FW: Content Classifications / Taxonomies

 

On 6/14/05, Tim Denby <[email protected]> wrote:
> Metadata? Thesauri? Taxonomies? Topic Maps!
> http://www.ontopia.net/topicmaps/materials/tm-vs-thesauri.html

Alex replied:
> I'd also like to point out that one of the stronger messages in that 
> article (and the same would happen if you asked any librarian, too, or 
> me, or Clay Shirky, or any knowledge representation professional) is 
> that taxonomies and other means of simple classification schemes are 
> dead ends. Faceted classification based on thesaurii seems a good 
> option, but really; don't lock yourself into thinking anything can be
classified as only one thing.

That is true, if you're in a situation where you'll need a constantly
evolving classification scheme or if your requirements are mostly unknown,
out of the techniques described topic maps are the most flexible and will
give you a much more solid foundation than the other techniques.

But topic maps are pretty complex and will require a great deal of planning
and thought and engineering to use gracefully.  Lars does a good job of
describing the viable applications and pros and cons of each technique he
looks at.  If the requirements are well-known, can be addressed by a simpler
technique, and are not likely to change, or if your project just doesn't
have the engineering resources to create a topic-map-based system, topic
maps might not be the appropriate route to go.

And as Lars points out each technique is a subset of the more sophisticated
technique: taxonomies can be expressed as thesauri, thesauri can be
expressed as faceted classifications, and faceted classification can be
expressed as topic maps.  So any route you choose, you can move up to topic
maps if the requirements change.  If your developers understand the range of
techniques and plan for expansion to a more sophisticated approach you can
minimize the cost of scaling up.

Tim
Ektron, Inc.