(19-07-2026, 06:40 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.oiin or an a eor or ch aiin a chor oiin qo a oke ok y chor ch o dan dan or a n oiin qo qo oke dan or or a iin ch ch aiin y chor eor eor o oke daiin a iin ch ch oiin daiin a n ch o eor oiin dan a a a n ch dan or aiin ain oke ok y ch aiin a n chol y dan or aiin a chor ch ch eor oiin oke oke daiin or ain y eor a in y chor ch eor eor oke a iin in y ch o oke or eor aiin y oke ok y ch eor oiin s or or
There are no problems with the algorithm. Is there a problem with me.
When I tried to encrypt your text manually, the result matched this one. The algorithm works flawlessly.
Perhaps if I hadn't rushed and tried to do it in 5 minutes, I wouldn't have made such a mistake...
In any case, @oshfdk, sorry. As you can see, I tend to make mistakes.
However, I don't consider this a reason to criticize the theory of numerical encryption in general, as it aims to try to reproduce the VMS text rather than to find out how quickly I can solve anagrams.
(20-07-2026, 07:07 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.However, I don't consider this a reason to criticize the theory of numerical encryption in general, as it aims to try to reproduce the VMS text rather than to find out how quickly I can solve anagrams.
I'm not sure one needs a reason to criticize various aspects of different theories. For example, it's clear from this experiment that the numerical encryption is not very convenient. And even the author of the solution cannot manually decode it, so I think it's safe to say it may not be a very practical cipher. It's also not clear whether this cipher can reproduce the statistics of the Voynich MS. Even with a large number of nulls one has to propose a way to place these nulls that would produce a result similar to the Voynich MS. I don't think it's enough to say "in principle nulls can reproduce any distribution, so this explains all the statistics". I think it has to be shown that some specific way of placing nulls, randomly or pseudorandomly or according to some simple rules, will produce a result similar to the Voynich MS. In general, there should be an explicit algorithm, or better a piece of code, that receives some plaintext and produces a good likeness of Voynichese to prove that a cipher actually a good candidate.
(20-07-2026, 07:24 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view. (20-07-2026, 07:07 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.However, I don't consider this a reason to criticize the theory of numerical encryption in general, as it aims to try to reproduce the VMS text rather than to find out how quickly I can solve anagrams.
I'm not sure one needs a reason to criticize various aspects of different theories. For example, it's clear from this experiment that the numerical encryption is not very convenient. And even the author of the solution cannot manually decode it, so I think it's safe to say it may not be a very practical cipher. It's also not clear whether this cipher can reproduce the statistics of the Voynich MS. Even with a large number of nulls one has to propose a way to place these nulls that would produce a result similar to the Voynich MS. I don't think it's enough to say "in principle nulls can reproduce any distribution, so this explains all the statistics". I think it has to be shown that some specific way of placing nulls, randomly or pseudorandomly or according to some simple rules, will produce a result similar to the Voynich MS. In general, there should be an explicit algorithm, or better a piece of code, that receives some plaintext and produces a good likeness of Voynichese to prove that a cipher actually a good candidate.
I totally agree with you!
It is not difficult to create rules for nulls. For example,
ch and
yk/
yt will most often be placed before the digrams
ol/or and
eor/eol/ey.
Final units can be converted to
y or
m.
The final
e in the words can be followed by a
y.
There is a possibility that Currier A has something similar.
The sequence of eos an on on f2r looks like the raw results of my encryption. Especially an and on, which appeared regularly in the texts.
Et littĕris multōrum et sermōne omnium perfertur ad me incredibĭlem tuam virtūtem et fortitudĭnem esse teque nec anĭmi nec corpŏris laborĭbus defatigāri. Me misĕrum! Te, ista virtūte, fide, probitāte, humanitāte2 in tantas aerumnas propte
ch a /// ch ch eor oiin dan or or a /// chor chor ch oiin o daiin daiin dan /// ch a /// eor oiin o oke daiin a a /// chor o oke daiin daiin or /// chor ch oiin oiin oiin qo an a a /// ok y /// daiin a /// oiin oke daiin dan or or or a a ok iin in /// chor ch daiin y /// chor chor ch ch oiin daiin or a /// ch a /// chor ch ch oiin o oke daiin or or an a ok /// eor eor a a /// chor ch cph a a /// oke a iin /// oke daiin or or y /// oke a iin /// eor oiin oiin q o or iin /// chor eor oiin o dan or in in y /// ch oiin or or ain an a ok y y /// daiin a /// chor eor oiin daiin daiin or a /// ch a /// ch eor or y /// chor chor ch ch oiin or a /// or an a ok /// ch ch oiin q or a in y /// chor ch ch oke daiin or aiin a n y /// oke or /// ch ch eor oke y y /// chor eor oiin oke daiin a n y /// ch oiin oiin qo q a /// daiin a /// eor eor oke or or or a ok iin
In the case of Latin, there are a lot of repetitions.
