HCI Bibliography Home | HCI Conferences | HCII Archive | Detailed Records | RefWorks | EndNote | Hide Abstracts
HCII Tables of Contents: 89-1a89-1b89-2a89-2b91-1a91-1b91-2a91-2b93-1a93-1b93-1c

Proceedings of the Fourth International Conference on Human-Computer Interaction

Fullname:Proceedings of the Fourth International Conference on Human-Computer Interaction
Editors:Hans-Jorg Bullinger
Location:Stuttgart, Germany
Dates:1991-Sep-01 to 1991-Sep-06
Publisher:Elsevier Science
Standard No:ISBN 0-444-88775-X; hcibib: HCII91
  1. HCII 1991-09-01 Volume 1
    1. Congress II: Design and Implementation of Interactive Systems: COGNITIVE ENGINEERING; Advances in Cognitive Engineering
    2. Congress II: Design and Implementation of Interactive Systems: COGNITIVE ENGINEERING; Research on User Modelling
    3. Congress II: Design and Implementation of Interactive Systems: COGNITIVE ENGINEERING; Design and User Modelling
    4. Congress II: Design and Implementation of Interactive Systems: SPEECH RESEARCH AND TECHNOLOGY; Design and Evaluation of Speech Interfaces
    5. Congress II: Design and Implementation of Interactive Systems: SPEECH RESEARCH AND TECHNOLOGY; Research Issues of Speech Input
    6. Congress II: Design and Implementation of Interactive Systems: SPEECH RESEARCH AND TECHNOLOGY; Speech in Multimodal Applications
    7. Congress II: Design and Implementation of Interactive Systems: STANDARDIZATION; Development of Standards
    8. Congress II: Design and Implementation of Interactive Systems: SOFTWARE ENGINEERING -- METHODS, TECHNIQUES AND TOOLS; Software Engineering and HCI
    9. Congress II: Design and Implementation of Interactive Systems: SOFTWARE ENGINEERING -- METHODS, TECHNIQUES AND TOOLS; New and Better Technologies for Software Engineering
    10. Congress II: Design and Implementation of Interactive Systems: SOFTWARE ENGINEERING -- METHODS, TECHNIQUES AND TOOLS; Applications of Techniques and Tools
    11. Congress II: Design and Implementation of Interactive Systems: USABILITY EVALUATION; Practical Experience of Usability Evaluation
    12. Congress II: Design and Implementation of Interactive Systems: USABILITY EVALUATION; Techniques for Usability Evaluation
    13. Congress II: Design and Implementation of Interactive Systems: USABILITY EVALUATION; Development of Usability Metrics
    14. Congress II: Design and Implementation of Interactive Systems: INTERFACE DESIGN; User Interface Design Methodology
    15. Congress II: Design and Implementation of Interactive Systems: INTERFACE DESIGN; Interface Design Examples
    16. Congress II: Design and Implementation of Interactive Systems: INTERFACE DESIGN; Advances in Interface Design
    17. Congress II: Design and Implementation of Interactive Systems: COMPUTERIZED WORK AND THE IMPROVEMENT OF WORKING CONDITIONS; Flexible Shop Floor Operations and Office Automation
    18. Congress II: Design and Implementation of Interactive Systems: COMPUTERIZED WORK AND THE IMPROVEMENT OF WORKING CONDITIONS; Workshop: Improvement of Working Conditions for Computerized Work: Where Are We and Where Will We Go?

HCII 1991-09-01 Volume 1

Congress II: Design and Implementation of Interactive Systems: COGNITIVE ENGINEERING; Advances in Cognitive Engineering

How We RAMmed Userfriendliness into Obsolescence BIBAK 403-408
  Jack F. Gerrissen
It is described why and how we involved office workers (from the application domain) in product development as an interface with reality. In a way, this enabled us to actually interact with the application domain on the basis of feature concepts and partial prototypes. Presence of the office workers in the product development activity also supported consensus management in multidisciplinary teams.
Keywords: System design, Product development, Concept formation, Project management, Software engineering
Promoting the Optimization of Preventive Maintenance Strategies in Complex Production Systems by Behavior-Outcome-Feedback BIBAK 409-413
  D. Gude; K.-H. Schmidt; A. Seeber
This contribution is concerned with the optimization of preventive maintenance strategies in complex production systems. In a simulated maintenance task the quality of the strategies depended especially on two cognitive requirements: knowledge about the maintenance strategy and knowledge about the problem space. Based on these results, a method is suggested that is expected to promote the optimization of preventive maintenance strategies by improving the knowledge in these domains. This method is called Behavior-Outcome-Feedback (BOF) and is characterized by a periodical feedback of the maintenance strategy and the resulting efficiency of the production system.
Keywords: Preventive maintenance, Efficiency, Knowledge, Qualification, Feedback
User Interaction -- A Framework to Relate Tasks, Users and Designs BIBAK 414-418
  Peter Johnson
A framework for mapping between user task models and user interface architectures is presented. A particular form of task modelling (Task Knowledge Structures) is outlined and a method of task analysis (Knowledge Analysis of Tasks). Task Knowledge Structure models can be used to represent existing user tasks and in terms of the changes designs bring to user tasks. The paper considers how task models and user interface architectures can be related.
Keywords: User interaction design, Task knowledge structures, Task analysis, User interface architectures
Browsing Cognitive Task Spaces Instead of Working on the Desktop: An Alternative Metaphor BIBAK 419-423
  Edmund Eberleh
Some of the limits and problems of the desktop metaphor are highlighted. Based on theory of goal-directed behaviour and on experimental results about mental representation of computer supported tasks, an alternative metaphor for the user interface is proposed. It maps an abstract globe on the screen and the whole functionality of the system is arranged on the surface of the globe and within the globe according to some principles. A first empirical evaluation by means of judgements of utility and usability showed acceptance of this kind of interface.
Keywords: Graphical user interface, Desktop metaphor, Goal-directed behaviour, Cognitive task space, Interaction style

Congress II: Design and Implementation of Interactive Systems: COGNITIVE ENGINEERING; Research on User Modelling

