I was waiting for that. You've shown that you have painted yourself into a corner.
How so ? Because I (foolishly, in hindsight) make an assumption that the people in the discussion are actually interested in the discussion, rather than debating-team style point scoring ?
Lol. I've seen it myself, you haven't. More corner painting.
You just described - even admitted yourself - that it would not look completely normal, but rather "too perfect". I certainly know that anyone I saw who had completely flawless and pale skin, without even a freckle, would stand out.
You do realize that you have strayed completely from the "stupid, idealistic, bleeding-heart-leftists" profiling system that you initially proposed, right?
No, I have not. *You* are the one who has continually made the assumption that whenever I refer to a "profiling system" I am referring to the (apparently utterly broken by design - I've never experienced it so I can't say based on first hand experience) system currently implemented by US airlines, or something else worthlessly simplistic based on skin colour and religion.
Allow me to quote from the paper you just read -- El Al's security system works from the assumption that every passenger is a threat
Allow me to quote from the same paper:
Upon entering the airport, a passenger is asked a series of specific questions including "Who paid for your ticket?" "What is the purpose of your travels?" and "When did you book the flight?" The questions are specifically designed to evoke an observable reaction. As passengers answer the questions, the officer carefully scrutinizes tone of voice, body language, and quickness of response. If the answers are unsatisfactory, a different officer will ask the ticket holder a different set of questions.
That is no longer about, "directing scarce resources where they are most likely to produce a positive result." El Al gives EVERYONE massive amounts of attention -- As El Al's president, David Hermesh, stated, "If you're a passenger on El Al, most likely you will be observed from the minute that you left your car or you have been dropped off."
But, clearly - as described in the paper - some passengers are given more attention than others, based on criteria. Ie: they are profiled.
Holding up El Al as an example anything even remotely feasible for the majority of the world, especially given the comparitively miniscule risk, is just silly.
Likewise, implying we should simply ignore all their experiences is stupid.
Your theory has morphed from racial and cultural profiling to behavioural analysis under dissection, although you still appear to have faith in some sort of massive system that is itself either information starved or big brother incarnate.
My theory was *never* based solely on racial and cultural profiling. It certainly considered them as part of the entire process, but never proposed them as the only, or even deciding, factors. Just becuase I didn't write a ten page essay describing every nuance of a properly run airline security system in my first post, doesn't mean I hadn't considered the "clever" ways you came up with to circumvent it.
I'll agree that some level of behavioural analysis has potential, although I can even think of simple ways to mess with that (e.g. $300 round of botox shots to the face kills visible evidence of anxiety, another $300's worth of shots to the sweat glands stops perspiration, and another $300 will prevent migranes in case the terrorist is subject to them too). If I can come up with that in about 2 minutes, I am sure somebody spending years on the effort can do much better.
The position you appear to be advocating, is that since every single workable security procedure that can be envisaged, can also be subverted, then we shouldn't make any attempts at airline security at all. This does not appear to me to be a reasonable solution.
Compared to the number of Muslims in the world a vanishingly small number of them are potential terrorists, yet they seem worthy of our attention, don't they?
You're moving the goalposts.
Remember, the 9/11 plane hikackings were just that - hijackings - not bombs.
I think making that distinction in this discussion is specious, to say the least. Although perhaps I should have said "cause damage with planes" to avoid allowing the semantic fallback. My bad, I suppose.
I really have to explain this?
You need to come up with an explanation which actually use *complex* and *combined* criteria to bias the search and allow for the additional random searches, rather than simplistic and individual criteria.
No proper profiling system is going to use just the skin colour, or just the flight destinations. It's going to look at everything from how the ticket was booked, to a passengers flight history compared to the average, to the type of bags they're using, to how they act when questioned at a security point. Most importantly, it needs to do so without the passenger ever knowing they're being "scored".
Don't get me wrong, I'm not arguing that intelligence lead targetting of resources is a bad thing, far from it - it's the only effective way of dealing with the enourmous number of people and risks. I'm arguing that assuming white people (or any other arbitrary category like young or old or married or single) pose a greater or lesser risk is foolish.
You just contradicted yourself. First you say that targeting based on criteria gleaned from intelligence is a good thing, then you're saying that selecting based on specific criteria is a bad thing. Which is it ?
Think about it - how many times have bombs actually been found by security screening? Compare that to how many times plots have been uncovered and thwarted by the traditional covert security services and police using conventional techniques like infiltration and comms interception.
Oh, I agree completely. I think the chances of discovering a bomb as part of the security checkpoint are slim at best. But I also think if there wasn't any security checking at all, the number of bombs (or just other weapons) getting carried onto planes would be *dramatically* higher.
Security checkpoints mean that getting a bomb past them requires planning and preparation. Planning and preparation take time, during which - hopefully - the people responsible can be located and caught using the resources you mention.
Ahh the old "people like me" line. Please don't generalise, it's offensive and makes you look kinda dumb.
I'm generalising because pretty much every single reply to my posting has said basically the same thing - "but they could just get some white guy to waltz right through the security check", as if there's a queue of white guys a mile long just itching to get onto a plane and blow it up.
But again, I digress. I assume that a white person can be "dropped in" because I don't see why it would be so difficult. There's nothing (that I can see) which would make a white Muslim any less likely to be a bomber than a non-white one. Sure there are less white Muslims, but there are enough to form a pool to draw candidates from.
Really ? The number of "white" suicide bombers - or even candidates - seems to be pretty small. There are certainly a reasonable number of white Muslims out there, but not many of them seem very keen to go out with a bang.
Despite what certain parts of the media would have us believe, suicide bombers are a distinct minority in the Muslim world. My point here is that when you take a demographic that is already relatively tiny (white Muslims) and then lay another relatively tiny one on top of it (suicide bombers) you end up with a very, very small group of people.
And, as I mentioned, don't neglect the possibility that the bomber may be unknowing or unwilling.
First, I think it's a dubious proposition that you can do much to prevent agents from measuring their susceptibility to additional scrutiny.
Ensuring there are also random searches in addition to targeted ones and putting significant effort into making sure a target never knows whether or not he is being searched due to triggering a flag condition, would help immensely. For example, searches should be carried out in a "blind" fashion, with the searcher *not* being the person who selected the searchee.
Unless you can perform extra searches unbeknownst to the searchee, it's going to be difficult to battle the strategy of simply going "dark" (to prevent contact with other agents setting off a new profiling flag), traveling repeatedly over some period of time, determining whether one is being searched more than average, and then using this to determine whether one is a good candidate for carrying out a mission.
Certainly, given sufficient time, the criteria could eventually be determined and possible steps taken to circumvent them. However, it may well be that in the process of researching these criteria, they begin to become interlinked. For example, the "tester" might need to fly the system ten times before they start to get an idea of what types of clothes, baggage, payment methods, etc are considering "normal". But if they were to be averaging, say, a flight every week doing this, then suddenly didn't fly for three months (while preparing for an attack) they might find on their attack run that their absence from the system has resulted in them triggering a different flag condition.
Anyway, my biggest problem with all this is that it's essentially impossible to measure the actual effectiveness of any of these measures. Terrorist events are so rare that you can't just divide hits by total attempts and come up with a percentage. Add to this that most of the details of the measures are (often by necessity) kept confidential, it's very difficult to watchdog the system for abuse.
