| 00:11.33 | RobotGuy | Where are the hammer.cfg and hammer.ocd files? |
| 00:12.24 | RobotGuy | I can't remember where I got them. |
| 00:16.56 | *** join/#edev dijenerate (n=dijenera@72.22.133.141) |
| 00:25.28 | *** join/#edev prpplague (n=dave@70.249.150.251) |
| 00:27.24 | *** join/#edev RobotGuy (n=RobotGuy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 00:27.38 | RobotGuy | Welcome back prpplague |
| 00:27.58 | RobotGuy | prpplague: Where do I find the hammer.cfg and hammer.ocd files? I can't remember. :( |
| 00:28.47 | prpplague | RobotGuy: no clue where you put them |
| 00:28.53 | prpplague | RobotGuy: normally they are on the cd |
| 00:28.58 | prpplague | RobotGuy: or on the wiki |
| 00:29.36 | RobotGuy | prpplague: Ah, OK. Am so used to looking on the Wiki that I didn't even think to look on the CD. |
| 00:31.07 | prpplague | you figure out the route issue? |
| 00:32.18 | prpplague | ho diddy ho ho |
| 00:32.47 | ds2 | Mmmm patents |
| 00:32.51 | SpeedEvil | You need more than mere research! |
| 00:32.56 | RobotGuy | Patent troll? |
| 00:33.08 | SpeedEvil | Simply showing prior art doesn't actually make you automatically win. |
| 00:33.13 | prpplague | SpeedEvil: yea well, thats my part of it |
| 00:33.16 | SpeedEvil | :/ |
| 00:33.26 | RobotGuy | prpplague: I haven't figured out the route issue yet. May just go with the generic addresses if I can get it to work. |
| 00:33.33 | prpplague | SpeedEvil: we have to show that the patents don't apply |
| 00:34.14 | SpeedEvil | I mean that you have to convince a court. |
| 00:34.22 | prpplague | RobotGuy: are you asking what a patent troll is? |
| 00:34.33 | prpplague | SpeedEvil: indeed, that's why i have to do the research |
| 00:35.22 | prpplague | SpeedEvil: http://trolltracker.blogspot.com/2008/01/4-interesting-new-cases-from-last-2.html |
| 00:37.06 | prpplague | every linux user of 802.11 should be concerned |
| 00:37.18 | prpplague | as WIAV claims that 802.11 violates their patents |
| 00:38.08 | SpeedEvil | prpplague: And RIM eventually won their patent case and got the patent thrown out. |
| 00:38.19 | SpeedEvil | Still had to pay damages. |
| 00:38.56 | SpeedEvil | Patent system is _so_ broken. |
| 00:39.01 | prpplague | yep, problem is that these patents don't apply to 802.11 |
| 00:39.09 | SpeedEvil | Last silly one I found was when I was researching SAW displays. |
| 00:39.32 | SpeedEvil | They were wondering about how to seal from the environment, but not absorm the accoustic wave. |
| 00:39.48 | SpeedEvil | The solution was what any tech would have found in a couple of hours. |
| 00:40.05 | SpeedEvil | You wander round the lab trying various stuff as gaskets till you find one that works. |
| 00:40.19 | SpeedEvil | This should not be patentable. |
| 00:40.28 | prpplague | SpeedEvil: well in this we aren't waiting |
| 00:40.36 | prpplague | SpeedEvil: don't know if you looked at the URL |
| 00:40.52 | SpeedEvil | checking now |
| 00:41.00 | SpeedEvil | missed it first time round |
| 00:44.46 | SpeedEvil | qI don't see why the ...mesh... one would apply |
| 00:45.25 | prpplague | http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 00:45.27 | SpeedEvil | However - I'm not sure why the other one wouldn't - on a cursory reading |
| 00:45.29 | prpplague | SpeedEvil: exactly |
| 00:45.47 | prpplague | SpeedEvil: the other is based on geographical location |
| 00:46.00 | prpplague | SpeedEvil: 802.11 doesn't do any geographical roaming |
| 00:46.40 | prpplague | SpeedEvil: what aspect of the other one seems to apply to 802.11? |
| 00:46.56 | ds2 | they both use RF energy and are spreadspectrum |
| 00:46.58 | ds2 | =) |
| 00:47.00 | SpeedEvil | As I understand it, the preface of the patent has no force at all. |
| 00:47.11 | SpeedEvil | It's simply text that is not closely examined. |
| 00:47.16 | SpeedEvil | The claims are the important part |
| 00:47.35 | SpeedEvil | I could of course be completely wrong, it's happened at least once. |
| 00:47.39 | prpplague | SpeedEvil: ok what part of the claims seem relevant to 802.11 |
| 00:47.41 | prpplague | SpeedEvil: ? |
| 00:48.07 | prpplague | SpeedEvil: remeber, their allegation is that ALL 802.11 is a violation of their patent |
| 00:48.22 | prpplague | " |
| 00:48.22 | prpplague | In a packet communication system wherein stationary nodes are assigned an absolute coordinate-based address, " |
| 00:48.36 | prpplague | 802.11 does not use any coordinate-based routing |
| 00:48.40 | SpeedEvil | However, on further reading - that's not how you usually login to anAP. |
| 00:48.55 | SpeedEvil | It was my understanding that the preface has no legal force. |
| 00:48.57 | SpeedEvil | Just the claims. |
| 00:49.08 | SpeedEvil | I don't remember where I picked this notion up. |
| 00:49.16 | SpeedEvil | s/was/is/ |
| 00:49.32 | adyer | I am under the same impression - the claims are the 'meat' |
| 00:50.23 | SpeedEvil | But ' |
| 00:50.25 | SpeedEvil | intermittently transmitting link acquisition packets to one or more of said stationary nodes to determine the quality of the possible communication link with them; ' |
| 00:50.35 | SpeedEvil | That's not quite how it works is it? |
| 00:50.53 | SpeedEvil | It's if the node can pick up the beacon packet broadcast by the AP. |
| 00:51.32 | SpeedEvil | The first claim could actually be read as searching for authorised nodes that you can login to though. |
| 00:51.47 | SpeedEvil | I think this might apply to ad-hoc networks. |
| 00:52.20 | prpplague | item 5 is where the "meat" is |
| 00:52.29 | prpplague | http://www.patentfizz.com/fizzdisplay.php?patno=5400338 |
| 00:52.33 | adyer | methinks asking for legal advice on irc is about the same as asking on slashdot :-) |
| 00:52.34 | prpplague | 802.11 does not do that |
| 00:52.44 | prpplague | adyer: i'm not asking for advice |
| 00:52.58 | prpplague | adyer: we have lawyers, i'm just doing research to hand them |
| 00:53.02 | SpeedEvil | ARP? |
| 00:53.14 | SpeedEvil | (item 5) |
| 00:53.22 | SpeedEvil | with MACs from the wifis |
| 00:53.27 | SpeedEvil | The server would be a router |
| 00:53.53 | adyer | are you trying to invalidat the patent through prior art or just say it's not applicable? |
| 00:53.57 | prpplague | SpeedEvil: i can see how that might apply |
| 00:54.07 | prpplague | adyer: not applicable |
| 00:54.32 | SpeedEvil | 1994! |
| 00:54.33 | prpplague | adyer: because if it is applicable, that means anyone who implements 802.11 owes these guys $$ |
| 00:54.41 | SpeedEvil | There _has_ to be prior art. |
| 00:55.11 | SpeedEvil | There were 2.4G network cards around then - if I've not screwed my timelines. |
| 00:56.00 | prpplague | SpeedEvil: thats just it, everything thats implemented in ieee was researched and those who held patents on the implemented tech were given letters of assurance |
| 00:56.18 | prpplague | SpeedEvil: you can't implement a free standard under threats of licenses and lawsuits |
| 00:56.54 | SpeedEvil | That assumes that the research was perfect. |
| 00:57.28 | SpeedEvil | If a patent got missed, or they screwed up on the applicability, it's quite possible for it to infringe. |
| 00:57.37 | adyer | i suspect you have to go through, come up with the most perverse interpretation of each claim and figure out why it doesn't apply |
| 00:59.34 | ds2 | when was that patent issued? |
| 00:59.39 | SpeedEvil | I think claim one clearly does not apply to AP mode - as there is no 'call and response' |
| 00:59.41 | SpeedEvil | 1994 |
| 00:59.57 | ds2 | so drag it out for 4 more years and... ;) |
| 00:59.58 | adyer | could also have been a 'submarine patent' |
| 01:00.06 | SpeedEvil | I'm unsure about AP mode. |
| 01:00.08 | SpeedEvil | err |
| 01:00.10 | SpeedEvil | ad-hoc |
| 01:01.09 | prpplague | ds2: hehe we are sueing them |
| 01:01.15 | prpplague | ds2: everyone else has been paying them |
| 01:01.18 | prpplague | ds2: to go away |
| 01:01.18 | SpeedEvil | I think claim 2 could apply - if you construe 'authorisation packet' to mean 'login attempt' |
| 01:01.31 | SpeedEvil | err |
| 01:01.47 | SpeedEvil | link acquisition packet |
| 01:01.49 | prpplague | ds2: and the issue is, if it is a valid patent, then need to take it up with iee |
| 01:02.23 | ds2 | prpplague: why would it come down you guys? shouldn't marvell or whoever made the chip + the nonopensource firmware be the one to get sued? |
| 01:02.41 | prpplague | ds2: exactly!!! |
| 01:03.10 | prpplague | ds2: they've been slowly hitting all the small guys, and they are paying up to make it go away, when they have enough money, they attack the bigger fish |
| 01:03.15 | SpeedEvil | I think claim 3 clearly does not apply - ' |
| 01:03.17 | SpeedEvil | receiving at any number of stationary nodes a link acquisition packet from said roaming node and transmitting a response packet to said roaming node from each stationary node that received said link acquisition packet; ' |
| 01:03.43 | SpeedEvil | wifi just doesn't do that in either AP or ad-hoc mode as I understand it. |
| 01:04.00 | SpeedEvil | Even if you construe acquisition packet as 'attempt to login' |
| 01:06.38 | prpplague | i was hopin groklaw would pick this and get some geek feedback |
| 01:07.56 | SpeedEvil | I think claim 3 clearly fails. I'm not sure if this means claim 5 fails as it defines claim 3 further |
| 01:08.03 | adyer | might try ars technica, boing boing, or slashdot |
| 01:08.08 | MonMotha | prpplague: isn't that SOP for patent trolls? |
| 01:08.16 | prpplague | MonMotha: yep |
| 01:08.24 | prpplague | MonMotha: they just weren't counting on us |
| 01:08.32 | MonMotha | that seems to be what normally happens |
| 01:08.39 | MonMotha | eventually one of the "small fish" decides not to roll over |
| 01:09.05 | prpplague | let's just say aml has been preparing for an issue like this for a long time |
| 01:10.07 | prpplague | got a really cool lcd from a game at walmart |
| 01:10.10 | prpplague | they are clearance |
| 01:10.12 | prpplague | $15 |
| 01:11.06 | adyer | anybody know the stty magic to make the tty code not buffer input? I want to do 'tail -f /dev/ttyS1' and chars as they are typed on the other side |
| 01:12.26 | ds2 | eh? |
| 01:12.42 | ds2 | why would one want to use tail -f on a tty device? |
| 01:13.08 | adyer | to keep the port open and see what gets sent |
| 01:13.16 | MonMotha | heck, just cat it, and even that's almost a uuoc |
| 01:13.18 | SpeedEvil | tail is line oriented |
| 01:13.21 | SpeedEvil | cat |
| 01:13.47 | ds2 | tail -f doesn't make sense; cat does |
| 01:13.52 | adyer | cat seems to be line buffered |
| 01:14.30 | ds2 | prehaps tip or kermit might be more to your liking then? |
| 01:15.06 | adyer | it's running on a target, so I'd have to build it and move it over |
| 01:15.22 | ds2 | kermit builds pretty easily |
| 01:15.24 | prpplague | adyer: try the toybox netcat, it works wonderfully |
| 01:15.33 | ds2 | tried stty raw ? |
| 01:15.50 | prpplague | http://www.miuchiz.com/index_1.0.asp |
| 01:16.07 | prpplague | these things are on clearance at walmart for $15 |
| 01:16.19 | prpplague | they have an awesome little color display |
| 01:17.02 | prpplague | http://www.amazon.com/Pawz-Spike-MIUCHIZ-Virtual-Companions/dp/B000F901FK |
| 01:17.28 | MonMotha | tiny |
| 01:17.49 | ds2 | what rez? |
| 01:18.42 | adyer | raw mode with cat seems to work well enough |
| 01:19.37 | adyer | now I can check IR serial off the list :-) |
| 01:26.23 | *** join/#edev landley (n=landley@cpe-72-177-114-214.austin.res.rr.com) |
| 01:41.25 | *** join/#edev RobotGuy (n=RobotGuy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 01:41.58 | RobotGuy | prpplague: I am having trouble getting openocd working now, and I have the library for ftdi stuff. |
| 01:46.38 | prpplague | RobotGuy: what seems to be the problem |
| 01:46.48 | prpplague | landley: nail board's should be back wed. |
| 01:47.01 | prpplague | landley: we'll do the assembly and start testing around friday |
| 01:47.11 | landley | Cool. |
| 01:47.40 | prpplague | SpeedEvil: yea, i just noticed it was $8 |
| 01:49.29 | prpplague | SpeedEvil: oh wait the one for $8 is monochrome |
| 01:49.43 | prpplague | SpeedEvil: http://www.amazon.com/Spike-Miuchiz-Electronic-Handheld-Game/dp/B000OD7YHA/ref=pd_bxgy_t_img_b |
