@oshfdk, you asked me to use an algorithm to encrypt some large text. I asked Gemini (the author of the code) and tried to apply large passages to it myself. It works, you can use it
I tried to encrypt Shakespeare's Sonnet 18:
oke y /// oiin o oke daiin an a iin y /// ch a /// daiin or or y /// chor ch oke dan or ain an a /// or a ok /// ch y /// chor ch /// eor oiin o or a ok iin /// chor eor qo dan /// eor aiin a in y /// ch oiin qo daiin or a a /// chor chor ch eor oke a /// eor oiin q dan dan or a iin /// daiin or or y /// chor eor a ok iin /// chor ch ch cPh or y /// chor chor eor oiin cph a a a in /// eor eor o dan a /// ch eor eor a n y /// ch oiin oiin or or y /// ch s or iin y
ch oiin o oke or a /// o oke daiin or or /// dan or a iin y /// chor ch oiin oke daiin or a ok /// chor chor eor o dan iin /// ch daiin or iin y /// chor chor chor oiin cPh daiin a n /// chor chor chor ch eor dan /// eor qo a a n /// chor ch oiin oiin che dan a ok iin y /// chor eor cph o oke daiin or a /// eor eor s or a a iin /// y /// eor s or a a iin /// chor eor y /// ch or ok iin y /// chor eor iin y /// chor ch y /// ch o daiin dan dan a n /// ch eor oiin q daiin or a /// chor ch oiin qo a ok y y y
ch y /// o oke oke /// ch eor oiin daiin a a iin iin y /// chor ch y /// ch oiin oke a a n y /// ch eor eor a n y /// oke a iin /// chor eor a ok iin /// ch oiin qo a a ok /// chor cPh o ok /// chor chor ch daiin /// chor oiin oiin qo q daiin or /// ch eor a /// oke a iin /// eor oiin o daiin /// ch a /// ch ch or or iin in y y /// oke or /// chor oiin daiin in y /// chor eor y /// chor oiin daiin dan a in y y /// chor daiin iin /// chor ch /// oke or /// chor eor oiin oke daiin or or iin in y /// ch eor oiin oke or a a n /// eor eor oiin a a iin iin
chor daiin ok /// eor o oke daiin or aiin a /// ch eor oiin qo oke or y /// chor ch y /// chor o dan or iin /// oiin oiin oke a a a iin /// chor chor chor ch ch eor eor q oke or a /// o aiin iin /// и /// chor ch daiin or y /// ch or or in /// ch o oke or ok in y
The peculiarity of the cipher is that when working with English text, it becomes very monotonous in terms of "words" (there are a lot of units like eor or, etc. Nulls are powerless in this case).
In fact, this text can be made "usable" without much effort, just by combining some words, and it will already be quite beautiful, for example.
q oke or).
What exactly is difficult about this cipher? Recognizing some letters and deciphering anagrams. Speaking of complexity, I think we need to consider that our decrypter knows all the rules of the cipher. His only obstacle is anagrams.
(30-07-2026, 07:37 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.@oshfdk, you asked me to use an algorithm to encrypt some large text. I asked Gemini (the author of the code) and tried to apply large passages to it myself. It works, you can use it
I tried to encrypt Shakespeare's Sonnet 18:
I'm not sure I understand what this was supposed to demonstrate. This certainly doesn't look like Voynichese with repeats like
a a iin iin and similar. If this was to run statistical tests, then a large text of at least 10000 symbols is needed, so this one is still short. Also to avoid any misunderstanding, I'm not going to be the one running the statistical tests. If you provide some generated text and show some statistics produced by one of the tool that people from the forum created, and the statistics look interesting, I think someone will be willing to reproduce them.
(30-07-2026, 07:37 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.What exactly is difficult about this cipher? Recognizing some letters and deciphering anagrams. Speaking of complexity, I think we need to consider that our decrypter knows all the rules of the cipher. His only obstacle is anagrams.
It's good you confirm there is an obstacle.
(30-07-2026, 08:40 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.This certainly doesn't look like Voynichese with repeats like a a iin iin and similar. If this was to run statistical tests,
You don't understand the cipher.
His task is not to reproduce the text itself, but rather to reconstruct the supposed encryption algorithm. In other words, to let you know what Voynichese means to me.
For statistical texts, I need to work with nulls. I'm trying to achieve this with Gemini.
(30-07-2026, 08:40 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.It's good you confirm there is an obstacle.
By the way, Gemini himself had a negative opinion about this cipher, saying that it would take too long to decrypt a large text if it was done manually.
One way or another, this obstacle exists.
As for the "dissimilarity," I can say that if I make a few minor changes to this text, it will be quite similar to VMS.
