just hack into Shell, and programm the stuff i wantt_ Oct 13 15:12:17 can i compile games like supertux for openmoko? Oct 13 15:12:38 der_papst_623: Supertux has already been showing running, so yes Oct 13 15:13:24 good Oct 13 15:13:48 oh yes, is there something like a joystick for openmoko? Oct 13 15:13:49 i'm just asking cause i need the geo with wifi, cause my cellphone died, and i can't wait to december Oct 13 15:13:51 der_papst_623: the only problem is that you don't have any keys Oct 13 15:14:05 i can't imagine, playing supertux with a touchsscreen Oct 13 15:14:39 Vegar: ok, thats an answer... Oct 13 15:15:09 der_papst_623: you could, of course, use a bluetooth keyboard or joystick or whatever Oct 13 15:15:22 kristian-m, the new package should be online Oct 13 15:15:27 Vegar: steering wheel! Oct 13 15:15:33 Oooh! Oct 13 15:15:38 kristian-m, I like the green a lot :P Oct 13 15:15:41 You put it on a steering wheel holder. Oct 13 15:15:52 And use the accelerometers to pickup direction. Oct 13 15:16:08 (do not drive while playing) Oct 13 15:16:08 or you can use the GPS to detect movement Oct 13 15:16:17 i meand a joystick to plug in... Oct 13 15:16:48 it would be funny to build a small joystick for it Oct 13 15:17:01 you could probably connect it to the USB port Oct 13 15:17:13 (02 has powered USB, doesn't it?) Oct 13 15:17:15 Ghiottone yeah green is really bright - try blue on red it looks like 3d somehow... Oct 13 15:18:05 kristian-m, yeah right :) Oct 13 15:18:13 i'm on my way to a electronic engineer, but i've never written a driver... Oct 13 15:18:21 kristian-m, you get x,y coords for each touch? Oct 13 15:18:47 Ghiottone when ever the gdk component gives me an event, yeah Oct 13 15:18:48 kristian-m, is there a way to know how wide the touch is (ie if I use my finger tip or the whole finger) Oct 13 15:19:18 Ghiottone not directly - you could however find out if a second finger touches the screen Oct 13 15:20:01 kristian-m, at the same time? I tought the screen wasn't multi touch Oct 13 15:20:02 Ghiottone you could detect a jump in the x,y coords while the pen is still in pressed state Oct 13 15:20:11 Ghiottone its not Oct 13 15:20:12 kristian-m, ah I see Oct 13 15:20:18 Ghiottone try two pens Oct 13 15:20:32 kristian-m, it could be perhaps used to guess the size of the brush Oct 13 15:20:33 Ghiottone the coord will be exactly between them Oct 13 15:21:26 Ghiottone hmm, that would be hard - the touches would have to happen in the right order Oct 13 15:21:29 can i use kde on openmoko? Oct 13 15:21:29 or is it not fast enough for it? Oct 13 15:21:48 it's probably doable Oct 13 15:22:09 kristian-m, or maybe guessing the greatest distance Oct 13 15:22:26 kristian-m, between all the points touched Oct 13 15:22:38 kristian-m, until pen is pressd Oct 13 15:22:41 Ghiottone well i can only detect jumps if they are really of great distance Oct 13 15:23:01 Ghiottone otherwhise it'll look as if the pen was moved Oct 13 15:23:39 Vegar, ... so :) Oct 13 15:23:39 kristian-m, yeah I see now. not really doable Oct 13 15:24:08 if i have 20 minutes of battery life, but can show compiz running on my phone to my friends, ill be happy Oct 13 15:24:43 cb22: bling addicts :) Oct 13 15:24:54 kristian-m, could you confirm pyPenNotes package is working? Oct 13 15:25:05 Ghiottone the only thing i could imagine would be to use 2 pens, keep one where it is and then measure the action ot the other - you could use it to zoom - but would have to be sure about the jump as those happen in nomal operation too when the device is busy Oct 13 15:25:07 * ScaredyCat does a super e and makes you all small Oct 13 15:25:10 kristian-m, the icon is at the end of the list as it start with a lowercase letter Oct 13 15:26:12 cb22: Not to mention having it interact with the accelerometers ;) Oct 13 15:28:24 ScaredyCat super e? €? Oct 13 15:30:34 Ghiottone okay, it's working thanks Oct 13 15:30:36 super e in compiz-fusion Oct 13 15:31:29 ScaredyCat ahh :-) Oct 13 15:32:24 kristian-m, great! thanks to you Oct 13 15:41:06 is information regarding gps on the wiki up to date? Oct 13 15:41:34 Sup3rkiddo: Probably not, as the decision apparently happened less than 24 hours ago Oct 13 15:42:27 abraxa_, hmm, thats why i asked, i mean, i am looking forward to create some python bindings for it..looks like i have to wait a bit longer Oct 13 15:43:07 Sup3rkiddo: Why? You'd want to create bindings for gpsd anyway Oct 13 15:43:26 abraxa_, what is the decision taken ? Oct 13 15:43:34 daltona: See topic ;) Oct 13 15:43:44 ok :( shame on me... Oct 13 15:43:59 abraxa_, sorry, thats what i meant, Oct 13 15:44:05 * Sup3rkiddo is kinda drunk Oct 13 15:44:14 Sup3rkiddo ;) Oct 13 15:44:16 Sup3rkiddo, ooh, python gps bindings! Oct 13 15:44:24 still nothing about GTA01 GPS availability ? Oct 13 15:44:29 And well, gpsd is not OM-specific so you can already start Oct 13 15:44:30 Sup3rkiddo, missed the ballmer curve? Oct 13 15:44:42 heh Oct 13 15:45:01 * daltona is using a BT GPS while waiting his NEO GPS to work.. Oct 13 15:45:59 man, 1 1/2 bottles of beer and i am a bit clunky..stupid indian beer Oct 13 15:47:03 those bottles of iadian beer are 1liter aren;t they? Oct 13 15:47:17 indian Oct 13 15:47:21 ScaredyCat, 650 ml.. Oct 13 15:47:35 ahh.. you must be a pussy then ;) Oct 13 15:47:48 lol, yeah...and that too i had lager Oct 13 15:48:14 * ScaredyCat sips lemonade Oct 13 15:50:53 * cb22_ sighs... Oct 13 15:51:01 my 4gb microsd and my bt keyboard just arrived yesterday Oct 13 15:51:12 now all that is missing is the phone :( Oct 13 15:51:16 can anyone confirm if they are python bindings for gpsd, wouldnt want to reinvent a wheel Oct 13 15:51:39 http://www.perrygeo.net/wordpress/?p=13 Oct 13 15:52:02 looks like there are Oct 13 15:52:31 they look pretty cool too Oct 13 15:52:33 import gps Oct 13 15:52:39 session = gps.gps() Oct 13 15:52:53 print session.fix.latitude Oct 13 15:54:13 oh dang, no i am without work.. Oct 13 15:54:22 s/no/now Oct 13 15:55:21 probably a good job if you're drunk Oct 13 16:04:14 hi Oct 13 16:15:45 lo Maledictus Oct 13 16:25:54 hey cb22_ Oct 13 16:26:07 I read the gps news. Oct 13 16:26:14 nice hey Oct 13 16:26:31 it will speak standard nmea Oct 13 16:26:41 pretty cool Oct 13 16:26:47 I'm so glad to hear that :D Oct 13 16:27:08 any news on the freebsd front? Oct 13 16:27:11 I guess the hardware is getting stable then Oct 13 16:27:18 i'm busy csupping current Oct 13 16:27:22 no, sorry. Oct 13 16:27:35 hey you can use the new RELENG_7 branch Oct 13 16:28:02 any reason why the kernel config is so minimal? Oct 13 16:28:16 like thinks like ohci, etc are disabled? do they break it? Oct 13 16:29:22 It really isn't interesting what is in the kernel right now, we are not to the point where that will be. Oct 13 16:29:49 ah right Oct 13 16:29:56 does the serial console bit work yet? Oct 13 16:29:57 anyone got a link to the new GPS chip info? Oct 13 16:30:12 rwhitby, http://www.u-blox.com/products/atr0635.html Oct 13 16:30:16 rwhitby: I just googled for ATR0635 Oct 13 16:30:26 rwhitby: info on the chip or the news? Oct 13 16:30:49 wait.. Oct 13 16:30:59 in a nutshell: speaks nmea, -160dbm rx, WAAS (+europe version, +asia version), DGPS, Oct 13 16:31:07 galileo ready? Oct 13 16:31:13 CEO of ARM on radio 4 being interviewed now Oct 13 16:31:22 is it galileo ready?! Oct 13 16:31:47 galileo? Oct 13 16:32:05 cb22_: i added the info to http://wiki.openmoko.org/wiki/GPS Oct 13 16:32:11 cool Oct 13 16:33:12 I'm trying to figure out what to add to http://wiki.openmoko.org/wiki/GTA02_Openness Oct 13 16:33:26 Now I've got my neo working with GSM manually, except for the microphone. I've tried gsmheadset.state and gsmhandset.state, but to no avail. Oct 13 16:33:42 shew, successfully got out of helio contract Oct 13 16:33:57 Maledictus, the GPS info? Oct 13 16:34:26 So it's been confirmed officially that the HH won't be in GTA02? Oct 13 16:34:32 yep Oct 13 16:34:39 A link to documentation Oct 13 16:34:41 Damn. Oct 13 16:35:09 SpeedEvil: why damn? Oct 13 16:35:17 We're very happy with the u-blox/Atmel ATR0635 GPS Receiver, and thus Oct 13 16:35:17 have made the decision to use this chip. Oct 13 16:35:39 so the gps chip should be able to keep operating with the main cpu off Oct 13 16:35:40 cb22_: wheres that from? Oct 13 16:35:42 Because the HH is potentially much more flexible than a device that simply speaks NMEA. Oct 13 16:35:52 Maledictus, the mailing list Oct 13 16:36:02 Maledictus: "Bi-weekly