Again, Miller and Richards ( ) have pointed to the lack of alternative visualisations. In contrast with more traditional paper-based reconstructions which can disguise missing or uncertain data with an appropriately selected viewpoint or the careful positioning of trees or smoke, the computer reconstruction is presented in more absolute terms. Consequently, even if the model is developed to convey an 'impression of possibilities' ( ) it more often conveys an impression of truth, and indeed, given the increasing emphasis on authenticity of materials and lighting, an impression of reality. It is still rare for alternative reconstructions to be presented, and while there are obvious costs involved in the presentation of multiple versions of a model in traditional paper-based publications, no such costs are associated with electronically delivered models yet alternative possibilities are still largely absent. There are attempts to create time-enabled 3-D models, but these are in their early stages. One example is the Rome Reborn project, which intends to create a model of the city from the Iron Age to late antiquity capable of displaying alternative reconstruction hypotheses ( ) but currently available models do not offer such alternatives.
Discussions of the construction of 3-D models inevitably refer to the ability of the user to modify the model according to changes in hypothesis, but the 'user' in this context is almost always the creator of the model, not the viewer, and the viewer usually sees only the final homogenised result. Few models have the level of flexibility and degree of viewer control as the reconstruction of the Roman theatre at Canterbury ( ) in which the height of the building, the number of seats, and the presence of other features can be altered interactively by the viewer. Whether this is necessarily an ideal approach to the flexible modelling of alternatives will be returned to later.