| 01:50.18 | SpeedEvil | I've gotten more interested in 'new' stuff. |
| 01:50.32 | SpeedEvil | Spares for portable game consoles can be quite cheap. |
| 01:50.38 | MonMotha | I have a nice 8" widescreen TFT I ripped out of somebody's old portable DVD player that bit the bucket |
| 01:50.41 | MonMotha | should be fun |
| 01:50.56 | SpeedEvil | As in $14 for a 2.8" TFT backlit display delivered |
| 01:51.34 | RobotGuy | prpplague: http://www.pastebin.ca/868001 |
| 01:52.22 | prpplague | RobotGuy: when you compiled openocd you didn't specify the correct options |
| 01:52.33 | prpplague | RobotGuy: did you use the openocd off your cd or off the wiki? |
| 01:52.36 | RobotGuy | prpplague: I tried both of the ftdi options |
| 01:53.39 | prpplague | http://elinux.org/Hammer_Board_Software_System |
| 01:56.19 | RobotGuy | prpplague: I'm trying to compile the one from the wiki. |
| 02:00.23 | prpplague | wow anyone ever heard of a LED with only 50k hours usage? |
| 02:00.39 | MonMotha | prpplague: cheap blue ones, yeah |
| 02:03.25 | prpplague | hmm |
| 02:03.34 | prpplague | this looks like a sweet little lcd |
| 02:03.50 | prpplague | cstn |
| 02:04.07 | MonMotha | give me TFT or give me death! |
| 02:04.59 | prpplague | for $15 i'm happy with the parts |
| 02:05.57 | SpeedEvil | For $15 you can get a 2.8" TFT with backlight |
| 02:07.00 | *** join/#edev chouimat|away (n=dieu@r2351064.cidc.net) |
| 02:07.46 | prpplague | SpeedEvil: where? |
| 02:07.49 | prpplague | SpeedEvil: http://www.powertip.com.tw/products_2.php?product_id=1171061940&area_idbk=1170780372 |
| 02:07.58 | prpplague | SpeedEvil: found the pinout |
| 02:09.35 | SpeedEvil | :) |
| 02:11.45 | prpplague | SpeedEvil: average joe can't get a usable tft for less than $20 |
| 02:12.06 | SpeedEvil | dealextreme.com |
| 02:12.12 | SpeedEvil | NDS screens |
| 02:12.18 | prpplague | SpeedEvil: i said usable |
| 02:12.24 | prpplague | ~lart SpeedEvil |
| 02:12.24 | purl | pries SpeedEvil's back open with a screwdriver and flashes a new bootldr to SpeedEvil |
| 02:12.33 | SpeedEvil | Why is it unusable? |
| 02:13.22 | prpplague | SpeedEvil: lcd connector + power supplies pushing that way beyond the $20 mark |
| 02:13.43 | SpeedEvil | True |
| 02:13.47 | SpeedEvil | dn't quite remember. |
| 02:13.53 | SpeedEvil | oops |
| 02:14.26 | prpplague | SpeedEvil: for $15 i get the lcd + connector + powersupply parts |
| 02:14.38 | SpeedEvil | http://home.comcast.net/~olimar/DS/jumbotron/ |
| 02:14.43 | SpeedEvil | Is worth a look. |
| 02:14.47 | SpeedEvil | If only for the insanity. |
| 02:14.53 | ds2 | 97x32? that is rather... limiting |
| 02:18.21 | prpplague | actually this one is 98x67, just same pinout |
| 02:24.49 | RobotGuy | openocd will not build for me. :( |
| 02:27.15 | RobotGuy | prpplague: I found a package for Ubuntu and installed that openocd and now get this error: |
| 02:27.20 | RobotGuy | Info: openocd.c:92 main(): Open On-Chip Debugger (2007-07-31 19:00 CEST) |
| 02:27.20 | RobotGuy | Error: ft2232.c:1509 ft2232_init(): No matching layout found for flyswatter |
| 02:29.40 | ds2 | at that rez, I'd use a 40x4 alphanumeric display |
| 02:29.43 | ds2 | easier to deal with |
| 02:30.07 | prpplague | RobotGuy: thats because the page you got probably doesn;t have the latest build with flyswatter support |
| 02:30.40 | prpplague | ds2: just as easy for the hammer to deal with a small graphical as a char based |
| 02:32.04 | prpplague | for $15 i'm happy with the parts |
| 02:32.11 | ds2 | but coding wise, a /dev/ttyXYZ is easier then a /dev/fb0 |
| 02:32.40 | prpplague | ds2: matter of opinion |
| 02:35.34 | prpplague | http://www.powertip.com.tw/products_2.php?product_id=1171150132&area_idbk= |
| 02:35.39 | prpplague | not bad resolution |
| 02:36.01 | RobotGuy | prpplague: I got it off the wiki. |
| 02:36.24 | RobotGuy | prpplague: Wait, no this one is from a package for Ubuntu. |
| 02:36.36 | prpplague | RobotGuy: you said you got a package for ubuntu |
| 02:36.59 | RobotGuy | prpplague: Yes, this latest one is from an Ubuntu package, sorry. Confused. |
| 02:40.09 | RobotGuy | It can't find ftdi.h |
| 02:40.30 | *** join/#edev bbradley_ (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 02:40.56 | *** join/#edev redblue (i=star@ppp227.108-253-207.mtl.mt.videotron.ca) |
| 02:47.58 | RobotGuy | prpplague: Removed ubuntu openocd. Still can not compile openocd from wiki |
| 02:48.27 | prpplague | RobotGuy: whats the error? |
| 02:50.46 | prpplague | RobotGuy: if it can't find ftdi.h you didn';t copy it over |
| 02:50.51 | prpplague | RobotGuy: from the lib dir |
| 02:51.19 | RobotGuy | prpplague: I did copy the .h file to /usr/include |
| 02:51.30 | prpplague | RobotGuy: wrong place |
| 02:52.28 | prpplague | 5) copy the fd2xx.h and WinTypes.h files from libftd2xx directory to the openocd/src/jtag/ directory |
| 02:53.27 | RobotGuy | prpplague: http://www.pastebin.ca/868049 |
| 02:55.31 | prpplague | RobotGuy: wrong library |
| 02:55.46 | prpplague | 10) then configure the build using the command: ./configure --enable-ft2232_ftd2xx |
| 02:56.20 | prpplague | RobotGuy: maybe you should start from scratch - http://www.elinux.org/Hammer_How_to_use_Jtag |
| 02:56.52 | ds2 | prpplague: have you consider approaching the local colleges to setup a mentoring/project class where the student gets hands on experience developing something they propose and having professional mentoring to get it implemented? |
| 02:57.12 | prpplague | ds2: to what end? |
| 02:57.38 | ds2 | prpplague: real education for students and a potential captive audience for the product =) |
| 02:58.09 | ds2 | a lot of schools just don't offer real hands on experience |
| 03:01.12 | prpplague | ds2: oh you mean for the hammer board stuff? yea, the boss is in talks with the university of north texas engineering department |
| 03:01.41 | MonMotha | hey, if I were interested in learning how to design plastic injection molded enclosures, anybody have any references? |
| 03:01.45 | prpplague | ds2: personally, i'd love to teach, but the way the education rules are here in the US, i'd never be able to |
| 03:02.08 | prpplague | MonMotha: our IE at the office is pretty in on that |
| 03:02.13 | prpplague | MonMotha: i can ask in the morning |
| 03:02.18 | MonMotha | prpplague: would be great |
| 03:02.36 | prpplague | MonMotha: we use solid works to do all of our plastic designs and SLA's |
| 03:02.46 | MonMotha | we have nobody to do that at our company, and we can't justify hiring on a ME just for enclosure design |
| 03:02.54 | MonMotha | as the "hardware guy", it would fall to me |
| 03:03.05 | MonMotha | yeah, I've heard lots of people like solidworks |
| 03:03.25 | prpplague | MonMotha: to be perfectly honest, it's a whole field on its on, i don't think you are going to have the time to spend learning |
| 03:03.33 | MonMotha | prpplague: I'm aware of that |
| 03:03.39 | prpplague | MonMotha: you might be better office outsourcing |
| 03:03.52 | MonMotha | well, we do that currently with great results, but it's extremely expensive |
| 03:04.03 | MonMotha | for complicated stuff, it's nice, but it would be good if I could handle simple designs |
| 03:04.27 | prpplague | MonMotha: indeed, be prepared for a steep and expensive learning curve |
| 03:04.43 | MonMotha | well, expense would be covered by the company, at least |
| 03:04.50 | MonMotha | I wouldn't (directly) see it |
| 03:05.42 | prpplague | MonMotha: indeed, be expecting multiples of 10k to get started |
| 03:05.57 | MonMotha | about as expensive as my ECAD software |
| 03:06.12 | prpplague | MonMotha: i mean for proto's |
| 03:06.29 | prpplague | MonMotha: you gonna need to do alot of SLA's to get used to the design needs |
| 03:06.37 | MonMotha | oh, yeah. We noticed that it's MUCH more expensive to do prototype plastic than prototype PCBs |
| 03:07.44 | MonMotha | SLA is really expensive, and tooling for small volume is also really expensive |
| 03:07.47 | RobotGuy | prpplague: It's going to work OK now. |
| 03:08.36 | RobotGuy | prpplague: I use Alibre Design Expert for 3D CAD. |
| 03:09.05 | prpplague | RobotGuy: yea alibre works well for very small designs |
| 03:09.13 | RobotGuy | It's MUCH less expensive that Solid Works, and I can't see it missing any features. |
| 03:09.25 | RobotGuy | prpplague: How small is small? |
| 03:09.38 | prpplague | RobotGuy: hehe, yea try designing handheld with alibera and you'll go nuts |
| 03:10.13 | RobotGuy | prpplague: All of a handheld? Why would Alibre not be able to handle that? |
| 03:10.37 | MonMotha | prpplague: like trying to design a multi-BGA 8 layer PCB in Eagle? :) |
| 03:10.39 | MonMotha | you can do it... |
| 03:11.31 | prpplague | MonMotha: hehe indeed |
| 03:12.04 | prpplague | RobotGuy: about 1/2 way thru the M7100 alibre would complete come to stand still |
| 03:12.25 | RobotGuy | prpplague: Have you tried v10 of Alibre? |
| 03:13.14 | prpplague | RobotGuy: i have not, but our IE has, we have a ton of legacy products on alibre, so he has to convert them over to solid-works on occasion |
| 03:13.52 | prpplague | the other major problem is that alibre isn't supported by most of the plastic houses |
| 03:14.30 | RobotGuy | prpplague: Hmmmm |
| 03:14.54 | prpplague | RobotGuy: for hobbists alibre is fine, production stuff, solid works is the only game in town |
| 03:15.02 | RobotGuy | prpplague: Well, there is no way I could ever afford Solid Works and their support charges. |
| 03:15.08 | prpplague | RobotGuy: indeed |
| 03:15.12 | MonMotha | cost of doing business |
| 03:15.27 | RobotGuy | prpplague: It took me a whole year to pay for my Alibre Design Expert. |
| 03:15.31 | prpplague | RobotGuy: just like eagle is penlty fine for hobbists, but for production orcad is where it's at |
| 03:15.39 | RobotGuy | prpplague: I hate Eagle. |
| 03:15.57 | prpplague | RobotGuy: what do you use? |
| 03:16.17 | RobotGuy | prpplague: Nothing at present. I want EasyPC by Number One. |
| 03:16.36 | prpplague | never heard of it |
| 03:16.37 | MonMotha | prpplague: I used to use OrCAD. When given the choice, I ended up switching to Altium Designer (the descendant of Protel) |
| 03:16.48 | prpplague | MonMotha: ahh |
| 03:16.55 | RobotGuy | www.numberone.com |
| 03:17.17 | MonMotha | it's still under heavy development |
| 03:17.36 | MonMotha | they have some neat 3d features that I'm hoping may evolve into a full-blown enclosure designer |
| 03:17.51 | MonMotha | they also have a whole bunch of firmware/FPGA integration that I'm not currently using |
| 03:18.48 | prpplague | RobotGuy: doesn't appear to be available for linux |
| 03:19.53 | ds2 | MonMotha: look into rhino |
| 03:20.22 | MonMotha | prpplague: Designer doesn't run on Linux, either, sadly |
| 03:20.29 | MonMotha | then again, neither does OrCAD |
| 03:20.53 | RobotGuy | prpplague: Neither is Alibre |
| 03:21.23 | MonMotha | sadly, very little professional CAD software is available for anything other than Windows at this point |
| 03:21.31 | MonMotha | which is hilarious, since it all used to only run on UNIX |
| 03:21.37 | prpplague | eagle and geda both run on linux |
| 03:22.06 | ds2 | if you got the budget, I think ProE is available for Linux |
| 03:22.06 | RobotGuy | prpplague: I need to install gEDA. |
| 03:22.25 | prpplague | ds2: yea? |
| 03:22.30 | prpplague | ds2: thats interesting |
| 03:22.40 | ds2 | last I checked, unless they withdrew it |
| 03:22.52 | ds2 | but a ProE seat costs as much as a basic car :( |
| 03:23.08 | MonMotha | ds2: I've heard not good things about the stability of ProE |
| 03:23.29 | MonMotha | prpplague: eagle and geda are hardly what I'd call "professional" CAD software |
| 03:23.30 | ds2 | I would say the same about alibre |
| 03:23.49 | MonMotha | geda is decidedly hobbyist at this point, and eagle is kinda in the boundary between hobbyist and casual professional use |
| 03:23.52 | ds2 | For most of what I care about, either Rhino or just hacking it out in MasterCAM is sufficient |
| 03:24.42 | ds2 | MonMotha: you plan to do the machining for the injection molds yourself? |
| 03:24.59 | MonMotha | ds2: no. We know some places that do that kind of tooling |
| 03:25.06 | ds2 | ah |
| 03:25.12 | MonMotha | I presume they take solidworks files or simliar |
