Real vs. Simulated Ecologies [was: Moravec article available]

From: Robert J. Bradbury (
Date: Wed May 03 2000 - 09:13:22 MDT

The Fermi Paradox argues that Robin is correct and intelligent life
capable of constructing and managing its niche is very rare, or
a very different set of ecological management tools, e.g. a
Morovecian simulation ecology takes over for the "natural" methods.

The Moravec (Robots) perspective seems to me to be still operating
in the "real" universe. I.e. the point of the simulations is to
allow the construction of something physical. However, if reasonable
goals of conscious evolution are to increase complexity as efficiently
as possible and minimize the "real" losses associated with the deaths
of bad ideas (taking a page from the Book of Burch that wants push the
morality of the mind as far as possible and still allow for any evolution
at all), then there would seem at some point to be the necessity for a
logical transition from operating in the "real" world to operating entirely
in the "virtual" simulation.

Some questions related to this would be:
(a) What are the (energy/entropy) costs of destroying information in
    the real world vs. the virtual world? [Reversible computing is
    going to produce the densest computing but it does so at the
    expense of minimizing the erasing bits.]
(b) Is the real world less expensive than the virtual world for
    "storing" an arbitrary amount of information? [It seems to
    me that a virtual world based on storage in photons should
    be cheaper than a real world based on storage in atoms.]
(c) Does a virtual simulation where any conscious sub-SIs are "informed"
    that their "beingness" may at any point be "suspended" or even
    "erased" satisfy the Burch "Morality of Mind" requirements?
    [I.e. you are in approximately the same boat we are in today
     in that death from "accidents" may occur, only in the moral-SI
     simulation it would be "painless".]

So, the elimination of unconscious species is of little interest to any SIs,
since it is a "process" of the universe, like the birth and death of stars.
Conscious species may be of interest since they may "join the club".
That leads to the allow vs. disallow interventions questions which
would seem to depend on the relative value of additional club members
vs. uniquely evolved club members (that may increase diversity).

If any of the above holds water, then we are looking at the Fermi
Paradox from the perspective of a "flat earth", not realizing
the full realm in which ecologies and minds can operate and thus
have a difficult time answering the question correctly.


On Wed, 3 May 2000, Robin Hanson wrote:

> Hal Finney wrote:
> >... we see here on Earth a diverse ecology which includes organisms
> >that cover a wide spectrum of rates of adaption. There are species
> >which are almost unchanged from hundreds of millions of years ago,
> >and others which are less than a million years old. ...
> >In that case we have a choice, not between adapt or perish, but between
> >adapt or stagnate. It may turn out that there is an ecological niche
> >for organisms who are wedded to their past, who refuse to change.
> OK. But keep in mind that if we look at *all* of the species that
> existed say 400 million years ago, we would likely find very few of
> them are still around. So the chances of making it to such a cozy
> niche may be rather small.

This archive was generated by hypermail 2b29 : Thu Jul 27 2000 - 14:10:25 MDT