Approaches to classification in publishing and knowledge management

Panel discussion — downloads available below

London, 20 June 2006

CLASSIFICATION OF KNOWLEDGE, and of the objects which contain it such as books and journals, has a long history, but is also a hot topic in the modern world of electronic collections and the World Wide Web. Indeed Tim Berners-Lee argues that the building of ontologies and software agents that can deal with them is central to the vision of the Semantic Web.

The subject goes by many names and has generated buzzwords such as taxonomies, ontologies, folksonomies and metadata, but the essential arguments are pretty much the same: how do we divide up and label the world or knowledge, does it have a hierarchy, what do you do about knowledge objects that seem to belong to several categories at the same time, and who decides? Can a controlled vocabulary be generated, and how does that help search and retrieval? How does one reconcile the classificatory judgements of experts with the way that the public and users see things?

Over the last year, the Electronic Publishing Specialist Group has been sponsoring an initiative known as ‘KIDMM’ within the British Computer Society — an ongoing project of discussions about the management of knowledge and information, data and metadata, in which people from many parts of the BCS’s Specialist Group community have become engaged.

Panel discussion

At this meeting, three participants in the KIDMM project, who in different ways have been professionally involved in classification for decades, spoke and took part in a panel discussion with the audience:

    Leonard Will, of former head of Library and Information Services at the Science Museum and now a consultant in information management

    John Lindsay, Reader in Information Systems at the University of Kingston, who has had a lifelong engagement with ‘infopolecon’ — issues of political economy as they affect knowledge classification.

    Nic Holt, Technical Architect for Fujitsu's Knowledge and Content Management Practice, and a visiting researcher at the University of Manchester’s School of Informatics.

The Chair of the panel discussion, Conrad Taylor, has prepared recordings of the presentations and discussion as two separate, strongly compressed MP3 files. You may also want to download the presentations slides and view them on screen while playing the recordings.

    Audio recording of panellists’ presentations as an MP3 file:
Download now (1 hour 2 mins — 25.3 MB)

    Audio recording of panel’s discussion with the audience as an MP3 file:
Download now (50 mins — 20.3 MB)

    Leonard Will’s presentation slides as a PDF file:
Download now (2.1 MB)

    John Lindsay’s diagram as a JPEG file:
Download now

    Nic Holt’s presentation slides as a PDF file:
Download now (1.5 MB)

   In addition, John Lindsay has written up some notes on the basis of his presentation. These are entitled ‘Pencil Test’ and can be found here within the KIDMM resources repository.