In order to understand the behaviour of ceramics and minerals at an at
omic level a wide variety of modelling techniques are presently used,
and one of the purposes of this workshop is to bring these together. A
t the most expensive end of the spectrum are the nb initio electronic
structure calculations, which can accurately model the anergies of bon
ds of varying strength. Al the other end are simple, almost naive, mod
els where interaction energies are approximated to be of either infini
te or zero strength. In between there is a variety of empirical scheme
s of varying degrees of complexity. Which approach is most appropriate
depends on what is wanted from a model, that is, what are the questio
ns that are being asked. In this introductory review I comment on the
virtues of different approaches, and highlight the comparisons with re
ference to a number of studies where combinations of different approac
hes have been used to good effect.