Ask HN: Tech crowd – What are your ideas for solving gun violence using tech?
45 comments
This problem is inherently American - it's not that other countries have access to some secret tech that's keeping them gun-violence-free. Saying "let's not make this political" about an entirely political issue is laughable. So is "let's fix gun violence with ML".
To actually answer the question with tech solutions:
- accurate fact-based media with no spin
- accessible education
- opportunities for people to be exposed to cultures, lifestyles and mentalities different from their own
Educated, informed and open-minded people are less likely to shoot up a school, or to buy into other people's "right" to do so.
To actually answer the question with tech solutions:
- accurate fact-based media with no spin
- accessible education
- opportunities for people to be exposed to cultures, lifestyles and mentalities different from their own
Educated, informed and open-minded people are less likely to shoot up a school, or to buy into other people's "right" to do so.
Agreed. But we are where we are now and have to face reality. The reality is, the political situation with guns is, and has been, stuck between a rock and a hard place. I want to keep those political arguments out of this thread and focus on how tech can help.
I think politically, you can get every side to agree that if technology could alert authorities that somebody is holding a gun in hand and walking up to the entrance of a retail store, it would be an improvement and can potentially save lives.
but yes, I agree with each of your bullet points. All three of those will have benefits far beyond solving gun violence as well.
I think politically, you can get every side to agree that if technology could alert authorities that somebody is holding a gun in hand and walking up to the entrance of a retail store, it would be an improvement and can potentially save lives.
but yes, I agree with each of your bullet points. All three of those will have benefits far beyond solving gun violence as well.
While it remains between a rock amd a hard place, people will be killed. I know that you're trying to steer the conversation towards the forum's purpose, but I'm not sure technology has anything to do with it, anymore than technology can stop the globe from increasing in temperature in its current state.
One of the gunmen was stopped within 60 seconds. People still died. Unless the tech can precisely point out terrorists before they leave the house, it risks being just a bandage. Even if it did, what whitepaper could possibly give you the confidence that the system's tolerance is effectively adjusted? That's some iRobot level stuff, and that was... not without quirks.
It'd probably be easier to just change the dang government.
One of the gunmen was stopped within 60 seconds. People still died. Unless the tech can precisely point out terrorists before they leave the house, it risks being just a bandage. Even if it did, what whitepaper could possibly give you the confidence that the system's tolerance is effectively adjusted? That's some iRobot level stuff, and that was... not without quirks.
It'd probably be easier to just change the dang government.
There is a political component, but I would say that the real problem is social. American society is broken in some fundamental way. If I had to put a guess at a root cause, I would say that too many Americans do not see "the other" as human beings. (While the right may be more likely to resort to guns, "othering" also happens on the left.)
Since I’m not a fan of people downvoting myself without an explanation here’s the explanation of why I did it.
You should avoid making statements such as “while the right maybe be more likely to resort to guns” without offering any sort of data to back that up.
Statements like this offer such information as a premise and lead the conversation towards the wrong direction because they have no base in reality.
You should avoid making statements such as “while the right maybe be more likely to resort to guns” without offering any sort of data to back that up.
Statements like this offer such information as a premise and lead the conversation towards the wrong direction because they have no base in reality.
The statement was tangential, and matched (so far as I can tell) most peoples' perception. As such, I didn't feel the need to document it.
You seem to be saying that "the right may be more likely to resort to guns" has "no base in reality". That would surprise me, for two reasons. First, gun culture seems to be much more prevalent in parts of the country that lean right. Second, the topic was mass shootings, and there have been several prominent mass shootings lately that have been done at least in the name of the right, whereas the Dayton shooter is the only one that I know of that was claiming positions of the left.
Neither of those are hard data. On the other hand, you seem to be the one claiming that what "most people know" is wrong; do you have data?
You seem to be saying that "the right may be more likely to resort to guns" has "no base in reality". That would surprise me, for two reasons. First, gun culture seems to be much more prevalent in parts of the country that lean right. Second, the topic was mass shootings, and there have been several prominent mass shootings lately that have been done at least in the name of the right, whereas the Dayton shooter is the only one that I know of that was claiming positions of the left.
