Re: Key questions about common upper ontologies
Pat and Danny,
I sympathize with Pat's hopes, but I agree with
Danny that we are not going to get there by a
brute-force legislation of one universal ontology.
There are two case studies that indicate where
the "sweet spot" or the "low-hanging fruit" is
available, and Danny is closer to it than Pat:
1. This year, the Cyc project will celebrate its
20th anniversary. It is the closest thing yet
to one universal ontology, but so far, Cyc has
not led to a single deployed application of its
very richly axiomatized ontology. Over 70 million
dollars has been spent on Cyc, but all that money
has come from research grants and contracts --
not one penny from an actual application.
2. At the other extreme is WordNet, which is the
most widely used lexical resource in the world
-- partly because it is freely available for
anybody to use. WordNet could certainly be
improved, and there are lots of other lexical
resources that do better in one area or another.
But they all share the same characteristic:
they focus on descriptions of how words are
actually used, not on legislated definitions
about how they should be used.
Since OpenCyc has been released, many groups have
started to use it, but they are using it more like
WordNet than like the fully axiomatized Cyc. That
is partly because OpenCyc does not have all the
axioms even for those terms that have been released.
However, I am also aware of some research projects
that have access to the full Cyc system, and many of
them are only using the hierarchy in a WordNet-like
approach, rather than the fully inferential way that
the Cyc axioms are designed to support.
Pat said
PC> And a standard upper ontology is perfectly feasible.
> In all of the discussion we have endured on this list
> about how different people might represent the same concept
> differently, I have seen many examples of how some of us
> might **prefer** one representation rather than another,
> but not a single example of how one purpose **requires**
> one representation and another purpose **requires** a
> logically inconsistent representation.
I agree with the first line, since Cyc is an example that
it is possible to create such a thing. But the fact remains
that after 20 years and over 70 million dollars, nobody
has successfully used the Cyc definitions and axioms for
a single deployed application.
Pat asked the following question:
PC> We need not debate whether it is **possible** to
> represent the same thing in logically inconsistent ways
> -- it does seem so, if one never encounters a situation
> where the differences have practical consequences -- but
> is there actually a **need** to do so? For what purpose?
Nobody (except Microsoft, perhaps) deliberately tries to be
inconsistent with anybody else, but what is important for
one application is often irrelevant for another. For example,
we might say that a ball is spherical. But no actual ball
is really spherical, and a soccer ball, baseball, tennis ball,
or golf ball has very clear and well defined indentations,
ridges, and roughness that deviate from a sphere. And a
football or rugby ball are not even designed to be spherical.
A ball bearing is one of the few things for which a perfect
sphere is ideal, but none of them reach that goal. And
nobody even attempts to make a perfectly spherical meatball.
These examples could be repeated endlessly in any field you
care to look at. For more examples, see the Knowledge Soup
chapter (Ch 6) of my KR book.
PC> So why haven't we agreed on one yet? This is an
> interesting issue, but it probably should be left
> to a different thread.
No. This is part of the central issue, which Jim stated
in his original questions:
Jim S> Would it be desireable and feasible for a large and
> diverse organization (with many domains), with a compelling
> need for data and semantic interoperability, to adopt
> or develop a common upper ontology, plus multiple
> domain ontologies, and to eventually mandate it within
> the enterprise for those systems requiring interoperability?
Note the two words "desirable and feasible". It is clearly
desirable, since people have been trying to agree on a common
ontology since the time of Aristotle, but every attempt has
led to failure. Before attempting to develop yet another
version of Aristotle's or Kant's or Peirce's or Cyc's
approach, it is essential to explore the reasons for the
previous failures and how they can be avoided.
I'll end by quoting Danny's conclusion, which I support:
DA> Successful communication does not require the whole abstract
> language or concrete terminology. This is advantageous to the
> adoption of any SUO, as people are more likely to be willing
> to augment their existing languages/vocabularies than replace
> them with a whole new language infrastructure.
John