Proceedings of OZCHI91, the CHISIG Annual Conference on Human-Computer Interaction
Time and place:
OZCHI is Australia’s leading forum for research and development in all areas of Human-Computer Interaction. OZCHI attracts an international community of practitioners, researchers, academics and students from a wide range of disciplines including user experience designers, information architects, software engineers, human factors experts, information systems analysts, and social scientists.
The purpose of this paper is to review a number of important international initiatives concerning ergonomics regulations, and in particular to consider the European Directive on display screen equipment. This Directive is driving the development of ergonomics based user interface standards in Europe. Since European standards are based on International Standards, the regulatory requirements in Europe are likely to have world-wide impact, not only on international suppliers but also on employers as computer users in other continents use the European experience as the basis for their expectations and demands. Other standards activities are briefly discussed.
Usability testing is an excellent tool for determining whether a software product is easy to learn and use. This paper describes our experiences with incorporating usability testing into the design process for a graphical user interface to an electronic mail system. The procedures necessary to conduct the usability test and to maintain a constant focus on end users in the initial design phases are described.
Current research into user interfaces has largely concentrated on the end-users, and more recently on tools for interface developers -- few if any tools are available for the human factors analyst to evaluate user interfaces. Rather than building a UIMS (User interface management system) from scratch (an expensive and multi-disciplinary effort), we have concentrated on developing software for the evaluation part of an UIMS, We have modified and combined existing software to create three functional modules: screen creation, recording, and compiling. The tool enables human factors specialists to design various text-based user interfaces and then test these on users (record time, keystrokes and task sequence). A first prototype version of the user interface testing tool has been completed. It is presently being tested. The tool requires no programming knowledge.
User-interface design requires constant interaction with the user, both to develop and clarify the requirements specification, and to test out ideas for solutions. The involvement of real users is often not possible or desirable in a teaching situation, but the commonly used alternative of a written problem specification supplemented by "clarifications" from a tutor provides neither the richness of interaction nor the flexibility to cope with the range of problems raised by different solutions to the same problem. This paper describes and attempts to evaluate an alternative approach, in which a staff member role-plays the user, meeting with each design team regularly during the design process.
The effectiveness of any software, and educational software in particular, depends upon its usability. It is unlikely that learning will occur where students (users) have difficulties using a particular learning package. Software usability is typically evaluated by observing and questionning users, or evaluating their mastery of the package under test conditions. In the study presented in this paper various aspects of student use of a piece of educational software have been logged by the system in actual use. This has provided a large quantity of data which has been correlated against several factors including student background, success in mastering the subject matter presented and the type of task undertaken. The results of the study provide guidelines not only to improve this software package but could be applied to other such products.
This paper attempts to share the author's experience in some of the problems faced in teaching user interface design at the Centre and how CAI can assist the students in learning the subject. A discussion of a prototype CAI system which was recently developed for teaching the subject will be presented. The author will also attempt to discuss the future of a CAI system for user interface at the Centre.
The last session of the OZCHI91 conference is a panel presentation and discussion about HCI education and training. It follows three papers presented on educational studies and activities and is intended to raise questions and (hopefully) find answers on the subject of what (if anything) needs to be done to ensure that HCI practitioners are well-educated and trained.
This paper concerns the difficulty of including human factors, a major component of quality, within the software development process. It acknowledges unsuccessful efforts of the past which are indicated through a study. There exists a large gap between understanding and implementing the key principles of usability. Understanding the cause of this chasm can give software developers the perspective and awareness needed to create positive change. These changes can lead to improved product usability and enhanced product development life-cycles.
Models play important roles in design, as they can provide useful representations of particular aspects of systems (such as task demands, or error types). Models are frequently developed and used by groups of people with different interests, in order to facilitate a "common view" of a system under design or review (and this process may be implicit or explicit). Three examples of explicit models used recently in local industry are described within a framework, the purpose of which is to illustrate some of the range and utility of models in interaction design processes.
This paper discusses the range of end user participation needed for a developing system during the design process. Relying upon interviews and customer representatives to obtain user input is inadequate. The methodologies inferred by this paper have resulted in optimized system usability, a dramatically high degree of system acceptance, and a high degree of customer satisfaction.
This study examines the relevance of user interface design guidelines, such as the extensive set developed by Smith and Mosier (1986), to the relatively sophisticated user interface of an existing language-based editor. The aims of the study were to ascertain the potential impact of the advice offered by the guidelines on this type of software product and to investigate whether any issues, relevant to this and future versions of the editor, were not addressed by the guidelines.
