The library information ecosystem is comprised of a complex web of applications, scripts, and workflows that handle the range of library operations including acquisitions, cataloging, circulation, and analysis. At the UC Davis Library, for example, there are 15 non-cataloging systems that exchange data directly with the ILS on a regular basis, and an additional 25 systems that drive library operations which depend on the cataloging data in the ILS. Any conversion strategy must deal not only with the transformation of cataloging data, but also with the various points of exchange and interaction between all of these systems.
Each of the above systems is also intimately tied to human workflows. As noted elsewhere, library operations are performed by highly trained, specialized staff with well-established workflows. Altering the tools used by employees could have drastic impacts on quality, efficiency, and speed. Additionally, retraining of staff could be necessary, adding significant cost to the transition.
Adding to this complexity is the fact that the majority of the software systems deployed by libraries are licensed software applications provided by external vendors. This means that the vendors themselves must alter these applications to work with Linked Data graphs instead of MARC records, or complex connectors must be built for native Linked Data systems to exchange information with non-Linked Data systems.
The Phase One transition plan is designed to mitigate the costs and risks associated with the transition by establishing a minimally viable Linked Data infrastructure upon which a future, more complete transition can be accomplished. A Phase One conversion consists of the following primary steps:
Figure 7: Transition phase one