X-Message-Number: 4303
From:  (Thomas Donaldson)
Subject: Re: CryoNet #4288 - #4294
Date: Thu, 27 Apr 1995 00:02:25 -0700 (PDT)

Hi again!

In response to Mr. Clark's comments:

I note that he does not actually give a definition or refer to one. I also
note that he SEEMS to equate "information-processing machine" with a 
computer. I think that some further definitions might sharpen my question.

My main problem with thinking of neural nets as computers comes from the fact
that they aren't programmed in any traditional sense. There is no special
loading of any sequence of directions to which the machine will respond. 
Instead they are TRAINED: given a large or small set of examples, and told
when their response is correct or not. I don't believe that such machines
were in the minds of either von Neumann or Turing when they did their 
work on those machines we would now refer to as computers without question.

And I'm unhappy with the notion that because each can (in a primitive way)
imitate the other, that must make them equivalent. The quality of the
imitation should have some weight. Furthermore, the equation of "computer"
with "information processing machine" has a problem: just what kind of
processing consists of "information processing"? Is a printing press an
"information processing machine"?

In any case, I didn't ask my question to do philosophy. I wanted to know 
what the standard definition of computer was. And if there is no standard
definition, I'd like to know that too.

			Best and long long life,

				Thomas Donaldson


Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=4303