[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Sukhotin's algorithm etc

I thought that this is worth sharing with the list.

On 20 Jan 2001, at 9:07, Mark Perakh wrote:
> Gabriel, in your list five VMS characters which supposedly are vowels are
> shown in green, and one in white. Those five which are in green I found in
> the conversion table on Rene's site, but the white character I did not
> found  in that table. Characters that correspond to R and T in Currier's
> alphab have both some similarity to that white character, but not exactly.
> ???  Mark

The story goes like this. Currier created an alphabet that does not 
cover all the characters in the vms. EVA does :-), so I converted 
from EVA to Currier and those EVA-chars which do not exist in 
Currier's (which is written in capitals) were left in EVA (which is in 
lower case).
So the character in white similar to a @ (EVA <u>) is not present 
in Currier alphabet's, but it is still present in the ms. 
EVA u seems like a squashed EVA <on> or even <an>, so it may 
not be a character on its own, perhaps just a funny ending <a> 
since it appears at the end of words
- (this reminds me that at primary school I was taught to write a 
different shape of z (extending below the line) when it is word final. I 
do not know if this silly detail is taught anywhere else)-

Anyway, <u> appears about 6 times, always word ending:

<f11v.4> ...-ctho.tchey.tu
<f35v.6>    shol.tcheey.chkcheeu.chcthaiin
<f71r.C3>   ...okeeody.oteey.chekeu{?}.okeol
<f86r6.C7>   {00:00}okas.epar.chir,u.oteo,t[o|e]idy...
<f89v1.5>  ....cheo,kchey.qokoiin.du
<f105v.11>     roees.aiiin.ol.okaiin.os.aiin.chckhodu.

Moreover, the Z of the FSG alphabet recognised by Sukhotin's 
algorithm cannot be a vowel, because Z is a suffix to indicate a 
complex gallow (in EVA) c*h, where * is one of the 4 gallows. At 
most, the Z could be thought as an overimposed <ch> over the 
gallow, but in any case, it is not sequentially correct.

I hope this helps.