Police are telling ShotSpotter to alter evidence from gunshot-detecting AI | Hacker News Reader

Police are telling ShotSpotter to alter evidence from gunshot-detecting AI

vice.com

516 pointsdanso334 comments

> Greene found a fifth shot, despite there being no physical evidence at the scene that Simmons had fired. Rochester police had also refused his multiple requests for them to test his hands and clothing for gunshot residue.

This is the same that giving queues to a dog to bark to be able to inspect someone or a vehicle. New tech, old tricks. No technology will work if the people using it is corrupt.

It's no wonder why the cops love it. Although the actual usefulness is questionable at best (much like the dogs) ShotSpotter will cook up "evidence" on command to help with the investigation (much like the dogs).

Any lawyers in the house who could comment on their refusal to test his hands?

Can they refuse to run a test because they're afraid it might produce exculpatory evidence?

Not a lawyer, but in my experience they don't need to do any tests. They usually have a policy that requires them to make a "thorough" investigation. Nobody holds them to that. For example, a trooper knew there were third party witnesses to a crime and did not seek them out. Granted, that trooper also withheld actual exculpatory evidence. Either way, nothing happened, because who watches the watchers?

IANAL, but curious and searched.

> A number of techniques designed to detect gunshot residues (GSR) on the hands of a suspect or victim have been developed. These techniques range from the now-discredited paraffin test to the more modern techniques which use instrumental analysis or scanning electron microscopy. The limitations of all GSR techniques are that the residues can be removed by rubbing or washing the hands and usually must be collected soon after the firearm is fired, yet even valid GSR tests are not conclusive.

https://www.ojp.gov/ncjrs/virtual-library/abstracts/forensic...

Very few forensic techniques are conclusive for a case. They are usually conclusive as in the test was either positive or negative. Then the forensic expert has to give testimony as to what it means. Just look at the medical experts in the Chauvin trial. It's just like many news articles today - the facts don't lie, but people can misapply them to say different things.

The Supreme Court ruled that 5th amendment protects you from incriminating yourself even if you're innocent, and in some states that extends to investigations of your body.

Never talking to or cooperating with police, unless legally required as advised by your lawyer, is what's known in legal circles as "smart."

That's not applicable here, because the defendant was asking police to timely collect additional evidence for its exculpatory potential.

I was referring to the suspect declining to be tested for gunpowder residue, and the ensuing tangent.

Not being in or having interactions with the criminal justice system, but having been involved in _many_ technology post-mortems, why wouldn’t you want to collect exculpatory evidence? Without it, and in the best case an innocent individual goes free, but with a cloud of uncertainty. In the worst case an innocent individual is convicted of a crime and the perpetrator goes without consequence.

If LEOs and prosecutors are incentivized based on raw arrest or conviction counts, then the consequence for being wrong should be astronomical (say, the sentence of the wrongfully convicted times a ≥ 1x multiplier).

If you start from a premise that law enforcement is searching for the truth above all else, your conclusions make sense.

I would think that as part of enforcing laws, truth seems like it would be relatively important.

The idea of incorrectly convicting someone for a serious crime not only is an injustice to the innocent, it leaves the community in an unsafe and unwitting position regarding an actual perpetrator. In those cases it would be better to have no law enforcement involvement at all.

The aftermath of many crimes, in America at least, would have been better with no law enforcement involvement at all. It's really that bad here.

Anyone who is acting to infringe upon real justice is most apt to either be a sociopath who only cares about keeping score or a lot more likely a moral person but not ethical person who regards their job as getting criminals.

If they truly believe you are guilty then convicting a larger portion of guilty people even via impure means is a net good for society and incidentally for their career.

"... why wouldn’t you want to collect exculpatory evidence?"

In my experience, it's mostly that they are lazy, incompetent, and/or have their mind made up. In some cases, it has to do with costs (eg they aren't going to DNA test a petty theft crime scene).

Contrary to TV, forensics are actually used very rarely in the criminal justice system. Most of the ones that are used are fingerprints and drug tests.

Nit: a queue is a line waiting for something (stuck in the queue), a cue is a signal to perform some action (that's my cue)

Eh not a nit - I didn’t understand it until your context was added

Yeah, I had my (completely clean) car searched because "the dog signaled". OK, sure.

It would be more honest and to just say cops have the right to search you at will.

> It would be more honest and to just say cops have the right to search you at will.

Except they explicitly do NOT have that power. They found a "trick" to pretend like they are staying within the law, while really not.

Exactly. Hence

> It would be more honest

They don't have the "right" to, they have the "ability to do so without repercussions".

There is a view, called “legal realism”, that what the courts and law enforcement collectively do _is_ the law itself. The rules as they are written down are useful in predicting or convincing how judges or other actors in the system will act, but that it all they are.

What they should be required to do is keep records on a dog's record of false positives.

If a dog fails to perform better than 95% I see no reason to use that as probable cause.

Better yet, let's just get rid of pretext stops. Almost all traffic stops could be dealt with by the officer mailing a ticket. Safer for cops, safer for the public, but a lot harder to go fishing for felonies

They treat K9s almost like officers (eg if you injure or kill one it will be a felony, likely similar to assaulting an officer, but varies by state). They should also then treat them like officers and keep Giglio records (not that they are good at that for human officers...).

Unless you're a cop, in which it's usually just a minor disciplinary infraction.

The Supreme Court ruled, unanimously, that this sort of thing isn't required:

https://en.wikipedia.org/wiki/Florida_v._Harris

It's going to take actual legislation to get this fixed, which is not something I think we have any reason to be optimistic about.

It's true, but it's also true that it's important for people to say "no" and assert their rights. Granting permission to search or willingly waiting for the K9 (assuming it's not already on the scene) eliminates a lot of recourse that people would otherwise have access to in the courts.

> or willingly waiting for the K9

Is there an option there? When the cop says "we're gonna get a dog to sniff your car", is it actually within your rights to just say "sorry, but I don't feel like waiting for a dog to get here, cya"?

Edit: huh, looks like they can't hold you for a K9 unit to arrive, assuming they have otherwise completed the traffic stop (based on the original purpose of the stop) and don't have reasonable suspicion to hold you longer: https://en.wikipedia.org/wiki/Rodriguez_v._United_States

(In the Rodriguez case, the lower court unfortunately still ruled against Rodriguez because the cop, based on the common understanding at the time, would have reason to believe that prolonging the stop was legal.)

> is it actually within your rights to just say "sorry, but I don't feel like waiting for a dog to get here, cya"?

Rather, just bluntly state that you don't consent to being held and ask if you can legally depart. If they tell you to stay, sort it out in court. Doing anything else seems like a really bad idea to me.

No amount of new technology, training, equipment, or funding with change this. US policing is fundamentally a broken system.

It still blows my mind how police are allowed to turn off body cams while on the clock. That should be considered an offense like a government official burning public documents, because that's basically what is going on when you turn off the video. It's uncanny how the cameras are almost always off whenever something controversial happens.

It depends on the jurisdiction. In some police departments, turning off a body camera is a punishable offense, up to and including firing.

It's up to your city leaders to enforce this, and override the local police union's resistance, if possible.

There has to be bathroom breaks right?

From what I've seen, the cameras don't really catch those angles to give video on that, and even if they did, that's a problem easily solved through a piece of fabric around the problematic area (as opposed to on the camera, which could be accidentally left in place). Beyond that, requiring sounds from the bathroom be audible is not too much to ask, IMO.

If it came down to it being absolutely essential, you could have one office take the camera off (but leave it running), and hand it to their partner to point at the bathroom door as that officer enters and uses the bathroom, and they put it back on as they exit the bathroom.

The whole point of a camera system is to be a reliable audit trail of interactions that police have been in, for both their benefit and the public's benefit. Every exception you allow that lets things happen without the camera is a problem for an audit, so you try to reduce those as much as possible.

"If it came down to it being absolutely essential, you could have one office take the camera off (but leave it running), and hand it to their partner to point at the bathroom door as that officer enters and uses the bathroom, and they put it back on as they exit the bathroom."

This is probably the most foolproof. The fabric idea wouldn't be a good idea considering many public restrooms are multi-person. So the concern is with the privacy of others, not just the officer.

That's a good point. At the same time I'm not sure what level of privacy people expect in a public bathroom. Anything you don't want someone seeing should be done within a stall. At best I would consider a public restroom as having increased privacy, but by no means being private.

I understand that's probably a hard sell though, as people have some idea that it's verboten to record in the open space of a public restroom (it's definitely frowned upon, but I think police cameras are a good exception to have, especially if they need to enter the restroom to perform their jobs because that's where a disturbance is), so whatever gets more camera use in the end I'm on board with, from a practical standpoint.

"At the same time I'm not sure what level of privacy people expect in a public bathroom."

They expect that the law should be followed, which stipulates no recording. Companies have tried for a long time to have security cameras in bathrooms, but it's not allowed. Now if they're responding to a call, then maybe that's an exception. Or they could change the law to allow recording in the bathroom, but there will probably be a ton of public pushback.

I'm not saying the law should be ignored. I'm just not entirely sure why the law exists as it does, or think that it's ill written to account for certain situations. A large bathroom with twenty people (or more! bathrooms at large events or in large airports can be big) in it is not what I would consider private.

I think (but could be wrong) what most people actually care about is that their stall is private (or in the case of urinals, that someone isn't to the side recording what they consider a small private wall space area), but since that's hard to account for and some stalls/urinals already have poor privacy to the rest of the bathroom, we got what we have now.

Honestly, to me it seems like a lot of bother to hide bits of our body that we should all be a lot less concerned about shame about in the first place, but I'm not holding my breath on any change there in the U.S.

"or think that it's ill written to account for certain situations."

I agree with this, for many laws really.

Perhapes we should just require the camera to be turned on and on their person for them to have the legal privileges and protections that we grant to law officers. If they turn it off or leave it behind for example during a restroom break then they are, under the eyes of the law, a civilian until it is turned back on. and any evidence "found" somewhere that they had gone without the camera while off the clock as it were is either inadmissible or must have a hire bar to be able to be used. so on planting drugs on a piss break to just walk in again with the camera on and "find it"

This seems like the most straightforward way to handle this.

> At the same time I'm not sure what level of privacy people expect in a public bathroom. Anything you don't want someone seeing should be done within a stall.

If a man at a baseball game went into the restroom and up to the trough holding a smartphone that was clearly recording, he would undoubtedly find trouble. People expect not to be recorded in the bathroom.

It depends, I suppose. Others might egg the guy on to do something really stupid. Just search for Wrigley Field trough diving...

I would rather there be a way to turn off audio than have the camera come off its officer.

It's enough to have the ability to turn off audio. That's useful for restroom breaks, but also for when officers are not interacting with the public, since they should have some privacy for cooler talk (well, maybe they should, or maybe they shouldn't, but I've not thought enough about that myself).

I would say it should be off for bathroom breaks of course. The main issue isn't these edge cases, but that these cops are shutting them off when they are approaching a person. Its not like they were taking a poop then the call came in and they had to pull up their pants and run and didn't have time to flick it on. They see a situation about to play out where they could end up looking bad in a court of law so they try and remove a potential source of evidence that could be used against them.

Sometimes there are also privacy laws that would need to be addressed at the state level. For example, many places prohibit the camera from being active in a private residence unless permission is granted by the resident.

Unless police see what they believe is a crime in progress, they aren't supposed to enter a private residence without permission anyway, right? I'm not sure how that can't be easily dealt with even in the problematic states by saying that permission for the police to enter the premises is also permission for them to record.

They could be responding to a non-criminal call, like a welfare check. Permission to enter isn't considered the same as permission to record in some states.

