Author: Thorsten Czub
Date: 17:21:16 04/14/98
Go up one level in this thread
>Wait a little, Thorsten, before shooting at me so easy. After all, I am >supposed to be the reductionistic guy...:-) I share with you the >feeling that laws cannot tell us everything about the world, but here we >are not talking of a wide, undefinite and infinite entity like universe, >the dynamics of air or anything like that, we are talking of a closed >problem with simple rules. There is no SIMPLE problem. We don't know why planes fly. There is a prejudice that tries to tell the people WHY planes fly, but nobody really knows. If the prejudice that people have would be true, planes could not fly on the back-side, they would instantly fall to earth. Since this is not the case ... Sorry. Chess. "simple rules", "simple problem". No infinity (as if this makes it easier. Finite or infinite , this does not make anything easier). > We are talking of chess, not of the 41ยบ >simphony by Mozart. To use the artgument against reductionismo as you do >is to be a reductionist of the worst kind. Even reductionism must be >understood in a wide, not reduictionistic context. There are things that >can be reduced and there other that not. There are problems that can be >solved aND other that not. Ches is limited in essence even if the number >engaged are great. No doubt that chess is limited. >I say more; even is possible that chess cannot be >solved on the ground of counting, but less because of a sustantial >reason that for a problem of speed of calculation. If you have few >pieces on the board is prefectly solved by counting, no doubt about. > as ending databases >shows clearly. When a program uses that tables, he does not use strategy >at all and even counting, as much as all is previously caklculated and >counted. I agree here. Indeed. >Finally, Thosrten, it is worth of consideration the fact that in >practiical terms chess does not need to be solved in the theoretical >sense of the word, from first move to the last possible move; it is >enough to get a mate or a sustrantial advantage in te contexct of a >position and that can be done throught calculation and in fact is done >that way most of the time. There are many ways to open a door. You can e.g. use a key. That has a good statistic. Many people use keys to open door. If you have MORE than one key, you need to find out which one is the right key. On the other hand, you can kick the door open. Or fire with a gun into the lock. In fact using the right key is enough to open doors. You are so right my dear friend. I cannot deny your statements. >All this meanms I rejecta strateghy? Not at all; general advives as >those stragegy gives arws neccesary in the intermediate zone between >short tactical horizon and the fully theretical extension of all the >moves that are possible. In other words, beyond 10 or 12 moves you need >some general signs and that are provided by strategy. > >Fernando The mistake is that if you have x programs calculating in the range from 1-30 and they come to a result y, to believe that this Y is REAL and the KEY for the door. All you have proven is that the programs have SIMILAR strategies and will therefore open the door with the same key Y. This does not imply that there are no other ways. Or that y is more succesful than z. Or or or or or.
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.