Issue Date: April 21, 2014
A Virtue Of Chaos
“Where is the wisdom we have lost in knowledge?” T. S. Eliot asked. “Where is the knowledge we have lost in information?” Presaging the cant of later generations of business consultants, Eliot arrived in his 1934 play “The Rock” at a key insight: The mere gathering of information does not impart knowledge leading to wisdom, and may be a step in the wrong direction. But it would seem that the utterances of poet and consultant alike have faded into obscurity in the 21st century.
With the ascendancy of “big data” and the business world’s burgeoning confidence that large data sets can be mechanically crunched into meaningful revelation, the emphasis has shifted from the attainment of knowledge to the agglomeration of facts and figures. The virtue of extracting clarity from chaos, it seems, has given way to making a virtue of stoking chaos.
Not surprisingly, big data is now the target of a mounting backlash. Recent articles in the popular and business press have raised questions about the technology and methodology underpinning the trend. Many scientists, business executives, statisticians, and average people on the street seem to have grown tired of what they see as the most recent in a parade of “next big things” in their respective worlds.
Some have found solace in humor, pointing to the kerflooey results of big data crunching in our everyday lives. Facebook comes to mind. A recent check of advertisements that the social media giant determined would interest me revealed barber’s insurance, funding for start-up companies, online doctoral courses, Converse high-top sneakers, and coffee mugs with pictures of dogs. I had an interest in one of these product categories about 20 years ago (the sneakers), but none currently interests me in the least. (The doggy mugs are obviously the result of all the pictures of pets posted by my “friends.”)
Well, Facebook is Facebook. But big data is also at work in important fields including finance, drug discovery, and health care. Major pharmaceutical companies are taking it very seriously, partnering just as enthusiastically with data banks as they are with academic research labs. Big investments are being made, and they shouldn’t be dismissed.
One of the biggest is at the Icahn School of Medicine at Mount Sinai in New York City. The only health care enterprise in Fast Company magazine’s recent ranking of the top 10 innovators in big data, Mount Sinai is in the process of expanding a supercomputer it calls Demeter, named after the Greek goddess of the harvest, which supplements another system called Minerva, named for the Roman goddess associated with wisdom and healing. Together, the two machines will put in more than 100 million computation hours per year.
According to Joel Dudley, a member of the department of genetics and genomic sciences at Mount Sinai, researchers there are entering as much relevant information as possible into the system, including genomics data, clinical data, and phenotypic data. What the computing power affords that’s new, he says, is connectivity across these modalities.
“What big data will help us do is reconnect everything,” Dudley says. “It is a losing game to focus on one kind of data.” A patient’s genome scan, lifestyle information, electronic medical records, zip code, and much more can be brought together and crunched with reams of comparable data on other patients in efforts to develop precision therapies, cut hospital costs, and educate caregivers, he says. “What I hope is that data will allow physicians to take a more holistic look at patients’ health.”
Dudley says he likes the term big data, but Mount Sinai is not playing a game of “my data is bigger than yours.” Wall Street or Facebook would laugh at the modest Minerva, he says. Math matters more, and what distinguishes Mount Sinai from Facebook is the level of sophistication of the data processing. What matters most is the patient, however. The computers at Mount Sinai are not in a cloud. They are in the same building as the clinic.
Not all the data are in about what role supercomputers will play in drug discovery and health care management. Dudley acknowledges that public disdain for big data in retail and social media is transferable and that many people are opposed to having a health data résumé following them to job interviews and the like. Sharing data between privately owned supercomputers may be a competitive issue, he says. But Dudley believes the real competition will be in data analysis.
The deflation phase can be delicious when it comes to megatrends. But it can be just as dangerous as the hype phase. A blanket dismissal of a packaged concept will blind us to its beneficial aspects. Ham-fisted pushback can start us on a short, easy path around the actual problems.
Big data is not Big Brother. Nor is it going away. It just isn’t quite here yet, and it’s natural for something so potentially huge to create anxiety in the early days. Eliot might quip that April is the cruelest month.
Views expressed on this page are those of the author and not necessarily those of ACS.
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society