AVP partnered with our colleagues at Berea College Special Collections & Archives to support three major data migrations into new archive and preservation management systems.
The team at Berea had previously selected and begun implementation of their new systems. Our role was to support data transformation and migration from legacy systems into the new systems.
Microsoft Word Finding Aids to Archon
After analyzing a wide array of Berea’s 200+ text-based finding aids, we developed a custom data scraping script to extract the box list inventories from each Word document. The script transformed that information into delimited text data that could be imported directly into Berea’s Archon instance.
Additionally, we customized Berea’s Archon import functionality to support a custom hierarchy during the import process.
ContentDM Digital Objects to Preservica
Berea has been moving many TBs of digital content into the new Preservica instance as part of a long-term digital preservation plan. Berea also uses Preservica’s Universal Explorer interface to provide public access to digital objects.
Berea coordinated with us to move thousands of digital objects from their existing ContentDM site to their new Preservica instance. Working within the framework of Preservica’s import mechanisms, we coordinated with Berea to determine requirements for the organization of the content, extracted files and metadata from ContentDM, and created custom SIP packages for Preservica. See some of the results here.
FileMaker Pro Database to Preservica
Because Berea had so many TBs of digital content to ingest into Preservica, the majority of the digital objects were shipped to Preservica on hard drives and imported directly to Berea’s storage buckets. After this process, Berea determined that useful metadata for more than 20,000 files could be attached to the objects now that they were stored in Preservica.
To achieve this goal, we worked with Berea and with Preservica to design a custom metadata import process that allowed us to map metadata out of Berea’s FileMaker Pro databases, transform the metadata according to Berea’s preferences, extract UUIDs of all target files in Preservica, and package metadata according to Preservica’s requirements.
We also built a custom script to achieve import into Preservica at scale, updating 1,000 records at a time (as opposed to one-at-a-time via the standard GUI for metadata editing). For example, a text search for the “Coon Creek Girls” now brings back many hits from Berea’s deep audio archives!
We thank Berea and Preservica for the opportunity to work on these challenging projects!