OpenMoko community update" on the community list Oct 13 16:36:09 SpeedEvil: but it's closed and I'm happy that the focus is on open ;) Oct 13 16:36:29 It's closed at the moment. Oct 13 16:36:35 time to subscribe there, only have kernel right now Oct 13 16:36:55 Up until the point where it was announced that the HH may not be in GTA02 there was good progress going on reverse engineerign. Oct 13 16:37:13 To the extent that we may have had a driver in December - a fully opensource one. Oct 13 16:37:19 oh, how wonderful. I hate re Oct 13 16:37:27 the GPS protocols are documented in http://www.u-blox.com/customersupport/gps.g3/ANTARIS_Protocol_Specification(GPS.G3-X-03002).chm Oct 13 16:37:34 The issue for me is the eventual capabilities. Oct 13 16:37:45 rwhitby: thank you :) Oct 13 16:37:59 SpeedEvil, out of interest, what would the hammerhead be able to do that a nmea only chip cant? Oct 13 16:38:03 The raw nature of the HH chip means that you can do lots of smarts in software. Improving accuracy. Oct 13 16:38:12 hmm. is there another format than .chm? Oct 13 16:38:15 no Oct 13 16:38:28 ok Oct 13 16:39:15 let me see if I can get it into PDF and attach it to the GPS page Oct 13 16:39:18 cb22_: several things. Wake up for 1s every 3 min (on timer) get fix, go to sleep. Post-correct tracks that have been downloaded for sub-meter accuracy. Oct 13 16:39:45 cb22_: download realtime very local correction information, again for sub-meter accuracy. Oct 13 16:39:45 is it true that the neo can only last for a few hours off the charger? Oct 13 16:39:51 mattzerah: at the moment. Oct 13 16:40:00 mattzerah: power managment is not done yet. Oct 13 16:40:17 apparently Oct 13 16:40:33 i raised this issue back in june and people didn't seem to think it was important Oct 13 16:41:20 how much power did the HH consume? Oct 13 16:41:21 They've all been busy working on the gta02 hardware, I'm hopeful we'll see some progress on the kernel and basics soon. Oct 13 16:42:09 cb22_: also NMEA GPSs output the current position. Any errors that the GPS discovers can only be implemented by a jump back to the correct position. If you delay producing the position, then you can filter both back and forward in time for a much better accuracy of a track - even absent other correction info. Oct 13 16:43:18 Maledictus: I can email you a PDF if you like. Oct 13 16:43:39 SpeedEvil: seems like atr0635 has it's own proprietary protocoll too Oct 13 16:43:46 rwhitby: mn@bsdgroup.de thanks you :) Oct 13 16:43:59 -s Oct 13 16:44:25 Maledictus: I need to look at that yws - I'd also appreciate an email rwhitby - ublox@mauve.plus.com Oct 13 16:44:36 Thanks! Oct 13 16:44:44 hopefully, power management is extremely important in "does everything" devices Oct 13 16:44:57 Of course it is. Oct 13 16:45:14 In some ways we could really do with an embedded controller. Oct 13 16:45:28 The minimum power mode of the CPU is really quite high. Oct 13 16:45:44 Maledictus, SpeedEvil: sent - can you please put this somewhere public and link to it from the wiki GPS page? Oct 13 16:46:15 speedevil: even in standby? Oct 13 16:47:13 SpeedEvil: how does the UBX binary protocol look? Oct 13 16:47:34 mattzerah: no - not in standby. There is no mode suitable for doing really slow stuff. Oct 13 16:47:54 rwhitby: ok, i'll look Oct 13 16:48:00 rwhitby: I'm not going to be able to look at it for an hour or so. Oct 13 16:48:18 up to an hour or so Oct 13 16:49:13 looks pretty detailed, but I have no idea what I am looking at :-) Oct 13 16:49:21 mwester: you said that they've been working on gta02 hardware... why are openmoko developers working on gta02? Oct 13 16:49:47 cause that's the one they are going to sell to the mass market Oct 13 16:50:06 Maledictus: and the other reason for the 'damn' - I have wasted a small chunk of my life on trying to figure out the protocol. Oct 13 16:51:16 rwhitby: but neo1973 is not developed by the openmoko team... i thought. Oct 13 16:51:23 sure it is Oct 13 16:51:36 SpeedEvil: that's unfortunate :) Oct 13 16:51:38 SpeedEvil: are you sure it's such a waste? It's not like all GTA01s will self-destruct when 02 is out. Oct 13 16:51:50 roh: cesarb might be interested in the wiki rework, fyi - he seems quite experienced with wikipedia work Oct 13 16:52:13 Vegar: it's a waste in that there is probably several hundred hours more work to get a workable GPS driver. Oct 13 16:52:42 Vegar: And the utility of that for a device which only has a volume of 3K is debatable. Oct 13 16:52:46 SpeedEvil: There are other devices using the global locate chip Oct 13 16:52:50 hehe Oct 13 16:53:01 Yes, there are. Oct 13 16:53:17 But in many ways the open-source driver is half of it. Oct 13 16:53:21 The other half is: Oct 13 16:55:47 http://wiki.openmoko.org/wiki/Server:A-GPS Oct 13 16:56:55 It's not impossible that with a good implementation of the open-source driver, the app that provides the information for this could go into every OM phone, and enable the user to simply click 'yes' in order to contribute. Oct 13 16:56:59 GTA02v3, the third revision of GTA02, is still being verified and Oct 13 16:56:59 tested. This will take another couple of weeks, at which point we'll Oct 13 16:56:59 make GTA02v4. Oct 13 16:57:06 that means the v3 had bugs? Oct 13 16:58:04 For much of the other stuff with the HH - tomtom GPSs forex - you can't do this, as they have no external comms. Oct 13 16:58:38 PBeck, nope Oct 13 16:58:47 its still being tested Oct 13 16:58:50 as you said Oct 13 16:59:06 PBeck: which hardware hasn't? Oct 13 16:59:32 cb22_: i think they have finished the verifing and testing process? Oct 13 16:59:38 dunno Oct 13 16:59:47 Maledictus: gta0v3 Oct 13 16:59:51 *gta02v3 Oct 13 17:00:12 i have no idea Oct 13 17:00:24 It was previously indicated that there will be external testing of GTA02 - before it goes on sale Oct 13 17:00:46 cb22_: the above posting was from Michael Shiloh Oct 13 17:00:52 So - it's not going to be this batch that you can buy Oct 13 17:01:09 SpeedEvil, maybe they will be nice and send one over to you :) Oct 13 17:01:16 SpeedEvil: yes, but why we make v4? Oct 13 17:01:26 SpeedEvil: when v3 has no bugs Oct 13 17:01:40 PBeck: v3 may have no bugs, but may lack features. Oct 13 17:01:53 For example, they may have found an easy way to connect a FPGA. Oct 13 17:02:09 (kidding) Oct 13 17:02:19 SpeedEvil: why they no integrate them in the first version? Oct 13 17:02:45 PBeck: because they were not specified. Oct 13 17:02:53 SpeedEvil: when the make a new phone then its not a gta02v4 ;) Oct 13 17:03:15 PBeck: you typically go for a revision that you think looks right. Oct 13 17:03:23 gta03 or what ever for a new device Oct 13 17:03:39 PBeck: but it will be a very rare device that has no bugs, or is not missing features that you realise would be nice when you get the hardware. Oct 13 17:04:13 PBeck: so - you add the features, and hope you don't introduce more bugs. Or a part becomes unavailable, and you've got to respin. Oct 13 17:04:31 SpeedEvil: why should the integrate new features now? I think that could be for a new device Oct 13 17:04:33 PBeck: or you find out that there is a much better part you could use, and need to respin Oct 13 17:05:06 PBeck: consider a more featured PMU chip that has an embedded controller, and can do all of the USB negotiation on its own, even with a flat battery. Oct 13 17:05:29 SpeedEvil: you has never a optimal solution Oct 13 17:06:01 No. However there are points at which it makes sense to respin, rather than stick with existing flawed designs. Oct 13 17:06:08 theres no way i could get gps working on gta01, is there? Oct 13 17:06:28 Unfortunately, until the 'real soon now' occurs, no. Oct 13 17:06:42 SpeedEvil: yes, but i think that ideas should be in a new device and not on a bug-free device that can be produced :) Oct 13 17:06:52 +used Oct 13 17:07:20 feeping creaturim Oct 13 17:07:26 feeping creaturism, even Oct 13 17:07:29 how is the nick of Michael Shiloh on irc? Oct 13 17:07:59 There is feeping creaturism - and then there is fixing nastinesses in the hardware you have to work round. Oct 13 17:08:08 SpeedEvil: http://www.u-blox.com/customersupport/gps.g4/ANTARIS4_Modules_SIM(GPS.G4-MS4-05007).pdf is an interesting read Oct 13 17:08:12 what means feeping creaturism? Oct 13 17:08:30 i dont find a translation ^^ Oct 13 17:08:40 rwhitby: ah - can download that. Oct 13 17:08:49 Sup3rkiddo: There are ways, but the openmoko team has not made that solution available, apparently due to contract problems. Oct 13 17:09:02 PBeck: creeping featurism Oct 13 17:09:10 mwester, oh. ok. Oct 13 17:09:18 PBeck: the tendancy to add more features because they are individually cheap or easy to add. Oct 13 17:09:28 PBeck: without considering the overall price, complexity, ... Oct 13 17:09:28 http://en.wikipedia.org/wiki/Creeping_featurism Oct 13 17:10:26 ah ok thanks SpeedEvil and cb22_ Oct 13 17:10:37 cjb_ie ;) Oct 13 17:10:47 tab is horror :) Oct 13 17:16:47 fresh email on the community mailing list Oct 13 17:19:19 "Apparently we have resolved the legal issues with the driver for the GPS chip on GTA01, and the problem right now is that we switched from OABI to EABI. The original driver from Global Locate / Broadcom worked for OABI but is not quite working for EABI. Work is proceeding on this." Oct 13 17:19:43 So OpenMoko should be able to distribute gllin for use in an OABI chroot right now. Oct 13 17:20:50 rwhitby: Yes, however roh doesn't have the license at hand that needs to be packaged with it Oct 13 17:21:02 rwhitby: do you know which chip of the range it is? Oct 13 17:21:45 SpeedEvil: http://www.atmel.com/dyn/products/product_card.asp?part_id=3932 Oct 13 17:22:58 Sorry - I'm having assorted network problems :/ Oct 13 17:23:14 http://www.atmel.com/dyn/resources/prod_documents/doc4928.pdf Oct 13 17:30:06 From the impression I get - this is a much more 'cooked' reciever, which is good for software complexity, but bad for flexibility. It would be ideal if it was flashable - but it's not. Oct 13 17:30:30 I need to work out the limits to the limited configuration that can be done. Oct 13 17:30:40 I need to work out the limits to the limited configuration changes that can be done. Oct 13 17:32:30 abraxa_: The source? :o Oct 13 17:32:48 jorgenpt? Oct 13 17:33:02 abraxa_: Is the source of gllin being distributed, or the binary? Oct 13 17:33:15 binary I expect Oct 13 17:33:19 Binary Oct 13 17:35:26 how is the finger keyboard on the neo?.. would you be able to type on ssh ? Oct 13 17:37:22 not really Oct 13 17:42:16 warning to everyone, dont click "I feel lucky" on google while searching for glinn..thank you Oct 13 17:43:19 glinn or gllin? Oct 13 17:43:35 you should type gllin (global locate linux) Oct 13 17:44:03 Vegar, theres a good chance someone would type glinn, i did and got a "OMG" from my dad a few minutes ago Oct 13 17:44:15 hehe Oct 13 17:45:23 but thankfully i told him about the typo and now he is sleeping knowing his son what he thought to be..heh Oct 13 17:45:36 oops...ain't Oct 13 17:45:46 strange, my google doesn't even have "i have lucky anymore" Oct 13 17:45:47 worst typo eva Oct 13 17:46:05 anyway, the top google result for glinn for me is http://memory-alpha.org/de/wiki/Glinn Oct 13 17:46:39 Henryk: that's because your google prefers german results Oct 13 17:46:51 (hl=de in the URL) Oct 13 17:46:55 yeah, sometimes I hate it for that Oct 13 17:47:58 (Even worse: The firefox search field was initially set to return only german language results. I didn't notice that at first, but wondered why all my searches did return so few results. When I finally noticed I basically had to redo half a day worth of searches) Oct 13 18:01:09 apart from certification problems, https://www.angstrom-distribution.org/unstable/sources/liberation-fonts-ttf-3.tar.gz is not found *at all* (so much for the ttf-liberation-0.2 bitbake package) Oct 13 18:08:08 Kero: it's httpS Oct 13 18:08:15 change it to http and it will work Oct 13 18:08:37 god knows why someone put https Oct 13 18:09:01 When I'm using ssh over usb, it seems to disconnect / reset the connection once in a while (every ssh terminal freezes) .. is that a known problem? Oct 13 18:12:33 ScaredyCat: erm. in which file? Oct 13 18:13:17 1 sec lemmi find it Oct 13 18:13:56 Mammux72: Never seen that before - maybe your cable is bad? Oct 13 18:14:38 ttf-liberation_0.2.bb Oct 13 18:14:49 in openembedded/packages/ Oct 13 18:15:07 in openembedded/packages/ttf-fonts Oct 13 18:16:02 just change the SRC_URI Oct 13 18:16:41 ScaredyCat: that's a different one. Oct 13 18:16:50 from redhat Oct 13 18:17:41 well, once you have it in your sorces, and add the .lock then it doesn't matter Oct 13 18:17:45 sources Oct 13 18:18:14 if you need it I can email the files to you Oct 13 18:18:39 Hey, guys. I'm interested in your opinion on an e-book project... I'm sure there are lots of opinions available for free here :-) Oct 13 18:18:47 I have the file, I added the .lock and a .md5 Oct 13 18:18:56 worked for a previous file, not for this one Oct 13 18:19:28 * ScaredyCat searches for dodgy file Oct 13 18:19:38 I'm interested in contacting gutenberg.org, and seeing if we can provide a smartphone-friendly ebook catalog, with e-books auto-formatted for Neo1973. Oct 13 18:20:02 isn't all their stuff essentially in the public domain anyway? Oct 13 18:20:42 Auto-formatting for devices is insane. Oct 13 18:20:43 yes... Oct 13 18:20:46 As you can't. Oct 13 18:21:06 Does the user want to read in sunlight, or at night. Oct 13 18:21:21 Yes, it's public-domain. There is an auto-formatter for the iPhone e-book reader that I used while I had an Iphone. It worked well for most texts. Oct 13 18:21:31 Kero: I don;t know which .bb file is trying to get that uri - none of mine contain it Oct 13 18:21:32 80*25 chars, or 8*2 Oct 13 18:21:35 The main thing was to break it into chapters, and create an index. Oct 13 18:21:44 And add backreferences. Oct 13 18:21:50 I find that so useful. Oct 13 18:21:52 One wants to have the source in some nice meta-format to convert to everything else... Oct 13 18:22:13 Any proper name is backreferenced if it hasn't occurred in 10 pages. Oct 13 18:22:18 Kero: someone else complained earlier today, so I did a test rebuild to chek, fine here... Oct 13 18:22:22 back to the first occurrance. Oct 13 18:22:34 However, I had to run the gutenmark and gutensplit apps manually... a huge pain. I'd like to just browse books on-line from the phone, and select what I want. gutenmark and gutensplit could run on the server before download starts. Oct 13 18:22:38 * ScaredyCat dives into a bath Oct 13 18:23:18 ScaredyCat: scariest, when I take the s out of https of redhat, it downloads http://downloads.openmoko.org/sources/liberation-fonts-ttf-3.tar.gz and builds succesfully. Oct 13 18:23:21 SpeedEvil: backreferences sounds very nice. Oct 13 18:24:22 xyzzy_bill: for me - I have minor memory problems with chars. Oct 13 18:24:37 If I put a book down for a week, I can really lose the plot (literally) Oct 13 18:24:59 SpeedEvil: Users mostly read inside during day, or at night. I have the exact same problem with books -- I'm really not quite normal that way. Oct 13 18:25:16 simply being able to click on names and get the first occurance, rather than having to read the beginning again is nice. Oct 13 18:25:25 SpeedEvil: I have to read a book in a few days, so that I don't loose track of the plot. Oct 13 18:25:32 xyzzy_bill: I mean that you can't define it for a device. Oct 13 18:25:43 xyzzy_bill: only for a user/device/environment Oct 13 18:25:47 SpeedEvil: Ok, that sounds very cool, and quite doable. Oct 13 18:26:01 xyzzy_bill: and then there is stuff like 'Wheel of time' Oct 13 18:26:10 where there are eleventy-billion chars. Oct 13 18:26:48 (though the wheel has fallen off the axle) Oct 13 18:27:16 well, at least then you can compress the hell out of the database without fear of having to update it... Oct 13 18:28:18 'k, build continues. Still, these kind of things are utter crap :( Oct 13 18:29:39 So, I don't even see an e-book reader listed on the projects.openmoko.org site. Is there really no e-book reader yet? Oct 13 18:30:02 I use lynx for book reading. Oct 13 18:30:21 what fontsize do you use on neo for ebooks? Oct 13 18:30:29 xyzzy_bill: i think i got something to read djvu running some weeks ago Oct 13 18:30:30 I haven't actually done much reading. Oct 13 18:30:38 Though I did use FF to read a bit. Oct 13 18:30:44 buz: user-selectable font-size. I use larger fonts - my eyes are getting old. Oct 13 18:30:59 yeah but what are YOU using Oct 13 18:31:16 The "Books" reader on iPhone was quite good. I read several texts until Apple borked my phone. Oct 13 18:31:36 how? Oct 13 18:31:41 But, you had to jump through hoops to get a book onto the phone in a "Books" friendly format. Oct 13 18:31:50 Ah. Oct 13 18:32:11 Minimally marked up HTML is my favourite format. Oct 13 18:32:17 Let