I have a question related to the answer to this question: Morphism of ringed spaces not induced by homomorphism of rings.
I understand the entire argument but I'm having trouble how I can see that the map of global sections $f^\#(\operatorname{Spec} A)$ is the natural inclusion.