Note that with a suitable system in place, opportunity for abuse by individuals should be relatively rare. Going back to my previous example, if the "selector" and the "searcher" are different people (ideally unknown to each other), then that removes one motivation the "selector" might have to only select certain types of people. Another way might be having the "random" checks triggered by a computer, rather than a person.
I can agree it's very difficult to assess the true effectiveness of such a system because of the relative rarity of terrorist events. However, it seems silly to me to waste resources on individuals that we can - to a large degree objectively - say have a very low probability of being a bad guy.
No, YOU moved past that point when you said your example was only a quickly made up hypothetical one.
My example was only ever a quickly made up hypothetical one and never even suggested as representative of a comprehensive, full profiling system. I would have thought that was blatantly fucking obvious to anyone with sufficient intelligence to compile a posting to Slashdot, mainly because of its incredibly simplistic and stereotypical nature, but clearly I've underestimated the bounds of human stupidity yet again.
It is completely indistinguishable from very fair caucasian skin - if anything it is too perfect, the result is near flawless skin that many women would kill for for since all of the pigmentation imperfections are removed. Looks a little weird on a guy, but very nice on a woman, or a female impersonator.
Right. So it doesn't look especially natural at all, is what you're saying. As I suspected it wouldn't.
Not that it matters since you are the one who said that it was only something you made up without much thought.
No, it doesn't matter because your "simple plan" is an "only in the movies" idea. Something makes me think you saw "White Chicks" a few too many times.
Again the presumption that the characteristics that matter can be profiled.
So long as the bad guys continue to exhibit certain common characteristics, *they can be*.
This line of thinking seems to run through all of your examples and most of your presumptions. It is NOTHING like chosing between the guy with the gun and the guy without a gun or the high-crime area vs the low crime area, etc, etc. On one hand you argue that "profiling" uses all kinds of subtle, hard to guess criteria but everytime you either list a possible criteria or make up an example, its all about blatant obvious stuff.
Probably because my examples are not meant to be comprehensive.
The problem is that all of the stuff that "stupid, idealistic, bleeding-heart-leftists" object to when profiling is discussed - the kind of stuff you listed in your hypothetical example about young, single, middle-eastern men from saudi - are all characteristics that don't matter. They are easily circumvented, and have such high rates of false positives, that they produce results that are less effective than completely random choices.
They're only circumventable if you have a suitably large pool of people who don't match them at your disposal.
If there is something I'm missing here, please list just two such profileable attributes that really are both meaningful and non-obvious.
It is my understanding that suitably well trained behavioral analysts can identify people who are acting "unnaturally", based on both passive observation and active questioning.
Just because some criteria are "obvious", doesn't mean they should be ignored. Not leaving your car keys behind the sun blind is an "obvious" way to avoid having your car stolen, but don't think a thief won't look there on the off chance you did.
Objectivity has nothing to do with it. Testability is all that matters and when you introduce randomness into the system, that only supports my point that random is more effective than targeting statisticly meaningless characteristics.
I was referring to the way criteria would change over time. However, a well designed security system would certainly use random as well as targeted flagging.
Furthermore, a formal and comprehensive proof is not a requirement - there are heuristics that can rapidly narrow the search space - like reading web sites where travelers post their experiences in discussions like this one. You don't have to find out what it takes to get stopped, all you have to do is find out what it takes to not get stopped.
The current system defaults to "not getting stopped". Therefore, the only reliable way to circumvent it would be to determine what it takes to get stopped - otherwise you could never be
Because you're the only person talking about "everyone being the same", in some misguided belief that successful terrorists are statistics rather than rational individuals carrying out a plan.
It's a well known phenomenon that recent converts to a belief or religon are often amongst it's most fierce advocates. For example, people who recently gave up smoking often will be more vocal in it's condemnation than those who never smoked in the first place. Likewise, those who convert to Islam from non-muslim backgrounds are actually surprisingly likely to be in the extreme minorities who would take part in such acts.
Certainly. But my argument is that there relatively few of them, not that they don't exist.
That, and of course the fact that muslims are not the only terrorists in the world (Tim McVeigh was Christian, so are the IRA, UDF, etc) [...]
The context of this discussion is - as far as I know - people trying to blow up aeroplanes in the US. Muslim extremists are, *by far* the most likely candidates for such an act.
[...] and the idea of targetting certain ethnicities is bad.
When there is a strong correlation between between ethnicity and other important factors, why ?
It doesn't make finding the bomber any more likely and it gives the planners a way of lowering the likleyhood of detection.
This conclusion is far from certain. You need evidence to support it or, at the very least, a reasoned argument that can be addressed.
If you know that a white person is even 10% less likely to be stopped you just use a white person and the odds are in your favour.
Why do people like you continue to work under the assumption that a "white person" can just be "dropped in" with the snap of some fingers ? What evidence is bringing you to the conclusion that some "white person" (your term, not mine) is equally as likely to be a suicide bomber ?
Chakrabarti and Strauss's Term Paper on the subject.
I haven't read the whole thing, however, based on a brief skim its conclusions are not relevant to this discussion because the paper is a critique of the *current* profiling system (or, at least, whatever system was current in 1999), a system that relies solely on criteria that can be determined at the ticketing desk. It also points out several flaws in that profiling system which could be rectified (for example, making it easy for the passenger to determine they had been flagged).
Added to that, it also assumes that *only* profiled searches will be used, whereas a properly run system would use both profiled and random searches.
The paper also seems to give disproportional weighting to "non traditional" terrorists. Three exceptions is not a reliable pool of data.
There is a distinction here that needs to be made between a properly run profiling system and a badly run profiling system. The "El Al" standard they talk about is better example of the latter. The paper certainly appears to be a valid critique of the CAPS system in place at the time, but I would be extremely cautious about extending its conclusions any further than that.
It is plenty easy compared to the planning involved in 9/11 and others like oplan bojinka - the hardest part is just sitting around waiting for the bleaching to take effect, and you don't even have to do the entire body, just the visible parts.
You seem to have forgotten the part where the person needs to change their behaviour so they don't look like a foreign man pretending to be a swedish woman.
You're also (still) ignoring the part where you're only circumventing a few of potentially dozens of criteria.
And even *that* is assuming your skin-bleaching, etc process produces a result that looks natural.
Meanwhile you've avoided justifying your position that it isn't so easy to use a profiling system against itself. Just what part of "find out the rules, and then avoid the rules" is so complicated?
Indeed. I mean, just what's so hard about interstellar colonisation ? It's simply a matter of finding habitable planets and dropping a few thousand people on them...
If you thinking finding out the rules is the hard part, then consider the phrase "test run."
