--- Day changed Mon Jan 19 2009
01:36 -!- wrldpc [n=ben@173.48.207.37] has joined #hplusroadmap
02:20 -!- parodyoflanguage [n=kevin@mmds-216-19-34-153.twm.az.commspeed.net] has quit [Read error: 60 (Operation timed out)]
02:21 -!- genehacker [n=chatzill@wireless-128-62-174-156.public.utexas.edu] has quit [Read error: 110 (Connection timed out)]
03:46 -!- ferrouswheel [n=jp@121-73-144-159.cable.telstraclear.net] has quit [Read error: 104 (Connection reset by peer)]
07:25 -!- fenn [n=foobar@c-98-223-187-235.hsd1.in.comcast.net] has quit [Remote closed the connection]
07:32 -!- fenn [n=foobar@c-98-223-187-235.hsd1.in.comcast.net] has joined #hplusroadmap
09:25 -!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has joined #hplusroadmap
09:36 < samrose> hey kanzure
09:36 < kanzure> hi
09:37 < samrose> Marcin is asking me about recommendations for open source CAD software. So, although I have preferences, I thought I would ask around and give him multiple perspectives
09:37 < samrose> what do you use/recommend?
09:37 < kanzure> BRLCAD. :-)
09:37 < samrose> 3D CAD software, that is
09:37 < kanzure> http://brlcad.org/
09:38 < samrose> any good resources that you can point novices to about learning BRL-CAD?
09:38 < kanzure> There's a lot of good tutorials on that website.
09:38 < kanzure> Also, there's #brlcad on freenode.
09:39 < samrose> just found this, too http://gpwiki.org/index.php/BRL-CAD:Tutorials
09:43 < samrose> gracias kanzure
09:44 < kanzure> There's more, but I can never remember how to spell advocado. 
09:44 < kanzure> The #brlcad folks get kind of ranty about all of the other F/OSS CAD systems not contributing code to BRLCAD..
09:52 < samrose> Does that thing work?: http://avocado-cad.sourceforge.net/
09:55 < kanzure> Not much. I tried it out a little bit, don't remember. fenn might know.
09:55 < kanzure> it might be biased, but #brlcad will know :/
09:57 < samrose> ok, thnx
10:01 < samrose> well, go avoCADo up and running, will check it out too
10:03 < samrose> not much to it yet.
10:04 < kanzure> yeah, if you need help compiling brlcad-
10:04 < kanzure> http://heybryan.org/brlcad.html
10:27 < samrose> I have brlcad compiled on one pc
10:28 < samrose> not on new laptop yet though
11:27 -!- genehacker [n=chatzill@wireless-128-62-174-156.public.utexas.edu] has joined #hplusroadmap
12:27 -!- ybit [n=heath@unaffiliated/ybit] has quit [Read error: 104 (Connection reset by peer)]
13:36 < wrldpc> anybody in here make instrumental music?
13:36 < kanzure> I know a few who do.
13:36 < kanzure> What's up?
14:01 < genehacker> want to start a band?
14:01 < genehacker> a software band or circuit bent band?
14:02 < genehacker> Kanzure turn the bible into horrible horrible music
14:17 < wrldpc> i'll rap
14:17 < wrldpc> how does one pronounce kanzure?
14:17 < kanzure> I'm not sure it's pronouncable
14:43 < genehacker> can zur
14:46 -!- ferrouswheel [n=jp@121-73-144-159.cable.telstraclear.net] has joined #hplusroadmap
14:48 < kanzure3> http://www.ldraw.org/Article48.html
14:49 < kanzure3> ^ metadata stuff for lego parts library
14:49 < kanzure3> http://www.ldraw.org/Article131.html   intro to creating build instructions (but with images)
14:49 < kanzure3> Building Instruction Images (BILs)
14:49 < kanzure3> hrm..
15:38 < kanzure3> http://www.python.org/dev/peps/pep-0362/
15:45 < kanzure> Why aren't the man pages generated from function signatures? A significant portion of man includes the common C header files for just about every useful function.
15:48 < kanzure> in fact, isn't this what the .lib files that the gnu linker deals with are for?
15:59 < kanzure3> http://www.linuxjournal.com/node/1060/print ELF dissection.
16:19 < genehacker> so you're trying to find a new way to write instructions?
