SUO: Re: "CAD Data Models", "FOL Type Stuff", & Translators

[Jon Awbrey <jawbrey@oakland.edu>]

¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤

Michael Uschold wrote:
> 
> [Matthew West, to] Phil Jackson said:
> 
> > The focus of TC184/SC4 is technical data exchange and integration.
> > This can be from property data to CAD data to product catalogues.
> >
> > There is quite a lot that has gone on in the area (I think) you
> > are talking about.  ISO 10303-50 (mathematical representation)
> > and ISO 10303-45 (material properties) come to mind.
> >
> > Of course the basis is data models, rather than FOL type stuff,
> > but as I am increasingly realising the gap between the two is
> > narrower than most people think.  It is just that very few
> > try to cross it.
> 
> This is a good observation, and I am one of those few.  I recently attended
> a mini-conference on product data exchange where most of the emphasis was on
> CAD data.  I have been struggling a long time to figure out what exactly the
> difference is between "data models" and "FOL type stuff", and to understand
> the extent to which the AI/FOL techniques have anything to offer that is of
> value in that arena.  I have not made a lot of progress, I'm afraid ...

One way to think about the series of transitions that we observe as we
pass up and down the spectrum (quantum or continuous) from data bases,
through data models, via deductive data bases, all the way up to the
kinds of full-fledged (non-plucked) species of rational animations
that we may yet hope to find or to make in our ultimate forms of
logical representations, is to view this as analogous to the
relation between the data of our experience and the laws of
our most general theories that we manage through the steps
of abductive and inductive so-called "inference", however
often these non-demonstrative styles of reasoning may
appear to be little more than just plain "dumb luck".

Just A Thought,

Jon Awbrey

¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤~~~~~~~~~¤