ASIS&T IA Summit, 2003
ASIS&T IA Summit, 2003 |
For any professional to be good at what they do, they must have a firm grounding in the theories that form the foundations of their discipline. In Library & Information Science, however, there are multiple bodies of theory that form the basis of our intellectual work.
I have always endeavored at syncretism (the synthesis of different approaches or philosophies) in all I do. For example, during my graduate education in Library & Information Science (LIS), I did not focus my studies on any single sequence of courses. Instead, I attempted to gain a holistic view of the LIS field by taking classes that ranged from Ecological Information Systems to Database Design to Thesaurus Construction.
Below, I describe one such example, in which I synthesized theories of knowledge organization and the practice of information architecture.
During my last term in graduate school, I took a thesaurus construction course (LIS 537) where our task was to select an intellectual domain and to research, design & build a 100-term thesaurus for that domain. My group chose Buddhism.
Thesaurus construction involves a great amount of learning, and my colleagues and I first set out to absorb as much as we could of Buddhist literature. We checked out and bought books, scoured the web, and pored over articles from academic journals. After this initial frenzy, we determined that we could not, in 10 weeks, digest it all nor limit ourselves to 100 terms. So we decided to attempt a cursory survey of Buddhism from the perspective of a novice. Thus, we titled our project "The Beginner's Buddhist Thesaurus".
Traditionally, the creation of a thesaurus would involve harvesting terms from literature, writing them down on cards (known as "Soergel cards"), and tracking the relationships between all the terms on those cards. If any term or its relation to other terms changed, the cards themselves had to be edited by hand. This was all incredibly time consuming. Most of our colleagues were using this technique with a little help from spreadsheet software, but I was determined to save our group time and effort. So, over a weekend, I built a web-based thesaurus management system in Zope. It was messy and slow, but it worked. And it is still working.
The tool freed us to experiment with the relationships of terms in the thesaurus and, as a result, explore the structure of Buddhism as a knowledge domain. Throughout this process, we found ourselves discussing deep philosophical concepts, which we otherwise would never have considered.
I feel that I learned, in addition to all the Buddhist concepts and thesaurus construction techniques, much about project management and teamwork through this experience. In retrospect, it may not have been the best use of my time to build a web application just for this project. However, if I ever need to build another thesaurus from scratch, it's there.
 Presenting my poster at the IA Summit |
My final paper in Classification Theory (LIS 535) represented my attempt to theoretically justify the use of faceted classification in web site design and content management. After that class, however, I wanted to put that theory into practice. Using the concepts culled from that paper, I created a conceptual design for a content management system based on facet analysis. When the ASIS&T Information Architecture Summit rolled through Portland, Oregon, I saw a unique opportunity to extend my intellectual argument from LIS 535. I submitted a proposal, was accepted, and authored a poster titled "Using Faceted Classification to Provide Structure for Information Architecture".
In the end of the paper, I suggested that a pragmatist, context-based approach should use faceted classification to create poly-hierarchical, multidimensional data structures, which could be used in the design and management of digital libraries.
Faceted classification is, in a nutshell, the combination of all the epistemic approaches to designing classification systems. Using a technique called facet analysis, a collection of documents is described from as many perspectives (facets) as is necessary. Within each facet, a hierarchical list of concepts (isolates) is arranged. The result is a multidimensional representation of knowledge that more closely matches the state of real human knowledge.
My idea was to use faceted classification as a basis for generating web site designs and for content management. I envisioned a faceted classification system paired with a collection of digital documents classified in that system. Each facet would provide a different way of viewing the collection, so each facet could be linked to a different navigation method. The order of the facets and hierarchies in each facet could be rearranged at any time, directly affecting how the collection was accessed.
This is not a new idea. Several web sites, such as Wine.com, Epicurious.com, and Flamenco employ systems similar to what I have described. I tested the idea myself in my graduate assistantship with the PETTT zBento project. However, I did not feel that the link between classification theory and information architecture had not been properly established in the Information Architecture community.
Information architecture (IA) is the art and science of organizing structures for document storage, use, and access -- most often in a digital environment. Since there are obvious commonalities between IA and LIS -- especially knowledge organization and representation -- a good number of professional information architects hold MLIS degrees. So, in January of 2003, I submitted a poster proposal to the American Society of Information Science & Technology (ASIS&T) Information Architecture Summit, which was to take place in Portland, Oregon, in March.
My proposal was accepted, and I went to work on the poster. It was to be presented as a case study of zBento, so I credited two of my PETTT teammates, Eric Maddox (zBento developer) and William Washington (zBento usability engineer), as 2nd & 3rd authors. However, I completed all of the intellectual work, graphic design, and production of the poster myself.
 Information Architecture Summit Poster: Faceted Classification and IA |
At the IA Summit, the poster was well received, in my opinion. I was able to chat with several people in the IA community about the poster, including Peter van Dijck, developer of XFML, an XML schema for representation and sharing of faceted metadata. Some conference attendees asked difficult questions, and others had some difficulty understanding the point of the poster. I suspect this was due to the fact that many in the IA profession have never heard of faceted classification.