HCI engineering design principles : acquisition of class-level knowledge
Stephen Cummaford · OpenGrey (Institut de l'Information Scientifique et Technique) · 2007
This research addresses a general problem, which is characterised as a lack of validated HCI design knowledge, supported by guarantees of application. Long & Dowell's (1989) conception of HCI as an engineering discipline characterises such knowledge as HCI Engineering Design Principles (EDPs), which offer prescriptive design support to solve the general design problem of HCI, and can be validated, such that guarantees of application may be ascribed. EDPs may thus be considered a solution to the general problem identified. The technical aim, addressed by this research, is the specification of initial EDPs. The EDPs are considered 'initial', as they have not yet been validated, and as such are not supported by guarantees of application. However, their specification makes progress towards the general solution of EDPs supported by guarantees of application. In order to support development of EDPs, this research extends Long & Dowell' s conception of EDPs, by specifying a conception of the general EDP as class HCI design knowledge, that is, EDPs have a class of design problems as their scope, and support specification of a corresponding class of design solutions; and a conception of the general design solution delivered by application of the general EDP. The thesis then identifies a strategy and a method for the development of EDPs, which involve the construction of class design problems and corresponding class design solutions, and the identification of EDP components from commonalities and non-commonalities between these classes. The method is operationalised in two cycles of research, to construct two initial EDPs for business-to-consumer electronic commerce transaction systems. The two initial EDPs achieve the technical aim of the research, and as such, make progress towards the general solution of acquiring EDPs supported by guarantees of application. The requirement for additional research, sufficient to develop guarantees of application, is then discussed