Indeed, let's consider it. You need to find a statistically significant number of "volunteers" to test the system, all of whom must be essentially identical in all potential criteria - physical attributes and mannerisms (covering everything from way they dress through to how much they sweat when they're lying), flying history, check-in time, method of payment, etc, etc. Then you have to start varying attributes - only one at a time, however - over a statistically significant period of time, to determine what the profiling criteria might be. But you also have to account for randomness, where a search may have been performed not because of any flag being raised, but because the security guard at the time was just having a bad day. At the end of this process (which could itself potentially introduce a feedback loop - for example, by resulting in an "abnormal" amount of flying for that particular type of person) you presumably have a team of people who will only be vulnerable to random searches, as they will not trigger any flags. You then need to carefully balance how many you send - enough to completely the mission, but not so many as to risk one being picked up in a random search and thus blowing the whole operation.
This whole thing, also, hinges on the two *massive* assumptions that a) you can actually assemble a large enough team of people that don't trigger any flags and b) that profiling criteria are both completely static and completely objective.
So, no, it's not just a matter of "find out the rules, and then avoid the rules". Proper profiling systems don't just use simplistic and individual criteria like "brown skin" or "one way ticket". They glue together a whole range of criteria both objective (major physical attributes, payment method, flight type) and subjective (behavioural cues, "gut feeling").
Racial and cultural profiling is all about focusing on characteristics that millions of harmless people also share [...]
Actually its about focusing on the characteristics that the bad guys share. Since the vast majority of people are very similar, however, they're also going to share attributes with the general population.
With random searches there are no loopholes and you don't give the people you most want on your side reason to distrust you.
But you end up wasting equal amounts of resources on targets that do not represent equal levels of threat. It's like saying if you walk into your house being robbed by two men, one of whom is unarmed and one of whom has a pistol, that you should consider each to be of equal threat to your person, when this is almost always going to be the wrong thing to do.
Based on a swag of 1 muslim male per 200 passengers (total muslim population in the usa being approximately 1-2%, 2x-3x higher in europe) and the number of actual terrorists apprehended in any context the numbers are more like 0
The point, which everyone else got except for you, is that since Islam is a religion targeting "the Muslims" on the basis of skin color is really, really stupid.
Well, that explains why I didn't get it, since I made no such suggestion.
Sure, 'everyone profiles', that is until they get smart enough to realize that 9 times out of 10 they are completely off-base in their judgements. You can take a look at a person across the room and make up some set of assumptions about them based an the way they look, dress, and carry themselves, but try talking to them and you'll often feel foolish as you realize just how wrong you were.
So you're saying that 9/10 people you talk to, you are completely wrong about in every respect ?
I'm not going to try and say I've never made a bad assumption about someone - hell, even people I've known since they were children - but I can pretty confidently say my fuckup rate is a hell of a lot lower than 90% - and the more information I have about them, the lower it gets.
Everyone "profiles", all the time. They do it when they decide whether or not to walk down a dark alley with someone. They do it when they decide whether or not to believe something someone else says. They do it whenever they decide where to sit on the bus. "Profiling" based on statistical data and behavioural experts is a hell of a lot more objective and scientific than the processes the vast majority of people use every day to live their lives.
Perhaps it would be better to take a look at yourself and the culture that we live in before inditing others for theirs.
I have no problems at all with other cultures living however the hell they want. Personally, I'm mostly in favour of leaving the entire Middle East alone. The problems come when they don't want to leave us alone.
"Profiling" - performed properly has been proven MATHEMATICALLY to be less effective than a purely random search.
Link, please. I can't make a comment about something I haven't evaluated.
Now when he/she attacks, instead of say 10 officers doing random tests, you have say 8 officers doing random tests cuz the other 2 are doing profiling, which oh how convenient we know we don't match.
You're making the same wrong assumptions everyone else who has replied has - that profiling criteria are easy to deduce and simple to evade.
Now, filter a bit for things like a woman who is in their 20s and you'll increase the percentage a lot. Never the less, I'd imagine there's many Swedish women who fit the profile.
And I wouldn't disagree.
I'd argue, however, that relative to the number of young Muslim males who would be prepared to blow themselves up, the number is tiny. Which is the point.
The only two major factors that Arabs have over Swedes is that there's more Arabs than Swedes and countries like the US are directly occupying Arab countries.
You don't think things like religious beliefs, cultural beliefs, social standards, media influence, governmental policies, economics, etc are "major factors" ?
And just how many "young single "middle eastern" men with one-way tickets flyng out of Saudi Arabia" or from anywhere else for that matter have been taking down planes since 9/11? What, none? How many have been correctly arrested because they were profiled? What, none? How many have been arrested because they were incorrectly profiled? More than a few. Seems to me that the evidence is against your position.
Presumably this is your example of "proof by assertion" ?
Sorry, but proof by assertion is just proof by bullshiting. I laid out just how easy it is to circumvent, you are just waving your hands.
The example you "laid out" is in no way, shape, or form "easy". Plus, it only manages to circumvent a few possible criteria that I happened to think of off the top of my head - not all that I could think of, and certainly not all that someone whose job it was, could think of.
Hardly. We still get the occasional news story about some high and mighty politician having to be subjected to the same bullshit the hoi polloi has to suffer.
And the fact that it's considered news should show you how far out of touch they are - and have been for some time - for any attempts to drag them back into the real world to be meaningful.
And by that reasoning, deciding to eat a PB&J sandwich or not is a life or death decision. I'm sorry, but you've really gone past the edge of credible with that.
So you wouldn't consider the decision whether or not to walk through a high crime neighbourhood with a history of violent and fatal robberies or not to be a "life and death decision" ?
So, if accuracy is not important, what is the point?
No-one said accuracy wasn't important. They said accuracy could be traded off for speed. Being 90% accurate is still better than being 10% accurate.
Truly random searches *do* provide the best security if you are not searching everyone. Without a profile to avoid, bad guys will always have a chance of being singled out, and that will make them nervous...and if you have well trained security people, they will notice that person and single *them* out for extra checking. The threat of a search can be just as effective as the search itself!
So you're saying we should profile the nervous looking people ?
In my experience, responses referring to the argument being responded to as "ridiculous", "over the top", or "irrelevant hyperbole", [...]
And just to refresh, let's remind ourselves what some of those "arguments" were:
Ah, you're RIGHT! You mean like white Baptists from Mississippi! We must bar them at once from boarding an airplane!
[...]
But if you do want to ban those religions with a history of violence from their members, you'd deal a pretty severe blow to air travel within the US. After all, most who travel are self-identified "Christians", it's going to be a hell of a time to search all of them. And a history of violence? I mean, look at those people, from the Crusades, to the Inquisition, to their modern-day terrorist acts like abortion clinic bombings! This is obviously a religion whose believers hate peace and believe terrorism is alright! We can't let a single one of these people on a plane without a search, who KNOWS what such a dangerous lunatic might do?
[...] are highly correlated with the intended response: "I label your view stupid and idiotic, and desire to make no effort to understand it, despite it being potentially rational.
I label the arguments above stupid and idiotic because they *are*. The tiny part of it that was a reasonable idea was addressed both previously and in the direct response.
If someone has a rational point, then they should be able to make it *with a rational argument*. The "take it to a ridiculous extreme example" should be used as a last resort, not a first attack.
I am incapable of seeing past your style of writing and choose instead to attack your style of writing in a personal way while simultaneously leaving all objective points you attempted to make unaddressed."
I am far from incapable of seeing past the "style of writing". I simply don't shy away from giving it the label it deserves.
