I think 'over-engineering' is too broadly used. Clearly some solutions are over-engineering, but most are built by skilled and experienced engineers who know th

I think 'over-engineering' is too broadly used. Clearly some solutions are over-engineering, but most are built by skilled and experienced engineers who know that requirements change / evolve. In this (common) scenario, not predicting it - and creating the necessary 'over-engineered' solution - will lead to a failed project. Alternatively, predicting the wrong thing to 'over-engineer' is equally problematic. Without these gambles, the project will often fail too. It's a really rather difficult problem.