Re: Advice on evaluation

[Dave Dyer <ddyer@xxxxxxxxxxx>]

 There's  a big gray area between "stably alive" and "stably dead",
and what occurs there always devolves to minutia.  Approximations
and quick calculations are never satisfactory, but they are usually
all that is feasable.  It's dangerous to set about making better
approximations unless they are a LOT better, because the way programs
and people lose games is to believe things which later turn out
to be untrue.  

 If your goal is a stronger program, I think your time will be better
spent trying to avoid the gray areas entirely, and on the other hand, 
making the gray areas smaller.  Except in very limited tactical situations, 
trying to cut too close to the edge of the cliff  is sure disaster.