that be rendered however. Oct 13 18:32:25 Fixed pages - PDF is _insane_. Oct 13 18:32:54 depends on the subject Oct 13 18:32:59 True. Oct 13 18:32:59 SpeedEvil: Yes, PDF is quite bad. The "Books" reader just used basic formatted text, but we lost all the pictures. Oct 13 18:33:07 I mean for stuff like novels. Oct 13 18:34:20 For stuff like textbooks, with complex diagrams, you need much more complex formatting. Oct 13 18:34:27 I liked several "Books" features. There was a button for next page, prev page, contents, and it remembered which chapters you read, so even if you lost your place, you could quickly find it again. Oct 13 18:35:02 There was also a list of titles currently on the phone, which was nice. Oct 13 18:35:19 I think there is an app page. Oct 13 18:35:22 search the wiki for book Oct 13 18:35:24 or ebook Oct 13 18:37:40 I found a link talking about possibly porting an ebook reader: http://wiki.openmoko.org/wiki/Reader Oct 13 18:38:33 I'm just floundering around trying to find where I can contribute well. I could write and/or port an e-book reader. Would that be a high-value task? Oct 13 18:39:15 alright, who broke the build :P Oct 13 18:39:27 webkit again >_< Oct 13 18:40:02 xyzzy_bill: for me, an offline capable rss reader would be worth more Oct 13 18:40:04 but that's jsut me Oct 13 18:40:28 think reader that downloads all links in a stream for later offline reading Oct 13 18:40:29 buz: I agree that would be good. Oct 13 18:40:31 Kero: that's what I told you! Oct 13 18:40:59 it's just daftness on the .bb file authors part Oct 13 18:41:11 xyzzy_bill: maybe some thing wrapping together httrack and webkit would do that just fine Oct 13 18:41:45 httrack is by far the best app to save offline copies of pages Oct 13 18:42:34 buz: can webit be used to just render html, while I add the buttons/controls I want to the app? Oct 13 18:43:15 xyzzy_bill: You might want to talk to zecke if you're interested in this task as he's the author of the rss reader used on OpenMoko Oct 13 18:44:06 buz: have you seen wwwoffle? http://www.gedanken.demon.co.uk/wwwoffle/ Oct 13 18:45:57 xyzzy_bill: i thought that was the point of webkit ;) Oct 13 18:47:11 I think an e-book reader is basically different than an RSS reader. A webkit based version sounds like a good idea, especially if webkit is already going to be on all our phones. The post just now by rtyler worries me... Oct 13 18:47:13 rwhitby: if you want a complete solution for browing, wwwoffle is ok, but if you want to store pages for offline viewing or maybe even crawl them, httrack is what you have to beat Oct 13 18:47:41 ScaredyCat: to take the S out of URL2, because URL1 does not work, so it uses URL3, which is unequal to either-without-s ? did you really say that? :) Oct 13 18:48:05 hmm, for ebooks djvu would be better though. it's basically finished and now only needs a gui that'll work on 640x480 Oct 13 18:48:40 i'd prefer ebooks in some sane format, i dont consider djvu to be sane Oct 13 18:48:50 (unless we're talking about scanned stuff) Oct 13 18:49:01 yeah, sane is markup Oct 13 18:49:14 yeah Oct 13 18:49:24 maybe even use webkit to render some sane xhtml Oct 13 18:50:34 Many gutenberg e-books are already provided in html, and those that are just text can typically be automatically split into chapters with a table of contents. A webkit based reader does sounds pretty sane to me. Oct 13 18:51:54 yeah and rendering html to different screen sizes is very simple Oct 13 18:51:58 not so for djvu or pdf Oct 13 18:53:09 and if you're really lazy, you can do without any custom moko gui work Oct 13 18:53:12 just do it in xhtml, too Oct 13 18:53:29 Should I try to talk to the minimo guys, since they already ported webkit? Oct 13 18:53:51 isnt minimo gecko based??? Oct 13 18:54:08 beats me. I was wondering why everyone seems to hate minimo Oct 13 18:54:34 What's the difference between webkit and gecko? Oct 13 18:54:36 because they develop firefox for windows ce, basically Oct 13 18:54:51 webkit: safari (konqueror soon), gecko: firefox, mozilla et al Oct 13 18:55:07 Is webkit better? Oct 13 18:55:07 on a phone, you rather want webkit Oct 13 18:55:13 why? Oct 13 18:55:19 it's faster Oct 13 18:55:29 i'm guessing it's also smaller Oct 13 18:55:39 and more widely supported, it's used by adobe, apple, kde, trolltech, nokia now Oct 13 18:55:42 Is anyone doing a webkit based OpenMoko browser? Oct 13 18:55:46 (kde soon, anyway) Oct 13 18:55:53 i think so, have seen screenshots on planet Oct 13 18:56:13 webkit compiles on the platform and is used by the feedreader, at least to a degree Oct 13 18:56:40 The RSS feedreader? Oct 13 18:56:52 I'm just trying to figure out who to talk to about webkit port. Oct 13 18:56:54 yes Oct 13 18:56:55 yeah but it doesnt support offline reading afaict Oct 13 18:57:11 zecke is the author of the feedreader Oct 13 18:58:15 xyzzy_bill: yes, webkit is already routinely built and used on OpenMoko (and tends to break builds for lots of people :) Oct 13 18:58:43 Ok, Thanks. How can I contact zecke? Oct 13 18:58:46 (s60 browser IS using webkit, right?) Oct 13 19:00:26 I found zecke@selfish.org on Google Oct 13 19:02:52 Thanks, guys, for the insights. Oct 13 19:03:01 gotta go feed the kids :-) Oct 13 19:13:24 so, anyone got an idea about how the OpenGL ES on the SMedia is actually going to work? Oct 13 19:16:19 cb22_ not till the driver is done'ß Oct 13 19:16:37 cb22_ then it can so egl1.2 or so i think Oct 13 19:17:04 so will it be very special? Oct 13 19:17:13 or could i type in glxgears and have them appear? Oct 13 19:17:26 * cb22_ has no knowledge about drivers... Oct 13 19:26:41 Hey, I just registered for the ebook project on projects.openmoko.org. Hopefullly, I'll be able to make our phones just a bit cooler! Oct 13 19:34:33 xyzzy_bill: hopefully, gta02 will ship on time, so i can actually use your app ;) Oct 13 19:36:42 cool, the magic of opensource, everyone helps Oct 13 19:36:47 and makes the phones cooler Oct 13 19:36:53 buz: Heck, I can't read that 2.8" screen... I figure I'll write the app, and hope tat the GTA-03 has a 3.5" :-D Oct 13 19:37:12 make that 4" widescreen for all i care Oct 13 19:37:46 800x480? Oct 13 19:37:52 using that 800x480 screen from the toshiba phone Oct 13 19:38:09 and a sliding qwerty keyboard Oct 13 19:38:51 and mimo hsdpa Oct 13 19:38:55 i wonder what would happen if you wrote a soft ipod interface Oct 13 19:39:20 technically, you can do it on most touchscreens Oct 13 19:39:28 could even improve on it Oct 13 19:43:49 buz: methinks you can do two finger scaling, but not rotation Oct 13 19:44:02 one finger rotation is good enough Oct 13 19:44:13 i once tried it on the drawing app on my p900, it does work Oct 13 19:44:28 besidesm you can do better than rotation on a touch screen Oct 13 19:45:04 i've long been of the opinion that a sliding scale would be far preferable to a wheel Oct 13 19:45:19 slide further from the center to accelerate scrolling, stop when you take your finger off Oct 13 19:45:42 scrolling through long lists tends to be literally painfull on the ipod Oct 13 19:47:27 buz: It's not all that pleasant with the current implementation of the finger scroll container either Oct 13 19:48:08 yeah thats why i would base scrolling speed on distance from center of scroll bar Oct 13 20:01:59 abraxa_: well, I'm an admin on en.wikipedia.org, so it's no wonder I have some experience with mediawiki-based wikis ;-) Oct 13 20:02:42 * juri_ bows in the presence of greatness. ;P Oct 13 20:03:44 cesarb: Ahh, hehe... well, in case you missed it: roh was looking for wiki-experienced folks to reorganize the OM wiki Oct 13 20:04:10 i'm a trac user myself. *shrugs* Oct 13 20:05:29 whatever happened to the asus eee? that would be the perfect companion to neo Oct 13 20:10:03 http://www.engadget.com/2007/10/12/asus-eee-pc-ships-this-month-199-but-not-for-you/ Oct 13 20:10:48 that sucks Oct 13 20:11:42 nabend Oct 13 20:11:54 re hhf Oct 13 20:12:33 hi woglinde Oct 13 20:13:21 does anyone have a count or guesstimate of shipped gta01? Oct 13 20:35:25 re Oct 13 20:35:51 hhf423, @ # of gta01, i heard 3000 or so Oct 13 20:36:46 the first batch was something like 2000 or already 3000? they should now be selling the second production run, right? Oct 13 20:51:34 Anyone in the last several days happen to have a ticket number? Oct 13 20:52:34 what kind of ticket numer? Oct 13 20:53:24 I mean a ticket number as is given out with orders. However I'm not sure if they