| 03:25.24 | ds2 | worse case, you can export to IGES |
| 03:25.30 | MonMotha | yeah |
| 03:25.54 | ds2 | i wouldlike to know more about the injection side... took a bunch of classes for the machining side |
| 03:26.00 | MonMotha | our current procedure for our enclosure people is that I export a STEP file (I can do IGES, too, but they prefer STEP) of my board and they go from there |
| 03:26.16 | MonMotha | then we hammer back and forth until there's a PCB that fits in an enclosure |
| 03:26.24 | ds2 | hahaha |
| 03:26.45 | ds2 | with IGES, you should be able to have a enclosed machined out for testing before talking to the mold makers |
| 03:26.50 | ds2 | or STEP files |
| 03:27.05 | MonMotha | well, the STEP file I make only describes my PCB, not any enclosure |
| 03:27.11 | RobotGuy | My compressed initrd is < 1 MB now, but it will get a bit larger. |
| 03:27.13 | ds2 | ah |
| 03:27.33 | ds2 | do you really need a STEP file for the PCB? one place I know uses DXFs for that |
| 03:27.56 | MonMotha | my ECAD software does do 3D stuff for the PCB itself |
| 03:28.12 | MonMotha | so it saves the enclosure people from having to generate models or at least height info for all the stuff on the boards |
| 03:28.27 | MonMotha | since I already have it |
| 03:28.29 | ds2 | that's another complexitiy layer for the library maintainers |
| 03:28.38 | MonMotha | library maintainers being me |
| 03:28.47 | MonMotha | I'm the only hardware guy on staff |
| 03:28.57 | MonMotha | our projects are mostly software projects |
| 03:29.10 | MonMotha | sometimes they run on custom hardware, sometimes COTS |
| 03:29.36 | MonMotha | I've done everything from designing ARM based machines to reverse engineering parts of a PC104 board we had with incomplete docs to writing drivers |
| 03:29.49 | ds2 | sounds fun |
| 03:29.58 | MonMotha | I certainly run the gambit in terms of jobs |
| 03:30.38 | MonMotha | I spec out what parts go on the boards, add them to my libraries, do the EE side, do the layout, and I write software when needed (due to time or just because I am the most familiar with what's going on) |
| 03:31.02 | ds2 | i would like to move more toward that |
| 03:31.19 | MonMotha | as far as the 3d stuff, I mostly just put a height dimension into the library when I create the symbol and footprint |
| 03:31.24 | MonMotha | I don't make a full model |
| 03:31.28 | ds2 | okay traffic is not going to get any better, guess I should start home :/ |
| 03:36.58 | prpplague | hmm the pinouts don't match |
| 03:48.08 | RobotGuy | prpplague: I did get another build error on the native toolchain: http://www.pastebin.ca/868079 |
| 03:48.19 | RobotGuy | Damn |
| 03:54.34 | RobotGuy | Wow, the Virtual COGs i.MX21 COG isn't much more powerful RAM and speed wise than the Hammer. It does have 64 MB RAM and is 266 MHz, designed for multimedia. |
| 03:54.55 | RobotGuy | I really like the stacking architecture. |
| 04:21.30 | landley | RobotGuy: http://lwn.net/Articles/265240 |
| 04:21.37 | landley | I mentioned how unionfs seemed called for... |
| 04:26.00 | RobotGuy | landley: Hi. I don't subscribe to that site. |
| 04:26.28 | landley | Ah, right. |
| 04:26.41 | landley | http://lwn.net/SubscriberLink/265240/141622cee00aaf74/ |
| 04:26.44 | landley | Try that. :) |
| 04:31.50 | *** join/#edev DrZimmerman (n=theo@2001:5c0:8fff:fffe:0:0:0:6af1) |
| 04:43.04 | RobotGuy | landley: I do understand what UNIONFS is and what it does. I just don't think it is going to solve my particular issue. |
| 05:00.32 | ds2 | UnionFS is great until you get NFS involved. |
| 05:07.49 | RobotGuy | ds2: I think what I will end up doing is copy everything from /etc to /usr/etc, rm -rf /etc, and then ln -s /usr/etc etc |
| 05:08.12 | ds2 | I would not do that |
| 05:08.18 | ds2 | but it is YOUR board, not mine ;) |
| 05:08.57 | RobotGuy | Why would you not do that? |
| 05:09.56 | ds2 | moving /etc around may have unexpected effects |
| 05:10.03 | RobotGuy | Like what? |
| 05:10.13 | ds2 | like not being able to login again |
| 05:10.26 | RobotGuy | That's a reflash. |
| 05:10.57 | ds2 | why are you doing that anyways? |
| 05:11.10 | RobotGuy | Why am I doing what? Reflashing? |
| 05:11.25 | ds2 | trying to get rid of /etc |
| 05:11.37 | RobotGuy | I'm not trying to get rid of /etc. |
| 05:12.07 | ds2 | then why are you blowing it away then? |
| 05:12.12 | RobotGuy | I don't want config files for software I build to be put into /etc. |
| 05:12.25 | RobotGuy | Because it seems the only way to accomplish what I want to do. |
| 05:12.29 | ds2 | then don't put it there |
| 05:12.35 | ds2 | compile time option for a lot of packages |
| 05:12.39 | ds2 | run time option for others |
| 05:12.42 | RobotGuy | Nobody seems to understand. |
| 05:13.00 | RobotGuy | No way to set that globally for a software build that I am aware of. |
| 05:13.24 | ds2 | that's typical of "standard" *nix software |
| 05:14.02 | RobotGuy | Just because it is typical, doesn't make it right. :) |
| 05:14.09 | ds2 | but how many custom packages do you want to have? |
| 05:14.52 | RobotGuy | I want a lot of software from buildroot. |
| 05:14.52 | ds2 | a lot of packages don't default to /etc unless it has been molested by the distros |
| 05:14.53 | ds2 | look at the ./configure they run, it is probally specified there |
| 05:14.54 | RobotGuy | Well, buildroot packages do default to /etc. |
| 05:15.32 | ds2 | buildroot runs ./configure |
| 05:15.35 | ds2 | with options |
| 05:15.46 | RobotGuy | ds2: That would require editing every single configure script. Oh, I could certainly do that, and it would likely be the best solution. |
| 05:16.09 | ds2 | nah... not every |
| 05:16.23 | RobotGuy | I don't see anywhere to set global configure script options. If there were a way, I could just set -prefix=/usr |
| 05:16.54 | ds2 | try it, learn from it |
| 05:16.58 | RobotGuy | ds2: To do it right, I'd have to edit every configure script to allow for global options. |
| 05:17.11 | ds2 | I don't think it is that drastic |
| 05:17.22 | ds2 | but I don't use buildroot |
| 05:17.43 | RobotGuy | Try what? What I am proposing to do by editing every configure script? |
| 05:18.18 | RobotGuy | There could be an option for additional global configuration options. |
| 05:18.48 | RobotGuy | Set to NULL everything happens as it does right now. Set to not NULL, the options change the build behavior as specificied. |
| 05:45.40 | landley | ds2: I don't use buildroot either, but that's because I consider doing _anything_ with it to be drastic. :) |
| 05:46.50 | RobotGuy | landley: What is so bad about buildroot? Why would TinCanTools choose to use it if it is so bad? |
| 05:47.18 | landley | I don't like the architectural approach. |
| 05:47.35 | landley | Tincantools uses it because lots of work has gone into it. |
| 05:47.38 | CP|home | prpplague really really like buildroot |
| 05:47.45 | CP|home | he's been using it since it was invented, pratically |
| 05:48.04 | landley | And because it's the pet project of Eric Andersen who was uClibc and BusyBox maintainer when buildroot forked off. |
| 05:48.14 | landley | I've been _following_ it since it was invented. Not the same as using it... |
| 05:49.47 | RobotGuy | OK, so is there something better out there than buildroot? |
| 05:50.14 | CP|home | depends on your point of view |
| 05:50.16 | RobotGuy | I have to use buildroot to at least some extent because I am an evaluator for TCT. |
| 05:50.44 | RobotGuy | I also want a Sledge Hammer when they are available. :D |
| 05:50.57 | ds2 | landley: =) |
| 05:51.04 | CP|home | The Zaurus guys used to use buildroot - they transitioned to something called OpenEmbedded about 5 years ago |
| 05:51.14 | CP|home | I use OpenEmbedded, its powerful but complex - huge learning curve |
| 05:51.21 | RobotGuy | I am just looking at OpenEmbedded now.:) |
| 05:51.32 | CP|home | wouldn't recommend it if buildroot is taxing your patience |
| 05:51.35 | RobotGuy | I just finished installing all the prerequisits. |
| 05:51.53 | ds2 | learn to do it by hand first then come back and decide |
| 05:51.59 | RobotGuy | I never said buildroot is taxing my patience. It just doesn't do some things I wish I did. |
| 05:52.26 | RobotGuy | ds2: I am not new to build environments. I've run FreeBSD, Gentoo Linux, etc. |
| 05:52.41 | ds2 | then do it by hand =) |
| 05:52.52 | RobotGuy | I've also done LFS and BLFS. |
| 05:54.47 | RobotGuy | I might try landley's firmware thing. |
| 05:59.26 | *** join/#edev harpal (n=Harpal@124.125.76.8) |
| 06:12.41 | RobotGuy | Nobody has told me yet why buildroot should not be used. |
| 06:14.34 | landley | If you can get buildroot to do what you want, have fun with it. |
| 06:14.34 | landley | I couldn't. |
| 06:17.00 | ds2 | a simple shell script should sufficie |
| 06:17.32 | RobotGuy | buildroot almost does what I want now. |
| 06:18.23 | landley | My build system _is_ a series of shell scripts. |
| 06:18.32 | landley | one to download source. |
| 06:18.45 | landley | One to build host tools to add to the environment (which you can skip if you don't need it). |
| 06:18.50 | landley | One to build a cross compiler. |
| 06:19.01 | landley | One to build a minimal native environment with that cross compiler. |
| 06:19.10 | landley | One to package the result into an ext2 image. |
| 06:19.13 | landley | And one to run it under qemu. |
| 06:19.26 | landley | Plus there's an init script you can run under qemu to set up networking and populate /dev from /sys and so on. |
| 06:48.07 | RobotGuy | Well, I can't say I think much of the OpenEmbedded documentation, and the website needs work. However, I am getting it setup. |
| 07:03.26 | ds2 | hahah |
| 07:03.54 | ds2 | just jump in head first, if you survive and rise up.... you'll be that much more better off |
| 07:05.56 | *** join/#edev steliosk (n=Stelios@85.72.92.83) |
| 07:08.31 | anxt | landley: is it public? |
| 07:24.48 | landley | anxt: which, my build system? |
| 07:24.51 | landley | http://landley.net/code/firmware |
| 07:24.56 | landley | It has releases and everything. |
| 07:24.58 | landley | Mercurial archive. |
| 07:24.59 | landley | Mailing list. |
| 07:25.09 | landley | I blog about it. |
| 07:25.11 | anxt | is it free for me to use for commercial purposes? |
| 07:25.13 | landley | Doesn't get much more public than that. :) |
| 07:25.15 | landley | Yup. |
| 07:25.21 | anxt | ahh ok i will check it out then |
| 07:25.29 | anxt | buildroot is giving me headaches |
| 07:25.39 | landley | The scripts themselves are gplv2, but you probably don't ship those, so it doesn't really matter what the license on 'em is. |
| 07:25.52 | landley | The resulting images it builds are under whatever licenses the source packages are under. |
| 07:26.36 | landley | To clarify: modify the scripts all you want. Add whatever to 'em you want. I don't care. |
| 07:26.46 | anxt | fair enough |
| 07:26.50 | landley | If you ship the build scripts to customers, they're gplv2. |
| 07:26.54 | landley | If you use 'em in house, have fun. |
| 07:26.56 | anxt | i just am always wanting to be on the up and up |
| 07:28.18 | RobotGuy | ds2: I usually do jump in with both feet and both hands, when I do decide to do something. :) |
| 07:28.20 | landley | (I think this is more or less how buildroot is licensed anyway. :) |
| 07:29.06 | anxt | the company i work for and myslef believe in integrity |
| 07:29.34 | RobotGuy | I've already got my initial OpenEmbedded setup completed. |
| 07:29.50 | anxt | i am doing the openembedded now |
| 07:33.12 | anxt | ./build.sh x86 |
| 07:33.17 | anxt | lets see what happens lol |
| 07:34.48 | anxt | landley: thx for link :) |
| 07:35.57 | *** join/#edev CP|home (n=nobody@www.crouse-house.com) |
| 07:38.28 | landley | You're welcome. |
| 07:38.40 | landley | Lemme know if something breaks, I'm interested. |
| 07:39.01 | anxt | arch/um/sys-i386/user-offsets.c:7:22: error: asm/user.h: No such file or directory |
| 07:39.25 | anxt | is that part of the linux kernel tree? |
| 07:39.38 | anxt | lemme locate |
| 07:40.17 | anxt | locate doesn't take regexp? |
| 07:40.57 | anxt | i swear this embedded thing is different |
