User blog comment:LittlePeng9/Random Turing machines/@comment-5529393-20130406111205

Very nice! I was thinking of using Goodstein sequences to get to epsilon_0, I hadn't thought of Beklemishev's worms. It's nice to see it done in under 100 states. Perhaps I'll give a shot at programming it with two symbols.