From: Hans Aberg (haberg@math.su.se)
Date: Mon May 23 2005 - 16:30:24 CDT
At 12:17 -0700 2005/05/23, Kenneth Whistler wrote:
>We need to steer clear of the dangers of assuming that anything
>which is text*like* must be encoded as characters and interchanged
>in plain text in Unicode.
>
>There is no prima facie case for standardizing characters for
>undeciphered scripts (even if they can be demonstrated to actually
>be *scripts*). There is no obvious *need* to generate and
>transmit plain text involving undeciphered scripts. The main
>potential users -- a usually small number of decipherers -- can
>use other means to meet their needs.
It seems me that the basic Unicode range should mainly focus on
"characterizing" the atomic semantic units of natural scripts,
ancient, regional and modern. If a set of symbols has not been
reliably deciphered, according to this principle, it should not be
added.
The reason, though, that good folks fight so hard over getting all
kinds of symbols added to Unicode, it seems me, is that if one does
not succeed with that, one is left out in the cold. The private range
gives no guarantees of anything. I can look through already defined
private ranges and deliberately put my characters there, just in
those places, it seems.
Unicode could provide more ranges of symbols, not as tightly held as
the basic one, helping to sort out symbols clashes, and helping to
provide glyphs for rendering. I think that if Unicode would do that,
much controversy would go away.
-- Hans Aberg
This archive was generated by hypermail 2.1.5 : Mon May 23 2005 - 16:31:12 CDT