Why we will not be visiting the US in the forseeable future

[drkitten]

And how do you look for "suspicious" activity without observing innocent people as well as the bad guys?

Hmmm..... Thousands and thousands of flight students across the country, but this particualr group of students is not interested in learning about landings. Suspicious?

Basically, by good old-fashioned police work.

 

[Katana]

No, it only looks like a non sequitur to someone who misses the point. Politeness, as Godmode describes it, is deference. The free and brave do not kneel so easily.

Hope that helps. Tomorrow we'll tackle the concept of "irony."

May I request that you add "slippery slope" to tomorrow's lesson plan, as well. Then again, some have that all down pat. Never mind.

I'll go back to believing in anything that I'm told will make me safer (because everyone knows I never question anything), supporting blanket investigations (if that's what some people think taking fingerprints qualifies as), and calling everyone else a coward now.

Meanwhile, to those of you who believe that that's what's happening, carry on with the blanket statements and mischaracterizations.

 

[drkitten]

I'll go back to believing in anything that I'm told will make me safer (because everyone knows I never question anything), supporting blanket investigations (if that's what some people think taking fingerprints qualifies as), and calling everyone else a coward now.

Katana, under what circumstances would you no longer support, for example, requiring people to surrender their fingerprints, banking records, and email when they entered the country?

BPSCG is correct that such information, if properly used, could be of tremendous help in an anti-terrorism investigation -- or a criminal investigation generally. The problem, however, is that at some level of (in)accuracy, the amount of work it creates to deal with the haystack of false findings will actually take manpower away from whatever needles might have been turned up.

For example, here's some data on the accuracy of fingerprint identification:

A recent report by National Institute of Standards and Technology (NIST) showed that one-fingerprint identification systems had an accuracy rate of 98.6 percent, while the accuracy rate rose to 99.6 when two fingerprints were used and 99.9 when four, eight and ten fingerprints were used.

Obviously, this will change somewhat as technology changes -- but let's look at this in more detail. If fingerprint identification is 99.9% accurate, that means that one person in a thousand will be mis-identified as any given suspect. With four million visitors from the UK alone trying to enter the country, this means that you'll have to investigate four thousand people for every person on the terrorist watch list, each year. How much time and money is DHS budgeting for this kind of wild goose chase? Can you think of something else that you would want to do with that kind of money, something that might perhaps be more effective?

 

[Katana]

Katana, under what circumstances would you no longer support, for example, requiring people to surrender their fingerprints, banking records, and email when they entered the country?

I have already said that I am uncomfortable with the credit card requirements. As for general bank records, they should not be required. I have said that I do not support requiring people to surrender their e-mail addresses.

As for fingerprints, I wouldn't support requiring them if the harm to individuals and the cost outweighed any potential benefit. What point is that? Well, that's likely to be a value judgement. If you never catch a terrorist trying to board a plane, does that mean that the system was a waste and that you spent all of this money on an unnecessary tool, or did it serve as a deterrent?

What is too much harm to individuals? I come up with phrases such as "significant disruption to a person's travels", "emotional trauma", "significant embarassment". Those don't mean much at the end of the day, though. Since false positives are going to happen, I would want to know that there is a smooth process for confirming the identity of those flagged in a timely manner so as not to cost them financially or personally, which would be unlike what we seem to have with our "no fly" lists. In other words, there's work to be done there.

BPSCG is correct that such information, if properly used, could be of tremendous help in an anti-terrorism investigation -- or a criminal investigation generally. The problem, however, is that at some level of (in)accuracy, the amount of work it creates to deal with the haystack of false findings will actually take manpower away from whatever needles might have been turned up.

If this requiring fingerprinting is at this level of inaccuracy, then it is unsupportable, but do we know that it is?

For example, here's some data on the accuracy of fingerprint identification:


Obviously, this will change somewhat as technology changes -- but let's look at this in more detail. If fingerprint identification is 99.9% accurate, that means that one person in a thousand will be mis-identified as any given suspect. With four million visitors from the UK alone trying to enter the country, this means that you'll have to investigate four thousand people for every person on the terrorist watch list, each year. How much time and money is DHS budgeting for this kind of wild goose chase? Can you think of something else that you would want to do with that kind of money, something that might perhaps be more effective?

