Consider, for example, a vast interconnected electrical energy grid for the North American continent. By 2050 this grid will interconnect several hundred million local sites. There are two key aspects of this future grid that will make a huge difference: massive long-distance electrical power transmission, and local storage of electrical power with real-time pricing.
He points out that we need better battery technology, so that people can rely on power generated and stored locally. We also need more efficient long-range transmission, "permitting, for example, hundreds of gigawatts of electrical power to be transported from solar farms in New Mexico to markets in New England. "
He suggests that nanotechnology can provide the solution to these technical challenges. If so, then the electrical grid will start to look more like the Internet. It will be more robust, and people will be less dependent on any one source of supply.
UPDATE: Marty, in a comment, points to this article about an energy Internet. It differs in its technology emphasis from the Smalley piece. In a way, that serves to underline the issue in the discussion question.
For Discussion. How can the government best determine where to spend money on energy research, for example on hydrogen vs. nanotechnology?