I will discuss recent progress and open challenges in connecting the dynamics of line defects to the large-scale evolution of facets, which give rise to free boundaries, on macroscopic crystal surfaces. At the microscale, a large number of differential equations is invoked for the positions of defects. At the macroscale, the surface evolution is plausibly described by a Partial Differential Equation (PDE) for the surface height profile. A goal is to derive boundary conditions for the PDE consistent with the motion of defects. I will address the (simplified) case of an evaporation-condensation process in a radial geometry, which provides some generic features of the more general problem.