[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: VMs: Voynich analysis

Hi Rene,

At 04:28 03/07/2003 -0700, Rene Zandbergen wrote: 
--- Nick Pelling <incoming@xxxxxxxxxxxxxxxxx> wrote:
> I think it would be a great step forward if you
> can even prove (or  disprove) that the word
> boundaries in the ciphertext correspond to word
> boundaries in the plaintext.

The idea that spaces might not be word boundaries
came up because the spaces always tend to follow
(and precede) certain characters with high
probability.

FWIW, I'm trying to reconstruct Cicco Simonetta's options circa 1465: he believes (from his own experience as a code-breaker) that mono-alphabetic ciphers are inherently crackable - but at the same time, doesn't have a huge wealth of alternate mechanisms to choose from.

What, then, is his #1 design aim? I suspect that, with the VMS, he explicitly intended to make the output code appear as much like existing codes - monoalphabetic ciphers, tachygraphic shorthand, and those few remaining Tironian notae still in use - as he could, so that code-breakers would attempt to break the code at the wrong level (and hence would fail)... essentially, to engage in a programme of *cipherbetic misdirection*.

I believe that he included a couple of word-terminating Tironian notae (EVA <d> and <y>) to help give the encoded text-stream the *semblance* of word structure - and so my specific prediction is that the chances of these actually coinciding with real word-terminating plaintext tokens are fairly minimal.But how to prove (or disprove) this?

It is not difficult to show that this qualitative
statement is actually incorrect (quantitatively).

Rather, it is not difficult to show that this qualitative statement is actually weak (quantitatively). Sure, there is some correlation, but at the same time there is weak correlation in lots of different directions - you can probably find (weak) statistical support for pretty much any vaguely plausible theory you devise.

However, there is also apparently some kind of complex rule-based system at the letter-to-letter coherency level, especially at the pair level (OKOKOK stuff). What can we infer from that?

Cheers, ....Nick Pelling..... 


______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list