Removing street signs, lights and arrows increases safety and road capacity(wilsonquarterly.com)
wilsonquarterly.com
Removing street signs, lights and arrows increases safety and road capacity
http://www.wilsonquarterly.com/article.cfm?aid=1234
14 comments
My perspective is that people are better at what they're doing when they're paying attention. Give them hints (signs) and shortcuts (straight roads), and their minds wander to something else, until something exceptional happens. There are few if any exceptionally good things that can happen in a car (most revolve around avoiding exceptionally bad things).
In becoming a "native" Boston driver, I have noticed that not all drivers pay equal attention. In other places I have lived, the attention that other drivers paid was not as obvious, because roads were straight and everyone was a "good" (rigid rule obeying) driver most of the time. The roads here are twisty, bumpy, often full of holes, merge in odd manners, have roundabouts, unconventional signs, and generally more difficult to drive. Yet I seem to see fewer road-side accidents on the roads that I drive every day.
Drivers new to the area see narrow lanes (more frequent lane violations), reduced traffic control (fewer clear opportunities to change lanes or turn), and faster drivers as chaotic and aggressive. It is also scary for tourists to cross the street in the midst of this apparent chaos (you'll see natives walking right through). So you might say that Boston has been in this sort of traffic chaos for a rather long time, and it has never been significantly more dangerous in reality (certainly in perception). I can only conclude that drivers who rely on a rigid rule-set have difficulty in unexpected situations, but when the exceptional is unexceptional, they will develop a more sophisticated model.
The net benefit of signs might not be to increase safety, but to allow for lazier, more distracted drivers. Some signs are generally helpful (construction / traffic / drawbridge ahead ), but I would like to see a general reduction. People are smart, and only do what is required. We can tell them not to talk on the phone or "drive distracted" with little effect. Give them a couple of (hopefully temporarily) bad experiences while they are not paying attention, and they will certainly pay more attention all the time.
EDIT: I should also point out that many people refuse to drive in Boston, making the driving population more self-selective than other cities. I don't think that this is a problem, but as society becomes more reliant on driving it becomes one. People who do not drive here can still get around by means of the subway, cabs, or plain walking (Boston isn't geographically large). I can't say the same of most other American cities.
In becoming a "native" Boston driver, I have noticed that not all drivers pay equal attention. In other places I have lived, the attention that other drivers paid was not as obvious, because roads were straight and everyone was a "good" (rigid rule obeying) driver most of the time. The roads here are twisty, bumpy, often full of holes, merge in odd manners, have roundabouts, unconventional signs, and generally more difficult to drive. Yet I seem to see fewer road-side accidents on the roads that I drive every day.
Drivers new to the area see narrow lanes (more frequent lane violations), reduced traffic control (fewer clear opportunities to change lanes or turn), and faster drivers as chaotic and aggressive. It is also scary for tourists to cross the street in the midst of this apparent chaos (you'll see natives walking right through). So you might say that Boston has been in this sort of traffic chaos for a rather long time, and it has never been significantly more dangerous in reality (certainly in perception). I can only conclude that drivers who rely on a rigid rule-set have difficulty in unexpected situations, but when the exceptional is unexceptional, they will develop a more sophisticated model.
The net benefit of signs might not be to increase safety, but to allow for lazier, more distracted drivers. Some signs are generally helpful (construction / traffic / drawbridge ahead ), but I would like to see a general reduction. People are smart, and only do what is required. We can tell them not to talk on the phone or "drive distracted" with little effect. Give them a couple of (hopefully temporarily) bad experiences while they are not paying attention, and they will certainly pay more attention all the time.
EDIT: I should also point out that many people refuse to drive in Boston, making the driving population more self-selective than other cities. I don't think that this is a problem, but as society becomes more reliant on driving it becomes one. People who do not drive here can still get around by means of the subway, cabs, or plain walking (Boston isn't geographically large). I can't say the same of most other American cities.
There are other costs though.
Let's suppose that a thirty minute drive where I'm impelled to pay attention all the time is safer. It still would take a lot out of me and make me less productive at work.
It's like the traffic circle where people felt less safe. The fact is you need to increase both actual safety and the feeling of safety, otherwise people are less happy and quite rightly aren't going to support the changes.
Let's suppose that a thirty minute drive where I'm impelled to pay attention all the time is safer. It still would take a lot out of me and make me less productive at work.
