Sorting Things Out

From ActiveArchives
Jump to: navigation, search


Classifications and its Consequences

By Geoffrey C. Bowker, Susan Leigh Star

http://mitpress.mit.edu/books/sorting-things-out

ISBN:9780262522953

Introduction: Histories of Classifications

Over the past 300 years (beginning with the ineffable Leibnitz) there have been a number of sweeping encyclopedic visions for storing all knowledge in a single form — be this through perfecting language (Slaughter 1982), classification systems (for example, Melvyl Dewey's library and industry schemes), or modes of knowledge organization (for example, Otlet (Rayward 1975)). These schemes have found their historians, but their shadow side appears to be discovered anew each generation. This side is the barrier to complete knowledge systems, notably in the following forms:

  • Data entry as work. No matter how good the scheme, its scope is limited by the fact that data entry is never an easy task, an there are never enough resources or trained personnel to make it happen. Not only will there inevitably be mistakes with respect to the internal structure of whatever classification one is representing, there will also inevitably be cultural variation with respect to how it is interpreted as well as culturally biased omissions.
  • Convergence between the medium and the message. Within any society there are a limited number of technologies for storing information (from ledger books to file cards to computer databases). The information that gets stored is at best what can be stored using the currently available technology: the encyclopedia comes to mirror the affordances of its technological base. In this process, people naturalize the historically contingent structuring of information; they often begin to see it as inevitable.
  • Infrastructural routines as conceptual problems. No knowledge system exists in a vacuum, it must be rendered compatible with other systems. The tricky, behind the scenes work of ensuring backward and sideways compatibility is not only technical work, it challenges the very integrity of any unifying scheme. Such work, however, is itself often classed as "mere" maintenance and deemed unworthy of public, historical pride of place.

— Chapter 3, pp. 107-108

Personal tools
Namespaces

Variants
Actions
Navigation
Toolbox