So the question remains: what is the point of a street? To answer that, you have to ask the larger question: what is the point of a city, the entire infrastructural network?
I don't even know where this dates back to, but I don't think anybody can argue with the answer: that cities (and their subservient subset of streets) are the platform for social and economic exchange.
Traffic planners point to speed as "efficiency." Which is odd, since all data points to their work being reactive to the inertia of their own creation. The recent Univ of Toronto study pinpoints this failing: for every doubling of lane capacity, Vehicle Miles Travelled (VMTs) double as well. There is a one to one ratio that emerges over time. You're driving further because you're moving further away.
Either the bigger roads made it insufferable to live near them (completely rational response) or they made it convenient to live further away (also rational). So if you believe that we are mostly rational actors within a system, traffic planning, which always suggests roads must be widened (to be "improved" - their term, their metrics), this points to the demand model of where people choose to locate further and further away from central core areas where real efficiency exists, the synergy of proximity, that Richard Florida and Ed Glaeser have recently been pointing to. People in proximity are more productive.
Furthermore, it is hardly efficient to be travelling long distances to do daily tasks energy and cost wise, for individuals as well as the public bodies spending on all that excess infrastructure. You begin to see why everyone is broke but those positioned to profit off of dysfunctional, disconnected systems. Oil companies and those facilitating "virtual" connections as a de facto replacement for legitimate interactions: Google, Apple, eBay, etc.
So what IS an improved road network? The last few posts have been examining some of those metrics where commercial opportunities are maximized through the creation and manipulation of proper networks. Now would be a good time to recall the work of Donald Appleyard, a transportation planner tragically and ironically killed crossing a street by an onrushing car.
Appleyard measured the amount of close social contacts, friends and acquaintances along certain streets in San Francisco, ie Social Exchange. As you can see in the diagram above, streets designed to move as many cars as possible have many fewer social contacts than those streets with calmer vehicular traffic, 3x more friends per person and 2x more acquaintances. Furthermore, you can visualize the reluctance to cross the street, as the diagram shows how much more "tethered" the two sides are on less trafficked streets. What would be interesting is the number of pedestrians per day on each street and add this to the total "traffic" number. They very well might be even.
However, it should be noted that he measured this in the heart of San Francisco neighborhoods, in densely interconnected grid streets. A functional network. This doesn't apply to a cul-de-sac where there is virtually no traffic and you may not know anybody at all. There is too much noise in dysfunctional systems. No order emerges. And that is half the point.