jmh comments on Problems with p-values