I think this article isn't talking about random vs. profiled (which is what the mit paper is about), but random vs. 100% (where there 100% has been analyzed carefully to be sufficient to detect large bombs, etc.).
He was hand-waving at the end about the 100% but that's not a solution. They are doing it randomly because 100% is simply not possible. My point was that in a situation where 100% is not possible, as in real life, any system is potentially worse than randomness.
Maybe this just means we need to come up with screening mechanisms that can be applied to 100%, but given the current capabilities, choosing random screening is better than profiling.
But if you read the article, I believe the point was that if the terrorist gets caught under a random system, the terrorist still achieves a positive result for the terrorists (the govt becomes forced to shut down aviation and then apply the maximum screening to everyone, causing expensive chaos and terror of its own).
Getting caught this time was a huge win. By all accounts, the guy who got caught was a nobody. Had he even been to the camps for training? For the cost of a pair of explosive underpants and the life of one shmuck, AQ is once again top-of-mind in the west --- not to mention the tens of millions of dollars of disruption the stunt caused.
There is a practically limitless supply of shmucks out there for AQ to weaponize. All they have to do is get better at converting them. What evidence do we have that this will be a long-term operational problem for AQ? Everything I see indicates that they will get better at it, not worse.
This is also why they aren't shutting down traffic lights. A failed attempt to shut down traffic lights wins nothing. Nobody is viscerally afraid of darkened traffic lights. In fact, until it happens, nobody is going to be viscerally afraid of someone taking out the grid. But everyone is afraid in their gut of exploding planes. Just the threat --- just 5% of the threat --- is enough to wreak havoc.
