16-05-2026, 07:26 PM
Hi Ed,
Thank you for reading through my post and going into detail on the presentation and replication issues you are experiencing. You beat me to the next post, and asked the right questions here for the next section.
The “translation” framing in the viewer was an early idea that morphological words could be attached to the tokens once they were decomposed to the interior structure. I realized quickly that this was an incorrect interpretation and removed the visual layer, but left it there programmatically. It should have been stripped completely. The viewer should simply be looked at as an early demo and not as the canonical proof layer. Any content layer work is now kept separate from the machine and apparatus layers.
You are correct that the role assignment is effectively 1 to 1 at the word type level. This is expected, and not necessarily a gap. Each word is a concatenation of the selector state: word_t = concat(prefix, gallows, dial0,dial1,..., dialN, suffix). The word is the machine state readout, and not a lookup into it. The determinism chain shows entropy collapsing from 13 bits to 0 as each selector is added. By the time decomposition is complete, exactly one word can appear at that position and one role. The underlying concentric decomposition is not original to my work, Stolfi’s 2000 analysis identified the same layered word structure (crust-mantle-core) from frequency analysis. What I extended was his resolution, those three layers decompose further into the 8 prefix positions, 90 suffix positions and 8 interior dial positions with LCG arithmetic.
For the C10 Slot Purity test, MACHINE_PROOF.md §8 is core to some of the additional work and theories I have. This will help validate some of the theories I am still making regarding this aspect. Please see the attached MACHINE_PROOF S8.txt regarding this request. Please let me know if you have any follow up questions or feedback on this additional data that I can clear up.
The token discrepancy is related to my adversarial approach in the corpus and testing methodologies. I maintain a copy of the IVTFF and the Voynichese.com transcriptions and run the battery of tests on both. IVTFF has 37,671 tokens and tested with lag -1 it hits 37,670 / 37,670. Voynichese.com maintains a corpus of 37,886 raw tokens. I filtered out the non Voynich tokens (?, !, etc) from the Voynichese.com version and lines < 2 to obtain the final token count of 37,046 for testing purposes. During lag testing at -1, the normalized Voynichese.com corpus gives 37,045/37,045. This does make it difficult to present a single number appropriately, and I will find a way to correct this in the presentation layer. The Voynichese.com layer represents the majority of the early work, before settling to IVTFF as the primary corpus.
![[Image: 237f7bd38cedcd03bc73ab2d6a42fb95.png]](https://i.gyazo.com/237f7bd38cedcd03bc73ab2d6a42fb95.png)
![[Image: 0adeaf31e81842d300bdeb2fa178fc9a.png]](https://i.gyazo.com/0adeaf31e81842d300bdeb2fa178fc9a.png)
For the za’irajah comparison, I have two matrices, one for the architectural features, which has 16, and one for the procedural and operational procedures from the Muqaddimah which contains 20 features. I referenced one image and then posted the second image. I added the other image here for reference. The references for the abjad encoding, the pharmaceutical use, the input process, the way to decipher the lookup tables, specific look up table navigation all are coming from the Muqaddimah. Without the physical za’irajah to verify I am feature matching off of these values that the book calls out. The modular arithmetic (LCG D_{k+1} = a·D_k mod N), multi tier selectors, the tashdid binary doubler, and elemental letter groups mapped to the detents are the ones that carry the weight. I cite Link’s work on the structural description, but not the full procedure and the incompleteness of his look up table is a fair point. I have been working on a version of the look up table recovery process to some degree of success.
![[Image: 69855a19caa1438f26789efac5ccbc02.png]](https://i.gyazo.com/69855a19caa1438f26789efac5ccbc02.png)
The idea that the za’irajah was a pharmaceutical device or at least had some usage for this aspect and was in scope for the device, came from the Muqaddimah and the operating poem itself. You are not allowed to view links. Register or Login to view. Pages 182 to 187 of the book are the poem itself and page 188 of book three calls out the poem’s usage and the reference mentions the look up table. The title of that is Ibn Khaldun's statement to the operating poem, and called out the science of medicine and the crafting of alchemy. The poem calls out alchemy, the science of medicine, elixirs and mixtures and patients being cured in almost every line and then directly references to the look up table values, making this more probable. Unfortunately, this and the other book I mentioned that is in Morocco are the only true first or even translation references we have that I am aware of. I realize I am just referring back to the same document you point out is a small sample.
