Roundabouts: The widening gyre (2013) | Hacker News Reader

Roundabouts: The widening gyre (2013)

economist.com

21 pointsrfreytag16 comments

I used to work near here:

https://www.google.com/maps/search/8th+and+townsend,+san+fra...

And people would regularly go the wrong way around the damn thing.

Hmm, no surprise - its missing any form of signage beyond stop signs. Despite being so much more common, you still get clear signifiers in the UK. e.g. you get "to the left please sir" blue and white direction signs on approach for all round-a-bouts:

https://goo.gl/maps/nRtNC

And for more major intersections, you get massive "go left you muppet" black and white chevron signs as well:

https://goo.gl/maps/0KXoM

There is obvious ambiguity between curvacious T-junctions and roundabouts so you really need some "go right you dude" and "I said right jerk" signage especially given how less common they are.

The other problem is, crossing with light signals are even worse than roundabouts.

Well, cars are the worst invention of our time but that for another discussion.

Interestingly that crazy looking magic roundabout serves a purpose. The (seemingly correct) theory was that such a setup would slow people down and make them cautious, thereby dispensing with the need for traffic lights and allowing safe self-organisation.

The UK is replacing many of its roundabouts with traffic signals in order to control the flow where one route across the junction dominates traffic density.

Simple domination doesn't seem like an adequate explanation. Is it the case that the dominant traffic pattern effectively monopolizes the roundabout making it difficult for other traffic to enter for significant periods of time?

Yes. It's also an issue of the traffic traversing the roundabout at great speed, making it very difficult to join.

I don't understand why they don't add traffic signals to the problem roundabouts, operated when required. It would seem less disruptive.

That sounds like a poorly designed roundabout to me. One recently added to my relatively small city has a speed limit of 20 MPH (yes, I'm in the US) and much more than 20 MPH seems too fast to me simply as a result of the radius of the roundabout.

It is routes such as ring roads where effectively it is a crossroads with one of the crosses being 4-6 lanes wide at 40mph and the other being 2 lanes approaching at 30mph, (though not always such a speed disparity).

So if you arrive from the 2 lane 30mph road you could have a very long wait time.

One of the issues with roundabouts here (France) is that almost nobody is using their turn signals or use the wrong one (they activate the left one entering the roundabout and keep it on, making you think they'll keep on turning around when they're about to exit the roundabout.) When you're waiting to enter a roundabout it makes you wait a bit longer when a car is already turning around.

Signalling incorrectly on roundabouts is pretty common in the UK. In fact, I would say only a minority of drivers signal correctly when turning right at a roundabout.

I used to get pretty frustrated by it. I can't do anything about other people's bad driving habits.

I now try and pay more attention to the speed at which they approach the roundabout, and their lane positioning/angle. I probably take more "chances" because I drive an auto with a bit of grunt.

This article talks about gridlock but somehow fails to address throughput, which is supposed to be the upside over the conventional 4-way stop.

http://nextstl.com/2013/10/mythbusters-tackles-four-way-stop...

Reply on news.ycombinator.com