I begin to study ko fights.
When I became shodan I realized, to my surprise and dismay, that I knew next to nothing about go. Oh, I had a pretty good intuition about the game
, but I was playing by the seat of my pants.
One thing I decided to research was ko fights. Kos are not independent plays, but involve plays elsewhere, in what I call the
ko ensemble. One of the first things that I studied was the relationship between kos and the
environment (other plays on the board besides ko threats), although I did not use the term at that time. This led me to the idea of
komonster -- again, a term I did not use until much later, after learning about Professor Berlekamp's idea of
komaster. A komonster is not only able to win a ko, but to reduce the size of plays that the
koloser is able to get in exchange for the ko (the
ko exchange). Nowadays we say that the komonster is able to reduce the
temperature of the environment (or
ambient temperature) before winning the ko.
The "half point ko" at the end of the game illustrates komonster effects, even though it is usually played at a lower ambient temperature than the size of the ko to begin with. Now I knew that, in theory, the size of each play in such a ko is 1/3 point (by territory scoring), not 1/2 point, but on the other hand, if you don't know who will win the ko fight, it's a 50-50 chance, and a half point makes sense. But my study of the komonster gave me a different understanding of the half point ko.
Suppose that we do know who will win the final ko. Then the koloser gets zero in the ko exchange. That means that taking the ko is enough for the komaster to win it. So instead of there being three moves between winning and losing the ko, there would be only two moves, in practice. Each move would gain 1/2 point, making for a half point ko.
Later on I learned about Chinese (area) scoring, in which each stone gains one point, so that each play in the final ko gains 4/3 points, not 1/3 point. Each dame gains 1 point, as well. But suppose that the komonster can reduce the ambient temperature from 1 point (dame) to 0 (no dame). Then the value of each move in the ko would become 4/2 = 2 points. That is the basis for my little problem.
I have to go now. More about that problem later.
