The use of smoothing splines and the method of generlized cross validation (GCV) for smoothing discrete noisy data from an unknown but smooth curve is reviewed. The use of 'plaque mince' or Laplacian smoothing splines with GCV for smoothing discrete noisy data from an unknown but smooth surface is described. A numerical algorithm for this (non-trivial) computational problem is described, and an example from a Monte Carlo study is presented to show how the method works on simulated data. The results are extremely promising. Some design problems are briefly mentioned. Some conjectures are made concerning optimality properties of Laplacian smoothing splines and Laplacian histosplines.