During Bret Victor’s talk, I loved learning about William Playfair and how he invented the bar chart and other graphical methods to represent data. Related to these methods are “explorable explanations,” abstract representations that show how a system works or a way for authors to see what they are authoring without the black box of code.
Data visualizations are a powerful representation that is suited for VR. Though there are some visualizations that have been developed in VR, they usually rely on the game engine to navigate between charts or they will have some irrelevant motion like the bars rising in a bar graph when it is first loaded. I think with VR we can do more to incorporate the different modes of understanding that Bret Victor mentioned. For instance, we can build upon our spatial understanding to understand quantities, time, associations between nodes of information, or even how the charts are organized (like a library of books). We can build upon our aural understanding through having audio explaining the data and walking the user through it at a level specific to the user’s experience.
VR can make an data visualization an interface to information that makes the data accessible and easy to understand through abstraction. However, there is also potential for it to unpack the layers of abstraction and show how the data visualization has been made or even give the context behind the data. For instance, if there is a chart showing the amount of snowfall, could the user be immersed in the environment showing the snowfall and the data visualization of its levels? Data visualizations are a person’s stories of that data, so they are already created in mind with a specific objective for their audience. The trouble with these visualizations is that they tend to dehumanize the context behind that data, so VR really has the ability to use its potential for immersion to help the audience better understand the story. However, it is important for VR to not exploit this potential and to falsify the data through creating a specific immersive experience that causes a different perception of that data. I also think using VR to visualize data relates to the dynamic models that Bret Victor discusses at the end of his talk. Imagine data being updated in real time and seeing how the representation changes: the bar increasing, a point on a line graph being added, etc.