▲ 102 points
back
55 comments
Related: ShotSpotter: listening in on the neighbourhood https://computer.rip/2024-03-01-listening-in-on-the-neighbor... (3 months ago, 523 points, 399 comments: https://news.ycombinator.com/item?id=39576974)
A lot of problems can only be solved by a combination of solutions A and B. If A and B have never existed together, it's trivial to write research papers "proving" that A and B are ineffective. Of course they are.
> combination of solutions A and B
That's not what ShotSpotter is selling. They're selling A. They tell you to buy A and it will reduce crime, increase arrests, or both.
What exactly do you think B is, anyway?
Your idea not to isolate variables is exactly what leads to bad science.
The OP never said ShotSpotter is selling A and B. The OP is making a general statement. Note I could see ShotSpotter combined with something else as being an effective crime fighting tool. I could also see ShotSpotter never working well.
What is an example of the "something else"?
Everything works better if you combine it with an unnamed, unknown thing that makes it betters. It's a nonsense assertion.
Aggressive law enforcement.
Where did I say to not isolate the variables? In fact, part of isolating the variables is understanding that any conclusions you make while testing A are exclusive to A and adding more variables might completely change the results. That's good science.
The issue with this tech is the following:
1. Get congressmen to earmark grant funds for shot spotter.
2. Purchase & Install
3. …
4. Cops get a report of loud noises.
The only useful thing I’m aware of is that these are installed in stadiums or other mass gathering venues and can triangulate location of a shooter in a chaotic environment.
I am not sure why this comment is being down voted. It's true. A good example is the combination of psychotherapy, and psychiatric drugs is the best solution for many mental illnesses. Using either one alone is less effective.
What is the "A" and "B" here?
Gun control and public healthcare.
I was under the impression Illinois had some of the tightest gun control in the country?
You can drive 1 hour, buy AR-15 in wallmart and go back without any border controls.
Gun control needs to be a nation-wide policy, and it needs to be implemented with all the common-sense measures like buying back the guns from population - see Australia for one example where it worked.
ShotSpotter seemed (in Chicago, at least) to be a huge source of graft. Before word got out of the cancellation there were some nodes installed in the more right wing or blighted wards last minute.
Scope is really important when people sign a contract. All ShotSpotter did was generate overtime hours for CPD without having any noticeable effect to crime. The amount of money ShotSpotter gives to Ald. Napolitano or the media to stoke anxiety about the city saving on SaaS bills is insane…
why does the mayor keep equivocating about its cancellation then after specifically running on the issue.
he turned it into schrodinger's cancellation.
This is not because of the technology but because of broken governance.
When someone is shooting other people, that person must be removed from society if that society is to remain functional.
The fact is that American cities simply do not do this. Japan does it. Singapore does it. And if the US incarcerated every person who fired a bullet at a person for the remainder of their natural lifespan, we could do it too.
Aside from a couple of dysfunctional police departments, everyone wants to catch violent criminals and put them away—no one is arguing against that. The problem is that they can’t be found for a multitude of reasons, this is also compounded by how easy it is to acquire bullets and firearms in the US.
I think California tried to tackle this issue by fingerprinting bullets (microscopic stamps I think) which can then be traced back to the person who bought them, but these people just end up buying bullets from other states and transporting them to California.
This is a very complex issue, prosecutors can’t just wave a magic wand and selectively put these people away. You have to find them, and build a case against them that has proof beyond a reasonable doubt that they were the perpetrator(s).
“Everyone wants to catch violent criminals and put them away”
I assure you this is not true in Memphis and many other cities with District Attorneys that are focused on elimination of jail time for nearly all crime. The DA blames gun policy, but then releases violent criminals (with real victims) back out on minimal bond. It is really sick. It’s not a complex issue.
100% this, with the modern redefinition of the word “racist” a lot of statistical correlation is being being reinterpreted as causation and people in control are attempting to manually “fix” systems that aren’t broken. It’s gotten so bad that crime itself is no longer punished in the name of “reparations”
In fact, the technology has never got to the point where it works reliably: https://en.wikipedia.org/wiki/Microstamping
Fill the prisons with police, you say? Sounds good to me.
Your worldview has produced Chicago. Mine has produced Singapore.
Singapore has some of the strictest gun control laws in the world regarding ownership and possession. If you want America to be more like Singapore, start by repealing the second amendment, not increasing policing.
> In addition, Singapore society is highly regulated through the criminalisation of many activities which are considered as fairly harmless in other countries. These include failing to flush toilets after use,[90] littering,[91] jaywalking,[92] the possession of pornography,[93] the sale of chewing gum,[94] and, prior to December 2022, sexual activity such as oral and anal sex between men.[95]
> Singapore retains both corporal punishment (in the form of caning) and capital punishment (by hanging) as punishments for serious offences. For some offences, most notably trafficking in drugs above a certain specified quantity, the imposition of these penalties is mandatory.
I’d rather hear gunshots and deal with property crime every day than be subjected to the above.
> I’d rather hear gunshots and deal with property crime every day than be subjected to the above.
I wouldn't. Which is a good reason to not have the same laws everywhere, I suppose!
Now ask the mothers of the recipients of those gunshots.
The laws sound ridiculous but they're generally not enforced. For example, I came across some truly disgusting public toilets in Singapore (granted, this happened rarely).
Sex between two men, while previously banned, was not enforced either. It is not uncommon to see openly homosexual men in Singapore.
Mmm, chicago
> This is not because of the technology but because of broken governance.
I mean it has a pretty big thing to do with it. It objectively doesn't work
Well... if you don't use the data from shot spotter in an actionable way and the mal-incentives for murder / illicit firearm possession don't change... yeah a tool meant to observe and record instances of gun violence probably won't prevent it.
