Re: declaredAs

Pat Hayes wrote:
>...
 > I think you are mixing up asserting - a speech act, claiming a sentence
> to be true by saying (or publishing) it - and declaring in the sense I 
> think its being used here, analogous to a declaration in a programming 
> language. The latter has some kind of special significance for parsing 
> and errors.
>   I agree RDF doesn't have explicit assertion or denial, etc., but there 
> is an emerging social/Web consensus that publication is tantamount to 
> assertion, which I think is what you are talking about here.

This reminds me of an issue I've been worrying about and am interested
in opinions about.

We've developed and are using a system to create RDF documents from
spreadsheets. We hope to release it (rdf123) for others to use this
week.  It has a number of features that we think improve on previous
spreadsheet to RDF translators.

The basic idea is that we generate a graph for each row in a
spreadsheet table using a template.  The RDF graph for the entire
spreadsheet is the union of the row subgraphs.  For a number of
reasons, we've chosen to represent a template as an RDF graph.
Here is the template for one of our simple examples.

   <rdf:RDF>
    <rdf123:Map rdf:about="">
      <dc:creator>Lushan Han</dc:creator>
      <dc:title>employee map file</dc:title>
    </rdf123:Map>
    <foaf:Person>
       <foaf:name>Ex:$1^^string</foaf:name>
       <foaf:mbox>Ex:$2^^string</foaf:mbox>
       <emp:supervisor>
          <foaf:Person>
             <foaf:name>Ex:$4^^string</foaf:name>
          </foaf:Person>
       </emp:supervisor>
       <foaf:phonenumber>Ex:$3</foaf:phonenumber>
    </foaf:Person>
   </rdf:RDF>

where the Ex:$N symbols refer to the Nth cell in the spreadsheet row.

One thing that nags at me is that the RDF template document doesn't
really mean what it purports to say.  One thought I had is that we
could avoid possible semantic lawsuits if there was a way to assert
that the (entire) RDF document is 'a work of fiction'.  The pragmatics
of this would be that, unless you knew what what you were doing (i.e.,
in this case, how to process an RDF123:Map) you should probably ignore
the document. This is the approach taken for human documents.

Has something like this ever been explored or suggested?

Tim

-- 
  Tim Finin, Computer Science & Electrical Engineering, Univ of Maryland
  Baltimore County, 1000 Hilltop Cir, Baltimore MD 21250. finin@umbc.edu
  http://umbc.edu/~finin 410-455-3522 fax:-3969 http://ebiquity.umbc.edu

Received on Sunday, 12 August 2007 23:40:34 UTC