Cognitive Task Allocation: A Design Methodology BIBAK 424-428
  Sotiris A. Papantonopoulos; Gavriel Salvendy
Cognitive task allocation employs task analysis and demand/resource matching to identify and match task requirements and human and computer performance resources. The paper introduces a formal, quantitative, and domain-independent decision model of cognitive task allocation.
   Demand/resource matching is modeled as an Analytic Hierarchy Process. By means of the Analytic Hierarchy Process task functions (Level 1) are analyzed into their constituent cognitive processes (Level 2); performance criteria are set for each cognitive process (Level 3), by means of which the capacities of the human, computer, or interactive human/computer controller (Level 4) are evaluated and compared. The Analytic Hierarchy Process then integrates partial judgements of relative human and computer capacity into a global weighted average indicating the relative capacity of human and computer to perform the overall function. The Analytic Hierarchy Process was applied and evaluated in the design of cognitive task allocation in production scheduling of a flexible manufacturing system.
Keywords: Task analysis, Task allocation, Methodology
Direct Manipulation and Command Language Interfaces: A Comparison of Users' Mental Models BIBAK 429-434
  Kishore Sengupta; Dov Te'eni
User interfaces that employ direct manipulation techniques have been found to be superior to command language interfaces with respect to measures of user performance. However, there is little evidence on how users' mental models are shaped by the interface they use. This study compared mental models of users who worked with direct manipulation interfaces with those using command language interfaces. Information on users' mental models was inferred through verbal protocols and computer logs. Results indicate that users given direct manipulation interfaces form better models of the task and device, i.e., use fewer production rules, attain better task-to-device mapping and carry out fewer unnecessary steps in executing their tasks. The implications of the results are discussed in terms of learning, transfer of knowledge between tasks, and efficiency of performance.
Keywords: Command language, Direct manipulation, Mental models, User interfaces
An Adaptive Interface Design Using Neural Networks BIBAK 435-439
  Nong Ye; Gavriel Salvendy
The neural networks' ability to learn by examples is combined into the design of an adaptive interface. This adaptive menu based interface between the user and the UNIX system can accommodate differences in user's experience about the system and individual differences among users.
Keywords: Adaptive interface, Neural networks, Menu design, Learning, UNIX

Congress II: Design and Implementation of Interactive Systems: COGNITIVE ENGINEERING; Design and User Modelling

A User Modelling System BIBAK 440-447
  F. Arcieri; P. Dell'Olmo; E. Nardelli; P. Vocca
A good cooperation between users and interactive systems often requires an adjustment of the man-machine interface to the specific user needs and requirements. Adaptive systems able to change their behaviour according to the characteristics of different classes of users have been proposed. They are typically based on user models and are targeted at specific application realities. In this paper we propose and discuss a general architecture for a User Interface Management System integrating application-independent user modeling capabilities. An object oriented prototype implemented to show the feasibility of the proposed approach is also presented.
Keywords: Adaptive systems, User models, User profiles, Stereotypes, Interactive systems
A Query Facility for Schema Integration BIBAK 448-454
  Ulla Merz; Roger King
The user interface design and evaluation of an interactive schema integration tool is presented. The tool consists of a browser of the database schemas and a query facility to specify the required data and their definitions. Results from evaluating the user interface suggest that a query facility should provide functions that match the user's preferred query writing strategy and a browser of the data definitions has to provide different visual representations and search strategies to assist the user in interpreting the meaning of the data.
Keywords: User interface design, Query facility
User Centred Requirements Modelling for a Multi-User Spreadsheet BIBAK 455-459
  Charanjit K. Sidhu; Jill Hewitt; Neil Watkinson; Stephen Furner
Since their introduction, electronic spreadsheets have become a very popular tool in the office environment. This paper describes the application of the Generic Office Reference Model to carry out requirements capture for the introduction of spreadsheets in a teleconferencing environment.
Keywords: Generic office reference model, Requirements capture, Spreadsheets, Multipoint teleconferencing, CSCW

Congress II: Design and Implementation of Interactive Systems: SPEECH RESEARCH AND TECHNOLOGY; Design and Evaluation of Speech Interfaces

