SKOS as a set of design patterns

ACTION: Jeff to write one or two paragraphs on VIAF design discussions
[recorded in http://www.w3.org/2010/11/18-lld-minutes.html#action06]

 

The point I wanted to make on VIAF's ontology evolution is easier to see
in the case of MADS where something similar has happened:

 

http://www.loc.gov/standards/mads/rdf/#t1-5

 

Like the MADS ontology, the original VIAF ontology existed in isolation.
Also like MADS, VIAF happened to model portions in ways that
instinctively mimicked patterns in SKOS. (Similar patterns also exist in
FRAD/FRSAD.) The VIAF ontology was eventually streamlined by aligning
itself more to the SKOS patterns, but the historical traces are harder
to see. The traces are easier to observe in MADS because the alignment
was neatly done in a separate mads2skos ontology and imported. 

 

The moral for me is that SKOS is not merely an element set for
"controlled vocabularies", it  also a data design pattern analogous to
software design patterns. Simply following the approximate form of these
patterns isn't good enough for the future, though. We also need to
express the alignments in the ontologies to make them recognizable and
interoperable.

 

Jeff

 

---

Jeffrey A. Young
Software Architect
OCLC Research, Mail Code 410
OCLC Online Computer Library Center, Inc.
6565 Kilgour Place
Dublin, OH 43017-3395
www.oclc.org <http://www.oclc.org> 

Voice: 614-764-4342
Voice: 800-848-5878, ext. 4342
Fax: 614-718-7477
Email: jyoung@oclc.org <mailto:jyoung@oclc.org>