| 07:41.14 | *** join/#edev RobotGuy (n=RobotGuy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 07:42.17 | anxt | landley: anyway that is where it broke for me |
| 07:42.40 | anxt | i could try a diffent arch |
| 07:43.13 | anxt | same error :( |
| 07:51.01 | landley | Yeah, user mode linux sucks. |
| 07:51.10 | landley | I'm in the process of migrating away from it, going to gene2fs. |
| 07:51.16 | landley | For the moment, you can skip that step. |
| 07:52.00 | landley | In host-tools.sh comment out the user mode linux build |
| 07:52.13 | anxt | k |
| 07:52.42 | landley | Around line 50 there's an "if [which linux]". |
| 07:52.51 | landley | Change that to "if false". |
| 07:52.52 | anxt | yeah i see it |
| 07:52.58 | anxt | oh that works |
| 07:53.03 | anxt | easier than vi # :) |
| 07:53.09 | landley | User Mode LInux has always been a pain in the ass. |
| 07:53.26 | landley | I'm using it as a better fakeroot, but all I'm using it for is to do a loopback mount without root access. |
| 07:53.40 | landley | I assume you can package up an ext2 image from a directory full of files yourself? :) |
| 07:53.55 | anxt | maybe |
| 07:54.01 | landley | (That's the package-mini-native.sh script). |
| 07:54.07 | anxt | i have an ext2.fs |
| 07:54.09 | landley | You can also read that script to see how it works. |
| 07:54.12 | anxt | i could mount it format it |
| 07:54.23 | landley | Half the reason to make it all driven by bash scripts is you can read them to see what it's doing. |
| 07:54.25 | anxt | the whole geometry thing throws me for a loop |
| 07:54.35 | landley | Don't worry about the geometry. |
| 07:54.39 | landley | I'm not partitioning it. |
| 07:54.51 | anxt | so my next step? |
| 07:54.53 | landley | qemu has a -kernel option that lets you boot a kernel without a bootloader. |
| 07:55.01 | landley | re-run ./build.sh |
| 07:55.04 | anxt | surely the toolchain hasn't built |
| 07:55.09 | landley | It hasn't. |
| 07:55.15 | anxt | without arguments? |
| 07:55.23 | landley | ./build.sh i686 |
| 07:55.30 | anxt | ahh i was using x86 |
| 07:55.49 | landley | It would have barfed when it ran cross-compiler.sh |
| 07:56.01 | landley | Because there's no sources/configs/x86 |
| 07:56.05 | anxt | we may target an arm board |
| 07:56.14 | landley | ./build.sh armv4l |
| 07:56.16 | anxt | but i would like to see how this works |
| 07:56.19 | landley | or ./build.sh armv5l |
| 07:56.32 | anxt | pretty colours |
| 07:56.51 | landley | You'll notice that all the differences between the targets are contained in those config files. Mostly setting environment variables. |
| 07:56.53 | anxt | now it says launching smurf DDoS |
| 07:57.00 | landley | Shouldn't. :) |
| 07:57.04 | anxt | please enter ip: |
| 07:57.14 | landley | What did you run, again? |
| 07:57.23 | anxt | should i put in my IP ? |
| 07:57.27 | landley | no. |
| 07:57.32 | anxt | j/k lol |
| 07:57.35 | anxt | i couldn't resist |
| 07:57.39 | anxt | sorry |
| 07:57.43 | landley | The colors let you know what stage is running, at a glance. |
| 07:57.52 | landley | It's just an echo doing an ansi escape sequence. |
| 07:58.00 | anxt | how long did this take you to put together? |
| 07:58.05 | landley | This build system? |
| 07:58.13 | anxt | yeah |
| 07:58.21 | landley | I've been working on it on and off since 1999, but most of that was learning. |
| 07:58.23 | anxt | i was thinking of writing my own |
| 07:58.30 | anxt | but now i don't need to |
| 07:58.36 | landley | I started over from scratch (for the... third time?) in 2006. |
| 07:59.02 | landley | ./build.sh is a wrapper that calls the other scripts in order. |
| 07:59.07 | anxt | what we do at my work is build appliances that stream vorbis to our server |
| 07:59.19 | anxt | currently it is running from a cd |
| 07:59.25 | anxt | and working fairly well |
| 07:59.36 | landley | If you wanted to, you could go "./download.sh && ./host-tools.sh && ./cross-compiler.sh i686 && mini-native.sh i686 && package-mini-native.sh i686 |
| 07:59.37 | anxt | but i figure we could bring it down to a board |
| 07:59.53 | landley | Cool. |
| 08:00.18 | anxt | it would help drastically in shipping weight |
| 08:00.24 | landley | With User Mode Linux ripped out, the package-mini-native.sh script will fail. |
| 08:00.32 | landley | but it'll still have created the host root filesystem and tarred it up. |
| 08:00.37 | landley | That'll be in the build directory. |
| 08:00.52 | anxt | so i will have a kernel and a root? |
| 08:00.56 | landley | Yup. |
| 08:01.02 | *** join/#edev lyakh (n=lyakh@p57BD08F4.dip0.t-ipconnect.de) |
| 08:01.14 | anxt | for a bootstraep loader i need to make a partition? |
| 08:01.25 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 08:01.27 | landley | The kernel should be called build/zImage-i686 and the root should be build/mini-native-i686 |
| 08:01.32 | anxt | infact i am already set up for that |
| 08:01.42 | landley | Getting your board to boot it is your problem. |
| 08:01.50 | anxt | i made a qemu debian install and i use grub and another partition |
| 08:01.54 | landley | The package-mini-native.sh script packages it up for qemu. |
| 08:02.08 | anxt | mostly i am doing proof of concept right now |
| 08:02.22 | landley | There's a run-i686.sh script (put in the build directory) that would boot the ext2 image and kernel to a shell prompt under qemu. |
| 08:02.31 | anxt | nice |
| 08:02.37 | landley | But the packaging step for that currently uses user mode linux, which is an evil brittle piece of... |
| 08:02.38 | landley | Ahem. |
| 08:02.45 | anxt | shit? |
| 08:02.50 | landley | I've been using User Mode Linux since 2005, and it KEEPS BREAKING. |
| 08:02.58 | landley | I keep fixing it, and it keeps breaking in new and interesting ways... |
| 08:02.59 | landley | Sick of it. |
| 08:03.09 | landley | And I'm only using it for something that's easily replacable. |
| 08:03.13 | anxt | there was a thing about 2 years ago |
| 08:03.21 | anxt | linux kernel vs cdrecord |
| 08:03.22 | landley | (Right now you don't need it at all... if you're willing to run package-mini-native.sh as root.) |
| 08:03.30 | landley | Normal user can't do a loopback mount. |
| 08:03.35 | landley | anxt: cdrecord lost. :) |
| 08:03.37 | anxt | ahh |
| 08:03.48 | anxt | now i see where you are coming from |
| 08:04.15 | landley | My early build scripts had to be run as root. |
| 08:04.22 | landley | These days, I consider that unacceptable. |
| 08:04.35 | anxt | yes and no |
| 08:04.44 | landley | I may only lobotomize my laptop once every year and a half when running large complex things as root... |
| 08:04.54 | landley | But after the first 5 years (3 lobotomizations) it gets old... |
| 08:05.08 | anxt | my laptop is mac os x |
| 08:05.15 | anxt | which is tolerable |
| 08:05.27 | anxt | work gave me a powerbook |
| 08:05.29 | landley | Mirell sent me a patch to make FWL run on a MacOS X host. |
| 08:05.50 | anxt | i prefer to work on my linux box |
| 08:05.50 | landley | I'm slowly integrating it, but he worked around several things that I'm choosing to fix in a more thorough way. |
| 08:05.54 | anxt | it is more familiar |
| 08:06.10 | landley | My laptop's linux, but macosx is interesting. |
| 08:06.32 | landley | Which version did you download, anyway? |
| 08:06.39 | landley | The hg snapshot or the most recent release? |
| 08:06.44 | anxt | i installed ubntu linux ona friends laptop |
| 08:06.47 | anxt | he is loving it |
| 08:06.50 | landley | (Both should work, as far as I know...) |
| 08:06.54 | landley | Kubuntu 7.10 here. |
| 08:07.07 | anxt | release |
| 08:07.11 | landley | Ok. |
| 08:07.12 | anxt | 3.0.0 or something |
| 08:07.15 | landley | That's the safe choice. :) |
| 08:07.24 | landley | 0.3.0. |
| 08:07.24 | anxt | htats what i figures |
| 08:07.42 | anxt | i would like to fold libogg libvorbis and ices into this |
| 08:07.50 | landley | ok. |
| 08:07.56 | landley | Do you want to cross compile them, or build them natively? |
| 08:07.58 | anxt | and php |
| 08:08.07 | anxt | prolly build native |
| 08:08.11 | landley | My default, what this is doing is making a minimal native development environment. |
| 08:08.13 | anxt | within the new box |
| 08:08.25 | anxt | and then rm the gcc |
| 08:08.30 | landley | It's 7 packages: uClibc, busybox, gcc, binutils, make, linux, and bash. |
| 08:08.49 | landley | Back in 2005, I got that to rebuild itself under itself from source code. |
| 08:08.56 | anxt | well if i get get wget working i am laffing |
| 08:08.59 | landley | Unfortunately, busybox seems to have bit-rotted a bit since... |
| 08:09.03 | landley | busybox has wget. |
| 08:09.15 | landley | When you boot to a shell prompt in mini-native, run qemu-setup.sh |
| 08:09.19 | anxt | busybos ar is not working |
| 08:09.26 | landley | That sets up networking and mounts /sys and /proc and populates /dev and some other stuff. |
| 08:09.29 | anxt | there is no -u for ar in busybox |
| 08:09.35 | landley | busybox ar is read only. |
| 08:09.39 | landley | I used the binutils version. |
| 08:09.41 | anxt | broke libogg |
| 08:09.56 | anxt | i read the makefile for a bit didn't see it |
| 08:10.00 | landley | I'm in the process of debugging the "rebuild itself under itself" stage now. |
| 08:10.16 | landley | just make sure the binutils ar is first in the $PATH. |
| 08:10.32 | anxt | lol that might work! |
| 08:10.50 | landley | I honestly did make this work back in 2005. (Before the restart.) |
| 08:10.59 | landley | The old version was based on Linux From Scratch 3.x. |
| 08:11.08 | anxt | funny thing is i am getting paid to do this |
| 08:11.12 | anxt | lmao |
| 08:11.33 | anxt | i love tweaking linux out |
| 08:11.33 | landley | It built a minimal system and chrooted into it (LFS chapter 5), and then built the final system (LFS chapter 6) under that. |
| 08:11.47 | landley | The minimal system I put together was uClibc and busybox based, these 7 packages. |
| 08:12.00 | RobotGuy | landley: Ah yes, I remember doing that for LFS. |
| 08:12.06 | landley | Instead of doing a chroot, I booted up User Mode Linux and did a hostfs mount of the appropriate host subdirectory. |
| 08:12.26 | landley | So I could run code as root without having to be root, via UML. :) |
| 08:13.02 | landley | Then I got interested in cross compiling, so instead of building uClibc toolchains I built cross compiler uClibc toolchains. |
| 08:13.09 | landley | And instead of using UML I used qemu. |
| 08:13.21 | landley | But to get an ext2 image, without root access, I sucked in UML as a quick hack. |
| 08:13.28 | landley | And haven't fixed it yet. It's on my todo list. :) |
| 08:15.09 | anxt | i don't know what uml is |
| 08:15.17 | landley | User Mode Linux. |
| 08:15.21 | anxt | ahh |
| 08:15.25 | landley | The above "pain in the ass". |
| 08:15.32 | anxt | lol |
| 08:15.39 | landley | It's... brittle. |
| 08:15.46 | anxt | well it is still compiling |
| 08:15.52 | landley | Good sign, I suppose. |
| 08:15.54 | landley | what color is it? |
| 08:16.02 | anxt | i had some issues with gentoo and lipmpv4 |
| 08:16.11 | anxt | and that other fragile one |
| 08:16.17 | anxt | purple? |
| 08:16.26 | anxt | ffmpeg iirc |
| 08:16.28 | landley | It's building the mini-native then. |
| 08:16.30 | anxt | it liked to break |
| 08:16.32 | landley | The cross compiler build is orange. |
| 08:16.41 | landley | The mini-native build is purple. |
| 08:16.54 | anxt | i am installing debain in an arm |
| 08:16.55 | landley | So it's successfully built a cross compiler, and is now building you a root filesystem with that. |
| 08:16.57 | anxt | it should be done |
| 08:17.02 | anxt | then it will speed up |
| 08:17.09 | anxt | nice |
| 08:17.12 | landley | If you just want to cross compiler more stuff, you can edit mini-native.sh. |
| 08:17.26 | anxt | i will |
| 08:17.28 | landley | If you want to natively compile more stuff, boot mini-native under qemu and build packages natively in there. |
| 08:17.33 | anxt | but first i wanna boot this |
| 08:17.40 | landley | You may have to bootstrap your way up a bit (build zlib and such first). |
| 08:17.58 | anxt | ok cool |
| 08:18.34 | anxt | i will rubn into ar issues again? |
| 08:18.39 | landley | if you hit a blocking issue, keep in mind I have done this before, so email it and I'll probably recognize it. |
