Re: VMs: An idea

["Jeff" <jeff@xxxxxxxxxxxxxxxxxxxxxx>]

You misunderstand what I am attempting to do. The numbers will indicate
where the letters fall in relation to an attempt to decipher using the
cylinder type decipherment. I could have chosen letters but there aren't 128
of them. Call them tokens instead. Strings of numbers 1 to 3 digits long
will stand in place of the assumed letters. These will then be analysed for
entropy and whatever else can be used to see how the values change as
different attempts are made. It would be like having a language in a 128
letter alphabet. It would still show a pattern, though not consistent with
the plaintext language. As each group of values is counted it will either
show which tokens are likely to be vowels or consonants or it would quickly
become apparent that the method is wrong, as the statistics would not form a
normal curve.

If abbreviation is involved as NIck Pelling believes then we are lost anyway
and have to think of a more specific attack method. Also in nulls or even
codebook numbers are involved then the whole method assumes another level of
complexity altogether.

This would need to be done on a variety of transcription methods. I was
recently sent a file by Robert Teague which I am looking into at the moment.
This will form the basis of the attack.

Jeff



----- Original Message -----
From: "Gabriel Landini" <G.Landini@xxxxxxxxxx>
To: <vms-list@xxxxxxxxxxx>
Sent: 24 January 2004 10:26
Subject: Re: VMs: An idea


> On Saturday 24 Jan 2004 12:04 am, Jeff wrote:
> > I have thought about numbering the 8 by 16 grid with numbers from 1 to
128
> > when trying the rollong cylinder method. That way an analysis could be
done
> > using the numbers as strings for frequencies and also as a way of
plotting
> > the information graphically. Can anyone think of any other method that
> > might show up interesting patterns? Spectral analysis perhaps?
>
> No, assigning arbitrary numbers to symbols is precisely what one should
not
> do.
> Number 10 implies some quantity which is two times number 5, and it would
be
> incorrect to assign quantities to symbols because that quantity has not
> physical meaning.
> A=1, D=4. What is the logic behind claiming that D is 4 times A?
>
> Cheers,
>
> Gabriel
>
> ______________________________________________________________________
> To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
> unsubscribe vms-list
>

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