Re: [xml-dev] Ten Years Later - XML 1.0 Fifth Edition?

[Elliotte Harold <elharo@metalab.unc.edu>]

Michael Kay wrote:

> I do agree with you that the potential set of characters is unbounded and
> one has to draw an arbitrary line somewhere. The problem is that different
> people would draw it in different places. I don't really see how you propose
> to get consensus on a subset.
> 

The set is large but finite. XML 1.0 actually represents a pretty damn 
good consensus that goes way beyond what any other language has ever 
achieved.

If we were to issue a new version of XML now, then Unicode 5.0 would 
suffice for all our lifetimes and likely way beyond. In fact, Unicode 
3.0 pretty much took care of the last cases anyone is ever remotely 
likely to need.

We simply don't need an indefinitely extensible set of name characters, 
and I think creating such a set would be actively harmful to 
interoperability. I prefer a fixed enumeration of all the legal name 
characters that does not include placeholders for characters that may be 
added in the future.

-- 
Elliotte Rusty Harold  elharo@metalab.unc.edu
Java I/O 2nd Edition Just Published!
http://www.cafeaulait.org/books/javaio2/
http://www.amazon.com/exec/obidos/ISBN=0596527500/ref=nosim/cafeaulaitA/