Conor Myhrvold and David Feinleib
Who is Sandy Pentland?
MIT Media Lab Professor Alex 'Sandy' Pentland develops technology to measure, analyze and predict human behavior. Research of his own direct doing called Reality Mining uses data from cell phones and badges to understand how people communicate – the results of which companies and governments are already starting to use to improve their organizations. With sensors and cell phones, Pentland is monitoring the pulse of society.
In many respects, Pentland is also the grandfather of Google Glass , a prototype heads up display that literally makes digital data a lens through which we see society in the “real” world. The idea is to use the types of big data already gleaned from smart phones to help make real-time decisions. Some of Pentland’s former students – and present day Google X Lab employees – have helped make the Project Glass program a reality.
Pentland is also head of the MIT Media Lab Entrepreneurship program, and has co-founded or served in an advisory role to several big data analytics startups such as Ginger.io and Sense Networks. In addition to his many other roles, this gives him a unique perspective into the intersection of Big Data and entrepreneurship.
He was also named as one of the world’s most powerful data scientists on Forbes.
Pentland’s 3 Big (Data) Insights:
From his MIT Media Lab office, Pentland shared three key insights about Big Data:
1) Big Data is about people.
SP: Big Data is principally about people, it’s not about RFID tags and things like that. So that immediately raises questions about privacy and data ownership.
I mean, this looks like a nightmare scenario unless there’s something that means that people are more in charge of their data and it’s not something that can be used to spy on them. Fortunately as a consequence of this discussion group at the World Economic Forum, we now have the Consumer Privacy Bill of Rights which says you control data about you. It’s not the phone company, it’s not the ad company. And interestingly what that does is it means that the data is more available because it’s more legitimate. People feel safer about using it.
2) Cell phones are one of the biggest sources of Big Data. Smart phones are becoming universal remote controls.
SP: Cell phones have a long way to go in getting smarter. As they get smarter they’re more your universal remote control. You use them for everything. You browse of course but you’re now paying bills with them, you’re using them to take the T [Boston's public transit system]. Not so much in this country but in other parts of the world, your phone is the way you interface through the entire world. And so it’s also a window into what your choices are and what you do.
We [the group] used some of the very first smart phones. Before that I ran the Wearables Experiments, where we decorated people with computers and sensors and things, before they had even cell phones. So that’s, for instance, where Google Glass came from. My students have now gone and finally built the things!
3) Big Data will be about moving past averages to understanding patterns at the individual level. Doing so will allow us to build a Periodic Table of human behavior.
SP: We’re moving past this sort of Enlightenment way of thinking in terms of markets and competition and big averages and asking, how can we make the information environment at the human level, at the individual level, work for everybody?
The way we think about our culture, our politics, our institutions, is in terms of these big aggregates that are pre Big-Data. They’re things that in the 1700s people could think about and observe out their window. Now we can look at the actual patterns of interaction, of exchange between people.
We are on this boundary between the descriptive science, pre-science, and the sort of scientific method we’re familiar with. We’ve had all of these sort of intuitions, and heuristics, and ways we’ve sort of learned to make things work, and now we suddenly have the data to begin to build the periodic table of human behavior.
And we haven’t done it yet. We don’t really know how all the pieces fit together and what the data is telling us. And that’s the sort of grand and glorious scientific effort that needs to happen before we can get to a point where we understand the building blocks of human behavior.
Conor Myhrvold is a frequent contributor to the MIT Technology Review and other publications. David Feinleib is the author of The Big Data Landscape. For more information, contact him at dave@thebigdatagroup.com.
