The Voynich Ninja

(26-05-2019, 10:50 PM)Emma May Smith Wrote: You are not allowed to view links. Register or Login to view.

Quote:This would mean that every page has it's own topic.

Now you mention it.

Keep in mind that there is also the gradual evolution from Currier A to Currier B. Moreover, tokens with high structural similarity appear preferably in close vicinity of each other. In other words, the chance that two tokens share structural similarities decreases with the distance between these two tokens.

(27-05-2019, 07:40 PM)Torsten Wrote: You are not allowed to view links. Register or Login to view.Keep in mind that there is also the gradual evolution from Currier A to Currier B. Moreover, tokens with high structural similarity appear preferably in close vicinity of each other. In other words, the chance that two tokens share structural similarities decreases with the distance between these two tokens.

Hello Torsten,

I like the title of the article "A possible generating algorithm..." but I don't agree with the emphasis on meaningless "self-citation". Isn't it an argument from ignorance to assume that only meaningless "self-citation" is possible or probable to account for these findings? They could be explained as well by the rule of least effort in a cipher where each vord (Voynichese word) is chosen among many possibilities and reusing a vord with little or no modification is quicker than generating a completely different vord from scratch. (Exactly the type of cipher that I have been studying since 2017.)

(28-05-2019, 12:06 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.Hello Torsten,

I like the title of the article "A possible generating algorithm..." but I don't agree with the emphasis on meaningless "self-citation". Isn't it an argument from ignorance to assume that only meaningless "self-citation" is possible or probable to account for these findings? They could be explained as well by the rule of least effort in a cipher where each vord (Voynichese word) is chosen among many possibilities and reusing a vord with little or no modification is quicker than generating a completely different vord from scratch. (Exactly the type of cipher that I have been studying since 2017.)

I don't know about such an argumentation. Which statement are you referring to?

(28-05-2019, 12:41 PM)Torsten Wrote: You are not allowed to view links. Register or Login to view.I don't know about such an argumentation. Which statement are you referring to?

In the conclusion : "Why are we then convinced that the hoax hypothesis can be the only correct interpretation of the VMS?" A cipher, significantly more complex than mono-alphabetic substitution is ruled out only because of "the historical context" (1). Steganography is "not totally dismissed" but "compelling arguments against it do exist". The arguments being a rhetorical question "Why, e.g., would someone hide a genuine secret within an obviously “suspicious” book?" (2) and "high regularities of the VMS text significantly limit the maximal amount of information possibly hidden within the “container”, virtually rendering it useless." (3)

1) Fair enough but the ink is not dated and it is not unconceivable that a more complex cipher was invented or derived from an existing cipher, known since antiquity, just once. A cipher/steganography/abbreviation hybrid could be designed to cover a wide range of ciphertext:plaintext ratios for variety and increased security. (I have such a system in mind, yet unpublished.)

2) They certainly tried hard to make it look mostly like a normal European 15th century manuscript (illustrations, glyphs imitating common Latin abbreviations) so it does not look suspicious at first glance, not any more than any manuscript written in an unknown foreign language and script. Only those who had studied all existing languages and scripts would have found it suspicious, and not many people had such an encyclopedic knowledge.

3) Useless is a (too) strong word. Someone willing to write dy nearly 7000 times, and other common patterns, instead of abbreviating them to a single glyph obviously did not care much about the amount of information that they can fit on a page.

I think McCrone determined that the VMS ink was most likely added close to the time of the radiocarbon dating (I'm going by memory, so it might be a good idea to double-check).

(28-05-2019, 08:50 PM)-JKP- Wrote: You are not allowed to view links. Register or Login to view.I think McCrone determined that the VMS ink was most likely added close to the time of the radiocarbon dating (I'm going by memory, so it might be a good idea to double-check).

Yes it might be. You are not allowed to view links. Register or Login to view.

