Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

RE: CG: Re: Re: Whole and Parts (and boundaries)



Title: RE: CG: Re: Re: Whole and Parts (and boundaries)

Doug

 

I think that  on the problem of long border crossing we have at more than the two “major ontological categories” you propose, or at least that some of the concepts might have to be extended.  I agree that the 2 that you propose are a useful starting point:

 

  1. an arbitrary physical border between tracts of land that are respectively
    claimed by two human social sovereignties. 
  2. the social border between those social sovereignties. 

 

If I have these modeled I should be able to answer quite a few questions about long borders.

But I can think of exceptions to this way of defining long border in the first sense.  Two sovereignties may claim territory that intersect and thus we have a “disputed border” in which there may be little or no physical border BETWEEN the tracts.  So perhaps I have a different category or a sub-type.

 

I don’t know if you followed Sowa’s earlier link to an article on “knowledge soup”  In his discussion  he offered some examples of problems with tight, neat definitions of concepts.  Applying some of his problem types to the border issue I came up with the following examples.

 

Overgeneralizations:  (example is Birds fly, but what about penguins?)  Borders separate territories, but what if it is a disputed border?

 

Abnormal conditions:  You cross borders, but what if there is

 a fight at the border? What if there is a terrorist alert?  Your passport has expired? There is an earthquake?

Incomplete definitions:  Is the concept of “no man’s land” included?  Does it cover boundaries between things like the atmosphere and “outer space”

 

Conflicting defaults:  You can see the Demasiado Corazon soap opera in Mexico, but not on US stations.   Except you can get it via a satellite broadcast.  Just as you can a TBS show in Mexico.

 

The point might be that when we try to build our formal models we are not typically including all this knowledge that humans know as adaptive agents working pragmatically based on interacting with the world.

 

Even in infancy we seem to distinguish 2 types of object categorization. One is perceptual categorization, which is part of perceptual processing based on perceptual similarity of one object to another. As we develop we create perceptual schemas of what objects look like.

 

Older infants develop a conceptual categorization, that seems to be based on what objects do.  One way of thinking of this more abstract type is as a restructuring of perceptual information into conceptual form.  One basis for this is the experience of  paths that objects take and the interactions among objects along these paths. Experience creates a simple mental model of a notion of kinds, such as animals, vehicles, furniture, plants etc.. Underlying this kind of categorization seems to be functional roles played  in events, rather than the physical appearance of the objects. Speculating we might propose that evolution has selected us to be able to build schemas involving such broad categories to operate effectively in the world. Unanticipated applications come up all the time and we have to accommodate to the new demands.  We can speculate about pragmatic based processes to do this.  If some concept becomes the “expectation” part of a sufficiently reliable schema, then this concept is compared with the concepts at the basic level of abstractedness and a process tries to determine the common parts between the concepts. If the difference is less than a certain amount, then a new abstract concept might be added.   Sowa has a more structured discussion of a pragmatic cycle.

 

 

While human cognition and knowledge can do this, our formal cognition and knowledge bases still have difficulty with this.  For one thing we haven’t fleshed out architectures for the pragmatic cycle.

 

Gary Berg-Cross

-----Original Message-----
From: Douglas McDavid [mailto:mcdavid@us.ibm.com]
Sent: Tue 7/19/2005 7:31 PM
To: Gary Berg-Cross
Cc: John A. Bateman; Murray Altheim; owner-standard-upper-ontology@IEEE.ORG; sowa@BESTWEB.NET; standard-upper-ontology@IEEE.ORG
Subject: RE: CG: Re: Re: Whole and Parts (and boundaries)

With respect to the problem of long border crossings, it seems to me that
we have at least two major ontological categories in play here.  One is the
arbitrary physical border between tracts of land that are respectively
claimed by two human social sovereignties.  The other is the social border
between those social sovereignties.  The former is a fictional plane
(height and length, but no depth) separating the territory of two
geopolitical jurisdictions.  This aspect of the boundary is physically
crossed in a second or less, once one is in a socially negotiated position
to do so.  The other boundary is a power nexus which must be negotiated
prior to making the final, physical movement through the arbitrary
geophysical demarkation plane.

Yes?


