Earlier quoted context omitted.
Did you miss the part about "light on two sides of every room"? These evolved designs seem optimised for nothing that actually matters to human comfort.
I took this as an initial exploration of genetic algorithms for architecture. As such, the idea is more interesting than the specific parameters and weights chosen here. You'll notice as you scroll down that this article _does_ vary the parameters and weights. For instance, the one that optimizes for windows tends to create interior courtyards. I could imagine adding parameters and weights for thresholds, areas for c…
One could also use a supervised classifier and have people rank the resulting designs. They can know what they like, but not why.