The Why & How of Crafting a Repository Retention Plan

At our institution, the repository collects a broad range of material, from syllabi to digitized archival works to exhibition photography. We actually collect far more self-deposited student work than faculty research. Student works are used to support assessment and accreditation processes, allowing external parties to audit our course work and obtain examples. However, resource constraints make storing such a massive amount of student work in perpetuity difficult. These items are rarely referenced after several years, can suffer from poor descriptive metadata, and their value depreciates from an institutional perspective. We designed a retention policy aligned with our accreditation cycles and criteria for identifying works to be retained, such as those that won campus awards or are involved in legal disputes. We also developed retention procedures for contacting the owners of works which were planned for removal in advance and end user documentation on exporting their own works as well as understanding our metadata schema. This presentation covers the technical aspects of using REST APIs to implement complex retention procedures in a repository that does not support them natively. We will also discuss the challenges we continue to face with executing our procedures. These challenges tend to be less technical than communicative, such as handling email bouncebacks in a programmatic manner and adjusting our procedures to suit changes to institutional email addresses for alumni.