haven't changed. Oct 13 20:53:31 changed order format Oct 13 20:55:30 We could simply ask the next new owner to appear in this channel to tell his gta01 serial, or an aproximation of it Oct 13 20:55:43 They have serials? Oct 13 20:55:59 yes Oct 13 20:56:11 * SpeedEvil contemplates getting up. Oct 13 20:56:23 I have gta01 bv4 with serial 000234 Oct 13 20:56:33 from 20070420 Oct 13 20:56:46 where would the serial be printed? Oct 13 20:57:01 Henryk: sticker on back side of device Oct 13 20:57:03 mine is from the end of 200707, and has serial in the 700s Oct 13 20:57:18 my gta01 bv3 is from 20070207 and has 000193 number Oct 13 20:57:56 s Oct 13 20:58:15 * cesarb wonders if this will turn in a slashdot-like "my ID is lower than yours routine", where each low-id user is countered by an even more low-id user Oct 13 20:58:27 s/n 8B7600121? Oct 13 20:58:31 hmm, the S/N starts with 8A, not with 00 Oct 13 20:58:45 mine starts with 8A too Oct 13 20:59:02 Well - I have 121 - so I beat you both. Oct 13 20:59:07 :) Oct 13 20:59:08 lol Oct 13 20:59:10 SpeedEvil: :D Oct 13 20:59:22 Dunno if that's the first and second batch of 500. Oct 13 20:59:48 cesarb: i think my gta01bv2 doesn't have a serial number. but there's a hand written sticker that says "#11" Oct 13 20:59:52 SpeedEvil: or perhaps separate assembly lines running in paralell Oct 13 20:59:57 Henryk hrhr Oct 13 20:59:58 Perhaps. Oct 13 21:00:01 Henryk: lol Oct 13 21:00:07 Henryk: gta01bv2 were internal only ;) Oct 13 21:00:09 Henryk yep.. thats a too-early one Oct 13 21:00:12 It's never been cleared up what the production was for GTA01 IIRC Oct 13 21:00:25 how the production was done. Oct 13 21:00:30 * cesarb sees a new wiki page: [[GTA01 serial numbers]] explaining (or wildly guessing) the serial number format :D Oct 13 21:00:30 bv3 was first sent to devels outside of openmoko Oct 13 21:01:19 ;D Oct 13 21:01:22 the gta01bv4 s/n ends in 1008, date code 20070724 Oct 13 21:01:29 SpeedEvil: I think bunnie's somewhat recent posts (even if they are not OM/neo related) help a lot visualize how things are done in the chinese factories Oct 13 21:02:33 ces Oct 13 21:02:46 : I was meaning more was it two batches of 500, or what. Oct 13 21:05:05 I assume you mean the ones with girls and lots of bins of different SMD parts? Oct 13 21:06:06 SpeedEvil: or the more recent ones with a mix of manual labor and places-6-parts-per-second-and-doesn't-even-get-tired machines Oct 13 21:06:12 i think that's only the case for small batches, there's just no chance humans can keep up with robots for any mid to large scale production Oct 13 21:06:20 No. Oct 13 21:06:53 buz: the bunnie posts say otherwise... just read them, they are very long but also (Score:5, Interesting) (in slashdot notation) ;-) Oct 13 21:06:58 And it's not going to be that long (IMO) till even the cheapest of the cheap stuff - 'mini CD radio' for example goes to SMD. Oct 13 21:06:59 url? Oct 13 21:07:19 The one I bought a year ago is still completely through hole. Oct 13 21:07:24 mhh cd player aren't using smd? Oct 13 21:07:34 i could have sworn the ones i recently trashed were Oct 13 21:07:40 buz: it's a mini CD player 'tower' system. Oct 13 21:07:54 mine was more like ultra cheap ghetto blaster Oct 13 21:07:58 That was sold for maybe $40us with a free phone. Oct 13 21:08:07 whereas ultra cheap <30EUR Oct 13 21:08:21 Or ratehr was a free gift with a phone. Oct 13 21:08:39 8cm*30*30cm speakers, and unit. Oct 13 21:08:54 oh mine only had like 2x5cm speakers Oct 13 21:09:04 and sounded worse than many a laptop Oct 13 21:09:13 buz: http://bunniestudios.com/blog/ Oct 13 21:11:06 buz: about the same. It sounds OK surprisingly enough. Oct 13 21:11:23 mine decidedly sounded like utter crap Oct 13 21:11:28 (as i say, like most laptops ;) Oct 13 21:12:05 Err - with proper speakers. Oct 13 21:12:14 I'm mostly just using it as a computer amp. Oct 13 21:13:19 ah well, building solid state amps isnt THAT hard Oct 13 21:13:35 i run low end studio headphones from my laptop Oct 13 21:13:52 my gf hates them Oct 13 21:14:00 (sound insulation is very good ;) Oct 13 21:14:01 i love my sennheisers Oct 13 21:14:13 i want to have their babies. Oct 13 21:14:30 mine are sony, believe it or not, they sounded better than more than double priced sennheisers Oct 13 21:14:38 what you using? cjb_ie ? Oct 13 21:14:47 (and that comes from someone who usually has nothing but scorn for sony) Oct 13 21:14:49 mookid: HD25, and HD25SP Oct 13 21:15:01 ah well THOSE were out of my price range ;) Oct 13 21:15:01 HD595's here :) Oct 13 21:15:16 the sony mdr v6 can easily take it up with and HD5XX tho Oct 13 21:15:20 any Oct 13 21:15:37 lovely rich sound with the 595's - perfect for aphex twin and the like Oct 13 21:15:38 and come in at a measly 80$ Oct 13 21:15:59 well the sony's are pretty flat, but that's not too much of aproblem if you use equalizer Oct 13 21:16:54 nothing beats Neo stereo speakers! Oct 13 21:18:27 right now i'm using a pair of horrid no-name in-ear things Oct 13 21:19:18 because i have no 3.5mm - 1/4" adaptor here to plug the HD25SP's into the laptop :( Oct 13 21:20:06 ok, at least I have a bootin, accessible image again. Oct 13 21:21:48 just wondering why it claims to be Connected to Network. No SIM inside... Oct 13 21:22:14 GSM spec allows for emergency calls without sim Oct 13 21:22:20 would that be related? Oct 13 21:22:28 cjb: whihc is not possible in all countries of course. Oct 13 21:22:41 yup, it is of course operator dependent. Oct 13 21:22:45 cjb_ie: possibly Oct 13 21:22:49 (UK does not allow SIMless emergency calls) Oct 13 21:23:42 SpeedEvil: huh. i thought they used to Oct 13 21:24:41 cjb_ie: just that until now, openmoko seemed to fail on about anything I tried with gsm related things. Oct 13 21:24:53 cjb_ie: improvement is good :) Oct 13 21:25:56 Kero: even worse: It claims to be connected even when you enter a wrong PIN :-) Oct 13 21:28:25 so the applet is still a delusion, hm? Oct 13 21:29:35 no, does work Oct 13 21:29:55 booting with SIM now, let's see :) Oct 13 21:31:33 so, whats the magic firmware update for the GSM modem? Oct 13 21:32:48 gsm firmware update can be probably handy for bv3 users Oct 13 21:33:22 i dont suppose anyone could use this tool and hack some gsm firmware? Oct 13 21:35:14 the firmwares probably are signed, quite possibly encrypted Oct 13 21:35:57 and probably update will not be published anyway Oct 13 21:37:18 Hm, wasn't there a email on the device-owners about this in the last few days? Oct 13 21:37:51 Saying they already have the license to distribute the firmware (but still not the firmware updater)? Oct 13 21:38:06 yes Oct 13 21:38:09 yes, there was. community-mailinglist even Oct 13 21:38:20 * cesarb should track community more often Oct 13 21:38:24 * cesarb goes to the web archives Oct 13 21:39:02 "Update: GPS driver for GTA01 and GSM firmware for GTA01" from Michael Shiloh Oct 13 21:39:09 cjb: in the sporadic tests I've done over the past decade or so with GSM phones (about 5) a SIMless emergency call has never connected. (scotland.) Oct 13 21:39:20 GSM !=GPS Oct 13 21:39:24 cesarb: http://lists.openmoko.org/pipermail/community/2007-October/011031.html Oct 13 21:40:03 SpeedEvil: True. So? Oct 13 21:41:11 hrw|tv: thx Oct 13 21:41:55 fwiw I used this emergency feature once successfully Oct 13 21:42:01 Sorry - I'm perhaps not following the discussion properly - I thought the original query was about GSM firmware updating. Oct 13 21:42:49 SpeedEvil: yes it was. And the answer happened to be in the same mail as an update about the GPS situation. "Update: _GPS_ driver for GTA01 and _GSM_ firmware for GTA01" Remember: GPS != GSM :-) Oct 13 21:42:50 SpeedEvil: the email message hrw|tv posted discusses both Oct 13 21:42:52 Henryk, read the subject again, it references both Oct 13 21:43:01 I mean SpeedEvil Oct 13 21:43:10 anyway, I think it's clear enough now ;) Oct 13 21:43:12 s/posted/linked to/ Oct 13 21:43:12 cesarb meant: SpeedEvil: the email message hrw|tv linked to discusses both Oct 13 21:43:25 Ah. Oct 13 21:43:57 For annoying reasons I'm having to do wierd stuff to get URLs - for some reason copy/paste has broken Oct 13 21:44:24 I'm not bothering chasing it down as I'm about to format the drive and completely reinstall. Oct 13 21:44:43 * cesarb wonders if the firmware is in fact an EDGE one with the EDGE bits disabled, or rather misdisabled causing it to not work with the 3G SIMs (because it would use them in