Voice as Interface: A Critique BIBA 463-467
  Dylan M. Jones; Philip Tucker
The revolution in information technology has made relatively little use of the auditory channel of communication. This is perhaps not surprising given that the tasks for which the computer has been used hitherto are typically ones for which there was already a corresponding manual visual version. However the prevalence of the manual visual interface may soon be diminished. As computers become more powerful and more functionally diverse, devices are being developed that can generate and perceive speech. Thus research into the effects of modality on communication behaviour is becoming increasingly salient to the human factors of interface design.
   Using voice in the human computer interface has several potential advantages but the realisation of these advantages will depend upon the sophistication of the technology and the appropriateness of its implementation.
   Speech liberates the user by allowing movement away from terminal while input and output continue. Moreover it allows interaction when the hands and/or eyes are busy. The auditory channel is well suited to bringing urgent information to the attention of an operator, regardless of the direction of the current visual focus and is not reduced even if the user is relatively remote from a terminal. Many physically handicapped users, hitherto disenfranchised by need for sight, finger dexterity and the paucity of specially adapted interfaces, come within the catchment area of speech-based computer users.
   In its most highly developed form the speech based computer could take the form of a telephone hand-set, perhaps coupled with all ordinary domestic television set. Some of the computer's intelligence may be local, but the greater burden of processing could be centrally located at a remote site.
   Yet despite the availability of speech I/O devices for some years, their introduction into day-to-day use has been relatively slow. While technological sophistication has clearly played a role in determining the development of the visual and auditory channels in the interface, a human factors perspective is also required to further explain the emphasis on the visual/manual medium.
Cognitive Evaluation for Multimedia Presentation Method -- Optimal Presentation Timings of Text and Speech BIBAK 468-472
  Ken'ichi Kamijo; Toshimasa Yamazaki
The authors propose two cognitive evaluation methods for multimedia presentations. One is to examine the cognitive process on the bases of the human memory. The other is to test subjective responses using instructions involving from the perceptual level to the cognitive level. The authors attempted to obtain optimal discrepancy between presentation timings (time lags) of text and speech, as a case study on cognitive evaluation, in the following two experiments with Japanese word presentation.
   Experiment 1 examined memory effects induced by the time lags using an incidental learning paradigm. The words, presented synchronously in the orienting task (subjective evaluation), resulted in high recognition rate. In Experiment 2, two subjective evaluation tasks, which were concerned with the impairment for word understanding due to presentation timings (cognitive level) and the existence of the time lags (perceptual level), were carried out. The subject's word understanding was disturbed by more than 0.2 second time lags, and more than 0.35 second time lags were perceptible. Moreover, we found out a non-symmetric interaction between visual and auditory functions at the cognitive level. That is, the impairment for the word understanding due to time lags in speech-presented-faster-than-text conditions was larger than the reverse conditions. The other hand, there was no such interaction at the perceptual level.
Keywords: Multimedia, Recognition, Speech, Subjective evaluation, Text
Dialogue Design for Speech Interfaces in the Office Context BIBAK 473-477
  James Monaghan; Christine Cheepen; Jill Hewitt
The two main objectives of this paper are, firstly, to demonstrate how the use of multi-perspective modelling techniques in the office environment can maximise the functionality and usability of systems designed for that context, and, secondly, to illustrate just how such techniques work in the case of the human factors aspects of an automatic dictation system. The production of the four-perspective model of a particular office situation based on the various views is linked to the generation of more abstract office representations as the basis for scenario-building, followed by the reanalysis of the results of targeted data-collection projects.
Keywords: Dialogue, Speech interfaces, Office, Scenario
Outline of a Comprehensive Assessment Methodology for Speech-Oriented Applications BIBAK 478-484
  Paulus H. Vossen
Information and communication systems with speech input or output interfaces will soon become available in various domains of professional and public life [1]. The basic technology for speech input, i.e. speech recognition algorithms and devices, has matured and is ready for implementation. The same holds for speech output technology, i.e. algorithms and devices for speech reproduction or synthesis. Furthermore, research on large-vocabulary, speaker-independent, continuous-speech recognition makes steady progress, and techniques for adequate handling of prosodic features of natural speech are under development.
   Although it will take some years before the results of these theoretical and technical advances will be applicable and visible, one should not wait any longer with the development and testing of comprehensive assessment methodologies for systems incorporating speech technology. As the temporal-acoustic interaction mode is quite different from the spatial-visual one, it is neither possible nor advisable to rely on existing design guide-lines for graphical user interfaces. This paper describes the main components of a framework and strategy for design-oriented system evaluation, and shows how this methodology has been introduced and will be worked out in a large-scale project aimed at the development of speech applications.
Keywords: Speech assessment, Assessment methodology, Speech interface, Interface design, Design methodology, Human factors

Congress II: Design and Implementation of Interactive Systems: SPEECH RESEARCH AND TECHNOLOGY; Research Issues of Speech Input

Consequences of Discrete Speech Using Speech-to-Text Technology BIBK 485-490
  Peter Day; Andreas Grunupp; Klaus-Peter Muthig
Keywords: Speech recognition, Speech-to-text technology, Task demands, Discrete speech, Human factors
Facial Expression Graphics Feedback for Improving the Smoothness of Human Speech Input to Computers BIBAK 491-497
  Tomio Watanabe; Akira Higuchi
Voice-reactive visual feedback with both cartoon-face-like graphics and model-based analysis-synthesis facial graphics, can be used to make the human-to-computer speech input environment seem warmer, more interpersonal, and more natural for the speaker. First the facial expressions used by human listeners in face-to-face human interactions were identified. Then computer graphics were used to simulate these expressions, and human subjects evaluated the effectiveness of these graphics. A model system using voice-activated input and providing feedback to the user using these graphics was tested with the object of developing more user-friendly computer systems.
Keywords: Human interface, Facial expression, Visual feedback, Speech input, Computer graphics
The Recognition Coefficient: Methodological and Statistical Issues Concerning Measurement of Speech Recognition Accuracy BIBAK 498-505
  Paulus H. Vossen
The quality of a speech recognition device depends on a large number of factors, e.g. hardware and software, vocabulary choice and word templates, pattern matching algorithms and decision rules. The ultimate test however is, whether the device accurately recognizes the words spoken by a user of the system. This paper is concerned with the methodological and statistical issues of speech accuracy measurement from a theoretical and a practical point of view.
   First we will examine some traditional approaches to measuring the accuracy of automatic speech recognition and discuss their weaknesses. Next we will present an improved approach and measure which lack the aforementioned deficiencies. Finally we will discuss the implications of our investigations for the practice of design and evaluation of speech recognition interfaces.
Keywords: Speech recognition, Recognition accuracy, Accuracy measurement, Confusion matrix, Recognition coefficient

Congress II: Design and Implementation of Interactive Systems: SPEECH RESEARCH AND TECHNOLOGY; Speech in Multimodal Applications

Software for Multi-Modal HCI Including Speech for Skilled Industrial Workers BIBA 506-511
  B. Scherff
A novel programming method for the work shop was described. A major element in the development of the programming method for welding robots involved testing whether speech input could satisfy technical and industrial science requirements. Therefore an experimental programme has been developed and the results are discussed.
SESAM: A Prototype Multimedia System Combining Computer Animation with Speech Dialogue BIBAK 512-516
  M. Niemoller; A. Aktas; U. Harke; U. Leiner; K. Zunkler
This paper gives an introduction to the objectives of our multimedia project. The longterm goal of the project is to develop a computer animation environment with a speech dialog interface. The combination of computer animation and a speech dialog interface in a multimedia system will enhance the work effectivity of a user in future. We sketch the state-of-the-art in computer animation and speech understanding and describe the current state of our prototype.
Keywords: Multimedia, Computer animation, Speech recognition
Speech and Other Modalities in the Office Environment: Some Research Results BIBAK 517-524
  F. L. van Nes
Research was carried out on the application of speech in three areas of man-computer communication: instruction, voice commands for system control and annotation of documents. As to instruction, learning was found to proceed equally fast with speech and written text; a number of subjects preferred speech. Secondly, in speech-to-text conversion, subjects preferred voice commands to manual commands for layout and typographic control, although text input was slower with voice commands. Thirdly, voice annotations are more readily made than text annotations, but processing times may be longer for voice than for text annotations. In conclusion, speech is a valuable medium for human-computer interaction, provided the applications are carefully chosen and a proper user interface is made.
Keywords: Multimodal interfaces, Office environment, Speech recognition, Voice annotation, Voice manual
An Evaluation of Speech Operated Word Processing -- A Task Based Approach BIBAK 525-529
  Mary Zajicek; Jill Hewitt
This paper evaluates and compares the usability of different word processors with speech or keyboard driven interfaces and explores the possibility of building a generic task-based vocabulary for word-processing by speech. The Speech and Language Technology Group based at Hatfield Polytechnic have developed their own speech driven word processor, VWP, which incorporates features specially designed to increase the usability of a speech input system. This is compared with the commercial word processors Wordstar and WordPerfect where a speech interface has been provided through the RABBIT system which is a transparent speech driven user interface developed by the group.
Keywords: Speech, Word processing, Generic, Task, Usability

