The longer I have watched the technology world develop, the more cautious I’ve become about predicting what will happen next. Products that seem destined to catch on often fade and vanish; companies, ideas, and even business leaders with the odds stacked against them somehow prevail. In the early 1980s, Xerox introduced its Star computer, which had most of the intuitive, easy-to-use features that Apple offered a few years later with huge fanfare in the Macintosh (and that Microsoft eventually added to Windows). Now the Star is long forgotten, while Apple and Windows live on. Ten years ago, as the much-loved Word­Perfect was being eclipsed by Microsoft Word, and Lotus 1-2-3 by Microsoft Excel, it would have seemed foolish to bet that Adobe, a Microsoft competitor with nothing like its resources, could establish its PDF (Portable Document Format) as a standard. Now PDF technology is ubiquitous, thanks to luck and to Adobe’s decision to distribute Reader software free, as part of its “Adobe everywhere” strategy.

Technologies developed originally for one purpose often prove most important in other, unforeseen ways. The most famous example is the Internet itself. Its novel system of communication—messages are broken into many tiny “packets” of information, each of which follows the least-congested path toward the destination the instant it is sent—was intended to let the network survive damage after a nuclear attack. So far its real importance has been allowing the worldwide Internet to expand constantly with amazingly few bottlenecks.

My nominee for the innovation now mutating into the most interesting new forms is a technology called Broadband over Power Lines, or BPL (about which I first wrote in The New York Times two years ago). For years it has been touted as the answer to a problem that, it now appears, will be solved in other ways. But meanwhile it is providing a useful new service for consumers. It could even play a part in addressing environmental and national-security concerns.

The concept behind BPL is that the same wires that carry electric power to homes and offices—at higher voltage across trunk lines and at (usually) 110 volts inside a building—can also carry high-speed Internet traffic. Somehow this seems harder to believe than the fact that a single wire from a cable-TV company can handle e-mail and Web signals while also bringing in TV programs. When you look at a Boeing 747 on the ground, you don’t think it can fly; when you stare at an electric outlet, you don’t think e-mail can come out of it. But it can. The reason, to oversimplify, is that a wire can simultaneously carry signals at a variety of frequencies, from sixty cycles per second for electric power to 3 million cycles per second and up for Internet data.

The initial promise of BPL was to help solve the Internet’s “last mile” problem, which until recently seemed a major concern. The last mile is the distance between main data lines and individual homes, businesses, and schools. It was seen as a social problem, because extending coverage to remote areas was costly and slow. Cable-TV companies have had an advantage over phone companies in solving last-mile issues, since cable for TV, which can carry Internet traffic, already reaches many homes, far more than are close enough to a phone switching station to receive DSL service. But since virtually all homes and other buildings are already connected to the electric grid, BPL could in theory offer near-universal broadband coverage at very low capital cost. Toward this end the Federal Communications Commission has issued rulings and standards to promote the spread of BPL.

This way of delivering Internet service, known in the trade as “Access BPL,” will probably be one part of America’s eventual broadband system. But it is rarely discussed anymore as a cure-all. In part this is because it is being complemented, leapfrogged, and in some cases outflanked by other technologies, including new forms of wireless strong and fast enough to eliminate last-mile problems by broadcasting from transmission towers to surrounding neighborhoods.

Access BPL has also been hindered by ferocious opposition from the Amateur Radio Relay League, or ARRL, which represents ham-radio operators. Whenever electric current is sent through an unshielded wire, the wire becomes an antenna that broadcasts radio waves. The ARRL points out on its Web site and in letter-writing campaigns that the signals resulting from BPL transmission can create static on the frequencies used by hams (as normal electric transmissions do not). The more electric lines are used to transmit data, the group warns, the harder it will be for hams to do work that matters—such as coordinating rescues, as they did after Hurricane Katrina. Some BPL companies have “notched” their transmissions to avoid frequencies that would interfere with hams, but this makes BPL slightly less efficient. For these and other reasons, “there’s going to be no single solution for covering the last mile,” says Theodore Schell, a former Sprint executive who is now involved in BPL startups. “Power lines will be one very effective tool in a toolbox of wireless, fiber, and other forms of last-mile transport.”

What, then, is the point of discussing BPL? For the average user, BPL offers about the only way to work around one of the rarefied discontents of the modern computer age: not having broadband access wherever you want it.

