Author: Robert Hyatt
Date: 10:10:03 10/29/03
Go up one level in this thread
On October 29, 2003 at 11:19:46, Sune Fischer wrote: >On October 29, 2003 at 10:13:10, Robert Hyatt wrote: > >>I didn't bring it up, I just pointed out that there were _plenty_ of examples >>of contrived results to favor a company over a single instance of a chess >>program. > >I'm far from convinced there is problem nowadays. >There might be a slight conflict of interest, but if the book guy is pro enough >he should be able to walk the fine line. > >>I don't mind sharing ideas, or even making suggestions for the book. But to >>put someone on creating a book with a significant effort, then letting multiple >>programs use that book, seems unreasonable to me. As I said, a typical chess >>game lasts 50 moves. I have played _many_ games where the first 20-25 moves >>come from the book. That is a _major_ influence on the result, since the book >>supplies 1/2 of the total moves played in the game... >> >>That is _direct_ involvement in the final outcome, providing just as many >>moves as the actual chess engine. If three programs can share the same >>book, should we not allow the same program to enter three times with three >>different books? Same thing _exactly_. > >Think of the book as being delivered by sub-contract, just like the tyres are in >Formula 1. OK. But think of the engine as being delivered by sub-contract. I'm a chess-player, not a programmer, so I don't know how to write an engine. But I do know how to develop a great (and novel) opening book. Can I enter and compete? > >You have few manufactores that supplies the whole lot. >Kind of the same senario where a sub contractor can theoreticly have great >incluence on many teams. > >The alternative is that everyone starts producing their own set, which is a huge >project in itself and would draw too many resources from the "interesting" >parts. > >>I disagree. When I compete in anything, I want a level playing field. Whether >>it be chess, checkers, tennis or fishing. >> > >Yes so do I, but before we attempt such a thing we must try to define what is >fair and what is not. I don't mind a single person for an opponent. I don't mind a team as an opponent. What I do mind is sharing members between teams, which _could_ produce a different program for each team member. That maximizes that team's chance of producing a #1. > >>>Why not? >>>I definitely feel cheated, I would like to use the Kure book as well :) >> >>I rest my case. I would too... >> >>:) >> >>Chessbase distributes a crafty binary for their GUI. Shouldn't I therefore >>be able to use their book? :) > >I guess if you want to play in their GUI and all, connecting through UCI, walk >around in a big "I love Chessbase T-shirt" etc. Maybe you could? > >-S. I might "could". But I definitely _wouldn't_...
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.