#qi-hardware IRC log for Tuesday, 2011-05-10

qi-bot[commit] Werner Almesberger: atrf-xtal/atben.c: mention that the AT86RF230 driver can also be disabled http://qi-hw.com/p/ben-wpan/d184d1d01:06
qi-bot[commit] Werner Almesberger: tools/lib/atben.c (atben_open): corrected comment claiming the clock was 316 MHz http://qi-hw.com/p/ben-wpan/89bcd3c01:06
qi-bot[commit] Werner Almesberger: atrf-xtal/atben.c: added missing #include; corrected garbled comment http://qi-hw.com/p/ben-wpan/180603e01:06
qi-bot[commit] Werner Almesberger: atusb/fw/Makefile (.PHONY): added development-only phony targets http://qi-hw.com/p/ben-wpan/c5992bc01:06
qi-bot[commit] Werner Almesberger: atusb/fw: fixed USB bus reset handling by software http://qi-hw.com/p/ben-wpan/eeb3af601:06
qi-bot[commit] Werner Almesberger: atusb/fw/Makefile (dfu): pass transfer size to hide DFU non-compliance http://qi-hw.com/p/ben-wpan/0db361d01:06
qi-bot[commit] Werner Almesberger: atusb/fw/README: updated for complete firmware upload process http://qi-hw.com/p/ben-wpan/1f9943501:06
wpwrakheh, how do we unban the unfortunate DocScrutinizer ?01:22
wpwrakwolfspraul: do you know how to unban someone at qi-bot ? we have DocScrutinizer meekly scratching at the door after he got himself kicked :)01:30
kristianpaulI knew it a bit with op may be dagerous someday..01:31
wolfspraulno idea, but this is something new, so it's fun :-)01:31
wolfspraulhey qi-bot, don't misbehave, eh?01:31
wolfspraulhow did DocScrutinizer get kicked?01:32
kristianpaulwpwrak: i think kick dont mean ban actually01:32
wpwrakhe did a bunch of nick changes01:32
kristianpaulwpwrak: http://paste.debian.net/116506/01:32
mthin gmenu2x, do we have to support 24bpp? the non-padded version, I mean, as opposed to 32bpp of which 8 bits are unused01:33
mthkristianpaul: he was banned and then kicked01:34
wolfspraulmaybe it will just time out?01:34
kristianpaulah +b01:34
kristianpaulwolfspraul: can you do op?01:34
wpwrakmth: draw, then quarter. traditional justice ;-)01:34
wpwraker schlaegt  /mode #qi-hardware <user>   vor01:35
kristianpauli think is matter of aply -b flag to his nick..01:35
wpwrakoops  /mode #qi-hardware <user> -q01:35
mthsomeone needs to become operator first01:36
wpwrakgrmbl. LANG didn't survive the context switch, sorry01:36
kristianpaulah.. :-)01:36
mthafaik you ask ChanServ to make you operator01:36
kristianpaullets ask qi-bot ;)01:36
mthkristianpaul: be careful, it you're not super polite it might ban you ;)01:37
kristianpaulmth: you mean this 24bp is obsolte in the code?01:37
wolfspraulwell I'm not an op either01:37
wolfspraulthere is a console somewhere, lemme see...01:38
mthkristianpaul: I'd like to remove it because it would simplify the code, but I don't know if there is still hardware we'd like to support that uses non-padded 24bpp01:38
kristianpaulmth: not for now i think :-)01:38
mthafaik the NanoNote uses 32bpp and the Dingoo uses 16bpp01:38
kristianpaulthat will mean break upstream..01:39
kristianpaulcant it became optional instead?01:39
mthit's quite possible no device supported by gmenu2x uses this format01:40
mththis code is from SFont, which is about 10 years old01:40
mthmaybe older when it was first written01:41
wolfsprauldon't know, this eggdrop thing is too cryptic for me01:41
wolfspraullet me just restart qi-bot, maybe that helps...01:42
wpwrakwolfspraul: maybe you can ask chanserv to give you op status ?01:42
wolfspraulbah there's so many settings and flags in all this01:43
wolfspraulI have no idea...01:43
kristianpaul /msg ChanServ op #qi-hardware wolfspraul01:43
kristianpauljust try..01:43
wolfspraulyou mean I should type that?01:43
wolfspraulyou are not authorized to perform this operation01:43
wpwrakdefeated by the bots ;-)01:44
kristianpaulsooner or later..01:44
mth"/msg ChanServ access #qi-hardware list" lists vegyraupe and qi-bot as having special access rights01:47
kristianpaulha wpwrak is the owner01:48
kristianpaul /msg qi-bot  op _password_01:50
wolfspraulyou mean I should type that?01:51
kristianpaulso that should give you op, so you can do more01:51
kristianpaulacordinf to http://www.egghelp.org/commands/irc.htm#msg_op01:51
wolfspraulnothing happened01:52
wpwrakman vs. machine, 0:3 ;-)01:53
mthwolfspraul: I guess you could ask freenode staff to add you to the access list for gaining op privileges01:54
mthI don't know how they check the validity of such requests though01:54
kristianpaulthats not the best option..01:54
wolfspraulno no, that's not needed01:54
wolfspraulwhen vegyraupe left we cleaned this all up01:54
mthwell, it is useful to have someone who is regularly on the channel that can become op01:54
wolfspraulthe channel is registered, and qi-bot is one of the founders etc.01:54
wolfspraulso everything is rooted by qi-bot, and qi-bot is 100% under control of the turandot.qi-hardware.com server01:55
wolfspraulnow... the only problem is how cryptic this all is, with irc messages, freenode ChanServ, eggdrop qi-bot on turandot etc.01:55
wolfspraulhundreds of commands, flags, modes, etc.01:55
mthah, so it is possible to become op via qi-bot, but no-one knows how?01:55
wolfspraulI have no overview.01:55
kristianpaulwe can kill qi-bot and suplantate it for a moment01:56
wolfspraulI tried logging into the server and accessing the eggdrop console, but it asks me for a username/password I don't know01:56
wolfspraulwho is causing the problem to DocScrutinizer now actually, ChanServ or qi-bot?01:57
kristianpaulChanServ i think01:57
kristianpaulqi-bot: just executed..01:57
wpwrakphew. at least the rest of us can talk again :)02:11
wolfspraulbut DocScrutinizer seems to have been kicked again for 'nick flood'02:11
wpwraknow .. how to disable the bot's vengeful memory ? ;-))02:11
mthit seems qi-bot does enforce the ban after all...02:12
kristianpaulmth: yeah..02:12
wpwrakthe bot may remember. maybe to prevent someone from slipping through the ban if the corresponding enforcer drops off the channel02:12
kristianpaulban time is 12002:16
wolfspraul120 what?02:18
wolfspraul120 minutes?02:18
kristianpauli guess02:18
wolfspraulinteresting - nvidia bought icerasemi.com, makers of baseband chipsets02:18
wolfspraulif the bot remembers, I am sure we can cause a little amnesia to the bot. the question is what is faster - just wait or find out on the server where/how the bot remembers?02:19
wpwrakso, no Sarah Connor award for you ?02:22
wpwrakseems that DocScrutinizer went to bed. so i guess the wait-for-timeout won't hurt. well, maybe it'll disturb his dreams02:22
DocScrutinizersorry for nickflooding02:29
wpwrakwhee, the return of DocScrutinizer !02:34
wpwrakand the Sarah Connor award for wolfspraul ! ;-)02:35
larscthe terminator wants to kill him?02:35
wpwrakhe boldly resisted the cruel and inhumane rule of the machines02:42
DocScrutinizeryou should add some entries to your access list ;-D02:46
wpwrakyeah, e.g., give channel op all those who have root on the server hosting qi-bot. that would be very natural mapping ;-)02:47
DocScrutinizerchanserv role MANAGER seems ok (and also is default for new access list entries afaik)02:50
DocScrutinizernah, actually chanop is default02:50
DocScrutinizerMANAGER has +f02:51
qi-bot[commit] Werner Almesberger: BOOKSHELF (dfu): added DFU specification http://qi-hw.com/p/ben-wpan/487011f02:53
qi-bot[commit] Werner Almesberger: atusb/fw/README: boot loader flashing time was a bit too optimistic (15 -> 30 s) http://qi-hw.com/p/ben-wpan/6179f3102:53
qi-bot[commit] Werner Almesberger: atusb/fw: implemented ATUSB_RESET http://qi-hw.com/p/ben-wpan/42dae8d02:53
qi-bot[commit] Werner Almesberger: tools/Makefile.common (CFLAGS): added -g http://qi-hw.com/p/ben-wpan/27c1c7b02:53
rjeffriesanother mention of wpwrak 's Most Excllent work:02:58
wpwrakkewl. let the masses roll in ! ;-)03:01
DocScrutinizer"hidden channel (doesn't show up in channel list)" - is that intentional?03:03
wpwrakmaybe that's why we got fewer "which nVidia card should i buy ?" type of visitors lately03:04
DocScrutinizerthe /topic invites this03:04
wolfspraulthat's not intentional03:06
wolfspraul'hidden' in which list?03:07
wolfspraulit's not hidden on the qi-hardware.com homepage :-)03:07
DocScrutinizerhidden on freenode's IRQ channel list03:23
DocScrutinizervirtually each client has a way to show these lists03:24
DocScrutinizerbut on my list it shows up.03:26
DocScrutinizermaybe because I am a user of this chan03:26
DocScrutinizeror Konversation is wrong on interpreting the channel flags03:27
DocScrutinizercould someone search for a chan tsthid please03:29
DocScrutinizerin his channel list03:29
DocScrutinizer+s (secret channel)   This channel will not appear on channel lists or WHO or WHOIS output unless you are on it.03:33
DocScrutinizer*** Kanalmodi: +cgnprst03:33
xiangfusomething wrong with the #qi-hardware07:45
xiangfu[15:45] == #qi-hardware Cannot join channel (+r) - you need to be identified with services07:46
xiangfuaw, cannot join the #qi-hardware07:46
xiangfusame with [http://webchat.freenode.net/?randomnick=1&channels=qi-hardware]07:46
xiangfucan not join #qi-hardware. :(07:47
kyakxiangfu: someone set the +r mode, only registered nicks can join07:51
kyaki'd say, +s is enough to hide this channel from unrelated users :)07:52
kyak+s = hide from the channel list07:54
kyakwell, if it even bothers someone to kick out ocassional users who saw "hardware" in the channels name :)07:55
kyakthey are usually all right and don't stumle in here very often07:56
-:#qi-hardware- [freenode-info] channel flooding and no channel staff around to help? Please check with freenode support: http://freenode.net/faq.shtml#gettinghelp08:01
kyakwolfspraul: we should have regular ops among users here, and preferrably one or two persons taking care of qi-bot :)08:02
wolfspraulkyak: feel free to give yourself whatever super power you need08:03
kyakwolfspraul: the problem xiangfu sees is coming from "+r" mode, whoever set it for whatever reason08:03
wolfspraullet's remove it08:03
wolfspraulmy problem is that there is really a _myriad_ of options in IRC land08:03
wolfsprauland then helpful 'bots' whether they are called ChanServ, or qi-bot, or whatever adding even more on top08:04
kyakwolfspraul: do you have a password for bot's console or i may as well reset it?08:04
xiangfuI am also trying to login qi-bot :)08:05
wolfspraulsomeone this morning said the #qi-hardware channel was 'hidden'08:05
wolfspraulwell, maybe it was (don't know where), and maybe now it's not, but as a side-effect it's now invite-only?08:05
kyakit's not invite only right now, only registered (with nickserv) nicks can join08:06
xiangfuand also it's 'hidden' now08:07
kyakso, what do we want to do?08:09
kyakchange the nick-fllod limit08:09
wolfspraulI don't even know what that exactly is/means08:10
wolfspraulsince DocScrutinizer seems to have accidentally run into it, sure, let's raise it08:11
wolfspraulbut I don't know what it means :-)08:11
wolfspraulxiangfu: was it hidden before?08:11
xiangfuwolfspraul: no08:11
kyakthat's exactly what DocScrutinizer ran into08:11
