Another Day, Another Airport

Sitting on the floor under a gleaming array of unused payphones at the Miami Airport some days ago, I watched two federal air marshals check in for my flight. I wasn't hiding, just taking advantage of the only electrical outlets I could find. There were two of them tucked under there, and two of us, white middle-aged dudes splayed out on the carpet with our laptops plugged into the grid. Our cellphones handy on the rug beside rendered the row of callboxes obsolete, good for nothing more than bumping heads on. The guy next to me on the floor was talking to his girlfriend on his about his chances of getting on the flight as a standby. (Slim to none). At the gateside check-in counter, about twice as far away, the flight attendant was giving away the last bulkhead seat. I want to make it quite clear that I wasn't spying, or even listening in; I was sorting through aerial photographs from Belize and trying not to be distracted by the swirl of conversations going on around me.

There are two prevailing and generally conflicting approaches to a philosophy of computer security systems. One calls for transparency and the other for secrecy. The idea that security could be transparent seems counterintuitive, but proponents of this view argue that transparency drives innovation. It seems self-evident, on the other hand, that it is a good idea to draw the curtains closed so that thieves will not know if you are home or not.

This dichotomy, which is part and parcel of the open-source debate, is a favorite topic of a favorite website, boing-boing. They come firmly down on the side of transparency, arguing, for instance, that anyone in the world who wants it should have access to the source code for the new Windows operating system, to poke and prod and jab at it at will, searching for holes and flaws. The collective brain-power of the world is thereby harnessed and any weaknesses are exposed before they can be exploited by malicious virus-writers wreaking havoc or enslaving your computer as a 'bot. In the open source model democratically selected innovations are suggested and incorporated, improving everyone's experience.

A couple of years ago one of boing-boing's links pointed to some interesting articles by computer scientist Matt Blaze, who applied the logic of open source transparency to old-school, analog locks and safes, dismembering them and then explaining not only how they work, but how to crack them by exploiting the specific design weaknesses he observed. Beyond the inherent fascination of safe-cracking and its Raymond Chandleresque associations, what was most interesting about this was the outrage that his work inspired in the locksmithing fraternity, who apparently still view themselves as a sort of medieval guild of upright and universally virtuous keepers of ancient and mysterious secrets which they have sworn a blood oath never to reveal. To them it was scandalous that Blaze would reveal "trade secrets," and put out into a public forum information that would allow any idiot with a Makita and a good drill bit to bust into one of the more popular makes of safe.

The argument for transparency is made compelling through Blaze's work. If, without recourse to "proprietary information," Professor Blaze manages to demystify locks and explain how the layman can pick his way through them then surely these locks need to be improved upon. For lock and safe companies to assume that only honest locksmiths are privy to the workings of their tumblers is absurd, and an altogether insufficient guarantee of the security I expect when I install a lock on my front door. It is therefore beneficial for the community as a whole to be made aware of weaknesses in security systems, so that they may realistically asses the level of protection they enjoy and take any further measures to enhance their own security and that of others.

Let us now consider this line of inquiry as it applies to airport security. Had anyone seen fit to crow about how easy it was to board an aircraft with a box-cutter before September 11th, 2001, history might have been altered. Similarly one wonders if it occurred to those guarding the airports to announce that a part A part B mixing of liquid chemicals on board a jet might pose a threat. Were they caught by surprise when this was attempted last year? It is difficult to imagine that such threats had never been conceived of by professional security personnel, and it is all to easy to imagine someone raising them in a meeting only to be told to keep quiet, that remedies were too expensive, that the security system seems to be functioning just fine without your alarmism, thank you, so just keep that to yourself. As a result the confiscation of your nail-scissors and shampoo today have the feeling of closing the barn door somewhat after the departure of the cattle.

But not so the hypothetical presence of air marshals, who seem to me to be a legitimate security enhancement. The marshals, we have been told, ride on our flights incognito, ready to leap into action should a threat present itself. We have been told about them because this acts as a deterrent; hijackers and would-be terrorists are unlikely to want to try their luck, armed only with a paperclip or a dessert spoon, against professional law enforcement agents authorized to carry guns on board an aircraft. If, however, the terrorists are able to identify the air marshals ahead of time, the advantages the latter have are greatly lessened, so we are therefore not told who the air marshals are, nor whether there are certain to be any on a given flight. The importance of their remaining hidden is precisely why the dress-code requirement for air marshals was dropped last year.

I am not proposing that transparency in the context of airport security can be directly equated to the computer security model. But one of the best arguments for transparency in computing is the implied invitation to any and all to propose and devise improvements to the system. How different from this is the atmosphere around the long lines waiting to clear security at an airport in the post 9/11 era. In Miami signs inform that "all jokes will be taken seriously" and that it is a "federal offense" essentially to do so much as talk back to a luggage screener. The hundreds of thousands of innocents who pad along in their socks through the endless metal detectors think, I imagine, much what I am thinking as I wait, holding my trousers up with one hand: just give me the patience to get through here without pissing anybody off or drawing any attention to myself. What would the reaction be if, instead of shuffling through the metal detector like a prisoner, one paused and began to examine it, to rub ones hands over its surfaces and peer at its electronics? Or walk around behind the baggage screener to have a good hard look at what the bags reveal when X-rayed? And a parallel level of transparency in the airport environment may not be desirable; we certainly don't want random citizens attempting to bring creative varieties of weapons through the checkpoint only to claim after being caught that they were only testing the system. Nonetheless we must have a mechanism by which the public might propose improvements based on what they have seen in their long bored wait in line.

Why such a lengthy, windy preamble? Why can't I get to the point already? Because I am scared to propose my modest innovation in airport security in today's climate, lest I be perceived as a criminal conspiring to undermine your in-flight safety by highlighting flaws in the system. But trust me on this, people; I only want things to be better and safer, for you and for me, even if I am annoyed by the uniformed rent-a-cops constantly bellowing at me to remove my shoes and place my laptop in a tray.

While looking up briefly from my seat on the carpet in the Miami airport, I saw two clean-cut, thirtyish men approach the check-in desk. One, caucasian, hair cut in the high and tight military style; the other african-american, with mirrored shades giving him the ominous vibe of Wesley Snipes in a vicious role. Almost simultaneously the duo pulled matching small black zippered wallets from the pockets of their smart-casual windbreakers, and thrust them over the counter to the attendant. Half the size of passports, the wallets were of such a size and proportion that they could contain only one form of identification: a badge. Wesley turned his mirrored lenses towards me and I wondered if he wondered what I was wondering. The attendant looked at her computer screen. There came a quick flurry of keystrokes.

"You're just catching a ride up, or...?" asked the attendant. We were flying to New York.

"No, no, we're working the flight," said the guy with the buzz-cut.

They backed away from the desk. No paper of any kind seemed to have been exchanged. I was trying not to stare, so I might have missed something. But if those guys weren't air marshals, then just slap the cuffs on me right now and haul me away to the loony bin. When, in living memory, has a passenger approached the gateside desk without presenting a boarding pass? To ask where the men's room is, or whether lunch will be served? The computer is never consulted for mundane information like this. Without tickets in their hands my two clean-cut undercover dudes might as well have been waving guns and wearing matching t-shirts emblazoned FEDERAL AIR MARSHAL.

I have, of course, no idea how the system works. Perhaps the marshals just circulate amongst the gates looking for suspicious characters before deciding at the last moment which flight to board, and therefore have no tickets. I don't want to blow the budget, but why not issue them reusable paper ticket jackets and generic boarding passes to present to the gate staff, or even air marshal identification cards with the look, dimensions and heft of a US passport? This simple ruse would have fooled me.