- From: Dave Beckett <dave.beckett@bristol.ac.uk>
- Date: Thu, 06 Sep 2001 16:34:57 +0100
- To: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
- CC: Jeremy Carroll <jjc@hplb.hpl.hp.com>
>>>Jeremy Carroll said:
>
> I had another comment:
>
> In section 5. Infoset Conformance we say "no items are added or removed."
>
> I think this is incorrect.
>
> xml:lang is not described in the grammar, and can legally appear on any
> element (I think).
>
> I think xml:space is removed; I suspect all xml:* attributes are removed.
XML 1.0 says:
This specification does not constrain the semantics, use, or
(beyond syntax) names of the element types and attributes, except
that names beginning with a match to (('X'|'x')('M'|'m')('L'|'l'))
are reserved for standardization in this or future versions of this
specification.
-- http://www.w3.org/TR/REC-xml#sec-logical-struct
Examples are xmlns, xml:space, xml:lang and xml:base. Different
documents define what they mean and how they affect XML.
The WD-to-be does not add or remove anything from the infoset - it
currently lists various patterns of infosets that match RDF/XML. The
next stage is to say what bits of graph they create. At that point
it might be an infoset=>infoset transformation if we choose to
continue down this route.
The xml:lang stuff in RDF/XML is referred to outside the EBNF grammar
productions. The prose in the grammar chapter have not yet been all
merged, rewritten or considered in how they impact this document.
We furthermore have issues with xml:lang (and literals) that have not
yet been decided.
Dave
Received on Thursday, 6 September 2001 11:34:58 UTC