Re: [Indic-computing-devel] Devanagari Unicode encodings - revision

[Dr. U.B. P. <pav...@vi...>]

> From:           	Hariram Pansari <hrp...@ya...>
> Devanagari Unicode encodings are based on ISCII-91
> encodings. This works on 2 tier system, Encodings -
> and glyphs of OTF.
> 
> "Devanagari lipi ka itihas", if searched, a best
> solution may come out, for making the encodings to a 
> SINGLE TIRE SYSTEM, i.e. encodings=fonts, with perfect
> sorting order which will be best fit for Devanagari
> Databases operations also.

Can you please elaborate a little more on this? How are you 
going to fulfil the requirements of almost 15,000 individual 
glyphs (=encodings, according to you) if we have to consider one 
glyph for every possible combination. Sure, you might have 
thought about this and might be having a solution. Or I might 
not have fully understood your approach. Please elaborate.
 
> Solve of many problems faced now, like, unscientific
> sorting order, duplicate/multiple forms/sequences of
> letters/conjucts/matras etc,  could also be get
> easily.

The biggest failure of ISCII is in sorting. It has only one 
table catering both for the code chart and for sorting. Unicode 
has takena better approach. It has different tables for 
character code and sorting.

Have you taken a look at the Unicode collation algorithm and the 
colation tables (www.unicode.org/tr10)?

Rgds,
Pavanaja
-----------------------------------------------------
Dr. U.B. Pavanaja
Editor, Vishva Kannada
World's first Internet magazine in Kannada
http://www.vishvakannada.com/

Note: I don't worry about pselling mixtakes