Can Data Be Resilient?

photoBefore attending the NOAA Coastal GeoTools Conference in North Charleston, South Carolina (March 30-April 2), I had not spent much time thinking about data resilience. A brilliant scientist now working for ESRI, the leading company in geographic information services, drew my attention to this important question. But first, a note about the delay:

Almost a month ago, I passed along a link to an article in the Post &Courier of Charleston, South Carolina, that reported on two presentations at the conference, one of them mine. I promised more material from that conference, but the following week, illness took hold of me for a day or two, and soon after, I was consumed with preparations for a trip to the American Planning Association’s National Planning Conference in Seattle, April 18-21. The next day, I flew on to Denver and attended a two-day Project Advisory Committee meeting for the Kresge Foundation, which is pursuing its own program concerning community resilience. Now that all that is over, and I have a modicum of free time, I want to circle back to report on both conferences.

Let me start with a fascinating presentation by Dawn Wright, for 17 years a professor of geography and oceanography at Oregon State University before becoming chief scientist for ESRI in 2011. Wright was the main plenary speaker at the conference on its opening day, March 31. (March 30 was devoted to a series of special interest meetings.) Wright launched into her speech by referring to information as the “fourth branch of government” as a means of underscoring its growing importance in the digital era. Like many good speakers these days, Wright was also a fountain of recommendations about great new books, including The Fourth Paradigm, available for free download from Microsoft Research. Wright noted that we are now afloat in data, to the point where she introduced the term zetabytes, which equal one billion terabytes, which not so long ago seemed like massive units of data, but Wright says we are approaching 40 zetabytes in the current digital universe. But what does this mountain of data really mean for users? How useful is it?

Wright said this emphasizes the importance of metadata, that is, data about the data, and that we are facing a huge problem in “dark data,” data that is not tagged or properly analyzed, making its utility more problematic. “Just making data and code available is not good enough,” she stressed. “We need to be more open and transparent about what we do with them. There is a need for interoperability and cross-walking.” She then said that ESRI would practice what it preached, citing its sharing of work flows and use cases at www.esriurl.com/workflows. She also noted that ESRI has joined forces with the U.S. Geological Survey in creating the global ecological land units map.

She then stressed the importance of digital resilience, noting that “human communication skills are still paramount.” She cited a recent Washington Post op-ed article by Fareed Zakaria, “Why America’s Obsession with STEM Education is Dangerous.” The acronym refers to science, technology, engineering, and mathematics, the collective focus of some recent education policies; Zakaria’s point was not that STEM education itself is wrong, but that a parallel focus on the liberal arts helps create students, and eventually adults, who not only have technical or scientific skills but the ability to ask and articulate fundamental questions. This led Wright to state that we need to learn how to “read deeply” in order to ask ourselves the tough questions.

The days when scientists can isolate themselves in ivory towers are over, Wright seemed to be saying, as she stressed the need to “write compellingly” and “think critically to analyze ideas.” For programmers, she stated that, “If you can organize your thoughts, you can organize your code.”

But there was a still larger theme in a world where the job environment is constantly shifting and evolving. Training, she said, “is not just for your first job, but for your sixth job.” Critical thinking is critical for navigating these transitions in life when some job skills are obsolete within a decade. She added that there are still not enough people “squarely in the community” of the emerging field of data science. She questioned why there was still no formal accredited academic degree in coastal or ocean data management, for example, and her own book on the subject, Ocean Solutions, Earth Solutions.

She also underlined the importance of knowing the “design story behind a product,” a subject she took up the next day in a session on “story maps,” a translational tool for users that allows science students to tell their story. In that session, she noted positive change in that “students now want to escape the ivory tower.” She highlighted her central point once again by stating unequivocally that, “If you’re not speaking up as a scientist now, you’re doing a public disservice.”

My personal take on Dawn Wright’s presentations is this: If data is to be worth something, if it is to be resilient, it must be interpreted. And only educated, knowledgeable professionals are in a position to do this. It is no longer enough, if it ever was, just to crank out data and hope it speaks for itself. That is the route to impoverishing public discussion, which, it seems to me, suffers enough already in an era of sound bites and conspiracy theories.

 

Jim Schwab