Re: SUO: *Date 15 Mar 2002 -- Obstacles To Integration
Jon --
You make a very good point in your daily SUO note below, which in my mind
highlights exactly what I have understood to be the value of the current
renewed interest and application of the ancient and philosophical
discipline of ontology to practical commercial and scientific ventures.
I have thought of the dictionary problem as an informal version of Gödel's
concept of incompleteness. A typical dictionary is a closed system of
words, defined in terms of the words within the same dictionary. This
introduces a well-known problem of circularity, which, if not formally
'incomplete', is generally regarded as less than fully satisfying. Using
your rubric, it seems that a dictionary might be seen as a system of
empirical lexical propositions that is necessarily incomplete, and calls
out for reconciliation by an external, rationalized system of concepts.
My particular interest in this approach stems from the domain of
technology-based information systems for business enterprises, or more
generally 'enterprises' of all types. The problem I have seen in this
domain is a form of natural language ambiguity, where terms (words and
other lexical elements) exhibit synonymy and polysemy. This seems to be
impossible to work out in the context of the empirically discovered terms
of the business itself. What this problem requires is an external
framework of terminology (like your notion of a rational, concept-driven
mode of cognition) that can stand as a reconciler of the ambiguities
inherent in the 'found', natural business language itself. I have used the
word 'ontology' to designate such a reconciling framework of concepts that
stands outside the empirically discovered 'lexicon' of terminology.
http://researchweb.watson.ibm.com/journal/sj/352/mcdavid.html
This is exactly a problem of semantic integration. You have provided a
framework whereby I can talk about natural (business) language as the
empirical data, and a more formal ontology as a rational, concept-driven
structure that provides disambiguation of meaning within a particular
business domain. I hope this isn't too much of a departure from the point
you are trying to make here.
Doug McDavid
Member, IBM Academy of Technology
mcdavid@us.ibm.com
"Imagine all the people ... living life in peace."
Jon Awbrey <jawbrey@oakland.edu>@majordomo.ieee.org on 03/14/2002 09:00:08
PM
Please respond to Jon Awbrey <jawbrey@oakland.edu>
Sent by: owner-standard-upper-ontology@majordomo.ieee.org
To: Stand Up Ontology <standard-upper-ontology@ieee.org>
cc:
Subject: SUO: *Date 15 Mar 2002 -- Obstacles To Integration
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤
| Tell me, good Brutus, can you see your face?
|
| No, Cassius, for the eye sees not itself
| But by reflection, by some other things.
|
| Shakespeare, 'Julius Caesar', 1.2.53-55.
Obstacles To Integration
One of the heuristics that I always have trouble remembering
in time is the one that I know as "Analyzing the Resistance".
It goes like this: When you find yourself having persistent
difficulties making progress on a problem, then maybe it is
time to take the load off the accelerator and take a look at
the obstacles or other impedances that might be opposing your
forward motion.
In that spirit, I would like to look at some of the obstructions
to inter-communication and inter-operation tasks, which for ease
of discussion I cross-index under the rubricon of "integration".
In order to bring a focus to these thoughts, I will restrain myself to
a single sector of integration tasks, namely, that of coordinating what
we might call the "empirical" and the "rational" approaches to any given
domain of phenomenal puzzles or practical problems, and for the point of
application I will concentrate on the general class of "dictionary tasks".
I believe that the dictionary task is badly misconceived in many present
efforts, and the reasons for this lie not so much in the problem domains
proper as in ourselves, to be specific, in the lack of integration that
we achieve between our various "cognitive styles". Most acutely in the
present focus, this amounts to a less than optimal coordination between
the "empirical", or data-driven, and the "rational", or concept-driven,
modes of cognitive specialization or other means of habitual operation.
What do you see when look through a dictionary?
If you see definitions, then recognize your "rational" face.
If you see usage data, then recognize your "empirical" face.
If you see no way to see both faces in a single moment then
see the gaps it shows you gaping in your stereoscopic sight.
Jon Awbrey
¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