I've thought about this a lot as I see more and more reckless driving in the areas I live in. Surveillance is generally a net negative, but it's also bad when you see people speeding around schools, rolling through stop signs, and running red lights. We seem to have a worst of all situations where traffic is getting increasingly difficult to enforce, driving is getting more dangerous year by year, and we're terrified of government overreach if we add any automation at all to enforcement.
I don't know the solution, but I do know that in the US we've lost 10-15 years of progress when it comes to traffic fatalities.
> Surveillance is generally a net negative, but it's also bad when you see people speeding around schools, rolling through stop signs, and running red lights.
The fact that these cameras are already pervasive and the problem of bad drivers hasn't been solved anywhere doesn't give me a lot of hope that these cameras are the solution to that particular problem.
It seems like police can do a lot to increase enforcement without the need of these devices. We have evidence that they've been doing less traffic enforcement so maybe start there. Increasing our standards for driving tests (some of which were eliminated entirely over the first few years of the pandemic) would probably help. Automatically shutting off/disabling or limiting the use of cell phones (all of which come with sensors that can detect when you are going at speeds you'd expect while in cars) might help. Bringing physical buttons and dials back to cars instead of burying common functions in touchscreen menus might help.
There's a whole lot of places to look for solutions to safer roads before we have to resort to tracking everyone's movements at all times.
In the US, a traffic ticket is an indictment of a crime (says it on the ticket. I wish I didn't know that fact).
That means that you have a right to trial/appeal, and the accuser (the cop) needs to show up, if you request a trial.
Traffic cameras can't accuse you of a crime, so they are considered civil infractions (no points, but also means they are a bitch to appeal). They can issue realtime civil citations, though.
ALPRS can't do either. They are forensic tools; not enforcement tools.
I believe in the UK, a camera can convict you of a crime, so they can issue severe tickets. They wouldn't really be able to do that, in the US.
In my county (Suffolk, NY), they just stopped all the redlight cameras. I doubt they would do so for ALPRs.
Also, I think some ALPRs are private. There's a shopping center, not too far from here, that's in a relatively high-crime neighborhood. They have cameras and ALPRs, all over the parking lots.
> Automatically shutting off/disabling or limiting the use of cell phones (all of which come with sensors that can detect when you are going at speeds you'd expect while in cars) might help.
I can’t think of a way to implement this that wouldn’t ban passengers from using their phone while riding in a vehicle. Which could be even a bus or limousine.
I don't disagree, but I can totally imagine a society where this inability is perfectly acceptable because it severely reduces the #1 killer of people from 5-55yo. I don't think we live in that society, if Apple and Google flipped a switch tomorrow to do that people would freak out, but I could imagine a rational, fictional society that had different shared values.
Not entirely.
The phones can defect if there are other phones nearby, so a single phone in a car on a highway going 75mph could be assumed to be a driver, but that is still just an assumption.
A lot of people would be fine with that. Drivers are impaired while on the phone, even hands-free. Not to mention texting while driving!
I kind of picture the cellular telcos doing this. Maybe buses and trains come with wifi hotspots allowed to connect. Otherwise auto passengers could use their devices offline, maybe read an ebook or something. Not the end of the world.
Lots of cars now come with a WiFi hot spot as part of their offerings. There's no way to prevent the driver from also connecting to it and circumventing whatever ill conceived notion this is
Even connected to wifi a cell phone canstill use the wireless network. Even airplane more won't actually stop your phone from connecting anymore. GPS data can also be transmitted in the background over wifi back to apple/google and/or the device manufacturer.
If they really wanted to push this they could do it directly in the baseband chipset and bypass the OS entirely when deciding to lock down the device to some kind of "travel mode" with limited functionality (such as no texting or no browser)
Not that I'm advocating for that sort of thing, but it's good to keep in mind that we don't really own the cellular devices we pay for and that even in the rare case we have root we can't stop them from doing what they want to our devices as long as they control the closed hardware.
I mentioned buses and trains, and was thinking that only those mobile wifi hotspots would be permitted, whitelisted for 5g service. Hotspots in (human driven) cars would not. That might encourage some people to take the bus?
I agree with the other poster about this being more workable in a fictional society with different shared values.
The standards for evidince, processes for enforcement and court side of things are not set up for cheap enforcement of "that clearly ain't right" behavior. They're set up for revenue enforcement of easy to prove but not necessarily bad in abstract offenses.
Police can't substantially increase enforcement overall because that would just cause bad political optics, say nothing of stops that needlessly escalate to being newsworthy in a bad way. They'd necessarily issue a hundred petty bullshit tickets for every deserved ticket for legitimately bad behavior. It just wouldn't work. It would be like trying to plow a field with the ripper on the back of a bulldozer. It kinda looks similar but it's wrong for the job.
And all of this is based on the assumption that we're trying to enforce things that the broad public agrees need strict enforcement, not whatever the original comment wants.
They have done automated enforcement a few times and it always sucks because the systems don't use discretion.
Someone going 40 in a 30 and swerving around other cars gets treated the same as someone going 40 when the road is empty. Someone slowing to 1-2mph before safely rolling through a stop sign get the same ding as someone blowing through it at 30mph.
If AIs can somehow learn how to take all this footage and enforce the spirit of the law (citing dangerous driving) instead of the letter I will fully support it.
It's an opportunistic solution to a budget problem.
Local governments have extra money from property price bubble increasing tax revenues.
At the same time there are great open source image analysis models for companies to put on a pole using cheap android hw and a solar panel and make bold claims about solving all crime, then they can sell that data again to 3rd parties like insurance. Also can start buying access to more video feeds from ring cameras etc and resell that. They'll hire someone to make a ux to integrate it into PDs later, for a premium of course.
We have very few alternatives to driving in the US so we have very lax driver training and testing.
Across the US we have roads and infrastructure that encourage speed right next to decaying pedestrian infrastructure. It's very difficult to get state DOTs to roll back or do traffic calming. They often prohibit the use of bollards or barriers near these roadways.
In a lot, not all, physical changes to the environment could drastically reduce traffic fatalities without surveillance.
Yep, the US has built itself a car-centric society over the last century, and there's essentially no way to change that now without burning everything down and starting over from scratch. They can't make driver training too strict because too many people would fail, and then have no way to get to work, and then who knows what those people would do. They can't turn cities pedestrian-friendly without tearing down everything there now and just the legal hurdles there are immense. I just don't see a way for this to change meaningfully in a lifetime, outside of a few little pockets here and there (like some towns or cities creating extremely limited pedestrian-only zones, which would then need a lot of parking to be useful for people, basically like an outdoor shopping mall).
3. major thru-way between shops, apartments, and the rest of the city
4. great visibility
this is a recipe for 50mph. the speed limit is 25mph. If you do the speed limit, you WILL be tailgated. If you do ~35, you're risking a ticket. There will still be people doing 45-50 and weave through the lanes.
also in my town, the main thru-way is a route dating back to the 30s. There are red lights at major intersections and they WILL turn red even if no one is there. They're designed to slow people down. HOWEVER if you speed and run a yellow light, you'll hit ALL the lights green! It shaves significant time off your trip, is easier on your car, is more enjoyable, and requires less attention. It's a system designed to make people speed and run reds.
Where I used to live, I could get from one side of the city to the other in a maximum of 30 minutes. the lights were designed to keep traffic flowing at 30-35mph. It ENCOURAGED you to go no faster, or you'll have to slow down and come to a stop. This also kept traffic flowing so you felt like you HAD to focus on driving. They also did things to encourage bicycles and make things safer for pedestrians.
I agree. It's frustrating that we have ended up in a reality where vehicle movement is heavily tracked, but we're not using that technology to do the most obvious and productive thing.
My city spent a few million dollars installing Flock cameras to all its municipal parking garages in a matter of months, but has been hemming and hawing over adding a few speed cameras for years, despite petitioning the state for an allowance do so back in 2023.
Traffic enforcement cameras don't even have to become the networked surveillance system that Flock offers. Most are still cameras triggered by radar rather than perpetually recording all drivers.
I live in NYC. People used to be afraid of double parking. Like you I regularly see the same bat-shit driving and no one seems to care to say or do anything. It's bonkers.
NYC should have been the model to follow. Instead of flock cameras, cities should have bounty systems: record a video of a speed violation with a plate, and get 10% of the ticket revenue. Enforcement would explode.
We could of had a system where we used the technology we already had in our hands to democratize speed enforcement, instead of corporatizing it
NYC already tried Snitching as a Service during COVID, and it went terribly. I grew up with a neighbor who would constantly record people and call the cops over every little perceived infraction. Everyone in the neighborhood hated her, including the cops. I do not want to live in a society that encourages those people.
These cameras are currently not used at all for traffic/speed enforcement. The best they would do is track more serious crimes like hit-and-runs by photographing cars in the area.
