Should surface science exploit more quantitative experiments