Neither of those are hard data. On the other hand, you seem to be the one claiming that what "most people know" is wrong; do you have data?
Most people perception really have no importance here or anywhere where science is used. That is a fallacy. If most people think that I’m blue it doesn’t make me blue.
Gun culture is more prevalent in the right. Yes. Now saying that right wing people are more likely to resort to guns is a different statement. Would you agree?
That’s the kind of nuance that makes that statement a perverse premise in my opinion.
No. I don’t have any data, but I’m not making any claims so the burden of proof is not on me :)
Gun culture is more prevalent in the right. Yes. Now saying that right wing people are more likely to resort to guns is a different statement. Would you agree?
That’s the kind of nuance that makes that statement a perverse premise in my opinion.
No. I don’t have any data, but I’m not making any claims so the burden of proof is not on me :)
You are making a claim. You are claiming that both sides a equally likely to resort to guns and as such the burden of proof on that claim lies with you.
I don't know the solution, but I think I have an idea of where to research the root causes. And maybe it can be crowdsourced.
From what I understand, these types of violent outbursts come from a feeling of isolationism paired with mental health issues. These people see those around them as being part of "the problem" and flock to online communities that "understand them". These communities become cesspools of reinforcement of their twisted "solutions", and they lose all connection with the real world.
To combat this, we need to get them some connections so they can get over that feeling of isolationism. Unfortunately, it's really hard to help someone who doesn't want help, so we need to be a bit tactful.
This idea isn't really fleshed out, but here are some notes:
- social app where people can identify individuals they are worried about - people get connected based on mutual concern, and each can share how closely they are related to the individual - these groups are paired with a certified psychologist who will act as a mentor for the group - the group finds ways to expand the support group of the target individual and find healthy outlets for that individual - if the target individual accepts counseling, the support group can help fund it
I'm not a psychologist, so I'd need to discuss it with a few domain experts before trying to put this into production. That being said, there may be room to include gun stores and police into the mix to improve the ability of the community to prevent harm.
From what I understand, these types of violent outbursts come from a feeling of isolationism paired with mental health issues. These people see those around them as being part of "the problem" and flock to online communities that "understand them". These communities become cesspools of reinforcement of their twisted "solutions", and they lose all connection with the real world.
To combat this, we need to get them some connections so they can get over that feeling of isolationism. Unfortunately, it's really hard to help someone who doesn't want help, so we need to be a bit tactful.
This idea isn't really fleshed out, but here are some notes:
- social app where people can identify individuals they are worried about - people get connected based on mutual concern, and each can share how closely they are related to the individual - these groups are paired with a certified psychologist who will act as a mentor for the group - the group finds ways to expand the support group of the target individual and find healthy outlets for that individual - if the target individual accepts counseling, the support group can help fund it
I'm not a psychologist, so I'd need to discuss it with a few domain experts before trying to put this into production. That being said, there may be room to include gun stores and police into the mix to improve the ability of the community to prevent harm.
Violence has generally fallen since the early 90's, or at least data collection of violence has fallen, and the suspected reason is technology. Cell phones, internet, video games, might not only lower street crime (by changing drug distribution) but also, through social atomization, by giving people alternative social outlets than ones full of people they'd get into altercations with.
So one angle is direct social technology designed to somehow lower the crime rate. Just spitballing here, how about some improved version of NextDoor?
Another angle: Make it easier to talk to the police. For example, in Baltimore, a lower proportion of murders are getting solved than a few years ago because people won't give police information, and there is now a higher rate of murders, too. To improve the rate at which people report crimes, make it much easier for people to talk to the police. Have the social networks expose the local police department as part of the community in some way. (I don't really know how to make this work.)
So one angle is direct social technology designed to somehow lower the crime rate. Just spitballing here, how about some improved version of NextDoor?
Another angle: Make it easier to talk to the police. For example, in Baltimore, a lower proportion of murders are getting solved than a few years ago because people won't give police information, and there is now a higher rate of murders, too. To improve the rate at which people report crimes, make it much easier for people to talk to the police. Have the social networks expose the local police department as part of the community in some way. (I don't really know how to make this work.)
