[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Suggested guidelines for using local types. (was Re: Enlightenmentvia avoiding the T-word)
- From: Francis Norton <firstname.lastname@example.org>
- To: "Fuchs, Matthew" <email@example.com>
- Date: Thu, 06 Sep 2001 17:37:21 +0100
"Fuchs, Matthew" wrote:
> The "conceit" of XML Schema is that Schema processing happens on the infoset
> of the well-formed document parse. Therefore, all namespace details have
> been settled before XSDL starts its business. As such, XSDL is basically
> concerned with qualified and unqualified - not prefixed or unprefixed. But
> your post reemphasizes the point that it's the instance author's job to make
> his/her intent clear.
Sure, there's no chance of XSDL getting confused - it's namespace aware.
Likewise, XSLT will put in the necessary xmlns="" attribute (I notice
that both saxon and msxml put it on the local element itself, not the
parent - presumably in case the parent is itself in a non-blank default
namespace). It's humans we have to worry about, since XML is supposed to
be both human and machine readable.
As you say, the instance author should make the intention clear, and I
think slapping an explicit xmlns="" attribute on unqualified local
elements would do the trick nicely, or at least as nicely as possible.