#qi-hardware IRC log for Wednesday, 2011-11-16

wolfspraulgood morning qiots01:08
wolfspraulDocScrutinizer51: congrats on your success! that is really fantastic news and they should be proud to have you01:09
wolfspraulmeanwhile I hope you keep throwing a snippet of wisdom over the fence into our ocean of cluelessness sometimes :-)01:09
DocScrutinizerlet's wait and see what results from today's interview01:23
wolfspraulah, still interviewing?01:23
wolfspraulI thought the deal was sealed already :-)01:24
DocScrutinizermaybe I can push a bit on Ericsson actually selling "open" LTE modules, even when my possible position as "Master of $arbitrary bus" doesn't suggest I could do much in that direction01:25
DocScrutinizernot yet sealed, just highly likely to get signed next week01:27
wolfspraulyou can try but I doubt you can get any openess beyond lip service01:29
wolfspraulI think raster at Samsung is also happily rationalizing himself as a freedom fighter inside Samsung with Samsung oh-so-wonderfully embracing open real-soon-now (tm)01:29
wolfspraulthe reality of those corps is that they are under huge pressure to deliver continuous and demanding financial results, every quarter01:30
wolfspraulalmost like a bank with a small engineering leftover :-)01:30
DocScrutinizerhaha, no I don't think I can save the world :-D01:30
wolfspraulif I were to work for such a company on the mgmt side, I would find 'open' too risky01:31
wolfspraulrather just focus on the main and largest customers that are really driving volume today, and think about how to serve them and share investment costs among them01:31
wolfspraulI do see some pressure here and there though towards open APIs/SDKs etc.01:32
wolfspraulI guess the spiraling development costs cause some pain...01:32
wolfspraulinteresting times ahead, you will have fun I'm sure01:32
wolfspraul(if it works out)01:32
DocScrutinizerbut look, they build chips for LTE, I'm (or maybe will be) in the hw/sw R&D team and have full access to specs and src. I might be able to evaluate value of a particular chip for open hw/sw, and help on whatever obstacles get in the way on exploiting e.g some USB stick with that LTE module in it01:33
wolfsprauloh sure01:33
wolfspraulit's great for your brain and they will not be able to silence you :-)01:33
wolfspraul(and will not want to I'm sure)01:33
wolfspraulso you will learn very valuable things for everybody, us poor Qi fellows included01:33
wolfspraulfor LTE and other high-end RF standard, I am most interested whether there is something that allows spontaneous networking among nodes01:34
DocScrutinizerhah, I guess that could only be done by some hacker firmware written for my own pleasure at a weekend ;-D01:35
DocScrutinizerloock, the very basic standards work with distincr bands for up- and down-stream, so two mobile equipments are fundamentally not capable of talking to each other01:37
DocScrutinizerlook*01:37
DocScrutinizerboth send on band A and receive on band B01:37
wolfspraulyes, I know01:38
DocScrutinizerthough those bands are only some dozen MHz apart usually, the standard will not work for that OOTB and without severe hacks01:38
wolfspraula lot of designs are inherently client-server01:38
wolfspraulit's built like this from day 1, because that's the only thing they think about anyway :-)01:39
wolfspraulpeer2peer sounds like some hippie fantasy...01:39
DocScrutinizerthen at least GSM is TDMA means it *needs* a central pacemaker clock01:40
wolfspraulbut that doesn't change my thinking that I am interested in direct networking between peers01:40
wolfspraulat least the pile of things to look at shrinks dramatically once you make that requirement, true? :-)01:40
wolfspraulI find it 'asymmetric' (to use Werner's words) that I need a megacorp to wirelessly network with my friends01:41
wolfspraulso I'm interested in p2p rf options, if we can get that to work well it may allow us to make unique and good products01:42
DocScrutinizerthere are dedicated "free" bands for such projects01:42
wolfspraulyes01:42
wolfspraulbut we need chips, dedicated for that or hacked/modified01:43
wpwrakDocScrutinizer: hey, sounds like good news. pre-congratulations ! :) when will they let you know their decision ?01:45