I don't mean to sound like I'm nit-picking, is what if it's better than 99.9%? It very well could be once you're fingerprinting all 10 fingers. People usually don't report farther out than one decimal point, and no one is going to say that anything is 100% accurate. What if it was 99.99% accurate? Would I tolerate, say, a 1 in 10,000 error rate with an efficient system in place to confirm an individual's identify once flagged? Yes. 1 in 1000? No.

I wonder how much each false positive would cost. Some will obviously be easier to clear up than others. I wouldn't think that it would be too costly to deal with a 20-year-old woman whose prints match a 40-year-old Iraqi man's. Obviously, they won't all be that easy, but I would think that a fair number would be pretty straightfoward.

 

[Luke T.]

Hmmm..... Thousands and thousands of flight students across the country, but this particualr group of students is not interested in learning about landings. Suspicious?

Basically, by good old-fashioned police work.

File this under "Be Careful What You Ask For".

Post-9/11, it was, "Why didn't Bush act on the terrorist threat?"

After the next terrorist attack, it will be, "Why didn't Bush (or whoever is President) act on the terrorist threat?" And the answer will be, "I was busy poring through 11 million Verizon phone records and chasing down people who bought the Koran on Amazon.com."

 

[drkitten]

I don't mean to sound like I'm nit-picking, is what if it's better than 99.9%?

Then the source that I cited is wrong. It woudn't be the first time that a reporter got a number wrong. But at least we've now got a number that we can discuss instead of some abstraction.

It very well could be once you're fingerprinting all 10 fingers. People usually don't report farther out than one decimal point, and no one is going to say that anything is 100% accurate. What if it was 99.99% accurate? Would I tolerate, say, a 1 in 10,000 error rate with an efficient system in place to confirm an individual's identify once flagged? Yes. 1 in 1000? No.

Even one in 10,000 would have 400 false positives per person on the list per year for each person on the list, coming in from England alone.

Going beyond this, of course, is the question of what DHS does with the data once they've gathered it. For example, now that you know that this person is not on the suspicious-person list, and you know it belongs to the person with the passport named "Sir Raymond Luxury-Yacht" -- what then? What purpose does it serve to store Sir Raymond's fingerprints on a hard drive somewhere outside of Washington DC, or to give that information to MI5? The potential for abuse -- the potential downside -- is obvious, but I have a rather difficult time seeing hte upside. It will create a records-management nightmare, make it that much more likely to find false positives once the new fingers go into the database, and do nothing to further the ostensible purpose of fighting terrorism.

That gets back to the "believing in anything that I'm told will make me safer" issue. If this will make you safer -- and the benefits outweigh the costs -- then of course the program should be created. But as far as I can see, the additional workload that results from this sort of program will actively reduce the effectiveness of the US's anti-terrorism efforts. Why do you believe that this program will be beneficial?

 

[drkitten]

File this under "Be Careful What You Ask For".

Post-9/11, it was, "Why didn't Bush act on the terrorist threat?"

After the next terrorist attack, it will be, "Why didn't Bush (or whoever is President) act on the terrorist threat?" And the answer will be, "I was busy poring through 11 million Verizon phone records and chasing down people who bought the Koran on Amazon.com."

Um, yes. That would be kind of my point, yes.

 

[boooeee]

If fingerprint identification is 99.9% accurate, that means that one person in a thousand will be mis-identified as any given suspect. With four million visitors from the UK alone trying to enter the country, this means that you'll have to investigate four thousand people for every person on the terrorist watch list, each year. How much time and money is DHS budgeting for this kind of wild goose chase? Can you think of something else that you would want to do with that kind of money, something that might perhaps be more effective?

Is that necessarily the case? I couldn't determine that from the link.

Does a 99.9% accuracy rate mean that one in a thousand will be identified as a terrorist? Or that one in a thousand will be mis-identified? In other words, they read my fingerprint and mis-identify me as drkitten. However, since drkitten isn't on a terrorist watchlist (I'm assuming), then I can fly without a problem.

I'm not completely sure how the system works, but it doesn't seem you can take the 99.9% success rate and directly apply that to the situation being discussed.

 

[Polaris]

I'm imagining a terrorist completely stumped by the email address requirement.

"Dammit, Hajib, I gave them my email address that I use to communicate with AQHQ! We're f-ed!"

I tell ya, the gubmint is brilliant.

As for the fingerprint requirement, I am with you all the way, Godmode. Maybe if enough tourists decide it just isn't worth visiting the US any more, the Schutzstaffel, er, I mean, the Department of Homeland Security will get wise and back off.

