[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
RE: [xml-dev] The prefix "fgdc" for element "fgdc:fgdc" is not bound
- From: noah_mendelsohn@us.ibm.com
- To: "Michael Kay" <mike@saxonica.com>
- Date: Wed, 12 Dec 2007 18:33:48 -0500
Michael Kay wrote:
> I gather the decision arose from some metaphysical debate about
> the intended
> semantics of namespaces. I'm no fan of that kind of discussion:it's up
to
> users, not specification writers, to decide what their data
> actually means.
> Specification writers should confine themselves to describing
> observable and
> testable behaviour.
I wasn't part of those discussions, but I think the reasoning was along
these lines: there was a hope that a given qualified name with a non-null
namespace would in fact have some particular use. So
{http://example.org/booknamespace,title} might be the title of a book,
regardless of the document in which it appeared;
{http://example.org/peoplenamespace,title} might be used for titles like
Mr., Ms., etc. There's clearly no hope for that with unqualified names,
since they are clearly used in all sorts of contradictory ways. Note
that, modulo some real messiness about the difference between QNames and
URIs, these names are sort of grounded in the Web, so if you wanted to
find information about {http://example.org/booknamespace,title} you'd
have a fighting chance of doing it by trying an HTTP GET from
http://example.org/booknamespace.
Ironically, and the reason I'm so aware of this, the fact that XML Schema
allows elementForm="qualified" on locally scoped names pretty thoroughly
undercuts this view. FWIW, I think elementForm="qualified" is on balance
OK. It's what people want in many case. Still, it means that you can
have expanded name {http://example.org/peoplenamespace,xxx} used for
completely different things, and with different validation rules, even in
the same document. Some members of the XML community were strongly
opposed to that, and I can see why. I happen to think that allowing it is
the lesser of the evils.
Noah
--------------------------------------
Noah Mendelsohn
IBM Corporation
One Rogers Street
Cambridge, MA 02142
1-617-693-4036
--------------------------------------
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]