It seems to me that there is an interesting phenomenon here that applies across a wide range of topics. Basically, what was an acceptable risk level in the past may not be acceptable in the future.

• Sailing ships. 100 years ago there were still square-rigged ships where people had to climb hundreds of feet into the air to work the lines, in bad conditions--night, winter, storms--without safety lines.

• War. Western soldiers now expect to have body armor in addition to helmets and boots.

• Railroad locomotives. Sure, a big old steam engine is "fun," but would would actually want to work in one nowadays, with no cab temperature control, lousy signaling, no visibility whatsoever...not to mention the problem of people on the tracks.

• Cars, where the death rate in the 1930s was astounding.

So a technical solution like a subway platform or a railroad ROW or a highway or the observation deck on a skyscraper or the failsafe system in an elevator--anything related to safety--that might have been acceptable in the past now becomes unacceptable. That cost must be taken into account for future designs, and as a retrofit cost for existing systems.

New subways have barriers so you can't fall on the rails. You could do that with train platforms, also. The ROW could be protected throughout its length--at considerable cost. If there's enough money, you can always find a way to improve the safety of a system.

I wonder whether this effect is going to be the end of cars. Since they are so deadly, at some point--maybe as a result of the gun debate in the U.S.--the question will come up "why are we building more of these slaughterhouse roads when we could build trains instead?"

Not to mention football/soccer; the game will change considerably when heading is inevitably outlawed...