Session I
Monday, October 28, 2:00-3:30PM

Organizing Information in Digital Libraries:
Classifying Digital Materials

Session Notes

David Levy, Geoffrey C. Bowker, Marcia Bates

General Session:

Levy:

• There's a distinction between subject cataloging and classification; folk/formal classification; and local/global classifications
• People are now using tools like WebCompass to classify things by themselves

Bates:

• Classification in U.S. libraries are used to linearize, but in Europe it's much more flexible
• Purpose of subject headings is not to give people multiple access points, but to complement classification schemes

  - they happen to allow multiple access, but they were not INTENDED to do this. Multiple access was not its original purpose.

• Folk classification = how our minds naturally work; can't deny this happens
• Formal classification = when it's written down

Bowker:

• There's a relationship between folk and formal classifications

  - formal grow out of folk classifications

  - formal categories are in tune with folk because people need to recognize how this works, or else it won't take hold

Bates:

• Indexes are usually thought of little or not at all when systems for description are constructed, but users think about indexes all the time

  - the user is expected to represent their queries the same way the indexer does

  - the user may not know as much as the indexer does, or may not be on the same wavelength as the indexer (e.g., under the Dewey decimal scheme, "military life" is under "public administration")

• There's a manual for indexers that's an adjunct to the Dewey classification scheme

  - it represents the nuances that have been worked out so indexers know where to put things

• Catalogers know that there is a problem, so they have to work it out to make things more unified

  - users don't have the time to do this

• Conditions that users/indexers go through are different

  - edges need to be smoothed out (edges of boundaries)

Bowker:

• Two huge formal classification systems often clash

  - sometimes, each group doesn't talk to each other

  - (e.g., international classification of disease and Ayurvedic medicine)

• So, the differences are not only between user/indexer, but between indexers

Levy:

• Usually, indexers are mediators
• "Why fix it if it ain't broke?" We think it's broken because of the avalanche of information

  - claim is to take the humans out of the group

Bates:

• You would think that we could come up with a classification system that CAN be implemented into a formal classification system.
• Not intending to get rid of mediators - but we should be more sympathetic to the user.
• We should use information expertise to make things easier.

Questions/Comments from Session Participants:

1. "There's so much information, though...if there's a lot of information, it won't be codified/indexed because of lack of resources, etc."

2. "There are so many users."

   - classification schemes are differently produced

   - the kinds of people that are using classification schemes somewhat add to the scheme itself

Bates:

"But there's a difference between creating a classification scheme and using a classification scheme."

Session Participant:

"But terms are codified by many users."

Bowker:

• Classification systems are very important things

  - they make a positive impact in the world, and are political interventions in the world (e.g., DSM IV (Diagnostic and Statistical Manual of Mental Disorders)

• Doctors find themselves having to use classification systems they don't believe in, just as a lingua franca, so they can talk about the work they're doing and to get reimbursements for treatments.
• Classification scheme sets into place a playing field - but the methods used within the scheme differ

  - e.g., schizophrenia (Freudian vs. Skinnerian)

  - they use the same scheme, but have different ways of dealing with it

Bates:

"There's an activity and there's a representation."

Last Updated: Jan. 15, 1997