4 ms·

We drove through this last week. It's not done yet but it looks very confusing. If they were everywhere people would get used to it but with just one I think it

We drove through this last week. It's not done yet but it looks very confusing. If they were everywhere people would get used to it but with just one I think it may cause problems. And they can't be too common because they are absolutely huge compared to a regular intersection.

yurishimo 3y ago
They are bigger but when people learn how to use them (only takes two or three encounters) then they massively increase throughput. Here in the Netherlands, a roundabout is always the first choice and only if space is a limiting factor is a light used. But nowadays, the lights are really smart and can sometimes change to green for a single approaching car if there is space.
- andrepd 3y ago
  Also, roundabouts (as well as many city streets) have only 1 lane, and relatively narrow at that.
- pi-e-sigma 3y ago
  Which is ironic because roundabouts were initially sold as solutions making road intersections more safe, not as means of increasing throughput. The reduction of crashed and injuries comes from decreased speed of cars. So obviously it's at a cost of throughput
  - andrewflnr 3y ago
    It's only obvious if you're confusing throughput with latency. Also, you're comparing with stoplights, a system where you often stop dead rather than slow down. So it's obvious to me that increasing safety and througput at the same time is very plausible.
    - pi-e-sigma 3y ago
      That's indeed so obvious that in Europe there is a new trend of reverting roundabouts back to standard intersections. Because in reality roundabouts slow everything down to unbearable speed. EDIT: you make a mistake assuming that just because traffic going through a roundabout doesn't completely stop that it automatically means that the throughput is higher than a stop-and-go traffic through normal intersection. And also assume that traffic going through a roundabout never stops, which is simply not lined up with the observable reality
      - andrewflnr 3y ago
        I'm not making any such assumptions. You're the one making bold claims about "obvious" causality, while I'm just saying it's not really obvious either way (and if you think that's a lie, check my phrasing more carefully).
        pi-e-sigma 3y ago
        Your initial claim: >So it's obvious to me that increasing safety and througput at the same time is very plausible. Your current claim: >I'm not making any such assumptions If you don't make any assumption how can you even say that something is 'very plausible'? Based on what exactly then? Gut feeling? If so then it's not 'plausible', it's just your gut feeling dressed in nice wording
        andrewflnr 3y ago
        > check my phrasing more carefully "plausible" is not nearly as definite as "safety obviously comes at a cost of throughput", and is furthermore not a statement about causality, but a claim that improving safety and througput don't immediately rule each other out; a contradiction needs a strong proof, since so many things that appear contradictory turn out to be compatible with enough imagination. Absent that proof, compatibility is plausible. But sure, maybe I should have put air quotes around "obvious", since there's no real obvious correct conclusions here. "Just your gut feeling dressed in nice wording" is another great way to describe your initial claim about roundabouts, too, which was my main point to start with.
  - yurishimo 3y ago
    Both can be true. If you have cars coming from all directions that can easily merge with each other, it makes sense that throughput would also be increased since the other 3 directions don't need to stop to make a left turn from any direction. Safety might be factor number 1, but a roundabout will sustain higher traffic loads than a traffic light, assuming that all directions are being fed with equal amounts of vehicles. For this reason, a lot of arterial roads in NL still use traffic lights because they prioritize traffic moving in a single dimension (x or y, but not both). The smaller intersecting roads need to wait on the light during rush hour since the priority is traffic on the larger road. When two roads of a similar size/traffic mix, a roundabout can be used to facilitate changing directions. Here is an example near Eindhoven in the south part of the country: https://maps.app.goo.gl/SYMnP9jwNEj1ytx2A https://maps.app.goo.gl/SYMnP9jwNEj1ytx2A There's also a fairly large roundabout about 1-2km north of that intersection if you're interested. It's increased size provides more throughput and slows down drivers coming from the highway and allows a smooth transition down to the lower speed allowed within the city limits.
coolhand2120 3y ago
According to the article this was a statistical outlier for a specific type of collision, likely because of the long flat area around the intersection. It's better to have a confused person drive the loop and have to circle back because they went the wrong way vs. initiative driver on a long haul zone out into intersecting traffic. Worse case they slam on their breaks for a rear end collision, which sounds less severe. I guess we shall see if it improves things, seems like they can't get much worse the way it is now.
- modeless 3y ago
  Rear end collisions were the largest category of collision at the previous intersection, according to the local paper report I saw. I expect that people will try to jump the curbs after they realize they're in the wrong lane. Some people will sadly do anything to avoid a wrong turn. But hopefully the collisions will be at lower speed. I'm sure locals will get used to it but I think a big part of the problem with this intersection is that it's used by a lot of people who pass through very infrequently.
pkulak 3y ago
Driving is so dangerous and difficult that we have an entire licensing system set up, but we’ve also made it necessary to live, so we can’t deny anyone a license and we can’t do anything that requires drivers to learn anything new, because maybe they won’t feel like it. Wonderful society we’ve built here.

