Don Dailey wrote:
> If you play a
game of go, it almost always matters who's turn to move it is.
This is so also for positional superko.
If it's
my turn to move it's a different position than if it's your turn to
move, even if the stones are sitting in the same exact location.
Technical remark: the position is the same, the situation differs.
Positional superko seems illogical to me for that reason.
"Inconsistent with what a current situation describes" would be a more
appropriate description than illogical. So you perceive the situational
superko rule's kind of legality decision making to be like the
move-making player's kind of legality decision making. This is a
reasonable argument in favour of situational superko or natural
situational superko.
I don't buy the argument that it's elegant because it depends only what
one can see on the board because it requires what has already happened
too.
Ok, let us be more precise here: What is currently regarded (to be then
compared with something having happened earlier) can be seen visually
on the board. Every such earlier position could also be seen visually
on the board at its time. The comparison to earlier positions is then
not visual, not even under positional superko.
Under situational superko, the entire situation cannot be seen visually
because its who-has-the-turn component cannot be seen.
There is no superiority here.
The superiority is there for those that can memorize positions more
easily than situations. Since go players study positions, patterns,
diagrams in books, etc. much more than situations, by far the
most can memorize positions more easily than situations.
If it's not logical it's not elegant.
I think you mean: not consistent, so not elegant.
> not elegant if it's contrived and arbitrary.
Inhowfar contrived and arbitrary?
have heard
arguments that NO superko rule quite does the job, they all apparently
have their ugly corner cases.
If one takes a superko rule first and then accepts its strategic effect
on positions, then there is no need to perceive anything as ugly.
If one starts with 100 specific positions, which one calls ugly for the
sake of it, of 0.012 * 3^361 possible positions and then takes a superko
rule and looks whether some of these ugly positions can occur, then one
will notice that yes. For every superko rule.
Is this what you mean by ugly corner cases? It does not require ko
rules. E.g., hane-sekis are also pretty ugly. To avoid such, one needs
at least a "Passing is prohibited if an opposing string is on atari."
rule (which must be made consistent with the used ko rule(s), e.g., by
overriding them).
--
robert jasiek
_______________________________________________
computer-go mailing list
computer-go@computer-go.org
http://www.computer-go.org/mailman/listinfo/computer-go/