We will add nulls (unfortunately, they cannot be highlighted in this font, so you will identify them in special cases by comparing them with the original text):
cha /// char cheor oiin dan chor chor a /// chor chor choiin o daiin daiin ydan / cha / cheor choiin cho okey daiin g* / chor qokeor daiin daiin / chor choiin oiin oiin qo s* / oky / daiin am / oiin okedaiin dan chor oiin* okaiin in / chor chy daiin / chor chor choiin chodaiin otey* / cha / chor chy choiin okeodaiin chor chor chan okar / cheor cheor g* / chor chcphy g* / okeaiin / okedaiin or or y / okeaiin / cheor choiin oiin qo oraiin / chor cheor choiin odan dor* / choiin or or dain dan oky chay / daiin a / chor cheor oiin daiin daiin or dy / cha / cheor or y / chor chor cha chor oiin / okaor dan / chy choiin qo oraiin / chor ytchy chdaiin okeor aiin dy n / okeor / chy cheor okey yky / chor cheor oiin okedaiin ain / choiin oiin qoky qoky am / daiin ar / cheor yteor okeor or or okaiin
*The last e can be turned into a g;
an + a + a = s (VI + V + V = XVI);
or + a = ote (IX + V = XIV); y is a null
in + in + y = d (II + II + I = V);
As @oshfdk says, deciphering this text is a real headache. In fact, I agree with him. But I believe that if there are hints in the text that immediately reveal certain frequent positions (for example, to quickly recognize addition, etc.), then this process will be faster.
Some of the complications and additions of zeros are repeated in the text, so it is possible that the author has compiled a list of clues to the cipher that allows you to quickly arrive at the numbers themselves.
Yes, decryption is complicated, but it can be simplified...
(22-07-2026, 03:35 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.Yes, decryption is complicated, but it can be simplified...
To check if it's plausible would be nice to have a demonstration how this encryption scheme recreats statistical properties of Voynichese. Is it possible to create a script that would convert at least 10000 symbol of plaintext to Voynichese using this scheme?
(22-07-2026, 03:56 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view. (22-07-2026, 03:35 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.Yes, decryption is complicated, but it can be simplified...
To check if it's plausible would be nice to have a demonstration how this encryption scheme recreats statistical properties of Voynichese. Is it possible to create a script that would convert at least 10000 symbol of plaintext to Voynichese using this scheme?
I doubt that it can be done now. At a minimum, you need to figure out how the text would enter nulls.
I'm also in a hurry to take back my words at the moment - I started to doubt the statistics myself
It doesn't have all the Voynichese symbols yet.
But otherwise, he parodies Currier A quite well. There is even something similar to LAAFU. I'm sure of it
(22-07-2026, 03:56 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.To check if it's plausible would be nice to have a demonstration how this encryption scheme recreats statistical properties of Voynichese
Also - when I saying "it can be simplified", I assume that it is possible to create a list of so-called "difficult words" that make decryption difficult. For example, a list of all possible variations of what
s can be (
an +
a +
a or otherwise), how to recognize nulls, etc.
I don't have these "nulls-rules" yet. For now, they just complement the single
or and
eor.
But okay, it's quite easy to identify garbage
ch's (as you can see in the example, they are out of order and form impossible combinations, such as chy), but in other cases of addition, you will need to introduce rules...
(22-07-2026, 04:31 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.But otherwise, he parodies Currier A quite well.
Here's the same problem I saw in the Bavarian (space vowels) thread. "It parodies Currier A quite well" is your subjective evaluation that may or may not be shared by others. For example, I don't see any particular similarity to Currier A looking at your examples, they look quite odd to me. The only way to settle this, I think, is to generate a large enough corpus in a reproducible automated manner.
(22-07-2026, 07:25 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.For example, I don't see any particular similarity to Currier A looking at your examples, they look quite odd to me.
What are the differences? How different is the text with nulls?
One similarity is that words from Currier A easily pop up, especially oiin, which appears in various forms (shoiin, soiin, etc.) in Botany. Or not?
I also asked @JoeyB to try to improve the code (I don't have the opportunity to work with Claude). I'm counting on him
However, you shouldn't expect any supernatural results - this cipher is merely illustrative, showing similarities but not exact matches. This is just to warn criticism...
In any case, you can use this cipher to study Voynichese, but not to decrypt it.
But there are some matches, for example, in Currier A, I often see combinations of
iin and
y. Perhaps the author was trying to "equalize" all the numbers in the string by using zeros to avoid giving away any clues. Otherwise, it would be easy to see the boundaries between the words.
So first we need to define the "similarity" criterion...