[topicmapmail] Are Facets Really Simple After All?

[Murray Altheim]

Carlo Moneti wrote:
> On 2003.11.30 11:25 Thomas B. Passin wrote:
>
>>If this is a reasonable way to look at facets, then I do not see why 
>>using facets would be different in any essential way from classifying 
>>using a single hierarchy.  A given topic could be related to one of the 
>>faceted terms, say "location", by an association called, perhaps, 
>>"hasFacet".  You would find out which particular facet tree that 
>>"location" is in by walking up to the root of its tree to arrive at 
>>"Geography", whose type would be "Facet".  Of course, a real 
>>implementation would use some kind of shortcut to avoid actually walking 
>>the tree each time.
>>
>>Is there something I am missing here, or is it really this simple?
>
> I have come to the same conclusion as Thomas. Just last night I resolved 
> the following topic and association types:
> 
> topic type to define a facet (#facet-root)
> topic type to define a facet category (#facet-category)
> association type for parent/child relation of facet-categories 
> (#facet-parent-child)
> association type for topic as member of a facet-category (#facet-member)
> 
> This goes a long way in addressing my concern (perhaps needless) about 
> polluting topic info with facet info, which seems to me like application 
> infrastructure detail rather than true topic info. True topics won't be 
> polluted in their occurrence data (as Murray also worried); and facet  
> hierarchies, being differentiated from others, can be processed distinctly 
> (turned on or off, as Murray pointed out). Also, many facet hierarchies may 
> be generic enough to have a double use as non facet-related hierarchies.

It's this last notion that leads me to avoid using explicit facet relations
in the hierarchy itself. Any topic in an ontology may be used as a facet of
another, and I think the inherent (non-facet) relations of each entity are
more important and should predominate. E.g., in my authoring ontology I have
a whole taxonomic hierarchy of entities from "location" all the way up to
my root ("thing"), plus each of these entities may have non-taxonomic relations
that are important as well. Attaching a facet to an entity entails not only
the facet's hierarchy, but any of those other relation types.

I guess put simply, I don't think there really is a distinct facet hierarchy
apart from any hierarchies or relations a facet by its use already entails.

> This approach is convenient in that it addressess easily and intuitively (I 
> think) the mapping of facet info from XTM to a relational database that 
> some application will query to generate a faceted navigation experience. 
> Now, perhaps this isn't ambitious enough. Perhaps there is in theory a way 
> to inform a topic map and topic map engine of an equivalent of the 
> aforementioned application logic and db queries, so that the topic map 
> engine can generate a faceted navigation experience. This, however, is 
> beyond my knowledge. Is anyone working on this, or theorizing such 
> potential?

The reason I brought this thread up is because I am working on an
implementation of FC-over-TM.

Murray

......................................................................
Murray Altheim                    http://kmi.open.ac.uk/people/murray/
Knowledge Media Institute
The Open University, Milton Keynes, Bucks, MK7 6AA, UK               .

    "Mr Bush, you have to admit it's a pretty remarkable thing for a
     man just to go to a hotel room door and open it and have a woman
     standing there and have sex with her," said Marshall Davis Brown,
     lawyer for Sharon Bush. "It was very unusual," Bush said.
     http://www.nzherald.co.nz/storydisplay.cfm?storyID=3536318