Harvard Doesn't Publish Science Fiction

 Hans Moravec
 The Robotics Institute
 Carnegie-Mellon University
 Pittsburgh, PA  15213
(412) 268-3829

Copyright 1987 by Hans P. Moravec


	Mea culpa: reality and fantasy blend in my mind.  My major
source of income is as a scientist, and, fortunately for all involved,
on my very best behavior I can pass as such.  What's more, it's fun.
Working out problems slowly, carefully and rigorously gives insight, a
sense of solidity, and of a job well done. Often it sheds light in
unexpected directions, and opens new doors. But it has limitations.
To do it, you must already understand the problem in a detailed way.
But many interesting questions are too nebulous and slippery for such
intimacy. My burden is that often I'm insensitive to the difference.

	Consider the question of intelligent machines. While the
entire idea was once viewed suspiciously by most of the scientific
community, the research has, by now, spun off enough practical results
to be reasonably tolerated.  On my best behavior, for my PhD thesis, I
wrote a program that enabled a robot to see well enough to cross a
cluttered room, building a map of it along the way [there's a joke
here - it took the robot five hours to make the transit, its million
dollar computer brain churning furiously the whole time].  At the same
time I couldn't help extrapolating that modest reality to very
immodest lengths. {\it Why are robots so much worse than animals at
the simplest things?  When can we expect that to change?  Will they
ever be a good as humans at most things?  Will they be better?
Cheaper?  Will they be able to carry on their own further improvement
without our help.  Is there some way we can avoid being left in the
dust?  What will the world be like after this happens?}

	By making some tentative assumptions and calculations, I was
able to conclude: {\it The computers are too small - insectlike
now. Slowly but steadily, 1000 fold every 20 years. Yes, in about 40
years.  Yes, rapidly thereafter.  Certainly, much.  Especially without
our help.  Only if we join them by rebuilding ourselves in their
image. Very different, much bigger and more interesting.}  While many
of my colleagues saw a big difference between the reasoning behind the
robot driving program and the futurism, I found little
distinction. Maybe it's the result of reading too much science
fiction.  In any case, several papers and essays on both subjects were
published, with the robot results showing up mainly in technical
literature, and the futurism in science fiction outlets.  But there
was crossover, in both directions.

	Which brings us to Harvard. A new editor at Harvard University
Press had read several of the futurism articles, and in January of
1985 invited me to submit an outline for a book expanding on the
theme.  His timing was excellent - I'd just started writing such a
book, after ten years of procrastination.  It was an opportunity to
develop many new ideas.  Some were about the evolution of our
machines, but others were about surprises in the universe that might
await our superintelligent progeny.  When my editor read a first draft
he explained that the book would have to be passed by an academic
review committee, and Harvard, by policy, does not publish science
fiction.  He felt also that many of the physics and astronautics
chapters impeded the theme of the book.  So the second draft abandoned
several ideas.  Fortunately Jim Baen {\it does} publish science
fiction, and some of the orphan chapters have found a foster home
here.

	The first article addresses the possibility of computers so
fast they violate apparent physical limits.  The secret is exotic
materials science.

\begin{center}
Superdense Computers
/users/hpm/project.archive/general.articles/1986/dense.ltx
\end{center}

	The second idea concerns time travel {\it without} violating
accepted physical laws.  The innovations here are mostly psychological
and philosophical.

\begin{center}
Time and Alternity by Computer
/users/hpm/project.archive/general.articles/1986/time.ltx
\end{center}

	Finally, here's the core of an idea that, if correct, would
put a radically new light on the fundaments of our space and time,
incidentally explaining the most bizarre effects of quantum mechanics.
Someone should develop it mathematically someday.

\begin{center}
The Harmonies of the Sphere
/users/hpm/project.archive/general.articles/1986/wave.ltx
\end{center}

\end{document}