+- 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: The Cardan grille approach to the Voynich MS taken to the next level (/thread-3537.html)
RE: The Cardan grille approach to the Voynich MS taken to the next level - MichelleL11 - 06-08-2021
(06-08-2021, 06:24 PM)julian Wrote: You are not allowed to view links. Register or Login to view.I don't want to derail this thread by injecting GA stuff, so will report further on it in my blog :-)
When you explain in your blog post, can you also comment on the second wheel that wasn't used -- how does that work? Is the program told that any of the wheels can be nothing, too?
Thanks in advance for the further discussion of how you produced the wheels.
Michelle
RE: The Cardan grille approach to the Voynich MS taken to the next level - byatan - 11-08-2021
How the wheels would actually be used is is something that interests me, both in terms of dealing with null wheels / spaces and possibilities of creating something beyond simply a vocabulary.
RE: The Cardan grille approach to the Voynich MS taken to the next level - julian - 12-08-2021
(11-08-2021, 10:24 PM)byatan Wrote: You are not allowed to view links. Register or Login to view.How the wheels would actually be used is is something that interests me, both in terms of dealing with null wheels / spaces and possibilities of creating something beyond simply a vocabulary.
Yes, indeed - it's interesting to speculate. Suppose there were nine wheels, and each wheel is used when making a word 50% of the time: that would produce word lengths that match the distribution we see in the VMS, i.e. binomial of 9. It would also suggest, if the gallows glyphs took up one of the wheels, that words containing a gallows glyph would occur about 50% of the time ... which is approximately the case. But how would the nine wheels actually be used to encipher plaintext words? Do the rarer VMS words with more than one gallows mean that the wheels were traversed more than once?
RE: The Cardan grille approach to the Voynich MS taken to the next level - MichelleL11 - 12-08-2021
(12-08-2021, 01:55 AM)julian Wrote: You are not allowed to view links. Register or Login to view.(11-08-2021, 10:24 PM)byatan Wrote: You are not allowed to view links. Register or Login to view.How the wheels would actually be used is is something that interests me, both in terms of dealing with null wheels / spaces and possibilities of creating something beyond simply a vocabulary.Do the rarer VMS words with more than one gallows mean that the wheels were traversed more than once?
The thing that has caught my imagination is the way the wheels can be used to make the same Voynichese word using different wheel combinations (see Julian's blog post You are not allowed to view links. Register or Login to view.) - thus possibly encoding completely different plaintext words. It is so weird it just could be the explanation for the unusual whole word serial repetition, yet the seeming lack of repetition of other, longer whole word patterns seen in Voynichese. Could the authors have been so caught up in their "system" that they didn't realize it how difficult this would be for reversing the cipher?
I guess the million dollar question is how can we figure out now which wheels represent what plaintext letters? Sorry to state the obvious but I'm really not sure what the next step would be. Any ideas?
RE: The Cardan grille approach to the Voynich MS taken to the next level - julian - 12-08-2021
(12-08-2021, 04:16 AM)MichelleL11 Wrote: You are not allowed to view links. Register or Login to view.(12-08-2021, 01:55 AM)julian Wrote: You are not allowed to view links. Register or Login to view.(11-08-2021, 10:24 PM)byatan Wrote: You are not allowed to view links. Register or Login to view.How the wheels would actually be used is is something that interests me, both in terms of dealing with null wheels / spaces and possibilities of creating something beyond simply a vocabulary.Do the rarer VMS words with more than one gallows mean that the wheels were traversed more than once?
The thing that has caught my imagination is the way the wheels can be used to make the same Voynichese word using different wheel combinations (see Julian's blog post You are not allowed to view links. Register or Login to view.) - thus possibly encoding completely different plaintext words. It is so weird it just could be the explanation for the unusual whole word serial repetition, yet the seeming lack of repetition of other, longer whole word patterns seen in Voynichese. Could the authors have been so caught up in their "system" that they didn't realize it how difficult this would be for reversing the cipher?
I guess the million dollar question is how can we figure out now which wheels represent what plaintext letters? Sorry to state the obvious but I'm really not sure what the next step would be. Any ideas?
Nothing is obvious in this Voynich game :-) The authors probably thought their system was obvious, as they twisted their cipher wheels, and look where we are ...
One train of thought about wheels (or tables, or whatever they might be) is that *if* there are nine wheels being used, and *if* each wheel has a 50% chance of being used, then all we need to do is look at all the long VMS words to find out which glyphs are on each wheel. Because if a VMS word has length nine, we know that all the wheels were used. So if we look at all VMS words of length nine, we should be able to assemble the set of glyphs which were on each wheel.
But this must be too simple.
RE: The Cardan grille approach to the Voynich MS taken to the next level - Koen G - 12-08-2021
If the series of wheels can be traversed more than once and each wheel has a 50 percent chance of being used, then words of length 9 are more likely to be made up of multiple passes, right?
