Re-Engineering Library Data — the Long Way from ADABAS to UNIMARC

The integration of different library catalogues into a union catalogue is a desirable goal in order to extend the available information and to ease the user’s access as well as to improve the user friendliness. Such a catalogue integration is no straight forward process because of the diversity and peculiarities of the involved sourcesystems. This article describes some of the problems the authors encountered during their approach to re-engineer such a library catalogue and to prepare its data for later integration into a union catalogue and shows the three-phase approach that was used. It has proven useful to remodel the data and to store it intermediately as well as to reverse engineer the sourcesystem to reconstruct the conceptual data model.