According to recent surveys usability engineering methods are not much used in actual projects, because they neither provide clear goals and milestones nor evaluation metrics. Much of the user interface development starts from scratch or rough sketches and relies on tool-driven prototyping. It seems evident that one should try to build support for usability engineering methodologies and HF expertise into HCI tools as much as possible so that they would be widely disseminated in practice. The User Interface Management Systems (UIMS) and recently introduced rapid prototyping tools are promising to change actual practices. This paper is a state of the art survey of UIMS and UI prototyping tools and supporting methodologies. It also considers evaluation and selection criteria for HCI tools. Finally, a case study about tool selection for an application project is described, which highlights the practical arguments that affect the selection of a HCI tool.
A study aimed at understanding user interaction with map-based user interfaces is presented. The study concerns itself with user modelling issues, rather than with the utility of the user interface per se. In accounting for the observed behaviour (in terms of a conception based upon three constructs: states, processes and resources) current theoretical work in modelling HCI is extended.
The evolutionary development and evaluation of a graphical user interface for Unix electronic mail is described. Preliminary interface designs were tested by controlled experiments with groups of subjects. The results were applied to a gradually evolving interface which has supported the day-to-day e-mail requirements of a small group of active users over a three year period. Some experiences and outcomes of this action research are reported.
The generation of computer models from computerised tomographic scan and nuclear magnetic resonance data is well established. The hardware to allow rotation of the models and concentration on particular parts of a model runs sufficiently fast to be useful. Software is readily available to provide visualisation which increases appreciation and knowledge of the images. This will permit sectioning of the object and isolation of its parts. The components of a system that could be used by surgeons and others providing invasive treatment are in place. These techniques provide an effective ability to visualise pathological lesions, and so increase the accuracy of the approaches, and decrease damage to normal tissues. In part the role of this paper is to outline the components of the technique; in part, the intention is to reflect on the conditions needed for early adoption.
Developing a methodology for a self-service marketing terminal for a banking environment has special issues regarding the customer interface that need to be solved. Information other than technical and interface design is required to address these issues. Resources from within the organisation, for instance a marketing department, are often able to supply the material required. Whilst information may have been gathered for a different end, appropriate analysis can extract the relevant portions. Applying this approach to a prototype self-service terminal shows the value of a research-based approach to computer-human interface design. Developing a team of experts to conduct a heuristic evaluation of a system prior to usability testing can provide a low-cost indication of likely problems. With the right mix of team skills problems involving navigation, visual and language inconsistencies and information credibility can all be highlighted. Often these points (and more) are confirmed by a first, and even brief, round of testing.
Bednall (1990) demonstrated that when subjects search for targets in lists presented on a computer screen, the way in which the screen is formatted determines how rapidly targets are located. In the current paper, data from that study are analyzed with respect to strategies subjects adopt when searching lists. Search strategy was inferred from questionnaire data and an examination of both the effect of target serial position and task instruction on search time. The main findings were that search strategy was affected by the amount of structure provided on the screen and could also be influenced by instruction.
When linked to telecommunications systems, multimedia acquires a new level of functionality. Multimedia telecommunications constitutes a flexible, new medium which can facilitate human-human interaction. However, communicating via images represents a new paradigm for many professionals. The arts and humanities have an important, if neglected, role to play in this area. CREATE is an interdisciplinary creative research team working between the School of Humanities at UTS and the College of Fine Arts at UNSW to develop projects which examine operating methods, interfaces and applications for multimedia communications systems in conjunction with visually-based end users.
A "calculation task" model has been developed to simulate the effects of human error on simple calculation tasks when using an electronic calculator. Results obtained from analyses of the model were compared to existing survey data. It was found that a keying error rate of approximately 0.0015 was appropriate for a typical keyboard layout. The "calculation task" model was verified by comparing error magnitude results with appropriate survey data.
This paper deals with the importance of conducting User Needs Analyses prior to designing, modifying or purchasing computer systems. Cases are presented that illustrate the effects one might expect on systems usage and usability when such analyses are not carried out. It is argued that User Needs Analyses conducted from within a wide framework or organisational, job and task needs be incorporated as an integral part of usability testing and evaluation plans. A warning is issued against the strong tendency to concentrate on development and refinement of usability measurement tools which can, it is argued, lead to a state of usability myopia in which the bigger issues of understanding users, their tasks and needs are overlooked.
This paper presents our perception of an intelligent user interface. First this definition is discussed, then the corresponding functional framework is given. A software architecture model is also described as well as an elaboration environment with several specialized tools. This presentation is completed by two examples of interactive applications using this kind of intelligent interface.
This paper outlines some of the pitfalls, in terms of organisational change, that may be met when introducing usability as a major contributor to product quality. Issues addressed are: * ownership of product usability, how to prevent the "us and them" syndrome * project team involvement and commitment * senior management backing and awareness * getting and maintaining a user perspective * communication, planning and reporting * planning a usability program, starting from scratch Based on the experiences within Westpac, this paper offers practical advice on how to set up and ensure the success of a usability engineering program in an industrial setting.
23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added 23 Jun 2007: Added
Page maintainer: The Editorial Team