Sensor Nation, UK Style

At the time of the July 2005 London bombings, the average Londoner was on camera an average of 300 times a day. That didn't prevent an attack, of course. Surveillance cameras — the UK had an estimated 4.2 million of them in 2005 — can help the police solve a crime after it's committed, but no one is looking at them in real time.

A organization named Internet Eyes would change that. It has developed a scheme that would allow ordinary citizens to watch CCTV (closed-circuit television) video streams and sound an alarm when they see things that are, well, alarming. Internet Eyes holds out a reward of £1000.

According to the BBC report that broke the story, today London has "one camera per 14 people" and yet the police have "estimated that in 2008 just one crime was solved per thousand CCTV cameras in the capital. The deficit was partly blamed on officers not being able to make the best use of the many thousands of hours of video generated by CCTV."

It's hard to imagine the Internet Eyes program, which is slated to begin next month in Stratford-upon-Avon, will do much to improve that. While news reports make it sound as if all you have to do is see a crime and report it to collect your £1000, the reality is a lot more complicated and a lot less lucrative.

Viewers are anonymously monitoring random video feeds streamed from privately owned establishments. At no time can Viewers designate or control the video feeds they receive and the locations of the feeds are not disclosed.

The instant a Viewer monitors an event, an alert can be sent directly to the owner of that live camera feed. The alert is sent along with a screen grab, identifying the image you have observed. Only the first alert received by the camera owner is accepted.

The camera owner will then feedback (rate) the result of the alert. Their feedback is converted into points and entered into a Viewers monthly league table. At the end of each month the highest scoring Viewer will receive the reward money; this could be split in the event of a tie.

Viewers register for free with no recurring fees. Each Viewer has 3 x alerts per month allocated to their account for free. Viewers are able to ‘top up’ their alerts through PayPal if they so desire. The free allocations of alerts are limited to prevent system abuse.

So, for one thing, you don't get to choose your feed. Even if you would probably be a more attentive viewer viewing your own neighborhood or a favorite store, you probably will never get that chance. Then, your success depends on the utility rating the camera owner gives to the alert you send — that, plus an opaque points program. Finally, only a single thousand-pound prize is awarded, no matter how many crimes are "solved" by CCTV alarms that month. (Oh, and having to pay for your own alarms after the first three — nice touch. Still, people pay for their votes on American Idol, without any prospect of an award, so maybe they'll pay to play here as well.)

"Solved" is in quotes because it's not clear what it means for a crime to be solved by a CCTV camera. Back in August, privacy expert and security entrepreneur Bruce Schneier noted,

To me, the crime has to have been unsolvable without the cameras. Repeatedly I see pro-camera lobbyists pointing to the surveillance-camera images that identified the 7/7 London Transport bombers, but it is obvious that they would have been identified even without the cameras.

Then there's the question of how well people will assess that a crime is being committed by seeing a few seconds of grainy video stream by their eyes. Football referees often can't determine that a facemask was grabbed, or that a foot stepped out of bounds, when they watch higher quality video streams than those of the average street camera.

The BBC report noted that crimes aren't noticed in real time today because "viewing hours of mostly tedious and often poor quality images is a lengthy and unpopular job." If that's true when people who are doing it receive a regular paycheck, does it get any less tedious when done in the hope of a windfall hardly more likely than winning the lottery? (Still, people pay good money to engage in the same repetitive motions that factory workers have complained about for decades, in the hope of hitting a slot-machine jackpot.)

If the program is unsuccessful, few people will be watching. but if the program is successful, will the camera owners, and ultimately the police, have enough staff on hand to evaluate an alarm quickly enough to apprehend a suspect at the scene of the crime? It seems unlikely. The bottom line is that probably some people will watch the video streams, the police will be alerted to a couple of crimes, and Londoners' privacy will continue to erode.

Exactly one year before the London bombings, Spectrum published a special report entitled "Sensor Nation." In an article entitled "We Like to Watch," my colleague Harry Goldstein presciently wrote,

For entertainment, we gather in front of the tube for mass-mediated group therapy sessions called reality shows. Hundreds of millions of us around the globe tune in to watch people who eagerly endure excruciating plastic surgery; stab each other in the back for a chance to work for Donald Trump; or wolf down sea worms, cockroaches, and worse to survive on a desert island. For Generation Y, "Big Brother" is a reality television show, where, for a chance at winning half a million dollars, contestants volunteer to be cooped up in a house with total strangers and have their most private moments broadcast to a hungry audience.

It's not hard to imagine a near future of reciprocal transparency when all of us are watched and can watch right back. We're halfway there.

Advertisement

Tech Talk

IEEE Spectrum’s general technology blog, featuring news, analysis, and opinions about engineering, consumer electronics, and technology and society, from the editorial staff and freelance contributors.

Newsletter Sign Up

Sign up for the Tech Alert newsletter and receive ground-breaking technology and science news from IEEE Spectrum every Thursday.

Advertisement