Info+ is a long-standing data vis conference, held biannually in rotating locations. This year, it was hosted at Northeastern University in Boston (my alma mater), chaired by Pedro Cruz of Northeastern and Sarah Williams from MIT. The event was an action-packed three days of workshops, keynotes, seminars and social activities, and even included an art exhibition at the MIT media lab.
The conference was a dose of concentrated inspiration, with a head-spinning line up of back-to-back 10-minute seminars by leading designers in the visualization field. By the second day there were definitely some unifying themes emerging from the blur of inspiration and ideas.
You can find recordings and abstracts for all of the talks on the conference homepage. A few selected presentations are also linked below.
From communication-to towards communication-with
As someone who’s been in the data vis community for a long time, the biggest change I noticed was a shift in the general framing of data vis problems. Instead of Tufte-esque critiques of “proper” visualization techniques or discussion of misinformation and misleading graphics in politics, the conversation (at least in this conference) has shifted strongly toward more participatory practices in data vis.
Rather than talking about how to present data so that people will understand it, the focus was on how to have conversations—with people, using data—and how to include appropriate context and resolution to help them see how it fits into and reflects their lives. This was reflected in games talking about inflation at the grocery store and local biodiversity challenges in college classrooms, mapping inclusive and discriminatory spaces for marginalized communities to inform urban planning, and using info vis techniques to map informal transportation networks in developing nations.
When communicating with disenfranchised groups (like middle-schoolers impacted by extreme climate events and migrants hesitant about motivations behind the intervention), it can also be a challenge to overcome obstacles to communication, like self-censorship and diminished agency.
Visualizing marginalized perspectives
Across many talks, there was a focus on using data as a form of community expression, and using locally-generated data to capture experiences that are often left out of the dominant narrative. The conference exhibition included a project to record the important annual events for the Quecha people of the Amazon, organizing their year around important agricultural and cultural events.
Another team used conversations with migrants to improve shelters, focusing on designing features that will support them best in their transition. Data can also help to articulate deep-rooted structural inequalities, or something as “simple” as pronouncing someone’s name. It may also help us to question what we memorialize, how, and why.
Designing for impact
Some talks showed how to use data in a political context, as a tool for advocacy and creating change. One project focused on providing legal evidence to demonstrate systematic displacement in the West Bank, another created an archive of communities erased by urban redevelopment in Seoul.
A blanket woven from currencies served as an entry point into deeper discussions about economic impacts and the many reasons for migration, informing and humanizing policy decisions at the UN. One team collaborated with corporate sustainability offices to use biodiversity data to create better-informed sustainability policy and achieve more meaningful targets. Data can also help to illustrate what is lost when policies change, such as local shore changes for communities in the Mediterranean, and the pain caused by lost reproductive rights.
Advocacy is one form of impact; others take a more neutral approach. Some speakers discussed using data journalism to represent geopolitical conflicts in an unbiased but informative way. Others illustrated the importance of thoughtful visualizations focused on place and the need to keep things simple when dealing with the practical realities of fast-paced projects in a newsroom. Conversely, including details in your charts can sometimes make them better, more interesting, and more understandable.
New modes for visualizing data
Of course, the medium we choose also influences what we observe. The representation of time in social media platforms can shape and even distort our perceptions. Using different modes of visualization (including touch and sound) can help people engage with and better understand different habitats on the ocean floor.
Textiles have deep traditional roots and can evoke a softer expression of meaning, especially in a cultural context. Acoustic data can have profound emotional impact as well as quantitative meaning, and mixing auditory and visual explorations can encourage different modes of exploration, as well as creating more accessible tools.
Perhaps my favorite application of unexpected media was using folded paper as the basis for the conference identity, creating rich and nuanced visuals by simple physical means.
Seeing the big picture
Stepping back from day-to-day practices, we also considered how visualization can be a reflection of worldview. Framing is a critical step for a designer grappling to create a visualization, and our underlying theories of change influence both how we approach and how we talk about data visualization.
What I didn’t hear
Across the entire conference, there was almost no mention of AI. Presenters were definitely using AI technologies for certain kinds of data, but their talks were focused on the output rather than the tools. The one talk focused explicitly on AI considered whether it is helpful to use visualization as an input for AI learning, and what properties of a visualization might make it more interpretable and more useful for training an AI. I’m not sure if that was incidental or intentional, but it was a notable absence when so much of our current discourse is dominated by AI froth.
Reflections to take forward
Coming out of these many conversations, I found myself wondering which of the “theory of change” approaches are most effective, for which audiences, and when. Some speakers mentioned negative receptions: from the CDC when talking about data rhetoric and emotional visualizations, and from institutions of higher education when talking about faculty pay inequity. Many others discussed the tangible impacts of their work in shifting stubborn social and policy problems.
As always, the key lies in consciously framing your data and your analysis: in terms of the context, your purpose, the audience, and the people impacted and involved. Across many projects, we heard designers talk about how to define and redefine the problem as a critical step in getting to insight and achieving a successful design.
As a designer working in industry to create large platform software, I find that all design often gets simplified to UX. It was nice to step outside of that bubble for a moment and remember the many things that design does, and the different places that designers contribute. I do think there is an interesting conversation to be had between the perspective of creating large-scale tools to structure data exploration for decision making at scale, and the one focused on using bespoke and personalized data visualization for communication—either to or with—an audience once the analysis is complete.
Many of the unique, nuanced and contextual factors in a dataset can get blurred out when analyzing data at scale, and much of the big picture gets lost when focusing only on the particularities of a specific dataset. And yet, both the large and the contextualized cases come down to helping humans create big-picture conclusions by understanding nuances in the data. Building systems to accommodate large, unwieldy, and heterogeneous datasets to connect across these different scales requires insights from both sides. Perhaps that’s a topic for the next conference.
Erica Gunn is a data visualization designer at one of the largest clinical trial data companies in the world. She creates information ecosystems that help clients to understand their data better and to access it in more intuitive and useful ways. She received her MFA in information design from Northeastern University in 2017. In a previous life, Erica was a research scientist and college chemistry professor. You can connect with her on Twitter @EricaGunn.
- Erica Gunn
- Erica Gunn
- Erica Gunn
- Erica Gunn
