The Voynich Ninja

I can only support that, but then we’d have to implement some kind of procedure, and I have no idea exactly what that would look like—partly because, for example, VBM doesn’t accept spaces, and other ciphers have different specifications.

I actually find the interface of Ed / DG97EEB very interesting: You are not allowed to view links. Register or Login to view.. Here, you can upload VMS text generated using a cipher and test how well it matches the VMS:

Under “Statistics and Comparison: Upload and Compare,” you can enter a text and have it tested against the entire VMS or individual sections.

Of course, the values are only a guideline, partly because too many factors can influence accuracy (the number of glyphs used, for example), but it’s a rough and useful benchmark—because the “feeling” that something is Voynichese is usually far from reality. 

Here you can see how far you are from the reference values. 

The image is just a screenshot; Ed should add a “Download” button that lets you download the entire table as an image:

@JoJo_Jost, thank you very much! This will be much better than checking by eye

So far, there have been no successes in developing the cipher . I'm trying to diversify the alphabet itself, keeping its size below 25. I'm not a very creative person... BUT!
I noticed a peculiarity. At will, some of the numbers we have we can represent as a sum. This also does not complicate the decryption (in the worst case, the desired letter is guessed by context). 
In the Voynichese cipher, this can be a way to get rid of multiple repetitions, among other things. For example, the repetition of o o (XV XV; which sometimes occurs in my cipher) can be represented as XXX, i.e. as che (XX+X), or as eee (X+X+X). 
So, we can shorten the following word a bit: XVIII XV XV XII IV = oiin che ein (dan) ok. By the way, if I introduce XI as od, then the next unit will turn into y, not into n: oiin che ody ok.
Otherwise, it can be written as: okoiin cheody.
As I mentioned in the previous post, the alphabet can be modified to make it positional, creating the LAAFU effect. If a word appears at the beginning of a line, the numbers within it change according to their specific "initial" variations in the alphabet. 
It is also important to avoid making the cipher more complex by making such changes...
This summation will allow you to fit more words into a line and "compress" the encrypted text. The summation should be done within reasonable limits, otherwise the decryption will stop at the selection of possible summands. I think limiting the number of terms to two will make the task easier. If you see che in the text, you can quickly guess that it is a hidden repetition of o o or e e e.
It is also possible that the numbers are combined, as in my example with sh eor - sheor. To eliminate or further disguise a repetition in a word, you can also replace the numbers.
It's actually easy to do all this, but I'm not in the mood today...

P.S. Please don't count my laziness as a flaw in this theory!

The problem is that there isn't enough information in what follows the line-start markers to be able to infer such a “change” or anything else from it  .

It would be so easy, but unfortunately it isn't.

And yes, decrypting a cipher is extremely difficult—it's time-consuming, nerve-wracking, and requires a lot of patience—all that and much more....

(18-07-2026, 08:43 PM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.And yes, decrypting a cipher is extremely difficult—it's time-consuming, nerve-wracking, and requires a lot of patience—all that and much more....

In what place or meaning? I don't see any problem with matching glyphs with numbers or numbers with letters.
Of course, as the person who came up with this code, it will always be easier for me to do this, so I will ask you about the difficulties you may encounter. Let's take the texts I've already published as an example.

(18-07-2026, 08:43 PM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.The problem is that there isn't enough information in what follows the line-start markers to be able to infer such a “change” or anything else from it

Line start markers? Do you mean EVA s and gallows? I'm sorry, I didn't understand that.

Oh, I see—I misunderstood you... sorry...

(18-07-2026, 09:01 PM)JoJo_Jost Wrote: You are not allowed to view links. Register or Login to view.Oh, I see—I misunderstood you... sorry...

Rather, I should have asked your forgiveness
If you mean VMS, then yes, you're right. But at this stage, VMS is a reference point for our mini-cipher. We need to check how easy it is to achieve such effects.