wolfspraulit never occured to me that the channel was hidden, but now it indeed does look hidden to me as well :-)08:11
kyaklimit for nick changes/per time08:11
wolfspraulkyak: it seems we went to a higher hidden level08:11
kyakdo you want it to be hidden from channel list?08:11
wolfspraulof course not08:11
wolfsprauland I don't think it was, but maybe it became somehow?08:12
wolfspraulwhy should it be hidden?08:12
wolfspraulsure let's increase the nick flood08:12
wolfspraulcurrent limit is 5 changes in 60 seconds?08:14
wolfspraulI'm wondering how DocScrutinizer managed to run into that? :-)08:14
kyaknick flood threshold changed from 5:60 to 10:60 (10 changes in 60 seconds are allowed )08:14
wolfspraulmaybe increase to 10:60 ?08:14
xiangfuwe also needs '-r'08:15
kyakoh yes08:15
kyak+c stands for "no color"08:16
kyakdo we need it?08:16
xiangfui'd vote 'kyak' as op, for co-work with qi-bot :)08:17
kyak+p is for not seeing the channel in "whois" output08:17
kyakalso removed it..08:17
wolfspraulno color?08:18
wolfspraulno idea, I don't know what it means practically08:18
wolfspraulcan we remember these things in /home/eggdrop/eggdrop.conf ?08:18
kyak4like this08:18
wolfsprauloh - red!08:19
kyakwolfspraul: not sure if your client can handle color though :)08:19
wolfspraulit can - irssi with ratpoison window manager :-)08:19
kyakwolfspraul: yeah, this things i'm doing now are remembered in channel file, but can also be inserted inside the config file08:19
wolfspraulmaybe in the config file is better?08:19
wolfspraulput it where it's most sticky...08:20
kyaksure, i'll move it to config when we're done :)08:20
xiangfukyak: so we have to modify those channel add #qi-hardware {08:32
xiangfu    chanmode "+nt-likm"08:32
xiangfuthen restart the eggdrop , should be ok, right?08:32
kyakxiangfu: yep.. and some other.. i'll handle that08:33
xiangfuthanks. 08:33
wolfspraulkyak: oh, I didn't want to have op rights08:35
wolfspraulthe reason is that I'm worried I will accidentally turn on some bad big switch08:35
wolfspraulso in general I keep all my accounts at the privileges setting that I regularly and actually need08:35
wolfspraulso I feel good and relaxed using my accounts, not worried about a rm -rf / mistake08:36
kyakwolfspraul:  :)08:36
wolfsprauljust peace for my mind, means I can still work with a beer or two :-)08:36
wolfspraulif I would max out the privileges of my accounts everywhere, just for whatever sake, I'd go nuts08:36
wolfspraulkewl, thanks. back to normal happy user :-)08:37
wpwrakwolfspraul: DocScrutinizer ran into the limit because he did something he shouldn't have done in the first place. he keeps a lot of nicks (variations of Doc*), and he iterated through all of them, to "refresh" them with NickServ (an unused nick expires after some 10 weeks).09:10
kyaki think he should use "nick groups" (as seen in "/ns help group") for the nicks not to expire09:11
wpwrakwolfspraul: nothing wrong with this per se, but he shouldn't have done this while in a channel (kinda rude to flood all the people with a lot of nick changes)09:12
wolfspraulI see.09:13
wolfspraulwell it seems kyak has gotten everything back under control now :-)09:13
wolfspraulorder is restored09:13
wpwrakkyak: "Please note that grouped nicks expire separately from accounts" does this really help ?09:13
wpwrakyeah, at least we found that we had a power vacuum and that the evil machines had started to creep into it ;-)09:13
kyakwpwrak: hehe :)09:14
kyaki just realized nf_conntrack_irc won't work for SSL connections09:21
kyaki guess i'll assign a range of ports manually for DCC then09:22
dvdkhey kyak: did you know that you already lost 10 months of fulltime work to nanonote development?  https://www.ohloh.net/p/nanonote-software/contributors/2488364465258550  :)09:38
kyakdvdk: wow, impressing! how does it count?09:40
dvdkkyak: dunno.  09:40
Action: dvdk is rated at 11 person-months :)09:40
dvdkkyak: maybe lines of codes in all commits09:40
kyakit sounds like 10 calendar months09:40
kyakstarting from July 201009:40
kyakbut somehow it's counted as person :)09:40
lekernelsloccoung ?09:40
dvdkthe total cost of the project is (over)estimated here: https://www.ohloh.net/p/nanonote-software/09:41
dvdk213 person-years09:41
dvdk11 million us$09:41
dvdk(but counts openwrt upstream, too)09:41
Action: lekernel is rated more than half his age by sloccount09:42
dvdklekernel: better than twice your age :)09:42
lekernelwell, some code that I wrote in the floppy disk era isn't counted09:42
kyakdvdk: why are yo udoing it? :) is it just for statistics?09:42
Action: wpwrak should have kept those tapes and floppies ...09:43
dvdkkyak: just for fun, ohloh.net is like facebook for open-source projects09:43
wolfspraul11 million USD!09:43
wolfspraulwhere can we cash them out?09:43
dvdkkyak: you can see who is contributing where, with timelines etc.09:43
wolfspraulcan someone email ohloh?09:43
dvdkkyak: even nicer commit log than on qi-hardware.com09:43
dvdkwolfspraul: :)09:44
wpwrakmay be useful when talking to investors. "here's what we already put into it"09:44
wolfspraulnot sure about that argument.09:44
dvdkwolfspraul: looking for a nice .png icon for the ohloh, maybe the Qi symbol09:44
wolfspraul"look, we already irreversibly converted 11 million USD into electrons, don't you want to add your xxx USD on top as well?"09:44
dvdkwhere should i look?09:44
wolfspraulwe have all sorts of stuff here http://en.qi-hardware.com/wiki/Press_and_marketing_material09:45
wolfspraulother than that - whereever you see it, just right-click save-as :-) images.google.com etc.09:45
dvdkwolfspraul: right-click didn't work on the main page.  baad java script?09:46
dvdkgoing to use Nanonote.cc-logo-symbol-slide-vignet.svg09:48
wpwrakwolfspraul: "we have a community that has already donated 11M. don't you think this is a great force multiplier for your investment ?"09:48
dvdkwolfspraul: careful with the numbers, not sure whether ohloh only counts our branch, or includes upstream openwrt, too09:49
dvdknow with icon https://www.ohloh.net/p/nanonote-software09:52
wpwrakdvdk: are you suggesting our work is worth less than 11M ? ;-)09:52
kyakdvdk: it sure includes upstream openwrt with nbd wasting 73 man-months :)09:52
kyakthat's a bad factoid :)09:54
wpwrakmaybe i should convert all the huge files from the 3d scans to a format that looks like source, e.g., xwd, to get them counted as well. that should nicely boost the value ;-)09:55
wpwrakthe absence of a ya making itself felt ...09:56
wpwrakit shows on the mailing list and irc activity as well. the project is definitely slowing down.09:56
wpwrakheise just picked up the $25 computer. they're a bit slow. let's see if they have a boring enough day that they'd even scrape the bottom for things like ubb-vga :)10:15
lekernelsimply everyone picks up that junk...10:16
lekernelI wish the same happened for our stories, but for some reason it's just not the case10:16
kyakcause it sounds nice.. "a 25$ computer"10:16
kyakwho cares about details10:16
lekernelmaybe that's one factor...10:16
wpwrakkyak: yeah, while huge megacorps are struggling for a decade with the $100 computer, some clever hacker does the $25 computer in his basement. it's hard to argue with that ;-)10:17
wpwraklekernel: (our stories) let's be honest: the whole nanonote branch doesn't produce much in terms of stories these days. there are certainly people out there who never heard of it and would be interested, but you can't just keep on broadcasting the same message over and over until you're sure everyone has heard it.10:19
kyakwpwrak: plus, someone mentioned correctly in ML that the article is heavily exploiting the charity issue10:20
wolfspraulit's not junk, I think the story is good10:20
wpwraklekernel: for mm1, there's still the small issue that even we don't seem to have a clear marketing message for it :)10:20
lekernelmaybe we should cry that the semiconductor companies are evil. even if it's a mere lie, if that's the price to pay to get stories picked up ...10:21
wolfspraulI don't care at all about that 25 USD usb stick computer announcement, but the story is good imo10:21
wolfsprauldon't be negative unless you have to10:21
wpwrakkyak: charity is always good :) the nanonote could be a "$100 laptop for SMALL children" ;-)10:21
kyakhow about flooding Africa with charitable nanonotes? :)10:21
wpwraklekernel: get yourself sued by apple, then you have your story ;-)10:21
wolfspraulthe 'm1 shown to lgm audience' story is nice, I'm not sure we get enough juicy details for a release though, and a well written text10:22
wolfspraulwe know in a few days, last chance to release it is probably Thursday or so, hopefully then with some jaw-dropping details about reaction of the audience10:22
wolfspraul"people were stunned"10:23
wpwrakkyak: hmm, wolfgang could donate a few hundred to some charitable project ...10:23
kyakand only hope it would pay off10:24
wpwrak"mm1 at lgm" would probably become just "lgm" with a list of things happening there. but i think that's okay. gives a bit fewer visitors, but they may be better quality.10:24
wpwrakkyak: the real cost depends on the estimate of how many of the bens in stock will turn sour there anyway. the current demand seems fairly low.10:25
wpwrakkyak: but of course, you can play that card only 1-2 times. the number of units would have to be "significant", i.e., >> 100, even for a small operation like sharism.10:27
wpwrakwolfspraul: ah, how did things progress on the order front in the end ? you mentioned that a few came in on saturday. how did sunday and monday go ?10:28
wolfspraulI think about 6 on the first day, 4 on the second day, then zero for 2 days I think10:30
wolfspraulsome more with David for sure10:30
wpwrakkyak: a problem with charity is that you can't just airdrop a container of stuff and be done with it. things need someone to make them usable locally. so you need to find them, get them interested, and at least get them to think they can use it.10:30
wolfspraulpretty good I think, considering the context and all10:30
wpwrakphew. a lot of work for the marketing department per unit :)10:31
wolfsprauloh sure, all investment into the future10:31
wpwrakbut yes, the context is weakish. it's more a curiosity notice than a "go and buy now" item10:31
wolfspraulkyak: do you have a proposal for the headline of a milkymist story? just anything on your mind...10:32
kyakwpwrak: yeah, i already imagine how hard would it be to explain these poor kids how to run "MS Word" on Ben :")10:32
wpwrak(future) yeah, spreading the word. the next time, people will have heard of the project, etc.10:32
kyakwolfspraul: i'm not sure.. was there some party at lgm involving mm?10:33
wpwrakkyak: find a place where they're not brainwashed enough yet ;-) maybe some pygmies for whom anything larger would be too big anyway. they'll be genuinely grateful :)10:33
wolfspraulyes, but think completely free10:33
wolfspraulwhat always works is "milkymist one on final space shuttle mission"10:34
