X-Message-Number: 4236
Date: 18 Apr 95 16:59:59 EDT
From: Paul Wakfer <>
Subject: Learning instead of Attacking
While I am one of the strongest proponents of the philosophy that the
universe is "knowable"; that, in time, we can gain an understanding of
anything the exists; and that we should be positive, optimistic and
confident in our ability to deal with the world and each other, I am
beginning to find a bit tiresome, the narrow-minded, arrogance of some on
this list who often don't even know enough to know what they don't know.
As an example, I use the recent message #4232 from John K Clark
criticizing a previous message from Thomas Donaldson. John's writing seems
to oscillate between well reasoned exposition and silly, nit-picking
replies. But I use his message only as one of the most extreme examples of
what I see in the writings of many others.
In that message he again states what everyone here must now know by
heart (even the few who didn't already know it): that all computers
including neural nets are *computationally equivalent* to Turing machines.
But what neither he nor anyone else has done here is to refute, or even
address, the statement that neural nets (and parallel machines in general)
will always be able to do things faster in principle than Turing machines.
Since we live in a reality which includes time as a very important
attribute, this clearly makes neural nets fundamentally different from
other types of computers.
In correcting Thomas' spelling of Godel's name, John really shows his
ignore and arrogance. If he had any shred of common sense he would realize
that Thomas' is no dummy; he has been around a lot longer and seen a lot
more than John has; and he is not likely to make such a mistake and then to
repeat it, unless he has a good reason. John, if you would get the god-
damned chip off your shoulder and stop acting like such a conceited ass,
you just might actually learn a thing or two. In fact, "Godel" was born in
Czechoslovakia and the "o" in his name should have an "umlaut" (2 dots)
over it. "Goedel" is actually a *better* rendering into english than
"Godel". Furthermore, I have seen many books and papers where it *is*
spelled that way, although probably more older ones than current ones.
Finally, in reply to Thomas' very wise observation:
>And so Goedel's Theorem will be forgotten. That is, of course,
>the general fate of negative results.
John makes a flippant and derisive statement listing a few important
negative results that have not been forgotten. John, will you please try to
open your mind and widen your vision? Yes, there are a few negative results
which have survived and will continue to survive, mainly because they
actually say something very important and positive. Thomas may be wrong
about this particular negative result, frankly, I suspect that he is. But
the wisdom which comes from many years of reading and thinking about, and
in doing science and mathematics shows clearly that, in general, he is
right. The conjectures which are disproved get forgotten and the ones that
are proved get collected up, published and remembered.
While I have singled out John K Clark for my comments, I hope that
others who attack before even taking the time to realize that there just
might be a lesson to be learned, will in this instance, at least, learn a
lesson from my message.
-- Paul Wakfer --
*****************************************************************
Paul Wakfer 1220 E. Washington St. #24, Colton, CA 92324, USA
238 Davenport Rd. #240, Toronto, ON M5R 1J6, CANADA
Pager:416-446-9461 Phone:416-968-6291
*****************************************************************
Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=4236