| 08:18.47 | anxt | i was thinking of trying to statically link one |
| 08:18.49 | landley | The fix to the ar issue is keep the binutils version of ar. ;) |
| 08:19.14 | anxt | great success! |
| 08:19.15 | landley | I _think_ I'm keeping the binutils version in mini-native, but haven't checked. |
| 08:19.29 | anxt | waterpanties! |
| 08:19.43 | landley | (I may have to feed busybox a config file and tell it not to build stuff if it's doing things like overwriting the binutils ar.) |
| 08:19.50 | landley | Cool...? |
| 08:19.55 | landley | (waterpanties?) |
| 08:20.05 | anxt | oh it is from a dumb movie |
| 08:20.08 | anxt | borat |
| 08:20.11 | landley | ok |
| 08:20.13 | landley | Haven't seen it. |
| 08:20.19 | anxt | don't bother |
| 08:20.21 | landley | I must go fend off an aggressive cat for a bit. |
| 08:20.23 | anxt | lol |
| 08:20.27 | landley | She is being affectionate. |
| 08:20.32 | landley | And interfering with my typing. |
| 08:20.40 | anxt | feline bane |
| 08:20.51 | landley | Check out the website, I've got some design docs on there, and need to write more... |
| 08:20.59 | anxt | i will |
| 08:21.01 | anxt | and thanks |
| 08:21.31 | landley | You're welcome. |
| 08:21.40 | landley | I look forward to any feedback you have... |
| 08:22.00 | anxt | will do |
| 08:58.06 | *** join/#edev redblue (i=star@ppp137.108-253-207.mtl.mt.videotron.ca) |
| 09:25.00 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 10:52.07 | *** join/#edev hw_ (n=hw@p578b3905.dip0.t-ipconnect.de) |
| 11:23.35 | *** join/#edev dijenerate (n=dijenera@72.22.133.141) |
| 11:49.58 | *** join/#edev RobotGuy (n=RobotGuy@dsl093-038-072.pdx1.dsl.speakeasy.net) |
| 12:03.10 | *** join/#edev CP|home (n=nobody@www.crouse-house.com) |
| 12:28.06 | *** join/#edev dijenerate (n=dijenera@66.54.118.16) |
| 13:15.05 | *** join/#edev roxfan (n=dunno@183.217-200-80.adsl-dyn.isp.belgacom.be) |
| 13:39.50 | *** join/#edev bbradley (n=bbradley@82-32-35-42.cable.ubr05.azte.blueyonder.co.uk) |
| 13:55.13 | dbtid | is anyone here using (or familiar with) AT91RM9200 SPI accesses? |
| 13:55.33 | dbtid | i'm trying to write /dev/mtdblock/2, which is my u-boot environment, from linux |
| 13:55.49 | dbtid | i made it writable by removing the _WRITEABLE mask from mask_flags |
| 13:55.57 | dbtid | but the kernel comes back and says operation not permitted |
| 13:59.10 | *** join/#edev dijenerate (n=dijenera@72.51.67.14) |
| 14:10.27 | Mirell | Don't touch the kernel that way, it doesn't like it. |
| 14:12.04 | *** join/#edev prpplague (n=dave@mail.americanmicrosystems.com) |
| 14:13.23 | Mirell | ...I obviously don't know the answer to your question, sorry. |
| 14:13.29 | dbtid | sigh |
| 14:13.44 | Mirell | However, people are heading to work about now. |
| 14:13.58 | Mirell | So not much activity. |
| 14:14.06 | Mirell | In fact, I should do the same. *foof* |
| 14:16.31 | prpplague | hehe |
| 14:16.55 | dbtid | AT91RM9200 and SPI writes to dataflash, using kernel 2.6.20 |
| 14:17.10 | dbtid | i get operation not permitted, even though that partition is writable. |
| 14:17.17 | dbtid | when i try to write using dd |
| 14:17.20 | dbtid | i can read using dd |
| 14:17.25 | dbtid | writing doesn't seem to happen |
| 14:17.44 | prpplague | dbtid: sounds like the dataflash might be write protected |
| 14:17.52 | dbtid | i can write to it via u-boot |
| 14:18.31 | dbtid | i don't know if the chip has an internal write-protect that must be released |
| 14:18.34 | dbtid | that's worth looking into |
| 14:18.37 | prpplague | dbtid: you'll have to look at the schematic for the device, i suspect there is a write protect line that is tied to a gpio |
| 14:18.45 | dbtid | there isn't |
| 14:18.51 | dbtid | it's our board |
| 14:19.03 | dbtid | u-boot can write to it just fine |
| 14:19.17 | dbtid | but there might be an internal block |
| 14:19.29 | prpplague | dbtid: u-boot and the kernel are two different animals, jsut because something works in u-boot doesn't mean its configured for the kernel |
| 14:19.37 | dbtid | exactly |
| 14:19.40 | dbtid | hence my question |
| 14:19.50 | prpplague | dbtid: whats the part number on the spi flash? |
| 14:19.51 | dbtid | but see you were talking about a write protect |
| 14:20.05 | dbtid | since i did all the updates for u-boot to make it work, i know there's no physical write-protect |
| 14:20.15 | dbtid | the chip is a 011 version |
| 14:20.23 | dbtid | i'll have to look at the schematic; one sec |
| 14:21.24 | RobotGuy | prpplague: Greetings |
| 14:21.36 | dbtid | AT45DB011B |
| 14:21.48 | dbtid | the pin /WP is pulled high |
| 14:23.54 | prpplague | RobotGuy: greetings |
| 14:24.49 | prpplague | dbtid: the driver you are using for the linux kernel, did you write it? or was it a stock kernel driver? |
| 14:25.05 | *** join/#edev emil|k (n=emil@f054012199.adsl.alicedsl.de) |
| 14:25.17 | dbtid | stock driver, but i think perhaps ulf had made some changes |
| 14:25.26 | dbtid | i put an email onto the linux arm mailing list |
| 14:25.30 | dbtid | i'll see what comes back |
| 14:25.39 | dbtid | iirc it won't work from userland, but i don't recall why |
| 14:25.49 | dbtid | i thought perhaps going through the device it might work differently |
| 14:25.58 | dbtid | but there is an at91_dataflash_write entry point into the kernel |
| 14:26.03 | prpplague | dbtid: i'd have a look at the driver's WP section, i suspect that most people would use a gpio to /WP , so the driver might expect taht |
| 14:26.06 | dbtid | i'm just going to have to delve into it |
| 14:26.21 | dbtid | well, it's our board, and /WP is pulled to 5V |
| 14:26.26 | dbtid | so it's not write protected |
| 14:26.29 | dbtid | it's not tied to a GPIO pin |
| 14:26.52 | dbtid | AFAIK it's just something someone didn't get to finish |
| 14:27.20 | dbtid | i don't have my kernel sources available ATM, but AFAIK there's nothing about a WP in the kernel. |
| 14:27.31 | dbtid | there's a data structure that defines the access to the chip |
| 14:27.44 | dbtid | it used to have a WRITABLE mask set, meaning no writes allowed |
| 14:27.59 | dbtid | that's checked and writes won't happen if the bit is set |
| 14:28.12 | dbtid | but now i've turned off the bit, so it should allow the writes |
| 14:28.19 | dbtid | so some other part of the kernel is blocking it |
| 14:29.29 | prpplague | okie dokie |
| 14:32.53 | prpplague | dbtid: currently we don't have many people using the AT91 series here |
| 14:33.01 | dbtid | i'm kind of surprised |
| 14:33.16 | dbtid | unfortunately, i can't remember where i was that had people that were doing it |
| 14:33.16 | prpplague | dbtid: why? |
| 14:33.18 | dbtid | i thought it was here |
| 14:33.29 | dbtid | i know cbrake uses AT91's in some of his projects |
| 14:33.34 | dbtid | but he's used the SAM IIRC |
| 14:33.54 | prpplague | ah right, i forgot about cbrake using them |
| 14:34.17 | prpplague | dbtid: what kind of qty are you using for the AT91's? |
| 14:34.29 | dbtid | not much |
| 14:34.36 | dbtid | we buy them in lots of 1K I think |
| 14:34.41 | dbtid | but i'm not on that side of things |
| 14:34.51 | prpplague | dbtid: mind if i ask how much you are paying per item ? |
| 14:35.01 | dbtid | i have no way of knowing |
| 14:35.06 | prpplague | ahh |
| 14:35.06 | dbtid | i'm in engineering, not manufacturing |
| 14:35.20 | dbtid | if you really want to know, i can fire off an email |
| 14:35.41 | prpplague | dbtid: hehe, a engineer has to know if the parts he is planning to use are too expensive, hehe |
| 14:35.47 | dbtid | sigh |
| 14:35.54 | dbtid | i do software, not hardware |
| 14:36.03 | dbtid | i couldn't care really if it cost $0.50 or $5.00 |
| 14:36.06 | dbtid | not my concern |
| 14:36.10 | prpplague | dbtid: naw i was jsut curious as when we priced out the AT91 it was about double what some other SoC's were |
| 14:36.32 | *** join/#edev CP|home (n=nobody@www.crouse-house.com) |
| 14:36.47 | RobotGuy | prpplague: Did you look at Freescale? I can't remember. |
| 14:36.49 | dbtid | i think we were paying about $11 for it |
| 14:36.55 | prpplague | RobotGuy: yea |
| 14:37.07 | prpplague | dbtid: yea that sounds about what we were looking at as well |
| 14:37.17 | dbtid | it's reasonable for what we're doing |
| 14:37.28 | dbtid | we're gonna make a new box based on this same one here in the near future |
| 14:37.45 | prpplague | "new box" ? |
| 14:37.51 | dbtid | yes |
| 14:38.07 | dbtid | we have a multilink broadcast receiver for satellite signals we build now |
| 14:38.20 | prpplague | dbtid: ahh |
| 14:38.23 | dbtid | we're going to make a new device based on the same mobo we have in that one for something else |
| 14:38.27 | dbtid | the deisgn works |
| 14:38.34 | dbtid | the 9200 isn't the greatest chip in the world |
| 14:38.43 | prpplague | dbtid: ahh ok, yea i doubt price is much of a factor then |
| 14:38.43 | dbtid | but we have a lot of experience with it now |
| 14:38.48 | dbtid | no, it's really not |
| 14:39.04 | dbtid | $11 or $20 wouldn't make too much of a difference |
| 14:39.38 | prpplague | $11.01 or $11.05 is a big difference to us |
| 14:39.44 | dbtid | ouch |
| 14:47.24 | prpplague | RobotGuy: we looked at freescale as well, but their pricing and features weren't the range either |
| 14:48.09 | RobotGuy | prpplague: Ah, OK. That's too bad because I'd love to get my hands on an i.MX31 board. |
| 14:49.19 | prpplague | RobotGuy: thats why you can't find any inexpensive boards, the small shops don't want to invest in making boards with those |
| 14:49.23 | prpplague | RobotGuy: same goes for omap |
| 14:49.32 | RobotGuy | I'm still thinking about putting together a COG stack. |
| 14:49.55 | prpplague | RobotGuy: the min qty is so rediculously high no one can get into using them |
| 14:50.45 | RobotGuy | prpplague: I understand, and that's too bad. It keeps otherwise great chips out of the hands of people like me. |
| 14:50.58 | prpplague | ineed |
| 14:51.11 | *** join/#edev GPSFan (n=kenm@12.10.255.246) |
| 14:51.11 | *** mode/#edev [+o GPSFan] by ChanServ |
| 14:51.19 | prpplague | GPSFan: greetings |
| 14:51.21 | RobotGuy | prpplague: That's alright though, because I found Tin Can Tools. :) |
| 14:51.52 | prpplague | i'm hope'n that the nail board will be back by the end of the week from the proto house |
| 14:51.57 | GPSFan | prpplague: morning pilgrim.. |
| 15:00.56 | *** join/#edev chouimat|work (n=mchouina@209.217.106.98) |
| 15:01.41 | chouimat|work | morning |
| 15:03.50 | RobotGuy | prpplague: What again is the nail board supposed to be? |
| 15:04.22 | RobotGuy | I know it's really thin, like a fat pen. |
| 15:04.23 | prpplague | http://www.elinux.org/Nail_Board |
| 15:05.11 | chouimat|work | nice |
| 15:06.07 | RobotGuy | The case looks like a mini-coffin. |
| 15:08.35 | RobotGuy | I was going to go to bed, but I think I will just stay up now. I may have to go to the ER. |
| 15:11.25 | Mirell | RobotGuy: Wah? |
| 15:13.52 | RobotGuy | Mirell: My arm is starting to swell up due to some real bad scratches. |
| 15:14.22 | Mirell | Ow. |
| 15:14.23 | prpplague | MonMotha: ping |
| 15:15.02 | *** join/#edev CP|home_ (n=nobody@www.crouse-house.com) |
| 15:23.34 | prpplague | i don't people realize that WIAV is intending on sueing everyone who uses 802.11 |
| 15:24.02 | dbtid | hmm? |
| 15:24.26 | dbtid | low-power tv? |
| 15:24.53 | prpplague | dbtid: patent troll |
| 15:25.37 | prpplague | http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 15:25.46 | prpplague | http://trolltracker.blogspot.com/2008/01/4-interesting-new-cases-from-last-2.html |
| 15:25.48 | dbtid | ty |
| 15:26.05 | RobotGuy | prpplague: I only need automounter v4, right? |
| 15:26.18 | prpplague | purl: wiav is a patent troll with information at http://trolltracker.blogspot.com/2008/01/4-interesting-new-cases-from-last-2.html and http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 15:26.18 | purl | okay, prpplague |
| 15:26.31 | prpplague | RobotGuy: automounter for what? |
| 15:26.36 | prpplague | ~wiav |
| 15:26.37 | purl | well, wiav is a patent troll with information at http://trolltracker.blogspot.com/2008/01/4-interesting-new-cases-from-last-2.html and http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 15:30.13 | RobotGuy | prpplague: The Kernel Automounter for file systems and such. |
