Technocreep (6 page)

Human-computer interaction experts agree that “always on” video links will take some getting used to, and people may find them disturbing, especially if they are installed by the boss. They are, in an ­uncanny valley sense, human (because human eyes and ears are processing the images and sound remotely) and simultaneously non-human since they are devices hanging on the wall.

Many police forces are equipping their members with body-worn cameras to document arrests and other interactions with civilians. A year-long study in Rialto, CA found that the cameras resulted in “more than a 50% reduction in the total number of incidents of use-of-force compared to control-conditions.”
63
The authors suggest that the behavior of citizens may also have been modified: “Members of the public with whom the officers communicated were also aware of being videotaped and therefore were likely to be cognizant that they ought to act cooperatively.” One of the study's authors was Rialto's police chief, William Farrar. He reports there was some reluctance by his officers to wear the cameras, which they referred to as “Big Brother.” However, as Chief Farrar told the
New York Times
, he reminded them that civilians can use their own smartphone camera, “so instead of relying on somebody else's partial picture of what occurred, why not have your own?”
64

While the presence of cameras can make people more civilized, it can also have the opposite effect. Stories abound of bystanders pointing their smartphones at accident scenes and rapes in progress instead of calling for assistance. The Vancouver police blame cameras for fueling some of the violent rampages in that city, commenting that “the 2011 riot can be distinguished as perhaps the first North American social media sports riot.”
65
The police report goes on to say that “the acting out for the cameras seen in the 1994 riot was multiplied many times more in the 2011 riot by the thousands of people cheering the rioters on and recording the riot with handheld cameras and phones.”

Cameras have come a long way since Matthew Brady captured the horrors of the Civil War on glass plates in the 1860s. They have morphed from a heavy object that sat on a substantial tripod and needed multi-minute exposures into chip-sized sensors that fit in our smartphones, laptops, even a pair of glasses. The next step is quite likely to be a camera in a contact lens. Korean researchers have ­created a proof of concept of this and tested in on rabbits.
66
Mounting them on everything from traffic lights to dashboards to police officers changes cameras from something that we pull out, deliberately aim, and focus into an organic, omnipresent part of our environment.

Just as GPS chips became smaller and cheaper and are now installed in your smartphones and camera and perhaps soon in your newborn baby, camera chips will continue to proliferate. The 3D printing revolution will allow people to make a hollowed-out button and stick a tiny camera inside to sneakily capture your photo.

Then again, they might just wink at you while wearing Google Glass.

I saw Google Glass before it was even a twinkle in Eric Schmidt's eye.

As a technology writer and reviewer, I was sent demo versions of all sorts of products, including some that never made it to market. In the mid-1980s, a package arrived with one of the first heads-up television displays aimed at the consumer market. It was set of glasses with a tiny monitor and a prism that allowed you to watch TV while still participating in normal life.

The device, now consigned to the tech dustbin, did give me one moment of profound technocreepiness. I was testing it one night in my university office, using it to watch
60 Minutes
. The cleaning lady came in to empty the trash. I will never forget what happened next. I saw a chimera—an elderly lady's body with Mike Wallace's head grafted on top. I screamed. She screamed. It seemed like a dumb way to watch TV, so I sent the thing back and wrote a lukewarm review: it was also extremely uncomfortable to wear.

The introduction of Google Glass has brought this type of technology literally to the public's eye. All of a sudden, people are walking around with a device that enhances their ability to grab information out of the ether. Google Glass wearers can potentially recognize your face as they shake your hand, and then casually glance upwards to retrieve your kids' names and birthdays.

But what really alarms many is that Google Glass can also secretly take a picture, or record a video, and immediately upload it to the Internet, just by the wink of an eye or the raising of an eyebrow. Google Glass does have a light to indicate when it is taking a photo or recording video. People promptly found ways to subvert it.

Chris Barrett, one of earliest users of Google Glass, said he was having trouble using the device in bright sunlight. So, he designed a clip-on sunshade, a piece of plastic that can be run off on a 3D printer. He has even made the code for it freely available online.
67
In addition to blocking the sun, Barrett's creation happens to cover that pesky light that tells others you are recording them.

Barrett loves wearing his high tech eyewear in places it is not supposed to go. He has reportedly filmed inside an Atlantic City casino and, in a stroke of luck, apparently became the first person to record an arrest with Google Glass. This little documentary was quickly posted to YouTube.
68

Dozens of innocent bystanders, some of them children, appear in that video, and many are facing the camera. Coupled with massive facial image databanks, and advances in facial recognition software, they could probably be identified. A bizarre new social rule is emerging: if you are really trying to protect your privacy, you should stay away from arrests, car accidents, riots, and landmarks—anything that people are likely to photograph. Perhaps you should not go out at all.

The camera function of Google Glass and similar devices is really a giant social experiment to redefine where photography is acceptable and what behaviors will get you called a creepy “glasshole,” a cheeky term that even Google has started using.
69
Even in a public place, where photos are generally fair game, people have reported being very disturbed by strangers taking their photos; especially pictures of their children.

Gym locker rooms have long been off limits for cameras. Cell phones are often now banned there, too. While I was embedded with the Canadian Forces in Afghanistan, my cell phone was confiscated for fear that I would take a photograph, perhaps even a geo-tagged one, of something I should not. Casinos have always banned cameras, except for the fleet that they operate themselves. Their latest challenge is gamblers with tiny cameras hidden in their sleeves to watch the cards as they are dealt.
70
Google Glass is already forbidden at some sports arenas, movie theaters, and concert venues. Hospitals are implementing bans, as will many other workplaces. Even hotel lobbies may become off limits.

