From: John Hudson (tiro@tiro.com)
Date: Thu May 20 2004 - 21:45:05 CDT
Kenneth Whistler wrote:
> A character encoding standard is an engineering construct,
> not a revelation of truth....
Amen.
I begin to suspect that part of the problem -- the problem of interminable debate, not any
technical problem -- is due in part to different perceptions of the Unicode Standard. It
must seem pretty obvious to engineers that this is a standard for encoding characters and
that implementing support for the standard does not, per se, imply much of anything about
how users should encode text. This is perhaps less obvious to non-engineers -- i.e. to
users --, and understandably so given the typical representation of Unicode to this
audience: 'Now supports all the world's major living languages!'. It is evident from the
Phoenician discussion that a good number of people -- intelligent people, and experts in
particular fields -- expect UTC decisions on what characters to encode to influence user
decisions on how to encode specific texts. I don't think this expectation is unreasonable,
given their perception of the standard, and perhaps Unicode needs to do a better job in
conveying what the standard is and does and how it can be used.
There remains, in the Phoenician debate, much fuss about Unicode disunifying what a
particular set of people consider to be the same thing. Perhaps the point needs to be made
more strongly that for practical text processing purposes *unification or disunification
of Phoenician and Palaeo-Hebrew happens only at the point of encoding a particular text*.
There is no reason at all why Semiticists cannot simply totally ignore the proposed
Phoenician block. The important question then, it seems to me, is not whether to encode
Phoenician or not, but how to better communicate that the encoding of a particular set of
characters does not mean that they have to be used to encode particular texts or languages.
John Hudson
This archive was generated by hypermail 2.1.5 : Thu May 20 2004 - 21:45:44 CDT