Re: VMs: Re: entropy, shorthand systems, gallows characters

["Larry Roux" <LRoux@xxxxxxx>]

"the "s" in "sh", isn't an "s" at all"  - Correct!!  I have seen
examples where the sh is really ch with something above.  A better
representation might be c'h than sh.

This is the problem with trying to derive statistics and entropy
studies.  Without knowing the glyphs it is hard to determine what to
feed into the calculations.  Bad input bad output.

Lastly, shorthand also has glyphs for words.  While the symbol for
"this" and "and" are /sort of/ shortened forms, they really are symbols
for the words themselves.  Gregg shorthand has motions for "th" "sh"
etc.

So:  

"This is an apple"

may look like:
"[th]s s n apl"

I would think that shorthand would raise the entropy (that is, lowering
the ability to predict the next letter) because you are in effect
throwing out the rules of the particular language to increase speed. 
"n" may be in, on, or an.   Some efficiency is lost as one must
understand the context of a word within a sentence to de-shorthand it. 
It is doubtful that any two people writing shorthand would write the
same text in the same way too.  Heck, the same person might write the
text differently at different times.  






Larry Roux
Syracuse University
lroux@xxxxxxx


>>> nickpelling@xxxxxxxxxxxxxxx 3/28/2005 4:13:30 PM >>>
Hi everyone,

At 11:57 28/03/2005 -0500, Wayne Durden wrote:
>In short, I am not yet convinced entropy studies related to shorthand
>systems for speed to keep up with normal speech would apply to a
method
>designed to capture as accurately as possible the soundstream where
speed of
>transcription is not a factor, and the latter method might make most
sense
>if the scribe doesn't speak well the language he is capturing.

When looking at the VMs with shorthand in mind, don't forget to take
into 
account the various strategies, such as notae (extra symbols),
tachygraphy 
(symbols that are quick to write), phonetic rendering (f for ph, etc),

contraction (missing internal letters), abbreviation (missing terminal

letters), etc. Rngs r nt th nly frt! :-)

>Also, at one point in my notes I captured some discussion related to
the
>label with the Pisces illustration but I didn't capture the source
enabling
>me to go back to it and can't find it via Google.  It may have been
Pelling
>as I have that written down in proximity?  The gist of the comment was
that
>the label was too short to have make much sense in the context with
Pisces.

Googling for "otolal" will probably retrieve most of the posts you're 
looking for. :-)

FWIW, this word is almost an exemplar of what I imagine a verbose
cipher 
would look like, where "ot-ol-al" would most likely represent a set of

consonants (like PCS) expanded out into pairs. Other Voynichese letters
(in 
particular <s>) rarely appear in pairs (so, I think "os" pairs are
probably 
"or" pairs that have been incorrectly corrected: and the "s" in "sh",
isn't 
an "s" at all).

Cheers, .....Nick Pelling..... 


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