I have. Placing objects in an editor is obviously essential, I agree. I just expect to find code at the bottom driving it, loading the map etc, and it doesn't seem like any example projects or learning documentation for Unity is geared that way. Basically I expect to follow code at the bottom until it loads some asset, which may create a bunch of objects, and register some scripts. Then I return to the code which can then get those objects if it wants.

sure but like, you don't have that expectation of a browser, right? Like to write a web page, you don't create an instance of an HTML parser and call a parse method on some text, and then walk the tree and tell it to draw each element, right? Same deal. Just think of the scene as being a DOM in three dimensions.