[Beldin]

> MVPs are supposed to be prototypes that reveal the answer about a specific hypothesis that is critical to your strategy. It's only a positive tradeoff if you are rigorous about what knowledge you don't have about your customers and technology, and build exactly the thing that will answer the question.

Put differently (to put the misunderstanding to rest): MVPs aren't alpha- or beta-versions of software; don't treat them as such.