VMs: RE: All quiet on the VMS front

["GC" <glenclaston@xxxxxxxxx>]

> 	Glen, since you see in, iin, iiin and I have to
> assume ir, iir, iiir
> etc...,
> then why not extend that same encoding as separate
> characters to the 'c'
> family
> c8 cc8 ccc8, cs, ccs, cccs are quite common, although
> the cccb is fairly
> rare.

Actually you're following my track quite well.  cc is many times
written as two characters, c c, but many times connected, cc.  ccs
is a connected character many times, but occasionally there is a
space as such, c cs.  Sometimes even c c s.  On f1v there's also
an i cs.  Part of the problem with a fully stroke-based
transcription is that while one can globally reduce all c c or c c
s to one character, they aren't always written that way.  A
transcription that includes connectivity, say c cs, can be
globally reduced or expanded appropriately without having to
retranscribe.  The separation/connectivity information is missing
in the EVA transcription.

I know we've been around the block on the table with the mark over
it, EVA sh.  We see from the very beginning of f1v that this mark
is a signal of some sort.  The only time it appears to the left
centered AND backward that I can find is on this page.  After
that, it appears in the middle most of the time, but sometimes in
a teardrop in the middle, or in the form of an s to the right or
left, and the number and careful location of these make them
clearly intentional markings.  I think the same might be true of
the scc or ccs, and once in awhile the mark is even over the
center.  These are the patterns that need to be recorded in a
transcription IMHO, and patterns that are missing in the EVA
transcription.  The classic "picnic table" is in my opinion
misnamed, because most of the time this character has feet on it
"walking" in one direction or another, yet this appears to be
encoded only with an 'x' in the EVA 'H' transcription, which means
the "directional feet" information is missing.  (I'm wondering if
there are four forms of this character as well.)

> 	The reason I've focused on the 'stroke' is simply
> that the patterns do
> exist
> for both 'i' and 'c' - if we're calling iiin a
> character - then ccc8 should
> also be a character by the same reasoning.

We've all seen these patterns and pondered them, but we've yet to
record them accurately and test any of our theories on them.  The
pattern of four characters to a set is a rational visualization,
but what are the numbers associated with this phenomena, how far
apart do they appear, etc.?  The only way to know these things is
to encode more information into our transcription scheme, and I
think the EVA authors must also have realized that some things are
simply units of character, otherwise we'd see the "a" character
transcribed "ei" in EVA, etc.

I'm not calling "iiin" a character all the time.  when the four
strokes are connected, I'd say it's a character.  If they're
distinctly separate, I'm not so sure.  All I want to do is encode
them as a unit when they clearly appear as a unit, and encode them
separately when they appear separately.  After that a few
statistical manipulations might be used to determine whether these
are always individual characters or are most probably individual
strokes.  We already know from just looking that 'in' and 'iin'
are so common they are most probably individual characters just
like the 'a' character, but does 'i in' appear once in awhile as
well?  But wait - one of the "key sequences" also has an 'aiin'
clearly laid out as an individual character, so how often does the
'ain' and 'aiin' sequences demonstrate connectivity and how often
do they not?  Is there an 'aiiin' connectivity or even some
connectivity between the 'an' sequences that might bring these
characters into our assumed grouping of four variations for each
character set?  I won't know positively or be able to validate my
suspicions until the book is retranscribed and the appropriate
information encoded.

Unfortunately, as much as I've tried to base a transcription on
EVA characters, what I'm attempting to do differs so much from
that transcription scheme that I have to come up with another
entirely new scheme.  Even my old transcription is not suitable to
this purpose, so out the window it goes!  To me there are no valid
numbers or statistics without a major attempt at answering the
"character" question, and I've been working around this problem so
long it's finally time to tackle this problem as thoroughly as
possible.  I'm starting with the pages we have high grade images
of, to get a feel of the text where I can see the connective
strokes on a 21" screen.  An accurate transcription of the entire
book will not be possible without clear images, but there are 54
or so now available, and I've ordered a few more, (f1r for
instance), so it would seem that with these high-res images now
available, the time is right to begin a definitive study of this
problem.

GC