🚀 go-pugleaf

Do computers cheat at chess?

This discussion has been around many times. The argument
is that chess programs use internal representation of
the chess board, i.e. another board and/or an opening
book. A human player is allowed to bring neither an extra
board or an openings book. Therefore chess programs is
cheating and should not be allowed in human competition.

This is all very confused. The argument can be met at two
different levels. First, human players think of the board
position or different aspects of it, and they usually also
have learnt a few opening variations from a book. The same
thing should be alright for a chess program. Isn't it quite
obvious that memory, even say a disk, is an internal part
of the computer system? I think that one should compare a
human player with the entire system, not only with the cpu,
disk or (let's say) the housing. In this way, the analogy
argument does not say that computers cheat at chess.

But of course there is a good formulation of the argument.
I think it could be simply "Chess programs are not (even
comparable to) humans, so they should not be allowed in
human competition". This is both simple and, I think, valid.
The solution to the whole problem is to have different
classes, human, computer and open to all.

Jan Eric Larsson
Department of Automatic Control
Lund Institute of Technology
Lund, (Close to Copenhagen), Sweden

Path: !seismo!mcvax!enea!alibaba!regler!janeric

By the way:

An opening book doesn't usually give a program very much
in playing strength. Most tests actually points in the
opposite direction if anywhere at all. The main use of
the book in a program is to make its play more interesting
to humans. I would be very sorry if opening books in chess
programs were skipped.