That's pretty low.

 

[WildCat]

You know, way back in 1992 I went to Canada w/ some friends. Canadian customs took us out of our car and took us to separate rooms where we were asked about how much money we had on us, what credit cards we had, what we were planning on doing in Canada, where we were staying, etc. All while they they searched every square inch of the car and our luggage. I was expecting the strip search and body cavity search next, but they stopped just short of that.

Never been back to Canada, nor do I want to.

 

[drkitten]

Is that necessarily the case? I couldn't determine that from the link.

No, it's not. But since the 99.9% accuracy figure is rather questionable anyway, I just went with it to make a point. The definition of "accuracy" is actually a rather difficult one that the people in information retrieval hate to discuss, precisely because it ends up being half the paper.

However, my central point stands. Police, throughout history, have generally avoided doing this kind of blanket investigation where they test everyone from a very large group of people in whom a suspect might be present, for the very good reason that the number of false positives almost invariably creates an unwieldy number of leads that need to be investigated. In the case of fingerprint identification -- had we good enough data about the exact measure they're using -- we could even quantify the number of false positives we would expect, and given the number of people coming into the USA each year, the number of false positives will be huge. (If fingerprint technology were that good, why has DNA technology been such a revolution in criminal prosecution?)

As I said earlier, the potential for abuse is obvious. But the benefit to law enforcement of collecting that much hay to pack around each needle is (ahem) questionable.

 

[BPSCG]

"Hey, we got good fingerprints from Ahmad ("The Camel") Nasir's house. You know, the guy who's responsible for those bombings in Milan, Frankfurt, Rio, and Tokyo? We got two index fingers, two middle fingers, and his left ring finger!"

"That's great! Now airport security will be able to nail him when he tries to get into the U.S., using the two thumbprints they get from visitors..."

 

[drkitten]

"Hey, we got good fingerprints from Ahmad ("The Camel") Nasir's house. You know, the guy who's responsible for those bombings in Milan, Frankfurt, Rio, and Tokyo? We got two index fingers, two middle fingers, and his left ring finger!"

"That's great! Now airport security will be able to nail him when he tries to get into the U.S., using the two thumbprints they get from visitors..."

You think that Ahmad "The Camel" Nasir doesn't know how to fool a fingerprint detector?

 

[DanishDynamite]

In the good old days of this forum, right-wingers used to throw the following Ben Franklin quote at me as a magic charm, especially IIRC during gun control debates. I think I can fairly throw it back at this time:

"The man who trades freedom for security does not deserve nor will he ever receive either."

 

[BPSCG]

Hmmm..... Thousands and thousands of flight students across the country, but this particualr group of students is not interested in learning about landings. Suspicious?

That's not the point. You, and others here, seem to be under some kind of impression that we should observe only the bad guys. You can't tell if someone is a bad guy or a good guy until you observe him.

How do you expect to get the personal information you need about someone to determine he's a bad guy without observing him and collecting information about him?

You (and others here) seem terribly afraid that the government in general and Bush in particular are going to try to turn this country into a police state, blithely ignoring the fact that we enjoy as much freedom here as in just about any other country, and more than in most. Tony doesn't appreciate the irony of his claim that we're losing our freedoms while he calls the president "white trash." People here openly and seriously call for Bush's impeachment; do you honestly think you could do that in a genuine police state? Please.

Frankly, I'm a lot more scared of what terrorists want to do to me than of what my government supposedly wants to do to me.

Basically, by good old-fashioned police work.

...while our enemies use our modern technology to try to destroy us. You seem to think we're trying to fight some kind of organized crime syndicate, rather than an enemy who has literally declared war on us and vowed our destruction.

"Mister President, there's been a sneak air attack against American territory. Thousands of Americans are dead. What are your orders, sir?"

"Good old-fashioned police work. We'll find who did this and punish them. Even if it means we have to prosecute General Tojo himself."

 

[Katana]

Then the source that I cited is wrong. It woudn't be the first time that a reporter got a number wrong. But at least we've now got a number that we can discuss instead of some abstraction.

Fair enough.

Even one in 10,000 would have 400 false positives per person on the list per year for each person on the list, coming in from England alone.

I don't consider that an unmanageable burden, again with many of them being likely easy to clear up and some being false negatives as boooeee pointed out.

