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

Re: Issue http://www.w3.org/2000/03/rdf-tracking/#rdf-ns-prefix-confusion

From: Art Barstow <barstow@w3.org>
Date: Thu, 24 May 2001 13:30:22 -0400
To: Dave Beckett <dave.beckett@bristol.ac.uk>
Cc: w3c-rdfcore-wg <w3c-rdfcore-wg@w3.org>
Message-ID: <20010524133022.B17980@w3.org>

On Thu, May 24, 2001 at 02:47:07PM +0100, Dave Beckett wrote:

> I propose we resolve:

Thanks for enumerating the changes Dave!

I agree with #1, #2 and #6.

WRT #5:

> 5.  On output, all RDF attributes from The List MUST be emitted with
>     a namespace prefix; where the prefix is defined as in #2.

Given #2, and the existing grammar, a RDF serializer would emit 
illegal syntax if it did not add a namespace prefix to all RDF
attributes so I don't see need to add this sentence to the spec.  
[Alternatively, it be stated in a non-normative Appendix of
implementor's notes.]

WRT #3 and #7 they address the question - what should be done with 
unprefixed attributes that are not in The List or not other RDF 
attributes (e.g. li, _<n>, aboutEach*):

> 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.
> 
> 7.  Unprefixed attributes not on The List have no meaning in RDF
>     and MUST NOT be used to generate statements.   Applications MUST
>     skip the element containing such attributes and generate no
>     statements for the entire XML element and content.

"attributes" in #3 and #7 are propAttr's in the grammar, and
those attributes are not required to have a namespace prefix:

 [6.10] propAttr       ::= typeAttr
                        | propName '="' string '"' (with embedded quotes escaped)
 [6.11] typeAttr       ::= ' type="' URI-reference '"'
 [6.14] propName       ::= Qname
 [6.19] Qname          ::= [ NSprefix ':' ] name

Also, there is a warning in the spec regarding property names:

 [[
  It is recommended that property names always be qualified with a 
  namespace prefix to unambiguously connect the property definition 
  with the corresponding schema.
 ]]

We could argue about how useful a propName with no NSprefix
is but there is probably lots of RDF with such propNames. And
they were warned!

So, I would not add #3 or #7 to the spec.  I most certainly
would not add the second sentence in #7.

WRT #4:

> 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.

I'm torn here between the pragmatic position that DanC gave in:

 http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0172.html

and a sense that we need to be sensitive to the BC issue we might create.   

Art
---
Received on Thursday, 24 May 2001 13:30:26 EDT

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