Omniveillance After Boston

April 20th, 2013

In 2007, Google first launched Street View. This new technology stunned me. I was familiar with, and leery of  the national surveillance state, and the proliferation of cameras in places like London’s Ring of Steel. But this was different.  It was run by a private party. My article chose not to “focus on government surveillance or Fourth Amendment issues”. Never before had a technology, in private hands, enabled such a wide-ranging form of omnipresent surveillance. The existing privacy laws, embodied mostly in Prosser’s 4 privacy torts in the Restatement (Second) of Torts, were totally inadequate to protect any form of privacy in public. My first published article, Omniveillance, responded to my concern by offering new torts for the digital age. This article has had surprising longevity, as recent debates in Hawaii over a paparazzi tort implicated my research.

But one aspect of my work that remained largely fanciful was what omniveillance actually could become.

At the time, forms of public surveillance were somewhat tame:

More recently, average people attempt to become paparazzi with their digital cameras and cellular phone cameras, by posting photographs and videos of public occurrences onto the Internet with services like Flikr (a popular photo sharing web site), YouTube (a popular video sharing web site), and even user-generated news web sites.121 New York City now allows citizens to upload amateur videos or photographs of potential crimes to the police department’s web site.122 A new service known as geotagging, featured on the Apple iPhone 3G, automatically records the latitude and longitude coordinates wherever a photograph is taken.123 This mild form of omniveillance allows a user to quickly photograph a specific event and upload the recording onto the Internet so that the location where the photograph was taken becomes linked to that picture.

More troubling, however, I envisioned a future where Google installs cameras on “street corners throughout America.”

In the future, individuals could be able to visit Google’s web site and search for a specific time, date, and location. Users will thus be able to witness what happened at any place and to what people, at any moment in Google’s recorded history. Such pervasive human monitoring is the essence of the phenomenon this article has termed omniveillance. Omniveillance is a form of omnipresent and omniscient digital surveillance in public places that is broadcasted indiscriminately throughout the Internet.

This omniviellance would not be limited to cameras installed by Google, but instead would be bolstered by citizen armies equipped with mobile phones. In the future, everything will be recorded.

As distinguished from previous forms of public monitoring, this new form of surveillance will be omnipresent, as it can record vast areas of space over a very small period of time. It provides the users of this system with omniscience to know everything happening in a specific location at a specific time. Furthermore, this information will be indefinitely retained, and easily accessible. When future versions of this technology is properly implemented, it will be possible to enter a time, date, and location, and witness what happened at that moment as if you were there. It is a virtual time machine. In addition, using facial recognition technology, it will even be possible to search for a particular person’s location at any given recorded moment. Although Street View is currently only limited to urban metropolises, companies such as Everyscape seek to expand digital surveillance to rural locations such as ski-resorts, beaches, and other out-of-the-way places.124 Another friendly neighborhood surveillance provider called Earthmine is developing technology that can record all stores, restaurants, and other locations in public.125 No place would be safe from the unblinking eye of omniveillance.

What I didn’t appreciate in 2007 was how pervasive this technology would become. I distinctly remember thinking, would it be possible for someone to walk around with an iPhone and constantly film or record everything. No, I thought, that was silly. Why would anyone do that?

Well, I didn’t imagine Google Glass, the embodiment of the omniveillance ethos.


Though we aren’t quite yet to the point where everybody is recording everything through Google Glass, the response of the internet to the Boston Marathon Bombing, and attempts to use public and private recordings to identify the suspects, has been quite eye-opening for me. People at  Reddit and 4Chan produced herculean efforts to pick out the suspects from all freely-available media, recorded by private individuals. It was stunning. Even after the FBI released grainy images of the two suspects, one guy had a clear shot of the suspect, snapped with his iPhone. The state no longer has a monopoly on surveillance. No matter how many cameras the government has, there will always be more iPhones and Google Glasses.

But what does this mean for the future of surveillance in public.

Despite (what I am sure is) Boston’s elaborate surveillance network, and enhanced facial recognition systems, the government was unable to identify the two suspects on their own (see New York’s Big Brother nerve center). These people were walking through one of the most heavily populated areas in Boston, on one of the busiest days of the year, and the surveillance network failed the state. I will leave aside for the moment the question of whether these surveillance networks are worth the cost–both in terms of money, privacy, and liberty–as they were unable to solve the crime. But what is amazing here is how regular people, equipped with iPhones, who took pictures of the disaster, or Redditors pouring over hundreds of photos on facebook and twitter, were integral in helping to solve the crime! Not the government’s surveillance network. A regular–to quote Glenn Reynolds–“Army of Davids.”

