Comments on ``finite difference approximation errors in acoustic intensity measurements''