--- Day changed Tue Dec 02 2008
00:23 < kanzure> huh
00:23 < kanzure> what's going on here?
00:24 < kanzure> http://change.gov/ is on the creative commons
00:34 < bkero> I'm putting it into revision control
00:38 < kanzure> the entire site?
00:44 < bkero> Yes
00:44 < bkero> http://staff.osuosl.org/~bkero/change.gov.patch
00:45 < kanzure_> "
00:45 < kanzure_> When you fold a complicated, choppy, cascade-y chain of differential equations in on itself via recursion, it should either flatline or blow up.  You would need exactly the right law of diminishing returns to fly through the extremely narrow soft takeoff keyhole."
00:45 < kanzure_> http://www.overcomingbias.com/2008/12/recursive-self.html#more
00:45 < kanzure_> yay I'm finding I like Eliezer's writing today
00:45 < kanzure_> I usually don't read his blog
00:57 < kanzure> though it sounds like he was just reading off of my 
00:57 < kanzure> http://heybryan.org/recursion.html file
01:11 < fenn> unless it's a fractal
01:11 < fenn> or one of those not-quite-random-but-unpredictable patterns that wolfram is so fond of
01:15 < kanzure> basically he's saying that the planets have to be in alignment.
01:15 < kanzure> except instead of 'planets', he's using a folk psych ontology of sorts.
01:17 < kanzure> hrm, where'd my brain compiler go.
01:18 < bkero> kanzure: ghc?  It's still here somewhere
01:18 < kanzure> ?
01:22 < bkero> Joke :P
01:30 < kanzure> Anyway, I'm kind of surprised that he's talking about differential equations for recursive self-improvement keyhole plotting when he doesn't even have metrics for improvement really, except in the sense of Big O notation I guess, but even then.
01:30 < kanzure> it will probably turn into some dimensional analysis epigenetic search algorithm that checks for a plotted course for "ridiculous FTW here"
01:31 < kanzure> but that's impossible without those quantitative measurements I'd think..
01:35 < kanzure> I mean, ontologically you get to restrict yourself to the seven fundamental types of energy and various types of matter
01:36 < kanzure> and although the combinatorial space for machines to interconnect and convert those types are many, even just within electrical for the representation of turing machines, perhaps we get to say that RSI-stuff is a class of its own on the level of playing field as a turing machine
01:36 < kanzure> von Neumann machines and von Neumann universal replicators, for instance, but replicators!=recursive improvers
02:35 < bkero> http://yro.slashdot.org/yro/08/12/02/0120249.shtml
02:35 < bkero> This is why the EFF deserves all your money.
02:38 < kanzure> What the fuck.
02:38 < kanzure> That's a page out of Star Wars.
02:43 < bkero> lol awesome
02:43 < bkero> The legislation is an attempt to give the president the authority to terminate claims that the president has violated the people's Fourth Amendment rights.
03:30 < kanzure_> http://slashdot.org/comments.pl?sid=1045407&cid=25925939 <- Wait, what? His sister died at a Who concert? That's .. brutal.
03:33 < kanzure_> http://en.wikipedia.org/wiki/Hillsborough_disaster
03:33 < kanzure_> "The crush"
03:50 < kanzure> Can anybody posit a proper argument against the notion of RSI leading to things like archailects and >S3 creatures that basically are giant brains built up of a civilization ?
03:51 < kanzure> I mean, I'm trying to picture the concept of an S3 in a box, but all I get is Ede-the-man-as-a-hologram.
03:51 < kanzure> I guess that's an obscure reference to War in Heaven.
03:52 < kanzure> Spoiler: Ede the God, in a final blast of desperation while fighting some evil super moon entity, transmits radio signals to holocrons across the galaxy and downloads "himself", in a highly cut and compressed form. So, in the box, he might appear to be rather brilliant, but in truth it's like a child trying to be a bajillion year old old-man.
