Until the Snowden episode, sending agents to track down leads that were statistically worthless was worse than just a waste of resources, because the problem US they're actually trying to solve is "how do we monitor suspects whilst managing not to give away how much we know to journo-bloggers so passionate about their right to freedom of speech they might publicly question why the government is scrutinizing their search terms".[1]
The intersection of "pressure cooker", "backpack" and visits to overseas countries not known for sponsoring low-level terrorism is perhaps not the most obvious criterion for identifying terrorists, so presumably they'd also have to flag most of the ~12,000 people who searched for "bomb-making" last month, for example, and perhaps some more innocuous ones like independent searches for "martyr" and "white house" or "chechnya" and "aeroplane". That's a lot of resource.
Now even if they're pretty selective about who they actually pay a visit to, they're going to return a lot of false positives. No big deal, except that pre-Snowden the programme is supposed to be a secret and even within the subset of people flagging the "possible terrorist" alert they're going to investigate far more journalists (common occupation) than terrorists (rare occupation). Which is a bit of a problem if their modus operandi is instructing police to knock on people's doors and ask them whether they were searching for X on the internet. Was this really PRISM's raison d'etre?
The alternate hypothesis involves a freelance writer famous solely for her number of Twitter followers embellishing her police enquiry story with some very topical themes.
[1]If I worked in intelligence I'd solve that problem by not telling local cops what potential suspects had been doing on the internet...
Comments
Until the Snowden episode, sending agents to track down leads that were statistically worthless was worse than just a waste of resources, because the problem US they're actually trying to solve is "how do we monitor suspects whilst managing not to give away how much we know to journo-bloggers so passionate about their right to freedom of speech they might publicly question why the government is scrutinizing their search terms".[1]
The intersection of "pressure cooker", "backpack" and visits to overseas countries not known for sponsoring low-level terrorism is perhaps not the most obvious criterion for identifying terrorists, so presumably they'd also have to flag most of the ~12,000 people who searched for "bomb-making" last month, for example, and perhaps some more innocuous ones like independent searches for "martyr" and "white house" or "chechnya" and "aeroplane". That's a lot of resource.
Now even if they're pretty selective about who they actually pay a visit to, they're going to return a lot of false positives. No big deal, except that pre-Snowden the programme is supposed to be a secret and even within the subset of people flagging the "possible terrorist" alert they're going to investigate far more journalists (common occupation) than terrorists (rare occupation). Which is a bit of a problem if their modus operandi is instructing police to knock on people's doors and ask them whether they were searching for X on the internet. Was this really PRISM's raison d'etre?
The alternate hypothesis involves a freelance writer famous solely for her number of Twitter followers embellishing her police enquiry story with some very topical themes.
[1]If I worked in intelligence I'd solve that problem by not telling local cops what potential suspects had been doing on the internet...