The Voynich Ninja
Positional Mimic Cipher (PM-Cipher) - Printable Version

+- The Voynich Ninja (https://www.voynich.ninja)
+-- Forum: Voynich Research (https://www.voynich.ninja/forum-27.html)
+--- Forum: Analysis of the text (https://www.voynich.ninja/forum-41.html)
+--- Thread: Positional Mimic Cipher (PM-Cipher) (/thread-4921.html)

Pages: 1 2 3 4

RE: Positional Mimic Cipher (PM-Cipher) - quimqu - 12-09-2025

(11-09-2025, 09:16 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.
(11-09-2025, 08:27 PM)quimqu Wrote: You are not allowed to view links. Register or Login to view.These slight gliph differences could be perfectly the residuals marks, so, no need of a suplementary booklet with the residuals.
Yes. The main issue then is that the table is not really needed for this method to work, and we are back in the realm of expanded alphabet/microwriting solutions: starting with Newbold's, and including at least half a dozen proposals from some active participants on this forum.

Just to clarify: I am not telling that there are hidden symbols for the residuals, like Newbold. I am thinking of things like the slight difference between k and t, for example, could mean the same character but different residual.


RE: Positional Mimic Cipher (PM-Cipher) - oshfdk - 12-09-2025

(12-09-2025, 05:10 PM)quimqu Wrote: You are not allowed to view links. Register or Login to view.
Just to clarify: I am not telling that there are hidden symbols for the residuals, like Newbold. I am thinking of things like the slight difference between k and t, for example, could mean the same character but different residual.

I can't see any difference between "same character and different residual" and "two different characters". To me these two look equivalent, just a naming convention. Could you clarify a bit more?

RE: Positional Mimic Cipher (PM-Cipher) - quimqu - 12-09-2025

(12-09-2025, 05:17 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.I can't see any difference between "same character and different residual" and "two different characters". To me these two look equivalent, just a naming convention. Could you clarify a bit more?

Yes, in the EVA transliteration k and t are treated as different characters. But imagine that they both are k, the same character, and that the small loop.at the left is giving us the residual. So k would be k0 and t would be k1. Of course this would reduce the character set. And please, don't take it as I think it is right. I am just saying that the PM_cipher needs the residuals as the same glyph can cipher different original characters. I am thinking that maybe the residuals could be in a different booklet or "hidden" in the glyphs.

And I am not saying that this is the cipher. Just making theories...

RE: Positional Mimic Cipher (PM-Cipher) - oshfdk - 12-09-2025

(12-09-2025, 05:30 PM)quimqu Wrote: You are not allowed to view links. Register or Login to view.Yes, in the EVA transliteration k and t are treated as different characters. But imagine that they both are k, the same character, and that the small loop.at the left is giving us the residual. So k would be k0 and t would be k1. Of course this would reduce the character set. And please, don't take it as I think it is right. I am just saying that the PM_cipher needs the residuals as the same glyph can cipher different original characters. I am thinking that maybe the residuals could be in a different booklet or "hidden" in the glyphs.

I thought we were discussing your specific implementation, k and t are both used in the table as two distinct characters. If there is a need to add residuals on top of existing character set, it should be via more subtle changes to the shape, like in two different l's from You are not allowed to view links. Register or Login to view. below.

   

RE: Positional Mimic Cipher (PM-Cipher) - quimqu - 12-09-2025

(12-09-2025, 05:38 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.I thought we were discussing your specific implementation, k and t are both used in the table as two distinct characters. If there is a need to add residuals on top of existing character set, it should be via more subtle changes to the shape, like in two different l's from You are not allowed to view links. Register or Login to view. below.
Yes, in my implementarion I use k and t as different characters. I am not saying that they should fuse in one or that they mark the residuals. I am just saying that the residuals could be written in a lost booklet or hidden in the same MS. Your example is a perfect one. This could mark residual 0 and residual 1... I was just meaning that slight changes in the glyphs could hide the residuals. But this is not implemented yet in my solution, neither explored.

