[topicmapmail] Hierarchy PSIs

[Thomas B. Passin]

Bernard Vatant wrote:

> Sure enough, that was my point. There are hierarchies and hierarchies, and
> you'd better know which one you use. 

> ... I'm not sure what you mean exactly here, Murray, but if you want to make
> the point that there are a variety of theoretical contexts (theories) that
> a topic map (or any other knowledge representation) can commit to, and that
> having a single PSI for "hierarchical relations" without precision of this
> context is too much fuzzy, then I agree completely.
> 

Just to round things off, here is what Sowa says about hierarchies, on 
page 494 of "Knowledge Representation" -

*hierarchy*. A partial ordering of entities according to some relation.

He goes on to name some kinds of hierarchies, such as _type hierarchy_ 
(uses a _type-subtype relation_), _broader-narrower hierarchy_ (mixes 
_part-whole_ and _part_ hierarchies), etc.

Cheers,

Tom P