Re: VMs: Follow-up

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi everyone,

At 16:01 21/06/2004 -0400, "surfing t" wrote:
> I agree that we have to try something new. But on the other hand, wouldn't 
> the NSA cryptographers have realised the possibility of a verbose cipher?

"Verbose cipher" is just what we call it here: but note that we typically 
use it to describe what might be better called a "mixed verbose cipher" 
(ie, a cipher where both single characters and paired characters in the 
cipherbet stream represent single tokens in the plaintext stream).

Most of our analytical tools rely on there being some kind of linear 
relationship between plaintext characters and ciphertext characters - yet 
crucially, a mixed verbose cipher would undermine that basic assumption.

However, I don't recall seeing any cryptography / cryptology book or paper 
describing a systematic way to break a mixed verbose cipher - perhaps some 
of our more hardcore crypto listmembers might care to comment?

> Another interesting feature: On f75v you see two lines of text labels. The 
> two labels in each column appear to have the same number of glyphs at 
> least most of them appear to (although I'm not fluent in this script yet!) 
> - if it is true for almost all of the labels in that illustration, but not 
> for a few, maybe we are dividing the glyphs in a wrong way (if in fact 
> there is a system). What are your thoughts on this? Are the labels on 
> figures included in (some of?) the transcription and how are they counted?

Labels are typically labelled as "L" in the .evt transcription file. They 
don't really seem representative of the text, and I'd typically be happier 
with results where they are filtered out. Still, even within such short 
pieces of text, you might find it illuminating to see verbose-cipher-like 
substrings, like "qo" "or" "ol" "dy" etc. :-)

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

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