User blog comment:LittlePeng9/Random Turing machines/@comment-10429372-20140808191119/@comment-1605058-20140808193926

Neat! That seems right, provided there is no ridiculous disproportion between f_e_0 and hydra function :)