Bang & Olufsen is a manufacturer of high-end home electronics, mostly integrated audio and video systems, TVs, VCRs, radios, and the like. B&O wishes to offer the customer "the best of two worlds"the emotional experience and technical capability.
User interaction is one of the core competencies at B&O. The company's core competencies are design, sound, picture, user interaction, and system integration. These have been selected for their contribution to the excellence of user experiences and the uniqueness of B&O products, which is how B&O achieves market differentiation. This means the user interface of B&O products should attain a high level of excellence, which will make the product stand out when compared to others on the market.
At B&O, we want the interface to be transparent to the user, with a playful feeling that will make its use enjoyable and inspiring.
Using home electronicswatching television, listening to the radio or CDsis different from conventional human-computer interaction (HCI) in a workplace setting. Both the scope of possible uses, as well as the range of operations that are possible and necessary in order to control audio/video (AV) products, are very limited in comparison to the range of possibilities when using computers. However, in reality, most AV systems are controlled by designated computers, though they are usually small and well concealed. In any case, the looming invasion of cyberspace into the home by the introduction of digital TVs and radios, electronic program guides (EPGs), access to Internet via TV, and interactive media such as set-top boxes is making the problem of navigating in virtual environments, known from HCI and related fields, relevant to the field of home AV electronics.
There are some obvious differences between home electronics and HCI in the workplace. First, consumer electronics are intended for leisure activities. Users are thus typically inherently motivated to use the equipment, and their activities are directly related to the gratification of their own personal needs. Second, skilled operation of equipment in the workplace is presumed to require the acquisition of special knowledge, which is not presumed to be attainable by everyone. In contrast, home electronics are for nonprofessionals without special training or previously acquired knowledge. AV equipment for use in the home should be so easy to operate that all users can do so immediately.
Some developments in home electronics resemble developments in HCI. One example is the evolution of the interface from one (hardware) button one function, via command-based software control with prompts and feedback presented on one-line displays, to the selection of options on screen menus, and the recent use of simple graphics (for example, bar graphs showing the levels of brightness, color saturation, and contrast). This progression of development signals a new era of control by means of fully graphical displays.
Some developments in home electronics resemble developments in HCI.
Some people would perhaps deny the need for full-screen graphical displays to control such "simple" things as TVs and stereo units. It is still the case that the interfaces of many everyday utilities possess only rudimentary means of giving feedback on the immediate effect of operations. This may be adequate, say, for volume control, but presents a problem in relation to, for instance, tuning a receiver, or timer programming of a recording on a VCR. Here the achievement of the intended result may not be immediately perceivable, so other means of communicating the result of the interaction are needed. The increasing complexity of products with regard to programming possibilities, use of EPGs, browsing the Internet, and accessing film and music servers makes the need for advanced interface technology obvious.
A Question of the Right People, the Right Idea, and the Right Way
At B&O, development of user interfaces takes place in a team comprised of a team leader, a user interface designer, a psychologist, a member from "Idealand," an ad hoc associated software developer, a narrator, and an integrator.
The motivation for this range of different professions being involved in the development team is to ensure that several professional perspectives are brought to bear on the design issues in order to achieve a synthesis that will benefit the user. The establishment of a development team is a relatively recent occurrence, as part of an effort to increase the impact of the user's point of view on interaction quality.
The user interaction design group is responsible for the specification, development, and verification of the user interface. A few years ago, it was decided to establish a usability test facility in order to question and enhance the quality of products' user interfaces. Since the usability test is the responsibility of the development team itself, there is no need to "transfer" results between departments, as is typically done in other companies. This integration of test and development results in a high degree of synergy.
The team leader has the responsibility of assuring that the customer will have an excellent experience when operating a B&O product. So it is essential that the right team members are obtained. The team leader is enlisted to generate a consensus on the user interaction concept within the widely differentiated user interaction team. The team leader has to ensure external consensus with the concept manager, the product manager, the technical product manager and the project leader.
The task of the designer is to create the user interface so that it contains and expresses the idea and identity of B&O. The designer must maintain the idea of the interaction concept throughout the process, and make sure the user interface contributes to the product as a part of the user experience. The designer's competence, acquired through years of creative translation of senses and feelings to design experiences, is the basis of the user interaction concept.
