Dare Obasanjo wrote:
> 
> I should give a practical example of what I mean. Imagine if RSS 2.0 was an RDF vocabulary and you wanted to map it to RSS 1.0. In RSS 2.0 the pubDate value is an RFC 822 formatted date while in RSS 1.0 the dc:date field usually used to indicate dates is in the ISO 8601 format. Semantic Web technologies generally allow me to say <pubDate> is logically equivalent to <dc:date> but an application needs more than that. If it can only consume ISO 8601 dates (e.g. if it is a .NET Framework application using the System.DateTime class) then simply saying 'treat <pubDate> like <dc:date>" is not enough. You will need to write some real code to do this transformation.

What would be needed here is full data type support by the Semantic Web. By 'full' I
mean the ability to define operators (methods) on data types (aka classes). That would
move the procedural aspects into the data types and would allow the integration 'work'
to remain declarative.

For example: the relational world[1] allows you to use operators in your
result relation *descriptions*:

SELECT Name, unix_timestamp(Age) AS AgeInUnixSecs
  FROM Person

Along these lines you can express mappings between RDF properties as expressions,
without procedural code. "If you encounter pubDate apply expression X to get
dc:date".

Not sure if existing Semantic Web standards support such transformational
notions or if this is orthogonal and should/could be added.


Jan

[1] Note that the relational model allows arbitrary data types, the limitations
    that one faces with SQL/RDBMS is not a Relational Model issue.


> --
> PITHY WORDS OF WISDOM
> There is nothing more satisfying that having someone take a shot at you, and miss.
> 
> ________________________________
> 
> From: Ronald Bourret [mailto:rpbourret@r...]
> Sent: Tue 11/30/2004 10:23 PM
> To: xml-dev@l...
> Subject: Re: [xml-dev] Ontolgies, Mappings and Transformations (was RE: Web Services/SOA)
> 
> Dare Obasanjo wrote:
> 
> >  However the Semantic Web
> > related mapping technologies don't allow for the kind of complex and
> > messy mappings that occur in the real world.
> 
> Then what's the point? I should have hoped that Semantic Web
> technologies (and ontologies in particular) would have allowed me to say:
> 
> 1) Here's a general concept.
> 
> 2) Here's how you get there from my particular expression of that concept.
> 
> Without this, we'll never be able to automatically exchange / integrate
> information, since we'll never agree on a single set of metadata for a
> particular concept.
> 
> -- Ron
> 
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
> 
> The list archives are at http://lists.xml.org/archives/xml-dev/
> 
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://www.oasis-open.org/mlmanage/index.php>
> 
> -----------------------------------------------------------------
> The xml-dev list is sponsored by XML.org <http://www.xml.org>, an
> initiative of OASIS <http://www.oasis-open.org>
> 
> The list archives are at http://lists.xml.org/archives/xml-dev/
> 
> To subscribe or unsubscribe from this list use the subscription
> manager: <http://www.oasis-open.org/mlmanage/index.php>

-- 
Jan Algermissen
Consultant & Programmer	
http://www.jalgermissen.com