[HarfBuzz] Tai Tham NGA, SAKOT is not Kinzi

[Richard Wordingham]

On Thu, 2 May 2013 16:49:37 -0400
Ed Trager <ed.trager at gmail.com> wrote:

> I haven't looked at the แม่ฟ้าหลวง dictionary recently, but from what
> I recall, I was suspicious that the font, which I've seen used in a
> number of books about Tai Tham, had a number of limitations.

> In the case of Tai Tham, which has a strong manuscript tradition but
> not a long tradition of printed books, at least for regions like
> Northern Thailand, perhaps manuscripts provide stronger evidence
> (than do typeset books) for what "proper" or "conventional" Tai Tham
> orthography is supposed to look like?

Yes.  I much prefer hand-written sources.  It took me a while to work
out what was going on with apparent spellings such as <HIGH HA, MEDIAL
LA, NGA, SAKOT, WA> for _luang_ (Thai transliteration หลวง).
Handwritten text explained it - not only do superscript characters
invade the airspace of following base characters, but chains of
subscript characters also tunnel under following base characters.   One
thing I haven't got to grips with is the tendency for superscripts and
subscripts to be not so much aimed at base characters but aimed at
syllables.  Unfortunately, accuracy is then an issue, though it's better
than my aim when dotting i's.

It's not just manuscripts where there's a problem with internal
consistency.  The positioning of marks in the MFL is quite erratic, and
I'm still not entirely clear where mai sam should be written.

Richard.