Re: VMs: Blanks: trying to summarize...

[Rene Zandbergen <r_zandbergen@xxxxxxxxx>]

--- "Maurizio M. Gavioli" <mmg@xxxxxxxxxxxxxxxxx>
wrote:

> At 12:20 05/03/2004 +0000, Nick Pelling wrote:

> > [N:] However, AIUI Voynichese words generally
> > have a very low instance count, 
> > which is hard to reconcile with their being part
of
> > a language (whether 
> > real or artificial).
> 
> [M:] (AIUI? If it is an acronim, I cannot
> resolve it).

I'm just guessing: 'as I understand it' ?

I don't think that this observation is correct.
The word frequency distribution of the VMs
follows a normal Zipf curve. Gabriel has looked at
this extensively. OTOH, I am not sure if this is
conclusive. I have in the past looked at the growth
of the number of distinct words (i.e. vocabulary
size) as a function of text length, and this did
not indicate an excessively high number of 
words with low frequency. One has to compare this
with plain texts to be sure.

> [M:] 2) I don't know if someone tried to count
> words only in portions of running 
> texts, excluding for example texts, labels and
> sequences in the figures 
> which might have a different nature.

The word frequency count in the zodiac labels
(the nymphs holding stars) is completely different 
from the main text. These label words are indeed
far more 'unique' than words in the main text (but 
some zodiac labels do occur more than once).

> > [N:] I agree that there must have been a good 
> > reason
> > for spaces to be retained, which would seem to
> > run counter to normal cipher practice: hence, I 
> > suspect that this points to an unusual kind of
> > plaintext, or to an unusual 
> > kind of cipher process (and possibly to both).

Retaining spaces to me does not seem like something
strange. They make complete sense if the MS is
in an unknown language, an invented language,
or just made to look like a strange language.

Do we have examples of cipher texts from the
relevant time period?

> [M:] Are you hinting as half-spaces as distinct
> from spaces? Two kinds of spaces 
> with potentially different meanings? I think we
> have enough problems with a 
> single kind of spaces!!

I don't like the term 'half spaces'. What I have
seen in the MS are 'uncertain spaces'.
If one were to make a histogram of the distances
between two adjacent letters in the MS, one should
expect to see two humps (maxima). One for the case
'no space in beteen' and one for the case 'space
in between'. 

If, between the two maxima the frequency goes down
to zero, one can alway decide perfectly whether
there is a space or not. In reality, this does
not happen. Now this does not mean that there are
half spaces. It just means that the distributions
of 'no space' and 'space' overlap.
If there were such a thing as 'half spaces',
we should see three maxima. 

Cheers, Rene

__________________________________
Do you Yahoo!?
Yahoo! Search - Find what you?re looking for faster
http://search.yahoo.com
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying:
unsubscribe vms-list