Slashdot Mirror
Fewer Than 1% Arrested From TSA's "Behavior Detection"
An anonymous reader writes "Fewer than 1% of airline passengers singled out at airports using the much vaunted 'suspicious behavior detection' techniques are arrested, Transportation Security Administration figures show. The TSA program, launched in early 2006, looks for terrorists using a controversial surveillance method based on behavior detection and has led to more than 160,000 people in airports receiving scrutiny, such as a pat-down search or a brief interview. It has resulted in only 1,266 arrests, often on charges of carrying drugs or fake IDs, the TSA said. The TSA has not publicly said whether it has caught a terrorist through the program." In related news, the odds of sanity coming to the TSA plummeted today when Schneier said he's not interested in the top job there.
The fact that less than 1% of the people caught were doing something illegal would make sense if we can assume that the vast majority of the people flying are not criminals.
Let say that the detector was accurate 90% of the time, and 5% of the people who passed through the airport were doing something illegal. If one million people came through that airport, we could assume that:
1,000,000 people
50,000 criminals
- 45,000 detected
- 5,000 not detected
950,000 innocent people
-855,000 not flagged
- 95,000 falsely accused
140,000 people accused
- 67.8% are innocent
- 32.1% are guilty
Granted this is just a hypothetical situation, not based on actual statistics, but the example shows how that even a reasonably accurate system can look unreliable when searching for a needle in a haystack.
Of course issues of fairness and privacy are something else entirely is another issue entirely.
Oh, how wrong you are...
http://www.softpedia.com/progDownload/ToasterClone-Download-54367.html
-- i am jack's amusing sig file
Simplistically, this psalm expresses grief and revenge by those who had been captured by the Babylonians.
(And "blessed" means "happy", not "God condones this and will bless you")
Looking into it more, though, I learned of a larger historical context (Taken from here):
"It is important to remember that the curses of Psalm 137 are not originally the psalmistâ(TM)s curses. They are the Lordâ(TM)s curses which the psalmist has made his own. The destruction of Edom was the fulfillment of prophecy, particularly the prophecy of Obadiah. In Isaiah 13:16, which was written about 200 years before Babylonâ(TM)s fall, the destruction of Babylon was prophesied in almost the exact terms used in Psalm 137. The destruction of the children who were too young to be transported into slavery was a common practice in ancient warfare. Since this cruelty was apparently practiced by the Babylonians during their campaigns of conquest against Israel, Babylon would receive from its Persian and Median conquerors the same treatment which it had inflicted on Israel (Jeremiah 50:29; 51:56). "
Right, the true numbers should be:
160,000 people searched under the new anti-terrorist behaviour screening, 0 terrorists found. 1,200 arrests made for completely non-terrorist activity.
This doesn't indicate a ~1% success rate, it indicates a 100% failure rate; no terrorists were found.
Perhaps there are no terrorists to find, perhaps there are; but in either case, this method has found to be a complete failure over a sample size of 160,000 individuals.
Police are not here to protect you, nor do they have to. See Warren v DC, or a whole slue (sp?) of cases from the supreme court. The police are there to enforce laws after the fact, and that is their ONLY duty.
> slue(sp?)
slew