On Friday, a bunch of smart organisations (including the engine room, where I’m spending a lot of time these days) brought together 35 artists, researchers, academics, activists, journalists and more, to talk about Responsible Data Visualisation.

The day’s conversations and brainstorming turned out to be incredibly rich and varied. There will be more write ups and documentation coming soon, but before I forget, below are just some of the themes that got me thinking:

  • visualising uncertainty: instead of a supposedly clean and shiny visualisation which shows things we can never actually be sure of, why not focus on what we don’t know? It would be more honest, and actually conveys a more accurate picture of the data we’re trying to depict. In terms of visual language, it’s interesting to think about what this would actually look like; is it possible to still convey a strong advocacy message, for example, while admitting uncertainty? I wonder what it would take for journalists or social change advocates to take that step – essentially weakening the story they’re trying to tell with the visualisation, while being more up front about what is actually there.

  • visual literacy of the viewer: normally, when I’ve worked on data literacy themes, the “literacy” in question has been that of the person using the data. But what about the ability of the viewer to actually accurately understand what they are seeing? I see problems around low levels of visual literacy happen a lot with, for example, network visualisations. Thus far, essential context to visualisations that I’ve seen most often appears in a lot of footnotes at the end of the visualisation – which very few people read – or, gets ignored completely. Is there a visual way of helping a viewer understand what they’re seeing? Examples I can think of include visualisations where the viewer is asked to draw what they think they will see - like this NYTimes story on how family incomes affect children’s college chances. (For more on the data literacy themes we discussed, see Rahul’s blog post)

  • situating knowledge: this line of thinking was inspired first by Catherine’s fantastic blog post exploring what feminist data visualisation would look like- and then by the Databites talk that she and Mushon did last week. In brief; people with low levels of visual literacy often seem to understand a visualisation as “objective”, or “truth”. Clearly though, the perspective shown comes from somewhere, and someone. How do we bring back our perspectives to within the visualisation, making it clear where that knowledge comes from? Mimi and Catherine had a fascinating conversation which I had the pleasure of recording (coming soon!) about whether the action of situating knowledge must by necessity be reactionary, or whether it can simply… be.

  • legitimacy: if making visualisations is becoming easier and easier, and if many people viewing the visualisations understand them as “truth” - then what are the risks involved in making a visualisation? Patrick Ball of the Human Rights Data Analysis Group had a number of concrete examples of where the data that is being shown, really matters. In those cases, doing inaccurate data visualisations without understanding the potential misunderstanding that is being conveyed, can be dangerous. It could lead to people making important decisions based upon what they see – understanding how many people have been killed in a conflict zone, for example. On the other hand, though, being able to make visualisations is empowering – for example, seeing areas drawn out on a map. In both cases, I think it depends a lot on the expectation of the viewer; whether they expect to see information that will inform vital, life-changing decisions, or whether they expect to see a visualisation that is simply a first iteration.

Many other topics were discussed, and I had the pleasure of recording a series of video ‘conversations’ with a few participants, which we’ll be releasing in the coming weeks. Keep an eye out for them, and in the meantime, updates will be posted on the Responsible Data site!