Re: U+0BA3, U+0BA9

From: Kenneth Whistler (kenw@sybase.com)
Date: Fri Oct 24 2003 - 14:05:01 CST


Peter Jacobi asked:

> Can someone clarify the status of
> U+0BA3 TAMIL LETTER NNA and
> U+0BA9 TAMIL LETTER NNNA
>
> Comparing the glyph shapes with TSCII character tables
> it is quite clear that U+0BA3 is NNNA and U+0BA9 is NNA.
>
> This makes also a lot of sense for non-speakers of Tamil,
> because it correctly correlates the number of 'N's
> to the number of loops in the glyph.

Correlating the number of loops with the number of N's in
the name is irrelevant.

>
> I also got agreement from a local expert.
>
> Now I'm wondering, why is there no Unicode Erratum on this?

Because it is not an error. Those names have been thus since
1991 in Unicode 1.0, and have been thus since 1993 in 10646.

>
> Is it just because -for stability- character names are not allowed
> to change even if incorrect? Or is it an oversight?

Neither.

U+0BA3 TAMIL LETTER NNA is the retroflex n, usually transliterated
as n-underdot <U+006E, U+0323>.

U+0BA9 TAMIL LETTER NNNA is the distinct alveolar n, usually
transliterated as n-macronbelow <U+006E, U+0331>.

The 10646 naming conventions, which are stuck with A-Z for
transliteration, generally use doubled letters to indicate
retroflex consonants, particular for Indic languages. When
a third distinction needs to be made, as for Tamil, the
third name occasionally just gets a tripled letter, as is
the case for U+0BA9.

TSCII naming conventions may differ.

--Ken



This archive was generated by hypermail 2.1.5 : Thu Jan 18 2007 - 15:54:24 CST