- Data visualization allows scientists to share research with the wider public
- Award-winning Network Earth video examines resiliency in complex networks
- Strong back end technology enables dynamic visualization
For scientist and artist Mauro Martino, everything is data.
“Data is a way to describe ourselves, a way to talk about our relationships, a way to explore our planet,” he says. “It’s not just something that comes from the visual world.”
As director of the Cognitive Visualization Lab at IBM Research, Martino is a pioneer in the field of data visualization, which he describes as a mix between computer science and design.
His recent video, Network Earth, presents ideas of resiliency in complex networks in terms of climate change and ecosystem collapse. Its success was acknowledged with a 2017 US National Science Foundation (NSF) Vizzie award, which honors visual presentations of scientific ideas, or ‘science made beautiful’.
Awarded annually since 2003, the Vizzies celebrate the best photos, illustrations, posters, videos, and interactive apps that communicate scientific ideas to the public and encourage the exchange of scientific ideas around the world.
Martino’s goal in creating a visualization is to put the viewer in front of an experiment in the laboratory and impart a deep understanding of what the scientists have discovered. He sees himself as the director of a new genre of movie, where the actors aren’t people.
“The actors are just a different type of chart, a different type of database,” says Martino. “I combine all these different actors to build a new story that can be very engaging and very clear.”
From complex to familiar
People who work in visualization need to be able to jump easily from one field to another, Martino explains. They may work on financial market problems for one project and — as in the case of Network Earth — theoretical physics for the next.
Martino’s award-winning video is based on research by Albert-László Barabási, et al. that examines a complex system’s ability to retain basic functionality despite changes.
Barabasi’s work uses mathematical modeling to predict the failure point and applies to all kinds of systems, from social networks to epidemics, air-traffic control, or international banking infrastructure.
Running the networks simulations required about three months of computing hours on 256 cores in the Center for Complex Network Research at Northeastern University.
Martino chose to represent Barabasi’s theoretical networks through the symbiotic relationship between Australian trees and ants because these familiar organisms are easily understood by viewers.
“It can make the content feel closer to us, to make us feel more empathetic about these mutual relationships because they are something we see every day,” he says.
A significant story
Once he has decided on the best way to represent the science, Martino creates the visualization. He claims that this is the ‘easy’ part of the process, thanks to his diligence in building from scratch his own scalable dataviz technology.
“Through my lab, my team and I are building API and back end technologies to handle different calculations,” says Martino. He often reaches for those existing components when designing a new visualization.
“There are so many objects you have built in the past that you can change a little bit and use again and visually they will appear completely different,” he says.
When choosing a project to visualize, Martino also looks for research that will be of interest beyond the scientific community.
“A good story for me means something with an ideological message, a way for me to contribute to some ethical problem,” says Martino.
In the case of Network Earth, Martino hopes that his visualization will help more people understand that when it comes to climate change, we can’t trust our own sense of what is or isn’t normal to evaluate what is going on.
“It doesn’t matter that the planet looks the same as yesterday,” he says. “Everything is weaker. One day it can all collapse so quickly and in such a dramatic way, that there will be nothing to do. There’s not really a way to recover.”
