Computer Chess Club Archives


Search

Terms

Messages

Subject: More about Knowledge Vs Fast Searchers

Author: Fernando Villegas

Date: 12:41:28 05/22/98



This one seem to be an eternal discussion, with some friends here stuck
to his first positions without taking a look at the reasonable things
that other points. Discussion is also prolonged by the effect of badly
defined terms, confusion, etc. So let me clear a couple of things so far
my understanding on this matter goes:
a) Knowledge is not a universal, homogeneous thing, but a relation
between a thinking device and the environment. Depends, then, of the
nature of the first. Dogs, cats, monkeys and human beings has knowledge
not only different in degree but in kind. And all of them are necessary
and efficient for the sake of the purposes of monkeys, dogs, etc. The
same for computer chess devices should apply. NOT necessarily our
knowledge based in general positional concepts and simple enunciation of
past experience -the so called opening  theory- is what computer needs.
As much as they thinks crunching great numbers, so knowledge for them
should be a very different thing that our system of signals post.
b) To exclude something in the analysis of a problem, this or that piece
of knowledge, is in itself a kind of knowledge, in fact, the essential
knowledge. In maths-test it is used a device called “distracting”
precisely to evaluate the capability of the student for grasping the
essential of the problem and reject the unnecessary data and principles.
 A guy that is messing around with unnecessary data or trying to use
Euclidean theorems to solve a termodynamic problem is showing total lack
of understanding.
c) The speed of a program is in itself a measure of knowledge in
computer terms. Is very simplistic to see speed as a thing-as-such or as
a “das ding an sich” in Katinan terms: this is for Thorsten   :-) and
independent of knowledge. You do with great speed what you know well.
That is called “mastery” in any art or craft. It involves that a cluster
of efficient proceedings are being used to discard unrelevant issues on
the run..

Fernando



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.