A couple of years ago, Sandy Pentland, professor of media arts and sciences at MIT, handed out about 100 Nokia cell phones to MIT students and faculty. The phones were equipped with software that helped Pentland’s team log interactions between the people carrying them. Based on phone calls and the devices’ physical proximity to other people’s phones (as measured by Bluetooth), Pentland and researcher Nathan Eagle developed social-network models that were more accurate and more nuanced than those constructed from the subjects’ self-reports. A paper on the study is currently under review at the journal Nature.
Sifting through cell-phone data to get at the truth of people’s social interactions falls under the umbrella of an emerging field that Pentland has dubbed “reality mining.” And he thinks that social networks are just the beginning. The same techniques can be applied to other sets of cell-phone data to help people communicate more effectively, manage their time better, and even make their neighborhoods more livable. And it’s all thanks to the ubiquity of cell phones–the ultimate data-collection machines.
Technology Review caught up with Pentland to ask him about reality mining and its implications.
Technology Review: When you talk about reality mining, what do you mean?
Sandy Pentland: The real roots of it go back to early 1990s, when people first started talking about context-aware computing. Just look at a cell phone. It knows where you are, and this is obviously sort of useful. But the generalization is that maybe it can know lots of things about you. Take your Facebook friends as an example. The phone could know which ones you socialize with in person, which ones are your work friends, and which friends you’ve never seen in your life. That’s an interesting distinction, and reality mining can make it automatic. It’s about making the “dumb” information-technology infrastructure know something about your social life. All this sort-of Web 2.0 stuff is nice, but you have to type stuff in. Things are never up to date, and unless you consciously know about something, you can’t put it in. Reality mining is all about paying attention to patterns in life and using that information to help you do things like set privacy policies, share things with people, notify people when you’re near them, and just to help you live your life.
TR: What technologies are enabling reality mining now?
SP: Today’s cell phones are on us all the time, and they come with hardware that can act as sensors for your environment. For instance, if Bluetooth is turned on, then the phone can see and be seen by other Bluetooth devices. You can start to make a record of the Bluetooth-enabled devices you encounter throughout the day. Then you can figure out, based on the frequency [with which] you encounter other people’s Bluetooth phones, what sort of relationship you have with them.
The iPhone also has an accelerometer that could tell if you are sitting and walking. You don’t have to explicitly type stuff in; it’s just measured. And all phones have built-in microphones that can be used to analyze your tone of voice, how long you talk, how often you interrupt people. These patterns can tell you what roles people play in groups: you can figure out who the leader is and who the followers are. It’s folk psychology, and some of the stuff people may already know, but we haven’t been able to measure it, at such a large scale, before these phones.
TR: What could be some benefits to all this measurement?
SP: You can really see things in a way that you never could before–a God’s-eye view. One of the examples I’ve been stuck on recently relates to how transformative Google Earth has been. Imagine having something where you can see all the people moving around on a map. Think about SARS in Hong Kong. What if in a particular apartment building, nobody left for work that day? You could identify a major health problem in 12 hours instead of two weeks. Another example is the social health of communities. It’s known that social integration, or how well people mix, correlates with whether or not a community is thriving. With reality mining, you can actually see social integration, as it happens or doesn’t happen. Once everyone can see it, then you can start to have transparent political discussions. Why isn’t the mayor putting more sidewalks and crosswalks in this area? Could more community events make the area more livable?
TR: This all gets very creepy very fast. How do you provide a sense of privacy in a world where cell phones are constantly logging your life?
SP: That’s not a trivial thing. Do you really want your government to know about you to that level? It could stop SARS, but there’s a big trade-off there. You could make this a much more transparent world where that’s available to everybody. But we definitely need to talk about it and figure out a new deal for privacy–to use this data and not be abused. The typical way is to make sure people can opt into these services so that they aren’t mandatory. Another thing is to make sure the personal data is removed from the information that anyone other than you sees. It comes down to needing to have open discussions about the implications of these things. The people making policies don’t know what is possible, and they don’t necessarily make policies that are in our best interest. You know, excuse the example, but I’ve been in a downtown somewhere and I don’t know where the nearest bathroom is, or it’s raining and I’d like a taxi. I’d give up a little bit of personal information to find these things. There are times when those services are really valuable. These capabilities are coming, but we have to come to a new deal. It doesn’t do any good to stick your head in the sand about it.
TR: Right now, reality mining is mainly done as research projects. Where do you see the technology and applications in five years from now?
SP: One is personal health monitoring. This means that people get feedback on how well they’re doing. This is really important in elder care. I see a reality check [coming] on e-mail lists and spam. How about I never get e-mail from a person I’ve never met? You can build whole systems based on real physical experience so it defaults the right way. I’m not saying any of this would be completely automatic; you could adjust it as you needed to. There’s also the personal-coach aspect of this stuff. All of us have the experience of talking to other people in public, but it’s hard to see yourself as others see you. Reality mining will help us see ourselves and, in an anonymous way, compare ourselves to peers. And I see organizations and companies using this to help people collaborate more effectively and do their jobs more efficiently.