Author: Duncan Stanley
Date: 15:50:41 04/21/01
Go up one level in this thread
On April 21, 2001 at 18:32:32, Robert Raese wrote: I'm sorry I completely confused your name earlier .... >On April 21, 2001 at 13:02:48, Thorsten Czub wrote: > >>On April 21, 2001 at 06:21:31, Duncan Stanley wrote: >> >>>Oliver Roese argues that in Computer Chess the doctrine of "End justifies the >>>Means" has been taken to its full, absurd and logical conclusion. All that >>>matters is the WIN - the "how" of the "win-result" becomes immaterial; all >>>morality, all fairness, all human-ness is removed from the "means". >> >>this is against laskers doctrine. lasker especially was interested in the >>fairness. > >there is nothing unfair being suggested, what is the problem? It's an imbalance problem for computer-human chess. > >>>Can he really be right? How far does his cultural imperative extend? Is all >>>discourse a mere manoeuvre towards a win position by any means? Isn't his view >>>too intolerable to accept? > >this is not about teaching children morals, it is about how to WIN a chess >match. the match begins long before the players sit down at the chessboard. it >begins with the negotiation of rules and conditions of play. look at the >history of bobby fischer for how this works. defeating a human opponant >MENTALLY before the match starts is good strategy, regardless of what sport you >play. Which can be most important, since the human's brain must function effectively during the match. But, the *programmer's* brain has no part in the actual game, thus psychological warfare in human-machine tournaments is unbalanced and therefore 'unfair'. as a member of a competitive team (software,hardware,operator) the >operator MUST do all he can to make sure that his team is not disadvantaged by >rules and conditions. to fail to attempt this is to serve poorly. "just go >along with what is proposed" is a bad start. Conventionally in computer chess, and according to the unwritten rules of conduct, the operator should have no other role than move entry. He is supposed to be a passive messenger. In practice operators manage to have a great effect on proceedings. You have to realise that the operator closely identifies with the program. Except that he usually didn't program it, and doesn't understand how it works. But he spends much time trying to make an influence in the only ways that he can, clock, openings. The more he feels he can 'influence' the more he identifies with the 'polymorphic system'. I think in some cases this desire to influence and the identification felt can lead to direct cheating in order to achieve the win. Worst case scenarios can occur when a program crashes or there is some input error. Opponent operators can form cabals that try to ensure the TD will make a decision bad for the enemy program. I've seen this happen. > >>as long as chess computer programs are stupid but fast, >>the HOW is not important and the WIN counts much. >> >>with cars it is the same. important is the WIN, the maximum speed, >>not the HOW it is driving. >> >>with girls it is the same. with money. >>it is called materialism. >>all that counts is the measuring of something. >>not the quality. > >i don't see the dualism. it is not a matter of quality VERSUS quantity. for a >competitive chess program, WINNING is the measure of its "worthiness"... that is >the only "morality" there is for a chess program... win, win, and win... we >must make no attempt to humanize the machine, and we must do nothing to >undermine its strength... rather we must embrace its goal of WINNING and help it >to win games. That is the ideology used. No doubt about it. > >>>read up on your sun tzu: win the battle before it starts, your opponant is >>>trying to. chess is a war game, leave the moralists outside the gate when the >>>question is how to crush your foe. this is a game, but it is a serious game. >>>WIN IT. >> >>no. lasker was 27 seven years a world champion, but he never was unfair, even >>against tarrasch, and he would have many many reasons to take revenge against >>him. but - fairness and moral has something to do with the SENSE you see in the >>game. if you see NO sense in chess other then to win, you will not work out any >>moral. >> >>but for me, chess and computerchess is more than winning. >>chess and life have rules. if you don't accept the rules, >>and be unfair, unsportmenlike, and breaking morals, you will >>win, but it leads to capitalism and not to quality. it leads >>to quantity. > >competitive chess is only about winning. it can be about nothing else. do not >confuse competitive chess with chess for fun or chess for the beauty of chess. I hope I've made the point that the introduction of human-machine events adds a new dimension to this win at all costs scenario. The human cannot freak out a machine. The machine team can seriously disturb the human.
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.