From: Harvey Newstrom (mail@HarveyNewstrom.com)
Date: Tue Jul 02 2002 - 07:49:12 MDT
I just finished reading "Review: Stephen Wolfram's 'A New Kind of
Science'" by Ben Goertzel
<http://www.extropy.org/ideas/journal/current/2002-06-01.html>. It was
a wonderful review, and mirrors a lot of my own thoughts about the
book. However, a couple of points peaked my pessimism:
Goertzel says "The chapter on 'Processes of Analysis and Perception' was
a big disappointment to me, perhaps because it came so close to my own
area of research, Artificial General Intelligence. I found /no/ really
useful insights in this chapter." He then posits that "Maybe this is
just because Wolfram's intuition in these areas is weaker than in other
domains...."
This seems to be the reaction by most people. They seem to think the
book is accurate or even brilliant in areas of science of which they are
only peripherally acquainted. But where the book delves into their own
personal area of expertise, they suddenly notice a lot of deficiencies.
This reminds me of the old story about an Astronomer and an
Anthropologist discussing Velikovsky's "Worlds in Collision" which
proposed that ancient myths about dieties actually described planets
literally interacting throughout history. The Astronomer observed that
the book seems excellent in the area of history whereas the astronomy
was fanciful. The Anthropologist replied that he thought it was the
complete opposite. Similarly, this book seems to be in the same
situation. Everyone seems to praise it as brilliant in /other/ fields,
but find it totally lacking within their own field of knowledge.
The most disturbing problem I have with Wolfram's work is his apparent
desire to map his theory onto the Universe, rather than deriving a
theory from observations of the Universe. In fact, he seems quite
willing to skip real world observations, preferring to assume that his
simulations are more accurate. I even fear that there may be a bit of
circular deduction here. Wolfram seems to have started his research by
building cellular automata simulations. He then studied his own
simulations and found cellular automata underlying everything. He later
expands this into a theory that the entire universe is composed of CA.
But all he really discovered is that his own simulations were built on
CA's, which we knew already because of the way he built them. He seems
to have skipped the part about real-world experimentation or observation
of the Universe.
This is illustrated by Goertzel's note that "Perhaps the most
controversial part of the book will be his chapter on physics," which he
describes as "definitely the most speculative part of the book". Even
though physics is Wolfram's original specialty, the reviewer notes that
"The graph-rewriting systems he describes here are the least CA-ish
systems he considers anywhere in the book: here, more than anywhere
else, he seems willing to morph his modeling approach to match the
phenomenon under study." In the one area where Wolfram could
demonstrate some expertise, he seems to be the weakest. He has to
modify his methods and morph his modeling approach to get the results to
come out right. This should be the most supportable area of the entire
book, given Wolfram's expertise and the unbiased objectivity of physics,
yet this is where the theory seems to be least demonstrable.
It is for these reasons that I find the book to be a total failure as a
"New Kind of Science". Unlike real science, it does not derive a theory
from observations, it does not predict observations, nor does it produce
verifiable results. Instead, I would consider the book to be a modeling
methodology. Cellular automata can be used to create complex patters
from simple programs. As such, it might be a programming tool
applicable to VR or games, or a mathematical method for describing
complexity. But it seems to fall short of any rigorous definition of
"science".
-- Harvey Newstrom, CISSP <www.HarveyNewstrom.com> Principal Security Consultant <www.Newstaff.com>
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:08 MST