Squirrels and nuts: metadata and knowledge management

Abstract

There is a paradox in practical knowledge management in which the need to have an environment in which knowledge is organized in a more efficient manner needs to be balanced against the problem of “information overload”. Two approaches to resolving this paradox comprise: the “big software” approach, involving the use of a sophisticated search engine; and the “traditional information” approach, where content is organized beforehand using highly skilled people to spend time indexing, abstracting and isolating critical pieces of information within the bulk of any information source. It is proposed that a more practical approach would be to use techniques based on metadata, XML (Extensible Markup Language), tagging, data models and relational databases, arguing that, above a certain size of genuinely useful internal knowledge assets, the metadata approach is an essential management tool. While search engines performed reasonably well with the older, smaller databases, the increasing size of these databases means that size and density have become critical variables in knowledge management. Details of metadata and related techniques are described as a means of solving some of the problems involved in organizing knowledge in very large databases.