Seigo began work on a smartphone app (for Android) that combined Google Maps, real-time information about radiation levels, and publicly available data about wind currents. The resulting Winds of Fukushima app worked as a sort of living map that provided constant information not just on where radiation existed but also where it was going, in the form of bright blue arrows.
Of particular concern to Seigo was the safety of food and drinking water. The Winds of Fukushima app confirmed that radiation was spreading far wider than the government was indicating in news reports. Seigo began to buy his food from the south side of Japan and drink water imported from the United States.
Winds of Fukushima is hardly a technological miracle. It takes a very conventional stream of data (current wind direction), combines it with a second data stream (real-time readings of radiation), and makes this new, combined data available in a format that the public can easily find and use: a Google map. Its most revolutionary aspect is how quickly it emerged in the wake of the disaster. A decade ago, the task of coordinating among hundreds of Geiger counter–armed volunteers, building a platform for all of them to stream data, and finding a vendor willing to sell the software internationally was neither cheap nor easy. Thanks to communities of interconnected amateur techies, open APIs like Google Maps, and direct-to-market software vending platforms like the Android app stores, Seigo was able to build and publish Winds of Fukushima from a small Yokohama apartment in virtually no time at all. The app went live in the Android store about six weeks after the initial quake but it actually took Seigo only a few days to create it (though he admits he barely slept).
Pachube was started by an architect named Usman Haque who wanted to build a sensing feature into his building designs so that, years after construction, he and his fellow designers could log on to Pachube and get a sense of how the buildings were being used. He wanted to let the occupants, too, reconfigure their living environments around their actual use patterns, their living data. Today, the Cosm system that acquired Pachube allows developers to build apps, programs, and immediately derive insights off massive amounts of data coming from a suddenly awake world.
“Everyone gets insight into the environment around them, data contributors get applications that are directly relevant to their immediate environment, and application developers get access to a marketplace for their software,” Pachube evangelist Ed Borden remarked in a blog post.
A world that senses its occupants and shares that information may be one where people become much smarter about how they live. It’s also a world where information that is accessible only to government suddenly becomes available to hackers and activists. Depending on the content of that information, and the method you go about obtaining it, a simple civic act such as trying to fix your local sewer system can look provocative to the local authority whose power you just usurped, as another Pachube user named Leif Percifield discovered in New York.
Seeing the Hot Water Before It Hits the River
The date is April 18, 2012. Leif Percifield, a few of his friends, and I are in canoes in Brooklyn’s famous Gowanus Canal. A shy drizzle rains down on us as we paddle out over rusted bicycles, tin cans, and other bits of metal and plastic that have imbedded themselves in this canal bed. The air smells slightly of sewage, which is why we’re here.
We reach our destination, the portion of the canal that meets Bond and Fourth streets. Leif secures a shoe-box-size plastic container with a solar panel atop it to a mooring above the combined sewer overflow pipe, or CSO. Two long wires extend from the device; these are tipped at the end by a small sensor. Leif made the box, which is a prototype, the day before using off-the-shelf components (an Arduino motherboard) and parts he created himself with a printable circuit-board machine at Parsons School of Design. Leif plunges his hands into the water, elbow deep, to affix the sensor tip as close as possible to the pipe. When he’s done, he does a cursory clean of his hands and checks his iPhone.
“It works!” he says. The sunken sensor is now broadcasting the temperature and conductivity of the water. Hotter water, and water with more electricity conducting minerals, are sure signs of sewage runoff.
Just about everyone in New York takes for granted a few key facts about the Gowanus Canal. The most important of these is that it’s beyond fixing. Not only does sewage water run into the canal when it rains but the water is laden with decades’ worth of heavy toxic metals, which has earned its designation as a Superfund site, one of the most poisonous environments in the United States. The U.S. Clean Water Act says the city of New York is supposed to clean this place up, remove the metals, and keep sewage from spilling into its waters. But before that can happen New York City and the U.S. Army Corps of Engineers must conduct a feasibility study, which neither New York City, the EPA, nor the army are in any hurry to complete.
“The numbers that we have say that three hundred million gallons of sewage go into the Gowanus Canal a year,” says Leif. He adds that the numbers are based on computer models and he believes them to be flawed. Members of the community have accused the New York Department of Energy of tweaking the data in order to put off the costly work of fixing the storm runoff problem.
Leif’s goal is to get people in the city to participate in rehabilitating the canal. That’s not easy. But if he can map where flows are bigger or smaller, he thinks he can put together a more accurate assessment of what’s going on and essentially predict how dirty the water will be on any given day as a result of environmental factors. This information is of no real use to one person but a community can edit their water usage, their showering and flushing, based on the sewage water level. The name of Leif’s blog says it all: Don’t Flush Me.