DocScrutinizernext week01:45
wpwrakstart on december 1st?01:45
DocScrutinizernot exactly 1st I think, that's a bit too short for all the paperwork in a 7000 man company, you know how things are handled there01:46
wpwrakheh :) so probably january then01:47
wpwrakgood. so we have time to pick your brain some more :)01:47
DocScrutinizernah, they explicitly asked me if I could start pre-Xmess01:47
wpwrakfunny. they want to pay you for the holidays :)01:48
wpwrakah wait. they're all on the weekend this year. clever :)01:48
DocScrutinizerI think this year holidays are perfect for *them*01:48
wolfspraulthis looks interesting http://www.hughski.com/02:26
wolfspraulhttp://blogs.gnome.org/hughsie/2011/11/13/introducing-the-colorhug-open-source-colorimeter/02:26
wolfspraulhe's using geda, all sources there etc. https://gitorious.org/colorhug02:29
Action: DocScrutinizer feels like Scotty, shouting "Computer, search route to Hauptmarkt!" at N9 with voicetogoog running02:39
wpwrakxiangfu: heh, a wiki page with 2000+ entries, just explaining how to exit each application. that would be fun ;-))07:28
xiangfu07:28
xiangfuwpwrak, Hi. I have a kernel opps. needs your help.07:31
wpwrakwhat did you do to your kernel ? :)07:32
xiangfuwpwrak, those documents is not easy. hope we also have a magic 'apple' key can always go back to 'gmenu2x' maybe. :)07:32
xiangfuI trying very hard scp the demsg to downloads.qi-hardware.com07:33
wpwrakpastebin ?07:33
xiangfuthe networks(openvpn) sometime very slow. ping give your 50% package loss. and 600+ms. 07:34
xiangfuhere is hte dmesg: http://downloads.openmobilefree.net/tmp/dmesg07:34
xiangfuit's the ks7010 driver. 07:34
xiangfuI try to find out the call trace. seems all is not relate about ks7010 and mmc driver. 07:34
xiangfuI don't know how to find more info from kernel oops message. learning that today... 07:35
xiangfuhere is the pastebin: http://pastebin.com/kKM0iUC107:35
xiangfuall the kernel stuff can find at : http://fidelio.qi-hardware.com/~xiangfu/openwrt-xburst.full_system/build_dir/linux-xburst_qi_lb60/linux-3.0/, system.map vmlinux 07:36
wpwrakqueue_work_on, printk, queue_work, __mmc_claim_host, <module>, default_wake_function, ...07:42
wpwrakand on top enqueue_task_rt07:48
xiangfuis the 'ra    : 8004e1a4' make you find out queue_work_on?07:49
wpwraki just looked at the call trace07:52
kyakxiangfu: btw, the instruction for nupdf are outdated anyway, because it is part of the image, no need to copy :)07:54
xiangfuwpwrak, is this help? : http://pastebin.com/6r18tySh07:55
xiangfukyak, Hi. here is the i386 SDK, and toolchain: http://downloads.openmobilefree.net/NanoNote/OpenWrt-SDK-xburst-for-Linux-i686/compile-log/openwrt-xburst.full_system-11142011-1520/07:56
xiangfuis ready for next release. now I have to fix the SDIO wifi bug. then release the images :)07:56
xiangfuthis image include a new boot slash screen. and wpan.nn script file. :)07:57
kyakxiangfu: ah cool! will you copy the toolchain to qi-hw.com?07:57
wpwrakit's some null pointer that you're getting there. but i'm not sure where it comes from. you may want to add a printk to __mmc_claim_host to examine the content of *host07:57
wpwrakmaybe check if the pointer is sane at all07:57
xiangfukyak, yes. already. just not modify the 'latest' link. 07:58
rejonhttp://en.qi-hardware.com/wiki/Colorimeter07:58
rejonhttp://en.qi-hardware.com/wiki/Laser07:58
rejongotta document more things07:58
xiangfuwpwrak, ok. got it. will report later.07:59
wpwrakmaybe something is amiss with host->detect07:59
xiangfukyak, (nupdf) yes. it already included. :)08:11
xiangfurejon, (colorimeter) yes. cool. maybe it can connect to nanonote. then this colormeter don't needs any driver. or install any app in PC. just Colormeter + Ben nanonote is enough.08:17
wolfspraulI doubt that as it would most likely want to talk to a USB Host08:18
xiangfuwolfspraul, yes. it is USB Host. 08:18
rejoni need to reflash my nanonote08:19
rejoni hope i didn't brick it08:19
xiangfuwolfspraul, I want say is : if the device don't needs any driver  etc. just open box. powerup. then it working . will be cool. then current 'Colormeter' :)08:19
xiangfurejon, update your 'reflash_ben.sh' first. :)08:19
rejonyah08:19
rejonwill do08:19
