Hacker News new | past | comments | ask | show | jobs | submit login
Gunshot detection system transforms and raises issues (nytimes.com)
200 points by mindblink on May 28, 2012 | hide | past | web | favorite | 171 comments

Reading this feels like accidental time travel. When did this start happening?

$40,000-$60,000/sqmile/year is an extremely reasonable number. Chicago has 50+ officers per square mile, implying tens of patrolling officers per square mile; the fully loaded cost of a single police officer is six figures, with potentially explosive defined-benefit pensions and benefits just to make things fun.

Moreover, like in many big cities (read Peter Moskos _Cop In The Hood_ for details), the squad car patrol tactics used in Chicago do a terrible job of suppressing gun crime. Over a 24 hour period this holiday weekend, we had 25 gunshot injuries (god knows how many illegal firearms were discharged above and beyond that). The police can't be everywhere, and the patrol process keeps them mostly in rolling squad cars.

And obviously, the overwhelming majority of gun crime in the city of Chicago are concentrated in a small percentage of our square miles: Austin, Garfield Park, and (particularly) Englewood and Chatham. Which makes systems like this cost-effective to roll out, and, particularly, easy to pilot.

The technology is impressive. The precision of the information (location, time, pattern) can be transformative in how police responds to high crime areas. Right now, I can imagine that the quality of information degrade in a game of telephone from the caller to the dispatcher to the patroller (and even more so in HCAs). Meanwhile the patroller has to be balancing the issue of investigating a false positive and wading into a danger area.

Panoticon here we come....

A Chicago suburbs resident here: You would think that as the system scales up, your cost per square mile would drop, as you're automating as much of a analysis and detection as possible.

One little bird (presumably a Chicago cop):

By my totals 17 shot on Sunday, 7 of them in 011. With 11 more starting at midnight to 0800 monday morning. The blackberry was ringing all night. 2 of the "victims" dead so far.


Note the scare quotes on "victims." Our informant continues:

The city should embrace all the violence and make it a tourist attraction. See if you can come to Chicago and make it out alive! Have tour buses go through the ghetto and stop at all the crime scenes. Get your picture with the body! Only 20 bucks. Better make it 30, 10 goes to the alderman of the ward the body is in. That way the visitors can get a small taste of how business is conducted in Chicago. As a special treat you may also get robbed at gun point during your tour. And if you want to see the indigenous population acting normally in its own habitat start using your Iphones where you can be seen. But you will be charged extra if you get apple picked! End the tour with specially trade marked Chalkie shirts with all proceed going to hire some real leadership at the FOP so we don't get screwed on our contracts. And give one buck from each tour to a special fund for the kids. ITS FOR THE CHILDREN!!!

God I love the Internets...

My 25 came from the same source: Second City Cop blog. They're a collection of anonymous Chicago police.

The scare quotes refer to gang shootings.

SCC tends strongly towards alarmism and, often, hyperbole. Your odds of getting shot in Chicago, even if you go to Woodlawn for Lem's barbeque, is extremely low. Unfortunately, your odds of getting shot if you're a kid living 24/7 in Englewood or Humboldt Park are unacceptably high. Also: lots of racist cops. Blue collar job.

Well! If you know SCC, you must love the Box Chevy Phantom! And Chalkie! And I'm sure you've seen the famous Cabrini-Green thread:


And you must have loved this discussion of "Jared," "Brent," and their fellow Occupy bombers:


Extremely low? Compared to what? What it should be? What should be your odds of being shot in a major city, anyway? San Francisco is extremely safer than Englewood, and earlier this year my wife and kids still found themselves in the middle of some kind of Norteno-Sureno shootout. They didn't get hit, though! So no harm, no foul.

Alarmist, certainly. Certainly alarmist compared to your normal Chicago sources of information, which as I recall insisted that the beaches on a certain day last summer were closed due to "temperatures in the '90s:"


It's a fact that I'm not a blue-collar fellow and I don't have a blue-collar job or a blue-collar family. And no, I don't think that if I were posting comments on SCC, I'd be quite as cavalier in referring to "the animal," "mutts," etc. It does get the point across, however.

I've seen this in action and it is pretty impressive. I expect the next step will be that a quad-copter UAV will 'nest' on top of a ShotSpotter pole where is will stay charged. And when an alert comes in, if it is closest, it will launch and give the HQ a video feed of the scene in 15 - 20 seconds after the event.

I do have concerns about the use of conversations the microphones overhear but I know that in many neighborhoods events are not called in because the neighbor doesn't want the repercussions of turning in the local gang lord or his troops.

Prediction: When the system goes to dispatch the closest UAV to the disturbance, the launch system will return "500 - Destroyed".

EDIT: C'mon, downvotes? You do realize that vandalism will be a serious problem for this system.

I think you need to make a convincing argument that vandalism will be a problem. Paraphrasing from The Information, pg 144, there was also that concern when Morse and Vail set up their wires wrapped in yarn on 20-foot posts. Morse told Congress they could transmit 30 characters per minute and he also told them that the lines had "remained undisturbed from the wantonness or evil disposition of any one." There is even more ample opportunity for domestic terrorism and destruction of things in the US today yet things seem okay.

Of course if the UAVs did start getting destroyed, it's not a stretch of imagination to think about how they could be protected...

Drug dealers and gangbangers aren't stupid. If they see the cops installing these things it's only a matter of time before they'll find out what they are and start vandalizing or destroying them.

From the sound of it, if there's enough crime for these things to be cost effective then there are enough criminals around for somebody to get rid of the shot detectors.

I don't think the comparison to telegraph lines really makes sense. If telegram lines were used solely for reporting crime it would be a better analogy.

Every high-crime neighborhood in Chicago is studded with cameras with flashing blue lights. The enclosures are hardened. Destruction of CPD cameras has not been a major issue in the city. If the city can keep cameras operational, they can keep acoustic sensors (smaller, no line-of-sight requirements) secure as well.

My guess (just a guess) is that the "criminals will destroy the sensors" concern is totally overblown.

I heard about pissed of people in the UK putting a tire over a speed / red light camera post and lighting it on fire. Tires are harder to extinguish than your typical fire and it's usually too late before it can be put out. Eventually it becomes unprofitable for the third party contractor to stay there and they start going away.

You can put a tire over a speed camera without your picture being taken by it and instantly transmitted off-site.

It happens but not that frequently.

Well, my comment was specifically directed towards the idea of a UAV first responder system. It's not hard to imagine ways that the detection system itself can be hardened (or hidden).

> Drug dealers and gangbangers aren't stupid.

Dispute the premise.

The head capos might have some smarts, in a cunning sort of way. But the footsoldiers and unaffiliated who get into gunfights on public streets? Not so smart. Disorganized American urban criminals aren't at the level of Al-Qaeda or the Mafia.

You don't need to destroy them to make them ineffective. Just find a way to produce a high number of false positives legally. Or hell, get 10 mates to fire some shots in 10 different locations while you make your hit.

Or use a silencer.

