Re: computer-go: Most simple Go rules

[Don Dailey <drd@xxxxxxxxxxxxxxxxxxxxx>]

   >    How many different rule-sets do you think the programmers should
   >    implement?
   >
   >For tournaments they  should implement a  single  agreed upon ruleset.
   >But for commercial use programmers  should implement every ruleset  in
   >common use and provide toggling of elements of these sets.

   Do you really think you will get the Ing foundation or the Japanese go 
   association
   to change the rules they use for computer go?

Actually,  yes eventually.  If  that's what the programmers wanted and
they made  it  known.  It wouldn't happen  overnight  but  it probably
would happen sooner or later.  It seems to  me that people may already
be moving in that general direction.

You have more influence than most of us because you are well known and
respected as an author  of a strong  program.  That means that  if you
resist and tell everyone it's a stupid idea or  won't work, you can go
a long way towards discouraging the idea.  If you did the opposite, it
would  get  more consideration.  But  in  the end what  will matter is
whether most people really want this, not one or two.

Anyway, I'm only saying  that it wouldn't hurt  to be more open to the
idea.  I know implementing  it  can't possibly  be a major  concern to
you.  Or maybe  it could be, I  don't know what  dependencies you have
between  scoring, move generation  and when to  stop  the game, but it
would still add value to your program just having another option.

If it's an issue of how  hard it is  to implement to existing programs
or the extra work involved, I don't  think this should  be any kind of
deciding factor.  Computer Go  may have a  long future that makes this
consideration pretty much nil.

But I'll put this question to the group: What type of existing program
would Tromp/Taylor be really difficult  to implement?  I think  almost
all  of us agree  that for a  new program, Tromp/Taylor is the easiest
set to implement.


Don