wolfspraul"VJ killed by cargo load of m1 boxes at big music event"10:34
wpwrak"mm1 suspected in shuttle explosion"10:35
wpwrak"[...] unnamed sources have confirmed prior concerns about the power input protection circuit [...]"10:36
wolfspraulyou can also do this whole 'leak' nonsense10:36
wpwrak"mm1 effect makes you look like elvis"10:36
wolfspraulmaybe the leak was saying that Intel is secretly investing in the Milkymist SoC10:36
wolfspraulnah, I hate this stuff10:36
wolfspraulbetter something with people10:37
wolfspraulthat's why I like the m1@lgm story10:37
wolfspraulpeople people people10:37
wpwrak"secret annex to 1st testament found in mm1 bitstream"10:37
wpwrak"hardware-accelerated vj syntesizer cracks encryption of secret video of osama assassination"10:39
wolfspraulI think we can extract enough stories out of the video synthesizer as a product. what you can do with it, why experimenting with it is cool, etc.10:39
kyak"Milkymist: the only One survived the party"10:40
wolfspraulpost-pc era (although I would avoid that term)10:40
wolfspraulit can even be a text about video synthesis that only later gets to the sales point of m110:40
wpwraktelling the world that you're the harbinger of a new era has always been an effective means to gather people for your cause ;-)10:41
wpwrakyeah. some clever video synthesis hack can work quite nicely. establish the mm1 as the video lab workbench of choice. also works with the ce angle, cf. those PS3 supercomputers10:42
wolfspraulwell right now I focus on my lgm story. either I can get that complete and interesting and released, or I failed. the devil is in the detail, as usual.10:42
wolfspraulyes! [clever video synthesis hack]10:42
wpwrakwhere have previous lgm conferences been covered ?10:43
wolfspraulthe hack can be the reason you get the story, m1 just happens to be the tool for the hack10:43
wolfspraulno idea, I don't know how professional that whole lgm thing is10:43
kyakor, "Milkymist One sets the party free"10:44
wolfspraulfirst I need m1 to be used there, I need credible audience reactions, I need a little story (2-3 paragraphs) about how it went there, I need a picture, etc.10:44
wpwraklike the usrp gets traction just because half of all the rf hacks use it somewhere10:44
wolfsprauland I'm on the other side of the world, so now I can only wait10:44
wolfspraulthe draft is done10:44
wpwrakkyak: "mm1 banned from aircraft"10:44
kyakwpwrak: otherwise, pilots are partying all the time :)10:46
wpwrakkyak: "vj activist Sebastien Bourdeauducq arrested for selling supercomputer technology to cuba"10:46
wpwrakgrmbl. it's today. there should have been an announcement on the qi-hw list last week.10:48
wpwrakno link to LGM ?10:49
kyakthe man known as "he, who never sleeps, but only takes cat naps" got his house full of secret agents after triggering a limit of banned words in IRC10:49
kyakoh, i forgot to mentioned mm :)10:49
wpwrakwolfspraul: with the may news probably becoming part of the june news, how about pushing out any public events scheduled within the next 4-5 weeks ? e.g., sebastien has two events in the next four days already !10:52
wolfspraultwo events? don't know10:57
wolfspraulwpwrak: that news is not complete, the date is preliminary10:57
wolfspraulone by one10:57
DocScrutinizerwolfspraul: you are aware that a access list entry not automatically enables you to do "bad things" by accident?10:58
DocScrutinizeractually freenode suggests NOT to use +O (auto-op)10:59
DocScrutinizerfor a bot it is ok, 11:00
DocScrutinizer/msg chanserv help (access)11:00
wpwrakwolfspraul: i mean http://en.qi-hardware.com/wiki/Community_news_2011-05-0111:02
wpwrakwolfspraul: there you have two events listed for sebastien: amsterdam on the 12th, paris on the 14th11:02
DocScrutinizerwolfspraul: I'd feel uncomfortable with control over channel is completely depending on a working instance of a bot not related to the chan infra. And I never seen anybody doing a "typo" like /msg chanserve kickban wolfspraul 11:04
DocScrutinizeronly bots do that ;-P11:05
wpwrakwolfspraul: (news date) yeah. seems that mm1 may get mentioned only on the 13th. maybe it's not too late to post to the list, in case someone is in the area and wants to pop in11:05
wpwrakDocScrutinizer: most people probably just stick needles into their little wolfspraul doll ;-)11:06
wolfspraulwpwrak: oh you mean we should pre-announce m1@lgm? that's a good idea! the thing is - I am not there myself.11:07
wolfspraulso I don't want to make up all sorts of things, and maybe reality is entirely different11:08
wolfspraulI finished my part of the lgm story (the draft), now let's see11:08
DocScrutinizerwpwrak: by "anybody" I meant one of the nicks that can actually do, there are sometimes a lot, e.g /msg chanserv access #maemo list (which has been even longer at times)11:08
wpwrakwolfspraul: we should pre-announce all public events. what's the point of reading a week later that you just missed it ? :-)11:08
wpwrakwolfspraul: just write what rejon has told you he'll do ?11:09
DocScrutinizerwolfspraul: a access list entry is not like being root, it's more like having sudo permission11:09
DocScrutinizerjust don't use O11:09
DocScrutinizergrrr +O11:10
wpwrakwolfspraul: i'm not sure if he'll actually talk about the mm1 / include it in a talk. but i think he'll have one to demo it on some occasions, no ?11:10
wolfspraulI am not sure what will really happen, and it is just plain impossible/unthinkable for me that I write this 'story' completely like a novel, like how it could be.11:11
wolfspraulif we cannot make it complete, we have a draft for next time...11:11
wolfspraulI need the text of what actually happened, audience reaction (quote), picture11:11
wolfspraulnone of which I can provide, no matter how long I stare at my keyboard & screen in Beijing11:12
wolfspraulI could write out the text a bit more, talk about m1 availability etc. but that's hard as long as the main meat is missing, because it would follow after that11:12
wolfspraulso we see11:13
wolfspraultomorrow we know more11:13
wolfspraulJon showed up in #milkymist today asking how to update his m1 to the latest software11:13
wolfspraulthat's a good sign11:13
wpwrakno no, not a novel. an announcement. "rejon will attend LGM and have a MM1 with him"11:13
wolfspraultoo thin imo11:14
wpwrakor if there's a talk, "rejon will show the MM1 in his talk at LGM, Friday May 13, 17:00"11:14
wolfspraulthe idea was to use it in between talks11:14
wolfspraulwhich could be really cool11:14
wolfspraulat that time people would be free to walk up to it, the camera (which he has) could point to them, he could play some nice CC music11:14
wpwrakalso good. if you're curious about MM1 and you're near montreal, you may decide to drop in11:14
wolfspraulsebastien tried to rush the twitter feature into it, not sure whether it made it or not but that would be great too11:15
wolfspraulif Jon can get the twitter feature onto his m1 and working, people sitting in the audience can post/project messages11:15
wolfspraulall nice ideas11:15
wolfspraulbut now we need execution :-)11:15
lekernelI wonder what version Jon is running11:16
lekernelI fear 0.111:16
lekernelargh ....11:16
wpwrakwolfspraul: (twitter) you can leave this to him ;-)11:16
wolfspraulwpwrak: yes but with those details you see that I can totally not fake the actual text11:17
wpwraklekernel: your events on the 12th and 14th are firm ?11:17
wolfspraulbecause I simply don't know between horrible/nothing and spectacularly attractive, where it will be11:17
wpwrakwolfspraul: well, look at sebastien's entries at http://en.qi-hardware.com/wiki/Community_news_2011-05-0111:18
wolfspraulyes we will get some bitcoin client or so onto the Ben, goes well in line with the wallet idea11:18
wpwrakwolfspraul: they don't say all that much. you can do the same about LGM.11:18
lekernelwe should sell stuff on bitcoin :-)11:18
wpwrakthen ben could calculate bitcoins while in the pocket. a self-filling wallet ;-)11:19
wolfspraulkyak: you there? DocScrutinizer was asking about some certain powers to be bestowed upon him (in #qi-hardware)11:19
lekernelforbes headline "Want to buy socks or psilocybin without being traced? Bitcoins are for you."11:19
lekernelnot good in fact. now governments have a reason to stop bitcoin ...11:19
wpwraklekernel: seems that they want governments to stamp on bitcoins :)11:19
wolfspraulbitcoin seems like a great experiment to me, right now is the testing phase whether the software is bug-free :-)11:20
kyakDocScrutinizer: is it what you want?11:20
wpwraki somehow don't think they missed that angle when they wrote it11:20
DocScrutinizerkyak: I offered wolfspraul to add me on chnnel's access list11:20
wpwrak(the article)11:20
lekernelit could work e.g. for small milkymist goodies, like the jtag/serial adapter11:20
DocScrutinizerfor a fallback when qi-bot is down or whatever11:20
wolfspraulI don't even know what/where the 'access list' is, so if kyak knows and can do it - great.11:21
wolfspraulI will first get bitcoin software onto the Ben11:21
kyakDocScrutinizer: it's done :)11:21
wolfspraulso people can have a balance there, and can make transactions11:21
wolfspraulthen we can use it so people can reward creators of free works, very easily11:21
DocScrutinizerwolfspraul: /msg chanserv access #qi-hardware list11:21
wolfspraulwe could even sell a Ben with some bitcoins preloaded11:21
wolfspraulall hypothetical, this stuff is early but I think it's an interesting project11:22
DocScrutinizerkyak: ok, I'm a lazy chanop on this chan, so ping me if sth needs my anttentian, please11:22
kyakDocScrutinizer: sure :)11:22
wpwrakwolfspraul: how about "Jon Phillips will have a MM1 at LGM and plans to demo it between talks" ? that's about as non-committal as you can get :)11:22
wpwrakwolfspraul: add the two events from sebastien, plus an excuse that you're too overloaded to do a proper community news release this month, and you're done11:23
-ChanServ:#qi-hardware- kyak enabled the VERBOSE flag11:25
-ChanServ:#qi-hardware- kyak disabled the VERBOSE flag11:25
wpwrakstrace eggdrop | post-to-irc11:25
DocScrutinizerwpwrak: ??11:26
wpwrakDocScrutinizer: the "verbose flag" bit sounded a little scary :)11:27
DocScrutinizerit's been chanserv flag, so while I'm unaware of its particular meaning, I think it's not THAT bad11:27
DocScrutinizer/cs help flags11:28
DocScrutinizer -ChanServ- VERBOSE         Notifies channel about access list modifications.11:29
lunavorax_miniHi all11:29
DocScrutinizer/cs help set11:29
kyakwpwrak: just playing around with freenode's services :) they are a bit different from what i was used to11:31
DocScrutinizera completely safe and resonable thing to have, to notice when anybody is messing with access lists11:31
lunavorax_miniDid anyone read/heard about the Rasberry Pi ?11:31
DocScrutinizerin fact I'd keep it11:31
lekernel{#~#{[§ if this were #milkymist you'd be kicked11:32
wolfspraullunavorax_mini: what do you think about the project?11:32
wolfspraulRasberry Pi I mean11:32
lunavorax_miniI'm very sceptical.11:32
