Taxonomy issues

[Anita Richards]

Thanks to Carole and Sean, I think I understand what I was trying to say
earlier better now....


> Of course in biology there are different classifications and
> definitions -- there is no standard definition for what a gene is for
> example. DAML+OIL allows multiple definitions, and will report on
> definitions that are inconsistent. What you do then is an interesting
> story. And it depends on what you want to use the ontology / taxonomy
> for.

It is much worse in astronomy though because no-one dies or even sues if
we define something in ten different ways...  So I do not think the people
constructing ontologies should have to sort out genuine inconsistencies
(just avoid accidentally creating more ourselves!)

> So, onto the subject of contradictions and inconsistency. I say that
> all Widgets are blue, while you think that Widgets are red. So in our
> ontology, I say:
> 
>   Widget -> colour blue
> 
> and you say
> 
>   Widget -> colour red
> 
> If we also have some constraints that say that blue and red are
> mutually exclusive, then we have arrived at a contradiction. 

This implies that we should avoid mutually exclusive definitions within an
ontology for astronomy, because people can disagree about definitions, or
definitions can change with time.  For example, an optical astronomer
might say that all stars with an emission line at 500 nm is a green fairy,
and those with a line at 700 nm are red fairies.  A radio astromoner might
then use an insensitive survey to find that only red fairies had radio
emission and classify all radio stars as red fairies.  However if green
fairies do have weak radio emission, some radio stars in a more sensitive
survey could be misclassified.

So, suppose a star cannot actually be both, we (as in the ontology
authors) do not want to have to decide, and we do not want to throw out
the information either.

I suppose this is really the 'individual' case, but there are also cases
of contradictory definitions of precisely measured properties, e.g.
whether a red fairy is a variable star with a period of <400 days or <350
days and a green fairy has a period of >400 or >350 days.....

But I think the implications for an ontology user are the same: if I was
searching for all the green fairies in a galaxy, I would want a warning
about those which were classified as red and as green in radio and optical
catalogues respectively as well as those which were classified according
to an ambiguous definition (periods of around 375 days).

What I think would be useful is if we could constrain properties to be
"contradictory" as a description, of perhaps just as "linked", so that any
query which involved returning a description or list of (a) green
fairy(ies)  would automatically note any also classified as red....

> One solution to this is to use the classification as a "conceptual
> coat rack" and hang "extralogical" information off it. This is the
> kind of thing that the RuleML initiative is aimed at. This then allows
> us to state things about OWL classes that are not necessarily part of
> the ontological or intensional definition of the class, but which we
> want to make available to applications.

Um, is that similar to what I am suggesting above?  



- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
Dr. Anita M. S. Richards, AVO Astronomer
MERLIN/VLBI National Facility, University of Manchester, 
Jodrell Bank Observatory, Macclesfield, Cheshire SK11 9DL, U.K.           
tel +44 (0)1477 572683 (direct); 571321 (switchboard); 571618 (fax).