[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: VMs: Faces at the roots
Hi everyone,
At 00:27 29/04/2005 +0100, Jeff Haley wrote:
To digress a little. Nick's idea of the verbose nature of the system is a
very com{pelling} one (sorry Nick I couldn't resist.) Joking aside, it does
hang together and brings out the groupings that occur most frequently.
This not only increases the information content
...(as a result of shortening the length of the text, of course)...
but when graphed it
produces a nuch smoother curve and eliminates the high end peaks of
EVA.
...(specifically by absorbing the deceptive fake vowels EVA <o> and <a>
into their verbose groupings)...
On the downside it can never be a smooth solution in and of itself
and must be part of a multi table cipher.
Well... I'd say that multiple tables are probably part of the remainder,
but not all. Transposition ciphers (like Neal keys, line-initial
characters, interleaved lines, etc) all seem to have their place as well.
Why? Because as Jorge Stolfi
has shown VMS words have a hilly structure. The mid word peak seems
to suggest this quite strongly. If true then the words that break this rule,
to my mind, show the scibe being particularly nasty in his choice of
substitutions.
This "mid word peak" is where the information is apparently concentrated:
but part of this could just as well be caused by an artificial method used
for inserting spaces, or some kind of space transposition cipher, either of
which would be nasty in a completely different way.
To get back to Wayne's point. I would be very interested in the results of
work on the labels as these will be the most problemmatic to resolve. The
stickiest points are the appearance of (EVA) ot & ok at word starts.
I have wondered many times what the difference between "t" and "ot" is... I
have a few ideas, but it's all a bit too vague at the moment. :-|
Cheers, .....Nick Pelling.....
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list