Peter_de_Blanc comments on Bayesian Utility: Representing Preference by Probability Measures