Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

ShotSpotter: listening in on the neighborhood

computer.rip

Post reply on HN
Print view View original

301–310 of 443 posts

Re: ShotSpotter: listening in on the neighborhood

#301

Post by krisoft »

jakelazaroff wrote: post #207
resolutebat wrote: post #205

Earlier quoted context omitted.

"Incredibly unclear"? Really? Not the OP, but let's spell it out: Manicuring a lawn is expensive and serves no useful purpose, so most practitioners have disposable income and/or spare time, and care about keeping up appearances. Putting a car up on blocks implies it's being used for spare parts, which is generally not something the well-off need to engage in (they buy new and use professionals for repairs as needed)…

The reason it’s unclear is that nothing either of you have said has anything to do with policing. Edit: you added the correlation bit after I replied, but why do you believe that to be the case? If you have more police in an area, of course they’ll hear more gunshots there. That doesn’t necessarily mean there actually are more gunshots. What you’re actually saying is “poor people need more policing”, which is A) offe…

> If you have more police in an area, of course they’ll hear more gunshots there. That doesn’t necessarily mean there actually are more gunshots.

So let me get your point straight: your worry is that since wealthy areas have less sensors gun-crime in those areas will go under reported?

Idk about your neck of the woods but where i live if I hear a gunshot I call the police with a high probability. If i see someone brandishing weapons i do the same. And of course i call the police/emergency services if i see someone with a gunshot wound. These all create the statistical evidence independent of the sensor systems.

Re: ShotSpotter: listening in on the neighborhood

#302

Post by misja111 »

ryandrake wrote: post #203
Aurornis wrote: post #197

Earlier quoted context omitted.

> The more you monitor an area, the more samples you get, biasing the stats So by monitoring an area with high crime rate, you catch more crime, which results in more policing for the area with a confirmed high crime rate? And that's a bad thing?

By monitoring any area more, you find more crime, which results in that area being considered as "high crime" and in turn increasing monitoring--that's the cycle OP is talking about.

The devices monitor specifically for gunshots. I find it hard to believe that gunshots in wealthy area's often go undetected, with or without those devices.

Re: ShotSpotter: listening in on the neighborhood

#304

Post by bko »

mcmcmc wrote: post #279
throwaway9917 wrote: post #248

Earlier quoted context omitted.

The tone of your article makes me honestly, really angry. You know damn well that the reason the sensors are in those neighborhoods are because that's where people are getting shot. You even talk about how a school where some little kid got shot has a sensor, as if it's some sort of punishment for the lower income people there. Perhaps it's because the police and the city government want to deter or solve murders tha…

Did we read the same article? The main concern is that millions are being pumped into a surveillance program of dubious efficacy with zero accountability and clear biases. Budget that could be allocated to social programs that have a dollar for dollar higher impact on reducing violent crime is instead going into the police industrial complex, increasing surveillance on underprivileged communities instead of actually…

If I were in one of these high crime neighborhoods where people are getting shot, I would want more surveillance and police. This position that they're being exploited by surveillance is mostly a rich white cosmopolitan belief rooted in fantasy

> In fact, large majorities of residents in low-income “fragile communities” — including in both urban and rural areas — want more police presence, not less. In the more than a dozen low-income urban areas surveyed, 53% of residents want more police presence while 41% want the same — only 6% want less.

Not being shot is pretty low on the hierarchy of needs. And let's be real, it's a tiny percentage of people that are committing violent crime. Increasing the odds of correctly putting one person in jail prob reduces future crime greatly.

The criminal element is real and I'm doubtful that you can give someone who's killing people access to a food bank or job training and they'll just become a productive member of society. Being a violent criminal is almost certainly the least economical thing you can do. You end up killed or in jail in a short time span so to think someone rationally picks this as a career opposed to a minimum wage job is not realistic.

https://www.usatoday.com/story/opinion/2020/08/26/why-americ...

Re: ShotSpotter: listening in on the neighborhood

#305

Post by bnralt »

Aurornis wrote: post #73
bombcar wrote: post #8

Earlier quoted context omitted.

It sounds like shotspotters are likely to be near where shots are; and that the wealthier you are the more likely you’d NOT want to be where the shots are.

Putting surveillance for gun crime in areas with the highest rates of gun crime is obviously what’s happening. I think the author and some commenters are nervous about claims that gun crime is inversely correlated with neighborhood wealth, which creates some of these strange accusations that some other bias is at play. If you pull up a map of crime in most cities the inverse correlation with home prices is obvious. I…

I've noticed a lot of people start with the desire to not enforce laws, and work from there. So when there's more law enforcement activity in poorer areas, we get claims that the poor are being targeted and overpoliced. But when the wealthier people say they want more law enforcement in their area, the response is always "Why are safe areas like yours trying to take these law enforcement resources away from the poor areas who need it?"

The positions are entirely inconsistent, but that's because anything is being said in an effort to remove enforcement. This was stated pretty explicitly a few years back with "Defund The Police", and now that it's lost popularity it's being pushed covertly.

Ironically, for all the talk about needing to remove police to help the poor, it's the poor who have suffered the most from the massive increase in crime these policies have caused. Here's a community meeting from a poorer area a few months back, where the residents are complaining that people who live in safer areas are running an experiment in their areas that's leading to a massive spike in crime and death[1].

[1] https://wjla.com/news/local/gun-violence-shootings-crime-you...

