• More Blogs
• Observations »
• Bering in Mind »
• Cross-check »
• Extinction Countdown »
• Solar at Home »
• Expeditions »
• Guest Blog »
• Anecdotes from the Archive »

SAN JOSE, CALIF. (March 11, 2009) -- If you want to create a more sustainable world, you need to build better cities. If you want to build better cities, you need to understand the networks that make up cities. And if you want all of those networks to actually connect, you need to improve the "dumb grid" and make it a smart grid.

That, in a nutshell, was the message of the third day of the eTech conference. Perhaps unwittingly, Chris Luebkeman of the global design firm Arup encapsulated the theme in his morning keynote: “We need to build [X] for the elderly population,” he said, “because hopefully, we’ll all get old some day.” His X was cities -- specifically, the need to design cities so that seniors can easily get around without cars -- but X could have been networks, copyright, the environment, energy systems or experimental science itself. Improve the future. Build it better. Here’s how.

Luebkeman spoke of the tremendous challenges and opportunities that have arose with the mass urban migration patterns in the developing world. More than 600 million people in China are expected to migrate from the countryside to cities in the coming decades. It’s a number that can slip by, but consider: That’s equivalent to the combined populations of the U.S., Canada, Mexico, Australia, Britain, France, and the whole of Scandinavia. Think of all the infrastructure that has to be built, all the communities that must be designed. Will those migrants fall into sprawling favelas, or will we seize the opportunity to design an intelligent new urban infrastructure for the 21st century?

Answering that question and others requires tools that help us understand how cities operate. Andrea Vaccari, a research assistant at MIT’s Senseable City Lab, described a number of projects that tap into GPS-tagged Flickr images, cell phone data, and internet traffic to create real-time maps of activity in a city. What parts of a city are interesting to tourists? Just map out what they take the most photos of. Need to know if an “abnormal event” like a traffic accident just happened? Look for abnormal spikes in cell phone usage in a particular area.

Tony Jebara, the chief scientist of Sense Networks, is counting on our tribal roots to guide development -- sort of. He explained how his company is mapping the modern city by tracking the movements of people within it. Sense Networks starts with raw location data from taxis and GPS-enabled cell phones and the like. It integrates this data to produce heat maps of where people are at any given time.

But that’s only the beginning. Sense Networks tracks the movement of individuals throughout the week, and determines that a person who often goes from work in the financial district to a sports bar on Thursday night (for example) is likely to be similar to another person who moves from a different financial district in a different city to a different sports bar. Using this data, it assigns individuals to one of 20 different “tribes”—young & edgy, business traveler, or mature homebody, for example. It’s a semantic approach to location data, and it’s working: The company currently has 4 million users.

What can you do with this information? Say one person in a tribe goes to a restaurant in London on vacation and loves it. Others in the tribe are more likely to enjoy it as well, and so this nascent social network can help other travelers find good food. But you're probably anticipating the other obvious use: Advertisers are very interested in targeting ads to consumers based on very fine-grained data about who and where they are.

And today, no discussion of the technology of the future -- of cities or anything else, for that matter -- would be complete without a discussion of the smart grid. eTech was no exception, which made sense given the focus on networks: Energy networks are just another kind of network, in one way of thinking. So when Tom Raferty of Redmonk argued against the “myth” that renewable energy is expensive, he blamed a "dumb grid."

“In fact,” he said, “on the wholesale markets, the cheapest electricity is the one with the highest renewable content.” In these wholesale auction markets, renewable energy sources will bid into the grid at any price—after all, there’s no additional cost to the renewable producer to run the windmill. Contrast this to a fossil-fuel plant, which must consume material and has substantial startup and shut-down costs.

The problem, Raferty said, is that the current worldwide electrical grid is dumb. Example: In Texas, houses in a county directly next to one of the largest wind farms in the world may suffer rolling brownouts because that county pulls its electricity exclusively from a coal-fired plant. Raferty argued that every power generator must be linked to every power consumer—even if that generator is a solar panel on top of your house.

But perhaps more importantly, the grid should also transmit information about the power source—price information and environmental information. This way, all the appliances in your home “could subscribe to this pricing information and modify their behavior in response to those prices,” he said. “What if you fridge changed its thermostat when electricity was cheap? What if your clothes driers or dishwasher came on automatically at the time of night when electricity was cheapest?”

It’s the kind of future—technologically savvy, ecologically conscious—that the eTech crowd could believe in.

Photo of Andrea Vaccari (Senseable City Lab, MIT) by Michael Moyer/copyright Scientific American

Comments