Re: Unified Framework for Ontology
John,
The earlier note you include is right on the money. I agree
with all of this. It's a breath of fresh air.
On Monday 05 December 2005 00:22, John F. Sowa wrote:
>
> I am not proposing that *any* category be fixed for all time.
> On the contrary, my recommendation is to make the ontology
> modular so that you can replace theories as units -- where
> any theory could be as large or as small as you like.
That sounds good. But do our ambitions have to be so modest? Simply
identifying the need to let theories change is only half the story. We are
moving in the direction of dynamic change. Shouldn't we be concentrating
already on the process of change and not only on the need for change?
I'm not saying the task is completely trivial, or that no intermediate steps
are necessary, but shouldn't we already state the goal and identify the
problem for what it is---a search problem?
Even if we retain fixed formal systems temporarily for some purposes, it is
important to at least identify the knowledge problem and the search problem
as one and the same on a theoretical level.
The world's knowledge management industry is moving to the search model by
default anyway. Perhaps it is time the theory caught up with the practice.
If we would once allow our own paradigm to shift, knowledge management might
be full of insights into ways of managing the ad-hoc categories of search
results. Instead we are separated by a cultural divide which refuses to see
search results as categories with the same status as the carefully
hand-crafted categories of ontologies.
On the other hand the search industry may be able to provide us with some
clues how to navigate among a universe of potential theories and find the
theory appropriate to a given problem.
Viewed as a search problem we are not constantly in retreat. New perspectives
open up, we can start to move forward and look at the process of generating
theories, not ever further backwards away from any given theory.
Obviously I'm talking from the perspective of a model of language which works
this way (search.) But as your earlier note points out, isn't such a model
(language) what we want?
-Rob