Folks, John has some incomplete thoughts about the number of k-means clusters that are justified. Come along tomorrow at 11:00 to attack him viciously for his lack of completeness. David