In the US, violence has fallen since the 90s for sure. I've heard of other reasons too. Less lead paint leading to development disorders, more economic prosperity, and more access to food.
I think the number one way to reduce violence is economic prosperity. It wont cure it, but it goes a long way.
I think the number one way to reduce violence is economic prosperity. It wont cure it, but it goes a long way.
Greetings,
This is an inherently societal (and by extension, political) issue as some have pointed out. Realistically speaking, legislative solutions do not appear imminent. Technological solutions could still be of use as follows:
1) predicting/detecting behavioural problems well in advance of the individual(s) in question becoming locked into a violent path (this would be challenging on both technical and ethical/privacy fronts)
2) more narrow-window prediction/detection of violent intent, and/or materials that would constitute threat to life (guns, as a prime example) and immediate notification of first responders. Essentially what some folks call a "sense and respond" system.
3) Environmental deterrents/safeguards coupled to the detection systems described in (2) above.
A multi-pronged strategy would likely deliver the best outcomess w.r.t minimising the loss of life. Solutions at each level could also be re-purposed for emergencies that aren't gun related, so the overall benefit to society would be meaningful.
Disclosure: I founded a startup over a year ago with the express purpose of solving such problems (== preventing loss of life in violent scenarios in general) and had to shut it down due to lack of funding. I have thought about these issues for a couple of years now and have some specific ideas on the subject. Happy to discuss further with anyone interested in collaborating, pointing me to funding opportunities etc.
This is an inherently societal (and by extension, political) issue as some have pointed out. Realistically speaking, legislative solutions do not appear imminent. Technological solutions could still be of use as follows:
1) predicting/detecting behavioural problems well in advance of the individual(s) in question becoming locked into a violent path (this would be challenging on both technical and ethical/privacy fronts)
2) more narrow-window prediction/detection of violent intent, and/or materials that would constitute threat to life (guns, as a prime example) and immediate notification of first responders. Essentially what some folks call a "sense and respond" system.
3) Environmental deterrents/safeguards coupled to the detection systems described in (2) above.
A multi-pronged strategy would likely deliver the best outcomess w.r.t minimising the loss of life. Solutions at each level could also be re-purposed for emergencies that aren't gun related, so the overall benefit to society would be meaningful.
Disclosure: I founded a startup over a year ago with the express purpose of solving such problems (== preventing loss of life in violent scenarios in general) and had to shut it down due to lack of funding. I have thought about these issues for a couple of years now and have some specific ideas on the subject. Happy to discuss further with anyone interested in collaborating, pointing me to funding opportunities etc.
Tech solutions of any kind will just be a band-aid. This is a very sensitive topic so it is easy to get political but the fact is that we as a society in the US have collectively accepted the fact that this is the cost of our freedom to be able to bear arms (I personally don't believe that). But this is how it is with the society today. Until that mindset changes, nothing will happen. We can surely add some tech solutions to help a bit but I personally would like to see a mindset shift first. The actual solution can be a combination of things but lets start with the mindset.
I agree. The complete solution involves many things at the societal level, and no one solution (that is actually practical in achieving) will be the ultimate fix.
Unfortunately, in today's political environment, band-aids might be our best hope at making progress right now, in the short term.
Unfortunately, in today's political environment, band-aids might be our best hope at making progress right now, in the short term.
I’m certainly wrong here but my understanding is that any crime happens because of two main factors:
1. Someone is not thinking straight. It could be from an ill, a temporary loss of judgement and other related causes.
Technology could aid in identifying these scenarios and trying to prevent such occurrences by treating the disease.
2. People think they can get away with it. Or, in other words, the risk of getting caught is smaller when compared to the benefit of the crime.
In that case technology could be used to avoid that people get away with crimes, thus making it high risk to choose that option.
But still, even when the majority of crimes are punished (and they are not) there will still be some crime because there are people who think that getting something that they don’t have is worth more than capital punishment.
I was born in Brazil where only 2% of murders are solved. Crime is not bigger than it is right now because people, to some extent, are inherently good. There is no expectation of punishment there.
