In my editorial for the earlier special issue on Constructing User Interfaces (March 1982), I expressed the hope that the U.K. would increase its support for research in the area. Shortly after publication it did indeed do so and my group, which reported its recent results in that issue, was able to formalize its position as the Human-Computer Interface Research Unit. More recently, the Unit has moved from Leicester Polytechnic to Loughborough University, where its growing links with the HUSAT Research Centre and staff of the Computer Studies Department were able to become much stronger. It is interesting to note, for example, that Maguire, who contributed to the earlier issue, is now working in HUSAT.
The papers in this issue cover much the same ground as before but demonstrate, amongst other things, a particularly strong shift towards the use of rule-based approaches and to work both with the language Prolog and on developments of it. The first paper, by Clarke, sets the scene by discussing the nature of the human-computer interface; the second paper, by Gittins, takes a careful look at ICON-based solutions to constructing user interfaces; and the following paper, by Richards, Bez, Gittins & Cooke, considers approaches and aids for specifying interfaces. The next three papers discuss various issues in logic programming that have arisen because of considerations of human-computer interaction problems. Hinde reports on some aspects of the use of Fuzzy Logic; Edmonds briefly discusses some recent developments in interactive logicbased systems and Schappo and Edmonds discuss the role of the screen image in interactive logic-based systems. The following two papers, by Connoily, Edmonds, Guzy, Johnson & Woodcock and by Connolly, make use of logic programming, firstly in a knowledge-based approach to speech recognition and secondly in relation to functional grammars. Sbarma & Scrivener describe a technique for constructing models of 3-D objects, and Waldern, Humrich & Cochrane describe a workstation and some preliminary functions dealing with true 3-D representation. Finally, Roberts & Rahbari consider some novel ideas for user-interface construction.
It is again appropriate to increase the funding of research into user interface construction and indeed the U.K.'s Alvey programme is doing just that. It is hoped that the work described in this issue will consequently be continued at a rather increased level of effort over the next few years.