I think you greatly misunderstand what a suppressor actually does on a firearm.

what do you mean? I know how silencer (sound suppressor) works and I know (and was shooting) a gun which doesn't require a suppressor. The latter one makes no noise AT ALL, except for the mechanics.

Presumably the drug _dealers_, who are more likely to be on the receiving end of violence in their own neighborhoods, are going to want to keep those shot detectors unmolested.

"I think you need to make a convincing argument that vandalism will be a problem."

I think if the people having shoot outs are so brazen that they can do this and people aren't necessarily reporting them it's not a stretch to think that they will, just for fun, attempt to shoot down a UAV. Wearing a mask or course to protect their identity. Or try to damage it while it sits up on the pole with a rock etc. (Although I guess it could be protected from that fairly easily).

I don't know what thugs do for fun. I don't see why shooting a UAV would be any more fun or wise than taking an axe or chainsaw to a telephone pole to cut the power of a block you intend to commit a crime in. (Plus if you know what the UAVs are for, you know that your shooting will be heard and police can have a very short response time just from that. Add in redundant UAVs that fly higher and faster, you're not getting away.)

Another form of protection: they'd be less likely to shoot at something that shoots back. (Good luck making that legal for a private company though, even if the UAV managed 99% non-lethal accuracy!)

I think UAVs are impractical for other reasons (why not just put cameras everywhere instead of a fewer number of mobile cameras?), but I still don't think they'd be that much more a target for destruction than the electrical grid.

"I don't see why shooting a UAV would be any more fun "

Doesn't have to be the thug shooting it down. Can be a pain in the ass bystander in the hood or someone with a baseball or rock. Have you ever seen kids who throw sneakers up over electrical pole wires in the inner city?

"they'd be less likely to shoot at something that shoots back."

Absolutely positively not going to happen.

Huh. I'm pretty sure that shooting down a UAV would be fun, and I'm not even a criminal.

I know of army units that have a machine-gun-the-RC-planes day for fun.

> If you know what the UAVs are for, you know that your shooting will be heard..

If only there was a device which could be affixed to the barrel of a pistol which could reduce the decibel level associated with the firing process... ;)

Assuming these people did add UAV monitoring (a leap), and assuming a UAV was dispatched to every report of gunfire (another leap), it doesn't matter if the assailants shoot it down because they already know where the gunshot was and, with a decent response time (another leap), officers can be there in minutes.

Not really the best analogy: telegraph/phone lines sit there passively and pose no apparent immediate danger, while the aforementioned UAVs would present an active threat to wrongdoers’ wrongdoing and thus become targets...

A number of people have responded to the key points, but one that hasn't been mentioned is that we have a presumption of innocence. So if a UAV shows up, and you're holding a .45 caliber handgun, we still have to presume you are innocent of the shooting (you may have picked up the gun from the street for example). However, if you take aim at a UAV streaming video back to base and start shooting at it, well that can and will be used against you in a court of law.

There is another issue which is that a hand gun is a poor weapon for shooting down an aerial vehicle. A good marksman, or someone very lucky might do it, but the weapon of choice would be a shotgun. Shot guns on the other hand are not effective at range so in a street fight they aren't a good weapon (great for close quarters, poor for engaging at range).

So in a completely lawless sort of situation I could see teams of shotgun toting folks keeping the UAV population down but not in the typical urban setting in the US.

I wonder what happens when a low light camera meets a laser pointer. Does anyone know?

Direct laser hits make camera sensors pretty unhappy:


If camera toting quadcopters on microphone equipped poles become common, I'd expect West Oakland gangs to start handing out high powered laser pointers to all the local kids…

Well I don't know about low light, but lasers operate on very narrow color bands, so if you filter out the color of the laser in the video stream, you filter out most of the laser light. You also have to be very precise to actually hit that tiny sensor moving in the air from a distance.

Criminal charges.

Return fire?


Just so you know, shotguns are not the wildly inaccurate things as depicted in most TV shows and video games. Depending on the choke and ammo used, they can be accurate up to 50yds with Buckshot (depending on choke), and probably up to 100yds with slugs.

I would say the most damning thing about using them for the purposes you describe would be the awkwardness of carrying around a shotgun that is that accurate and hiding it.

I don't think it's easy to destroy a small moving object so far away that you can barely see it. The UAV don't has to fly too close to take photos or video and it can be very small. A more useful thing for the criminals would be a jamming device against the wireless signals of the UAV, but then they can make the UAV work without wireless (without the live stream it can still take photos and videos of the place).

This sounds like a case where the problem of collective action would actually work in society's favor: a single vandal would bear the full cost of destroying the sensor (e.g., the risk of getting caught), but the "benefits" of the destroyed sensor would be diffuse, spread out to all would-be shooters.


At this point you have to compensate the individual enough by, e.g. paying much and/or providing proper equipment[1].

[1] - I imagine a proliferation of cheap microwaves converted into portable battery-powered anti-electronics weapons. And how do you protect UAVs against that?

You do realize that vandalism will be a serious problem for this system.

I sure hope so. My dream is to start a DIY / Open-Source Hardware project to create a drone that can shoot down the government drones and UAVs.

Aerostats. The Diamond Age becomes ever closer to reality.

From the sounds of how violent the country is, I think Americans actually do need to have this level of control and suppression.

Combine it with one of these http://en.wikipedia.org/wiki/Phalanx_CIWS and the problem is solved !

Beatings will continue until morale improves.

> the system was not intended to record anything except gunshots

So, they could record conversations, but they won't for the time being. A classic example of using software to temporarily cripple the true capabilities of a computing device.

In the hands of DRM-wielding corporations, "defective by design" results in inconvenience and loss of users' freedom. In the hands of the surveillance state, the same technique results in a situation where citizens must simply trust the authorities to exercise restraint. Because the police could flip a switch at any time and record all sorts of conversations. Somehow I don't trust that the switch will remain un-flipped for long. And when it does get flipped, as it did in New Bedford, everyone will say it was just an accident.

Maybe, just maybe, we should accept systems like this as a necessary evil in certain cities where there's a lot of gun violence. Still, I don't like this. When it comes to the government, I'd much rather give them hardware that can't be unlocked "by accident" because there's nothing to unlock. The thing is, it's unrealistic to do that in all cases, and we have to hit a balance somewhere. Which is exactly why systems like this raise difficult issues.

Given that the microphones are placed in a public location, and that they record noises heard in the street, one could argue that it's a reasonable level of surveillance.

The real concern will be when governments start to identify their citizens by their voice signatures, and then use that to track their whereabouts.

That's not the worst of it. Consider the full implications here. Imagine a worst case scenario where gunshot detection is used as a pretext to put microphones all over city streets, every block say. Now imagine that those recordings are stored forever and properly time / location tagged, not a technologically difficult problem.

Then imagine what is possible with such data. First off, you could use triangulation and advanced filtering between multiple microphones to be able to pinpoint the source of each sound and separate it out from the background. You could, as you say, identify individuals by their voices. You could track their wherabouts. You could monitor who they are talking to and when. You could learn so much about their lives by monitoring all of their conversations in "public". In the worst case scenario of the government turning into a police state this is a frightening level of surveillance.

