12 ms·

How Sweden Has Redesigned Streets to Route Around Bad Behavior

97 points by bane 10y ago

StavrosK 10y ago
The article fails to answer the question its title poses. How has Sweden redesigned streets? "Uhh, it kinda has, and fewer people die. Also, cameras." Okay, thanks.
- open-source-ux 10y ago
  Yes, it does lack detail. Here is the Wikipedia page of countries ranked by traffic-related deaths (the table headers are sortable) with data from 2013. Sweden has one of the lowest fatality rates per 100,000 inhabitants. But, interestingly, so does the UK with its much bigger population. Is there a common factor between these two countries that explains the low road fatality rates? Or maybe not? https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r...
  - refurb 10y ago
    Wouldn't you measure traffic fatalities by deaths per miles driven? Doing it by population completely ignores factors like the percent of people who actually drive and how much they drive each year. If you use deaths per inhabitabt it wouldn't surprise me that NYC has a very low rate as well (hint: many people don't own cars in NYC).
    - jontro 10y ago
      That statistic is also included in the table for some countries
- alkonaut 10y ago
  From article: speed limits and cameras. That's not the whole picture though. Infrastructure planning: we have few plane 4-way intersections on high speed roads, most have been replaced with roundabouts. Most wide highways with oncoming traffic (1+1) have been converted to 2+1 with central barrier, without changing the total width. This removes most high speed collisions. Still lots of accidents but they are now involving cars going in the same direction. The most important one though is probably much safer cars in recent years. This isn't policy of course, and only tangentially related to Sweden. It's basically helping the bottom line "for free". Also, not irrelevant: driving drunk is completely socially unacceptable, and the socially acceptable alcohol level is much closer to zero than the legally accepted limit.
joatmon-snoo 10y ago
Some details about the _how_ from an Economist article (http://www.economist.com/blogs/economist-explains/2014/02/economist-explains-16 http://www.economist.com/blogs/economist-explains/2014/02/ec...): Planning has played the biggest part in reducing accidents. Roads in Sweden are built to prioritise safety over speed or convenience. Low urban speed limits, pedestrian zones and barriers that separate cars from bicycles and oncoming traffic have helped. Building 1,500km (900 miles) of “2+1” roads—where each lane of traffic takes turns to use a middle lane for overtaking—is reckoned to have saved around 145 lives over the first decade of Vision Zero. And 12,600 safer crossings, including pedestrian bridges and zebra stripes flanked by flashing lights and protected with speed bumps, are estimated to have halved the number of pedestrian deaths over the past five years. Strict policing has also helped: less than 0.25% of drivers tested are over the alcohol limit. Road deaths of children under seven have plummeted—in 2012 only one was killed, compared with 58 in 1970.
- UweSchmidt 10y ago
  I always found those 2+1 roads stressful, as it required unpleasant delays behind very slow vehicles and puts a limit on how many cars can overtake before the other side gets the middle lane. Is this the price for saving lives, or do I misunderstand anything? Does the 2+1 road replace 1+1 roads (which is probably a lot more dangerous)?
  - RyanZAG 10y ago
    In Cairo, most of the roads during rush hour are so packed with cars that you can't drive faster than 5km/hour. Those packed roads with slowly moving cars have the lowest accident rate in the country. So yes, you can definitely reduce accidents by forcing everyone into high traffic situations where they have no choice but to drive extremely slowly. In IT security, you can also reduce hacked computers by making so many security regulations that nobody can actually use the computers. It does work, but nobody really gets their job done. Or at least, in the case of the state department, it works until your secretary of state moves all of her email onto her own server to avoid your regulations and you end up with a political firestorm. At some point you probably want to try find solutions that optimize safety and actually getting something done at the same time. Creating traffic isn't a very good solution in the long run, I think.
rvern 10y ago
