User Centered Collaboration for Archival Discovery (Part 2)

By the SAA 2017 Session 403 Team: James Bullen, Alison Clemens, Wendy Hagenmaier, Adriane Hanson, Emilie Hardman, Carrie Hintz, Mark Matienzo, Jessica Meyerson, Amanda Pellerin, Susan Pyzynski, Mike Shallcross, Seth Shaw, Sally Vermaaten, Tim Walsh

Insights from Discussion Groups

  • In discussion group 1, we went through a few different discussion areas. We had an interesting conversation about how to navigate using a user-centered approach. We talked about a) how to balance changing user needs professional practices; b) the difficulty of being user-centered within an MPLP paradigm; c) the contrasting difficulty of being *too* detailed in our description, and that getting in the way of discovery; and d) shifting our reference model so that public services staff are more facile with using finding aids and assisting users in navigating minimal description.
  • Discussion group 4 began by discussing the range of concerns relating to the state of discovery at the participants’ institutions. Everyone recognized that their current discovery systems for archives were not ideal, and there was a common interest across the group in centralizing discovery within an institution or consortium. The group also spent a significant amount of time discussing specific known issues to implementing a new discovery system, including issues related to system integration, the reality that information about archival materials is spread across multiple platforms, and that abrupt transitions across platforms were jarring for users. We also discussed challenges to undertaking user-centered design and collaborative work, which included barriers related to administrative support, systemic IT issues, lack of knowledge of user experience design methodologies, and resources for these projects.
  • Discussion Group 5 began by discussing archival discovery at participants’ institutions.  There was a wide range of strategies being used to facilitate archival discovery, but none of the participants were happy with their current state.  In most cases, the discovery systems were too deeply connected to library technologies like the OPAC, or utilized static html websites.  Participants were frustrated that their systems didn’t meet potential users where they were (the open web) and didn’t provide desired opportunity for users to find materials or more effectively use the finding aid data.   Participants saw flaws in their systems that negatively impact users, and all saw user testing as something that should be done when designing and maintaining archival discovery systems.  There was, however, some concern that user testing is resource intensive and that many archives don’t have the tools or training to do it effectively and that, while we feel good about our attempts to include users on product development we don’t have a good sense of what the return on that investment actually is in most cases.  
  • Discussion group 6 first considered how to go from having good description in multiple tools to having an effective user experience searching across all of those tools.  A user centered design approach was appealing, but there was concern about lacking the staff time and expertise to take this on, as well as challenges of knowing the demographics of your users well enough to establish meaningful user personas and being able to prioritize across different user groups’ needs. Collaboration and sharing the work seemed to be the answer to lacking staff time and expertise, although formal collaborative efforts do require overhead to manage the logistics of the collaboration. We discussed ways to have more effective collaboration, such as sharing the results of our work (like user personas) online rather than writing journal articles, making room for smaller institutions in the conversation, and allowing for different levels of time commitment and expertise within a collaboration. Roles for institutions without technical expertise include providing feedback or replicating a test at your own institution using another institution’s method.


Working on a user-centered design project for your archives? Have questions about the topic? Chime in via the comments below!


User Centered Collaboration for Archival Discovery (Part 1)

By the SAA 2017 Session 403 Team: James Bullen, Alison Clemens, Wendy Hagenmaier, Adriane Hanson, Emilie Hardman, Carrie Hintz, Mark Matienzo, Jessica Meyerson, Amanda Pellerin, Susan Pyzynski, Mike Shallcross, Seth Shaw, Sally Vermaaten, Tim Walsh

At the SAA Annual Meeting, a group of archivists and technologists organized a session on collaborative user-centered design processes across project and institutional boundaries: ArcLight (based at Stanford University), the ArchivesSpace public user interface enhancement project, and New York University’s archival discovery work. Using community-oriented approaches that foreground user experience design and usability testing, these initiatives seek to respond to the documented needs and requirements of archivists and researchers. In an effort to continue the conversation about user-centered design in archives, we wanted to share a recap of the session and discussion reflections with the community.


Sally Vermaaten started off the presentations by outlining NYU’s staged design work on a new archival discovery layer. In the first phase of work, a team of archivists, technologists, and librarians conducted a literature review on usability of archival discovery systems, held an blue-sky requirements workshop with stakeholders, assessed several systems in use by other archives, and drafted personas and high-level requirements. In parallel with this design work, a Blacklight-based proof of concept site was set up that utilized code developed by Adam Wead. The results of the pilot and design work were promising but also highlighted the growing need to upgrade other archival systems including migration to ArchivesSpace. Because implementing ArchivesSpace would offer new mechanisms to access metadata via API and would change underlying data structures, it made sense to migrate to ArchivesSpace before a full redesign of the discovery layer.


At the same time, the team at NYU knew that the proof of concept Blacklight-based site already running in a test environment included several tangible improvements to search and browse functionality that could be polished and rolled out with minimal investment. NYU therefore decided to take a phased approach in order to put those usability improvements in the hands of users earlier. First, they used rapid user-centered design techniques to quickly iterate on the proof of concept site, including a heuristic analysis and wireframing, and were able to deploy significantly improved search functionality to users within a few months. Next, they focused on ArchivesSpace migration and once that system was live, ‘Phase II’ of archival discovery work, a holistic rethink of the archival discovery layer, was kicked off. Sally wrapped up her presentation by sharing some of the aspects of the NYU approach that proved most helpful in their process and encouraged other institutions to consider these strategies in archival discovery work:

  • sharing and re-using existing resources (user research, design work, and code
  • documenting user needs as an impetus for and input into a future systems projects
  • incremental improvement and proof-of-concept approaches.


Susan Pyzynski and Emilie Hardman discussed the ongoing collaborative work toward an enhanced public user interface for ArchivesSpace. The first Design Phase took place between March and December 2015. This phase produced a set of wireframes and a report by the design firm, Cherry Hill, which was contracted to establish initial plans for the PUI.  The Development Phase, which spanned January-June 2017, took this initial planning into account and fleshed out the firm’s work with both fully exploratory and structured comparative user tests. A selection of these tests and findings may be found here. This work yeilded a 2.0 test release of the ASpace PUI: Though it sounds conclusive, the Release Phase (summer 2017) is not the final work, though it has put forth a user-informed product:

With this new release Harvard University is pursuing an aggressive timeline, looking at a January release for the ASpace PUI, and plan to engage in further and more specific community-centered user testing.


Finally, Mark Matienzo presented on the design process used for ArcLight, a project initiated by Stanford University Libraries to develop a purpose-built discovery and delivery system for archival collections. ArcLight’s design process followed a similar model as Stanford’s design process for the Spotlight exhibits platform, but with morea higher amount of community input and participation. The ArcLight design process included input from thirteen institutions, and significant individual contributions from eleven individuals, including both user experience designers and archivists. After providing an overview of the design process, Mark presented on how requirements were identified and evolved through over time. Using the example of the delivery of digital objects within the description of a specific collection component, this review included looking at early stakeholder goals and investigating existing functionality in their environmental scan; identifying questions to ask in user interviews, and subsequent analysis of their answers; how those insights were reflected in design documents like personas and wireframes; and their eventual implementation in the ArcLight minimum viable product. Mark closed his presentation by discussing lessons learned about the highly collaborative process. This included the recognition of the value of very broad input, the time and effort needed to organize collaboration, and the importance of needing professional knowledge and expertise in user experience in creating certain kinds of design artifacts./