SUO: RE: RE: Re: Missing Ingredients
Tom Johnston wrote:
<snip/>
> I could be wrong. But all of us do this
> kind of thing, all the time, in all aspects of our lives. We think about
> what to do and, at some point, we act. The course of events will sometimes prove
> us to have acted too hastily. At other times, opportunities pass us by,
> the ship sails from the harbor without us, and so those courses of events
> prove us to have acted not hastily enough. Doing something before all
> questions are laid to rest is a risk, yes. But not doing something until all
> questions are laid to rest is also a risk. That's the background of my statement,
> none of your 1 - 3 above. As a pragmatist, I would think you might approve of it.
>
> Tom
<snip/>
Very well stated! This process itself, of reflecting, judging, acting, and
monitoring results, is what ought to be encoded into the top level of the
ontology (just below nil). All measures of effectiveness (MOEs) should
be instrumented in response to this process's needs for information.
The closest, lightly formalized thing I've found to a good description of
this process is at:
http://www.isi.edu/~marsella/marsella-aamas03.pdf
I posted that link earlier, but nobody responded. If we're looking for
an active ontology, we'll need an agent that 'improves' the ontology
over time. This link shows some anthropomorphic concepts that are
based on psychology research results.
Comments welcomed.
Rich