2 ms·

PRISM Stopped Najibullah Zazi From Blowing Up Backpacks in the Subway

6 points by daegloe 13y ago

mindcrime 13y ago
OK, check out this bit: "And so when that bell rings, they say, 'Hey, they hardly ever use this account, but it's associated with Rashid Rauf, who is al Qaeda's master bomb maker, behind the plot to blow up all the airplanes, 'Who's he talking to?' And when they find out the other IP address on the other end resolves to Aurora, Colo., outside Denver, it connects them to Zazi, it takes them to the plot to blow up the New York subways, it's all prevented. That's how a program like this is supposed to work." That, to me, weakens the government's argument that they need to collect all this data indiscriminately using PRISM. They had an email address that was associated with a known Al-Qaeda bombmaker... ok, great, go get a warrant and get permission to pull all emails sent to or from that address. That would have caught this guy just the same! Which takes us back to what I said before, which is that analyzing activity between random people isn't going to do anything to help catch a terrorist. Either you know somebody is "on the radar" or you don't. If two (or more) people are plotting with each other to explode something, you can analyze the metadata about their calling and emailing habits until the cows come home, and it's not going to tell you that they're planning to plant a bomb if none of them are current suspects. Now if they reach out to a known bombmaker or militant group or something, then OK, that would tell you something. But, in that case, you just go get a warrant. The one argument I could almost see for this PRISM thing to make a little bit of sense, is this: If there are too many "suspected terrorist" phone numbers and email addresses for the Verizon's and AT&T's and GMails of the world to monitor constantly, in real time, and sending an alert to the NSA if there is activity. So by getting a raw dump of everything, the NSA can then run their analysis using their super-duper fancy schmancy quantum computer, or whatever they have (probably just a huge Hadoop cluster, but who knows) to mine for the "hits". But even if that were the case, I'd still consider this PRISM thing to be way over-reaching and improper.

6 points by daegloe 13y ago

OK, check out this bit: "And so when that bell rings, they say, 'Hey, they hardly ever use this account, but it's associated with Rashid Rauf, who is al Qaeda's master bomb maker, behind the plot to blow up all the airplanes, 'Who's he talking to?' And when they find out the other IP address on the other end resolves to Aurora, Colo., outside Denver, it connects them to Zazi, it takes them to the plot to blow up the New York subways, it's all prevented. That's how a program like this is supposed to work." That, to me, weakens the government's argument that they need to collect all this data indiscriminately using PRISM. They had an email address that was associated with a known Al-Qaeda bombmaker... ok, great, go get a warrant and get permission to pull all emails sent to or from that address. That would have caught this guy just the same! Which takes us back to what I said before, which is that analyzing activity between random people isn't going to do anything to help catch a terrorist. Either you know somebody is "on the radar" or you don't. If two (or more) people are plotting with each other to explode something, you can analyze the metadata about their calling and emailing habits until the cows come home, and it's not going to tell you that they're planning to plant a bomb if none of them are current suspects. Now if they reach out to a known bombmaker or militant group or something, then OK, that would tell you something. But, in that case, you just go get a warrant. The one argument I could almost see for this PRISM thing to make a little bit of sense, is this: If there are too many "suspected terrorist" phone numbers and email addresses for the Verizon's and AT&T's and GMails of the world to monitor constantly, in real time, and sending an alert to the NSA if there is activity. So by getting a raw dump of everything, the NSA can then run their analysis using their super-duper fancy schmancy quantum computer, or whatever they have (probably just a huge Hadoop cluster, but who knows) to mine for the "hits". But even if that were the case, I'd still consider this PRISM thing to be way over-reaching and improper.