SUO: Negotiation Instead of Legislation
In my recent notes, I have explained why I believe that legislated
ontologies, such as CYC, SUMO, and others, are misguided and obsolete.
Some people have objected that we have to start somewhere and that
we should build on what has already been done. I agree with both
of those principles, but I disagree with the suggested starting
points and the suggested direction in which to proceed.
Instead, I believe that our starting point must be the existing
legacy computer systems that have been implemented over the past
50 years and that won't go away for a long, long time. Furthermore,
I believe that new developments must proceed in an evolutionary way
from where we are to where we want to be instead of an all-or-nothing
jump into the unknown.
In my keynote presentation for the Knowledge Technologies Conference,
which will be given in Seattle on March 13th, I make those points,
and I discuss successful implementations that have worked on major
systems. Following are the slides for the talk:
http://www.jfsowa.com/talks/negotiat.htm
At the end of the slides are some recommended readings for further
background about these issues.
John Sowa