**** BEGIN LOGGING AT Sun Mar 01 02:59:57 2009 Mar 01 03:01:34 Hello. Sorry I didn't keep up to date. what's the most stable+usable OS running on openmoko now? Mar 01 03:01:52 Where get docs to download & install them? I am using NEOFreeRunner Mar 01 03:03:04 file:///home/brian/Desktop/http:%252f%252fwiki.openmoko.org%252fwiki%252fMain_Page.desktop Mar 01 03:03:28 something like that Mar 01 03:03:36 http://wiki.openmoko.org/wiki/Main_Page Mar 01 03:04:22 that top link works for me:) Mar 01 03:05:08 2008.12 was working for me Mar 01 03:05:28 slaxxin: top link? I didn't find the word "operating system" on the page, and the search "upgrade" and "upgrade os" both return no results. Mar 01 03:05:41 slaxxin: could be easier to search if I knew the name of the OS. Mar 01 03:06:02 slaxxin: sorry I am stupid new user but I am old user of Linux, should be capable to upgrade if I follow the docs Mar 01 03:06:14 shr debian gentoo 2008.12 2007.2 Mar 01 03:06:33 fyp Mar 01 03:06:43 neopwn Mar 01 03:06:50 2008.12 wotked for me Mar 01 03:07:07 keyword "2008.12" worked for me Mar 01 03:07:09 look under downloads on the wiki link i gave u Mar 01 03:07:57 im flashing the latest shr now Mar 01 03:08:04 slaxxin Mar 01 03:08:18 ok i be quiet Dave Mar 01 03:08:22 :) Mar 01 03:08:59 Hi, can someone tell me how to start the gps modem from terminal pm SHR? dir: /sys/devices/platform/s3c240-i2c/i2c-adapter/i2c-0/0-0073/neo1973-pm-gps.0 doesnt exsits Mar 01 03:10:17 i might say something disturbing an get banned for life Mar 01 03:10:37 ? Mar 01 03:10:41 something offtopic Mar 01 03:10:45 Dave: what's wrong by answering stupid questions from me? Mar 01 03:11:12 its all my fault Mar 01 03:11:34 at least he didnt yell at me Mar 01 03:12:04 i know im supposed to ask for donations first Mar 01 03:12:53 o by the way flash ,, dont upgrade Mar 01 03:13:24 which os is installed Mar 01 03:13:38 donations? Mar 01 03:13:42 I gave at the office. Mar 01 03:14:13 slaxxin: I don't know what has been installed on my om but it was there by default. A bad version I guess because batter fail too quickly. Mar 01 03:14:48 does it have a dark seashore for a desktop? Mar 01 03:14:56 I want to do " Alternative : running from microSD card " because I am not confident to rewrite flash NAND in my life yet. Mar 01 03:15:10 ooo yes u can Mar 01 03:15:20 please flash Mar 01 03:15:28 slaxxin: okay if you suggest so Mar 01 03:15:33 too simple Mar 01 03:15:41 dave does too Mar 01 03:16:07 ./dfu-util -a rootfs -R -D '/home/brian/Desktop/moko/shr 2009/openmoko-shr-lite-image-glibc-ipk--20090228-om-gta02.rootfs.jffs2.nosummary' Mar 01 03:16:49 slaxxin: sure I can use 2008.9 version of dfu-util? http://downloads.openmoko.org/distro/releases/Om2008.9/dfu-util Mar 01 03:16:54 well dave may say that 2007.2 is the way to go Mar 01 03:17:01 slaxxin: since there doesn't seems to be a new version of dfu-util in 08.12 Mar 01 03:17:20 that will work Mar 01 03:17:52 slaxxin: wait, no source package? I am on linux+ppc thus I must compile from source (no available binaries for me) Mar 01 03:18:19 slaxxin: okay, find solution myself on wiki Mar 01 03:18:22 i gotta go get my slax cd Mar 01 03:18:59 slaxxin: by Mar 01 03:19:31 no ideas how to start the gps modem from a terminal on SHR? Mar 01 03:19:51 so when u get a working image maybe u can make n install file for others?:) Mar 01 03:46:27 Hi, why: cat /dev/ttySAC1 | grep GPGGA > file generates a empty file ? Mar 01 04:08:40 What I does? Mar 01 04:09:43 i don't know Mar 01 04:13:53 heh, playa Mar 01 04:43:07 http://wiki.openmoko.org/wiki/User:Harry Mar 01 04:43:18 folks~ Mar 01 04:43:24 give me a hand Mar 01 04:52:59 Hi, how i can start the gps modum on a SHR via a terminal? Mar 01 04:53:07 any ideas? Mar 01 04:54:06 TsaiHarry: do you have a idea? Mar 01 05:10:21 I am not engineer... :( Mar 01 05:12:34 cat /dev/ttySAC1 should work Mar 01 05:12:40 if nthng else has it open. Mar 01 05:12:54 there should be a 'manually using the GPS' or similar page on the wiki Mar 01 06:53:57 what is responsible for switching audio to a wired headset when a call becomes active? Mar 01 06:54:24 because I'm getting audio to the headset when I plug in to listen to music, but not when placing a call -- still comes out the speaker Mar 01 06:56:39 basically wondering if in my phone application I need to change the alsa state or if something underneath should be doing that already Mar 01 07:10:35 Has anyone used a USB to audio jack to listen to music on their freerunner? Mar 01 07:13:53 Like this one: http://www.talkietech.com/momiusb3head.html Mar 01 07:24:40 I'm pretty sure that adapter does not include a DAC Mar 01 07:24:49 moto does strange things with USB pins... Mar 01 07:34:13 sarahasharp it only works on moto devices. they have an custom asic which can switch the usb-plug to 13 or so different modes, including debug serial, usb, audio mono with mic, stereo out for music etc Mar 01 07:34:41 it gets switches via the resistor value on the id-pin and how the rest of the wires is connected Mar 01 07:35:04 e.g. d+ and d- short and a specific value on id to gnd sets it to charger mode Mar 01 07:35:18 I see. Thanks. Found a 2.5mm headset in my wirebox to use. Mar 01 08:28:01 Anyone around who has problems with GSM and Qi? Sounds very weird but please let me know how to reproduce it, i want to find the reason. Mar 01 08:34:24 Latest Qi, latest gsm0710muxd, kernel with the latest sysfs patches. Right after boot gsm0710muxd communicates with the modem without problems. Mar 01 08:43:14 Started ogsmd from ogsmd-new-timeout, no problems noticed. Mar 01 08:46:08 Hm, probably those who has problems with GSM and Qi are using pre-moko11 gsm firmware version? Mar 01 08:56:38 I flashed moko10, everything still works. Mar 01 09:15:25 good morning Mar 01 10:34:35 i installed openmoocow and cow don't make noise. in terminal there is "unable to open audio no available audio device". i'm using shr Mar 01 10:34:49 what is wrong? Mar 01 10:58:53 how do you use dfu-util when there are more than one device detected? Mar 01 11:00:04 dns53: via neotool (-; Mar 01 11:00:06 yeah i should do that Mar 01 11:00:33 or just add parameter: --device 0x1457:0x5119 Mar 01 11:03:52 i wonder what the other device is Mar 01 11:06:22 in my case it was usb bluetooth dongle, but it could be anything else Mar 01 11:08:58 it's been a while since i had to flash anything anyway, i just use qi and a sd card but you do need to flash android to the phone Mar 01 11:13:25 does anyone know when the htc magic will be available? Mar 01 11:15:34 jOERG_rw: Hm, interesting. People on ML are complaining that they are barely heard during GSM conversations. Yet calypso has AGC enabled... Mar 01 11:19:44 PaulFert`: i had to set volume from 2 to 3 Mar 01 11:40:08 lindi-: Mic2 Capture Volume? Mar 01 11:40:15 PaulFert`: yeah Mar 01 11:40:31 lindi-: do you have cap fix applied btw? Mar 01 11:40:54 PaulFert`: not yet Mar 01 11:41:18 i got sick and haven't had time to visit the shop where they sell 100 uF capacitors Mar 01 11:42:11 lindi-: Hm, calypso's integrated AGC should make Mic2 Capture Volume setting not that important (that is, you get the best SNR with the higher setting, but other than that it should work with any). Mar 01 11:42:45 PaulFert`: is there a way to disable that AGC btw? Mar 01 11:45:38 lindi-: i think yes, but i do not remember the command right now. BTW, Joerg said that calypso deep sleep mode makes a 4mA difference, not much comparing to your overall consumption. Mar 01 11:45:38 PaulFert`: would be interesting to test houd disabling AGC affects buzz Mar 01 11:45:43 PaulFert`: since the other side keeps hearing buzz only when I am not talking and the AGC starts amplifying buzz Mar 01 11:46:05 PaulFert`: ok, any idea where the consumption comes from? Mar 01 11:47:01 lindi-: that means you still have too good gsm network conditions. If you go farther, you'd probably have the other side complaining