I feel like Vladimir Arnold echoed a similar sentiment, but I can't find the quote. He was complaining that a first course in ODE and PDE tends to teach methods

I feel like Vladimir Arnold echoed a similar sentiment, but I can't find the quote. He was complaining that a first course in ODE and PDE tends to teach methods for finding exact solutions, even though equations with exact solutions are vanishingly rare in practice. Or maybe it was Rota again, who knows? Edit: nope, Rota again. Lesson one in your link :)