SPEC - Text (I18N)

Jan Hardenbergh ([email protected])
Thu, 20 Apr 95 15:19:00 E


A loosely defined desing process allowed the tremendous process VRML has
made in less than a year, but adding text makes me nervous.

- "just add some text" seems like a recipe for disaster.
- Waiting to see what happens in HTML does not close VRML 1.0 soon.
- Adding no text does not seem to be popular.
- Getting the group up to speed on all of the I18N issues in a
timely fashion seems unlikely.

Perhaps side stepping the whole issue by providing a symbol library
and indexing scheme. There would be a 1:1 match between index and
symbol (like an X11 font file.) Conceptually, it would be a URL
with a VRML Switch statement in it. Each child node would have some
graphics and a Translate node...

More URLs
(What IANA considers to be valid "charsets")
ftp://ftp.isi.edu/in-notes/iana/assignments/character-sets
Very recent discussions about these issues by Glenn Adams, editor
of the upcoming Unicode revision.
http://www.stonehand.com/glenn/csets.html

If we wanted to do something, I would argue for specifying the
coded character set to be ASCII instead of ISO-Latin-1, since ASCII
is a subset of most other character sets (including ISO-Latin-1 and
ISO 10646). This should be an attribute of the text.

This looks interesting (from the IANA URL above) :
> Name: ISO-10646-UTF-1
> MIBenum: 27
> Source: Universal Transfer Format (1), this is the multibyte
> encoding, that subsets ASCII-7. It does not have byte
> ordering issues.
> Alias: csISO10646UTF1

Tho, it probably has character switching semantics (as does EUC, ISO2022)?

If we get into 16 bit characters, we are into binary data and
need to think about byte ordering.

YON, [email protected], Jan C. Hardenbergh, Oki Advanced Products 508-460-8655
http://www.oki.com/people/jch/ =|= 100 Nickerson Rd. Marlborough, MA 01776
Imagination is more important than knowledge - Albert Einstein (1879-1955)