Re: VMs: Feedback on: Are there samples of dotless Arabic in existence?

[Jacques Guy <jguy@xxxxxxxxxxxxxxxx>]

3/9/03 2:02:11 AM, "Don Latham" <djl@xxxxxxxxxxx> wrote:

>But suppose the dots were leading or following characters? entropy
>flattened, but word length perhaps longer.

Hmmm... yes, but not by much, I think (the entropy). There are 6 
possibilities in Arabic:

1. one dot above
2. two dots above
3. three dots above
4. one dot below
5. two dots below
6. no dots.

I imagine Sukhotin's algorithm would give the dots as vowels.
But it clearly gives <o> as a vowel. So the dots would have 
to precede the ... "ur-consonant", for want of a better word.

If that is so, the "ur-consonant" for R/Z would only ever
be preceded by the "one-dot-above vowel" (when Z), and
by nothing, or the "no-dot vowel" when D.

Likewise the ur-consonant "J,H,KH" would only be preceded
by "one dot below" (J), "one dot above" (KH) or "no dot" (H).

A dead giveaway would be the "ur-consonant" F/Q which would
_always_ be preceded by something: "one dot above" (F) or
"two dots above" (Q). Unless, of course the author had
decided to do away with "one dot above" (or two) since this
"ur-consonant" is always written with one or two dots.



>Also, wrt the astronomy section,
>if the ms were after 1580 or so, would not the great supernova event of the
>16th century be in there somewhere?

I thought it was only recorded in the Chinese annals?



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