Re: Anderson & Kuhn's "Improved DFA" paper