3 ms·

This article is all well and good, but the BIG problem are traffic lights. Traffic lights are undoubtedly the problem. What can replace lights? Roundabouts.

This article is all well and good, but the BIG problem are traffic lights. Traffic lights are undoubtedly the problem. What can replace lights? Roundabouts. Plain and simple OR a startup that can compete with the current traffic light system.

rewqfdsa 11y ago
Are you trolling? A startup for traffic lights? Versus actual civil engineers who aren't idiots and who have probably fully explored the solution space of traffic handling arrangements? Traffic is a political problem, not an engineering problem.
- dajohnson89 11y ago
  Unfortunately civil engineers are hired/contracted by municipal governments. They^H^H^H^H Their employers are usually clueless and/or apathetic about really solving problems. After all, the taxes keep rolling in regardless.
stephengillie 11y ago
Having seen some roundabout implementations, I'm no longer convinced. Traffic circles require a large flat space, and can't easily replace stop lights on steep hills or narrow crevasses. A roundabout replaced a set of traffic lights at a 3-way intersection in a crevasse next to SR-18 in Auburn, WA. It's a tight fit and difficult to navigate, and trucks regularly jump the curb. After driving it, I really wish they could change it back.
- Animats 11y ago
  Roundabouts may work very well for self-driving cars with some short-range communications capability. Merges can be negotiated, both at roundabouts and on-ramps.
  - stephengillie 11y ago
    Roundabouts allow only simple traffic interactions, that's kinda the point. Nothing complicated like a left turn can happen. Why would special communication be needed?
Piskvorrr 11y ago
Living in a city which has both hi-tech traffic lights (dynamically selected phases), dumb traffic lights (two alternating phases, one for each direction) and low-tech roundabouts (give way on entry), I doubt any of these is a catch-all solution. There are roundabouts which would be better off as traffic lights, uncontrolled intersections where roundabouts would help, and traffic lights which are actively creating problems. For example, I regularly drive across a four-way roundabout; there are strong, almost continuous traffic streams coming from the north and west entrances, exiting mostly east and north. This means that while traffic from the west still mostly manages to merge, traffic from the south is regularly backed up, as there are only a few "slots" left in the roundabout traffic flow where it could merge. This is a spot that would be better served with traffic lights. So, I don't think there's one "The Problem" in traffic engineering, generally solvable.
serge2k 11y ago
> a startup that can compete with the current traffic light system. The startup. HNs cause of, and solution to, all of life's problems.

rewqfdsa 11y ago

Are you trolling? A startup for traffic lights? Versus actual civil engineers who aren't idiots and who have probably fully explored the solution space of traffic handling arrangements? Traffic is a political problem, not an engineering problem.

dajohnson89 11y ago
Unfortunately civil engineers are hired/contracted by municipal governments. They^H^H^H^H Their employers are usually clueless and/or apathetic about really solving problems. After all, the taxes keep rolling in regardless.

stephengillie 11y ago

Having seen some roundabout implementations, I'm no longer convinced. Traffic circles require a large flat space, and can't easily replace stop lights on steep hills or narrow crevasses. A roundabout replaced a set of traffic lights at a 3-way intersection in a crevasse next to SR-18 in Auburn, WA. It's a tight fit and difficult to navigate, and trucks regularly jump the curb. After driving it, I really wish they could change it back.

Animats 11y ago
Roundabouts may work very well for self-driving cars with some short-range communications capability. Merges can be negotiated, both at roundabouts and on-ramps.
- stephengillie 11y ago
  Roundabouts allow only simple traffic interactions, that's kinda the point. Nothing complicated like a left turn can happen. Why would special communication be needed?

Piskvorrr 11y ago

Living in a city which has both hi-tech traffic lights (dynamically selected phases), dumb traffic lights (two alternating phases, one for each direction) and low-tech roundabouts (give way on entry), I doubt any of these is a catch-all solution. There are roundabouts which would be better off as traffic lights, uncontrolled intersections where roundabouts would help, and traffic lights which are actively creating problems. For example, I regularly drive across a four-way roundabout; there are strong, almost continuous traffic streams coming from the north and west entrances, exiting mostly east and north. This means that while traffic from the west still mostly manages to merge, traffic from the south is regularly backed up, as there are only a few "slots" left in the roundabout traffic flow where it could merge. This is a spot that would be better served with traffic lights. So, I don't think there's one "The Problem" in traffic engineering, generally solvable.

serge2k 11y ago

> a startup that can compete with the current traffic light system. The startup. HNs cause of, and solution to, all of life's problems.