A Cognitive Dimensions view of the differences between designers and users of theorem proving assistants

This paper discusses some observations on the differences between what the designers of theorem proving assistants (TPAs) think about the systems they designed and what the users of those TPAs actually find. A questionnaire based on the cognitive dimensions framework was sent to designers and users of a sample of TPAs. The aim of this work is to be able to identify specific areas that designers of TPAs need to devote extra attention to. It was observed that the cognitive dimensions of closeness of mapping, visibility and juxtaposability and perceptual cues are of particular significance. 1.Introduction A mismatch between the views of the designers and the expectations of the users of computer systems has been cited as one factor that contributes to lower levels of system usability. This is not a recent development in the field of HCI but rather a long-standing and very much alive issue in system’s design. Gingras and McLean (1979) described a study of designers and users of a large information system in the operational area of a major Canadian company in which they investigated the issue of designer orientation. The authors used semantic differential techniques to profile the members of the IS design group and the corresponding user population. When these profiles were compared with each other, it was found that the designers’ images of an actual user were significantly different from the users’ self image. However, they reported that the users’ profile of the designers did not differ significantly from the designers’ self profile. The authors concluded that when designers claim to be user oriented, they are, but what they have in mind are people that look very much like themselves. Research into cognitive approaches to users modelling has, without doubt, greatly contributed to an understanding of what occurs at the interface. Frameworks that originated from such research, such as, the GOMS model (Card, 1980), Backus Naur Form in 1981, Task-Action Grammar (Payne and Green, 1986), and others have been in use for some years. However, the mismatch problem between designers and users still persists. This paper does not aim to investigate this issue but rather to report on it from the theorem proving world. The following section gives some background to the case study. Section 3 exhibits and analyses the results, and the final section puts the case study into perspective and draws some conclusions. 2.Preview of case study The work reported in this paper is a continuation of a questionnaire-based study that utilised Green’s Cognitive Dimensions’ framework (Green, 1989), Green & Petre, 1996) to analyse the usability of theorem proving assistants (TPAs). The framework comprises a set of dimensions or generic attributes that are relevant to successful interaction in terms of influencing how the user think about the system and their tasks. The dimensions were mapped into corresponding TPA’s usability issues to devise a questionnaire. The questionnaire was sent to the designers and users of 27 theorem provers. A response rate of 63% was reported together with the results of the case study in (Kadoda, 1997a, b). The application of the cognitive dimensions framework to analyse the usability of TPAs was successful in producing a set of trade-off relationships between the dimensions that could be incorporated in the early stages of