computer-go: Winning strategy speculation and loosing games.

[Yannick Delbecque <yannick_delbecque@xxxxxxxxxxx>]

Vlad Dumitrescu wrote:
> 
> Does anyone know whether there are any speculations/theorems/thoughts about
> wheter there is a winning strategy for any of the players in Go? I know of
> some games where partial proofs are made...
> 

I know people that have proved that the center point is a winning
strategy for the 5x5 case, and worked a lot to prove the same thing for
the 7x7 case. In general (and 9x9 in particular) my bet would be that
the center is the winning starting point. But this intuition rest, exept
for my own go experience,  on the many exemples i've seen where the
"center of symmetry" is the winning play in a game, and in all cases the
rest of the winning sequence involved some sort of "mirror play" which
does not work well in go because of kos. 

Some recent comments made me wonder if anyone has tried to make "worst"
go program. The task is not so obvious...It could probably be usefull in
some way to help make good go programs. Maybe someone even tried to make
a tournement of such bad go programs? 

Yannick Delbecque