There are also the so called CAPS that could be used as residuals.. who knows?... I really don't know. 

But I'd prefer to go step by step and see if the PM_cipher can be a good match in terms of entropy, JSD, bigram correltaion, etc.. and then imagine different solutions (even if I prefer to stay objective).

RE: Positional Mimic Cipher (PM-Cipher) - Jorge_Stolfi - 12-09-2025

(12-09-2025, 01:14 PM)quimqu Wrote: You are not allowed to view links. Register or Login to view.I’m not comparing raw Vietnamese (or natural languages) against Voynich with JSD. All my JSD results for the cipher are EVA vs EVA.

Ops, sorry, I skipped the beginning of the thread.  

So you are varying the encoding table, attempting to get the same digraph statistics as the VMS?

It would be notable if you can do that.  However, I would question your assumption that the plaintext was in Latin and in the "standard" spelling system.  

The distribution of lexeme (word type) lengths seems to be incompatible with the source language being Latin or any Indo-European, Finno-Ugric, or Turkic language, if the encoding approximately preserves the length of words.  A Semitic language (Arabic, Hebrew, Aramaic,Ge'ez, Berber, etc) seems more compatible, since it should have root words limited to 3-4 consonants, but it stll may require a non-standard splitting of articles and affixes ("walaalddallin"  -> "wala al ddalin").

Moreover, in ciphers of the time it seems that it was common to encode digraphs like "ch" and doubled consonants like "ss" and "cc" as single letters, because otherwise they could be keys for cracking the encryption.

But do let's see how close one can get to the digraph distribution even apart from those objections.

All the best, --jorge

RE: Positional Mimic Cipher (PM-Cipher) - quimqu - 13-09-2025

(12-09-2025, 09:17 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.So you are varying the encoding table, attempting to get the same digraph statistics as the VMS?

That's it. I iterate to get a H2 at around 2.1 (as the Voynich) and the same digraph statistics. By now I get the best JSD with Romeo and Juliet (0.29) but English has quite a lot of long words.

(12-09-2025, 09:17 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Moreover, in ciphers of the time it seems that it was common to encode digraphs like "ch" and doubled consonants like "ss" and "cc" as single letters, because otherwise they could be keys for cracking the encryption.

This is a great idea to reduce the original word length. I worked with EVA digraphs (for example "qo" to cipher one single original letter) but I think it's wort to try to encode the most used digraphs of the text in a single EVA char. 

Thank you!

RE: Positional Mimic Cipher (PM-Cipher) - quimqu - 13-09-2025

Edited, I will update in a next post

RE: Positional Mimic Cipher (PM-Cipher) - Jorge_Stolfi - 13-09-2025

(12-09-2025, 01:14 PM)quimqu Wrote: You are not allowed to view links. Register or Login to view.One question, can I lowerize the capitalized chars, or they are really different (eg. O vs. o)?

I forgot i answer this one, sorry.  Yes, capitalization in Vietnamese is pretty much as in English. Proper names, sentence starts, titles, etc.

RE: Positional Mimic Cipher (PM-Cipher) - quimqu - 13-09-2025

(12-09-2025, 09:17 PM)Jorge_Stolfi Wrote: You are not allowed to view links. Register or Login to view.Moreover, in ciphers of the time it seems that it was common to encode digraphs like "ch" and doubled consonants like "ss" and "cc" as single letters, because otherwise they could be keys for cracking the encryption.

Well, working with digraphs is giving worse results in H2 and JSD. In fact, original double letters like "ss" will be hidden, as they should be ciphered with different EVA characters as they have different positions. I reached a JSD of 0,15 working with digraphs, but as there were repetitions digraph/simple chars, the deciphering became impossible. So I kept working with simple chars.

Optimising a bit my code, i still have the best results in English (Romeo and Juliet and Culpepper), with a JSD of 0,269, with a 28% of words generated contained in the Voynich EVA transliteration. You can see here the matrices Voynich / Ciphered Romeo and Juliet:

[Image: TXsjq7C.png]


I would be happy to hear any comment and learn from you.