aam at astro.caltech.edu
Fri Sep 27 12:22:23 PDT 2002
The Taxonomy issue is a perfect context to illustrate the scoping
methodology that Topic Maps have (and other KO/KR manifestations have to
a varying extent under different names):
Taking up the examples that George provided, for each object rather than
putting it in a bin, we just leave it as an object (not very helpful) but
then have multiple scopes which tell us what all it has been / can be
classified as under different contexts. Each such scope, for a given
object, comes with a pointer that tells you where the "classification"
came from and that can take you to more links whcih tell you how dubious
or reliable the classification is.
Thus when you search for a particular scope within objects (radio Es in
the NGP region), you get all objects which have been defined with that
scope (rdio Es), and in addition you get to see all the other scopes
(classifications) those objects may have coming from other sources.
Decreasing positional error bars or multiple imaging (for example, under
better conditions) or other reasons may revise the classifications too and
the adopted system will have to have room for that.
Ashish Mahabal, Caltech Astronomy, Pasadena, CA 91125
http://www.astro.caltech.edu/~aam aam at astro.caltech.edu
Signatures are a waste of bandwidth.
More information about the semantics