The Schlesinger Library at the Radcliffe Institute for Advanced Study, part of Harvard University, hosts the premiere collection of resources for research on the history of women in America. The Long XIX Amendment Project was launched in part to celebrate the 100th anniversary of women’s suffrage in America. Schlesinger selected NewCity to plan and develop a search portal that integrates suffrage collections from libraries around the United States, enabling new scholarship and conversation around gender and voting rights.
On the 100th anniversary of women’s suffrage in America, it was vitally important to create a tool for scholarship on this topic. Research materials are abundant, but they are dispersed in various library collections, databases, and archives. Pulling them together in one place would benefit academic researchers, librarians, and archivists – not to mention anyone seeking to understand, and continue the fight, for equality.
No Ordinary Search Tool
A robust search tool presents significant technical and user experience challenges.
Schlesinger wanted to bring together archival materials from libraries around the U.S. Due to the broad scope of this project, we understood that there would be very little standardization of these collections.
From a technical standpoint, we needed to use a database that didn’t rely on a rigid structure like SQL – while still enabling complex searches.
From a user experience standpoint, we needed to understand our primary audience of historians and archivists. How would they use the search portal? What they would expect to find? How we could best introduce them to the scope of available sources?
Fortunately, we were partnered with a client team possessing a rare combination of technical expertise, creative thinking, and empathy:
Amy Benson: Librarian/Archivist for Digital Initiatives
Jennifer Weintraub: Head of Digital Services
Pablo Morales Henry: Senior Developer and Archivist for Born-Digital Materials
Rachel Guberman: Digital Humanist for the Long 19th Amendment Project
After a project roadmap was in place, we used Agile sprints to prototype and implement the system.
During the initial research and discovery phase, we ran usability tests on other search portals to get a sense of what academic researchers expect. Then we led Schlesinger through a 3-day participatory design sprint – conducted as remote workshops. We emerged from the sprint with several ideas for the search portal, which we narrowed down to a few concepts for prototyping and testing.
This was the first of 3 planned releases:
- A prototype mockup of the search portal
- A working version of the editorial content and management tools
- Delivery of the fully functional search portal and database
These led to a successful launch of the system in August 2020.
Prototyping and Learning
Not every search is like Google’s; and every searcher is not the same either.
One of our key findings from the first prototype was that simply putting a keyword search box in front of users (like Google) isn’t always the best strategy. You can generally assume that Google has indexed all the publicly available documents on the planet. But a library archive is different. A first-time user doesn’t necessarily know the scope of available content. This impedes a researcher’s ability to search effectively.
We needed an intuitive way to introduce them without getting in the way. Librarians, by their nature, want to be extra helpful – but a lot of people just want to jump in and start using the system. Designers of many similar systems often go to extremes: either a guided step-by-step tour… or they just throw people into the deep end. What we wanted was a middle ground.
Finding the Right Balance
Our next iteration included introductory messaging such as:
- About the Materials: Where we gave people the option to learn about the scope of the collection
- Submission Criteria: An indication that other materials can be added to the collection
Understanding that some users still may not know how to get started, below the search box we listed “20 items at random to get you started.” This gave first-time users a sense of what to expect in the collections.
Later iterations would fine-tune the messaging for instructions and help text.