Adapting Cataloging Workflows to a Batch Processing Environment: e-Forum Summary

This article was co-written by the e-Forum moderators.

An ALCTS e-Forum on “Adapting Cataloging Workflows to a Batch Processing Environment” was held November 14–15, 2017. Ethan Fenichel (Florida Atlantic University), Ariel Turner (Kennesaw State University), and Susan Wynne (University of Iowa) hosted the discussion.

Word cloud of e-Forum summary text

Participants from many types of libraries shared experiences with a variety of Integrated Library System (ILS) platforms, including Sierra, Aleph, Alma, Voyager, Millennium, Symphony and Worldshare Management Services, as well as homegrown platforms. Common tools and services include MarcEdit, Python, Backstage Library Works, OCLC Collection Manager (WCM), SQLite, and OpenRefine.

Record quality was the major topic of discussion on day one. Participants described common errors or issues encountered in vendor or knowledgebase (KB) records, as well as methods for evaluating quality. A law librarian discussed a collaborative effort within the American Association of Law Libraries to certify vendor record quality. Determining “good enough” quality varies among record sets and scenarios. Participants use local guidelines and/or tools, such as MarcEdit or scripts, to evaluate quality. Issues of duplication and whether a record set accurately matches the library’s access or holdings are other considerations involved in evaluating quality.

We also discussed reasons for choosing among vendor records, an ILS-specific KB, or WCM—or in some cases, abandoning MARC records altogether when metadata is available in a discovery tool. Several respondents mentioned collaboration with public services staff when selecting records.

On the second day of the e-Forum, participants shared advice and local practices for batch loading and editing. MarcEdit was a favorite tool for many participants. MarcEdit’s Task Lists functionality was popular for automating repetitive tasks.

ILS-specific tools are also used for batch editing. For Millennium and Sierra, participants discussed Create Lists and Global Update as essential tools. In Alma, some use normalization rules during batch imports to add or remove fields or make other corrections and enhancements. Participants discussed considerations for matching, merging, or overlaying records.

For those new to batch processing, participants recommended talking with vendors, consulting the ILS’ documentation or electronic mailing list, consulting the MarcEdit electronic mailing list and YouTube tutorials, and—most importantly—talking with colleagues. Other tips included marking up a sample record printout with edits you’d like to do in batch, testing new import profiles with small sample sets, ensuring a way to backtrack, and maintaining thorough documentation.

Many libraries use local fields in collections for collocation, statistics, or other purposes. One has experimented with adding administrative data such as open access or dates of access data during batch loads.

We wrapped up the second day with discussions on training and hiring in relation to batch loading and processing. Some participants indicated that batch loading and processing is now integrated into training plans for more staff, or noted the increasing importance of hiring staff with aptitude and interest to learn batch processes. Willingness to experiment and problem solve is important for staff working with or managing batch processes.

Over the two days there were over 200 posts to the e-Forum, with many active participants sharing advice and perspectives on the topic of vendor records and batch processes.

Submitted by Ethan Fenichel, Ariel Turner, and Susan Wynne

Resources

Articles

Presentations

Working Groups

Learning Resources

Workflow Documents

Tools

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.