A journey from archives to UX

About a year ago, I began my pivot to UX. I had been working as an archivist for over ten years, after shifting away from a faculty career in ethnomusicology. I was embarking on a journey from academia to industry, with a large suitcase of ideas in tow.

Along my journey, I found the nascent ResearchOps community. Research repositories caught my attention, and I made the connection to archival practice right away. UX researchers had identified the need to store their work for future (re)use - and they wanted to elevate their work by making it easily findable. This is the problem space that librarians and archivists occupy on a daily basis.

I’d like to make a few connections between what archivists do, and the needs that UX folks have. Here I’ll be focusing on a few aspects of archival description and archival processing - as they relate to UX research repositories.

‍

What archival description is - and how it can help

Archivists use a handbook called “Describing Archives: A Content Standard” (aka, “DACS”). As you might imagine, this handbook goes into great detail about the practice of archival description. But stay with me! The principles in this handbook are gold for research repositories. I’m going to talk about two of these principles here: standards and context.

First of all, DACS is an “output neutral standard”. This means that DACS can be applied to any kind of material - not just old letters and photographs. And why do we need standards? Here’s the line of thinking: good archival description allows people to find what they’re looking for. And there has to be a common understanding of what “good description” entails - a “standard” implies this common understanding. Establishing common understandings (standards) is crucial for your organization’s research repository.

For example, let’s say you decide that each item in your repository needs to be tagged with a date (fyi, you should definitely do this). However, first you have to define what “date” means. Date the data was collected? Date of the report-out? Or simply, year and quarter? How you decide upon a definition of “date” depends on how people at your organization will be searching for items by date. You may end up choosing two different date fields: “date of data collection” and “[year] + [quarter]”.

Now let’s talk about context. Context is very important to UX researchers - no one wants their insights to be “taken out of context” (altering the significance and value of the insight). Archival description considers context to be essential to the understanding of archival records’ authenticity and reliability. What gives archival records context? A common set of descriptors. In order to be most effective, a research repository also needs a common set of descriptors - whether that’s a taxonomy or metadata that is applied to each item in the repository.

Above all, archival description is tailored to the needs of the institution and the users - the description(s) in your repository should also be based on your users and your organizational needs. For a more detailed run-down of archival description see this video.

More product, less process

In the past, archival processing (the practice of describing and arranging material in a collection) was very laborious and time-consuming. Some archivists would describe every single item in a collection! Imagine writing up a description for every piece of paper in an 80-box collection. Nope.

Nowadays, the concept of “more product, less process” (or, “MPLP”) is the standard way to process an archival collection. Rather than describing every single item, each group of items is described (e.g., letters, photos, writings). This way, public access to a collection occurs quickly.

I bring this up because it is very easy to get tangled in the weeds when setting up a research repository. It can become a seemingly endless process (especially when developing a taxonomy). Rather than planning for every potential search that someone might perform in your repository, keep things focused and streamlined. The goal is access.

Toward human-centered repository design

One of the primary functions of a repository is to prevent the duplication of research efforts. A UX researcher is asked to address a problem. They do a quick check in their team’s files, maybe mention it to a colleague or two. They don’t see anything - and the clock is ticking - so they move on.

Your job, when setting up a repository, is to make that look-back as quick and seamless as possible. Make sure your UI is friendly to *all* of your users (not just researchers). Try to make your tags self-explanatory. And make sure that your design allows the user to “know what they are looking at” when they find something in a repository. You don’t want a product manager to have to sift through a 40-page research report to identify the insights/findings.

And finally, elevate the researcher by using your repository to tell the “story” of the research study. How did the project start? How did it end? What were the twists and turns along the way? What might you have done differently? 

Distill your story into learnings for the next person who addresses the problem. This will not only save time and frustration - you will begin to create a corpus of knowledge and experience that your future colleagues will build upon.