That's interesting - I had assumed that there was no maximum to the scalar
values in Unicode, just that each encoding had its limits.
In my copy of The Unicode Standard Version 3.0, I can't find an explicit
statement that scalar values in Unicode are only in the range U+0 to
U+10FFFF - but this could be deduced from the fact that all the encodings in
the Standard (UTF-8 and UTF-16) are limited to this range.
It looks like UCS-2 and UCS-4 are defined in ISO 10646. Does that standard
restrict the valid range of UCS-4 to 0..10FFFF? If not, does this represent
a significant divergence between Unicode and ISO 10646?
(Sorry if these are well-known issues - I've only been on the list for a
couple of months)
Thanks!
- rick
-----Original Message-----
From: Kenneth Whistler [mailto:kenw@sybase.com]
Sent: Tuesday, 18 December 2001 10:09
To: CHohberger@zebra.com
Cc: unicode@unicode.org
Subject: Astral planes (was: RE: Plane One use, was Re: HTML Validation)
Clive said:
> But I never did figure out if everything above Code Plane 16 was above
> or still below the Heaviside Layer... ;-)}
As for the cats, in "Up up up past the Russell Hotel, up up up to the
Heaviside layer"?
Actually, my surmise, given the fact that the code points past U+10FFFF are
forever inaccessible, is that they lie beyond the event horizon. Not merely
the event horizon of local black holes, as for Planes 2 and 3, which will be
endless sinks for CJK encoding energy, pulling hapless encoders in and
shredding them with infinite tides as they approach the danger zone.
No-- rather, the event horizon of the entire character universe, beyond
which we can detect nothing.
Perhaps, however, from the murmer of chilled microwaves emanating from the
vicinity of the noncharacters U+10FFFE and U+10FFFF, at the far nether
reaches of the astral planes, we can find patterns that will allow us to
interpret the earliest history of the character universe.
--Ken
This archive was generated by hypermail 2.1.2 : Tue Dec 18 2001 - 13:16:41 EST