Wired Magazine wrote of Hans Monderman, [He] is one of the leaders of a new breed of traffic engineer - equal parts urban designer, social scientist, civil engineer, and psychologist. The approach is radically counterintuitive: Build roads that seem dangerous, and they'll be safer. Donald Norman of "The Design of Everyday Things" coined the phrase, a sign is a sign of a bad design.
A roadway, like any engineered artifact (say a computer system) should be intuitive to the user and afford the user with obvious indications of it's proper use. When key elements of the design have to be explained then obviously, less is obvious. There are two different ways of expressing this concept; the first is that people shouldn't need signs / directions to properly navigate, understand, or use your solutions. The more signs or documentation you need, the less well-designed is the product.
Another way (and the way I prefer to express this) says that people have brains, and the more you engage the brain correctly, the less external information is needed to properly convey the proper use of a solution. In Matthew May's book, In pursuit of Elegance he states that there is a counterintuitive element of nature at work in many systems. As we add documentation such as "Slippery when wet", "Children at Play", "Deer Crossing" and so forth, we actually cause the brain to disengage from risk analysis.
In December of 1995, the State of Montana removed the speed limit signs from its state highways and instead posted notifications to drive at speeds which were reasonable and prudent. The State Police were not particularly enthralled with this move and continued to cite drivers who exceeded 80-90 miles per hour. For the next five years Montana recorded its lowest traffic related fatality rates in twenty five years. Coincidence? When one of the speeding tickets was challenged, the Supreme Court of Montana ruled that "reasonable and prudent" was unconstitutional, and in 2000 the state went back to posted speed limits. In the next year, road fatalities in Montana rose 111%, and hit all times highs the following two years.
Why would this be? Could it be that when you are solely responsible for the speed at which you are traveling, you take extra precautions, stay more engaged, and use your brain more fully? In the case of the Montana highway speed control, offering less instruction not only enabled faster travel, but safer travel as well.
Hans Monderman takes this same approach to designing intersections. He removes traffic signals, and many of the cautionary signs. He replaces them with well-designed roundabouts. I have traveled through New England where they have used Rotary intersections (Americanized roundabouts), and have always approached them with trepidation. Only recently did I begin to realize that my apprehension was a direct result of having to engage the brain more fully - I was now completely responsible for navigating the intersection. Here is a time lapsed video of a busy roundabout. Note the elegance of the traffic flow.
Let me emphasize that neither Monderman nor I am advocating the indiscriminate removal of signs, directions, or documentation. Rather, the suggestion is that well-designed solutions don't need them. If a solution is in need of placards, signs, and documentation, then maybe the solution isn't so well designed.