EDGE mode but the mode is disabled elsewhere...) Oct 13 21:45:12 SpeedEvil, xchat on windows? Oct 13 21:45:26 SpeedEvil: ouch. Sounds like something I'd see happening on windows :| Oct 13 21:45:46 nah, bitchx on linux Oct 13 21:46:03 * cesarb wonders what would cause copy-paste to break Oct 13 21:46:09 cesarb nope. ti calypso cannot do edge. Oct 13 21:46:12 * cb22_ wonders why we cant just have edge :( Oct 13 21:46:38 there is no edge-firmware thus. it was 'just' a regression in the upstream ti binarys Oct 13 21:46:52 roh: the enfora IIE has a rather similar set of AT commands (even including the %EM commands), and it's also EDGE... it's possible that the firmware is the same, only with parts ifdefed out Oct 13 21:46:58 Is there any software for openmoko, that changes the IMEI adress ? Oct 13 21:47:08 cesarb sorry.. thats bullshit. Oct 13 21:47:22 roh: that's only wild guessing on my part ;-) Oct 13 21:47:41 cesarb ive read the datasheets. it is 2g only. Oct 13 21:47:42 (wild guessing can sometimes be fun!) Oct 13 21:47:42 maybe we could bribe some ti employees to 'leak' the code :) Oct 13 21:47:53 cesarb, indeed, wild guessing is very fun Oct 13 21:48:29 Bluezz: there isn't, since it's on the GSM chipset side, and it's very unlikely that one will be developed, as the neo is still too fringe to be interesting to the kind of people who write these Oct 13 21:48:51 The more wild and improbable the theory, the funnier it can be Oct 13 21:49:06 Bluezz: that would in fact make it illegal to sell in the UK. Oct 13 21:49:14 As long as it has at least a slight possibility (no matter how improbable) of being true Oct 13 21:49:14 Bluezz: once the first person abused it. Oct 13 21:49:22 no applets (what's new), but could enter PIN and actually have a connection. Oct 13 21:49:35 cb22_: no - for some reason pastes out of xterm are not working. Oct 13 21:49:41 SpeedEvil: this means all the common phones where IMEI-changing methods have already been found cannot be sold there??? Oct 13 21:50:02 cesarb: no - I mean if there was an approved method by the maker. Oct 13 21:50:34 cesarb: If you can say 'the evil hackers did it' you're OK. Oct 13 21:50:38 SpeedEvil: ah. but changing the IMEI isn't that useful, unlike ethernet where some things need special MACs Oct 13 21:50:57 cesarb: apart from fraud. Oct 13 21:51:34 well.. besides fraud i do not see much useful application for an imei changer Oct 13 21:52:13 roh: some very esoteric bug-chasing scenarios (for instance, a carrier wants to find out why a specific set of IMEIs keep getting dropped somewhere) might also need it Oct 13 21:52:37 * cesarb is wild guessing again :D Oct 13 21:52:44 cesarb these people do not need to mod mobile handsets so often. they have proper equipment Oct 13 21:53:12 the only reason people change imei in masses out there is because its stolen devices and the imei is in the blacklist. Oct 13 21:53:26 some countries in the eu actually have one Oct 13 21:53:37 ^^ and in south africa too Oct 13 21:53:55 but here the networks are evil, they can block your phone if you piss them off Oct 13 21:54:09 second question, Can I emulate one sim card with openmoko, so I can switch btw them with out changing the simcard it self? Oct 13 21:54:41 hmm, in principle yes, in practice no Oct 13 21:54:42 Bluezz no. the simcard reader is directly connected to the gsm modem. Oct 13 21:54:56 Henryk, in principle? Oct 13 21:55:08 so no access there from the app-cpu at all without the gsm modem Oct 13 21:55:18 aka no sim simulation Oct 13 21:55:29 you should be able to reconnect the simcard reader to some I/O pins on the neo. then all you need to do would be to extract the keys from two cards. oh, and figure out the algorithm and implement it. Oct 13 21:55:31 ;-) Oct 13 21:55:38 how do I answer an incoming call? quite a few buttons, but none of them seems to work... Oct 13 21:56:45 roh: I wonder if the firmware update would also make it work with GSM850 Oct 13 21:56:58 it better Oct 13 21:57:08 cesarb thats another question i cannot yet answer Oct 13 21:57:30 hmm, actually, constructing a second sim holder and a software controlled switch shouldn't be too hard of a self-made project Oct 13 21:57:42 roh: I expected you couldn't answer it. But that doesn't stop me wondering ;-) Oct 13 21:58:38 roh: What is the meaning of life? Oct 13 21:58:39 Henryk that would not give you anything existing multi-sim adapter can already do. Oct 13 21:58:45 Henryk: you would need a way for the calypso chipset to re-initialize the SIM - i don't think its there. Oct 13 21:59:04 Writchie: switch gsm off, switch connected sims, switch gsm on Oct 13 21:59:32 have you been able to switch gsm off and on reliably? Oct 13 21:59:35 SpeedEvil that i know... sex drugs and rock'n'roll^welektronic music^w^w^w^w^w 42!1!!!1! Oct 13 22:00:00 Writchie: everything's bound to be better in the future. or at least supposed to be. Oct 13 22:00:04 does not all Bv3 had same IMEI? Oct 13 22:00:43 in principle on and off with work. In future, there's a better way especially with Bluetooth remote sim. just need to wait a couple of years Oct 13 22:01:15 well.. Bluetooth remote sim is something which desperately needs some bored hackers to destroy that braindead scheme Oct 13 22:01:27 well yes, the bluetooth way would be even better. i really like the idea of enslaving an existing phone Oct 13 22:01:41 kinda nice idea from the usability idea.. but totally flawed from the security standpoint Oct 13 22:02:06 roh: how's that Oct 13 22:02:37 well.. your sim card needs to transfer data which should not be on rf via bt. that easy Oct 13 22:03:16 roh: the critical part stays in the chip - BT is just a transport. Oct 13 22:03:20 and the authentication beeing, well.. optimized for usability and not inherent security Oct 13 22:03:40 roh: in the wiki there's the alternative idea of using existing cheap bt capable phones and simply enslave them: connect to the phone via serial and do remote control with AT commands, transfer voice via bluetooth Oct 13 22:04:30 Henryk well.. that does mean they don't do bt sim access at all. only the pricy nokia devices did that at all afaik. Oct 13 22:04:50 and that was mainly for the even more expensive carkit they sell Oct 13 22:05:03 Henryk: the d-bus layer i'm working on will do that - but its a little tricky to make the BT headset part work properly with most phones. Oct 13 22:05:04 roh: right. that's one of the reasons for this el cheapo scheme: basically nobody implements sim access Oct 13 22:06:47 in reality its much cheaper just to use the rf thats near the sim (aka same device) and use bt for connecting the car as a handsfree-kit. Oct 13 22:07:59 i think the remote sim motivation is also driven by other sim apps and authentication Oct 13 22:08:04 voila, usecase served, problem solved. much more standardized, less security hassle (no sim card traffic on the air) and the devices for that are already on the market and getting cheaper every day. evne 200E car stereos can do that nowadays Oct 13 22:08:29 Writchie there are no other sim apps. we don't do simtoolkit Oct 13 22:08:49 roh: i'm speaking generally Oct 13 22:09:18 roh: you don't do simtoolkit by philosophical choice Oct 13 22:27:52 mm, hanging up from the dialer (before call is answered?) seems to destabilize gsmd. Oct 13 22:28:00 restarting gsmd etc, works, tho. Oct 13 22:28:42 (I removed ttySAC0 from bootargs; there's one more patch for bug 788 today, then that might not be necessary anymore) Oct 13 22:32:03 * Kero is still baffled by the bitbake (or mokomakefile) peculiarities. Oct 13 22:32:27 but this seems at least a workable starting point. finally. Oct 13 22:59:59 anyone know a place with good, clear photos of the neo? Oct 13 23:00:14 flickr Oct 13 23:00:21 search on neo1973 Oct 13 23:00:29 ty Oct 13 23:00:53 cb22_, http://damaestro.us/photography Oct 13 23:02:13 i think i still have the originals laying around somewhere, if you need a higher resolution Oct 13 23:04:21 daMaestro, nice Oct 13 23:04:26 that screen looks beautiful Oct 13 23:04:41 yes, it is really nice Oct 13 23:05:31 The screen is insane. Oct 13 23:05:56 I was playing with it a while back, and realised that I had an 80*25 term open - with other stuff onscreen too. Oct 13 23:06:00 And could read it. Oct 13 23:06:26 this phone is like crack for me :( Oct 13 23:07:11 grr i think my ip-kvm is b0rked, /me moves to server room Oct 13 23:16:49 * cb22_ wonders what openmoko would look like in a