Liam,
I didn't mean to imply that anyone was trying to change XML itself, only the
"all" part of "all XML files" by enabling XML encodings that are not
considered text. I never really thought of the binary XML efforts as a way
to simply define one more encoding to use, but that does make sense.
Is it possible to define such an encoding, though, while leaving Appendix F
of the XML Rec. (non-normative as it may be) alone? I must plead ignorance
on some basic facts of what has often been called "binary XML", so if
there's a particular FAQ I should read just point me to it.
Bob
-----Original Message-----
From: Liam Quin [mailto:]
Sent: Friday, March 12, 2004 11:59 AM
To: DuCharme, Bob (LNG-CHO)
Cc:
Subject: Re: [xml-dev] How does XML work with Linux
On Fri, Mar 12, 2004 at 11:00:46AM -0500, DuCharme, Bob (LNG-CHO) wrote:
> And because all XML files are text files (despite the best efforts of some
> people to change that) they're particularly portable between operating
> systems.
And with XML 1.1 you can even move them to/from sysetms that use
the Unicode line ending (NEL) thingy.
Of course, you have to ask what is a text file -- UTF-8? UTF-16?
UCS-32? Big 5? At one time Unix defined a text file to contain
only 7-bit ASCII characters (the 8th bit being used for parity,
and in effect stored as 0).
As for "best efforts", I'm not seeing a strong demand to change XML.
I'm seeing people who want a standardised way to interchange XML
with an "efficient encoding", but that's not the same as changing
XML itself.
Liam
--
Liam Quin, W3C XML Activity Lead, http://www.w3.org/People/Quin/
http://www.holoweb.net/~liam/