More details on Wikipedia: https://en.wikipedia.org/wiki/Vision_Zero https://en.wikipedia.org/wiki/Vision_Zero.
mannykannot 10y ago
It is beyond me why the traffic strategist quoted in the second paragraph feels that it is necessary to absolve distracted drivers. It is not a prerequisite for what they are doing.
- jessriedel 10y ago
  It's just a terrible article. > But unlike almost every other scheme to make roads safer, Vision doesn't try to blame the victim or the perpetrator. Oh right, because the car safety tech responsible for the lion's share of the lives saved over the past few decades is all about blaming the perpetrator.
  - geon 10y ago
    I think you misread that. It is not about blaming the driver, in the sense that we all know driver will continue to drive poorly. So if we want change, we can't just blame the bad drivers and go on designing cars and roads the same way. Instead we need to design the roads, knowing that bad drivers will be using them. We in Sweden have some of the worlds highest standards for drivers tests. You could blame bad drivers and raise the standards even more. But you know what? Some people will drive without even having a license. You can't stop people like that unless you ban cars altogether.
    - jessriedel 10y ago
      No, I think you misread me. I am not arguing that we should do more blaming, I am arguing the writer is wrong to assert that this is some miraculous new insight. Most lives saved in recent decades have come from improved safety tech, which has nothing to do with blame and everything to do with minimizing the bad effects of bad driving.
- ajmurmann 10y ago
  It's not a prerequisite for what they are doing, but it's a prerequisite to get political buy in to go down that round rather than spend more effort on trying to enforce behavior that seems to be unenforceable. Of course you could try both at the danger time in theory but that seems like a much harder sell given most political systems.
ChuckMcM 10y ago
I was struck by their camera system, it would be interesting to have traffic cameras that recorded but did not attempt to enforce traffic violations. Then if you're in court for something there might be an additional piece of information about how safe of a driver you are given how often you've been recorded violating various traffic rules. In Sunnyvale there are a number of the 'automatic' speed signs (they show you your speed and flash if it is over the limit) and on the streets near me they do seem to improve compliance when they are active. Of course given that they are two lane streets (one lane in each direction) one compliant driver can keep the entire block compliant :-).
- yxhuvud 10y ago
  The cameras do catch some speeders, but the areas with cameras is very marked out and visible by the drivers.
- prawn 10y ago
  I read about one place that trialled red lights that were shorter for roads where the drivers had kept under the speed limit. Push over the limit and you'll get a longer stop at the lights. Defeats the purpose of speeding.
  - Coincoin 10y ago
    We have a similar system on a specific road over here and what end up happening is there is always that "idiot" who doesn't know about the system and just triggers the penalty for everybody else on the road.
  - mjevans 10y ago
    Won't this just trigger people to try to go faster to 'make the light'? Which is the issue in the first place; the lights not being timed up to /reward/ driving at the indicated speed.
Etheryte 10y ago
Having both kmh and mph in two sentences next to one another is very confusing, articles should use one or the other as a primary unit and include the other, if needed, in parenthesis or an aside.
vanderZwan 10y ago
As a Dutchman living in Stockholm I can tell you one thing: they should fix their goddamn bike lanes. Sure, they have them, which makes them better than 90% of the rest of the world. They're still shit. There's no separation of car- and bike traffic. To turn left a cyclist first has to cross parallel car traffic, making this unsafer for cyclists (and annoying for the people in cars, for that matter). Key issues: - the cyclist and driver should meet at a (near) 90 degree angle, so neither requires eyes in the back of their head - separate the goddamn lanes so neither car nor cyclist feels uncomfortably close to the other - don't require extra space because every square meter counts in a big city Well guess what, we've figured this out decades ago in the Netherlands: https://www.youtube.com/watch?v=FlApbxLz6pA https://www.youtube.com/watch?v=FlApbxLz6pA https://www.youtube.com/watch?v=5HDN9fUlqU8 https://www.youtube.com/watch?v=5HDN9fUlqU8 https://www.youtube.com/watch?v=41XBzAOmmIU https://www.youtube.com/watch?v=41XBzAOmmIU
