Classical techniques in analytic mechanics, such as the Lagrangian and Hamiltonian approaches, do not extend themselves to systems with unilateral contact and kinetic (sliding) Coulomb friction. Most current approaches formulate the contact forces as a solution of a linear complementarity problem (LCP). However, when friction is present the solution set of an LCP may be nonconvex, leading to theoretical and computational difficulties. This talk will describe how Gauss' Principle of Least Constraint can be used to formulate the contact forces as the minimum Euclidean norm solution of a convex quadratic program. Example problems and experimental results will be discussed.