>Does anyone have the url for this article?
>
>Steve KK
>
Steve: NYTimes.com is a free subscription site. Make sure to write down
your password because it asks you to reup at long intervals.
Here's the article--copied from the site. Quite legal.
June 11, 2002
Did This Man Just Rewrite Science?
By DENNIS OVERBYE
ou can try this at home.
Take a sheet of graph paper that has been divided into grids. Color a
square in the middle of the top row black. Drop down to the next row.
Now invent a rule that will decide if a square should be black or white,
based on the square above it and that square's neighbors — for example,
that a square should be the same color as the one above it unless that
square has a black neighbor. Go across the second row filling in squares
accordingly, then repeat the process, following the same rule, for the
third row, the fourth row, and so on.
There are 256 rules you can concoct to play this simple game. Most will
create a boring or repetitive pattern. But at least one rule will cause the
page to explode into complex, ever-shifting patterns. You will have created
a so-called universal computer, equal in its computational
sophistication to Apple's jazziest laptop. Given the right starting
pattern, and the right rule, according to Dr. Stephen Wolfram, a former teenage
particle physicist and software entrepreneur who has been doing this at
home for the last 10 years, those lines and shapes cascading downward
can be made to pick out the prime numbers, compute pi, calculate your
income tax, or model the evolution of a star — anything a real computer
can do.
This insight is the jumping-off point of Dr. Wolfram's glossy 1,263-page
book, "A New Kind of Science," published a month ago by Dr. Wolfram
himself to the accompaniment of articles comparing Dr. Wolfram to Isaac
Newton. The book holds a No. 2 Amazon.com ranking. It returns to
the arena a prodigy who published his first physics paper at 15, earned his
Ph.D. from Caltech at 20 and two years later, in 1981, became the
youngest MacArthur "genius" fellow. In 1988 he founded Wolfram Research
Inc. to market his program, Mathematica.
"A New Kind of Science" may be the scientific publishing event of the
season, but whether it is a revolution in science as well must await the
judgment of Dr. Wolfram's peers. So far, some seem amazed by his courage,
others by his chutzpah. In the book Dr. Wolfram argues that the
ability of such a simple system to engage in complicated-looking behavior
means that scientists have underestimated nature, seeking complex
reasons where simple ones will do.
As a result, he says, science has been going in the wrong direction. Most
systems of even modest complexity, he concludes, are so complicated
that they are beyond the grasp of mathematical formulas. Science should be
looking for a simple program, not a T-shirt's worth of equations, if it
wants to explain the universe, a project, he says, that would redefine our
understanding of space and time, evolution, intelligence, free will, and
philosophy, as well as physics.
"If the whole history of our universe can be obtained by following definite
simple rules," Dr. Wolfram writes near the end of his book, "then at
some level this history has the same kind of character as a construct such
as the digit sequence of pi. And what this suggests is that it makes no
more or less sense to talk about the meaning of phenomena in our universe
than it does to talk about the meaning of phenomena in the digit
sequence of pi."
In conversations over the last fews weeks, computer scientists and
physicists expressed admiration for Dr. Wolfram's "engaging" and "beautifully
written" exposition of ideas, like the power of algorithms, that they say
are underappreciated by society.
Edward Fredkin, a physics professor at Boston University and the
Massachusetts Institute of Technology's Media Laboratory and a longtime
proponent of something called "digital physics," viewing nature as a
computer, said: "For me this is a great event. Wolfram is the first significant
person to believe in this stuff. I've been very lonely."
Dr. Gregory Chaitin, a mathematician at the I.B.M. Watson Research
Laboratory in New York, called it "a remarkable book," a courageous
attempt to create a new way of viewing the world.
But many scientists, while evincing admiration for some of his results,
also said they wondered how far the book went in a fundamental sense
beyond what had been said before, including by Dr. Wolfram himself in the
1980's, before he left academia.
Dr. J. Doyne Farmer, a professor of computer science at the Santa Fe
Institute and the University of New Mexico, called it "a well-written
summary" of ideas that have been around for a long time.
"We've known for 50 years that complexity can come from simpler and simpler
things," Dr. Farmer said, adding that it had been shown that even
bouncing billiard balls could be a universal computer.
Others said Dr. Wolfram was blurring facts and conjectures. "I don't feel I
can recommend the book to nonexperts because a nonexpert won't be
able to tease out what's really known from what's speculated," said Jaron
Lanier, a computer scientist, musician and virtual reality pioneer. Dr.
Wolfram's critics also complain that he has an almost chronic inability to
share credit for ideas — all the history and names of other scientists have
been relegated to a 349-page section of "General Notes" in the back of the
book. Even Mr. Fredkin, who said he was a longtime friend, said Dr.
Wolfram had "an egregious lack of humility."
The idea that complex things can arise from simple ones is as old as
Euclid, who built a whole geometry out of a few axioms and logic, but the
giant on whose shoulders Dr. Wolfram is most securely standing is the
English mathematician Alan Turing. In 1936, Mr. Turing and Dr. Alonzo
Church, a Princeton mathematician, showed that in principle any
mathematical or logical problem that could be solved by a person could be
solved by a so-called Turing machine. As envisioned by Mr. Turing, it was
like the head of a modern tape recorder that would move back and
forth along an endless tape reading symbols inscribed on it and writing new
ones. Moreover, a so-called universal Turing machine could emulate
any other conceivable computer.
