Re: VMs: Gallows mega-theory...

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi Rene,

At 06:34 16/12/2003 -0800, Rene Zandbergen wrote:
> Some things which any theory will need to explain:
> = Eva-k and Eva-t can usually be interchanged, in
>   the sense that a swap will usually lead to another
>   valid VMs word.

Good so far - in fact, this would be a strong prediction of the GMT 
(gallows mega-theory).

> = Eva-f and Eva-p can usually not be changed into
>   Eva-k or Eva-f, then leading to a valid word.

Also good so far, as <f> and <p> would *define* a sequentialised Neal key 
(which would be quite different from *using* such a key).

> = ckh and cth look like k and t written 'through'
>   a ch symbol. It is not automatically clear that
>   there should be a relation between them, but it
>   is interesting to note that cth and ckh can be
>   interchanged in the same manner as t and k.

As yet, I'm far from sure what the relationship between t/k and cth/ckh is: 
but (as you point out) it is very suggestive that both pairs share the same 
kind of use-symmetry. Perhaps it will become a little clearer as the GMT 
evolves... fingers crossed. But at the very least, ckh/cth seem to parallel 
the mechanisms I suggest (rather than to compete or to conflict with them), 
so that's not a bad starting point. :-)

> Another thing you can test is whether the type
> of multiple-alphabet cipher you propose can turn
> a regular (say) Italian or Latin text into something
> with the patterns of VMs. My opinion is that this
> is more or less the opposite of what one would expect
> to see (and the opposite of the aim of early
> encryption systems) so this will be a very hard test.

I really don't think the VMs' plaintext will prove to be a regular Italian 
or Latin text, but rather some kind of semi-systematic / semi-adaptive 
shorthand system. However, (despite my efforts to date) this remains much 
more vague than I'd like - tying this down more accurately should make the 
task much easier.

If the VMs is encoded, then it should be reasonably clear (from its 
internal structure and word formation) that it is encoded in a way which 
hasn't been encountered - yet, to be reasonably consistent with its 
apparent dating evidence, we should expect its complexity to arise from the 
combination of multiple simple ciphers (rather than from a single 
computationally complex cipher).

> > What do you think? :-)
>
> To be honest, I don't think so :-/
> To be even more honest, I can't come up with
> anything more promising myself right now....

Fair enough. AFA I'm concerned, my GMT really isn't intended to be the last 
word on the VMs... but it might just (with a bit of luck) turn out to be a 
good step towards the solution. At least it directly accounts for many of 
the VMs' unusual features (rather than pushing them to one side)! :-)

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


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