[no subject]

["Anders, Claus" <canders@xxxxxxxxx>]

	 Matthew Skala wrote:
	 
	> As well as being prohibitive in terms of time (there are a *lot*
of
	 > possible permutations of an alphabet), this approach assumes that
the VMS
	 > *is* a simple letter-for-letter substitution cipher.  We have
strong
	> theoretical reasons to believe that it isn't - the statistical
properties
	 > of the text are in some ways quite different from those of text
in a known
	 > language encrypted by simple substitution.  If the VMS were a
simple
	> substitution cipher, there would be much easier ways to break it
than
	> brute-force anyway.

	Maybe it's a simple     letter-for-letter substitution cipher with a
substution depending on the position of the character within the word, for
example:
	Text: 
		This Book is written in chinese 
	gives 
		Tikv Bpqn it wskwxjt io cikqixk ...
	where I simply replace the nth char. in the word with same letter+n
in the alphabet (with an one to one char mapping). With this approach I get
produce the same code for the same word but double characters in the
original text are code to different chars and a same character  sequence in
the code was coded from different char in the org. text.
	However this coding scheme cannot explain the abundance of single or
double char "words" in the VMS.
	Cheers 

	Claus  

===================================
Claus Anders

debis Systemhaus GEI
Pascalstr. 8
52076 Aachen, Germany

phone:(+49) 2408/943-781          Fax: -430  
mailto:CAnders@xxxxxxxxx
===================================