Apart from the fact that the devices are apparently designed not to be capable of recording conversations, and can clearly be improved to make it even more difficult to record conversations, "not deploying gunshot detection" isn't the only privacy control that cities can employ; cities can just make it illegal to collect raw audio.

The article mentions that the devices did manage to record conversations in at least one case. If so, they are clearly capable of recording conversations, just not optimized for it.

Also, how do you distinguish gunshots from background noise and triangulate the location of shots without first collecting raw audio from multiple devices and analyzing it?

Indeed, there's clearly some disinformation going on in the article:

"James G. Beldock, a vice president at ShotSpotter, said that the system was not intended to record anything except gunshots and that cases like New Bedford’s were extremely rare. “There are people who perceive that these sensors are triggered by conversations, but that is just patently not true,” he said. “They don’t turn on unless they hear a gunshot.” "

So apparently "the sensors", "They don’t turn on unless they hear a gunshot.". How, exactly, do they "hear a gunshot" if they're not (yet) turned on?

I suspect the truth is there's some software configuration that inhibits _recording_ of the sensor data until a gunshot-like event occurs (though if _I_ were designing this system there'd be at least a 30second or so buffer, so I could archive the sounds that if heard _before_ a gunshot as well as afterwards). But I'd hesitate trust that "configuration" to be particularly secure - much like the TSA "pornoscanners" - which in spite of claims of it being impossible, seem to be able to record images for the amusement of the operators and their friends…

They could be designed to run some local preprocessing and only record/forward audio when a gunshot seems significantly probable. (Not that I'd bet they actually do. We need public pressure to get them made that way.)

If I recall correctly, privacy laws are based more on reasonable expectation of privacy rather than where the recording device is located. I can't climb a telegraph pole and photograph someone inside through a second floor window, for example.

I don't know how this would affect the legality of surveillance though.

>The real concern will be when governments start to identify their citizens by their voice signatures, and then use that to track their whereabouts.

And thanks to a recent heavy-hitter at the box office, the public has already had it decided for them that it's unethical! (Unless it's used once for a really really bad guy.) But apart from that this is an awesome system for its purpose. Keep it out of my home (they've already got phone-tapping for that anyway) and I'm fine with it.

ShotSpotter cannot be "unlocked" to record conversations by accident. These sensors are placed far away from where people might be talking and this is intentional. Conversations are just one more source of noise to be filtered out.

Engine sound is another source of noise, but ShotSpotter can record engine sound and use it to figure out whether the shooter was in a car. Why would conversations be any different? Noise filters can be turned on and off.

This is no different from modifying the nude scanner at the airport to display a generic human form rather than the actual outline of your body. The machine still has the capacity to display what your private parts look like.

Most likely they use the fact that the shot location moved, not the engine noise.

How? With multiple shots it's easy as hell. But from one it's quite hard?

You could hear the firing pin hitting the cartridge (harder to hear than engine running) then you should hear the gasses exiting barrel somewhere around 0,001 second later. Then calculate doppler... Sounds kinda hard.

I bet the doppler shift from two different listening microphones would give you a pretty easily calculation for how fast the gun was moving (at least relative towards/away speeds with respect to the two microphones) - I'd guess 3 or more microphones listening to the same gunshot could pinpoint both direction and speed.

I was thinking about the Doppler too, but the Doppler shift of the bullet while in flight doesn't really do it. And the time-frame to do it while inside gun is really really small.

I highly doubt they're providing that service for single shots. Besides, I can't see too many thugs firing a single shot from a car in commission of an assault or homicide. They're not exactly known for their accuracy at the best of times.

Ah, that makes sense.

Locks are mostly superficial. I've seen too much footage recorded from security camera monitors with phones to think that any monitoring system is secure or locked when people are involved.

On the growth of such systems: it is almost inevitable that sensors end up getting placed everywhere. You have hundreds of sensors in your phone right now, many of them are uploading data to websites at various times.

Sensors will be placed on almost every utility as a matter of safety and efficiency. It is important to know when a pipe breaks or if there is a small leak in a gas main. Look up any given utility company and they will probably have some "smart" this or "integrated" that scheme in development.

There are cameras everywhere in most cities now, it is a matter of time before there are microphones too. Identifying who is where and who is speaking is a programming problem that has mostly already been solved. The two will be put together in the interests of safety and fighting crime. Someone might even invoke the terrorism word to speed up the process. Small-scale at first, in key places and important settings. Then rolled out across entire cities and states.

I am almost entirely visible to security cameras from the moment I step outside in the morning until I reach my office. Eventually, these systems will be integrated and process data effectively. The question is, what happens then?

“If the police are utilizing these conversations, then the issue is, where does it stop?” he said.

Libertarians consistently make slippery-slope arguments when most everybody else is just happy that some immediate problem is solved. This line of debate is freaking getting old, and I'm the first person to do it.

The problem is that it always is a slippery slope. No bullshit. Changes take place over years or decades, so there's no single time you can raise an alarm. Right now it's gunshots. Next it will be car sounds -- estimating speeders and the conditions for traffic accidents. Then somebody will work out screaming. Then, perhaps conversations. And let's not forget that the systems will be justified by talking about the horrendous inner city. In actuality the vast majority of the time these systems will be used in places nowhere like that. When you read stories like this remember that these guys are selling equipment just like any other startup. You're getting their best pitch.

Big cities need this stuff, so it's a good thing for them. (Although I imagine we'll just start seeing a lot of silencers). What concerns me is that 90%+ of the time there's no crime being committed, save for discharging a firearm. So there's all these thousands of "criminals" discharging firearms that haven't been arrested before but could be now. Yay? Is it always a good thing with the grip of the state tightens, as long as we can point to something good coming of it?

Hopefully the cops will be so overloaded with gunshots they'll ignore the system and use it only for forensic purposes. But I doubt it. Instead I imagine we'll see these discharge numbers added to the crime reports for cities in an effort to secure more funding for even more police presence. Whether that's a good thing or not is debatable. There's obviously a real problem in Chicago and several other cities, but the rest of the country not so much.

Most of these slippery slope arguments are made by people who seem to have no actual experience with any of the issues at hand apart from reading 1984 and Fahrenheit 451. This gives them about as much credibility as someone whose information comes from watching Jack Bauer on 24 and Fox News.

You cannot come to reasonable conclusions about issues like this without seeing what's going on in the real world. This is not a math problem we're talking about; you can't sit around thinking and extrapolating and expect your conclusions to mean anything. It's like trying to run a scientific experiment without any data.

Do you know anything about criminal law? Do you know anything about the legal checks that are in place that prevent your doomsday scenario from happening? Are you aware of how evidence rules are applied in real cases? I'm sure you've read stories about the most egregious abuses, because they are the ones that pop up in the feeds that you read, but do you extrapolate from these that all cases are like this?