16:36 < kanzure3> no, this would be for interoperability. 
16:36 < kanzure3> when a compiler compiles your program, each function has a "signature" that describes what it wants as input and what it gives as output
16:36 < kanzure3> so I'm just wondering about just using a large registry of these signatures paired to compiled binaries
16:42 < genehacker> ok
16:43 < genehacker> so you want to write instructions for anything?
16:43 < kanzure3> for the packaging format? yes, it would be nice to have a computational representation of instructions
17:13 < genehacker> I am a bit lost on what you are doing
17:16 -!- wrldpc [n=ben@173.48.207.37] has quit []
18:06 -!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has quit [Read error: 110 (Connection timed out)]
18:11 < genehacker> Fenn, UAVs might be cheaper than cars
18:13 < genehacker> or fedex delivery vans
18:33 -!- PeerInfinity [n=someone@216.36.180.162] has joined #hplusroadmap
18:35 -!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has joined #hplusroadmap
18:46 -!- ybit [n=heath@unaffiliated/ybit] has joined #hplusroadmap
18:48 < genehacker> http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=01656362
18:48 < genehacker> check it out
18:48 < kanzure> ?
18:49 < genehacker> it's about the UAV delivery system me and Fenn have been talking about
18:49 < kanzure3> "A fixed-wing aircraft for hovering in caves, tunnels, and buildings"
18:49 < genehacker> yeah
18:49 < kanzure3> ":Micro air vehicles (MAVs) are small bird-sized aircraft with applications in reconnaissance, search-and-rescue, airborne agent and pathogen detection, and target acquisition. Fixed-wing MAVs cannot hover and thus, are not able to fly in tight, enclosed spaces. Rotary-wing platforms can hover but are limited by endurance. This paper presents a fixed-wing MAV with a secondary flight mode (i.e. hovering) allowing it to fly in caves, tunnels, and buildings. The sensing and control system used to achieve autonomous hovering is also described. This is, to the best of our knowledge, the first documented success of autonomously hovering a fixed-wing MAV in the open literature"
18:50 < genehacker> it can avoid crashing into things with optic flow sensors
18:50 < genehacker> in an urban environment
18:52 < genehacker> you could retrofit an off the shelf RC plane with off the shelf sensors and controllers, set it up to hover real low and drop a package to deliver it
18:57 < genehacker> though the customer might have to provide the weight of the package
18:59 < genehacker> dang plane cost about $300, Inertial measurement unit cost $1495
19:00 < kanzure> gyro?
19:00 < genehacker> you could call it that
19:01 < genehacker> something that tells the plane it's orientation
19:01 < kanzure> oops, wrong device
19:01 < genehacker> but things like that are cheaper nowadays
19:01 < genehacker> "It weighs
19:01 < genehacker> just 30 grams and is comprised of three triaxial accelerometers
19:01 < genehacker> and angular rate gyros as well as three
19:01 < genehacker> orthogonal magnetometers."
19:02 < genehacker> it has gyros
19:02 < genehacker> wonder if a wiimote could be used?
19:32 -!- wrldpc [n=ben@c-98-217-191-173.hsd1.ma.comcast.net] has joined #hplusroadmap
20:16 -!- samrose [n=samrose@c-24-11-214-181.hsd1.mi.comcast.net] has quit ["Ex-Chat"]
20:27 < genehacker> I wonder how the wiimote figures out orientation
20:28 < kanzure> it doesn't
20:28 < kanzure> there are sensors that you set up on your wii
20:28 < kanzure> stereo.
20:29 < kanzure> triangulation.
20:29 < kanzure> "black magic"
20:29 < genehacker> how doe it do it when you hold it like a steering wheel and it can't see the sensor bar?
20:29 < kanzure> there's a sensor bar?
20:30 < genehacker> I knew about the triangulation
20:30 < genehacker> it's not a sensor bar, it's infrared LEDs set a constant distance apart from each other
20:30 < genehacker> do you have a Wii?
20:30 < kanzure> no
20:31 < genehacker> my parents have one
20:31 < genehacker> I heard the wiimote has tilt sensors in it
20:32 < genehacker> one of the add ons for it, has a 2 axis angular rate sensor
20:33 < genehacker> interesting, for small quantities of accelerometers, it's cheaper to hack them from a wiimote than buy them from the manufacturer
20:33 < kanzure> I'm bored.