Congress II: Design and Implementation of Interactive Systems: STANDARDIZATION; Development of Standards

Standards Relevant to European Directives for Display Terminals BIBAK 533-537
  Nigel Bevan
The European Directive on the "minimum health and safety requirements for work with display screen equipment" contains minimum ergonomic requirements for the equipment, environment and operator/computer interface. National legislation which implements the Directive is required by the end of 1992, and may make reference to relevant standards. ISO 9241 is being developed as a multi-part standard which has a similar scope to the Directive, and provides a potential means for interpreting some of the more general requirements of the Directive which include: "software must be easy to use" and "the principles of software ergonomics must be applied".
Keywords: Usability, Evaluation, Guidelines
Standards as a Means of Influencing Interface Design BIBAK 538-542
  Tom Stewart
This paper reports a short case study on the introduction of internal user interface standards in the information technology department of the headquarters of a large UK organisation. It discusses the relationship between such in-house standards and current developments in European and International standards.
Keywords: Standards, User interface, Case study, Organisation
Developing Recommendations for CAD User Interfaces BIBAK 543-547
  Andreas M. Heinecke
The reference model for CAD systems developed by the Gesellschaft fur Informatik (GI -- the German membership organization of IFIP) is a frame for classifying the functionality of CAD systems. Whereas the reference model regards the user interface as one of several modules of the CAD system, the user interface appears to the user as being the whole system. This is the reason why an additional task working group on CAD user interfaces has been established by the GI in order to develop recommendations for the design of CAD user interfaces. Proceeding and preliminary results of the working group are described.
Keywords: CAD, Design of user interfaces, Guidelines, Standardization

Congress II: Design and Implementation of Interactive Systems: SOFTWARE ENGINEERING -- METHODS, TECHNIQUES AND TOOLS; Software Engineering and HCI

Software Engineering = Human Factors Engineering? BIBA 551-555
  Peter Gorny
In this introduction to the conference stream "Software Engineering -- Methods, Techniques and Tools" some fundamental issues of the software developing process and the underlying attitudes of software developers are investigated. The author stresses the necessity of a change of the technology-centered production process approach towards the human-oriented attitude of architects.
Dialogical Software Design BIBAK 556-560
  Jurgen Pasch
Software development is not merely a mathematical or technological challenge, but a complex social process, in which the kind of communication and cooperative, creative interaction of the participants determine the quality of the collaboratively developed product. Qualified design is not primarily tied to given guidelines, but is guided by insights emerging in the design process and by the quest for quality shared by all participants. In this paper I discuss the role of models in the design process. The results of a field study show that a consensual definition of the situation is a prerequisite for dialogical design. During the design process the participants developed techniques of mutual contradictions.
Keywords: Design, Quality, Method
Control vs. Creativity: Software Engineering at a Crossroads BIBA 561-565
  G. Bradley
This paper looks at the issues raised by integrated project management support in software development environments. In particular, the difference between prescriptive and constraint-based process models is analyzed, and the suggestion is made that an object-oriented approach is one of the most promising paths to the practical application of software engineering principles in software projects.
Design and Experimental Evaluation of a New Graphical Multi-Process Debugger BIBA 566-571
  Sia Maleknasri; James D. Foley
Most program development tools do not provide for testing of multi-process programs. We have developed the Concurrent Process Environment Monitor (CPEM), which facilitates debugging of UNIX/C concurrent programs. CPEM monitors programs comprised of many concurrent processes, and informs the user of the status and interactions of the processes.
   CPEM presents information either graphically or in a more traditional textual form. A controlled experiment was conducted to assess the relative effectiveness of the two CPEM presentations. The graphical CPEM was found to be superior to the textual CPEM in helping programmers debug two different concurrent programs. Furthermore, programmers overwhelmingly preferred the graphical presentation to the equivalent textual interface.

Congress II: Design and Implementation of Interactive Systems: SOFTWARE ENGINEERING -- METHODS, TECHNIQUES AND TOOLS; New and Better Technologies for Software Engineering