(28-05-2019, 08:15 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.In the conclusion : "Why are we then convinced that the hoax hypothesis can be the only correct interpretation of the VMS?" A cipher, significantly more complex than mono-alphabetic substitution is ruled out only because of "the historical context" (1). Steganography is "not totally dismissed" but "compelling arguments against it do exist". The arguments being a rhetorical question "Why, e.g., would someone hide a genuine secret within an obviously “suspicious” book?" (2) and "high regularities of the VMS text significantly limit the maximal amount of information possibly hidden within the “container”, virtually rendering it useless." (3)

You write it yourself. We say that steganography is "not totally dismissed" and we conclude from the facts we know that the hoax hypothesis can be the only correct interpretation of the VMS. This is something else then your statement that "only meaningless 'self-citation' is possible". 

Our main point is that the VMS text contains too many regularities. The text responds to the page of the manuscript. There is also the line as a functional unit. The shape of each glyph corresponds to the previous glyph. Last but not least the tokens respond to previously written tokens on the same page. For instance on page page You are not allowed to view links. Register or Login to view. you can find numerous instances of tokens containing <oke> or <ote> like <qokeedy>, <qokedy>, <okedy>, <otedy>, <qokeey>, <okeey>, <qokey>, <qotedy>, <okeedy> ... (see You are not allowed to view links. Register or Login to view.). What can a string like <oke> or <ote> mean if he is repeated so often? Or with other words, where is the information hidden?

(28-05-2019, 08:15 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.1) Fair enough but the ink is not dated and  it is not unconceivable that a more complex cipher was invented or derived from an existing cipher, known since antiquity, just once. A cipher/steganography/abbreviation hybrid could be designed to cover a wide range of ciphertext:plaintext ratios for variety and increased security. (I have a possible system in mind, yet unpublished). 

Indeed, if somebody presents new information it might be necessary to find a new explanation. But until now this information is not available.

(28-05-2019, 08:15 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.2) They certainly tried hard to make it look mostly like a normal European 15th century manuscript (illustrations, glyphs imitating common Latin abbreviations) so it does not looks suspicious at first glance, not any more than any manuscript written in an unknown foreign language and script. Only those who studied all existing languages and scripts would find it suspicious, and not many people had such an encyclopedic knowledge.

You interpret this word differently than we do. With suspicious book we mean a book that nobody is able to read. 

(28-05-2019, 08:15 PM)nablator Wrote: You are not allowed to view links. Register or Login to view.3) Useless is a (too) strong word.

We chose the word because in our eyes the text contains too many regularities.

Quote:Thread  "[Article] A possible generating algorithm of the Voynich manuscript"

The tool is used to demonstrate that by using this method it is possible to reproduce the statistical key properties of the Voynich manuscript.

At least the word lengths seem to differ from the VMS ( You are not allowed to view links. Register or Login to view., Figure 3  ). The "self-citation text generator" does not produce text with the "bell-like" curve as in the VMS.

(31-05-2019, 09:53 AM)bi3mw Wrote: You are not allowed to view links. Register or Login to view.

Quote:Thread  "[Article] A possible generating algorithm of the Voynich manuscript"

The tool is used to demonstrate that by using this method it is possible to reproduce the statistical key properties of the Voynich manuscript.

At least the word lengths seem to differ from the VMS ( You are not allowed to view links. Register or Login to view., Figure 3  ). The "self-citation text generator" does not produce text with the "bell-like" curve as in the VMS.

This looks interesting, thank you!
Do you think could provide histograms for both the VMS and the generated text in the same format? That would make it easier to visually compare the two...

(31-05-2019, 09:53 AM)bi3mw Wrote: You are not allowed to view links. Register or Login to view.

Quote:Thread  "[Article] A possible generating algorithm of the Voynich manuscript"

The tool is used to demonstrate that by using this method it is possible to reproduce the statistical key properties of the Voynich manuscript.

At least the word lengths seem to differ from the VMS ( You are not allowed to view links. Register or Login to view., Figure 3  ). The "self-citation text generator" does not produce text with the "bell-like" curve as in the VMS.

1) You compare a bar chart with a line chart. This doesn't make sense.

2) The relevant chart for the token length distribution at stolfis page is "Figure 1 - Token length distributions"

3) See page 2: "We use the algorithm to create a 'facsimile' of the VMS 'Recipes' section."

4) See page 15: "The line is a Gaussian fit with mean 5.73 and width 3.02. The corresponding values for the VMS 'Recipes' section are 6.09 and 3.30, respectively."

5) Please, read the paper.