RE: SUO: RE: Re: More KIF-ified Ontology Content
Pat,
. Further comments interspersed below, prefaced "GH2>
".
.
*
.
.
.
Cheers Graham Horn
National Data Standards Unit
Australian Institute of Health and Welfare
================================================
Phone: +61.2.6244.1094
Fax: +61.2.6244.1199
Email: Graham.Horn@aihw.gov.au <mailto:graham.horn@aihw.gov.au>
-----Original Message-----
From: pat hayes [mailto:phayes@ai.uwf.edu]
Sent: Friday, January 19, 2001 6:41 AM
To: Horn, Graham
Cc: standard-upper-ontology@majordomo.ieee.org
Subject: RE: SUO: RE: Re: More KIF-ified Ontology Content
> > . Until you asked this question, I hadn't ever
> > considered intransitive verbs. I suspect they still can be
> > expressed with either a direct or indirect object, or both.
> > Eg. I thought a thought. If I ran, it will have had some
> > attributes, such as manner, speed, and location. I'll
> > think on this further, but fir now, I'll keep with the
> > proposition that actions are:
> > * relations; and
> > * 4-D.
>
>MW: My philosophical problem with this is that relations are
classifications
>of tuples, and tuples are essentially some special sets, so you have
>activities as sets which are 4D, where as sets are timeless.
>
>GH> I'm afraid I don't know the necessary theory to appreciate why
>relations are classifications of tuples.
The SUO will consist of axioms in a logical notation. The basic semantic
rules for interpreting such logics specify that a relation symbol denotes
the extension of the relation, ie a set of n-tuples (for an n-ary relation)
of individuals (or whatever the relation is a relation on: in the proposed
new SUO-KIF, we will allow relations on relations also.)
GH2> (Eyes cloud over at the word extension.)
This is a very basic assumption about the semantics of relations. In a modal
logic, a relation symbol denotes something more complex, in that it has an
extension in every possible world (ie can be thought of as a function from
possible worlds to relational extensions), but it is still essentially
extensional in nature.
>Activities ARE 4D objects, we sometimes REPRESENT them as relations. This
>distinction is very important. One of my reasons for not using relations to
>represent activities, is because of this confusion.
>
>GH> This makes me again wonder whether what you are calling relations
>aren't, in fact, types of relations rather than instances of them?
Maybe it would be useful if you were to tell us what YOU mean by 'relation'?
I havnt been able to follow several of your messages on this topic. For
example, I don't know any sense in which a relation could be said to be
4-dimensional (or indeed n-dimensional for any n).
GH2> Here is an example.
GH2> From a business modelling perspective, suppose we have a supplier, a
customer and a product. Relations(hips) between the supplier and customer
include:
* advice of conditions for a transaction,
* order for product,
* delivery of for the product. delivered
* billing for cost for the product. delivered
* payment of money for the product. delivered
GH2> In this case all the relations are communications. If we used mail,
all would use the same transport mechanism. These days, more likely 4 would
be information transfer (especially if one accepts money as a form of
information), and the other one a physical delivery.
GH2> In other cases the relations could be other sorts of processes, such
as transformation (eg. during manufacture). As I mentioned before, they
could also be processes like learning, and interpersonal relations.
GH2> But in each case the relation has time related characteristics, such
as stages and/or varying magnitudes. Many also have spatial attributes, ie
they take place at/between/around/etc. locations.
GH2> I wonder whether we can get to a stage where we can appreciate each
other's concepts of relation. In my case, the next thing I need to
understand is the meaning of "extension", as you used it above.
GH2> I feel there is an important aspect to this. That is that the SUO
should be able to be comprehended by its users. Only that way, will they be
able to use it with confidence of being accurate.
GH2> Therefore, I suggest we are going to have to find means of
describing the underlying principles in ordinary, everyday language that the
non-cognoscenti are, at least in principle, able to understand. I believe
our target user population should not need to be graduates of high-flown
philosophy or logic courses to be able to understand the SUO.
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 <mailto:phayes@ai.uwf.edu>
http://www.coginst.uwf.edu/~phayes <http://www.coginst.uwf.edu/~phayes>