How do you know when an encrypted book is solved?

[doranchak]

If there are a multitude of possible solutions, then it's not reproducible.

Among the many wrong codebreaking claims I've encountered, there are some whose steps are comprehensible and some that aren't.

In the former case, at least we can repeat the steps that lead to the proposed solutions and then demonstrate why they lead to other plausible ones.

In the latter case, they are "not even wrong", as Wolfgang Pauli liked to say.

[-JKP-]

The basic initial steps might be reproducible, but if the result is different for each person because of increasing degrees of freedom as you walk through the steps, I don't think we can call the solution reproducible.

Depending on the nature of the proposed solution, there might be some wiggle room that produces a small amount of variability that might be acceptable. It depends on what method is needed to produce the end result and the complexity of the underlying message (e.g., polyglot would be harder to decipher than a single language; a straight narrative easier than poetry). If a majority of it rings true, I think most people would be satisfied with that.

[nickpelling]

Comprehensible is all too rare: the norm is reprehensible, sadly.