Re: [Problems to Attack] Re: On Game Space Size

[Mousheng Xu <moushengxu@xxxxxxxxxxxxxxxx>]

At 10:21 PM 5/6/99 -0700, David Fotland wrote:
>A is connected to B or to C, but not both.  If A connects to B, then C is
>no longer
>connected to D, etc.

David,

	Yes. I have been thinking about this recently. When the connection of A to
B is actually established, the connection of A to C or the connection of C
to D might be affected. So you need to adjust your connection assumptions
after every move.
	There are two ways that I can think of to reduce the hassle: 
	1) for every possible connection between two groups, list all the critical
positions that may affect the connection. A connection assumption is
updated only  when one of the critical positions are affected. 
	2) use a larger pattern to include multiple groups. I guess you've already
thought about it.

	Professor Chen has a team of a couple of 5d (Chinese amateur ranking)
programmers. They should have a lot of research results on such issues. If
there were a magazine that could publish such results, or if there were
some money to encourage people to publish their results, with no doubt go
programming could be facilitated. 

	Thanks.

-- Mousheng