03:52 < kanzure> which is more of a fake than anything.
03:53 < kanzure> I suppose calling it a "civilization" makes it difficult in argumentation to convince others because "civilization" usually conjures up other images
03:53 < kanzure> rather than things pulled from Zindell and Orion's Arm.
03:56 < kanzure_> http://www.orionsarm.com/sophontology/toposophy.html
03:56 < kanzure_> or rather: http://www.orionsarm.com/sophontology/toposophics.html
04:01 < kanzure> oh, I guess that's the same as fenn's fractal comment. bleh. nevermind.
04:39 < gene> so Kanzure could you give me a brief overview of that paper you scooped?
05:37 < gene> kanzure?
05:39 -!- Irssi: #hplusroadmap: Total of 22 nicks [0 ops, 0 halfops, 0 voices, 22 normal]
06:25 < gene> kanzure...
08:47 -!- xp_prg2 is now known as xp_pr
08:47 -!- xp_pr is now known as xp_prg
13:26 < kanzure_> http://makerfaire.com/austin/2008/blueribbons/
13:26 < kanzure_> Yay, the Ponginator made it
13:30 < kanzure_> http://box2dflash.sourceforge.net/ <- what Fantastic Contraption is probably using..
13:31 < kanzure_> haha
13:31 < kanzure_> "Thanks to those who have supported this project with their generous donations: .... Colin Northway"
13:31 < kanzure_> Colin being the guy who gets all teh monies from Fantastic Contraption
13:31 < kanzure_> hehe.
13:32 < kanzure_> oh, #box2d
13:35 < fenn> toposophics.html is disappointingly focused on the physical size of the intellect
13:39 < kanzure> another thing I noticed yesterday with OA is that they are of the mentality of the ai 'ruling' the 'lesser intellects'
13:39 < kanzure> OA was originatedd with the idea of ai not being a necessarily evil thing or something
13:40 < kanzure_> http://www.box2d.org/screenshots.html look at #2. hrm..
13:40 < kanzure_> comparing to fantasticcontraption.com that is
13:41 < kanzure_> the flash cuts out for me when the ads to the right reload, and loading the swf file just redirects me to the ad-ridden page
13:44 < kanzure> in light of OM, sourceforge is a jerk for using a good name
13:51 < kanzure> "make: error: g++ not found." <- Have I really not been using g++ in so long?
15:02 < kanzure> I don't get it. What's so open source about the IEEE-NIST VMAC?
15:02 < kanzure> There's no code, or anything.
15:04 < kanzure> apparently the three organizers have written a 'framework' (the three from NIST)
15:04 < kanzure> USARSim, MOAST, unreal game engine
15:05 < kanzure> teams: Zoran Duric, Eugene Yu, Rahul Simha, Darby Thompson, Alex Stefanovski, Ali Eydgahi, Payam matin, Mark Bivens, Chris Epps, Lee Long, George Dimitoglou, Xinilian Liu, Brandon Burger, Kristopher Reese, Eric Walton, Keith Hargrove, Therese Andeme, Kashun Davis
15:05 < kanzure_> http://usarsim.sourceforge.net/
15:06  * kanzure_ leaves
15:27 < bkero> Hehe frameworks
16:13 < kanzure__> Hello.
16:15 < kanzure__> "Kinematically accurate models of new robots can be added using vehicle classes from the Karma Physics Engine, a rigid multi-body dynamics simulator that is part of the Unreal development environment. "
16:16 < kanzure__> "The standard version requires the Unreal game engine bundled with Unreal Tournament 2004. By using a game engine, the simulator achieves high fidelity on commodity hardware. The Unreal runtime engine version uses a modified vehicle class." 
16:16 < kanzure__> Does this mean unfree software is required?
16:18 < kanzure__> http://sourceforge.net/projects/moast/ "The Mobility Open Architecture Simulation and Tools (MOAST) framework aids in the development of autonomous robots. It includes an architecture, control modules, interface specs, and data sets and is fully integrated with the USARSim simulation system."