all the time, no matter whether you speak or not :) Mar 01 11:47:29 PaulFert`: ok :) Mar 01 11:48:48 lindi-: 35mA is still huge... I forgot who and how measured the consumption last time (probably Werner). Are you sure in your measurements, do you use a current-meter for that? Mar 01 11:49:07 PaulFert`: I am only reading the "current_now" sysfs node Mar 01 11:49:16 PaulFert`: i am still not sure if that is a reliably way to measure it or not Mar 01 11:49:37 s/reliably/reliable/ Mar 01 11:49:37 lindi- meant: PaulFert`: i am still not sure if that is a reliable way to measure it or not Mar 01 11:49:46 lindi-: if you have a multimeter handy, it worth to try using that to compare. Mar 01 11:50:58 PaulFert`: I have but I can't figure out how to measure it when the battery is connected Mar 01 11:52:43 PaulFert`: got any tips on how to do that? Mar 01 11:52:46 lindi-: do you have a lab PSU? Attaching to the battery terminals on FR itself is easy (strip the wire, bend, twist). But using the battery itself as the current source is not that easy. If you still want to use the battery as the power source, i think duck tape to attach wires to terminals is ok. Mar 01 11:54:04 hmm Mar 01 11:55:17 lindi-: you can even ignore the middle terminal (you won't get columb counter readings, but you won't really need them). Mar 01 11:55:23 ok Mar 01 12:00:21 h to the ello Mar 01 12:02:08 lindi-: once i wanted to measure current consumption of an iPAQ and it required all 7 terminals of the battery to be attached. We shaped a piece of wood with a knife and used nails instead of terminals. Installed this wooden piece in battery compartment, somehow attached the wires to the battery itself and we could reliably measure in all modes. So this is definetely doable without much effort even in much tougher conditions (7 terminals instead of 2). Mar 01 12:02:51 Get some copper tape, and a strip of card Mar 01 12:03:17 cut the copper tape into strips as wide as the battery terminals Mar 01 12:03:42 stick these strips on the card, wrapping round the end. Mar 01 12:03:59 now, cut off the end with scissors, and solder to the other end Mar 01 12:04:51 Now, cover other end in insulation tape, poke card carefully into between battery and frame, and put battery in fully Mar 01 12:06:49 SpeedEvil: good advice, i think it'll work with FR. AFAIR with iPAQ it was impossible or not practical because of the casing. Mar 01 12:07:00 works with 01 Mar 01 12:07:02 so should Mar 01 12:07:13 the bats are mechanically identical? Mar 01 12:07:24 SpeedEvil: yes Mar 01 12:07:30 Why doesn't openmoko have 3g support? Mar 01 12:07:37 oipat: because we hate you Mar 01 12:07:51 oipat: also, because sourcing a suitable 3g chipset proved difficult. Mar 01 12:07:56 Oh, SpeedEvil, you're now meaner than me :) Mar 01 12:08:11 oipat: vendors only want to sell to makers that are going to make >100000 phones. Mar 01 12:08:12 SpeedEvil: ok. patent/legal issues? Mar 01 12:08:29 SpeedEvil: ok. thanks :) Mar 01 12:08:34 oipat: they're not really interested in supporting lower volumes as they don't make significant money, and cost just as much to admin Mar 01 12:08:51 oipat: but FR has a usb host, so attaching an external 3G modem is possible. Mar 01 12:09:13 oipat: if apple is buying 10 million chipsets, and you want to buy 3000, and it costs 50 hours of their engineering time to setup both deals, ... Mar 01 12:10:07 Makes sence. Doesn't really matter that much to me anyways. Was just curious :) Mar 01 12:10:08 SpeedEvil: large vendors are doing their own design, based on reference. Look for Raster's mail where he explained most of this stuff. Mar 01 12:10:11 oipat: also other issues like open-source compatibility raise their heads. You can't actually buy many of these chipsets from 'normal' vendors, you have to talk to the maker - who will only deal in large volumes. Add other issues - power and space needs, and it gets nasty Mar 01 12:10:31 paulproteus: I mean chipset as in the chips, not modules Mar 01 12:11:10 wiuth 3g Mar 01 12:11:13 u either buy modules Mar 01 12:11:15 PaulFert`: Anyway - the point holds - though makers do do some of their own designs, the parts they design with may well not be readily available. Mar 01 12:11:17 which are pricey Mar 01 12:11:27 or u buy a reference design and make it al lyourself off the design Mar 01 12:11:30 raster: and large Mar 01 12:11:34 this requires much higher skill than a modulke Mar 01 12:11:42 its cheaper ... EXCEPT the entry price tag Mar 01 12:11:53 the entry price to use a design is high Mar 01 12:12:03 they may chharge $1,000,000 just to get your foot in the door Mar 01 12:12:09 that is just to get the design Mar 01 12:12:15 before you can even make 1 unit Mar 01 12:12:20 THEN they will also charge per unit too Mar 01 12:12:21 the rights to the design Mar 01 12:12:23 less than a module Mar 01 12:12:27 so if a module is $100 Mar 01 12:12:38 a per unit royalty may be $20, $30 or $40 Mar 01 12:12:41 depending Mar 01 12:13:39 so you need your $1,000,000 before you make a single product Mar 01 12:13:39 thats a LOT of mulah Mar 01 12:13:39 now Mar 01 12:13:39 lets say you expect to sell 20,000 units Mar 01 12:13:39 thats a lot more than gta02's sold is my guess Mar 01 12:13:39 but lets be generous Mar 01 12:13:39 Which means that $1M/50 is the number of phones you need to breakeven on that decision Mar 01 12:13:40 lets assume $20/unit royalty Mar 01 12:14:06 thats STILL $70/phone Mar 01 12:14:17 minus the price of the module vs chips Mar 01 12:14:23 and thats how the big boys do it Mar 01 12:14:36 (nokia, motorola, samsung, lg, sony-ericsson etc.) Mar 01 12:14:40 but they dont make 20,000 Mar 01 12:14:43 Though of course, this hurts much less if you're nokia. Mar 01 12:14:46 they make 10,000,000 Mar 01 12:14:50 with the same reference design Mar 01 12:14:56 sone u make 10mil Mar 01 12:15:00 As you pay that $1M first, and make 30 models of phone with aht one. Mar 01 12:15:10 the cost of a 1mil "entry fee" averages out at $0.10 per unit Mar 01 12:15:19 And next year you make another 30, slightly more downmarket. Mar 01 12:15:21 so really you pay less than $21 per unit Mar 01 12:15:22 ... Mar 01 12:15:26 where om would pay $70 Mar 01 12:15:44 rememebr that in the hw biz to allow for brekage, warranty, margin, design costs and production costs Mar 01 12:15:48 take any hw base cost Mar 01 12:15:51 multiply by 3 Mar 01 12:15:54 and thats retail Mar 01 12:15:59 if u really are top notce in quality Mar 01 12:16:01 maybe x2 Mar 01 12:16:06 Added to the fact you'll be paying also _lots_ more for hardware - maybe 2.5* that of nokia, due to volume as well. Mar 01 12:16:10 so om have to add $210 Mar 01 12:16:18 nokia have to add $62 Mar 01 12:16:35 and nokia do this much more - so their quality and ability to produce working phones is higher Mar 01 12:16:43 so really they just need to add maybe $40 in retail Mar 01 12:16:46 om need to add $200 Mar 01 12:16:50 thats a pretty unfair game Mar 01 12:17:06 just to make the same thing (with 3g-chipset-wise stuff) Mar 01 12:17:11 Gotta find someway to cheat usually Mar 01 12:17:21 SpeedEvil: not even accoutning for the rest of the device Mar 01 12:17:27 i am even being generous here Mar 01 12:17:38 only covering licensing costs Mar 01 12:17:45 not actual production costs Mar 01 12:17:56 and the only way to "hceap" is a module Mar 01 12:17:58 I disagree with the above - in that at that price point, you might choose modules, not integration. Mar 01 12:17:58 cheat Mar 01 12:18:00 i mean Mar 01 12:18:07 s/price/volume/ Mar 01 12:18:07 SpeedEvil meant: I disagree with the above - in that at that volume point, you might choose modules, not integration. Mar 01 12:18:11 SpeedEvil: of course! Mar 01 12:18:31 i am just takeing the example of "if everyone says "but nokia ship 3g on cheeap phones! what dont om!" Mar 01 12:18:37 yeha Mar 01 12:19:12 you cant play that game on a level playing field Mar 01 12:19:12 