I have to wonder, do you seriously think the post I responded to put forth its arguments in a reasonable, balanced and appropriate fashion ?
Correct me if I'm wrong, but that seems to be an appropriate analysis of this potentially useful discussion, as similarly pointed out by the other guy's comment on unknowingly wasting your time.
You seem to be completely ignoring the basic points of the messages that you are responding to.
You seem to be ignoring the honkin' great big IFs that they rely on to make them meaningful.
If an intelligent attacker can figure out the criteria that you are using for profiling, then they will _deliberately_ look for an agent who will bypass that profile.
And there they are. Firstly, that the criteria can be reliably and completely detected and, secondly, that a suitable candidate can be found to evade them.
(And this is still talking solely about relatively obvious profiling criteria like ethnicity, religion and prior travelling habits - we haven't even started talking about behavioural profiling yet.)
Also, on rereading the post I was replying to, it seems I also missed an opportunity to point out how ridiculous the idea of "finding a young swedish radical blonde convert and buying her a round the world ticket with a credit card becomes incredibly easy" is - without even going into all the other profiling criteria that could potentially flag her.
The only part that statistics will play in this scenario is how long it takes for them to find a matching recruit or unknowing patsy.
No, statistics gives you certain criteria that prior terrorist attacks and suspects have met. Your (flawed, to say the least) assumption is that every single subsequent terrorist will meet none of these criteria and, hence, those prior occurrences should be ignored.
Profiling is useful _only_ when the potential targets you are attempting to profile to don't or can't change their behavior in response.
Yes. I have never suggested otherwise. In case you haven't noticed, suicide bombers - and potential suicide bombers - that aren't young, disaffected Muslim males usually of "middle-eastern appearance", aren't exactly common.
Since they can and do change their behavior [...]
Another bad assumption that you are making is that the profiling criteria will never change to match.
[...] encouraging the use of profiling simply makes you _think_ you're being efficient about your searching, when you're really opening up a hole in your security for an intelligent attacker to exploit (since your searching resources are being concentrated based on your profile).
Only if profiling is your one and only line of defense. Something I have also never suggested.
What is it with you people and using incredibly bad assumptions to support blinkered arguments ? Why does every shred of thought, rationality and intelligence fly out the window as soon as the suggestion is made that maybe - just maybe - not everyone is the same ?
I apologize, it seemed as though you were interested in a discussion on the issue and knew a few things.
I am.
If I'd known you were more interested in namecalling I'd not have wasted your time.
I calls it how I sees it. If someone makes a stupid suggestion, I see no reason why I shouldn't call it stupid.
Likewise, when someone spouts ridiculously over the top and irrelevant hyperbole instead of trying to enter into a rational, balanced and reasonable discussion, I'll call them an idiot.
In the future, if you want a "discussion on the issue", I suggest you don't start with the kind of post you did.
Given two years it is possible to convert a dark-skinned, black-haired muslim man into a "young single woman flying out of sweden."
Which would explain why there's so many young Swedish women (or people pretending to be) committing terrorist acts, right ? I mean, 9/11 was what - 5 years ago ? Iraq was invaded when ? Over 3 years ago ? Surely we should be seeing some of the results of these "two year" conversions by now ?
Anything you can come up with to base your profiling on can be used to work the system. All it takes is to figure out what the profiling rules are. Then all you are left with a big false sense of security.
Sorry, but circumventing the system is nowhere near as easy as you would like to pretend it is.
Plus, by profiling you make it very easy for certain people to "opt out" - like politicians and other members of government and law enforcement while the rest of the poor shlubs have to suffer through it. It's extremely important that these people not become even more sheltered from the realities of normal life (and the effects of the their own policy decisions) than they already are.
Way, *way* too late for that to be relevant.
They do it as a heuristic that saves time as a trade off for reduced accuracy.
Exactly. Which is precisely what has to be done when what you're doing involved millions of people and tight schedules.
But most rational people don't make life-or-death decisions based on profiling.
Sure they do. One example: deciding how to get home late at night.
You do, however, make the very good point that profiling is just using heuristics. Maybe if it was called 'Heuristic selection', fewer people would get their knickers in a twist about it.
Dude. Three names: Timothy McVeigh, Theodore Kaczynski, and Richard Reed. Three terrorists who would not set off your criteria.
At no point did I ever suggest my criteria to be exhaustive, nor such a process to be infallible.
You'll never get all of them. No sane person would argue otherwise. The objective is to get *most* of them.
More than just bleeding heart liberalism here, just common sense. Islam is a religion, not a race or ethnicity.
I'm sure you think you've made a point here, but I have no idea what it is...
You start looking for brown Muslims, and they'll start sending white ones. (Remember the white guy in the Al Qaeda video released around Election Day, 2004?)
Assuming they can find enough "white ones" to keep sending them. Last I checked, radicalised Islamic converts from the west were still pretty thin on the ground. Those willing to commit suicide, even rarer.
If we're going to bother to make a secure system, we might as well go through the trouble of making it work for real.
No-one will ever make a usable system 100% secure.
It can, however, be a valid cultural tag. A very, very simplistic one & not a guarantee, but...
That was my point:).
Skin colours gives a correlative, not causative, relationship.
However, religion doesn't necessarily leave any detectable marks.
Most religions tend to have a noticable impact on people's behaviour (assuming they're suitably devout). Language, dress, what they eat, read, watch, etc.
What would you do about a suicidal/homicidal Atheist? I was involved (many years ago) in a FIDO chat with Madalyn Murray O'Hair's grand-daughter Robin when she suddenly stopped posting. It turned out later that she'd been murdered (along with Madalyn) by David Roland Waters, an Atheist working for American Atheists as an office manager and typesetter. He evidently did it in order to be able to steal some gold coins. What if he'd wanted to blow up an airplane instead?
No system will ever be infallible, and I would never suggest otherwise.
However, I will point out that your suggestion profiling is concentrating only on religious beliefs (and maybe skin colour) is flawed.
Absolutely not-it is simply suggesting that if such a profile is used, any wannabe terrorist will find a young person to fly out of Sweden and make sure to purchase round-trip tickets in an attempt to duck the "profile." And probably succeed.
Except young Swedish women aren't exactly queuing up to commit suicide via exploding aeroplane, just in case you haven't noticed.
Suggesting that such people do not exist and such a plot would not be possible is the true folly.
Fortunately, I am making no such suggestion. What I *am* doing, is pointing out that such a plot is _vastly_ less likely to eventuate than the aforementioned one involving a group of disaffected Muslim youths.
As to idealistic-leftist-etc., I know many people (including several in law enforcement) from MANY backgrounds who agree profiling is ineffective and dangerous, and who don't have the slightest bit of the beliefs you listed. Surely you can make your point without name-calling or presuming about your opponents.
I'm afraid I can't come up with any other description for people who think everyone is equally likely to be a religiously fanatical suicide bomber than "stupid".
I'd be more than willing to bet there's *at least* as many "experts" out there who think profiling (done properly) would be affective, as there is who would disagree.
I'll freely admit to it! However, most people (including me) profile on objective and non-bigoted criteria.
Which is precisely what proponents of profiling are suggesting should be done.
