Discussion of "A possible generating algorithm of the Voynich manuscript"

[ReneZ]

Torsten,

I have no reason to doubt your statistics so I trust they are correct.

I do find the auto-copying hypothesis not sufficient as a method for constructing the Voynich MS text. Not every word in the MS is a copy with edit distance 0 or 1 from a recent word, for any reasonable definition of 'recent'. The percentage of such words for different definitions of 'recent' could be computed but this has not been done. What if more than 50% of words is not a copy with edit distance 0 or 1 from within the last 10 lines? Then more than half the text is not explained.


My biggest problem is with the conclusion. It does not follow that the correlation between edit distance and distance in the text is evidence that the text is meaningless.

Anyway, the whole discussion about an alternative experiment to simulate a high level graph as given in figure 4 is at best a sidetrack.

This is not a sidetrack. It is an example how a meaningful text can also exhibit this unusual property, and demonstrates that the conclusion is not valid.

It is of course perfectly fine to be of the opinion that the text is meaningless, but this does not follow from the evidence presented.