[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Re: Syntax + object model

• To: "Simon St.Laurent" <simonstl@s...>, <xml-dev@l...>
• Subject: Re: Re: Syntax + object model
• From: "Jonathan Borden" <jonathan@o...>
• Date: Sun, 25 May 2003 17:54:46 -0400
• References: <r01050400-1026-97D1A6BC8E4C11D7AA160003937A08C2@[192.168.124.11]>

Simon St.Laurent wrote:

> jonathan@o... (Jonathan Borden) writes:
> >What I am saying is that you are not making -- what I consider -- a
> >fair or valid statement about ontologies *in general* as opposed to
> >the above two projects.
>
> I disagree, but since you cut my discussion of specific ontologies,
> there isn't much further to say.  Hang out at knowledge tech conferences
> and listen, and you'll find plenty of rhetorical overreach.  I'll be the
> jerk reading Feyerabend in the corner.

Since I've not ever been to a KT conference, I'll take your word for it --
on the other hand similar rhetorical overreach was common in the heyday of
XML conferences as well. In neither case does it mean that there isn't a
core reality to both technologies.

>
> >For what I consider a significant class of problems, ontologies *are*
> >the right answer for vocabulary development for XML. (e.g., see:
> >http://www.cogsci.princeton.edu/cgi-bin/webwn)
>
> Sure, for word searches, that's great.  If I just want to mark up a
> document or an invoice, I'm not nearly as excited about that vision.

If the documents are components of a patient's healthcare record, I get
pretty excited about that vision. Similarly folks like John Cowan might get
excited about similarly marking up news stories related to heathcare. The
National Library of Medicine has done a great job 'marking up' biomedical
scientific articles using such a vision (i.e.:
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed), now you may call
that 'word search' but that is what google etc. are all about -- which alot
of folks find pretty important -- of course I am not claiming (hardly) that
ontologies encompass all of computer science, just to point out that there
is a large and widely acknowledged as important class of problems that
benefit from an ontologic approach. Since those problems also lend
themselves to using XML as a document interchange format, the intersection
between XML and ontologies is significant.

>
> >When you write about
> >"ontologists" coming 'round xml-dev, I wonder who you are talking
> >about? I don't consider myself an "ontologist" though I'm involved
> >with ontologies -- I don't know of any real ontologists who post
> >regularly on xml-dev, though perhaps there are a few who read it.
>
> I suspect 'real' ontologists have better things to do.  Still, about
> once a month, we get people here who talk about RDF as the right way to
> do vocabularies.  Roger Costello's recent piece on OWL was fairly
> memorable, and it's been a popular trend lately.

Are you suggesting that Roger's posts are offtopic?

> Maybe it's just the
> latest version of RDF condescension toward XML?
>

It is an all too common misconception that RDF is about ontologies.
Admittedly there are relationships between RDF Schema and OWL, but the
primary relationship between _RDF_ and OWL is that OWL is a language encoded
in RDF _syntax_. You might gasp at that choice of syntax -- and it has been
a somewhat controversial one as evidenced by discussions of the OWL WG --
but nonetheless RDF and hence RDF/XML is the official syntax in which the
OWL language is interchanged. There are other presentation syntaxes for OWL
including non-RDF XML (see:
http://lists.w3.org/Archives/Public/www-webont-wg/2002Dec/att-0295/01-OWL-XM
L-Schemas.html)

In any case I hardly detect any sort of "RDF condescension toward XML" in
Roger's posts. I *have* heard XML condescension among discussions involving
senior computer scientists who, for example, prefer s-expressions etc., but
this is a really tough group -- what can one say? ***

Jonathan

*** I usually chide them that if the world really cared what they had to
say, it would already be using Lisp :-))) ****
**** I can say this (in jest) as a former Lisp hacker (so calm down)

• References:
  • Re: Re: Syntax + object model
    • From: "Simon St.Laurent" <simonstl@s...>

• Prev by Date: Re: Syntax + object model
• Next by Date: RE: Does SAX make sense?
• Previous by thread: Re: Re: Syntax + object model
• Next by thread: Re: Syntax + object model
• Index(es):
  • Date
  • Thread