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

[nickpelling]

I suspect that deletions are also a tad problematic. Why would a hypothetical autocopyist often replace word-final -dy with word-final -y (which surely to anyone's eyes amounts to a deletion) but almost never with word-final -d? There's no obvious secondary stroke harmony restriction here, so I guess this constraint must be an additional (and arbitrary) rule that limits what a hypothetical autocopyist can do.

Additions, too: for even though a word-final -y can be replaced with a word-final -dy in the autocopying schema, presumably there's some other additional rule that prohibits word-initial d- from being replaced by word-iniitial dy- ?

The github source has yet other special cases for page-initial gallows, line-initial letters, and line-final letters (e.g. -m). So from my perspective, the autocopying hypothesis doesn't travel particularly light: it needs a lot of assumptive baggage to help it shape its output closer to what we actually see.