Why starting from differential equations for computational physics?

The computational methods currently used in physics are based on the discretization of differential equations. This is because the computer can only perform algebraic operations. The purpose of this paper is to critically review this practice, showing how to obtain a purely algebraic formulation of physical laws starting directly from experimental measurements. In other words, we can get an algebraic formulation avoiding any arbitrary process of discretization of differential equations. This formulation has the great merit of maintaining close contact between the mathematical description and the physical phenomenon described.