Crowdsourcing a real-time solution to air terrorism | Hacker News Reader

Crowdsourcing a real-time solution to air terrorism

markmaunder.com

21 pointsmmaunder29 comments

"Give passengers a way to quickly and discreetly report suspicious activity to officials, both before boarding and during the flight. A button that simply reports suspicious activity within a 3 passenger radius may be all that’s needed."

Yeah, then every time anyone with a thick beard boards a 747 they'll get 200 reports. That will solve the problem.

Also passengers who don't speak English is not cause for suspicion on a flight from Nigeria.

Never mind that the passenger in question attended University College London: He speaks English.

And if acting nervous on a plane is reason enough to move a person to the back of the plane to conduct a full-body search... well, let's just say the airlines had better have a significant rubber-glove budget.

No, they just reuse the rubber gloves. The security people at LHR doing the body pat-downs are a great example. I watched one sneeze all over his gloves, and then immediately pat someone down. Mmmm, body fluids...

So thereby it becomes clear who the gloves are protecting. Not the passengers.

It's funny because they are supposedly looking for (sharp?) syringes, but are only protecting themselves with a 1mm thick piece of latex.

...On further consideration, I can't think of any prominent airline terrorists of recent years who have not spoken English. The shoe bomber and the liquid bomb plot were British, and the 9/11 guys trained at flight schools in the U.S. midwest.

So, if history is any measure, a total inability to speak English is actually a great indicator that someone is not a terrorist. If the author's proposed system worked exactly as his example goes, it would not only have failed to catch any past terrorists, but it would add huge amounts of false positives due to the large number of similar-thinking people finding a lack of English suspicious.

Do you guys even know what the spoken language of Nigeria is? It's english.

Amplification: lots of languages are spoken in Nigeria. One of them is English.

No, english is the most widely spoke common language of Nigeria. Same situation as in India.

I also thought of the "panic button" idea, but then thought better of it since it would be comprehensively abused. You'd have kids pressing it, sleeping passengers accidentally knocking it and so on. It would be nearly impossible for the flight attendants to determine whether the button press was genuine, accidental, malicious or simply pressed by someone who was bored on an 8 hour trans-Atlantic flight and wanted the small amusement of watching his neighbour being body searched.

Why not post all the x-ray images on mechanical turk and give $1000 to anyone who can catch a weapon the TSA misses. That way we'll have an army of people trying to improve the TSA's awful success rate. I suspect the only reason we don't already do this is that it would expose the fact that the TSA catches only a tiny fraction of banned weapons.

No, you'll have an army of people falsely pointing out random shapes as weapons, hoping they'll win the lottery.

3 wrong guesses and you are banned for 3 months.

Which would cause people to ignore things that might be weapons.

You know that, even on mechanical turk, you'd still have to pay people something to look at the x-ray images, whether they find a weapon or not?

What you're describing would still cost a significant amount of money (even at, say, a penny per image?) and your screeners would have even less of an incentive to actually find any weapons than the current TSA does.

Yawn

Another ineffectual "solution" to a problem that cannot be solved. The idea of some nosy, self-important passenger on a flight deciding who is and is not a terrorist is ludicrous.

I think one reason that crowd sourcing works as well as it does on the web is because there is an actual crowd. What this article is referring to is more like "small group sourcing".

The benefit of the crowd is that there are generally enough people to counter act false alarms, prejudices, etc (not always, but generally). The proposed system would have only a few people one which to draw information rather than the hundreds, thousands or millions that these kinds of operations usually are deployed towards. Basing information on just a handful of people exposes many flaws that are generally evened out over actual crowds of people.

Is air terrorism a problem? More people have died from the pilot flying the plane into the ground than have died in terrorist attacks. And that "more people" is "not very many".

This seems to be a way to launder the racial profiling that we're unwilling to say we'll do through making the decision to profile diffuse among an arbitrarily large crowd of people who are not subject to feedback loops for doing it.

Phrased like that, I'm not sure if I oppose it or not.

Sidenote: This is a really good line: [T]he TSA continues to treat passengers as self-loading cargo that may harbor a terrorist.

This article was impossible to read on 1024x768 screen because the wrapping DIV was set to a width of 1000px.

Particularly irritating because it's a really interesting post.

try http://lab.arc90.com/experiments/readability/

Keep this Readability bookmarklet on your browser toolbar.

http://lab.arc90.com/experiments/readability/

I just changed the CSS with firebug, but thanks for the link

You say Solution as if there's really a problem? What's the statistic, one incident for every 16m miles? The real problem is the culture of fear these kinds of ideas perpetuate, ie letting the "terrorists" win in their main goal, which is to increase fear.

Agree.

I'd put both words -- "the terrorists" -- in quotes. There definitely are individual terrorists in the world, but the notion of a unified red team of "the terrorists" has motivated a lot of bad decisions.

I've always thought that elements of the terrorist infrastructure -- plots, operatives, muscle, money, connections -- are sort of cloud-provided and available to pretty much anybody capable of implementing the API. Al Quaeda is sort of the 37 Signals of terror: they aren't the central nexus, they're a media-savvy early adopter whose core competence is marketing and have a happy little Italian restaurant-scale kill-the-Jews-and-Crusaders business, and they inspire people all over the globe to quit the day jobs, try a little lean jihad, iterate rapidly, and blog about the whole process.

Sadly the I in API does not stand for Interface and we are totally unwilling to discuss what it does stand for so War on Terror it is.

Does it stand for "Basque"? No, wait, that doesn't start with "I". Oh! "I.R.A." No, wait, that's itself an acronym. It's not FARC... no, hrm, not "Shining Path", either. And clearly not Aum Shinrikyo. Wait, waitwaitwait... you couldn't mean "Islamism", could you? But what about all the terrorist organizations that aren't Islamist?

See, that is exactly why the War on Terror naming doesn't make sense (+). All of the above are terrorists (and contemptible), none of the above are even a remote threat to American air travel, and approximately none of our War on Terror related activities target them except incidentally.

http://upload.wikimedia.org/wikipedia/commons/3/39/Emblema_S...

Reply on news.ycombinator.com