Re: VMs: Noise or data ?

[Nick Pelling <incoming@xxxxxxxxxxxxxxxxx>]

Hi everyone,

At 22:07 20/05/2003 +0200, Petr Kazil wrote:
> I find the current discussion quite fruitful.
> I think it's the first time someone said:
>
> (e) It's a one-way unbreakable encryption.
>
> Somehow I find this idea both feasible and appealing.

An unbreakable code shouldn't need hiding - and the pair-like nature of the 
VMS' cipherbet makes me *extraordinarily* suspicious that something is 
being hidden there. So, I think this is a good alternative theory (thanks 
Graham)... but I guess it's probably wrong. :-)

Instead, I suspect that the major analytical effort in constructing the VMS 
may well have gone into *hiding* its core code(s). Whatever they/it turn/s 
out to be, I predict we'll find them wrapped in a number of steganographic 
mechanisms - but our stats-driven cryptological toolbox isn't really geared 
up for breaking steganography (by definition, if it can't see it, it can't 
break it).

I suppose what I'm trying to say is:-

(f) It's a very well-hidden coding system - and we're simply not used to 
looking for that kind of thing.

(Though - strictly speaking - my original (a)-(d) related to why the *idea* 
appears never to have spread beyond this one manuscript). :-)

So: data at the correct level of analysis, noise at the wrong level of 
analysis! :-)

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


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