Data Visualization is an Art Form
As many people know, more data is being collected online. This allows for more exploration and interpretation for people. Although most people would assume big data is for engineers or scientists, some artists have begun to find creative uses for big data. With this new audience taking a dive into data, a new form of art has come about. What was once initially viewed as a technical job is now being altered with customization and creativity. That brings me to the question at hand:
What is data art and how do you do it?
Data is growing at an accelerating rate
Since data is increasing everyday, there needs to be a way to harness said data. According to the World Economic Forum, “By 2025, it’s estimated that 463 exabytes of data will be created each day globally.” Some technologists compare five exabytes to all the words ever spoken by mankind. Imagine saying every word in the human language — how long would that take you? Yes, there is a whole career path to crunching this data and learning stuff from it — but there’s a neglected side nobody really considers.
Artists!
Questions like, “how does one turn 500 million tweets sent a day into something more than an CSV file full of tweets?” or “how does one take the 5 billion searches made on Google per day and effective display what people search?” come to mind. As businesses start to use mass data to its full potential, the need for displaying big data in an easy, digestible form is ever more prominent.
So now what is data art?
Data art creates forms of artistic works from large datasets like tweets and search history. It can take in databases, raw data, data collected through search engines, calculations, statistics, and more to create art that can be interpreted. Due to the rise of accessibility to data, artists are beginning to get into this field as technology has enabled them to manipulate and create such works.
As artists become more immersed with data, they are starting to do what data analysts and scientists would be doing to utilize the data: processing the data and exploring what the data is. Once the artist has cleaned and started to understand the data, he/she will need to choose what best form of visualization will represent the data at hand.
Below are images that visualize taxi trip data in San Francisco, CA from 2009 for a span of 30 days. The images represent different tweaks in a programming library to produce various displays of the data. The objective in each of the versions of this visualization was to be able to show how taxi cabs in San Francisco move.
For a live website view of the final work, please go to sf-trips.netlify.app/.
Note: The 2009 dataset I used didn’t have a time interval small enough to make the taxis not jump around, which is why they don’t conform to the road. I ultimately liked the effect, though, and felt it would be a pain to try and interpolate more intelligently, so I kept it as is.
In all these pictures displayed above, the data populating the piece of work is the same, but the visualization is drastically different dependent on a few factors: the trail length of the taxi cabs, the background of the map, the size of the taxi cabs, and the viewpoint of the audience. These are a few of the many types of customizations an artist can choose when displaying this type of coordinate data. Even better, this is one of many visualizations that data can be represented by — another data artist seeing the same data might use an interactive heat map to showcase the overall density of taxi paths.
What does this mean for the future?
The art of telling a story with data can come in many forms. Like mentioned, with the access of data becoming easier with time, artists are now able to take data visualization to the next level. Artists will be able to create more complex and immersive experiences. The field of big data is no longer just for data scientists, but data artists as well.
Works Cited
Jeff Desjardins, Founder and editor. “How Much Data Is Generated Each Day?” World Economic Forum, https://www.weforum.org/agenda/2019/04/how-much-data-is-generated-each-day-cf4bddf29f/.
