The way the author describes static and dynamic typing seems to imply that static == compiled and dynamic == interpreted. I probably misunderstand something, can someone help me clear things up?
Dynamic typing boils down to runtime tagging of any value that could be referred to by a "dynamically-typed" variable. That's not "interpreted" in a strict sense, but it requires dispatching on every single value access, which will definitely affect performance.
That is typically the case, but it doesn't have to be. You could have an interpreted language that determines all the types when the code is first parsed and generates errors before it starts executing. You can also have a statically compiled language that defers (some or all) type checking until runtime.
