[EAS]Thoughts/URLs re. attack
pjk at design.eng.yale.edu
Wed Sep 12 14:25:23 EDT 2001
Mail*Link® SMTP Thoughts/URLs re. attack
Dear Colleagues -
I'm forwarding some thoughts from Phil Agre, several very pertinent to
engineering design and ethics. Also more URLs.
And I was horrified to learn, as you probably have, that airport
security people get minimum wage and have typical annual turnovers in
excess of 100%.
Date: 9/12/01 1:10 PM
From: Phil Agre
I've enclosed some more URL's related to the attacks on the east coast.
Although I want the attackers caught as much as anyone, I'm concerned by
some of the language I'm hearing, including some nearly fascist rhetoric
about America being "soft". I was happy to hear George Bush emphasize
that civil liberties will be protected. If you understand the attack as
an assault on freedom, then it hardly makes sense to diminish freedom as
We do need to improve security, but we should not understand the need
for heightened security in a broad, vague way as a cultural imperative.
We do not need a police state, and we should not militarize our society.
Rather, we should view security as a design problem. We have an opening
now, a brief window when the airlines cannot undermine improved security
in their own commercial interests. Maybe we can also force Microsoft
to design its products in a secure way, rather than exposing us to the
severe information security problems we've seen in the last few months
with its fundamentally shoddy architectures. We should take advantage
of this opening to redesign our aircraft, buildings, software, and
institutions in a rational way. Consider some examples:
* Look at the doors between aircraft cabins and the cockpit. Anyone
could knock down those doors. Of course, just fixing the doors isn't
enough, but it's an example of the concrete design problems that we can
address. We have a chance to completely rethink the interior spaces
of aircraft, which could benefit dramatically from the attention of
an industrial design firm.
* We also have a chance to implement long-delayed proposals for things
like fuel tank safety. How well do we understand the entire life cycle
of jet fuel, surely one of the most dangerous substances in existence?
* Next-generation digital aircraft electronics should be rethought more
deeply for their contributions to security, as well as their security
vulnerabilities, before their architecture is set in stone. Right now
the controllers on the ground have far too tenuous an idea of where the
planes are, especially in emergency conditions. It's absurd that an
attacker can simply turn the tracking devices off.
* Many airports predate modern security procedures, with the result
that the security arrangements are crammed into spaces where they
don't belong. The physical design of the conveyor belts on the luggage
scanners is terrible, and the signs are useless. And have you actually
looked at the video display from the X-ray unit? The whole system can
be redesigned to be more meaningful, more reliable, and less frustrating
-- another job for real industrial designers.
* How did the incentives get set up to pay the airport security people
minimum wage? Who's allowing the airlines to use security procedures to
play out their conflicting agendas about baggage size? The institution
of airport security needs to be redesigned. One approach would be to
federalize it; those who don't like to federalize things are invited to
come up with designs of their own.
* Another area that needs to be redesigned is the identification
system for airport, airline, security, and law enforcement personnel
in airports. As it is, anyone can wave any badge-like object at
anyone else and go wherever they want. Identification systems that
would be unacceptably invasive for the general public are reasonable
for employees in security-sensitive environments. Identification
systems in general are a slow-motion catastrophe, and simplistic
proposals like a national ID are a poor substitute for fine-grained
attention to the details of how identities get administered in
practice. Identification also has an information-design angle that
is usually neglected, given the small, cryptic, hard-to-read markings
on most identity documents.
In short, we need an analytical approach and a design approach.
Vague abstractions are counterproductive. It is useless to ask
"how much of our civil liberties do we need to give up?" or "is our
intelligence capacity too constrained?" or "we need more security,
but how much is enough?". We should look at problems concretely, in
specifics. Seeming tensions between privacy and technology routinely
disappear once problems are considered concretely and in detail.
So the question is not "can biometrics solve the problem?", since
biometrics, as such, in general, can't solve anything. It is entirely
conceivable that specific biometric technologies can play a specific
role in a systemic redesign of the security systems at airports and
elsewhere, including online. Indiscriminate use of biometrics to
identify everyone and everything is useless, and it's also dangerous
if it's simply pasted on top of dysfunctional institutions, or if it
substitutes for concrete, analytical thinking.
Wall Street Journal coverage
(appears to be available without a subscription)
Yahoo links to news stories etc
online mechanisms for donating to the Red Cross
mailing list to connect people who can volunteer or provide resources
aircraft flight tracks
front pages of 50 newspapers' coverage of the attack
Current Awareness via Streaming Audio/Video
Speech/Transcripts/Statements from US and Foreign Leaders
Anonymous Remailer Operators Start to Take Remailers Offline
Middle East Newswire
Two Planes Hit Twin Towers at Exactly the Worst Spot
Security Experts Knew a Major Attack Was Possible
Insurance Cost for Terrorist Attack to Near $1 Billion
Reports: Boston Investigators Find Evidence in Attacks
civil engineering aspects of the building collapse
online discussion site for pilots
Rescuers Struggle at Pentagon
Why the Killers Threaten World Prosperity
In Shock, Teachers Downplay Tragedy
EBay Cancels Auctions of Attack-Related Items
(some idiots were actually gathering rubble in order to sell it on eBay)
More information about the EAS-INFO