it just doesnt apply due to low volumes Mar 01 12:19:12 so you are driven to the "buy a module" world anyway Mar 01 12:19:12 if u buy a module prices are hgovering in the $100/unit land anyway Mar 01 12:19:15 For some stuff, it's almost cheaper to employ people to buy cheap phones and strip them :) Mar 01 12:19:24 but unlike a "use a ref design 3g" u dont have the $1m upfront Mar 01 12:19:27 raster: at 5k? Mar 01 12:19:43 SpeedEvil: depends who u buy it from Mar 01 12:19:57 i have seen them for > 250eur in lots of 100 Mar 01 12:20:03 raster: random question. Do you happen to know of DECT module suppliers? Mar 01 12:20:18 i have heard rumors of some as low as $70 inf u buy a few k Mar 01 12:20:31 but thats is COSt price Mar 01 12:20:40 actual cost of integrating a module onto your board Mar 01 12:20:48 the wiring costs Mar 01 12:20:54 software costs? (unknown) Mar 01 12:21:14 as for dect - nup Mar 01 12:21:44 You might be able to somewhat finesse the retail margin - go down to 100% markup or lower - if you ship direct to end-users in a low-cost manner. Mar 01 12:21:46 raster: Hey, another random question :) How're you, having fun at home again? Mar 01 12:22:01 That utterly relies on 'word of mouth' though. Mar 01 12:22:25 SpeedEvil: yes. that helps. the less betwene u and retail price the customer pays - the less it has to be padded Mar 01 12:22:47 but low volume is the enemy to competing in the 3g world with the big boys Mar 01 12:22:50 pretty simple Mar 01 12:23:00 unless your customers will be happy to pay $900 per phone Mar 01 12:23:08 Open-source 3G SDR stack! Mar 01 12:23:11 :) Mar 01 12:23:12 if u accept the device will be outrageously expensive Mar 01 12:23:14 you can do it Mar 01 12:23:29 the other problem is also ensuring u ave an open hw interface to it Mar 01 12:23:32 Cover it in diamonds! Mar 01 12:23:39 that alonw already trims down the options Mar 01 12:23:48 SpeedEvil: to be hoest - that almost makes sense Mar 01 12:23:52 raster: I know Mar 01 12:23:56 if u are going to make it fuck-arese sepxnesive due to components Mar 01 12:24:02 raster: it's one market for some of my stuff I'm contempleting. Mar 01 12:24:04 may as well justify it by making it gold plated anyway Mar 01 12:24:08 etc. etc. Mar 01 12:24:13 raster: cases cost - in low volume. Mar 01 12:24:16 use the top of the line in materials Mar 01 12:24:20 justify its silly price tag Mar 01 12:24:26 raster: a waterproof/... case doesn't actually cost that much more. Mar 01 12:24:31 and yes Mar 01 12:24:34 cases are expensive Mar 01 12:24:36 in low volume Mar 01 12:24:45 at one point u are better off just getting it cnc milled Mar 01 12:24:49 compared to a plastic mold Mar 01 12:24:51 raster: or cheat Mar 01 12:25:05 PaulFert`: doing good here :) Mar 01 12:25:07 busy tho Mar 01 12:25:07 raster: find a mobile with a suitable widely available replacement case, and buy 1000 from ebay Mar 01 12:26:43 SpeedEvil: that is an option Mar 01 12:26:46 tho even then - not cheap Mar 01 12:26:50 and it really limits u Mar 01 12:26:56 ad u now also need the screen int he right place Mar 01 12:26:59 yes. Mar 01 12:27:04 all internals wired out to the exact same places Mar 01 12:27:05 etc. Mar 01 12:27:10 and the buttons - you still need to make the insides the same shape Mar 01 12:27:15 yup Mar 01 12:27:25 i'd go for the cnc mill myself Mar 01 12:27:32 at least for a few hundred units its not too bad Mar 01 12:27:42 maybe $20-$40 per case? even if complex Mar 01 12:27:53 mill some of it out of aluminium for strength Mar 01 12:27:56 some bgits out of plastic Mar 01 12:27:58 for rf Mar 01 12:28:01 combine Mar 01 12:28:25 this iof course presupposes u relented to the "well i know its going to be expensive" way of thinking Mar 01 12:28:30 u are paying a premium for all parts Mar 01 12:28:55 cpu, ram, board, lcd, 3g etc. Mar 01 12:28:55 why stry and skimp on the case? Mar 01 12:29:01 it is the case that possibly most of all visible differentiates you Mar 01 12:29:06 if ua re going to buy a case Mar 01 12:29:10 why not just fine an oem phone Mar 01 12:29:13 form china somewhere Mar 01 12:29:21 that just uses another os and "hack it open" Mar 01 12:29:26 then resell with linux opened uop etc. Mar 01 12:29:42 China may even be more 'open'. Mar 01 12:29:55 'will you give me docs if I buy 50K phones' Mar 01 12:30:05 "whatever u want - just buy 50k" Mar 01 12:30:31 (even tho their upsteram supplier forbids passing on such docs.. if they make a sale.. happy happy.. whatever sells" Mar 01 12:34:40 freesmartphone.org: 03seeseekey * r336 10/trunk/ (python-elementary/ umaf/): Mar 01 12:35:38 raster: btw, you know "asian" mentality much better than i do, can you explain why OM TPE stuff is so untransparent and sometimes making horrible mistakes? Is it culture-specific issue or not? Mar 01 12:39:18 TPE? Mar 01 12:39:35 oh Mar 01 12:39:54 PaulFert`: hmmm. where do i begin. Mar 01 12:40:14 PaulFert`: in this regard i am not sure i'd intertwine asian and this. Mar 01 12:41:17 raster: that's why i'm asking about it, trying to understand what's going on. Mar 01 12:41:53 well 1. a whole hardware department that has zero to do with community or open source Mar 01 12:41:53 they come from and are hw enigneers from hw companies Mar 01 12:42:00 they work within their team and mostly speak just chinese (mandarin - taiwanese dialect) Mar 01 12:42:07 some speak english to a level Mar 01 12:42:11 some better, but most work Mar 01 12:42:30 so 1. group of people who are simply not used to doing anything in public Mar 01 12:42:38 nor doing anything community or in the open Mar 01 12:43:06 most of them have grown up and worked in places where u are locked down with nd'as, security and the "dont you dare tell anyone what you are doing" kind of world around them Mar 01 12:43:27 its a total cultureshock and goes entirely against everything they have learnt to do otherwise Mar 01 12:43:32 teaching an old dog new tricks is hard Mar 01 12:43:56 now add to this- they would hjave to learn those new tricks in a new language most of them dont know well and are not confident in... Mar 01 12:45:14 right. # echo fic-gta01 31 >> /etc/opkg/arch.conf; opkg install gllin*ipk Mar 01 12:45:22 FSO 5.1 is that worthy of a ticke? Mar 01 12:45:56 raster: Ok, that explains lack of transparency in TPE HW EE department. But it doesn't explain the same about other departments and it doesn't explain horrible mistakes and stupidity. Mar 01 12:47:12 well u want my honest opinion? Mar 01 12:47:15 I wouldn't have asked otherwise Mar 01 12:47:17 PaulFert`: are there horrible hw mistakes? Mar 01 12:47:24 managament in om (specifically sean etc.) are NOT from the oss community Mar 01 12:49:43 they are not inside it - they say the things to appease - but they do not participate Mar 01 12:49:43 they really imho dont have a desire to either Mar 01 12:49:43 open code != working well with the open source community. Mar 01 12:49:43 it is a desire to leverage the community to make money - fair enough. by all means Mar 01 12:49:43 Nessecary, but not sufficient. Mar 01 12:49:43 but not really "drink the koolaide" Mar 01 12:49:43 :( Mar 01 12:49:43 being open about decisions is just not in their culture either Mar 01 12:49:43 its not in their ethos or being Mar 01 12:49:43 no fault here Mar 01 12:49:43 but thats just how it is Mar 01 12:49:43 their decisions are theirs and will be made with or withotu community knowing or being involved Mar 01 12:49:46 involving the community just will make the process lower, more painful and cumbersome Mar 01 12:51:48 now thats a small group Mar 01 12:51:50 then u have software Mar 01 12:51:54 lindi-: i'll tell you about some hw mistakes a bit later. I just don't understand how a professional can do that and why it takes them like a year to admit and change something that is obviously wrong. Mar 01 12:52:06 software has basically seen almsot everyone come and go over time Mar 01 12:52:08 i can