Two lines with You are not allowed to view links.
Register or
Login to view. that are similar to the previous encryption results:
dShor chol Shol cThol Shtchy chaiin ?
choraiin (!)
cham
otchol chol chy chaiin qotaiin daiin Shaiin
Examples of my text:
chol cTheol oiin qoteor aiin aiin chey
chor Sheor qokor qotain dy
This is, by the way, the very first attempt...
(!) - it looks suspiciously like ch or aiin
(30-07-2026, 08:49 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.You don't understand the cipher.
His task is not to reproduce the text itself, but rather to reconstruct the supposed encryption algorithm. In other words, to let you know what Voynichese means to me.
For statistical texts, I need to work with nulls. I'm trying to achieve this with Gemini.
It is true that I don't understand what you are trying to achieve. As far as I understand, you think this cipher may be similar to the cipher used for the Voynich manuscript, but what makes you think so? I mean if it produced some close resemblance to Voynichese, this would be an argument in its favour. But it doesn't in its present state.
(30-07-2026, 09:01 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view. (30-07-2026, 08:49 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.You don't understand the cipher.
His task is not to reproduce the text itself, but rather to reconstruct the supposed encryption algorithm. In other words, to let you know what Voynichese means to me.
For statistical texts, I need to work with nulls. I'm trying to achieve this with Gemini.
It is true that I don't understand what you are trying to achieve. As far as I understand, you think this cipher may be similar to the cipher used for the Voynich manuscript, but what makes you think so? I mean if it produced some close resemblance to Voynichese, this would be an argument in its favour. But it doesn't in its present state.
As I said, a similar result can be achieved.
My cipher is, you could say, a very lightweight version of VMS. I assume it works the same way.
I don't understand the direction of your criticism either. Are you dissatisfied with the theory in general or with this particular cipher?
And why do you think my goal is to reproduce the text? I'm showing you how to create such a text.
(30-07-2026, 09:25 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.As I said, a similar result can be achieved.
Until it's reliably achieved, this remains to be seen.
(30-07-2026, 09:25 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.I don't understand the direction of your criticism either. Are you dissatisfied with the theory in general or with this particular cipher?
And why do you think my goal is to reproduce the text? I'm showing you how to create such a text.
I said that I don't understand your goal, I literally mean it, I don't understand what you are trying to do with this experiment. If it's not to reproduce the statistics of the Voynich MS and not to attempt deciphering the Voynich MS, then what is it?
(30-07-2026, 10:25 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.If it's not to reproduce the statistics of the Voynich MS and not to attempt deciphering the Voynich MS,
It's still a long way off. But at least the goal now is to show that Roman numerals have the same effect and look similar.
(30-07-2026, 10:25 PM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.Until it's reliably achieved, this remains to be seen.
Exactly. But for now, we have what we have.
"show that Roman numerals have the same effect and look similar."
But I dont think you show anything of sorts? I mean, there are different ways to demonstrate that A looks similar to B. You just show A and say "look, it's simlar to B", while it's not, as least not to me and judging by general lack of interest not to many other people either. Same thing as happening in the Bavarian thread, I think. If you don't want to run statistical tests yet, ok, but maybe at least show how you can automatically produce some text (say, 500 characters) that we won't be able to identify among three samples of Voynichese.
(31-07-2026, 09:04 AM)oshfdk Wrote: You are not allowed to view links. Register or Login to view."show that Roman numerals have the same effect and look similar."
But I dont think you show anything of sorts? I mean, there are different ways to demonstrate that A looks similar to B. You just show A and say "look, it's simlar to B", while it's not, as least not to me and judging by general lack of interest not to many other people either. Same thing as happening in the Bavarian thread, I think. If you don't want to run statistical tests yet, ok, but maybe at least show how you can automatically produce some text (say, 500 characters) that we won't be able to identify among three samples of Voynichese.
As I understand it, you want the perfect result right away.
It's not here yet.
(31-07-2026, 09:04 AM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.as least not to me and judging by general lack of interest not to many other people either.
Okay, I'm going to delete my account due to a lack of interest in my versions.
I don't see the point in continuing my work here due to the lack of interest from others... Hehe
Check out other theory threads, such as Ruby's Greek Thread, Bavarian Thread, Enochian Thread, etc. I don't see anything wrong with conducting my own research, even if it turns out to be incorrect. The lack of interest from others is not a criterion of truth for me.
(31-07-2026, 09:04 AM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.If you don't want to run statistical tests yet, ok
How?
On what?
If we're talking figuratively, it's like carrying an empty canvas to a gallery. There's literally nothing for statistical tests yet, and we need to improve the algorithm.