Comment on the interpretation of inductive probabilities

A recent defense of Jaynes' information-theoretical approach to statistical mechanics is rejected, and an earlier critique of this approach is extended.