Sure, you have a high-speed Internet connection coming to your house, via cable, DSL, or some other means. And you feed that connection into a router, so that you can link other computers to the Internet directly with Ether­net cable or via a home WiFi network. If you’re in a modern dorm or office building, Ethernet cable has already been snaked through the walls to provide access points in every room. But most houses aren’t pre-wired, and most large or multi- story houses have dead spots that the WiFi coverage can’t reach.

This is where another form of the technology, “In-House BPL,” comes in. It allows you to create an Internet access point, or another WiFi hotspot, wherev­er you have an electric outlet. Considering the difference it makes, it is simple to set up.

You buy one adapter that converts your incoming Internet signal to a form that will travel through your home electric system. These adapters are about the size of a thick cell phone, and good ones are available for well under $100 (and prices are beginning to fall as the devices become more popular). I have both the Netgear XE102 and the Linksys PlusB10; a full list of suppliers is available on the Web site of the industry’s standard-setting organization, the HomePlug Powerline Alliance. You connect the adapter to your router or modem by Ether­net cable, and you plug the adapter into any outlet on the wall.

From that point on, your home wiring is also an Internet carrier. You can plug in a second, similar adapter anyplace else in the house and have an instant Ethernet port (yes, that means spending another $50 to $100). Or you can plug a different model, like the Netgear WGX102 (similar size and price), into any socket and create a new WiFi zone. (The adapters now available can create either an Ethernet port or a WiFi hotspot, but not both.) You can move one of these adapters from room to room or leave several of them permanently in place. On early models, the transmission speed through the electric wires was noticeably slower than through a direct connection; BPL technology is improving so fast that now there is little speed penalty. These in-home systems have escaped opposition from the ham-radio operators, both because they radiate less interference than transmission on outdoor lines and because they are “notched” to protect the hams.

That’s the benefit to the individual. The potential benefit to the country comes from another emerging BPL feature: “smart grid” operations.

The nation’s electric system is indispensable to modern life but in many ways old-fashioned. As anyone who has lived through a long post-storm power outage knows, it can take electric companies hours or days to figure out exactly where the problem is and how many customers have been affected. “It’s a one-way broadcast of energy,” says Bill Berkman, the chairman of a BPL company called Current Communications. “The companies have no knowledge whatsoever about whether the power is getting to consumers, whether enough is getting there, how the load is balanced across the system.” His company’s business includes installing small sensors on transformers and utility poles, which send status reports back to the head office over the power lines using BPL data-carrying technology. “You can monitor the voltage, the tilt of the pole, the oil pressure in the transformer drum—these all sound boring, but for a utility they are huge,” Berkman says. The company has installed smart grid systems for utilities in California, Ohio, Texas, and elsewhere. The industry trade press is full of reports about potential huge savings to utilities through automated meter reading, easier maintenance, and other advantages of smart grids.

The real value of a self-sensing electrical system, according to Berkman and others, lies in the electrical generating plants it may keep from being built. This benefit comes from the possibility of using a smart grid’s monitoring capacity to detect exactly when, where, and why power use is surging—and to shift or limit it in precise ways, to avoid costly spikes in demand. The bane of the power business is the “peak load” phenomenon—providing the extra 5 percent to 10 percent of power to run air conditioners on a summer afternoon or electric heaters in the winter. Meeting that small increase in demand is disproportionately expensive, because the extra generating capacity is usually more polluting and less efficient than the normal plants. “If you can reduce the peak even a little bit, you can reduce costs a lot, especially in some regions of the country,” says Susan Tierney, a former utility regulator in Massachusetts (and my sister).

In principle, smart grid systems should be able to do just this. Devices on equipment with high power demand—air conditioners and refrigerators in the home, a wide variety of commercial equipment—could, with the customer’s agreement, automatically slow or shut these appliances down briefly as peak demand neared. Or they could report, via thermostat readouts or some other measure, real-time price changes, so the customer could decide what to turn off. Many new power plants are built to meet peak-load demand. In much of the country building new plants is difficult or impossible because of environmental or other regulatory factors; in all parts of the country, power generation creates greenhouse gases or nuclear wastes. If BPL can reduce those, it’s got my attention—as it’s already got my customer loyalty.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.