When history and statistics currently show that terrorists are overwhelmingly young muslim males, then focusing more attention on young muslim males is neither subjective, nor bigoted, it's mathematics.
Also, "everyone" is not a government agent, which changes the rules 100%. I am -personally- allowed to hate black people, or women, or those with red hair, and refuse entirely to associate with or speak to them. (Note: I don't -support- this behavior in any way, I'm just noting it is allowed!) However, if I work at the DMV, and one of those people walks up to my desk, I -must- give them a driver's license if they meet the criteria for one.
Because handing over a drivers license and letting someone onto a plane are such similar situations...
Your criticism is based on a flawed assumption - that focusing on certain cultural and religious beliefs is done without justification, and is inherently subjective and bigoted.
However, any profile we use will by definition be able to figure out and subvert (are 90% of the searches against Arab-looking guys? Find someone white! Or black! Are 90% of the searches against males? Find a woman!).
Your circumvention techniques assume that suitably different people can simply be drop-in replacements. Again, I'll point out that this assumption is false.
Therefore, statistically and psychologically, the safest way is to make sure EVERYONE knows they have a chance of getting a search, be they Grandma or Mohammed in the turban.
What statistics are you using to support your argument that Grandma is equally as likely to be a suicide bomber as Mohammed *right now* ?
What psychology are you using to support your argument that people prepared to commit suicide by explosion will be deterred from doing so by the possibility of being discovered before they actually board the plane ?
Ah, you're RIGHT! You mean like white Baptists from Mississippi! We must bar them at once from boarding an airplane!
How many white baptists from Mississippi have blown up aeroplanes lately ? How many of them are calling for - or at least condoning - the destruction of the western world ?
Wait, did that just sound silly? I bet it did! Why? Because it's a hideously bigoted statement, but it's bigoted against those who are more like the picture of "us" (bit different flavor of Christianity, same skin color).
Slashdot Mirror
How so ? Because I (foolishly, in hindsight) make an assumption that the people in the discussion are actually interested in the discussion, rather than debating-team style point scoring ?
Lol. I've seen it myself, you haven't. More corner painting.
You just described - even admitted yourself - that it would not look completely normal, but rather "too perfect". I certainly know that anyone I saw who had completely flawless and pale skin, without even a freckle, would stand out.
You do realize that you have strayed completely from the "stupid, idealistic, bleeding-heart-leftists" profiling system that you initially proposed, right?
No, I have not. *You* are the one who has continually made the assumption that whenever I refer to a "profiling system" I am referring to the (apparently utterly broken by design - I've never experienced it so I can't say based on first hand experience) system currently implemented by US airlines, or something else worthlessly simplistic based on skin colour and religion.
Allow me to quote from the paper you just read -- El Al's security system works from the assumption that every passenger is a threat
Allow me to quote from the same paper:
That is no longer about, "directing scarce resources where they are most likely to produce a positive result." El Al gives EVERYONE massive amounts of attention -- As El Al's president, David Hermesh, stated, "If you're a passenger on El Al, most likely you will be observed from the minute that you left your car or you have been dropped off."
But, clearly - as described in the paper - some passengers are given more attention than others, based on criteria. Ie: they are profiled.
Holding up El Al as an example anything even remotely feasible for the majority of the world, especially given the comparitively miniscule risk, is just silly.
Likewise, implying we should simply ignore all their experiences is stupid.
Your theory has morphed from racial and cultural profiling to behavioural analysis under dissection, although you still appear to have faith in some sort of massive system that is itself either information starved or big brother incarnate.
My theory was *never* based solely on racial and cultural profiling. It certainly considered them as part of the entire process, but never proposed them as the only, or even deciding, factors. Just becuase I didn't write a ten page essay describing every nuance of a properly run airline security system in my first post, doesn't mean I hadn't considered the "clever" ways you came up with to circumvent it.
I'll agree that some level of behavioural analysis has potential, although I can even think of simple ways to mess with that (e.g. $300 round of botox shots to the face kills visible evidence of anxiety, another $300's worth of shots to the sweat glands stops perspiration, and another $300 will prevent migranes in case the terrorist is subject to them too). If I can come up with that in about 2 minutes, I am sure somebody spending years on the effort can do much better.
The position you appear to be advocating, is that since every single workable security procedure that can be envisaged, can also be subverted, then we shouldn't make any attempts at airline security at all. This does not appear to me to be a reasonable solution.
You're moving the goalposts.
Remember, the 9/11 plane hikackings were just that - hijackings - not bombs.
I think making that distinction in this discussion is specious, to say the least. Although perhaps I should have said "cause damage with planes" to avoid allowing the semantic fallback. My bad, I suppose.
I really have to explain this?
You need to come up with an explanation which actually use *complex* and *combined* criteria to bias the search and allow for the additional random searches, rather than simplistic and individual criteria.
No proper profiling system is going to use just the skin colour, or just the flight destinations. It's going to look at everything from how the ticket was booked, to a passengers flight history compared to the average, to the type of bags they're using, to how they act when questioned at a security point. Most importantly, it needs to do so without the passenger ever knowing they're being "scored".
Don't get me wrong, I'm not arguing that intelligence lead targetting of resources is a bad thing, far from it - it's the only effective way of dealing with the enourmous number of people and risks. I'm arguing that assuming white people (or any other arbitrary category like young or old or married or single) pose a greater or lesser risk is foolish.
You just contradicted yourself. First you say that targeting based on criteria gleaned from intelligence is a good thing, then you're saying that selecting based on specific criteria is a bad thing. Which is it ?
Think about it - how many times have bombs actually been found by security screening? Compare that to how many times plots have been uncovered and thwarted by the traditional covert security services and police using conventional techniques like infiltration and comms interception.
Oh, I agree completely. I think the chances of discovering a bomb as part of the security checkpoint are slim at best. But I also think if there wasn't any security checking at all, the number of bombs (or just other weapons) getting carried onto planes would be *dramatically* higher.
Security checkpoints mean that getting a bomb past them requires planning and preparation. Planning and preparation take time, during which - hopefully - the people responsible can be located and caught using the resources you mention.
Ahh the old "people like me" line. Please don't generalise, it's offensive and makes you look kinda dumb.
I'm generalising because pretty much every single reply to my posting has said basically the same thing - "but they could just get some white guy to waltz right through the security check", as if there's a queue of white guys a mile long just itching to get onto a plane and blow it up.
But again, I digress. I assume that a white person can be "dropped in" because I don't see why it would be so difficult. There's nothing (that I can see) which would make a white Muslim any less likely to be a bomber than a non-white one. Sure there are less white Muslims, but there are enough to form a pool to draw candidates from.
Really ? The number of "white" suicide bombers - or even candidates - seems to be pretty small. There are certainly a reasonable number of white Muslims out there, but not many of them seem very keen to go out with a bang.
Despite what certain parts of the media would have us believe, suicide bombers are a distinct minority in the Muslim world. My point here is that when you take a demographic that is already relatively tiny (white Muslims) and then lay another relatively tiny one on top of it (suicide bombers) you end up with a very, very small group of people.
And, as I mentioned, don't neglect the possibility that the bomber may be unknowing or unwilling.
