Recent Posts

The Rockefeller Archive Center is excited to announce the release of our Collections Data API! As part of our commitment to open access, we’ve made our archival description widely available as structured data to open new avenues of access and inquiry through data visualizations, statistical analysis, and other applications that expand the questions that people be asked about our collections.

In the last few years, the Digital Strategies Team has built many – 25, by last count – custom applications to support the work of RAC across all function areas, ranging from acquisition of digital records to providing access to digitized content (and everything in between). Because our colleagues rely on these applications to get things done, we want to ensure their ongoing stability, and also facilitate our ability to confidently make changes to these applications in response to user needs or security threats without breaking core application functionality.

Over the course of the last six months, the Digital Strategies team undertook three related projects to meet these goals:

• Enhanced management of application dependencies to ensure that applications are built on up-to-date and secure dependencies.
• Broader and deeper use of continuous integration (CI) methodologies and tools in order to ensure that applications remain in a working state as changes are made over time.
• Expanded use of continuous deployment (CD) methodologies and tools in order to decrease the time between application feature development and deployment, reduce the amount of manual labor required to deploy an application, and help to document and disseminate knowledge about application deployment.

These projects, which shared the common overall steps of setting a baseline and then implementing tooling to maintain or improve that baseline over time, have helped us level up our shared knowledge of systems, and what it takes to keep them running.

At the end of last year, the Rockefeller Archive Center undertook a tightly scoped initiative to digitize and make available online a large group of archival records in support of a Research and Education team project. Given that some of these documents were dated as recently as the 1970s, we felt that it would be necessary to perform a scan for personally identifiable information in all documents after a certain date. However, the substantial number of documents, some 100 pages or longer, preemptively made any thought of manual review impossible. We had to find a tool to automatically search OCR’d PDF for PII and that could create a brief report on any potential matches.

We recently launched a new feature on DIMES – called a “minimap” – which allows users to jump directly to search matches within collections. This feature was conceptualized and implemented over the course of several months through the efforts of colleagues across the RAC and represents a bit of an unusual approach to a common design problem in archives, so it seemed worth writing about the process behind it in more detail.

The D-Team has worked over the past couple of years to integrate processes for digitized and legacy born-digital content into Project Electron infrastructure. In addition to updating policies and workflows, we have developed a microservice to package bags; an application to store, calculate, and assign PREMIS rights statements; and a microservice to create image derivatives and IIIF Manifests.