John G. Halstead comments on Deference for Bayesians