Re: [xml-dev] RE: Namespace use cases

[Michael Ludwig <mlu@as-guides.com>]

Kurt Cagle schrieb:

> I'd also contend that while most of the people on this list are
> used to the notion that a prefix is essentially just a temporary
> placeholder for a namespace, for many people, especially those
> struggling with XML, a prefix is perceptually carved in stone.
> The "xsl" namespace is for XSL, the "xs" namespace is for XSD
> (or XSDi), the "atom" namespace is for atom

Good point. The namespace-aware XML processor doesn't care when
the XSL namespace is mapped to the "xs" prefix - but it confuses
a lot of humans, like me. It is carved in the mind-stone.

> and these namespaces are immutable. I suspect a part of this
> comes from the fact that in most of the languages that most
> people are exposed to, the closest thing to a namespace is a
> class, and class names in nearly all languages (except XML) ARE
> immutable.

Most languages do not work when arbitrarily mixed with other
languages. (Ask javac to compile Perl.) XML is not a language in
that sense, but, as Ken Holman explains it, a labeled hierarchy.

Being able to combine and recombine instances of this hierarchy to
unforeseen new instances, unmediated by any central registry or
authority, where each component maintains its local and global
identity (its name and nationality) requires the flexibility to
remap prefix tokens along the way.

How could such a mechanism have been made simpler?

Michael Ludwig