barely talk about who is there now Mar 01 12:52:28 ooh Mar 01 12:52:34 admitting to a mistake == loss of face Mar 01 12:52:38 and loss of face is BAD Mar 01 12:52:50 thats a personality + culture thing Mar 01 12:53:03 i think =westerners are more apt to say "oops. my bad!" Mar 01 12:53:37 anyway - software direction at least whil i was there changed every 3 weeks Mar 01 12:53:38 totally Mar 01 12:53:45 some new shiny thing to chase Mar 01 12:53:56 not based on actual reality Mar 01 12:54:36 how can u be transparent when u barely know what the flavor of the month is? Mar 01 12:54:36 hmm, i wish u-boot would show 'current_now' Mar 01 12:55:07 not to mention i found my hands pretty tied as software technical decisions (that werre already made in existing upstream code) wer being dictated to be different by management Mar 01 12:55:10 specfically sean Mar 01 12:55:48 and that means that those in softwar ecan tbe transparent Mar 01 12:56:38 as they had only 1 choice - to do as instructed. i probably was one of the most transparent there - but i found myself pretty much caught between community and talking with you - knowing just what you want and not being able to do it. Mar 01 12:58:53 after over a decade of being part of the community - listeing to users and dealing with their problems - iv'e gotten good at filterign down just what they want even from a few words. but i also have had the power to DO somethnig about them Mar 01 12:59:31 or at least "put it on the todo list" Mar 01 12:59:31 or mull over it in terms of a bigger picture and come up with the solution to solve 20 things in one swoop. Mar 01 12:59:47 but i was neutered on that front Mar 01 12:59:56 i was being told to do things i have historically found to be bad practice Mar 01 13:00:03 and that lead to big big big problems Mar 01 13:00:19 and i just couldnt do it Mar 01 13:00:34 policy was not do do such things at least in e development - i created such policy Mar 01 13:01:09 then being told to revers that policy because a non-technical person decided so. without experience in the field. Mar 01 13:01:58 i do not know if things have changed - i'd hope so, but i would guess a lot of lack of transparency is that i know sean doesnt like people who disagree with him. Mar 01 13:02:18 so chances are over time the group has been chosen for people who agree and do as instructed without complaint. Mar 01 13:02:37 thus you are less likely to see openness from such a group Mar 01 13:02:39 thats a guess Mar 01 13:03:01 nothnig i said here is new - it's all pretty well defined by community mailing list archives Mar 01 13:03:02 :) Mar 01 13:05:09 of course others who are there might just not want to get stuck in the firing line between community etc. :) Mar 01 13:05:16 Yep, that explain pretty much everything. Except some real weirdness in HW. Some stuff just doesn't make sense, can't be justified by management requirements and stuff (like management doesn't say whether you need a pull-up or pull-down or change BOM without altering the scheme to make a mess or to lose track of serial numbers etc). Mar 01 13:05:25 raster: speaking of which, did you notice that all links to lists.openmoko.org of last summer now point to different articles? Mar 01 13:06:08 lindi-: nup. i dont really trawl through them. but there are other list archives around.\ Mar 01 13:06:23 raster: yep but wiki has hundreds of links to list archives Mar 01 13:06:31 and now they all point to wrong articles Mar 01 13:06:43 lindi-: likely a technical glitch Mar 01 13:06:48 not purposeful Mar 01 13:06:51 I reported this to admin-trac.openmoko.org/ a few months ago but got no response Mar 01 13:06:54 b ut it does mean info is lost Mar 01 13:07:01 now admin-trac.openmoko.org does not reply Mar 01 13:07:30 raster: info is somewhat lost since wiki often links only to the articles without telling their title Mar 01 13:07:35 PaulFert`: chances are theat they themselves dont know and if they do the question is lost somewhere along the way Mar 01 13:08:16 raster: ehm, HW EE stuff doesn't know how to design properly? Mar 01 13:08:33 has anyone here setup a netconsole over usb sucessfully on the Freerunner Mar 01 13:09:30 PaulFert`: there is a tonne of turnover in om Mar 01 13:09:35 in personell Mar 01 13:09:50 come back 6 mo later and chances are 50% of the people are not the same Mar 01 13:10:07 u cant often find info as the person who did it/knows is gone Mar 01 13:10:21 any company with high turnover is going to have a very bad time Mar 01 13:10:32 if u are a small startup - high turnover is going to kill u Mar 01 13:10:39 u have few proceedures and so on in place Mar 01 13:10:50 these proceedures and "paperwork" are what slow big companies down Mar 01 13:10:54 but they can handle turnover Mar 01 13:11:01 small co's cut corners Mar 01 13:11:04 startups do Mar 01 13:11:22 they are in a tight spot - so turnover is going to tickle all your corner-cutting badly Mar 01 13:13:34 om needs to stem turnover on staff. its just not sustainable. Mar 01 13:13:50 i hope they fix it Mar 01 13:16:30 raster: now i see... Thanks for the clarifications again! :) Mar 01 13:19:01 dont get me wrong - i hope om succeeds Mar 01 13:19:34 the "open devices are a workable thing and the community and users should be given the opportunity to do whatever they like with their products" flag needs to be flown Mar 01 13:19:39 and it needs to succeed Mar 01 13:19:56 and if om fails thats a huge blow to the whole idea of open devices Mar 01 13:20:00 truly open ones Mar 01 13:23:56 interesting, GPS (gta01) could not get a fix within my appartment, but coming bck from outide, it's still having a fix after a few minutes. Mar 01 13:24:15 second screen in the zhone gps app remains empty, what's supposed to be there? Mar 01 13:24:32 Kero: getting fix requires a lot more data xfer than keeping one Mar 01 13:24:41 and a large stream of very slow reliable data Mar 01 13:26:42 raster: which makes a lot of sense, but I do remember my gta01 loosing fix earlier. so it's a nice improvement (or just luck today) Mar 01 13:29:25 it seems to have lost fix now. Mar 01 13:34:56 just recoverd fix. nice Mar 01 13:47:08 apt, make everybody in the world some coffee Mar 01 13:47:09 * apt brews a nice hot cup of coffee for everybody in the world some Mar 01 13:47:26 apt, make everybody in the world coffee Mar 01 13:47:27 * apt brews a nice hot cup of coffee for everybody in the world Mar 01 13:47:31 muuch better Mar 01 13:48:18 if only my freerunner could make coffee, I'd have use for it Mar 01 14:09:25 freesmartphone.org: 03mickey 07cornucopia * redb406cc5fe0 10/libfsoframework/ (8 files in 2 dirs): create MasterKeyFile Mar 01 14:09:26 need your help Mar 01 14:10:30 http://wiki.openmoko.org/wiki/User:Harry Mar 01 14:11:10 your support is very important for me Mar 01 14:11:38 help me to fill the survey Mar 01 14:14:55 TsaiHarry: how's your research, how many responses have you already got and how many still needed? Mar 01 14:15:24 I only received 70........ Mar 01 14:16:36 I hope over 100 or more.... Mar 01 14:16:55 I'd say give it up - at least here. Mar 01 14:17:12 mailing list Mar 01 14:17:25 few of these people actually talk, I think they bought OM, logged into IRC, and then tried holding their breath until the software was ready. Mar 01 14:17:44 aaahahahaahahhaaahaa Mar 01 14:17:48 .................. Mar 01 14:18:10 so people is login here for waiting? Mar 01 14:18:20 they can eamil to me... Mar 01 14:18:24 I think most people have just left it logged Mar 01 14:18:27 mostly complaining actually Mar 01 14:18:28 :) Mar 01 14:18:36 and aren't actually reading it on a regular basis Mar 01 14:18:38 or idling Mar 01 14:18:44 hoping one day will say the magic words Mar 01 14:18:50 too bad.... Mar 01 14:18:51 "it's aliiiiiiiiiive!" Mar 01 14:19:34 this is different with my understand when I worked for OM... Mar 01 14:20:27 TsaiHarry: what were you working on at OM btw? Mar 01 14:20:27 om internally knows little of what actualyl goes on in the community Mar 01 14:20:27 om internally knows little of what actually goes on in the community Mar 01 14:20:41 and if they do - they figure it out months later after much flamewar action Mar 01 14:20:42 :) Mar 01 14:20:43 oh..I worked for OM since 2006, very early stage Mar 01 14:21:43 so I should say sorry to everybody..... Mar 01 14:22:04 nah - no need Mar 01 14:22:13 because I sold many neo1973 and freerunner to community......... Mar 01 14:22:39 i dont think the blame is at your feet :) Mar 01 14:22:40 raster: btw, do you think it'd be better if OM highlighted that GTA02 is not yet ready for everyday use for "end-users" in the time of launch? Mar 01 14:22:45 If you were just selling stuff, it's not your fault. Mar 01 14:23:07 my fault... Mar 01 14:23:16 PaulFert`: yes. i wouldnt have done it Mar 01 14:23:36 i wouls have said "if you get a prompt after boot... thats about as much as we guarantee" Mar 01 14:23:38 :) Mar 01 14:23:39 because..I am the guy done each sales task from company... Mar 01 14:23:52 raster: but was it intentional or did Sean believed himself that it'll be ready really soon? Mar 01 14:23:53 i go for underselling Mar 01 14:24:06 and make sure those that do buy get what they expected or more Mar 01 14:24:09 not the other way around Mar 01 14:24:31 PaulFert`: he believed Mar 01 14:24:43 Belief isn't enough. Mar 01 14:24:50 correct Mar 01 14:24:50 actually in all reality i think it was hoped Mar 01 14:25:00 but saying its a developer only device would harm sales Mar 01 14:25:02 Well - to be fair. Mar 01 14:25:18 raster: indeed it seems that people without skills & time seem to be buying the phones and they can only complain about things not working :/ Mar 01 14:25:26 thus say nothing and hope for the best that the "hopes" that updates will fix it Mar 01 14:25:26 OM - with 10 or 50 times the funding might have succeeded with a _lot_ more than it's had Mar 01 14:25:32 and indeed - updates would fix it... Mar 01 14:25:45 :) Mar 01 14:26:19 for a "product"...those problem should not happen... Mar 01 14:26:36 SpeedEvil: om with the same funding and a few different people would have succeeded with a _lot_ more.... :) Mar 01 14:26:54 raster: I agree Mar 01 14:27:00 om' initial software tries were pretty attrocious Mar 01 14:27:03 at gta01 time Mar 01 14:27:12 and it needed a total fresh redo Mar 01 14:27:21 (you know the first ones) Mar 01 14:27:21 raster: however - 2007.3 or so actually more-or-less worked Mar 01 14:27:26 (pre 2007.x) Mar 01 14:27:30 yeah Mar 01 14:27:34 it just needed fixing Mar 01 14:27:37 like labels not icons Mar 01 14:27:39 SpeedEvil: but it lacked the undocumented echo fix Mar 01 14:27:39 or better icons Mar 01 14:27:42 raster: it took a long time for the software to get back to that state after changing massive stuff Mar 01 14:27:50 a _hugely_ long time. Mar 01 14:27:53 SpeedEvil: and the undocumented deep sleep disable fix Mar 01 14:27:53 gsmd needed a lot of hard core stabilisiation Mar 01 14:28:15 and kernel was below par too Mar 01 14:28:19 but that swas gta01 Mar 01 14:28:24 gta02 incredibly complicated things Mar 01 14:28:39 glamo alone imho cost 9 months of engineering time Mar 01 14:28:54 simplky because suspend/resume, sdio to card, video, etc. etc. now all went through glamo Mar 01 14:28:59 and it really just was a mess Mar 01 14:29:07 withotu glamo it'd be much easier Mar 01 14:29:19 GTA01 - with some tweaks'd have been mostly adequate unfortunatley. Mar 01 14:29:19 i rememebr suggestions Mar 01 14:29:45 keep 2007.x replace one app at a time, keep gsmd - fix it up. and put in solid infra betwene all of thses for glue thats currently missing" Mar 01 14:29:55 there was a virtual keyboard protocol Mar 01 14:29:58 itwas a real hack Mar 01 14:30:10 that was improved in illumes implementation for example Mar 01 14:30:50 better centralised power management too Mar 01 14:30:50 that still is part of fso and only partly done in 2008.x Mar 01 14:30:54 but illume keyboard works with only one wm so I can't use it with icewm :( Mar 01 14:31:04 raster: do you still developing OM project? Mar 01 14:31:05 but really u had unstable kernel, unstable drivers, unstable userspace, and veryt little solid infra Mar 01 14:31:17 thus - why it didnt take 2 or 3 months from gta02 release to "stable os" Mar 01 14:31:45 TsaiHarry: ummm... not really - i just try stuffon my gta02 every now and again - but frankly the hardware is just too bad :( Mar 01 14:31:57 2g just is useless to me as a phone Mar 01 14:32:10 the indented screen is annoying and loses a lot of screen touch space Mar 01 14:32:24 and i have never had it reliable enough in its suspend/resume to bother more than toying Mar 01 14:33:01 and realistically - i have much bettre hw with me Mar 01 14:33:05 thats not from om Mar 01 14:33:16 its got stable kernels Mar 01 14:33:21 and much more horsepower Mar 01 14:33:26 higher resolution screens Mar 01 14:33:28 etc. Mar 01 14:33:35 and the design is nicer Mar 01 14:33:43 hehe... Mar 01 14:33:58 om as a distro is very far from upstream oe/angstrom Mar 01 14:34:16 and i'm in the game of supporting lots of devices with lots of hardware, screens etc. Mar 01 14:34:42 but as time goes on my gta02's just spend more tiem turned off Mar 01 14:35:43 what i do eventually finds its way back into om - but i'm not involved in that - i work for different people with different priorities :) Mar 01 14:36:52 raster: what's your email? Mar 01 14:39:10 TsaiHarry: raster@rasterman.com :) Mar 01 14:39:14 ok Mar 01 14:39:19 cool Mar 01 14:39:32 i'd love to see someoen produce some actually GOOD hardware Mar 01 14:39:37 like modern stuff Mar 01 14:39:42 omap3530 maybe Mar 01 14:39:46 thats open Mar 01 14:39:48 3g Mar 01 14:39:49 wifi Mar 01 14:39:56 all the things the nerds really want Mar 01 14:39:59 big screen Mar 01 14:40:02 keyboard Mar 01 14:40:04 oh~ where can buy one?? Mar 01 14:40:10 but still small enough to fit in a pocket Mar 01 14:40:13 i wish Mar 01 14:40:18 i've been looking into it Mar 01 14:40:27 i have an idea of what is needed - the components Mar 01 14:40:37 how to minimise software work (build on work elsewhere) Mar 01 14:40:38 raster: 3g stuff is probably not easy without having to pay a lot of patent rojalties? Mar 01 14:40:42 but the problem is basically the usual Mar 01 14:40:44 need cash Mar 01 14:40:54 lindi-: i know. :) trust me. i've looked Mar 01 14:41:05 i have an idea of what it might cost and what can be done Mar 01 14:41:35 raster: and if something breaks there we are probably not given source code even under NDA Mar 01 14:41:44 never Mar 01 14:41:47 thats will NEVER happen Mar 01 14:41:52 well not for the gsm firmware Mar 01 14:41:52 so fixes like moko11 firmware won't be possible Mar 01 14:42:04 that will just never happen Mar 01 14:42:13 never say never ;) Mar 01 14:42:21 in the end the best u can hope for is all driver componets in or above the kernel are open Mar 01 14:42:33 but like all hardware these days Mar 01 14:42:45 it runs some level of software (Firmware) Mar 01 14:43:00 as such - if theres a bug in that - consider it a silicon bug u can fix without solder Mar 01 14:43:02 but no more Mar 01 14:43:12 and it wont happen due to regulation Mar 01 14:43:18 and regulation is thanks to govt + telcos Mar 01 14:43:21 it wont change Mar 01 14:43:27 raster: but we got moko11 fix? Mar 01 14:43:39 not in the time any company can stay alive in to try make it 100% open Mar 01 14:43:55 lindi-: and if u get a 3g module with firmware that has bugs - fixes can happenh Mar 01 14:44:11 u need to make sure u get a 3g module from a supplier that will allow re-diistribution of a firmware fix Mar 01 14:44:16 than canbe flashed with an open os Mar 01 14:44:24 so choose your 3g chip wisely Mar 01 14:44:25 :) Mar 01 14:44:27 anyway Mar 01 14:44:32 sleepy bo bo's time Mar 01 14:44:37 land of nod for me Mar 01 