Ways of arrangement : the basic operations of form-making

Abstract

Making forms is essentially a matter of arranging things, and arranging things is essentially to establish spatial relations among selected elements. The thesis provides a minimal set of basic operations believed to be sufficient for constructing any given configuration. These basic operations can aggregate to make compound operations handy to designers. Both the basic and the compound operations are called 'arrangement moves'. Two kinds of basic moves are distinguished: the generic moves, which construct only generic relations such as 'connection', 'separation', etc.; and the ordering moves, which are characterized by using virtual 'lines' as references in establishing spatial relations. A physical design is viewed as finding a correct arrangement that satisfies given constraints. Ordering moves are viewed as an operational foundation that makes such exploration of formal arrangement possible. The thesis demonstrates that arrangement moves can describe any individual form by reconstructing it; arrangement moves can also describe any family of forms by formulating rules governing the form family. It is further demonstrated that the basic arrangement moves have inherent properties capable of constructing inference rules for perceiving spatial relations. Based on the fact that arrangement moves can sufficiently construct forms, representing rules of forms, and perceiving spatial relations, it is of particular interest to the development of a computational design system that can do arrangements, know form rules, and can check arrangements against rules.