It's easy to blame people. Thing is, it's a natural state of traffic flow for jams to occur - it doesn't necessarily fall to human error or impatience. You can have gridlock with as few as four cars, in low traffic volume, with each car following correct road procedures: Please Register or Log in to view the hidden image! Each of these cars is proceeding as considerately as reasonable. Each leading driver stops to make their left turn, and each trailing car stops behind. Notice that the road ahead is clear - it is low traffic volume. The drivers are on a clear road, and thus the trailing drivers have no reason to expect they are about block a side street (like they might expect in high volume, slow traffic). This can happen before anyone has time to realize it will result in gridlock. Sure, it's a simple example, and relatively easily to sort out with time, gear-shifting and space (as long as there's enough of all three), but the principle remains - it is not necessarily about people being rude, impatient or neglectful.
Dave: Still nothing that a bit of strategizing doesn't solve. There are times when one shouldn't mechanically follow traffic rules. My friends and I have an inside joke: "Written on a tombstone: Here lies John Smith who had right of way."
It's not about blaming people. If traffic jams and collisions are seen as something that is usually caused by the drivers (as opposed to objective givens), then we can reasonably hope to reduce them and drive accordingly. But if we see traffic jams and collisions as usually caused by powers outside of human control, that is a hopeless outlook. "The speed was too high, this is why the collision happened. It had nothing to do with the driver."
I'm confused. You said it's not about blaming people, then you said it's (usually) caused by the drivers. It's a self-defeating outlook if you automatically blame the wrong cause. Faulty anology. As my example shows, the drivers did nothing wrong. Sure, lots of jams occur due to people. I'm just counter-pointing the self-reinforcing view that this kind of image - and its resulting opinions - engender. The OP asked why does this happen? It is not always, or only, about people.
Automating driving does not change any of the basics of traffic jams; if you have more traffic than the road can accommodate, you slow down, period. You would simply move from human traffic jams to computer-assisted traffic jams. (If you doubt this, ask yourself if Internet congestion has ever been a thing.)
Blaming and determining causes are not the same thing. Of course. It wasn't an analogy, it was an example of the hopeless outlook outlined above, about traffic jams and collisions being caused by powers outside of human control. People differ in what they think can reasonably be expected from drivers. When I went to driving school, we were taught to think at least one car ahead, one car behind, one car left, one car right; and that one has to be prepared to improvise if necessary. Have you ever been to Istanbul? Driving there is ... an adventure, to say at least. Officially, they do have the same traffic rules as in Europe, but they are far from driving in line with them. Basically, they tend to just drive. You have to be a very skilled driver to be able to drive there safely. http://istanbul.usconsulate.gov/driver_safety_briefing2.html I didn't go to driving school in Turkey, but I kind of wish I did. I'm sure I'd be an even better driver. I never said it always was about people. But for the most part, it seems to be.
Whle I grant they are not synonymous, I think they may be in this situation, since it comes down to what actions we take to alleviate it. If we were to think the cause is people, when in reality, it's not (or at least not entirely), we would likely form faulty or less-than-effective remediations.
Yes, just like people will always need to get around even in their automated cars. The point though, is that all the traffic routing on the Webernets is automatic. Thus automating a process does not mean congestion is eliminated. You wouldn't fix the internet by limiting what _people_ can click on would you?
That's not a fair way to put it! Internet congestion is a complex mixture of technological givens and people's online behavior, both of which are changeable.
Just like traffic. Please Register or Log in to view the hidden image! You don't limit the people, you address the design and tech issues. To limit the people is to defeat the very purpose of the thing - which only exists to serve them.
What do you mean by "limiting people"? In the long run, and if finances and other givens permit -- sure, design and technological issues should be addressed. In the short run, in immediate situations, it's still down to people -- and expecting more from them.
Designing things so that ordinary people acting reasonably will foul them up in easily predictable ways is designing them badly. This seems to be a difficult lesson for technocrats to absorb. People's online behavior is not under technological control. Unless of course you really intend to establish that kind of control over people's online behavior - the Chinese can provide you with the lessons of their years of experience along that line, as can the Catholic Church. Some friends of mine lived in mainland China for while, back when bicycles were the standard transport (of everything), and they brought in lights for night cycling (no streetlights), and they were confronted by the police - who were in charge of changing people's behavior to fit the requirements of the designers of things. The policeman was angry with them, and said this to teach them the selfish error of their ways: "What if everybody put lights on, what then, hey?" They had no answer.
I was in a European town where they did this: Before: Please Register or Log in to view the hidden image! and after (the same road, from the other direction): Please Register or Log in to view the hidden image! Please Register or Log in to view the hidden image! There's still two-way traffic on that road, but it now appears to run more smoothly than before, even though there are more things to watch out for.
I said it was changeable -- esp. in terms of skills. Like in the above traffic arrangement: all those boxes with plants and those light-yellow boards make people drive more carefully, with more applied skill. So far, the result appears to be fewer collisions and jams, quite paradoxically. Not sure what your point is.
Better design got better results, quickly and at comparatively low cost, from the same people. The original problem was not the deficiencies of skill or character in the people, and trying to improve them would have been a much more difficult task. Agreed?
It appears that the new design called for and brought out the good in people, the better skills and character traits (and arguably, once they are brought out, they are more likely to stick). It appears that the new design (indirectly) improved people's skills and character traits. How fabulous is that! There appear to be bascially three groups of drivers: 1. conscientious drivers who drive skillfully regardless of traffic arrangement or situation, 2. drivers whose quality of driving can be influenced by traffic arrangement or situation, 3. poor drivers whose quality of driving can not be influenced by traffic arrangement or situation. The first group are not a problem and the third group are, for all practical intents and purposes, impossible to improve. The second group is probably the most numerous one and the one that traffic designers target.
In a simple and obvious way, they are at fault. They are controlling their cars. But I agree that there's probably more to it than that. End traffic or end traffic jams? My expectation is that traffic jams would still occur, even if all human drivers were replaced by robot vehicles. Traffic reminds me of cellular automata, where complex structural and behavioral phenomena emerge from lots of interacting individuals each following simple rules. https://en.wikipedia.org/wiki/Conway's_Game_of_Life Traffic jams seem like they would be an excellent subject for the applied mathematicians and computer modelers. (They probably already are.) Create a board with lots of intersections, merges, routes, lane changes and metering elements like traffic lights, then introduce large numbers of pieces moving around it according to simple rules. Then sit back and watch where backups and jams occur. I expect that mathematical formulations of traffic flow could be abstracted from that. (And maybe they have been, this research program I'm suggesting is so obvious that I expect it's already been done.)