| 15:32.25 | prpplague | dbtid: basically WIAV says that 802.11 violates their two patents and they've been hitting up people for "license" fees |
| 15:32.25 | RobotGuy | Busybox's route command is lame. |
| 15:32.53 | RobotGuy | prpplague: That is crazy. Wireless has been around for quite awhile. |
| 15:33.31 | prpplague | RobotGuy: yea, and the IEEE 802.11 standard was cleared patent wise from the start |
| 15:35.24 | RobotGuy | prpplague: So WHERE do these WAIV people get off trying to extort money from companies that have been making wireless stuff for awhile? |
| 15:36.44 | prpplague | RobotGuy: that's why they are called patent trolls |
| 15:36.59 | prpplague | RobotGuy: pretty popular way of making money these days |
| 15:37.01 | RobotGuy | WAIV is just betting that a lot of companies, especially smaller ones, won't risk litigation and will settle up with them. |
| 15:37.42 | prpplague | RobotGuy: yep and they've been doing well untill they tried to hit us up |
| 15:38.12 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 15:38.12 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 15:39.13 | RobotGuy | prpplague: WAIV has tried to hit TCT up? |
| 15:39.26 | prpplague | CosmicPenguin: you guys at amd hold any classes on designing hardware with the geode? |
| 15:39.41 | prpplague | RobotGuy: hehe you didn't read the url's very well |
| 15:40.12 | CosmicPenguin | prpplague: no - but we have full service assistance |
| 15:40.35 | prpplague | CosmicPenguin: ? like we can send the schematic to you guys for review or something? |
| 15:40.35 | CosmicPenguin | prpplague: usually what happens is that people end up taking the CPU and chipset pages from our reference schematicsa and drop them into their own design |
| 15:40.42 | CosmicPenguin | we have all the libraries and everything to help you do that |
| 15:40.43 | prpplague | ahh |
| 15:41.55 | prpplague | RobotGuy: AML is parent company of TCT |
| 15:42.11 | RobotGuy | prpplague: OH! I had forgotten that. |
| 15:42.16 | CosmicPenguin | prpplague: yeah - you can send the design to us for review - in fact, we recommend that |
| 15:42.45 | prpplague | CosmicPenguin: we've been using a third party board for our geode stuff |
| 15:43.00 | prpplague | CosmicPenguin: just working out the details for possibly doing a custom board |
| 15:43.22 | prpplague | CosmicPenguin: not sure if we want to get into the x86 board dev area |
| 15:44.52 | RobotGuy | prpplague: It would be nice to have a stackable board set, something like Virtual COGs has created. |
| 15:44.53 | prpplague | CosmicPenguin: i was looking at some of the ETX modules |
| 15:45.09 | prpplague | RobotGuy: you mean for the hammer? |
| 15:46.02 | RobotGuy | prpplague: Maybe for Sledge Hammer. I really like the stacking system of adding on new modules. |
| 15:46.38 | prpplague | RobotGuy: no reason you can't stack with the hammer, pretty easy to do with .1" board to board conectors |
| 15:46.52 | CosmicPenguin | yeah, those stackers are fun |
| 15:47.25 | RobotGuy | prpplague: True, but I am thinking more along the lines of a Carrier board in a square form factor with all the connectors and stacking headers for adding more modules. |
| 15:47.42 | RobotGuy | (even additional Hammers) |
| 15:47.50 | prpplague | RobotGuy: yea like i said easy to do |
| 15:48.05 | *** join/#edev emil|k (n=emil@e179235076.adsl.alicedsl.de) |
| 15:48.14 | RobotGuy | prpplague: For you, sure, but not for me and I am sure others like me. |
| 15:48.29 | prpplague | RobotGuy: why? its just 40 pins? |
| 15:49.02 | RobotGuy | prpplague: I am not just thinking about the pins, but about the connectors and maybe additional peripherals. |
| 15:49.23 | prpplague | i guess i'm not understanding what you are asking for |
| 15:50.07 | RobotGuy | prpplague: I have been eyeing the COG stack stuff as a possibility for creating my portable computer/terminal/controller instead of getting a small laptop. |
| 15:50.30 | prpplague | CosmicPenguin: i suspect the cost for doing a geode board is gonna be out of our range, but i gotta research it anyway |
| 15:51.02 | CosmicPenguin | you need to find yourself an field engineer |
| 15:51.05 | CosmicPenguin | you probably already have one |
| 15:51.14 | RobotGuy | prpplague: Something with a nice stacking display board for instance. Like the Portlable Computing COG. |
| 15:51.39 | prpplague | CosmicPenguin: an AMD field engineer? |
| 15:51.44 | CosmicPenguin | yeah |
| 15:52.27 | prpplague | RobotGuy: you can add all of that already to the hammer by stacking |
| 15:52.51 | prpplague | CosmicPenguin: i doubt we have a field engineer assigned |
| 15:53.05 | prpplague | CosmicPenguin: we've never purchased anything directly from AMD |
| 15:53.30 | RobotGuy | prpplague: I know it can be added, but there aren't any stacking Hammer boards now. |
| 15:54.03 | prpplague | RobotGuy: sorry i don't understand what you are asking for |
| 15:54.20 | CosmicPenguin | prpplague: I thought you guys had a Geode design before? Was I wrong? |
| 15:54.33 | prpplague | CosmicPenguin: we used a third party board |
| 15:54.44 | prpplague | CosmicPenguin: off the shelf device we just intergrated into the product |
| 15:55.21 | CosmicPenguin | oh, okay |
| 15:55.30 | prpplague | CosmicPenguin: basically it was an attempt to see if we could get any cost savings by using off the shelf products |
| 15:55.44 | prpplague | CosmicPenguin: with a resounding no |
| 15:59.38 | CosmicPenguin | I'm not 100% sure how that whole process works |
| 16:00.03 | RobotGuy | prpplague: Buildroot question.. How hard would it be to add a global parameter for additional build options (like --prefix, etc)? |
| 16:01.07 | prpplague | RobotGuy: not hard at all, but i assume you want to use a prefix for installation of files, and thats usually done in the package makefile |
| 16:02.42 | RobotGuy | prpplague: Yes, but I want it to be used for ALL packages, not just a single one for instance. |
| 16:03.02 | prpplague | RobotGuy: to what end? |
| 16:04.04 | RobotGuy | prpplague: I don't want config files going into /etc for software that goes into /usr, because that is my thumb drive. |
| 16:05.33 | prpplague | then it doesn't sound like you want it for all packages just for the ones that put stuff in /usr |
| 16:05.52 | prpplague | RobotGuy: why are you doing two seperate builds? |
| 16:06.32 | RobotGuy | prpplague: One is for my initrd. The other is for software that goes on my thumb drive. |
| 16:07.22 | prpplague | RobotGuy: sorry guy, i guess we aren't in sync today, because i'm not understanding what you are trying to do |
| 16:08.04 | RobotGuy | I seem to not be in sync with anyone lately. :( |
| 16:09.01 | RobotGuy | prpplague: I have a GREP build error: http://www.pastebin.ca/868614 |
| 16:10.43 | prpplague | yea gettext is always a nightmare |
| 16:10.53 | RobotGuy | prpplague: I don't like some of what busybox has/does. |
| 16:11.14 | prpplague | okie dokie |
| 16:11.30 | *** join/#edev ico2 (n=ico2@modem-1676.cougar.dialup.pol.co.uk) |
| 16:11.32 | CosmicPenguin | There are some grumbly things in busybox, but I've never had a problem with the grep |
| 16:12.12 | prpplague | RobotGuy: http://buildroot.busybox.net/cgi-bin/viewcvs.cgi/trunk/buildroot/package/grep/grep.mk?rev=19477&r1=19061&r2=19477 |
| 16:12.15 | prpplague | CosmicPenguin: me either |
| 16:12.21 | RobotGuy | prpplague: I'm being a tester too, finding broken things. I want to replace most of what busybox does with the standard utilities. |
| 16:12.35 | prpplague | RobotGuy: uh why? |
| 16:12.47 | *** join/#edev ccherrett (n=chris@S0106000c413c715f.ed.shawcable.net) [NETSPLIT VICTIM] |
| 16:12.48 | prpplague | RobotGuy: thats the whole purpose of busybox is to go the other way |
| 16:13.04 | prpplague | RobotGuy: to replace standard utils with smaller ones |
| 16:13.42 | prpplague | RobotGuy: you know you don't have to have any special access to check browse the buildroot source |
| 16:18.17 | *** join/#edev samuelr (n=samuelr@dsl093-061-094.pit1.dsl.speakeasy.net) |
| 16:19.10 | RobotGuy | prpplague: I don't like some of what busybox does. |
| 16:24.10 | prpplague | RobotGuy: okie dokie |
| 16:24.57 | prpplague | RobotGuy: if you run into a problem with buildroot, its a good idea to check the buildroot repo first to see if someone else has already fixed the problem |
| 16:28.17 | CosmicPenguin | and patches are always welcome - those guys have never been shy about taking enhancements |
| 16:28.28 | CosmicPenguin | their whole thing is just sharing as much of the backend stuff as possible for size |
| 16:34.58 | RobotGuy | prpplague: When will the Hammer stuff be in buildroot so we can use the latest versions of buildroot? |
| 16:35.51 | prpplague | RobotGuy: you can add it in any time you want, there is nothing specific other than the sub dir with the basic rootfs skeleton |
| 16:36.21 | prpplague | RobotGuy: most users of buildroot tend to get a snapshot and stick with it |
| 16:36.37 | prpplague | RobotGuy: generally buildroot is too much of a moving target to keep maintaining |
| 16:37.04 | RobotGuy | prpplague: Don't the buildroot folks fix things or take patches from others that fix things? |
| 16:37.21 | prpplague | yea we do |
| 16:38.34 | prpplague | CosmicPenguin: any rules on what kind of drivers go into linux/drivers/misc ? |
| 16:38.58 | CosmicPenguin | I don't believe so |
| 16:39.09 | CosmicPenguin | Assuming they really don't belong elsewhere |
| 16:41.11 | prpplague | CosmicPenguin: yea, i had a look there, it seems this particular driver looks to fit with whats there now, but i was just curious about it |
| 16:43.06 | *** join/#edev T0mW (n=Tom@24.238.86.158.res-cmts.sth.ptd.net) |
| 16:53.02 | prpplague | T0mW: hey stranger |
| 16:54.35 | chouimat|work | hi prpplague CosmicPenguin T0mW |
| 16:54.47 | prpplague | greetings |
| 16:55.03 | T0mW | prpplague: ain't no stranger than you, but I'm getting better with age. |
| 16:55.24 | CosmicPenguin | amusing |
| 16:55.25 | T0mW | chouimat|work: ooo! you have a *job*?! |
| 16:55.43 | chouimat|work | T0mW: for 2 months now ... www.axentra.com |
| 16:55.59 | T0mW | CosmicPenguin: didn't say which way I was going with age though |
| 16:57.28 | T0mW | prpplague: I seem to be having trouble with that EN28J60 driver that I will have to look into. It seems that many times it won't initialize the hardware into a working state. Sometimes I have to completely power the unit off before it will work. |
| 16:58.03 | T0mW | prpplague: how stable have you found it to be? |
| 16:58.06 | prpplague | T0mW: yea, i'm aware of that issue, i suspect that it is an init timming problem |
| 16:58.13 | prpplague | T0mW: i've seen it occur |
| 16:58.40 | prpplague | T0mW: if you have the driver as a module you can load the module, remove it , what 10 seconds, load the module again and it will init |
| 16:59.10 | T0mW | prpplague: ok, good. I have a new board it is on so this points me towards the driver rather than a noise issue with power supply or electrical issues with the SPI signals. |
| 16:59.32 | prpplague | T0mW: yea |
| 16:59.53 | prpplague | T0mW: that driver was quick and dirty, so i'm sure there is something on the init side that might be a problem |
| 17:01.29 | T0mW | prpplague: I corrected the problem of "console not found" when booting. if /dev dir is empty, then there is no "console", populate the /dev dir with: console, ttySAC0, and null devices instead of leaving it blank, the problem goes away and you see the startup messages from rcS + rc |
| 17:01.48 | prpplague | T0mW: hehe, indeed |
| 17:01.59 | prpplague | T0mW: gotta have some device nodes, hehe |
| 17:02.21 | T0mW | prpplague: yes, I saw that it does a lot of punching of bytes into registers without any loops to wait for things |
| 17:02.30 | T0mW | prpplague: "experience" |
| 17:02.57 | T0mW | actually, experience is assumed by the amount of trivia you learn, heh |
| 17:03.02 | prpplague | hehe |
| 17:03.50 | T0mW | well, I've got postgresql running in this new board. I've got to test it and see how it will behave. |
| 17:04.24 | Mirell | Hmm.. |
| 17:04.35 | T0mW | a 200MHz clock + 32bit wide bus to the RAM does make for a snappy little system. |