- lacampbell 10y ago
  In New Zealand they have a nasty habit of painting a bit of the road green, then painting a bit of the footpath green. So the "cycle lane" effectively veers on and off the road, or over busy driveways where no car is looking for cyclists on what to them is the footpath. It's extremely dangerous to follow the cycle lanes our government conjures up. I imagine they have some quota to meet like "X metres of cycle lane".
  - vanderZwan 10y ago
    In general I'm flabbergasted as to why countries don't just look at what has been tried in other trailblazing countries with issues like this. In the case of cycling, the Netherlands and Denmark are so far ahead of the rest of the world that catching up should be a matter of asking us about everything that we've tried, about did (not) work and why, and then stealing the good ideas.
    - chrismealy 10y ago
      The degree of "not invented here" among city officials is astounding. I think a lot of them don't really want to succeed.
geon 10y ago
Another change recently was more fine grained speed limits. You now have speeds for each 10 km/h instead of 20. Some roads had their speed limit raised, like from 90 to 100 km/h. I think it has created a higher sense of fairness, and the limits are respected more as a whole.
henrikschroder 10y ago
> the speed limits are set low, at 30 kmh, or 18.6 mph. Pet peeve: 30km/h = 20mph. 30.0km/h = 18.6mph. 50km/h = 30mph, not 31mph. It's so annoying when people don't do significant digits correctly when converting between metric and imperial, it completely breaks the flow.
- karthikb 10y ago
  30 kph is a speed limit and thus an exact number, not an approximation to 1 SF. Thus, 18.6 mph is more correct than 20 mph.
  - masklinn 10y ago
    > 30 kph is a speed limit and thus an exact number, not an approximation to 1 SF. Thus, 18.6 mph is more correct than 20 mph. No, because the imperial speed limit corresponding to 30km/h is 20mph, and imperial drivers would have encountered that and thus would have a clear and absolutely trivial frame of reference. Likewise 50km/h and 30mph. 18.6mph and 31mph is not correct, it's worthless pedantry completely missing the point to the disservice of the reader.
    - karthikb 10y ago
      I was responding to parent who was making the argument based on significant figures. Speed limits are exact (though enforcement may not be!). But I completely agree that from a practical perspective, 30 kph = 20 mph, 100 kph = 65 mph, and so on.
      - henrikschroder 10y ago
        > Speed limits are exact And this is exactly the kind of worthless pedantry that masklinn called out.
- tomjen3 10y ago
  My pet peeve is that it is way too slow. Sorry but I don't want to spend an extra day a month in traffic so retarded teens can send sms back and forth without ending in a collision.
cmurf 10y ago
If only software designers accepted more blame for bad design, admitting that it's hard, rather than always considering that it's user error.
Mz 10y ago
If a car almost hits a pedestrian when the car is turning right on a red, whose fault is it? According to Matts-Åke Belin, Sweden's traffic safety strategist, the blame is on whoever designed the intersection. This. So much this. This was kind of how I raised kids, moderated forums, etc. When people cannot get along without serious drama, someone running the system has some work to do. Trying to find someone to blame -- "who started it" -- is a sucky approach. I generally want to figure out how to put a stop to it, not who to point fingers at. And it is an approach that gets uncommonly good results when followed faithfully/to its logical conclusion.
kristianp 10y ago
How are they going to get to zero by 2020? That's an infinite improvement in 3 years.
- throwaway7767 10y ago
  I don't think anyone reasonably expects them to eliminate fatalities in traffic. It's just a goal to work towards, or in other words it's a way to market the initiative.
Buge 10y ago
>New York is actually pursuing a version of Vision Zero (though not without a few snags), so perhaps, one day, cars will no longer be able to turn right on red, Right on red is already illegal in New York City. https://en.wikipedia.org/wiki/Turn_on_red#North_America https://en.wikipedia.org/wiki/Turn_on_red#North_America