Re: ShotSpotter: listening in on the neighborhood

#306

Post by tomas789 »

mszcz wrote: post #273
hcfman wrote: post #259

Earlier quoted context omitted.

I developed a TDOA based sound localization system for the Raspberry Pi https://medium.com/@kim_94237/tdoa-sound-localization-with-t... and with manual input you can greatly improve the accuracy from what an automatic system could do due to noise and signal degradation. However it’s very time consuming. The service level response would not be met if doing this on many cases I’m sure. Meaning the default accuracy of l…

This looks awesome. I’ve been wondering if it would be practical/feasible to create something similar but portable, self-contained for quieter, „local” sounds, like for instance locating a termite-like bug(s?) that’s been terrorizing my sanity for years now hiding in wooden beams.

Check out “sound camera” on youtube.

Re: ShotSpotter: listening in on the neighborhood

#307

Post by sokoloff »

splintercell wrote: post #298
smt88 wrote: post #174

Earlier quoted context omitted.

How would you know no one called it in? Also ShotSpotter has false positives. You have no way of knowing if you (or it) even heard a gun or one of many other things.

What's the problem with false positives in this case? I'm pretty sure, "don't bother with this technology which reports shots to the cops, because it has false positives" is a very weak argument.

If the false positive rate is 5%, it’s probably not much of a problem in terms of the overall balance of effectiveness.

If the false positive rate is 95%, it’s pretty obviously going to be a diversion of police resources from whatever else they could be doing if not responding to zero-value alerts.

Re: ShotSpotter: listening in on the neighborhood

#308

Post by bko »

nonrandomstring wrote: post #280
nonrandomstring wrote: post #262

Earlier quoted context omitted.

Previously done a lot of research into firearm and explosive acoustics (and the DSP to locate and categorise) and can say this is a _hard_ problem. Military versions for use in wide open spaces have a much better chance than in urban areas. Less than five percent seems unlikely, but I wouldn't be surprised to hear "less than half the time".

The frustrating problem with people down voting without any explanation is that simply ignoring the knowledge of experts and hiding whatever has emotionally enraged you does not further the discussion.

I think people don't "trust the experts" when they read the "experts" saying things like this:

> The reader can probably infer how this coverage pattern relates to race and class in Albuquerque. It's not perfect, but the distance from your house to a ShotSpotter sensor correlates fairly well with your household income. The wealthier you are, the less surveilled you are.

This sets off the bs alarms to me. The author knows this is a gross distortion of the rationale and is playing dumb. So what else are they lying about?

Re: ShotSpotter: listening in on the neighborhood

#309

Post by skippyboxedhero »

bnralt wrote: post #305
Aurornis wrote: post #73

Earlier quoted context omitted.

Putting surveillance for gun crime in areas with the highest rates of gun crime is obviously what’s happening. I think the author and some commenters are nervous about claims that gun crime is inversely correlated with neighborhood wealth, which creates some of these strange accusations that some other bias is at play. If you pull up a map of crime in most cities the inverse correlation with home prices is obvious. I…

I've noticed a lot of people start with the desire to not enforce laws, and work from there. So when there's more law enforcement activity in poorer areas, we get claims that the poor are being targeted and overpoliced. But when the wealthier people say they want more law enforcement in their area, the response is always "Why are safe areas like yours trying to take these law enforcement resources away from the poor…

The funniest part is having a (presumably) wealthy person knowledgeably inform poor people, who are the victims of these crimes, that they are "overpoliced".

From the point of view of someone who can roll into a poor area, say they are "overpoliced", and roll out again...yes, it probably does that seem way. That is because they aren't likely to get shot so the police are the problem, not the bullets.

For the person who can get shot, the bullets are the issue (this isn't limited to this area, you are seeing in multiple policy areas that wealthy people enthusiastically advocate for the state to be rolled back like some kind of giga-Milton Friedman...as with libertarians, they aren't the ones that have to deal with the consequences of this).

Re: ShotSpotter: listening in on the neighborhood

#310

Post by sokoloff »

modriano wrote: post #276
hamburglar wrote: post #198

Earlier quoted context omitted.

> These devices don't send audio over the network unless a shooting-like noise is detected. How does this jibe with the fact that the police can apparently request a ShotSpotter operator review of audio recordings for up to 30 days if a known shooting is missed by the system? How does this jibe with the fact that the system has apparently at least twice recorded voice conversations that were used (and thrown out in o…

The website says the audio buffer on the sensor holds the last 30 hours before overwriting. So police have less than 30 hours to flag a missed shooting so shotspotter staff can check sensors for audio around the shooting time. Regarding the court cases, I literally included the entire linked info for one of those Court cases in the post you're responding to. For both of the cases, the verbal exchange was at the same…

It seems plainly to be a wiretap under the laws of my state (MA): anything “capable of transmitting, receiving, amplifying or recording a wire or oral communication.”

With some additional carve outs, none of which appear to apply to ShotSpotter.

I’m not arguing that properly scoped and protected ShotSpotter couldn’t be allowed by MA legislators, but it sure doesn’t fall apart under scrutiny by my reading of the law.

The court seems to agree in the text you copied in your upthread quote:

> However, the court found that the exchange was an “oral communication” and that the recording was a prohibited “interception” under the Massachusetts Wiretap Act because the defendant had no knowledge that the exchange was being recorded.

Which I think is the proper interpretation of the written law.

Post reply on HN