16:19 < bkero> UT2004 is unfree, yes.
16:22 < wrldpc_> where are the logs for #hplus?
16:23 < kanzure__> there's a #hplus?
16:24 < kanzure__> http://heybryan.org/~bbishop/docs/logs.zip are some old #hplusroadmap logs, I have some more recent logs in an irssi log file, but I no longer have complete access to complete logs.
16:24 < wrldpc_> ok
16:24 < kanzure__> any particular reason for asking? sharing with somebody?
16:25 < wrldpc_> the presidential schema to remove culpability for amendment violations or something ..
16:25 < kanzure__> bkero: There is talk on the internet, in some forums, of a free UT server app. Be this true?
16:25 < kanzure__> wrldpc_: huh?
16:25 < wrldpc_> and you also had a GEM this morning/last night, B.
16:25 < wrldpc_> I can't remember the context or the content .. I just recall it as being extremely well worded.
16:25 < kanzure__> GEM?
16:25 < kanzure__> genetically engineered machine?
16:29 < bkero> kanzure__: The UT dedicated server.  I don't think it's free.
16:30 < kanzure__> http://udn.epicgames.com/Two/UnrealEngine2Runtime22262002.html#Downloading
16:30 < bkero> That source code is just for the scripting language
16:30 < wrldpc_> gem as in precious stone ;D
16:31 < kanzure__> a well worded gem stone. Nope, don't remember anything good in the last 24 hours. But I also don't have my logs in front of me.
16:31 < kanzure__> bkero: How is this open source then. 
16:32 < kanzure__> USARSim, I mean.
16:32 < fenn> wrldpc_: http://yro.slashdot.org/yro/08/12/02/0120249.shtml
16:32 < wrldpc_> ty
16:33 < kanzure__> http://superb-east.dl.sourceforge.net/sourceforge/usarsim/USARSim-manual_3.1.3.pdf
16:34 < kanzure__> lies: http://www.washacadsci.org/capsci08/presentations/ieeemadhaven.pdf
16:42 < kanzure__> "Sython, a language for privacy-preserving programming in which developers don't get to see private data (open-source)."
16:48 < kanzure__> Meh. The VMAC participants seem boring. Except maybe Keith Hargrove.
16:49 < kanzure__> I don't know how this is virtual manufacturing automation just by focusing on automated ground vehicles.
16:49 < kanzure__> I guess it's somewhat related, but I don't expect manufacturing environments to be so visually messy and noisy in the sense of things not being where they are supposed to be.
16:53 < kanzure__> http://www.youtube.com/watch?v=1maJZ0U8ULg Fantastic Contraption speed run
17:09 < kanzure__> hah! hahah.
17:10 < bkero> kanzure__: I can have open source software that interfaces with a proprietary module.
17:11 < kanzure__> You'd still be a jerk.
17:11 < bkero> Sure
17:12 < bkero> I never claimed USARSim wasn't open, nor run by jerks. :P
17:13 < kanzure__> http://heybryan.org/~bbishop/2008-12-02_box2d.PNG <--- bwahah
17:13 < kanzure__> Theo Jansenn's mechanism in box2d
17:17 < kanzure__> yeah, so fantasticcontraption is like one of the tests for box2d but with extra graphics
17:17 < kanzure__> (except using the box2d version for flash)
17:26 < bkero> http://en.wikipedia.org/wiki/Mike_the_Headless_Chicken 
17:28 < kanzure__> yeah, he's famous, but I don't get the whole "feeding with an eyedropper". Where's the mouth and throat?
17:31 < kanzure__> brb'
17:44 < fenn> so are you going to do some gear simulations in box2d? :)
17:54 < kanzure__> Maybe. Actually what I really want to do is a clone of fantasticcontraption. I've figured out how to do it too, I just need to go remember how to wire up glut with the mouse callback function.
