Re: VMs: Playfair cipher and triplets

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi Jim,

At 20:25 19/01/2004 -0800, Jim Gillogly wrote:
> The statistics aren't anything like a polyalphabetic or digraphic
> cipher... at least not using the things that look like letters to
> be the unit of the cipher.

I've tried out several hundred types of the kind of pairings/groupings one 
might expect for a verbose cipher, and have yet to see any resulting stats 
even remotely like those of a progressive-key polyalphabetic or a 
Porta-style digraphic cipher at any level - in fact, there's structure 
almost everywhere you look. :-o

BTW: from Alberti's codewheel-change every few words, I infer that the "you 
need roughly 30 letters to break a monoalphabetic cipher" rule must have 
been reasonably well-known circa 1470: and if you accept (from the 
historical evidence) that this is the VMs' rough date of production, the 
only reasonable inference is that the VMs (if some kind of cipher) must 
involve some kind of cipher change every few words.

The only problem here is that this is what Steve Ekwall has been saying all 
along. :-)

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


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