Cassandra wrote:
However, if you wanted to define "territory", but not "seki-group", you would have to incorporate something (that jann does not want to call "exception") in your set of conditions for defining "territory".
I don't see why. What I wrote above seems to work without this:
jann wrote:
It seems more robust to define territory first (transformable to pass-alive). Then dead stones are only those in territory, and seki is unnecessary.
BTW, I think the only reason to exclude territory in sekis is implicit in this approach. Otherwise explicitly excluding them seems pointless and unnecessary complication (burden for no gain). So defining seki in either way seems a bad idea.