A simple deterministic verbose cipher for a short English plaintext made with 3 tokens per ciphertext word, no nulls, is already hard to crack. See the thread about You are not allowed to view links. Register or Login to view.. I think I got the tokenization right (mostly, a few errors in low frequency tokens is not really a problem) but making sure that there is no other possibility is not trivial. Then there are 3 alphabets to map to the tokens. That's a lot for such a short ciphertext: frequencies are not helpful except for the few most frequent tokens. There are possible cribs like "the" and "that" but they don't help much.
For 34 symbols (letters + space) in the plaintext and also a very short ciphertext (if there are different settings to find for every paragraph) it would be even harder.
For 34 symbols (letters + space) in the plaintext and also a very short ciphertext (if there are different settings to find for every paragraph) it would be even harder.