The poem for easy reference.
‘O student of medicine and of Jâbir's science (alchemy),⁸⁵⁵
And you who know the extent of the powers in succession,
If you want the science of medicine, there must exist a
relationship
For the laws of scales, which ⁸⁵⁶ will hit upon a (possible) way.
III, 153
Your patient will be cured. The elixir will be right.
The soundness of the mixtures of your composition will
be apparent.’
In relation to your independent convergence, I found 9 named suffixes (dy, in, ey, ol, ar, hy, al, ky, am) that came from the original morphological decomposition and one miscellaneous catch-all that handled everything else. The misc accounts for 28.4% of tokens, roughly ~10,506 tokens in the corpus. It contains 81 hidden classes, split by terminal 2 characters of the word. Not of the suffix label, of the full word form, since misc words end at different positions. The disjointness criterion at Jaccard = 0 for all pairs means each 2-character ending maps to a completely distinct interior pool. This confirmed mechanically the distinct wheel positions. Focusing on my theory that this is OAT/Letterist technology in the Eastern abjad, 90 appears firmly in the position of ص (sad) and is present in the za'irajah lookup table headers. In abjad cosmology, both are boundary values but 90 closes tens register and 9 closes the units.
Jaccard disjointness test:
# For each pair of misc sub-classes A, B:
s1 = {interior(w) for w in misc_words_ending_in_A}
s2 = {interior(w) for w in misc_words_ending_in_B}
inter = len(s1 & s2) # shared interiors
union = len(s1 | s2) # all interiors combined
J(A, B) = inter / union # = 0.000 for all 28 tested pairs
The expansion rule is:
def expanded_suffix(word, suffix):
if suffix != "misc":
return suffix
end2 = word[-2:] if len(word) >= 2 else word # terminal 2 characters of word
return "misc_" + end2
Thank you,
Sean Rice
Thank you for reading through my post and going into detail on the presentation and replication issues you are experiencing. You beat me to the next post, and asked the right questions here for the next section.
The “translation” framing in the viewer was an early idea that morphological words could be attached to the tokens once they were decomposed to the interior structure. I realized quickly that this was an incorrect interpretation and removed the visual layer, but left it there programmatically. It should have been stripped completely. The viewer should simply be looked at as an early demo and not as the canonical proof layer. Any content layer work is now kept separate from the machine and apparatus layers.
You are correct that the role assignment is effectively 1 to 1 at the word type level. This is expected, and not necessarily a gap. Each word is a concatenation of the selector state: word_t = concat(prefix, gallows, dial0,dial1,..., dialN, suffix). The word is the machine state readout, and not a lookup into it. The determinism chain shows entropy collapsing from 13 bits to 0 as each selector is added. By the time decomposition is complete, exactly one word can appear at that position and one role. The underlying concentric decomposition is not original to my work, Stolfi’s 2000 analysis identified the same layered word structure (crust-mantle-core) from frequency analysis. What I extended was his resolution, those three layers decompose further into the 8 prefix positions, 90 suffix positions and 8 interior dial positions with LCG arithmetic.
For the C10 Slot Purity test, MACHINE_PROOF.md §8 is core to some of the additional work and theories I have. This will help validate some of the theories I am still making regarding this aspect. Please see the attached MACHINE_PROOF S8.txt regarding this request. Please let me know if you have any follow up questions or feedback on this additional data that I can clear up.
The token discrepancy is related to my adversarial approach in the corpus and testing methodologies. I maintain a copy of the IVTFF and the Voynichese.com transcriptions and run the battery of tests on both. IVTFF has 37,671 tokens and tested with lag -1 it hits 37,670 / 37,670. Voynichese.com maintains a corpus of 37,886 raw tokens. I filtered out the non Voynich tokens (?, !, etc) from the Voynichese.com version and lines < 2 to obtain the final token count of 37,046 for testing purposes. During lag testing at -1, the normalized Voynichese.com corpus gives 37,045/37,045. This does make it difficult to present a single number appropriately, and I will find a way to correct this in the presentation layer. The Voynichese.com layer represents the majority of the early work, before settling to IVTFF as the primary corpus.
