Looking around here in the winter, we could make some major changes to how roads are laid out as well. For example, most of the freeways are just one lane during and right after a heavy snowfall. The cars are also traveling tens of miles an hour under the speed limit. I've even seen two-lane roads become one-lane roads, because people only want to drive where everyone else has proven it's safe to drive.

Driving in the snow is much more dangerous than driving in perfect weather. It's only natural that people are going to change their driving habits. This article has a good premise, but poor execution.

"This article has a good premise, but poor execution."

Makes sense, but I take the view the premise is flawed. Its one of those "oh, that seems so obvious" type premises that describe the skin of the onion but fails on pulling back the layers. It seems to happen a lot while dealing with complex systems. The fun part of this one is that it seems like they don't realize they are dealing with a complex system. The illustrating pictures even show some cracks and the author skims over them.

If the premise is "we can learn how traffic flows from watching traffic patterns" and that's accomplished by watching where the tire marks go, then it seems pretty good to me. However, watching where the snow disappears first from the roadway isn't an accurate measure, because people drive differently in the snow.

Another example based on the premise in the article: you can find the best racing line on a race track by looking at where the track is darker, because rubber and any leaking fluids will stain the track directly under where the cars are always driving. That's where racers will be driving during normal conditions, but it'd be inaccurate to say that the race track should only be exactly that line, because drivers will occasionally go off that line to pass, avoid an accident, or participate in an accident. In both racing and normal driving, the best line isn't always the only line.

That's what I meant by the premise is good, but the execution is bad. It's just a faulty measurement.

Don't get me wrong, I think your reasoning is sound and I think I could argue your view with pretty good confidence. Your example is quite good and I like the idea of taking the totally opposite driving circumstances to prove the point.

but.....

I took the premise as "there is a lot of wasted space on the road and we can reduce it" with a methodology of "removed snow will show us what is not wasted". So, I thought the premise was hogwash, but the execution fit their twisted premise.

i think that's exactly the point. They refer to their suggestions as "traffic calming" measures.

So "traffic calming" = congestion, frustration, and dangerous driving for bikes and cars. Not a great idea in my book to spend more money (and have a higher continued upkeep cost) to make the roads more dangerous and less useful. I suppose if they put an island there then they understand the snow is going to get piled on it and then expand into the rest of the road, right?

Where I live they recently installed some new intersections that are now sharp corners and they're now really dangerous for everyone. You need to go so slow to make the turns that nearly everyday I see accidents that are narrowly avoided. They don't "calm traffic" instead they just make it congested which means it's dangerous for people in cars, bicyclists, and pedestrians.

Have you ever seen a big European intersections? Planning one with the well-being of pedestrians in mind causes zero problems for vehicles, see [1].

[1] http://commons.wikimedia.org/wiki/File:Stra%C3%9Fenkreuzung_...

They aren't doing this here, they are using a flawed methodology to (imagine) change a road.

Fair enough. Pedestrian-friendly remodeling of intersections is also being done here in Europe, but AFAIK always by widening up the entire intersection and sacrificing parking space. This is IMO a sensible approach, since parked cars are dead space and thus should be absent from a busy intersection.