"This is exactly wrong. Profiling does not increase safety whatsoever. Bruce Schneier argues it much better than I would:"
You may say it doesn't 'increase safety', but Israel is a great example of it working. The attacks, hijacks, and deaths have nose dived to nearly 0. They have a hybrid system that involves both random checking and profiling.
I'm not saying we should lock up anyone that fits a specific racial profile. The problem is that when we use data and risk assessment to question specific people that may be suspicious, it's seen as racist if the person is anything but white.
The two articles you linked mention hijackers that happened pre-911. I don't think people are hijacking planes and diverting them to central america nor do we have IRA militants bombing planes.
We need a hybrid system between random checking and profiling. Random checks work so far, but will not help with long lines because everyone needs to be checked equally.
To reduce long lines, we either need to scale-back security (which is a bad idea) or use information and data to make the process more efficient (IE: profiling).
Some profiling is a good thing, but we can't completely rely on it.
You may say it doesn't 'increase safety', but Israel is a great example of it working. The attacks, hijacks, and deaths have nose dived to nearly 0. They have a hybrid system that involves both random checking and profiling.
I'm not saying we should lock up anyone that fits a specific racial profile. The problem is that when we use data and risk assessment to question specific people that may be suspicious, it's seen as racist if the person is anything but white.
The two articles you linked mention hijackers that happened pre-911. I don't think people are hijacking planes and diverting them to central america nor do we have IRA militants bombing planes.
We need a hybrid system between random checking and profiling. Random checks work so far, but will not help with long lines because everyone needs to be checked equally.
To reduce long lines, we either need to scale-back security (which is a bad idea) or use information and data to make the process more efficient (IE: profiling).
Some profiling is a good thing, but we can't completely rely on it.