Software engineers have learnt to read the reaction to coding agents as a split between people who love the craft and people who care about the product. Seen through that lens, mathematicians upset about OpenAI’s Navier–Stokes result look like craft-lovers in denial, moving the goalposts. I think that misreads them. “After Math”, by Silvia De Toffoli and Eamon Duede, argues that mathematics was never after the answer: the understanding is the product. Then I extrapolate to research software, where I’d argue the same holds for code that establishes scientific results.