Re: QNames and the XPointer namespace scheme

/ Elliotte Rusty Harold <elharo@metalab.unc.edu> was heard to say:
| The recently released xmlns() scheme draft
| <http://www.w3.org/TR/xptr-xmlns/> makes it even clearer that
| namespace bindings from the XML document in which the XPointer appears
| do not apply to the XPointerr. This draft explicitly states:
|
| The XPointer Framework on which this specification is based dictates
| that the initial namespace binding context prior to evaluation of the
| first pointer part consists of a single entry: the xml prefix bound to
| the URI http:/www.w3.org/XML/1998/namespace.

You are correct. I was not on the linking WG several years ago when
this decision was reached and I had forgotten the fracas that arose at
the time.

It flies so counter to my intuition that I never even considered it
and failed even to see it when it was written in black and white
phosphor before me.

We are now in a position where some specifications explicitly use the
in-scope namespaces (XPath) and some explicitly do not (Schematron,
XPointer).

I think this means that a finding on the subject of QNames can do
little more than point out the issues and highlight the considerations
that must be made when choosing an implementation strategy.

I will revise the finding accordingly.

                                        Be seeing you,
                                          norm

-- 
Norman.Walsh@Sun.COM    | Words--so innocent and powerless they are, as
XML Standards Architect | standing in a dictionary, how potent for good
Sun Microsystems, Inc.  | and evil they become, in the hands of one who
                        | knows how to use them!--Nathaniel Hawthorne

Received on Monday, 15 July 2002 11:40:53 UTC