Hacker News Reader

Comment by notthemessiah | Hacker News Reader

Reminds me of another bombing incident, where the FBI was warned by the Russian FSB that the Boston Bombers, the Tsarnaevs, had associated with Islamist militant extremists: https://www.reuters.com/article/us-usa-explosions-boston-con...

Yet, millions of dollars were being poured into "Fusion Centers", efforts to coordinate activities between intelligence agencies and local law enforcement: what were they doing? Turns out they were busy monitoring Occupy Boston, same as Fusion Centers across the country: http://www.justiceonline.org/unaware_of_tsarnaev_warnings

They always ask for more surveillance funding/powers, yet they never have a good answer when it comes to what they do when they're granted such powers.

View on HN

I sympathize with the goal of not giving law enforcement more surveillance powers, but I wonder how valid the anecdotal "connecting the dots" stories of law enforcement failures are.

For example, it's conceivable that law enforcement or intelligence agencies regularly get a huge volume of urgent tips, 90% of which are totally incorrect or useless, and that they nonetheless manage to stop 95% of super-bad acts like terror attacks. Then the public hears about the 5% that they don't stop, and especially hears about the 0.5%, say, where a probably-unreliable tip was not successfully used to prevent a super-bad act.

In order to argue convincingly that agencies are actually bad at using the powers or information that they already have, we would probably want to know a lot more about base rates (and about thwarted or deterred crimes).

Edit: This comment downthread https://news.ycombinator.com/item?id=25579252 just before mine makes a somewhat similar point (although I was emphasizing the idea that we should want more data for this argument, while the other comment emphasizes the idea that there are more fundamental or important grounds on which to oppose expansions of surveillance powers).

The problem (luckily!) is, that the thing they are trying to detect happens extremely rare. And when you try to detect rare events, even if your methods are 99.9% accurate you will drown in false positives.

I once did the maths on facial recognition tech which was meant to recognize terrorists at Berlin main station. With the known number of commuters there and an accuracy of 99.9% (which is extremely optimistic) they would have over 600 false positives per terrorist if I remember correctly. The system they had installed there however wont even reach 80%...

Imagine you have an alarm that rings twice every day, and is real and serious only once a year. How likely are you to react in time?

I think this is the fundamental problem with the "we need more data and more rights"-strategy of most police and law enforcement.

I am convinced what they actually need is more personel with a good education.

Edit: what this kind of data is better at recognizing are broad movements in society like demonstrations, protests and such things.

> Edit: what this kind of data is better at recognizing are broad movements in society like demonstrations, protests and such things.

Yes, this is their real purpose. They're not rolling out these wide ranging systems at extraordinary expense when they're at all unsure they will actually solve the problem they're trying to solve.

It’s probably a good thing that vague accusations from an upset ex aren’t enough to have all of your property searched by the police.

about that 95% success rate of law enforcement

https://miami.cbslocal.com/2017/11/09/tsa-fails-bomb-weapons...

2017: "more than 70 percent of the time, undercover Department of Homeland Security investigators were able to get through TSA checkpoints with mock knives, guns, and explosives.

Just two years ago, testing found a 95 percent failure rate."

Most TSA employees are actually not law enforcement officers, if I understand this page correctly (and if it's correct): https://en.wikipedia.org/wiki/Transportation_Security_Admini...

EDIT: see this: https://www.tsa.gov/blog/2016/07/03/tsa-myth-busters-do-tsa-...

>They always ask for more surveillance funding/powers, yet they never have a good answer when it comes to what they do when they're granted such powers.

Their remit is to protect the state, not you. Occupy movements / class consciousness pose a real threat to the state and therefore is of far higher concern than some lunatics bombing some plebs doing a marathon.

If Warner's girlfriend had told them he was planning to use the bomb to take out a major node in their mass surveillance infrastructure one imagines they might have pursued the investigation a little further.