xiangfurejon, make sure you use the latest version of reflash_ben.sh08:20
rejoni'm about to take my markers to the nanonote08:20
rejonto improve the case08:20
rejonhttp://fabricatorz.com/2011/11/2nd-round-sharism-this-year-in-beijing/08:22
rejonsharism beijing tomorrow08:22
rejonmore milkymist demos08:22
rejonshould sell some nanonotes on the spot08:22
rejonxiangfu, wolfspraul 08:23
rejonsharism presents beijing will now be monthly08:23
rejonkeep moving locations08:23
wpwrakrejon: how do you demo the M1 ?08:24
rejonwill just leave it running in the background08:25
wpwrakrejon: and to what kind of people ?08:25
rejonwhat i would really like is the image support to work08:25
wpwrakhmm, that's not so cool :-(08:25
wpwrakthat's the USD 500 screensaver08:25
rejonso it could just cycle through the twittersphere, and pull images08:25
rejonwpwrak, ha08:25
rejonha, lots of these people already seen the demo08:25
rejoni just keep it with me all the time08:25
rejoncasual08:25
wpwrakyou should show it off. play it. like an instrument08:25
rejonnot that kind of event08:26
wpwrakjust for a moment08:26
wpwrakshow why it's cool08:26
rejon:)08:26
rejoni like it in the bg08:26
rejoni'm not really into the foreground, performance ;)08:26
wpwrakbah08:26
wpwrakcourage. you can do it ! nothing sells if it's kept in back of the storage room :)08:27
wpwrakyou should try it. it's fun. you can also let others play with it. use my MIDI RMX. that one has all the basics08:28
wpwrakof course, you may need to upgrade your travel case if you also add a MIDI controller :)08:30
xiangfuwpwrak, maybe we add a demo patch. that support current Android/iPhone OSC application.  08:30
xiangfuwpwrak, since everyone in the meeting will be using iPhone or Android phone. :)08:31
xiangfuwpwrak, oh. no. it have to modify the osc source code to accept the different osc message.08:31
xiangfunot just 'patch'08:32
wpwrakyes,why not. i don't know the droid/iPhone app world, but if there's some nice multifunctional controller there, that ought to work08:32
wpwrakuse MIDI-over-OSC ?08:32
wpwrakwell, you need a PC anyway, to go from WLAN to ether. so you can also do the translation there. liblo is your friend :)08:33
wpwraki was pleasantly surprised how easy the whole MIDI to OSC translation was. i think it took me longer to get that controsc to build that it took me to write midi2osc :)08:36
xiangfuwpwrak, for you it's easy. not for everyone. at least not for me :D08:43
wpwrakxiangfu: naw, just look at it. you can find all the ingredients conveniently in controsc, oscsend, and - for osc to osc - oscdump08:46
oogaboogaanyone have a fake or broken or toy CPU for sale?14:11
kristianpaulnice https://www.yubico.com/personal-use19:53
kristianpaulbut still closed hw it seems, i would no trust that thing..19:58
qi-bot[commit] Werner Almesberger: USB analysis tools (master) http://qi-hw.com/p/wernermisc/ed8593019:59
kristianpaulnow i see why people keep saying nanonote a password wallet ;)19:59
wpwrakand that thing probably isn't a password wallet anyway20:01
kristianpaulhi Artyom 20:03
Artyomhello kristianpaul :)20:05
kristianpaulArtyom: how is your milkymist hacking going?20:13
ArtyomVery exciting ;)20:14
ArtyomI've spent a lot of time installing all required software for linux ;)20:14
ArtyomI tried to run carlos project but with no success :(20:15
ArtyomToday I found Zet-processor project and I will try to use their SDR SDRAM controller20:15
ArtyomI also spent a lot of time studing documentation on milkymist.org site. 20:16
ArtyomNow I have understanding of how many things are done in milkymist SoC20:17
ArtyomI must say that I'm impressed by the amount of work that was done by lekernel and the rest.20:18
ArtyomThe more I study milkymist SoC the more I like it :) I'm a little disappointed by the fact that I didn't know about it earlier ;)20:20
ArtyomAnd how is your success with using front-end clock?20:27
kristianpaullate... sorry... damn unexpected meetings..22:30
kristianpaulArtyom, basically the same, clock seems to be OKAY but system dont boot, i know i had let something uncofigured..22:31
kristianpaultryin to find what..22:31
--- Thu Nov 17 201100:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!