The spectacular and fatal train crash on the MetroNorth line in New York earlier this month brought about renewed calls for a faster installation of positive train control systems across the nation's rail network.
The expensive and expansive fix, long sought by federal authorities, allows for a combination of technologies to monitor trains and automatically slow them if disaster is impending. The technology is expensive, costing hundreds of millions of dollars for MetroNorth alone, for example.
It also will take years to get installed, test and get working on a daily basis.
The nation's railroads are under the gun to install the systems by the end of 2015, but it could take many of them longer to meet the requirement.
And, on a sheer, day-to-day basis, positive train control really isn't the answer to safety that it would appear to be.
There is no doubt it would improve the chances of prevention of catastrophic accidents such as the one in New York, where the engineer simply zoned out for a few moments and hit the brakes far too late to slow the passenger train to hit a sharp curve; or one earlier this year, where a runaway oil train drifted into the Canadian town of Lac Megantic and exploded, the result of nobody at all being onboard while the crew took a break without the engineer making sure enough brakes were set; or the Los Angeles commuter train head-on wreck in 2005 where a cell-phone-distracted engineer missed a signal.
Yet, it won't improve the chances of preventing other sorts of fatalities and accidents, such as those at unsafe crossings, those caused by vandalism, and those resulting from people being on the tracks.
Spending more on maintenance also would prevent derailments that occur from deteriorating track or defective train equipment.
Massive train accidents are spectacular in an unfortunate way, and they gather the attention of the nation to call for the most expensive of safety features.
But safety could be as simple as spending a bit more on employment than infrastructure and putting a second engineer into the cab. It might not solve every problem, but surely a crew dedicated to safe operation would back one another up as is required in airliners. And, even there, multiple pilots and redundant safety systems do not solve all the safety problems that can arise.
Automation in aviation has led to a generation of pilots with less plane-handling manual skills because the computers have been doing the flying of the big aircraft for years.
Diligence and systems and humans backing one another up can improve safety, but errors will still result so long as humans are in charge of human-designed systems, no matter how many billions of dollars are spent.