Going beyond this, of course, is the question of what DHS does with the data once they've gathered it. For example, now that you know that this person is not on the suspicious-person list, and you know it belongs to the person with the passport named "Sir Raymond Luxury-Yacht" -- what then? What purpose does it serve to store Sir Raymond's fingerprints on a hard drive somewhere outside of Washington DC, or to give that information to MI5? The potential for abuse -- the potential downside -- is obvious, but I have a rather difficult time seeing hte upside. It will create a records-management nightmare, make it that much more likely to find false positives once the new fingers go into the database, and do nothing to further the ostensible purpose of fighting terrorism.

What purpose would it serve to have an "innocent" person's prints on file? Well, are there applications that could protect against identity theft? What if someone using your information was scanned and didn't match?

Having said that, what bothers me more about this is that it isn't being used on everyone flying. It could identify many people (in addition to potential terrorists) that would be better taken into custody and could potentially uncover cases of identify theft, especially if broadened to include Americans. Given our large illegal immigrant population and issues around identity theft, I would support this, again if it was sufficiently accurate. Of course, this would require that we all be the first ones to submit prints so that anyone else using our identity would be compared to ours. Hmm.

That gets back to the "believing in anything that I'm told will make me safer" issue. If this will make you safer -- and the benefits outweigh the costs -- then of course the program should be created. But as far as I can see, the additional workload that results from this sort of program will actively reduce the effectiveness of the US's anti-terrorism efforts. Why do you believe that this program will be beneficial?

As for whether I believe that it will be beneficial, I believe that it has the potential for great benefit. I'm not quite as convinced as you that the extra workload will be of sufficient magnitude to undermine completely the program's benefits, but you absolutely raise valid concerns.

My irritation stemmed from the generalization that anyone supporting any component of a program such as this must be all for everything done in the name of security. I'm not one such a person. I appreciate your concerns, I share some, and I have a bit more perspective because we're actually exchanging thoughts rather than insults. Having said that, I also see the potential in this with the caveats I mentioned.

I just wish that discussions about these matters didn't always do such a quick free-fall into name-calling and generalizations about what all Americans or all non-Americans think, feel, or do.

 

[BPSCG]

You think that Ahmad "The Camel" Nasir doesn't know how to fool a fingerprint detector?

You think that technology that doesn't work perfectly 100% of the time should be discarded?

 

[BPSCG]

I just wish that discussions about these matters didn't always do such a quick free-fall into name-calling and generalizations about what all Americans or all non-Americans think, feel, or do.

Good post, even if you are traitorous slime (*ducks, runs...*)

 

[Katana]

Good post, even if you are traitorous slime (*ducks, runs...*)

Damn. Wait, Beeps! I take it back.

No, wait.

I...can't...do...it.

Damn.

Forgiiiivvve meeee!

 

[Region Rat]

Katana, under what circumstances would you no longer support, for example, requiring people to surrender their fingerprints, banking records, and email when they entered the country?

BPSCG is correct that such information, if properly used, could be of tremendous help in an anti-terrorism investigation -- or a criminal investigation generally. The problem, however, is that at some level of (in)accuracy, the amount of work it creates to deal with the haystack of false findings will actually take manpower away from whatever needles might have been turned up.

For example, here's some data on the accuracy of fingerprint identification:



Obviously, this will change somewhat as technology changes -- but let's look at this in more detail. If fingerprint identification is 99.9% accurate, that means that one person in a thousand will be mis-identified as any given suspect. With four million visitors from the UK alone trying to enter the country, this means that you'll have to investigate four thousand people for every person on the terrorist watch list, each year. How much time and money is DHS budgeting for this kind of wild goose chase? Can you think of something else that you would want to do with that kind of money, something that might perhaps be more effective?

drkitten,
Do you take more issue with the collecting of the fingerprints, or the accuracy of the data once taken? If the data was considered correct, would it still not be worth the effort?

What about retinal scanning?

A retinal scan involves the use of a low-intensity light source and coupler that are used to read the blood vessel patterns, producing very accurate biometric data. It has the highest crossover accuracy of any of the biometric collectors, estimated to be in the order of 1:10,000,000.

Would you have a problem with having your retinal scan on record, as a way of proving that 'you' are in fact 'you'? If the method was considered foolproof (1:10,000,000), would you support identifying US citizens and those wishing to enter in this way, or would you consider it an infringement on personal privacy?