Automating Record Loading—An Implementation Story

Abstract

If loading record files is consuming more and more of your library’s time and effort, learn from our experience in automating record loading at the University of Tennessee. Like most libraries, the University of Tennessee Libraries (UTL) has loaded files of order and bibliographic (bib) records into our Integrated Library System (ILS) for years. In 2012, we automated this process by writing loaders that look for these record files and load them into our Ex Libris Aleph ILS without staff intervention. We began the project with Yankee Book Peddler (YBP), one of our major vendors, and the result has been positive and significant—with a few bumps in implementation. This paper covers the process from three perspectives: systems, vendor, and technical services. From a systems librarian, hear about writing the loaders—the decisions we made and factors we considered. From a vendor, see how vendors can support this kind of project, and learn a few tips to make the process easier. From a technical services librarian, learn how staff worked to implement the process, check the results, and revise our workflows. We’ll share the problems we encountered, and the changes that we—systems, vendor, technical services—made together to get the project on track. Why Did UTL Want to Automate Record Loading? Efficiency, of course, is the easy answer—like most libraries, we are constantly challenged to do more. And like most academic libraries, we are moving away from buying print books title-by-title and into buying e-book collections. Every e-book package we buy comes with a record loading task, and every file of e-book collection records is different. Loading collection records requires staff to analyze and sometimes edit the file. We needed to automate our routine record loading so that we could move staff to the more complex task of loading records for collections. Automating routine loads meant that we wouldn’t have to train more staff and that we could limit the number of people who had the permission level necessary to load record files in our ILS. Automation also allows staff to focus their attention on parts of the process that require their expertise in acquisitions or cataloging rather than the minute details of things like filenames and dates or how to maneuver around a server on the command line. Besides making staff more productive in the use of their time and expertise, automation makes the process much less prone to error. Why Did We Think We Were Ready to Automate? We had at least a year of experience in manually loading most of these files, and we had manually loaded some of the files for more than 5 years. We chose to start with one vendor, YBP, so the record files were the same and did not require editing. We had established a record profile, tweaked the profile, and were happy with the records—we didn’t need to review the records, just load them. We had established a regular schedule with YBP, so that our bib files were posted on Fridays, and our invoice files were posted on Saturday. A regular schedule was crucial to the success of our loaders—the bibs had to be loaded in our ILS before the invoice loader ran. We also have a successful history of automating processes within our Aleph ILS thanks to an innovative, dedicated, and productive programmer in the Systems department, Mike Rogers. After serving as an integral team member for our Aleph implementation in 2003, Mike took responsibility for developing and implementing procedures and eventually training staff in the Technical Services unit in bibliographic record and EDI invoice loading. Loading records in Aleph