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

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

From: Dave Beckett <dave.beckett@bristol.ac.uk>
Date: Wed, 30 May 2001 16:27:01 +0100
To: w3c-rdfcore-wg@w3.org
Message-ID: <24455.991236421@tatooine.ilrt.bris.ac.uk>

Previous threads can be followed from
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0166.html

following the actions recorded in the meeting at
  http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/att-0195/01-rdfc25May.html#N77

I have amended the previous resolutions (#3-#5,#6 deleted in above)
to give the following with new #3-#9 expanding on grammar and
document changes.


1. The current RDF/XML syntax uses the following attributes in the syntax:

     about aboutEach aboutEachPrefix
     ID bagID
     resource
     parseType
     -- List of RDF attributes (henceforth The List)

     The remaining concepts are not in the list because:
       a. Seq Bag Alt Property Statement
	 These are rdfs:Class-es and can never be used as attributes
       b. RDF Description   
	 Syntax only things that have no current use as attributes
       c. li _<n> subject predicate object type value
	 Not allowed to be used unprefixed according to the grammar

     Note Re: aboutEach aboutEachPrefix
       Note that there are other issues relating to aboutEach and
       aboutEachPrefix.  The resolution of those issues may affect
       the inclusion of these attributes in The List.


2.  Namespace qualification MUST be used for all RDF attributes
    on The List, where the namespace URI is 
    http://www.w3.org/1999/02/22-rdf-syntax-ns#

    The meaning of the attributes is defined by the appropriate RDF
    M&S sections and is not modified here.


3.  The updated grammar productions required for #2 in
      http://www.w3.org/TR/REC-rdf-syntax/#grammar
    are as follows.

      [6.6] idAttr ::= ' rdf:ID="' IDsymbol '"'

      [6.7] aboutAttr ::= ' rdf:about="' URI-reference '"'

      [6.8] aboutEachAttr ::= ' rdf:aboutEach="' URI-reference '"'
			      | ' rdf:aboutEachPrefix="' string '"'

      [6.9] bagIdAttr ::= ' rdf:bagID="' IDsymbol '"'

      [6.11] typeAttr ::= ' rdf:type="' URI-reference '"'

      [6.18] resourceAttr ::= ' resource="' URI-reference '"'

      [6.32] parseLiteral ::= ' rdf:parseType="Literal"'

      [6.33] parseResource ::= ' rdf:parseType="Resource"'


4.  Namespace qualification MUST be used for all property attributes.
    
    Discussion: this is to address the point Art Barstow raised in
      http://lists.w3.org/Archives/Public/w3c-rdfcore-wg/2001May/0175.html
    when he pointed out the grammar allowed non-prefixed properties.

5.  The updated grammar production required for #4 is:

      [6.19] Qname ::= NSprefix ':' name


6.  As a consequence of #4 the following paragraph is removed from
    http://www.w3.org/TR/REC-rdf-syntax/#grammar

     [[
      It is recommended that property names always be qualified with a 
      namespace prefix to unambiguously connect the property definition 
      with the corresponding schema.
     ]]
    in the annotated M&S copy this is http://ioctl.org/rdf/ms/rdfms#215

    since this is now required, not optional.


7. There are certainly other changes needed throughout the rest of
   the document such as all the examples and all other mentions of
   the attributes - there are probably over a hundred of these.


8.  Non-prefixed RDF elements (NOT attributes) are allowed when a
    default XML element namespace is defined with an
    xmlns="http://www.w3.org/1999/02/22-rdf-syntax-ns#" attribute.


9. The grammar changes required from #8 in
      http://www.w3.org/TR/REC-rdf-syntax/#grammar
    are all productions mentioning an rdf: element:
      6.1, 6.3, 6.25, 6.26, 6.27, 6.29
    which need to be changed to also allow the unprefixed form, for
    example:

       [6.1] RDF ::= ['<rdf:RDF>' ] obj* ['</rdf:RDF>']
                     | ['<RDF xmlns="http://www.w3.org/1999/02/22-rdf-syntax-ns#">] obj* ['</RDF>']

    and so on for the other productions, but this is rather verbose
    to express in this BNF and might be better addressed by more
    words.

Dave
Received on Wednesday, 30 May 2001 11:27:05 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 3 September 2003 09:35:54 EDT