DocScrutinizerkyak: I suggest you keep VERBOSE for chanserv11:32
wolfspraullekernel: see :-)11:32
wolfspraulyou spoke too soon11:32
kyakDocScrutinizer: sure, let's keep it11:32
wolfspraullunavorax_mini: now you and lekernel can be eternal friends...11:32
-ChanServ:#qi-hardware- kyak enabled the VERBOSE flag11:33
wolfspraulis that thing even open/copyleft hardware?11:33
wpwraklekernel: you'd make a good rapper with all that hate ;-)11:33
wolfspraulI didn't look into the details11:33
lekernelthere are no details, just an empty page11:33
wolfspraulwill they publish the schematics, layout, bom?11:33
lunavorax_miniwolfspraul, I'm sceptical mostly because it looks like a "toy for the rich" and nothing else. I don't think the Raspberry can achieve it's "education" goal.11:34
wpwraklekernel: how about an anti-arduino, anti-rasberry rap video, lyrics by lekernel, video my milkymist ? ;-)11:34
lekernelI'm not anti arduino11:34
-ChanServ:#qi-hardware- DocScrutinizer (jOERG_rw) set flags +V on kyak.11:34
-ChanServ:#qi-hardware- DocScrutinizer (jOERG_rw) set flags -V on kyak.11:34
wolfspraullunavorax_mini: I would of course agree with you on that.11:34
DocScrutinizerok, works (like that )11:34
wolfspraulthe best thing for poor people right now is probably a cheap android smartphone11:34
wolfspraulonce you have that, and it may only cost 50 USD or so, then basically you are first limited by lack of knowledge/understanding, not by lack of technology11:35
lunavorax_miniwolfspraul, indeed, but still android phones aren't that open to understand programming and stuff.11:35
wolfspraulso you can grow a lot personally with that thing, if you have the motivation to do so11:35
wolfspraulwell yes, but I'm just saying from the practical perspective of a poor 12-year old in the Philippines, Pakistan, Indonesia, China, etc.11:36
wolfspraulif that's the only thing I can afford for my kid, I'll get him/her a cheap Android phone11:36
lunavorax_miniYes wolfspraul and I agree with you.11:36
wolfspraulnot an OLPC, not a NanoNote, and not a (only announced even) Raspberry Pi11:36
lunavorax_miniI have been wondering once about a Nanonote based OLPC ;P11:37
lunavorax_miniSugarUI is heavily python based iirc ?11:37
wpwrakwolfspraul: or make something really cheap, like that $25 device, but design it with composite video out. something you can connect to a TV, including analog TV. you can find TVs even in the most impoverished places.11:37
wolfspraulyou want to run OLPC software o NanoNote?11:38
wpwraklekernel: and also USB keyboards ;-)11:38
lunavorax_miniwpwrak, totally agree11:38
wolfspraulwpwrak: it's not mobile.11:38
lekernelwpwrak: the problem is, OMAP and many other similar critters don't have composite out :-)11:38
wolfspraulpoor people have to walk a lot, travel a lot, they are kicked around a lot11:38
wpwrakwolfspraul: doesn't have to be mobile. makes it cheaper.11:38
lunavorax_miniNope wolfspraul I was thinking about using both concept of the OLPC and hardware from the nanonote.11:38
wolfspraulI've seen kids in Vietnam with amazing daily schedules, and the only chance they have for learning is on some crazy walk or bus ride somewhere between paid jobs and work at home.11:39
wolfspraulso once you dive into the practical real life, you realize how important it is that the battery lasts long, and the device is light (100-200g at most), and rugged/robust11:39
lekernelwpwrak: but sure, for charity, composite out wouldn't look totally stupid like the touted 1080p HDMI11:39
wpwrakwolfspraul: okay, ADS kids may depend on mobility :)11:39
wolfspraulconnecting to a TV sounds wrong to me11:39
lunavorax_minilekernel, :) I feel less alone11:39
wolfspraulyes but it's not good enough. A cheap Android phone still beats it hands down.11:40
lekerneloh, certainly11:40
lunavorax_miniIf it's for learning, composite should be ok, as well as vga.11:40
wpwraklekernel: yeah, HDMI is a bit over the top. nice to have if it comes without much extra cost, but it shouldn't be the entry bar as well.11:40
wpwrakwolfspraul: anyway, regarding the events, have i convinced you to announce them to the list ? :)11:41
lunavorax_miniHDMI should seriously be left alone, even more when it come to non-capable hardware.11:41
wpwrak(if bens were that hard to sell, we'd still wait for our first customer ...)11:41
wolfspraulwpwrak: no because I totally don't know what it is.11:41
wolfspraulI cannot write a credible text about something I don't understand.11:41
DocScrutinizerwho will have a VGA monitor is his hut in Sahel?11:41
wolfsprauland collecting tiny bits of facts one by one is hugely time consuming.11:41
wolfspraulthat's why editors of publications just copy/paste11:41
wpwrakwolfspraul: is there anything in  "Jon Phillips will have a MM1 at LGM and plans to demo it between talks" you'd disagree with ?11:41
wolfspraulnot because they are cheap, but because there is no other choice11:42
lunavorax_mini(I hate netbooks review when people start asking if it can play 1080p while the screen is 1024x600 :) )11:42
wolfspraulthat's way too thin11:42
wolfspraulthat's no news item, that's more like a cry for help11:42
wpwrakDocScrutinizer: maybe not a monitor, but a TV for sure. also, old monitors may very well be cheap enough that people do/can own them11:42
wolfspraulno I need whoever is somewhere first hand to write up a credible, emotional, well written text about what is going on11:43
wpwrakwolfspraul: it's a summary. one line per event, like sebastien's two events11:43
DocScrutinizernot really, as there's no market for those critters in a poor country11:43
wolfspraulFIRST HAND, that's the key11:43
wpwrakwolfspraul: for the details, you have the link to the event site11:43
wolfspraulI can put some structure around it and shop/push it around to sites11:43
wolfspraulI'm focusing on that lgm thing right now11:43
lekernelwpwrak: are you talking about the mailing list? or LGM PR?11:43
wpwrakwolfspraul: no no, not a press release. i mean a post to the qi-hw list.11:43
DocScrutinizerCVBS or even RF is the key feature. See ZX81 ;-D11:44
wpwraklekernel: just the list. cancel this month's community news but make sure the perishable items are salvaged.11:44
wpwrakDocScrutinizer: RF is expensive ;-))11:44
lunavorax_miniComposite is less headbanging than RF11:44
DocScrutinizernot really11:44
wpwrakDocScrutinizer: consider the scrap value of all the metal of that modulator can :)11:45
DocScrutinizerif you build that RF-modulator directly on board, it's about 1$ BOM11:45
wpwrakmust be several whole cents11:45
wolfspraulman, we also had kristian paul using it at the labsurlab11:45
wolfsprauland guyzmo at the lua workshop in Paris11:46
wpwrakwolfspraul: yes yes, i'm quite aware that events are badly under-reported11:46
wpwrakwolfspraul: but instead of making a cult around it, i propose to fix that problem ;-)11:46
kyakwolfspraul: with milkymist, you are so much focused on offering it to "libre" comminuty. Why not start PR'ing and offering it directly to interested communities of Dj's/Vj's, whatever?11:47
wpwrakwolfspraul: and once you start properly announcing events, more people may give you lines you can actually use without too much guesswork11:47
wpwrakwolfspraul: like lekernel already does11:47
lekernelkyak: I'm doing that in parallel... but even within the libre/engineer/hackers/hardware communities it's already badly under reported11:48
lekernelI'm writing a 6-page article for Elektor atm...11:49
wpwrakwolfspraul: also, events aren't necessarily about a specific talk or such. often, the real value is in a hallway conversation. or meeting like-minded people from the area.11:50
lekerneland those two events are much more about the VJ stuff than hacking11:50
wpwraklekernel: if they allow people to have a look at mm1 in real life and maybe shake hands with its creator, they may be more than satisfied with what they get :)11:51
wolfspraulfor those community news, no worries, I will go through this, clean it all up, and properly push them out11:51
wolfspraulobviously it won't be on 05/0111:51
wpwrakwolfspraul: not unless i get that ubb-timemachine to work ;-)11:52
wolfspraulso anything added to that page will be properly handled, if a little delayed this month (or merged into 06/01)11:52
wolfspraulbut the community news cannot become 'good stories', my only goal so far is to keep a record of what's going on, and to highlight the different people11:52
wpwrakthe events have an expiration date. they can't wait much longer. particularly not lgm, but also lekernel's two events should be announced immediately11:53
wpwrakyes yes, agreed11:53
wpwrakthe community news play a different role than the press releases11:53
wolfspraulit also serves as a good starting point to read through the entire community news backlog http://en.qi-hardware.com/wiki/News11:53
wolfspraulthe community news can only be like a list, one line per item. it is too hard to bring out the emotional part, or small details, from the distance11:54
wolfspraulthose things have to be written first hand, just like a good blog post11:54
wolfspraulthe community news can then link to the blog post, and write a 1-line summary of what happened11:55
wpwrakone role of the community news is to summarize and condense what happened in a period of time. plus, since you're already got people's attention, announce future things they're likely to be interested in.11:55
wpwrakthat's where the event calendar comes in11:55
wolfspraulthe blog post itself can definitely be shopped around as 'news' to whatever site11:55
wolfspraulif the text is written well, they may copy/paste and done11:55
lunavorax_miniI think I have an image to illustratre the idea I had of a "Ben Nanoboard"11:59
lunavorax_miniThe Trim-Slice computer is exactly what I was thinking about.11:59
lunavorax_mini(except for the wireless stuffs and the HDMI out)12:00
wpwraklunavorax_mini: by the way, did you see my explanation why "fewer features" doesn't necessarily mean "less cost" ?12:03
lunavorax_minierr no sorry12:03
lunavorax_miniWas it earlier in that conversation ?12:03
wpwraklunavorax_mini: a few days ago :)12:04
lunavorax_miniDamn I feel stupid :( I can't remember it12:04
wpwrakthe example was: if you wanted to get a ben without the keyboard, that would be less than a regular ben. so you may expect it to be cheaper as well.12:04
wpwrakbut in reality, it would be more expensive to provide it, because someone would have to take an existing ben, open the box, open the ben, remove the keyboard, then close it all up again.12:05
lunavorax_miniHum yeah but I was thinking about a brand new product, not a modified one wpwrak 12:05
wpwrakworse, if you wanted a ben with a smaller RAM chip, it would be even more expensive, because someone would have to change the board, buy the new chips, make a new pcb, and get the board smt'ed.12:06
wpwrakpcb and smt have a fixed setup cost in the order of maybe USD 3000 or more.12:07
wpwrakso that "reduced ben" where you would try to save maybe USD 1 would in truth have a cost of maybe USD 10'000 in the end.12:08
