Spot The Terrorist

Jay Walker achieved fame and fortune as an internet pioneer (, then notoriety and considerably less fortune as an icon of the dot-com bust. But his legacy might one day be a sweeping scheme for homeland security that doesn't earn him a buck. For the past few months, Walker has quietly been visiting key figures in Washington, D.C., to brief them on an idea he calls US HomeGuard. It is audacious, ingenious and a little bit scary. Basically, it attempts to protect chemical plants, reservoirs and airports--all targets where terrorists could get horrifying results with relatively little effort--by a system involving 10 million Webcams and a stay-at-home army of up to a million watchful citizens.

Did I say a little bit scary? I'm getting hives just typing this.

But scarier still is not doing something to protect ourselves against demonstrably real threats. We found this out, of course, on September 11. Walker himself saw the tragic columns of smoke as he was driving down the West Side Highway to visit a branch of his business-incubation company, Walker Digital. It was located on the 21st floor of the Woolworth Building, looking straight into the towers.

Afterward, recalls Walker, "we asked ourselves, 'What can we do?'" Eventually he and his team focused on the seemingly intractable problem of keeping hostile trespassers away from the perimeters of tempting "soft" targets--like chemical plants in populated areas or reservoirs that serve cities. Not surprisingly from a dot-com innovator, the solution relies on the distributed power of the Internet. Walker also claims it's financially self-sustaining, non-intrusive to ordinary citizens and a potential morale builder for the hundreds of thousands who will get a chance to actually do something about terrorism.

The first part of the system involves installing rugged outdoor Webcams along the fences protecting the sites. At $1,000 or so each, they would be equipped with heat sensors, microphones and loudspeakers. They'd be connected to the Internet, probably by Wi-Fi. Every five seconds they'd grab a picture of their little slice of forbidden perimeter and send it, encrypted, to a US HomeGuard data center--operated by an independent company that runs like a security utility. Computers would decrypt the picture and compare it to the previous shot. If there's any change, the picture would get sent to at least three "spotters."

These would be adult citizens who've taken a brief online training course in recognizing interlopers, based on digital pictures. The pay for this part-time work would be $8 to $10 an hour. If any of the first round of spotters saw something suspicious, the system would "flood the zone" by sending more pictures from that camera and those around it to 10 new spotters. If this group confirmed the alert, the professionals at the data center would take over. They could confront the trespassers via the speakers in the Webcam ("Hey, you!") and, if necessary, contact local authorities. Result: vastly improved protection.

First hurdle: the price tag. Walker envisions US HomeGuard's protecting 47,000 facilities with about a quarter-million miles of perimeter. At his estimated $50,000 a mile for setup, that's about $12 billion, for starters. Who's going to pay? According to Walker, the facilities will absorb the cost, and happily. Their insurance rates will plummet, he says, and the facilities will be protected not only against catastrophic destruction but garden-variety trespassers who are a constant nuisance.

Then there's the Big Brother issue. When Walker unveiled his plan earlier this year in an off-the-record talk at the fabled TED tech conference, the idea of people-scanning cameras on the Net "creeped everybody out," says one attendee. Walker thinks this is a bogus issue, since cameras will be pointed only to areas where people aren't supposed to be. "I don't see a problem with that," says Marc Rotenberg of the Electronic Privacy Information Center. But Rotenberg worries about whether a future iteration of the system might include facial recognition and other features that could track ordinary people.

One criticism that Walker takes more seriously is that the very size of the system is something people might balk at. Indeed, the concept of recruiting a surveillance force of a million people is on its face horrifying. But equally depressing is a fact we can't dodge: without bold and innovative schemes to protect ourselves, we're sitting ducks.

Walker's goal right now is getting the federal government to spend $40 million to build a prototype to see if the idea works. If a test is successful, a working system could be in place in less than a year, he says. Working with Steven Hofman, a well-connected Republican Party adviser, he's been briefing key people in Congress, the executive branch and private sector. People are shocked, he says, to learn that he's not after big government contracts and will take no profits from its efforts. I'd be curious to see if Jay Walker's big scheme works. And a little bit scared if it does.