[CodiePetersen]

Well it's not too crazy when you think about it. The web is just encoding location and size data for them. Any medium that generated the same results every time would do the same thing. I think because of their smaller intelligence it just seems more profound than it actually is because it represents a significant portion of their capabilities.

We see this in some types of machine learning. Echo state networks and reservoir computing essential just encode the data with their random connections.

So I think it's cool, but not that surprising. The spider is just learning the world through its web just like a musician learns music through a specific instrument. The difference is its not a significant portion of our intelligence and we have extra space/power to learn the world through different tools. Language, writing,reading, computing, mathematics, etc.