Re: CG: Re: Re: Whole and Parts (and boundaries)
Murray,
I am very skeptical when anybody makes claims about what
anybody else thinks -- especially some typical or average
man in the street. That was the problem with Chomsky's
policy of trying to build theories of language use based
on somebody's "intuition".
>My purposes come down to trying to model the way people
>actually think about boundaries when discussing them, in
>a layperson's sense. I realize a generalized definition
>is all but impossible, so I suppose the closest thing I
>could come up with is "Murray's Ontology" in the same
>sense that Cyc is "Doug Lenat's Ontology" (and I do
>remember him saying there is some sense to that notion)
>as a notion of "common sense ontology," but I'm not
>trying to stress that terminology, which is overloaded.
In NLP, people have found that anybody's intuition about
language -- even a professional linguist's -- is likely
to be wrong. So they are now using large corpora of actual
language as it is used by people who are trying to do something
real -- communicate with other real people about real problems.
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.
If you want to do something useful, look at examples where an
answer would have some significant effect (i.e., a difference
that would affect what somebody would actually do that really
mattered). Idle parlor games about what people might or
might not say are useless for any serious kn. rep. task.
>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.
Do you really have any data about that? Who have you asked?
I've considered Barry's mereotopological
>Again, for people to make accurate mereological and
>topological statements in a pseudo-English. If there
>is a greater need for precision I expect that the
>ontology can be modified to suit.
I am very dubious about the project of creating detailed
ontologies outside the context of actual tasks that have
any practical dependencies on the answer.
John