20:33 < genehacker> they cost $60 individually
20:35 < genehacker> the Iphone has fairly accurate tilt sensors hmm....
20:36 < genehacker> http://www.engadget.com/2007/08/28/iphones-tilt-sensor-hacked/
20:37 < genehacker> http://www.youtube.com/watch?v=CRcld5aAN2E
20:37 < kanzure> it's odd how getopt focuses so much on parsing rather than defining
20:37 < kanzure> I guess parsing is what the programmers needed at the time more than anything
20:37 < genehacker> youtube's sound remove music thing is way too strict
20:37 < kanzure> eh?
20:38 < genehacker> shit, sound was off
20:38 < genehacker> I failed
20:48 < genehacker> so you know arduino?
20:48 < kanzure3> yes
20:50 < kanzure3> is just c.
20:50 < genehacker> could I use arduino in a UAV?
20:50 < genehacker> something with a realtime control system
20:51 < kanzure3> that depends entirely on how long you're willing to wait between sense-process-react
20:51 < kanzure3> physically, an arduino is small enough if that's what you mean
20:52 < genehacker> not long
20:56 < genehacker> I want to be able to have a UAV fly to a site, go up into a prop hang above a drop zone target, fly down to about a foot above the target and drop a small package and go back to base to refuel and pick up another package
20:57 < genehacker> you can use arduino on a UAV
20:57 < kanzure3> right, if it is ok to fly blind even better
20:58 < genehacker> well, something like this wouldn't fly blind it would use something like optic flow sensors, sonar, or lidar to avoid obstacles
20:59 < kanzure3> as long as you can wire up all of the devices into the board, that's fine.
21:01 < genehacker> a delivery UAV might also have to have a video camera on it sending a video feed to a human to check on things
21:02 < kanzure3> don't know how much bandwidth the arduino can handle, but you might as well just pair the video separately on its own radio freq or something
21:02 < genehacker> that's what you do
21:03 < genehacker> hmmm... I wonder how you could monitor the video on a bunch of UAVs
21:52 -!- wrldpc [n=ben@c-98-217-191-173.hsd1.ma.comcast.net] has quit []
22:01 < genehacker> I estimate something like a delivery UAV should cost about $1600
22:09 < genehacker> and that it should achieve break even at about 40 deliveries at best
22:09 < genehacker> that's pretty good
22:23 < fenn> $40 delivery is pretty steep
22:23 < fenn> today i drove 12 miles to get a tiny 8mm rubber disc (shutoff valve washer)
22:23 < kanzure> don't you feel stupid.
22:24 < fenn> no, not really
22:24 < kanzure> isn't it inefficient though
22:24 < fenn> yes of course
22:24 < fenn> i'll feel stupid when i go to put it in and it doesn't fit
22:24 < fenn> ;)
22:24 < kanzure> ah.
22:24 < kanzure> so, I just thought I had this wonderful burst of inspiration
22:24 < genehacker> no it's $20 per delivery
22:25 < kanzure> where I'd just put each getopt option into a separate file and then do a compatibility matrix table
22:25 < genehacker> it's not for rubber disks
22:25 < kanzure> where the rows would be each option, and each column is anoption
22:25 < kanzure> and then a binary value at the x,y coord for compatibility (i.e. if one option can't be used with aonther)
22:25 < genehacker> it would take 40 deliveries to break even at about 200
22:25 < genehacker> it would take 40 deliveries to break even at about $20 per delivery
22:25 < kanzure> but this doesn't work for multiple option combination logic stuff (i.e., -x and -y and -z but not -y and -c)
22:26 < genehacker> erm $40 per delivery
22:26 < genehacker> that
22:27 < genehacker> is about the price for very fast delivery with a local courier service for a small item
22:35 -!- PeerInfinity [n=someone@216.36.180.162] has quit []
22:35 < genehacker> fast delivery as in under 2 hrs
22:46 < fenn> I can get a pizza for less than $40
22:48 < fenn> kanzure this sounds like something one would write in haskell
22:48 < kanzure> I've never been motivated to learn haskell. what's so special about it?