This isn't about politics it's about understanding what a functioning society looks like.
Considering the false positive rate was 80%, getting something useful out of it is going to be hard
Agree with your first point, but I think most if not all politics boils down to defining what a functioning society looks like.
‘Everyone is trying to save the world, they just can’t agree on how.’
As it were.
Though some do seem to think the way to save the world is for them to get filthy rich and everyone else to die.
It does a great job picking up voices though
It's incredible how technosolutionism keeps coming back in Law Enforcement in either a reliance on the Chilling Effect or as a 'new' way of Predictive Policing, while they never work
Cool tech in theory. I didn't read how the study was conducted but I'll assume the results are correct. The note on the time to reporting makes me wonder about victim survival likelihood. I didn't see that in there, but I assume that 93 seconds could be the difference for someone who was shot.
I don't understand how police could show up later. This technology doesn't replace humans calling, right? I didn't have time to check out that other report, it's very long.
The guiding principle of science is to never assume that the results are correct.
Two observations: this is a highly political topic and there is highly political language in the linked article. Its conclusions also rely on the integrity of a multitude of referenced studies.
What this article and topic deserves is a colonoscopy level research critique / review.
I don't see political language at all. I also don't understand why this would be a politicized topic. There's a for-profit corporation selling snake-oil to cities, and then there are people (like the researcher/author) who want to reduce crime and save taxpayer money.
It's important to note that this researcher would easily get published regardless of what she found. If ShotSpotter works at all, that's a huge result! If it doesn't work, that's also very notable! And if the evidence is mixed, it still contradicts every other unbiased study that shows that ShotSpotter doesn't work. In a way, this was the most boring result.
Plus the author seems to go out of her way to be balanced, writing things like:
> " None of the seven prior investigation evaluations identified by my colleagues and I found evidence that case clearance increased following ShotSpotter deployment. While these findings read as an indictment of the technology, future research should determine if the manner by which police implement and utilize ShotSpotter impacts the likelihood that program goals will be met."
>I don't see political language at all.
The article links to openly political articles. Whatever the supposed context (excuse), this is a political act in the context of a research presentation. As if this is not enough, the article quotes unsupported statements of Brandon Johnson as well as counterarguments (though the latter are mostly burred in the article link). Which is unnecessary to present their data and is unequivocally political. That they are "both siding" the political argument should not be an argument: a. because it is on its face inappropriate in such a presentation and b. such a tack tries to hurdle the obvious critique question that would follow. Which is whether both sides are presented evenly. Are they acting as political journalists or trying to present data? Give us a break. The authors can write what they wish, of course, but don't ask us to turn a blind eye to the politics that are inserted into the article. Whatever the excuse.
> I also don't understand why this would be a politicized topic.
The topic is openly politicized, as everyone can see from the fact that it is hotly debated on the political stage. It is difficult to believe that you are being serious.
>There's a for-profit corporation selling snake-oil to cities, and then there are people (like the researcher/author) who want to reduce crime and save taxpayer money.
Okay, I see.
I rest my case.
What a waste of time.
>It's important to note that this researcher would easily get published regardless of what she found. If ShotSpotter works at all, that's a huge result! If it doesn't work, that's also very notable! And if the evidence is mixed, it still contradicts every other unbiased study that shows that ShotSpotter doesn't work. In a way, this was the most boring result.
It reads to me like you don't understand the relevant principles under discussion when it comes to research.
>Plus the author seems to go out of her way to be balanced, writing things like:
Is this your idea of research critique? Because this is the type of analysis, generally speaking anyway, that is accomplished in actual competent research critique. And a lot more. Which was the sum of my comment.
We must be in agreement that what this topic calls for is intensive critique and review of all research.
I'll especially be interested in whether or not the metrics are even meaningful, let alone if the results are high evidence, or if the research design itself is faulty. Of course, I wouldn't dare suggest a political reason for that should I see obvious faults in it. Like I said, I'd await a real critique / review and let that speak for itself.
Also there have been a history of ‘industry’ influenced studies in the opposite direction.
The whole thing is solidly dirty.
Yeah multi million dollar decisions, lifesaving decisions, this is intense.
So showing up 1-2 minutes earlier does indeed increase survival rates. For instances in which someone is actually shot, I estimate that -2 minutes = 1% increase in survival, https://link.springer.com/article/10.1007/s12061-020-09362-3. So in that case, ShotSpotter will save a life around every 100 shooting victims it covers.
Part of the rub is that people being hit are not all that common. (It may be false positives, or just no recovered shells on scene, but many of the ShotSpotter calls result in a nothingburger.) So what has happened in some cities is that ShotSpotter has increased workload in going to calls, so may have had negative spillover in that regards.
As a paramedic, drive time isn't necessarily the biggest factor, I feel, it's response times for EMS, particularly for wounds that are not easy to tourniquet.
After that, it's also generally not distance or speed of drive time - usually the factor that determines speed to the hospital is "opticom", aka traffic light pre-emption. I can move a lot more smoothly if my opticom is making the entire "lane" of intersections green for me, even if traffic is garbage.
There's also a self-selection in that study. Reality is most GSWs do not NEED to go to a L1 Trauma Center (the difference between L1 and L2 in most cases is as simple as "teaching facilities", the difference between these and L3 is who is required to have certain people physically at the hospital 24/7 versus "on-call").
That's something I didn't consider, you're right. I guess they have to weigh lives saved versus cost. A decision I am glad I don't have to make. Thanks for finding that data.
one of the purported justifications for the shot spotter system is that, in cities where gun violence is prevalent, people don't call the police when shots are heard.
If it's anything like loud mufflers where I live if you called every time you heard one you'd never be able to go about your day.
>This technology doesn't replace humans calling, right?
I thought that was the problem: that people in certain neighborhoods didn't call to report shootings.