Re: VMs: Re: Quick question

["Jeff" <jeff@xxxxxxxxxxxxxxxxxxxxxx>]

Rene Zandbergen r_zandbergen@xxxxxxxxx wrote
on 05 January 2004 16:18
Subject: Re: VMs: Re: Quick question


> Dan Gibson wrote:
>
> > Your  question about left-right text is very good.
> > There are MANY things
> > about the VMS that hint that the base language was
> > not European, but Middle
> > Eastern in origin. The artwork, however points to a
> > Medieval origin. Unless,
> > as some suspect, the manuscript ?? Middle Eastern,
> > but illustrated or
> > re-illustrated by a European scribe at a later date.
>
> I have always loved that scene in "the Name of the
> Rose" where the young chap is completely confused
> by a book that doesn't contain words but 'worms
> and fly poops'. His first encounter with the Arabic
> script. I don't remember if that was the
> movie (English) or the book (Italian).
>
> Many important sources in the Middle Ages would
> have been written in various langauges using the
> Arabic (orother) script. One good way of making
> a hoax (or a clever cipher) would be to take
> such a text and 'transcribe' it. The change of
> writing direction (if needed) does not worry me
> at all.
>
> > Has anyone worked out a frequency table for the VMS
> > using EVA, and has this
> > frequency table been applied to as many languages as
> > possible, checking for
> > mirror as well as normal layout?
>
> Most simple statistics that have been produced
> about the VMs text are independent of the writing
> direction. In particular the often published,
> anomalous entropy values are.
>
> The difficulty in checking the VMs (in my opinion,
> or rather "in my case") are obtaining usable
> source texts for comparison in electronic form.
>
> Cheers, Rene
>

One idea would be to use the triplet idea to count the occurances of unique
triplets, both forwards and backwards in the subject languages. This will
show peaks and troughs and how regularly they occur. Each language should
have its own distinct pattern over a large enough sample. As I showed in a
previous email. Italian has an undulating characteristic, whereas the VMS
has dramatic spiking. This dramatic spiking, I feel would hinder direct
comparison. It is as if certain parts of the data are compressed as has been
suggested by Nick Pelling. That is if the triplet idea is a no go.

Just a note for Nick. Would a full run of triplets showing spike patterns be
of any help in determining where any abbreviations are located?

The other way to look at it of course is that the flattened areas may be the
abbreviations and the spikes may be data expansion in a verbose manner.

What an intriguing thought!

Jeff


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