Hacker News Reader

Comment by omarhaneef | Hacker News Reader

My first thought was: how is NYC not on this list since it has cameras on every block, and has for decades.

And then I saw this: "Atlanta was the only place in the U.S. to crack the top ten, with 15.56 cameras per thousand residents."

So they are looking at cameras per 1000 residents. This seems like the wrong way to count. If you point a camera at a block, you see everyone on that block whether there is one person or 10.

They should compute it as percentage of public area that is under surveillance, I think.

View on HN

What matters more than the number of cameras is how/if the cameras are integrated. I don't much care that the local gas station has cameras tied to a 30-day DVR. I expect that. I don't expect them to be connected to facial-recognition AI, networked to the national police, and be feeding my movements into a citizenship score system that will impact my future job prospects.

While not networked to the national police I expect that many commercial security cameras do have facial recognition to identify suspected thieves and VIPs along with tracking movement habits through the store which is shared with advertisers and product manufactures to justify cost for higher profile shelf space.

>> tracking movement habits through the store

My understanding is that was a passing fad a few years ago. It was hyped by some startups but commercial users found it less than helpful compared to traditional metrics. I met a grocery store consultant (a real job) at a security conference. He said they used cameras to deal with traffic issues (even load between aisles) and predict cashier/customer ratios. They had abandoned the face/eye-tracking tech. Eye-level shelf space is better. They don't need an algorithm to keep telling them that.

They don't need CV for this. They can use beacons embedded in smart devices to tell which part of the building you're in. Worked with one of these guys.

They need cameras for theft. Tossing a person detecting algorithm on top is basically free (especially if you don't need it to do fancy things like track faces).

How would one calculate that score?

Investigative journalism.

Parked too far from pump -1. Didn't check air pressure in tires -1. Drove too quickly -1. Yawned too much, driving while tired -10 points. Did have correct national flag in rear window +1. That is all easily within current AI abilities.

You can ask Alibaba https://en.wikipedia.org/wiki/Zhima_Credit

Well, Atlanta has integrated over 10,000 cameras and 200 pole mounted automated license plate readers. The per capita statistics are distorted by the relatively small boundary of the City of Atlanta, but there are still a lot of cameras.

[https://atlantapolicefoundation.org/programs/operation-shiel...]

[https://www.muckrock.com/foi/atlanta-325/2018-vigilant-data-...]

I saw a "best cities for public transit" list recently that had Reno, NV up near the top. Which was obviously baffling, and sent me to find out why. I eventually found that one of the major factors was "miles of transit line per capita and another was something like "annual mileage per transit vehicle" (possibly per-capita?).

The goal was to evaluate density of coverage and frequency of arrivals. But that completely neglected population density, so it ended up rewarding cities for urban sprawl.

This feels like the same type of mistake, where per-capita count neglects both population density and total coverage. There's a massive difference between every shop having its own camera (even if police access them) and something like London's Ring of Steel, where the camera layout and even streets have been designed to maximize coverage and identification rate in the most populous areas.

Yeah, it also conflates private/publicly owned cameras. As someone who lives in London, I think we get unfairly portrayed as a surveillance state, because really the vast majority of cameras are inside shops and businesses, not on the street, and not connected to any central databank. I personally don't find cameras like these troubling. Once you get out of central London, most streets don't have cameras on them.

And besides, you'd need a public walkway license to look at the walkway, and then a camera license to use the camera, and then a license license to display the license...

(I'm joking, but the covert surveillance and public scaremongering that London did to enforce TV licenses was a little absurd.)

> I personally don't find cameras like these troubling.

It’s interesting to hear this difference. Here in Germany there are often cases where the courts forbid fake cameras because they "create an atmosphere of surveillance".

Berlin has made that list too, though.

It’s not as if cameras are illegal here. You just need to have good reasons for them and have a very narrow area it surveils.

For example many of those fake cams getting forbidden were from landlords and residents sued. Now if there were good reason like many break-ins in the area, the same landlord would have been allowed to have a real camera there.

This is a tricky problem, but I don't think limiting the numbers to public cameras would work. I'm not sure any numerical approach would.

In Atlanta, for instance, the cameras surveilled by the police are 90% privately owned. And this isn't just "with a warrant" or "when a shopkeeper reports a crime and provides tapes", but nonstop, integrated access to their feeds.

