VMs: Resending this...

[Elmar Vogt <elvogt@xxxxxxx>]

-------- Original Message --------
Subject: Resending this...
Date: Fri, 22 Apr 2005 17:49:05 +0200
From: Elmar Vogt <elvogt@xxxxxxx>
To: vms-list@xxxxxxxxxxx


(Silly me had tried to fire this message from an old, discontinued account
-- no wonder it wouldn't get through.

So, here are my thoughts on another possible solution for the VM (yawn...)
again.)

Hi all,

Since we all appeared to be a bit uneasy with the current transcription
systems lately, and we're all unsure over what constitutes a VM character
and what not, I had decided I'd set up my own transcription system in a whim:

http://www.beamends.de/hidden/voynich_st.pdf

The basic idea is that the Voynich characters are "decomposed" into a setof
strokes, and the _strokes_ are transcribed -- ie a character like "d" canbe
decomposed into a "c" and an "l", and we transcribe the "cl" combination.

A little while later it struck me -- was this the way the VM was
"enciphered"? (And this is, what my mail is really about!)

Could it be that the following happened:

We take a look at the latin alphabet (or whichever the source language for
the VM was), and notice that the letters are made up of a finite set of
graphical elements -- a vertical line like an "i" or an "l", for example,a
"c"-like arc, etc.

Now, we take our source text, and decompose it letter for letter into the
elements.

Let's assume we use "i" for a vertical stroke beginning at the letter
baseline, and an "l" for one starting somewhat lower. Let "-" denote a
horizontal stroke, and "n" an n-like vertical arc. Finally, let an "s" bean
"s".

Then, the plaintext word "decompose" would become something like --

"ci c- o nn lo o s c-", or "cic-onnloosc-" for short.

Now, replace each of the ciphertext elements with a funky-looking character
you made up -- pronto, you have an encoded text.

It is straightforward to note, that --

*) This encoding scheme would of course fool all simple attempts at
frequency analysis,

*) Words would become longer due to the encoding, and you would notice the
absence of very short words,

*) You would have a limited set of ciphertext "letters" which could follow
others,

*) And in general notice a higher entropy, because several ciphertext
letters are needed to encode one plaintext letter.

We all know these features of the VM.

OTOH, we also --

*) Would have still no clue to the significance of the gallows, and why
every sentence would begin with one,

*) Would have still no understanding why several letters are confined to
word beginnings/endings.

What do you think? Would it make sense to pursue this further?

Fire away.

Cheers,

    Elmar


--
Elmar Vogt / Königswarterstr. 18 / 90762 Fürth / GERMANY
elvogt@xxxxxxx / www.beamends.de / Tel.: (++49/0)911 - 31 52 58
Agilmar v. Sevelingen: CUR CANIS TESTIS SUIS LINGET? -- POTEST!

"Es gibt viele, die nicht reden, wenn sie verstummen sollten, und
andere, die nicht fragen, wenn sie geantwortet haben." (J.G.A. Galletti)

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