Doug McDavid

Business Transformation Architect
IBM Academy of Technology - http://www-306.ibm.com/ibm/academy/index.html
mcdavid@us.ibm.com
408-927-1565 (IBM tie-line: 457)


|---------+-------------------------------------->
|         |           "Gary Berg-Cross"          |
|         |           <gary.berg-cross@em-i.com> |
|         |           Sent by:                   |
|         |           owner-standard-upper-ontolo|
|         |           gy@IEEE.ORG                |
|         |                                      |
|         |                                      |
|         |           07/14/2005 01:44 PM        |
|---------+-------------------------------------->
  >---------------------------------------------------------------------------------------------------------------------------|
  |                                                                                                                           |
  |       To:       "John A. Bateman" <bateman@UNI-BREMEN.DE>, "Murray Altheim" <m.altheim@open.ac.uk>                        |
  |       cc:       <sowa@BESTWEB.NET>, <standard-upper-ontology@IEEE.ORG>                                                    |
  |       Subject:  RE: CG: Re: Re: Whole and Parts (and boundaries)                                                          |
  >---------------------------------------------------------------------------------------------------------------------------|





I generally agree with John Bateman's contextual discussion on  goal and
functions affecting what boundary means (“ontological characterizations.”
As John described it).  The experimental work on robot-human interactions
in spatial settings provides real data as I suggested in my earlier message
drawing on epigenetic robotics.

I think that his statement
>"It took us three hours to cross the border."…..Sounds pretty thick to me.

suggests how linguistic data gives very useful indications.  It’s thru our
own mental models which we use in turn for

> modelling decisions relevant for building
>ontologies, but only when listened to.

There is a degree of commonality to our own experience based mental models
as designers of ontological models from insights offered by linguistic data
we have to be aware that our lifelong developed mental models will differ.
That’s one reason that letting robots develop their own models rather than
building them in offers some check on unconscious bias.  Of course we have
realize that our general design principles for developing cognition may
also be flawed and have to be reformulated.

Gary Berg-Cross


      -----Original Message-----
      From: owner-standard-upper-ontology@ieee.org on behalf of John A.
      Bateman
      Sent: Thu 7/14/2005 7:57 AM
      To: Murray Altheim
      Cc: sowa@BESTWEB.NET; standard-upper-ontology@IEEE.ORG
      Subject: Re: CG: Re: Re: Whole and Parts (and boundaries)



      > I have no real data about people talking about boundaries,
      > but I have some info about people talking about walls --
      > i.e., whether a wall is part of a room or a boundary
      > between rooms or whether part of the wall is part of the
      > room or part of the wall is part of the boundary.
      >
      > If you ask them, people have an incredible number of
      > different thoughts ranging from confusion, to irritation,
      > to lengthy and irrelevant discourses on the topic -- and
      > essentially all of those responses are useless for any
      > serious kind of application.

      These kinds of issues are ABSOLUTELY ESSENTIAL for
      any one to make any statements about 'how people
      do things'. As John S. says, if you haven't collected
      (or have access to) the data where people are actually
      going about tasks trying to solve problems involving
      the concepts, then it is better to say nothing.

      In my own group, putting people in situations where they
      have to talk about and solve spatial problems is
      the fundamental methodology without which we say
      nothing about 'how people' do stuff
      (http://www.sfbtr8.uni-bremen.de/i1). On the basis
      of these experiments (usually robot-human
      interactions in spatial settings),
      we propose ontological characterisations.
      Absolutely obvious, is that characterisations depend
      on goal and functions. So even the word 'boundary'
      is not yet sufficiently defined to say much that
      is sensible, but as soon as it is used in context,
      that complexity usually vanishes.

      Example:
      > For example, when lay people (as opposed to geographers
      >>or politicians) talk about the boundary between two
      >>countries, they generally don't think about the thickness
      >>of the boundary.

      Absolute nonsense. Data:

      "It took us three hours to cross the border."

      Sounds pretty thick to me.

      As soon as one then attempts to 'explain away' the
      usage (e.g., "well, they were not strictly talking about
      the border but about the road up to the crossing point,
      which is actually a line of no thickness"), I think
      the point has been missed. Either you use the
      linguistic data or not. If not, fine, then we
      can go with boundaries defined mereotopologically,
      fiat or otherwise). If linguistic data
      ('how people think about...'),
      then it should be done properly. That's my
      only point here I think. I believe the
      linguistic data do give very useful indications
      of modelling decisions relevant for building
      ontologies, but only when listened to.

      John B.