17-02-2026, 09:36 AM
(This post was last modified: 17-02-2026, 10:26 AM by Koen G.
Edit Reason: Removed huge quote
)
Below, I am not stating anything as fact, it is simply my opinion based on my reading of the data presented by Mr. Dulsen from my perspective and point of view.
Hi @Dunsel You haven't just plotted data, you have visualized the Volvella's behavior.
To answer your "Big Question" immediately:
"Why does a lexical 'engine' make a drastic switch like 'ed' if the vocabulary isn’t actually changing much?"
The answer that fits your data perfectly is structural:
Because the "Engine" (The Core) remained the same. Only the "Rim" (The Outer Ring) was swapped. Let me use your numbers to prove why this must be a mechanical state change rather than a linguistic one.
1. The "Switch" vs. The "Drift"
Your cumulative chart ''Cumulative ED Count: 0ED vs ED+'' shows a flatline followed by a vertical ascent. As you rightly questioned regarding natural language: "If there’s a natural-language example where a very high-frequency digraph shows this kind of behavior, I’d genuinely be interested to see it."
There isn't one! Even your control test with Culpeper's "rk" showed a much softer distribution (76% coverage).
The ed bigram is not a word part, it is a Hardware Component (likely a specific suffix abbreviation on the Outer Ring) that was physically absent in the 0ED configuration and installed in the ED+ configuration.
2. "Repairability" = Mechanical Adjacency
This is your most crushing piece of evidence:
"Latin (Caesar) ... Within ≤2: 69.31%"
"English (Dracula) ... Within ≤2: 66.47%"
"Voynich ... Within ≤2: ~93.7%"
You asked why the vocabulary is so repairable. It is because you aren't measuring linguistic typos, you are measuring Gear Teeth. In a mechanical Volvelle, "Word A" and "Word B" are often just one click apart on the dial.
Click 1: Produces chol
Click 2 (Next tooth): Produces chor
Click 3: Produces shor
The "1 Edit Distance" you found in 82% of cases is simply the Physical Adjacency of the glyphs on the rotor. The Volvella doesn't "invent" new words, it just rotates to the neighbor. That is why 93.7% of the text is so similar, it is generated by a closed loop of neighboring states.
3. The "Engine" Didn't Change
You noted:
Precisely. The Core Rotors (Roots/Prefixes) were never changed. That is why the "high-frequency backbone" (daiin, ol, chol) is 100% shared. The only thing that changed was the Output Filter (The Suffix Ring/Modality), which determines whether the machine appends an ed (abbreviation for -dum???) or leaves it blank.
Currier A and Currier B are not dialects. They are Setup A (Ring 1 + Ring 2 + Ring 3A) and Setup B (Ring 1 + Ring 2 + Ring 3B). You have mathematically proven the existence of Volvella's "Settings Dial".
Alicia
Hi @Dunsel You haven't just plotted data, you have visualized the Volvella's behavior.
To answer your "Big Question" immediately:
"Why does a lexical 'engine' make a drastic switch like 'ed' if the vocabulary isn’t actually changing much?"
The answer that fits your data perfectly is structural:
Because the "Engine" (The Core) remained the same. Only the "Rim" (The Outer Ring) was swapped. Let me use your numbers to prove why this must be a mechanical state change rather than a linguistic one.
1. The "Switch" vs. The "Drift"
Your cumulative chart ''Cumulative ED Count: 0ED vs ED+'' shows a flatline followed by a vertical ascent. As you rightly questioned regarding natural language: "If there’s a natural-language example where a very high-frequency digraph shows this kind of behavior, I’d genuinely be interested to see it."
There isn't one! Even your control test with Culpeper's "rk" showed a much softer distribution (76% coverage).
The ed bigram is not a word part, it is a Hardware Component (likely a specific suffix abbreviation on the Outer Ring) that was physically absent in the 0ED configuration and installed in the ED+ configuration.
2. "Repairability" = Mechanical Adjacency
This is your most crushing piece of evidence:
"Latin (Caesar) ... Within ≤2: 69.31%"
"English (Dracula) ... Within ≤2: 66.47%"
"Voynich ... Within ≤2: ~93.7%"
You asked why the vocabulary is so repairable. It is because you aren't measuring linguistic typos, you are measuring Gear Teeth. In a mechanical Volvelle, "Word A" and "Word B" are often just one click apart on the dial.
Click 1: Produces chol
Click 2 (Next tooth): Produces chor
Click 3: Produces shor
The "1 Edit Distance" you found in 82% of cases is simply the Physical Adjacency of the glyphs on the rotor. The Volvella doesn't "invent" new words, it just rotates to the neighbor. That is why 93.7% of the text is so similar, it is generated by a closed loop of neighboring states.
3. The "Engine" Didn't Change
You noted:
"This is a very smooth growth rate. This suggests that there was no big shift between 0ED and ED+. ... it's still the same base 'engine' chugging along"
Precisely. The Core Rotors (Roots/Prefixes) were never changed. That is why the "high-frequency backbone" (daiin, ol, chol) is 100% shared. The only thing that changed was the Output Filter (The Suffix Ring/Modality), which determines whether the machine appends an ed (abbreviation for -dum???) or leaves it blank.
Currier A and Currier B are not dialects. They are Setup A (Ring 1 + Ring 2 + Ring 3A) and Setup B (Ring 1 + Ring 2 + Ring 3B). You have mathematically proven the existence of Volvella's "Settings Dial".
Alicia