"Snowden’s biggest single victim," a senior intelligence official told The New York Times, isn't his revelation of the National Security Agency's existing surveillance system. Instead, it's the NSA's push to broaden that ability to cover all of the scannable traffic on the internet. The public relations damage done to the agency has made an expansion of its ability to read data a non-starter. For now, anyway.
In a document released to coincide with the president's press conference last Friday, the NSA described the breadth of its existing data-scanning ability. As has been documented extensively, the agency monitors traffic on fiber-optic cables entering the United States (and elsewhere), monitoring it in near-real time for indicators of terror activity. But, it insists, it doesn't actually monitor that much.
So by its own assessment, the NSA can access ("touch") about 29.2 petabytes of traffic a day, reviewing about seven terabytes of that data. As author and professor Jeff Jarvis notes, this probably isn't just the NSA stumbling around, reviewing content. "Keep in mind that most of the data passing on the net is not email or web pages," he writes. "It's media." That media — videos and photos — isn't reviewed by the NSA. Messages and web traffic, text content, is. And a lot more text (and metadata) fits into that seven terabytes than photos. Seven terabytes is over a third of the contents of the Library of Congress, according to WolframAlpha.