[iDC] please make comments regarding semantic overlay term

Danny Ayers danny.ayers at gmail.com
Mon Mar 17 10:30:48 UTC 2008


On 17/03/2008, Paul Prueitt <psp at ontologystream.com> wrote:
> The "maths" behind OIL are not mature, but are getting close, and yes
>  those system can be useful.
>
>  The barrier that is not understood is still there.
>
>  The statement should be:
>
>  There is an alternative to OWL, topics maps being one example of an
>  alternative, that is not explored

Ok, RDF now has greater mind share, but there's nothing stopping
people exploring topic maps further. (OWL is based on the Description
Logics formalism, as far as I'm aware TMs don't have such a
foundation).

It's worth noting that there's significant overlap between topic maps
and RDF, to the extent that to all intents and purposes topic maps can
be expressed in RDF.

See:
A Survey of RDF/Topic Maps Interoperability Proposals
http://www.w3.org/TR/rdftm-survey/

Living with topic maps and RDF
http://www.ontopia.net/topicmaps/materials/tmrdf.html

> because of the fixation by people
>  who somehow have been nominated as the authority.

...and perhaps because it works?

>  We all lose because of the non-exploration of topic maps, and of
>  other and even better alternatives such as n-ary patterns.

Better in what respect? For what purpose?

The binary relations of RDF are a very good fit with the links of the
Web, and n-ary relations can be expressed in RDF (though they do get
tricky at times), see:

Evolving the Link (one of mine, I'm afraid)
http://dsonline.computer.org/portal/pages/dsonline/2007/06/w3web.html

Defining N-ary Relations on the Semantic Web: Use With Individuals
http://www.w3.org/TR/swbp-n-aryRelations

>  The paradigm shift is held back because of the over investment in
>  something that objectively should not work to the extend that we need
>  for collective knowledge sharing.

As far as I can see, the Web has proven itself very effective for
collective knowledge sharing, even though machine processing of
human-readable text is severely limited. While the Semantic Web
languages are crude in expressive ability compared to human language,
they are considerably more tractable. Having a more computer-friendly
approach to information representation on the Web would seem to me a
step forward.

I really don't see any systematic hold-back of collective knowledge
sharing. Ok, there have no doubt been many technical errors in detail
(arguably RDF/XML syntax being the biggest). Education and outreach
hasn't been handled very well, though I think things have improved
considerably in the last year or two.

>  Perhaps you will agree..?

Semantic Web technologies may be far from ideal, but they do appear to
be the best bet right now. They do enable improved knowledge sharing
over what we've currently got, and are finding deployment *now*. They
should help provide technologies more to your liking in whatever phase
follows :-)

Cheers,
Danny.

-- 
http://dannyayers.com
~
http://blogs.talis.com/nodalities/this_weeks_semantic_web/


More information about the iDC mailing list