> So to answer your question, nothing (in this context), because the Turing machine is a more useful model for thinking about computation itself. This isn't ob

> So to answer your question, nothing (in this context), because the Turing machine is a more useful model for thinking about computation itself. This isn't obvious to me. It seems to me that a TM is more useful for thinking about computation from an operational perspective. But this isn't the only, or arguably even the most effective, way to think about computation. The lambda calculus is much more useful for thinking about computation from a denotational perspective.