W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > June 2003

Unrecognised elements in RDF/XML starting with "xml"

From: Dave Beckett <dave.beckett@bristol.ac.uk>
Date: Wed, 4 Jun 2003 15:30:38 +0100
To: RDF Core <w3c-rdfcore-wg@w3.org>
Message-Id: <20030604153038.727209f6.dave.beckett@bristol.ac.uk>

I got some comments from PFPS
on his confusion with attributes that begin with 'xml' and that lead
into a an editorial clarification of 6.1.2 of the syntax specification.

This led into noticing that we say nothing about what to do with
element names starting with 'xml' (case independent actually, I'm

Way back we made a decision on unrecognised attributes recorded in item 12
which was to ignore such attributes after dealing with the ones
we care above.

I don't recall seeing another XML specification that says much, if
anything, about XML names starting with 'xml' either, they assume you
can read the XML specifications and take it's description that these
as reserved and use that.

So I guess, in terms of proposals.  We can either:

1) continue to say nothing, the XML specification is sufficient.

2) modify the above decision (and clarify it's wording slightly) to:

  The WG resolves that unrecognized XML names should be ignored.

  then add a couple of new tests to record that such as:

   <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
    <rdf:Description rdf:about="http://example.org/thing">
      <xml:foo xmlfoo="blah">xyzzy</xml:foo>
      <ex:prop1 xmlnewthing="anything">stuff</ex:prop1>

  giving 1 triple
    <http://example.org/thing> <http://example.org/schema#prop1> "stuff" .

3) Explicitly allow them.  Maybe people do want to do:
  or things like that?

  with the same test above but with 2 triples, the second:
  <http://example.org/thing> <http://www.w3.org/XML/1998/namespacefoo> "xyzzy".

FYI, I see the W3C RDF Validator/ARP2 does #3.

Received on Wednesday, 4 June 2003 10:31:12 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:24:23 UTC