Donec eu mi sed turpis feugiat
XV-XIV-V-IV-III / XXI-V / XIII-IX / XIX-V-IV / XXI-XX-XIX-XVIII-XVI-IX / XXI-XX-IX-VII-VI-V-I / XXI-XX-IX-VII-VI-V-I
Integer turpis arcu, pellentesque eget, cursus et, fermentum ut, sapien
XX-XVIII-XIV-IX-VII-V-V / XXI-XX-XIX-XVIII-XVI-IX / XXI-XVIII-III-I / XXI-XX-XIX-XVII-XVI-XIV-XII-XII-V-V-V-V / XX-VII-V-V / XXI-XXI-XIX-XIX-XVIII-III / XX-V / XXI-XX-XVIII-XIV-XIII-XIII-VI-V-V / XXI-XX / XIX-XVI-XIV-IX-V-I
Special changes in the alphabet: in the case of XV XIV, you can write it either as o oke or as qo ke (the second o disappears; in the case of qokey, you understand that the ke part is oke, i.e. XIV. If it doesn't fit, it's a null gallows). Also d is no longer null. Now it can replace V. We also turn en into ey. Also, an (VI) at the end can be replaced with dy (also VI). In general, it can be replaced not only at the end, but it is better to avoid this for now. 


qo oke a ote iin / choky / daiin or / eor chok / chol sh eor oiin s or / chor sh or ain an / sh oiin ot ain a a / chor sh eol oiin s or / chor oiin iin n / chol sh eor cph eol / dan a a a a / sh ain a a / sh ain a a / chor chor eor eol oiin iin / sha / chor sh oiin oke daiin daiin an a a / chor sh / eor qo oke ol a n 
1). During the encryption process, I noticed that the first word can be represented in a slightly different way:
qo oke d ote iin - qokedy otey ytey - the last units were distributed among the remaining words. The te part in ytey is null. By getting rid of it, you will get otey yy, which is equivalent to ote iin.
or
qo oke a ote iin - qokey oteaiin - a simpler option, but boring 
2). The peculiarity of the word choky is that the y element makes it clear that it is not chok (XXIV), but XXI V, which were glued together.
Let's make some minor cosmetic changes, keeping in mind the "no more than two terms" rule:
qokey otey ytey / choky / daiin or / chokeor / chol sheor soiin or / chor shor chain dy / shoiin otain cha am / chor sheol soiin or / chor oiin daiiin / chor sheor cpheol / dan cheey / shain shey / chor chor ycheor cheoly oiin y (ycheor cheoly y - iin) / shar / chor shoiin okey daiin daiin chan ches / chor sheor qokeol dy 

You can see that the changes made have almost erased the line between real symbols and their null counterparts. This can indeed make decryption more difficult, but not by much - once you've converted all possible numbers into letters and solved the anagram beforehand, you can figure out where the nulls, additions, and so on are.

(18-07-2026, 10:07 PM)ololololo Wrote: You are not allowed to view links. Register or Login to view.You can see that the changes made have almost erased the line between real symbols and their null counterparts. This can indeed make decryption more difficult, but not by much - once you've converted all possible numbers into letters and solved the anagram beforehand, you can figure out where the nulls, additions, and so on are.

Can you actually read this cipher off the page? I mean, can a person who spent some time using this cipher look at "chor shoiin okey daiin daiin chan ches / chor sheor qokeol dy" and after thinking for a few seconds at most get the Latin original? I have no idea how your cipher works exactly, but the description looks insanely complicated. It looks like it might take more than a minute to read just a single phrase. This looks prohibitively complex to me, not very suitable for a book. Not impossible, of course, just not very practical.

(19-07-2026, 12:29 AM)oshfdk Wrote: You are not allowed to view links. Register or Login to view.Can you actually read this cipher off the page? I mean, can a person who spent some time using this cipher look at "chor shoiin okey daiin daiin chan ches / chor sheor qokeol dy" and after thinking for a few seconds at most get the Latin original? I have no idea how your cipher works exactly, but the description looks insanely complicated. It looks like it might take more than a minute to read just a single phrase. This looks prohibitively complex to me, not very suitable for a book. Not impossible, of course, just not very practical.

@ oshfdk 

Just a quick note : Your question contains an as-yet-unproven assumption, namely that the cipher must be quick to read.

Since, if it is a cipher, it is almost certainly a complex one (otherwise it would have been cracked already), it would always take some time to decipher it.

There’s no reason to assume that it has to be quick to read, especially if it involved knowledge that was dangerous at the time. In this case, the main concern was preventing it from being decrypted in order to protect one’s life. The time it took to decrypt it likely played a lesser role.