It's like the traffic circle where people felt less safe. The fact is you need to increase both actual safety and the feeling of safety, otherwise people are less happy and quite rightly aren't going to support the changes.
Not just make you less productive at work, but these roads are inherently roads where you drive more carefully (read slower). Driving slower means longer commutes, and can be correlated to the same economic problem as road congestion (read slower). Richmond estimates that it has saved 18 million dollars from the economic cost of congested roads by providing public transit. There's a lot of money to be lost in the time.
Sometimes I wonder if A/B tests we run suffer from this syndrome.
Maybe temporary in the sense that once the driver is out of the naked area, back to the "forgiving road" where any obstacle that might possibly hinder the flow of speeding automotives has been removed, the alertness probably falls back to zero again.
I don't think it's temporary in the sense that everything goes nicely for the first year and then everyone gets back to normal and accidents increase again.
I think that once you're sloshing through flocks of living people with your car and never get a chance to actually clear your way and speed up over 10-20km/h, I suspect that you will always become alert whenever you hit that area.
Conversely, walking around in an area where there are 1.5 ton blocks of moving metal without ever getting a chance to actually become absorbed in your thoughts and ignore the environment that you're in all together, you'll certainly feel less secure because you never know what's coming next and from where.
Because there's no "known" in the equation and speeds are forcibly slow, people are both more careful and more able to stop their vehicles before anything happens.
I don't think it's temporary in the sense that everything goes nicely for the first year and then everyone gets back to normal and accidents increase again.
I think that once you're sloshing through flocks of living people with your car and never get a chance to actually clear your way and speed up over 10-20km/h, I suspect that you will always become alert whenever you hit that area.
Conversely, walking around in an area where there are 1.5 ton blocks of moving metal without ever getting a chance to actually become absorbed in your thoughts and ignore the environment that you're in all together, you'll certainly feel less secure because you never know what's coming next and from where.
Because there's no "known" in the equation and speeds are forcibly slow, people are both more careful and more able to stop their vehicles before anything happens.
The other major possibility is the road markings where distracting drivers. Understanding the rules while driving is more complex than simply not hitting people.
I would much rather someone trying to make a U turn thinking “Is this safe?” vs. “What does the sign say?”
I would much rather someone trying to make a U turn thinking “Is this safe?” vs. “What does the sign say?”
In some ways i agree, but in others i disagree.
I'm a person who loves traffic lights, because the traffic light specifically tells a road user You can go or no you cant go. nobody has to do any guesswork as to who is allowed to go.
I see the whole "more alert because its more dangerous" every day. There used to be a single lane roundabout that i traveled through every day that they have now changed to a double lane roundabout. I am not sure of actual crash statistics related to it, but i see that people are more alert, because it is far more dangerous to what it used to be.
I'm a person who loves traffic lights, because the traffic light specifically tells a road user You can go or no you cant go. nobody has to do any guesswork as to who is allowed to go.
I see the whole "more alert because its more dangerous" every day. There used to be a single lane roundabout that i traveled through every day that they have now changed to a double lane roundabout. I am not sure of actual crash statistics related to it, but i see that people are more alert, because it is far more dangerous to what it used to be.
I'm a person who loves traffic lights, because the traffic light specifically tells a road user You can go or no you cant go. nobody has to do any guesswork as to who is allowed to go.
And yet about 1/4 of all traffic accidents happen at a traffic light within 5 seconds of a light change. Why? Because one person tried to cheat the rule, while another followed the rule without checking whether it was actually safe to do so.
And yet about 1/4 of all traffic accidents happen at a traffic light within 5 seconds of a light change. Why? Because one person tried to cheat the rule, while another followed the rule without checking whether it was actually safe to do so.
Strong agreement on this point. Roads like River Drive in Philadelphia and Storrow Drive in Boston have twisty, blind corners and a 35 mph speed limit, but a regular flow of traffic that is at least 20 mph faster than that. It seems that once people become familiar with a road they fly along way faster than is actually safe. They promote their speed to their level of incompetence, if you will.
On the other hand, it makes me consider (but doesn't remotely convince me of) the possibility that all of Boston's rotaries maybe aren't the dumbest road designs possible.
On the other hand, it makes me consider (but doesn't remotely convince me of) the possibility that all of Boston's rotaries maybe aren't the dumbest road designs possible.