14:44:37 Night. Mar 01 14:44:39 niteo! Mar 01 14:44:53 nox: no, you don't. Mar 01 14:44:54 oops Mar 01 14:45:40 ahoi Mar 01 14:49:03 * Dave throws a snowball at raster Mar 01 14:49:17 Not while he's asleep! Mar 01 14:52:20 :D Mar 01 14:54:01 Hm, I just installed shr unstable (with latest kernel) and alsa can't find any soundcard on my gta02v6, so I have no audio at all. What's up? Mar 01 14:59:50 JFYI: A reboot fixed my "no sound" problem Mar 01 15:11:22 freesmartphone.org: 03daniel 07zhone * r9c204f11bfc3 10/src/zhone: Mar 01 15:11:22 freesmartphone.org: Remove timer that automatically sends unsent messages Mar 01 15:11:22 freesmartphone.org: This was built into zhone at a time where sending SMS was flaky at best Mar 01 15:11:22 freesmartphone.org: and so resending messages that failed to send was a necessary evil. Mar 01 15:11:22 freesmartphone.org: Now this is not the case any more and this "feature" does more evil than Mar 01 15:11:24 freesmartphone.org: it helps. If anything we need a watchdog that keeps trying to send that Mar 01 15:11:26 freesmartphone.org: particular message until it has been sent successfully. Mar 01 15:12:09 hehe Mar 01 15:12:30 wow Mar 01 15:12:43 CVirus has evolved through the alphabet ;) Mar 01 15:13:02 and no comment by Dave on this subject yet :) Mar 01 15:15:14 LOL Mar 01 15:15:39 Davey, we're going to be the next internet meme .. trust me :-p Mar 01 15:44:20 freesmartphone.org: 03mickey 07cornucopia * ra2131fc4e6cc 10/ (12 files in 4 dirs): start with logging class Mar 01 15:51:54 give me a hand... fill the survey http://wiki.openmoko.org/wiki/User:Harry Mar 01 16:04:57 slaxxin Mar 01 16:10:27 Superdave Mar 01 16:11:23 sigi try an older kernel Mar 01 16:14:25 :D Mar 01 16:29:54 openmoko Mar 01 16:31:18 http://wiki.openmoko.org/wiki/User:Harry give me a hand fill the survey!!! thanks Mar 01 16:39:14 openmoko Mar 01 16:39:55 <|Marco|> .org ? Mar 01 16:40:38 openmoko Mar 01 16:41:37 <|Marco|> yeah ? Mar 01 16:41:46 openmoko Mar 01 16:42:43 what are you guys babling? :) Mar 01 16:50:47 open moko moko Mar 01 17:07:11 http://wiki.openmoko.org/wiki/User:Harry give me a hand..fill the survey :D Mar 01 17:31:47 shr works with thewr2 kernel an not the r3 kernel Mar 01 17:31:51 sound Mar 01 17:34:52 <|Marco|> has there been any more talk about power usage/battery life ? Mar 01 17:45:53 |Marco|: no. I think the talk is all done, it's been discussed every possible way -- now it is the time for someone to do something about it, but that activity seems pretty much stalled. Mar 01 17:46:35 The simple solution is to make it consistent. Turn off after 1 hour in all circumstances, till recharged. Mar 01 17:47:07 And to be honest, I would much rather that people invest time to make the device work properly; it's quite silly for the device to run on battery for 1 month if it can't reliably resume, or if you can't answer it when it rings, etc, etc, etc. Mar 01 17:47:45 Of course, Openmoko is working on the gta03, so any effort would have to be done by the community. Mar 01 17:48:49 mwester: I have to disagree Mar 01 17:49:03 I am working on the FR and will continue to do so Mar 01 17:49:11 to make it reliable as a phone Mar 01 17:49:14 and usable Mar 01 17:50:04 <|Marco|> mwester: I understand Mar 01 17:51:43 mirko-paroli: perhaps I misunderstand then; are you working on kernel and low-level userspace for power improvements (cesarb's cpu-freq stuff for example)? I though paroli was a dialer app. Mar 01 17:52:57 mwester: well I am not the only one working on it, you are right paroli is UI but the rest is also being worked on Mar 01 17:53:17 mwester: how can I make a UI for a reliable phone witout the rest being reliable Mar 01 17:55:25 ok. AFAICT, andy is doing gta03 and 2.6.29-rc-way-too-early-to-be-useful, werner has been very helpful recently, and seems to have been spending the most time of anyone on "old" products, but even his latest commits have dried up in the past week, and those we've seen are all gta03-related. Balaji -- who knows? He commited to get the gta01 working, but that's not yet done, I understand. Nelson seems to work on very specific subsystems, such a Mar 01 17:55:25 s the touchscreen, but I'm not even sure he's Om. Mar 01 17:56:23 So from a community-visibility-point-of-view, we see andy's "stable" kernel broken, commits all happening on the gta03, and a mad insane rush to the latest -rc kernel sources and all the disruption to the user-space apps and users that causes. Mar 01 17:57:41 If there's someone actively working to stabilize the existing kernel on the gta02, and make it work on the gta01, it would be nice to hear a report from them -- I think there are many people who would step up to assist (look at what happened when the community and Werner worked on the serial port problems -- that should be the model for all work!) Mar 01 17:59:26 i am completely on your side here Mar 01 17:59:57 there is a piece of hardware which imho is not too bad Mar 01 18:00:00 of course it could be better Mar 01 18:00:09 <|Marco|> true Mar 01 18:00:12 but it should be exploited first Mar 01 18:00:46 I am trying to help the exploitation along Mar 01 18:01:07 and to move some focus on what we all have in our hands right now Mar 01 18:01:38 so all I am saying is, that generalizing doesn't help and is not fair minded Mar 01 18:01:39 <|Marco|> mwester: as you stated: ".. or if you can't answer it when it rings .." is this in connection with dbus/wm being slow ? Mar 01 18:01:52 freesmartphone.org: 03mickey 07cornucopia * re5a95ebef3c2 10/libfsoframework/ (7 files in 2 dirs): formatted log output ok Mar 01 18:02:31 |Marco|: that's one of the many problems. Sometimes it won't even resume properly, other times the touch-screen is just non-responsive -- there are just a lot of different bugs and potential failures. Mar 01 18:02:38 mwester: i just booted stable kernel and it worked. Booted the latest andy-tracking and it works as well. FYI Mar 01 18:03:11 the slow dbus is what has kept me from using SHR, though -- qt extended is almost as good for phone (has few other features that I like though), and it responds much more quickly. Mar 01 18:03:28 mwester: as to resume, i agree, i've seen it once or twice. It didn't resume on any event, including PWR button and USB. But it's very rare. Mar 01 18:04:15 PaulFert`: I'm going by the mailing list -- where apparently stable won't build. Mar 01 18:04:31 mwester: won't build or won't run? Mar 01 18:04:35 PaulFert`: "very rare" is not acceptable -- "almost never" is. :D Mar 01 18:04:46 PaulFert`: the mailing list said won't build. Mar 01 18:04:53 <|Marco|> I have not seen the resume problem yet, but I have not been testing it activly. Mar 01 18:05:15 mwester: but i've just built it without _any_ tricks. Just fetch, checkout the branch and build. Mar 01 18:06:11 "bug: Closed [worksforme]" <---- the most infuriating status in a bug-tracker. :D Mar 01 18:06:21 mwester: Ah, it was stable-tracking that failed to build Mar 01 18:07:00 Don't know, I'm speed-reading the lists as I'm on another project these days and can't spend the many hours per day I used to :( Mar 01 18:07:40 I know, the guy tried to build stable-tracking by mistake. Mar 01 18:08:50 mwester: another project == daytime job or something free software related? Mar 01 19:18:30 does this channel also handle Qtopia? Mar 01 19:19:04 It's anything you can get to run on your openmoko device. Mar 01 19:19:10 AS well as the openmoko OS Mar 01 19:20:29 yes it is frustrating Mar 01 19:24:14 because I installed the Qtopia and want to try syncing address data Mar 01 19:24:29 so I installed qtopia-desktop Mar 01 19:24:39 What's the resolution of the touchscreen sensor part? Mar 01 19:24:56 rhkfin: in principle, it can be quite accurate Mar 01 19:25:10 rhf: you can get 2-3 pixel accuracy IIRC - when I measured it. Mar 01 19:25:18 with a stylus. Mar 01 19:25:30 but in qtopia desktop the sync function is deactivated