22:48 < kanzure> I hear of these them there 'atom' thingies
22:49 < fenn> well they use lots of annoying mathy words
22:49 < fenn> dealing with large sets is sort of the basic idiom of the language
22:49 < fenn> or at least that's my foggy recollection of my primitive understanding of it
23:01 < genehacker> you wouldn't want to deliver pizza with a UAV fenn, unless your life depended on getting pizza
23:01 < genehacker> but a more reasonable price would be $20
23:02 -!- ybit [n=heath@unaffiliated/ybit] has quit [Read error: 104 (Connection reset by peer)]
23:02 < genehacker> break even happens at 80 deliveries
23:03 < genehacker> which seems fairly reasonable to me
23:04 < fenn> huh. integrated circuits really are "magic crystals"
23:05 < genehacker> huh?
23:05 < genehacker> I guess they are
23:05 < fenn> i was just looking at wireless chips on sparkfun, "These are the latest gems from Laipac Tech" and started thinking
23:07 < genehacker> diode lasers are really magic crystals
23:07 < genehacker> diode pumped solid state lasers are even more magical
23:07 < fenn> er, what's the difference?
23:09 < genehacker> diode pumped lasers are a diode laser pumping a special crystal setup
23:10 < fenn> so they're more crystally, not more magical
23:10 < genehacker> hmmm... if I made a super-fast UAV capable of 30 minutes or less delivery times, then I might be able to charge $50 per delivery and breakeven in a day
23:10 < genehacker> well they're pretty magical
23:10 < genehacker> it's like several crystals
23:11 < fenn> oh, speaking of optics..
23:11 < genehacker> http://en.wikipedia.org/wiki/Diode-pumped_solid-state_laser
23:11 < fenn> i was wondering why we dont see any mobile head up displays that use fiber optic cable to pipe the image to your eye/glasses
23:11 < genehacker> you'd need a thick cable
23:12 < fenn> like a laparoscope or borescope, they sell for $200 at harbor freight
23:12 < genehacker> link?
23:12 < fenn> http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=91565
23:12 < fenn> er, but in reverse
23:13 < fenn> no big eyepiece or anything
23:13 < genehacker> resolution I guess
23:13 < genehacker> price
23:13 < fenn> you could do much higher resolution than tiny LCD displays
23:13 < fenn> or do you mean in the optics sense
23:14 < fenn> and $200 isnt that much, especially considering the parts cost would be much elss
23:14 < fenn> *less
23:14 < genehacker> resolution is how many fibers you have
23:14 < fenn> are you sure about that?
23:15 < genehacker> for the most part
23:15 < fenn> can't you pipe a whole image down one fat cable?
23:15 < genehacker> maybe if you have some weird adaptive optics at the end
23:15 < fenn> why adaptive?
23:16 < genehacker> to compensate for distortion from reflecting in the cable
23:16 < genehacker> I don't know
23:16 < genehacker> you might look into it
23:16 < fenn> what if i had some kind of multiply jointed periscope
23:16 < fenn> actually it only really needs to go like 5 inches
23:17 < fenn> if the whole setup is compact enough to fit in like a hat or whatever
23:17 < fenn> i'm just trying to work around the "big thing clamped to your face" idea
23:17 < genehacker> you really need to use one of those DMD chips
23:17 < fenn> yes
23:18 < genehacker> you know a chip covered with a whole bunch of tiny mirrors?
23:18 < fenn> I came up with this idea years ago
23:18 < genehacker> you could get high resolution with that
23:18 < fenn> DLP chips are still expensive..
23:18 < fenn> wah
23:19 < genehacker> this one company was able to make a really tiny high res display out of a DLP chip
23:19 < fenn> who?
23:19 < genehacker> also made it display color too, using reflected sunlight
23:19 < genehacker> I forgot, it was a while ago
23:19 < fenn> and how tiny is tiny?