The Development of Human-Computer Interface Descriptions from Organisational Models BIBAK 572-576
  J. S. Medes
This paper presents an organisational model of a typical software development team. This is used to illustrate the interaction of tasks and actors in that team that need to be supported by any proposed integrated project support environment (IPSE). The conclusion is drawn that an advanced interface management system is the key to providing integrated project support for interworking in any advanced software development group. Further, an explicit and comprehensive organisational model is essential for defining the human-computer interface requirements in such environments. The conclusions are generalised to cover any groups of users accessing multi-user application software in closely integrated working environments.
Keywords: Organisational models, Integrated project support environment (IPSE), Cooperative working
Structured Design of User-Interfaces and Knowledge-Based Design BIBA 577-581
  Axel Viereck; Egbert Schlungbaum; Peter Gorny
The authors propose a principled design of human-computer interaction following a software ergonomics phase model, which separates the conceptual design decisions affecting the work organisation on the background of the tasks, user characteristics and qualifications from the structuring of the dialog for functional requirements and from concretizing the general cognitive considerations regarding the screen layout and dialog sequencing, while the realization will transfer all design decisions into a software product or prototype. The paper describes briefly a knowledge based system for user interface development support presently under development which will counsel software developers in regard all human factors design decisions.
The Use of Co-Operation Models for Specification and Design of User Interfaces BIBAK 582-586
  R. Krickhahn; M.-J. Schachter-Radig; K. Streng
The quality of user interfaces (Ui/f) becomes more and more the crucial point for the acceptance and usability of software systems. Despite user interface toolkits and UIMS's, the necessary effort to develop state-of-the-art Ui/f's increases constantly.
   This paper discusses two major issues in developing good user interfaces in the context of several research projects as well as customer projects NTE is involved in.
  • what should the Ui/f look like and how should it behave?
  • how can these specification be transformed into an effective user interface
       design? It is shown how the analysis of co-operation behaviour in a certain environment (such as a company or an organization) is used for the specification as well for the design of Ui/f's. A representation schema for co-operation modelling is described in detail, and the transformation process towards the design based on this schema is explained.
    Keywords: Knowledge based systems, Co-operation analysis, Ui/f design, Structured development, Interaction techniques
  • Congress II: Design and Implementation of Interactive Systems: SOFTWARE ENGINEERING -- METHODS, TECHNIQUES AND TOOLS; Applications of Techniques and Tools

    Visual Programming with Reusable Specifications Described by a Conceptual Data Model- and Constraint-Based Language BIBAK 587-591
      Katsumi Okamoto; Masaaki Hashimoto
    This article proposes a visual programming system with resuable specifications described by a conceptual data model- and dependency constraint-based language. The visual HCI (Human-Computer Interaction) with the language improves the comprehensibility and extensibility of specifications to be reused. Moreover, the HCI assists in macroscopic comprehension using the conceptual schemata, and supports the extraction of a program specification to be converted into a C program.
    Keywords: Visual programming, Software reusability, Conceptual data model, Entity-relationship model, Program specification description language
    Selective Information Hiding: A Debugging Technique to Address Some of the Problems of Novice Programmers BIBAK 592-596
      Ray Waddington; Roger Henry
    We discuss a finding from our observations of novice computer programmers. This finding reveals one of the major barriers to effective debugging performance: novices are unable to distinguish between that information provided by the programming environment which is relevant to diagnosing their bugs, and that which is irrelevant to the task. We hypothesise that if irrelevant information is hidden from the novice programmer automatically, it is likely that debugging performance will improve.
       We propose a debugging technique designed to do this. We label this debugging technique selective information hiding. It can be implemented using existing compiler technology. A debugging aid that uses this technique hides from the view of the user those program's expressions (and their value) which were not defined or referenced in the most recently executed statement(s) of a program. We present the design of a prototype user interface to a debugging aid which performs selective information hiding.
       We discuss future research directions and relate this work to the wider context of user-centred design of software development tools.
    Keywords: Debugging, Novice programming skills, Selective information hiding
    Assessing Usability Evaluation Methods in a Software Development Process BIBAK 597-601
      Nobuko Kishi; Yosuke Kinoe
    Several usability evaluation methods are examined to assess their effectiveness in a practical software development process. Four criteria were used: the times at which an evaluation can be conducted, the type and the number of usability problems it can detect, the workload it involves, and the variations in measurement caused by evaluators. These criteria were applied to four evaluation methods that were used in an actual development process: a simplified check-list, mock-up prototyping, a formal analysis of memory load, and a verbal protocol analysis. As a result of the examination, several methods of categorizing usability evaluation methods were proposed -- formal/heuristic, specification/system, and hypothetical users/real users. With the above criteria and categorizations, it was concluded that a single evaluation method could not have discovered all the usability problems, and that several evaluation methods skillfully combined can detect a larger number of usability problems in an actual development process.
    Keywords: Usability evaluation, Usability testing, Software development process, Verbal protocol analysis, Formal analysis
    Automatic Testing for the Applications with Direct Manipulation Interfaces BIBAK 602-606
      Gang Lu
    This paper describes a tool in which we test automatically our applications built on top of OSF/Motif graphical toolkits. The tool itself is also based on X Window, so a good portability is guaranteed. The testing tool allows to generate test cases in a WYSIWYG style. The testing tool provides also classical facilities like editing test cases and setting stop points.
    Keywords: Interface testing, Interface building, Tool, X window, Direct manipulation

    Congress II: Design and Implementation of Interactive Systems: USABILITY EVALUATION; Practical Experience of Usability Evaluation

    Applying Formal Verbal Protocol Analysis to Practical Usability Evaluation BIBAK 609-613
      Yosuke Kinoe
    Verbal protocol analysis can be an effective means for evaluating a product's usability. We carried out the practical usability evaluation of an integrated CAD system by applying a formalized verbal protocol analysis method, the VPA Method. This method is expected to provide a stable and reliable analysis procedure for maintaining consistency in the data analysis of bulk verbal reports from users.
       This study showed that the "thinking-aloud + the VPA Method" was an effective technique for relatively large-scale practical usability evaluation. A wide range of usability issues, including users' general requirements for a product, were identified on the basis of the rich verbal protocol data, according to the standard analysis procedure of the VPA Method. For the development team, the persuasiveness of the evaluation results was enhanced by the overall effects of open testing using the thinking-aloud method, data analysis using a formalized verbal protocol analysis procedure, and management of the testing by inter-disciplinary teamwork.
    Keywords: Usability evaluation, Usability testing, Verbal protocol analysis, Thinking-aloud, Practical study
    Procedures to Evaluate the Usability of Software Products BIBAK 614-620
      G. Zulch; J. Englisch
    Usability evaluation requires appropriate procedures according to request and scope at different occasions. For this purpose an evaluation system has been developed. The scheme of the evaluation system can be used for designing and carrying out adapted usability evaluations. An evaluation with ISO WD 9241-14 of three CAD user interfaces shows one application of the evaluation system.
    Keywords: Usability evaluation, Evaluation system, Procedures, Conformity testing, CAD
    User Interfaces for Public Information Systems BIBAK 621-624
      D. Felix; W. Graf; H. Krueger
    User-interfaces for public systems require special care because the user-group is very mixed. Everybody is a potential user and basically no training is possible. For this reason, a research-project was initiated to investigate the possibilities of introducing new technologies and using the technical possibilities available today to aid the users with their task. As an example of a complex system a ticket vending machine was chosen. The machine uses a 19" colour monitor with a touchscreen as input and output-media. The aim of the research was to investigate with what kind of structuring of the task the users feel best. Two different ways of structuring were prototyped: The first offered all selectable parameters on a screen with defaults for most of the values, the second presented each parameter on a separate screen, leading the user through the system step-by-step. A field experiment with a complete prototype of the ticket vending machine looked at user performance and reactions in a realistic environment (train station). The results show an over-all preference for the step-by-step path through the process of composing a ticket, but further analysis is still in progress.
    Keywords: Public, Information, Rapid prototyping, Touchscreen, Strategy
    A Human Factors Model for Evaluating Advanced Telescience System Throughput BIBAK 625-629
      Richard F. Haines
    The ability to carry out high quality scientific procedures remotely from one's laboratory or office is referred to as telescience. This paper presents a human factors oriented throughput (Tp) model useful for evaluating advanced manned telescience systems. The model's first two input parameters involve nominal and off-nominal predicted events. The first of these calls for a detailed task analysis while the second a contingency event assessment. The last two parameters involve measured human performance and continuous system performance events. Using digital simulations and identical, representative, random data an expression combining these four parameters was found which yielded the smallest output variance. Manned simulations are underway at Ames to further evaluate this throughput model.
    Keywords: System throughput, Validation, Telescience, Modelling, Human factors

