Distinguishing Characters

[nablator]

I'm not sure if this is the best way of putting this given the current state of our knowledge.

Sure. My opinion only, not a certainty.

It suggests that the Voynich cipher (if we assume it is one) is a conscious attempt to make substitution ciphers more difficult to crack.

My system is not better than a substitution cipher with homophones and nulls for any purpose other than having fun, possibly in the context of a game. Therefore of no practical value for "serious" cryptography. If it is more secure, it is mostly because we don't know the rules. Any system with unknown rules is nearly impossible to crack. I tried several variants of the antipolybius or zigzag path cipher but I believe that there is some added complexity in the way apparent ligatures and gallows should be interpreted, that are difficult to guess.

Even when the system was known, all ciphers and codes were as secure (i.e. not very) as the key or code book that had to be transmitted separately and could be intercepted.

An overly complicated system would be very cumbersome to implement and, given the mass of text, also very error-prone.

Yes. My design constraints for trying to replicate Voynichese were: nothing that requires more than a quick look at a table to encipher/decipher a letter or a few letters at a time. All ciphers/codes other than monoalphabetic substitution require more time-consuming or boring operations. In this sense, my system fits nicely in the gap.