Recent Posts

In late 2014, the RAC migrated from Archivist’s Toolkit to ArchivesSpace. Shortly thereafter, we started working with ArchivesSpace’s API by writing scripts, mostly in Python, to export EAD, produce reports, and otherwise help to manage archival description at scale. After use, they were filed away in a GitHub repository, which over time became a sort of junk drawer.

Flash forward to the end of 2019. While we’re still using the ArchivesSpace API, a lot of other things have changed. Our coding chops, both individually and collectively, have improved dramatically thanks to several years of sustained engagement with Project Electron development work. We’ve also helped to seed the development of efforts like ArchivesSnake, which took all of the disparate code archivists had written to interact with the ArchivesSpace API and combined the best of it into a single Python package. It was time, we thought, to open up the junk drawer and see if there was anything worth saving.

Last week I attended the #a11yTO Conf in Toronto. #a11yTO promotes digital inclusion for the design and development communities in the Toronto area, and the conference brings together accessibility experts and disability advocates from around the world to present on topics ranging from legal strategy, standards, and values to more technical talks on design and web development.

Back in August, Hillel shared a post describing the next phase of Project Electron, which focuses on enabling users to find and access our archival records. In my role as an archivist who thinks about UX design and usability, I’ve been looking at how we might improve and reimagine the user experience of archival discovery at the RAC in ways that leverage our data model and data services work and represent what we’ve learned from our user research. In this post I’ll share our first steps in thinking through a user-facing site for archival discovery through creating wireframes, conducting some participatory site mapping activities with our staff, and looking at how other archives are approaching discovery.

Author Cary Reich published one volume of a biography on Nelson Rockefeller, The Life of Nelson A. Rockefeller: Worlds to Conquer, 1908-1958 (Doubleday, 1996). The Rockefeller Archive Center was gifted 564 audio cassettes that contain all of the interviews Reich conducted, including recordings with the likes of Donald Rumsfeld, Henry Kissinger, and former US President Gerald Ford. However, these tapes were poorly annotated which created identification, description, and access challenges.

As Hillel mentioned in his post about the next steps in Project Electron, we’re working on three data services for our pipelines into a new archival description interface: fetchers, transformers, and indexers. Once the fetchers grab data from a specific data source, we need the transformers to, for lack of a better word, transform that data from its original structure to a new one that fits with our proposed discovery data model before we index it for researcher access. But how do we ensure that the transformers are creating structured data that fits our data model as it’s defined? We need some sort of validation process to ensure the standardized structure of our data, which is why we’ve created a rough JSON schema that we can validate transformed data against before pushing it into our index.