RE: SUO: RE: CYC event vs. SUMO Process -- really different?

[Richard Cooper <rich@valutech.com>]

Title: RE: SUO: RE: CYC event vs. SUMO Process -- really different?

Patrick Cassidy wrote
> >
> > 1. Require events to occur at locations (in regions) because
> >    there are events that are not spatial at all, e.g.,
> >    "what if Bob thought about Mary's birthday party?"
> >    is a hypothetical, with no spatial relationships at all.
> >    It shouldn't be necessary to represent this in any
> >    spatial way.
> >
>      I'm not sure which event you are referring to, the
> Bob thinking event or the Mary's Birthday party event,

All three events:
1. Mary's Birthday party
2. Bob's thinking about 1 above
3. The hypothetical about 2 above

All of these can, or can not, have relevant 4-space
coordinates, but need not have them. 

Mary's birthday party may be located at a specific
4-space coordinate when it occurs. 

Bob can think
about Mary's birthday party months before it occurs,
when there is no need to determine where Mary's birthday
party will be located if and when it occurs. 

Somebody (probably Mary) could be hypothesizing about
Bob's thinking about her party in yet another spatial
way - she repeatedly thinks about whether Bob will
remember her birthday and throw a party for her, and
that repeated sequence of Mary's thoughts can have
a distinct 4-space location at each element, or at
no elements. 

The point is that a 4-space location of all these
events is an unnecessary complication in any terse
representation of any one of these events.  Yet I should
be able to use temporal logic to conclude that Mary's
thinking about Bob's thinking occurs at a time that is
unrelated to Bob's thinking about Mary's birthday party
because there is no reason to conclude these two events
are related in any way. 

But I can use common sense reasoning to conclude that
Bob's thinking about Mary's birthday party can occur
before, during, and/or after Mary's birthday party.

There seems no justification to force a 4-space relationship
onto situations in which it may not be appropriate.  Just
as there seems no justification for eliminating a 4-space
relationship onto those situations. 

The point is that 4-space relationships are modular;
they can be applied to nearly any situation, but are
largely irrelevant to the vast majority of situations.
Therefore 4-space is unrelated to most events, yet can
be related to nearly any event when appropriate.  Therefore
4-space representation is not justifiably a default add-on
to any general situation.  It is appropriate to engineering
problems, scene analysis, medical records, and many specific
classes of situations, but not to the general concept of
an event, which is far more general than 4-space representation.

HTH,
Rich

> but both of those (if they occurred) would have a spatial
> location.  With the data in the sentence, the spatial
> location is not known.  It is the rule rather than the
> exception that some necessary relations or properties of
> things mentioned in text (especially future events and more
> especially in sentences taken out of context) are not known
> from the text and their values in the representation of the
> text must be left empty, or filled with a default if the
> programmer chooses that tactic (e.g. Earth's Surface).
>     The fact that the events are hypothetical would need to be
> represented, perhaps by a wrapper around the sentence within
> the "if" clause, but the representation of hypotheticals
> is a different topic.  To say that an event marked as
> hypothetical cannot have a location is like saying that
> "Bob thought about mary's birthday party" cannot
> be a sentence because it is enclosed within another
> sentence.  Both logic and language can encapsulate
> structures within larger structures.
>
>
> > or
> >
> > 2. Make the mathematics of temporal logic unavailable to
> >    any class of events, whether spatial or nonspatial in
> >    description.
> >
>     I don't know what point you are making here.  The
> temporal location of events is point 3.4 of my suggestion,
> and temporal location is in both Cyc and SUMO.  What,
> more specifically, is the problem you see?
>
>
>    Pat
>
> =============================================
> Patrick Cassidy
>
> MICRA, Inc.                      || (908) 561-3416
> 735 Belvidere Ave.               || (908) 668-5252 (if no answer)
> Plainfield, NJ 07062-2054        || (908) 668-5904 (fax)
>                              
> internet:   cassidy@micra.com
> =============================================
>