You would expect the city would appreciate Leif’s efforts to better monitor the sewer system. But his relationship with local New York City authorities quickly became rocky. His previous project literally got him in a lot of hot water: he actually went into the city sewer system to fit it with a network of sensors.
“The air is not pleasant,” he says of the New York underground. “But I was thinking it would be putrid. Instead it was more acrid. And it was incredibly hot, twenty-five degrees hotter underground than aboveground. People use hot water, you know, and hot things come out of your body.”
The sensors he attempted to install were supposed to read the water level and a fast rise was a good indication of coming overflow. The experiment didn’t pan out. The sensors didn’t stay in place and the Bluetooth signal inside the sewer was too weak. The data was trapped. Had the sensor system functioned, he would have been the second person in history to be able to predict when the sewers were going to overflow. The first was Cynthia Rudin, an MIT researcher who figured it out with a statistical formula.
Leif’s project was simple, commonsense infrastructure stewardship. But when he posted a few pictures of his adventure online he immediately got a call from New York’s city officials. They ordered him downtown and made it clear that he was “in trouble” for what he had done. He was told to cease his activities. Leif believes this is because of how he was able to show how easy it is to get into the New York City sewage system.
Leif has grown better at working with the New York City Department of Environmental Protection but his experience reveals how complicated our relationship with authority becomes in this interconnected era. The program Leif put in place all by himself was very similar to ones in place in Maryland and Washington, D.C., to manage sewage run off in the Chesapeake Bay, but the latter are managed by local authorities with little citizen input so they’re less controversial (and arguably rather ineffective). Everyone can agree, at least publicly, that fixing sewage backup should be a top priority. But when citizens armed with sensor boards suddenly start outflanking government on government’s own turf, tensions can rise.
Most of us grew up in an environment where we comfortably assumed that local government always had more information than we did about
what was going on within our city, certainly the best data on the state of infrastructure. We also instinctively trust local government as the provider of information during an emergency, even when it’s an emergency in which we’re directly involved. See a fire? Call 911 and ask for services, wait for someone to come to where you are and tell you what’s happening. This is an inefficient way to collect and distribute information during a time of crisis.
The Internet of Things is ushering in a new era of proactive citizenry. It’s an era where much of the most important information during a fire, a flood, a citywide disaster doesn’t come from government but from you and your suddenly empowered neighbors, people like Gordon Jones.
Seeing the Fire Before You Are in It
In the summer of 2007 Gordon Jones was living in Charleston, South Carolina. A fire broke out at a nearby furniture store, killing nine firefighters, the largest number of firemen to die on duty since the 9/11 terrorist attack. An enormous memorial followed. Emergency workers from around the country came out to Charleston as the facts of the incident were reported on the news in rounds, like a funerary dirge. The public safety workers succeeded in pulling out several survivors from the blaze before the roof caved in on them.
Jones was working at the time for Global Emergency Resources (GER), a company that markets a software tool for monitoring ambulances and hospitals during emergencies. Watching the local coverage of the memorial service for the firefighters, he realized that the technology he was developing could have saved lives: “I said to myself, What if somebody, one of the people trapped inside the store, had a smartphone to broadcast what the scene looked like? That might have made a difference.”
It sounded like a worthwhile and potentially profitable start-up. Jones founded a company and shortly after announced the launch of the Guardian Watch app. Guardian Watch enables anyone with a cell phone to live-stream video and pictures of an event directly to emergency personnel. This may not sound that significant but think of an alarm system as nothing more than an information distribution network. Some alarm systems are better than others. Guardian Watch enables thousands of people to provide streaming visual data about a situation at an information transfer rate of hundreds of thousands of bytes per second, the average upload speed of a 4G or higher phone. Guardian Watch was the first iPhone app to take advantage of the smartphone’s full capabilities to give emergency workers a visual and auditory sense of what may be ahead of them.
“A picture is worth a thousand words and a video is worth a thousand pictures,” says Jones. This statement, though perhaps a bit corny, encapsulates why Guardian Watch really is a clear improvement over traditional emergency response systems. It delivers information that’s user-specific, varies depending on context, and moves at a speed and scale that make sense for emergencies—namely more and faster.
A decade ago, increasing the scale of information collection and distribution to the point where it would have made a difference to one of those Charleston firefighters would have been a daunting technological challenge. Today, the tools, platform, and infrastructure already exist and have been widely distributed. You’re carrying all of this around in your pocket.
The single biggest driver of the Internet of Things is the smartphone, that always-on, GPS-enabled sensor that more than 64 percent of the U.S. population carries around with them. We know that smartphones today make it easier to find restaurants, share experiences as they occur, shop, and study. Mobile technology makes data creation and curation possible anywhere, which means we’re creating and curating much more of that data more of the time.