17:54 < kanzure__> and then assign some keys for different types of bodies (which I've done)
17:54 < kanzure__> they kind of just appear out of nowhere at the moment though, oops
17:55 < kanzure__> once I have key bindings, I'd like to do an XML-ish format for the different types of bodies to specify
17:55 < kanzure__> mapped to the API ideally
17:55 < kanzure__> so if that happens then yes, I'll just write a converter to spit out the graphsynth gear stuff to the nonexistant box2d XML format
17:56 < kanzure__> the simulator is just as addictive as fantastic contraption, and there's not even any goals.
18:01 < bkero> kanzure__: check out protocol buffers or yaml
18:01 < bkero> They're both awesome and not as heavy or expensive as xml
18:01 < kanzure__> protocol buffers?
18:03 < bkero> http://code.google.com/apis/protocolbuffers/docs/pythontutorial.html 
18:04 < kanzure__> I must be out of practice. If I wanted to declare a global variable in C++, I go define it in a common (common to the source files in which I need it) header file, yes?
18:05 < kanzure__> ah, my extern was right
18:06 < kanzure__> Nevermind. :)
18:24 -!- Utopiah is now known as UtopiahGHML
18:53 < kanzure__> geeze, destroying bodies in box2d is unnecessarily complicated
18:54 < kanzure__> it's this messy web of pointers from just about every other object/body in existence within the simulation..
18:54 < xp_prg> kanzure__ I wrote a script to turn every bbf file into a big json file!
19:21 < kanzure__> Heh. Took a little too long to get a delete method going.
19:46 < xp_prg> kanzure__ did you see what I said?
19:46 < kanzure__> yes.
20:42 < kanzure__> http://heybryan.org/~bbishop/2008-12-02_box2d_3.PNG <--- *grins*
20:43 < kanzure__> No motors yet. Only 'water'-type connector.
20:51 < fenn> why are you making a fantastic contraption clone?
21:02 < kanzure__> I'm not sure
21:05 < kanzure__> well, yes I am
21:05 < kanzure__> my "super secret plan" is to load up graphsynth constructions
21:05 < kanzure__> constructions.
21:10 < kanzure__> and using some of the rougelike-algorithms from the other day that I linked to (roguelikedevelopment.org), some automatic level generators, then do some automatic design (first in a vacuum, then plus selection -- vacuum = design up to 20 nodes, and then see which ones work)
21:16 < kanzure__> todo: joint-to-joint, rigid/solid joints, motors, joint delete function (or, rather, general object indexing), and then yaml + gxml + some level specification format (ok, yaml too.)
21:16 < kanzure__> http://heybryan.org/~bbishop/docs/fantastic/ has a 10 MB zip file
21:18 < kanzure__> it's kind of buggy because of a lack of checks on pointer indexes.
21:18 < kanzure__> *indices
22:17 -!- kanzure___ is now known as kanzure_-
22:41 < kanzure_> http://www.physorg.com/printnews.php?newsid=147353055 Zipf
22:42 < kanzure_> Zipf's law re: connectivity following a negative curve when on a log-log scale
22:44 < kanzure> Hi fenn_.
22:44 -!- fenn_ is now known as fenn
22:44 < fenn> alo
22:44 < kanzure> hah, Slashdot article - "Teacher sells ads on tests"
22:45 < kanzure> wonder if each ad covers the ink cost ..
22:45 < kanzure> "but I was distracted by the punch the monkey ad!"
22:45 < fenn> want to try and git-clone git://fennetic.net/cxf2cnc for me?
22:46  * fenn just poked a hole in the firewall
22:46 < kanzure> yep, works
22:46 < fenn> yay
22:54 < xp_prg> kanzure have you used my script yet?
23:08 < kanzure> No.
23:13 < xp_prg> how come? :(
23:45 < kanzure> biotechdoc?
23:45 < kanzure> Hrm.