> In the light of the present discussion it is interesting to > note that this too presents a "algorithm-generated consensus > classification" Certainly not of the sort that I was describing. Aloha, Rich