Re: computer-go: Perfect play

[Mike Gherrity <gherrity@xxxxxxxx>]

>   As I was thinking again about the issue of perfect play some
>   interesting thoughts occurred to me. I believe that there is a
>   fundamental difference between chess and Go that hasn't be thought
>   about when trying to translate stones to elo-ratings.
> 
>   Here is the difference:
>   In chess you either lose a game or win it(or draw it). In Go you
>   also win or lose or draw, BUT, if you lose, the question is, by how
>   much did you lose, and according to that amount you can adjust
>   handicap stones to try to make the score equal in the next game.

This is also an issue for the game of RoShamBo (paper-rock-scissors).

	http://www.cs.ualberta.ca/~darse/rsbpc.html

In fact, it is clear in these tournaments that the Glicko (son of ELO)
rating of a player depends on the ratings of the other players.  So I'm
not sure talking about an absolute rating for perfect play even makes sense.

		mike