Date: Tue Mar 01 2005 - 20:50:27 CST
Peter Constable wrote:
> > - what gets put in the Web page code
> I'm not sure since I don't know exactly what the nature of the
> distinctions you want are.
> and what rather are one-off glyph variants that are best handled
> other ways.
*What* other ways could there be, other than a language tag for each local
script type -- *and* a language tag for each variation of that local script
Like Old Cretan Doric is a set of variant glyphs. And then there's also,
within Old Cretan Doric, a variant upsilon. I can only see this as two
language tags "language: Old Cretan Doric", and "language: Old Cretan Doric
Alt". There might be "Alt 1", "Alt 2", "Alt 3", "Alt 4" for another script
with 5 forms of theta.
If there *is* some other way of selecting that variant Old Cretan Doric
upsilon, from a Web page -- while otherwise using the main "Old Cretan Doric"
language -- then what is it?? Because that could be used to select all the
variants of standard Greek script in the first place.
If that technique doesn't exist (as you've previously indicated), than we can
discuss what is going to be done only with "language" tags. For our
discussion, let's say an utterly dependable scholarly process has defined 20
sets of local versions of "Greek Script", including "Alt" versions of those
Would adding 20 "local Greek Script variants" to the list of Microsoft
language tags be a reasonable thing to do or not? And the same process, for
example, for Berber script (mentioned by another list member), etc. (Though
*definitely* not for cuneiform!).
And actually -- why are we talking about "language" tags rather than "script"
tags? (Couldn't an OT font detect either "language" or "script" as the basis
of performing glyph swapping?)
This archive was generated by hypermail 2.1.5 : Tue Mar 01 2005 - 20:36:33 CST