wpwrakthey key here is volume. all costs are composed of one-time costs and per-unit costs. with electronics, per-unit costs are very small but one-time costs are huge.12:09
wpwrakso if you can't make a lot of units, they will be expensive, no matter how limited their functionality.12:10
wpwrakthe ben is only so cheap because the pricing is based on a 3000 units volume. and a large portion of the development and setup cost has been borne by the original dictionary and is already written off.12:11
lunavorax_miniHum ok12:11
lunavorax_miniI don't know if there was a misunderstanding or if I didn't understood something well.12:11
wpwrakmaking a new ben from scratch with a 3000 units quantity would mean a significantly higher price.12:12
lunavorax_miniBut if I understood correctly, you game me the exemple of a product already been made.12:12
lunavorax_miniHum ok12:12
wpwrakyes. a product already made has all the one-time costs covered. you may still pay them off, but you've at least already solved the financing.12:13
DocScrutinizerHAHAHA skype == M$ now12:13
wpwrakjust a basic board would be less expensive to engineer and produce than a complete ben. but it would still cost significant money. (and people's time)12:13
lunavorax_miniNot Skype12:14
wpwrakDocScrutinizer: kewl. i used it about twice in my life so far. i think that's a good number that doesn't really need to change ;-)12:14
lunavorax_miniOk XMPP can do what Skype does, but Skype is so "hype" and "cool" you know :(12:14
DocScrutinizerwpwrak: same here12:14
DocScrutinizerwpwrak: actually I abandoned skype betatester crew when those morons managed to delay linux version with ALSA for a year and then failed on all you can probably fail on when inplementing ALSA12:16
lunavorax_miniI'm lost with the audio server stuff in Linux DocScrutinizer, iirc a lot of people hate pulseaudio12:17
DocScrutinizerI'm one of them :-D12:17
lunavorax_miniThat's why I was suprised to see "Gnome3 is pulseaudio only now"12:17
DocScrutinizerI came to hate everything Lennart Poettering aver touched ;-D12:18
lunavorax_miniIs it me or does Lennart Poettering doesn't like PulseAudio either...??12:19
DocScrutinizerLennart *invented* PukeAudio12:20
lunavorax_miniHahaha PukeAudio12:20
DocScrutinizerand pushed it down the throat of evry distro he could find12:20
zearhey guys12:21
DocScrutinizerby telling lies about ALSA12:21
zearthis guy post a comment into my old nn yt vid, but i don't know how to reply12:22
zearmaybe you guys will know the answer12:22
DocScrutinizeractually by now I seem able to tell a particular system is from / with Lennart P. by the particular way it sucks12:22
zearthe comment about "EMF emission"12:22
lekernelpukeaudio is making 'clicks' and other forms of unwanted digital noises all the time when listening to music on my system12:22
lekernelquite annoying12:22
DocScrutinizerin fact L.P. calims to know everything (better) but evidently has no clue about some of the stuff he likes to mess up 12:24
DocScrutinizerand all of his crap is makimum intrusive and targeted to obsolete proven concepts12:25
DocScrutinizerand quite usually is messing up your carefully tailored systems12:25
zearso, can anybody help that guy from comment of my yt video?12:27
lekernelweird comment indeed... ask what for?12:28
lekernelhe could also just add some shielding himself :)12:28
zearCan EMF be actually dangerous for health in a device powered from a battery? Common sense tells me it can't be possible :)12:30
wpwrakzear: it should be possible to construct a battery-powered device that damages your health :)12:32
wpwrakzear: i mean, we have lasers of a few dozen mW that can hurt your eyes, so maybe a little MASER could do the trick for EMF ;)12:33
zearwpwrak, i mean, one that passes CE/FCC tests ;)12:33
wpwrakzear: oh, i'm sure it can pass all sorts of tests as well. as long as you declare its function properly, i.e., to harm people ;-)12:34
zearwell, yes ;)12:34
zeari guess the only potential risk that guy from the comment could get from a netbook is when he falls asleep with in on his lap and the netbook overheats ;)12:35
zearthis shouldn't be a case with non-x86 devices though12:36
zear*with it12:36
mthmy sister once rented a room from a guy who didn't want wifi because of the radiation... but he did use a cell phone12:45
lunavorax_mininon-x86 doesn't overheat ? è_è12:46
lunavorax_minimth, I would have done the contrary, no phone but wifi12:46
DocScrutinizerthey psychosomatic effects can be devastating12:46
lunavorax_miniThey do have some effects but what can you do to avoid theses waves ? They are everywhere now.12:47
DocScrutinizerluckily there are such things like shielding mats aginst geo-radiation for your bed, and similar things that could easy those folks' lifes12:47
DocScrutinizerin fact there's been no documented case under lab conditions where a test subject could tell when a TX been active and when not12:49
wpwrakDocScrutinizer: a metal shield, applied between head and the rest of the body may also bring relief from this and other ailments. i think the french call it guillotine ;-)12:49
lunavorax_miniDammit I forgot how bloated was Ubuntu, I have tons of packets to remove.12:49
wpwrakDocScrutinizer: did the flickering TX LED give it away ? ;-)12:50
DocScrutinizeryeah indeed, actually the flickering LED alone in a completely empty plastic box can cause headache for those "sensitive to EM rays"12:51
DocScrutinizertoldya, the psychosomatic effects are devastating12:51
mthlunavorax_mini: yes, but it seems people most afraid of radiation also know least about it12:52
DocScrutinizerthat's related12:52
lunavorax_miniOf course mth I understand.12:52
zearlunavorax_mini, none of my arm or mips handhelds have ever overheated just from using them12:55
lunavorax_miniiirc, my gp2x wiz and the ben are getting real warm when using gmu for too long.12:56
lunavorax_miniI'll test it again one day.12:56
zearlunavorax_mini, really? I haven't had wiz in hands for more than 5min, but dingoo or gp2x would never get hot from either playing or listening to music12:57
zearthey get hot only if you leave them on the sun (dingoo especially) ;D12:57
wejpyeah, my ben does not get warm at all, even when listening to music with gmu for hours12:58
wejpsame with gp2x and dingoo12:58
lunavorax_miniThe dingoo has a remarquable battery life for a linux-powered device12:58
wejpyeah the dingoo's battery life is pretty good12:59
wejponly the pandora has a longer battery runtime12:59
zearlunavorax_mini, nope, mine is only a year old and the battery lasts for 30-60min maybe13:00
zearbut that's pretty much what cheap chinese components do ;)13:01
wejp30 minutes?13:01
zearwejp, yea :(13:01
wejpmine is over a year old but still lasts pretty long13:01
zeari've charged it to maximum, then copied some oggs on it and used it with gmu for about 30-40min13:01
zearthen the next day i wanted to listen to music when on my way to uni, but surprise - the battery is fully deplated13:02
lunavorax_minizear really ? I used the dingoo for 3days with a single charge.13:02
lunavorax_miniNot 24h/24 but still.13:02
zearlunavorax_mini, well, mine used to last for days when it was new13:02
zearit degraded quite rapidly though13:02
wejpdid you have the battery discharged for a longer period, zear?13:03
zearwejp, hmm.. actually, yes13:03
zearlike 2-3 weeks13:03
wejpthat can kill li-batteries :(13:03
zearoh, great13:03
wejpi know, it is quite a challenge to keep all the batteries charged at all times :/13:04
zearbut dingoo draws a little current all the time to keep the RTC alive, so unless you charge it once in a while, it is destined to die13:05
wejpyeah :|13:06
wejpon other devices i usually remove the battery13:06
zeari didn't know that, gonna remove the battery from my nn13:11
wejpyeah :)13:12
lunavorax_miniOh ! The nn drains battery even turned off ?13:13
wejpnot very much, but to be safe, better remove the battery when not using it for a longer period13:14
wpwrakwolfspraul: speaking of conferences, i'm not sure if anyone already mentioned that rejon and I should each have a talk at FISL this year.13:15
wpwrakwolfspraul: (porto alegre, brazil, and of june/beginning of july)13:16
wpwraki.e., in mid-winter ;-)13:16
zearlunavorax_mini, it does so to keep the RTC up to date13:17
zearbut with the nn you're lucky you can remove the battery13:17
lunavorax_miniI realise everyday that I always have my nn with me. That's strange.13:17
lunavorax_miniOnly down side is that I lost two feets :/13:18
lunavorax_miniI mean on the nn13:18
zearnn is too precious to me13:18
zearto carry it around13:18
wpwrakno schedule details yet, not have we worked out exactly what we'll talk about. i think rejon's will be about the free hw culture more at a higher level and he'll have mm1. i'd focus more on the nanonote and workflow/tools side.13:18
zeari took it once or twice and i dind't even notice when i lost the rubber feet13:18
zeari carry dingoo everywhere with me, since that's the cheapest of my handhelds ;)13:19
wolfspraulwpwrak: rejon mentioned it to me and it's great. we also need to do at least a decent job on publicity on that one...13:19
wpwrakaye. and before the event :)13:20
wpwrakhmm. tuxbrain is in silent mode once more. i kinda wonder what his testing plan for SMT is ...13:29
DocScrutinizertopic batteries, RTC: I hope Qi moved away from those nasty LiIon backup batteries used in e.g FR and N90013:48
DocScrutinizeruse a supercap, or a dedicated capacitive bupbat13:49
wpwrakor if it has to be a battery, just make it non-rechargeable13:50
wpwrak(and removable)13:50
DocScrutinizeryeah, like CR1405 or what it's called13:51
wpwrakor if there's a bit of space, 203213:51
DocScrutinizerwill last longer than those messy LiIon coin-cels13:51
wpwrakyeah. a battery that's inserted after SMT will not die during SMT, that's already a big plus ;-)13:52
DocScrutinizerjust had to mention it13:52
wpwrakyou're warning the ghosts of 9/11 about the danger of fire ;-)13:53
DocScrutinizeras it's been a bit of experience you only gather with time and running into the trap13:53
DocScrutinizeralmost all FR and most of N900 have broken bupbat13:54
wpwrakyes, you learn from your mistakes. we certainly learned a lot at openmoko, especially in the area of the BUB ;-)13:54
wolfspraulDocScrutinizer: with 'broken' you mean they don't charge anymore?13:56
wolfspraulin which way do they break?13:56
DocScrutinizerthe idiocy with thos LiIon bupbat is they are susceptible to deep discharge (will definitely break) but no way to disconnect them13:56
DocScrutinizeryes, don't hold charge, leak...13:57
wpwrakDocScrutinizer: (will break) if the soldering itself doesn't kill them ... they're also a great tool for lowering your yield. well, if you care to QC them :)13:57
DocScrutinizerso, by all means avoid LiIon technology for soldered bats13:58
wpwrakdo not solder batteries :)13:59
DocScrutinizersupercaps are great and should have a waaay better duability13:59
DocScrutinizersee above linked tayo yuden part13:59
