#qi-hardware IRC log for Thursday, 2012-01-26

qi-bot[commit] Werner Almesberger: m1/patches/rtems/mem-fixes.patch: fix mdump and medit shell commands (master) http://qi-hw.com/p/wernermisc/19462f401:06
wolfspraulplanet was stuck (permission problem on server, my bad)11:02
wolfspraulthanks to Harald for pointing it out to me, I guess he misses the planet mosts more than I do? how is that possible???11:02
wolfspraulthanks Harald!11:02
wpwrak"save the planet" ;)11:08
qi-bot[commit] Werner Almesberger: m1/ledravaganza/ramp.pl: exponential ramp generator for ledm (master) http://qi-hw.com/p/wernermisc/94e955815:05
viricwpwrak: do you like this? http://www.openscad.org/19:31
Jay7where is tuxbrain? :)19:39
Jay7I see error on his site :)19:39
viricthe bot can be asked19:45
viricbut I don't know of him since a long time ago19:46
wolfspraulerror on his site?20:04
Jay7user warning: Unknown column 'token' in 'field list' query: UPDATE dpr01_captcha_sessions SET token='e50a89d13561813d263afaf9907b5a94' WHERE csid=1071880 in /home1/tuxbrain/public_html/sites/all/modules/captcha/captcha.module on line 216.20:16
whitequarkpersonality multiplication?..20:34
viricmaybe we'd have to ring him20:35
wolfspraulI think more or less we have to move forward without him, which is sad20:43
larsc14 weeks?20:45
wolfspraulhe's fine, but as much as he loved tuxbrain.com he had to take up a full-time job, and with his kid etc. it's becoming very hard right now to be active in open hardware20:45
wolfspraulhis spirit is alive though, one day he will be back I'm sure20:45
wolfspraulI'm a little worried to send EU NanoNote customers to him nowadays, as I can also barely get email responses, but he says he is still filling orders...20:46
wolfspraullarsc: you have a job now, let's see when you also had a kid ;-)20:47
larscthat will hopefully still take a few years20:49
viricwolfspraul: ah ok!21:16
viricImpressive, that he is so busy.21:17
viric14 weeks without looking at irc.. I've never been so busy21:27
kristianpaulyou have kids?21:28
viricno no :)21:29
viricbut I know people with kids that enter irc. Some! :)21:29
wolfspraulI think I've heard once that 80% or so of FOSS developers leave the scene after they have kids21:34
larschehe and only come back after the kids have left home21:35
wpwrakviric: (openscad) nice idea but it's hard to avoid artefacts. see also this thread: http://lists.en.qi-hardware.com/pipermail/discussion/2011-July/008476.html21:37
viricah I remember your tests21:38
wpwrakit seems that a lot of studies that try to gauge the overall happiness of people find that, for men, there's a negative correlation with having kids ...21:38
wpwrakwolfspraul: i hope you're one of the outliers ;-)21:39
wolfspraulyes I'm fine21:39
wolfspraulbut I'm very childish myself I think :-)21:39
wpwrakhehe ;-) it probably also helps not to be in the traditional employee role. less conflict between limited control over your own life yet being responsible for another one.21:40
wolfspraulso... let's play!21:40
wpwrakmore usb-midi fun ? ;-)21:40
wolfsprauldocumenting keyboards and mice fun21:40
wolfspraula rather exotic game, I admit21:41
wpwrakoh :) what kind of documenting are you doing ?21:41
wpwraki have a pile of mice still waiting for me to return to M1 usb testing. what i was thinking of was making a register of them, with the following information:21:44
wolfspraulhere http://en.qi-hardware.com/wiki/Milkymist_One_accessories#keyboard21:44
wolfspraulit's not much but cleanup is tedious21:44
wpwrakname, description of key characteristics, picture, lsusb -v, HID report descriptor dump, and its status with M121:45
wolfspraulI want to add a little picture for each to make it easier to get a feeling once the table gets longer21:45
wolfspraulof course I don't plan to document all keyboards or mice but the idea is that we have some baseline and can document our quality efforts, for ourselves and others21:46
wpwrakyou may also need some of the dump information. some mice may bear the same name yet differ internally21:46
wolfspraulwhen I'm finished with keyboard and mice I will continue with the midi controllers21:46
wolfspraulbut it's a wiki, and I really only do what helps me right now understand and guide the quality effort21:46
wolfspraulI have 5 keyboards right now, and 4 mice21:46
wpwrakdo you have a means to generate wiki pages from files ?21:46
wolfspraulxiangfu has a few more keyboards and mice21:46