Mar 01 19:25:42 pixel there is height/640 or width/480? Mar 01 19:25:50 rhkfin: yes Mar 01 19:26:12 thanks Mar 01 19:26:18 what pass? Mar 01 19:29:11 what tweaks should I do to a newly installed shr-unstable? Mar 01 19:30:00 does your sound work? Mar 01 19:30:05 fredrin? Mar 01 19:30:29 not tried yet.. Mar 01 19:30:50 which kernel? Mar 01 19:31:10 latest.bin? Mar 01 19:31:58 no, 28-rc4 i think Mar 01 19:32:24 right now i can't log into the phone :( the usb0 on my laptop wont come up, must do some resarch of what's wrong Mar 01 19:33:06 I installed wicd from opkg.org and think something broke Mar 01 19:38:07 fredrin: shr should be restarted at least once to get usb networking Mar 01 19:38:43 i just get this from dmesg Mar 01 19:38:45 [64185.816240] usb 6-1: new full speed USB device using uhci_hcd and address 2 Mar 01 19:38:45 [64186.048188] usb 6-1: new low speed USB device using uhci_hcd and address 3 Mar 01 19:38:45 [64186.169269] usb 6-1: device descriptor read/64, error -71 Mar 01 19:38:45 [64186.504339] usb 6-1: new low speed USB device using uhci_hcd and address 4 Mar 01 19:38:45 [64186.917254] usb 6-1: device not accepting address 4, error -71 Mar 01 19:38:46 [64187.032084] usb 6-1: new full speed USB device using uhci_hcd and address 5 Mar 01 19:38:48 [64187.441166] usb 6-1: device not accepting address 5, error -71 Mar 01 19:38:50 [64187.441612] hub 6-0:1.0: unable to enumerate USB device on port 1 Mar 01 19:40:19 fredrin: did you try rebooting FR Mar 01 19:42:19 freesmartphone.org: 03tim.niemeyer 07framework * r75552dc873f0 10/framework/subsystems/ogpsd/nmea.py: ogpsd: Fix 2D/3D fix reporting in NMEA Mar 01 19:42:52 paste.bin please Mar 01 19:47:57 small problems reboot, big problems reflash Mar 01 19:49:46 fredrin: no usb networking after first boot is a known problem, probably udev-related. Mar 01 19:50:48 PaulFert`: It's been working for a while, but today it did not work anymore, I think it got something to do with wicd, the only program that I installed Mar 01 19:50:59 *food* Mar 01 19:51:00 fredrin: oh with a usb cable Mar 01 19:54:47 only luck ive had is the r2 kernel on shr Mar 01 19:54:54 .28 Mar 01 19:58:00 slaxxin: i've had usb networking luck with all the kernels i tried, including 2.6.24 and some self-compiled andy-tracking etc. Mar 01 19:58:24 mee too Mar 01 19:58:41 things are a mess as we speak Mar 01 19:59:13 slaxxin: i just compiled the latest andy-tracking and i'm using it right now. Mar 01 20:01:59 paul will u link me:) Mar 01 20:02:28 is that 8608601? Mar 01 20:03:10 slaxxin: 86247495c3d585349e2a43de416d9275a48bda8c Link you with what? Mar 01 20:03:23 where is that kernel? Mar 01 20:03:38 an thanks Mar 01 20:04:35 slaxxin: it's on my FR and on my hard drive. I compile all the kernels i use :) Mar 01 20:07:01 i dont know how to compile as im just a usr Mar 01 20:10:00 mvuets: Хм, ничего не видел.... Давай еще раз. Mar 01 20:10:11 Disregard that :) Mar 01 20:16:01 unicode anyone? =P Mar 01 20:16:41 vininim: that's koi8-r i guess, so unicode guys are out of luck :) Mar 01 20:30:33 uImage-gta02-gc7283981_mwester-stable.bin Mar 01 20:30:40 shr Mar 01 20:30:44 latest Mar 01 20:32:28 ah, now usb networking is working again, I guess I'll stay away from wicd for a while Mar 01 20:33:14 slaxxin: what nice with that kernel? Mar 01 20:33:15 fredrin: i thought you ensured that usb0 on freerunner itself is properly up Mar 01 20:33:58 y no white screenod Mar 01 20:34:46 PaulFert`: how can I unsure that? Mar 01 20:34:58 wat? Mar 01 20:35:13 ensure, lol Mar 01 20:35:15 fredrin: with ifconfig or ip commands. Mar 01 20:53:51 slaxxin: i have sound, playing the latest Prodigy album on the FR now, sweet Mar 01 20:56:29 but pythm is a bit b0rked... Mar 01 20:57:51 fredrin: using the headset? Do you feel the lack of bass? Mar 01 20:58:03 just using the internal speaker Mar 01 20:58:21 pretty good speaker to be a phone Mar 01 20:58:46 but I read about the lack of bass problem Mar 01 20:59:05 but plan to solve that by buying a bluetooth head set Mar 01 20:59:06 fredrin: ah, so it doesn't matter much. Don't forget that all 3.5mm-2.5mm 3-ring converters are not suitable if you decide to try headphones. Mar 01 20:59:13 anyone know about a good one? Mar 01 20:59:59 fredrin: i'm not sure anyone got A2DP reliably working with high quality. Probably it's ok, i don't know. Do you have a link? Mar 01 21:00:39 PaulFert`: not yet, I'll do some internet window shopping now :) See if I can find something Mar 01 21:00:49 fredrin which kernel u useing? Mar 01 21:01:09 fredrin: i'm talking about getting decent results with freerunner and A2DP Mar 01 21:01:21 Linux om-gta02 2.6.28-rc4 #1 PREEMPT Sun Feb 8 19:53:16 CET 2009 armv4tl unknown Mar 01 21:01:34 where is it ? Mar 01 21:01:46 oo Mar 01 21:01:51 feb 8 Mar 01 21:02:12 rc4 huh Mar 01 21:03:01 I just took it from here one day: http://build.shr-project.org/shr-unstable/images/om-gta02/ Mar 01 21:04:17 so can u post it Mar 01 21:04:25 sure Mar 01 21:04:33 its not there Mar 01 21:05:52 or is it the latest.bin? Mar 01 21:06:23 it's named uImage-om-gta02-latest.bin Mar 01 21:06:28 ok Mar 01 21:06:56 slaxxin.latest Mar 01 21:07:19 lol Mar 01 21:07:22 Dave Mar 01 21:07:26 I call upon you! Mar 01 21:07:35 here i is Mar 01 21:07:36 Hey, Da Maestro is back! Mar 01 21:07:41 slaxxin: http://folk.uio.no/fredrin/uImage-om-gta02-latest.bin Mar 01 21:07:49 slaxxin: if you want it Mar 01 21:07:52 :o Mar 01 21:07:52 ok Mar 01 21:08:02 er du en nordmann? Mar 01 21:08:06 ja Mar 01 21:08:08 ! Mar 01 21:08:11 :D Mar 01 21:08:12 i argentina Mar 01 21:08:16 hahaha Mar 01 21:18:04 whitescreened on me Mar 01 21:18:10 MAN Mar 01 21:18:25 slaxxin: do you use Qi? Mar 01 21:18:35 yup Mar 01 21:18:45 do i need to stop? Mar 01 21:18:56 backup? Mar 01 21:19:03 I had the same problem with Qi and wihtescreen Mar 01 21:19:06 ok Mar 01 21:19:07 ic Mar 01 21:19:08 went back to u-boot Mar 01 21:19:25 but it sucks a bit because the long suspend/resume time :( Mar 01 21:19:26 i c that dave has been keeping me in the dark Mar 01 21:20:05 is there a newer Qi? Mar 01 21:20:33 newer than what? Mar 01 21:20:59 is there only 1 version of Qi Mar 01 21:22:13 nah, looks like more http://people.openmoko.org/andy/ Mar 01 21:23:47 GTA02 -> qi-s3c2442 is the one Mar 01 21:28:27 slaxxin: http://git.openmoko.org Mar 01 21:32:52 any cool splash screens out there? Mar 01 21:41:50 i don't think SHR likes Qi so much Mar 01 21:42:22 fredrin: what's wrong with SHR? Mar 01 21:42:45 it wont suspend now that i switched to qi Mar 01 21:42:52 not sure why Mar 01 21:43:20 even if i push the power button and click on the suspend icon it wont suspend Mar 01 21:43:28 fredrin: does GSM work? Mar 01 21:44:03 fredrin: please check GSM functionality Mar 01 21:44:22 maybe reboot 1 time Mar 01 21:44:24 hi! anybody knows if I can install openmoko over a ks20? Mar 01 21:45:05 xota: what's ks20, what's openmoko? Mar 01 21:45:06 I'll start with the reboot Mar 01 21:46:25 PaulFert`: ;? Mar 01 21:46:53 xota: please clarify your question. What is "ks20" and what is "openmoko"? Mar 01 21:48:27 openmoko a OS for mobile devices and ks20 a LG mobile phone? Mar 01 21:49:14 xota: the short answer is: No. Mar 01 21:49:58 PaulFert`: does not look like it Mar 01 21:50:23 mwester and the long answer? ;) Mar 01 21:50:29 xota: there is no OS called "openmoko" at all. BTW, does ks20 have a working Linux? Mar 01 21:50:40 fredrin: GSM doesn't work? Ok let's investigate it. Mar 01 21:51:05 fredrin: what's the output of s3c24xx-gpio H0 and the same for H1, H2, H3? Mar 01 21:51:19 well... openmoko is the full set: hw and sw? Mar 01 21:51:32 xota: not at the moment Mar 01 21:51:54 what's then? I think that lg ks20 doesn't have a working linux Mar 01 21:51:56 PaulFert`: how do i check that? Mar 01 21:51:59 xota: does a working kernel fro ks20 exist? Mar 01 21:52:16 fredrin: type s3c24xx-gpio H0 and provide me with output. Mar 01 21:52:23 xota: why do you ask