I see what you mean. This is about very complex structures and interactions.
But in fact, I believe with enough time you can get used to any level complexity and find your way through it. It is a bit like a big city: however complex it is (intertwined roads, subways, highways, etc.), if you live there long enough, you will get a precise mental model. Same with software projects: after some time, abstraction will not be those foreign short lived unstable concretions that you hold in your head and that vanish away if someone says hello. They will be good old friends and the sole invocation of their name will instantly call a lot of solid knowledge to the bar.
So in this line, I'd say clever guys are just faster and more agile (maybe both are the same thing). Above a certain level of tolerance to abstract thinking, anyone can understand and work on big complex projects, given enough time.
There is another difference between good and great developers, though. It is the "best path-finding skill". For instance, given enough time, I would probably be able to write a Python script solving most sudokus, but I fear I wouldn't find as elegant and straightforward a path as Norvig's. http://norvig.com/sudoku.html
But in fact, I believe with enough time you can get used to any level complexity and find your way through it. It is a bit like a big city: however complex it is (intertwined roads, subways, highways, etc.), if you live there long enough, you will get a precise mental model. Same with software projects: after some time, abstraction will not be those foreign short lived unstable concretions that you hold in your head and that vanish away if someone says hello. They will be good old friends and the sole invocation of their name will instantly call a lot of solid knowledge to the bar.
So in this line, I'd say clever guys are just faster and more agile (maybe both are the same thing). Above a certain level of tolerance to abstract thinking, anyone can understand and work on big complex projects, given enough time.
There is another difference between good and great developers, though. It is the "best path-finding skill". For instance, given enough time, I would probably be able to write a Python script solving most sudokus, but I fear I wouldn't find as elegant and straightforward a path as Norvig's. http://norvig.com/sudoku.html