Under the best of circumstances metadata should be created according to specific collection needs but unified across collections wherever possible (e.g., to facilitate greater discovery and/or improve an institution’s ability to manage the content efficiently and effectively across their repository environment). Institutions should choose metadata schemas and approaches that meet focused use cases, and create metadata that they can functionalize and sustain. The range of metadata managed by an institution can often be complicated by legacy approaches. Over time, different curators at different moments in time may have made different choices about metadata approaches, leading to institutional inconsistencies across content sets. As an institution begins to ready its collections for long-term preservation, it can impose consistency through mapping its metadata into a common format and enriching it for preservation using the tools described above. As the tools continue to improve, this remediation will come into reach for a broader range of institutions, including smaller and less resourced institutions.
Case Study: Georgia Tech
Georgia Tech participated in the NEH-funded Chronicles in Preservation project. They exported three digital newspaper collections from their DSpace repository to package them with BagIt and send them to project preservation partners.
They ensured that the DSpace assigned Handle ID for each object was exported and stored. These Handle IDs are essential for future recovery.