Reimagining data visualization

This past December, I took a workshop, “Data Visualization as Activism” led by Sam Lavigne and hosted by Pioneer Works. It really got me thinking about how data is typically visualized - clean lines, minimalist color pallets, minimal text, following a long tradition in infographics + data visualizations. This type of representation seems to detach itself what what the data might actually represent. A bar graph of the number of lives lost each day during the covid pandemic hardly conveys the real impact each of those lost lives had, nor will a chart of numbers really encapsulate how many people are currently incarcerated + their lived experience.

So how can we go about reimagining how to visualize data? How can we interrogate the meaning behind the data?

I started looking into the work of artists + designers + thinkers who had some interesting approaches to data visualization.

Tom Scott went on road trips to visualize the difference between one million and one billion dollars.. Reddit user WhiteCheeks represented population density of animals through dot portaits of the type of animal.. Developer Matt Korostoff made data visualization to show Jeff Bezos’s wealth as well the number of incarcerate people in the United State.. Artist Mimi Ọnụọha created The Library of Missing Datasets, a physical repository filled with empty files.

The workshop, alongside some of these approaches to data viz, inspired me to start playing around with data visualization beyond using D3.js, to the point where I’m hoping to create at least one data viz project each week this year (perhaps ambitious- check back with me in a few months!) I also made a data visualization from the covid-19 api and using water consumption datafrom Open Data NYC.

Sure, this might be a more…artistic approach than what many data scientists/engineers are comfortable with. I recognize that. These tend to fall into descriptive visualizations rather than trying to perform mathematical analysis. But I think data visualization often falls into the trap that it is objective. But the questions of who collects the data + how can never be free from bias - just look at how the datasets that were used for AI technology reinforced existing racial biases.

I do realize that the traditional data visualization methods (graphs, charts, etc made with D3, Plotly, etc) often can be quicker, easier to implement. Given the rapid pace of our news cycle, this can really help the public better understand dense information + data sets in a timely manner. And I recognize that coming from a visual arts background, I’m far more excited about the visualization aspect than the data crunching + math that goes into creating + analyzing data - so perhaps to some of you who are more practiced with using data, this might all sound a little naive. I’m hoping to learn more about data science + using python or R in that capacity this year…though perhaps a little later on in the year.

But I also think there is value in reimagining what might be. Design can speculate about how things could be—to imagine potential futures, dream of alternatives to what already exists…the “what if”s that can open debate and discussion about challenging the status quo. And isn’t that what innovation is?