WHEREAS MOST BOOK-STYLE indexing is a stand-alone process requiring only the item in hand, collection indexing is a long-term process dependent on the use of controlled vocabularies for maintaining consistency between indexers. In traditional bibliographic database and library indexing, this is done through authority files for names and thesauruses for subject terms. On intranets and websites the same basic tools may apply, although name authority control is less likely, and the controlled vocabulary may be a taxonomy used for navigation as well as for term selection. Ontologies and topic maps are important for more automated approaches to information access on the web. For an introduction see Warner (2002).

Name authority files

In library catalogues and other collection indexes, decisions about the form of names are recorded in authority files that list the preferred forms of names and alternative names from which references can be made. The alternative names may be earlier or later names, pseudonyms, or fuller forms of the name, and often include birth and death dates to distinguish different people with the same name. Authority files either provide cross-references or work behind the scenes enabling automatic searches for different forms of names.

Authority files may also include the source of information about the name. For example, Library of Congress Authorities (authorities.loc.gov) has a record for my great-great aunt, Alice M Browne, with a reference to Ryce, John, her pseudonym, quoting a book by RL Wolfe on nineteenth-century fiction that states: ‘John Ryce is the nom de plume of Alice M. Browne’.