Taking this opportunity, I’d like to share what, in my opinion, should be included in the characteristics of a VMS cipher:
1). The limitation of basic glyphs. After the author created a basic set of strokes such as o, a, in, k, t, p, etc., he noticed that these characters were not enough. To fill the empty positions, he created bigrams such as ok-ot, yk-yt, or, ar, che, etc. That is, a digram or a combination of characters can be considered as a single character.
2). Ambiguity of symbols. Perhaps the symbol p could well mean something, but at the same time it is used by the author to mark paragraphs. Similarly with the symbol s, which, in addition to the space/line break character, could mean something else.
3). Nulls. Yes, they come from the same alphabet. If they actually exist, I suspect they can be quite easily identified by certain characteristics (for example, repeated symbols or words may simply be placeholders).
4). The Voynichese vord can be represented as a number, but not as a word. That is, for example, if we get the equation chey = XXXI or daiin = XVIII, then we won’t be able to draw any clear conclusions from the resulting number. Simply because the alphabet encrypts not letters, but numbers. Accordingly, there may be no visible connection to the plaintext at all.
5). One number can be represented in different ways. In my cipher, the number VI could be represented both as an and as dy. Such ambiguity can quite complicate the deciphering process.
As I pondered the idea of a numerical cipher, I came to the following conclusions. I can’t yet be more specific about them. If it turns out that Voynichese is a book cipher by Stolfi, with no connection to Roman numerals, then these rules can be forgotten.
1). The limitation of basic glyphs. After the author created a basic set of strokes such as o, a, in, k, t, p, etc., he noticed that these characters were not enough. To fill the empty positions, he created bigrams such as ok-ot, yk-yt, or, ar, che, etc. That is, a digram or a combination of characters can be considered as a single character.
2). Ambiguity of symbols. Perhaps the symbol p could well mean something, but at the same time it is used by the author to mark paragraphs. Similarly with the symbol s, which, in addition to the space/line break character, could mean something else.
3). Nulls. Yes, they come from the same alphabet. If they actually exist, I suspect they can be quite easily identified by certain characteristics (for example, repeated symbols or words may simply be placeholders).
4). The Voynichese vord can be represented as a number, but not as a word. That is, for example, if we get the equation chey = XXXI or daiin = XVIII, then we won’t be able to draw any clear conclusions from the resulting number. Simply because the alphabet encrypts not letters, but numbers. Accordingly, there may be no visible connection to the plaintext at all.
5). One number can be represented in different ways. In my cipher, the number VI could be represented both as an and as dy. Such ambiguity can quite complicate the deciphering process.
As I pondered the idea of a numerical cipher, I came to the following conclusions. I can’t yet be more specific about them. If it turns out that Voynichese is a book cipher by Stolfi, with no connection to Roman numerals, then these rules can be forgotten.