I say all this because I recently had the very eye-opening experience of serving on a juror in a five-week criminal trial. I realized that many of the prejudices that the technorati crowd has about the justice system are totally unfounded, at least for the case that I sat on. Our jury was not a clueless bunch of simpletons, our prosecutor couldn't get whatever he wanted by just saying "think of the children," and the judge was not an apathetic overworked bureaucrat who let anything slide.

What concerns me is that 90%+ of the time there's no crime being committed, save for discharging a firearm. So there's all these thousands of "criminals" discharging firearms that haven't been arrested before but could be now. Yay?

Are you really arguing that discharging firearms is a victimless crime? (edit: I'm talking about dense urban areas).

Given that I spent the better part of my childhood firing a .22 and a .306 after school at whatever I decided was worth shooting (not to mention fall season, when whatever was worth shooting usually had four legs), I would argue that discharging firearms is not a crime. And it most certainly fits _the_ definition of "victimless crime." The only parties that you are indirectly benefitting would be Walmart (or wherever you pick up your ammunition).

Were you doing that in a dense urban area?

I had neighbors. I was taught (as most gun users are) to be aware of both what I was shooting and, and, importantly, what was _behind_ what I was shooting at.

The Rural/Urban/Suburban thing is irrelevant - what's important is the safety of the discharge.

Regardless - the point I was trying (and clearly failing) to make is that there is nothing illegal regarding the discharge of a firearm. The crime is doing so irresponsibly and in an inappropriate location.

Firing a firearm in an urban area in public is, as far as I can tell, almost always illegal. No one is trying to invent a system that detects every gun shot ever. No one cares if you shoot in a shooting range, or in your expansive back yard, or in the wilderness.

Could you please find me one example of a public space in a major metropolitan area where it is legal to discharge firearms? That is the issue here.

"Regardless - the point I was trying (and clearly failing) to make is that there is nothing illegal regarding the discharge of a firearm."

What? "Discharging a firearm" (shooting) is illegal in almost every situation in urban areas. If you think that you can go shooting at beer cans in, say, Central Park, as long as you make sure to only shoot at a downward angle and with a sand hill behind your targets, you're in for a rude awakening. It is unsafe to shoot in all but the most controlled circumstances (e.g. on a shooting range) in urban environments, and even suburban ones.

Who's thinking about deploying gunshot detection systems in rural areas? If nobody, then what does it matter? Gunshot detection systems in cities are almost by definition targeting inappropriate discharges.

I don't really care about gunshot detection systems. I'm trying to bring a bit of balance to a thread that had the unqualified statement, "Are you really arguing that discharging firearms is a victimless crime?"

I wouldn't be surprised to believe that some percentage of the HN audience might actually _believe_ that discharging a firearm is a crime, so I was seeking to inform them that it is not the act of discharging a firearm that is a crime. It's doing so in a negligent manner, or in an inappropriate location. Check your state/municipal bylaws.

It's simply not practical in the sort of conversation to continually provide context for every statement as you make it. It's the reader's job to remember what was said in the parent posts and keep that context in mind.

Functional style of writing?

I was making the argument in the context of this discussion about gunshot detection systems in dense urban areas. I've clarified this in my original comment.

A healthy skepticism of placing recording microphones on street corners all over a community is a natural reaction to an egregious abuse of civil liberties. It probably doesn't mean much to you, because you're probably white, and probably live in a "good" neighborhood. So this won't affect you.

The slippery slope is already iced -- the systems are already being used to record conversations. But I guess that's ok, since it's just affecting those people on the wrong side of the tracks.

I don't believe you could come up with a better definition of "victimless crime" than "discharging firearms within the boundaries of a dense urban area." "Discharging Firearms" is akin to "Driving Car" or "Operating Forklift" or "Owning Swimming Pool" - all of these acts can potentially be dangerous, but, responsibly done, the risks can be mitigated. (Though, statistics seem to indicate it is inordinately more dangerous to own a swimming pool than to discharge a firearm, in terms of accidental deaths of children per year of doing so)

Listen, I think many of the people responding to you are sympathetic to your argument (as in, pro gun ownership, not being alarmist, etc - I certainly know I am), but you're hurting the cause more than you are helping it by comparing "discharging firearm" with "owning swimming pool". Your Freakonomics example was about gun ownership, not shooting it. I don't see how you can reasonably argue that shooting a gun is equally safe or unsafe than is "owning a swimming pool".

Of course the risks of firing a gun can be mitigated and controlled, but to do so, one needs a lot more rules and procedures (both on the individual and societal level) for guns than for forklifts and swimming pools. If you disagree with that, I'm afraid I (or, I suspect, some or most of the other people responding to you) won't be able to have a real discussion with you since our fundamental assumptions would then appear to be so far apart that we'd have to regress to a much more fundamental level and clarify those assumptions first before it would make sense to come back to the relatively high-level argument at hand.

My bias, I guess, is growing up with the sound of gunfire on a daily basis from family and neighbors. I didn't really think it was a big deal - just became part of the background noise. The concept of "Gunshot detection systems" just seemed ludicrous to me. I didn't actually realize until I googled a bit that it actually _was_ illegal to discharge a firearm within certain city limits - I would have argued (and lost) that the law was "negligent discharge of a firearm"

I guess if it's illegal to do so, then a "Gunshot Detection System" has use. In the same way that Red Light cameras and Speeding Detectors serve a function in managing those laws.

There are lots of places in the Bay Area where you can safely discharge a firearm within city limits. At least I know I need to review the various city/municipal bylaws to see whether it's legal.

What? Even if you are the most responsible and conscientious gun owner whose bullets will never damage any people or property, discharging a firearm in a dense urban area is severe noise pollution and causes distress to people who hear it because they have every reason to believe that they may be in danger. And there is practically no legitimate reason for it outside of self-defense.

I live in the northeast and own several guns. I would never even consider discharging them in an urban area. Even most suburban areas are unsafe. From my point of view, if I'm not at a gun range or on a giant expanse of rural/hunting land, the gun is locked up.

A gun locked up in a gun safe is surely safer than swimming pool. But a gun being shot in the backyard of a suburban home is definitely less safe than a swimming pool.

My neighbors live roughly 60 feet away from my house. If I went outside and shot a racoon that crossed my property, I'm pretty sure I would be in prison.

I could feel the outrage in your comment and felt compelled to reply. I don't usually reply because, quite frankly, it's impossible to accurately convey your argument to 40K+ readers and most of the time the thread underneath my comments is just an indication of how poor a job I did with communication.

You've constructed quite an elaborate straw man ad hominem around me, so let's clear some things up. I do not watch Fox News or listen to talk radio (although there's nothing wrong with either). I do not have a feed that provides me with shallow-reasoned libertarian tripe. In fact, I scan more than 30 editorials and commentaries daily from sources as diverse as the foreign language press, Mother Jones, and the NYT. I also have many friends in the law enforcement community and think they do a great job. I am somewhat familiar with the system as it exists today. Nowhere near an expert, but not a theorist either.

Now that we've straightened out that bit, let's get to your argument. You cannot come to reasonable conclusions about issues like this without seeing what's going on in the real world.