wolfspraulI think I will leave the soc+binary on my m1 that you gave me21:47
wolfspraulcall it 2012-01-26 werner interim build21:48
wolfspraulmy feeling is maybe after your next round of usb fixes, most keyboard and mice will just work21:48
wpwrakyeah, i hope so. we had some regressions but that should be fixable21:49
wpwrakfor a proper USB devices database, i think we should generate the wiki content and keep the raw data in git. cut out the manual processing.21:51
wpwrakplus, that way it would also be easy to run searches, diffs, and such21:52
wpwrak(searches as in "do we have another device with this of that quirky report format variation ?")21:53
qi-bot[commit] Werner Almesberger: m1/BUILD-CHEAT-SHEET: use 2011/12/01 version for now (master) http://qi-hw.com/p/wernermisc/65555f721:55
wolfspraulusb devices database sounds very scary21:55
wolfspraulin my opinion that could quickly become a massive distraction of our very limited resources21:56
wolfspraulinstead, all I will do is to have a table of known supported and known unsupported devices21:56
wpwrakwell, "database" = a set of files21:56
wolfspraulover time collected from core users and contributors21:56
wpwraki'm not talking about SQL ;-)21:56
wolfsprauland also delete from that table if we loose track of whether something works or doesn't21:56
wolfspraulthe idea is to give people a quick feeling and guidance as to what works and what doesn't21:57
wolfspraulquality of the table goes above quantity21:57
wpwrakyes, that's what the generated table should contain21:57
wpwrakplus, there should be links to lsusb -v and the HID report descriptor. that way, one can see some of the technical background.21:57
wpwrakwith a set of files, all this is easy to manage. in wiki, it's like surgery through a rectal probe ...21:59
wolfspraulI'll do this in wiki right now22:01
wolfspraulit's the quickest and easiest way, and allows for temporary format diversity, later cleanup, etc.22:01
wolfspraulbut if you put your data into files, why not22:01
wpwraklsusb -v -d xxxx:yyyy >critter-lsusb is so much easier than wiki copy & paste :)22:02
wolfspraulI'm not trying to convert you to wiki, not at all22:03
wpwrakplus any confusion that may come up if someone mis-copy&pastes22:03
wolfspraulplease do what works best for you22:03
wolfspraulcollecting information on M1 accessories will be messy though, unless we do it in a very centralized way which will not happen22:03
wpwrakwell, that's what i was aiming for ;-)22:07
wpwrakgetting the report descriptor is a bit tricky, though. has several prerequisites. basically a "developers only" operation.22:08
wpwrakand of course, you need linux :)22:08
wpwrakwell, maybe we can add some hack to m1 in the future to dump that data to a file, then do the formatting offline22:09
whitequark(surgery through a rectal probe) finally I know how this is translated to English22:57
whitequarkin russian, you have two analogues22:58
whitequark1) cut one's tonsils through one's rectum22:59
wpwraknow, when enjoying social talk at parties, you can finally explain to the ladies present what you do for work (-:C22:59
whitequark2) cut one's tonsils through someone other's rectum22:59
wpwrakoh, didn't know #2 ;-)22:59
whitequark#2 is soooo applicable to flash23:00
whitequarkits compiler has a hack which translates finally clauses as catch() ones (emitting invalid bytecode in the process)23:00
whitequark_and_ its VM has a hack which translates these catch() clauses back to finally ones (ignoring invalid bytecode in the process)23:00
whitequarkI don't even have an adequate obscene comparsion for that.23:02
whitequark... have I damaged your brain? Sorry for that :D23:03
wpwrakmaybe compiler and VM are the work of two different groups of developers ? :)23:06
whitequarkthe VM is called tamarin23:10
whitequarkand I guess that the developers share some genetic relations with the animal23:10
whitequarklike 100%23:10
whitequarkyou know, "a million monkey theorem"23:11
whitequarkand I won't even start to tell you which kinds of relations do developers of the compiler have with the developers of VM23:11
whitequarkand which relations exactly I would like to impose on both groups23:11
wpwrak(relations) the same two grandparents ? :)23:17
whitequarkyou have a very innocent mind don't you? :D23:18
whitequarkanyway, I need some sleep23:28
whitequarkand I hope I'll never dream of _that_23:28
--- Fri Jan 27 201200:00

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