| 17:05.22 | prpplague | wow, postgresql? |
| 17:05.31 | T0mW | prpplague: license |
| 17:05.32 | prpplague | why not sqlite? |
| 17:06.25 | T0mW | prpplague: not sure if sqlite would be appropriate for what could be in that database. I would expect to have as many as 200..500 rows of data, or more, at any one time. |
| 17:07.10 | prpplague | yea, sqlite handles that |
| 17:07.34 | T0mW | prpplague: the primary purpose of that board is to host the SQL server, run the SPI circuit to 14 cards, and run some other processes. |
| 17:08.30 | T0mW | prpplague: I would also need a to provide the data to MicroSoft clients (ODBC???) |
| 17:08.49 | prpplague | T0mW: yea sqlite has an ODBC filter iirc |
| 17:08.51 | T0mW | do database joins, etc |
| 17:09.57 | T0mW | prpplague: I can always change to sqlite later, I have not bothered to research it. |
| 17:10.25 | prpplague | T0mW: yea, get your proof-of-concept working first, hehe |
| 17:11.09 | prpplague | T0mW: the 2440 has 289 pins the 2460 jumps up to 416 |
| 17:11.23 | T0mW | prpplague: I have some resistance from others in the client company regarding the use of an SQL connection to gather the Alarm data from these units. They are pushing for using an proprietary, wacko, protocol that they developed. |
| 17:12.58 | T0mW | prpplague: I'm not interested in finding out what the limitation of the proprietary protocol would be. Nor do I want to waste time writing my side of the driver that would be needed, etc. |
| 17:16.09 | prpplague | T0mW: hehe |
| 17:22.03 | prpplague | i love the fact that samsung no longer provides datasheets |
| 17:23.34 | SpeedEvil | The shiny website makes up for it! |
| 17:24.20 | T0mW | prpplague: so WTF are we now supposed to get datasheets? |
| 17:24.36 | SpeedEvil | That reminds me - I sent a request to access some datasheets a few days ago |
| 17:24.52 | SpeedEvil | tomw: it's possible to find most of them - partnumber datasheet in google |
| 17:25.26 | SpeedEvil | finds most stuff. Unless it's critical, or you're picky, and like want the guaranteed version with all current errata. |
| 17:26.38 | prpplague | T0mW: no, clue, we've been trying to get updated datasheets for about a week now we no luck and we are are an "official" user of samsung products |
| 17:27.54 | prpplague | "We provide limited information for product information security. After checking your information, we will let you know by email whether you are authorized to have the access to the product information within 24 hours." |
| 17:28.33 | SpeedEvil | They haven't got back to me either - though I'm a nobody. (I went through the automated procedure) |
| 17:29.14 | SpeedEvil | Your rep can't get datasheets? |
| 17:31.09 | SpeedEvil | I wonder if they could be seeing pressure from huge OEMs to restrict information to restrict the ease of hacking devices. Doesn't seem likely. |
| 17:31.36 | T0mW | SpeedEvil: exactly |
| 17:31.50 | prpplague | SpeedEvil: i.e. apple with the iphone |
| 17:31.54 | SpeedEvil | Yeah. |
| 17:32.11 | SpeedEvil | The notion seems ... flawed. |
| 17:32.19 | prpplague | i know mattel wasn't very happy with our hacking of the juicebox and the usage of the datasheets from samsung |
| 17:33.28 | SpeedEvil | p2p search 'secure digital card' or 'bluelab csr' |
| 17:33.30 | SpeedEvil | or ... |
| 17:38.05 | prpplague | i love the fact that samsung's request form is IE only |
| 17:38.18 | prpplague | SpeedEvil: ? |
| 17:38.24 | SpeedEvil | It worked for me (apparantly) in FF/linux |
| 17:38.53 | SpeedEvil | I mean - secure documents/packages are available on peer-peer networks |
| 17:39.27 | prpplague | SpeedEvil: you were able to request documents via samsung request page using something other than IE? |
| 17:39.34 | SpeedEvil | yes. |
| 17:39.52 | prpplague | SpeedEvil: i tried firefox on both linux and windows and it wouldn't let me add anything to the request list |
| 17:39.55 | SpeedEvil | At least I got through to the 'ok - we got your request, we will email you within the week' one |
| 17:40.01 | SpeedEvil | ff1.5 |
| 17:40.14 | prpplague | must be a java thing |
| 17:40.52 | prpplague | SpeedEvil: we have to get our documents legally |
| 17:40.55 | *** join/#edev chouimat|work (n=mchouina@209.217.106.98) |
| 17:40.59 | prpplague | SpeedEvil: business you know |
| 17:42.12 | SpeedEvil | prpplague: I wasn't meaning that was a solution. I just downloaded them to see if security was real as I suspected it wasnt'. |
| 17:42.35 | SpeedEvil | I was meaning that if the documents leak onto P2P the security is pointless from an anti-hacking point of view |
| 17:47.47 | *** join/#edev chouimat|work (n=mchouina@209.217.106.98) |
| 17:56.44 | *** join/#edev gandhijee (n=user@mail.win-ent.com) |
| 18:05.39 | *** join/#edev ico2 (n=ico2@modem-2969.chimpanzee.dialup.pol.co.uk) |
| 18:15.35 | prpplague | SpeedEvil: ahh gotcha |
| 18:21.47 | T0mW | prpplague: either I am losing sense of time with all this documentation crap I'm forced into writing today, or you had a very short lunch. |
| 18:22.17 | T0mW | :( |
| 18:22.30 | prpplague | T0mW: just drove to the house, checked the mail, and had a cup of coffee |
| 18:22.41 | T0mW | uhhuh |
| 18:23.21 | T0mW | drat, it is snowing |
| 18:24.05 | prpplague | hehe |
| 18:24.43 | *** mode/#edev [+o prpplague] by ChanServ |
| 18:55.47 | T0mW | prpplague: anything good in the mailbox? |
| 18:56.31 | prpplague | T0mW: bills, bills, bills |
| 18:56.37 | T0mW | :) |
| 18:57.12 | prpplague | T0mW: i got a nastygram from a debt collector, it seems my ex-wife never switched the phone over to her name after our divorce |
| 18:57.30 | prpplague | T0mW: and when she kicked the bucket she had a $300 phone bill that was never paid |
| 18:57.40 | prpplague | T0mW: so they've been calling me |
| 18:58.02 | prpplague | T0mW: gotta send them a certified copy of the court order back in 2003 |
| 18:58.16 | T0mW | prpplague: don't you have a declaration of severence from the divorce court? |
| 18:58.26 | T0mW | yeah |
| 18:58.28 | prpplague | yea |
| 18:59.22 | T0mW | formality, the debt collector will harass you until told to stop in the hopes of collectiing on it from you to "make them go away". heehee |
| 18:59.43 | prpplague | T0mW: yea, fortunately her estate still has some $$$ |
| 18:59.50 | prpplague | T0mW: so they probably will pay it off |
| 19:00.15 | T0mW | prpplague: or, you could sue the estate for fraud |
| 19:00.51 | prpplague | T0mW: no need that'd just take money out of the hands of my sons' hands |
| 19:00.58 | T0mW | nod |
| 19:29.43 | *** join/#edev davidc__ (n=davidc__@S01060002b360aacd.vc.shawcable.net) |
| 19:35.23 | *** join/#edev coreyfro (n=coreyfro@63.204.219.4) |
| 19:40.18 | *** join/#edev TuDo_Raptor (n=LC@i577BD368.versanet.de) |
| 19:40.49 | *** part/#edev TuDo_Raptor (n=LC@i577BD368.versanet.de) |
| 19:41.02 | gandhijee | 4/j #gentoo |
| 19:41.33 | T0mW | prpplague: have you used sqlite in a non-MMU processor? e.g. ARM7TDMI ?? |
| 19:41.53 | prpplague | sorry no |
| 19:42.13 | T0mW | hmm, I see the license is public domain? |
| 19:44.01 | T0mW | prpplague: I'm thinking of making the ARM7TDMI cards use SQL for their internal database. |
| 19:44.12 | prpplague | yea |
| 19:45.02 | T0mW | well, I have 256K of Flash + 512K RAM in them... |
| 19:45.38 | *** join/#edev hmmhesays (n=Neg@24-119-176-74.cpe.cableone.net) |
| 19:49.52 | prpplague | hmm, doesn't this patent sound like the one apple and starbucks was trying to get ? http://www.patentstorm.us/patents/5003472.html |
| 19:50.01 | prpplague | T0mW: pretty tight |
| 19:50.35 | T0mW | hmm |
| 19:54.25 | *** join/#edev gandhijee (n=user@mail.win-ent.com) |
| 19:56.37 | prpplague | ho ho hum |
| 20:02.42 | hmmhesays | pretty much |
| 20:05.49 | hmmhesays | i'm having a hard time with libpgp-error |
| 20:07.48 | *** join/#edev ico2_ (n=ico2@modem-1075.cougar.dialup.pol.co.uk) |
| 20:09.39 | T0mW | prpplague: sqlite only works with an O/S |
| 20:09.58 | prpplague | thats kinda what i figured |
| 20:10.09 | T0mW | yeah, it is not selfcontained |
| 20:10.27 | T0mW | well, moving on... |
| 20:13.56 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 20:13.56 | *** topic/#edev is a channel for the discussion of hacking small consumer electronic devices, mostly ARM based; Hammer Development boards now available at http://www.tincantools.com/ |
| 20:21.56 | *** join/#edev purl (i=ibot@pdpc/supporter/active/TimRiker/bot/apt) |
| 20:21.56 | *** topic/#edev is a channel for the discussion of hacking small consumer electronic devices, mostly ARM based; Hammer Development boards now available at http://www.tincantools.com/ |
| 20:24.48 | *** join/#edev landley (n=landley@adsl-76-197-33-46.dsl.austtx.sbcglobal.net) |
| 20:25.39 | landley | prpplague: so, did those nail prototypes ever show up? |
| 20:25.56 | landley | You mentioned some progress last night, but it was late enough my brain doesn't seem to have retained the details... |
| 20:30.42 | ds2 | wow SFE shipped things the next day.. major improvement |
| 20:35.31 | *** join/#edev _lifeless (n=patrick@90.150.222.136) |
| 20:37.56 | Mirell | Mmm.. |
| 21:10.20 | cbrake | prpplague: any idea where to get 8385 SDIO dev boards? |
| 21:10.39 | cbrake | prpplague: we have one zcom adapter board but had to send scouts into China to find that one, but we need another |
| 21:11.05 | cbrake | prpplague: I just ordered a Kodak camera SDIO wifi card as it supposedly has an 8686 in it |
| 21:35.35 | prpplague | landley: they aren't scheduled to ship from the board house until friday |
| 21:35.45 | prpplague | landley: which means we should get them monday or tuesday |
| 21:36.21 | prpplague | cbrake: you just need a 8385 sdio module? |
| 21:36.26 | landley | prpplague: So what did the final specs turn out to be? |
| 21:36.28 | cbrake | prpplague: yeah |
| 21:36.29 | landley | (Is there a web page?) |
| 21:36.49 | prpplague | landley: no web page as of yet, i've been totally swampped at work |
| 21:36.50 | cbrake | prpplague: with an SDIO adapter board to plug into a dev board |
| 21:36.53 | landley | Now that I'm not sick, adding capabilities not easily available from the hammer boad isn't worth the money. |
| 21:37.14 | landley | Keeping in mind things like RTC and USB2 for the successor to the hammer board probably is, though. :) |
| 21:37.15 | cbrake | prpplague: a consumer grade SDIO wifi card might be good enough if I knew of one that had a 8385 |
| 21:37.40 | prpplague | landley: yea the rtc is easy but, the usb2 will be more costly, but we'll look at it |
| 21:37.50 | landley | I keep meaning to upgrade my thyme.html documentation but I've sort of ground to a halt until I get an actual nail boad to play with, or at least more info to reorganize. :) |
| 21:37.52 | prpplague | cbrake: you just wanting to do some 8385 dev? |
| 21:38.08 | prpplague | landley: i'll try to post more specs and info today |
| 21:38.20 | landley | usb1 is tolerable, just be up front about it. |
| 21:38.28 | cbrake | prpplague: yes, we now have two engineers working on the wifi stuff, and trying to find enough hw to go around |
| 21:38.30 | landley | Don't _surprise_ people with lack of usb2, and it's ok. |
| 21:38.38 | prpplague | landley: i've been preparing for this lawsuit so i've been totally pre-occupied |
| 21:38.44 | landley | lawsuit? |
| 21:38.49 | prpplague | ~WIAV |
| 21:38.49 | purl | it has been said that wiav is a patent troll with information at http://trolltracker.blogspot.com/2008/01/4-interesting-new-cases-from-last-2.html and http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 21:39.20 | cbrake | prpplague: plus, I'd like to find something to have on hand long term for testing after this project is finished. |
| 21:39.21 | prpplague | landley: they claim all of 802.11 is in violation of their two patents |
| 21:39.35 | prpplague | cbrake: zcomax wasn't able to get you some dev modules? |
| 21:40.06 | prpplague | MonMotha: ping |
| 21:40.47 | prpplague | MonMotha: http://www.efunda.com/designstandards/plastic_design/plastic_intro.cfm |
| 21:41.33 | cbrake | prpplague: zcomax dropped the ball twice -- kept send us modules instead of adapter boards. So finally customer calls up their agent in TW and he finds one in China and ships it to us !!!! |
