In typical usage, a managed object context for Rails wouldn't be that useful. In a stateless HTTP request, objects are typically created and deleted within a short period of time. Your example in the Rails console is atypical: after you finish the move, your request would be complete and you'd destroy the show object. The next request would load a new show object.

The typical Cocoa application, on the other hand, has a much longer object lifecycle and benefits from a context in which objects are managed for you.

-ch