On my way home, I noticed at a stoplight across the street from Apple Park that the driver in the lane next to me had his phone mounted up high in landscape mode and was watching the Simpsons. Just absolutely unhinged behavior lately.
Here in Washington DC we've set up many different types of traffic cameras. One side effect is a reduction in the number of negative interactions the average person has with the police. There are communities where the cops treat people like piggy banks and spend tons of time aggressively enforcing minor traffic violations. That has negative effects on trust, not to mention the number of fatal interactions that have stemmed from traffic stops. Other jobs that can be done by the police in other places are handled by Parking Enforcement, the Department of Buildings, etc.
The recent presence of federal agents and soldiers has reversed some of the hard-fought gains in trust, but my broader point still stands: more automated enforcement of traffic laws has positive effects in how people interact with the police. This effect needs to be balanced against the harms of increased surveillance.
Your examples don't seem to represent the leading causes of fatalities from traffic accidents in the US, which remain distracted driving (phones and crappy touchscreen controls, etc) and drunk driving. People rolling through a stop sign / performing a 'California' stop are nowhere near the massive security concern to authorize the ideal surveillance state. To deal with the bigger issues you would have to film peoples actions in their cars and run it all through ai to accuse them of driving while drowsy. That would be an incredible and lazy failure. Also, fatalities per mile had a massive surge in 2020-2021, but they have been steadily dropping since then (2024 in wikipedia estimated a 1.27 per 1m miles, about the same as 2008). If the trend continues, in a year or two we will be back to the early 2010s lows without draconian measures.
The solution, as always, is better infrastructure and support at multiple levels, not beating everybody with a stick.
I get the vibe from his comment that the visible deviance from rules of the state is what he's really concerned with and that the death and destruction is just the reason everyone else is suppose to care.
A camera doesn't stop those acts though, it may only discourage those who know about it at a huge cost of privacy and rights.
How about we build better infrastructure and regulate vehicles since those do actually stop this behavior. Most of those red lights and stoplights in the US should be roundabouts. Narrower lanes and other traffic calming measures should be much more pervasive. Vehicle size, specifically bumper height is out of control.
Compare US traffic and pedestrian deaths to the rest of the world, or at least a lot of EU countries. Its embarrassing.
Thanks for this CDC link! So the pedestrian death rate and the overall road traffic death rate is about three times higher in the US compared to other high-income countries... Mind-boggling.
> Vehicle size, specifically bumper height is out of control.
Many manufacturers are now selling pre-lifted trucks. Here in Texas around a third of the vehicles on the road are pickup trucks, and about half of those have been lifted beyond standard height either from the factory or aftermarket, another third of vehicles are SUVs, most of which are significantly larger than necessary to be fit for purpose for the driver and occupants.
This situation was /caused/ by government regulation and it can be fixed by government regulation. It's absolutely absurd the gargantuan vehicles most people drive in the US, and the fact that we let people turn their vehicles into monster trucks and then operate them on public roads with impunity. I don't care how small someone's dick is, they don't get the right to drive a truck down the highway that can literally drive /over/ a modern standard sedan/hatchback. The continuing absurdity has turned into an iterated prisoner's dilemma which has resulted in more and more people buying SUVs and crossovers who by every measure do NOT need them. Absolutely is out of control, and it negatively impacts everyone, including the drivers of these vehicles.
Eventually we’ll have autonomous vehicles which will mitigate many of these issues, but will the surveillance infrastructure then be reduced? Probably not.
My observation has been that the danger increase is a combination of three things:
1. Lack of situational awareness/awareness of surroundings. This manifests in various ways like improper merging, improper turns into traffic, turning across multiple lanes, and left-lane hogs.
2. Frustration becomes aggression. This also manifests in multiple ways, but primarily is seen through tail-gating (in response to left-lane hogs) and swimming through traffic (in response to left-lane hogs), as well as road-rage (mostly in response to the other misbehavior).
3. Constant distraction. I have seen SO MANY drivers literally watching videos on a tablet, playing games, or otherwise driving at high rates of speed (70mph+) while fully engaged in something other than driving. It's at epidemic proportions.
The issue is, by and large, not people doing rolling stops (which by the way have never been proven to cause an increase in accidents when performed properly) or speeding (generally, some exceptions like school zones do matter). Running red lights is definitely a problem that is more common place, and is likely a part of all 3 of these these.
I feel like to a large degree all of this is a symptom of a wider societial issue where everyone acts in selfish and self-centered ways, completely ignoring their impacts on others, and moving from moments to moments where they can engage with their phone/social-media, to the exclusion of all else. I don't think phones/social-media /caused/ the problem, I think it exacerbates it though. Every aspect of our society is worsening because the revealed behavior of our population is one of lack of care or outright disdain for everyone else around them and an absolute obsession with serving their own interests above all else. That this manifests in driving is unsurprising.
IMO the issue is that there's little to no enforcement outside of speeding. I see people get pulled over for speeding in a straight line even if it's relatively safe to do so, but never for talking on their phone, messing with the touch screen in their vehicle, left turns from the right lane and vice versa, failure to keep lane, failure to yield, failure to signal, cutting people off, driving a vehicle unfit for the road, etc.
It usually takes about 5 minutes of driving to observe someone doing something that I would pull them over for. I don't think cops need all this automated surveillance, they just need to drive around and be proactive.
> Surveillance is generally a net negative, but it's also bad when you see people speeding around schools, rolling through stop signs, and running red lights.
Speeding and running red lights can be combated without affecting the privacy of innocents on the road. A debate can and should be had about the placement of radar traps though, many are straight off highway robberies trapping people who don't notice a speed-limit sign that is visually hard to notice.
I would actually be for civil asset forfeiture in these cases. The laws of physics don't care about due process; sue the car and save the video. Let the owner pay the fines and if they can prove some other driver did it, let them file civil suit against them to recoup the fines. Driving on public property is not a right.
We cracked down on driving under the influence with changes from DWI to DUI. In the 10-15 years you mention, the prevalence of distracted driving from mobile devices has gotten out of hand. There's no field sobriety test that can prove one was distracted by a device. That makes this much more difficult to crack down on.
Only if you're driving next to the car. Speeding obviously can be detected from a distance. Driving under the influence also can be seen from a distance.
I could see having a cop and a camera stationed in a place working as a spotter that then radios to other cops up ahead to pull over in an old school speed trap set up. However, that would take a lot of man power and coordination that I doubt most departments would care to do. We've seen plenty of YT videos of people trying to make a point of catching distracted drivers. The cops just need to give a shit, and want to do something about it.
It doesn’t take a targeted trap or driving next to. It’s like 1 in 6 drivers at any time; it’s super easy to spot if you’re looking. One cop could pick off drivers no problem without any coordination.
Part of the public pushback is that people almost always drive the “feels like” speed and not the posted speedlimit. We build 6 lane roads and then wonder why people go 50mph when it’s 35 posted, it’s because it’s 6 lanes and 35 feels slow. Cities profit from this in the form of speed cameras, which is why they’ve been outlawed in a lot of places.
The driver blithely keeping with the flow of traffic is not the one I am worried about. It is the one who is aggressively trying to cut through the flow of traffic while putting everyone and themselves in danger that I worry about.
I love when cities time their lights so that aggressive drivers just get hit with waiting at a red light while driving the speed limit means hitting greens for long stretches.
Yes, and about twice I've seen this done really right, wherein they post signs of the synchronization speed (e.g., "Lights Timed for 35mph"). I just get in sync with one light and adjust speed and it feels almost magic to go a few miles hitting every green light (it kind of is the macic of math).
It'd be cool if more roads were implemented that way.
I have the same in my area, but instead the lights are synchronized to slow traffic as much as possible. They literally coordinate to make you stop at as many lights as possible and grid lock at rush hour so the highway doesn't get flooded.
I walk my kids to school, but when I do drive the 1.1 miles there are no less than 12 stoplights/stop signs.
There's a part of downtown here that is known for being set up like this. However, it's been a bust while there's been a lot of construction blocking lanes so nothing moves at speed.
Sure, some people will get hit with the reds. Some people will learn the timing and learn that you can run just one red and then hit all greens, or that it will save you 5+min if you hammer down and pull a "clearly not ok" pass to get around some idiot who's going too slow to make the green timing.
So yeah, you're reducing speeding, number go up, pat yourself on the back. But you're also increasing the incidence of something rarer than speeding, but way worse.
This is the same problem that 4-way stops at roads that don't deserve them create, you're basically teaching people that the signage is bullshit.
Upping the speed limit reduces the incidence of the latter because just about everyone has a "fuck this I'm weaving" threshold and the number of people who hit it goes down when you reduce the incidence of rolling clusters of traffic caused by the handful of people who religiously follow the speed limit even when not appropriate.
If the speeds aren’t appropriate for the built environment, then the limits should be changed or the environment should be changed. Enforcement of the law should be consistent regardless of the quality of the law.
The speeds are appropriate for the roads generally. I mean, at the lowest level speed limits are a matter of social consensus so the broad public is tautologically correct.
The problem is that knocking the magic number on the sign down by 5-15 and then simply not enforcing it too seriously results in less screeching Karens harassing the politicians who then harass the bureaucracy than taking a hard line about "well akshually this is the engineered speed for the road".
The speed cameras in San Francisco have to result in lowered speeding over the 18 month period they're active. If they don't, they will be pulled. Seems pretty well-designed. Perhaps the fines are weak but it's good that they're there.
Just because a road "feels" like it can handle more speed does not mean that it is. The wider streets are built to handle the volume of cars, not necessarily meant to become a speed way. There are several 6 lane roads in my area while being wide and well built still have many intersections only controlled by stop signs for the smaller streets with multiple intersections controlled by stop lights at the larger cross streets.
People unable to recognize this and only driving by the feels are the problem. Hand wavy comments like yours suggesting using the feels as being okay do not help the situation
Basing your speed on what the road looks like may not be “okay” or “legal”, but it’s what people do. It’s just not useful to claim that individual people are the problem when this is something that is overwhelmingly true across the entire population—-a broader solution than individual responsibility is the only thing that will actually work.
Speeds are fundamentally a tradeoff between risk and reward. In nominally democratic societies we place these thresholds based on some approximation of social consensus. The general public literally cannot be wrong because their rough consensus, the fat part of the bell curve if you will, is what determines what the right speed is.
Your comment is literally the principal Skinner "no it's everyone else who's wrong" meme.
Um, no it’s not. If you think that roads are built with no consideration for anything other than if the infrastructure will support traffic at a certain speed then you’re just not thinking about things. Speed limits are set with many factors in that decision. Things like noise and safety are major parts of that. Someone else has already suggested a lame reason as a noise complaint made by a single person, but cars moving faster make more noise. Cars moving faster limits the time a car at a stop sign can safely navigate causing traffic at cross streets.
Regardless of what you think, speed limits are not set in place just to ruin your day because you can’t leave on time and constantly need to “make up time”. They are not arbitrary decisions just because you haven’t considered all of the factors involved.
Now you're relying on Jimmy "Buck" Rawgers born and raised in Billton county working at the DOT setting the speed limits and the traffic light cadence.
How many roads are 35 when they should be 50 simply because some local yokel asshole made a stink at city counsel 10 years ago and now it's impossible to change?
The solution is to have you abused by the system so that you realize that petty deviance is not something the dragnet should be brought to bear on if it must exist at all.
Your opinions are directly counterproductive on these petty issues. You ask for the state to use the jackboot. The jackboot just makes people hate the state and think in terms of "will I get caught". If not for you people trying to force compliance on this, that and the next thing the state would be in higher standing in people's minds and voluntary "when it matters" or "because it's the right thing to do" compliance would be higher. Sure, we could add more jackboot, but that costs money and no democratic-ish system is gonna allocate a bunch of money to do stuff everyone hates.
Not over the long term, no. There may have been a recent uptick in the post-pandemic US but it's mostly just noise. Fatalities per mile driven have been going down markedly in recent decades. Driving was twice as dangerous in the 80's as it is now.
You are incorrect. Fatalities in the US leveled out in the early 2010s and have been climbing since then. In all other developed nations they continued trending downwards.
This is not a statistical anomaly that can be handwaved by pointing out that things were worse 40 years ago. Roads in the US are uniquely lethal and getting moreso.
Sigh. I hate that phrasing. But OK, fine: you are misreading me, misanalysing the data, or just plain spinning to mislead readers.
Fatalities per capita and per mile driven go steadily downward until covid, and maybe there's a bump after that: https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in... If you have numbers (you don't cite any) showing otherwise, they are being polluted by demographic trends (the US having higher population growth doesn't say anything about driver behavior).
> Roads in the US are uniquely lethal and getting moreso.
So spinning it is. Would you rather drive in Germany in 2002 or the US in 2025? Seems like "uniquely lethal" doesn't really constitute a good faith representation of the truth.
According to the link that you posted, the roads in Germany in 2002 were quite a bit safer than the roads are in the USA in 2025. And they don’t have speed limits. Absolute no brainer to me.
Anyway, not to pile on but you are absolutely incorrect. Forgive the phrasing.
German roads absolutely do have speed limits. Only certain rural sections of the Autobahn don't, but that's not representative of the country as a whole, or even the Autobahn as a whole.
Did you open the wikipedia article you linked? The first image contradicts you, see the caption:
> Per capita road accident deaths in the US reversed their decline in the early 2010s.
Amusing that you accuse me of bad faith framing and then pose a nonsense question like this:
> Would you rather drive in Germany in 2002 or the US in 2025?
I cannot time travel and neither can you. The comparison that matters is US in 2025 vs other developed nations in 2025, and with that framing the US is uniquely lethal.
Of course, a good faith reader of my comment would understand this, but we already know that's not you since you did the research and have decided to be wrong anyway.
One thing this surfaces nicely is how scale changes the privacy model. Individually these look like “mere observation,” but once you can reconstruct routine movement patterns across counties, the data starts behaving more like long-term tracking than casual surveillance.
Retention is one good line between enforcement and tracking.
If we wanted to solve this problem while retaining the benefits, what we'd do is have stiff penalties for warrantless retention and required yearly independent audits of systems.
I understand why these statistics may be interesting, but all I really want to see is a map of the locations of the ALPR cameras. I would add an easy link to that data on this site.
This is great work. Once ALPR coverage is dense enough that you can’t go anywhere without generating a permanent record, the “mere observation” argument falls apart. Mapping this openly is one of the few ways communities can actually understand what they’ve signed up for.
The county lists are wrong, at least they are for my state of New Jersey. We have 21 counties, not 27. Is it picking up the bordering counties that might have overlapping contracts or something?
It pulls counties from OSM administrative boundaries of level 6, which according to the OSM wiki, is "State counties and county equivalents, Territorial municipalities" (https://wiki.openstreetmap.org/wiki/Tag%3aboundary=administr...). It does lead to some weird oddities, like counties with under 10 homes... I'd rather not manually correct it, since I want to rely on pure OSM data. (Unless you mean there is an actual bug in the processing and there's counties listed that aren't in the right states...)
I'll add a link to the OSM relation for the county to each county page, so you can see the source data on OSM to verify/edit.
One of those counties is Rockland, which is in NY. I wonder if it's counting bordering counties for states, since the assumption may be that the closest essential services for some Rockland residents may be in New Jersey.
Ah! This looks like a bug with cross-border calculations. Ideally it would not show a Rockland County, NJ - it might cross state borders to perform calculations but they should all have a 'home state' that matches to the right county. Thanks for the example case, I'll work on a fix
I know at least in Rockland's case, their power utility extends slightly into NJ. Can see it on statewide power outage trackers like this one: https://projects.nj.com/data/outagetracker/
I don't see how this provides actionable info for the individual. Unfortunately, this is just going to be a dashboard for pro-surveillance elements to see "how are we doing in our neck of the woods?", or a sales tool for Flock to find untapped markets.
My goal is to provide actionable statistics for any 'deflock' movements in a certain county, by being able to point to specific statistics on surveillance. If even one motivated person uses my data to petition, I'll be happy; it doesn't have to be for the average person. There's tons of these movements, too. Deflock Olympia just succeeded: https://www.yelmonline.com/stories/commentary-olympia-joins-...
Also, another answer to this is that there is no overarching goal; I just wanted to build a large scale data analysis pipeline for fun :) I am no stranger to side projects to distract me from finals unfortunately.
Fair enough. I am sorry if my original post was too harsh or overly critical; my original thinking was simply that those would organize against surveillance are probably already aware of it, but perhaps this would be useful in the cases you mention, if the metrics could be used as "ammo", as you suggest.
Not too harsh at all! I just wanted to fill the gap between "man, those cameras seem to be everywhere" and "XX% of people are surveilled on the average commute". It's true that this is not a particularly ambitious project, just a small niche I wanted to fill.
I think we have a surplus of "awareness" tools/websites that are great at what they do, but not much "rubber meets the road" tools to guide the user in actually taking action based on the information presented. I, for one, feel a huge sense of fatigue at the amount of awareness I have of problems I don't have the tools/strategies/knowledge/time to solve.
(This is not a negative comment about this post btw, more just commentary on how the fire hose of "look at how bad this is in excruciating detail" can be overwhelming.)
100% coverage seems like an inevitability in a country where filming in public is a constitutionally protected right and networked ALPR capability is possible (if not regularly offered yet) in commodity doorbell cameras.
> 100% coverage seems like an inevitability in a country where filming in public is a constitutionally protected right a
It really doesn't have to be though. The rights of individuals to record in public doesn't have to translate to the right of corporations (flock, amazon, etc.) to do it without restriction. Time, place, and manner restrictions on our rights already exist, it just needs to be found that this manner is unacceptable as an imposition on our freedom which should be protected under the fourth amendment.
If a home or business owner sets up Ring cameras, is it fair to say Amazon is recording in public? That feels like blaming Canon for the behavior of a paparazzi, but perhaps there are reasons those aren't equivalent I'm not aware of.
> hat feels like blaming Canon for the behavior of a paparazzi, but perhaps there are reasons those aren't equivalent I'm not aware of.
The difference between ring cameras and paparazzi using a canon camera is that the photos recorded to film/local storage can't be automatically compiled with the footage captured from everyone else's canon camera to create databases of people and track their movements, activities, attributes, etc.
It really depends on where the data goes and who can access it. I'd even go so far as to say that keeping that data on the cloud is fine as long as the data is encrypted, amazon doesn't access it beyond storage and deliver to the customer (meaning that they can no longer mine it for personal data) and amazon cannot give access to anyone else (including police who should have to request footage directly from the camera owners).
We seem to have strayed into your opinions on how things ought to be, while my comment was meant to acknowledge the reality (unfortunate though it may be) that we live in.
The law doesn't make this distinction you're making. If you replace Canon with "iPhone/Android with cloud photo backup enabled" then your issue with Ring cameras applies to all smartphones. Maybe you'd prefer that not be the case, but it is.
In any case, I'd prefer we not get hung up on this lossy analogy since neither activity is restricted by current law which is my actual point.
Our current reality is that the photos taken with my Canon aren't being handed over to police without my knowing it, and aren't used to track people. The same can't be said about footage from people's ring cameras.
It's not clear to me that photos taken with an iphone with cloud backup enabled are compiled into national databases that track people's movements like ring and flock cameras are today either.
Currently flock and ring cameras are tracking people at a scale that canon cameras just aren't. Therefore it is fair to blame Amazon and flock but not cannon for the manner in which amazon and flock uses the footage their cameras collect to violate our rights.
Only the government is bound by the fourth amendment, as long as the people setting up the cameras and running the network are private citizens it's fair game (correct me if I'm wrong).
Imagining a universe where companies are also bound by that is an interesting thought exercise. Many products (cloud photo backup, foursquare style "check-ins," location sharing with friends, etc) would be simply impossible because the aggregated data amounts to comprehensive surveillance.
I'm a private citizen. On my house we have an ALPR Axis camera pointing down the street (in addition to Axis cameras around the whole perimiter.) And when the police ask, we almost always provide them with data. I feel perfectly justified doing this, and we've helped solved several crimes.
It is an inevitability, because a critical mass of municipalities are going to roll these out such that there isn't a practical route to take through any major metro without being recorded.
If I was American I would certainly be using this tool as a consideration when moving / buying a new house.
I’ve watched a lot of the coverage by Benn Jordan on Flock cameras and their inherent vulnerabilities, and it’s deeply concerning.
The applications of these technologies far outpaces appropriate checks and balances, and the increasing fusion between law enforcement, intelligence and private industry is largely ignored by the larger population.
Thanks for developing this, it’s important to visualise the virus-like spread of these technologies and see where it is concentrated.
I had a thought a while back about companies like Tesla, with cameras on the road and driving models that could classify bad drivers, being in a position to at least a) avoid those drivers if they are encountered on the road and b) record/report them to the police.
Then I had an intrusive thought of a small squad of cybertruck 'enforcers' running around autonomously, tracking these drivers down via the live network of incoming video and doling out punishment to the chief offenders.
My dream is huge drones that get dispatched and harpoon dangerous drivers cars and fly away with them. "Oh a lifted truck with a modified exhaust just swerved onto the shoulder to pass me at 85mph, and there it goes"
How come any area that has enough homes in the data set and ALPR have Veterinarians as the most surveilled, then Hospitals, then Libraries, usually over everything else, including food and church?
The strange implication is that they're watching the vet office traffic to find people who are getting treated by vets instead of doctors?
also my parish reports 0.0% across the board, and all the parishes near me. you have to get on the coast to get above 25%.
Vets/hospitals are far less common (and the former probably suffers from less tags as hospitals are more important) so the distance one must travel increases, so higher likelihood of crossing one. Especially compared to how common everything else is.
If you check deflock.me, would you say that 0.0% aligns with what you expect?
There's 6 cameras in the metro area per deflock.me, 3 at each Lowe's, and that's it. It's very easy to not get in range of those, at least on "my side" of the metro. How do we know that is all the ALPR in an area? Or rather, what's the confidence? I'd assume Home Depot would also have them, for example.
note: maybe i need to restart firefox, but deflock.me is the slowest "map" based site i've seen since keyhole in the late 90s
The best thing you can do is keep an eye out and tag them manually. The second best is a FOIA to your county government - there's some good examples on deflock.me and templates on muckrack. But private ones are not going to be FOIA-able.
I didn't mean to allude to this, but i heard that a lot of vet clinics are being bought by PE firms. The same way that a certain internet company is buying all the legal offices (and web medicine sites)
What about Ring cams? Is there a deRing site?
Also, the flock cams are beyond ALPR, I saw a video before about some investigation and one of the things was how flock was used to recognize a car without a plate. Flock is an automated car recognition system, the plate is just part of it. Makes me wonder if having extremely unique or “cloaking” car design will fool it.
So you have a set of positions for some services (coffee shops, groceries), a set of positions for cameras and a set of positions for homes.
But how you model people actually driving each day and their paths? Because the accuracy of your conclusions seems to heavily depend on the accuracy of modelling.
I don't model daily paths as in 'coffee before work, then groceries on the way home'; I do straight shots from residences to each of these amenities. I don't know of a better way to do it than this; any more complicated model that tries to model 'daily routines' risks losing simplicity, as well as straying too far from actual driving behavior, and my main goal is extremely simple statistics.
When it's against the law to not be on an ALPR camera, only outlaws won't be on ALPR cameras! Well, in SF, this is generally true because there is no enforcement of a missing license plate.
Number plates are just one of the privacy tracking technologies. Any modern connected car infotainment system will report and have that data sold or anything that has Bluetooth can be tracked.
At least as per Carpenter v. United States, that data requires a warrant, not just any cop/LEO in the country typing in a license plate with reasoning as 'investigation'. That's a much better standard.
I've thought about this a lot as I see more and more reckless driving in the areas I live in. Surveillance is generally a net negative, but it's also bad when you see people speeding around schools, rolling through stop signs, and running red lights. We seem to have a worst of all situations where traffic is getting increasingly difficult to enforce, driving is getting more dangerous year by year, and we're terrified of government overreach if we add any automation at all to enforcement.
I don't know the solution, but I do know that in the US we've lost 10-15 years of progress when it comes to traffic fatalities.
> Surveillance is generally a net negative, but it's also bad when you see people speeding around schools, rolling through stop signs, and running red lights.
The fact that these cameras are already pervasive and the problem of bad drivers hasn't been solved anywhere doesn't give me a lot of hope that these cameras are the solution to that particular problem.
It seems like police can do a lot to increase enforcement without the need of these devices. We have evidence that they've been doing less traffic enforcement so maybe start there. Increasing our standards for driving tests (some of which were eliminated entirely over the first few years of the pandemic) would probably help. Automatically shutting off/disabling or limiting the use of cell phones (all of which come with sensors that can detect when you are going at speeds you'd expect while in cars) might help. Bringing physical buttons and dials back to cars instead of burying common functions in touchscreen menus might help.
There's a whole lot of places to look for solutions to safer roads before we have to resort to tracking everyone's movements at all times.
In the US, a traffic ticket is an indictment of a crime (says it on the ticket. I wish I didn't know that fact).
That means that you have a right to trial/appeal, and the accuser (the cop) needs to show up, if you request a trial.
Traffic cameras can't accuse you of a crime, so they are considered civil infractions (no points, but also means they are a bitch to appeal). They can issue realtime civil citations, though.
ALPRS can't do either. They are forensic tools; not enforcement tools.
I believe in the UK, a camera can convict you of a crime, so they can issue severe tickets. They wouldn't really be able to do that, in the US.
In my county (Suffolk, NY), they just stopped all the redlight cameras. I doubt they would do so for ALPRs.
Also, I think some ALPRs are private. There's a shopping center, not too far from here, that's in a relatively high-crime neighborhood. They have cameras and ALPRs, all over the parking lots.
> Automatically shutting off/disabling or limiting the use of cell phones (all of which come with sensors that can detect when you are going at speeds you'd expect while in cars) might help.
I can’t think of a way to implement this that wouldn’t ban passengers from using their phone while riding in a vehicle. Which could be even a bus or limousine.
I don't disagree, but I can totally imagine a society where this inability is perfectly acceptable because it severely reduces the #1 killer of people from 5-55yo. I don't think we live in that society, if Apple and Google flipped a switch tomorrow to do that people would freak out, but I could imagine a rational, fictional society that had different shared values.
Not entirely. The phones can defect if there are other phones nearby, so a single phone in a car on a highway going 75mph could be assumed to be a driver, but that is still just an assumption.
And the driver will just carry two phones, and be even more distracted than before. Cool
A lot of people would be fine with that. Drivers are impaired while on the phone, even hands-free. Not to mention texting while driving!
I kind of picture the cellular telcos doing this. Maybe buses and trains come with wifi hotspots allowed to connect. Otherwise auto passengers could use their devices offline, maybe read an ebook or something. Not the end of the world.
Is a driver impaired if they are using their phone to stream music (or screen-off Youtube listening) to their car while driving?
Lots of cars now come with a WiFi hot spot as part of their offerings. There's no way to prevent the driver from also connecting to it and circumventing whatever ill conceived notion this is
Even connected to wifi a cell phone canstill use the wireless network. Even airplane more won't actually stop your phone from connecting anymore. GPS data can also be transmitted in the background over wifi back to apple/google and/or the device manufacturer.
If they really wanted to push this they could do it directly in the baseband chipset and bypass the OS entirely when deciding to lock down the device to some kind of "travel mode" with limited functionality (such as no texting or no browser)
Not that I'm advocating for that sort of thing, but it's good to keep in mind that we don't really own the cellular devices we pay for and that even in the rare case we have root we can't stop them from doing what they want to our devices as long as they control the closed hardware.
I mentioned buses and trains, and was thinking that only those mobile wifi hotspots would be permitted, whitelisted for 5g service. Hotspots in (human driven) cars would not. That might encourage some people to take the bus?
I agree with the other poster about this being more workable in a fictional society with different shared values.
The standards for evidince, processes for enforcement and court side of things are not set up for cheap enforcement of "that clearly ain't right" behavior. They're set up for revenue enforcement of easy to prove but not necessarily bad in abstract offenses.
Police can't substantially increase enforcement overall because that would just cause bad political optics, say nothing of stops that needlessly escalate to being newsworthy in a bad way. They'd necessarily issue a hundred petty bullshit tickets for every deserved ticket for legitimately bad behavior. It just wouldn't work. It would be like trying to plow a field with the ripper on the back of a bulldozer. It kinda looks similar but it's wrong for the job.
And all of this is based on the assumption that we're trying to enforce things that the broad public agrees need strict enforcement, not whatever the original comment wants.
They have done automated enforcement a few times and it always sucks because the systems don't use discretion.
Someone going 40 in a 30 and swerving around other cars gets treated the same as someone going 40 when the road is empty. Someone slowing to 1-2mph before safely rolling through a stop sign get the same ding as someone blowing through it at 30mph.
If AIs can somehow learn how to take all this footage and enforce the spirit of the law (citing dangerous driving) instead of the letter I will fully support it.
The answer is to take the human out of the equation, and have the computer drive. Comma.ai works well enough. Tesla is mostly there. Waymo works.
Which is also bad for privacy
This! Things keep getting worse and worse, and we keep getting more surveillance. It's clearly not the answer!
so what do they seek to curb? freedom of movement?
It's an opportunistic solution to a budget problem.
Local governments have extra money from property price bubble increasing tax revenues.
At the same time there are great open source image analysis models for companies to put on a pole using cheap android hw and a solar panel and make bold claims about solving all crime, then they can sell that data again to 3rd parties like insurance. Also can start buying access to more video feeds from ring cameras etc and resell that. They'll hire someone to make a ux to integrate it into PDs later, for a premium of course.
“Crime.”
https://www.forbes.com/sites/thomasbrewster/2025/09/03/ai-st...
We have very few alternatives to driving in the US so we have very lax driver training and testing.
Across the US we have roads and infrastructure that encourage speed right next to decaying pedestrian infrastructure. It's very difficult to get state DOTs to roll back or do traffic calming. They often prohibit the use of bollards or barriers near these roadways.
In a lot, not all, physical changes to the environment could drastically reduce traffic fatalities without surveillance.
Yep, the US has built itself a car-centric society over the last century, and there's essentially no way to change that now without burning everything down and starting over from scratch. They can't make driver training too strict because too many people would fail, and then have no way to get to work, and then who knows what those people would do. They can't turn cities pedestrian-friendly without tearing down everything there now and just the legal hurdles there are immense. I just don't see a way for this to change meaningfully in a lifetime, outside of a few little pockets here and there (like some towns or cities creating extremely limited pedestrian-only zones, which would then need a lot of parking to be useful for people, basically like an outdoor shopping mall).
100% agree
my local middle school has their school zone on:
1. four lane highway
2. dedicated turning lanes
3. major thru-way between shops, apartments, and the rest of the city
4. great visibility
this is a recipe for 50mph. the speed limit is 25mph. If you do the speed limit, you WILL be tailgated. If you do ~35, you're risking a ticket. There will still be people doing 45-50 and weave through the lanes.
also in my town, the main thru-way is a route dating back to the 30s. There are red lights at major intersections and they WILL turn red even if no one is there. They're designed to slow people down. HOWEVER if you speed and run a yellow light, you'll hit ALL the lights green! It shaves significant time off your trip, is easier on your car, is more enjoyable, and requires less attention. It's a system designed to make people speed and run reds.
Where I used to live, I could get from one side of the city to the other in a maximum of 30 minutes. the lights were designed to keep traffic flowing at 30-35mph. It ENCOURAGED you to go no faster, or you'll have to slow down and come to a stop. This also kept traffic flowing so you felt like you HAD to focus on driving. They also did things to encourage bicycles and make things safer for pedestrians.
I agree. It's frustrating that we have ended up in a reality where vehicle movement is heavily tracked, but we're not using that technology to do the most obvious and productive thing.
My city spent a few million dollars installing Flock cameras to all its municipal parking garages in a matter of months, but has been hemming and hawing over adding a few speed cameras for years, despite petitioning the state for an allowance do so back in 2023.
Traffic enforcement cameras don't even have to become the networked surveillance system that Flock offers. Most are still cameras triggered by radar rather than perpetually recording all drivers.
I live in NYC. People used to be afraid of double parking. Like you I regularly see the same bat-shit driving and no one seems to care to say or do anything. It's bonkers.
NYC should have been the model to follow. Instead of flock cameras, cities should have bounty systems: record a video of a speed violation with a plate, and get 10% of the ticket revenue. Enforcement would explode.
We could of had a system where we used the technology we already had in our hands to democratize speed enforcement, instead of corporatizing it
NYC already tried Snitching as a Service during COVID, and it went terribly. I grew up with a neighbor who would constantly record people and call the cops over every little perceived infraction. Everyone in the neighborhood hated her, including the cops. I do not want to live in a society that encourages those people.
Seen this recently, the Citizens filming other people and their cars in public, presumably for their crimestopper social media app.
I definitely agree it can go too far. Maybe only allow bounties for active, dangerous crimes like speeding, drunk driving, and racing?
How do you imagine this working in a world where video editing is so easy?
No Stasi. No Salem witch trials. Thank you. Do not need to retread that ground.
Pitting people against each other in the service of the state is likely to cause problems at scale and in the long run.
Frankly, I think it's a miracle that nobody has been beaten into a coma or killed over NYC's bounty program yet.
Interesting. 'But your honor, it is AI enhanced. I wasn't speeding.'
'Weird, I got this footage here from another angle and it shows you did. We also got the data from your car. You were speeding son.'
These cameras are currently not used at all for traffic/speed enforcement. The best they would do is track more serious crimes like hit-and-runs by photographing cars in the area.
Ah, that's helpful and something for me to learn more about. Thanks for the info.
On my way home, I noticed at a stoplight across the street from Apple Park that the driver in the lane next to me had his phone mounted up high in landscape mode and was watching the Simpsons. Just absolutely unhinged behavior lately.
Here in Washington DC we've set up many different types of traffic cameras. One side effect is a reduction in the number of negative interactions the average person has with the police. There are communities where the cops treat people like piggy banks and spend tons of time aggressively enforcing minor traffic violations. That has negative effects on trust, not to mention the number of fatal interactions that have stemmed from traffic stops. Other jobs that can be done by the police in other places are handled by Parking Enforcement, the Department of Buildings, etc.
The recent presence of federal agents and soldiers has reversed some of the hard-fought gains in trust, but my broader point still stands: more automated enforcement of traffic laws has positive effects in how people interact with the police. This effect needs to be balanced against the harms of increased surveillance.
[1][https://www.nbcwashington.com/news/local/transportation/poli...]
Your examples don't seem to represent the leading causes of fatalities from traffic accidents in the US, which remain distracted driving (phones and crappy touchscreen controls, etc) and drunk driving. People rolling through a stop sign / performing a 'California' stop are nowhere near the massive security concern to authorize the ideal surveillance state. To deal with the bigger issues you would have to film peoples actions in their cars and run it all through ai to accuse them of driving while drowsy. That would be an incredible and lazy failure. Also, fatalities per mile had a massive surge in 2020-2021, but they have been steadily dropping since then (2024 in wikipedia estimated a 1.27 per 1m miles, about the same as 2008). If the trend continues, in a year or two we will be back to the early 2010s lows without draconian measures.
The solution, as always, is better infrastructure and support at multiple levels, not beating everybody with a stick.
I get the vibe from his comment that the visible deviance from rules of the state is what he's really concerned with and that the death and destruction is just the reason everyone else is suppose to care.
A camera doesn't stop those acts though, it may only discourage those who know about it at a huge cost of privacy and rights.
How about we build better infrastructure and regulate vehicles since those do actually stop this behavior. Most of those red lights and stoplights in the US should be roundabouts. Narrower lanes and other traffic calming measures should be much more pervasive. Vehicle size, specifically bumper height is out of control.
Compare US traffic and pedestrian deaths to the rest of the world, or at least a lot of EU countries. Its embarrassing.
https://www.cdc.gov/mmwr/volumes/74/wr/mm7408a2.htm#F1_down
Thanks for this CDC link! So the pedestrian death rate and the overall road traffic death rate is about three times higher in the US compared to other high-income countries... Mind-boggling.
> Vehicle size, specifically bumper height is out of control.
Many manufacturers are now selling pre-lifted trucks. Here in Texas around a third of the vehicles on the road are pickup trucks, and about half of those have been lifted beyond standard height either from the factory or aftermarket, another third of vehicles are SUVs, most of which are significantly larger than necessary to be fit for purpose for the driver and occupants.
This situation was /caused/ by government regulation and it can be fixed by government regulation. It's absolutely absurd the gargantuan vehicles most people drive in the US, and the fact that we let people turn their vehicles into monster trucks and then operate them on public roads with impunity. I don't care how small someone's dick is, they don't get the right to drive a truck down the highway that can literally drive /over/ a modern standard sedan/hatchback. The continuing absurdity has turned into an iterated prisoner's dilemma which has resulted in more and more people buying SUVs and crossovers who by every measure do NOT need them. Absolutely is out of control, and it negatively impacts everyone, including the drivers of these vehicles.
Eventually we’ll have autonomous vehicles which will mitigate many of these issues, but will the surveillance infrastructure then be reduced? Probably not.
War is peace.
Surveillance is safety.
> driving is getting more dangerous year by year
My observation has been that the danger increase is a combination of three things:
1. Lack of situational awareness/awareness of surroundings. This manifests in various ways like improper merging, improper turns into traffic, turning across multiple lanes, and left-lane hogs.
2. Frustration becomes aggression. This also manifests in multiple ways, but primarily is seen through tail-gating (in response to left-lane hogs) and swimming through traffic (in response to left-lane hogs), as well as road-rage (mostly in response to the other misbehavior).
3. Constant distraction. I have seen SO MANY drivers literally watching videos on a tablet, playing games, or otherwise driving at high rates of speed (70mph+) while fully engaged in something other than driving. It's at epidemic proportions.
The issue is, by and large, not people doing rolling stops (which by the way have never been proven to cause an increase in accidents when performed properly) or speeding (generally, some exceptions like school zones do matter). Running red lights is definitely a problem that is more common place, and is likely a part of all 3 of these these.
I feel like to a large degree all of this is a symptom of a wider societial issue where everyone acts in selfish and self-centered ways, completely ignoring their impacts on others, and moving from moments to moments where they can engage with their phone/social-media, to the exclusion of all else. I don't think phones/social-media /caused/ the problem, I think it exacerbates it though. Every aspect of our society is worsening because the revealed behavior of our population is one of lack of care or outright disdain for everyone else around them and an absolute obsession with serving their own interests above all else. That this manifests in driving is unsurprising.
IMO the issue is that there's little to no enforcement outside of speeding. I see people get pulled over for speeding in a straight line even if it's relatively safe to do so, but never for talking on their phone, messing with the touch screen in their vehicle, left turns from the right lane and vice versa, failure to keep lane, failure to yield, failure to signal, cutting people off, driving a vehicle unfit for the road, etc.
It usually takes about 5 minutes of driving to observe someone doing something that I would pull them over for. I don't think cops need all this automated surveillance, they just need to drive around and be proactive.
> Surveillance is generally a net negative, but it's also bad when you see people speeding around schools, rolling through stop signs, and running red lights.
Speeding and running red lights can be combated without affecting the privacy of innocents on the road. A debate can and should be had about the placement of radar traps though, many are straight off highway robberies trapping people who don't notice a speed-limit sign that is visually hard to notice.
I would actually be for civil asset forfeiture in these cases. The laws of physics don't care about due process; sue the car and save the video. Let the owner pay the fines and if they can prove some other driver did it, let them file civil suit against them to recoup the fines. Driving on public property is not a right.
We cracked down on driving under the influence with changes from DWI to DUI. In the 10-15 years you mention, the prevalence of distracted driving from mobile devices has gotten out of hand. There's no field sobriety test that can prove one was distracted by a device. That makes this much more difficult to crack down on.
Sure, though it is very easy to visually identify drivers looking at their phones. (Unless they have illegal tint.)
Only if you're driving next to the car. Speeding obviously can be detected from a distance. Driving under the influence also can be seen from a distance.
I could see having a cop and a camera stationed in a place working as a spotter that then radios to other cops up ahead to pull over in an old school speed trap set up. However, that would take a lot of man power and coordination that I doubt most departments would care to do. We've seen plenty of YT videos of people trying to make a point of catching distracted drivers. The cops just need to give a shit, and want to do something about it.
It doesn’t take a targeted trap or driving next to. It’s like 1 in 6 drivers at any time; it’s super easy to spot if you’re looking. One cop could pick off drivers no problem without any coordination.
Part of the public pushback is that people almost always drive the “feels like” speed and not the posted speedlimit. We build 6 lane roads and then wonder why people go 50mph when it’s 35 posted, it’s because it’s 6 lanes and 35 feels slow. Cities profit from this in the form of speed cameras, which is why they’ve been outlawed in a lot of places.
The driver blithely keeping with the flow of traffic is not the one I am worried about. It is the one who is aggressively trying to cut through the flow of traffic while putting everyone and themselves in danger that I worry about.
I love when cities time their lights so that aggressive drivers just get hit with waiting at a red light while driving the speed limit means hitting greens for long stretches.
Yes, and about twice I've seen this done really right, wherein they post signs of the synchronization speed (e.g., "Lights Timed for 35mph"). I just get in sync with one light and adjust speed and it feels almost magic to go a few miles hitting every green light (it kind of is the macic of math).
It'd be cool if more roads were implemented that way.
I have the same in my area, but instead the lights are synchronized to slow traffic as much as possible. They literally coordinate to make you stop at as many lights as possible and grid lock at rush hour so the highway doesn't get flooded.
I walk my kids to school, but when I do drive the 1.1 miles there are no less than 12 stoplights/stop signs.
There's a part of downtown here that is known for being set up like this. However, it's been a bust while there's been a lot of construction blocking lanes so nothing moves at speed.
Sure, some people will get hit with the reds. Some people will learn the timing and learn that you can run just one red and then hit all greens, or that it will save you 5+min if you hammer down and pull a "clearly not ok" pass to get around some idiot who's going too slow to make the green timing.
So yeah, you're reducing speeding, number go up, pat yourself on the back. But you're also increasing the incidence of something rarer than speeding, but way worse.
This is the same problem that 4-way stops at roads that don't deserve them create, you're basically teaching people that the signage is bullshit.
Upping the speed limit reduces the incidence of the latter because just about everyone has a "fuck this I'm weaving" threshold and the number of people who hit it goes down when you reduce the incidence of rolling clusters of traffic caused by the handful of people who religiously follow the speed limit even when not appropriate.
If the speeds aren’t appropriate for the built environment, then the limits should be changed or the environment should be changed. Enforcement of the law should be consistent regardless of the quality of the law.
The speeds are appropriate for the roads generally. I mean, at the lowest level speed limits are a matter of social consensus so the broad public is tautologically correct.
The problem is that knocking the magic number on the sign down by 5-15 and then simply not enforcing it too seriously results in less screeching Karens harassing the politicians who then harass the bureaucracy than taking a hard line about "well akshually this is the engineered speed for the road".
The speed cameras in San Francisco have to result in lowered speeding over the 18 month period they're active. If they don't, they will be pulled. Seems pretty well-designed. Perhaps the fines are weak but it's good that they're there.
Just because a road "feels" like it can handle more speed does not mean that it is. The wider streets are built to handle the volume of cars, not necessarily meant to become a speed way. There are several 6 lane roads in my area while being wide and well built still have many intersections only controlled by stop signs for the smaller streets with multiple intersections controlled by stop lights at the larger cross streets.
People unable to recognize this and only driving by the feels are the problem. Hand wavy comments like yours suggesting using the feels as being okay do not help the situation
Basing your speed on what the road looks like may not be “okay” or “legal”, but it’s what people do. It’s just not useful to claim that individual people are the problem when this is something that is overwhelmingly true across the entire population—-a broader solution than individual responsibility is the only thing that will actually work.
Speeds are fundamentally a tradeoff between risk and reward. In nominally democratic societies we place these thresholds based on some approximation of social consensus. The general public literally cannot be wrong because their rough consensus, the fat part of the bell curve if you will, is what determines what the right speed is.
Your comment is literally the principal Skinner "no it's everyone else who's wrong" meme.
Um, no it’s not. If you think that roads are built with no consideration for anything other than if the infrastructure will support traffic at a certain speed then you’re just not thinking about things. Speed limits are set with many factors in that decision. Things like noise and safety are major parts of that. Someone else has already suggested a lame reason as a noise complaint made by a single person, but cars moving faster make more noise. Cars moving faster limits the time a car at a stop sign can safely navigate causing traffic at cross streets.
Regardless of what you think, speed limits are not set in place just to ruin your day because you can’t leave on time and constantly need to “make up time”. They are not arbitrary decisions just because you haven’t considered all of the factors involved.
Now you're relying on Jimmy "Buck" Rawgers born and raised in Billton county working at the DOT setting the speed limits and the traffic light cadence.
How many roads are 35 when they should be 50 simply because some local yokel asshole made a stink at city counsel 10 years ago and now it's impossible to change?
I agree. And Flock doesn’t help one bit with reducing reckless driving.
The solution is to have you abused by the system so that you realize that petty deviance is not something the dragnet should be brought to bear on if it must exist at all.
Your opinions are directly counterproductive on these petty issues. You ask for the state to use the jackboot. The jackboot just makes people hate the state and think in terms of "will I get caught". If not for you people trying to force compliance on this, that and the next thing the state would be in higher standing in people's minds and voluntary "when it matters" or "because it's the right thing to do" compliance would be higher. Sure, we could add more jackboot, but that costs money and no democratic-ish system is gonna allocate a bunch of money to do stuff everyone hates.
> driving is getting more dangerous year by year
Not over the long term, no. There may have been a recent uptick in the post-pandemic US but it's mostly just noise. Fatalities per mile driven have been going down markedly in recent decades. Driving was twice as dangerous in the 80's as it is now.
You are incorrect. Fatalities in the US leveled out in the early 2010s and have been climbing since then. In all other developed nations they continued trending downwards.
This is not a statistical anomaly that can be handwaved by pointing out that things were worse 40 years ago. Roads in the US are uniquely lethal and getting moreso.
> You are incorrect
Sigh. I hate that phrasing. But OK, fine: you are misreading me, misanalysing the data, or just plain spinning to mislead readers.
Fatalities per capita and per mile driven go steadily downward until covid, and maybe there's a bump after that: https://en.wikipedia.org/wiki/Motor_vehicle_fatality_rate_in... If you have numbers (you don't cite any) showing otherwise, they are being polluted by demographic trends (the US having higher population growth doesn't say anything about driver behavior).
> Roads in the US are uniquely lethal and getting moreso.
So spinning it is. Would you rather drive in Germany in 2002 or the US in 2025? Seems like "uniquely lethal" doesn't really constitute a good faith representation of the truth.
According to the link that you posted, the roads in Germany in 2002 were quite a bit safer than the roads are in the USA in 2025. And they don’t have speed limits. Absolute no brainer to me.
Anyway, not to pile on but you are absolutely incorrect. Forgive the phrasing.
German roads absolutely do have speed limits. Only certain rural sections of the Autobahn don't, but that's not representative of the country as a whole, or even the Autobahn as a whole.
In 2002, significant portions of the Autobahn were unlimited. That's changed in the past 25 years, of course.
Did you open the wikipedia article you linked? The first image contradicts you, see the caption:
> Per capita road accident deaths in the US reversed their decline in the early 2010s.
Amusing that you accuse me of bad faith framing and then pose a nonsense question like this:
> Would you rather drive in Germany in 2002 or the US in 2025?
I cannot time travel and neither can you. The comparison that matters is US in 2025 vs other developed nations in 2025, and with that framing the US is uniquely lethal.
Of course, a good faith reader of my comment would understand this, but we already know that's not you since you did the research and have decided to be wrong anyway.
That's thanks entirely to government-required safety features, not skilled drivers.
There is picture of a cyclist. He is holding a stick.
The cyclist puts the stick between his spokes.
The cyclist lies on the road, cursing the government, for causing this.
One thing this surfaces nicely is how scale changes the privacy model. Individually these look like “mere observation,” but once you can reconstruct routine movement patterns across counties, the data starts behaving more like long-term tracking than casual surveillance.
Retention is one good line between enforcement and tracking.
If we wanted to solve this problem while retaining the benefits, what we'd do is have stiff penalties for warrantless retention and required yearly independent audits of systems.
I understand why these statistics may be interesting, but all I really want to see is a map of the locations of the ALPR cameras. I would add an easy link to that data on this site.
https://ringmast4r.github.io/FLOCK/
https://deflock.me/
DeFlock is the first link in the "Related ALPR & Flock Projects" section, it contains a map.
This is great work. Once ALPR coverage is dense enough that you can’t go anywhere without generating a permanent record, the “mere observation” argument falls apart. Mapping this openly is one of the few ways communities can actually understand what they’ve signed up for.
The county lists are wrong, at least they are for my state of New Jersey. We have 21 counties, not 27. Is it picking up the bordering counties that might have overlapping contracts or something?
It pulls counties from OSM administrative boundaries of level 6, which according to the OSM wiki, is "State counties and county equivalents, Territorial municipalities" (https://wiki.openstreetmap.org/wiki/Tag%3aboundary=administr...). It does lead to some weird oddities, like counties with under 10 homes... I'd rather not manually correct it, since I want to rely on pure OSM data. (Unless you mean there is an actual bug in the processing and there's counties listed that aren't in the right states...)
I'll add a link to the OSM relation for the county to each county page, so you can see the source data on OSM to verify/edit.
There are also not 53 states.
One of those counties is Rockland, which is in NY. I wonder if it's counting bordering counties for states, since the assumption may be that the closest essential services for some Rockland residents may be in New Jersey.
Ah! This looks like a bug with cross-border calculations. Ideally it would not show a Rockland County, NJ - it might cross state borders to perform calculations but they should all have a 'home state' that matches to the right county. Thanks for the example case, I'll work on a fix
I know at least in Rockland's case, their power utility extends slightly into NJ. Can see it on statewide power outage trackers like this one: https://projects.nj.com/data/outagetracker/
California includes a number of counties which are actually in Baja California and Baja California del Sur.
yeah we have 8 counties in CT, not 14. The names are also wrong.
As far as I know there's no easy fix to this. It's the counties as reported by OpenStreetMaps' administrative boundaries.
I don't see how this provides actionable info for the individual. Unfortunately, this is just going to be a dashboard for pro-surveillance elements to see "how are we doing in our neck of the woods?", or a sales tool for Flock to find untapped markets.
My goal is to provide actionable statistics for any 'deflock' movements in a certain county, by being able to point to specific statistics on surveillance. If even one motivated person uses my data to petition, I'll be happy; it doesn't have to be for the average person. There's tons of these movements, too. Deflock Olympia just succeeded: https://www.yelmonline.com/stories/commentary-olympia-joins-...
Also, another answer to this is that there is no overarching goal; I just wanted to build a large scale data analysis pipeline for fun :) I am no stranger to side projects to distract me from finals unfortunately.
Fair enough. I am sorry if my original post was too harsh or overly critical; my original thinking was simply that those would organize against surveillance are probably already aware of it, but perhaps this would be useful in the cases you mention, if the metrics could be used as "ammo", as you suggest.
Not too harsh at all! I just wanted to fill the gap between "man, those cameras seem to be everywhere" and "XX% of people are surveilled on the average commute". It's true that this is not a particularly ambitious project, just a small niche I wanted to fill.
I think we have a surplus of "awareness" tools/websites that are great at what they do, but not much "rubber meets the road" tools to guide the user in actually taking action based on the information presented. I, for one, feel a huge sense of fatigue at the amount of awareness I have of problems I don't have the tools/strategies/knowledge/time to solve.
(This is not a negative comment about this post btw, more just commentary on how the fire hose of "look at how bad this is in excruciating detail" can be overwhelming.)
100% coverage seems like an inevitability in a country where filming in public is a constitutionally protected right and networked ALPR capability is possible (if not regularly offered yet) in commodity doorbell cameras.
> 100% coverage seems like an inevitability in a country where filming in public is a constitutionally protected right a
It really doesn't have to be though. The rights of individuals to record in public doesn't have to translate to the right of corporations (flock, amazon, etc.) to do it without restriction. Time, place, and manner restrictions on our rights already exist, it just needs to be found that this manner is unacceptable as an imposition on our freedom which should be protected under the fourth amendment.
If a home or business owner sets up Ring cameras, is it fair to say Amazon is recording in public? That feels like blaming Canon for the behavior of a paparazzi, but perhaps there are reasons those aren't equivalent I'm not aware of.
> hat feels like blaming Canon for the behavior of a paparazzi, but perhaps there are reasons those aren't equivalent I'm not aware of.
The difference between ring cameras and paparazzi using a canon camera is that the photos recorded to film/local storage can't be automatically compiled with the footage captured from everyone else's canon camera to create databases of people and track their movements, activities, attributes, etc.
It really depends on where the data goes and who can access it. I'd even go so far as to say that keeping that data on the cloud is fine as long as the data is encrypted, amazon doesn't access it beyond storage and deliver to the customer (meaning that they can no longer mine it for personal data) and amazon cannot give access to anyone else (including police who should have to request footage directly from the camera owners).
We seem to have strayed into your opinions on how things ought to be, while my comment was meant to acknowledge the reality (unfortunate though it may be) that we live in.
The law doesn't make this distinction you're making. If you replace Canon with "iPhone/Android with cloud photo backup enabled" then your issue with Ring cameras applies to all smartphones. Maybe you'd prefer that not be the case, but it is.
In any case, I'd prefer we not get hung up on this lossy analogy since neither activity is restricted by current law which is my actual point.
Our current reality is that the photos taken with my Canon aren't being handed over to police without my knowing it, and aren't used to track people. The same can't be said about footage from people's ring cameras.
It's not clear to me that photos taken with an iphone with cloud backup enabled are compiled into national databases that track people's movements like ring and flock cameras are today either.
Currently flock and ring cameras are tracking people at a scale that canon cameras just aren't. Therefore it is fair to blame Amazon and flock but not cannon for the manner in which amazon and flock uses the footage their cameras collect to violate our rights.
> filming in public is a constitutionally protected right
As with everything, there's much nuance to this "right".
https://consumerrights.wiki/w/Common_Questions,_Arguments,_%...
Only the government is bound by the fourth amendment, as long as the people setting up the cameras and running the network are private citizens it's fair game (correct me if I'm wrong).
Imagining a universe where companies are also bound by that is an interesting thought exercise. Many products (cloud photo backup, foursquare style "check-ins," location sharing with friends, etc) would be simply impossible because the aggregated data amounts to comprehensive surveillance.
I'm a private citizen. On my house we have an ALPR Axis camera pointing down the street (in addition to Axis cameras around the whole perimiter.) And when the police ask, we almost always provide them with data. I feel perfectly justified doing this, and we've helped solved several crimes.
Most people would do the same in your situation, we should expect citizens to want to help victims of crime, especially crimes against the person.
However I think there’s a significant difference between a single household and a centralised network of cameras across dozens of states.
For me the core issue of this is private enterprise holding gigantic amounts of PII, and the forms that is taking.
Sounds like a rough neighborhood, stay safe.
Hi enemy! :)
It is an inevitability, because a critical mass of municipalities are going to roll these out such that there isn't a practical route to take through any major metro without being recorded.
If I was American I would certainly be using this tool as a consideration when moving / buying a new house.
I’ve watched a lot of the coverage by Benn Jordan on Flock cameras and their inherent vulnerabilities, and it’s deeply concerning.
The applications of these technologies far outpaces appropriate checks and balances, and the increasing fusion between law enforcement, intelligence and private industry is largely ignored by the larger population.
Thanks for developing this, it’s important to visualise the virus-like spread of these technologies and see where it is concentrated.
From just a few days ago:
https://news.ycombinator.com/item?id=46170302
(Repeating: in a few months sites like this will be replaceable with a static HTML page that says "yes, you've been tagged by an ALPR".)
I had a thought a while back about companies like Tesla, with cameras on the road and driving models that could classify bad drivers, being in a position to at least a) avoid those drivers if they are encountered on the road and b) record/report them to the police.
Then I had an intrusive thought of a small squad of cybertruck 'enforcers' running around autonomously, tracking these drivers down via the live network of incoming video and doling out punishment to the chief offenders.
My dream is huge drones that get dispatched and harpoon dangerous drivers cars and fly away with them. "Oh a lifted truck with a modified exhaust just swerved onto the shoulder to pass me at 85mph, and there it goes"
Yeah, and then you get harpooned for reading an article about Kamela Harris five years ago.
As an aside, there are currently companies that do this but for delinquent auto loans to expedite the repo process.
Or the Tesla could just record where it is at all times.
I’m thinking more the Tesla could identify bad drivers around it and fire off a notification.
How come any area that has enough homes in the data set and ALPR have Veterinarians as the most surveilled, then Hospitals, then Libraries, usually over everything else, including food and church?
The strange implication is that they're watching the vet office traffic to find people who are getting treated by vets instead of doctors?
also my parish reports 0.0% across the board, and all the parishes near me. you have to get on the coast to get above 25%.
Vets/hospitals are far less common (and the former probably suffers from less tags as hospitals are more important) so the distance one must travel increases, so higher likelihood of crossing one. Especially compared to how common everything else is.
If you check deflock.me, would you say that 0.0% aligns with what you expect?
i suppose that makes sense.
There's 6 cameras in the metro area per deflock.me, 3 at each Lowe's, and that's it. It's very easy to not get in range of those, at least on "my side" of the metro. How do we know that is all the ALPR in an area? Or rather, what's the confidence? I'd assume Home Depot would also have them, for example.
note: maybe i need to restart firefox, but deflock.me is the slowest "map" based site i've seen since keyhole in the late 90s
The best thing you can do is keep an eye out and tag them manually. The second best is a FOIA to your county government - there's some good examples on deflock.me and templates on muckrack. But private ones are not going to be FOIA-able.
The quality of ALPR tagging does probably lag behind true counts - for example, Williamsburg, VA has 28 tagged on OSM, but 32 are listed in the transparency log (https://transparency.flocksafety.com/williamsburg-va-pd). Unfortunately not much can be done except spotting them out and about (or wardriving with a BLE beacon scanner: https://www.ryanohoro.com/post/spotting-flock-safety-s-falco...)
The ones nearest me are at Home Depot, FWIW.
Perhaps there's some large chain of vets run by a PE firm where the PE firm did a blanket deal.
I didn't mean to allude to this, but i heard that a lot of vet clinics are being bought by PE firms. The same way that a certain internet company is buying all the legal offices (and web medicine sites)
What about Ring cams? Is there a deRing site? Also, the flock cams are beyond ALPR, I saw a video before about some investigation and one of the things was how flock was used to recognize a car without a plate. Flock is an automated car recognition system, the plate is just part of it. Makes me wonder if having extremely unique or “cloaking” car design will fool it.
I swear it's about 75% of houses I come across that have ring (or similar) cameras
So you have a set of positions for some services (coffee shops, groceries), a set of positions for cameras and a set of positions for homes.
But how you model people actually driving each day and their paths? Because the accuracy of your conclusions seems to heavily depend on the accuracy of modelling.
I don't model daily paths as in 'coffee before work, then groceries on the way home'; I do straight shots from residences to each of these amenities. I don't know of a better way to do it than this; any more complicated model that tries to model 'daily routines' risks losing simplicity, as well as straying too far from actual driving behavior, and my main goal is extremely simple statistics.
https://archive.ph/et1uD
Data accuracy can be a problem. It lists 115 counties for Illinois, which is news to me since Illinois has 102 counties.
For example, Kenosha County is in Wisconsin, not Illinois.
When it's against the law to not be on an ALPR camera, only outlaws won't be on ALPR cameras! Well, in SF, this is generally true because there is no enforcement of a missing license plate.
Number plates are just one of the privacy tracking technologies. Any modern connected car infotainment system will report and have that data sold or anything that has Bluetooth can be tracked.
Bluetooth? You're overthinking this. We've been mandated to carry 4-5 transmitters per vehicle broadcasting their UIDs at 315 MHz since the mid 2000s:
https://en.wikipedia.org/wiki/Direct_TPMS
I have an SDR in a facility that is a public parking lot. It picks up TMPS incidentally and I'm definitely able to track individual vehicles.
Including the camera and microphone in our pocket that pings every cell tower every time we move.
At least as per Carpenter v. United States, that data requires a warrant, not just any cop/LEO in the country typing in a license plate with reasoning as 'investigation'. That's a much better standard.
Recent car tracking discussion here https://news.ycombinator.com/item?id=46097624
To an extent this data is within our reach to stop (buy old car, unplug cellular modules, etc). With ALPRs the only option is moving.
I'm surprised an app like Citizen hasn't tried to match Flock with a dashcam that automatically shares recordings.
Can you think of any use cases apart from government surveillance?
Small nitpick: you're writing OpenStreetMaps instead of OpenStreetMap (no trailing s) everywhere.
Thanks, pushed a fix for the home page!
Mass surveillance in the "Land of the Free" [tm]. LOL