Yes, they could change the laws about recording. In many cases, not just for police, they probably should.

> They could be responding to a non-criminal call, like a welfare check.

Armed law enforcement responding to calls without an evident need for armed law enforcement is itself a problem.

Probably true, and goes to complaints from police officers that they're also expected to be counselors as well as enforcers.

On the other hand, asking someone to go into a situation that may put themselves at risk (for some cases of welfare checks) without appropriate training or equipment (or leaving the distinction of which is which to some third party not on the scene) doesn't seem like a good idea either, and at the point where you have a force that's equipped to protect themselves and possibly restrain others, that begins to sounds a lot like a police force, so I can see why they just use the police.

There's probably some solutions along the lines of different shifts for those with different training with different load outs, or additional trained personnel in sets of police, but all of those also have some problems I can see and don't address officers being dispatched that are prone to use force because of a prior altercation that day/week or trauma at some prior date.

> Armed law enforcement responding to calls without an evident need for armed law enforcement is itself a problem.

It’s fairly obvious that in a country where citizens are free to be armed, police must be armed too.

I think what's being discussed is not that police should be unarmed, but that police should not be the first choice to respond to situations where they currently often are.

Yes, police should not be the welfare patrol. However they are, because they are out and about and often would be the nearest public authority to respond to such a need, and governments don't want to fund the social programs that they legislate. It's the same reason teachers are imposed upon to be defacto child-welfare social workers.

Police should enforce the law/investigate crimes.

Teachers should teach.

If we're going to have social welfare programs, we need to provide personnel to do that work, not pile it upon other professions as an unwanted secondary responsibility.

> we need to provide personnel to do that work, not pile it upon other professions as an unwanted secondary responsibility

There could be police who want it as a responsibility as long as they are trained and compensated. That is another possibility.

We can do what we want with our own money, we don't have wait for politicians to allocate it. Give directly to causes that help people. People have the money (power)! Tax yourself! Fund social programs that replace this function of policing directly. If government does not work we can route around it. Create Kickstarters for good causes.

> It’s fairly obvious that in a country where citizens are free to be armed, police must be armed too

Even granting that for the sake of argument, it is less obvious that government-public interactions not for the purposes of arresting criminals or otherwise responding to or preventing apparent actual or imminent crime need to involve police.

Totally agreed.

That said, there is no also no reason at all not to have specialized police who do actually have mental health and de-escalation training.

But again I agree - there should be other options for government/public interaction.

Firing isn't decertification and the punishment isn't even that bad.

>It's up to your city leaders to enforce this, and override the local police union's resistance, if possible.

It's only possible in jurisdictions where DAs and/or Mayors can get elected without an endorsement from law-enforcement, and look at SF for an example of what police do when one slips through the cracks.

By far, most US government is ruled by law enforcement, not the other way around, and this isn't even touching the nationwide police rebellion we've been enduring.

> It's only possible in jurisdictions where DAs and/or Mayors can get elected without an endorsement from law-enforcement, and look at SF for an example of what police do when one slips through the cracks.

As someone who doesn't live in SF, what are some of the examples? This is a good faith question, but I'm not really sure what I would google to know what you mean.

The 1975 SF police strike, which ended when the police detonated a bomb at the mayor’s house.

Wow, I'd never heard about this, but just read the Wikipedia blurb[0]. It doesn't appear to be established fact that the bomb was indeed placed by the police, but it doesn't seem like it would have been surprising if they had.

Pretty shameful that Mayor Alioto override the Supes and caved to the police's demands rather than calling in the National Guard and taking the entire striking force into custody. But I guess people don't always make the best decisions while their lives are being threatened...

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

Google "Chesa Boudin," the current DA, who is ensuring that criminals run rampant throughout the city, if not inviting them to do so, if you listen to certain opinions.

Find a news story about the current crime rate in SF that isn't framed by comments from carceral interests: the police, the police union, former police officers, former hardline prosecutors, and aspiring hardline prosecutors. Many if not most stories only talk to those people.

You may be surprised at the difficulty in finding an even-handed appraisal.

As a non-SF resident, doing useful research inspired by 'google the current DA' is pretty laborious. I found https://www.sfgate.com/politics/article/Chesa-Boudin-San-Fra... which seems sensible on the face of it, and doesn't sound at all like he's "ensuring that criminals run rampant throughout the city".

The general problem of media assuming good-faith information from police is widespread and SF is no different from any other location in the US (or Canada for that matter).

Sorry about that, I must have assumed the story was as big everywhere as it is here (I know).

This is probably a decent start: https://www.washingtonpost.com/opinions/2021/06/14/bogus-bac...

In theory, there should be no controversial actions when the cameras are on, for the simple fact that everyone can see/hear what happened. The policy can be controversial, but whether the actions were consistent with policy should be evident.

There are also potentially privacy considerations... for example, in jurisdictions where body-cam data is a matter of public record, should officers be recording while inside someone's house where that person has a reasonable expectation of privacy?

And even assuming the footage isn't automatically public, what about the evidentiary value against the house's occupants? Currently, police need a warrant to execute a search of someone's residence, but there's a "plain view" exception: if while the officer is in the house anyway (either because the occupant let them in, or because they entered due to exigent circumstances like a violent crime in progress), if they happen to see evidence of a crime in plain view while there -- a gun, drugs, whatever -- they don't need a warrant to use that as evidence. If the officer has a camera on the whole time, that potentially means the officer could use not just what they happened to see in the moment, but retroactively whatever can be seen in the footage after the fact, which likely significantly expands the extent of plain-view evidence that can be used without a warrant. Some have asserted that video-recording itself constitutes a fourth-amendment search if the officer is someplace where the person has an expectation of privacy, though I don't think this question has been adjudicated.

Also, in two-party-consent states, capturing body camera audio might also require either a warrant or consent for the capturing of at least some kinds of interactions.

I think body cams are net-good, but it's more complicated than it seems at first blush.

> for example, in jurisdictions where body-cam data is a matter of public record, should officers be recording while inside someone's house where that person has a reasonable expectation of privacy?

Public records laws already generally anticipate that public records will frequently contain sensitive, protected data and have general processes (involving redaction or denial of unprivileged general public access, depending on the circumstances) for handling public records to which that applies (source: spent a couple decades in government dealing with public records containing PHI, other PII, and other legally restricted-access information.)

If you listen to a police scanner for a bit, you'll probably hear at least a few social (security number, full name) pairs. If you're in range, a $20 SDR dongle let's you listen and record on your own, or you can just google your way to an online stream from someone else with an SDR.

If you watch BWC footage on the completely public COPA (the civilian office of police accountability, an organization empowered by the govt of the city of Chicago to investigate complaints against CPD) case portal [0], you'll see a lot of intimate, unredacted footage, BWC and otherwise.

I don't think the privacy laws have caught up with the tech. Or maybe the privacy is sacrificed in the name of transparency and accountability. I was not in the room where these decisions where made, so I couldn't say.

[0] https://www.chicagocopa.org/data-cases/case-portal/

At the very least if any situation requires official capacity, the officer shouldn't have the camera off. The camera should have features like "toggle off for 5m" or "turn off persistent but you can't act as LE".

The situation remains that law enforcement has no real checks on their own power, and they've moved to restrict or block such checks on their power.

Right, but also consider that what you're describing is a matter of law, and we're already talking about changing law (to require body cameras to be always-on), so why not change the other stuff too?

Change the "plain view" exception to only apply to stuff the officer notices while physically there; if they see it in footage later, it's not admissible in court, and not grounds for a search warrant. Change the recording-consent laws to not apply to police body cameras.

And regarding privacy, have retention policies: if there's no legal hold on some body cam footage (part of an active investigation or court proceeding, etc.), it gets deleted after 90 days (or whatever time period sounds reasonable). That way we don't have recordings of people's private residences living in storage indefinitely.

No, there are no "privacy considerations" to body camera footage. Those only into play when the footage is made public. When you cite "privacy," you're letting the police dictate the terms of the discussion.

The footage must be recorded and retained, with the decision to edit and/or release it left up to courts or other independent authorities.

As for two-party laws, those have to go. They don't have a place in a society where the people need the ability to hold police and other officials accountable. If the police are doing nothing wrong, they have nothing to fear, right? Isn't that what they tell us?

The party in question that might object is the person being recorded, not the officer.

Too bad. The camera, like the officer himself, is supposed to be there for the individual's protection. Or, more accurately, the officer's job is to protect and serve those in his or her charge, while the camera's job is to protect society. That's a compelling interest if there ever was one.

The courts can decide when it's appropriate for the footage to be viewed where privacy concerns exist. They can't do that if the footage never existed in the first place... but they will cheerfully take the officer's word regarding what he saw. It's misguided at best to think that this state of affairs is preferable from a privacy and security standpoint.

Police officers go to the bathroom. They take lunch breaks. They go to doctors appointments. They notify next of kin after a death.

> The courts can decide when it's appropriate for the footage to be viewed where privacy concerns exist

Any data that is recorded will be made public eventually, one way or another, which is why it shouldn't be recorded in the first place.

Again: too bad. There are other professions besides law enforcement, if you can't deal with the requirements.

It's certainly one of those jobs that I could never do. I have nothing but respect for ethical, professional LEOs. They are among the people who will benefit the most from having a full AV record of their interactions with the public.

I don’t think anyone has, or should have, a reasonable expectation of privacy when there is a uniformed officer there - for exactly the reason you say, the plain sight exception, the fact officers are armed agents of the gov’t, etc.

If they could see it (real or imagined) and arrest you for it, having a camera to prove if they could or could not even actually see it seems like a pretty reasonable safeguard!

I don’t think it is complicated at all. At least not based on the examples you gave here.

Of course the video will be admissible as evidence. It’s a lot better in fact than relying on the officers memory.

I’d much rather innocent people are exonerated based on video than that guilty people go free because a police officer happened to miss the evidence of a crime.

What are we even talking about here?

If you are being stopped by police, you are generally already not doing that voluntarily, so saying you need consent to record audio or video in such a situation strikes me as weird too.

I believe when video exists (or at least is used in a trial), there's an obligation that the plaintiff receives a copy. I don't know how common this actually is, but I've seen speculation that in some circumstances (e.g. domestic and/or sexual violence) this could cause problems for the victim.

> should officers be recording while inside someone's house where that person has a reasonable expectation of privacy?

The officer has the opportunity to plant drugs if he is not being recorded.

Body cams are a silly idea in the first place. If you're so distrustful of your police that you feel the need to place cams on them, then you need to go back to the drawing board and address the root problem - the trustworthiness of your police force.

There's no reason to treat cops as second-class citizens. There's no evidence to suggest widespread power-abuse, despite all the FUD from BLM. Are there some bad apples? Sure, but there are bad apples in every profession and steps like that only punish good officers and make corrupt officers more careful when they perform misdeeds.

You cannot neuter the power of every profession of authority. All you're going to do is build instinctive and reflexive mistrust of anyone in authority before they've done anything wrong, just like the reflexive mistrust BLM has whenever anyone interacts with an African-American.

What? A body cam is useful for the same reason that AWS CloudTrail and other audit logs are useful-- auditing? What part of policing necessitates that it is done in secret?

> but there are bad apples in every profession

A "bad apple" software engineer probably over-reports hours or is neglectful in their job. A "bad apple" cashier probably skims off the top of sales. A "bad apple" executive probably enriches their personal wealth.

A "bad apple" cop probably results in ruining the lives of other people through wrongful arrests, excessive force, or other means.

Some people simply should be held to a higher standard because of the responsibilities of their position. When you can quite literally be the arbiter of life and death at times, you should be held to a higher standard.

