[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: VMs: Proving Gibberish

Hi Marke,

At 10:15 17/08/2004 +0100, Marke Fincher wrote: 
From a pedantic point of view it will be impossible to prove something
is gibberish if it's creation involves a random element.  You will not
be able to prove that the random element is actually random and contains
no information.

However, if the process that created it was deterministic (like moving
a grill in a consistent pattern over a table of characters), you can (at
least in theory) reconstruct the process and by virtue of proving there
was no choice involved, show that there is no information contained in
the message.  Although you would have the further possibility that there
could be information in the underlying 'table of characters' !!!

If the creation process involved arbitrary human choice you will similarly
not be able to prove it is gibberish, although the evidence might be beyond
reasonable doubt.

The problem with the hoax hypothesis is that randomness (and especially sustained randomness) is hard for humans to do without leaving some kind of "fingerprint" - and as it is, Rugg's general idea is that the contents of the tables explain all the fingerprints we recognise, but used in such a clever way so as not to leave any clear fingerprints of its own.

In the general time period we're considering here (say, pre-1600), few had any understanding of the nature and structure of probability - the most notable was Girolamo Cardano. Even so, according to Peter Bernstein's (1996) "Against the Gods: The Remarkable Story of Risk", Cardano's work on dice wasn't actually published until 1663 (p.49), so limiting the field of possible VMs' creators to those that had a reasonably modern understanding of probability pre-1600, we may well be left with a short-list of one - Cardano himself.

AIUI, Rugg argues that, to get his postulated grille-and-table mechanism to produce what we see (but without any obvious mechanism-based fingerprints), the grille must be moved randomly (ie *inconsistently*) over the table - which brings us right back to the first problem, of how to produce randomness satisfactorily... which he does not really address in his treatment. Oh, well. :-o

BTW, Cardano was born in Milan in 1500, so perhaps there is (after all) actually one possible connection with the VMs that Gordon Rugg & I do happen to agree on. :-)

   By the way, is it just me or is there a high chance that
a VMS word is similar to one that is immediately above it??

We've probably all seen instances where this appears to be true (I was looking at the first para of f3r yesterday, where EVA <damo> appears twice about a line apart), but you'd need to construct a properly designed statistical test to decide how significant a feature this actually is.

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


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