Data visualization for science: the next frontier?

Large media organizations are struggling to find a viable business model in the digital age. Newspapers are laying off journalists and photographers. But one department is in the throes of a huge expansion: data analysis and visualization. 

Every day, designers, data miners, journalists and computer programmers are applying new techniques to visualize data. A leading light of this new movement, Moritz Stefaner, says data visualization is the 21^st century's photojournalism. In a complex world, it makes the invisible visible providing cutting insight into things like algorithmic trading, climate change, tax evasion and government shutdown. Amanda Cox, Graphics Editor at the New York Times, articulates why the media is making such a big investment: “Data visualization can change how you think about the world.”

Swamped in a deluge of data, this is the attraction visualization holds for editors. Moreover we are all facing information overload. Our attention span is reducing and we need new tools to make sense of the world.

Is there anything science and policy can learn from how the media and high-tech sectors are using visualization? In November 2013, a unique two-day workshop* (see footnote) organized by Greg McInerny, Senior Research Fellow in Information Visualization for the Biological Sciences at Oxford University, attempted to do some thinking on how data visualization is used by scientists and policymakers and how it could be harnessed as an effective tool to increase impact.

What is data visualization?

But first of all, what is data visualization? As an emergent field, it has a range of definitions.

Nathan Yau, in his 2013 book “Data Points”, described it as “a representation of data that helps you see what you otherwise would have been blind to if you looked only at the naked source.”

“The best visualization evokes that moment of bliss when seeing something for the first time, knowing that what you see has been right in front of you, just slightly hidden,” he added.

Workshop co-organizer Min Chen emphasised other characteristics of the medium, explaining that it is something that unveils underlying structures, which helps humans to solve problems. It can reveal patterns of distribution, clusters, anomalies and correlations. In an era of big data, it will inevitably become an increasingly important cognitive tool.

Why should science and design work together?

“Consider the central role that visuals play in examining data and models, all the way to engaging less-expert audiences and communicating that information beyond science,” workshop organizer Greg McInerny told Future Earth.

“Why should we confine information visualization to the final stages of research projects?” he said. “Instead, a joined-up approach should embed information visualization into the work programmes of science and policy in order to leverage the greatest benefits for all involved,” arguing that an early involvement will produce both better visualizations and lead to new thinking in the design of research.

Moritz Stefaner kicked off the first day of the workshop – which is funded by the Tansley Initiative – by describing some of the challenges that designers face in trying to partner with with science. His concerns were echoed by Kim Rees, founder of Periscopic, based in Portland, Oregon. 

Top of the list was an innate suspicion that scientists have of design – fear that designers will misrepresent their research, but also what Stefaner called “Dumb Blonde Syndrome”, the idea that if something looks good, it is suspect. Another was the idea that science is focused on being comprehensive, insisting on an overwhelming quantity of detail, whereas design and good communication is all about simplicity. Finally, the reputational metrics of both communities are very different. Designers get recognition through awards and design publications, scientific reputation is measure through publications and their impact. “A paper in Nature, for example, will not get me any extra work,” he said.

A strong theme at the workshop was how to engage the new Intergovernmental Platform on Biodiversity and Ecosystem Services (IPBES), and several presentations and break-out sessions drilled deeper into what the workshop community could do to pro-actively make that happen.

Another recurring theme was translation. Many participants argued that the goal of a good visualization is to translate a set of data into something accessible for a general audience. A pragmatic definition of translation is: making content that is available in one language accessible in another language.

But this is superficial, as any literary translator can attest. All translation is a compromise. A perfect rendering from one cultural worldview to another is impossible. But the imperfect translation that is possible may be preferable to not doing the perfect translation that is not possible. This is a challenging condition for science to embrace, and it lies at the heart of the science/design divide.

What’s next

The workshop left participants with a sense of excitement at the potential for visualization in the next decade, and identified some major gaps in the interface where science meets policy. Data visualization, specifically for IPBES, may help bridge that gap, they said. They agreed that a part of the task ahead is to inform and educate users such as policymakers of the potential of this new medium.

Complex problems can only be solved by the concerted efforts of mulit-disciplinary teams and enabling these collaborations is one of the most important missions of Future Earth.

In the words of Silicon Valley venture capitalist Steve Jurvetson, an investor in Tesla Motors and SpaceX: “When academics move from one discipline to another…that’s really where you find interesting breakthroughs at the interstices between academic disciplines that had been formerly sheltered with their systems vernacular and ways, politically and otherwise, not to talk to each other.”

__________________________________________________________

This was the first in a series of workshops funded by NERC – the UK's main agency for funding research in the environmental sciences – through the Tansley Initiative. It was organized by Greg McInerny and Min Chen of Oxford University, Denise Young of the International Council for Science icsu.org and Owen Gaffney of the International Geosphere-Biosphere Programme. igbp.net

The workshop was co-sponsored by Future Earth.