Conversely, a lot of the cameras in London may be private, unconnected, and indoors, which I agree is much less worrying. But the Ring of Steel does far more surveillance per-camera-per-capita than any shop camera, since it has unified design meant to maximize how many people are identifiably captured on camera.

Central London was a surveillance state more than 20 years ago.

Since 1984, one might say.

The one time I was in London fifteen years ago, it felt like 1984. I was sitting in a bus and there were at least five or six TVs showing several camera perspectives.

Including one directly at the back of my head, which showed some early signs of balding.

It was the most unpleasant bus ride I have ever had. London is nuts.

Five or six TVs on a bus? I live here and I've never seen that. What some busses do have is a single TV screen that shows multiple feeds - either at once or flipping between them. I've never been particularly sure of the purpose or the screen (dissuading violence?), but it is useful for seeing if there are any seats upstairs.

It's the posters that creep me out

http://www.adsavvy.org/wp-content/uploads/2008/12/watchful_e...

http://www.adsavvy.org/wp-content/uploads/2008/12/creepcamz....

Even better would be the amount of seconds each person is surveilled during their time in public space. Summed as an absolute number, and as a ratio of the total time they spend in public space.

A camera in a busy street or at a chokepoint is worse (or more effective, depending on your viewpoint) than one in a cul-the-sac, even if they cover the same amount of space.

Good point, I think it's even more complicated though.

Consider not only how many cameras there are and where, but also to what level of detail they are able to see, who has access to what they see, and as mentioned elsewhere ITT, the other systems that they are connected to.

e.g. a grainy CCTV cam in a convenience store is a single camera, as is one in a Gorgon Stare surveillance plane, but those do not generate equivalent amounts of surveillance.

As you suggest, NYC is intensively and effectively surveilled by the NYPD. The first bit of Steven Rambam’s talk at HOPE 2016 covers this in detail, starting around 3:30:

https://youtube.com/watch?v=FHwl6AyL6js

Seriously, this is dumb. NYC alone gets 60M visitors.

NYPD has a well integrated and pervasive intelligence system. I’m sure Atlanta is no slouch, but the money spent by NYC just dwarfs anyone who isn’t the Feds.

They also say

> Chicago had the highest number of individual cameras amongst U.S. cities, at 35,000.

So we at least have an upper bound.

Actually if you dig in you can find a spreadsheet [0] with all the info. NY has 11k cameras total. If you sort it by just number of cameras NY ranks as #38 (as opposed to #58 by cameras/1k people)

Also digging in, they don't have data on LA. Though they do mention that there are 300-400k private cameras. Though the top 3 cities (all in China) have well over a million.

What's also interesting is the correlation between the crime and safety indices. Some of those cities have low crime indices (under 20 is good, above 60 is high[2]) and high safety indices (higher == better). The correlation is really off here.

[0] https://docs.google.com/spreadsheets/d/1bKBFiVXNzrgtW95j66Tp...

[1] https://www.comparitech.com/vpn-privacy/the-worlds-most-surv...

[2] https://www.numbeo.com/crime/indices_explained.jsp

I agree with your assessment. I'm in Atlanta and I'd be hard pressed to say I see too many cameras outside of roadway cameras, which we have LOTS of. I'd wager road cams make up a very health proportion of the cameras counted by this article. Pedestrian cameras, not so much.

Note the population figure for Atlanta though: 501,178. They're only counting cameras and people within the city limits, not the area that most people broadly call "Atlanta".

Even when looking at raw camera numbers, Chicago has 35,000 vs NYC's 11,000 cameras.

Ah, interesting.

I suspect they mean the wider NYC area. Manhattan itself is just shy of 34 square miles, and I think it is probably more surveillance dense per square mile than any other place in the US, but that may not apply to the wider NYC area.

NYC is still not on the list by # of cameras. If you read the rest of the article you see Chicago is #1 in that category, with 35,000 total cameras.

the point of the parent comment was that the # of cameras doesn't matter. New york is small but dense. a single camera on a busy corner will see more people in a day than multiple cameras somewhere else.

you're talking about number of people covered, but area covered might be an even better metric.

I think the number of cameras per 1000 capita is actually a good metric, it's easier to calculate and it probably is doing a good job on average.

Agreed. By the per capita metric my pants have 2000 cameras per 1000 inhabitants.

Reply on news.ycombinator.com