Presidents being shot. Planes flying into buildings. A nuclear weapon or any bomb being detonated inside of the country are all things that I do not want to happen. I don't know how to prevent these without SIGINT dominance.

If these events can be prevented without the chance of any human using the collected data for any other purpose and the data only being stored for a very short period of time, then I'd have to say I'm on the side of the NSA.

I think that corporations and law-enforcement agencies using intelligence and mass collection techniques are much more of a threat.

Yes, the sharing and data-retention of the NSA should be greatly limited and controlled and monitored in my opinion, but eliminating their ability to protect the country (their ultimate goal) might not be worth it.

The result of the "base rate fallacy" in the context of the bulk data collection program is that the false positives will swamp the true positives in any general search results from the data-set.

One of the thing that always bugs me is that whenever privacy/security issues come up is that everyone seems to assume that the authorities are naturally going to gravitate to the least efficient, most invasive means of carrying out their job, simultaneously implying that it's preferable to have (in this example) 100 terrorist operating amongst us and not do anything than to put our heads together and propose an intelligent solution.

Have you flown in the past 10 years? Shuffling through TSA checkpoints in my socks makes me feel like authorities do gravitate to stupid measures.

I agree about the TSA checkpoints, though. There are plenty of examples on both sides of the arguments. My point is that the general assumption seems to gravitate to "they'll implement it stupidly, so they shouldn't have that capability at all". It's not necessarily a binary choice between no security and stupid implementation.

There, fixed that for you. Now, I'll admit, a large part of the time the owner is also the driver, but with speed cameras it is not the "driver" who receives the fine, but the _owner of the vehicle_.

The one difference with "speed cameras" and the NSA's data mining is that measurement of the rate of travel of a vehicle (assuming a properly calibrated and non-malfunctioning speed measurement device) is so close to 100% accurate as to be considered 100% accurate.

No "connect the dots, infer here, infer there" data mining in the NSA's dataset will ever be even within several orders of magnitude of accurate as compared to a properly functioning, properly calibrated, speed measurement device.

And that underlying, non zero, error rate is what triggers the base rate fallacy issue in statistical analysis. When an occurrence of X is extremely rare in a very large population without X, then the test for X has to be impossibly accurate for the results to not be swamped by false positives. And this is a completely counter intuitive result, which is why it is so often overlooked.

Pertinent quote from the Shiner article at https://www.schneier.com/blog/archives/2006/03/data_mining_f...:

"Let's look at some numbers. We'll be optimistic. We'll assume the system has a 1 in 100 false positive rate (99% accurate), and a 1 in 1,000 false negative rate (99.9% accurate).

Assume one trillion possible indicators to sift through: that's about ten events -- e-mails, phone calls, purchases, web surfings, whatever -- per person in the U.S. per day. Also assume that 10 of them are actually terrorists plotting.

This unrealistically-accurate system will generate one billion false alarms for every real terrorist plot it uncovers. Every day of every year, the police will have to investigate 27 million potential plots in order to find the one real terrorist plot per month. Raise that false-positive accuracy to an absurd 99.9999% and you're still chasing 2,750 false alarms per day -- but that will inevitably raise your false negatives, and you're going to miss some of those ten real plots.

This isn't anything new. In statistics, it's called the "base rate fallacy," and it applies in other domains as well. For example, even highly accurate medical tests are useless as diagnostic tools if the incidence of the disease is rare in the general population. Terrorist attacks are also rare, any "test" is going to result in an endless stream of false alarms."

Exactly as it should be. It's your responsibility as a vehicle owner to not let bad drivers use your vehicle.

RE base rate fallacy - true, but I guess that's why they want even more data. A single test may swamp you with false positives, but positive results from many different sources can correlate into useful info.

I read the fine print in everything that I sign. I never noticed any such language in either the state driving regulations, or the restrictions imposed by my automobile insurer. I did notice obligations to operate one's vehicle in a safe and prudent manner at all times.

This also doesn't match with real-world numbers that we know on the NSA metadata program. Schneier was saying immediately after 9/11, the FBI was getting thousands of tips per month, which may very well have been the case back then. But the Presidential Review Group report stated that in 2012 a total of 12 tips were passed to the FBI[1]. To me, that indicates that they've become much better at getting rid false positives.

Schneier's hypothetical program also isn't analogous because it starts by scanning everyone to determine if one is or is not a terrorist, but the actual NSA metadata program (according to Review Group, same article cited above) starts out by examining people the NSA has already identified as terrorists through other means - the controversy isn't in the accuracy rate, it's in the method of collection.

[1] http://www.huffingtonpost.com/geoffrey-r-stone/nsa-meta-data...

I don't think it's necessary, but if they didn't require that then people would be blaming them for not taking enough action.

The times I've flown it hasn't been too much of a hassle, though I suppose it's highly dependent on the airports you pass through and whether or not it's an international flight coming into the US.

Schneier has written on this problem a number of times.

I suspect this problem may be caused by the well-known problems many people have with Bayes Theorem and how to properly use it.

Just look at the Ford Pinto. It was calculated that it would be cheaper to pay the families of victims than it was to fix the exploding gas tank.