What I got from the article was not that naked roads should be everywhere, but actually that they absolutely shouldn't be everywhere. They should, however, be considered, as should a great many other possibilities, in the context of the specific environment that a traffic engineer is designing for. Instead we too often see a bias towards signs and signals and too much copy+paste of what has been done elsewhere.
It's possible that they will...but also possible that they won't. Given the evidence presented, we can't know. Has anyone done a followup study?
We could have street signs that turn off when safety goes down then, and turn back on when measurements show that people are getting used to it
You have it EXACTLY right. Most Indian roads have exactly this sort of system in place, so it's a pretty good comparison.
http://timesofindia.indiatimes.com/india/India-is-worlds-roa...
Not so great of an example.
The model is like every classic "We turned up the lights and people became more productive. And then we turned down the lights and they became more productive. And then..." It's flawed.
http://timesofindia.indiatimes.com/india/India-is-worlds-roa...
Not so great of an example.
The model is like every classic "We turned up the lights and people became more productive. And then we turned down the lights and they became more productive. And then..." It's flawed.
parent is referencing the Hawthorne Effect http://en.wikipedia.org/wiki/Hawthorne_effect
I used to think shared space was a great alternative to car-centric streets but after reading David Hembrow's blog about bicycling in the Netherlands for a while it's clear that shared space is just a half-measure. If you want a society where people other than (mostly male) daredevils in their 20s and 30s can cycle then you need segregated biking facilities. Holland's percent of trips by bike (modal share) is around 30% (America is about 2%). It got there by having world-class segregated cycling facilities, not shared space.
David Hembrow on shared space: http://hembrow.blogspot.com/2008/11/shared-space.html
If you have the time check out John Pucher's "Cycling for Everyone: Lessons for Vancouver from the Netherlands, Denmark, and Germany": http://www.sfu.ca/city/city_pgm_video020.htm
David Hembrow on shared space: http://hembrow.blogspot.com/2008/11/shared-space.html
If you have the time check out John Pucher's "Cycling for Everyone: Lessons for Vancouver from the Netherlands, Denmark, and Germany": http://www.sfu.ca/city/city_pgm_video020.htm
What definitely does not work is drawing cyclist lanes onto the road. In Germany at least I think for a while it was believed that cycling lanes should be drawn on the roads rather than on the sideway.
The result is that the cycling lanes become very dangerous, because car use them for short term parking all the time (it is not unusual to have a parking car every 100 meters). That would mean cycling around the car and hitting the normal road. I think it is safer to just stay on the road all the time, rather than cycle in and out.
The parking on the cycling lane is actually illegal, but people don't care. Sometimes I consider just calling the police for every car that parks there, but I suspect they wouldn't really bother.
The result is that the cycling lanes become very dangerous, because car use them for short term parking all the time (it is not unusual to have a parking car every 100 meters). That would mean cycling around the car and hitting the normal road. I think it is safer to just stay on the road all the time, rather than cycle in and out.
The parking on the cycling lane is actually illegal, but people don't care. Sometimes I consider just calling the police for every car that parks there, but I suspect they wouldn't really bother.
I don't see why this is a big deal. Philadelphia has some recently added cycling lanes that I use all the time. If a car is parked you just look and cautiously veer around it. Sometimes you have to wait for a few seconds if there's heavy traffic and the space is narrow. But usually there's even enough room to get around a car without actually going back into the normal lane, since our bike lanes are converted from fully sized car lanes, which are bigger than cars.
And of course people park in them short-term with their hazards on. I'm not sure whether it's illegal or not (in Philly people park with impunity long-term in the turn lanes sometimes). But if it is it shouldn't be, there's no better alternative. Park in the non-bike lane, and make other cars go around them into the bike lane? No idling at all?
And of course people park in them short-term with their hazards on. I'm not sure whether it's illegal or not (in Philly people park with impunity long-term in the turn lanes sometimes). But if it is it shouldn't be, there's no better alternative. Park in the non-bike lane, and make other cars go around them into the bike lane? No idling at all?
I believe the big deal is that you are asking "regular folks" to weave in and out of traffic on a bicycle on the way to work. Not to mention all of the other hazards with the bike lane, at least in DC area, such as being the wasteland for crap left over after the plowed snow banks melt. I'm usually comfortable on even the worst roads but wouldn't want my wife to bike to work on many of them.
As a cyclist my feeling is that the risky situations are the ones were you cross into and out of traffic, or do anything unusual. So the cycling lane would be more risky than just driving straight on on the road. In the latter case, at least car drivers can see you from far away and prepare for overtaking you.
