SUO: Re: Montologies
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o
Chris_1 & Chris_2,
It's possible that I was twisting 2 or 3 threads together
in my head, but my ultimate concern here is a stock issue:
What is the best way to achieve a capacity for coping with change
and a facility for dealing with diversity in logic-based intelligent
software systems, with or without an endoskeleton of ontological ribs?
Two themes that lie constantly in the background are these:
1. Keeping Faith, not alienating our real or imagined,
passed, pressing, and prospective user communities.
2. Saving Time, doing what hasn't been done already.
From my present standpoint, and for the forseeable future,
I am personally focusing on the problems that affect the
development of a "Scientific Technical Ontology" (STO_1),
most especially, a "System Theoretical Ontology" (STO_2).
Whether it concerns change and time or anything else,
I'm usually tempted to say that we can't formalize
common sense, that it's a contradiction in terms.
But that is not exactly so. There is a process
for deriving scientific theories out of common
intuitions, but -- and here only the greatest
degree of self-control keeps me from writing
this all in capitals --
It takes about 2.5 millennia --
and --
It's already been done --
So common sense can be formalized in pretty much the same sense that
a moose can be hunted, killed, and stuffed, but if we want ontology
to be something more than a venery and a taxidermy of the intellect,
then I think that we've got some thinking to do about the precise
nature of the formalization arrow.
Long consideration of the problem has convinced me that
we must study how the quantitative version of the problem
was solved if we want to have any chance of success with the
qualitative or logical analogue.
Two observations are immediate, if you can call
the decade that it took me to make them immediate:
1. The historically found solution did not gel out
of "ordinary language analysis". We will never
wring a working theory of qualitative change and
process out of ordinary language concepts alone.
2. Our best models of complex phenomena all work by way of
a multi-stage linkage, the simplest case being a 2-step
process where the logic is used to prove theorems about
abstract mathematical objects, while the mathematical
objects serve as the icons of the external objects.
Jon Awbrey
o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o~~~~~~~~~o