97 points by bane 10y ago

The article fails to answer the question its title poses. How has Sweden redesigned streets? "Uhh, it kinda has, and fewer people die. Also, cameras." Okay, thanks.

open-source-ux 10y ago
Yes, it does lack detail. Here is the Wikipedia page of countries ranked by traffic-related deaths (the table headers are sortable) with data from 2013. Sweden has one of the lowest fatality rates per 100,000 inhabitants. But, interestingly, so does the UK with its much bigger population. Is there a common factor between these two countries that explains the low road fatality rates? Or maybe not? https://en.wikipedia.org/wiki/List_of_countries_by_traffic-related_death_rate https://en.wikipedia.org/wiki/List_of_countries_by_traffic-r...
- refurb 10y ago
  Wouldn't you measure traffic fatalities by deaths per miles driven? Doing it by population completely ignores factors like the percent of people who actually drive and how much they drive each year. If you use deaths per inhabitabt it wouldn't surprise me that NYC has a very low rate as well (hint: many people don't own cars in NYC).
  - jontro 10y ago
    That statistic is also included in the table for some countries
alkonaut 10y ago
From article: speed limits and cameras. That's not the whole picture though. Infrastructure planning: we have few plane 4-way intersections on high speed roads, most have been replaced with roundabouts. Most wide highways with oncoming traffic (1+1) have been converted to 2+1 with central barrier, without changing the total width. This removes most high speed collisions. Still lots of accidents but they are now involving cars going in the same direction. The most important one though is probably much safer cars in recent years. This isn't policy of course, and only tangentially related to Sweden. It's basically helping the bottom line "for free". Also, not irrelevant: driving drunk is completely socially unacceptable, and the socially acceptable alcohol level is much closer to zero than the legally accepted limit.

joatmon-snoo 10y ago

Some details about the _how_ from an Economist article (http://www.economist.com/blogs/economist-explains/2014/02/economist-explains-16 http://www.economist.com/blogs/economist-explains/2014/02/ec...): Planning has played the biggest part in reducing accidents. Roads in Sweden are built to prioritise safety over speed or convenience. Low urban speed limits, pedestrian zones and barriers that separate cars from bicycles and oncoming traffic have helped. Building 1,500km (900 miles) of “2+1” roads—where each lane of traffic takes turns to use a middle lane for overtaking—is reckoned to have saved around 145 lives over the first decade of Vision Zero. And 12,600 safer crossings, including pedestrian bridges and zebra stripes flanked by flashing lights and protected with speed bumps, are estimated to have halved the number of pedestrian deaths over the past five years. Strict policing has also helped: less than 0.25% of drivers tested are over the alcohol limit. Road deaths of children under seven have plummeted—in 2012 only one was killed, compared with 58 in 1970.

UweSchmidt 10y ago
I always found those 2+1 roads stressful, as it required unpleasant delays behind very slow vehicles and puts a limit on how many cars can overtake before the other side gets the middle lane. Is this the price for saving lives, or do I misunderstand anything? Does the 2+1 road replace 1+1 roads (which is probably a lot more dangerous)?
- RyanZAG 10y ago
  In Cairo, most of the roads during rush hour are so packed with cars that you can't drive faster than 5km/hour. Those packed roads with slowly moving cars have the lowest accident rate in the country. So yes, you can definitely reduce accidents by forcing everyone into high traffic situations where they have no choice but to drive extremely slowly. In IT security, you can also reduce hacked computers by making so many security regulations that nobody can actually use the computers. It does work, but nobody really gets their job done. Or at least, in the case of the state department, it works until your secretary of state moves all of her email onto her own server to avoid your regulations and you end up with a political firestorm. At some point you probably want to try find solutions that optimize safety and actually getting something done at the same time. Creating traffic isn't a very good solution in the long run, I think.

rvern 10y ago

More details on Wikipedia: https://en.wikipedia.org/wiki/Vision_Zero https://en.wikipedia.org/wiki/Vision_Zero.

