Alt-BEAM Archive

Message #03710



To: Wilf Rigter Wilf.Rigter@powertech.bc.ca
From: Steven Bolt sbolt@xs4all.nl
Date: Tue, 25 May 1999 23:16:38 +0200 (CEST)
Subject: [alt-beam] Re: BEAM and machine evolution


On Mon, 24 May 1999, Wilf Rigter wrote:

> A reasonable reply from Steven raising interesting questions:
>
> 1. Cell like Nv neuron circuits could scale in size and numbers to be
> fabricated by standard IC technology.

Despite their being around for quite a while now, the MicroCores,
Bicores and other members of the Nervous Net menagerie have not
been scaled beyond very modest levels. No method or theory which
moves things in that direction has been published. Imho you are
contributing to BEAM myth here. A little mystery and myth no doubt
furthers the Cause, but let's hang on to some credibility as well.

> 2. Like any technology Beam is searching for a "killer application"

To take it where? Imho BEAM is neither too hot nor too cold, but
just right at entry level mechatronics. Your best bet is horizontal
expansion, and that needs a growing number of local workshops and
hobby clubs, rather than a killer app. Robots like Mark T's
Walkman, Zoz' Lurch, the good looking little guys made by Paul,
Darrell, Jim and many others should be quite sufficient to draw
people in (people and `bots mentioned at random).

> Take the capaciflector circuit which is about the same level of
> complexity as a microcore and has a real feel of
> tinkering/optimizing about it.

Well... BEAM could certainly do with more and better documented
circuitry. Keep up the good work!

> 3. Yes, use a biological model for optimum growth:

:)

> 4. Self assembling or adapting machines.
> Self learning adaptive programs
> Neural networks
> Automated factory building factories
> Trial and error correcting feedback loops

Not all the above are in the same league, for real, or BEAM...

> If I can think of an autonomous modular scalable and not so
> predictable self-replicating and robot building automated factory
> with robots gathering raw materials I can see that machine
> evolution is possible.

Far fetched - and hardly BEAM.

> Sure, the trick is to get from Beam to there.

I hope not :)

> Do we really want such machine life to compete with us for resources?

Frankly, such machine `life' is sufficiently far-fetched to not
bother me at all. I worry about cars, bulldozers and large aircraft
competing with me for resources...

Best,

Steve

----------------------------------------------------------------------
# sbolt@xs4all.nl # Steven Bolt # popular science monthly KIJK #
----------------------------------------------------------------------






------------------------------------------------------------------------

eGroups.com home: http://www.egroups.com/group/alt-beam
http://www.egroups.com - Simplifying group communications



Home