Hacker News Reader

Comment by ikeyany | Hacker News Reader

Where is the conspiracy? I thought the US government had been surveiling its cities via aerial technology. Is the conspiracy that these are unsecured?

View on HN

I don't think people have internalized the fact that the US gov has surveillance drones in holding patterns over most US metropolitan areas. Every time I've seen someone poke around one of those persistence ADSB/radar maps, they have gone through the stages of discovery: "What are those circles?" "Surveillance drones." "Haha, no, really, what are they? Planes circling the airport?" "No, that (points to thin circle) is a plane circling the airport. The heavy circles are surveillance drones. Really." "Uhh.... (shifts uncomfortably) so where do you want to go for lunch today?"

Have an example of this? Would be interesting to see it.

Seconded.

BuzzFeed News had a big feature and analysis here: https://www.buzzfeednews.com/article/peteraldhous/spies-in-t...

This isn't about drones, but here is an article from last year about the DoD using surveillance balloons that use radar from 65,000 ft to generate high-res images of the ground:

https://www.businessinsider.com/us-government-testing-survei...

It's so funny how you can whip up conspiracy theories just by adding superfluous adjectives. Instead of just saying "radar" you say "wireless radar device", as if prior radars used wires, and then you say it creates "high-resolution images using radio pulses", as if this did not apply to every radar ever built.

We have been using aircraft to track vehicles since the invention of aircraft. The people and their government have a legitimate interest in doing so. Nobody has a right to go about in a vehicle anonymously and privately.

https://www.modbee.com/news/local/news-columns-blogs/bee-inv...

Where did I say there was no legitimate interest to do this? I am in agreement with you in saying this surveillance is not new and is not a conspiracy.

It's possible for something to be not new and not a conspiracy but also not good and not right, all at the same time.

Then it should have been posted in technology, security, or news subreddits.

We shouldn't expect the centuries old philosophical/legal/procedural balance we've crafted between investigatory powers and personal rights to remain valid and correct in the face of a 99.999% decrease in the cost of surveillance and corresponding explosion in scope.

Naturally, authorities didn't object to the technological expansion of their power. It's up to us to figure out if/when/how to push back. Step zero is admitting that it actually happened, admitting that we should reset our heuristics, and calling out the ludicrous argument "this is how it's always been," because, well it isn't.

Not a map but it being done openly in Baltimore.

https://www.baltimoresun.com/maryland/baltimore-city/bs-md-c...

This book is the best source on the topic: https://www.amazon.com/Eyes-Sky-Secret-Gorgon-Stare/dp/05449...

There are twitter bot accounts that post live when they detect aircraft flying in circles:

@SkyCirclesSF @SkyCirclesLA @SkyCirclesDC

I don't think people really think about it.

Another example: most cities have systems that can pinpoint a gunshot to a block radius anywhere in the city so that police can respond quickly.

It's funny. I remember a court case where Shotspotter had a pinprick on one block and the news said "the shooting actually occurred a block away". Haha, that's nuts. Crazy accurate.

The expert testimony did have the engineer saying "The accuracy is invented by sales and marketing. The pinpoint is just a starting point. It's in a radius around that." Which I found amusing because it seemed obvious to me.

It did improve my ability to sell, though. These caveats are useless when you're selling. You can always explain afterwards.

Oakland has shotspotter and my understanding is that it's been considered a failure

Really? Why?

In DC, I think it's been pretty successful because lots of gunshots do not result in police calls.

Looking into it more it's less clear than I remembered. OPD considered scrapping it in 2014 saying the costs weren't justified and they'd rather have more money for their helicopter. Other cities have said they're not confident in its effectiveness but that doesn't seem to be a large factor here. They ended up keeping it.

https://www.sfgate.com/crime/article/Oakland-cops-aim-to-scr...

This is an interesting system. I have worked with this technology on large and small scales, and it's a lot less accurate than you'd think. You have no hope of sending a police cruiser to check out a single gunshot based on the outputs of this system. But a single gunshot won't generate a police call in many cities as it is, so no change there. But a gang war? Sure, that will get the phone lines going and also generate many data points to help pinpoint the location using this system. So many factors affect the speed of sound through air, that the problem becomes one of those logarithmic "How many more microphones do you want to install to eke out this tiny bit of accuracy?" kinds of things. I'm not sure it's useful to point to what cities are saying about it, either. The technology itself is a Hard Problem, but I'm also 100% sure many law enforcement bureaucrats will use any excuse to keep money flowing into their budgets, even if the money is earmarked for something that doesn't work (or doesn't work well).

