[XeTeX] Hyphenation in Transliterated Sanskrit

[Philip TAYLOR (Webmaster, Ret'd)]

Mojca Miklavec wrote:
> On Mon, Sep 12, 2011 at 12:09, Philip TAYLOR (Webmaster, Ret'd)
> <P.Taylor at rhul.ac.uk> wrote:
>> I wish I understood more about the "duplicate apostophe" problem, in order
>> to be able to offer a more directly relevant (and constructive) comment :
>> Google throws up nothing relevant.
> Users type ' (U+0027) and expect the proper apostrophe (U+2019) to
> show up in final PDF. Knuth just replaced the character (you cannot
> get U+0027 in pdfTeX, except in typewriter font). In XeTeX
> mapping=tex-text does that, but not all users use that one, so we need
> to support both variants.

OK, (sort of) understood.  But does a Unicode-aware user /really/ type  (U+0027)
[APOSTROPHE] if if he/she wants (U+0219) [ RIGHT SINGLE QUOTATION MARK]
other than through habit/laziness ?  I can quite see an ASCII-based TeX user doing
just that, but those that are capable of entering real Unicode must surely be aware
of the multiplicity of apostrophe-like characters [1] available to them, and be capable
of choosing the correct one, must they not ?

** Phil.
--------
[1] Including, but not restricted to, APOSTROPHE, RIGHT SINGLE QUOTATION MARK,
LEFT SINGLE QUOTATION MARK, PRIME, MODIFIER LETTER PRIME,
SINGLE HIGH-REVERSED-9 QUOTATION MARK, ...