Crosswalks, mapping tables, and normalisation rules: when we don’t even share the same vocabulary for authority control

Abstract

Institutional repositories typically make it easier for users to type in new metadata than to look up a standard vocabulary. To provide browse functionality, repository administrators often create ad hoc systems of authority control.

Integrating with external systems both highlights and compounds the problems with this approach. An external system harvesting from multiple repositories might deduplicate records, or filter results based on access rights: difficult when each repository has a different local standard. On the other end of the process, repositories sourcing records from external systems must carefully ‘map’ key fields and terms. Even then, local authority control is likely to suffer when confronted with inconsistent capitalisation of a standard subject vocabulary or wildly varying forms of author names.

This presentation will take as a case study the Lincoln University flow of metadata from research information management system to institutional repository to library discovery system and open repository aggregators. It will cover challenges we encountered including author names, ANZSRC keywords, item types, departments, and Creative Commons licence names, and offer practical solutions towards improving how our systems talk to each other.