[info] [croquet-user] Wow. Network monitoring in 3D...

[Eugen Leitl]

----- Forwarded message from Les <hlhowell at pacbell.net> -----

From: Les <hlhowell at pacbell.net>
Date: Fri, 20 Apr 2007 11:59:02 -0700
To: croquet-user at duke.edu, PSHELDON at flash.net
Subject: Re: [croquet-user] Wow. Network monitoring in 3D...
X-Mailer: Evolution 2.8.3 (2.8.3-2.fc6) 
Reply-To: croquet-user at duke.edu, Les <hlhowell at pacbell.net>

Hi, Paul,
	I have also had that thought.  Our minds appear in most cases to
operate like a relational database, but with many many linkages.  I
think of it like a wire and styrofoam ball construction, with millions
of styrofoam balls and wires linking them together, not just a few
connections, but perhaps thousands from some areas, like you home links
to the books in your library, your computer, its programs, tools, car,
magazines, newspapers, etc.  I once tried to generate an equivalent
structure, using linkages that could change, and also be updated, along
with a "success count" which would raise the linkage on that node's
search order, to try to optimize the access of data.  It kind of worked,
but the machine was woefully inadequate.  Based on the results I saw, it
would take about 2TB of memory to get anything useful, and about a
400Ghz processor to make it work.  But today, if you could somehow
implement it using a CAPP architecture, it might be possible given
today's processor speeds.  The linkages and the optimization would
become problematic, because speed would then be dependent on the
locality of the data being linked to some degree, and on the tasking of
the specific processor where the data was resident.  But Croquet offers
much of the required functionality.  I am not familiar with creating
arbitrary linkages with objects to visualize how the architecture could
deal with the link table, nor how the link table would be constructed to
deal with platform variances and internet protocol linking.  That
doesn't mean it couldn't be done, just that I personally don't have all
the required information to make it a reality.  

	Another requirement would be movement of information to help optimize
search paths.  I even postulated a "sleep and dream" state to help sort
out new information.  That architecture would constitute "creative
thought" resulting from linking new concepts with existing knowledge,
and postulating changes in the internal model of the environment.
Strategies that appeared successful could be retained in a "possibility
memory" and when acted upon update the linkages and the success counts
for the paths through the thought matrix.

	Strategies that proved failures would be backtraced and have the
success counts decremented, signifying that the path was less productive
than some other path.  Over time, the matrix would self organize, and
pruning could take place based upon negative values of success counts.

	What then would be the "data".  I feature it as something like 20
questions postulations.  It could be a destination (a place), an action
(put the book on the chair), a literary composition ( a poem or a book),
or perhaps a design.  In other words something either physical,
philosophical, creative, or movement.  Or could be the embodiement of a
process.  Each node would be a deconstructed portion of the final
overall "thing", whatever it might be.  The machine so proposed would
need to be able to collate and evaluate information, say over the
internet, to have input from the environment, like camera's or weather
sensors, produce actions, say product movement by robotics, and
replicate findings either physical or via media in some form.  All of
this is somewhat possible today.  If you think of a world of balls of
thought, linked to other worlds with balls of thought, with messaging
and event parsing, you are pretty close to the basic construct.  In
addition one could visit the world(s) and interact with the on-going
thought processes.  

	Any thoughts?

Regards,
Les H
On Fri, 2007-04-20 at 10:18 -0800, PSHELDON at flash.net wrote:
> I fiddled with hyperlinking "Course in Physics" with 
> my own derivations and external references.
> 
> I wondered at a portrayal of a new market
> that wouldn't be competitive but working with
> publishers to make collective knowledge more
> useful.
> 
> I had to scan in pdf's but alerted them
> that they might like to make the books
> free and instead sell the networking of ideas.
> 
> How can one represent what is going on
> in an individual or collective mind?
> Something about a network of ideas
> or something more primative and wonderful.
> 

----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE