- From: John Cowan <jcowan@reutershealth.com>
- Date: Wed, 26 Jan 2005 14:58:54 -0500
- To: Paul Grosso <pgrosso@arbortext.com>
- Cc: public-xml-core-wg@w3.org
Paul Grosso scripsit:
> Can someone point me to this wording or other
> wording that may mislead people into thinking
> they could define their own attributes in the
> XLink namespace?
Well, it seems I've put my foot into it (and now will be XML-cored).
What I really *should* have complained about was not that XLink implicitly
allows you to make up your own xlink: attributes, but that it does not
say you are non-conformant if you use xlink-typed elements in bogus ways.
So you can have a stray xlink:type="title" element outside a link, and
still be conformant, because the conformance clauses are only positive:
you do conform if you use Xlink elements in normal ways, but you are not
non-conformant if you use them in unexpected ways.
Here are the place where I think the language should be strengthened to
speak of non-conformantness rather than mere undefinedness.
4.2: Other combinations have no XLink-dictated significance.
5.1 (after the bulleted list): Subelements of the simple or
extended type anywhere inside a parent extended-type element have
no XLink-specified meaning. Subelements of the locator, arc, or
resource type that are not direct children of an extended-type
element have no XLink-specified meaning.
5.1 (next graf): If other XLink attributes are present on the element,
they have no XLink-specified relationship to the link. [I think
the next sentence should perhaps be left alone, though, because
it has a suggestion for higher-level resolution.]
5.1.1: If a resource-type element has anything other than an
extended-type element for a parent, the resource-type element
has no XLink-specified meaning.
5.1.2: If a locator-type element contains nested XLink elements, such
contained elements have no XLink-specified relationship to the
parent link. If a locator-type element has anything other than
an extended-type element for a parent, the locator-type element
has no XLink-specified meaning.
5.1.3: If an arc-type element has anything other than an extended-type
element for its parent, the arc-type element has no
XLink-specified meaning.
5.1.4: If a title-type element contains nested XLink elements, such
contained elements have no XLink-specified relationship to the
parent link containing the title. If a title-type element has
anything other than an extended-, locator-, or arc-type element
for a parent, the title-type element has no XLink-specified meaning.
5.2 (after the second snippet): If a simple-type element contains
nested XLink elements, such contained elements have no
XLink-specified relationship to the parent link.
Also remove the first assumption in Appendix B.
Doing these things means that my RELAX NG schema will reject only
non-conformant documents (I think); currently it also rejects documents
that are technically conformant but have constructs with no XLink-specified
meaning.
--
They do not preach John Cowan
that their God will rouse them jcowan@reutershealth.com
A little before the nuts work loose. http://www.ccil.org/~cowan
They do not teach http://www.reutershealth.com
that His Pity allows them --Rudyard Kipling,
to drop their job when they damn-well choose. "The Sons of Martha"
Received on Wednesday, 26 January 2005 19:59:24 UTC