Let there be meaning

[oshfdk]

Let there be no connection to meaning!

I mean, no connection to meaning is implied when I talk about textual structures and statistics. Meaning is certainly external to these and lies in the structure between the text and other objects, features or behaviors.

[ReneZ]

There is a problem, I use the file that I downloaded, I think, about 2 years ago when I first decided to look at the Voynich manuscript statistics. I assumed that it would be uniquely identifiable from its top header.

The header in full reads:

#=IVTFF Eva- 1.7
# ZL transliteration file, updated from EVMT project
# Version 1r of 11/04/2020

Could you help with identifying the proper name of this file, by which I should refer to it?

No problem, really.
This is the older version "1r" of the "ZL" file, or more elaborately: the Zandbergen-Landini transliteration, as explained at You are not allowed to view links. Register or Login to view.

It is still available at the 'legacy' files page: You are not allowed to view links. Register or Login to view.
The latest published version, 3a, has a number of mistakes corrected.

Using the ZL file means that you have to cope with a lot of special symbols (&nnn; notation) and alternative readings ([ab] notation). This can be avoided by using the IT flle (incomplete and less accurate) or the RF file (quite new, but complete and less complicated). These are all found in the table:
You are not allowed to view links. Register or Login to view.

[oshfdk]

This is the older version "1r" of the "ZL" file, or more elaborately: the Zandbergen-Landini transliteration, as explained at You are not allowed to view links. Register or Login to view.

It is still available at the 'legacy' files page: You are not allowed to view links. Register or Login to view.

I'll update the text. Should I reference Renè Zandbergen and Gabriel Landini as the authors of the transliteration file?

[ReneZ]

If you reference the file, it is sufficient for me.

[oshfdk]

• 10 NUMBERS FROM 1 TO 500:
133   499   165   144   207   215   491   361   336   400

By the way, your example with numbers is perfect to demonstrate what I'm talking about. Imagine, that we have a number from 1 to 500, let's take the first number in your sequence: 133. What is the probability that another number starting with 13 will appear among the other 9 numbers? There are 11 numbers between 1 and 500 that start with 13 (13, 130-139), and there are 9 independently generated numbers, so the probability is 1 - (1 - 13 / 500) ** 9, or 21%. If another number starting with 13 appeared in this sequence, I'd call this curious, but not very remarkable.

On the other hand, the probability of finding two numbers sharing the same 2-digit prefix in this list is very high, I won't compute it here, but we can try adapting the Birthday paradox formula for this if needed. Edge cases like prefixes above 49, etc make it a bit complicated, but I won't be surprised if it's actually above 50%. And as you can see, 499 and 491 in your list share a two-digit prefix. 

The probabilities of the co-occurrence of three prefixes from 72r3 and 67r2, given we chose labels on 67r2 before starting the experiment, is 0.58%. It's the same as the probability of any chosen number between 1 and 500 to appear among the first 4 numbers of the random sequence.

So, imagine you fix a number between 1 and 500 in your head, say 432 and generate 4 random numbers as above, and on the first try they go:

113 499 432 369

That's the level of statistical significance we are talking about. Not bullet-proof, and as I write in my text, particle physicists would laugh at this, but for the purposes of identifying meaningful patterns in the Voynich manuscript I find this more than enough.

[nablator]

On a more fundamental point, I am concerned that a "stateful one to many cipher" may be helpful when trying to decode text, but will not be able to explain how any plain text, when encoded using this, will have the properties that Marco Ponzi already pointed out.

No one claimed that it, alone, explains the properties that MarcoP You are not allowed to view links. Register or Login to view..

If "distance-based ciphertexts" are possible, I'd like to see concrete examples (in the other thread) of ciphertext that looks a bit like Voynichese to get a better appreciation of what would cause these properties to appear and how encoding would work in practice: how many ad hoc suppositions are needed and how constrains that rule Voynichese statistical properties could be enforced.

[oshfdk]

On a more fundamental point, I am concerned that a "stateful one to many cipher" may be helpful when trying to decode text, but will not be able to explain how any plain text, when encoded using this, will have the properties that Marco Ponzi already pointed out.

No one claimed that it, alone, explains the properties that MarcoP You are not allowed to view links. Register or Login to view..

Actually, I do claim this, quite explicitly. A distance based cipher has very good explanation for many of the properties of Voynichese. On the other hand, most one-to-many ciphers can explain a lot of properties. Distance based ciphers just play nicely into the irregular spacing patterns and irregular character shapes..

However, until I have some specific evidence, I don't claim Voynichese is a distance based cipher. I have just started on my fourth text in Voynich marathon, dedicated to identifying the properties of possible one-to-many encoding. I really have no good answer so far, it's better to stay open-minded.

If "distance-based ciphertexts" are possible, I'd like to see concrete examples (in the other thread) of ciphertext that looks a bit like Voynichese to get a better appreciation of what would cause these properties to appear and how encoding would work in practice: how many ad hoc suppositions are needed and how constrains that rule Voynichese statistical properties could be enforced.

Thank you for the reference to Marco's post, I forgot about it. I've added some information in that thread.

[nablator]

Actually, I do claim this, quite explicitly.

No, I wrote alone, as if all "stateful one to many cipher" should show these properties: of course they don't. Some do, some don't. Marco and René expressed their concerns as if there were no additional requirements.

[oshfdk]

Actually, I do claim this, quite explicitly.

No, I wrote alone, as if all "stateful one to many cipher" should show these properties: of course they don't. Some do, some don't. Marco and René expressed their concerns as if there were no additional requirements.

I see, I misunderstood this comment. Anyway "stateful one to many character based additive cipher" is an umbrella term for almost any stream cipher. Any encoding that you can apply without knowing the whole source message in advance will usually conform to this very abstract definition. Even simple substitution with multiple tables selected via prefixes is one of those. I hope that using the most generic properties of these ciphers it will be possible to attempt breaking the Voynich encoding without assuming any scheme in particular.

[ReneZ]

On a more fundamental point, I am concerned that a "stateful one to many cipher" may be helpful when trying to decode text, but will not be able to explain how any plain text, when encoded using this, will have the properties that Marco Ponzi already pointed out.

No one claimed that it, alone, explains the properties that MarcoP You are not allowed to view links. Register or Login to view..

I did not think that this was claimed. I rather thought that it had been ignored (or not fully understood - no bad intentions assumed).

I believe that the cipher method of Leonell Strong could be considered an example of such a cipher, though it certainly includes additional (and unrealistic) complications.

A criticism of this method included that it was not considered realistic that many different plaintext words would, by chance, result in the same cipher text word (e.g. Eva chedy).