This is exactly what I said: that real world tactical considerations are all that matter to some folks. The existing situation is that we have no gunshot detectors. So the question becomes, aside from structural political considerations, "Do such gunshot detection systems provide more societal value than harm?"

This might surprise you, but I'm all in favor of gun control laws for the inner cities. As population densities rise, the local government by necessity must take more control over citizen's lives. Note that I said "local government" and "by necessity". So the only problems I would have with these systems is whether they are being selected by the majority of the local people and whether or not they are deemed necessary by those people.

Are you really arguing that discharging firearms is a victimless crime? (edit: I'm talking about dense urban areas). No, I'm not, but since you brought it up, let's take a look at the real world data. According to the article, the vast majority of gunfire in a city doesn't cause death or major bodily injury. That looks pretty victimless to me. Personally, for a device that can cause bodily harm or injury to only one other person? I'm comfortable with an safety rate of around 99.9%. Others may have different preferences. The point here is that you can never have a perfectly safe conditions. The only question for folks to ask is how much risk is acceptable.

My problem is that we are making sweeping decisions for the entire country without realizing it. When you have a negative law, like "you can't own a gun", it only impacts the area where it is passed. But when you introduce new technology not covered by laws, especially technology meant to fix a social problem, you're setting a precedent for everybody. I'd argue that for 99% of the land area potential future gunshot detection systems will cover they will be neither wanted and controlled by the people nor deemed necessary by them. They'll become like automatic radar systems: another income stream and another intrusion into what used to be a mostly harmless affair.

One of the things that we don't talk about is how many laws are broken all of the time and it doesn't matter much to society at large. Things like the majority of these urban firearms discharges, people who lie to federal agents, speeding, or so on. To me, it looks like the purpose of these urban firearm discharge laws are just to have another crime to tack on to a suspect when sending them off to jail. I think the data shows that. But in either case, we have all kinds of laws, like speed limits, that are broken all of the time. Do we want to computerize our system of justice such that they are all followed all of the time? I'd give a vigorous "hell no!" to that idea. The system grew in complexity based on the idea that it was porous. You make it airtight and you'll have massive civil unrest on your hands. My opinion only, for what it's worth. The system of law we have evolved is not a system of morality or a way to lead your life. It's just a bunch of ad-hoc rules that are mostly consistent and worked by mostly great folks. But society couldn't function if you enforced 100% of all the laws all of the time. If you want to automate enforcement, we're going to have to get rid of a lot of legal code.

> I could feel the outrage in your comment

I think you're mistaking a clue bat for outrage. You sound like me when I was in high school: high on concern, low on experience. I don't know if you're young or not, but I'm basically saying what I needed to hear back then.

Also, my comment was not intended to offer unqualified support for this program. But if we're going to debate it, let's debate it in terms of the real world and not in a philosophical vacuum. The real world has checks in place like wiretapping laws and evidence rules. Are these perfect? Maybe not, but you do not even acknowledge that they exist.

> According to the article, the vast majority of gunfire in a city doesn't cause death or major bodily injury. That looks pretty victimless to me.

Gunfire in dense urban areas is harmful even if it does not hit anything: http://news.ycombinator.com/item?id=4036259

> To me, it looks like the purpose of these urban firearm discharge laws are just to have another crime to tack on to a suspect when sending them off to jail.

Suppose you lived in a high-crime area. Are you saying that it's totally fine with you if people drive down the street shooting guns as long as they don't hit anything? If you hear gunfire, how do you decide whether you are in danger or not?

> If you want to automate enforcement, we're going to have to get rid of a lot of legal code.

There's nothing in the article AFAICS about automating enforcement, just detection. I'm not sure how you're going to charge someone with discharging a firearm without an officer showing up. How would you even know who to charge?

The ratio of random gun shots to injury in a city is above 1/1,000 shots fired even excluding intended targets so by your own argument it's rather dangerous.

Also, most peoples risk threshold is far below that for good reason. If you take a 1/1,000 risk of death per day (as a sum of all activity's) you have less than a 50/50 shot of living 2 years. On an hourly basis assuming 15 active hours a day 1/100,000 risk of death per hour = 50/50 chance of living 13 years. Really, 1/100,000 per hour is about the rational threshold for dangerous vs safe but 1/500,000 is closer to average.

As to speeding laws that's a side issue. With it's own set of arguments that has little to do with gun control.

> Although I imagine we'll just start seeing a lot of silencers

I wouldn't bet on this. Suppressors are highly controlled Federally, and completely illegal to own in Illinois (and Iowa and Michigan and a number of other states as well). You may think that doesn't matter to criminals, but most illegal firearms are purchased legally at some point-- not gonna happen for suppressors. And they aren't mix-and-match; you need a threaded barrel for your gun specifically. (There's a nice little scene in Goodfellas about this problem.)

But let's say you get ahold of one. You've got a line with a crooked Class 3 dealer in Indiana (this is someone who is also allowed to sell machine guns and grenade launchers) and he decides he's gonna risk his life by selling to a gang banger. Fair enough. So you get home, you load up your 9mm, screw on your shiny new suppressor, and go outside to take a potshot at some asshole you got beef with. Two minutes later, a couple of squad cars show up, lights flashing.

What happened? Well, you got your suppressor, but you're still using your same old bog-standard black-market ammunition. The can turns the hearing-damaging bang of your gun into a merely very loud bang, but it does nothing about the supersonic crack your bullet makes. You were quieter, but far from quiet; a ShotSpotter that was close enough easily picked you out.

If your dealer was looking out for you, he would have mentioned this, and recommended you switch to a subsonic ammunition. If he was really looking out for you, he'd also mention that this ammunition has significantly reduced range and stopping power, and probably does not have enough operating pressure to reliably cycle your automatic-- meaning you need to cock your gun every other shot like Jack Bauer. Sounds pretty much ideal for your inner-city gang fight, right?

And lest we forget, when the fuzz finally catches up to you, whether with the ShotSpotter or with good-old-fashioned police work, you are not going to get slapped with possession and sent upstate. As soon as that suppressor gets entered into evidence, with its nicely filed-down serial number, they're going to make a phone call, and the ATF is going to come to town with some serious Federal prosecution heat. They will hang you up by your toes until you tell them who sold the thing to you, and then they will find that person and make them severely wish that they had not.

So... yeah. Don't bet on it.

Suppresors are not hard to make. Due to selection pressures (jail/death) criminals in the aggregate are crafty. They will figure it out.

Have you ever made or used one? For the record, neither have I, but in the rural environment I grew up in, it was common 50 years ago to make them for poaching purposes, so I heard a lot of stories and spoke to many people who have made or used home-made ones.

For one, while it's not 'hard' in the same sense that writing an OS isn't 'hard' (after all, it's a widely studied design, many have been made/written, there is much information available, so how hard can it be to write an OS, really?), it still requires tools that aren't that commonly available (a lathe, for example - how many people do you know that have one and would be willing to make a class 3 device with it?).

