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

[nickpelling]

This is the main difference I have with Torsten's description of the text and my perception of the text, and I'm glad you brought it up Nick, but I still like what Torsten is doing, regardless of whether I agree with an autocopying hypothesis.

I also kind of like what Torsten is doing, but at the same time I can't help but suspect that he has significantly oversold his autocopying hypothesis. In my opinion, the only reason that his generator produces even moderately Voynich-like output is that he has had to add in a large number of special cases and conditioning constraints, for which he has no obvious extrinsic justification beyond 'they seem to make things work better' (e.g. his 'ligatures' seem to me to fall exactly in this category).