W3C home > Mailing lists > Public > w3c-rdfcore-wg@w3.org > May 2001

Re: Issue #rdf-ns-prefix-confusion

From: Dan Connolly <connolly@w3.org>
Date: Thu, 24 May 2001 10:05:30 -0500
Message-ID: <3B0D233A.1FC731C9@w3.org>
To: Dave Beckett <dave.beckett@bristol.ac.uk>
CC: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
Dave Beckett wrote:
http://www.w3.org/2000/03/rdf-tracking/#rdf-ns-prefix-confusion
[...]
> 3.  Unprefixed attributes are deprecated - they SHOULD NOT be used in
>     the syntax from this date and WILL be forbidden in the next RDF
>     syntax document.
> 
> 4.  On input, unprefixed attributes from The List MAY be accepted.
>     If accepted, these attributes MUST be handled as if they were
>     written with a prefix as defined in #2.

What does "on input" mean?
The RDF M&S spec doesn't specify software modules;
it specifies a language. Resolution of
this issue requires deciding, about some XML documents,
whether they're in the RDF language or not, right?

My original 26 Apr 2000 message included a test case,
taken from the spec [namspace declarations filled
in here for completeness]:

=========
  <rdf:RDF
	xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
        xmlns:s="http://example/stuff#"
        xmlns:v="http://example/vocab#"
	>
    <rdf:Description about="http://www.w3.org/Home/Lassila">
      <s:Creator rdf:resource="http://www.w3.org/staffId/85740"/>
    </rdf:Description>

    <rdf:Description about="http://www.w3.org/staffId/85740">
      <v:Name>Ora Lassila</v:Name>
      <v:Email>lassila@w3.org</v:Email>
    </rdf:Description>
  </rdf:RDF>
=========

The question is: is that document an RDF document or not?

btw... it's also available at
  http://www.w3.org/2001/04rs22/confusedPrefixes.rdf

Whereas
	-- We have heard from an editor of the original
	spec that he never intended unqualified attributes
	to be part of the syntax

	-- tools that support the above syntax only
	do so because of buggy namespace support; they
	don't accept
		<otherNs:ClassName about="#xyz"/>
	but only
		<otherNs:ClassName rdf:about="#xyz"/>

	-- tools differ in their interpretation of
	unprefixed attributes in RDF in general
	and in the case of rdf:resource vs. resource in particular.

	-- the cost of supporting both about= and rdf:about=
	in the future is significant: consider XPath queries
	like
		<xslt:template match="/rdf:RDF/*[rdf:about="$myURI]">
		...
	and try fidgeting with it to accept about= on elements
	from the RDF namespace. It's doable, but it's a pain.

I propose that no, this is not an RDF document. It is
not part of the language specified in the 19990222 RDF spec.
Any suggestion that it does in the spec an error; I propose we
update the errata page to say so and notify www-rdf-interest
that any documents that use this syntax are in error.

I just wrote a little XSLT transformation to make the fix:

  http://www.w3.org/2001/04rs22/fixAttrs.xsl

and documented it to some extent:
  http://www.w3.org/2001/04rs22/#fixatt


> I've tried to capture the concensus on deprecation but I'm unsure if
> it comes across in the above.  Specifically, #3 and #4 seem to
> clash.  Do we allow unprefixed attributes for backwards comp. at all
> in future or forbid them in the future?

I consider them forbidden since 19990222, and I ask the
WG to agree.

-- 
Dan Connolly, W3C http://www.w3.org/People/Connolly/
Received on Thursday, 24 May 2001 11:05:36 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:35:51 EDT