But also, the home made ones aren't that good. Yes, you can make a silencer for a .22LR good enough to not be heard several miles away when you're hunting at night. But those purposes are very different from silencing a caliber powerful enough to be used against humans; meaning at the least a .38 but preferably 9mm. Furthermore it's not so easy to affix silencers to hand guns and still have the gun be concealable, you have problems with semi-auto weapons to get the next bullet into the chamber, etc etc.

So, all in all, having to use silencers, even if they would work 100%, would still cause a significant barrier for perpetrating gun crimes.

One of the primary business activities of street gangs are selling drugs smuggled in from Latin America. Do you really think that it's a stretch for them to import illegal gun accessories?

I think you severely overestimate the resources of most criminals. The majority are not the criminal masterminds you make them out to be. If they were that smart they would be doing something else.

And most of the market for weapons flow South; drugs flow north.

> And most of the market for weapons flow South; drugs flow north.

Oh really? "Weapons from the north" are semi-autos. They have autos. (No, they're not convertible.) They get them from their military (which is supplied by the US) and similar sources.

A source that says "x% of traced" doesn't tell you how they decided to trace. (Hint - they have an agenda.)

> They have autos. (No, they're not convertible.)

It's a side point, but many semi-automatic weapons can in fact be converted to fully-automatic mode with drop-in parts. Makes sense; if you think about it, it's actually harder to make an auto stop firing after the first round.[0] Manufacturers do take steps to make the parts incompatible, but only with limited success.

Of course those drop-in parts themselves are considered to be machine guns by the ATF, and are thus absurdly expensive and hard to get ahold of. A fully legal and transferrable drop-in auto sear[1] for an AR-15, which is a few ounces of machined steel that lets you use the automatic M-16 parts, would have had to be made before 1986, and these days goes for >$5000.

[0] See http://en.wikipedia.org/wiki/Sputter_Gun . The ATF's original definition of a machine gun was a firearm which shoots more than once with a single trigger pull. Someone entirely too clever for his own good came up with the idea of making a gun with no trigger, which fired continuously so long as it was loaded. These were, for a brief period, completely legal to build and sell (although none were sold).

[1] http://www.quarterbore.com/nfa/dias.html . See also the Lightning Link, http://www.quarterbore.com/nfa/lightninglink.html which is literally just a couple of interlocking strips of metal. Cutting a strip of metal in that way is a Federal felony that can nab you >10 years even if you do not possess the actual rifle. They really do take this stuff very seriously.

> Of course those drop-in parts themselves are considered to be machine guns by the ATF

In other words, it takes a machine gun to "convert" an "not machine gun" into a machine gun.

Yes, machine guns are easy to make. However, the claim is that mexican drug gangs, which have automatic weapons, are importing semi-autos.

I note that they are importing US made cars and using them. What should we do about that?

If they are getting weapons from the military (and they probably are as well since the Zetas are ex military) then why are they smuggling in from the US? [1]


This is a good point, but as my sibling points out there are suppressors and there are suppressors. It's certainly well in the ability of a competent amateur to make an effective suppressor for .22LR -- I understand you can use PVC or such materials -- but try that with a larger caliber and it'll take your finger off with a decidedly non-silent noise.

You're going to need a lathe, at least -- a machine shop would be better. Even if you copy an existing design (the science is mostly understood, but it's still easier to make one that doesn't work than one that does), the metallurgy is important-- you really need someone who knows what they're doing.

And that person had better hope that no one outside of the crew knows that they know what they're doing, because as soon as those off-brand suppressors start showing up on the streets, finding that person will become the number one priority for a number of very highly trained forensic experts. They really do take this stuff very seriously.

Slippery-slope arguments come into play almost any time you use technology to solve social problems. It doesn't matter that the slippery slope is a logical fallacy, if the entities involved (i.e., people) aren't logical ones. "Libertarians" harp on this stuff because historically, the cost of letting governments do anything they want runs into the hundreds of millions of lives.

A more enlightened approach would be to ask why Chicago has so much trouble with gun crime, given that many other American cities have higher per-capita rates of firearm ownership both legal and otherwise.

One thing's absolutely certain: the real problem, whatever it is, will not be solved with digital signal processing.

I listened to a set of lectures from the Teaching Company a while back on the debates that occurred when the United States ratified its constitution. The opponents made all sorts of slippery-slope arguments: that there would be a standing army, that the federal government would grow past it's ability to pay (which the supporters actually acknowledged), that the Congress would become politicians for life, and so on.

At the time, these guys were dismissed as being totally wacko (my words, not the lecturer's). What kind of ridiculous slippery-slope arguments were these guys making, anyway?

Of course, all of these predictions did come true, it just took a long time. But when they were made it was all too easy to discount them as being far-fetched. This trend in public discourse continues to this day.

If a founding principal of the United States was that legislators shouldn't become politicians for life, the US Constitution could have included term limits for the legislative branch. How seriously are we to take the "debates that occurred when the United States ratified the Constitution"? The opinions expressed in those debates aren't laws.

> I listened to a set of lectures from the Teaching Company a while back on the debates that occurred when the United States ratified its constitution.

I'd like to listen to these lectures. Please try to remember what they're called. Thanks!

I'm guessing he's referring to a course being offered on the Federalist Papers, for what it's worth.

This system isn't deployed in Chicago; I've used Chicago as an example because I live here and am familiar with it. Every major American city has a gun violence problem.

I don't understand your glib "digital signal processing" sentence. Technology won't help police forces become more effective? Strong disagree.

Technology won't help police forces become more effective? Strong disagree.

It's not a "police problem." It's a cultural one.

I think you're being unfair. The statement could have come from anyone. Moreover, in the article it is not associated with any particular group. Why do you have to point at libertarians, or any other group for that matter?

I've kind of been waiting for this. This could provoke really interesting renaissance of the crossbow among street gangs. And other deadly melee weapons might get big leaps of progress. There might be a telescopic sword coming or something similar.

Here's something from India. From times when weapons had to be carried secretly. https://en.wikipedia.org/wiki/Urumi

My forecast is that in the long run, bystanders will get less damage. But real murders get harder to investigate as there will be no gunpowder residue. And crossbow is easier to manufacture by oneself than firearm and that makes murder weapon tracking harder.

I would bet on bows before crossbows (speed, easier to hide). Quite a few hunters here are learning bow because the number of rifle deer licenses this season have been cut in half, but a lot of bow licenses remain.

You don't have to have a leaf spring in the bow as is custom right now. With spiral spring, a "crossbow" can look more like a big pistol.

I think people should consider the upsides to these technologies in addition to the privacy concerns. Faster police response could lead not just to saved lives, but also to much better suspect identification. Our current methods for identifying suspects using evidence gathered long after a shooting hall are complete trash. Of the wide array of forensic techniques used by police, only DNA is remotely reliable. If a system like this meant the difference between apprehending a fleeing suspect a couple of minutes later and picking up a random ethnic minority days later based on highly flawed identifications, this technology could have major benefits that must be weighed.

Potential upsides. Rarely realized.

