Robin Hanson:
<I'd expect an advanced computer to notice if some external force
started to much around with its innards, and to do something about it.
So if the world around us is an advanced computer, it knows about us and
chooses for some reason to allow our activities to continue. Either we
are part of this computation, or it works around us.>
Can anything humans do really be considered the application of an
external force? If the universe is some sort of giant irreducible
computation, we are, of course, part of it. Compared to any 'higher
levels' of this system, we are perceived, perhaps, at the same level
that we perceive our own intestinal fauna.
Robin:
<How much physics do you know Steve?>
Most of the people writing about the universe as a 'computation' or "sea
of information" are physicists; e.g., Paul Davies _The Mind of God_,
Fred Alan Wolf _The Dreaming Universe_, Brian Josephson, David Bohm,
Nick Herbert, and so on (though some of these do tend to get a little
mystical).
Robin:
<I'm not sure you realize how very much we do know about all that
information streaming around us. We know which systems are exchanging
bits with each other, and how fast, and where this arrangement changes,
etc. We also know a lot about designing computers, carefully arranging
the routes and speeds of bit exchanges. And these two really look
nothing alike!>
Our current computers are optimized for serial, not parallel
computations, have inadequate associative memories, and are pitiful at
communications.
What we know is only a tiny fraction of what we have yet to learn. How
much do we really know about the initial conditions of the many chaotic
biological and economic systems that we're still just identifying?
I just happened* to reread something that physicist John Cramer wrote
(in his 3/92 Analog column): "Chaos can be removed, reduced, or steered
by monitoring and feedback or by introducing periodic perturbations of a
carefully chosen frequency."
(* Thanks to Damien Broderick for the URL:
www.npl.washington.edu/AV/av_index_sub.html)
And then, Robin:
<Last year the Econ Nobel went for Game Theory, this year for Mechanism
Design, both areas I've been specializing in. Maybe next year they'll
do signaling games.>
I read the Academy page and noted that they said: "Incomplete and
asymmetrically distributed information has fundamental consequences,
particularly in the sense that an informational advantage can often be
exploited strategically." I assume that by "Mechanism Design" and
"signaling games" you're referring to approaches for controling these
chaotic, natural systems.
But, as Cramer noted in that same article:
<What's going to happen when traders realize that they can steer and
control the fluctuations of the market by introducing periodic
perturbation ... when the trick is learned and many operators are doing
it, the effects will be cancel out and the Market, 'efficient' as
always, will go its own way.>
Anyway, while studying "the routes and speeds of bit exchanges", might
give us clues on how to control or optimize them, it doesn't tell us
much about the application being performed or its relation to other
applications. I'm certaintly not going to be just looking at the
circuitry and electron flows in my computer to find out what it's doing.
Another example: when we look at a Penrose tiling (like the cover of the
latest Science News), we can either notice the individual tiles
themselves and the rules by which they can be connected, or we can see
the overlapping and repeating patterns that they make on the larger
scale. Both views have utility.
I think Lyle's point ('made-to-order cells', Monday, October 14, 1996
4:43PM) is relevant here:
<Eric Drexler envisions little assembler arms picking atoms up and
putting them in place, one by one. In cells, atoms are guided into
place by their three-dimensional environment ... It is possible to do
the same thing with cellular machinery that Edison and others did with
electrical machinery a hundred years ago.>
Nature uses a cellular automata with genetic algorithms approach. The
Drexlerian approach requires a pre-defined, step-by-step program. While
this might be the most efficient, it's certaintly not the most flexible.
I suspect we'll eventually combine both approaches where appropriate.
Steve Witham:
<I would expect a colonized universe to look exactly like a barren one.>
Robin:
<To me, the idea that this circuit diagram is actually an optimum design
for an advanced computer seems completely crazy.>
What is "an optimum design"? Would it do away with redundancy? Would it
be based on crystalline rather than fractal patterns?
Robin:
<It seems to me that the outcome of all this strategizing is that the
best strategy is just to expand as aggressively as possible. By the
time anyone comes to destroy any one system you have used it well,
converting it to a billion probes or much more within a few years. That
huge economy gives you all the better chance to learn new technology, so
some of your descendants become as advanced as your advanced enemies.>
How is this different from Vernor Vinge's Blight? The image I get from
this is a homogenous nanofog out to conquer the universe, leaving in its
wake only a Black Goo, or less, converting everything to its
pre-programmed optimal design. How would such a system have any
creativity and what "new technology" could it ever develop?
I'd expect a colonized universe of this type to look very barren, more
like a black hole or a 'premature' Omega Point.
Mark Crosby
"My purpose in writing _Mind Tools_ has been to see what follows if one
believes that everything is information. I have reached the following
(debatable) conclusions: 1) The world can be resolved into digital bits,
with each bit made of smaller bits. 2) These bits form a fractal pattern
in fact-space. 3) The pattern behaves like a cellular automaton. 4) The
pattern is inconceivably large in size and in dimensions. 5) Although
the world started very simply, its computation is irreducibly
complex.... And where is this huge computation taking place? Everywhere;
it's what we're made of."
- Rudy Rucker