Charlie Steiner comments on Protocol evaluations: good analogies vs control