I hear the "bad apple" thing a lot when it comes to discussions of bad cops, and it's usually in the context you're using it, e.g., "sure, there are some bad apples, but you can't hold everyone responsible for them."

Yet the actual old folk wisdom was "one bad apple can spoil the barrel," or as Benjamin Franklin rephrased it in Poor Richard's Almanack, "the rotten apple spoils his companion." The entire point is that we don't get to just say "it's one bad apple." A barrel with one bad apple in it quickly becomes a barrel of bad apples.

The uproar we've seen isn't just about example after example of police brutality, poor treatment of minorities, and abuse of power. It's about the overwhelming resistance among the police to change, to accountability, even to self-examination. It's about how the reaction, by and large, to "bad apples" isn't to say "we need to get rid of that one," but rather to say "we need to protect our own."

We don't treat police as second-class citizens; we entrust them with extraordinary power and latitude. In return, shouldn't we be holding them to an extraordinarily high standard?

I'd agree.

What seems to be missing is the populist support on both sides.

Imagine you're a police officer standing next to a 'bad apple' - you should be put in a situation where helping to prosecute the 'bad apple' is the easiest option for you to take. I don't think it's that easy. I've had plenty of jobs where I was working with an arseshole, but I've never taken it to HR. Maybe most I've ever done, is just ensured I don't have to work near them - solves my problem.

In the US you don't have populism, you have Red Team Populism and Blue Team Populism.

Police reform has been tagged Blue Team populism, so Red Team populists oppose it on principle.

I think that's true, but I also think there's unique, or at least highly specific, aspects to police culture which also come in here. It's come out that a lot of police training is very explicitly "us vs. them" in its approach -- reinforcing the message that every call you go out on could be your last, every person you interact with could be a dangerous criminal, you should be prepared to do whatever it takes to get the bad guys and other officers are your only line of defense. So it's not as much "if I take this guy's misconduct to HR, they won't do anything anyway" as "if I take this guy's misconduct to HR, I will have betrayed my fellow officer." And it's that, that "thin blue line" mentality, that really needs to be tackled.

> If you're so distrustful of your police that you feel the need to place cams on them, then you need to go back to the drawing board and address the root problem - the trustworthiness of your police force.

I think we definitely should do that, agreed, but when we're talking about people who are empowered to deprive others of their freedom -- or even their lives -- I want more than just "we've fixed our hiring practices and training to ensure our officers are trustworthy". Sure, ok, great, but I want more than just your word on it. I want controls put in place to reduce the possibility of abuses, and when there are abuses, I want it to be difficult for there to be no record of those abuses.

> There's no reason to treat cops as second-class citizens.

This sort of verification and monitoring would not be unique to a police force. The whole "second-class citizens" thing is just FUD and an attempt to distract from the problem at hand. Any government agency that deals with sensitive information has controls in place to try to stop bad employees from doing bad things. Some companies do that too, depending on the kind of stuff they deal with. That extends to other sorts of controls, like at most tech companies, most employees should not have access to billing information, or even deployment permissions for services not owned by their teams. Some of this is just to prevent mistakes, but these controls are also there to prevent malicious employees from doing bad things. I believe some of the ISO and security certifications my company has requires controls like this, even.

I suggest you read this book: https://en.wikipedia.org/wiki/The_End_of_Policing

which goes into great detail many different times where police abuse power, and why things need to change.

There is implicit reason to mistrust anyone handed a deadly weapon and the ability to kill people or take away their freedom. 1 in 100 bad cops is more than enough reason for all of them to wear body cams because that one can trivially ruin any number of lives.

It's farcical to suppose that recording their conduct an essential element of accountability is a cross too heavy to bear when life and death and citizens freedom are on the line. Everyone should in fact have a reflexive mistrust of authority because the majority of the human race lacks much of an internal moral compass and can be said to be "good" mostly within the confines of a function system that holds people accountable for meeting objective standards and worse power attracts the corruptible.

You could do well to delete half the buzzwords from your post and come up with reasons in their place.

It shouldn’t even have an on/off button.

Although I agree with you in general, I think this is for practical reasons. It's not helpful to record 24/7, outside of police operations with regard to battery capacity and limits to bandwidth and data storage.

One could propose a system where recording is triggered automatically when leaving the patrol car (BLE dongle) and can only be reset by a centralized authority.

Recording 24/7 is a simple problem- they turn off when placed on a charging dock in the station. With how microSD is progressing I don't think it's necessary to disable in a patrol car. Perhaps it could somehow communicate with the clock in/out system and only be disabled when the officer is not clocked in to work? Then, lunch would be off the clock and off the camera.

I don't see how it's too different from stores recording their cashiers 24/7.

How about hot swappable batteries with a charger in the patrol car?

Can we actually charge the cam through the officers uniform while he sits at his seat?

That's a (likely problematic and limiting) technical solution to a social problem. Turning it off inappropriately should be subject to review and meaningful action. Any toggle should be near impossible to trigger unintentionally.

> It's uncanny how the cameras are almost always off whenever something controversial happens.

It's because the body cameras that are bought by police departments aren't marketed as devices that keep cops honest, but as devices that protect them from the lying public. That's why they're sold with features that allow the wearer to turn them on and off, and with buttons that trigger the camera to only capture small clips of video. Cops only want to turn them on when they feel like they might need to prove something in court later.

Have you never heard of the 5th Ammendment?

>It's uncanny how the cameras are almost always off whenever something controversial happens.

I've seen this claim pretty frequently, but never any actual data, so I started tracking the stats in my own city.

Since I started ~January, there have been 25 police shootings/use of force events(this includes things like tazing, bean bags, and punches/kicks that require medical treatment. It doesn't include violence against animals or desk pops/negligent discharges). 23 out of 25 events had bodycam footage. The two that didn't both involved off-duty officers, which I think is a reasonable exception to body cam rules. There were no instances of "malfunctioning" body cams, dead batteries, or officers who "forgot" to activate their cams.

Even though 6 months of data from one city isn't enough to make any sweeping statements, I haven't found any studies more thorough than my own spreadsheet. If you know of any please send them my way!

Here's a recent report for a large department that apparently takes body cam footage seriously. 22% failed to have cameras on for "use of force" incidents... which seems like the most extreme example. If the usage was higher in any other circumstances you would have to wonder why, because it would seem sinister.

https://www.latimes.com/california/story/2021-07-20/22-of-la...

>22% failed to have cameras on for "use of force" incidents

IMO that's a an oversimplification compared to what the report actually says[1]. The 22% number isn't how common it was for officers to not have their cameras on, it was how common it was for officers to have cameras that were off OR the camera was on but they were late in hitting the record button[2]. Situations where the officer just straight up didn't record were more rare (that happened 3.7% of the time and if you exclude one instance that was considered justified it's only 3.2%).

Funnily enough, LAPD is the department whose data I was tabulating (over a slightly different time period than this study). I think there's 2 main discrepancies between my numbers and the OIG report: I only counted if at least one officer present had body cam footage, not if every officer had body cam footage. Also, I still counted it if the footage of the shooting/use of force was in the 2 minute buffer.

[1]: https://a27e0481-a3d0-44b8-8142-1376cfbb6e32.filesusr.com/ug...

[2]: Keep in mind that just because they were late to hit the record button doesn't mean the shooting wasn't captured on camera. If an officer talks to witness to ask where a suspect went, radios for help, activates the camera, finds the suspect and then a shooting occurs, they'd be considered non-compliant (e.g. https://www.youtube.com/watch?v=UDThHbrtQE4&t=355s). And even if the officer doesn't activate the camera until after a shooting, it'll still be captured in the 2 minute buffer (e.g. https://www.youtube.com/watch?v=WT4Ss6ASOrE&t=3m30s)

Wait, so they're supposed to be on all the time (that they are actively patrolling) so that the 2 min buffer exists, but as long as they turn on the cameras immediately before using lethal force it's just fine and really shouldn't count? I mean close to 2 minutes is kinda the minimum here. All they have to do is press the record button sometime before they pull their weapon or at least within a few seconds of discharging it to have 1:55 saved.

Otherwise, they can inform their victim/perp that they are going to execute them and then turn the camera on for the few seconds before as a warning to anyone who would challenge them. Is this not a corner case you considered?

A) BWC is constantly recording to a circular buffer. It doesn't "start recording" when an officer presses the button; it just writes the buffered data to memory and adds new video data to that file until ~30-60 seconds after the officer ends the event. This has resulted in some amusing footage where actual bad cops plant evidence, then "start" their BWC, and then get fired when the footage is reviewed, released, and goes viral.

B) At least for CPD, the main thing BWC footage does is exonerate officers from unfounded complaints. Most officers have really embraced BWC as it regularly shields them from having to pay out on lawsuits. And when footage should exist on an event number but doesn't, that leaves an officer pretty wide open to having to pay out of pocket.

C) What instances are you thinking of regarding your claim that cameras are "almost always off whenever something controversial happens"?

There was a high profile instance of B a couple of days ago that I watched today.

https://roguerocket.com/2021/07/26/wisconsin-police-deny-pla...

The fact that they can get away with refusing to collect evidence like that is absolutely infuriating.

> No technology will work if the people using it is corrupt.

Not if the people above them are not. Sadly, the people above them need corrupt police because corrupt police is easier to control than the good one, and their members can also turn themselves into private police serving the same politicians who saved their asses. It's a "do ut des" scenario in which everyone has something to gain, except common people.

>No technology will work if the people using it is corrupt.

Put it on a blockchain then.

Edit: I'm serious. Blockchain is exactly meant for application inside zero-trust environments. If you think it's impossible to put a video on blockchain, just have software making hash-fingerprints at regular time intervals and save those on a blockchain to make tampering with the video evident.

Ouch:

> Over the years, ShotSpotter’s claims about its accuracy have increased, from 80 percent accurate to 90 percent accurate to 97 percent accurate. According to Greene, those numbers aren’t actually calculated by engineers, though.

> “Our guarantee was put together by our sales and marketing department, not our engineers,” Greene told a San Francisco court in 2017. “We need to give them [customers] a number … We have to tell them something. … It’s not perfect. The dot on the map is simply a starting point.”

In fact, 89% of ShotSpotter deployments turn up no evidence of gun-related crime: [1]

1. https://endpolicesurveillance.com/

I'm not sure why this is an important stat. Isn't it good if there aren't gun crimes happening in a lot of places?

Check the link. What it's saying is that most times the police receive a notice of a gunshot, it's a false positive. In Chicago this leads to 60 dead-end responses each day. Very few systems are worth that trouble

I 100% dont think it should be used as evidence but this seems like a decent rate for responding. There's 12,000 total officers in Chicago. Even if only 3,000 are on duty at any time that's not a huge inconvenience if there are still even dozens of cases a year you get to the scene 5 minutes before you would otherwise.

If you track the number of calls to 911 that are "dead-ends" because there are no arrests I'm sure it would be high too.

It also doesnt say the error rate for when the police aren't called. If criminals know that if they shoot a gun there is a 50% chance a police car will drive through 5 minutes later - wouldn't that deter you?

Yeah, in fairness the steam controller had its perks. I didn't see the appeal of the two pads, but it was the first I can recall to do half/full pull triggers. Their customization also allowed you to do some crazy things. Personal favorite of mine was setting the gyroscope to act as right joystick, but only while zoomed in. This let you fine adjust your aim with about as much precision as with a mouse.

In the end though every game had to be translated to the controller and there was always enough of a change to make it feel unintuitive. If Valve had paid an intern to go through the top games each day and make official layouts for games that roughly lined up with each other I'd probably be using one now.