Maybe our streets are just not as big as yours in Philadelphia.
Maybe our streets are just not as big as yours in Philadelphia.
From my experience with cycling I can say that cycling on painted cycling lanes feels 2x safer than on shared space, and cycling on bike-only roads feels 10x safer (by far the worst problem on separate roads is dogs). The problem with shared space is that cars don't keep enough distance from the side. They go around you instead of driving in a straight line. This is especially a problem with trucks and buses, because they are so big.
What I usually do is cycling without hands on the handlebars. This keeps the cars at a much safer distance.
What I usually do is cycling without hands on the handlebars. This keeps the cars at a much safer distance.
This is absolutely correct. The cycle lanes in Amsterdam took up relatively little space, had physical dividers, and at busy intersections there would be separate small traffic lights for cyclists - they get the green light about 10 seconds before the cars, which is usually enough time to get through the intersection and makes life easier for both cyclists and drivers.
I just wanted to double-check your comment about "ten seconds" - really that much? As in, timed with a watch, or it "feels like" ten seconds?
I'm only asking because ten seconds is actually quite a long time at an intersection, and it fascinates me that such a concession to bike safety would be made - and work.
I'm only asking because ten seconds is actually quite a long time at an intersection, and it fascinates me that such a concession to bike safety would be made - and work.
Feels like. I dithered between 5 and 10, but although it was a decade ago that I lived there I remember thinking that dutch drivers were remarkably patient. Mind you, traffic laws there are such that motorists are almost automatically at fault if they hit a pedestrian or cyclist, so they have an incentive to be careful.
One more thing: the author of the article, Tom Vanderbilt, wrote a terrific book called "Traffic" and has a excellent blog (at http://www.howwedrive.com/). It was from his blog that I found out about David Hembrow's blog.
I used to think shared lanes were good too, until I was hit (in SF) and my cousin was killed (in London) on the same day.
Good article, but lacks some pictures.
Drachten:
http://maps.google.com/maps?client=opera&q=Drachten&...
Makkinga:
http://maps.google.com/maps?client=opera&q=Makkinga&...
Drachten:
http://maps.google.com/maps?client=opera&q=Drachten&...
Makkinga:
http://maps.google.com/maps?client=opera&q=Makkinga&...
Thanks for the pictures. I can definitely see how the traffic circle in Makkinga is safer than just a through street.
I didn't expect to see a bunch of Japanese labels on the maps of these German towns, though... :)
I didn't expect to see a bunch of Japanese labels on the maps of these German towns, though... :)
Alex Tabarrok linked to a video of a British town's experiment with this on a roundabout: http://www.youtube.com/watch?v=vi0meiActlU&feature=playe... . They saw a 75% decrease in waiting time at the intersection.
Now just do something about auto insurance since everyone doesn't worry about smacking into someone else when they know that their car will be paid for afterward. That will really raise people's alertness.
I can see this working for traffic circles / roundabouts, as long as people know how to use them, and you have them. They're far simpler to use than a regular intersection. They're exceedingly rare in Wisconsin, for example, especially in any residential area that's more than about 20 years old. The cost of replacing intersections with them (especially as they're often bigger, to handle fire trucks, busses, and whatnot) would be astronomical. In the meantime, I'll take stop signs to nothing at intersections. There are too many rude people in the world, and it only takes one to ruin someone else's life.
So local traffic knows local roads?
Signs and indicators are for people who are not from the area.
Signs and indicators are for people who are not from the area.
It's a long article and I didn't read all of it, but the changes they made seem more fundamental than that:
A year after the change, the results of this “extreme makeover” were striking: Not only had congestion decreased in the intersection—buses spent less time waiting to get through, for example—but there were half as many accidents, even though total car traffic was up by a third. Students from a local engineering college who studied the intersection reported that both drivers and, unusually, cyclists were using signals—of the electronic or hand variety—more often. They also found, in surveys, that residents, despite the measurable increase in safety, perceived the place to be more dangerous. This was music to Monderman’s ears. If they had not felt less secure, he said, he “would have changed it immediately.”
That the locals who know the roads perceived the intersection as more dangerous indicates to me there's something more than "people familiar with roads don't need signs."
