Re: [xtm-wg] The Nature of Things...

[<guy.a.lukes@frb.gov>]

I have been investigation the philosophy of Peirce, that is sometimes
mentioned on this list, and wanted to relate it to some of the ideas that
are coming out of the XTM modeling group.

First Peirce makes the distinction between:

ICON      Is an intersection of property values given meaning by capturing
a quality of experience
INDEX          Is a token given meaning by pointing to an existent thing or
event
SYMBOL    Is a type given meaning by a structured network of typed
connections to other symbols

A symbol can be treated (cast) as an index or an icon.  An index can be
treated as an icon.  An index cannot represent the distinctions of a
symbol, and an icon cannot represent the distinctions of either an index or
a symbol.  This suggests a class inheritance relationship between the three
concepts.

These three concepts are also very close to the topic map TOPIC, OCCURRENCE
and FACET concepts, where ASSOCIATIONS and their ROLES provide the
structure of the network of connections between symbols.  These connections
provide an interpretation of the relationship between symbols.  Different
interpretations will require a different network of associations.
Therefore, Peirce talks about how a coherent set of associations needs to
be linked to an interpretant. Although there may be many different
interpretations for the same set of symbols, each interpretation would have
to be in its own sub-network of associations.

In Topic Maps this idea of interpretant is very close to the concept of
SCOPE.  Scope provides a discriminant to disambiguate some distinction of
interpretation.  In theory, this could be done by establishing connections
to detectable property values (iconic), establishing a connection to an
existing thing or event (indexical), or establishing a connection to an
abstract type (symbolic).  In this case, since an interpretant is trying to
disambiguate associations between topic/symbols, a connection to an
abstract type by a typed link seems the appropriate choice.  In this case
scope just becomes a built in role with predefined semantics.

If there are going to be structured levels of interpretation then
ASSOCIATION should inherit from TOPIC.  This would allow second order
associations (structuring interpretation) to organize first order
associations (cast as topics).  These first order associations could then
structure normal topic "primitives" in the standard way using their
association interface.

While all of this could be exported as RDF for transport, it would loose
the inherent semantics for the distinctions that are intrinsic to a Topic
Map.

While a specific example would probably help to make this a lot clearer, I
think I should probably stop and see if anyone else finds these ideas
interesting, useful  or misguided.

Guy

PS
I have never seen an example of XTM that included the use of facets.  I
would be particularly interest in an example that includes an association.
If facets are to be replaced by something else, an example of the new
syntax would also be interesting.




-------------------------- eGroups Sponsor -------------------------~-~>
Create your business web site your way now at Bigstep.com.
It's the fast, easy way to get online, to promote your business,
and to sell your products and services. Try Bigstep.com now.
http://click.egroups.com/1/9183/4/_/337252/_/973172313/
---------------------------------------------------------------------_->

To Post a message, send it to:   xtm-wg@eGroups.com

To Unsubscribe, send a blank message to: xtm-wg-unsubscribe@eGroups.com