FW: [Ontology] UCDs vs ontologies?

[Tony Linde]

[forwarded from dm to semantics]

Hi Sebastien,

> That way, instead of having to define zillions of slots (i.e. at least 
> one per new subclass of Property) and writing:
>
> MyConcept hasContact Contact
> MyConcept hasHow How
> MyConcept hasWhat What
> ... and as many as there are different possible properties
>
> we could simply write things like:
>
> MyConcept hasProperty Property  (with multiple cardinality, this would 
> cover all the above: no need to predefine all possible cases)

This is in line with something I read when I first started looking into
ontologies a few years ago, that one way of doing things is to have <10
properties of the generic "has a", "is a" variety.

However, since then I've also read that providing more specific properties
can also reduce confusion if the property has a human-readable /
understandable name. If someone sees that class "Cow" has property
"hasLegs", then you can write Cow hasLegs Legs (4), whereas if "Cow" had
"hasProperty", then you could end up with Cow hasProperty Beak.

Of course, if the ontology goes the first route of having only a few generic
properties, then one only has to be omnicient for each class, so
"hasProperty" could be restricted to "value", "unit", "WhereWhen" or "Legs"
for any given class.

So: many properties or few properties? I could try out two different
versions while the VOEvent ontology is still quite small and see what people
think.

Elizabeth