mannykannot 10y ago

It is beyond me why the traffic strategist quoted in the second paragraph feels that it is necessary to absolve distracted drivers. It is not a prerequisite for what they are doing.

jessriedel 10y ago
It's just a terrible article. > But unlike almost every other scheme to make roads safer, Vision doesn't try to blame the victim or the perpetrator. Oh right, because the car safety tech responsible for the lion's share of the lives saved over the past few decades is all about blaming the perpetrator.
- geon 10y ago
  I think you misread that. It is not about blaming the driver, in the sense that we all know driver will continue to drive poorly. So if we want change, we can't just blame the bad drivers and go on designing cars and roads the same way. Instead we need to design the roads, knowing that bad drivers will be using them. We in Sweden have some of the worlds highest standards for drivers tests. You could blame bad drivers and raise the standards even more. But you know what? Some people will drive without even having a license. You can't stop people like that unless you ban cars altogether.
  - jessriedel 10y ago
    No, I think you misread me. I am not arguing that we should do more blaming, I am arguing the writer is wrong to assert that this is some miraculous new insight. Most lives saved in recent decades have come from improved safety tech, which has nothing to do with blame and everything to do with minimizing the bad effects of bad driving.
ajmurmann 10y ago
It's not a prerequisite for what they are doing, but it's a prerequisite to get political buy in to go down that round rather than spend more effort on trying to enforce behavior that seems to be unenforceable. Of course you could try both at the danger time in theory but that seems like a much harder sell given most political systems.

ChuckMcM 10y ago

I was struck by their camera system, it would be interesting to have traffic cameras that recorded but did not attempt to enforce traffic violations. Then if you're in court for something there might be an additional piece of information about how safe of a driver you are given how often you've been recorded violating various traffic rules. In Sunnyvale there are a number of the 'automatic' speed signs (they show you your speed and flash if it is over the limit) and on the streets near me they do seem to improve compliance when they are active. Of course given that they are two lane streets (one lane in each direction) one compliant driver can keep the entire block compliant :-).

yxhuvud 10y ago
The cameras do catch some speeders, but the areas with cameras is very marked out and visible by the drivers.
prawn 10y ago
I read about one place that trialled red lights that were shorter for roads where the drivers had kept under the speed limit. Push over the limit and you'll get a longer stop at the lights. Defeats the purpose of speeding.
- Coincoin 10y ago
  We have a similar system on a specific road over here and what end up happening is there is always that "idiot" who doesn't know about the system and just triggers the penalty for everybody else on the road.
- mjevans 10y ago
  Won't this just trigger people to try to go faster to 'make the light'? Which is the issue in the first place; the lights not being timed up to /reward/ driving at the indicated speed.

Etheryte 10y ago

Having both kmh and mph in two sentences next to one another is very confusing, articles should use one or the other as a primary unit and include the other, if needed, in parenthesis or an aside.

vanderZwan 10y ago

As a Dutchman living in Stockholm I can tell you one thing: they should fix their goddamn bike lanes. Sure, they have them, which makes them better than 90% of the rest of the world. They're still shit. There's no separation of car- and bike traffic. To turn left a cyclist first has to cross parallel car traffic, making this unsafer for cyclists (and annoying for the people in cars, for that matter). Key issues: - the cyclist and driver should meet at a (near) 90 degree angle, so neither requires eyes in the back of their head - separate the goddamn lanes so neither car nor cyclist feels uncomfortably close to the other - don't require extra space because every square meter counts in a big city Well guess what, we've figured this out decades ago in the Netherlands: https://www.youtube.com/watch?v=FlApbxLz6pA https://www.youtube.com/watch?v=FlApbxLz6pA https://www.youtube.com/watch?v=5HDN9fUlqU8 https://www.youtube.com/watch?v=5HDN9fUlqU8 https://www.youtube.com/watch?v=41XBzAOmmIU https://www.youtube.com/watch?v=41XBzAOmmIU

