From: Lee Corbin (lcorbin@tsoft.com)
Date: Wed Jul 24 2002 - 23:05:19 MDT
First, I should make it clear that I am not a security professional, and
although that doesn't necessarily make my points invalid, it does show
that my points need not be taken entirely seriously. Of course, people's
remarks since the dawn of time have always been understood and taken
in the context of just who is speaking, but it's interesting that on the
Extropians list, deference to authority often rises to an explicit level.
Hal writes
> We have occasionally discussed the use of profiling as a security
> measure in airport security, attempting to identify those people
> who would be more likely to be terrorists. Harvey Newstrom writes
> > As a security professional, I really must insist that standards require
> > search of everyone or random searches. You cannot let guards try to
> > detect the possible "guilt" of people by looking at them... Security
> > profiling must be based on individuals, meaning behavior... or situation.
> > Groupism that includes or excludes whole genders or races will instantly
> > fail because the bad guys then have a magic profile that will let them
> > through.
> http://swissnet.ai.mit.edu/6805/student-papers/spring02-papers/caps.htm.
> The article... certainly calls into question
> the methods currently being used by the airlines.
Yes, any person with reasonably good judgment who regularly flies
will tell you their opinion of the "idiotic" current procedures.
Any truly determined and resourceful terrorist can get a bag filled
with explosives through checking, and meet his maker along with the
other passengers. Likewise, tests have shown that experienced people
also can get weapons aboard. The "zero tolerance" being pursued by
the present American administration makes no sense to me.
I happen to find the most instructive part of the above article to be:
"El Al... receives daily threats, [yet] has not experienced a
terrorist incident in over thirty years. Its success is mainly
due to its tight on-site security. Each passenger each time he
or she flies is psychologically evaluated. Carryon bags are
checked multiple times. In essence, El Al's security system
works from the assumption that every passenger is a threat,
and treats him or her accordingly."
By asking good questions the well-trained and highly
intelligent Israeli security personnel succeed where
ours fail. The article is, to me, rightly explaining
why we cannot automate security in the way that the CAPS
system attempts. As Harvey and Hal point out, it makes
it rather easy for an intelligent terrorist to evade the
precautions.
The key word in the above paragraph is "intelligent",
which I used twice. There just isn't any substitute
for intelligence.
It's also interesting to wonder where "psychological evaluation"
ends and "profiling" begins. The article mentions the case
of the "80-year-old grandmother from Texas" which the CAPS
system algorithmically profiles as a non-threat. My guess is that
the intelligent El Al security agents doing the psychological
examination do not fail to note the age and even the ethnic
composition of the subject, and that this undoubtedly plays
a part in their assessment, although again, all I can do is
guess since I'm not a security professional.
"While El Al does keep a database of individuals' nationalities, genders, criminal
records and flight histories, this tracking should not be confused with the a
CAPS-like profile. Unlike CAPS, El Al's system only makes a determination of
the risk of a passenger after a security agent has questioned him or her.
CAPS, on the other hand, is a prior system focused on predicting who will become a
terrorist. El Al, through its system of psychological analysis and advanced
baggage screening, has found success in determining who is a terrorist."
Although this is obviously good advice, the U.S.'s situation is not
that similar to Israel's in several ways, and the U.S.'s needs for
security, are, in my opinion, not as severe.
I think one of the reasons our society has become too risk aversive
is "innumeracy", as John Allen Paulos describes it. On any one day
approximately 40,000 airplane flights take place in the U.S., and
so more than a million per month. What should be our desired rate of
terrorist take-over of airplanes? The U.S. should probably be aiming
for the rate of about one-in-a-million flights or so, which implies
that the current security precautions are excessive. Widespread innumeracy
guarantees that most people cannot intuitively understand the magnitude
of the inconveniences inflicted upon tens of millions of travelers, all
to prevent one or two hijackings per month. The number "tens of
millions" simply doesn't mean anything on an intuitive level to
non-professionals.
As a professional mathematician, I am capable of assessing the
statistical implications of large numbers in our daily lives,
and I should appreciate others deferring to my judgment about
such matters.
Lee Corbin
P.S. I must point out that the ironical tone of some of my remarks
above may not be apparent to all, and for that I express my regrets.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:39 MST