yurishimo 3y ago

They are bigger but when people learn how to use them (only takes two or three encounters) then they massively increase throughput. Here in the Netherlands, a roundabout is always the first choice and only if space is a limiting factor is a light used. But nowadays, the lights are really smart and can sometimes change to green for a single approaching car if there is space.

andrepd 3y ago
Also, roundabouts (as well as many city streets) have only 1 lane, and relatively narrow at that.
pi-e-sigma 3y ago
Which is ironic because roundabouts were initially sold as solutions making road intersections more safe, not as means of increasing throughput. The reduction of crashed and injuries comes from decreased speed of cars. So obviously it's at a cost of throughput
- andrewflnr 3y ago
  It's only obvious if you're confusing throughput with latency. Also, you're comparing with stoplights, a system where you often stop dead rather than slow down. So it's obvious to me that increasing safety and througput at the same time is very plausible.
  - pi-e-sigma 3y ago
    That's indeed so obvious that in Europe there is a new trend of reverting roundabouts back to standard intersections. Because in reality roundabouts slow everything down to unbearable speed. EDIT: you make a mistake assuming that just because traffic going through a roundabout doesn't completely stop that it automatically means that the throughput is higher than a stop-and-go traffic through normal intersection. And also assume that traffic going through a roundabout never stops, which is simply not lined up with the observable reality
    - andrewflnr 3y ago
      I'm not making any such assumptions. You're the one making bold claims about "obvious" causality, while I'm just saying it's not really obvious either way (and if you think that's a lie, check my phrasing more carefully).
      - pi-e-sigma 3y ago
        Your initial claim: >So it's obvious to me that increasing safety and througput at the same time is very plausible. Your current claim: >I'm not making any such assumptions If you don't make any assumption how can you even say that something is 'very plausible'? Based on what exactly then? Gut feeling? If so then it's not 'plausible', it's just your gut feeling dressed in nice wording
        andrewflnr 3y ago
        > check my phrasing more carefully "plausible" is not nearly as definite as "safety obviously comes at a cost of throughput", and is furthermore not a statement about causality, but a claim that improving safety and througput don't immediately rule each other out; a contradiction needs a strong proof, since so many things that appear contradictory turn out to be compatible with enough imagination. Absent that proof, compatibility is plausible. But sure, maybe I should have put air quotes around "obvious", since there's no real obvious correct conclusions here. "Just your gut feeling dressed in nice wording" is another great way to describe your initial claim about roundabouts, too, which was my main point to start with.
- yurishimo 3y ago
  Both can be true. If you have cars coming from all directions that can easily merge with each other, it makes sense that throughput would also be increased since the other 3 directions don't need to stop to make a left turn from any direction. Safety might be factor number 1, but a roundabout will sustain higher traffic loads than a traffic light, assuming that all directions are being fed with equal amounts of vehicles. For this reason, a lot of arterial roads in NL still use traffic lights because they prioritize traffic moving in a single dimension (x or y, but not both). The smaller intersecting roads need to wait on the light during rush hour since the priority is traffic on the larger road. When two roads of a similar size/traffic mix, a roundabout can be used to facilitate changing directions. Here is an example near Eindhoven in the south part of the country: https://maps.app.goo.gl/SYMnP9jwNEj1ytx2A https://maps.app.goo.gl/SYMnP9jwNEj1ytx2A There's also a fairly large roundabout about 1-2km north of that intersection if you're interested. It's increased size provides more throughput and slows down drivers coming from the highway and allows a smooth transition down to the lower speed allowed within the city limits.

coolhand2120 3y ago

According to the article this was a statistical outlier for a specific type of collision, likely because of the long flat area around the intersection. It's better to have a confused person drive the loop and have to circle back because they went the wrong way vs. initiative driver on a long haul zone out into intersecting traffic. Worse case they slam on their breaks for a rear end collision, which sounds less severe. I guess we shall see if it improves things, seems like they can't get much worse the way it is now.

modeless 3y ago
Rear end collisions were the largest category of collision at the previous intersection, according to the local paper report I saw. I expect that people will try to jump the curbs after they realize they're in the wrong lane. Some people will sadly do anything to avoid a wrong turn. But hopefully the collisions will be at lower speed. I'm sure locals will get used to it but I think a big part of the problem with this intersection is that it's used by a lot of people who pass through very infrequently.

pkulak 3y ago

Driving is so dangerous and difficult that we have an entire licensing system set up, but we’ve also made it necessary to live, so we can’t deny anyone a license and we can’t do anything that requires drivers to learn anything new, because maybe they won’t feel like it. Wonderful society we’ve built here.