Comments on ´The role of analysis in an age of computers: view from the analytical side´

The commenter disagrees with the implication in the above-titled paper by R.E. Collin (ibid., vol.32, p.27-31 (Aug. 1990)) that analytical solutions are the only way, or even the best way, to check the accuracy of numerical solutions. He expresses surprise that no mention was made of verifying solutions accuracy by comparison with measured data. He argues that measurements provide what is often the best way of verifying the accuracy of a solution to a practical problem in electromagnetics, regardless of the extent to which the solution is analytical or numerical. He gives an example from the paper (radiation conductance of an open-ended microstrip line versus substrate thickness) to illustrate this point.<>