As I write this, the 19th Annual SIG/CR Classification Research Workshop on Classification and the Arts: Enduring
Practice, Alternative Strategies, and Contrasts with Other Domains has just
concluded. It was a small conference in
terms of the number of both active participants (6 papers and 1 poster) and
attendees (20), but rich in terms of ideas, discussion, and vitality.
In general, the
papers took an indexing approach, specifically a social tagging approach, to
subject access rather than a classification approach. A major drawback of social tagging is its use
of uncontrolled vocabulary, an issue addressed indirectly by several of the
papers. For example, one paper presented
the background use of a controlled vocabulary to clarify tags that are
identifiably ambiguous; another paper addressed the role of subject expertise
and indexing experience in achieving greater consistency of image
indexing. At the same time, a theme that
permeated the conference was the need to resist the inclination to insist on one
precise subject description of an image.
Rather than assume that knowledge organization structures should reflect
a singular view of reality, our goal should be to develop knowledge
organization tools that can reflect different social realities—your reality, my
reality.
But a plethora of
personal knowledge organization tools is not our ultimate goal. Consider, for instance, recommender systems
(like Amazon’s) that suggest new titles that might be of interest to you, based
on your previous purchases. But in
reality, it is not just your previous purchases that are taken into
account. Such systems also take account
of the purchasing patterns of many others.
They recommend title Y to you, after you have bought title X, because of
(among other things) the number of others who have purchased both X and Y. There is a role for collective knowledge
organization schemes (such as the DDC) working in tandem with personal knowledge
organization systems. That’s where mappings between knowledge organization
tools come in.
My own contribution
to the workshop came in the form of a poster (the text can be read here,
but if you’re ever in the Library of Congress, come see the wonderful job that
OCLC’s J.D. Shipengrover did with the graphic design) on chronological
organization of schools and styles of art.
This poster documents the thinking behind our October 2008 New and
Changed Entries (in Word and PDF formats). In essence, we are choosing
to expand under 709.05 21st century, 2000–2099 by decade by adding period notation from Table 1. Then, rather than trying to expand explicitly
for contemporary art styles or movements, we will give guidance for them
through mapping newly created arts-related headings from external
sources (e.g., Library of Congress Subject Headings [LCSH], Art and
Architecture Thesaurus [AAT] descriptors) to the appropriate DDC class.
Recent Comments