Month April 2008

“what they want is an automatic feed”

Another sign of growing state interest in the semantic web… According to a recent article in the washington post, “the FBI has created a network of links between the nation’s largest telephone and Internet firms and about 40 FBI offices and Quantico” as part of their Digital Collection System (also called “The Digital Collection System Network” or “DCSNet”). DCSNet not only represents a blurring between state and corporate interests — Sprint facilitates communication between FBI wiretapping rooms on the government’s behalf and companies like VeriSign are often used to carry out the actual wiretapping — but it also sheds light on how state/corporate-run surveillance is being embedded in everyday life.

The DCSNet, which was made possible under the Clinton Administration with the Communications Assistance for Law Enforcement Act (CALEA), allows FBI agents to instantly wiretap just about any communication device associated with a particular person of interest — with “point-and-click ease.” While “content” surveillance must be authorized by a court, the washington post explains how the FBI’s “transactional” surveillance does not need court authorization:

Wiretaps to obtain the content of a phone call or an e-mail must be authorized by a court upon a showing of probable cause. But “transactional data” about a communication — from whom, to whom, how long it lasted — can be obtained by simply showing that it is relevant to an official probe, including through an administrative subpoena known as a national security letter (NSL). According to the Justice Department‘s inspector general, the number of NSLs issued by the FBI soared from 8,500 in 2000 to 47,000 in 2005. [emphasis added]

That’s a 453% increase in just 5 years, and the Bush Administration is currently petitioning the FCC to expand CALEA to make it easier for the FBI to access more types of data. The ultimate goal is to achieve real-time tracking of everyday life. In order to accomplish such a task, not only must our laws and telecommunications infrastructure be “updated” but the human actors associated with surveillance practices must be removed… which brings us back to the emerging semantic web, with its desire to find, organize and distribute information between machines without the need for human intervention. As a telecom industry lawyer, who was interviewed by the washington post, put it:

…government officials now “have to rely on a human being at a telecom calling up every 15 minutes to send law enforcement the data… What they want is an automatic feed, continuously. So you’re checking the weather on your mobile device or making a call,” and the device would transmit location data automatically. “It’s full tracking capability. It’s a scary proposition.” [emphasis added]

Scary indeed. They want an RSS feed of everyday life.

(young) person of interest

What would it look like if we were to situate young people in the growing semantic web? A 2007 report from the U.S. Government Accountability Office (GAO) took a look at some of the data mining programs currently underway at the Department of Homeland Security. In their report, GAO offer a “Typical Semantic Graph” which represents the “data relationships and linkages” of a particular “person of interest” which can now be generated through a process called “semantic graphing.” GAO’s report defines semantic graphing as “a data modeling technique that uses a combination of ‘nodes,’ representing specific entities, and connecting lines, representing the relationships among them.”

So what might a “Typical Semantic Graph” for a young person of interest look like? Part work, part play – here is GAO’s “typical semantic graph for a person of interest” compared to my “typical semantic graph for a young person of interest”:

(young) Person of Interest :: GAO & GTD