Little Dictionaries: Making an open-source taxonomy for UX Research Repositories

Imagine a UX Researcher, new to their organisation and excitedly taking their first steps in a fresh project. The Researcher wants to add something worthwhile to their organisation’s collective knowledge of its users, so they decide to check if any useful information can be gleaned from research done by others in the past.

Cautiously, they approach the organisation’s Research Repository, which heaves and bulges with years of collected data and reportage that’s been contributed by a multitude of other researchers.

As they begin typing search terms or looking for words in the repository’s navigation for where to click next, they unconsciously reference a tiny personal dictionary that lives inside them. Because UX is a maturing field, their dictionary is not made up of universally accepted words based on centuries of academic discourse. Instead, their dictionary has been built up over their life through the process of listening to short course teachers, reading books, exchanging ideas with colleagues and connecting themes to words that simply make sense to them.

Will their internal dictionary match the vocabulary of their organisation’s Research Repository? Did the multitude of past researchers organise and annotate their data in a way that complies with the repository’s rules, or did they follow their own personal dictionaries?

Now imagine: How might we work towards a common taxonomy that we can easily add to our personal dictionaries?

Enter the ReOps #ResearchRepos Project

In August 2019 the ReOps community initiated the #ResearchRepos project to address a number of issues members were facing when dealing with the management of their research data. The project’s vision is:

“Research repositories present a huge opportunity for surfacing relevant insights from different projects, study designs, business functions and disciplines. However they are often poorly implemented, laborious and siloed.

ResearchOps can leverage the community to conduct research, map out the problem space for different users, construct shared definitions, taxonomies and models, prescribe good governance and guidelines, and finally develop a checklist to build, compare and evaluate better research repositories.”

The #ResearchRepos project is broken up into three streams:

• Research: Using UX research methodologies to understand researchers’ behaviours when managing knowledge
• Governance: Ensuring that the guidelines developed by this project comply with international standards
• Data and Taxonomy: Getting everyone’s little dictionaries working in harmony

The Data and Taxonomy stream’s hypothesis is:

Text-based interfaces rely on the user’s ability to recognise words and concepts that can be matched to their own personal definitions. The less time a user needs to make a correct match, the more successful the interface is. If a researcher can find what they need in a repository quickly, they will be happier to continue using it to find information, and they will also be more likely to format their own contributions in a way that makes it easier for others to find their insights.

Many dictionaries, one goal

Our goal to give researchers the best chance to share knowledge with peers and non-researchers within their organisation by creating a UX Research vocabulary that:

• is logical
• is memorable
• is easy to adapt one’s workflow to
• is easy to integrate with existing repositories and platforms
• is powerful enough to be used for data mining in the future.
• extends as many existing standards as possible
• is supported by industry, yet remains independent
• is open source

You’ve got to start at the beginning, and that’s where we are right now. Our first tasks are to:

1. Collect as many relevant words and terms that we can from reputable sources such as the Dublin Core Metadata Initiative (DCMI) and international statistical standards, throwing them together into a ‘word smorgasbord’
2. Collaborate with the #ResearchRepos research stream to add names to new concepts revealed by their research

After that we will:

1. Sort and refine the word smorgasbord, turning it into a vocabulary
2. Structure the vocabulary into a taxonomy and host it online for all to access and contribute to
3. Work with the project’s governance stream to ensure the taxonomy has the best chance of complying with systems around the world

Of the 3000 ReOps community members, 100 are actively working on the #ResearchRepos project, with 40 of those in the Data and Taxonomy stream. They represent the UX industry, repository platform providers, academia and the public service across all timezones. In order to manage this large and varied workforce, we are developing (and rapidly improving) workflows and processes that ensure everyone has the opportunity to make a contribution, and that the project keeps moving forward.

There will be disagreements and debate around which is the most appropriate word to define a concept: Our word smorgasbord currently has at least five synonyms for ‘User’ (and 3 for ‘synonym’)! Some long-held ideas will lose out while others become new norms.

When we are done there will be a well-documented and comprehensive taxonomy for UX Research, and a solid base from which new work can be done to experiment with creating an ontology to facilitate deep data mining.

But perhaps more importantly, we will have defined ways to collaborate at scale across the planet, setting the tone for the field of UX Research to mature faster, to the benefit of all.

If you’re interested in contributing your time to this project, or just want to join the community, please get in touch with us directly on Twitter or on LinkedIn. We also welcome everyone interested in ResearchOps to join the Community on Slack.