Re: Sanskrit nasalized L

On Sat, Jun 18, 2011 at 10:38 PM, Peter Constable
<petercon_at_microsoft.com> wrote:
> Candrabindu + visarga is not the same scenario as the one Andreas asked
> about. Yours will be fixed in future releases of Windows and Office.

This problem (of candrabindu + visarga causing problems) would not
arise in the first place if visarga (and spacing anusvaras in South
Indian scripts) were analysed as independent spacing characters
(GC=Lo) rather than as combining marks as I have previously pointed
out in L2/09-343 p 4 under the subheading "Whether to encode as
combining mark or not". (Further remarks on this are in L2/10-234 p
2.)

However, the Indic committee commented on this in L2/10-167:

"c. We recommend not accepting the request to change the general
category from Mc to Lo for
VEDIC SIGN ARDHAVISARGA, as well as for visarga, anusvara, and other
characters. This
change could significantly affect existing data and implementations.
However, a well‐formed
proposal that argues in favor of such a change, with a discussion of
the implications for current
data and implementations, would be welcome for consideration by the UTC."

And why I don't see anything specific in the UTC minutes, I think this
was approved as there is no note to the contrary.

I personally don't see how a change of GC=Mc to GC=Lo would have
adverse effects, but as I am not in the software industry I can't make
any authoritative comment on the implications of such a change.

-- 
Shriramana Sharma