W3C home > Mailing lists > Public > public-wai-ert@w3.org > April 2005

Re: RDF Structure (was Extension property: "using")

From: Charles McCathieNevile <charles@sidar.org>
Date: Thu, 14 Apr 2005 00:34:07 +1000
To: "Chris Ridpath" <chris.ridpath@utoronto.ca>, public-wai-ert@w3.org
Message-ID: <op.so6de5eow5l938@researchsft>

On Wed, 13 Apr 2005 18:16:33 +1000, Chris Ridpath  
<chris.ridpath@utoronto.ca> wrote:

> This is one topic that I'm still not clear on so please forgive my  
> ignorance. I assume that our EARL, which is RDF and XML, will have a  
> schema which states which elements and attributes can be present. The  
> schema may also state the order in which some elements must occur. It  
> may also state which elements may contain other elements. Are my  
> assumptions correct?


> I know that you can mix RDF documents but my concern is that it will be  
> difficult to process the EARL as XML because I won't know where things  
> are.

Right. Unless you produced it yourself, you cannot expect it to always  
look the same.

> For example, the file that passed or failed a guideline will always be  
> marked like:
> <earl:WebContent rdf:about="#subject">
>  <earl:reprOf rdf:resource="URI to the file"/>
> </earl:WebContent>
> Is that right?

No. It could equally be

<r:Description r:ID="subject">
   <r:type r:resource="&earl;WebContent" />

<r:Description r:about="#subject">
   <e:reprOf r:resource="URI to the file" />

which, in RDF, means *exactly* the same thing.

If we have a suggested XML schema - i.e. one that lines up elements and  
attributes in a particular way, for some subset of EARL information, then  
we will need to write a SPARQL query that constructs things in this way. I  
am not convinced that this is a worthwhile exercise, since processing as  
XML places serious limits on what is actually interoperable (these are  
being reduced slowly, but will continue to exist for the future) and  
encourages people to regard the strictly limited XML version as equal.

It also means we need to do a lot more work to define the things we want,  
since the XML version will not have the flexibility and mergeability of  
RDF (unless people dumb down to only using well-formed XML, in which case  
semantic interoperability just about vanishes).



Charles McCathieNevile                      Fundacion Sidar
charles@sidar.org   +61 409 134 136    http://www.sidar.org
Received on Wednesday, 13 April 2005 14:34:52 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:55:52 UTC