How quickly did the VMS leave the possession of those who understood it?

[rikforto]

Many changes in the character statistics and in the character shapes at the same time between A and B may be explained by using a different encoding with two distinct characters sets (typical A t and B t as two different codes) instead of saying that these are two different scribes that use two slightly different languages or codes for some reason.

The variation is much more extreme than that. The formation of the calligraphic building blocks and the distance between them varies quite significantly and predictably between Currier A and Currier B. For instance, we can be reasonably sure, based on how other medieval scripts were conceived, that a minim in Currier A is the same conceptual stroke as in Currier B, yet they vary predictably from each other. The same is true of the distance between letters, weights on the strokes, size of the strokes, qualities of lines and paragraphs like slope and break size, and a few other qualities that do not usually distinguish letters within a script. In this way the script is not "totally unknown"; it fits, albeit somewhat mysteriously, into the known craft of the time.

The VMS being what it is, I cannot totally rule out a single scribe (including a forger) having deliberately manipulated their letter formation to some end. But they are doing so in a way that looks an awful lot like standard handwriting variation in a group of distinct scribes, and Occam's razor cuts hard here, I think.