All the efforts look awful. There's no analysis whatsoever of the number of similar-degrees-of-freedom different-meaning decipherments that would seem equally

All the efforts look awful. There's no analysis whatsoever of the number of similar-degrees-of-freedom different-meaning decipherments that would seem equally credible, or likelihood of deciphering a random text "successfully" given that many degrees of freedom. Granted, what I'm complaining about isn't easy to compute. If I knew how to actually do it, I'd publish a paper.