Paul Crowley comments on How to come up with verbal probabilities