This is both awe-inspiring and creepy at once. It is amazing that regular people can pool their wisdom together and help solve one of the most gruesome crimes of our generation. But, and this is a huge but, now the government has complete access to a free surveillance ring, not bound by the strictures of the Fourth Amendment or any other limit. In Omniveillance, I stressed  how powerful this privately-owned surveillance network would be.

Additionally, with a subpoena, the government has ready access to a free surveillance network, further imperiling our civil liberties

No warrants necessary.

If this same disaster had happened ten years ago, without the Redditors and Instagrammers and YouTubers, the suspects may have gotten away. What will technology be like in ten years–will any crime go undetected? Who needs the Precogs from Minority Report if we have Postcogs on Reddit? And if crimes no longer go undetected, will there be any other activity that escapes the watchful eye of omniveillance?

UpdateWired writes a very similar piece about how private surveillance enabled the government to solve this crime:

When the smoke literally cleared on Monday, investigators had a huge problem and nearly no leads. No individual or organization claimed responsibility for the bombings that killed three and wounded more than 180. So they took a big leap: They copped to how little they knew, and embraced the wisdom of The Crowd.

Hiding in plain sight was an ocean of data, from torrents of photography to cell-tower information to locals’ memories, waiting to be exploited. Police, FBI, and the other investigators opted to let spectator surveillance supplement and augment their own. When they called for that imagery, locals flooded it in. They spoke to the public frequently, both in person and especially on Twitter. All that represented a modern twist on the age-old law enforcement maxim that the public’s eyes and ears are crucial investigative assets, as the Internet rapidly compressed the time it took for tips to arrive and get analyzed.

But the FBI and police have been reluctant to embrace what the hive mind can provide: it implies the authorities don’t always have the answers. Veteran law enforcement officers remember cases from the ’90s when the bureau clammed up to the public and local cops, at the expense of receiving greater public cooperation.

“The great advantage here is the number of cameras out there,” Rolince says. “Without the cameras, I don’t know where we are.” The cameras were everywhere. It wasn’t just the surveillance cameras looming on the tops of buildings at Copley Square. Bostonians and out-of-towners who came to the Marathon, one of the most celebrated civic events in the city, pulled their phones out throughout the race to feed their Instagram addictions and keep their Flickr pages current. It would become a reminder that the public enthusiasm for documenting their lives can outpace even the vast surveillance apparatus of the government.

If so, the call for public assistance helped get it over the hump. Within two days, DesLauriers received what he would describe as “thousands” of photos and videos, showing different vantage points of the Copley Square spectators. Once investigators arranged them by the time they were taken, they could piece together a mosaic of the scene, allowing them to check behavior they considered suspicious — and apply imaging tools to focus the accumulation of data.

Numerous law-enforcement sources were reluctant to get specific about those tools. (“We wouldn’t discuss this specific detail at this stage,” FBI spokesman Paul Bresson demurred.) But a former FBI technology official who didn’t want to be quoted by name cautioned that there isn’t a tech-intensive secret sauce behind the manhunt. Facial-recognition tools may be growing more advanced, but their limits are still on display: “If you don’t have a high-quality photo, you can’t use that,” the ex-official says. But there’s “all different types of [relevant] data — pixels, the cell tower data, any records that might exist.”

And there are messy precedents emerging from mass photo and video data emerging from an era of ubiquitous cellphone cameras augmenting police surveillance: the ex-FBI technology official says that “legal and ethical questions” cause investigators to hesitate before launching big data-mining projects, even with all the broad leeway they have to violate citizens’ privacy. That hesitation doesn’t have to apply when citizens volunteer the data.

“Nothing has really changed,” Bar-Tur says, “just the medium has changed.” That might be enough for a new model manhunt to emerge.

This confirms much of what I wrote above.

Update: The Times writes about how releasing these photographs helped close the case–and at a huge cost:

The authorities knew that broadly distributing the images — some captured by ubiquitous surveillance cameras and cellphone snapshots and winnowed down using sophisticated facial-recognition software — would accelerate the digital dragnet, but they did not realize the level of chaos it would create.

Intelligence and law enforcement officials said the authorities in Boston weighed the risks of some mayhem against their growing fear that time was slipping away, and that heavily armed and increasingly dangerous men, and possibly accomplices, could wage new attacks in the Boston area or beyond.

“Their posting of that video and photographs speaks to the fact they likely felt they had run into a dead end,” said Mitchell D. Silber, a former senior intelligence official in the New York Police Department, now with the investigative firm K2 Intelligence.

Federal authorities involved in the investigation had briefed administration and Congressional officials on their hopes to arrest the suspects early Thursday without revealing their hand.