While working as a TV journalist, I was asked to leave a posh hotel while trying to shoot a “stand up” in the lobby. The manager explained that “we need to protect the privacy of our guests—you just might catch somebody here with his mistress or something like that.” Legally, they have the right to do that in most jurisdictions.

Internet Rule 34 (“If it exists, there's porn of it”) strikes here with vengeance. Soon after Google Glass became available, the “first Google Glass porn” appeared online at the adult app store mikandi.com. Their inaugural offering features a cameo appearance by Ron Jeremy, holder of the
Guinness Book of World Records
title for “Most Appearances in Adult Films.” Now a senior citizen, he keeps all his clothes on in this short piece of point-of-view pornography. A censored version is available on YouTube.
71

There seem to be no limits to human stupidity when it comes to posting inappropriate images online. People have distributed photos of themselves in an unbuttoned airline stewardess uniform (Ellen Simonetti, the “Queen of the Sky” blogger who flew for Delta airlines) and taking a bath in the sink at Burger King (Timothy Tackett).
72
Both were fired from their jobs, though they went on to other careers, propelled no doubt by the notoriety from their online misadventures.

Simonetti and Tackett are consenting adults, but things are quite different when children are involved. Often they seem either unaware of or unconcerned about the risks of releasing questionable comments and images into the Wild West of cyberspace. In a CNN report called “The Secret Life of My Sixth Grader,” a mother creeps herself out by spying on the texts and Instagram photos of her eleven-year-old son, who, she notes, “has never let on that he is remotely interested in girls.”
73
The content of his messages and photographs soon convinces her otherwise. “Maybe it's the digital photo-filters,” she muses, “but the girls seem sexy beyond their years.”

A 2008 survey by the National Campaign to Prevent Teen and Unplanned Pregnancy asked teens aged thirteen to nineteen and young people aged twenty to twenty-six if they had ever posted or sent nude or semi-nude photos of themselves. Overall, 20% of the teens and 33% of the young adults answered in the affirmative, with more females than males admitting to this activity in both age groups.
74

Skype video chat, and sites like Chatroulette, created by a seventeen-year-old Russian lad, have brought inappropriate images into the video domain. Turn on your web camera and microphone and meet new friends from around the world, chosen for you randomly. They may or may not be wearing clothes.

In 2011, researchers working closely with Chatroulette introduced a filtering technology to try to cut down on the surprise of sudden, unwanted nudity. According to one report, the “filter technology and moderation (by human censors) results in the banning of 50,000 inappropriate users daily.”
75

Bare skin-finding technology works both ways, of course, and there are now apps that will mine photos posted online specifically for nudity or some close approximation. As just one example, Badabing!, available in the iTunes stores, claims it will save you the effort of browsing “endlessly through a friend's albums looking for beach or pool pictures.”
76
The developers call it “the only image social recognition app,” but it certainly won't hold that title for long.

The wildly popular smartphones app Snapchat would appear to address the problem of persistent photos, since images sent through it disappear within a few seconds. However, this may well be a false sense of security. Armed with the right forensic tools, experts have been able to mine smartphones for supposedly deleted Snapchat photos. There are also apps like SnapSave and SnapChat Save Pics that are explicitly designed to defeat the ephemeral aspect of Snapchat photos and videos. These apps work around the Snapchat system, so, as one says, “the sender of the snaps doesn't get notified” that you are not playing by the rules.
77

Even without these apps, someone can always point a camera or another smartphones at a fleeting Snapchat image and capture it for posterity. Just assume that if you send a photo, it can be grabbed and have a life that extends far beyond momentary ogling.

Increasingly, your face is becoming a key that unlocks a vast amount of personal information about you. This was brought to the public's attention when some users of a leading U.S. matchmaking site learned that their online personas were not as private as they thought they were.

Dating site users almost always include a photo, but typically ­register using a pseudonym, like sexybaby235 or hungguy404. The would-be lovers only reveal their true identity when someone of interest comes along. Carnegie Mellon University professor Alessandro Acquisti grabbed almost six thousand dating site profiles from
Match.com
and compared them against publicly available Facebook profile photos in the same geographical area using a facial recognition program called PittPatt. His goal was to “de-anonymize” people.

He immediately encountered an interesting research problem: just because the computer says two faces match up does not mean it is true. The human brain is still the world's best facial recognition engine, so Acquisti enlisted human reviewers to rate the computer's work. Amazon runs a business called Mechanical Turk. Its name pays homage to the mysterious chess-playing robot constructed in Europe in the late 18th century that secretly contained a chess-playing human being. Amazon's system is a high tech “piecework environment” where people agree to do menial tasks, like looking up a company's main office address, for a few pennies each. If you do enough of these, fast enough, you can actually earn some decent cash.

The researchers asked the Mechanical Turk validators to sort the computer's proposed facial matches into categories like “sure match” and “highly likely.” Acquisti's conclusion was that “about one out of ten dating site's pseudonymous members is identifiable.”
78

He freely admits that “no human being can really take the time of having one browser open on Facebook and the other browser open on the dating site, and hope to find matches.” However, as this experiment demonstrates, face matching can be automated. This project involved comparing more than 500 million pairs of faces, which would take any of us a pretty long time. A computer can do that in a flash and come back asking for more.

To illustrate how our faces are becoming excellent personal identifiers, he performed another study, this time on a U.S. university campus: “Passers by were invited to participate in the experiment by sitting in front of a webcam for the time necessary to take three photos, and then by completing a short survey. While a participant was completing her survey, her photos were uploaded to a computing cluster and matched against a database of images from profiles on the social networking site. Thereafter, the participant was presented with the images that the facial recognizer had ranked as the most likely matches for her photograph. The participant was asked to complete the survey by indicating whether or not she recognized herself in each of the images. Using this method we re-identified a significant proportion of participants.”