Several fundamental and unsolved questions should be answered or clarified before new and substantial progress can be made toward a more rigorous and complete solution of the general problem of electromagnetic scattering, and of the related problem of emission, by the ocean surface. Successively discussed are the limitations of the linear surface waves theory, the assumptions implied by and the inadequacies of the method used to obtain the spatial spectrum of the sea surface from the temporal spectrum, and the poor knowledge of the spectrum angular behavior. It is stressed that the sea surface is not uniquely determined by the wind speed and that it should be depicted by a more complete set of internal parameters. Relationships between some usual sea state parameters, such as H1/3, &sgr;, u*a, and U(z) are discussed. On the other hand the available methods for solving the electromagnetic scattering problem are briefly discussed, and it is shown that the assumptions implied by Brown's (1978) two-scale model, based on the boundary perturbation method, are roughly satisfied when applied to the sea surface. The contribution that particular scattering mechanisms could provide to the scattered power is estimated. The azimuthal behavior of the backscattering coefficient is taken as an illustration.