Hi again Ed,
Thanks for the reply, I'm having a really nice Sunday afternoon, to be honest haha!
Let’s start with point #3 because it made me chuckle. You say, "I know an AI response when I read it."
It is fascinating how "You are an AI" has become the modern version of "You are a Heretic." Would an AI cite specific page numbers from René Zandbergen’s 2021 papers to correct your math? Because that is exactly what I am about to do:
Let's dismantle the pillars of your critique using the documentation you ignored.
The Statistical Critique (You are fighting Zandbergen, not me)
Claim:
"With 17×19×23 teeth you get 7,429 combinations... leaving no room for redundancy."
Correction: You are creating a strawman with those numbers. If you look at Zandbergen, Page 6, he explicitly calculates the capacity of a standard 3-wheel system,
below is a direct quote (I like this pink colour):
"Three wheels with only 24 word fragments each, will give rise to 24x24x24 = 13,824 different words, already well in excess of the number of word types in the manuscript''
The VMS has ~8,000 word types. A simple 24-tooth system (which is small) generates nearly 14,000 combinations.
There is a massive surplus of ~6,000 slots. That is plenty of room for the "Redundancy" I described to create the Zipf skew. The math supports the machine.
---
Claim:
"The blank teeth also predict bimodal word lengths... VMS word lengths are smooth and unimodal."
Correction: This is factually incorrect.
Please see Zandbergen, Page 12, Figure 7. He demonstrates that a 3-wheel system with variable/null slots produces a distribution that is "exactly binomial" (unimodal). The mechanical interaction smooths the curve. It does not split it. You are predicting a bimodal output for a system that has been mathematically proven to generate a binomial one.
---
Claim:
"Coprime gears... make a specific testable prediction: periodic autocorrelation with cycle length 323 words."
Correction: You are assuming the machine runs infinitely without stopping. The cycle length is indeed 323 words if you never lift the pen. But the scribe hits a Carriage Return (Line Reset) every ~8-10 words.
The "Reset" breaks the cycle before the periodicity can become visible. The autocorrelation doesn't decay because it's "natural language", it decays because the mechanical cycle is interrupted and reset 30 times per page.
The Historical Critique (The "1600s" Myth) ahah this is the most fun one!
Claim:
"Mechanical Volvelles didn't actually exist until the 1600s... so either time travel is involved..."
Correction: This is a surprisingly common misconception.
- Ramon Llull (died 1315): Used concentric rotating wheels (Ars Magna) to generate combinations of concepts.
- Giovanni Fontana (c. 1420): A contemporary of the VMS, designed cryptographic instruments and bellicose machines.
- Leon Battista Alberti (1467): Explicitly described the Cipher Disk.
To say volvelles didn't exist until the 1600s is to erase three centuries of astronomical and combinatorial history. The VMS fits perfectly in the "experimental gap" between Llull and Alberti.
---
Claim:
"Section-level register differences... would need wheel swaps, which puts the instruction set back in play."
Agreement: Yes! Exactly!
Different sections (Herbal vs. Bio) use different wheels. That is the instruction set. Just as an Enigma machine operator changes rotors for a different day, the VMS scribe changed wheels for a different subject. That’s not a flaw in the theory, it’s a feature of the hardware.
Please allow me to delete what I mentioned earlier, “thank you for raising the level.”
(15-02-2026, 08:12 PM)nablator Wrote: You are not allowed to view links. Register or Login to view. (15-02-2026, 06:49 PM)AliciaNelPresente Wrote: You are not allowed to view links. Register or Login to view.It is a 3-Ring Stack configured as follows:
1. High Redundancy on the Middle Ring: Painting frequent roots like "chol" on 50% of the teeth solves the Zipf/Frequency issue.
2. Coprime Tooth Counts (Like 17 vs 19): Mismatching the rings solves the Length Autocorrelation/Clustering issue.
3. Line-Start Mechanics (Reset): Resetting the lever at the margin solves the Entropy/Positional issue.
How do they solve anything?
I remember Nick Pelling jokingly saying on his blog that it looks like a cipher wheel was stuck on "or" or "aiin". That would definitely explain some pages. 
(15-02-2026, 08:04 PM)DG97EEB Wrote: You are not allowed to view links. Register or Login to view....adding a 3rd... I know an AI response when I read it...
CopyLeaks doesn't detect it (0% AI content). It's very reliable usually.
As a coworker likes to say: ''I'm pointing at the moon, and you're staring at my finger''