For the za’irajah comparison, I have two matrices, one for the architectural features, which has 16, and one for the procedural and operational procedures from the Muqaddimah which contains 20 features. I referenced one image and then posted the second image. I added the other image here for reference. The references for the abjad encoding, the pharmaceutical use, the input process, the way to decipher the lookup tables, specific look up table navigation all are coming from the Muqaddimah. Without the physical za’irajah to verify I am feature matching off of these values that the book calls out. The modular arithmetic (LCG D_{k+1} = a·D_k mod N), multi tier selectors, the tashdid binary doubler, and elemental letter groups mapped to the detents are the ones that carry the weight. I cite Link’s work on the structural description, but not the full procedure and the incompleteness of his look up table is a fair point. I have been working on a version of the look up table recovery process to some degree of success.
The idea that the za’irajah was a pharmaceutical device or at least had some usage for this aspect and was in scope for the device, came from the Muqaddimah and the operating poem itself. You are not allowed to view links. Register or Login to view. Pages 182 to 187 of the book are the poem itself and page 188 of book three calls out the poem’s usage and the reference mentions the look up table. The title of that is Ibn Khaldun's statement to the operating poem, and called out the science of medicine and the crafting of alchemy. The poem calls out alchemy, the science of medicine, elixirs and mixtures and patients being cured in almost every line and then directly references to the look up table values, making this more probable. Unfortunately, this and the other book I mentioned that is in Morocco are the only true first or even translation references we have that I am aware of. I realize I am just referring back to the same document you point out is a small sample.
The poem for easy reference.
‘O student of medicine and of Jâbir's science (alchemy),⁸⁵⁵
And you who know the extent of the powers in succession,
If you want the science of medicine, there must exist a
relationship
For the laws of scales, which ⁸⁵⁶ will hit upon a (possible) way.
III, 153
Your patient will be cured. The elixir will be right.
The soundness of the mixtures of your composition will
be apparent.’
In relation to your independent convergence, I found 9 named suffixes (dy, in, ey, ol, ar, hy, al, ky, am) that came from the original morphological decomposition and one miscellaneous catch-all that handled everything else. The misc accounts for 28.4% of tokens, roughly ~10,506 tokens in the corpus. It contains 81 hidden classes, split by terminal 2 characters of the word. Not of the suffix label, of the full word form, since misc words end at different positions. The disjointness criterion at Jaccard = 0 for all pairs means each 2-character ending maps to a completely distinct interior pool. This confirmed mechanically the distinct wheel positions. Focusing on my theory that this is OAT/Letterist technology in the Eastern abjad, 90 appears firmly in the position of ص (sad) and is present in the za'irajah lookup table headers. In abjad cosmology, both are boundary values but 90 closes tens register and 9 closes the units.
Jaccard disjointness test:
# For each pair of misc sub-classes A, B:
s1 = {interior(w) for w in misc_words_ending_in_A}
s2 = {interior(w) for w in misc_words_ending_in_B}
inter = len(s1 & s2) # shared interiors
union = len(s1 | s2) # all interiors combined
J(A, B) = inter / union # = 0.000 for all 28 tested pairs
The expansion rule is:
def expanded_suffix(word, suffix):
if suffix != "misc":
return suffix
end2 = word[-2:] if len(word) >= 2 else word # terminal 2 characters of word
return "misc_" + end2
Thank you,
Sean Rice
(16-05-2026, 06:07 AM)DG97EEB wrote: You are not allowed to view links. Register or Login to view.Hi Sean,
Thanks for the detailed write-up and for sharing the viewer. I spent some time with the zip file and tried to reproduce the central claim. A few notes, roughly in order of how central they seem to the framework.
