I wasn't yet a UC Berkeley student during the 1964 Free Speech Movement, but I do remember one of the slogans,"Do not fold, spindle or mutilate." It was the phrase printed on the IBM card that the university handed out to students for class registration. That phrase became a rallying call by protesters because it associated big computing with the big university. Back then, computers were only used by big institutions like universities, the government and big companies.
We no longer associate computers with big institutions because most of us have our own personal computers or smartphones, which in many ways have more computing power than those giant mainframes that read and spit out IBM cards. But one thing most big enterprises have that we don't is access to "big data." Big data is a term for databases that are so large that it's possible to start making inferences based on connections and correlations.
Search engines like Google (GOOG) and Bing give us access to thousands of petabytes of data, but they're not designed to correlate that data in ways that give us a great deal of insight into how all those bits and bytes interact with each other. And that data is limited because much of the world's most interesting data is closely guarded by those who own it.
But social media is
We've long known that Facebook has a lot of information about its users, including location information, friend connections, likes, interests and photographs from more than a billion people. By any measure that's "big data." But until Facebook announced its new "graph search" tool last week, we didn't have the ability to dig into that data to make our own inferences.
I've had a chance to play with Facebook's social search and, aside from learning it's a bit addictive, I've already gained some insights. One of my first searches' was "movies my friends like," which led to me to a movie on Netflix (NFLX) I had never heard of but thoroughly enjoyed. I love Japanese food but thanks to graph search, I now know the name of the local sushi bar that's most popular with people who are actually from Japan. I also now know
Admittedly, Facebook's tool doesn't give us access to the entire database. To protect the privacy of its users, Facebook's new search tool only exposes information that you already have access to. Nothing is visible unless you're part of the audience which, typically, includes posts from friends and friends of friends along with what others post publicly. But that can still be a lot of data. If you have 245 friends (the average according to Pew Research) and each of them has 245 friends, you have somewhere south of 60,000 friends of friends, accounting for overlap. Add to that those users (myself included) who chose to designate a lot of their posts and their demographic information as "public," and you have an even larger database to tap into. Assuming Facebook remains popular, over the years this data will grow exponentially as will your number of connections.
What this means is that, over time, users will gain the ability to make their own inferences based on an extended network of people (your "social graph") who mean the most to you. There is plenty of research to show that people tend to "like" what their friends like and behave the way they perceive their peers to behave.
Norms research -- the basis of some public health campaigns such as smoking cessation and alcohol moderation -- suggests that people's behavior has a big influence on those in our social circles. If you think your friends don't smoke (regardless of whether they do), then you're less likely to smoke. It's also true for other activities, such as what movies or TV shows we watch, what restaurants we patronize and where we go on vacation. We're not sheep but we do engage in some "herd behavior," which isn't necessarily a bad thing, especially if the "herd" is encouraging you to engage in healthy and socially beneficial behavior.
There is reason to fear big data and plenty of reason to be concerned about the possible misuse of what we've posted on Facebook and other social media, which is why it's more important than ever to think about what you post and monitor what others post about you. But as I've already discovered from my use of graph search, there is also plenty to be gained.
Disclosure: Larry Magid is co-director of ConnectSafely.org, a nonprofit Internet safety organization that receives financial support from Facebook.