Re: CG: More meat for the ontological stew
I agree with John Sowa, Jean-Luc. This was a very interesting article.
Thank you for posting it. First off, I think that Shirky makes many valid
points, but I can't agree with his conclusions.
The example from Yahoo is one application of an ontology and it is a
particularly bad one. However, one could (and people are planning to)
use ontologies as the contents of tags or for many other purposes. So,
directing his criticism at ontologies rather than classification schemes
is ultimately off-target.
Shirky seems to have a general strategy of adopting a very constraining
definition of a particular project, pointing out a flaw that follows from
that definition and then saying that the project is doomed. He doesn't
seem to imagine that the projects could address those problems.
Oddly, his basic strategy seems to rely on the essentialism he rejects.
i.e. ...
Shirky: Ontologies are rigid, absolutist and essentialist.
Representations need to handle context and different point of views in
order. Therefore, ontologies are destined to fail.
Ontologist: True. We should make ontologies and the systems that use
them be able handle context, multiple points of view, and a variety of
conceptual frameworks so that they can express all of the things people
will want to express.
(Implicitly) Shirky: But that wouldn't be an ontology!
This article
http://www.shirky.com/writings/semantic_syllogism.html
follows the same argumentative style. In light of the inadequacies of
deduction for performing many important types of reasoning, Shirky gives
up on the project rather than (a la John Sowa and others) working on how
to address the challenge.
JPC
>
> JLD> Another recent paper, ontologies for "the masses"
> > don't work either:
> >
> > http://shirky.com/writings/ontology_overrated.html
>