more glossy style Oct 13 23:23:54 Hio Oct 13 23:35:13 cb22_: I, too, wonder what theme professionals can produce Oct 13 23:35:57 The 16-bit framebuffer is nice but really makes the gradients look inferior to photos and solids Oct 13 23:36:14 yeah Oct 13 23:36:51 * cb22_ thinks a gloss theme would look really cool Oct 13 23:36:55 I used gpe-gallery to browse pictures on the Neo and damn do they look good Oct 13 23:37:10 maybe with the crystal icons Oct 13 23:37:25 Crystal? Oct 13 23:37:30 everaldo Oct 13 23:37:41 Doesn't ring a bell ;x Oct 13 23:37:57 the everaldo icons? Oct 13 23:38:01 crystalsvg Oct 13 23:38:21 http://www.everaldo.com/ Oct 13 23:38:26 I'm an Xfce guy, I don't do themes ;) Oct 13 23:38:28 hehe Oct 13 23:38:30 Thankies Oct 13 23:40:50 Indeed... Oct 13 23:41:18 or even the oxygen icons Oct 13 23:41:22 http://www.oxygen-icons.org Oct 13 23:42:17 "To expand a little on what Harald wrote, thanks to TI we are now able to Oct 13 23:42:17 distribute the upgraded GSM firmware, but we're still working on the Oct 13 23:42:17 exact distribution terms for the firmware update program." Oct 13 23:42:19 yey! Oct 13 23:46:58 :) Oct 13 23:48:01 man, I can't wait for the GTA02 to be released, only I'll be in limbo land re: work visa in canada, so I can't have it delivered to some place without risking being 7000 miles away from that place when it arrives :( Oct 13 23:48:23 * cb22_ will order the GTA02 as soon as i see it Oct 13 23:48:44 no hesitation Oct 13 23:49:21 Yeah, I'll order one, too. And I haven't even gotten my GTA01, yet... it's in transit. Oct 13 23:54:03 Is the gsm chipset closed source or? Oct 13 23:55:18 bluezz: yes gsm is ran by binary system Oct 13 23:56:19 dcordes, all the gsm chipsets that are out in the market ? Oct 13 23:56:23 gsm devices have to be certified by official testing in order to become allowed Oct 13 23:56:49 so they are not allowed to be changed afterwards Oct 13 23:57:22 that's because gsm is an extremely cooperative system, any foul-playing device would be a major nuisance Oct 13 23:58:54 define nuisance pleas? Oct 14 00:01:41 bluelightning: people unable to make emergency calls Oct 14 00:02:09 bluezz: one user using one radio frequency to transfer data will wipe out 8 users. Oct 14 00:02:33 SpeedEvil, although it would be fun :) Oct 14 00:02:36 bluezz: GSM works well only because it is a scheduled service. Oct 14 00:02:54 bluezz: Other phones do not transmit when you do in your area. Oct 14 00:03:21 bluezz: this means that the cell-site can recieve your signal from a much longer distance than it could if it was uncontrolled. Oct 14 00:03:39 bluezz: degrading service for lots of people in the same cell, not following timing requirements, maybe even crashing the base station, getting unfair advantages (there are bits in the protocol for "I'm a emergency worker and I need to place calls even when others can't, so please kick other users out, if necessary" with no security in place other than that a normal handset wouldn't set those bits) Oct 14 00:04:56 Henryk: great, now some l33t h4x0r will read this channel's log and try to make his nokia set these bits Oct 14 00:05:27 Some of these problems can be fixed with a new protocol - with security. But the fundamental problem that radio is not point-point remains. Oct 14 00:06:16 SpeedEvil: can the new MIMO stuff make it look at least somewhat like point-to-point? Oct 14 00:06:24 cesarb: should actually be quite easy as they are stored on the sim card. i think there's no secure messaging in place, so you'll probably only need to perform a man-in-the-middle attack on your phone and sim card Oct 14 00:06:27 cesarb: to a very limited extent. Oct 14 00:06:50 cesarb: consider that a 1.8Ghz radio signal has a wavelength of 15cm or so. Oct 14 00:07:14 cesarb: this means that you need a 15cm antenna to get directivity over 60 degrees - at best Oct 14 00:07:26 And a 150cm antenna to get 6 degrees. Oct 14 00:08:09 SpeedEvil: well, cell tower antennas are big... Oct 14 00:08:27 cesarb: true - but mobile phone ones aren't. Cell towers are indeed doing htis. Oct 14 00:08:33 There is alot of software for crypting phonecalls, but is there software to secure ones simcard? Oct 14 00:08:37 (or at least look big, I don't know what's inside these white rectangles) Oct 14 00:08:51 and even in paranoid cases the IMEI+Simcard and any other information that is important Oct 14 00:09:21 bluezz: there's no need. the simcard is already protected by a pin Oct 14 00:09:36 i've always wondered, what is 'protected' Oct 14 00:09:43 (and yes, there's software to encrypt SMS, probably also on sim storage) Oct 14 00:10:24 is the sim protected as in encrypted with the pin, or does the phone impliment it Oct 14 00:10:33 Henryk, I doubt that is any strong encryption protection? Oct 14 00:10:48 bluezz: no, doesn't need to be. it's a smart card Oct 14 00:10:56 cb22_: the sim implements it Oct 14 00:11:57 but just thinking about the length of an average pin - if you could read the sim card, you could just try 99999999 pins Oct 14 00:12:12 a smartcard is in principle a small computer sans keyboard and display. you send it a pin and it either responds "ok" or "not ok, 2 tries left". But it actually is the decision of the card itself, not the phone. and it's quite hard to manipulate the card. you can't just take out the hard disk and put it into a different computer Oct 14 00:12:55 cb22_: the idea is that you can't read the. you can only communicate with it and it will decide whether it wants to talk to you and what information it gives to you on its own Oct 14 00:12:56 really? i always thought of pins as nothing more then storage devices Oct 14 00:13:14 +sim Oct 14 00:13:20 yeah Oct 14 00:13:58 cb22_: hard disks are also storage devices, but they have their own firmware Oct 14 00:14:07 cb22_: (and can also be protected by a password, btw) Oct 14 00:14:18 suppose so Oct 14 00:14:50 cesarb: you, but even then its quite easy, because rather macroscopic. in the worst case you'll take the platters out and read them with a laser Oct 14 00:15:34 Henryk: reading the platters on a modern hard disk isn't trivial. it would be easier to hack into the controller board and make it ignore the password. Oct 14 00:16:16 in order to manipulate a smart card you'll need something on the order of a scanning electron microscope. plus most modern cards are built to self-destruct should they be tampered with, that makes things even more complicated Oct 14 00:16:20 if the disk is properly made, the data will be encrypted based on the password. Oct 14 00:16:29 cesarb: yeah, or that Oct 14 00:16:37 Jiten_: unlikely, as it would take too long Oct 14 00:16:56 Jiten_: it would have to rewrite hundreds of gigabytes when changing the password Oct 14 00:16:58 Jiten_: you (or at least the regular user) will want to be able to change their password at will Oct 14 00:17:33 right... didn't consider that :) Oct 14 00:17:41 No - it can be gotten round. Oct 14 00:17:47 You have a user-visible password. Oct 14 00:17:52 And a real one Oct 14 00:17:59 you store the difference between them Oct 14 00:18:03 Henryk: I don't think a smart card can self-destruct, since it doesn't have a power source Oct 14 00:18:21 if the user wants to change the password, you simply store the new difference. Oct 14 00:18:30 SpeedEvil: sure, that's not a problem once you have the OS up. but you won't want to implement that in hard disk firmware Oct 14 00:18:42 Henryk: the one true tamper-proof thing is a IBM PCI coprocessor card, which has its own battery, and self-destructs by removing the power Oct 14 00:18:46 Henryk: why not? Oct 14 00:18:56 data encrypted with an unknown, random key, password used to unlock the key itself Oct 14 00:19:15 well, the hard disk just needs some specialized chips to do the encryption :) Oct 14 00:19:20 madwoota: that too. Oct 14 00:19:42 (of course you don't store the difference with the ascii password, but with a hash of it) Oct 14 00:19:53 of course :) Oct 14 00:20:12 cesarb: ok, self-destruct was an over-simplification. at least it will stop functioning. and there are some designs relying on chemical reaction with the atmosphere and stuff Oct 14 00:20:37 Henryk: nothing prevents you from using nitrogen. or argon. or a vacuum. Oct 14 00:20:45 cesarb: thermite Oct 14 00:21:12 Henryk: it's not possible to reliably self-destruct without a power source, and ideally you'd want to self-destruct on loss of said power source Oct 14 