Re: SUO: The Story So Far
Chris and Doug,
You both have correct views about some of the requirements.
The complete requirements must include the union of both.
Chris P. wrote:
>>... The problems we are
>talking
>> about arise when we start using a sufficiently accurate/formal/etc.
>> representational framework. To slip into a Hayes rhetorical mode for a
>> moment - if you are taking these as your examples, you have little or no
>> idea of the kinds of tasks that need to be done to produce something
>useful
>> for commercial/industrial systems.
Doug wrote:
>In any event, natural language and various practical classification schemes
>have been the way useful commercial and industrial work have always been
>conducted. I came to this list as a step in a long career of trying to
>refine these
>schemes and usages to make them more practical and useful in a world
>that is increasingly permeated and dependent on interoperating software
>mediations of human communication. In recent exchanges I am learning that
>this endeavor (IEEE SUO) was conceived by folks who have a different
>concern.
>I am still trying to really understand what that motivating concern is, and
>whether
>it has any relevance to the work I am trying to do. The words "useful
>commercial/industrial systems" give me hope for commonality. The rejection
>of all of natural language and existing classification schemes pretty
>effectively
>snuffs that glimmer of hope.
We need both. And the formal ontologies that Chris P. is
talking about must be able to support the concepts (or revised
or specialized versions theorof) that come from the lists of
terms that Doug is talking about.
We have been talking about these requirements for years, and
there is no question that we need both. I recommend the
glossary of definitions that I wrote (and which were based
on two years of discussion in the NCITS T2 committee):
http://www.bestweb.net/~sowa/ontology/gloss.htm
Please note the following two terms in that glossary:
terminological ontology.
An ontology whose categories need not be fully specified
by axioms and definitions. An example of a terminological
ontology is WordNet, whose categories are partially
specified by relations such as subtype-supertype or part-whole,
which determine the relative positions of the
concepts with respect to one another but do not completely
define them. Most fields of science, engineering, business,
and law have evolved systems of terminology or nomenclature
for naming, classifying, and standardizing their concepts.
Axiomatizing all the concepts in any such field is a
Herculean task, but subsets of the terminology can be used
as starting points for formalization. Unfortunately, the
axioms developed from different starting points are often
incompatible with one another.
formal ontology.
A terminological ontology whose categories are distinguished
by axioms and definitions stated in logic or in some
computer-oriented language that could be automatically
translated to logic. There is no restriction on the
complexity of the logic that may be used to state the
axioms and definitions. The distinction between
terminological and formal ontologies is one of degree
rather than kind. Formal ontologies tend to be smaller
than terminological ontologies, but their axioms and
definitions can support more complex inferences and
computations. The two major contributors to the development
of formal ontology are the philosophers Charles Sanders
Peirce and Edmund Husserl. Examples of formal ontologies
include theories in science and mathematics, the collections
of rules and frames in an expert system, and specification
of a database schema in SQL.
What Doug is talking about is a "terminological ontology" and
what Chris is talking about is a "formal ontology". According
to the above definition, every formal ontology is a special
case of a terminological ontology.
There is no conflict or inconistency. It just takes a lot
of hard work to get from a typical terminological ontology
to a formal ontology. The Cyc project has spent over 100
person years in trying to do that.
John Sowa