4 ms·

First, get past the obvious benefit of a seemingly unfair advantage for avoiding traffic tickets, such as reduced costs due to traffic fines or insurance premiu

by jcr 12y ago

First, get past the obvious benefit of a seemingly unfair advantage for avoiding traffic tickets, such as reduced costs due to traffic fines or insurance premiums. Then get past the obvious harm of a seemingly unfair advantage for avoiding traffic tickets, such as reduced freedom to drive "too fast" and reduced revenues for law enforcement budgets [1]. The end result of tech like this (including the ultimate step of speed governors preventing speeding) really seems to be all of us sharing safer roads, and fewer of us getting hurt or dying in traffic accidents. This seems pretty good at first sight, but unfortunately, it leaves open one problematic question; if everyone drove at the speed limit (and fully stopped at stop signs/lights --feasible to enforce with tech), would there be fewer accidents? Also, I wonder if there would be more or less traffic congestion? [1] http://www.networkworld.com/article/2226966/opensource-subnet/driverless-cars-could-cripple-law-enforcement-budgets.html http://www.networkworld.com/article/2226966/opensource-subne...

hcarvalhoalves 12y ago
> This seems pretty good at first sight, but unfortunately, it leaves open one problematic question; if everyone drove at the speed limit (and fully stopped at stop signs/lights --feasible to enforce with tech), would there be fewer accidents? Yes. Why is that a problematic question in the first place? It's physics.
- AnthonyMouse 12y ago
  > Why is that a problematic question in the first place? It's physics. It's not that simple. Example: Just before rush hour a significant number of cars get on the highway. If they travel at a lower rate of speed, fewer of them will have reached their destination as most businesses start to let out, which means there will be more cars on the highway rather than at their destination and therefore more traffic congestion. And traffic congestion is a primary cause of highway collisions.
manicdee 12y ago
Ways to reduce congestion without altering roads, signs, traffic light cycles, or work hours: 1) Everyone drives slightly below the speed limit 2) Pick a lane and stick to it 3) Merge late
- prawn 12y ago
  There's research showing that almost whatever you do (short of charging), congestion will remain the same. More roads makes for more drivers. More public transport takes cars off the road? Others will replace them due to the increased convenience.
  - AnthonyMouse 12y ago
    > There's research showing that almost whatever you do (short of charging), congestion will remain the same. More roads makes for more drivers. More public transport takes cars off the road? Others will replace them due to the increased convenience. That's clearly incorrect. Cars aren't going to magically appear in North Dakota just because you put in a twelve lane superhighway to nowhere. What the data in question is actually explaining is that traffic congestion reduces traffic. When you remove the cause of the traffic congestion you get more traffic. But this isn't Zeno's Paradox. There is a finite number of lanes capable of handling all the traffic you would see when there is no traffic congestion.
    - prawn 12y ago
      Obviously in extreme cases (30 lane freeways everywhere), it doesn't follow, but I was speaking of realistic changes.
softgrow 12y ago
The odd thing is that once the automated/autonomous vehicles reach above the 10% level the consensus is (cant find a stellar ref but http://www.usfav.com/publications/TAVI_8-CapacityPinjari.pdf http://www.usfav.com/publications/TAVI_8-CapacityPinjari.pdf will do for a start) that we'll see marked improvements beyond 10% expected in - less traffic congestion and increased roadway capacity beyond what can be obtained with human drivers - 28% reduction in fuel usage on the whole fleet - marked reduction in crashes beyond a 10% reduction expected
- jcr 12y ago
  Thanks for the link to that white paper. I've skimmed the intro, and then skipped to the end for their conclusions. I'll read the whole thing later tonight, but as expected, the conclusion states itself to be speculative, without supporting data, and in need of more research. From the pdf: > "many experts speculate that the overall benefits outweigh the potential negative externalities." > "Additional research is necessary on a number of aspects related to AVs, including ..." Like everyone, I really want to believe having more automated and autonomous vehicles on the road will result in fewer accidents, reduced congestion, and other benefits, but all we really have at the moment is (seemingly well reasoned) speculation since we lack supporting data. Until we have supporting data from repeatable experiments, whether the speculated benefits will both materialize outweigh the negative impacts should remain open questions. The open questions will get answered, eventually, but it will take a few decades and buying into the all the speculated hype at this early and unproven stage is harmful. I guess I'd rather be cautiously optimistic until we have more useful and repeatable results.

by jcr 12y ago

hcarvalhoalves 12y ago

> This seems pretty good at first sight, but unfortunately, it leaves open one problematic question; if everyone drove at the speed limit (and fully stopped at stop signs/lights --feasible to enforce with tech), would there be fewer accidents? Yes. Why is that a problematic question in the first place? It's physics.

AnthonyMouse 12y ago
> Why is that a problematic question in the first place? It's physics. It's not that simple. Example: Just before rush hour a significant number of cars get on the highway. If they travel at a lower rate of speed, fewer of them will have reached their destination as most businesses start to let out, which means there will be more cars on the highway rather than at their destination and therefore more traffic congestion. And traffic congestion is a primary cause of highway collisions.

manicdee 12y ago

Ways to reduce congestion without altering roads, signs, traffic light cycles, or work hours: 1) Everyone drives slightly below the speed limit 2) Pick a lane and stick to it 3) Merge late

prawn 12y ago
There's research showing that almost whatever you do (short of charging), congestion will remain the same. More roads makes for more drivers. More public transport takes cars off the road? Others will replace them due to the increased convenience.
- AnthonyMouse 12y ago
  > There's research showing that almost whatever you do (short of charging), congestion will remain the same. More roads makes for more drivers. More public transport takes cars off the road? Others will replace them due to the increased convenience. That's clearly incorrect. Cars aren't going to magically appear in North Dakota just because you put in a twelve lane superhighway to nowhere. What the data in question is actually explaining is that traffic congestion reduces traffic. When you remove the cause of the traffic congestion you get more traffic. But this isn't Zeno's Paradox. There is a finite number of lanes capable of handling all the traffic you would see when there is no traffic congestion.
  - prawn 12y ago
    Obviously in extreme cases (30 lane freeways everywhere), it doesn't follow, but I was speaking of realistic changes.

softgrow 12y ago

The odd thing is that once the automated/autonomous vehicles reach above the 10% level the consensus is (cant find a stellar ref but http://www.usfav.com/publications/TAVI_8-CapacityPinjari.pdf http://www.usfav.com/publications/TAVI_8-CapacityPinjari.pdf will do for a start) that we'll see marked improvements beyond 10% expected in - less traffic congestion and increased roadway capacity beyond what can be obtained with human drivers - 28% reduction in fuel usage on the whole fleet - marked reduction in crashes beyond a 10% reduction expected

jcr 12y ago
Thanks for the link to that white paper. I've skimmed the intro, and then skipped to the end for their conclusions. I'll read the whole thing later tonight, but as expected, the conclusion states itself to be speculative, without supporting data, and in need of more research. From the pdf: > "many experts speculate that the overall benefits outweigh the potential negative externalities." > "Additional research is necessary on a number of aspects related to AVs, including ..." Like everyone, I really want to believe having more automated and autonomous vehicles on the road will result in fewer accidents, reduced congestion, and other benefits, but all we really have at the moment is (seemingly well reasoned) speculation since we lack supporting data. Until we have supporting data from repeatable experiments, whether the speculated benefits will both materialize outweigh the negative impacts should remain open questions. The open questions will get answered, eventually, but it will take a few decades and buying into the all the speculated hype at this early and unproven stage is harmful. I guess I'd rather be cautiously optimistic until we have more useful and repeatable results.