VMs: Approaches to Reorganizing EVA For Analysis

[Koontz John E <John.Koontz@xxxxxxxxxxxx>]

For what it is worth, my discussion of my attempts to re-render EVA with
the (a)i* and (a)e* sequences treated as units and flourishes stripped
from their preceding bases and treated as separate units is at:

http://www.voynich.net/Arch/2004/09/msg00000.html
http://www.voynich.net/Arch/2004/09/msg00016.html

I believe that extended experiments of this sort have also been conducted
by Jorge Stolfi and John Grove.  No doubt others, too.

In the particular approach I report, I treated o as e + n-flourish, where
n-flourish is an entity representing the flourish that combines with EVA i
to make EVA n and (I assumed) with EVA e to make EVA b.  I've been meaning
to get back to this.

One thing I'd like to do is to try *not* matching the e- and i- attached
flourishes initially.  For example, dissolve EVA n into i and n and EVA b
into e and b, so that I'm not making the arbitrary shape-based assumption
that the n- and b-flourish are the same from the outset.  The problem with
shape-based assumptions is that the shapes are ambiguous.  Perhaps the
distributions will help me do the matches.  For my ideas on this, see
http://www.voynich.net/Arch/2004/09/msg00482.html.

Incidentally, treating the gallows as two-character sequences in
transcription would permit representing the oddities where other
characters are embedded within them.  This is already a principle in the
EVA transcription, since transcribing the space-delimited benches (ch and
sh) as two Latin characters in the EVA transcription is what allows
transcribing the more frequent cases where these bracket the gallows, cth,
sth, etc.

I'd also like to represent the "bar" ligature separately in analysis.
"Bar" is what connects "underlying" e e in EVA ch and "underlying" e +
s-flourish e in EVA sh.  Or putting it another way, EVA h is bar + e.
The problem with treating bar separately is that it appears that bars
occur in various contexts other than ch and sh.  These additional bars are
apparently represented in the files that Takeshi Takahashi uses to
generate the EVA-font page images at his Web site, but are missing in the
standard distribution of his transcription file (see comment at
http://www.voynich.net/Arch/2004/09/msg00418.html).  Apart from bar in h,
bar is otherwise represented in EVA transcriptions by capitalizing the
following character, and it appears that the standard TT EVA file
suppresses this capitalization.

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