W3C home > Mailing lists > Public > public-swd-wg@w3.org > September 2008

SKOS Comment

From: Johan De Smedt <johan.de-smedt@tenforce.com>
Date: Wed, 10 Sep 2008 21:20:27 +0200
To: "'SWD WG'" <public-swd-wg@w3.org>
Cc: "'Alistair Miles'" <alistair.miles@zoo.ox.ac.uk>
Message-ID: <48F5ABD8838C4DAC8C7D989A3AE8D806@TFJOHAN>


The mapping constructs are not practical for expressing the mapping between
concepts in the following use case.
(maybe this use case is not considered as part of the skos objective, then
please ignore the remark)

- ConceptSchema csA, csB, ... are used to classify content
  Typically csA, csB, ... are large and complex content schemes, used by
experts and/or expert tools
- Taxonomies tM, tN, ... are used in a an custom [i.e. target audience
specific] application
  Typically tM, tN, are made at hoc and are not used for direct content
classification (This exclusion need not be absolute)
- A seach expressed in terms/concepts of tM, tN, ... 
  must be resolved by a one-way mapping (tN & tM & ...) ->
boolean-expression-of(csA * csB * ... )

Multiple of these mappings allow content to be uniquely indexed (indexed
over csA, cs B, ...)
While supporting reuse of that indexed content in different search
strategies for different audiences
(user query specifies concepts from tM, tN, ...)

The current mapping does not address such mappings (with a boolean

Kind Regards,
   Johan De Smedt
mobile: +32 477 475 934
Received on Wednesday, 10 September 2008 19:21:13 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:07:54 UTC