wpwrakdunno about supercap durability. but their capacity is on par with the li-junk, so they can't be any worse14:00
DocScrutinizerthose are actually supercaps build to serve as drop in replacement for the LiIon 14:00
DocScrutinizerwpwrak: at least supercaps won't mind deep discharge14:01
DocScrutinizerand have a ~10^5 higer cycle count14:01
wpwrakall sounds promising. how about the soldering profile ?14:01
DocScrutinizeror maybe 10°314:01
DocScrutinizerwpwrak: not worse than LiIonMg coin14:02
DocScrutinizerwpwrak: not worse than LiIonFe(?) coin14:02
DocScrutinizerhttp://www.yuden.co.jp/us/product/pdf/capacitor01_e.pdf p.714:03
DocScrutinizerNB I'm no sourcing expert, so do the required tests and chatting with manufacturer by your own14:07
DocScrutinizerjust saying "this looks like a way better component"14:08
wpwrakyeah. for no, nobody is planning to add any bats or supercaps anywhere :)14:08
DocScrutinizerk, so it's for the record here, and for future planning14:08
wpwrakduly noted :)14:09
DocScrutinizersomebody told wolfspraul about charge balancing of LiIon in series?14:12
wolfspraulno - go on. what do I need to know?14:13
DocScrutinizeryou posted a few links to photos showing 3 LiIon cells in series14:14
DocScrutinizerthat's pretty dangerous practice14:14
wolfspraulI've since bought the larger version of this beast, and plan to take it apart as well, so more pics coming. probably the same idea inside.14:15
DocScrutinizeras there is always a weaker cell, that gets overcharged/reverse-charged on charging/discharging14:15
wolfsprauleven the voltages make no sense to me, because people plug in 12V to charge, and want to draw 12V as well14:15
DocScrutinizerthat's basically a strongly discouraged over-simplistic design14:16
wolfspraulofficially they say 12.6v on the label for charging (3*4.2), but practically people just plug a 12.0v supply in for charging14:16
wpwrakwolfspraul: why did you get it in the first place ? for halloween ? :)14:16
wolfsprauloh I can imagine that14:16
wolfspraulthis is Chinese engineering excellency at its best14:16
wolfspraulbrutally simple, brutally wrong, but still works somehow, sometimes, a bit14:16
wpwraka trophy then ?14:16
wolfspraulpartially to open and take pics14:17
wolfspraulbecause what we discuss now we can only discuss because I opened one14:17
DocScrutinizergood point14:17
wolfsprauland then I will keep one to use, or show people14:17
wolfspraulthat's all14:17
wolfspraulmy understanding was already that with a 12.0v supply, they would not get fully charged14:18
wpwrakwolfspraul: just make sure you don't end up on any no-fly lists ;-)14:18
wolfspraulmaybe that protects against Joerg's 'weaker cell' argument a little :-)14:18
DocScrutinizerprobably not, but yet one might/*will* get overcharged14:18
wolfsprauland then, on the discharge side, my understanding is that they will slowly loose voltage, if they are at 3.7 you only get 11.1V, not 12.0V14:19
wolfspraulso whatever is on the other side better runs on 11.1, not 1214:19
wolfspraulis that correct?14:19
wolfsprauland in addition, Joerg is telling me now that there is a more systematic problem with the series design, because there will also be a weaker and a stronger cell. and the weaker one will be treated badly.14:20
wolfspraulthat brings the number of things that are wrong with this already to a whole handful, I guess :-)14:21
wolfspraulit does work though, really14:21
wolfspraulyou can charge with 12.0, and I am able to operate a mini ccd camera on it14:22
wolfspraulI was told those cameras will run on anything between 8 and 12v, roughly14:22
wpwrakwolfspraul: it only has to work until you've entered your "all it well" on ebay ;-)14:22
wolfspraulso they cna drive the cells down...14:22
wolfspraulis there a practical danger that it may explode?14:23
wolfspraulthat'd be the only thing I'm worried about14:23
DocScrutinizeryeah, so two cells @4.2V and one on -0.4V reverse charge - extreme example14:24
wolfspraulXiangfu had a power supply explode on him a week ago or so14:24
wolfspraula real explosion14:24
DocScrutinizersearch channel log for "monster"14:24
wolfspraulhow can two cells be at 4.2 and one at -0.4?14:24
DocScrutinizerI provided a rather scary URL14:24
wolfspraulis that in charging or discharging mode?14:24
wolfsprauland the 3rd cell is completely broken?14:25
DocScrutinizerif the weaker cell already much weaker than the other both14:25
DocScrutinizerafter that it is14:25
wolfspraulhow is this done in a proper notebook battery? they also have a number of li-ion cells in those14:26
DocScrutinizerthe problem is this is a self-inducing vicious circle, as the weak cell always gets treated worse and thus gets weaker and weaker14:26
wolfsprauljust that they have a small battery controller with some ics and other parts14:26
wolfspraulyes I understood that14:26
DocScrutinizerthey use balancing, basically treating each cell on its own14:26
DocScrutinizerthat's why some battery packs have 7,8, 10 contacts14:27
DocScrutinizersimpler designs have bypass FETs14:27
DocScrutinizerto keep all cells on same voltage14:28
wpwrakwolfspraul: (xiangfu) so he has a faster PC now ? :)14:28
DocScrutinizerso the bypass FET shorts a cell with higher voltage than the weakest one14:29
wolfspraulI see14:30
DocScrutinizersuboptimal but works14:30
wolfspraulseems the monster explosion was also caused by the metal tube acting as a pressure build-up14:30
wolfspraullike a real bomb14:30
wolfspraulat least those 3 cells are only packed in a thin cardboard plus blue tape14:30
wolfspraulso it can't be that bad :-)14:30
DocScrutinizerthe cells are built semi-safe with a vent14:30
DocScrutinizerdoesn't help when you put them inside a pipe-bomb14:31
wolfspraulyes but I mean the wrap around the 3 cells is just thin paper14:31
wolfspraulI'm just thinking worst case here.14:31
DocScrutinizerworst case is your pack is catching fire14:32
wpwrakthat is of course until you decide to put them in a nice solid travel box ...14:32
wolfspraulyes, already thinking about that14:32
wolfspraulI have this nice metal case, fairly airtight I think14:33
wolfspraulI wouldn't want to test whether it can hold or vent the pressure from this kind of event inside...14:33
wpwrakjust let us know in advance what flights you take, and we'll all be safe ;-)14:33
DocScrutinizerconsider LiIon cells worth teir weight in loose blackpowder14:33
wolfspraulok I hold it in my hand when flying14:34
wolfspraulthat way the pressure/fire can directly go into the air14:34
DocScrutinizernot going to happen ;-P14:34
wolfsprauljust kidding14:34
DocScrutinizeryou'll either hold it OR fly14:34
wolfspraulisn't it fun to cut open this kind of little stuff you can buy on the street here :-)14:34
wpwrakat some companies, they have a policy that important employees shouldn't travel on the same flight. with wolfgang, the same policy would make sense, although for a very different reason ;-)14:34
wolfspraulunforgettable impressions14:34
wolfspraulDocScrutinizer: this is what it looks on the outside, in case you didn't see that yet http://www.dinodirect.com/rechargeable-lithium-battery-12v-4800ma-cctv-camera-d-12480.html14:35
wolfspraulanother model but surely the same idea inside14:35
wolfspraul8 reviews (4.5/5), 100 diggs14:36
wolfspraulseems to be selling :-)14:36
wpwraklife expectancy is overrated anyway ;-)14:51
rohmy advice about batteries: dont sell em if possible. let others do THAT dirtywork. adapt some (any) standard as it comes along. make it work. dont make batteries in own casings or mechanics. only invites pain.15:02
wolfspraulsure I totally agree. AA or AAA standard sizes still look like very good solutions to me.15:03
rohmmmh.. also something like using nokia batteries is fine. (rechargeables)15:04
wolfspraulthe industry can focus on optimizing them, you can buy them on every street corner of the world, they can be rechargeable15:04
wolfspraulI think even those sizes will disappear fast over the next few years15:04
wolfspraulthe trend to fully integrated batteries is very strong for phones, I think unstoppable15:04
rohthere are also some with more than one cell .. maybe look at the camera market15:04
rohAA and AAA is mostly bad becuase there are no good rechargeable ones and they eat loads of space (and weight)15:05
wolfspraulwhy are there no good rechargeable ones?15:05
rohwolfspraul: nopp. only apple is THAT stupid. (and their customers).. will come back. anf if only by force via EU-law or so.15:05
wolfspraulok, I will keep an eye on cameras :-) you mean digital still cameras? which brand?15:05
rohthink 'e-waste'15:05
wolfspraulthey changed the laws in the german congress (bundestag) to allow ipads, now they can play games and stuff and it's not so boring15:06
wolfspraulwo ein wille ist ist ein weg...15:06
qi-bot[commit] Xiangfu Liu: disable CONFIG_ALL, add Tile and gottet http://qi-hw.com/p/openwrt-xburst/646a75515:06
qi-bot[commit] kyak: config.full_system: add supertux, terminus-font http://qi-hw.com/p/openwrt-xburst/50f9a1c15:06
qi-bot[commit] kyak: config.full_system: build gcc-mips, make, binutils as modules. Remove http://qi-hw.com/p/openwrt-xburst/3af99c115:06
qi-bot[commit] kyak: config.full_system: remove "=m" packages http://qi-hw.com/p/openwrt-xburst/ee6568b15:06
qi-bot[commit] kyak: don't include ghostscript and fbgs into rootfs, build as modules http://qi-hw.com/p/openwrt-xburst/1c3204f15:06
qi-bot[commit] David Kühling: config.full_system: fix name of gnuplot package (was renamed a long time ago) http://qi-hw.com/p/openwrt-xburst/a88683715:06
qi-bot[commit] Xiangfu Liu: add zgv, MPlayer, brainless http://qi-hw.com/p/openwrt-xburst/c891a5315:06
qi-bot[commit] Xiangfu Liu: remove custum banner, focus on 100% upstream, http://qi-hw.com/p/openwrt-xburst/22c0efa15:06
qi-bot[commit] kyak: config.full_system: add setterm http://qi-hw.com/p/openwrt-xburst/766e01915:06
qi-bot[commit] kyak: config.full_system: include GNU tar http://qi-hw.com/p/openwrt-xburst/890af2e15:06
qi-bot[commit] Lars-Peter Clausen: [package] ks7010: Remove debug printks http://qi-hw.com/p/openwrt-xburst/2c9e6dc15:06
qi-bot[commit] Lars-Peter Clausen: [xburst] jz_mmc: Drop warning about spares irqs. http://qi-hw.com/p/openwrt-xburst/44fe47915:06
qi-bot[commit] Lars-Peter Clausen: [xburst] Improve mounttime. http://qi-hw.com/p/openwrt-xburst/5629baf15:06
qi-bot[commit] kyak: have sound modules built in kernel http://qi-hw.com/p/openwrt-xburst/0cd316915:06
qi-bot[commit] Xiangfu Liu: reflash_ben.sh, new option b k r, reboot device after reflash http://qi-hw.com/p/openwrt-xburst/9ef591515:06
qi-bot[commit] kyak: add sound options to default config, so that it won't popup http://qi-hw.com/p/openwrt-xburst/b33d46b15:06
qi-bot[commit] Xiangfu Liu: config.full_system: add openssh-sftp-server http://qi-hw.com/p/openwrt-xburst/858959415:06
qi-bot[commit] Xiangfu Liu: update opkg.conf, don't using /tmp keep packages information http://qi-hw.com/p/openwrt-xburst/d87c24615:06
qi-bot[commit] kyak: config.full_system: busybox enhancements http://qi-hw.com/p/openwrt-xburst/278a1a315:06
