Hacker News new | past | comments | ask | show | jobs | submit login

When I said that draconian measures were often opposite of what is needed to make roads safer, I was mainly referring to attempts to get people to drive slower on highways. Any attempt to force people to drive slower than the 85th percentile speed poses a danger to others on the road. If you do not believe me, try driving in the left lane at the 55mph speed limit in NYS in areas around NYC. You will undoubtably have many near collisions in just a 5 minute time frame from people cutting you off. Now, imagine a number of people being coerced to drive like that all the time while the rest do not. It is easy to see that there will be more collisions.

Doing draconian things that are ineffective has an opportunity cost that requires diverting time and effort from doing things that actually make a difference. Data on traffic cameras improving safety is mixed (mainly because of people flooring their brakes to avoid fines only to cause themselves to be rear ended). It also does not help that a number of places actually try to cause motorists to run red lights when traffic cameras are put into place by decreasing the amount of time used for yellow lights so that they can increase revenue. Interestingly, increasing the time spent with the light yellow decreases collisions at intersections and unlike traffic cameras, always has a positive improvement on collisions. Another option would be to eliminate intersections by adopting cloverleaf and/or diamond interchanges, which not only make traffic flows more efficient, but also improve safety (since you cannot run a red light, or have a collision caused by someone suddenly stopping upon seeing a yellow light). As for speed limiters, they are outright dangerous when they restrict people to speeds well below the 85th percentile.

That said, NYS has done a number of things to improve driver safety. They probably do not do as much as they would in less densely populated areas, but it is not hard to imagine other U.S. states adopting them. The real problem is that there are so many cars on the road that the average distances between them are very small. As long as the distance between vehicles remains non-zero, collisions are avoided. Efforts really should be focused on maximizing the distance between vehicles, rather than on minimizing the speed at which they travel. Raising the speed limits to the 85th percentile would help there. Car pooling lanes would also help. Modernizing public transport so that people do not need as many cars would also help.






> try driving in the left lane at the 55mph speed limit in NYS in areas around NYC

Why the left lane? I used to live in NYC and when traffic was light enough to actually drive the speed limit I generally stuck to the right lane to avoid inconveniencing other people who wanted to drive faster. Wouldn't people with court-mandated speed limiters simply stick to the right lane?

> Any attempt to force people to drive slower than the 85th percentile speed poses a danger to others on the road.

Perhaps they'd need the "ATTENTION: This vehicle's speed is monitored by GPS" stickers that I occasionally see on fleet trucks in the city.

Regardless, this seems to be an objection to selective enforcement methods, right? For example, if all drivers are subject to the same constraints (whether speed cameras, universal speed limiters, road diet, etc) then not only would this not increase discrepancies in speed, but it would likely decrease them. Does this mean you would support such measures?

> Doing draconian things that are ineffective has an opportunity cost

Agreed. But that applies to ineffective things whether or not they are draconian.

I do agree that stoplight cameras have mixed results -- typically reducing serious T-bone collisions while increasing rear-end collisions -- but to be clear, when I said "cameras" I was referring to speed cameras, not red-light cameras. On the topic of NYS, their school zone speed cameras seem to have been effective at reducing injuries caused by drivers in school zones.

> As long as the distance between vehicles remains non-zero, collisions are avoided. Efforts really should be focused on maximizing the distance between vehicles, rather than on minimizing the speed at which they travel.

Due to human reaction times and the physical limitations of braking, maintaining a time between vehicles is more relevant than distance. That's why defensive driving courses teach you to keep about three seconds between your car and the car in front of you, more if you're hauling a trailer. If you're driving at 8mph that's only about 10ft, but at 80mph that's 100ft. So while I agree that increasing distance is useful, the distance necessary to ensure safe operation is a direct function of speed.

I heard of a few cases in which so many drivers were already breaking the speed limit on an overengineered highway that raising the speed limit did not increase collisions e.g. in Michigan, Texas, British Columbia. But otherwise the data is pretty clear that making people drive more slowly improves safety.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: