Author: Erik Bergren
Date: 10:58:12 11/22/03
The Kasparov vs. Computer, and other such matches have a particular quality to keep in mind when interpreting them: >>Concern for the reliabilly of data is logical. Among the penalties >>for not doing that is: wasted time studing unreal data. (Terry McCracken said): > >Logic dictates not to draw conclusions from dubious data. >There is no indication, that the data I have is doubtful, but every indication >the internalization of conspirators are! Qualifying a "chess game" as non "doubtful data" or non "dubious data", is not specific enough for critical use. In the present case for example: Kasparov was able to practice play over and over against the computer (or a very similar version) before playing the games that counted. Kasparov learns from such games, but the computer did not. That is a higher degree of "aquaintance with the opponent" than is usual. Surely the games have a quality to them do to that. For example, they could contain assaults on weeknesses that Kasparov found in that computer and software. Thus: Such data(games) would(may) then not be "rigorous enough" (or "less rigorous") for forming general plans for winning at chess against other opponents. (Unless you interpret it carefully, conscienciously).
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.