RE: The Cardan grille approach to the Voynich MS taken to the next level - julian - 12-08-2021
(12-08-2021, 07:15 AM)Koen G Wrote: You are not allowed to view links. Register or Login to view.If the series of wheels can be traversed more than once and each wheel has a 50 percent chance of being used, then words of length 9 are more likely to be made up of multiple passes, right?
If multiple passes are allowed, yes. But that probably throws off the distributions, so I'm not sure it's valid. Also, I no longer think there are 9 wheels, I think there are 12 or maybe 13, after re-examining the statistics.
RE: The Cardan grille approach to the Voynich MS taken to the next level - AliciaNelPresente - 17-02-2026
(03-08-2021, 07:01 PM)julian Wrote: You are not allowed to view links. Register or Login to view.I suppose this may be simply a pictorial representation of Stolfi's grammar? It certainly doesn't get us any further, but I throw it in anyway. I am obsessed with the notion that a few cipher wheels were used to generate the text, perhaps geared together, and perhaps with cams that rotate adjacent wheels at certain positions :-)
Exactly, it would have taken us very further, Julian. I don't know you but, I am your fan!
---
Source: You are not allowed to view links. Register or Login to view.
Peer review: You are not allowed to view links. Register or Login to view.
The Unintentional Proof: How "Computational Attacks" Validated the Hardware
I have spent the last few hours auditing the entire history of "Computational Attacks" on the manuscript, specifically the work by JB (Julian Bunn), MarcoP, and the insights from this thread.
If you are planning to write a paper, I have good news: The data is already there. The tragedy is that the "Computational Attacks" didn't fail. They succeeded. They successfully characterized the machine with 96% accuracy, but the researchers kept interpreting the mechanical telemetry as linguistic failure.
Let me connect the dots of 15 years of research to show you why the Syntaxis Volvella (Differential Gear) is the only solution that satisfies JB’s own data.
1. The Wheels are Proven (August 2021)
In his post "Cipher Wheels - Genetic Algorithm", JB proved that a set of wheels can reproduce ~93% to 96% of VMS vocabulary.
He confirmed that the Binomial Word Length Distribution (first noted by Stolfi) is the natural byproduct of rotary wheels. He stopped because he couldn't generate "meaningful" sentences. He was looking for Latin plaintext. He should have realized the structure itself was the message.
2. The "Reset" Mechanism (December 2022)
In "Word Positions on the Folios – Part Deux", JB discovered something structurally impossible for a natural language but mandatory for a machine:
The word daiin has a massive affinity for Position 0.0 (Start of Line). Words like am and dy cluster at Position 1.0 (End of Line) > This is the Carriage Return.
Start (0.0): The scribe resets the Gallow/Lever. The gears return to Index 0 -> The Volvella outputs daiin.
End (1.0): The gears have rotated under tension. The volvella is physically restricted to short suffix states (-dy, -am).
The line isn't a grammatical sentence. It's a Mechanical Cycle.
3. The "Interlock" Proof (May 2021 - This Thread)
@lurker correctly identified why independent wheels (Slot Machine/One-Armed Bandit) fail:
"Stolfi is splitting common Voynich words... while deedy is essentially nonexistent."
If the wheels were independent (like JB's Genetic Algorithm assumed), deedy should exist!
The Volvella Solution: The Rings are Geared. If Ring A (Outer) is at d, the counter-rotation of Ring B (Middle) physically moves the -edy suffix out of the window.
deedy is not "rare." It is mechanically impossible due to the gear ratio. This is exactly what @MarkKnowles missed with his "One-Armed Bandit" joke. It's not a bandit, it's a Differential.
4. Language A vs B is a Hardware Setting (March 2013)
In "The Relationship Between Currier Languages A and B", JB found that the most common word in A (8am) becomes am in B.
Linguistic view: Did the author/scribes forget how to spell?
Volvella view: The scribe simply disengaged the Outer Ring (8) or changed the starting offset of the machine. It’s not a dialect change; it’s a calibration change.
Summary for the Forum:
JB, Marco, and the rest of you have done the hard work.
You proved it's Wheels (Binomial Dist).
You proved it resets every line (daiin at 0.0).
You proved the wheels are coupled (No deedy).
You have built the entire engine in your simulators. You just forgot to put the gears in Counter-Rotation. Once you do that, the "noise" becomes "phase", and the text becomes readable.
Alicia
RE: The Cardan grille approach to the Voynich MS taken to the next level - AliciaNelPresente - 17-02-2026
I am talking about: You are not allowed to view links. Register or Login to view.
Sorry
RE: The Cardan grille approach to the Voynich MS taken to the next level - nablator - 17-02-2026
(17-02-2026, 02:02 AM)AliciaNelPresente Wrote: You are not allowed to view links. Register or Login to view.deedy is not "rare." It is mechanically impossible due to the gear ratio.
When the facts don't fit the theory, change the facts.