This paper will critique Empirical Modelling by comparing its benefits against traditional methods of developing what EM calls Interactive Situations Models; that is models which describe the operation of an artefact and embody its behaviour. Models that can be called ISMs are often used as specifications for real-world artefacts that are embedded into the functionality of a device for example the behaviours of a digital watch or car locking system. There is a previous body of work within the Empirical Modelling library that has attempted to use tkEden to build and refine such behaviours. By taking these examples, as well as developing my own, this paper shows that Empirical Modelling's suitability to this area of development allows a designer to constantly rework his model, interactively test it, and by doing so capture the semantics of his artefact.
This paper briefly considers the application of the principles of Empirical Modelling to an analog mechanical computer of WWII, and puts forward the theory that this primitive early computer, that followed such different principles to today's all-purpose computers, had properties of its design in common with EM models. The computer and system in question are related briefly to the principles of Human Computing, and the outcome of the modelling exercise itself is critiqued.
This paper considers the possibility of getting robots to develop Empirical Modelling (EM) construals. This would allow robots to make models of their surroundings which would help them to per-form their tasks better. I look at the possibilities of robots in the home using EM to learn their envi-ronment and to help them locate where they are and where they need to go. The first section of this paper looks at the three main aspects of EM and discusses the benefits and difficulties of robots producing these. In the second section I discuss two scenarios (and their respective models) which a robot may find themselves in.
One of the key principles of Empirical Modelling is observation. The interface to a model is critical to a user's experience. This paper describes the various notations for describing graphics using Empirical Modelling techniques, including notations for specifying complex geometry as the application of functions on primitive objects. The Sasami 3D visualisation notation is discussed and an improvement to Sasami is described. This involves adding support for primitive objects such as cubes, cylinders and spheres, allowing the creation of 3D environments for models using a definitive notation.
This paper will explore the possibilities and problems associated with modelling digital logic circuits in a definitive language. This will be achieved by attempting to model a digital circuit (the logic, its visual representation and an editor) using a definitive notation. It is hoped that wires in a circuit can be modelled as dependencies in Eden - the output from the wire (and therefore input to the next component) is the output of the previous component. The methods used by the model (and possibly the efficiency of a dependency based simulation) will then be compared and contrasted with the methods used in the Java based OLS solution.
Definitions in Empirical Modelling are used to represent state, however it is possible to use definitions to describe behaviour, as shown in this paper. To show the differences and benefits of behaviour definitions a new language called Doste is implemented based entirely on these behaviour definitions. An existing EM model will be implemented in Doste for this purpose. Current EM tools have no ideal solution for modelling agency so the aim of this investigation is to see how definitions could be used for state and behaviour. Ultimately the idea is to come up with alternative and possibly better tools for Empirical Modelling which do away with procedures and at the same time allow greater integration with the operating system. This paper is meant as an introduction to these ideas and showing the possibility for future work in this direction.
This paper describes the crafting of a colour model of Schubert's "Erlkönig" composition using an Empirical Modelling approach. First of all, different methods for associating music and colours are discussed followed by an explanation of the model itself. The main interests of this study are to examine the relationship between visual cues and musical experience, and also the extent to which Empirical Modelling provides a means for non-theory based analysis of musical compositions. I am also interested in the creative process of Empirical Modelling and its usefulness in regards to human computing, as discussed in M. Beynon, S. Russ, W. McCarty, Human Computing: Modelling with Meaning.
This paper will present an approach to the visualization of quantitative dependencies. It will do this on the basis of an examination of factors, which determine the distance a driver will keep to the car in front in real world traffic situations. Though common sense poses that the proposed model is fundamentally valid it should not be understood as an attempt to predict real world behaviour and is not the core of this paper. It is rather a medium to give meaning to the proposed approach of visualization. The paper will discuss advantages of the given approach and discuss the application of EM for its implementation.