[Harfbuzz-indic] Unicode vs OpenType: nukta after vowel

[Jonathan Kew]

On 28 Jun 2011, at 21:37, Chapman, Christopher wrote:

> Bernard Massot wrote:
>> I personally agree with Unicode. I've never seen a nukta following a
>> vowel, and linguistics books I've read on Indian scripts don't speak of
>> that. Can you think of any real world use of such a thing?
> 
> Yes, sort of.  :-)
> 
> Take a look at the paper, "Recent Adaptions of the Devanagari Script for the Tibeto-Burman Languages of Nepal" by Michael Noonan, available here:
> https://pantherfile.uwm.edu/noonan/www/Recent%20Adaptions%20of%20the%20Devanagari%20Script%20for%20the.pdf
> 
> On page 15 there's an example of a "subscript dot" (which is thus arguably a nukta [remember that "nukta" just means "point" or "dot"]) being used to modify the vowel signs for use in the Limbu language. The dot appears below the consonant "ka" in the example although in a different position from where it would typically go when changing "ka" to "qa" in Hindi.  Perhaps this could be represented in Unicode as a U+0323 COMBINING DOT BELOW rather than a U+093C DEVANAGARI SIGN NUKTA, but I mention it because it is a NUKTA-like below-base dot that is modifying the vowel sign, not the consonant.

For minority-language uses of Devanagari script, I've also seen it used below one or more of the spacing vowel signs - I don't have examples on hand, it could have been any (or all) of AA, II, O, etc.

Users think of this as being "the same nukta sign" as they use to modify consonants; they're just using the script's well-known "modifier" symbol in new ways to express additional modifications.

JK