I'd rather pay for more cops.

These technologies are just used to put more distance between the corporate profits and accountability to the communities effected.

As a history lesson, dehumanizing intelligence services, preferring satellites and wiretaps to HUMINT, has crippled our capabilities. But it did enrich defense contractors, so it's all good.

This whole system deployed at scale appears to cost 0.5% of what patrol does. There's no reason cities can't do this and ramp up patrol.

Paying for more cops has its own problems. In a high-cop low-crime environment, e.g. most suburbs, cops spend most of their time harassing people for minor things or figuring out ways to increase ticket revenue.

I find it interesting that people's main concerns are the privacy laws. Isn't anyone concerned that there is a need for gunshot detection? I would have thought that the main issue was that there was enough gun related violence to make this product viable? As much as I love to see an innovative solution to a problem this seems a lot like its treating the symptoms and not the cause.

Won't somebody think of the children? The problem is that there's always a good justification for reducing rights. It makes law enforcement easier. We need to fight terrorism. We need to stop child rape. Etc, etc, etc. But if you let your rights be stripped away, even for a purportedly good reason, you aren't getting them back.

I enjoy following these structures of cludge upon subsidy upon moral bankruptcy.

I wonder at what point (in the adoption of this technology) it makes sense to invest in gun silencers.


I'm sure that silencers will still be detectable? They do make the gun softer but it's still noticeable if you know what you're looking for I'm sure.

"If nothing else, ShotSpotter has made it clear how much unreported gunfire takes place on city streets. In many high-crime urban neighborhoods, gunshots are a counterpoint to daily life, “as common as the birds chirping,” as Commander Mikail Ali of the San Francisco Police Department put it. But whether out of apathy, fear or uncertainty, people call the police in only a fraction of cases."

Or maybe, they got sick of calling the cops and not having them come. If the gunshots are that common, it wouldn't be uncommon for the police to ignore the complaints as a high-crime district.

Now they can ignore the automated reports instead of the called-in ones.

There's a potentially huge difference: if these outside company reports exist independently of the official PD story it becomes much easier for a reporter to start asking why the numbers are much higher. If the police are under pressure to get their stats down (i.e. cook the books) there will probably turnout to be a number of real incidents which were described as mistaken reports

I think the assumption with this is that all gunshots would be crimes in process and over time and with more ubiquity that might not be the case leading to a less effective system.

I would imagine that a criminal could use a system like this to do a DOS of police presence and/or send them on a wild goose chase so they can commit a crime in another area. In other words they get police to scramble to 5th and Main and they have another team to break into a place a few miles away with a non-gunshot crime.

Also, as anyone who has experience with home or business alarm systems knows false alarms get to be a "boy who cried wolf" issue and you begin to not take them as seriously as you should.

The false positive rate with home alarms is not the same as that of acoustic gunshot detection alarms. Your cat will routinely set off your interior motion sensor alarm; a loose door fitting will do the same to the front door.

If criminals want to use gunshot noise to scramble police presence to a specific area, they can obviously do that without gunshot detectors. The point of detectors is to make it easier to follow up on ideopathic gunshots.

You might also note that they did not give any sort of false positive rate - they mentioned how often it correctly identified gunfire, but not how often other sounds were incorrectly labeled as gunfire. Since gunfire is uncommon, even with a fairly high accuracy rate, it is still likely to wrongly identify other sounds as gunfire more often than making accurate identifications.

If you had carefully synchronised low intensity sound sources close to all detectors, you could certainly simulate a gunshot and choose where it was triangulated to by controlling the timing and intensity of the sounds emitted by your multiple sources. Such a simulation could certainly be used to trick the system into thinking that there are hundreds of simultaneous major gun fights going on - which would obviously inhibit the ability of the police to respond to a real one, with a lot less manpower than calling in so many fights.

However, it would probably be more effort than it would be worth, so it is probably more a sci-fi / crime plot than a scenario that is likely to come up often.

What's stopping people from doing the same via 911 right now?

Nothing. That's why police usually sends some patrol to the site to check things up.

If you would get a signal of multiple assault weapons firing at the same time, you would not send just one patrol.

It's not very likely. Unless someone manages to simulate gunfire accurately with a boombox or with a pipe and some layered gunpowder + lead shots.

I still think it would be far easier and more obvious to have a few friends phone in diversions. What you're describing might feature in a Batman film with a master villain!

That would only be effective if you could get all the police to go to a specific location. I don't think they typically send every single available police officer to the location of a gunshot.

"get all the police to go to a specific location"

Maybe. But I think it would depend on the particular department obviously and geography as well. You don't need all the police. You are decreasing manpower and ability to respond to other calls which can vary greatly depending on what else is going on. And this system might be installed in smaller towns with less backup resources as well.

It's not that hard to put a few timers on few automatic arms.

Reading about this in a country where there is tight gun control feels like reading about a dystopia future by Gibson or the like. It is the future, but not necessarily in a good way.

I would imagine ND has more firearms than Chicago with a smaller total population. FBI[1] has listed 9 murders in 2009 and 3 were firearm related. Of the 3 in 2008[2], 0 were firearm related. Gun control doesn't stop people from killing others, it just makes them use something else. The something else can be a lot worse.

1) http://www2.fbi.gov/ucr/cius2009/data/table_20.html

2) http://www2.fbi.gov/ucr/cius2008/data/table_20.html

I am not in the mood for a debate on gun control. I do agree that guns aren't a prerequisite for violet death but they do make the process significantly easier given they are designed for that purpose.

Your example compares apples with oranges.

The important question is "why?" not "how?" - concentrating on any tool is missing the real problems and solutions in life.

The tools isn't Chicago's problem or else it would cause the same problem in ND. My cousin is alive because he had a shotgun on him when attacked by abandoned dogs. I don't blame the dogs ("how"). I blame the owner's who abandoned them ("why").

If they designed the system (hopefully at the hardware level) so that it was incapable of recording voices I would have no problem with it.

That would probably be a good thing for a competitor to create: a shot detection system that's designed to address civil liberty concerns. I bet someone on HN could do it.

If you do that simply by damping wavelengths of speech, you possibly make opportunity to alter the sound of a gun so that it doesn't set the alarm of. Maybe more importantly the sound of a gunfire would not sound like anything familiar to human ear, so it would make it very difficult to manually pick the real shots.

But the problem is how can you verify that the system that is actually installed on all the building and telephone poles is actually one that is incapable of recording voices instead of a (really cheap) normal microphone.

I'm confused about the timeline in the NYTimes article. The gunfire was detected at 7:22:07, and tactical team arrives at 7:25:02 (Pacific). The article says, "Total elapsed time: 3 minutes, 55 seconds."

Isn't that a total elapsed time of 2 minutes, 55 seconds?

I wonder if false positives will take off e.g. personal attack alarms that make a sound like gun fire

My dad was using a powder-actuated concrete nail gun for some remodeling not long ago. It used .22 loads to drive nails. The sound it made was eerily identical to the real thing.

