Re: [NetBehaviour] Entropy

[marc garrett]

Hi Daniel,

Thanks for sharing with us some of the context, structure  decisions 
around using The Eliza program. I was intrigued of its origin and found 
this info about it on wikipedia...

ELIZA is a computer program and an early example (by modern standards) 
of primitive natural language processing. ELIZA operated by processing 
users' responses to scripts, the most famous of which was DOCTOR, a 
simulation of a Rogerian psychotherapist. Using almost no information 
about human thought or emotion, DOCTOR sometimes provided a startlingly 
human-like interaction. ELIZA was written at MIT by Joseph Weizenbaum 
between 1964 to 1966.

Also visited here (http://www.manifestation.com/neurotoys/eliza.php3) 
ELIZA first appeared in the 60's, some people actually mistook her for 
human.

 I'm a big fan of Rosa's work but knew little about the demoscene,
 so it was great to read about it from her perspective and consider
 its connection to Glitch Art.

I'm watching where Rosa  JonCates and other peer glitchers take it, 
noticing they are bringing in their own flavour and calling it 'Critical 
Glitch Artware', and not Glitch Art - adding 'software art' element to 
it all (it seems).

Oh yeah, before i forget - do you know http://runme.org/ ?

I'm sure you do... I thought that your recent project would fit well 
amongst their software art repositories platform...

 the breakdown of this data is visible as it runs: you see the
 data corroding as Eliza's speech breaks down.

I would like to see this in action.

  I would love to write a Linux compiler --
 will definitely let you know if/when it happens.

I look forward to the Linux compiler version of 'Entropy', in the 
meantime will hassle some one who has Microsoft/Windows and download it :-)

Wishing you well.

marc


  Marc,
 
  Thanks. I'm a big fan of Rosa's work but knew little about the 
demoscene, so it was great to read about it from her perspective and 
consider its connection to Glitch Art.
 
  A big part of the Glitch aesthetic to me is the loss of stability. I 
thought this would be particularly interesting to apply to the 
experience of programming, since programming often has a rigid quality. 
Programmers need to get things just right or the program will be 
defective if it runs at all -- there's little room for approximation. 
But, of course in the real world programs are buggy -- even when they're 
written perfectly (which rarely happens), they run on operating systems, 
use standard libraries, and these huge bodies of code written by other 
people have their own bugs and inconsistencies (luckily for us Glitch 
Art folks). All code is faulty somewhere, if you dig deeply enough.
 
  Entropy brings this chaotic nature to the forefront. When you code in 
Entropy, you need to let go of this sense of control -- the most you can 
hope for is that the person using your program gets the gist of what 
you're trying to do. To program in Entropy means treating all data as 
limited resources that have unspecified number of uses before they've 
veered far enough away from their original values that they're 
essentially random.
 
  The best Entropy programs are ones that are ambiguous and 
structurally flexible. So if a loop runs fifty times instead of 
forty-five, it won't crash -- they conform to its corrosive and 
approximating nature. When I wrote the Eliza program, since the same 
data is used over and over, the breakdown of this data is visible as it 
runs: you see the data corroding as Eliza's speech breaks down.
 
  Hope that answers your question
 
  I would love to write a Linux compiler -- will definitely let you 
know if/when it happens.
 
  -Daniel
 
 
  ___
  NetBehaviour mailing list
  NetBehaviour@netbehaviour.org
  http://www.netbehaviour.org/mailman/listinfo/netbehaviour

___
NetBehaviour mailing list
NetBehaviour@netbehaviour.org
http://www.netbehaviour.org/mailman/listinfo/netbehaviour

Re: [NetBehaviour] Entropy

[dave miller]

hi daniel

This sounds really great - especially like the drunk Eliza idea.

I did a couple of projects a couple of years ago that used the Eliza idea:
http://davemiller.org/index.php?nav_item=Gallerygallery_nav=buddy_rivers_live

http://davemiller.org/index.php?nav_item=Gallerygallery_nav=a_walk_in_the_park_with_paxton

Will have a good look through your project

cheers, dave



On 7 July 2010 04:15, Daniel Temkin dan...@danieltemkin.com wrote:
 Hi, I'd like to introduce my programming language, Entropy. Entropy was
 inspired by the Glitch Art aesthetic. Data decays as the program runs. Each
 value will alter slightly every time it's used, becoming less precise, and
 holding less meaning, until every string becomes a line of gibberish. Only
 by not accessing the data at all -- not reading from it or writing to it --
 can it be kept intact.



 My Entropy page: http://danieltemkin.com/Entropy.aspx



 Code samples can be found here: http://esolangs.org/wiki/Entropy



 I wrote the classic Eliza program using Entropy objects, results can be
 found here: http://danieltemkin.com/blog/post/Drunk-Eliza.aspx



 Thanks!



 -Daniel

 ___
 NetBehaviour mailing list
 NetBehaviour@netbehaviour.org
 http://www.netbehaviour.org/mailman/listinfo/netbehaviour

___
NetBehaviour mailing list
NetBehaviour@netbehaviour.org
http://www.netbehaviour.org/mailman/listinfo/netbehaviour

Re: [NetBehaviour] Entropy

[Daniel Temkin]

Marc,

Thanks. I'm a big fan of Rosa's work but knew little about the demoscene, 
so it was great to read about it from her perspective and consider its 
connection to Glitch Art.

A big part of the Glitch aesthetic to me is the loss of stability. I 
thought this would be particularly interesting to apply to the experience 
of programming, since programming often has a rigid quality. Programmers 
need to get things just right or the program will be defective if it runs 
at all -- there's little room for approximation. But, of course in the real 
world programs are buggy -- even when they're written perfectly (which 
rarely happens), they run on operating systems, use standard libraries, and 
these huge bodies of code written by other people have their own bugs and 
inconsistencies (luckily for us Glitch Art folks). All code is faulty 
somewhere, if you dig deeply enough.

Entropy brings this chaotic nature to the forefront. When you code in 
Entropy, you need to let go of this sense of control -- the most you can 
hope for is that the person using your program gets the gist of what you're 
trying to do. To program in Entropy means treating all data as limited 
resources that have unspecified number of uses before they've veered far 
enough away from their original values that they're essentially random.

The best Entropy programs are ones that are ambiguous and structurally 
flexible. So if a loop runs fifty times instead of forty-five, it won't 
crash -- they conform to its corrosive and approximating nature. When I 
wrote the Eliza program, since the same data is used over and over, the 
breakdown of this data is visible as it runs: you see the data corroding as 
Eliza's speech breaks down.

Hope that answers your question

I would love to write a Linux compiler -- will definitely let you know 
if/when it happens.

-Daniel
 
___
NetBehaviour mailing list
NetBehaviour@netbehaviour.org
http://www.netbehaviour.org/mailman/listinfo/netbehaviour