From: Jukka K. Korpela (jkorpela@cs.tut.fi)
Date: Fri Dec 07 2007 - 08:52:29 CST
Asmus Freytag wrote:
> Given the history of the character additions in question, which were
> added precisely to allow the STIX project to add them to their font,
> Jukka's answer is unnecessarily dismissive.
If I'm unnecessarily dismissive, how about the Unicode conformance
requirements:
"C3 A process shall not interpret an unassigned code point as an
abstract character."
Or the Unicode FAQ:
"Q: Can applications simply use unassigned characters as they wish?
A: No! No conformant Unicode implementation can use the un-encoded
values outside of the private use area."
If this is not meant to apply to unassigned code points for which an
assignment has been planned, then the exception should be written loud
and clear.
> that would result in unnecessary duplicate coding if documents get
> created both with PUA codes (for STIX fonts) and standard codes, for
> any font suite that comes along just a bit later.
This is still about proposed characters, and about characters with very
limited scope of use (though surely important enough in that scope), so
I don't see how it would hurt to keep things clear: they are now
non-Unicode characters, which can be represented in Unicode data by
private agreements; if and when a new version of the standard containing
them has been approved, they become Unicode characters that should be
encoded as indicated, and hence existing data containing them should be
recoded. It's after all a straightforward operation, needed for limited
amount of data only. Fonts will need similar adjustments.
The risk of having a character's position changed or removed is probably
very small if not negligible in this case, but an approach that uses
unassigned code points would set a bad example.
Jukka K. Korpela ("Yucca")
http://www.cs.tut.fi/~jkorpela/
This archive was generated by hypermail 2.1.5 : Fri Dec 07 2007 - 08:53:53 CST