Research into human-computer interaction (HCI) is mainly conducted by engineering psychologists, cognitive psychologists and computer scientists. The principal consumers of applied HCI research, on the other hand, are human factors practitioners and system designers and developers. The HCI researcher who believes his or her findings to be of practical relevance has therefore to consider the interface between researcher and practitioner as well as that between system and user: the products of HCI research must not only be relevant but also “user-friendly” to the practitioner. This problem is not merely one of communication between different professional communities, as the optimal route for the translation of research findings into terms that will be of practical use in the design process is itself a matter of considerable uncertainty and debate. Thus there are many instances in the research literature where apparently contradictory recommendations can all too easily be drawn from findings based on sound but, by its very nature, limited experimentation (e.g., compare the findings of Landauer et al., in press, Ledgard et al., 1980, and Scapin, 1981, on naming text-editing operations). One of the prerequisites for tackling both the communication problem and the translation problem is an understanding of relevant aspects of decision-making in design which influence the usability of the end-user interface. This is so for three reasons. First, an appreciation of the nature of design practice will at least help identify those areas where research input might have the greatest impact and allow researchers to direct their efforts towards them. Second, it may identify possible modifications to existing design practice which would allow research input to be used more effectively. Finally, it would be somewhat surprising if current design practice were not to furnish researchers with any insights into the underlying processes of users. The experience and skills of the practitioner should be a valuable source of information for the HCI researcher. For these reasons, we have been documenting some of the relationships between design practice and the usability of systems for use by non-experts. While there is considerable literature on programming behaviour (e.g. Mayer, 1981), reports of design behaviour are rare, other than occasional descriptions by practitioners of the interface design of their own products (e.g., Botterill, 1982; Smith et al., 1982). This paper focusses on the influence of the individual designer's decision-making. Evidence is taken from interviews with experienced system designers concerning design issues influencing the nature of the user interface which had arisen with systems they had recently worked on. For two of the systems usability investigations had been performed (see Lewis & Mack, 1982 and Hammond et al., 1983).