habryka comments on Quadratic models and (un)falsified data