1. Someone is not thinking straight. It could be from an ill, a temporary loss of judgement and other related causes.
Technology could aid in identifying these scenarios and trying to prevent such occurrences by treating the disease.
2. People think they can get away with it. Or, in other words, the risk of getting caught is smaller when compared to the benefit of the crime.
In that case technology could be used to avoid that people get away with crimes, thus making it high risk to choose that option.
But still, even when the majority of crimes are punished (and they are not) there will still be some crime because there are people who think that getting something that they don’t have is worth more than capital punishment.
I was born in Brazil where only 2% of murders are solved. Crime is not bigger than it is right now because people, to some extent, are inherently good. There is no expectation of punishment there.
The best line of code is the line of code you don't write.
The best gun is the gun you don't have.
Less guns less gun violence, it's not rocket science!
The best gun is the gun you don't have.
Less guns less gun violence, it's not rocket science!
This is not a technical problem.
There is no amount of technical solutions you can apply that will actually solve the problem.
So far as I can tell, all the other real solutions are ones that are considered inconceivable.
So long as all sides refuse to budge on the real things that could actually solve the real problems, I don’t see this situation changing. At least,not for the better.
There is no amount of technical solutions you can apply that will actually solve the problem.
So far as I can tell, all the other real solutions are ones that are considered inconceivable.
So long as all sides refuse to budge on the real things that could actually solve the real problems, I don’t see this situation changing. At least,not for the better.
Everything is a technical problem.
Armed police drones in the right spot could solve the problem, even if we do not want that for other reasons.
But everything is a technical problem.
Armed police drones in the right spot could solve the problem, even if we do not want that for other reasons.
But everything is a technical problem.
I have been thinking about solutions for a few months now. However, there doesn’t seem to be a solution that does not also create other problems (e.g. a system capable of detecting violent crime—or predicting it—is likely to encroach on privacy).
Another thought is that, systems that are/can be designed to recognize intent and predict events are (or would be) used to increase capital/revenue/market-share/etc.
I understand that people want to be both “free” and “safe.” But the conversation and/or conception of the relationship between freedom and security seems to be a zero-sum game.
I’m interested in what others think about this. Is it necessary that we exchange one for the other?
Another thought is that, systems that are/can be designed to recognize intent and predict events are (or would be) used to increase capital/revenue/market-share/etc.
I understand that people want to be both “free” and “safe.” But the conversation and/or conception of the relationship between freedom and security seems to be a zero-sum game.
I’m interested in what others think about this. Is it necessary that we exchange one for the other?
Tech like this https://smarttechfoundation.org/smart-firearms-technology/fi... ?
Very few of us who own guns actually want this tech, for a variety of reasons:
* The reliability isn't there. These are weapons. They need to fire in an emergency, when the trigger is pulled, 100% of the time, or as close as is possible. Fingerprint readers don't work 100% of the time, or even close.
One of the major trends in handgun design over the last 40 years has been the adoption of striker fired guns without manual safeties. They are enormously popular. Many people are buying them because they don't want to worry about disengaging a manual safety when grappling with someone, or while injured, and while dealing with an adrenaline dump. A fingerprint reader even less reliable.
The litmus test is probably whether or not you can get police to adopt these. They're more likely to draw their guns, and likely to do so at close range, and if you can somehow make them reliable enough to satisfy police, then ordinary people are likely to start buying them too. But I don't think Police are anywhere close to accepting these.
* Nobody expects them to survive long. Guns are some of the most durable consumer goods you can buy today. With proper care, they'll easily last decades if not centuries. Consumer electronics, on the other hand, rarely survive long. I don't want to buy a gun that will be a brick in 20 years.
Optics tend to have a shorter lifespan, and gun owners accept that as a trade off because it gives them something far more capable than iron sights. And most people still opt to have back-up iron sights so that a dead battery doesn't render their gun useless. But a fingerprint reader doesn't offer much in exchange for the problems it brings.
* A large fraction of firearm purchases are by hobbyists who already own multiple guns. Smart guns aren't interesting to us, since we already need to come up with a way to secure all our other guns.
