[topicmapmail] Superclass-subclass indentation in the Omnigator

[Robert Barta]

On Thu, Dec 26, 2002 at 11:17:38PM -0500, Thomas B. Passin wrote:
> > If you where right and Arthropods is-a-Subclass-of Phylum
> > than we have a contradiction that Crustacea is-a-Subclass-of Phylum.
> > Which it is not.
> >
> 
.....
> So we say "genus" but what we mean is something like "classes that are on
> the 4-end of a third-level association"  "genus" and "species" are, I am
> pretty sure,  roles in topic map terms and not classes.

I think what makes this confusing to all of us, is that we tend to
look at a particular concept and then try to think _what it is_.

So, for instance (not again this word!), 'charlie'. The definition "it
is an instance if it cannot be subdivided" is not helpful, especially
for those who know how orang-utan liver tastes (you can divide
charlie, believe me).

One beauty of TMs is that the only thing which really counts are
associations. They make the statements, they provide "meaning". But
they only do it through a _relativistic_ view of things. It mirrors
reality in the sense as every one of us is not defined by him/herself,
but only by the various views (relative behaviors):

   - how to deal with particular humans
   - how to deal with students (why do I make a difference here :-)
   - how to deal with your car
   - how to interact with your food
   -...

> Anyway, I am convinced that we should be clear that we always and only mean
> "a concerete instance" when we say "instance-of" or "is-a" to avoid any
> possible confusion about whether "is-a" as applied to a class is intended to
> result in an instance or a subclass.

In the above sense there is no need to distinguish between "here is a
concrete thing" and "there is something abstract" to make the
distinction between "instance-of" and "subclasses":

  love (emotion) # love is an instance of an emotion

  dominating-love subclasses love

  submissive-love subclasses love

As a consequence this means that the choice is arbitrary. But so are
all our theories.

\rho