From that point on, Windows 95 and the Internet were only matters of time
and transistor technology.
Over the years, computer scientists and mathematicians have labored to
whittle down the requirements for how simple a universal Turing machine
can be. Dr. Alvy Ray Smith, a computer scientist and a founder of Pixar,
described this process in an e-mail message as "fun and games" but
beside the point. "The fact is that computation is known to be a thing of
surpassing complexity generated from a simple set of rules. That's why we
do it."
Cellular automata, the grids of black and white envisioned at the beginning
of this article, were invented in the 1950's by Dr. John von Neumann,
physicist and mathematician at the Institute for Advanced Study in
Princeton as a way of developing models of biological reproduction. Although
they seem at first glance to bear little relation to Turing machines, work
by Dr. von Neumann later showed that at least one version of a cellular
automata, in which the cells could have 29 colors, was also a universal
computer.
By the early 70's, many more automata had joined the ranks of universal
computers, including the Game of Life, a cult classic invented by Dr.
John Horton Conway, a Cambridge mathematician.
Dr. Farmer said cellular automata had attracted attention because they were
like physics. "There are simple rules, local in nature, with information
propagated from one place to another. It's a very physicslike framework."
If it's easy to see the universe in cellular automata, it is also easy to
see cellular automata in the universe, which by the definition of a universal
computer is itself a universal computer. Every physical action, from the
crashing of a wave to the rusting of a tin can, can be thought of as a
computation in which the universe moves from some initial input moment by
moment to an outcome — from a cresting ridge of water to bubbles
of foam on the sand. For some scientists, like those who divide the globe
into a fine grid to calculate approximate numerical solutions to
horrendously complicated equations describing weather dynamics, this is a
convenient metaphor.
For others it is more. The notion of information has become a vital
component of modern theories of black holes and of the origin of quantum
mechanics. Mr. Fredkin said he had started mulling the universe as a
computer back in the 1950's but began publishing papers about it only
around 1980. "An equation is just a thing you write down on a piece of
paper. E=mc2 can't keep you warm," he explained. But programs are
different. "Put them in the computer and they run."
Dr. Wolfram joined the cellular automata fray in 1980. He devoted himself
to systematically studying what happened when simple programs were
allowed to go on running, something he said that other scientists like Mr.
Turing could have done but did not do. Indeed, he said in an interview,
scientists seeking predictable behavior had regarded the chaotic patterns
of some cellular automata as a "nuisance." But Dr. Wolfram said that as
far back as 1984 he suspected that these automata could be a universal
computer.
Dr. Wolfram said his book was an effort to build a big conceptual
structure, quite different from what's been there before. "I'm not
surprised some
people don't get it at the first reading," he said, adding that it might
have to be read many times, the way people pore over philosophy books.
At its core is what he calls the Principle of Computational Equivalence,
which is that all systems that exhibit more than simple behavior (like the
automata cells on this page) have equal computational powers — they are all
universal computers. This means, he conceded when asked in an
interview, that even a bucket of rusting nails should have, in principle,
the computational oomph of a desktop computer, a human brain or the
universe itself.
Among other things, this principle puts limits on science, he said. Since
no universal computer can outstrip any other, most things in the world are
inherently unpredictable — "computationally irreducible." Even if
scientists found the ultimate rule that runs the universe, they wouldn't
know what
it did or said about, say, the masses of quarks or the origin of life,
without running the program. There are no shortcuts.
Some computer experts said that this principle was just a fancy restating
of the Church-Turing hypothesis, which equates all universal Turing
machines. Dr. Wolfram says, "It is completely not." For one thing, he said,
that former theory was framed in terms of numerical mechanical
calculations, and many physicists doubted that it applied to natural systems.
Perhaps most important, Dr. Wolfram said, his principle implies that there
is no gradation between simple and sophisticated computers. Anything
above a small threshold is the same, thus enforcing a kind of rude cosmic
democracy.
"Maybe we are not any more complex than cellular automata," Dr. Wolfram
said, "but the universe is no more complex than us. We're not above
the universe, but we're not below it either."
Among other complaints, Mr. Lanier and other experts argued that Dr.
Wolfram had not defined the threshold of complexity. Moreover, the
problem in computer science today was not how to build a universal machine,
but what to do with it.
"The laptop in front of me is probably large enough to hold programs that
would utterly change not only my life, but just about anything else," Mr.
Lanier said. "It could cure diseases, design starships, and express an
exquisite theory of quantum gravity. Alas, I don't know how to find the right
programs to accomplish these things."
Dr. Wolfram said he had a good time writing his book. "I found it really
interesting and exciting to understand the foundations of things," he said,
adding, "I hope I succeeded in communicating the excitement I had in
figuring things out."
He also said he looked forward to facing a jury of his peers in a lecture
tour this fall. Invitations have been pouring in as word has spread that his
self-imposed exile from academia is over.
"I'm looking forward to not being a recluse," he said.
Copyright 2002 The New York Times Company
| Permissions | Privacy Policy
|