Guardian Watch already faces competition from other groups looking to leverage the information gathering and broadcasting technology of smartphones. A Silicon Valley–based start-up called CiviGuard takes the idea a step further. The platform integrates streams from Twitter, Facebook, and local emergency channels and presents the user with a “networked window” of an emergency situation playing out in real time. It gives geo-tagged advice that’s specific to an individual user based on a variety of variables, the most important of which is location. What that means is this: depending on the situation, a user may be told to stay where she is while a different user may be told stay away from that area. Most important, CiviGuard includes a scenario function to allow users to conduct virtual emergency simulations.
Imagine you’re in Manhattan and there’s just been a terrorist attack. Want to know which streets are most likely to become blocked when the news spreads? How your company’s supply chain will be disrupted? Where to find food and water while they’re still available on store shelves? CiviGuard will tell you and will do so based on a rapidly updating understanding of what’s going on around the city. And should CiviGuard not pan out, the Environmental Systems Resources Institute (Esri) can also build you a custom geographic information system that does all of the above, and can integrate it with population density, water tables, jurisdiction, and hundreds of other maps.
The same real-time broadcasting capability that will allow me to better navigate my way out of a disaster can be used for other purposes as well. The mapping of human behavior promises enormous benefits, but it also speaks to a future where invisibility and anonymity are no longer the default setting for life.
The Internet of Things is also the intersection camera that snaps a picture of my license when I try to beat the yellow light. It’s the smart electricity meter that California’s Pacific Gas and Electric Company now insists its customers use, allowing the utility to optimize energy delivery but also to better track individual energy use. If you’re a PG&E customer, the Internet of Things is the reason why your energy company can infer when you’re home and when you’re not based on when and how you use certain devices.10
In our rush to overlay data collection devices across the physical environment, we overlooked the fact that the same devices we use to perceive our environment can just as easily be turned on us.
We will be seen. We will be tagged. It’s happening.
Checked In. Your iPhone Knows Where You’re Going
Want to know how many people with smartphones are in terminal 4 of New York’s JFK Airport, standing on line to get tickets to The Daily Show, browsing the shoe store down the street? A company called Navizon sells a device that can track every phone using Wi-Fi within a given area. Just plug this device into a nearby wall outlet to monitor that action in real time. Because we know that more than 60 percent of the U.S. population now owns a smartphone, a couple of months’ worth of data will tell you how many people are likely to be in any area that you’re surveilling on any given day and time of the week. Leave the device plugged in for a few decades and you’ll have a reasonable estimate for how many people will be at a specific place, at a specific time, on a specific day of the year. This is the sort of information that big phone companies like Verizon and AT&T have at their fingertips. When you walk around with your cell phone on, you give these companies data about your location. AT&T and Verizon then strip that data of identifying information and sell it to city planners, commercial interests, and others. Verizon even claims the ability to build a demographic profile of people gathered together in a specific place for a specific thing, such as in a stadium for a rock concert or a sporting event.11
Navizon puts the same sort of capability in the hands of individuals with small budgets but larger time horizons. Navizon CEO and founder Cyril Houri is marketing the device as a way for entrepreneurs to do location planning. There are some limitations. Because the device measures Wi-Fi from smartphones, it’s also biased toward younger adults (18–25) who are—not surprisingly—more likely to own a smartphone than are people over age sixty-five. High-income earners also show up more often than low-income earners. But the current profile of smartphone users is not the future profile.
Navizon’s analytics system won’t disclose the names of specific people whom the device picks (unless those people opt in to the
Navizon buddy network) but the system can recognize individual phones. It has to, in order to count them. If someone follows roughly the same pattern every day, hitting work, the store (or the bar), then home in the same time window, the difference between tagging the phone and tagging the person effectively disappears. MIT researchers Yves-Alexandre de Montjoye, César A. Hidalgo, Michel Verleysen, and Vincent Blondel from the Université catholique de Louvain took a big data set of anonymized GPS and cell phone records for 1.5 million people, the sort of stripped-down location data that Verizon and AT&T sell to corporate partners to figure out the types of people who can be found at specific locations at particular times of day. The data consisted of records of particular phones checking in with particular cell antennas. What the researchers found was that for 95 percent of the subjects, just four location data points were enough to link the mobile data to a unique person.12
A growing percentage of smartphone users voluntarily surrender data about themselves wherever they use geo-social apps. Facebook, Twitter, and Google+ all have “check in” features that broadcast your location to people in your network. Other, more creative services will facilitate specific interactions based on what you’re looking to do wherever you happen to be.
An app called Sonar will identify the VIPs in the room; Banjo will tell you the names of nearby Twitter, Facebook, and Instagram users; a service called Grindr, launched back in 2009, will pinpoint the location of the nearest gay man who may be interested in a relationship—of either the long- or short-term variety.
The Naked Future Page 3