A story about Python... sort of

Anton Vredegoor anton at vredegoor.doge.nl
Thu Jul 10 11:53:43 CEST 2003


"Greg Ewing (using news.cis.dfn.de)" <g2h5dqi002 at sneakemail.com>
wrote:

>Hmmm. So, if the universe is open and contains an infinite
>amount of matter, then we could solve chess eventually,
>but we might have to wait a while for the limits of the
>observable universe to expand sufficiently...

IMO it all depends on the amount of low hanging fruit in the universe.
For example if at some  early stage in the game the Queen could be
captured cleanly, the rest would be just minor details, unless some
careless move would give the game away of course.

For example at the department of informatics in Maastricht, there was
someone (Victor Allis IIRC) who had a habit of inviting people in a
most friendly and convincing way into his "torture chamber" where one
could play a game of "gobang" or "four in a row" or something like
that (it's a long time ago, I'm not sure about the details anymore)
against an innocent looking computer program.

As time went by the program got stronger and stronger and the faces of
the people leaving the room got more and more somber. Winning a game
just resulted in the program not making the same mistake next time :-)

In the end he managed to prove that the program was invincible, not
because he could generate the whole game tree, but because he could
maneuver the game into some position that was always winnable for the
computer.

Anton




More information about the Python-list mailing list