W3C home > Mailing lists > Public > public-esw-thes@w3.org > March 2004

RE: faceted classification

From: Miles, AJ (Alistair) <A.J.Miles@rl.ac.uk>
Date: Tue, 23 Mar 2004 14:31:08 -0000
Message-ID: <350DC7048372D31197F200902773DF4C0494418A@exchange11.rl.ac.uk>
To: "'danny666@virgilio.it'" <danny666@virgilio.it>
Cc: "'public-esw-thes@w3.org'" <public-esw-thes@w3.org>


> Hi Alistair, folks,
> 
> fyi, there's a 'vanilla' XML format for facetted 
> classification with the 
> obvious acronym XFML [1]. It's primarily aimed at web site 
> data (such as 
> blogs). The model/syntax is close to parts of Topic Maps/XTM, and 
> considers facets to be mutually exclusive categories, which may form 
> part of a hierarchical classification.
> 
> I'm pointing to it now because there are some good refs on 
> facets around 
> that area [2], and I've a feeling facetting may be a very 
> powerful tool 
> in the SKOS kit. Also I'm wondering whether some of the data 
> that's been 
> marked up in XFML could be automatically translated into 
> SKOS/RDF. 

Hi Danny, thanks for the refs on this.  

It seems to me that data in XFML can go straight into RDF, using the
constructs of RDFS (and a bit of OWL) only.  E.g. I modelled the detergents
example from [1] using RDFS - see:
<http://www.w3c.rl.ac.uk/SWAD/rdf/facet_in_rdfs.txt>.  This approach is also
what the Siderean folks do [2] and what Nikki was suggesting this morning
[3].  I.e. a 'facet' in the XFML sense can be modelled as a rdf:Property
with a defined domain and range.

At the moment I'm trying to manage SKOS so it doesn't duplicate anything
that can already be modelled with RDFS or OWL - these three vocabularies
ought to be complementary.  Which is why I'm so keen to clear up the 'facet'
issue.  

Maybe someone should sit down and write a white paper on how to do 'faceted
classification' to support web catalogue browsing using RDF, RDFS and OWL
(including some XSLT to generate RDF from XFML)???

Al.

[1] http://www.miskatonic.org/library/facet-web-howto.html
[2] http://www.siderean.com/TechnologyWhitePaper.pdf
[3] http://lists.w3.org/Archives/Public/public-esw-thes/2004Mar/0070.html




That 
> the format yields well to XSLT is demonstrated at [3] - view source.
> 
> btw, XFML doesn't really have a mechanism for enforcing the mutual 
> exclusion, it's assumes the cataloguer will look after that.
> 
> Note too that Siderean Software have done a fair amount of work with 
> facetted classification and RDF [4].
> 
> Cheers,
> Danny.
> 
> [1] http://xfml.org/
> [2] http://xfml.org/links.html
> [3] http://www.alter.most.org.pl/fa/xfml/altermap.xml
> [4] http://groups.yahoo.com/group/facetedclassification/message/199
> 
Received on Tuesday, 23 March 2004 09:31:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:38:52 GMT