I think this is much les
I'm pretty sure that amendment doesn't apply to your dealing with private companies.
Ensuring there are also random searches in addition to targeted ones and putting significant effort into making sure a target never knows whether or not he is being searched due to triggering a flag condition, would help immensely. For example, searches should be carried out in a "blind" fashion, with the searcher *not* being the person who selected the searchee.
Unless you can perform extra searches unbeknownst to the searchee, it's going to be difficult to battle the strategy of simply going "dark" (to prevent contact with other agents setting off a new profiling flag), traveling repeatedly over some period of time, determining whether one is being searched more than average, and then using this to determine whether one is a good candidate for carrying out a mission.
Certainly, given sufficient time, the criteria could eventually be determined and possible steps taken to circumvent them. However, it may well be that in the process of researching these criteria, they begin to become interlinked. For example, the "tester" might need to fly the system ten times before they start to get an idea of what types of clothes, baggage, payment methods, etc are considering "normal". But if they were to be averaging, say, a flight every week doing this, then suddenly didn't fly for three months (while preparing for an attack) they might find on their attack run that their absence from the system has resulted in them triggering a different flag condition.
Anyway, my biggest problem with all this is that it's essentially impossible to measure the actual effectiveness of any of these measures. Terrorist events are so rare that you can't just divide hits by total attempts and come up with a percentage. Add to this that most of the details of the measures are (often by necessity) kept confidential, it's very difficult to watchdog the system for abuse.
Note that with a suitable system in place, opportunity for abuse by individuals should be relatively rare. Going back to my previous example, if the "selector" and the "searcher" are different people (ideally unknown to each other), then that removes one motivation the "selector" might have to only select certain types of people. Another way might be having the "random" checks triggered by a computer, rather than a person.
I can agree it's very difficult to assess the true effectiveness of such a system because of the relative rarity of terrorist events. However, it seems silly to me to waste resources on individuals that we can - to a large degree objectively - say have a very low probability of being a bad guy.
My example was only ever a quickly made up hypothetical one and never even suggested as representative of a comprehensive, full profiling system. I would have thought that was blatantly fucking obvious to anyone with sufficient intelligence to compile a posting to Slashdot, mainly because of its incredibly simplistic and stereotypical nature, but clearly I've underestimated the bounds of human stupidity yet again.
It is completely indistinguishable from very fair caucasian skin - if anything it is too perfect, the result is near flawless skin that many women would kill for for since all of the pigmentation imperfections are removed. Looks a little weird on a guy, but very nice on a woman, or a female impersonator.
Right. So it doesn't look especially natural at all, is what you're saying. As I suspected it wouldn't.
Not that it matters since you are the one who said that it was only something you made up without much thought.
No, it doesn't matter because your "simple plan" is an "only in the movies" idea. Something makes me think you saw "White Chicks" a few too many times.
Again the presumption that the characteristics that matter can be profiled.
So long as the bad guys continue to exhibit certain common characteristics, *they can be*.
This line of thinking seems to run through all of your examples and most of your presumptions. It is NOTHING like chosing between the guy with the gun and the guy without a gun or the high-crime area vs the low crime area, etc, etc. On one hand you argue that "profiling" uses all kinds of subtle, hard to guess criteria but everytime you either list a possible criteria or make up an example, its all about blatant obvious stuff.
Probably because my examples are not meant to be comprehensive.
The problem is that all of the stuff that "stupid, idealistic, bleeding-heart-leftists" object to when profiling is discussed - the kind of stuff you listed in your hypothetical example about young, single, middle-eastern men from saudi - are all characteristics that don't matter. They are easily circumvented, and have such high rates of false positives, that they produce results that are less effective than completely random choices.
They're only circumventable if you have a suitably large pool of people who don't match them at your disposal.
If there is something I'm missing here, please list just two such profileable attributes that really are both meaningful and non-obvious.
It is my understanding that suitably well trained behavioral analysts can identify people who are acting "unnaturally", based on both passive observation and active questioning.
Just because some criteria are "obvious", doesn't mean they should be ignored. Not leaving your car keys behind the sun blind is an "obvious" way to avoid having your car stolen, but don't think a thief won't look there on the off chance you did.
Objectivity has nothing to do with it. Testability is all that matters and when you introduce randomness into the system, that only supports my point that random is more effective than targeting statisticly meaningless characteristics.
I was referring to the way criteria would change over time. However, a well designed security system would certainly use random as well as targeted flagging.
Furthermore, a formal and comprehensive proof is not a requirement - there are heuristics that can rapidly narrow the search space - like reading web sites where travelers post their experiences in discussions like this one. You don't have to find out what it takes to get stopped, all you have to do is find out what it takes to not get stopped.
The current system defaults to "not getting stopped". Therefore, the only reliable way to circumvent it would be to determine what it takes to get stopped - otherwise you could never be
False.
Certainly. But my argument is that there relatively few of them, not that they don't exist.
That, and of course the fact that muslims are not the only terrorists in the world (Tim McVeigh was Christian, so are the IRA, UDF, etc) [...]
The context of this discussion is - as far as I know - people trying to blow up aeroplanes in the US. Muslim extremists are, *by far* the most likely candidates for such an act.
[...] and the idea of targetting certain ethnicities is bad.
When there is a strong correlation between between ethnicity and other important factors, why ?
It doesn't make finding the bomber any more likely and it gives the planners a way of lowering the likleyhood of detection.
This conclusion is far from certain. You need evidence to support it or, at the very least, a reasoned argument that can be addressed.
If you know that a white person is even 10% less likely to be stopped you just use a white person and the odds are in your favour.
Why do people like you continue to work under the assumption that a "white person" can just be "dropped in" with the snap of some fingers ? What evidence is bringing you to the conclusion that some "white person" (your term, not mine) is equally as likely to be a suicide bomber ?
I haven't read the whole thing, however, based on a brief skim its conclusions are not relevant to this discussion because the paper is a critique of the *current* profiling system (or, at least, whatever system was current in 1999), a system that relies solely on criteria that can be determined at the ticketing desk. It also points out several flaws in that profiling system which could be rectified (for example, making it easy for the passenger to determine they had been flagged).
Added to that, it also assumes that *only* profiled searches will be used, whereas a properly run system would use both profiled and random searches.
The paper also seems to give disproportional weighting to "non traditional" terrorists. Three exceptions is not a reliable pool of data.
There is a distinction here that needs to be made between a properly run profiling system and a badly run profiling system. The "El Al" standard they talk about is better example of the latter. The paper certainly appears to be a valid critique of the CAPS system in place at the time, but I would be extremely cautious about extending its conclusions any further than that.
You seem to have forgotten the part where the person needs to change their behaviour so they don't look like a foreign man pretending to be a swedish woman.
You're also (still) ignoring the part where you're only circumventing a few of potentially dozens of criteria.
And even *that* is assuming your skin-bleaching, etc process produces a result that looks natural.
Meanwhile you've avoided justifying your position that it isn't so easy to use a profiling system against itself. Just what part of "find out the rules, and then avoid the rules" is so complicated?
Indeed. I mean, just what's so hard about interstellar colonisation ? It's simply a matter of finding habitable planets and dropping a few thousand people on them...
