Some thoughts on the Navier–Stokes Millennium Prize Problem