The system will have to distinguish real firearms from other powder-actuated hardware. Otherwise it's going to become useless anywhere near a construction site.

But what if you shoot a person with a nail gun?

> But what if you shoot a person with a nail gun?

What if you stab a person with a knife?

It's off-topic. It was designed and built to detect gunshots from real guns.

I think the best way to distinguish between firearms and nail guns would be with a man in the loop listening to the sounds (which they appear to have). Construction doesn't sound like a firefight.

Wouldn't it make more sense to make the alarm send out a standardised frequency a la DTMF tones?

As a libertarian, I'm generally iffy about the expansion of surveillance, or of government activity in general.

But I try to be pragmatic about it. Gunshots fired in urban areas are one thing that really is the business of the police! There's always some tension between law and order versus liberty; the trick is to find ways to trade the latter for the former at as high an exchange rate as possible. This system strikes me as having a very good exchange rate indeed.

$40,000-$60,000 per year per square mile. How many gunshots per capita per year per square mile do need for that price to make sense as a way to spend tax dollars?

It seems expensive.

It is a tiny fraction of the cost of police patrols in probably any mid-sized metro area in the US, and an even tinier fraction of that cost in major metros.

On the other hand, a full police force can handle more than murder. This can't. That would be okay if murder was the function of a police force, but its really just another amongst many responsibilities of theirs.

Are we talking past each other? Note the words "tiny fraction".

The city of New York is 303 square miles. So for $12-18M, NYC could be completely covered by this system.

I suppose it depends how much time it shaves off of investigations.

Furthermore, I'm assuming they're not putting these in the middle of a Kansas corn field. Three people got shot outside of my apartment last year, I imagine they're going to prioritize that area over the upper east side.

That's probably a single police officer's salary, and it's probably more effective than adding a single officer to patrol a high-crime area.

I'm not 100% on the laws around this, but if a city was utilizing ShotSpotter, wouldn't they have to disclose the area(s) they were monitoring?

My city (Springfield MA) has a shotspotter system, and even though I have pretty high-up contacts among Springfield PD who know that my interest in the system is purely academic (I'm an acoustical consultant) they wouldn't reveal to me any of the transducer locations. In fact just trying to find a picture of a shotspotter transducer is difficult (the pic in the NYT link is only the second one I've seen).

The sensors are very easy to spot. They are small, white/pale-grey boxes about 4 inches square by 2 inches deep. They're usually placed on stop light struts or on free-standing poles by themselves. They're easy to mistake for emergency vehicle "light changer" systems, but the difference is these are on roads where there are no stop lights. Also, I've seen clusters of 3 or 4 of them attached to the same pole that a camera is attached to -- presumably the camera automatically pans and zooms in on the area as soon as a shot is heard.

They are small, white/pale-grey boxes about 4 inches square by 2 inches deep.

Are you sure about that? That doesn't jibe with the picture I've seen (http://i.imgur.com/XBevy.jpg) and doesn't sound optimal for an acoustic transducer.

I'm 99.9% sure that is NOT a shot spotter. We have Shotspotter in some areas Minneapolis, as well as city-wide wifi. Those little cans are all over the place and are part of the WiFi System. I took a photo of one just down the street from my house: http://i.imgur.com/OqYEA.jpg

The can looks like a BelAir Networks BelAir200 WiFi Node.

As another poster mentioned, that's most likely a wifi access point.

I spent the last few minutes poking around StreetView looking for a photo of the units I'm thinking of without luck (the dates Google lists on their StreetView seem quite incorrect), but a hard, flat surface would be perfect for a sound transducer. It'd also be easier to weatherproof than something with microphones dangling outside.

I spent the last few minutes poking around StreetView looking for a photo of the units I'm thinking of without luck

In my city I know that they periodically move the units so it may be that they were moved when the pics were taken.

but a hard, flat surface would be perfect for a sound transducer.

I would be surprised if the mic was embedded in a flat surface, that would cause all sorts of diffraction problems for a surface that wasn't (effectively) infinite, especially since there is a lot of high-frequency energy in gunshots. I suppose they could try to equalize them out, but that seems like unnecessary work. But again, I haven't seen a (confirmed) transducer.

It'd also be easier to weatherproof than something with microphones dangling outside.

edit: actually I suppose the mic could be offset just an inch or so from the enclosure, that would be enough to reduce diffraction problems while not being noticeable from a distance. I'd love to see one of these systems up close. That problem has long since been solved with metal diaphragm mics, especially at the price points they're targeting. Microphones in airport monitoring systems last for years.

edit 2: some more searching indicates that a unit is actually made up of an array of mics. In that case, a flat surface would make perfect sense.

my understanding is that anyone can record any visual or auditory info from any public place (in the US)

Unless you use a telephone to do it or are using a telephone when you're being recorded?

I understand that making a recording of a phone call without notification is strictly disallowed in many states legislation? (eg http://en.wikipedia.org/wiki/Telephone_recording_laws#United...)

Telephone systems are a public place?

If I'm walking down the sidewalk talking on my phone, overhearing my voice is not a violation of privacy.

>overhearing my voice is not a violation of privacy //

But we're talking about recording your voice. Which I think is a violation of my privacy vs just listening to me passively - now that may not be enshrined in any law but then the law doesn't IMO [exclusively?] dictate what is moral.

Think point is. Unless that voice is carried over telecommunications network to, say central processing center.

>The detection system, which triangulates sound picked up by acoustic sensors placed on buildings, utility poles and other structures, is part of a wave of technological advances — among them, license plate scanners, body cameras, Global Positioning System trackers and hand-held fingerprint identifiers

This technology also happens to be very useful for domestic counter-insurgence, something of interest to an economically polarized state during a time of economic transformation.

No, it really isn't. The first step in any insurgency is to compromise intelligence and surveillance systems, and these sensors are very fragile sitting ducks.

Insurgents do not attack prepositioned listening posts. Their tools are randomness and radio silence.

I wonder if this could also detect car crashes?

It is a fairly distinct sound. You could probably scale this out to a bunch of other uses. Detect for; cries for help, cars doing burnouts or loud parties at inappropriate hours. It would combine really well with CCTV if it could alert an operator to look in the area of a sound rather than have to send a patrol out for minor things.

The movie Minority Report is starting to look a hell of a lot more real. I didn't even know this technology existed, it's awesome but at the same time scary. I wonder what other technologies are out there like this nobody really hears about?

I've always wondered if a similar system could be developed to monitor and locate the sound of windows being broken?

You could have one in every neighborhood. It might also be good in parking lots/garages.

Another interesting application of this (which I believe is being done) is sniper spotting in warzones.

I've heard of it being mounted on humvees. This article says there are hand-held versions as well: http://www.defenseindustrydaily.com/scaneagle-shotspotter-sn...

They have to replay the sound because the gunshots are indistinguishable from the sound that lady's bones make when she contorts herself to sit at the terminal. (Look at her arm.) What an ergonomic nightmare.

phased arrays of microphones: the future is now

Time to hack it: Drive around with a 1963 Oldsmobile and get pulled over immediately.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact