3 ms·

Are you familiar with V2X? The concept is actually specifically to help pedestrians and bicyclists. It's essentially just a LTE or WLAN link (which most modern

Are you familiar with V2X? The concept is actually specifically to help pedestrians and bicyclists. It's essentially just a LTE or WLAN link (which most modern cars already have and that costs less than 100usd) that would allow intersection cameras, etc to report useful information back to vehicles. i.e. the city can have infrastructure watch for pedestrians on crossings (or even just know the crossings are active or the light is red) and report it to vehicles approaching. This makes assisted driving tech safer for everyone involved. It also generally just adds a lot of opportunities for QOL improvements. An example is parking. The city infra can tell the vehicle (or your phone) where the closest open parking space is relative to your destination and handle scheduling spaces so that you get the space the first time around. And importantly the cities that have been pioneering this tech have been pushing it while also pushing separated bike lanes, improved transit, more distinct street vs road distinctions, etc. This tech is something that has been in development since the 90s and it's now reaching maturity to the point where modern cars could adopt it today, older semi-modern cars could be trivially upgraded, and it can see widespread use in 10-20 years. Also it's not the first line of defense for pedestrians but rather one of many that can be rolled out now while also pushing other material benefits.

panick21_ 2y ago
There is a gigantic difference between 'this technology has some useful applications' and 'this is fundamental to improving road safety'. And there is even a more gigantic difference between cost and useful how useful it is. Again, we know how to make gigantic gains in road safety. Its not difficult, its proven technology, its cheap, it can literally be done quickly. The only question is will of politicans. That's it. First of all, its already incredibly absurldy expensive how the US currently does traffic. There are far, far to many traffic lights and traffic lights are expensive. Far to many lanes and far to wide, incredibly expensive. That again makes traffic signaling more expensive. Now in addition to traffic lights you need cameras everywhere, often covering 8 lane stroads, and then you need comptuers to process all that data. And do so correctly under difficult conditions. Installations of these system will be hugely expensive. You can already compare your avg traffic intersection between the Netherlands and the US. The US often only has 1 sensor while the Dutch have many more. US intersections still operate based on completely outdated signaling orders for the most part (not to mention completly unsave and unfriendly for anything but cars). So when in the US most towns can't even configure their basic signaling orders in an efficent way, but you want them to do complex video image analysis and messaging? How does that make sense? Again, we know how to make things cheaper and much, much safer. And it doesn't need fancy technology. We don't need people with PhD in data science to implement this. If you want to make an argument that some of this has some uses, sure, whatever. I honestly don't think anything you mentioned is really all that complelling but I guess its possible. > The city infra can tell the vehicle (or your phone) where the closest open parking space is This is great, if your goal as a city is maximum utilisation of parking space, but that shouldn't be the goal in the first place. If cities actually listen to experts on how to actually manage parking correctly, finding a parking spot wouldn't be hard in the first place. > And importantly the cities that have been pioneering this tech have been pushing it while also pushing separated bike lanes, improved transit, more distinct street vs road distinctions, etc. That just means that the lobbies pushing this stuff have successfully done their job despite far more important things. No city in the US has even begun to fully implement modern traffic practices. Playing around with this V2X thing and investing money into it is foolish. In the actual countries where they take safety seriously, you know where they actually successfully have reduced traffic accidents and death. In those countries you hear very little about V2X and the almost universal thing you hear from all the actually successful experts is that road infrastructure needs to be changed according to the newer standards. Most nations did gignatic damage to themselves in last 50 years and all of it needs to be undone. I'm sure those governments have V2X somewhere, but its simply not what most actual traffic engineers in those countries talk about. This V2X stuff is something car companies and lobbiest are primarly pushing. Its mostly popular with tech people. Most actually existing organisation for traffic safity are pushing what we know actually works. > It's essentially just a LTE or WLAN link Most vehicles can barley even software updates at all. And most old cars simply wont be updated with everything need to fully support this stuff. And even then most people on motorcycles and bikes don't have a great way of receiving that information. And there tons of old cars who want have it, so you can not realy on this for the next decades anyway. > This tech is something that has been in development since the 90s And designing streets without killing unbelievable amounts of people has been in development much longer and yet still traffic accidents are going up in the US not down. So maybe focus on that instead of fancy tech from the 90s. > Also it's not the first line of defense for pedestrians but rather one of many that can be rolled out now while also pushing other material benefits. The problem is that once this tech is there drives will relay on it. The idea that in order to be safe to cross, a whole complex change of senors, network tech and so on has to work is crazy. And talking about defense in depth is all fine and good, but if the actual first 4 lines of defense don't work, you are never gone stop anything. Literally every $ spent on this by the government or cities is 1$ less for infrastructure that is proven to save lives. I think people really don't understand the amount of destruction these issues are causing. The insane amount of people (and children) killed is one thing. But beyond that the cost of the current transportation patterns are actually insane. The amount of cost in terms of police, firefighters, the cost for delays and infrastructure damage. Not to mention the hilarious amount of cars that smash into people houses. The overhead of insurance. People times dealing with all that stuff. How much of this can V2X fix? Almost nothing is the answer. Its actually comical how bad it is. And even worse that this is done by doing road infrastructure the most expensive possible way. So its both expensive and does a bad job, its actually crazy. But sure lets invest in V2X when most towns can't even maintain the road in proper state or paint cross walks correctly. Great plan.
- CuriousCosmic 2y ago
  > Now in addition to traffic lights you need cameras everywhere, often covering 8 lane stroads, and then you need comptuers to process all that data. And do so correctly under difficult conditions. Installations of these system will be hugely expensive. I don't know where you live but I don't think I've been anywhere where 8 lanes is anything remotely close to standard. The busiest parts of most interstates may have 5 or 6 lanes and they don't have intersections but otherwise I don't think I've seen an intersection with more than 2, maybe 3 lanes (and an additional turning lane on each side). The only place I could think of like that is Texas but even then they are massively pushing changes to decrease car usage in general to the point they are investing billions in comprehensive high speed rail networks. And you don't need lots of compute power to do what these systems are doing. They are doing basic shape checks across a narrow column and you can do that at the crosswalk on a five dollar DSP. ----------- > If cities actually listen to experts on how to actually manage parking correctly, finding a parking spot wouldn't be hard in the first place. > That just means that the lobbies pushing this stuff have successfully done their job despite far more important things. No city in the US has even begun to fully implement modern traffic practices. Playing around with this V2X thing and investing money into it is foolish. The cities that are implementing this are listening to experts and they are solving the problem with infrastructure redesigns but again, that takes literal decades. Planning for changing an intersection to a roundabout or separating out a bike lane may take 5+ years before it even breaks ground and there's not the capacity to do that construction all at once anyways so you have to stagger it out. So you are at the point we are now. The ball is in motion but the "real fix" still has decades before it actually comes to fruition. This V2X system however is a decent bandaid while the actual fix rolls out and its full rollout window is 10-15 years rather than 50-75 years. So you can push this today and see some level of harm reduction while you wait for the actual fix to come around hopefully before you die. -------- > In the actual countries where they take safety seriously, you know where they actually successfully have reduced traffic accidents and death. In those countries you hear very little about V2X and the almost universal thing you hear from all the actually successful experts is that road infrastructure needs to be changed according to the newer standards. Most nations did gignatic damage to themselves in last 50 years and all of it needs to be undone. Yes. And the cities that have been pushing for this tech have been trying to make those exact same changes as well. The issue is that most of the changes that would ideally solve this problem won't be comprehensively rolled out for over 50 years. You can change construction standards and push for pedestrian first designs (which many of these cities are doing) however actually pushing those changes out to the streets takes decades of gradual construction and redesigning parts of the city. The difference is that this doesn't have to be a comprehensive solution. You can add it here and there in problem areas as you slowly roll it out but this tech is cheap. Probably 1000 USD per intersection. And you can add it without redesigning the entire intersection or the entire road. -------- > Most vehicles can barley even software updates at all. And most old cars simply wont be updated with everything need to fully support this stuff. And even then most people on motorcycles and bikes don't have a great way of receiving that information. And there tons of old cars who want have it, so you can not realy on this for the next decades anyway. Sure bikes and motorcycles will have limited exposure to this but they can use their phones as beacons in this case so that they are visible to others even if they can't consume the information. Modern vehicles could potentially get basic access to the system via an extension to Apple Carplay or Android auto and new vehicles would have it built in. But even then it'll still take time to roll out. HOWEVER it'll still be faster than the actual fix and both of them can be progressed when they can as forms of harm reduction. -------- > And designing streets without killing unbelievable amounts of people has been in development much longer and yet still traffic accidents are going up in the US not down. So maybe focus on that instead of fancy tech from the 90s. Sure but again, that takes time to roll out. Some cities don't care about it and that's a problem but the cities that do care about this are already trying to redesign streets to be pedestrian first but again, that takes 50+ years to do completely and costs an order of magnitude more than this bandaid. -------- > The problem is that once this tech is there drives will relay on it. The idea that in order to be safe to cross, a whole complex change of senors, network tech and so on has to work is crazy. That is not at all the argument being made. If you read the plan the DOT actually proposed, it's a very measured, reasonable plan that is focused on pushing a mitigation tool that also serves as a quality of life feature and it is explicitly designed to work without being comprehensive and gracefully handled degraded conditions. It is not the end all be all solution but it is something that the admin can throw money at today without slowing down proper infrastructure redesigns. -------- > Literally every $ spent on this by the government or cities is 1$ less for infrastructure that is proven to save lives. That is not true. To my knowledge most infrastructure redesigns are limited by capacity and route scheduling more than money. Capacity doesn't just come out of thin air and route scheduling is limited by existing routes so you can only change so much in a given area at one time. -------- > I think people really don't understand the amount of destruction these issues are causing Trust me, I am very much aware. > How much of this can V2X fix? It's not supposed to fix anything other than reduce pedestrian crossing deaths and bicyclist/jogger fatalities. Again it is a bandaid. It isn't being sold as a cure-all. It is just something that has been in the works for a long time that is now viable without being prohibitively complex or expensive. The economic and tech conditions are right for it to be able to roll out with marginal cost so there's no reason to not give it the little nudge it needs. -------- > But sure lets invest in V2X when most towns can't even maintain the road in proper state or paint cross walks correctly. Great plan. I'm not sure how you think this was even supposed to address that issue? This is a relatively minor amount of federal funding and isn't even coming out of anything close to the same funding pool as funds for maintaining major city road infrastructure or interstate infrastructure. And towns not being able to maintain the roads wouldn't even apply here since that's the state DOT not the federal DOT and that's a separate funding pool from a separate organisation entirely. ------- So I'm not sure what you expect here? Is your proposal "We should do absolutely nothing else ever until THE ONE TRUE FIX is complete and we are long since in retirement?" because that makes absolutely no sense. This is one thing that can be pushed forward that can roll out as a quality of life feature relatively easily and cheaply while also saving some lives along the way. It is by no means the only thing the DOT is doing. And it is even more by no means the only thing that cities are doing. And even more this isn't even anything close to eating up funds that could go to something else. So far this project has paid out less than 100 million USD and there's no chance it'll get anywhere close to 1 billion USD in funds spent. Considering the actual fix to this problem will cost hundreds of billions if not trillions of dollars invested over decades, this is a drop in the bucket that isn't even worth mentioning as "taking away money".
ikekkdcjkfke 2y ago
Wifi flesh detectors

panick21_ 2y ago

There is a gigantic difference between 'this technology has some useful applications' and 'this is fundamental to improving road safety'. And there is even a more gigantic difference between cost and useful how useful it is. Again, we know how to make gigantic gains in road safety. Its not difficult, its proven technology, its cheap, it can literally be done quickly. The only question is will of politicans. That's it. First of all, its already incredibly absurldy expensive how the US currently does traffic. There are far, far to many traffic lights and traffic lights are expensive. Far to many lanes and far to wide, incredibly expensive. That again makes traffic signaling more expensive. Now in addition to traffic lights you need cameras everywhere, often covering 8 lane stroads, and then you need comptuers to process all that data. And do so correctly under difficult conditions. Installations of these system will be hugely expensive. You can already compare your avg traffic intersection between the Netherlands and the US. The US often only has 1 sensor while the Dutch have many more. US intersections still operate based on completely outdated signaling orders for the most part (not to mention completly unsave and unfriendly for anything but cars). So when in the US most towns can't even configure their basic signaling orders in an efficent way, but you want them to do complex video image analysis and messaging? How does that make sense? Again, we know how to make things cheaper and much, much safer. And it doesn't need fancy technology. We don't need people with PhD in data science to implement this. If you want to make an argument that some of this has some uses, sure, whatever. I honestly don't think anything you mentioned is really all that complelling but I guess its possible. > The city infra can tell the vehicle (or your phone) where the closest open parking space is This is great, if your goal as a city is maximum utilisation of parking space, but that shouldn't be the goal in the first place. If cities actually listen to experts on how to actually manage parking correctly, finding a parking spot wouldn't be hard in the first place. > And importantly the cities that have been pioneering this tech have been pushing it while also pushing separated bike lanes, improved transit, more distinct street vs road distinctions, etc. That just means that the lobbies pushing this stuff have successfully done their job despite far more important things. No city in the US has even begun to fully implement modern traffic practices. Playing around with this V2X thing and investing money into it is foolish. In the actual countries where they take safety seriously, you know where they actually successfully have reduced traffic accidents and death. In those countries you hear very little about V2X and the almost universal thing you hear from all the actually successful experts is that road infrastructure needs to be changed according to the newer standards. Most nations did gignatic damage to themselves in last 50 years and all of it needs to be undone. I'm sure those governments have V2X somewhere, but its simply not what most actual traffic engineers in those countries talk about. This V2X stuff is something car companies and lobbiest are primarly pushing. Its mostly popular with tech people. Most actually existing organisation for traffic safity are pushing what we know actually works. > It's essentially just a LTE or WLAN link Most vehicles can barley even software updates at all. And most old cars simply wont be updated with everything need to fully support this stuff. And even then most people on motorcycles and bikes don't have a great way of receiving that information. And there tons of old cars who want have it, so you can not realy on this for the next decades anyway. > This tech is something that has been in development since the 90s And designing streets without killing unbelievable amounts of people has been in development much longer and yet still traffic accidents are going up in the US not down. So maybe focus on that instead of fancy tech from the 90s. > Also it's not the first line of defense for pedestrians but rather one of many that can be rolled out now while also pushing other material benefits. The problem is that once this tech is there drives will relay on it. The idea that in order to be safe to cross, a whole complex change of senors, network tech and so on has to work is crazy. And talking about defense in depth is all fine and good, but if the actual first 4 lines of defense don't work, you are never gone stop anything. Literally every $ spent on this by the government or cities is 1$ less for infrastructure that is proven to save lives. I think people really don't understand the amount of destruction these issues are causing. The insane amount of people (and children) killed is one thing. But beyond that the cost of the current transportation patterns are actually insane. The amount of cost in terms of police, firefighters, the cost for delays and infrastructure damage. Not to mention the hilarious amount of cars that smash into people houses. The overhead of insurance. People times dealing with all that stuff. How much of this can V2X fix? Almost nothing is the answer. Its actually comical how bad it is. And even worse that this is done by doing road infrastructure the most expensive possible way. So its both expensive and does a bad job, its actually crazy. But sure lets invest in V2X when most towns can't even maintain the road in proper state or paint cross walks correctly. Great plan.

CuriousCosmic 2y ago
> Now in addition to traffic lights you need cameras everywhere, often covering 8 lane stroads, and then you need comptuers to process all that data. And do so correctly under difficult conditions. Installations of these system will be hugely expensive. I don't know where you live but I don't think I've been anywhere where 8 lanes is anything remotely close to standard. The busiest parts of most interstates may have 5 or 6 lanes and they don't have intersections but otherwise I don't think I've seen an intersection with more than 2, maybe 3 lanes (and an additional turning lane on each side). The only place I could think of like that is Texas but even then they are massively pushing changes to decrease car usage in general to the point they are investing billions in comprehensive high speed rail networks. And you don't need lots of compute power to do what these systems are doing. They are doing basic shape checks across a narrow column and you can do that at the crosswalk on a five dollar DSP. ----------- > If cities actually listen to experts on how to actually manage parking correctly, finding a parking spot wouldn't be hard in the first place. > That just means that the lobbies pushing this stuff have successfully done their job despite far more important things. No city in the US has even begun to fully implement modern traffic practices. Playing around with this V2X thing and investing money into it is foolish. The cities that are implementing this are listening to experts and they are solving the problem with infrastructure redesigns but again, that takes literal decades. Planning for changing an intersection to a roundabout or separating out a bike lane may take 5+ years before it even breaks ground and there's not the capacity to do that construction all at once anyways so you have to stagger it out. So you are at the point we are now. The ball is in motion but the "real fix" still has decades before it actually comes to fruition. This V2X system however is a decent bandaid while the actual fix rolls out and its full rollout window is 10-15 years rather than 50-75 years. So you can push this today and see some level of harm reduction while you wait for the actual fix to come around hopefully before you die. -------- > In the actual countries where they take safety seriously, you know where they actually successfully have reduced traffic accidents and death. In those countries you hear very little about V2X and the almost universal thing you hear from all the actually successful experts is that road infrastructure needs to be changed according to the newer standards. Most nations did gignatic damage to themselves in last 50 years and all of it needs to be undone. Yes. And the cities that have been pushing for this tech have been trying to make those exact same changes as well. The issue is that most of the changes that would ideally solve this problem won't be comprehensively rolled out for over 50 years. You can change construction standards and push for pedestrian first designs (which many of these cities are doing) however actually pushing those changes out to the streets takes decades of gradual construction and redesigning parts of the city. The difference is that this doesn't have to be a comprehensive solution. You can add it here and there in problem areas as you slowly roll it out but this tech is cheap. Probably 1000 USD per intersection. And you can add it without redesigning the entire intersection or the entire road. -------- > Most vehicles can barley even software updates at all. And most old cars simply wont be updated with everything need to fully support this stuff. And even then most people on motorcycles and bikes don't have a great way of receiving that information. And there tons of old cars who want have it, so you can not realy on this for the next decades anyway. Sure bikes and motorcycles will have limited exposure to this but they can use their phones as beacons in this case so that they are visible to others even if they can't consume the information. Modern vehicles could potentially get basic access to the system via an extension to Apple Carplay or Android auto and new vehicles would have it built in. But even then it'll still take time to roll out. HOWEVER it'll still be faster than the actual fix and both of them can be progressed when they can as forms of harm reduction. -------- > And designing streets without killing unbelievable amounts of people has been in development much longer and yet still traffic accidents are going up in the US not down. So maybe focus on that instead of fancy tech from the 90s. Sure but again, that takes time to roll out. Some cities don't care about it and that's a problem but the cities that do care about this are already trying to redesign streets to be pedestrian first but again, that takes 50+ years to do completely and costs an order of magnitude more than this bandaid. -------- > The problem is that once this tech is there drives will relay on it. The idea that in order to be safe to cross, a whole complex change of senors, network tech and so on has to work is crazy. That is not at all the argument being made. If you read the plan the DOT actually proposed, it's a very measured, reasonable plan that is focused on pushing a mitigation tool that also serves as a quality of life feature and it is explicitly designed to work without being comprehensive and gracefully handled degraded conditions. It is not the end all be all solution but it is something that the admin can throw money at today without slowing down proper infrastructure redesigns. -------- > Literally every $ spent on this by the government or cities is 1$ less for infrastructure that is proven to save lives. That is not true. To my knowledge most infrastructure redesigns are limited by capacity and route scheduling more than money. Capacity doesn't just come out of thin air and route scheduling is limited by existing routes so you can only change so much in a given area at one time. -------- > I think people really don't understand the amount of destruction these issues are causing Trust me, I am very much aware. > How much of this can V2X fix? It's not supposed to fix anything other than reduce pedestrian crossing deaths and bicyclist/jogger fatalities. Again it is a bandaid. It isn't being sold as a cure-all. It is just something that has been in the works for a long time that is now viable without being prohibitively complex or expensive. The economic and tech conditions are right for it to be able to roll out with marginal cost so there's no reason to not give it the little nudge it needs. -------- > But sure lets invest in V2X when most towns can't even maintain the road in proper state or paint cross walks correctly. Great plan. I'm not sure how you think this was even supposed to address that issue? This is a relatively minor amount of federal funding and isn't even coming out of anything close to the same funding pool as funds for maintaining major city road infrastructure or interstate infrastructure. And towns not being able to maintain the roads wouldn't even apply here since that's the state DOT not the federal DOT and that's a separate funding pool from a separate organisation entirely. ------- So I'm not sure what you expect here? Is your proposal "We should do absolutely nothing else ever until THE ONE TRUE FIX is complete and we are long since in retirement?" because that makes absolutely no sense. This is one thing that can be pushed forward that can roll out as a quality of life feature relatively easily and cheaply while also saving some lives along the way. It is by no means the only thing the DOT is doing. And it is even more by no means the only thing that cities are doing. And even more this isn't even anything close to eating up funds that could go to something else. So far this project has paid out less than 100 million USD and there's no chance it'll get anywhere close to 1 billion USD in funds spent. Considering the actual fix to this problem will cost hundreds of billions if not trillions of dollars invested over decades, this is a drop in the bucket that isn't even worth mentioning as "taking away money".

ikekkdcjkfke 2y ago

Wifi flesh detectors