From: p_j_anderson@volny.cz
Date: Thu Feb 18 2010 - 08:29:24 CST
The trick is to find the right level of abstraction for a script. That's
what I'm aiming for, not encoding sounds specifically. I'm using sound
as a guide to cross-compare regional letter repertoires, allowing for
sound changes.
Unlike Gaelic or Fraktur flavours of Latin, there are no additional letters
with different behaviour in regional repertoires, or different typographical
conventions, or use of certain specific regional letters for academic
transcription. There's just a straight swap of -some- of the letters,
but in each case the writer has the same understanding: a Tifinagh "sh",
for example.
A common alphabet for the script works because the phonology of the language
variants is similar everywhere - what accounts for most of the variability
is the additional vowels of the Tuareg group.
There is understandable resistance among Berbers to being told which
letterforms to use. With this convention, areas can teach to their own
custom using different fonts but common tools.
With the existing encoding of variant letters in non-IRCAM repertoires,
regional repertoires are incompletely encoded, even for Algeria!
Of course, there are historical letters with presumably unknown or unusual
typographical conventions and meanings, which would need their own code
points if ever encoded, rather than being represented with a common
alphabet.
This archive was generated by hypermail 2.1.5 : Thu Feb 18 2010 - 08:30:41 CST