Program Review: Reality Taxonomy 
By Thom Shepard, tshepard@ll.mit.edu

The SLA Boston Chapter program, "Reality Taxonomy: Applying Taxonomy to the Enterprise" was extremely useful for me. While I have been developing taxonomies for several years at WGBH and have read dozens of articles on managing vocabularies, this was the first program I've attended that was dedicated exclusively to taxonomy. It was also refreshingly devoid of specific vendor influences (no sales pitches!). The presenters were folks who actually grappled with the problems of designing hierarchical vocabularies to serve specific clients, or – in Brandy King's case – a group of diverse users. The order of presentations also worked very well.

Lynda Moulton, in her presentation, "Making a Business Case for Taxonomy," addressed one of my pet peeves almost immediately: she presented a kind of taxonomy of taxonomies, explaining both denotation and connation of such slippery concepts as thesaurus, glossary, ontology, classification, index, controlled vocabulary, and others. I was especially gratified when she touched on some of the issues involved with new technologies, understanding that vendors may not fully understand the needs of its clients and that users of technology should maintain a kind of healthy skepticism when presented with technological promises.

Lillian Woon Gassie's presentation, "Project Design and Management," contained a great wealth of valuable information. I wanted to ask her to freeze a slide from time to time, just to absorb the details. [All briefings are now posted at http://www.sla.org/chapter/cbos/meetings/fy05/200504lincoln.htm.]  Lillian provided a good outline of how to manage a taxonomy in an enterprise environment: planning, scoping and usage testing. In particular, I liked her SMART goals: Specific, Measurable, Attainable, Realistic, and Tangible. Lillian also gave an excellent, but too brief demonstration of her team's Homeland Security Digital Library Taxonomy.

Brandy King's "Smart Search" was perhaps the most interesting presentation. Brandy illustrated what can be accomplished with little or no budget, but with lots of ingenuity. I think her presentation also opened up some of the limitations or pitfalls in designing taxonomy for a widely diverse group of people. Her use of the terms "aggressive" and “bullying” within the context of instant messaging was particularly telling.  Would such a diverse group of people, from parents to medical professionals, use the same search mechanisms (browse vs. keyword) and, if not, should a vocabulary for navigation aim to satisfy all possible users or just the users who might utilize the browsing features?

Betsy Cogliano's "Knowledge Zones" presentation, on applying a taxonomy to an enterprise web site with multiple contributors, did address the scope note question; her taxonomy or classification system apparently uses scope notes to a large extent. "While Betsy made good use of her limited time, I wanted a clearer sense of how the vocabularies are managed behind the scenes. For example, are terms, relationships and definitions handled in ASCII files, either plain text or XML, or some other text tagging system, or strictly within a back-end database?"

The use of case studies was successful, but a panel discussion might have been a more effective way of wrapping up the program. A follow-up program might be useful to focus on metadata exchange and federated searching. This would probably be more technical than the taxonomy program, touching on such technologies as XML and XSLT, as well as developing schemas and dtds. 

The meeting gave me many practical ideas and raised just as many questions for future investigation.

Keep a lookout for more programs in this area. Below are some of the suggestions attendees suggested for follow-up programs:

• User interaction with the taxonomies
• Integration of taxonomies in semantic searching or broader ontology work
• A workshop on process of taxonomy creation
• A similar program with other projects using taxonomies (or just an update on the projects presented) - more case studies
• A session on general project management

Return to Top