- From: Leonard Will <L.Will@willpowerinfo.co.uk>
- Date: Tue, 1 Mar 2005 21:20:24 +0000
- To: public-esw-thes@w3.org
In message <F5839D944C66C049BDB45F4C1E3DF89D18DBC9@exchange31.fed.cclrc.ac.uk> on Tue, 1 Mar 2005, "Miles, AJ (Alistair)" <A.J.Miles@rl.ac.uk> wrote > >Hi all, > >Thinking about the application of SKOS to things like weblog >categorisation schemes, and also to traditional classification schemes ... it >seems like the semantics of a 'categorisation' of an information resource >against a concept are often ambiguous. The two main meanings I have >come across are *subject* (i.e. the categorisation says something about >the subject of the information resource) and *type* (where the >categorisation says something about the type of the resource). > >Given that there exists this kind of ambiguity, it doesn't seem right to use >the skos:subject property across the board when dealing with weblog >categorisations or classification schemes. I'm thinking that we could >create a more general property, e.g. 'skos:rudeCategorisation' which >would be a semantically more appropriate vehicle for dumping any type of >ambiguous categorisations into RDF. The skos:subject property would >then be a sub-property of this. What do you think? This is a long-standing problem in cataloguing library and museum items. It is necessary to distinguish a description of what an item "is" from what the item is "about". In the library format, MARC21, <http://www.loc.gov/marc/bibliographic/ecbdsubj.html> there is a distinct subfield code for "form", $v, in most of the 6XX fields for subject information as well as the following distinct field for those who want to use it: 655 - INDEX TERM--GENRE/FORM (R) Terms indicating the genre, form, and/or physical characteristics of the materials being described. A genre term designates the style or technique of the intellectual content of textual materials or, for graphic materials, aspects such as vantage point, intended purpose, or method of representation. A form term designates historically and functionally specific kinds of materials distinguished by their physical character, the subject of their intellectual content, or the order of information within them. Physical characteristic terms designate historically and functionally specific kinds of materials as distinguished by an examination of their physical character, subject of their intellectual content, or the order of information with them. It does get rather complicated when you try to decide what distinctions to make or which types of form should be lumped together, "rudely", as you put it. A dictionary, an image, a novel, a person and a teapot can all describe what a resource "is" rather than what it is about, though it is also necessary to label resources that are "about" these things. Additional complications arise when you need to describe things at two levels, e.g. an image of a teapot, an image on a teapot, or a critique of a novel. In these cases you may have to create separate records for each level, and link them together in some way. This is getting on to cataloguing (i.e. "resource description"), though, rather than thesaurus construction. 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? 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 Tuesday, 1 March 2005 21:26:04 UTC