Hm, I'm not so sure. Take this for instance: https://en.wikipedia.org/wiki/Murder_of_Seth_Rich#Shooting_a...

Officers were able to locate and respond within a minute because of a system like this.

There is a very big difference between a system that can localize gunshot sounds and surveillance drones.

Assuming, of course, that the gunshot detectors are in fact special gunshot detectors and not a giant array of normal microphones with some algorithms applied to them.

> Assuming, of course, that the gunshot detectors are in fact special gunshot detectors and not a giant array of normal microphones with some algorithms applied to them.

Why is your statement based at all on that assumption?

I don't see what makes surveillance drones necessarily worse than human cops flying helicopters overhead with cameras (which I know for a fact occurs in my city). What makes the "drone" part problematic?

Scale, oversight. Would you mind if we had one cop per person and followed them around in public at 20 foot distance everywhere they went?

I mean, a human in a helicopter is NOT the same as Gorgon Stare[0]. That's what's problematic, to me, anyhow.

[0]: https://en.wikipedia.org/wiki/Gorgon_Stare

But if a human is piloting a helipcopter with a "gorgon stare"-esque technology attached, seems pretty much the same.

I guess I'm unsure what about the autonomous flying is what draws people ire, rather than just overhead surveillance as a concept.

Looks like they have an app for law enforcement. It does provide some audio.

>gunshot audio to gain situational awareness, and; precise location of the gunshot to search for evidence.

https://apps.apple.com/us/app/shotspotter-respond/id10968200...

Some thoughts on pole surveillance:

https://nccriminallaw.sog.unc.edu/pole-camera-surveillance-f...

It's not 100% consistent, but part of it is a reasonable expectation of privacy, compared to what the public would normally see. E.g. having a camera up in the air is not what the public normally sees. Which in some cases violated suspects' rights.

That means LE should have warrants for these cameras. Plus, they're not allowed to go on fishing expeditions.... These cameras aren't designed to only look at one place, but a wide area... Which means public oversight is required to keep LE honest. If there are no public records, then it's unreasonable search.

Your comment isn't totally true in the narrow sense that it's not being done domestically by the USG specifically (at least at the time I was reading about it).

The best book on the topic: https://www.amazon.com/Eyes-Sky-Secret-Gorgon-Stare/dp/05449...

The primary domestic use was by the Baltimore police department and it was supplied by an individual billionaire interested in funding the project.

Interesting fact in that book was that the initial spark for the project was someone in government watching Enemy of the State in the 90s and thinking "we should be building this". Somewhat comically the guy who worked on the camera effects in the movie also has a defense company and was asked to help build the real thing.

True if you don't count JLENS, and ARGUS, which have been deployed wherever they feel like for years. Further muddied is that there are federal grants to let local law enforcement buy this gear.

https://www.theatlantic.com/politics/archive/2016/03/the-rap...

https://www.baltimoresun.com/maryland/harford/aegis/bs-md-jl...

Always enjoyed that movie. I suggest everyone check it out!

> comically

Synergy!

IANAL. It's also surveying something openly visible from public space, so even distributing it accidentally isn't illegal, per se. Now, it's obviously fucked, and abuses and mistakes like this should make public officials think twice about building a surveillance state. My main hope is news outlets that pick this up tie it to the encryption debate.

Also worth noting that just about every redlight intersection has cameras mounted on top pointing at traffic in each direction.

It's pretty much impossible to navigate through a city anonymously in a vehicle.

Now with facial recognition software getting so advanced I'd water it's impossible to walk through a city anonymously as well.

They don't necessarily need aerial imagery when every intersection has HD cameras pointing at you.

They’re not cameras, they are metal/RF/infrared/whatever detectors used to detect vehicles waiting at the intersection so lights don’t stop traffic for no reason.

Source: I’ve had to call municipal governments to get the detectors rotated or angled properly due to them not detecting cars and therefore not changing the light to green.

Of course it’s technically possible they also contain cameras, but I have yet to see any proof.

Yes, optical detectors are often used for traffic control.

Traffic cameras which stream visible-light video are also very common. https://www.weatherbug.com/traffic-cam/

I'm not sure why you've been downvoted. Traffic-aware traffic lights are generally implemented using induction loops built into the road surface: https://en.wikipedia.org/wiki/Induction_loop

Today, many have cameras for other reasons. But this is not ubiquitous and the prevalence of these cameras varies from location to location. Many traffic lights don't have cameras (many don't even have induction loops, and work off timers instead.)