qi-bot[commit] kyak: config.full_system: more busybox options http://qi-hw.com/p/openwrt-xburst/66f199315:06
wolfspraulironically notebooks are still banned, but ipads are allowed now. argh.15:06
rohno specific brand. check voltage*amprs and do the math. price is more a volume thing there.15:07
rohargh. fucking botcrap15:07
wolfspraulI think the trend to fully integrated batteries will accelerate. we see.15:07
wolfspraulit stops after 20 lines15:07
rohwolfspraul: nope- people buy stuff explicitly with removeable ones. and i know loads of these people.15:07
wolfspraulfully integrated batteries may also be more eco friendly, because you have to return the whole device15:07
rohbatterys are much too small to last long enough.15:08
wolfspraulwell let's see15:08
wolfspraulI won't convince you to change your mind, for sure. we just see. I like replacable batteries.15:08
rohand there is nothing eco-friendly on it.  since people throw away working devices that way.15:08
wolfspraulbut there is a mega trend, tsunami size force.15:08
wolfsprauland that trend is towards fully integrated and completely closed devices, for a variety of reasons15:09
rohwolfspraul: you think? i only see apple devices. others try one devcice and abort the idea immedietly after.15:09
wolfspraulI'm talking about phones in particular now, and maybe also tablet devices.15:09
rohthats only 'hype-bullshit' .. look npast that.15:09
wolfspraulthe others are simply not that fast on the engineering side15:09
wolfspraulyou need quite strong vertical integration to pull off a battery integration15:09
rohall the tablet think is also still hype. ther is (and was) a market before. (just it had only bad devices)15:10
qi-bot[commit] Mirko Vogt: add <http://downloads.qi-hardware.com/software/mirror-openwrt-sources/> http://qi-hw.com/p/openwrt-xburst/d3d6c3c15:10
qi-bot[commit] Xiangfu Liu: gitignore vim temporary files (*~) http://qi-hw.com/p/openwrt-xburst/996641d15:10
qi-bot[commit] Xiangfu Liu: optimize for ben nanonote http://qi-hw.com/p/openwrt-xburst/00b686515:10
qi-bot[commit] Xiangfu Liu: [xburst] Improve mounttime http://qi-hw.com/p/openwrt-xburst/4006bc615:10
qi-bot[commit] Xiangfu Liu: nanonote optimize http://qi-hw.com/p/openwrt-xburst/945452815:10
qi-bot[commit] Xiangfu Liu:  Add-gfortran-compiler-support-to-the-toolchain http://qi-hw.com/p/openwrt-xburst/02a7dc815:10
qi-bot[commit] kyak: add kernel patch for setfont2 http://qi-hw.com/p/openwrt-xburst/a972bf115:10
wolfspraulbecause yes, batter integration causes many problems15:10
wolfspraulanother 20 :-)15:10
rohwolfspraul: imho  its a sign of bad ME-vision to make the battery not userreplaceable. shows that you were unable to solve that problem and not willing to cooperate with other companies by not using a(ny) snan[D[D[D[Ddart15:12
rohargh. and i have multiple seconds lag on this ssh session... bbl15:13
kyakxiangfu: did you just catch up with openwrt trunk?15:14
xiangfukyak: working trunk now....15:15
qi-bot[commit] juhosg: ar71xx: ag71xx: add debugfs entry for [rt]x_ring http://qi-hw.com/p/openwrt-xburst/e5061b115:16
qi-bot[commit] hauke: brcm47xx: update patches http://qi-hw.com/p/openwrt-xburst/26205d115:16
qi-bot[commit] jow: [package] base-files: retrigger usb coldplug after module loading, solves usb_modeswitch on boot and possibly others (#9352) http://qi-hw.com/p/openwrt-xburst/6f1485515:16
qi-bot[commit] juhosg: ar71xx: ag71xx: fix build error if debugfs is enabled http://qi-hw.com/p/openwrt-xburst/140888915:16
qi-bot[commit] juhosg: ar71xx: ag71xx: fix section mismatch warnings http://qi-hw.com/p/openwrt-xburst/0511a7a15:16
qi-bot[commit] juhosg: ar71xx: ag71xx: make switch register access atomic http://qi-hw.com/p/openwrt-xburst/bb14a2f15:16
qi-bot[commit] thepeople: Deployment of IPv6 has opened up many more prefixes than just the http://qi-hw.com/p/openwrt-xburst/28e2aae15:16
qi-bot[commit] thepeople: [PATCH] ipv6: restore correct ECN handling on TCP xmit http://qi-hw.com/p/openwrt-xburst/318f8f915:16
qi-bot[commit] juhosg: ar71xx: remove superfluous fifo_cfg overrides http://qi-hw.com/p/openwrt-xburst/5b6a15915:16
qi-bot[commit] juhosg: ar71xx: use phy0tpt trigger for the WLAN LEDs http://qi-hw.com/p/openwrt-xburst/a1c5a3d15:16
qi-bot[commit] juhosg: ar71xx: add uci config for the WLAN LED on the WRT160NL http://qi-hw.com/p/openwrt-xburst/b3a8c8a15:16
qi-bot[commit] juhosg: ar71xx: populate LED3 on the JA76PF board http://qi-hw.com/p/openwrt-xburst/b500b6715:16
qi-bot[commit] blogic: [PATCH v3] (respin) 802.1Q VLAN support for ADM6996M/ADM6996FC http://qi-hw.com/p/openwrt-xburst/10f03e015:16
qi-bot[commit] blogic: [kernel/modules] http://qi-hw.com/p/openwrt-xburst/a37581215:16
qi-bot[commit] Mirko Vogt: add <http://downloads.qi-hardware.com/software/mirror-openwrt-sources/> http://qi-hw.com/p/openwrt-xburst/9a2454c15:16
qi-bot[commit] Xiangfu Liu: optimize for ben nanonote http://qi-hw.com/p/openwrt-xburst/1bbdd7e15:16
qi-bot[commit] Xiangfu Liu: [xburst] Improve mounttime http://qi-hw.com/p/openwrt-xburst/4b7084315:16
qi-bot[commit] Xiangfu Liu: nanonote optimize http://qi-hw.com/p/openwrt-xburst/5971cea15:16
qi-bot[commit] Xiangfu Liu:  Add-gfortran-compiler-support-to-the-toolchain http://qi-hw.com/p/openwrt-xburst/bb1f5f915:16
qi-bot[commit] kyak: add kernel patch for setfont2 http://qi-hw.com/p/openwrt-xburst/ec0ff6015:16
xiangfukyak: now catch up with openwrt trunk15:16
xiangfukyak: I sync both 'master' and 'trunk' of openwrt-xburst.git with upstream.15:17
xiangfukyak: I found one error about 'pango' in buildhost. which is upstream package have updated.15:18
kyakxiangfu: nice15:20
DocScrutinizerdarn, seems I need to write my own IRC client. A lot of nice features I'd like to have are nowhere to be found15:20
xiangfukyak: I hope we will have some images at buildhost. then we can test and release :) (for include dvdk MPlayer works)15:20
kyakxiangfu: (pango) not nice :)15:20
kyakxiangfu: so you plan to make another backfire-based release?15:21
xiangfukyak: I will first check if both 'master' and 'trunk' compile fine in buildhost. 15:21
xiangfukyak: I have cleanup my build script file and add both branches to the cron jobs.15:22
kyakxiangfu: "trunk" won't compile fine without soem patches.. for gcc, for examples15:22
kyakxiangfu: i have some patches, but the question is how to separate openwrt-packages "for trunk" and "for master" :)15:25
xiangfukyak: create a 'trunk' branch in openwrt-package for keep your patches is ok. 15:26
kyakxiangfu: ok15:29
kyakxiangfu: i'll have to read soem git manual for creating remote branches :)15:30
xiangfugit push origin LOCAL_BRANCH_NAME:REMOTE_BRANCH_NAME 15:30
xiangfulike: there is a new branch, 'git checkout -b trunk', then 'git push origin trunk:trunk'15:33
kyakxiangfu: will git push origin & automatically create remote branch if it doesn't exist?15:35
xiangfukyak: yes.15:36
xiangfugit push origin trunk:trunk for example. 15:37
xiangfu'git push origin' will push all local branch to remote when they are have relationship. like :15:37
kyakok, what i'm already on trunk locally, do i still need to indicate it in LOCAL_BRANCH_NAME:?15:38
xiangfubetter do that.15:38
kyakso far, i created the trunk branch, checkout'ed to that branch and commited a change there15:38
kyakso now i'll do git push origin trunk:trunk15:38
xiangfuafter do that. there are some line in your .git/config like: http://pastebin.com/PhNL4PTR15:40
kyaknot like you've shown15:41
kyakxiangfu: well actually, i see my changed commited to trunk15:49
xiangfukyak: yes. 15:49
kyakjust for some reason the history of gcc-mips/Makefile is lost15:50
kyakmaybe i should've pushed the trunk branch without any changes at first15:50
kyakand then push my changes in it15:50
qi-bot[commit] kyak: alsa-lib: no need for this patch in trunk http://qi-hw.com/p/openwrt-packages/09d5d0215:53
xiangfukyak: I will search the 'git branch' about your .git/config tomorrow, have to sleep, see you.15:55
kyakxiangfu: good night :)15:55
qi-bot[commit] kyak: ncursesw is perfectly fine in openwrt trunk, no need to override http://qi-hw.com/p/openwrt-packages/1d1796515:56
qi-bot[commit] kyak: triggersad is now happy again! http://qi-hw.com/p/openwrt-packages/ed4867915:57
qi-bot[commit] kyak: guile: fix compilation in trunk http://qi-hw.com/p/openwrt-packages/ec6b1e716:00
kyak!note xiangfu btw, i never made plplot compile fine in trunk :)16:00
DocScrutinizerbtw (my chanop hat on) what's common notion now about channel status "hidden" (+s) ?16:07
DocScrutinizerI'd cancel that16:08
DocScrutinizeror better: I suggest to cancel that (-s), as it's quite uncommon for a channel that is for larger public to hide in chan-lists16:09
kyakhm, the +s is not set now.. or what do you mean?16:09
DocScrutinizernevermind then16:10
kyakall right :)16:10
wpwrakdoes anyone actually use that channel list ? :)16:29
kristianpaulis too big16:30
wpwrakexactly :)16:30
kyakpeople do you16:37
kyakthose who join here asking random questions about hardware16:37
kyakthey are coming from the channel list16:38
kristianpaulhide us !16:38
kristianpauli tought was from somelse like googling..16:38
wpwrakkyak: yeah, it's probably only those who get here via the list :)16:38
kyakkristianpaul: wolfgang mentioned that you were playing with eggdrop earlier today.. Do you mind joining the company of channel ops? :)16:40
qi-bot[commit] kyak: gnuplot-gfx: fix compilation issue in trunk http://qi-hw.com/p/openwrt-packages/17d05bb16:43
qi-bot[commit] Mirko Vogt: add <http://downloads.qi-hardware.com/software/mirror-openwrt-sources/> http://qi-hw.com/p/openwrt-xburst/d3d6c3c16:47
qi-bot[commit] Xiangfu Liu: gitignore vim temporary files (*~) http://qi-hw.com/p/openwrt-xburst/996641d16:47
qi-bot[commit] Xiangfu Liu: optimize for ben nanonote http://qi-hw.com/p/openwrt-xburst/00b686516:47
qi-bot[commit] Xiangfu Liu: [xburst] Improve mounttime http://qi-hw.com/p/openwrt-xburst/4006bc616:47
qi-bot[commit] Xiangfu Liu: nanonote optimize http://qi-hw.com/p/openwrt-xburst/945452816:47
qi-bot[commit] Xiangfu Liu:  Add-gfortran-compiler-support-to-the-toolchain http://qi-hw.com/p/openwrt-xburst/02a7dc816:47
qi-bot[commit] kyak: add kernel patch for setfont2 http://qi-hw.com/p/openwrt-xburst/a972bf116:47
qi-bot[commit] kyak: Merge branch 'trunk' of projects.qi-hardware.com:openwrt-xburst into trunk http://qi-hw.com/p/openwrt-xburst/08cc62a16:47
qi-bot[commit] kyak: config-2.6.37: enable battery, disable RNDIS http://qi-hw.com/p/openwrt-xburst/9fb09d616:47
qi-bot[commit] kyak: patches-2.6.37: support for Ben NAND partitioning http://qi-hw.com/p/openwrt-xburst/c80e80716:49
kristianpaulkyak: yeah, why not, different time zones is important too16:53
-ChanServ:#qi-hardware- qi-bot set flags +votsriRfA on kristianpaul.16:55
kyakkristianpaul: there we go :)16:55
kristianpaullots of flags i even ignore the function some of then ;)16:59
kyakkristianpaul: no, these are not flags, this is the actual word that has meaning :)17:04
wpwrakif you add enough flags, they eventually assume metaphysical properties. you can probably summon some demon with this one.17:14
