Hi Dennis,
I think there is a slight misunderstanding of my imagination:
It's not a substitution, I was thinking of an addition/overlay - glyph-like .
Imagine a constructed script, where certain important part are omitted.This parts are painted on translucent material, which will be put over the text for reading (ok, far fetched, but possible).
Hanzi/Kanji example (not VMS related, even if I do support the Chinese pizza theory somewhat):
Paint a lot of common radicals/strokes on a strip of translucent paper->template.
Write down your text in Hanzi and omitting all the strokes, you find on the template and add strokes, where you find the on the template, but not in your text.
Result: where there strokes both on paper and template:omit them.When there is a stroke only on the template: add it to the "character".IMO, with a little practise, you can read and write text quite fast.
It will maybe not work real good on Latin alphabets, but who says, the VMS script is Latin?
Again: even if good optical material didn't exist at VMS assumed creation time, sort of translucent material (thin oiled parchment for ex.) probably existed at that time.
Just some free spinning thoughts,
Cheers
Claus
(@nick: don't damn the VMS's author, where would be the whole fun? How many people did you learn to know by trying to solve the mystery?I never regretted, starting to look at the VMS)
-----Ursprüngliche Nachricht-----
Von: Dennis [mailto:tsalagi@xxxxxxxx]
Gesendet: Donnerstag, 31. Juli 2003 02:08
An: vms-list@xxxxxxxxxxx
Betreff: Re: VMs: Possible optical device
> "Anders, Claus" wrote:
>
> Each gallow
> will tell the reader, which side of the device is to
> used.
I think we've heard this idea, although not such
a device. This idea is that the VMs cipher is
polyalphabetic substitution, and the gallows letters
indicate where to change ciphers.
Strong, I think, proposed this idea in the
1940's but his solution was pretty clearly wrong, since
it
contained linguistic anachronisms.
In general, modern polyalphabetic substitution
ciphers wouldn't produce the character entropy profile
we see in the VMs, but they change
the cipher at each letter. If you only changed the
cipher
at each gallow and only had a few different ciphers
for the whole VMs, I'm not sure. I think the entropies
would
look more like a simple substitution cipher and closer
to
natural language, but that again is not like the VMs.
Of course,
that's assuming that the underlying natural language is
a
typical Western Indo-European one!
Dennis
--------------------------------
[This E-mail scanned for viruses]
______________________________________________________________________
To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying: unsubscribe vms-list