Encoding of Text in the Myanmar Script

[Richard Wordingham]

On Fri, 28 Oct 2022 12:54:52 -0400
Wunna Ko <wunnakoko at gmail.com> wrote:

> Richard,
> 
> There is no official Standard in Myanmar. There is no official
> encoding as well. But the Burmese users generally accepted Unicode as
> a standard.
> 
> Not sure what you are referring to when you mention TUS encoding.

The TUS encoding for Modern Burmese is Table 16-4 in the Unicode
Standard, currently (Version 15.0) on p673, in Chapter 16.  Now, if one
applies those rules to the _Mon_ word generally transliterated as
<ruy> 'to choose', one gets <RA, SIGN AI, SIGN U>.  However if one
applies the Microsoft Typography rules, one gets <RA, SIGN U, SIGN AI>.
One renders properly, while the other gets a dashed circle inserted.
Safari on iPhone and Edge on Windows 10 render different ones properly.
My key question is, has the Government ruled which is correct?

There's a similar issue with the sequence <SIGN AA, SIGN AI> v. <SIGN
AI, SIGN AA>, but slightly different - Apple rejects the first, HarfBuzz
accepts both (or a font corrects the renderer), and I think UTN-11
Version 4 is meant to reject the second as it accepts the first, but it
is not mentioned in Footnote 3 to the table on pp6-7.  This is also a
nasty one, as it is not beyond the wit of man to ascribe different
renderings and even meanings to the two sequences. 

Richard.