The technologies that will make cities smarter are error-prone and brittle. What does that mean for city energy data?
Maybe, if you’ve been living under a rock for the past seven years, you haven’t heard of Masdar City. This from-scratch city in the United Arab Emirates is being built to house more than 40,000 eventual residents, and the government of Abu Dhabi intends for it to become a global clean-tech hub.
Designed by architectural firm Foster + Partners, Masdar will use solar, wind, and geothermal energy to get as close to the goal of zero carbon as possible. Greywater will be recycled; biological waste will be used as fertilizer. Within the city, people will get around using personal rapid transit and electric cars.
Sounds nice, doesn’t it? Masdar is just one of a clutch of “smart cities” whose ambitious plans have been trumpeted since the mid-2000’s, their dreamlike architectural renderings splashed all over the internet.
The creators of these new cities make grand claims. New Songdo City in Korea offers “the ultimate lifestyle and work experience,” much like a five-star hotel or luxury car. PlanIT Valley, planned for Portugal, touts an “Urban Operating System” that will digitally unite and manage energy, mobility, access, and infrastructure — a kind of citywide brain that can control every movement of the city’s limbs and digits.
Smart-city marketing hype has proven near-irresistible, and the media’s coverage of these and other projects has been breathless. But there are so many things to be skeptical about, it’s hard to know where to begin.
How able are these city authors going to secure billions of dollars of funding despite a shaky global economy? What’s the likelihood that 225,000 people will relocate to a greenfield site in the hills of Portugal, mainly because it’s “smart”? Not least, what’s the chance that these cities, assuming they are eventually fully built out (a big if), will actually deliver on their heady techno-utopian promises?
Unpacking the Smart City Myth
Luckily, urbanist Adam Greenfield took it upon himself to unpack the myth and attendant problems of these smart cities in a new e-book. Conceding that these schemes may not even come to fruition, Greenfield argues that it’s nevertheless important to study the assumptions that lie behind them, and the language used to describe them, so we can better understand their strain of urban thinking.
After all, smart-cities rhetoric is now being used of countless existing metropolises, from Rio de Janeiro to New York to Bangalore, and smart-city technologies pioneered by companies such as IBM and Siemens are increasingly woven into urban environments.
Even when their code is clean, the innards of smart cities will be so complex that so-called normal accidents will be inevitable. — Anthony Townsend
What Greenfield discovers is that the smart city is bogged down in misconceptions about how cities operate (they’re not machines) and how its populace behaves — that the populace is always “in need of active management and incapable of making wise use of information about its own behavior.”
The smart city, unlike real cities, exists in generic space and time, impervious to local conditions and chronological change. It has top-down governance and is fatally overspecified: that is, it is designed with the belief that every future technology, use, or need has been accounted for at inception.
Such a belief is patently wrongheaded, and Greenfield offers the perfect example. The smart cities he writes about — which were conceived in the early to mid 2000s — propose embedding sensors in sidewalks and bus stops and using their data in networked city services, but despite this high degree of sophistication, fail to acknowledge “the one piece of networked information technology that citydwellers all over the planet already have ready to hand.”
Yep: The smart phone is all but absent from the smart city.
Big Energy Data, Big Questions
What does all this mean for city energy data? At IMT, we’ve long hoped and planned for a future in which building energy data is measured, collected, and published on an urban scale. For us, a smart city is a place where you can compare the energy efficiency of every building on your block with relative ease.
But reading Greenfield’s Kindle book prompted me to wonder: Are we liable to fall into these same traps?
We are probably safe from some. We’ve never believed that energy data should be anything other than free and open-source — and even the platform we recommend for collecting it, ENERGY STAR’s Portfolio Manager, is free and publicly (not privately) run. (One hallmark of the smart city is that it runs off a proprietary platform, Greenfield notes.)
We think energy data will be a powerful tool for city administrators, yes, but we locate the real transformative potential elsewhere — with the apartment-hunter, the small business owner, the real-estate investor, the environmentally-minded citizen. We don’t overlook the role of the smart phone in bringing this information to the public.
Other of Greenfield’s lessons, though, we should take to heart. Let’s not be victims of our own hype — so excited about the potential of the data we’re gathering that we lose sight of its limitations.
No dataset is ever “complete,” especially in the context of something as complex and ever-shifting as a city. And as Greenfield points out, there is no such thing as “just the data,” either; it’s always shaped by how it’s being measured. We should continually ask ourselves how we arrived at our metrics and whether they’re (still) the right ones for our purposes.
Technologies are buggy and brittle, as Anthony Townsend, another writer on smart cities — real, not hypothetical ones — points out. “Even when their code is clean, the innards of smart cities will be so complex that so-called normal accidents will be inevitable.”
What would be the worst accident for building energy data? Portfolio Manager crashing? In fact, we got a taste of this during the recent federal government shutdown, when U.S. EPA took the tool offline. As Greenwire reported (subscription required), the founder of an energy management company testified before the U.S. Senate that his business — and energy retrofits in general — were suffering as a result.
The wrong metrics could lead us to the wrong conclusions and misguided actions. Unintended consequences could arise from using a new method to clean the data or correlate it with some other data set.
The worst accident might be one we can’t anticipate. Above all, as we start to gather and analyze data from thousands of buildings in cities around the country, we must be willing to expect the unexpected.