00:21:21 cesarb: right. there's always a counter-countermeasure. but that's only getting more costly with each measure Oct 14 00:21:50 cesarb: i know. but for the typical smartcard usage that's quite enough Oct 14 00:22:00 * cb22_ sighs Oct 14 00:22:12 and all i asked about was coping a sim card :) Oct 14 00:22:17 cb22_: heh Oct 14 00:22:26 cesarb, pleas give me an information page about IBM PCI coprocessor card? Oct 14 00:23:15 bluezz: http://www.cl.cam.ac.uk/~rnc1/descrack/ Oct 14 00:23:58 I allways thougt that a sim card could be dumped down with dd Oct 14 00:24:24 (and yes, i've read the part in the anderson book where it relates the rumor that the united states president's atomic football is so think for the reason that the circuitry to destruct the nuclear codes on a tamper attempt will have enough time to zero the memory even when you attack it with a plasma charge that moves at serveral kilometres per second) Oct 14 00:24:31 bluezz: in particular http://www.cl.cam.ac.uk/~rnc1/descrack/ibm4758.html Oct 14 00:25:07 bluezz: dd only works for block devices, and a SIM would be more like a filesystem than a block device ;-) Oct 14 00:25:13 Henryk: doing it right is hard if you want to consider all attacks. Oct 14 00:25:32 bluezz: no, that's why it's a smart card and not a simple storage card. there have been however (and probably still are) some flawed designs which will 'leak' the key through one trick or another Oct 14 00:25:45 the ki key Oct 14 00:26:39 Henryk, when you refer to the key, you mean the memory of the card? Oct 14 00:27:01 bluezz: the idea behind smart cards is: you give it the PIN and then it'll decide whether it wants to work for you or not. but even with the right PIN you can't read the key. instead you'll give it the data you want to encrypt, decrypt, sign or verify and it will return the encrypted or decrypted data or signature or verification result Oct 14 00:27:24 cb22_: AFAIK there are ways to get at that key as long as you already passed the PIN barrier (quite easy for the SIM's owner who already knows the PIN) Oct 14 00:27:32 yeah Oct 14 00:27:34 read about that Oct 14 00:28:28 bluezz: smart cards usually have something like a file system with directorys (called DFs) and files (called EFs) that can have various types. they can store arbitrary data (like the phonebook or messages) (actually these are stored in record oriented files, like small databases) or keys that can be used by the card but not read back Oct 14 00:30:54 you smartasses overseas... Oct 14 00:31:00 Henryk, Is it passable to create a virual smartcard ? Oct 14 00:31:04 virtual even Oct 14 00:31:07 you crack the encryption between ATMs to get money Oct 14 00:31:17 here in south africa we just take guns and blow the thing apart :) Oct 14 00:31:24 bluezz: sure. or rather: depends on what you want to do Oct 14 00:32:39 Henryk, Create an virtual card on openmoko and have it stored encrypted Oct 14 00:33:17 bluezz: you don't want to do that because the storage on the phone is _way_ more unsafe then storing on the card itself Oct 14 00:34:26 cb22_: well, from what I've heard, the latest trick on the US is to use the default operator password on ATMs to get to the money Oct 14 00:34:30 but in principle you'll only need to connect the smartcard holder to I/O pins on the Neo (because it's normally connected directly to the modem), then somehow dump the key from an existing card (shouldn't be possible with modern cards). Oh, and implement the algorithm that's not publically available Oct 14 00:34:49 sounds like a walk in the park Oct 14 00:35:21 cb22_: a park in south africa at night while looking like someone who has lots of money and no guns. Oct 14 00:35:28 Henryk, because of the cellphones security issues or? Oct 14 00:35:30 hehe Oct 14 00:35:54 bluezz: yes. the card is already specially protected against attacks. the phone is not Oct 14 00:41:05 Henryk, is there anything you could recommend, with results in that nobody can read the phone number I used if the get hold of the phone? Oct 14 00:41:55 bluelightning: if the SIM is not unlocked, you can't get the SIMs phone number. Oct 14 00:42:08 bluezz: without the cooperation of the operator anyway. Oct 14 00:42:31 If you mean the phone number of the phone. Oct 14 00:42:34 SpeedEvil, well I said nobody :) Oct 14 00:42:40 bluezz: the sim has a PIN and can be set to request the PIN when powered on. I have not checked, but I think if it is set this way you won't be able to read the SIMs phone number Oct 14 00:42:45 SpeedEvil, phone number, and IMEI number Oct 14 00:43:24 bluezz: IMEI is pretty easy to get - if the phone is reflashed. Oct 14 00:43:33 (assuming you've pulled the sticker off) Oct 14 00:43:46 IMEI: no way. that's printed on the phone and the standard even requires you to print it if you enter a special key sequence (something like *06*# or so) Oct 14 00:44:13 Henryk: the standard does not however say that the phone has to allow anyone to press the key sequence. Oct 14 00:44:26 IOW, if it's locked. Oct 14 00:44:47 SpeedEvil: well ok. but in any case you can simply connect to the built-in modem directly and ask it Oct 14 00:44:52 bluezz: what are you worried about? Oct 14 00:45:05 Henryk: which involves opening hte case, so you may as well flash it. Oct 14 00:45:55 SpeedEvil, well there is alot of cryptophones, and programs for that, but none of them apply to secret the phones identity Oct 14 00:46:10 bluezz: why do you want to do this? Oct 14 00:46:10 or I could not find anything Oct 14 00:47:03 bluezz: IOW - who is the person who you do not want the identity revealed to. Oct 14 00:47:04 bluezz: because that's kind of pointless. if you want to blur the connection between your identity and your phone number, buy a prepaid card in a country that allows to do this anonymously Oct 14 00:47:07 Just an paranoid phone ide Oct 14 00:48:08 bluezz: the only pretty good solution out there is to find someone poor, hand them a large amount of money. Have them sign up for bidirectional satellite internet service. Take the dish away to your fortress of solitude, and use skype. Oct 14 00:48:20 And yes, you will have horrible delay. Oct 14 00:48:31 skype is evil Oct 14 00:48:41 skype or whatever. Oct 14 00:48:46 I agree. Oct 14 00:50:40 well, VoIP over tor does work, in principle Oct 14 00:50:58 Tor has problems against well-funded motivated opponents though. Oct 14 00:52:06 The bidir satellite solution is _really_ hard to find. Oct 14 00:52:18 Well the ide was to apply this to a simple mobile phone Oct 14 00:52:26 that you could carry around with you Oct 14 00:52:28 bluezz: you can't. Oct 14 00:52:35 not a satelite connection unit Oct 14 00:52:38 bluezz: the network must know who and where you are. Oct 14 00:53:34 * cb22_ sees black helicopters over the horizon Oct 14 00:53:35 SpeedEvil, Yeah I know the operator can allways look and find out how I am from there, but I am more intressted to hidde it in the phone, and that leaves the operator with X alternatives that have been at the same places. Oct 14 00:54:44 Define 'hidde it in the phone'. Oct 14 00:55:18 SpeedEvil: he want's the phone software to neither store nor print the own number Oct 14 00:55:26 SpeedEvil, any identification for the mobile phone, like IMEI and phone number. Oct 14 00:55:35 (or dialled or received numbers) Oct 14 00:55:50 bluezz: you can do that at the moment - barring people with debug boards. Oct 14 00:55:54 Henryk, Correct Oct 14 00:56:06 bluezz: with just a phone lock. Oct 14 00:56:18 And even then it's quite easy: Use an IMSI catcher and set the range down low Oct 14 00:56:18 and a sim-lock Oct 14 00:58:48 sim-lock? will that protect me if the operator has the simcard physicaly and want to look around inside it? Oct 14 00:58:57 No. Oct 14 00:59:19 nothing can protect you against the operator. just switch to sealed envelopes and snail mail Oct 14 01:00:00 And remember that all public mailboxes have concealed cameras, so wear a balaclava. Oct 14 01:06:38 But don't worry: It's not paranoia if THEY really are after you. Oct 14 01:08:37 It can be. Oct 14 01:09:05 You can be mentally ill - have paranoia. Just because you're right doesn't mean that you are not ill. Oct 14 01:15:47 Yes, isn't the saying "Just because you're paranoid doesn't mean they aren't after you!" Oct 14 01:16:30 yeah, I just wanted to sound a little more positive Oct 14 01:23:20 Anyone here play Diamond/pearl? **** ENDING LOGGING AT Sun Oct 14 02:59:56 2007