Is it possible to design a better stop sign?
slate.com
slate.com
The preponderance of stop signs on US roads is one of the most infuriating parts about American road engineering. (Along with such a large number of signs that it is not humanly possible to process them all when traveling at vehicle speeds.)
It really does condition drivers to look for signs instead of other people. If you run a stop sign, you'll get fined even if nothing happens. If you stop at a stop sign and then in blissful ignorance ram someone who didn't, no harm comes to you (unlike to the other person).
The book "Traffic" by Tom Vanderbilt is an interesting read about some counterintuitive things about how traffic works and about how differently engineers in different countries approach safety. I can highly recommend it.
EDIT: One of the most fascinating stories in the book is about how they redesigned some big 5-way intersection with a traffic light in some central square of a Dutch city by just removing everything and leaving a big empty square with some lines painted indicating the roundabout. Crashes went down, and traffic flow improved, because drivers were forced to pay attention to figure out how the whole thing worked.
Experience seems to indicate that the "safety" measures usually taken in the US (widening roads, removing pedestrians, etc) are totally counterproductive because they send subconscious signals to the drivers that this is a speedway which causes everyone to drive faster and pay less attention, which makes crashes more severe (not to mention relegates pedestrians, bicycles and other alternative means of transport to second-hand status.)
Incidentally, this is a pattern underlying many such European innovations--put a lid on the obvious problem without acknowledging the more insidious and possibly much worse problems that the solution creates.
Not to mention the effect on pedestrians, bicycles, etc. If you're worried about civilization slowing down and increase productivity, minimizing time stuck in traffic seems like a good start.
My issue is not with any particular intersection design. Maybe the roundabout was better in that Dutch city than a traffic light, and probably all things considered, some intersections will be better if drivers are forced to pay more attention. What bothers me is the assumption that there is no cost to forcing people to solve puzzles when they otherwise wouldn't have to. You can't have something so important as an afterthought.
I'm sure the rest of the people around you appreciate you making this risk assessment on their behalf as well.
To sell me on the idea of a new stop sign, you'd have to demonstrate that not only was the new design objectively better, but it is objectively better by a large enough margin to offset all the additional errors caused by drivers who aren't trained to look for a different design.
Would a different color work better?
No.
"Red-green colorblindness" is an inaccurate and misleading name; "green weakness" is a better description, and is what every doctor I've talked to uses as a casual name for the condition.
The best explanation I know of is to imagine the following: you're looking at a screen onto which is projected a square of pure red. Then, gradually, green is added, which will eventually cause the square to appear yellow (yes, red and green make yellow). When a person with "normal" color vision sees it as yellow, a person with the common forms of green weakness would still see it as a bit orange-ish.
In other words, the common form of "red-green colorblindness" doesn't cause a red sign to look green, or a green light to look red, or grass to look red, or any of that stuff. It mostly causes problems in the yellow part of the spectrum (one reason why I tend to avoid earth tones).
I clearly remember two pictures, one depicting very bright red berries on a bush and other with the berries colored green, almost invisible now.
So, is that a different condition?
Also, for what it's worth, on my MacBook the images in the aforementioned blog post look almost the same to me. But I looked at them on a different computer as well, a netbook where I tweaked the color settings, and there they don't look the same at all. (I still can't pass the Ishihara test on that one though... :-)
Not as many as I said, but still a significant fraction.
When people say "red-green colorblindness" they almost always mean deuteranomalous trichromacy (AKA deuteranomaly, "green weakness"), which is A) by far the most common form of "colorblindness", B) for practical purposes mostly affects the yellowish part of the spectrum and C) is typically so mild that most people who are deuteranomalous could go their whole lives without ever knowing it.
The problem is that the average person doesn't ever realize this -- just as you apparently believed that the most common form of colorblindness led to an inability to distinguish red from green -- and so there are a huge number of misconceptions floating around which, as one of those 6% of males who are deuteranomalous, I do my best to clear up whenever possible.
Since about 1973, a sort of fluorescent chartreuse yellow has been adopted as the color of fire engines in parts of the United States and elsewhere. The reason behind this is that chartreuse fire engines are more visible on the streets than the traditional red fire engines, especially at night...
However, everyone is conditioned to a semaphore scheme of GREEN go, RED stop, so I guess that can't be conditioned out.
Strong fluorescent colours, particularly the pink, do terrible things to my eyes in some circumstances particularly when they have contrasting writing on them.