Resources on cyberscholarship; classification systems

A few weeks ago I attended a talk by Geoffrey Bowker, professor and “Senior Scholar in Cyberscholarship” at the University of Pittsburgh School of Information Sciences. This was part of the Digital Libraries & Cyberscholarship Colloquium Series, which continues next Tuesday and again in December.

I don’t have time to write a post that would do justice to Prof. Bowker’s talk, so I just decided to link to his website, which can provide a better idea of the topics he has been investigating (data sharing, interoperability, classification, standards, meta-narratives…). The main subject of the talk I attended was “social and organizational features of emerging scientific cyberinfrastructures”, but now that I’ve skimmed this paper, “How things (actor-net)work: Classification, magic and the ubiquity of standards“, I’m more curious about the book Bowker co-authored with Susan Leigh Star: Sorting Things Out: Classification and its Consequences (MIT Press, 1999).

What I find most interesting about Prof. Bowker’s approach (as it was represented in his talk at SIS) is how historical it is, and how he draws his examples from various disciplines and makes connections between them, especially in terms of “storytelling” and organizational behavior. I think this type of analysis is especially enjoyable when applied to classification systems, since they have such a long history and are such dependable vessels of ideology. In fact, I’ve read another book that takes a similar approach: Glut: Mastering Information Through the Ages by Alex Wright (National Academies Press, 2007). Wright’s website offers an annotated bibliography to his book; this is a great resource for anyone interested in the history of human efforts to represent and organize knowledge and information.

“Popularity is not a semantic structure”

I just read this great article by Tom Reamy in KMWorld. It’s about popular (and widespread) misconceptions about taxonomies and folksonomies. I loved the attitude in this piece, and it expressed frustrations I’ve had with peoples’ blind love of folksonomies, and with the misconception that hierarchical classification systems maintained by experts are an outdated effort that only librarians still care about. Reamy emphasizes how a hybrid approach, using a taxonomy and user-generated terms, is where the real value lies. He also points out the myth that folksonomies allow us to break free from the authority of “those dictatorial librarians”:

…folksonomy sites do have a central authority, and it is the most oppressive and most dangerous type of central authority there is — the authority of the majority. Against the will of the people, there is no recourse, no way of insuring the rights of the minority[…] It seems to me that having a system in which there is a central group of authorities or librarians that you as a minority can appeal to might work better than letting the collaboratively emergent dictatorial majority unconsciously ride roughshod over the minorities.”

The other thing I wish people would shout from the mountaintops is that the LCSH is (are?) not a thesaurus. Mary Dykstra says it best in her 1988 rant in Library Journal: just because LC decided to use the terminology of thesauri (RT, BT, NT, UF) doesn’t mean the semantic relationships between the headings are on par with those in real thesauri. Citing the 1974 ISO standard on what constitutes hierarchical relationships between terms, Dykstra uses the example of the heading:
Oysters
NT Cookery (Oysters)

In LC’s defense, this subheading doesn’t appear to exist anymore. I checked some of Dykstra’s other examples:

Proposal writing in business
BT Contracts, letting of
(Contracts, letting of is now an RT, and the BTs are Business and Business writing)

Children
NT Television and children
(still in there)

Here’s one I found:
Fortune-telling by Chinese characters
BT Chinese characters

Fortune-telling and Chinese characters are different types of entities. Fortune-telling by Chinese characters is not a type of Chinese character.

A lot of these issues stem from the insane degree of pre-coordination in the LCSH. Headings often represent multiple concepts, while in thesauri, terms represent only one concept. “With the use of a thesaurus, several terms (analyzed) may be strung together (synthesized) according to syntactic rules to form a subject” (Dykstra, 1988). I’m not saying LCSH isn’t useful and that it’s not currently serving many of us (relatively) well. It’s just frustrating that many people seem to think the LCSH is representative of thesauri in general. Reamy makes a similar point at the beginning of his article, but his complaint is with the use of the term “taxonomy”, not “thesaurus”:

A fundamental flaw in the vast majority of articles on folksonomies and taxonomies is the almost universal use of the Dewey Decimal System (or Library of Congress Subject Headings) as the example taxonomy. Using the Dewey Decimal System as your example taxonomy shows that you have no understanding of taxonomy creation and use in today’s world.

This brings me to a question that keeps bothering me. What really is the difference between a classification system, a taxonomy, a thesaurus, and an ontology? A nice set of definitions is available on this Hedden Information Management site (the creator teaches at Simmons College). These are things I need to have burned into my brain, especially if I’m going to avoid being led astray by the many instances of the terms being erroneously used interchangeably. These tools are too important to be so confused with one another, especially by professionals.

lcsh

Cited/linked:
Dykstra, M. (1988). LCSH Disguised as Thesaurus. Library Journal 113 (4): 42-46.
Reamy, T. (2009). Folksonomy Folktales. KMWorld 18 (9): 6-8. <http://www.kmworld.com/Articles/Editorial/Feature/Folksonomy-folktales-56210.aspx>