On C10: "100% slot purity at lag=-1, 37670/37671 tokens"
I tested this against a ZLZI based transliteration with a four-slot prefix/gallows/core/suffix decomposition (37,465 tokens; within-line adjacency, n = 32,303 pairs). The natural readings I could derive from your description:
gallows[t] == gallows[t-1] (literal identity): 43.3%
gallows[t] given token[t-1] (best class given previous token): 66.4%
gallows[t-1] given (prefix, core, suffix) of token t: 64.9%
has_gallows[t] == has_gallows[t-1] (binary presence): 53.2%
gallows[t] == prefix[t+1] (literal display-register reading): 13.8%
shuffled-gallows null: 65.0% (SD 0.10 pp), so the best reading is 1.4 pp above null
naive first-character baseline given previous token: 33.4% entropy reduction, slightly stronger than the gallows version at 29.9%
The best reading I could construct is about 33 percentage points below 99.997%. The "1 exception out of 37,671" precision is also the part I'd most like to see derived: categorical data on a 9-value column doesn't naturally produce exactly one exception under any test I could construct from the text. If MACHINE_PROOF.md §8 is shareable, the rule and the result would both be straightforward to verify.
On manuscript_data.js
The file contains a type-consistent VMS to English substitution dictionary: 7,861 distinct word types, each mapped to one fixed English gloss, with zero word types showing more than one gloss and only 20 of 7,861 showing role variation. Multiple distinct VMS tokens collapse onto the same gloss: daiin = aiin = "reference"; chedy = shedy = "and"; ar = dar = "seek"; al = dal = "take"; qokeey = qokedy = "transform".
Applied to You are not allowed to view links. Register or Login to view. line 1 the result is "which seek type marker become reservoir formula reading ember become". Applied to You are not allowed to view links. Register or Login to view. line 1 it is "vessel purification ember the opening trace open vessel extract toward pathway vent outward quench fire vessel opening ember sustains". The output is not coherent English at the sentence level.
The forum post says "we cannot translate it"; the viewer file is, functionally, a complete word-by-word translation. Those two framings are contradictory, and it would help to know which one is current.
On the execution graphs
Role assignment in the data file is fixed per word type: every daiin is LBL, every qokeey is PRC, with only 20 of 7,861 types showing any variation. That makes the "FULL_EXECUTION" graphs a token sequence with one colour per word type rather than a per-position computation. If there is machinery that produces context-dependent role assignment, the token-to-role rule would be the thing to share, because the current visualisation cannot show it.
On the za'irajah comparison
Several of the 20 features in the Image 4 matrix ("Pharmaceutical use", "Humoral degree encoding", "Abjad numeral system", "Contemporary 1400–1440 CE") align with the proposed reading of the VMS by construction rather than independently. A version of the test restricted to device-independent geometric and procedural primitives, pre-registered before the VMS comparison, would be more informative.
Two further concerns with the historical comparator. David Link's reconstruction of the as-Sabti zairja (Scrambling T-R-U-T-H: Rotating Letters as a Material Form of Thought, in Variantology 4, eds. Zielinski and Fuerlus, Cologne: König, 2010) treats the procedure as deterministic and rule-governed but produces only a partial reconstruction; he writes in his Section 10 that he was unable to determine a coherent procedure for the reverse-table operation in the later cycles. The single example output in the Muqaddimah (the Idris / Holy Spirit answer) is one rhymed verse of around 47 characters produced from one question, not a 37,000-token continuous corpus, so the determinism claim and the corpus-generation claim are at different scales. Rosenthal's footnote (Muqaddimah trans. Rosenthal vol. 3, p. 197, fn. 880) describes the rationale of the letter-determination procedure and its relationship to the table as "by no means clear to me", and Link's reconstruction inherits that uncertainty for parts of the procedure.
I have also not found a primary-source attestation of the zairja as a pharmaceutical or medical device. Link's account, the standard scholarly reference, treats the device throughout as divinatory. If you are drawing on a medical-zairja tradition, a citation would help locate it.
Minor
The token count differs across your own materials. The forum post and the OpenSCAD comments cite 37,671; manuscript_data.js (generated 2026-04-07) gives 37,046. The C10 denominator does not match the data file. Worth reconciling.
None of this is a claim that the underlying structural intuitions are not worth pursuing. The four-slot decomposition in your wheel_spec (8 prefixes, 4 gallows, 9 named suffixes) tracks with what I've been working on independently - I'm curious how you landed on this configuration. The questions above are about the deterministic-generator claim and the za'irajah linkage specifically. If you can share the §8 rule and the source transliteration, I will run the same tests against your numbers and post the result.
Best,