If you thinking finding out the rules is the hard part, then consider the phrase "test run."
Indeed, let's consider it. You need to find a statistically significant number of "volunteers" to test the system, all of whom must be essentially identical in all potential criteria - physical attributes and mannerisms (covering everything from way they dress through to how much they sweat when they're lying), flying history, check-in time, method of payment, etc, etc. Then you have to start varying attributes - only one at a time, however - over a statistically significant period of time, to determine what the profiling criteria might be. But you also have to account for randomness, where a search may have been performed not because of any flag being raised, but because the security guard at the time was just having a bad day. At the end of this process (which could itself potentially introduce a feedback loop - for example, by resulting in an "abnormal" amount of flying for that particular type of person) you presumably have a team of people who will only be vulnerable to random searches, as they will not trigger any flags. You then need to carefully balance how many you send - enough to completely the mission, but not so many as to risk one being picked up in a random search and thus blowing the whole operation.
This whole thing, also, hinges on the two *massive* assumptions that a) you can actually assemble a large enough team of people that don't trigger any flags and b) that profiling criteria are both completely static and completely objective.
So, no, it's not just a matter of "find out the rules, and then avoid the rules". Proper profiling systems don't just use simplistic and individual criteria like "brown skin" or "one way ticket". They glue together a whole range of criteria both objective (major physical attributes, payment method, flight type) and subjective (behavioural cues, "gut feeling").
Racial and cultural profiling is all about focusing on characteristics that millions of harmless people also share [...]
Actually its about focusing on the characteristics that the bad guys share. Since the vast majority of people are very similar, however, they're also going to share attributes with the general population.
With random searches there are no loopholes and you don't give the people you most want on your side reason to distrust you.
But you end up wasting equal amounts of resources on targets that do not represent equal levels of threat. It's like saying if you walk into your house being robbed by two men, one of whom is unarmed and one of whom has a pistol, that you should consider each to be of equal threat to your person, when this is almost always going to be the wrong thing to do.
Based on a swag of 1 muslim male per 200 passengers (total muslim population in the usa being approximately 1-2%, 2x-3x higher in europe) and the number of actual terrorists apprehended in any context the numbers are more like 0
Well, that explains why I didn't get it, since I made no such suggestion.
So you're saying that 9/10 people you talk to, you are completely wrong about in every respect ?
I'm not going to try and say I've never made a bad assumption about someone - hell, even people I've known since they were children - but I can pretty confidently say my fuckup rate is a hell of a lot lower than 90% - and the more information I have about them, the lower it gets.
Everyone "profiles", all the time. They do it when they decide whether or not to walk down a dark alley with someone. They do it when they decide whether or not to believe something someone else says. They do it whenever they decide where to sit on the bus. "Profiling" based on statistical data and behavioural experts is a hell of a lot more objective and scientific than the processes the vast majority of people use every day to live their lives.
Perhaps it would be better to take a look at yourself and the culture that we live in before inditing others for theirs.
I have no problems at all with other cultures living however the hell they want. Personally, I'm mostly in favour of leaving the entire Middle East alone. The problems come when they don't want to leave us alone.
Link, please. I can't make a comment about something I haven't evaluated.
Now when he/she attacks, instead of say 10 officers doing random tests, you have say 8 officers doing random tests cuz the other 2 are doing profiling, which oh how convenient we know we don't match.
You're making the same wrong assumptions everyone else who has replied has - that profiling criteria are easy to deduce and simple to evade.
And I wouldn't disagree.
I'd argue, however, that relative to the number of young Muslim males who would be prepared to blow themselves up, the number is tiny. Which is the point.
The only two major factors that Arabs have over Swedes is that there's more Arabs than Swedes and countries like the US are directly occupying Arab countries.
You don't think things like religious beliefs, cultural beliefs, social standards, media influence, governmental policies, economics, etc are "major factors" ?
Presumably this is your example of "proof by assertion" ?
Sorry, but proof by assertion is just proof by bullshiting. I laid out just how easy it is to circumvent, you are just waving your hands.
The example you "laid out" is in no way, shape, or form "easy". Plus, it only manages to circumvent a few possible criteria that I happened to think of off the top of my head - not all that I could think of, and certainly not all that someone whose job it was, could think of.
Hardly. We still get the occasional news story about some high and mighty politician having to be subjected to the same bullshit the hoi polloi has to suffer.
And the fact that it's considered news should show you how far out of touch they are - and have been for some time - for any attempts to drag them back into the real world to be meaningful.
And by that reasoning, deciding to eat a PB&J sandwich or not is a life or death decision. I'm sorry, but you've really gone past the edge of credible with that.
So you wouldn't consider the decision whether or not to walk through a high crime neighbourhood with a history of violent and fatal robberies or not to be a "life and death decision" ?
So, if accuracy is not important, what is the point?
No-one said accuracy wasn't important. They said accuracy could be traded off for speed. Being 90% accurate is still better than being 10% accurate.
How does it deprive your of your "right to privacy" ?
So you're saying we should profile the nervous looking people ?
And just to refresh, let's remind ourselves what some of those "arguments" were:
[...]
[...] are highly correlated with the intended response: "I label your view stupid and idiotic, and desire to make no effort to understand it, despite it being potentially rational.
I label the arguments above stupid and idiotic because they *are*. The tiny part of it that was a reasonable idea was addressed both previously and in the direct response.
If someone has a rational point, then they should be able to make it *with a rational argument*. The "take it to a ridiculous extreme example" should be used as a last resort, not a first attack.
I am incapable of seeing past your style of writing and choose instead to attack your style of writing in a personal way while simultaneously leaving all objective points you attempted to make unaddressed."
I am far from incapable of seeing past the "style of writing". I simply don't shy away from giving it the label it deserves.
I have to wonder, do you seriously think the post I responded to put forth its arguments in a reasonable, balanced and appropriate fashion ?
Correct me if I'm wrong, but that seems to be an appropriate analysis of this potentially useful discussion, as similarly pointed out by the other guy's comment on unknowingly wasting your time.
Consider yourself corrected.
You seem to be ignoring the honkin' great big IFs that they rely on to make them meaningful.
If an intelligent attacker can figure out the criteria that you are using for profiling, then they will _deliberately_ look for an agent who will bypass that profile.
And there they are. Firstly, that the criteria can be reliably and completely detected and, secondly, that a suitable candidate can be found to evade them.
(And this is still talking solely about relatively obvious profiling criteria like ethnicity, religion and prior travelling habits - we haven't even started talking about behavioural profiling yet.)
Also, on rereading the post I was replying to, it seems I also missed an opportunity to point out how ridiculous the idea of "finding a young swedish radical blonde convert and buying her a round the world ticket with a credit card becomes incredibly easy" is - without even going into all the other profiling criteria that could potentially flag her.
The only part that statistics will play in this scenario is how long it takes for them to find a matching recruit or unknowing patsy.
No, statistics gives you certain criteria that prior terrorist attacks and suspects have met. Your (flawed, to say the least) assumption is that every single subsequent terrorist will meet none of these criteria and, hence, those prior occurrences should be ignored.
Profiling is useful _only_ when the potential targets you are attempting to profile to don't or can't change their behavior in response.
