It's providing an interface/binding to that underlying functionality that fits into the React model of specifying a UI declaratively. Like almost any framework or library, React tends to be easiest to work with if you can follow its conventions. There is more friction if you want to do something outside its comfort zone, so this kind of bridge is potentially quite useful for anyone who does have a need for 3D visuals in their React-based app.

Something about making 3D primitives into a DOM-like hierarchy with update-able props: https://github.com/toxicFork/react-three-renderer#implementa...

Still wrapping my head around it.

Because it's demonstrating the binding of WebGL to React using this library: https://github.com/toxicFork/react-three-renderer/