RE: rude categorisation

Not sure I understand where this conversation is leading. I absolutely
agree with all Leonard's points about using different metadata elements
to describe the "subject" and the "type" of a document. Other important
elements include its date (or period), its language, its provenance,
etc. For the purposes of categorisation, it is possible to use any of
these elements, or a combination of them, and all can be useful.

And, as Al points out, the term "categorisation" does not necessarily
mean sorting things by subject. If we want to limit it, we can instead
use the term "subject categorisation".

But we've moved away from SKOS itself into its application for indexing,
search and retrieval. Yes, interesting and important, but so are lots of
other aspects of search and retrieval - are we going to discuss all of
them? We're not thinking of extending SKOS so that it covers metadata
elements other than Subject, are we?

Cheers
Stella

*****************************************************
Stella Dextre Clarke
Information Consultant
Luke House, West Hendred, Wantage, Oxon, OX12 8RR, UK
Tel: 01235-833-298
Fax: 01235-863-298
SDClarke@LukeHouse.demon.co.uk
*****************************************************



-----Original Message-----
From: public-esw-thes-request@w3.org
[mailto:public-esw-thes-request@w3.org] On Behalf Of Miles, AJ
(Alistair)
Sent: 01 March 2005 21:55
To: Leonard Will; public-esw-thes@w3.org
Subject: RE: rude categorisation



Hi Leonard, 

> Rather than providing a property that allowed or encouraged people to 
> blur these distinctions, would it be better to provide two or more 
> distinct relationships between concepts and resources, to force 
> cataloguers to choose the correct one?

Yes absolutely.  The intended semantics of the skos:subject property
restrict it to the "about" relationship only ... let's think about
whether some other relationships might be usefully added to SKOS Core.  

But we can have our cake and eat it.  What we can do in RDF is have
'relationship hierarchies'.  So we could have a set of more specific
properties, each a sub-property of a more general property.  

Why bother to have a more general (fuzzy) property if we want to
encourage better semantics as you rightly suggest we shuold?  With this
fuzzier property we could dump existing (ambiguous) categorisation data
(which has some usefulness) into RDF with an automated transform.  (To
disambiguate existing ambiguous categorisations takes lots of human
effort.)  

I'm just concerned that, without a more ambiguous property in addition
to skos:subject, people will use the skos:subject property in a very
general way, and hence the intended semantics of skos:subject (and hence
the usefulness of the property) will be undermined.

Cheers,

Al.

> 
> Leonard
> -- 
> Willpower Information       (Partners: Dr Leonard D Will, 
> Sheena E Will)
> Information Management Consultants              Tel: +44 
> (0)20 8372 0092
> 27 Calshot Way, Enfield, Middlesex EN2 7BQ, UK. Fax: +44
> (0)870 051 7276
> L.Will@Willpowerinfo.co.uk               
> Sheena.Will@Willpowerinfo.co.uk
> ---------------- <URL:http://www.willpowerinfo.co.uk/> 
> -----------------
> 
> 

Received on Wednesday, 2 March 2005 09:41:37 UTC