In areas where the ground is covered in a foot of snow and ice all year those aren't particularly effective and they do use devices that look kinda like cameras and everyone assumes are cameras.

They're actually just small radar devices. It's a much simpler problem to solve across a wide variety of conditions to simply point radar at the ground and when the distance suddenly drops 4-8' assume a vehicle is there.

These devices don't have any sort of lens assembly which should be your first clue, but they also are pretty easily differentiated if you own a radar detector as well...

Traffic signal cameras for traffic light control are common. They used to be dumb monochrome analog cameras connected to simple "non-pavement object detected in box" processing units. There's been considerable mission creep since. Here's the promo video from Econolite's current product.[1] HTDV, WiFi, car, truck, bus, bicycle and pedestrian detection, connects to control center if the bandwidth is available.

CALTRANS has most of their highway cameras available from their web site.[2][3] CALTRANS has been at this since the 1980s, and the cameras are mostly somewhat old and low-rez.

[1] https://www.econolite.com/products/detection/autoscope-visio...

[2] http://cwwp2.dot.ca.gov/vm/streamlist.htm

[3] http://cwwp2.dot.ca.gov/vm/iframemap.htm

How often is this done? Do they also detect if a bike is at the intersection? Will they eventually cycle the light even if nothing is detected (just in case the sensor is not working)?

They are at almost every intersection in my state. I’ve heard people in bikes complaining about having hard time getting detected sometimes. I suppose it depends on how the light is programmed if it has a max time before it cycles, but I’ve had to go through intersections on a red light because I waited minutes and it didn’t turn green.

Video-based lane occupancy detection is widely used by traffic lights, but it's rare for these cameras to be connected to anything other than a trigger module in the signal cabinet. It is conceptually possible to also connect these cameras to a remote monitoring system, but this comes at a very high cost. My city has been one of the stronger proponents of doing so but has only connected a very limited number of intersections due to the expense.

The city (major suburb) I used to live in has done this with all the intersections that lead into the city, effectively making it impossible to enter or leave the city without them having a record of it. I know they are able to scan license plates from these cameras -- and do so automatically -- because it's how they are able to "catch" people trying to use those 3M strobe lights that ambulances and fire trucks use to make the lights turn green: An intersection detects the use of the strobe, but also is able to know that no (deployed) emergency vehicle is near that intersection, so it automatically "calls" the police department with the license plate of the offender. This information was buried on that city's website in a bunch of short videos the traffic engineers made about their new gee-whiz traffic management system.

Which, by the way, many municipalities are building (or have already built) "new gee-whiz traffic management systems" which are Orwellian nightmare machines. They seem to be flying well under the radar because everybody hates traffic, and especially because nobody is allowed into the new traffic management buildings for tours, to see the capabilities of the system that seem to be hooked directly into law enforcement's computers.

Do they directly say that they are using license plate reading? I find it far more likely that they are using the OPTICOM GPS solution which, in hybrid with OPTICOM IR, solves the same problem using cooperative radio equipment in emergency vehicles. Reporting on received IR preemption with no matching radio communication is an advertised feature of this system, whereas performing ALPR at the range and conditions of lane occupancy cameras is still largely experimental. ALPR virtually requires IR illumination, while lane occupancy cameras have no illuminators and are usually monochrome.

What do you mean by "conspiracy"? I'm not trying to be funny here, I think it's worth examining what precisely you mean by that word when you use it. It's a word with a specific meaning that is frequently abused in wishy-washy ways, often used to describe loony beliefs not grounded in reality, or beliefs that run contrary to the official narrative.

If the official narrative is that we aren't being constantly surveilled, then this submission pales in comparison to what should be on the front page.

But I don't think that is the official narrative, per se.

Is the official narrative posted on the official narrative report somewhere for reference?

Note that if you're using the term "conspiracy" in the strict sense, it's not contrary to the official narrative. The official narrative for 9/11 for instance, the 9/11 Commission Report, is terrorists conspired to crash airplanes.

In the colloquial wishy-washy sense, that report is not a conspiracy theory, despite being a theory about a conspiracy. However every other theory that disagrees with it is a conspiracy theory.

It's an opportunity for people to fein surprise. And people always like a little treasure hunt...

Reply on news.ycombinator.com