A year after the change, the results of this “extreme makeover” were striking: Not only had congestion decreased in the intersection—buses spent less time waiting to get through, for example—but there were half as many accidents, even though total car traffic was up by a third. Students from a local engineering college who studied the intersection reported that both drivers and, unusually, cyclists were using signals—of the electronic or hand variety—more often. They also found, in surveys, that residents, despite the measurable increase in safety, perceived the place to be more dangerous. This was music to Monderman’s ears. If they had not felt less secure, he said, he “would have changed it immediately.”
That the locals who know the roads perceived the intersection as more dangerous indicates to me there's something more than "people familiar with roads don't need signs."
So if one would believe in a universal "fair trade" principle, the cost you pay for security and fluidity is heightened attention from the participants in traffic. One would wonder how well would this scale: if drivers had to pay the same attention to every intersection, how long before they would start making mistakes?
So, you mean that the locals can ignore traffic signs because they know the place? Like in "oh, yes, there's a speed limit, but it's useless, see, the street is safe enough, screw that". That could be actual quote, I've heard it so many times. And I guess it's much better to remove such sign and make the street actually look like it's not safe so you have to pay attention.
And of course, routine and automatization are a grave danger for any driver. Especially at the roads where the conditions are constantly changing. Traffic signs would just provide a false sense of security.
And of course, routine and automatization are a grave danger for any driver. Especially at the roads where the conditions are constantly changing. Traffic signs would just provide a false sense of security.
People walking on the edge of a 100' drop tend to walk very carefully. The end result of this sort of reasoning is a steel spike sticking out of the steering wheel in the place where the airbag sits now.
If applying a steel spike sticking out of the steering wheel resulted in a measurable decrease in deaths from traffic accidents, would you support it?
That's a really good question. I think I should, but I find the concept somewhat disconcerting.
There are a fairly large number of single vehicle accidents so I don't think that it really would lead to a measurable decrease though.
Alternatives: everybody should drive cars with bodies made of hardened but very thin glass (I'd support that one).
Maximum weight of a personal car 750 Kg.
There are a fairly large number of single vehicle accidents so I don't think that it really would lead to a measurable decrease though.
Alternatives: everybody should drive cars with bodies made of hardened but very thin glass (I'd support that one).
Maximum weight of a personal car 750 Kg.
Dupont Circle in Washington DC could take a page from this. It's the exact opposite principle: inner and outer circles, with sets of inner and outer traffic and pedestrian lights at each incoming road. At rush hour, it's fortunate if one vehicle can make it through each light. Other traffic circles in the area, in my experience, work much better without such control.
Morro Bay, CA, is a small town that replaced a freeway offramp/frontage road 4-way stop intersection with a roundabout. Transit time through the intersection improved because you no-longer had to stop and wait for your turn. It seems like a good model for low to moderate volume intersections. Flow is increased and speed is reduced.
Roundabouts are the New Big Thing here in Sweden, they're building them everywhere. On average it seems to be an improvement, and I have a local anecdote where a very busy intersection was rebuilt and improved vastly (while "everyone" said that it was a bad idea before it was completed).
Here's a set of videos on YouTube with Monderman on site in Drachten explaining what's been done
http://www.youtube.com/watch?v=Xo3KWHqmDhA&feature=relat...
http://www.youtube.com/watch?v=Xo3KWHqmDhA&feature=relat...
It's true that some are just eye-catchers that distract.
But that bridge sign in the article? It's not for when you CAN see the bridge, it's for all those times in inclement weather when you CAN'T.
But that bridge sign in the article? It's not for when you CAN see the bridge, it's for all those times in inclement weather when you CAN'T.
And are speeding down the road, secure in your knowledge that even though you cant see, if the road turned, there'd be large reflective warnings beforehand?
what a well written and thought provoking article; i love articles on technical topics that feel like art.
Copied from the other thread on this topic.
Vietnam: http://www.youtube.com/watch?v=mr5Gssaxl6g
Ikaruga: http://www.youtube.com/watch?v=i-zSi0xt9fY
Vietnam: http://www.youtube.com/watch?v=mr5Gssaxl6g
Ikaruga: http://www.youtube.com/watch?v=i-zSi0xt9fY
Even the article provides a hint in this direction:
"They also found, in surveys, that residents, despite the measurable increase in safety, perceived the place to be more dangerous. This was music to Monderman’s ears. If they had not felt less secure, he said, he “would have changed it immediately.”
It is possible that once people get used to it, they'll feel more secure -> less alert -> things'll go back to normal.