Deployments I assume means police response, not deploying the system.

> 89% of ShotSpotter deployments turn up no evidence of gun-related crime

If ShotSpotter is limited to registering the the sound of gunshots, then it is limited to firearm-discharge-related crime which is a subset of gun-related crime.

That site says 14% of alerts result in a case report being filed and 10% result in a gun-related case report. But even a 10% alert -> case report rate is better than I would've expected for a big city. That still seems like it'd be extremely useful, even if it means some wasted man hours.

And that's case reports - not necessarily detection true positives. If someone fires a gun at the sky and picks up the shell casings and drives away, with no witnesses or cameras and an inconclusive/indiscernible audio recording, police likely won't be able to file a report.

Also, I'm guessing there are some confidence indicators for each alert: those case report rates probably don't capture the full story even when accounting for hypothetical inconclusive true positives. Police are likely (and should be) obligated to investigate every detection, including every detection rated as low-confidence.

Plus that's just for Chicago. I doubt they actually achieve their claimed 97% accuracy anywhere, but it could be higher in other areas. Especially areas where loud noises are less common.

The main issue seems to be when it's used as evidence. I think it shouldn't ever be admissible as evidence of anything. And to mitigate some of the other issues that site raises, it'd be nice if there were some kind of "no fishing expeditions" law, where police responding to a gunfire alert must exclusively look for evidence of a fired gun in that vicinity and not use it as an excuse to scavenge for low-hanging fruit by harassing / arresting people in the area for unrelated matters.

"That still seems like it'd be extremely useful, even if it means some wasted man hours."

I think the least wasteful approach would be to build rapport with the community and rely on their reports. Probably not an easy thing, but the most valuable things rarely are.

Still doesn't stop the phenomena of everyone thinking "oh, someone else will call."

Shotspotter is a very useful tool IMO.

> the phenomena of everyone thinking "oh, someone else will call."

This generally doesn't happen. It originated as a slander against the people who lived around where Kitty Genovese was killed, as an after-the-fact excuse for the lack of police response.

> While there was no question that the attack occurred, and that some neighbors ignored cries for help, the portrayal of 38 witnesses as fully aware and unresponsive was erroneous. The article grossly exaggerated the number of witnesses and what they had perceived. None saw the attack in its entirety. Only a few had glimpsed parts of it, or recognized the cries for help. Many thought they had heard lovers or drunks quarreling. There were two attacks, not three. And afterward, two people did call the police. A 70-year-old woman ventured out and cradled the dying victim in her arms until they arrived. Ms. Genovese died on the way to a hospital.

https://en.wikipedia.org/wiki/Murder_of_Kitty_Genovese#Accur...

There's evidence around how the majority of gunshots in most cities are not followed by 911 calls.

For instance, the murder of Seth Rich in my home city (ignore the stupid conspiracy theories) was not followed by a call but was found by shotspotter which is how it was responded to.

Just because that particular story was a myth doesn't mean the bystander effect isn't a real phenomenon.

Also, that's still two out of 38 people who called. If only 36 people were there, without those two, it's possible no one might have called.

Obviously that's also necessary, but even if you have the best possible rapport and reputation with the community, you won't always get a report. Also, even the fastest report will be slower than a gunfire detector report.

It's hard to tell, as many of those cases may have been created anyways. That is, you may get an alert and 911 calls about the incident. You'd need to subtract all those cases from the rate to see the improvement.

This tends to be the issue with a lot of "AI" technology tied to police work. It does work to some extent, but it mostly tells you things you already know.

That's very true. Though, while accounting for that, it's also shown that gunfire detectors result in faster response times as well. So even if police were to have received a 911 report, the gunfire alert still could've led to them getting there faster than they would have otherwise.

> Though, while accounting for that, it's also shown that gunfire detectors result in faster response times as well.

That seems incompatible with roughly nine out of ten shotspotter reports ultimately being false positives.

What do you mean? I was referring to faster response times to true positive shootings. I can see how it sometimes might indirectly lead to slower response times to other kinds of crime, but I imagine probably not violent crime, unless they have awful triaging.

Two things:

1) False positives lead to under reaction. This is the "boy who cried wolf" issue.

2) Extra false positives put strains on resources, as they produce spurious false reports that must be dealt with in parallel with other requests.

Combine these two, and I'm dubious that shotspotter would actually make the police more responsive to real shootings.

>1) False positives lead to under reaction. This is the "boy who cried wolf" issue.

Alert fatigue is a big problem for any alerting system, but here the baseline is 911 reports. If they receive a 911 report for a shooting, they're going to arrive after the report. If they receive a ShotSpotter alert before a 911 report for the same incident, then they're probably going to arrive as soon or sooner than they would have otherwise. If they receive an alert but don't receive a 911 report, then any response time will be faster than no response.

Not sure if these are backed by data, but there are many quotes of law enforcement saying response times are decreased. I could paste a bunch of quotes but I know it doesn't really mean much.

This study used "gunshot victim prehospital time" as a proxy; arguably a more important proxy:

>Use of ShotSpotter detection technology decreases prehospital time for patients sustaining gunshot wounds - https://pubmed.ncbi.nlm.nih.gov/31425474/

Some other studies indeed found it did not seem to decrease response time, though.

Another factor is that it tries to precisely locate the source of the shot. It's hard for humans to infer the location and direction of a supersonic bullet's source if they're going just by hearing and didn't see the incident, while in theory an automated system could determine it more accurately. I don't know how helpful this is empirically, though.

So you may be right. I think most of the value probably comes from cases where they otherwise wouldn't have received a 911 report, or only would have received one much later.

> If they receive a ShotSpotter alert before a 911 report for the same incident, then they're probably going to arrive as soon or sooner than they would have otherwise.

This seems true, and if it happens a lot that would imply that the system is worthwhile.

> If they receive an alert but don't receive a 911 report, then any response time will be faster than no response.

Assuming it’s real, yes. If it’s not, it’s taken a cop on a wild goose chase when they could have been doing something else.

> Not sure if these are backed by data, but there are many quotes of law enforcement saying response times are decreased. I could paste a bunch of quotes but I know it doesn't really mean much.

Some of these are certainly true and in good faith, but as a general rule I don’t trust quotes from cops about whether or not they need new powers or tools since their answer is always “yes”. Sometimes they’re right, of course, but we need to evaluate more unbiased sources of data to correctly determine what tools are worth it.

It’s a bit like asking me if I need an M1 to do my job better. I might be incidentally right, but I also have extra motivations too.

And that’s before we talk about tools that are more ambiguous. Remember that this data will be used to prosecute people too, if it’s not accurate then the consequences might be worse than just sending out on pointless calls.

> This study used "gunshot victim prehospital time" as a proxy; arguably a more important proxy:

Yeah, that’s compelling.

> Another factor is that it tries to precisely locate the source of the shot. It's hard for humans to infer the location and direction of a supersonic bullet's source if they're going just by hearing and didn't see the incident, while in theory an automated system could determine it more accurately. I don't know how helpful this is empirically, though.

If it works, I agree. The article implies that it is not accurate in determining the location of a gun shot. The extremely high false positive rate also could be from the system accurately detecting a gunshot and inaccurately placing its location.

At a cost of $33 mil a year for the city of Chicago, I personally can’t help but wonder if there’s something else that could be done with that money to reduce the crime rate.

Also, as a nit: most of the sound from a typical gunshot comes from the bullet “uncorking” from the barrel and suddenly releasing the trapped high-pressure gases all at once, generating a shockwave. This is precisely why suppressors (“silencers”) work, they slow down and cool off these gases before releasing them.

The supersonic crack is a factor, but it’s usually not most of the sound. Typically you will not be able tell a difference between subsonic and supersonic ammunition in the same caliber, it’s all too loud to notice. It’s only once it’s suppressed that you can tell.

Many common handgun calibers are naturally subsonic too, such as the fairly popular .45ACP, as well as heavier 9mm loads. The march 22lr ammo I use is naturally subsonic (faster, trans sonic ammo is less accurate), and it’ll still destroy your hearing without ear protection on.

I know nothing about firearms, but my naive assumption/half-remembered factoid was that the supersonic speed is partly what contributes to what I believed was a common phenomenon of it being difficult to infer where a gun was fired from based on the sound alone. (Not sure if that actually is a common phenomenon or just a trope.)

Maybe any fast, loud noise is difficult to place and the supersonic crack is pretty irrelevant, though? Or maybe sounds tend to just be hard to place in general, similar to how eye witness reports tend to not be very reliable?

>At a cost of $33 mil a year for the city of Chicago, I personally can’t help but wonder if there’s something else that could be done with that money to reduce the crime rate.

Even if one assumes all the other parts are unreliable, I think the most important question is what percentage of confirmed gunfire was identified by the system which otherwise would never have been discovered, or only discovered more than 30 minutes later.

If there are even a few cases like that, then I think that could lead to someone potentially being saved when they'd otherwise face certain death; potentially due to bleeding out over a long period of time with no one nearby to help. To me, that alone would justify the cost.

This other HN comment is one example:

>For instance, the murder of Seth Rich in my home city (ignore the stupid conspiracy theories) was not followed by a call but was found by shotspotter which is how it was responded to.

Given enough instances like that, it seems pretty plausible that in at least one case, someone who would've died could be rescued.

>Alert fatigue is a big problem for any alerting system, but here the baseline is 911 reports. If they receive a 911 report for a shooting, they're going to arrive after the report. If they receive a ShotSpotter alert before a 911 report for the same incident, then they're probably going to arrive as soon or sooner than they would have otherwise. If they receive an alert but don't receive a 911 report, then any response time will be faster than no response.

I'd like to suggest that it might lead to slower response times in the cases of false negatives: someone calls in a report and the ShotSpotter doesn't make an alert. If ShotSpotter tends to be over sensitive police might assume if they don't get an alert from it something definitely didn't occur and take their time

I can see how this could be a risk if it were an assumption they make, but my guess is all PDs would consider human reports far more reliable than ShotSpotter alerts. I'm sure in all or almost all areas, a 911 report of a gunshot is much more likely to be a true positive compared to any automated alert. A lack of a report doesn't imply a lack of an incident, but the presence of a report implies a high likelihood that there is an incident.

A system could be considered generally "sensitive" in some way (e.g. seemingly firing in response to any loud noise) while actually performing poorly in terms of statistical sensitivity (the percentage of actual gunshots that the system detects and alerts on with high confidence).

Having an 11% success rate seems pretty impressive to me. Much better than the rate of false panic alarms, for example. And I'm sure a lot of those ShotSpotter alerts were accurate, but the police didn't get there soon enough.

I think for the purposes of sending a car over to drive through a street its pretty good.

The article discusses instances where it is used in trial and for that it is obviously pretty weak - surprised it can be used at all.

I can see a use to confirm timing, or approx location, or as backing up a witnesses testimony. But it shouldn't be used as stand alone evidence.

It shouldn't have anything to do with getting there soon enough. If they find shell casings, victims, or other evidence that a crime occurred (usually even discharging a firearm in a major city is considered a crime with very limited defenses), then they should still be collecting evidence and logging a report. Now, they may not be doing that because they don't want to have high crime numbers that are unsolved, or even dragging down the reputation of the safer areas (I've heard of officers doing this).

I was thinking of cases where someone fires a few shots and then everyone runs off - odds are the shell casings aren't going to be seen unless they do a careful sweep of the area, and the victims all left pretty quickly. No idea how often that actually happens though.

Yeah, but if the people running off aren't calling in, then the technology isn't going to make any impact until the public relations change.

That's the ouch? Not literally rewriting history to create evidence for a murder trial?