kristianpaulwpwrak: ;-)17:20
lekernelwpwrak: http://hackaday.com/2011/05/10/bit-banging-vga-from-an-sd-card-slot/17:40
lekernelwpwrak: see? hit the pyramid at the top :-)17:41
lekernelnow I will flood those slashdot lamers who rejected my milkymist story with more submissions ...17:41
kristianpaulwpwrak i a faraon now ;)17:41
kristianpaulwait lekernel , why not share your submissions with us for a moment before submit...?17:42
kristianpauli think 17:42
vladkorotnevand is this adapter THAT easy to make at home?17:43
vladkorotnevUnfortunately my NN did not arrive yet :(17:44
wpwraknice. hackaday picked up ubb-vga as well :)17:44
kristianpaulnow what is left..17:45
vladkorotnevCan I buy a UBB or I have to make one myself?17:46
kristianpaulIf you are skilled making PCB and good sourcing the 0.8mm thin sheet..17:48
kristianpaulAlso you can hack a uSD-to-SD adapter17:48
wpwrakvladkorotnev: tuxbrain sells them: http://www.tuxbrain.com/en/oscommerce/products/188017:49
vladkorotnevI'm not. I'm more a programmer than an engineer :P I can make some computer accessories, but not PCBs17:49
vladkorotnevwpwrak: 10x!17:49
wpwrakhe also sells single pieces: http://www.tuxbrain.com/en/oscommerce/products/189817:49
vladkorotnevOh, they are cheaper one-by-one :P17:50
wpwrakalthough i think it's a good idea to get more than one. first of all, something may go wrong. second, there may be more cool things to build in the future.17:50
wpwrakthe EUR 35 are for all ten ;-)17:51
wpwrakwell, it would be EUR 30 for you. then your customs would add vat and such for you.17:52
vladkorotnevEUR 35 = 1400 RUR, plus S&H. Not so cheap, but not so expensive, I think :P17:52
wpwrakyeah, the board could be super-cheap, but a lot of people would have to buy one17:52
vladkorotnevI'm from Russia, so postage is a problem17:53
wpwrakyou could probably bring the price down to half an Euro each at big volumes17:53
wpwrak(big = thousands)17:53
vladkorotnevI still worry for my NN, and if I order these UBBs... they may get lost :(17:54
vladkorotnevlol, I'm not making so much electronic stuff to buy  > 1000 pcs.17:54
wpwrakhackaday gave us a nice peak. already 172 visitors from there. let's see how the day continues :)17:55
vladkorotnevI think 10 would be enough ;)17:55
vladkorotnevBTW, can I reflash my NN using a Mac?17:55
wpwrak(10) i would calculate about 5 for each project you develop. fewer if you just build an existing project.17:57
vladkorotnevi think I'd first try out these awesome existing projects and then move on to developing myself17:58
wpwrakexcellent plan ;-)17:59
DocScrutinizerwpwrak: story topics: how to break copy-protection in video streams18:18
wpwrakwith ubb-vga ? :)18:18
DocScrutinizererr, mm18:19
DocScrutinizercreating HD-video from old super8 and vcr, using mm and kalman filters18:21
DocScrutinizer(with example: apollo11 pistures from moon and how they'd look today)18:22
kristianpaulhttp://experimentaltv.org:8000/plataformacero.ogg mur.at (satellites for the poors)18:25
wpwrakhmm, does anyone know a quick way to convert a dynamically linked executable to static ? (without recompiling)18:54
wpwrak(or redoing the linking of all the objects)18:54
kyakyeah, copy all shared libs in the apps dir and use LD_LIBRARY_PATH :))18:58
wpwrakah well, just rebuilt it. even seems to work ;-)19:04
kyakthis is the rare case when the most obvious solution is also the best one and, most impotantly, it works ;)19:07
kristianpaulcool ratpoison is the screen fot the X ;-)19:46
DocScrutinizerwpwrak: LD_PRELOAD all the libs, from a "local" storage <project>/libs/*20:34
DocScrutinizerooh, kyak was faster than me20:34
wpwrakyeah, that sucks in too many ways. the objective is to keep things simple ;-)20:35
DocScrutinizerwell, took me like 3 years of loose watching until I ran into a howto-statically-link20:36
qi-bot[commit] Werner Almesberger: avrdude/README: added build instructions for making a static executable http://qi-hw.com/p/ben-blinkenlights/cd3dbc020:48
qi-bot[commit] Werner Almesberger: atusb/fw/usb/dfu.c (dfu): lowered bwPollTimeout from 1 s to 100 ms and explained why http://qi-hw.com/p/ben-wpan/fc1a33420:48
qi-bot[commit] Werner Almesberger: atusb/fw/boot.c (main): don't time out until there is an application http://qi-hw.com/p/ben-wpan/ddd002220:48
qi-bot[commit] Werner Almesberger: atusb/fw/Makefile (dfu): our functional descriptor is fine, no need for -t http://qi-hw.com/p/ben-wpan/fd9154620:48
qi-bot[commit] Werner Almesberger: atusb/fw: use the unique serial number of the ATmega8/16/32U2 for iSerialNumber http://qi-hw.com/p/ben-wpan/85f60de20:48
dvdkwpwrak: do you know since which kernel version do we have the /proc/pid/pagemap files?21:13
dvdkbecause there are none on my nanonote (kernel 2.6.32)21:13
dvdkmaybe disabled in kernel config?21:13
wpwrakchecking ...21:24
wpwrak2.6.32 has it21:25
wpwrak>= 2.6.2521:26
wpwrakaccording to Documentation/vm/pagemap.txt :)21:27
dvdkwpwrak:  hmm, it's missing on my nanonote.21:28
dvdkls /proc/self/21:28
dvdkauxv             fd               mounts           root21:29
dvdkcmdline          fdinfo           mountstats       stat21:29
dvdkcoredump_filter  limits           net              statm21:29
wpwraktime to build your own kernel ?21:30
dvdkwpwrak: just waiting for the next firmware :)21:31
dvdkwpwrak: wasn't even able to find out which config option enables that feature21:32
dvdk"monitor", ok wouldn't have searched for that. 21:35
dvdkok, it's set for my ubuntu kernel, which explains why it's present there21:35
dvdkhmm, another git problem.  21:38
dvdkafter git pull, i still have kernel 2.6.32 in openwrt-xburs.git21:38
dvdkwhile the web-interface shows 2.6.3721:38
dvdkwrong branch?21:39
dvdksays i'm on th emaster21:39
wpwrakdoes git pull complain in any way ?21:40
dvdkmaybe master is the wrong branch.21:40
dvdkbecause here it's also 2.6.3221:40
dvdkbut this commit by kyak shows 2.6.37:21:41
Action: dvdk hates git21:41
wpwrakah wait, i think master is upstream21:41
wpwrakdon't hate git - hate those who use branches ! ;-)21:41
dvdkok, so i have to switch to trunk branch !?21:42
wpwrakjz-2.6.38 shoud be the right one21:42
qi-bot[commit] Lars-Peter Clausen: MIPS: JZ4740: Convert to new irq functions http://qi-hw.com/p/qi-kernel/e7c93ce21:43
qi-bot[commit] Lars-Peter Clausen: MIPS: JZ4740: GPIO: Use common irq chip for all gpios http://qi-hw.com/p/qi-kernel/bd6196121:43
qi-bot[commit] Lars-Peter Clausen: MIPS: JZ4740: Set one-shot feature flag for the clockevent http://qi-hw.com/p/qi-kernel/b0a533621:43
qi-bot[commit] Werner Almesberger: Merge branch 'jz-2.6.38' of projects.qi-hardware.com:qi-kernel into ben-wpan http://qi-hw.com/p/qi-kernel/f87f20d21:43
wpwraksee, it's easy ;-)21:43
dvdkwpwrak: no such branch listed with the git browser!21:43
wpwrakgit branch -a21:43
wpwrakwhy trust anything less than the command-line utility ? :)21:44
dvdk remotes/origin/HEAD -> origin/master21:44
dvdk  remotes/origin/history21:44
dvdk  remotes/origin/master21:44
dvdkno jz..3821:44
dvdktime for rm -rf?21:44
wpwrakoh wait. i'm using qi-kernel.git21:44
wpwrakno idea what you guys have in openwrt-xburst21:45
dvdkwhat command is used to switch branches 'git branch trunk' doesn't work?21:45
wpwrakyes, git checkout21:45
dvdkdoesn't work21:45
dvdkdoes nothing21:46
wpwrakqi-kernel is a different project21:46
dvdktrying checkout origin/trunk21:47
dvdk(does one have to understand that?)21:47
dvdkok, now i have a config 2.6.3721:47
wpwrakheh :)21:47
dvdkkernel is building.  maybe it'll be done in the morning21:55
wpwrakhuh ? how long does a kernel build take for you ?21:56
dvdkwpwrak: i'm afraid 'make' in the openwrt toplevel dir does a little more than just rebuild the kernel21:56
wpwrakyou''re picking an impressively scenic route ;-)21:57
dvdkcool, now it compiles CMake using C++21:57
wpwrakona decent pc, the kernel should build within a few minutes21:57
dvdkmy computer is named 'snail'21:58
dvdkfor a reason21:58
dvdkoh, one of many passes (re)building the gcc toolchain just started21:59
wpwrakzx81, running an ia32 emulator written in java and swapping to cassette tape ? :)21:59
dvdks/java/java script/21:59
wpwrakeven better ;-)21:59
wpwrakyou should install the host toolchain and then just build straight from the qi-kernel project22:00
dvdkwpwrak: http://www.tramm.li/i8080/22:01
wpwrakuseful :)22:01
dvdkstill waiting for the ia32 emulator to boot :)22:01
wpwraktoo bad i throw all those old cp/m floppies away some 20 years ago ;)22:01
wpwrakit's probably busy compiling the NOP instruction ...22:02
dvdkcool, here you can play space invaders on 8080, emulated via java script:22:03
wpwrakhere are kernel build instructions for the ben: http://projects.qi-hardware.com/index.php/p/ben-wpan/source/tree/master/install/INSTALL-Ben22:03
wpwrakfor openwrt and jlime22:04
wpwrakand here are two useful scripts: http://projects.qi-hardware.com/index.php/p/wernermisc/source/tree/master/bin/22:04
wpwrakmknnk and nnui22:04
wpwrakmknnk just sets up make such that you get a ben build22:05
wpwraknnui does the mkimage incantations and such as well22:05
dvdkwpwrak: well, doing 'make -j5' then searching for any uImage.bin that has a current timestamp.  hope that does it.22:06
dvdk... still gcc compiling22:06
dvdkok time to go to bed22:06
wpwrakah, you have a quad-core then ?22:06
dvdkwpwrak: yup, underclocked, because the mini-itx chassis doesn' like so much heat (and the powersupply is insufficient etc)22:07
wpwrakthat would be: make -j5 ARCH=mips CROSS_COMPILE=mipsel-openwrt-linux- vmlinux.bin22:07
wpwrakhow much RAM ?22:07
dvdkit's only 1.8GHz22:07
wpwrakokay, probably enough22:07
dvdkpropably have to go down to 1.5 once the whether gets hotter over here22:07
dvdkmanual thermal throtteling22:08
wpwraknowadays, they make nice big towers that aren't all that noisy ...22:08
dvdkalways an eye on the hdd temp.  usually over 50C it's not going to live long22:08
wpwrakand they stay cool. around here, we get around 40 C in summer22:09
wpwrakdo not buy cheap WD disks ;-)22:09
dvdki like to save space.  big towers is so backwardish :P22:09
wpwrakbetter fry the cpu every now and then ;-)22:09
dvdkit's a WD "green" disk, to stay within power limits22:09
dvdkcompile error.  i hate it22:10
dvdkok, gotta go22:10
kristianpaulwpwrak: minutes? ha, on my laptop it take almost a day and at the desktop computer, at least 4 hrs..22:16
wpwraka kernel compile ? or the whole distro image dance ?22:18
kristianpaulminimal distro22:19
Action: kristianpaul need get a computer with at least two cores..22:19
wpwrakthat's why you should build the kernel directly if all you need is a kernel ;-)22:19
kristianpaulokay, i'll squeeze that kernel next time i compile it :-)22:35
xMfffun to see milkymist on slashdot, looks like the pr machine is working :)23:38
xMffthe initial comments are a bit depressing though, but thats expected from the slashdot armchair engineering crowd23:38
wpwrakwhoopie !23:40
--- Wed May 11 201100:00

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