Re: New Voynich Stuff

[Jason Morningstar <mornj@xxxxxxxxxxx>]

Hi All, 

I've posted to the list several times, and have been following the various
threads for quite a while.  The "polysyllabic information retrieval" stuff
is mine, from an IR course I took earlier this year.  I'm currently
working on a literature review related to the VMS, and intend to make it,
in some fashion, the focus of my thesis.  

I would tend to discount the content of those pages, which I generated
before I had seen some of the EVMT low entropy stuff.  

Best Regards, 

Jason

----------
Jason Morningstar
School of Information and Library Science
UNC Chapel Hill


On Sat, 28 Oct 2000, Dennis wrote:

> 	Any opinions on the "genuinely worthwhile stuff"? 
> 
> 	There' a lot to consider at Tugba's site.  
> 
> 	I'm not sure what Joyce means by "polysyllabic", but
> it sounds like the natural languages with low phonemic
> entropy that I & Gabriel considered in "Understanding
> the Second-Order Entropies of Voynich Text".  Her note,
> "To date, no one has attempted to use information
> retrieval formulae to investigate the possibility of an
> unconventional (non-western) underlying structure" is
> of more interest.  I wonder what information retrieval
> formulae she has in mind.  
> 
> 	On this page which I didn't notice:
> 
> http://ils.unc.edu/%7Emornj/inls172-01.htm
> 
> she proposes to poll us list members on our information
> needs.  
> 
> Dennis
> 
> Dennis wrote:
> 
> >         First the genuinely worthwhile:
> > 
> > -----------------------------------------------------------------------------------------
> > A very interesting site:
> > 
> > Statistical Analysis of Voynich Manuscript
> > http://www.cs.jhu.edu/~tugba/voynich/
> > 
> > by Tugba Onal Suzek,
> > http://www.cs.jhu.edu/~tugba/
> > Grad Student at John's Hopkins
> > 
> >         Quite a few statistical analyses, some of which we
> > haven't done, such as beam search, and languages we
> > haven't investigated, such as Czech, Hungarian,
> > Slovene, Estonian, and Romanian.
> > 
> > Another interesting site:
> > Univariate Terms Project
> > INLS 172, Jason Morningstar
> > 
> > http://www.ils.unc.edu/~mornj/inls172-02.htm
> > 
> >     "To date, no one has attempted to use information
> > retrieval
> > formulae to investigate the possibility of an
> > unconventional (non-
> > western) underlying structure. It is possible that
> > measures derived
> > from this investigation will allow researchers to
> > dismiss this avenue
> > of approach altogether. Alternately, the results might
> > signal a need
> > for further research. No matter what the outcome, the
> > groundwork laid
> > by this project will prove useful for future research
> > into the
> > possibility of an logographic source.
> > 
> >     "It is important to note that a correlation with a
> > polysyllabic
> > language model does not imply that the manuscript was
> > necessarily
> > written in Hawaiian or Yupik - given the commonly
> > understood
> > provenance of the Voynich manuscript, this would be
> > impossible.
> > However, the 14th through 16th centuries saw the
> > creation of many
> > artificial languages in Europe, including those of John
> > Dee and
> > Hildegard Von Bingen. It is possible that the
> > underlying Voynich text
> > is similarly constructed, but using a novel
> > polysyllabic structure."
>