Topic Maps, Scope and Subject Identity [RE: Taxonomy issues]

[Bernard Vatant]

Honorable astro-semantic folks

I've been just notified today the existence of this forum by Steve Pepper, one of the
Topic Maps "gurus", who gently remembered that I had both the hat of topic maps "expert"
and the one of (life-long amateur) astronomer. As a matter of fact, I've had recently some
exchanges with Ashish Mahabal about Topic Maps in the VO project, and I'm happy to jump on
the occasion to bring more Topic Maps flavor in this semantic pot :))

About Taxonomy issues, I want to confirm that Topic Maps "scope" is exactly intended to
handle those situations of multiple views on a single object: multiple names, multiple
classifications, multiple interpretations ... as Ashish has pointed.

Another issue that can be tackled by scope is controversial or hypothetical object
identity, e.g., "Is IR-Source A the same "object" than X-Source B?"

What TM approach will do in such a case is defining each of those as a distinct "subject"
(of conversation), distinct from each other and of any "object" out there - a distinction
in fact very crucial to Astronomy! Each "subject" is represented as a specific topic, say
"IR_A" and "X_B".

Information about each of those subjects (classifications, observations, data,
interpretation, papers, ...) will be attached to the respective topics. If at some point
in time the two sources are somehow "proven" to be different "manifestations of the same
object", say "Quasar_Z", you can either:

-- Merge the three topics, but keep "IR_A" and "X_B" characteristics distinct through
scope mechanism.

-- Keep the topics classified as "sources" independent of the topic classified "object",
and link them through "object-manifestation" associations, asserting that, e.g.

In (the scope of) IR bandwidth observations, "Quasar_Z" is identified as "IR_A".

If you want to add the information that this identification is still controversial, it can
be added as a scope element too ...


Hope that helps

Bernard

-------------------------------------------------------------------
Bernard Vatant
www.universimmedia.com

Consultant - Mondeca
www.mondeca.com

Chair - OASIS TM PubSubj Technical Committee
www.oasis-open.org/committees/tm-pubsubj/
-------------------------------------------------------------------