Been watching a lot of disaster shows lately, specifically engineering disasters. Partly it's the morbid fascination with destruction, but another big part of it is the problem-solving that goes into a) figuring out what went wrong, & b) figuring out how to prevent it from happening again.

Every single time, a failure is down to *multiple* factors: a series of failures that all manage to align when unforeseen circumstances make failure not just a possibility, but an inevitability. Factors include human negligence or ignorance, lack of training/education, under-designing, building with materials inadequate for the job (either from cutting costs or because they don't know yet that the materials aren't enough)...

...design ideas that seem good at the time, but then environmental pressures prove more than the design itself can handle, changes made to a structure while it's under construction that aren't properly documented... lots of reasons, really. And of course, every safety policy is written in blood.

Not every engineering failure ends in deaths, but plenty do; and even when there aren't deaths, there are often serious injuries, and of course the destruction of materials, structures, and building equipment (like cranes or tugboats or barges, etc.). Hence the saying about policy being written in blood: all too often, people have to die before something is made truly safe.

And when something fails, it isn't always down to human error: sometimes it really is that the humans who built something *just didn't know* that the materials weren't suited to the environment, or that a sexy new design wasn't going to cut it at a particular location.

Or, rather, it isn't down to human *negligence* or carelessness. Sometimes it is, definitely. A lot of times, it isn't. And it's interesting to me to see how often the solution to an engineering problem is to *over*-engineer the new structure. Like, figure out how tough the environment will be on the structure, then make it 5x or 10x stronger than it needs to be.