> But after the 11:46 p.m. alert came in, a ShotSpotter analyst manually overrode the algorithms and “reclassified” the sound as a gunshot. Then, months later and after “post-processing,” another ShotSpotter analyst changed the alert’s coordinates to a location on South Stony Island Drive near where Williams’ car was seen on camera.

> Motherboard’s review of court documents from the Williams case and other trials in Chicago and New York State, including testimony from ShotSpotter’s favored expert witness, suggests that the company’s analysts frequently modify alerts at the request of police departments—some of which appear to be grasping for evidence that supports their narrative of events.

Marketing games are absolutely nothing compared to this.

This sounds suspiciously like evidence tampering. If only the legal system would be so dogged in pursuit of justice for the cops & analysts involved in this conspiracy.

Any time I see "accuracy" reported as a single number, I assume it's B.S. and "for marketing purposes only". If you're serious, give me the precision and recall.

>If you're serious, give me the precision and recall.

The thing is decision-makers have no idea (and no care to know) what those are.

Maybe these people shouldn't be decision-makers then. We need both greater transparency in decision-making processes, and greater education amongst the population about statistics, causal inference, and the scientific method. In order for the citizenry to hold decision-makers to account and obtain a higher standard we need:

1. Citizens to have the time and motivation to seek out and examine primary evidence.

2. Citizens to have the skills to understand and critique that evidence.

3. Robust mechanisms to allow citizens to replace people in power.

This is an inherent problem with the specialization of labor. Those who run police departments have more relevant specializations to their job role than information science. This isn't a problem with statistics or science in particular -- it's an issue with every specialization. This is only going to get worse as popular opinion of the role of higher education is starting to shift from a more liberal-arts focus to a job-placement focus.

Correct and that's unlikely to change, but it should be grounds to throw out the prosecution. Yes, this will make it harder to convict real criminals. Too bad.

Precision is an attribute of the deployed system. The same tech in an affluent suburb will have different precision deployed in an inner city. Ran into this exact problem in biometrics, where NIST publishes tech error rates like FNMR and FMR, and system owners don't even know precision is a thing. They expect the vendor to tell them if it will work, when it depends how many true and false events you throw at the system.

I've never understood why ML folks insist on using precision, which depends on the base rate. Why do they not use probability of false alarm as in traditional detection theory?

It doesn't really depend on the base rate though. The idea is that a high precision model will have all the targets at one end of the distribution, regardless of base rate.

Precision is P(T|R), where R is the event that a thing is reported, and T is the event that the thing is true. By Bayes' rule, P(T|R) = P(T)/(1+Pfa/Pd). So in terms of the traditional statistics Pd and Pfa, it is proportional to the base rate.

This is all devastating. These kinds of devices should use a a cryptographic tramper-evident log, and those logs should be available to the public in real time for archival purposes. That's the only way to avoid tampering. The same should be used for things like, e.g., rolls of who voted at every precinct.

That sounds like fraud

> ShotSpotter’s claims about its accuracy have increased, from 80 percent accurate to 90 percent accurate to 97 percent accurate. According to Greene, those numbers aren’t actually calculated by engineers, though.

there’s the problem right there.

a large number of us have seen, first hand, how a sales and marketing team will happily distort the abilities of something we build. and we’ve all seen first hand how clients will lap that up and then turn around and blame the engineers when it doesn’t live up to what they were led to believe.

as his quote says, those numbers aren’t actually calculated by engineers, they’re calculated by the sales and marketing department and he happily pushes this nonsense because “… we have to tell them something.”

meanwhile people are dying from this/being prosecuted and the accuracy rates are pulled out of thin air by the marketing/sales department.

>> and then turn around and blame the engineers when it doesn’t live up to what they were led to believe.

Any examples of how shotspotter technology has not lived up to or has failed in what it was designed to do?

I'm not really that familiar with the technology, all I know is that if there are shots fired within the vicinity of a shotspotter it registers those shots. Does it have the capability of triangulating exactly where the shot took place?

We have them all over Minneapolis and there's several twitter accounts that monitor police activity and I've seen a lot of posts saying:

shotspotter was activated at 33rd and Hennepin, police located several bullet casings nearby and are actively searching for the shooter

Stuff like that.

What's the difference between having 80% and 97% accuracy?

> Any examples of how shotspotter technology has not lived up to or has failed in what it was designed to do?

You're literally quoting on an article where they moved the location of a sound by over a mile, and reclassified it from a firework to a gunshot. They tried to use this obviously falsified evidence to throw a man in jail, and had to withdraw the evidence once this was pointed out. Sure seems like an example of it not living up to what it was designed to do.

Even if it turns out Williams is guilty and he did shoot that man, Shotspotter was just blatantly wrong about what happened and where it happened.

Also, 86% of the time it detects a gunshot police find nothing.

https://chicago.suntimes.com/2021/5/4/22417660/shotspotter-a...

> We have them all over Minneapolis and there's several twitter accounts that monitor police activity and I've seen a lot of posts saying:

Ooookay? This is practically the definition of how the plurality of anecdote is not data.

> What's the difference between having 80% and 97% accuracy?

When it's being used to send armed representatives of the state to arrest and prosecute people? The difference is a lot of people arrested and prosecuted based on faulty evidence.

This comment would be a lot better if it dropped the antagonism. They were just asking questions – questions I agreed with, for what it's worth.

They didn't quote from the article.

> They were just asking questions

Yes they were JAQing off.

> questions I agreed with, for what it's worth.

You agreed with the question “what is the difference between 80% and 97%”?

Not op but I had that question.

In my case the actual question is what is an accurate reading?

The good faith question isn’t “what’s the difference between 80% and 97%?” That’s a pointless question at best, a rhetorical trick excuse police and prosecutorial misconduct at worst.

A better question would be along the lines of “how accurate would shot spotter have to be to justify both its use by police and courts, and the cost imposed on cities for its use?” Also “how much do we trust shotspotter’s assertion of their own accuracy?” is a good follow up too.

This is tough. Pretty clearly what's happening here is shot spotter is producing a probability distribution about the places where a shot may have been fired, but they're not showing that information, they're showing a point. That's broken.

It's a shame because setting up some microphones around the city to detect a firearm discharge seems like a great idea. The problem is people are overselling what it is doing.

Something that jumps out at me in the article is this:

  Paul Greene, ShotSpotter’s expert witness and an employee of the 
  company, testified at Simmons’ trial that “subsequently he was asked by 
  the Rochester Police Department to essentially search and see if there 
  were more shots fired than ShotSpotter picked up,” according to a civil 
  lawsuit Simmons has filed against the city and the company. Greene found 
  a fifth shot, despite there being no physical evidence at the scene that 
  Simmons had fired. ...

Like, what shotspotter is doing is attempting to resolve the origin of a sound wave in a city. The shockwave from a gun blast must reverberate across buildings all over the place, more than likely you're getting reflected waves rather than waves from the source. You're going to get echos and reverb and it's going to be messy and you're going to be imprecise.

Would shotspotter's AI misclassify noises? Yah. Of course. If it's like any other classifier it's likely listing out a list of probabilities of what something is and, for the sake of simplicity, just saying it's the highest probability output. It's just a filtering step. If you know a gunshot occurred, you update your priors.

Should shotspotter be able to detect the number of gunshots exactly? Maybe? How to differentiate between echos and original? I'm not shocked it can detect the wrong number of shots.

Since it misclassifies noises, isn't exactly sure of the origin, and can read echos as additional shots, is it useless? No.

This is an unreliable system which police and prosecutors misrepresent as reliable.

It might not be useless in the strictest sense of the word, but it ought to be taken away from them.

This is the issue: it's unreliable, and its results presented as a matter of fact.

This is also the latest I've heard of, in a long line of the criminal justice system presenting what's often closer to fiction as an indisputable fact of someone's guilt. One of the cases that made me incredibly angry was the junk arson science surrounding the conviction and execution of Cameron Todd Willingham in Texas in 2004:

https://innocenceproject.org/cameron-todd-willingham-wrongfu...

But I also think of how unreliable eyewitness testimony is, and yet people's recollections of events (when we know, and have known for so long, that memory is plastic!), sometimes years later, are treated as fact:

https://www.ncsc.org/trends/monthly-trends-articles/2017/the....

I don't know how to solve this, since the "proven" part of "innocent until proven guilty" seems like it's built on a house of cards, and the only people with a direct incentive to change it are suspects (i.e. the people that the criminal justice system has already decided could've done it, and how can you trust a criminal's motivations? eye roll)

Restrict its use to the original intention: detecting potential events police might want to immediately respond to. It (and any other gunfire locator) seems like it'd probably be an extremely useful and publicly beneficial technology even if it had a low true positive rate.

The problem is allowing it to be used as evidence in criminal cases. That's the part that should be taken away.

This exactly. I see Shotspotter and tech like facial recognition / license plate readers as the same. Its a lead, a clue for detectives to follow up on and validate / verify. It shouldn't be evidence on its own. Just a source that's slightly more reliable than a human witness.

The arguments for facial recognition and license plate readers are much more complex. In the case of a gunfire detector, you're specifically looking for (supposed) activity that indicates the most serious possible crime might have occurred seconds ago.

Facial recognition and license plate readers are used to automatically track individual humans wherever they go so that they can be quickly arrested if they're on a to-arrest list. I'm not entirely sure how I feel about that, even if it's all taking place in public, open areas, but it's a totally different category from just knowing when and where a gun may have been fired. It's real-time crime detection vs. alleged criminal detection.

Thats very true. I also have reservations around facial recognition / ALPR systems if they are saving data of people who are not currently "flagged" related to a crime. I can see the usefulness to find stolen cars, or track down murder suspects but only after they are entered as a suspect. There shouldn't be a historical database of previous movements. There should also be states/federal guidelines around who can be entered.

This seems to be intentional:

> Rather than defend ShotSpotter’s technology and its employees' actions in a Frye hearing, the prosecutors withdrew all ShotSpotter evidence against Williams.

Ducking the test and withdrawing the evidence means prosecutors can continue to claim that this stuff hasn't been deemed unreliable by experts.

A quirk of our legal system that allows dishonest people to continue with impunity.

which courts could easily fix...but don't.

> shot spotter is producing a probability distribution about the places where a shot may have been fired, but they're not showing that information, they're showing a point. That's broken.

>This is an unreliable system which police and prosecutors misrepresent as reliable.

I would argue in american policing, that broken is the point. People (cops, judges, juries, the public) don't engage with anything criminal justice related as probabilistic. If they showed the probabilities, they wouldn't have a product specifically because it would not be able presented as reliable.

It's product market fit, like the parallel citizen thread, that serves malicious and anti-social goals. But it is profitable.

It's the same deal with breathalyzers, which don't measure blood alcohol and cannot estimate it to the levels of precision the manufacturers want everyone to believe.

and a pattern of withdrawing the evidence when the results are challenged so as to avoid setting precedent.

The "evidence" can't be reexamined either. The good states require a blood test. At least it's more accurate and can be examined by the defense.

Yeah, it's a really useful tool. There are a lot of gunshots in Chicago (last year there were over 4000 incidents (in Chicago) where someone was shot, and there are far more gunshots where no one is hit), and in many of those shootings, no one calls 911. If someone is hit and rendered unable to call 911 or get help, if no one calls 911, that shooting victims is likely to become a homicide victim. ShotSpotter gets someone to check the area out, and it has saved a number of people from bleeding out.

I don't think it should be used as evidence of the location of a shooting (often shootings are only discovered when the victim shows up at a hospital), as locations are triangulated based on the time different sensors detect a gunshot and the distance sound travels is impacted by building geometry and materials. But it's a very useful tool that saves lives.

Regarding the locations of ShotSpotter coverage, as the pricing model is proportional to area covered (there's a physical network that needs to be maintained and contacts with owners of buildings with sensors that need to be paid), so sensors are put in the areas with lots of shootings. If you want to know why there is such extreme racial homogeneity in the most violent, least well resourced parts of Chicago, I highly recommend the book "The Color of Law" [0] on redlining.

[0] https://www.epi.org/publication/the-color-of-law-a-forgotten...

Do you have any citations for your claims about the Chicago system being responsible for saving a number of people? I'm looking for reading on that topic. Thanks ahead of time.

None that are public (that I'm aware of), but it's just a matter of time. We just got ShotSpotter alerts put up on the Chicago data portal.

But to make it less difficult to believe, I'll point you too LEMART training [1], an emergency first aid course that many patrol officers elect to take, and that many officers carry a first aid kit [2]. Considering ShotSpotter alerts officers (many of which are capable of administering emergency first aid) to shootings (many of which have no corresponding 911 call for service), it's not too hard to accept the weak claim that a nonzero number of lives have been saved. I won't speculate on the number as I haven't done a rigorous analysis, but I know of a lot of shootings where the responding officer was notified by ShotSpotter, administered first aid (normally a tourniquet), and the victim survived.

Edit: just found a public number for the number of officers who have received LEMART training. Over 6000 officers as of 1/21/2020 [3]. For reference, there are about 13k sworn CPD officers.

[0] https://data.cityofchicago.org/Public-Safety/Violence-Reduct...

[1] https://www.chicagocopa.org/wp-content/uploads/2018/04/Polic...

[2] https://directives.crimeisdown.com/diff/efaff1a7e94a8ad642c7...

[3] https://twitter.com/chicago_police/status/121974785846900326...

"... and in many of those shootings, no one calls 911."

It's sad that our society (can we really call it that?) has come to this - relying more on a piece of technology because you can't depend on your fellow citizens.

There's no point in calling the police when you hear a gunshot anywhere near Chicago. For one, you probably heard fireworks. For another, calling them doesn't accomplish anything. What are they even supposed to do about random gun shots?

I'm not asking about shots off in the distance. With the population densities of cities, people should be close enough to know.

Discharging a firearm is a crime in most cities, with limited defenses. They should be collecting evidence like shell casings and investigating if another crime was committed (check for victims). If you have people illegally shooting firearms, there's a decent chance they arent legally allowed to own them either. Catching these people for this crime, could prevent others.

Maybe the US should try to sell less weapons ..

Being unarmed in the current environment is not advisable.

Before defending it too strongly, take a look at their article on where the pickups are being placed [0] and in the section on Chicago zoom in on the map. It's ASTONISHING how well it lines up with the demographics. Looking through on the map, it looks like the census blocks with both ShotSpotter and >=50% white are one in CA-17 (Dunning), two in CA-60 (Bridgeport), two in CA-61 (New City), five in CA-41 (Hyde Park, University of Chicago area), five in CA-56 (Garfield Ridge) and four in CA-64 (Clearing). Those last 9 census blocks are part of a little largely-white enclave around and to the west of Midway airport and just north of the rail yards. (CA refers to Community Areas in Chicago, see [1])

I think it's notable that there are a total of 19 majority-white areas in the city with ShotSpotter tech - not with widespread installations, areas that have them at all. If you want to shoot someone in Chicago and not have it picked up by these devices, just head to Lincoln Park - looks like it's a couple miles from there to the nearest audio pickup.

[0] https://www.vice.com/en/article/88nd3z/gunshot-detecting-tec... - don't forget to click the "visible layers" button and turn on the "gun crime" locations layer as well.

[1] https://en.wikipedia.org/wiki/Community_areas_in_Chicago

Seems like most of the benefit could trivially be kept without much of the downside if we simply made in inadmissible as evidence which seems to be advisable as the system is designed to allow a human to rewrite history as part of normal operations. It seems irredeemably broken for evidentiary purposes nor would it be reasonable to ever in the future trust the work of any company who would so arrange matters.

All parties involved ought in present matter down to peons if they were aware of it and didn't come forward should go to jail.

It's a shame because setting up some microphones around the city to detect a firearm discharge seems like a great idea.

It's an idea, yes, but a *great* idea? There are so many technical, social and civil rights issues wrapped up in this idea on its face, I think its performance history should be part of whether it gets to enjoy the label of being a "great idea."

Since it misclassifies noises, isn't exactly sure of the origin, and can read echos as additional shots, is it useless?

Furthermore, since it can and does record human voices speaking, and creates permanent records into the private hands of a for-profit company that is allowed to sell that data to 3rd parties, does it actually cross a line from being useless to being harmful? It might.

Right... but I don't classify its performance of "saying 5 gunshots when their was 4" a failure.

If it can say "I heard a gunshot, sounded like it came from around here." and a gunshot occurred somewhere around there, with only a few false positives. That's fantastic. In my view, that's "great".

It's just not iron clad evidence, prosecutors are overstating its statistical power. That needs to be checked.

Yah absolutely. But the thing is gunshots are pretty loud noises compared to human speech. There's no reason to think a microphone needs to be sensitive enough to capture human speech to capture a gunshot, nor placed in a location that is well suited for capturing human speech, nor would a place that IS well suited for capturing human speech a good location for capturing gunshots.

If I was placing a microphone to capture the sound of gunshots, I'd be putting it on the tops of streetlights, telephone polls, and the corners of buildings. You're not going to be hearing people from there.

So if we see that behavior of ShotSpotter sticking microphones in inappropriate places for public surveillance purposes, then I think they should be taken to task for it.

> Right... but I don't classify its performance of "saying 5 gunshots when their was 4" a failure.

Where there four? The part of the recording containing the fifth shot disappeared, there is no way to reexamine it as there where no steps taken to protect the original data against tampering. If your method is so bad that no one else can reproduce your results and the data you base your results on just keeps changing and disappearing to suit the expected result then it has the scientific rigor of a fortune teller.

I think the problem isn't the technology or the idea, it's the business model and the associated incentives. ShotSpotter's customers are the police, and the objective of police is to secure convictions, so ShotSpotter is incentivized to help them do that, not to produce true and accurate predictions. Exonerating a defendant isn't going to win them more business.

> the objective of police is to secure convictions

I mean... that's the problem point it would seem

Right. Shotspotter or the generic equiv should be purchased by the city, not the police. The data (raw and interpreted) should be managed in the public, as its a public interest to know about firearms (including the police's) being discharged.

> Shotspotter or the generic equiv should be purchased by the city, not the police.

Generally, the police are part of the city, and the sole and deliberately centralized city agency with a mandate that makes something like ShotSpotter relevant.

So, of course, if it is bought “by the city”, it is through and under control of the police department.

I'm arguing though that due to the conflict of interest, it should be at least one level up IRT purchasing, access to interpretive/raw data, and decision making on which of that data they want to share etc. Remove funding and responsibility from the police to run this system/vendor, and move it into the "office of public data accountability" e.g. which can serve its data equally to the public, watchdog groups and the police.

Overselling + altering reports after the fact to fit a prosecution's preferred theory of the events.

How is this tough?!

The system detected a loud noise as fireworks... that should be the end of it!

Let the cards fall where they may, if cops go out they go out, they go out, otherwise that's that.

You don't get to go back days later and reclassify it as a gunshot based on ambiguity inherent to your product, and you don't get to go back and move the origin.

This is not tough at all, it literally should work exactly like it works today, and no one is allowed to change the initial call manually.

After all, who is helped by retroactively marking it as a gunshot but a DA?

If the concern is building up statistics or something, ShotSpotter is still recording that it heard a sound. Nothing is stopping an out of band note that says "we believe this recording was tied to so and so incident at so and so location"

The key difference though is that ShotSpotter is no long claiming to have recorded that incident directly. Which matters when people's freedom is at stake.

If we want to rely on this technology as being unbiased and accurate, then it must be a purely one-directional flow of information. Data should only flow from SpotShotter to the police, and never in the other direction. No questions, no inquiries, no anything. As soon as you do, you introduce bias or at least the ability for bias to exist, intentional or otherwise.

If the SpotShotter technology is as good as they claim, there should never be a need. And if the police can't make a case with what SpotShotter gives them, they shouldn't rely on it.

It will always flow back, at least in the form of dollars spent.

It also flows back in terms of political contributions, lobbying, and political appointments. Right now a former ShotSpotter VP is being nominated to lead the ATF.

But also this should flow to the public in real-time for archival and should be tamper-evident to prevent alterations. Like a blockchain.

> then it must be a purely one-directional flow of information.

That's absolutely not possible.

I don’t understand how they get to just walk away without any consequences.

prosecutor1: sir we got evidence that defendant1 was behind the shooting

defendant1: sir I was miles away from this area - let’s verify the evidence

prosecutor1: you know what, never mind - poker face

Court: poker face

Defendant1: poker face

>“Rather than defend the evidence, [prosecutors] just ran away from it,” he said.

Worse: this is one of the most discriminatory issues in the criminal justice system. The prosecutor can and will use that "evidence" unless and until the defence challenges it. But poor people with overworked lawyers don't have the time to challenge anything so they go down based on BS.

The same thing happens with Stingrays apparently: as soon as someone asks where the lead came from for the arrest, its case dropped. But if you can't afford a lawyer with time to chase down every detail you go to jail on illegal evidence.

Why is ShotSpotter using "AI" to analyze this? It's time-based triangulation of a loud noise.

It must be nice to make a living on a grift like this.

Because gunshots are rare and if you're just triangulating loud noise you're gonna be overwhelmed by false positives. Even if you do basic filtration to try and tease out the sound of a gunshot vs everything else you're still gonna get overwhelmed by false positives.

I don't know enough about the procedures, but would citizens calling 911 to report gun shots help coroborate and help weed out false positives?

Many citizens can't tell the difference between gunshots and other explosive sounds like fireworks.

If ShotSpotter can't either then it effectively useless.

It's not useless, sometimes it can be used to fuel a prosecution against an innocent person who happens to be a racial minority.

Police locking up black people in America is the reason police in America were created.

> Police locking up black people in America is the reason police in America were created.

Formal policing in America has a long and sometimes infamous history, but it definitely did not begin with "locking up black people"; it varied based on the locale.

For example, in NYC, it goes back to the Tammany Hall days and was notoriously corrupt.

In the west (where there were very few black people, and those that were there were free), police forces began with various marshals and sheriffs who were hired, typically by the merchants, to tame the wild western towns.

According to Time, "The first publicly funded, organized police force with officers on duty full-time was created in Boston in 1838."

https://time.com/4779112/police-history-origins/

The Boston police were started to defend the rights of property owners, not citizens.

In the South, post-Civil War, police used gold star badges because they were adapted from the gold star badges used by slave patrols.

You should learn more about the history of policing in America: https://podcasts.apple.com/us/podcast/the-history-of-america...

>In the west (where there were very few black people

According to Smithsonian Magazine, 1 in 4 cowboys were black. I suspect popular culture (movies) are what's to blame for this misconception. Many fled to the West after the Civil War to find work.

https://www.smithsonianmag.com/history/lesser-known-history-...

If I remember correctly from a radio story, the police can listen to a recording of the shot report before investigating to confirm it sounds like a gun.

Maybe it gets harder around june/july/new years, but the rest of the year has very little firework activity.

I live in Chicago. We have fireworks year round, some neighborhoods more often than others. Usually you will hear fireworks far more frequently than you hear gunshots.