then? Mar 01 21:52:57 PaulFert`: I think no, I've just bought the ks20 terminal and like to see alternatives to WM OS, but I don't have the terminal. I will receive it next week Mar 01 21:53:27 I get "F0" Mar 01 21:53:42 xota: if there is a Linux for device, then porting FSO to it will be appreciated. If there's no working kernel, you're out of luck. Mar 01 21:53:53 fredrin: What about H1, H2 and H3? Mar 01 21:54:09 but maybe the problem is that the pin code dialog never poped up? Mar 01 21:54:29 fredrin: that's a consequence, not the root of the problem. Mar 01 21:55:25 fredrin: what GSM firmware version do you use btw? Mar 01 21:55:38 PaulFert`: :) k H1 = "0R", H2 = "1", H3 = "F1R2 Mar 01 21:55:41 PaulFert`: :) k H1 = "0R", H2 = "1", H3 = "F1R" Mar 01 21:55:43 :) Mar 01 21:56:07 what is this data? Mar 01 21:57:02 fredrin: that's very important data and it looks bad. What GSM firmware version are you using? Mar 01 21:57:16 the default one Mar 01 21:57:18 fredrin: can you confirm with ps ax |grep gsm0710muxd that it's running? Mar 01 21:57:30 but I'm using a Norwegian sim-card in Argentina Mar 01 21:57:40 the phone network here is pretty unstable Mar 01 21:58:34 PaulFert`: It looks stalled: 1414 ? S 0:00 /usr/sbin/gsm0710muxd Mar 01 21:58:34 Mar 01 21:58:42 fredrin: it's ok. Mar 01 21:58:54 fredrin: provide me with cat /sys/class/i2c-adapter/i2c-0/0-0073/dump_regs |head -2 output (only the second line is necessary) Mar 01 21:59:47 15 47 ff 01 00 01 00 00 02 08 6b 01 00 0a 1b 02 Mar 01 22:00:20 fredrin: now issue s3c24xx-gpio H2=F Mar 01 22:00:45 k Mar 01 22:01:12 fredrin: now what does s3c24xx-gpio H2 outputs? Mar 01 22:01:27 F1 Mar 01 22:02:29 PaulFert`: could this have anything to do with switching from u-boot to Qi? Mar 01 22:02:37 fredrin: that's better :) If your pin dialog doesn't come up, kill gsm0710muxd (and it'll be restarted by frameworkd automatically) Mar 01 22:03:04 fredrin: it can be that, yes, because you're using an old kernel (and i know, that you don't compile yourself). Mar 01 22:03:45 PaulFert`: K, now it suspended and killed my ssh connection, just wait a sec Mar 01 22:05:21 fredrin: it should be possible to resume by pressing power button Mar 01 22:05:59 yes, but the ssh connection dies Mar 01 22:06:33 no pin dialog, I'll do a reboot Mar 01 22:06:37 fredrin: no Mar 01 22:06:49 fredrin: if you restart, you'll lose the workaround :) Mar 01 22:06:50 ups Mar 01 22:07:56 fredrin: ayt? Mar 01 22:08:16 ayt? Mar 01 22:08:19 fredrin: don't try to suspend from now on while we're investigating, ok? ;) Mar 01 22:08:24 fredrin: Are You There Mar 01 22:08:33 PaulFert`: yupp Mar 01 22:08:56 fredrin: tell me when you're ready Mar 01 22:09:46 ready, Mar 01 22:10:18 fredrin: so, you have ssh access and s3c24xx-gpio H2 shows F1? Mar 01 22:10:59 no H2 shows 0R Mar 01 22:11:20 fredrin: that means you restarted, well... Do s3c24xx-gpio H2=F again :) Mar 01 22:11:57 now it says F1 Mar 01 22:12:24 fredrin: ok :) Now kill ophonekitd and see if gsm0710muxd disappears after that. Mar 01 22:13:32 PaulFert`: ophonekitd is not runningg Mar 01 22:13:47 fredrin: check if gsm0710muxd is running Mar 01 22:14:45 it is Mar 01 22:15:21 fredrin: kill it Mar 01 22:16:01 killed :) Mar 01 22:16:32 fredrin: check again it's not running and start it manuall "gsm0710muxd -v" Mar 01 22:17:25 k Mar 01 22:18:03 fredrin: well, pastebin the output Mar 01 22:20:45 fredrin: you can kill it if you like btw, it'll produce the output indefinetely. Mar 01 22:21:30 did that Mar 01 22:21:37 fredrin: hm, i'm now sure s3c24xx-gpio H1=F is also needed Mar 01 22:22:03 http://pastebin.com/m60113b72 Mar 01 22:22:35 should I set that before I start gsm071muxd? Mar 01 22:23:05 fredrin: oh, it's not really needed. Mar 01 22:23:40 fredrin: just start ophonekitd and it'll ask frameworkd for GSM and frameworkd will start gsm0710muxd and it'll work. You pastebined the good output. Mar 01 22:24:43 This is with H1 = F1 Mar 01 22:24:44 http://pastebin.com/m1bb96dcd Mar 01 22:25:11 fredrin: the same good output Mar 01 22:26:22 fredrin: i suggest you to add these commands to your startup scripts (it's better to do for both H2 and H1) before ophonekitd is started. I guess it'll make GSM (and suspend :) )work. Mar 01 22:26:25 one output is after I set the H1 = F1 Mar 01 22:27:09 to set both H1 and H2 to F1? Mar 01 22:27:18 fredrin: both to F Mar 01 22:27:22 ah, k Mar 01 22:28:35 fredrin: F means special pin function. In this case it's hardware flow control (not really needed until you upgrade your gsm firmware which is very good to do anyway) and the other is data to the modem (which is strictly necessary for bi-directional communication). Mar 01 22:29:20 so, if I upgrade the firmware, my problems could go away? Mar 01 22:29:21 :) Mar 01 22:29:52 fredrin: to make your problem go away i suggest you to add these commands to your startup scripts (it's better to do for both H2 and H1) before ophonekitd is started. I guess it'll make GSM (and suspend :) )work. Mar 01 22:30:16 fredrin: you can check right now by starting ophonekitd again. Mar 01 22:30:59 it seg faulted Mar 01 22:31:15 fredrin: export DISPLAY=:0 Mar 01 22:31:20 fredrin: and then start it :) Mar 01 22:32:11 of course :) Mar 01 22:32:25 fredrin: still segfaults? Mar 01 22:32:41 nope, and pin dialog came Mar 01 22:32:54 sweet Mar 01 22:33:09 fredrin: cool, thanks for interactive debugging session :) Mar 01 22:34:21 :) thanks for the help Mar 01 22:34:44 fredrin: don't forget to modify your bootscripts to include this workaround. Mar 01 22:35:03 doing it now Mar 01 22:35:20 fredrin: that's a kernel issue, it doesn't initialize the lines properly. Fixed in new versions that didn't reach the distributions yet. Mar 01 22:36:50 PaulFert`: in what file should i put the workaround? Mar 01 22:37:05 fredrin: before ophonekitd start. Mar 01 22:37:16 fredrin: i'm not an SHR user, so i don't know exactly. Mar 01 22:37:32 k Mar 01 22:37:49 fredrin: you can add it to the frameworkd startup file (look in /etc/init.d/) Mar 01 22:40:18 ok Mar 01 23:28:15 Openmoko: TRANSFORM! Mar 01 23:28:52 lulz Mar 01 23:45:59 TRANSFORMING Mar 01 23:49:01 transform into something usable, please like a cell phone Mar 01 23:54:46 mrmoku|away: lol for the workaround for fixing e high cpu usage Mar 01 23:54:57 it will be very usable Mar 01 23:55:04 after there is a working webkit browser Mar 01 23:55:11 http://blog.shr-project.org/2009/03/time-for-testing.html Mar 02 00:01:22 Hi, any idea which browser is used on this screenshot: http://scap.linuxtogo.org/files/e4340718ef28604518012670269d0063.png Mar 02 00:01:24 ? Mar 02 00:02:54 looks like webkit Mar 02 00:03:10 (the thing apple's safari is made of) Mar 02 00:04:13 and konqueror? Mar 02 00:04:48 niv: yes, i know.. i think it is a gtk-webkit browser, but it would be nice to know what its name is. ;) Mar 02 00:05:15 lindi-: konqueror is still khtml, which is was webkits starting point Mar 02 00:15:50 here's a list of webkit/gtk+ based browsers [http://trac.webkit.org/wiki/ApplicationsGtk], but it doesn't seem to be one of the ones listed.. it might be based on om2007.2 s browser, though Mar 02 00:51:06 I'm writing this atm btw: http://people.defora.org/~khorben/share/DeforaOS/surfer/surfer-hackable1-daily-gtkhtml-fullscreen.png Mar 02 00:51:26 (based on GtkHtml 2, so no javascript and half-decent rendering but quite fast) Mar 02 00:51:48 cool Mar 02 01:03:37 khorben: Nice. So you are using hackable:1? Mar 02 01:08:24 SHR unstable's crapping out on me. Mar 02 01:08:45 ophonekitd doesn't run, it won't suspend... Mar 02 01:08:49 Pythm quit working... Mar 02 01:08:53 It was working fine an hour ago. Mar 02 01:11:31 frameworkd won't stay on. Mar 02 01:43:44 slaxxin: sound from the headset aswell in Shr, tested now Mar 02 01:43:56 weak bass though Mar 02 01:44:16 a bit lagg when you switch between headset and internal speaker Mar 02 02:42:49 Hmm **** ENDING LOGGING AT Mon Mar 02 02:59:57 2009