    Congress II: Design and Implementation of Interactive Systems: USABILITY EVALUATION; Techniques for Usability Evaluation

    An Integrated Design and Evaluation Tool for Dialogue Systems in Vehicles BIBAK 630-634
      G. Nirschl
    A program tool is presented that integrates design and evaluation of dialogue systems in motor-cars. The design component is realized with a prototyping tool for user interfaces. The evaluation component is based on an approach modelling the driver's task-related knowledge by means of production systems. Cognitive complexity measures of driver-vehicle interactions, such as visual distraction, learning and execution time are derived. The integration of design and evaluation is achieved by generating the production system models of the driver's tasks automatically when the tasks are executed with the prototyped dialogue system.
    Keywords: Driver-vehicle interaction, Cognitive complexity, Driver modelling, Dialogue evaluation, Rapid prototyping
    User Interface Design and Evaluation -- Application of the Rapid Prototyping Tool EMSIG BIBAK 635-639
      Annette Kaster
    Rapid Prototyping involves the creation of software prototypes of user interfaces for demonstrating and evaluating design concepts. The rapid prototyping tool EMSIG supports a system developer in the following steps: to simulate a physical system, to design an interface for this system, to test the quality of the design and to evaluate alternative designs. The paper describes briefly the tool and an experimental investigation, where EMSIG had been used to simulate a trim system of a submarine as well as to construct alternative interface layouts for comparison and evaluation.
    Keywords: User interface design, Rapid prototyping, Rapid prototyping tool, Simulation, Evaluation
    Comparison of Interaction Techniques BIBAK 640-645
      H. Widdel; J. Kaster
    An experiment was conducted to investigate the effects of different interaction techniques on dialogue performance. Results show a time advantage of using pull-down menus in comparison with form-filling interaction techniques, when a textprocessing task had to be performed, which required a high amount of knowledge. A micro-analysis of the data demonstrated very short action sequences on keystroke-level but a high error rate in this context. Time differences were insignificant, when tasks required less knowledge.
    Keywords: Interaction technique, Menu, Form-filling
    Analysis and Evaluation of Cognitive Tasks Using an Information Transition Model BIBA 646-650
      Kenji Itoh; Takao Enkawa
    In the present study, we propose a two-stage method for analyzing and evaluating cognitive tasks based on human information processing, and apply it to the operation of a high-speed train. In the analysis stage of the proposed method, human activities in performing a task are modeled using a proposed modeling scheme called Information Transition Model. Next, in the evaluation stage, potential problems in the existing system are identified through calculation of the reliability and work load based on the constructed task model.

    Congress II: Design and Implementation of Interactive Systems: USABILITY EVALUATION; Development of Usability Metrics

    What is Usability? BIBAK 651-655
      Nigel Bevan; Jurek Kirakowski; Jonathan Maissel
    The paper relates different approaches to usability based on the product, the user, ease-of-use, actual usage and the context of use, and proposes that usability should be defined as the ease of use and acceptability of a product for a particular class of users carrying out specific tasks in a specific environment. Criterion levels for measurements of attitude and user performance determine whether the design of the product is successful in achieving usability. Diagnostic evaluation of usability problems may be based on analysis of user interaction or comparison of product attributes with guidelines.
    Keywords: Usability, Evaluation, Guidelines
    Indicators of Usability Based on Performance BIBAK 656-660
      Ralph Rengger
    As part of the ESPRIT MUSiC Project, literature recently published on measuring the performance of Human Computer Interaction (HCI) was I surveyed. 'Indicators of usability based on performance' were identified and assigned to generic groups. Using the experience of NPL and the HUSAT Research Institute in Loughborough, coupled with evidence from the literature survey, a set of indicators of usability based on performance were proposed for validation. The NPL work was jointly supported by the CEC and by the UK Department of Trade and Industry.
    Keywords: Usability, Performance, Metrics, Measures, Evaluation
    The Value of Psychophysiological Measures in Human-Computer Interaction BIBAK 661-665
      Marion Wiethoff; Albert G. Arnold; Edo M. Houwing
    In testing interface usability several aspects are relevant: effectiveness, efficiency and satisfaction. In this paper, emphasis is on user efficiency, and this is operationalised by the ratio of mental effort and performance. Valid and reliable indices for mental effort investment in a human computer setting may be physiological measures. An inventarisation of psychophysiological measures has been completed in the context of an ESPRIT project [1]. Heartrate variability, temporal indices of respiration and cortisol excretion are discussed in this respect. It is concluded that heartrate variability may be an appropriate indice for mental effort in human computer interaction settings, and that saliva cortisol may be an appropriate to measure of feelings of lost control.
    Keywords: Usability, User efficiency, Interface evaluation, Physiological measures, Mental effort
    Predictive Metrics for Usability BIBAK 666-670
      Dirk Gunsthovel; Tom Bosser
    Predictive metrics of usability are developed to permit the prediction of aspects of usability from formal specifications of user tasks and device functionality. We use the SANE model of cognitive skills as a basis for calculating various indicators of usability, mainly relating to the complexity of user procedures. The validation process is analogous to the validation of psychometric tests, but limited by the impossibility to draw representative samples from the space of design alternatives.
    Keywords: Usability, Metrics, Evaluation, Predictive

    Congress II: Design and Implementation of Interactive Systems: INTERFACE DESIGN; User Interface Design Methodology

    A Usability Engineering Approach to the Development of Graphical User Interfaces BIBAK 673-677
      R. Gimnich; K. Kunkel; L. Reichert
    We outline a methodological framework for the development of (direct manipulation) user interfaces, which enables to integrate usability issues into the software development process from the very beginning. The focal point of this approach is a psychologically based task analysis documented by means of a formal task description, which serves as the central reference for designing and testing the usability of the user interface under development. A syntax-driven graphical editor delivers support for the method.
    Keywords: Usability engineering, User interface development, Methodological framework, Formal task description, Graphical task description editor
    XIT -- A Multi-Layered Tool for User Interface Design BIBAK 678-683
      Jurgen Herczeg; Hubertus Hohl; Thomas Schwab
    XIT is a tool for building user interfaces for the X Window System. Different system layers are provided corresponding to different abstraction levels on which user interfaces may be constructed by user interface programmers, application programmers, or end users. The system includes toolkits with object-oriented programming interfaces based on Common Lisp and CLOS as well as graphics-oriented, interactive tools. This paper describes the different components of XIT and how they are employed to design user interfaces.
    Keywords: User interface toolkits, Graphical user interfaces, Construction kits, Visual programming, Rapid prototyping
    U-Face: A User Interface Design System Based on Multiview Model BIBAK 684-688
      Tomonari Kanba; Osamu Hashimoto
    A novel user interface design methodology called the Multiview Model (MVM) is proposed, and a user interface design system, U-face, developed with this methodology is shown to contribute to improved user interface quality. The MVM focuses on the full variety of user interface aspects and consists of three operational components: design, operational simulation, and verification. U-face is a design system mainly for use with menu-driven application software on generic terminal screens. It provides design views, a simulation view, and verification views. The design views consist of a screen layout view and an operation rule view. The simulation view is used to check each operation, such as runtime, step by step. The verification views include displays of a screen transition network, a mode sequence diagram, and a key-binding graph. Simulation views and verification views are automatically produced from a designed interface. U-face represents an important new step in interface design because it allows designers to verify various aspects of their creations from points of view different than those used in the design process itself. It may be expected to contribute significantly to the development of improved interface quality.
    Keywords: Design tool, Iterative design, Multiview model, Rapid prototyping
    Action Facilitation -- A Non-Formal Approach to the Design of Usable Systems BIBAK 689-693
      Albert G. Arnold; Bert Zwep
    In this contribution a number of approaches to the design of usable systems are considered. A distinction is made between formal and non-formal approaches. The formal approaches are characterized by the use of models containing formal notations. The action facilitation approach is presented as an instance of a non-formal approach. The implications of this approach for user oriented interface design and evaluation are discussed. Further, the question is raised whether a fruitful integration is possible between formal approaches and the action facilitation approach.
    Keywords: Human-computer interaction, Usability, Formality, Evaluation

    Congress II: Design and Implementation of Interactive Systems: INTERFACE DESIGN; Interface Design Examples

    A Reasoning-Level User Interface for Spatial Layout Planning Problems BIBAK 694-698
      P. Banerjee; C. L. Moodie; R. L. Kashyap
    An user interface is developed for spatial layout planning problems with embedded reasoning ability to refine the decision making process in multiobjective optimization from a human problem solving orientation. A combination of a "communication-oriented" approach as well as a "problem solving-oriented" approach is used. The embedded reasoning capabilities leads to a reasoning-level user communication. The problem solving involves a technique known as "causal ordering" to reduce the entire multiobjective problem, with many nonlinear objectives, into sets of single objective linear optimization problems. The nonlinear objectives are not explicitly stated but are derived from the structure of the problem.
    Keywords: User interface, Object-oriented programming, Qualitative reasoning, Facilities planning
    Computing Access in Public Spaces: Design Lessons Learned BIBAK 699-703
      Rachelle S. Heller
    "Every designer wants to build a high quality system that is admired by colleagues, celebrated by users, circulated widely and frequently imitated" (Shneiderman, 1987, p 8). Interactive multimedia in public spaces represents the biggest challenges for the designer and the implementor. This paper identifies the characteristics for computing access in public spaces, highlights design issues, and describes the implementation of one system.
    Keywords: Public access, Interactive multimedia, Design
    Criteria for Designing User Interface of CIMple, A Toolbox for CIM Consultants, Planners and Vendors BIBAK 704-708
      N. Banerjee
    This paper gives an overview of a user interface design guideline defined for the ESPRIT Project CIMple aiming to construct user-driven and configurable tool set supporting planning and implementation of CIM in SMEs.
    Keywords: User interface design guideline, CIM planning desktop, CIM, SME, User-driven and configurable tool set
    Ethnographic Workflow Analysis: Specification for Design BIBAK 709-715
      Danielle Fafchamps
    Ethnographic workflow analysis is a design methodology developed to study information-related behavior in the work place. Designers in the Physician Workstation Project used this methodology to study the work practice of physicians. This paper describes a minimal conceptual framework and techniques for data collection and analysis, presents ethnographic data collected in the teaching and private clinics of two health care institutions, and illustrates the translation of ethnographic data into functional specifications for the design of the physician workstation interface.
    Keywords: Clinical applications, Design methodologies, Ethnography, Observational studies, Physician workstations

    Congress II: Design and Implementation of Interactive Systems: INTERFACE DESIGN; Advances in Interface Design

