What are the characteristics of Labelese?

[-JKP-]

Hyde & Rugg's approach and Torsten Timm's approach are almost completely opposite. It worries me when I see people lump them together.


Hyde & Rugg came up with a grille theory (I don't know whose idea it was originally, but the example on the Web was posted under both their names) and applied it to VMS text. They did NOT really study the text. Their grille example is fundamentally flawed because it assumes a certain structure that is not there. I won't discuss the specific reasons here, because it's a long conversation about how Voynichese works, but the way they divided up the parts of the tokens is incorrect not just in details, but in generalities.


Torsten Timm, on the other hand is LOOKING at how VMS text is constructed and making an effort to describe it. He is looking at patterns, he is looking at similarity, he is looking at proximity (Hyde & Rugg never even addressed the issue of proximity).

Timm' algorithms may not generate perfect Voynichese, but he IS studying the patterns and their possible interdependencies and Hyde & Rugg simply did not do that. The grille idea is logically flawed and does not demonstrate knowledge of the actual patterns that occur.

Timm's description of the text is in a completely different ballpark from the Hyde & Rugg experiment.