lacampbell 10y ago
In New Zealand they have a nasty habit of painting a bit of the road green, then painting a bit of the footpath green. So the "cycle lane" effectively veers on and off the road, or over busy driveways where no car is looking for cyclists on what to them is the footpath. It's extremely dangerous to follow the cycle lanes our government conjures up. I imagine they have some quota to meet like "X metres of cycle lane".
- vanderZwan 10y ago
  In general I'm flabbergasted as to why countries don't just look at what has been tried in other trailblazing countries with issues like this. In the case of cycling, the Netherlands and Denmark are so far ahead of the rest of the world that catching up should be a matter of asking us about everything that we've tried, about did (not) work and why, and then stealing the good ideas.
  - chrismealy 10y ago
    The degree of "not invented here" among city officials is astounding. I think a lot of them don't really want to succeed.

geon 10y ago

Another change recently was more fine grained speed limits. You now have speeds for each 10 km/h instead of 20. Some roads had their speed limit raised, like from 90 to 100 km/h. I think it has created a higher sense of fairness, and the limits are respected more as a whole.

henrikschroder 10y ago

> the speed limits are set low, at 30 kmh, or 18.6 mph. Pet peeve: 30km/h = 20mph. 30.0km/h = 18.6mph. 50km/h = 30mph, not 31mph. It's so annoying when people don't do significant digits correctly when converting between metric and imperial, it completely breaks the flow.

karthikb 10y ago
30 kph is a speed limit and thus an exact number, not an approximation to 1 SF. Thus, 18.6 mph is more correct than 20 mph.
- masklinn 10y ago
  > 30 kph is a speed limit and thus an exact number, not an approximation to 1 SF. Thus, 18.6 mph is more correct than 20 mph. No, because the imperial speed limit corresponding to 30km/h is 20mph, and imperial drivers would have encountered that and thus would have a clear and absolutely trivial frame of reference. Likewise 50km/h and 30mph. 18.6mph and 31mph is not correct, it's worthless pedantry completely missing the point to the disservice of the reader.
  - karthikb 10y ago
    I was responding to parent who was making the argument based on significant figures. Speed limits are exact (though enforcement may not be!). But I completely agree that from a practical perspective, 30 kph = 20 mph, 100 kph = 65 mph, and so on.
    - henrikschroder 10y ago
      > Speed limits are exact And this is exactly the kind of worthless pedantry that masklinn called out.
tomjen3 10y ago
My pet peeve is that it is way too slow. Sorry but I don't want to spend an extra day a month in traffic so retarded teens can send sms back and forth without ending in a collision.

cmurf 10y ago

If only software designers accepted more blame for bad design, admitting that it's hard, rather than always considering that it's user error.

Mz 10y ago

If a car almost hits a pedestrian when the car is turning right on a red, whose fault is it? According to Matts-Åke Belin, Sweden's traffic safety strategist, the blame is on whoever designed the intersection. This. So much this. This was kind of how I raised kids, moderated forums, etc. When people cannot get along without serious drama, someone running the system has some work to do. Trying to find someone to blame -- "who started it" -- is a sucky approach. I generally want to figure out how to put a stop to it, not who to point fingers at. And it is an approach that gets uncommonly good results when followed faithfully/to its logical conclusion.

kristianp 10y ago

How are they going to get to zero by 2020? That's an infinite improvement in 3 years.

throwaway7767 10y ago
I don't think anyone reasonably expects them to eliminate fatalities in traffic. It's just a goal to work towards, or in other words it's a way to market the initiative.

Buge 10y ago

>New York is actually pursuing a version of Vision Zero (though not without a few snags), so perhaps, one day, cars will no longer be able to turn right on red, Right on red is already illegal in New York City. https://en.wikipedia.org/wiki/Turn_on_red#North_America https://en.wikipedia.org/wiki/Turn_on_red#North_America