"I didn't want to change society. I wanted to give society a chance to determine if it should change itself." -- Edward Snowden, in a recent Washington Post interview.
Source: Laura Poitras, Praxis Films, Prism (screenshot via Wikipedia).
Edward Snowden has been one of 2013's most prominent and most polarizing figures. In his retelling, he's already won his battle against the encroaching surveillance state by simply giving them the chance to choose their future. But in a world where cameras are as common as computer chips, and where machines can learn who you are, what you like, where you go, and how you behave with greater levels of precision each passing year, the real war for privacy is already won by those who demand every citizen expose themselves to the watchful eyes of greater powers. Snowden's document dump may have only shown us what many long suspected -- that a world governed by ubiquitous technology is a world where privacy is a thing of the past.
The notion of privacy, as we understand it, is a relatively recent one. For thousands of years, with only a few localized exceptions, humans lived in small bands with few barriers between each other. They ate communally, farmed and hunted communally, ate around a communal campfire, and even squatted over a communal latrine. It was pretty hard not to know what your neighbor was up to, because you saw them every day, often for much of the day. This is the environment Internet pioneer Vint Cerf hearkened back to in a speech at the Federal Trade Commission's Internet of Things workshop last month. His comments -- "In a town of 3,000 people, there is no privacy," and more prominently, "privacy may actually be an anomaly " -- point out the fundamental flaw in the pro-privacy movement in an era of increasingly communal digital lives.
Think about that. Privacy may actually be an anomaly. The anonymity longtime Internet users took for granted since the early 90s is already fading into history, replaced by persistent online identities that follow us from site to site thanks to the efforts of Facebook (FB +1.33%), Google (GOOG +2.53%), and other Internet giants to embed themselves in our daily lives. Over 4.5 billion people click the "Like" button on Facebook each day (go ahead and click Like for this article now, you'll find the button floating at the top left of the page), and 350 million new photos flow into Facebook's servers daily from roughly 1.2 billion users worldwide.
That's a powerful example of the basic human need for close community, even in an era of global connections. Every Like you click and every photo you upload is you reaching out to your self-selected community of Facebook friends and saying "I'm here and I matter, and this is what I'm thinking about right now." Before the Industrial Revolution, you'd gather around the community well or church or firepit to talk about the same basic things. Most people want to talk about themselves and each other; otherwise Facebook wouldn't have 1.2 billion users sharing on a regular basis.
What profits the watchers?
Facebook updates and Google search histories haven't been the targets of Snowden's acts, but as technology continues to improve, corporate data-collection efforts will increasingly become indistinguishable from the digital dragnet cast across the Internet by the NSA and other surveillance agencies. It's already difficult to tell where Google ends and where the NSA begins, particularly in light of Snowden's revelation that the surveillance agency has actively worked to infiltrate the search giant's servers at a very basic level . Nor should anyone feel particularly reassured when NSA backers claim that no identifying information is collected by the agency's "routine" data-gathering efforts. A Stanford Security Lab research project recently showed that metadata (phone numbers and basic call records) can easily be matched to individuals or businesses, often with simple searches on Facebook and other major sites that retain user information. As we share more of ourselves with the commercial Internet, we wind up sharing ourselves with the surveillance web as well, intentionally or not.
Source: SHARE Conference via Flickr.
What sort of world do we have in store as surveillance becomes truly pervasive? I recently considered the possibility that tomorrow's children simply won't comprehend the notions of privacy that today's anti-surveillance crusaders grew up with, and that by exposing them (and ourselves) to social media at ever-earlier ages, we come to accept a reality in which being watched, watching others, and wanting to give others something to watch or see are all simply a part of our daily lives. When I was growing up, we watched TV. Now we watch each other. Some forms of entertainment ask us to give parts of ourselves in exchange for inexpensive access. TV asked us to give away a part of our creative imaginations. The Internet asks us to give up part, or all, of our private lives.Future battles for privacy probably won't take place over how much data winds up in corporate or government hands, but over what corporations and governments are obligated or allowed to do with that data. Even this more limited battleground may be stacked against privacy crusaders; there's little indication that either big business or the world's surveillance states have any intention of loosening their chokehold on the flow of information.
The truth is that it's becoming easier with each passing year to give up our private lives without really thinking about it, and the real driver of this loss of privacy is not the government, but data-hungry businesses that ask us to trade parts of our lives for free access to information (Google) or entertainment (Facebook). Hardware solutions, like cameras and GPS chips, will keep improving in quality while simultaneously becoming cheaper, and these are not the focal point of today's most advanced surveillance efforts. Rather, it's what happens to the data our hardware creates that determines how much of our lives become part of the surveillance web. The software solutions to this hardware-driven data explosion are becoming dramatically better, and it's data-driven companies like Google and Facebook that are leading its development.
Thinking machines
Machine learning is one way of approaching the classic Big Data problem -- how do you make sense of a flood of information, much of which is dreck and noise? Google has used it to train its systems to recognize images on YouTube videos, and also uses it to improve voice recognition accuracy. IBM's (IBM 1.44%) Watson platform , which was so successful at trivia shows, uses machine learning techniques to answer complex questions and has been put into service as a resource for diagnosticians. If the medical drama House were remade a few years from now, Watson would probably solve the show's major problem within the first sequence, leaving viewers to enjoy 50 minutes of uninterrupted sexual tension and interpersonal sniping without all that fancy hospital talk. Facebook, too, is entering the machine learning field in a big way, as it's recently hired New York University machine-learning guru Yann LeCun to lead an effort to derive meaning and business value from the 350 million photos its users upload every day.
Source: Mike Herbst via Flickr.
Computers that can diagnose problems, identify voices, and analyze images are only some of the works in progress that take the data billions of people willingly create every day and translate it into a form that companies can use to identify and target you. This sort of software is extraordinarily difficult to create at the moment -- an advanced form of machine learning called "deep learning " claims fewer than 100 graduate students worldwide, and salaries of $250,000 are not uncommon for these rare gems. But high remuneration and intense demand tend to combine to swell professional ranks, so it's reasonable to expect deep learning experts, and other machine learning pros, to be far more common in a decade than they are today. By that point, such advanced solutions to the Big Data problem are likely to be far more commonplace, too.
Most people have a deep human need to share themselves with a community. With few exceptions, we give away bits of ourselves as the price of entry to the global online community, and most of us do so quite willingly. Society's already determined itself to be OK with this arrangement, which is something privacy advocates like Edward Snowden don't seem to fully appreciate. The data genie is already out of the bottle, so to speak. Instead of trying to cram it back inside, let's ask it instead to keep us safe from those -- government and corporations included -- that would use it to do us harm.