| 21:41.37 | chouimat|work | prpplague: wow (the WIAV thing) |
| 21:41.54 | prpplague | cbrake: thats odd |
| 21:42.02 | prpplague | cbrake: they go us our modules like overnight |
| 21:42.16 | cbrake | prpplague: yeah, that is odd |
| 21:42.19 | prpplague | cbrake: how quickly do you need them? |
| 21:42.29 | cbrake | prpplague: no rush, just be a nice to have |
| 21:42.34 | prpplague | cbrake: do they need to go into a sd slot? |
| 21:42.43 | prpplague | cbrake: or you want them wired? |
| 21:42.51 | cbrake | prpplague: the would be preferable, but I could sky wire something as well if I had to |
| 21:42.55 | *** join/#edev landley (n=landley@adsl-76-197-33-46.dsl.austtx.sbcglobal.net) |
| 21:42.57 | prpplague | landley: re |
| 21:43.08 | cbrake | prpplague: I just don't want to skywire that 50 pin connector on the zcomax module by hand :-\ |
| 21:43.12 | prpplague | landley: you look at the url on the WIAV case? |
| 21:43.28 | prpplague | cbrake: no need for 50pins of course, just 6 |
| 21:43.32 | cbrake | prpplague: too bad about the patent stuff, that is really nasty |
| 21:43.42 | cbrake | prpplague: good point, that might be doable |
| 21:43.42 | landley | Sigh. |
| 21:43.44 | landley | Yeah, I looked at the URLs, then Konversation crashed. |
| 21:43.53 | prpplague | cbrake: well we sued them first |
| 21:43.57 | landley | One of the many, many, many bugs in kubuntu 7.10 is if you type too quickly after switching windows, the window you switched to will segfault. |
| 21:44.04 | prpplague | cbrake: so hopefully we'll get it squared away |
| 21:44.21 | landley | I thought it was a kmail bug, but it's the whole of kde. Seems to be in the window manager or qt or something. |
| 21:44.31 | prpplague | cbrake: the thing is they are saying all of 802.11 is in violation of their patents, so that means you guys might be next |
| 21:44.36 | prpplague | landley: ahh |
| 21:44.39 | landley | And with qt switching to gplv3, I may be stuck with this buggy old version until I finally break down and switch to xfce... |
| 21:44.42 | cbrake | prpplague: heh |
| 21:44.53 | cbrake | prpplague: you can't even turn around any more |
| 21:45.01 | landley | Anyway, waht I was about to say is "kill this chongsoo park guy". |
| 21:45.12 | landley | You may take that as literally or metaphorically as you like, this being texas. |
| 21:46.04 | landley | Did you ping the SLFC and OIN to see if there's any way to deny this guy the right to use Linux even for personal use? |
| 21:46.36 | prpplague | landley: no not yet, i don't have any contacts there, so i don't know where to begin |
| 21:47.11 | SpeedEvil | prpplague: what do they say - if you can mention it? Is it simply 'you are in violation of patent # and #, pay us $x'/ |
| 21:47.15 | landley | Email gpl@busybox.net to ask the SFLC for advice. Feel free to tell 'em I sent you. |
| 21:47.29 | landley | I dunno if they'll _have_ any advice. It's not quite their focus. |
| 21:47.48 | landley | But they might have a good way to make this guy's life more difficult. And you'll definitely want to alert the Open Invention Network to this troll too. |
| 21:48.02 | prpplague | landley: i'm waiting on some feedback from IEEE to see if they believe its a valid patent for the 802.11 spec |
| 21:48.05 | landley | Oh, and email pj@groklaw.net. |
| 21:48.09 | landley | She'll definitely want to know about it. |
| 21:48.23 | landley | Patents on public specifications are a separate issue. |
| 21:48.28 | prpplague | landley: yea, i was thinking of that, but wasn't sure what i was allowed to do as an employee of aml |
| 21:48.42 | prpplague | landley: i'm waiting on the lawyer's ok for groklaw |
| 21:49.01 | landley | Well, _somebody_ should email pj those two links. :) |
| 21:49.34 | prpplague | i think GPSFan submitted it to groklaw. but i haven't seen anything there |
| 21:49.46 | landley | prpplague: point your lawyer at http://www.abanet.org/antitrust/committees/intell_property/june21.html |
| 21:49.59 | prpplague | SpeedEvil: without too much details, yea |
| 21:50.08 | landley | There's an old 1996 Dell case where Dell submitted technology to a standards body and didn't disclose they had a patent on it. |
| 21:50.15 | landley | Then they went to enforce that patent and got slapped down. |
| 21:50.16 | SpeedEvil | :/ |
| 21:50.26 | chouimat|work | landley: cool :) |
| 21:50.40 | prpplague | landley: yea, i can't find where the original owners submitted it to ieee |
| 21:50.40 | landley | So the question is, did whoever this patent came from ever participate in 802.11 or something similar? |
| 21:51.04 | landley | Maybe you can make something out of a failure to object. |
| 21:51.04 | prpplague | landley: the original owners didn't but its succesors did |
| 21:51.16 | prpplague | proxim |
| 21:51.20 | prpplague | and terrabeam |
| 21:51.21 | landley | For one thing, the ieee standardizing this _without_ knowledge of their patent is a darn strong piece of evidence as to "obviousness". |
| 21:51.36 | prpplague | indeed |
| 21:52.02 | landley | So with a big standard like this relying on such a patent, either the Dell case applies or the obviousness test applies. |
| 21:52.04 | landley | One or the other. |
| 21:52.32 | prpplague | right, thats what i've been gathering evidence for |
| 21:52.41 | landley | Cool. |
| 21:52.48 | landley | I should leave you to it then. :) |
| 21:53.15 | prpplague | hehe, np, i'm a little worn out with it right now |
| 21:53.24 | prpplague | landley: i'll try to update the nail board page |
| 21:53.34 | landley | Between following rambus, the sco case, all the busybox lawsuits, having IP law as a hobby since the 90's, and a few other things, I've accumulated a large pile of strange legal trivia. :) |
| 21:53.42 | landley | ok |
| 21:54.05 | chouimat|work | landley: it's not what I call a fun hobby :) |
| 21:55.20 | prpplague | hehe indeed it would be nice if _someone_ emailed pj |
| 21:55.59 | SpeedEvil | prpplague: is there a list anywhere of 'these are the patents we (the 802.11 committee) considered for applicability, and rejected'? |
| 21:56.10 | landley | chouimat: http://www.fool.com/portfolios/rulemaker/2000/rulemaker000501.htm |
| 21:56.16 | landley | And 0502, 0503, 0504, and 0505. |
| 21:56.24 | landley | Week long series I wrote, back in 2000. |
| 21:56.53 | landley | Reviewed by The Motley Fool's in-house lawyers, although they missed a couple things I've since noticed. :) |
| 21:59.40 | Mirell | So bored. |
| 21:59.53 | landley | Mirell: you have a qemu image. |
| 22:00.07 | prpplague | one sec phone |
| 22:00.07 | landley | Play with C in the qemu image. |
| 22:00.38 | chouimat|work | landley: playing with vista in qemu is more fun |
| 22:01.00 | prpplague | SpeedEvil: no, no list like that |
| 22:01.02 | Mirell | landley: Heh. |
| 22:01.13 | Mirell | landley: Easier to just SSH to my home machine |
| 22:01.13 | prpplague | SpeedEvil: they have a partial list of ones that apply to 802.11 |
| 22:01.22 | prpplague | SpeedEvil: but its not complete |
| 22:01.25 | SpeedEvil | Be nice to find that list - they must have generated it |
| 22:01.53 | landley | Mirell: if they allow that, sure. :) |
| 22:01.54 | prpplague | SpeedEvil: the ones that don't have patent listed just have a "we aren't saying which apply but we won't sue" |
| 22:02.47 | Mirell | landley: I'm SSHed into the UT machine currently, so of course. |
| 22:07.44 | *** join/#edev ico2_ (n=ico2@modem-935.chameleon.dialup.pol.co.uk) |
| 22:14.45 | SpeedEvil | prpplague: assuming that the originator of this patent was part of the process. |
| 22:15.02 | SpeedEvil | If it wasn't, then they presumably never signed anything. |
| 22:15.08 | prpplague | SpeedEvil: they were not |
| 22:16.20 | SpeedEvil | I was assuming there must have been some sort of discovery process by the 802.11 committee to search for patents by parties not agreeing to the patent release. |
| 22:16.55 | SpeedEvil | I was assuming there must have been some sort of discovery process by the 802.11 committee to search for patents by parties not involved in the patent release. |
| 22:18.01 | prpplague | SpeedEvil: correct |
| 22:18.16 | prpplague | SpeedEvil: hence my request to IEEE to identify these two patents and if they apply to 802.11 |
| 22:18.42 | SpeedEvil | Sorry - I missed that part. Good luck. |
| 22:18.54 | SpeedEvil | (on behalf of all of us) |
| 22:23.58 | adyer | SpeedEvil just use a reverse biased LED :-) |
| 22:24.23 | prpplague | cbrake: i'll see what i can dig up, i'll have to clear it with the boss |
| 22:24.31 | *** join/#edev Redhatter-Out (n=stuartl@203.94.56.252) |
| 22:24.47 | SpeedEvil | adyer: I need 50Mhz + and decent sensitivity. |
| 22:24.49 | cbrake | prpplague: ok thanks. |
| 22:24.59 | SpeedEvil | adyer: (laser ranger) |
| 22:26.11 | ds2 | I wonder if there is precedant where if the previous holder of a patent failing to enforce it is a implicit license to use that cannot be revoked by later holders? |
| 22:33.57 | prpplague | ds2: there is |
| 22:35.36 | adyer | SpeedEvil sounds like fun (especially the rcv/timing parts) |
| 22:53.35 | SpeedEvil | adyer: not actually doing timing as such. Measuring frequencies at which round-trip path delay is an integer number of wavelengths, then computing range from that |
| 22:54.07 | *** join/#edev sunrunner20 (n=jamesdoe@pool-96-226-38-56.dllstx.fios.verizon.net) |
| 22:55.16 | landley | ds2: I think the stronger defense is that the 802.11 people inventing technology withough being aware of this patent is proof of obviousness. |
| 22:55.32 | landley | The standards body did a patent search and found many other patents, so they did due dilligence. |
| 22:55.58 | landley | Which shoots down the "oh, they got it from us, they just pretended they didn't" argument. |
| 22:58.06 | SpeedEvil | If patents were that simple, there wouldn't need to be patent lawyers. |
| 22:58.58 | ds2 | landley: that argument is for the patent being invalid not merely it doesn't apply |
| 22:59.20 | landley | ds2: yes. |
| 22:59.30 | landley | If you're going after a patent troll, don't just defend yourself. Break them. |
| 23:00.04 | landley | The best defense is to take off and nuke the entire site from orbit. Only way to be sure. |
| 23:00.18 | landley | Like IBM did with SCO. |
| 23:00.41 | landley | "Blacken the skies of Utah with lawyers, so that the smoking crater and the lamentations of their women will serve as an example unto future generations." |
| 23:00.54 | landley | Only moreso. |
| 23:01.32 | landley | Of course this only triggers for malice, not for accidental license infringements and such (like the SFLC guys are pursuing). |
| 23:01.44 | landley | But patent trolls are malicious by nature. |
| 23:01.49 | landley | They need to be weeded out of the gene pool. |
| 23:03.07 | landley | If somebody trolls me with a software patent portfolio, I want them to wind up pulling back a bloody stump. |
| 23:03.26 | landley | This may not always be achievable, but it is my natural goal. Not merely defense. |
| 23:04.07 | landley | Recruit groklaw. |
| 23:04.50 | prpplague | yea, i'm hope'n pj will get wind of the case |
| 23:05.17 | landley | prpplague: what were those two links again? |
| 23:05.25 | prpplague | ~WAIV |
| 23:05.43 | prpplague | oops looks like purl is dead |
| 23:05.51 | prpplague | ~WAIV |
| 23:06.01 | prpplague | purl: WIAV |
| 23:06.01 | purl | i guess wiav is a patent troll with information at http://trolltracker.blogspot.com/2008/01/4-interesting-new-cases-from-last-2.html and http://www.setexasrecord.com/news/205753-texas-company-seeks-judgement-to-stop-potential-patent-suit |
| 23:06.14 | prpplague | helps if you type the name properly |
| 23:07.55 | *** join/#edev CosmicPenguin (n=nobody@163.181.251.103) |
| 23:07.55 | *** mode/#edev [+o CosmicPenguin] by ChanServ |
| 23:08.01 | prpplague | CosmicPenguin: greetings |
| 23:10.41 | landley | I hath poked pj, we'll see if it gets lost in her inbox... |
| 23:14.38 | prpplague | landley: i'll be back online in a little bit to post the latest pics and specs for the nail board |
| 23:14.49 | prpplague | landley: thanks |
| 23:16.22 | *** join/#edev redblue (i=star@ppp099.109-253-207.mtl.mt.videotron.ca) |
| 23:18.45 | *** join/#edev dijenerate (n=dijenera@72.51.67.14) |