You don't mean this. If you did, you would support mandatory exercise regimens and diet monitoring, and aggressive eugenics programs that include mandatory sterilization of folks at very high risk for nonviable offspring.

> [Pinto anecdote]

The facts seem to disagree with you. Read page eight (listed as page 1020 in the text) of this PDF: http://www.pointoflaw.com/articles/The_Myth_of_the_Ford_Pint...

You can't compare dying in a terrorist attack to the personal choice of not exercising or eating unhealthy.

On page 1037 of the Schwartz Paper you linked to it stated that the California Supreme Court's opinion was that it actually encouraged auto manufacturers to consider the safety trade-offs in the sake of cost. I fundamentally disagree with this.

I personally think that consumer responsibility to research every safety aspect of every product they purchase is bullshit. There should be no corners cut when it comes to safety and if there are then it should be made known to the customer before purchasing the product.

The majority of people and decision makers disagree with me, but that's the great thing about this country and the internet, freedom to disagree and participate in open discussion.

Indeed - after all, the risk from a terrorist attack is so many orders of magnitude less likely we can often ignore it completely. It's noise, and deserves little special preparation.

On the other hand, heart disease and other diet related problems are one of the largest risks faced.

I should have the choice to decide whether or not I am unhealthy, it doesn't affect anyone but myself (and my family if they are dependent on me).

The government should not have the choice on whether or not to protect citizens, that's their job, citizens are depending on them to perform that action.

While I know people would be healthier and "better off" by eating healthy, I have no right to force people to eat a certain way.

If I want to smoke cigarettes, drink alcohol, or smoke marijuana, that should be my choice, not yours to make for me, as long as it doesn't affect you.

If I want to ask the government or a company to delete all information they have about me, I should have the right to do that, just my opinion of course.

> The government should not have the choice on whether or not to protect citizens, that's their job, citizens are depending on them to perform that action.

And this:

> If I want to smoke cigarettes, drink alcohol, or smoke marijuana, that should be my choice, not yours to make for me, as long as it doesn't affect you.

Makes perfect sense to me, do you still think that they are mutually exclusive statements?

What's that - you still want the freedom to drive? Even though it involves a fairly significant (0.01%) risk of dying because of things not under your control? Well, some people want the freedom to not be monitored all the time, even though it involves a much smaller than 0.01% risk of dying because of a terrorist attack. And all people should, and would, if terrorism didn't capture the imagination so vividly.

Maybe you're right, maybe we're both partially correct, it's a complicated issue and I do appreciate the discussion.

You also have no right to go through their private communications unless you have a direct probable cause. (IANAL, the exact terms are probably different, but you get the idea)

Except that the chances of each occurring are wildly different.

Chance of dying from terrorist attack in the US from 2007 to 2011: 1 in 20 million [1].

Chance of dying from choking from inhalation and ingestion of food: 1 in 3,649 [2].

20,000,000 / 3,649 => 5,480.

You are five thousand eight hundred forty times more likely to die from choking on your breakfast/lunch/dinner than to die from a terrorist attack. The difference is so extreme that you'd be better off ignoring the terrorist attack threat and instead installing a feeding tube so you no longer need to chew and swallow your food. You'd statistically live longer that way than worrying about protecting yourself from terrorist threats.

[1]http://swampland.time.com/2013/05/06/chances-of-dying-in-a-t...

[2] http://www.nsc.org/news_resources/injury_and_death_statistic... linked from here: http://www.nsc.org/news_resources/injury_and_death_statistic...

Chaotic events happen. People are terrified of them but having billions of dollars worth of surveillance state does not reduce the risk one iota.

I don't think they can be prevented with "SIGINT dominance" either.

But more importantly, to me: the idea of consitutional and human rights is that it was agreed, roughtly, that "these things are required to live a dignified, free life". Not literally, but I do think that is the rough idea. So taking away from that is akin to killing everybody. It's not even burning the village to save the village, it's burning the whole village to save a tiny part of it.

In a follow-upp comment you said "when we are talking about peoples lives, it's worth saving as many as possible in my opinion" and I agree. For example, stop mass surveillance, stop selling weapons like candy, persecute war criminals (I mean the ones having a ball and getting invited to gala dinners on home soil, not the ones around the globe in areas of geostrategic interest), that sort of stuff. Otherwise, I'll stick with believing a lot of "security" stuff is just lip service, a way to transfer tax payer money into private pockets without achieving anything other than more control of the domestic population. I'm not talking about cogs in the machine, mind you, but the machine itself. I don't doubt many people in the NSA or elsewhere have good intentions, I just think that's worth very little in the bigger scheme of things.

This simply can't/won't happen. Human nature dictates that this data can't be held and dealt with responsibly.

But I do disagree.

Do not assume mass surveillance can happen "without the chance of any human using the collected data for any other purpose". That's fairy tales. Look up "Loveint",think about everyone you ever dated and their exes. Commercial use is not unheard of, either. Even if - if! - nothing much darker goes on yet, what would stop corruption in the future? That's a much greater threat than terrorism.