From: Robin Hanson (rhanson@gmu.edu)
Date: Wed Jul 18 2001 - 09:03:09 MDT
Robert Bradbury wrote:
>I'm getting the feeling that people haven't read my
>papers in detail, or I'm not making myself clear in
>my messages, so I'll make *one* more try:
Oh, I'd bet it will be much more than one more time.
Maybe not this week, but over the many years to come. :-)
But I'm happy to revisit memory (and future) lane again
with you.
Btw, if you're going to complain that we aren't reading
your papers, you should point out which paper(s) of yours
are most relevant. It isn't obvious to me when I look at:
http://www.aeiveos.com/~bradbury/CurrentWork.html
>... Natural selection *FAILS* when the S-curve of the singularity
> flattens out at the top with the complete optimization of
> all local matter at the atomic engineering scale.
>Just as humans have *almost* trumped natural selection on
>the Earth (no more bears in Europe or tigers in India...),
>JBrains & MBrains will trump natural selection in the universe.
>They "FIX" the hazard function so it is virtually nonexistant
>(at least until the protons decay). After you trump natural
>selection, the logical game is self-directed conscious evolution
>(which may rely on natural selection for internal development
>strategies, but it isn't visible externally). ...
This summary of your position seems completely inadequate to me,
and I'm guessing to most of the rest of us. I suspect that you
really need to write a paper directly on this specific topic,
where you elaborate your argument in detail. We just don't get
it from your many quick summaries.
Here are some examples of ambiguities.
Consider very simple artificial life sims where the creatures
reach a complete optimization of their local "matter". How
does natural selection "fail" in this context?
Is there some analogous claim about conscious evolution that
you could verify in a simple artificial life sim, where some
creatures could self-modify?
Humans have been self-conscious for millions of years, but
also evolved a lot during that time. So consciousness seems
compatible with ordinary evolution.
Humans are still changing, with some types reproducing more
than others. Some types of our machines also get copied more
than others. Some types of AIs & uploads will also presumably
reproduce more. This makes it seem that humans and their
descendants are still participating in an evolutionary
variation and selection. So how can you say that humans
have trumped natural selection?
I think some bacteria are immortal. That is, they divide
and may get killed, but are physically capable of living
forever in a friendly environment. How is this different
from your claim that JB/MB have zero hazard functions?
My main point here is to show you just how wide the
communication gulf here is. And, for better or worse, it
is your job to bridge most of that gulf. Maybe not here
and now, but at some point.
Robin Hanson rhanson@gmu.edu http://hanson.gmu.edu
Asst. Prof. Economics, George Mason University
MSN 1D3, Carow Hall, Fairfax VA 22030-4444
703-993-2326 FAX: 703-993-2323
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:08:52 MST