Makes me wonder if most of the "give us more surveillance power" claim is nothing more than a convenient excuse surveillance apparatus give to "excuse" their lack of result.(and sometimes failures) Probably starting somewhere from the bottom/middle managers excuse, then that excuse being used by above manager, rinse and repeat to the top. Then politician with different incentive(voter satisfaction) granting that power just for their own game.

"oh if you just gave us the power by unshackling us from regulation, we could've done better! I promise, it's not related to our lack of efficiency or ingenuity!"

The FBI seems to have been caught lying about not having heard of Warner before the bombing: https://twitter.com/onekade/status/1344154240189661185

No, surely the FBI wouldn't lie. That's nothing like them :)

The only war you’re really good at fighting is the last one.

I don’t know much about the Fusion Centers, but post-9/11, the failure of intelligence and law enforcement to communicate was regarded as a major reason why the plot was never uncovered. In totality, there was a lot of signs something was coming but different organizations had different pieces so nobody could see the “big picture”.

So I’m just wondering if the focus on Fusion Centers and not on intelligence from foreign powers is just our gov’t “skating after the puck” instead of “skating where the puck will be”.

The military and security services have deployed vast, illegal domestic spying networks, on the basis that if USians give up their privacy, they will be able to exchange it for safety.

Here we see that not only is that untrue, but also that even with tips they can't seem to get basic, non-surveillance policework right.

Remind me again why we tolerate these abuses in the name of safety? We and they now know they aren't being used for safety, if indeed that was even possible.

What else could and will these massive surveillance powers be used for?

Providing “jobs” and enriching executives basically.

It is important to keep in mind that when security is working, you don't see any evidence of it. We don't actually know how many terrorist attacks have been stopped by this surveillance. It may well be a lot. It wouldn't surprise me at all to learn that many have been. Which is why it's so important that we do not ground criticisms of surveillance in arguments about efficacy. Surveillance is effective, or at least, it can be. Surveillance is wrong because it concentrates too much power in the hands of the government.

We don't actually know how many terrorist attacks have been stopped by this surveillance. It may well be a lot.

Actually, we do, because the FBI issues press releases when they succeed.

Here's a relevant paper from an FBI study of "lone offender terrorism" [1] This guy fits the profile - white, some college, no major criminal history, not employed, marijuana user.

[1] https://www.fbi.gov/file-repository/lone-offender-terrorism-...

This is a list of people who successfully carried out attacks, it looks like. Not people who's attacks were foiled.

Because that second list would be empty.

They would be shouting it from the rooftops if they were actually stopping attacks.

Domestic surveillance is going to tell you about Americans on American soil. Are you proposing that people in this category are getting quietly disappeared? To my knowledge we have to either charge these people with terrorism (definitely news), or wait until they’re abroad and do a drone strike (not viable with an imminent attack in the US). And the latter strategy was a huge deal the one time it’s known to have been done intentionally (al-Awlaki).

There are tons of prosecutions of this sort. Consider the recent plot to kidnap Governor Whitmer as just one example among dozens.

According to the feds, they had CIs inside that organization.

No illegal mass surveillance necessary for that, unless the CIs being there were illegal parallel construction in the first place.

> To my knowledge we have to either charge these people with terrorism (definitely news), or wait until they’re abroad and do a drone strike (not viable with an imminent attack in the US).

Your knowledge is incorrect.

Insofar as a person is an enemy combatant in law, they can be militarily detained without charges regardless of where they are caught and what their citizenship is (and, if they are charged and tried, it can be by military commission), see, Ex Part Quirin, 317 US 1 (1942), concerning exactly this question with regard to German saboteurs in WWII, two of which were US citizens and all of whom were captured on US soil.

Also, insofar as they are enemy combatants, they may be engaged by the military with force anywhere in the world, regardless of citizenship and location. It would be silly if the military couldn't fight active enemies when they were in US territory but had to wait for them to leave, regardless of citizenship (for one thing, that would have made fighting the Civil War difficult for the Union.)

A familiar situation to anyone who worked in a large organization of any kind. What’s really needed on the ground and what gets funded by the top are often only tangentially related.

Reply on news.ycombinator.com