W3C home > Mailing lists > Public > www-rdf-logic@w3.org > June 2003

RE: Presentation Syntax - why?

From: Dickinson, Ian J <Ian.Dickinson@hp.com>
Date: Tue, 17 Jun 2003 21:23:07 +0100
Message-ID: <5E13A1874524D411A876006008CD059F05F814F6@0-mail-1.hpl.hp.com>
To: "'Jim Hendler'" <hendler@cs.umd.edu>
Cc: www-rdf-logic@w3.org

Jim -
You wrote:
> I still don't understand this at all.  The only OWL in 
> XML you will ever encounter is indeed RDF/XML with OWL 
> vocabulary.  The stuff in the presentation syntax would 
> be mapped to that before it would be expected to be used - 
> otherwise it violates the assumptions of OWL (i.e. that RDF/XML 
> is the exchange language for OWL documents) - nothing in this note 
> changes that.
You could apply the same analogy to N3.  Since the official exchange syntax
is rdf/xml, we would - by this argument - never encounter n3 documents
because everyone will diligently transliterate n3 to rdf/xml before
publishing.  In fact, it's pretty easy to find n3 content.  If a notation is
useful, people will use it ... and it will appear web pages, meta-data
repositories, email archives, etc etc.

So, we must assume that the new presentation syntax will leak out.  I
encounter some XML metadata.  Is this rdf/xml, owl presentation, vanilla xml
or what?  I want to be "liberal in what I accept, and conservative in what I
produce" ... so I should try to figure it out.  Maybe it's easy to tell from
the presentation syntax that that's what I've got, and that consequently
it's easy to apply the stylesheet and move on.  But it's too simple to say
that tool vendors and ontology/metadata consumers will never have to handle
this case.

I'd also point out that if one of your use cases is to provide a syntax that
can be handled with a streaming parser, then necessarily you're expecting it
to be published raw (otherwise you've thrown away the advantage).

I don't deny that encouraging the adoption of OWL for building ontologies is
a worthwhile goal.  We just need to be clear about the costs as well as the


Ian Dickinson    HP Labs, Bristol, UK     mailto:Ian.Dickinson@hp.com
Received on Tuesday, 17 June 2003 16:23:36 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 2 March 2016 11:10:40 UTC