Yes. I have never suggested otherwise. In case you haven't noticed, suicide bombers - and potential suicide bombers - that aren't young, disaffected Muslim males usually of "middle-eastern appearance", aren't exactly common.
Since they can and do change their behavior [...]
Another bad assumption that you are making is that the profiling criteria will never change to match.
[...] encouraging the use of profiling simply makes you _think_ you're being efficient about your searching, when you're really opening up a hole in your security for an intelligent attacker to exploit (since your searching resources are being concentrated based on your profile).
Only if profiling is your one and only line of defense. Something I have also never suggested.
What is it with you people and using incredibly bad assumptions to support blinkered arguments ? Why does every shred of thought, rationality and intelligence fly out the window as soon as the suggestion is made that maybe - just maybe - not everyone is the same ?
The current US Government (ie: George Bush) is about as protectionist as they come...
I am.
If I'd known you were more interested in namecalling I'd not have wasted your time.
I calls it how I sees it. If someone makes a stupid suggestion, I see no reason why I shouldn't call it stupid.
Likewise, when someone spouts ridiculously over the top and irrelevant hyperbole instead of trying to enter into a rational, balanced and reasonable discussion, I'll call them an idiot.
In the future, if you want a "discussion on the issue", I suggest you don't start with the kind of post you did.
Which would explain why there's so many young Swedish women (or people pretending to be) committing terrorist acts, right ? I mean, 9/11 was what - 5 years ago ? Iraq was invaded when ? Over 3 years ago ? Surely we should be seeing some of the results of these "two year" conversions by now ?
Anything you can come up with to base your profiling on can be used to work the system. All it takes is to figure out what the profiling rules are. Then all you are left with a big false sense of security.
Sorry, but circumventing the system is nowhere near as easy as you would like to pretend it is.
Plus, by profiling you make it very easy for certain people to "opt out" - like politicians and other members of government and law enforcement while the rest of the poor shlubs have to suffer through it. It's extremely important that these people not become even more sheltered from the realities of normal life (and the effects of the their own policy decisions) than they already are.
Way, *way* too late for that to be relevant.
They do it as a heuristic that saves time as a trade off for reduced accuracy.
Exactly. Which is precisely what has to be done when what you're doing involved millions of people and tight schedules.
But most rational people don't make life-or-death decisions based on profiling.
Sure they do. One example: deciding how to get home late at night.
You do, however, make the very good point that profiling is just using heuristics. Maybe if it was called 'Heuristic selection', fewer people would get their knickers in a twist about it.
At no point did I ever suggest my criteria to be exhaustive, nor such a process to be infallible.
You'll never get all of them. No sane person would argue otherwise. The objective is to get *most* of them.
More than just bleeding heart liberalism here, just common sense. Islam is a religion, not a race or ethnicity.
I'm sure you think you've made a point here, but I have no idea what it is...
You start looking for brown Muslims, and they'll start sending white ones. (Remember the white guy in the Al Qaeda video released around Election Day, 2004?)
Assuming they can find enough "white ones" to keep sending them. Last I checked, radicalised Islamic converts from the west were still pretty thin on the ground. Those willing to commit suicide, even rarer.
If we're going to bother to make a secure system, we might as well go through the trouble of making it work for real.
No-one will ever make a usable system 100% secure.
Stunning as it might seem, they're not *only* looking for terrorists.
Good old fashioned illegal immigrants, drug smugglers, money launderers and criminals are still around.
That was my point :).
Skin colours gives a correlative, not causative, relationship.
However, religion doesn't necessarily leave any detectable marks.
Most religions tend to have a noticable impact on people's behaviour (assuming they're suitably devout). Language, dress, what they eat, read, watch, etc.
What would you do about a suicidal/homicidal Atheist? I was involved (many years ago) in a FIDO chat with Madalyn Murray O'Hair's grand-daughter Robin when she suddenly stopped posting. It turned out later that she'd been murdered (along with Madalyn) by David Roland Waters, an Atheist working for American Atheists as an office manager and typesetter. He evidently did it in order to be able to steal some gold coins. What if he'd wanted to blow up an airplane instead?
No system will ever be infallible, and I would never suggest otherwise.
However, I will point out that your suggestion profiling is concentrating only on religious beliefs (and maybe skin colour) is flawed.
Except young Swedish women aren't exactly queuing up to commit suicide via exploding aeroplane, just in case you haven't noticed.
Suggesting that such people do not exist and such a plot would not be possible is the true folly.
Fortunately, I am making no such suggestion. What I *am* doing, is pointing out that such a plot is _vastly_ less likely to eventuate than the aforementioned one involving a group of disaffected Muslim youths.
As to idealistic-leftist-etc., I know many people (including several in law enforcement) from MANY backgrounds who agree profiling is ineffective and dangerous, and who don't have the slightest bit of the beliefs you listed. Surely you can make your point without name-calling or presuming about your opponents.
I'm afraid I can't come up with any other description for people who think everyone is equally likely to be a religiously fanatical suicide bomber than "stupid".
I'd be more than willing to bet there's *at least* as many "experts" out there who think profiling (done properly) would be affective, as there is who would disagree.
I'll freely admit to it! However, most people (including me) profile on objective and non-bigoted criteria.
Which is precisely what proponents of profiling are suggesting should be done.
When history and statistics currently show that terrorists are overwhelmingly young muslim males, then focusing more attention on young muslim males is neither subjective, nor bigoted, it's mathematics.
Also, "everyone" is not a government agent, which changes the rules 100%. I am -personally- allowed to hate black people, or women, or those with red hair, and refuse entirely to associate with or speak to them. (Note: I don't -support- this behavior in any way, I'm just noting it is allowed!) However, if I work at the DMV, and one of those people walks up to my desk, I -must- give them a driver's license if they meet the criteria for one.
Because handing over a drivers license and letting someone onto a plane are such similar situations...
Your criticism is based on a flawed assumption - that focusing on certain cultural and religious beliefs is done without justification, and is inherently subjective and bigoted.
However, any profile we use will by definition be able to figure out and subvert (are 90% of the searches against Arab-looking guys? Find someone white! Or black! Are 90% of the searches against males? Find a woman!).
Your circumvention techniques assume that suitably different people can simply be drop-in replacements. Again, I'll point out that this assumption is false.
Therefore, statistically and psychologically, the safest way is to make sure EVERYONE knows they have a chance of getting a search, be they Grandma or Mohammed in the turban.
What statistics are you using to support your argument that Grandma is equally as likely to be a suicide bomber as Mohammed *right now* ?
What psychology are you using to support your argument that people prepared to commit suicide by explosion will be deterred from doing so by the possibility of being discovered before they actually board the plane ?
Ah, you're RIGHT! You mean like white Baptists from Mississippi! We must bar them at once from boarding an airplane!
How many white baptists from Mississippi have blown up aeroplanes lately ? How many of them are calling for - or at least condoning - the destruction of the western world ?
Wait, did that just sound silly? I bet it did! Why? Because it's a hideously bigoted statement, but it's bigoted against those who are more like the picture of "us" (bit different flavor of Christianity, same skin color).
No, it's stupid because it