Re: SUO: Re: Proposed SUO Content Outline
Douglas,
I'd like to agree heartily with your comment below about backing up
suggestions with axioms. I suspect it's a lot easier for people to
disagree about abstract concepts when using English to describe different
approaches. There's just too much room for ambiguity and
misunderstanding. If folks who are suggesting alternate approaches could
axiomatize those approaches, I suspect we could either reach agreement more
quickly, or at the very least, define precisely where differences of
opinion lie. The latter, in and of itself, would be a very valuable
scientific result.
Adam
At 09:02 AM 3/3/2001 -0800, Douglas McDavid wrote:
>Pat --
>
>I'm going to rethread this, putting my original reply to John Sowa at the
>bottom,
>and some responses inserted within your reply to me.
>
> > That is a good question, and your point is well taken. But look where
> > it leaves us. We will need to write axioms which say that spatial
> > coordinate systems exist, without saying anything in particular about
> > any spatial coordinate system.
>
>Actually, the few statements I made were meant to imply that there are
>a lot of things we can say about coordinate systems. Some are very
>general, and would apply to all coordinate systems. Some are more
>particular, and would allow us to differentiate different kinds of
>coordinate systems.
>
>Many of those things that would be said in an upper ontology would
>have to do with point of view. This would articulate under what
>conditions a particular (type of) coordinate system would be
>appropriate.
>
>My expectation coming into this group was that an SUO would be used
>to express these kinds of issues. In general I would like to see us
>spending less time trying to convince each other of some particular
>point of view, and more time working on how to reconcile multiple
>points of view, through axioms (or whatever representation) that
>articulate what is common, what needs to be differentiated, and for
>what purpose.
>
> > How do we do that? What can be
> > usefully said about spatial coordinate systems in general, or about
> > the general concept of a spatial coordinate system? I do not mean to
> > imply that this task is impossible, but I do think it is considerably
> > harder than the task of describing one or two coordinate systems and
> > the mappings between them. At the very least, such a limited exercise
> > might be a very useful preliminary study before attempting to write
> > the mathematical masterpiece of generality required for a universal
> > theory of coordinate systems in the abstract. The basic
> > methodological point being that these 'upper' theories, if they are
> > to be nontrivial, are likely to be very difficult to get right. We
> > are trying to break the sound barrier before we know how to walk.
>
>Maybe I'm overly naive, but I didn't think I was proposing anything
>like a mathematical masterpiece. My view of the SUO is that it carves
>information spaces into useful areas of particularity. I see it being used
>as an organizing and indexing mechanism that acts as a framework for myriad
>particular ontologies. I believe the world is ready for a huge new
>emphasis
>on the *meaning* within software-based information systems. We're only
>recently in this position, with the critical mass of computing power and
>interoperation among enterprises of all types now in place. I think that
>significant progress can be made on this program of meaning reconciliation
>in a matter of months, rather than decades. I have a sinking feeling that
>this puts my out of step with the most prolific members of this group, for
>whom some issues of fundamental point of view need to be resolved before
>any progress can be made. My hope is that we can deal with many issues
>without making particularistic commitments at the uppermost level.
>
>Does this make sense?
>
> >
> > Pat Hayes
> >
> > ---------------------------------------------------------------------
> > IHMC (850)434 8903 home
> > 40 South Alcaniz St. (850)202 4416 office
> > Pensacola, FL 32501 (850)202 4440 fax
> > phayes@ai.uwf.edu
> > http://www.coginst.uwf.edu/~phayes
> >
> >
> > >John --
> > >
> > >It seems to me that an upper-level ontology should strive to avoid most
>of
> > >the problems that you outline below. The kinds of things that an
>uppermost
> > >ontology could say, that would actually be useful to organize subtending
> > >ontologies would be things along the lines of:
> > >
> > >There are spacial coordinate systems.
> > >There can be any number of such coordinate systems.
> > >Each coordinate system exists in relationship to some specified
>reference
> > >point.
> > >Coordinates from one system can be mapped onto coordinates of another
> > >system.
> > >Some coordinate systems are based on fixed units of length.
> > >Some coordinate systems are based on dividing a geometrical figure into
>a
> > >fixed number of units.
> > >Etc.
> > >
> > >I wonder why anything in an upper ontology would ever be so particular
>as
> > >to require reference to any specific spacial coordinate system
>whatsoever?
>
>
>Doug McDavid
>
>Member of IBM Academy of Technology
>mcdavid@us.ibm.com -- 916-549-4600
-----------------
Adam Pease
Teknowledge
(650) 424-0500 x571