Louis Kates’s research while affiliated with TU Dresden and other places

proto is an R package which facilitates a style of programming known as prototype pro- gramming. Prototype programming is a type of object oriented programming in which there are no classes. proto is simple yet retains the object oriented features of delegation (the proto- type counterpart to inheritance) and object oriented dispatch. proto can be used to organize the concrete data and procedures in statistical studies and other applications without the necessity of defining classes while still providing convenient access to an object oriented style of programming. Furthermore, it can be used in a class-based style as well so that incremental design can begin with defining the concrete objects and later transition to abstract classes, once the general case is understood, without having to change to object-oriented frameworks. The key goals of the package are to integrate into R while providing nothing more than a thin layer on top of it.

To the extent that object-orientation has influenced languages for statistical computation, the approach taken to date has been predominantly based on the notion of classes as the key organizing principle; however, there exists another possibility – the object-based or prototype approach to object-oriented programming. The use of prototype-based programming could provide a fundamental alternative for statistical computation. Prototype programming is particularly suitable for situations where there are a small number of objects and exceptions are frequent. This is often the case in statistical applications suggesting that this approach may have widespread applicability to statistical analyses. As an example, the statistical language R can be employed to compactly describe and construct a variety of prototype-based approaches at a high level. We also explore a number of alternative approaches within the prototype-based paradigm.