[topicmapmail] Two Models of Facets

[Murray Altheim]

Carlo Moneti wrote:
[...]
> I wonder if semantics is causing problems here. After implementing a 
> faceted classification navigator/search engine to better understand faceted 
> classification, I came to the following (street language) interpretation 
> vis-a-vis facets vs properties:
> 
> 1) The  property, "blue eyes", belongs to Jim specifically (I agree).
> 
> 2) "Blue eyes" is the value of  a property called, perhaps, "eye-color";.
> 
> 3) The property, "eye-color" is general to all persons.
> 
> We can create an eye-color facet with several categories, e.g., 
> R,O,Y,G,B,I,V. In the "Persons" records,  the property "eye-color" value is 
> used to associate the record of (topic of) person Bill to category B of the 
> eye-color facet.

I think informally this agrees with my understanding as well.

> Properties and facets are clearly different, but it seems fair to say that 
> facets are built from properties, or should I say, property-classes?  
> Anyway, if someone says, "Facets are basically properties", it's probably 
> OK for general conversation, though it may obscure the subtle distinctions 
> that may help clarify how to define facets in topic maps. 

The difficulty is of course that in trying to develop a standard
way for people to create and use facets and/or properties, to
distinguish the differences (since there are differences), it's
important to define and clarify, rather than conflate definitions,
i.e., "street language" won't quite do here.

When you say "property-classes" you're quite close (I think) to what I'm
getting at. The example of eye colour is a bit misleading because colour
is actually a continual spectrum. Nobody can be said to have exactly
"blue" eyes unless somebody defines "blue" as a specific value or a
range, and then, as we know, eyes can contain multiple colours. Mine,
for example, are both brown and green, listed as "hazel" (inaccurately)
on my driver's license, which didn't have a check box for "brown and
green".

Weight is a better example. When described for facets (as per my
working definition), "weight" is an abstract property, whereas
for a property (as per my working definition) of an individual,
"weight" is a measurement. You can't have an abstract property,
applied to an individual, and you can't have a measurement
applied to a class. So, I'm pushing the idea that the above idea
of "abstract property" (what you're calling "property-class")
might be our searched-for definition for "facet", which seems to
agree with that found in Faceted Classification.

BTW, thanks very much for your input -- I'd be curious to know
what you thought about Topic Maps' application to Faceted
Classification, since I'm currently in the process of trying to
see the overlaps for purposes of implementation.

Murray

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

   International lawyers and anti-war campaigners reacted with
   astonishment yesterday after the influential Pentagon hawk
   Richard Perle conceded that the invasion of Iraq had been illegal.
   http://www.guardian.co.uk/uk_news/story/0,3604,1089042,00.html

     "...if it's the case that international law doesn't permit
      unilateral pre-emptive action without the authority of the
      UN, then the defect is in international law."

   Thank God for the death of the UN, by Richard Perle
   http://www.guardian.co.uk/Iraq/Story/0,2763,918812,00.html