I think the author underestimates the inherent complexity in large scale systems. Yes, incidental complexity makes us skip tests that might have been good, but

I think the author underestimates the inherent complexity in large scale systems. Yes, incidental complexity makes us skip tests that might have been good, but it is the inherent complexity that explodes in our face when totaly apparently unrelated issues causes emergent behaviour. It has nothing to do with that it's a young field, it's just that we can create complexity that is not bound by laws of physics. I usually show 'business people' Conway's game of life, and they are always fascinated, and then I ask them how many rules WE have.