I think that depends on the city/neighborhood. We heard fireworks pretty regularly throughout the year when we lived in San Jose. And they became more frequent after the Flyod protests.

On the flip side, we heard far more gunshots than fireworks when we lived in Kansas City.

It's presumably some sort of machine learning to distinguish gunshots from firecrackers, backfires and other kinds of impulsive sounds one would find in a city. It just does not appear to work well.

Cincy PD disables alerts for ShotSpotter on the 4th of July and disables it during baseball/football games downtown because the pyro from the games causes false alarms.

It’s just marketing speak. If your tech product doesn’t use AI, machine learning or have an algorithm then it won’t look as bright and shiny to customers.

I m not convinced that simple triangulation works in a city environment. Sound reflected off of buildings can mess things up good. I can think of several ways in which a later reflection can be louder than the initial sound.

Triangluation works in these cases if you apply some DSP tricks (marketed in this case as AI).

Multipath interference can be mitigated to a large extent with multiple listening devices calibrated against a known acoustic impulse - i.e. intentionally firing something like a gun at a precise time & location known to the system beforehand.

Shotspotter concerns me - the city I live in has paid millions for a system and recently expanded it. I am not aware of _any_ data that has been published from the system, even fuzzed data. That doesn't pass the sniff test for me.

Wow, millions?! Do they have any cost/benefit data for this system?

Of course not. They're hoping that a noise will some day coincide with a crime so then they can write a press release lauding the technology and the smart decision maker for buying it.

The easiest way to be perceived as "doing something" is to spend money on it. Succeed or fail, you can still claim you tried. Much of the voting public is convinced that Democrats drink the blood of babies, so convincing them to look more closely and with skepticism is an uphill battle.

"Succeed or fail, you can still claim you tried. Much of the voting public is convinced that Democrats drink the blood of babies, so convincing them to look more closely and with skepticism is an uphill battle."

I'm not sure what you mean by this.

Find some like-minded people (perhaps including journalists), pool some resources, and use FOIA to sue. If your chief of police equivocates, call them a liar. Treat them like criminals, accuse them of running a protection racket.

That's falsification of evidence. Will there be any consequences? Who watches the watchmen?

You say falsification of evidence, they say manual review by an "expert" to correct errors from the automated process.

Reclassifying the type of sound is possibly reasonable. Changing the coordinates seems like falsification of evidence. Later in the article it mentions adding a fifth shot in a different event.

If shotspotter wants to be relied on for evidence, allowing technicians to go back and modify data seems like a recipe to quickly eliminate its use in cities and destroy its market share. Adding notes about manual review after the fact would be the only acceptable thing, but the originals should never be modified/deleted.

Sorry, but I’m an expert shooter. I can reasonably identify the round and sometimes the gun on the range next to me - I would say I have an excellently tuned ear for these things…

And not a chance I could do that based on microphone playback between firework and gunshot with some random mic in a city. 1/2 the “trick” is knowing the location, at my range I know the echo of a 12ga vs a 45ACP, the crack difference in a 223 or a 6.5prc, if I went to some random place and you shot one random shot, it would be ridiculously less accurate.

Shotspotter seems like a ton of bullshit to me after reading this article including…

> Both the company and the Rochester Police Department “lost, deleted and/or destroyed the spool and/or other information containing sounds pertaining to the officer-involved shooting,”

> And not a chance I could do that based on microphone playback between firework and gunshot.

I’m happy to agree with you on that point, since I don’t have the experience there.

My larger point is that I could reasonably say allow for shotspotter to review the sound, but the location I’m assuming is pretty darn accurate.

And yes, the article completely destroys their credibility as a reliable source of evidence.

Multipath effects, which for sound are huge in an environment with large flat surfaces, could create phantom origin spots at the fringes of the covered area.

So if your point is it’s too complex and variable for automatic processing; how does adding a human arbitrarily changing classification and location going to improve accuracy? Unless it’s the shotspotter tech put there shooting people, seems the company is very successful bullshit either technically or ethically.

If there was solid external evidence of when a shooting occurred (eg gps timestamped video of the event) I could see them doing it purely for calibration purposes.

Clearly those adjustments should not be admitted as evidence in a court case. ShotSpotter just needs to take the 'L' and they will have to prosecute without (or admit whatever was originally recorded as corroborating evidence).

>but the location I’m assuming is pretty darn accurate.

I remember reading about a similar technology used by soldiers in Afghanistan to find snipers, but Afghanistan is largely an open area. I don't see how it would be accurate when the acoustics of multiple buildings are in effect.

The idea is attractive because it's quite obvious how well ShotSpotter would work in a cleanroom environment. The part that's less intuitive is how far the standard deployment environment is from that standard and what margin the product has to accommodate.

If the source audio is not reviewable in court, none of it should be admissible as evidence. Algorithms are just laundering bad opinions like blood splatter analysis, so that there’s not even a personal reputation at stake.

"Liability laundering" does roll off the tongue more easily than "machine learning algorithm".

>Algorithms are just laundering bad opinions like blood splatter analysis

You have way too much faith in the legal system. So many pseudo-scientific classes of evidence exist:

- Field sobriety tests

- Ballistic forensics

- Blood spatter analysis, as you mentioned

- Bite pattern analysis

- Burn / arson forensics

- Polygraph

- Biometrics under less-than-perfect conditions (your fingerprint reader on your phone works well because it's near perfect conditions. Crime scene partial finger prints are usually insufficient. Same for facial recognition -- works with your iPhone FaceID sensor, doesn't work with the gas station 480x320 camera)

- Most DNA and hair analysis

Pretty much all hocus pocus. Courts don't care. Anything for the conviction. Anyone hoping for fair scientific analysis in a court of law today is in for a rude awakening.

To be fair, polygraph tests aren't admissible as evidence in court, but I completely agree with your larger point. You also left out drug sniffing dogs, which are just a prop for the justification of an otherwise illegal search, and apparently drug sniffing cops too. There was recently a case where a cop claimed in court to have smelled unburned pot from several cars away while driving on the interstate. This was fortunately thrown out but it's not guaranteed.

I agree with you completely, except for the idea that I have too much faith in the legal system.

> manual review by an "expert"

Sounds like that should be required before the results of automated processes can be used as evidence.

I'm sure it is, but the point here is that the manual review appears to be "aided" by the police "suggesting" where they believe the sound came from. And then, wow, wouldn't you know it! The manual review agrees. Case closed.

Journalists, courts, and citizen activists.

There won't be any consequences, because the police never face any consequences. There's no way to reform modern policing, the evidence shows little progress from decades of trying, which is why modern policing must be abolished and something entirely different needs to take its place. The police have resisted incremental reforms, so the only option now is drastic reforms.

This reads like a troll message to me, so I'm not sure I should even engage. But in the off chance it isn't, then if you actually want to convince people of something you should try:

* Not using ridiculous hyperbole

* Providing a well-reasoned solution instead of just stating what the problem is

>Providing a well-reasoned solution instead of just stating what the problem is

I did:

>modern policing must be abolished and something entirely different needs to take its place

All you did was state "this thing sucks and it needs to go." Oh and by the way, it should be replaced by "something."

That's not a well-reasoned solution. That's not a solution at all. All you're doing is complaining.

Put your money where your mouth is, and come up with what the "something" should be.

Okay, how about starting with a new department modeled after incredibly successful programs like these?

https://www.npr.org/2021/03/08/974941422/6-month-experiment-...

https://www.npr.org/2021/07/23/1019704823/police-mental-heal...

Both articles mention specifically that police were replaced _for certain calls_.

But that doesn't jive with your proposal to completely eliminate police and replace them with something else.

I'm not going to engage you any further since it's become clear you're either trolling or not really trying.

You have to start somewhere. They started somewhere. It worked.

Abolitionists did that all last year. It should be somewhat common knowledge by now for people who read news.

So I should assume that this guy's opinion is the same as everyone else's? A+ communication skills.

I mean, I'm not in the "Defund police" camp, personally, quite... but OP doesn't read like troll OR hyperbole, and I'm not sure why the reaction - this is an increasingly mainstream perspective.

Police resisting reforms, limited consequences to police misbehaviour, and limited progress in attempts to reform, are I think baseline facts by now. The notion that they need to be rebuilt from ground up with different culture/prerogatives/internal processes/priorities again is nowhere near "troll like".

>>Providing a well-reasoned solution instead of just stating what the problem is

That's a high bar and an unreasonable threshold to speaking out. A lot of times, recognizing / defining a problem is an important first step. It should not be suppressed solely for lack of immediate solution provided.

> OP doesn't read like troll OR hyperbole

Really? And how do you figure that? They literally opened with "There won't be any consequences, because the police never face any consequences." The entire planet can name at least one instance of a cop facing consequences for their behavior.

"There's no way to reform modern policing" is a strong assertion without very much to support it.

> That's a high bar and an unreasonable threshold to speaking out.

No, it isn't. _All_ systems are subject to abuse by bad actors. _No_ system is perfect. There are only trade-offs.

If you have no alternative solution it means you haven't considered the trade-offs, and if you haven't considered the trade-offs, you haven't actually spent time thinking about the problem.

It's easy to point fingers at what sucks. Everybody can do it, and everybody is doing it. If you're here on Hacker News, which is supposed to be a place for interesting conversation, I expect that you're going to do better than that.

Fair enough; if the aim is to increase the level of discourse on HN, that's a noble cause I'd have to support.

(Note, and entirely FWIW, I still think message could've been presented clearer / in a more friendly & productive fashion - it's usually better to coach and show than to yell & chastise).

Their post meets both criteria. Their description of the issue is not hyperbolic; it is a brief but reasonable and accurate description of the failure of American policing. Their description of the solution is well-reasoned: it is not unreasonable to suggest that we should prefer an institution facing intractable flaws be deconstructed and replaced with a new one.

I do think it's unreasonable to expect a detailed description of this solution off-the-cuff. A simple search would have provided you with that, though: https://defundthepolice.org/alternatives-to-police-services/

> Their description of the issue is not hyperbolic

Saying that police are never held accountable is, in fact, hyperbolic.

> Their description of the solution is well-reasoned

There's no reasoning beyond first order "this thing sucks, progress isn't fast enough, it has to go". No consideration is given to the second or third order effects of such a massive change. So no, it isn't "well-reasoned."

> I do think it's unreasonable to expect a detailed description of this solution off-the-cuff.

Hacker News is _supposed_ to be a place for curious and insightful discussion. Not just issuing complaints about the social issues of the day.

I made an attempt to engage in order to try and produce something substantive, but that fell flat pretty quickly because no effort was made to produce something well-reasoned.

> A simple search would have provided you with that, though: https://defundthepolice.org/alternatives-to-police-services/

A simple search yielding a page that ignorantly proclaims "But we have to remember that police do not prevent violence." Who responds to mass-shootings in progress? Did a cop not prevent someone from being stabbed in the Makhia Bryant case? Does solving cases and locking up the right people who are violet not prevent them from committing further crimes?

Your own link is agenda pushing, not a serious analysis of how & why policing has reached its current state, how this new system will avoid the same fate, and largely avoids talking about the potential problems of this new system.

>Hacker News is _supposed_ to be a place for curious and insightful discussion.

A reddit-style per-statement rebuttal would seem to be antithetical to that. You don't seem curious, you seem like you have an agenda.

EDIT: I'm going to further defend the link. It is exactly what you asked for: a detailed description of the ways in which policing is lacking as a public service and the initiatives that would replace it. That you singled out a single quote which you have ideological (and flawed) disagreements with does not change that face.

Splitting policing into multiple institutions with specific focuses (per the link: mental health, traffic, violent crime and gender-based crime, investigative services, and minor crime) would allow governments to optimize recruitment and training and to create new institution-specific cultures free from the baggage of modern policing. The link specifically points out the detrimental effect modern policing has on black citizens and communities of color and how they (along with the rest of the population) would benefit from the changes suggested. I have a hard time believing that you would be against this end, and assuming that you're not, I would love to hear your alternative solutions for countering the widespread abuse, corruption, and hostility police visit upon communities of color.

> The police have resisted incremental reforms,

That's not really the problem. Even where the police have embraced incremental reforms they've driven funding and responsibility into the police, pushing them into more domains for which they are inappropriate, and haven't dealt with the underlying problems because those aren't peripheral policy problems but fundamental structural problems of having a general purpose paramilitary force as a top level local government organization unto which almost all law enforcement and some peripheral tasks are layered fostering an insular warrior culture without any other strong focus. (Note that when the US was founded—before standing paramilitary domestic security forces were common enough to be a particular general concern—avoiding a very similar problem with military forces which inevitably would get leaned on for domestic security was a major reason for skepticism of standing armies in favor of relying on militia + cadre, and thus for the second amendment.)

Which also suggests what the “something else” to replace the police could be—given that abandoning professional permanent law enforcement for reliance on citizen militia/posses is probably not viable for modern society, one way to address that problem is to have specialized law enforcement entities in domain-specific agencies whose individual foci are narrower, enabling each to have a cultural focus (and substantive competence requirements) that are domain specific.

And if you look at US law enforcement beyond the local level, that's what a lot of it looks like already. The highly centralized paramilitary law enforcement structure where the top uniformed commander is a the head of a top level agency of the unit of government (and possibly independently elected and effectively unaccountable to the civilian government officials) is unique to city/county government.

I think you’re jumping to conclusions. We don’t know their standard process for reclassifying alerts. Maybe an analyst reviews every single alert and manual overrides are common. Maybe post-processing several months later is also standard practice.

I get how suspicious this looks, especially the changed geo-tag that occurred months after, perhaps at the behest of or influenced by the CPD’s investigation, but we should try to be as objective as possible before accusing people of committing felonies.

> We don’t know their standard process for reclassifying alerts.

and we don't because they withdrew evidence, rather than allowing outside review of said processes.

not really. the evidence is basically ShotSpotter's final analysis of the incident which they do using a combination of AI and human "experts". it's probably similar for DNA tests. there is still some human reviewing the results, possibly discarding error samples, etc.

>Motherboard recently obtained data demonstrating the stark racial disparity in how Chicago has deployed ShotSpotter. The sensors have been placed almost exclusively in predominantly Black and brown communities, while the white enclaves in the north and northwest of the city have no sensors at all, despite Chicago police data that shows gun crime is spread throughout the city.

But I thought there was nothing racially-biased about American policing.

Related: 11 years ago or so, 3 Tesla executives died when their small plane crashed in East Palo Alto, where the shotspotter system recorded audio of the crash: https://www.paloaltoonline.com/news/2010/02/19/shotspotter-s...

On one hand, it seems that Shotspotter lacks the accuracy to be used in court. On the other hand, it seems like a valuable probabilistic tool to help police perform detective work and collect real admissible evidence. The problem seems to confusing the two.

You can use an Ouija board to help find video evidence, but the video evidence still needs to show something incriminating.

In the Williams case, I feel like the key information is how likely ShotSpotter is to miss a gunshot completely.

They have video footage that shows the suspect and victim in the suspect's car alone at 11:46. If the victim was shot at 11:46 then this is extremely strong evidence of guilt, especially as it would also show the suspect lied. It seems reasonable to me that an algorithm might mistake a gunshot from inside a car as a firecracker out on the street. The classification aspect is not important. What the defence have to argue is not just that this sound is a firecracker, but the actual shot that killed the victim (which by the suspect's account happened in the same place), was not captured.

I'm not familiar enough with the case to guess at strategy.

The keys seems to be: If police have any evidence besides ShotSpotter to connect the time and place of the shooting with the video of the two of them together.

I don't they necessarily need other evidence for the time. The article and most of the comments here are focussing on its inaccuracy in geolocation and classification, but it seems to be an accurate recording of loud sounds that ring out across the city. A high false positive rate actually works against the suspect. He needs it to have missed one.

I think that really depends on the frequency of loud sounds detected. If possible gunshots are detected every 2 minutes, false negatives don't really matter in this case.

All technical support, including DNA, fingerprints, hair, fiber, etc. in law enforcement should be controlled outside of police departments and prosecutors, and should be made equally available to defendants as well as police and prosecutors.

With the level of conflict of interest, you can't call it "science."

Why didn't the judge find the prosecution in contempt of court? And why didn't this trigger an internal investigation inside the Chicago PD? Modifying evidence to fit a crime is itself a crime, withdrawn or not, and heads should be rolling right now.

The courts and the prosecutors and the police are all on the same team.

The adversarial, truth-seeking system you believe that this is is a fiction. It's a pipeline for money, nothing more.

Using privacy badger and noticing Vice pages now redirect to a 404 AFTER loading. Guess I'm not reading the article.

Privacy Badger, Firefox, and NoScript here, and I can read the article fine. I'd suggest you try a few js-blocking add-ons.

There needs to be a law that any technology used by law enforcement in the course of doing business is open-source.

I don't see how this isn't already the case given the lives at stake

> Rochester police had also refused his multiple requests for them to test his hands and clothing for gunshot residue.

GSR tests are total crap and in my experience lean very hard toward the false positive side of things, unsurprisingly nitrates are all over the place. Cops know this, so them not administering a test that can only hurt a suspect's chances with a jury - well the guys on the street certainly weren't out to frame him.

> The sensors have been placed almost exclusively in predominantly Black and brown communities, while the white enclaves in the north and northwest of the city have no sensors at all, despite Chicago police data that shows gun crime is spread throughout the city.

One country, two systems

Arbitrary death by firing squad? Check

Arbitrary expropriation of private property? Check

Arbitrary internment for one class of society? Recurring theme, but Check

Why is this considered "evidence" at all? Raw recordings of sound, with timestamps and locations of sensors, could be evidence, but something output from an AI?

Let's also consider that the speed of sound isn't constant, sound reflects off of everything, and can even be ducted by natural causes.

All you need to hear from this article is that no one outside of the company can independently audit the software

Any such solution should be dead in the water as a source of meaningful forensic evidence, period

How hard would it be to create a community network of crappy microphones to do what shotspotter does? You have a million phones that all have gps and microphones

Oh man the privacy folks are gonna hammer this comment hard.

I don’t think people firing guns have a reasonable expectation of privacy. Voluntarily submitting cellphone data is already used for several things like traffic congestion via Waze and related apps.

I am vary pro privacy, but as long as it’s voluntary I don’t have an issue with this.

It would depend strongly on whether the phone is analyzing the data from the mic and reporting to a server "possible gunshot heard, GPS coords X,Y, volume-level 60%, gunshot-like-score 75%" or whether the phone is sending raw recordings from the mic to a server and trusting that the server is only checking them for gunshot sound (not, say, saving your private conversations).

And of course it's difficult for the average user of an app to tell the difference, or even for a tech savvy user to tell without investing some time and effort.

Yes hopefully all the analysis would happen on device but that limits anything like a google home/alexa. Last I looked neither device allowed processing on device.

I am going to assume significant processing would need to happen on the device or your uploading audio 24/7 which would have heavy bandwidth implications. I don’t think uploading individual sub 2 second sound clips with time stamps is significant from a privacy standpoint, but it might be in the middle of a long gunfight.

I'd not put that on my primary device either, maybe leave a spare plugged in near an open window, etc

I'm a huge privacy advocate but it makes some sense. If the police and government is corrupt and using this tech, creating a non-corrupt open-source community driven alternative to combat the corruption is probably a net win despite the privacy issues. The real win is to destroy the Police surveillance state and tech but we as a people (the US) seem incapable of real policing change. :(

There are also the home assistance devices like Amazon Echos that have an always on Mic. Amazon could be a huge surveillance network for police with Ring video feed and a shot spotter network of echo devices if they went down that road.

If i was a programmer, I would be making this. a small alexa-like array of microphones on a RasPi for rough direction estimate, times a bunch of networked units microphones to get precise-ish TDoA, mounted on rooftops...

I've always wanted to do this to record and report on the number of gunshots I hear every night - hundreds go unreported - and to get an accurate direction less prone to reflections. Also to start an ML library to recognize thunder vs. fireworks vs. dumpster lids vs. planks of lumber vs. misfires vs. gunshots and other percussive noises.

I've always wanted to do this to record and report on the number of gunshots I hear every night - hundreds go unreported

If I were you, I'd expend 100% of my efforts into figuring out how to move to a safer neighborhood.

Despite all the news reports, in many cities in the USA people can and do live their entire lives without hearing a single gunshot.

And some places in the US hearing a gunshot isn't a cause for concern. People sighting in their hunting rifles, plinking cans, whatever.

If it meant an increased likelihood of catching the cops in a lie, I'd host such a device on my roof.

I'd just want an easy way to audit the outbound data so I could become confident that it was only recording cases of: "Holy crap that was an EXPLOSION" and not recording cases of: "Did you hear what that guy said in his back yard?"

It's not like anybody creating explosions has a reasonable right to privacy. Everybody heard it, it might as well be public record. This would just add timestamps.

How would the phone differentiate between real gun shots, and the gun shot noises my phone hears every day from the Television I watch, or video games I play?

You'd have to corroborate with the network I guess

television and video game gunshots don't sound like gunshots, nor do they meet the pressure threshold that a gunshot does.

But rather than argue that particular aspect, just have it be "off" when it's connected to wifi or moving faster than 30MPH.

Vice conveniently leaves out the fact that Toledo had a gun in his hand with the slide locked back which means he had emptied the clip. There is video of this.

Not to be pedantic but that's not what that means. That means that the slide was pulled back then released with the slide lock activated either manually or by an empty magazine. It's advisable to store pistols in this way with the magazine removed because it's generally the least threatening configuration a pistol can be in short of being disassembled.

Assumes facts not in evidence, to wit that he had a full clip at the outset of the incident.

There must be some laws against doing this stuff in the US? In the uk we have 'perverting the course of justice'.

We used a shot detection system… in Afghanistan. Not really sure it’s needed in the United States.

Stories like this are part and parcel why many people don't trust the police.

For some reason this isn't a felony offense by anybody involved.

Cops Lie. Here's a few examples.

https://www.rollingstone.com/culture/culture-news/nypd-shake...

https://popculture.com/trending/news/target-starbucks-tampon...

https://www.nytimes.com/2019/12/31/us/police-mcdonalds-coffe...

https://www.washingtonpost.com/news/morning-mix/wp/2017/08/1...

https://popculture.com/trending/news/nypd-called-out-falsely...

https://www.oregonlive.com/crime/2019/07/police-lieutenant-w...

https://www.foxla.com/news/deputies-arrest-kpcc-reporter-cov...

https://www.courier-journal.com/story/news/local/breonna-tay...

https://www.vice.com/en_us/article/939zda/how-stupid-do-cops...

https://www.vice.com/en_us/article/jmv94x/testilying-cops-ar...

https://www.ocregister.com/2020/09/25/oc-sheriffs-deputies-w...

https://www.cjr.org/special_report/officials-say-chicago-pol...

https://fair.org/home/6-elements-of-police-spin-an-object-le...

> Police said ShotSpotter, a surveillance system that uses hidden microphone sensors to detect the sound and location of gunshots.

Wait, what?

Reply on news.ycombinator.com