--- In qalam@yahoogroups.com, "suzmccarth" <suzmccarth@...> wrote:

>
> --- In qalam@yahoogroups.com, "Richard Wordingham"
> <richard.wordingham@...> wrote:
> >
> > --- In qalam@yahoogroups.com, "suzmccarth" <suzmccarth@...> wrote:
> > > Me too. Almost as depressing as reading the Uni... well you
> know
> > > what I mean to say but I don't want to be censored.
> >
> > The appropriate sanction would be for you to be forced to submit
> > corrections to Unicode!
>
> I tried in this post!
>
> http://groups.yahoo.com/group/qalam/message/6127

Say you're trying to prepare a coherent change proposal for submission
via http://www.unicode.org/reporting.html . That's why I am trying to
understand if there is a non-geographic unity to 'SE Asian scripts'.

> > On the 'Unicode and abugida' saga, the real problem is that the
> > concepts of abjad and abugida are almost irrelevant to what Unicode
> > needs to consider.
>
> That is a point - a map and an historical overview of national and
> political boundaries would be more useful. Then a timeline, with a
> list of available technologies by date.
>
> Anyway, I have been trying to sort out all the misnomers for
> Syllabics for another project and it led back to the "featural
> syllabary" term in Unicode. I wasn't looking for trouble, Richard,
> believe it or not.
>
> >The differences between an alphabet, a pure abjad
> > and a pure syllabary are immaterial for Unicode.
>
> True.
>
> > Implicit vowels are chiefly significant because ISCII coalesced the
> > two vowel stripping mechanisms (stacking and virama) - a neat idea
> > that has caused endless confusion.
>
> Yes, this has been very confusing for me.
>
> > One thing that is troubling me is the division of Indic scripts.
> > Can
> > the division between South Asian and South East Asian scripts be
> > anything but a geographical division.
>
> Legacy standards?

That would apply to the politics of Unicode. Thai could resist the
phonetically ordering ISCII principle and stick to its visual order.
Also, Thai logical order would be a disaster, for the same reason that
the way Thai has eliminated the consonant stack is a mess. (Splitting
Thai words into syllables - even approximately - is an art, and I
haven't heard of it being captured by an expert system. It fails
abysmally without phinthu ('Pali virama'), which is not used for the
Thai language, though it is used for Thai phonetic spellings as well
as for Pali.)

The price Thai pays for this victory is that the alphabetic sort in
Excel currently fails abysmally. Excel treats the Thai preposed
vowels like Devanagari independent vowels! (The sort works well
enough in MS Word!)

Tamil seems to be sticking out, but the ISCII model is workable for
Tamil. It's just a lot of unnecessary overhead.

> or is there a difference because of the difference in language
> families?

There are two historical differences. There's a division between
North Indian and South Indian scripts, and a difference between Indic
languages and the rest. There are also hiddden issues because
Devanagari was taken as the Indic language exemplar - Bengali might
have been better.

I'm not sure that Bengali is closely related to Devanagari - Bengali
has <e> on the left and split vowels for <o> and <au>. In these
respects it is like the South Indian scripts.

There is a language issue, in that the principle of a default vowel
only works (worked?) well for the Indic languages. In Pali the
orthographic syllable is quite similar to the phonetic one, especially
if one uses anusvara for homorganic nasals. Many non-Indic languages
have a tendency to avoid phonetic syllable splits within the consonant
stack. In that respect Burmese is quite similar to Tamil. However,
Burmese needs consonant stacks for its complex syllable onsets, and
tolerates them elsewhere in loanwords. (By contrast, Thai doesn't.
Pali in the Thai script requires that phonetic syllables be made up of
entire orthographic syllables.)

Other Indo-Chinese scripts seem to manage fairly well despite hardly
using the virama. It's obsolescent in Khmer and I haven't yet
confidently found it in the Thai version of the Lanna script, yet used
very heavily in Burmese.

> That is why Tamil is so different
> from other Indic scripts. It is from the same *script* family
> historically but there are so many conceptual differences. The
> concept of how the writing system works seems to vary from the Indic
> language family considerably.

Not really. There is a problem in that the Tolkappiyam does not
relate to the Indic script scheme, but that is about it. The main
points are that having largely simplified the system by eliminating
most conjuncts, it is then complicated by consonant-vowel ligatures,
and the handling of stacks is trivial in Tamil.

I wasn't joking when I suggested that visual Tamil could be handled as
a variant of Thai.

This doesn't help my search for a concept unifying SE Asian scripts.

Richard.