23:19 < genehacker> as big as the chip is
23:20 < fenn> i have some of these chips, they range from 0.5" to 1" diagonal
23:20 < fenn> (no docs unfortunately)
23:21 < genehacker> http://www.goldmine-elec-products.com/prodinfo.asp?number=G13751
23:21 < fenn> yes, that
23:21 < genehacker> $5
23:21 < fenn> the connector from the chip to the flex circuit is removed
23:21 < fenn> there are smaller ones that are still whole
23:22 < fenn> these guys http://www.goldmine-elec-products.com/prodinfo.asp?number=G15426
23:22 < genehacker> you can get a special connector for things like that, so it mounts on a PCB
23:22 < fenn> I love how they give me all sorts of documentation on where they got them, what the part number is, etc </sarcasm>
23:23 < fenn> the connector is easy, but i dont know what the protocol is or anything
23:23 < genehacker> I don't know what you could do about the scratches
23:23 < fenn> if it works it'd be good enough for me
23:23 < fenn> there is a glass or plastic plate over the actual chip
23:24 < genehacker> then you have a really annoying scratch going across your field of vision
23:24 < fenn> i have a 'really annoying' red line going down my monitor, i dont seem to notice it
23:24 < genehacker> you can't replace the plate, unless you have mad skills
23:24 < genehacker> ok
23:24 < genehacker> you're good then
23:24 < genehacker> what's the part number?
23:25 < fenn> ok please tell me what signals to send to the chip and on what pins
23:25 < fenn> no part number
23:25 < fenn> one sec, i have a hi res image, coming up
23:25 < genehacker> any number at all
23:25 < genehacker> look at it under the microscope there might be a number
23:26 < genehacker> or a datagrid
23:27 < genehacker> dang I forgot how this one tiny LED grid I had worked
23:30 < genehacker> http://focus.ti.com/download/dlpdmd/2503686.pdf
23:30 < genehacker> check out these dox
23:31 < fenn> http://fennetic.net/pub/irc/little-mems.jpg
23:31 < fenn> http://fennetic.net/pub/irc/big-mems.jpg
23:32 < genehacker> that's far from a slight scratch
23:33 < fenn> here's a picture someone else took http://fennetic.net/pub/irc/mems.jpg
23:33 < fenn> i dont even know if they are from the same company tho
23:34 < fenn> oh the white crap is just dust
23:34 < genehacker> big mems looks easier, more pins, more likely you can individually control each mirror
23:34 < fenn> i dont want to individually control each mirror 
23:35 < fenn> anyway it certainly doesnt have 500 or 1000 pins or whatever would be needed to multiplex each mirror
23:35 -!- wrldpc [n=ben@173.48.207.37] has joined #hplusroadmap
23:37 < genehacker> it forms a grid I think
23:38 < genehacker> you might get your self a USB microscope, an ultralow power supply, and some damn small test leads
23:39 < genehacker> and see if you can get any response
23:41 < genehacker> have fun
23:41 < fenn> i was going to just shine an led or laser on it and look at the reflection
23:41 < genehacker> good idea
23:41 < fenn> or just look at it, i bet you could see it change color or something
23:42 < genehacker> you see fenn, if you figure out how to control a DMD chip
23:42 < genehacker> then that means we can do lots of cool things
23:42 < fenn> geomatec's website doesnt say anything about mirror arrays (lots of laser scanning and lenses stuff tho)
23:42 < genehacker> on of them is do maskless lithography
23:43 < genehacker> http://www.eetasia.com/ART_8800416404_1034362_NT_a5fd45ee.HTM
23:43 < fenn> am i supposed to be able to read that?
23:43 < genehacker> maskless lithography means we can make our own microchips, do massive DNA synthesis, and make high res 3d printers
23:44 < genehacker> you can't?
23:44 < genehacker> oh well
23:44 < fenn> Please login or register with us to view this article
23:44 < genehacker> oh well
23:45 < fenn> any ideas for collimating the output from an LCD into a fiber optic cable? and doing it in a compact manner
23:45 < fenn> so ideally i'd have the LCD panel in a case with the fiber coming out coplanar with the panel
23:47 < fenn> is there such a thing as a fresnel prism?
23:48 < fenn> ooo magical internet: http://bencraven.org.uk/prism_specs.html
23:50 < genehacker> a fresnel prism
23:51 < genehacker> Fresnel is a bunch of prisms
23:52 < genehacker> sounds like you need lenticular 3d viewing sheets
23:52 < fenn> maybe i can use an lcd backlight
23:53 < fenn> heh http://www.fresnelprism.com/
23:53 < fenn> somehow i dont think it's quite right
23:54 < fenn> well i guess i could stack them, but it might end up just as thick as using a mirror