Energy efficiency has been called the low-hanging fruit for reducing carbon emissions, because it actually pays for itself. But it can be difficult to get people to take simple steps to save energy, and it's hard to maintain those savings over time. For example, people offset the savings from a new efficient refrigerator by changing their habits or buying a plasma TV.

A company called OPOWER has what appears to be a successful strategy for dealing with this human element of energy efficiency. It has increased participation in energy efficiency programs from about 5 percent to 80 percent for the utilities that have used it so far, and the savings appear to be sustained.

So far, since its founding in 2007 the company claims its program has saved over 90 million kilowatt hours of electricity. The company has even attracted the interest of the White House--last week President Obama dropped by, highlighting the company as a source of the green jobs he hopes to help create.

The company's approach is based on the idea that people want to fit in. OPOWER first lets people know how their energy use compares to that of their neighbors. Then for each billing period the company gives them a single tip that they can act on, also connected to what they're neighbors are doing, such as "Most people in your area keep their AC at 78 degrees." They also tell people how much energy they will save.

It's obvious how this could work with people who are using more electricity than average. But for those using less than average, the company also compares them to their "efficient" neighbors to motivate change.

The company makes use of public information (such as from the tax assesor's office and weather data) as well as third-party demographics to tailor results. If a person is renting, OPOWER doesn't recommend insulating the house, for example. It also has software that looks at energy use patterns to identify likely sources of energy waste. Power spikes during hot days greater than that shown in neighboring houses could suggest someone has an inefficient air conditioner.

It seems like a promising approach. But it's not going to solve the world's energy problems on its own. On average the program has cut energy consumption by a modest amount--about 2.5%.

The cost of solar power could be cut 30 percent without improving the performance of individual solar cells, says Daniel Alcombright, vice president for North America at Solon Corporation, based in Berlin.

At the ReTech 2010 conference in Washington, DC, last week, Alcombright laid out his company's plans for cutting costs and increasing power output, which will result in lower costs per kilowatt hour over the lifetime of solar panels, making it more competitive with conventional sources of electricity.

About half of the cost of solar power comes from the cost of the solar panel itself. The rest comes from installation costs, the costs of electronics for converting DC to AC for the grid, and other things like the cost of land and maintenance.

When Alcombright was first hired at Solon a few months ago, the company sent him to a solar installation project, to give him an idea of how things went. He saw plenty of ways things could be improved. For example, highly-paid electricians spend hours constructing assemblies for conduits, when such things could be built for less in a factory. Solon is working on fixing this, and making other changes: Larger solar modules with quick mount frames could also reduce overall construction costs. So could standardized plans for solar farms, so that each new project doesn't have to be engineered anew. Finally, low cost tracking systems and software for optimizing their performance in different locations and from season to season could increase power output from the same solar panels.

These sorts of inefficiencies are evidence of an industry that's still in its infancy--but they also suggest there's a lot that can be done to make solar cheaper, even without major advances in solar cell efficiency.

A key part of Energy Secretary Steven Chu's plan to revamp the U.S. Department of Energy and push forward new clean energy technologies is the "Energy Innovation Hub," a research center modeled on the legendary Bell Labs, which generated many key advances for computers and the Internet. But his plans ran into trouble earlier this year, as Congress proved reluctant to fund the eight hubs he had in mind. At one point it looked as if none would be funded, but in the end, three were, at $22 million each. On Tuesday the DOE announced details about the three hubs.

The first three hubs will focus on these problems: making fuels from sunlight, designing energy efficient buildings, and using computer modeling and simulation to develop better materials for nuclear reactors. More details are available for the sunlight to fuels program than the others--the department has issued a formal funding opportunity announcement to get proposals. The nuclear hub seems to be the next in line--DOE already had a workshop on the subject in early December. Details about all three can be found here.

In their announcement this week, the department put the hubs in context. One of the main congressional objections was that the hubs seemed to duplicate other new programs at DOE. A new set of Energy Frontier Research Centers and a new agency called the Advanced Research Projects Agency -Energy (ARPA-E) both fund research that could transform energy technologies. The frontier research centers are meant to tackle specific, basic science questions--the kind of basic research that most economists say the government should be funding, because industry won't. ARPA-E is also for funding risky research, but the focus here isn't on basic science. Rather it's on research that could lead to very big changes in energy, but that involves technology so different from existing technology that industry isn't likely to fund it, even if it doesn't require fundamental science breakthroughs. Early projects being funded here include a liquid battery that's quite different from today's batteries, which use either solid electrodes or electrolytes.

The key idea behind the innovation hubs is to bring together a larger group of researchers, "ideally under one roof," according to the announcement. The idea is that if you get enough brilliant people together from different disciplines, they're going to generate a lot of ideas really quickly, and just as quickly weed through them to keep from working on dead ends. The idea is to eliminate the problem, sometimes seen today in battery research for example, where isolated researchers work diligently to solve problems that other researchers have already solved, or perhaps more importantly, that other researchers have shown face insurmountable obstacles and so should be abandoned in favor of other approaches. As Chu has put it when describing Bell Labs, if you have an idea , chances are you'll find world experts in the relevant subjects down the hall, who you can run it by. This is key because energy-related problems tend to transcend scientific and engineering disciplines, requiring, for example, the collaboration often of physicists, materials scientists, mechanical engineers and microbiologists.

Of course a lot will depend on attracting the right people, which can be difficult these days with venture capitalists waiting to lure the best researchers to lucrative jobs in the private sector. That challenge will be made more difficult by the fact that each of the hubs was funded at $22 million for the first year, not the $35 million requested.