* There is a fear that once smart guns come on the market, they'll be the only thing we're allowed to buy. (New Jersey actually passed such a law in 2002). Because of this, anyone who tries to develop such a gun faces backlash. S&W tried, and the ensuing boycott nearly bankrupted the company.
* Finally, only about 1% of firearm deaths are accidental. The overwhelming majority of firearm deaths involve someone deliberately shooting themselves or others.
* The reliability isn't there. These are weapons. They need to fire in an emergency, when the trigger is pulled, 100% of the time, or as close as is possible. Fingerprint readers don't work 100% of the time, or even close.
One of the major trends in handgun design over the last 40 years has been the adoption of striker fired guns without manual safeties. They are enormously popular. Many people are buying them because they don't want to worry about disengaging a manual safety when grappling with someone, or while injured, and while dealing with an adrenaline dump. A fingerprint reader even less reliable.
The litmus test is probably whether or not you can get police to adopt these. They're more likely to draw their guns, and likely to do so at close range, and if you can somehow make them reliable enough to satisfy police, then ordinary people are likely to start buying them too. But I don't think Police are anywhere close to accepting these.
* Nobody expects them to survive long. Guns are some of the most durable consumer goods you can buy today. With proper care, they'll easily last decades if not centuries. Consumer electronics, on the other hand, rarely survive long. I don't want to buy a gun that will be a brick in 20 years.
Optics tend to have a shorter lifespan, and gun owners accept that as a trade off because it gives them something far more capable than iron sights. And most people still opt to have back-up iron sights so that a dead battery doesn't render their gun useless. But a fingerprint reader doesn't offer much in exchange for the problems it brings.
* A large fraction of firearm purchases are by hobbyists who already own multiple guns. Smart guns aren't interesting to us, since we already need to come up with a way to secure all our other guns.
* There is a fear that once smart guns come on the market, they'll be the only thing we're allowed to buy. (New Jersey actually passed such a law in 2002). Because of this, anyone who tries to develop such a gun faces backlash. S&W tried, and the ensuing boycott nearly bankrupted the company.
* Finally, only about 1% of firearm deaths are accidental. The overwhelming majority of firearm deaths involve someone deliberately shooting themselves or others.
I brought this tech up precisely because of S&W's experience supporting it in NJ; you reminded me of the state and company that took the heat and backed off - thank you
As you're here on HN I'd be disappointed if you found the engineering points insoluble. Although I take your point that you're purchasing for the long haul, perhaps are a hobbyist, and that's cool, engineered solutions gong forward needn't ruin the shooting utility for collectors and hobbyists. Yet there'll always be a market for the old stuff. Firearms are not the only area where the old stuff is collectible and sometimes superior (tube audio amps, LPs come to mind) The social disagreements aren't as easily susceptible to fixing as the engineering problems. The questioner raised the question as if the social factors were subsidiary when in fact they are primary and trump all other considerations.
As you're here on HN I'd be disappointed if you found the engineering points insoluble. Although I take your point that you're purchasing for the long haul, perhaps are a hobbyist, and that's cool, engineered solutions gong forward needn't ruin the shooting utility for collectors and hobbyists. Yet there'll always be a market for the old stuff. Firearms are not the only area where the old stuff is collectible and sometimes superior (tube audio amps, LPs come to mind) The social disagreements aren't as easily susceptible to fixing as the engineering problems. The questioner raised the question as if the social factors were subsidiary when in fact they are primary and trump all other considerations.
similar, I recall seeing one where you wear a ring, and the ring's proximity allows the firearm to work.
This isn't a problem that will be solved by technology anytime soon. There are more guns in the U.S. than people. Even if you had a perfect tracking system for legally sold weapons, there are plenty of people willing to sell guns illegally, just like there are many people willing to sell drugs illegally.
A shooter can keep loaded guns in his car trunk, park in front of an establishment, and run inside in 20 seconds. It will be a long time before we have AI that can identify and subdue a gunman within 20 seconds.
A shooter can keep loaded guns in his car trunk, park in front of an establishment, and run inside in 20 seconds. It will be a long time before we have AI that can identify and subdue a gunman within 20 seconds.
I am looking for a more scientific and statistical approach.
Demanding that our Gov make statistic about gun violence including demographic, geographic, social and economic data on that and then act strong to curb gun violence.
If there is a single group that is outlier causing more gun violence limit the access to the guns on this group.
Start from the low hanging fruit basic statistic.
We can't treat everyone as the same as nobody is equal to others.
Demanding that our Gov make statistic about gun violence including demographic, geographic, social and economic data on that and then act strong to curb gun violence.
If there is a single group that is outlier causing more gun violence limit the access to the guns on this group.
Start from the low hanging fruit basic statistic.
We can't treat everyone as the same as nobody is equal to others.
Well you have backed yourself into a corner. You only care about gun violence and start from the assumption that it will always happen. The only way to stop gun violence in this scenario is by increasing the efficiency of the police force which is an incredibly inefficient way of dealing with the problem at hand.
You could use tech to improve the educational opportunities and equality within a society, thereby increasing societal cohesion. People with more stake in society are less attracted to the fringes, and people with genuine mental health issues are more likely to receive help.
Hi, this was an idea https://gtruceversionzero.carrd.co/ I had few months back about a platform mediating between violent gangs in the UK. Could this be re-skinned? All thoughts welcomed
The problem is the shooter, so lets look at them. Can we make an ML based system that tries to find people likely to end up as a shooter, but instead of being some Orwellian nightmare, sends a person out to listen to them?
It's obvious that:
1. Criminal who wants to mass-hurt people would try to hurt group of people who cannot protect themselves.
2. Criminal who wants to mass-hurt people (and ready to die himself) will find a ways to [possibly illegally] acquire means to do that.
Should we disarm everyone in a hope that criminals will stop being criminals?
How many people die of knife wounds from the hands of bad guys vs. from guns? From drunk driving or road rage? Granted it's less CNN news-worthy (not mass enough). Should we stop selling knives? Cars?
1. Criminal who wants to mass-hurt people would try to hurt group of people who cannot protect themselves.
2. Criminal who wants to mass-hurt people (and ready to die himself) will find a ways to [possibly illegally] acquire means to do that.
Should we disarm everyone in a hope that criminals will stop being criminals?
How many people die of knife wounds from the hands of bad guys vs. from guns? From drunk driving or road rage? Granted it's less CNN news-worthy (not mass enough). Should we stop selling knives? Cars?
The worst (# deaths) multiple homicides in USA history were arson with no firearm involved.
I mean the obvious answer is Robocop / I, Robot. High frame rate cameras with deep learning algorithms trained from mass shooting footage that control sniper rifles on pan tilt zoom mounts. Someone starts doing something bad, BAM catches a .308 Winchester in their dome.
unfortunately, there is no IT solution to this...
Bullet proof armour, I think.
jimmyvalmer(2)
I have sat here all morning, as a software engineer, trying to think of ways that would HELP to solve gun violence.
I want to hear what others ideas are, from the tiniest, small impact ideas, to ones that would be a huge undertaking/thought impossible.
All I am seeing elsewhere today is, "ban guns..", "solve mental health..", "more strict gun control..", etc. We go through these same "solutions" every time a mass shooting occurs.
I DO NOT want to make this thread about any of those political/societal issues. I want to see what ideas are out there with regard to using technology. Please keep political blaming out. Pointing fingers is not the goal here.
And the more detail you have about implementing a tech solution, the better!
Some ideas I've had/thinking about:
- Using Machine learning image processing to identify weapons immediately in an area where there should not be any. Instantly setting off alarms/notifying the police when one is exposed. Recent examples, somebody walking down the street of a popular nightlife area with many bars, holding a gun. Or walking up to the entrance of a popular shopping center. Even if the authorities had 5 - 10 seconds more to respond, it have a major impact on the outcome. I think I've also seen some startups working in the space. Of course it won't be easy. But deployed in the proper areas, such as crowded nightlife areas, could make a huge difference.
- Better tracking of guns location and how guns are traded. This of course is already a law in many places, but I think there are many ways tech can be used to do it better. Including how we can source as much information as possible, within the limits of personal privacy, and put those points together. I don't have specific examples here, would love to hear peoples ideas! I believe there is a lot of low hanging fruit.