    Gestures in Multimodal Human-Computer Interaction BIBAK 716-721
      Karl-Heinz Hanne
    The basis of the work described in this paper is the attempt to build on well established interaction possibilities between humans, which are natural to users without the computer, and to bring some of advantages of natural interaction into human-computer interaction. This cannot be done in one step, but some of the paradigms and the human factors related findings or restrictions can serve as models for advanced and multimodal interfaces. One aim is to improve Human-Computer Interaction by gesture interaction in order to achieve multimodal interaction. Developed systems for gesture interaction (character recognition of handwritten text, and correction signs) are presented.
    Keywords: Multi-modal, Gestures, Recognition, Combined interaction, System
    Conceptual Models in the Design Process of Direct Manipulation User Interfaces BIBAK 722-727
      Th. Kuhme; G. Hornung; P. Witschital
    Conceptual models of interactive software systems are suited to explain the static appearance and the dynamic behaviour of user interfaces, but they usually do not incorporate the interactive design process of user interfaces. This paper illustrates which models are relevant for interactive design and how they relate to each other. An approach called "Direct Composition" of user interfaces is presented. With this approach the user interface objects comprise aspects of manipulation visualization, and construction, and can take different roles according to the models involved.
    Keywords: Conceptual model, User interface design, Direct manipulation, Direct composition, Dialogue model, Processing model, Interactive system, Design environment
    An Experimental Study of the Granularity and Range of the Undo Function in User Interfaces BIBAK 728-732
      S. Lenman; J.-M. Robert
    An experimental study was carried out on the granularity and range of the undo function in user interfaces. Eight subjects were asked to report on how they would undo word processing tasks they had just carried out. The main results show that several actions are integrated into UNDOs, that there are large between-subject differences in undo strategies and that layout and content modification actions were more often included in the reported UNDOs than other actions.
    Keywords: UNDO, Backtracking, Granularity, Range, Error recovery
    Integrating Usability Evaluation with Systems Design BIBAK 733-737
      Judy Hammond; Brendan McManus
    To improve the quality and usability of a finished system or product and eradicate or minimise the impact of usability problems identified during evaluation, requires a total integrated team approach to systems development. How is this to be achieved? This paper examines processes used in a real working environment and provides some solutions to this question. It describes evaluation techniques used by a large corporation in its quest to improve the usability and total quality of its products. Benefits are noted, particularly in terms of the integration of usability evaluation with systems design and its effects on the total product team.
    Keywords: Systems development, Usability evaluation, Usability testing, Human-computer interaction, Industry application

    Congress II: Design and Implementation of Interactive Systems: COMPUTERIZED WORK AND THE IMPROVEMENT OF WORKING CONDITIONS; Flexible Shop Floor Operations and Office Automation

    Task and User Adequate Design of Human-Computer Interfaces in Production -- The CNC Programming Example BIBAK 741-747
      K.-P. Fahnrich; M. Thines; C. Raether
    For CNC machine tools a graphical-interactive programming environment has been designed and implemented by means of a user interface management system. The system has been evaluated in the laboratory by experts. In a last step it has been evaluated in a field test carried out in a user organisation.
    Keywords: Graphical-interactive programming environment, CNC machine tools, System design, System evaluation
    Design of an Adaptable/Adaptive User Interface Management System in Production BIBAK 748-752
      Sinisa Zimek
    Motivated by the lack of ergonomic design and the deficiency of todays human interfaces in matching the needs of non-computer professionals in the area of production, an adaptable/adaptive user interface management system (uims) was designed. The paper describes the architecture model of an adaptable/adaptive uims containing an user modelling, a task modelling and a strategy component. Further, some useful representation notation for the components are discussed, well-suited in a software engineering tool for adaptable/adaptive uims.
    Keywords: Adaptable/adaptive user interface, Software ergonomy in production, User interface management, User modelling, Task modelling
    Iconic Interface for Interactive Fiction Retrieval in Libraries Based on a Cognitive Task Analysis BIBA 753-762
      Annelise Mark Pejtersen; Finn Nielsen
    The BOOK HOUSE is a system designed for fiction retrieval in public libraries based on a cognitive task analysis and intended to support casual, novice end users' retrieval and decision making. The structure of the work domain is represented by a metaphor familiar to users, which provides the proper associations for the information retrieval task. An iconic interface display of a complex, rich information context within which they can navigate freely without being constrained by the system provides the capability of information retrieval by means of visual exploration of icons representing book contents. Search procedures are performed through direct perception and manipulation of the iconic objects of the BOOK HOUSE.

    Congress II: Design and Implementation of Interactive Systems: COMPUTERIZED WORK AND THE IMPROVEMENT OF WORKING CONDITIONS; Workshop: Improvement of Working Conditions for Computerized Work: Where Are We and Where Will We Go?

    The Design of Computerized Work According to Human Needs -- R&D-Promotion of Software Design in the Work and Technology Programme BIBA 763-768
      Constantin Skarpelis; Heinz Thunecke
    Although the future of work looks much brighter than scenarios of the 70ies would make us believe much R&D activity is needed for the design of computerized work according to human needs. Principles of the "Work and Technology Programme" by the Federal Minister for Research and Technology are outlined focussing on the promotion of software design projects. Finally some future subjects for R&D-promotion in this field are discussed.
    Competence, Learning and Roles in the Workshop BIB 769-775
      Bengt Edgren
    Error Management or Error Prevention: Two Strategies to Deal with Errors in Software Design BIB 776-782
      Michael Frese
    New Approaches in Software Engineering for Interactive Systems BIBAK 783-787
      A. Beck; J. Ziegler
    Improvements of working conditions both for the software engineer and the user require more suitable methods and tools. Currently, most software engineers and programmers still use methods and tools developed in the 70's and earlier. What they need are practical methods and tools which reduce life cycle costs and gain productivity at the same time. Moreover, there is an increasing demand for systems, which consider the user's needs and are better suited for their tasks. The most important approaches of today are: user participation and prototyping, automated support of style guides, task analysis and modeling, CASE tools, UIMS and collaborative work, and object-oriented design. TASK, a method currently under development will be introduced.
    Keywords: Participation, Prototyping, Style guides, Task analysis, CASE