The coordinator's job is to ensure that the user interaction concept is in harmony with the product concept, the product identity, and B&O as a brand. The coordinator comes from "Idealand," which is the concept development group that creates all new product concepts at B&O, and represents the product concept in the interaction concept design process.
The psychologist ensures the transparency of the interface (that it lets the user do what is needed to achieve a particular goal without thinking about the operations), allowing the conscious focus to remain on the goal. This is done by applying a general psychological understanding of the user's activity in interaction with technology. In key areas, usability tests must be carried out with real users. Such tests are carried out partly in the usability lab and partly in field studies.
The software developer is an ad hoc associate in the group. This person comes from the field of technology or product area currently being developed; for instance AV, multimedia, or telecommunications. As such, the software developer has a special knowledge of the product area being worked on, and brings this knowledge into the concept of user interaction. The software developer must continuously assist with the development of models to probe the interaction concept. It is the software developer's responsibility to ensure that the interaction concept can be implemented, to map out requirements for construction, and to maintain product requirements.
The task of the narrator is to make sure that the user interaction concept is communicated to the user via the user's guide. The narrator works out the key elements of user guidance in cooperation with the other members of the team. The user's guide must support the users' interaction with the product, generating an understanding of the principles of the user interaction concept.
The integrator must ensure that the interaction concept is an integrated and consistent whole across all functions of the product, and that it is compatible with other B&O products. Consistency means that the same principles of operation must characterize the whole user interface. This consistency has a substantial influence on the transparency of the interface. The integrator guarantees that the interaction concept is written down as product requirements.
Development of the user interaction concept is initiated when an idea or a problem needs to be solved in a certain field of technology or product area. As an example, a software developer from the Multimedia department proposed an idea of operating traditional A/V sources from a PC, and demonstrated the technological possibility. The designer adopted this idea and created a design model, which was, in fact, the first proposal for a user interaction concept. The model now had a B&O look and feel, but could it be operated?
The model was examined by the interaction design team, through a number of meetings and lively discussions. The model was continuously updated and elaborated into working prototypes and later ended up as the finished concept of user interaction (specification of the user interaction concept), which was then incorporated in the product.
During the early 1990s, B&O decided to establish a usability lab as part of the user interaction design team. The test facility comprises a comfortable test room furnished as a living room, and equipped with four video cameras controlled from an adjacent control room, from which the tests can be monitored.
Apart from a few field studies, most direct involvement of users in the design process has up until now been achieved by standard usability tests, employing an experimental "think aloud" method. Test results are not communicated in formal reports, but are taken directly into the development process by the psychologist and by having designers and developers observe the test directly.
It has, however, proved problematic to handle usability issues by testing in the design phase, when the designers are trying to create a particular product. This is because B&O products must have the distinctive B&O identity, so "user-driven design" (see the Buur and Bagger article in this section) is out of the question. If, however, the test is carried out when the design has been concluded, only minor changes are possible. This situation has resulted in a decision to aim at separating usability testing and usability development in a more fundamental sense.
Regarding testing, we aim to establish a permanent panel of users who are able to test products in a way comparable to the sound and vision panels that today evaluate the quality of sound and pictures on B&O products. This may sound controversial, but we believe it's possible to operationalize the evaluation of ease of interaction in such a way that it becomes analogous to the assessment of sound and picture quality.
Usability is a more complex issue as the evaluation is not confined to perception in one sensory modality, nor to parameters of sensory modalities alone, but entails numerous motivational and cognitive factors; learning, habit formation, cultural stereotypes, issues of symbolic communication, reasoning, decision-making, trouble-shooting, and so forth.
It has proved problematic to handle usability issues by testing in the design phase, when designers are trying to create a particular product.
Because of this, it will be necessary to conduct tests with various categories of users in order to capture specific usability features. In order to overcome the limitations of the experimental method, other research methods will be necessary (field studies, interviews, and surveys), as well as the collection of feedback from the market. In addition to usability testing, it is also intended to try a "theory-driven" development of usability.
From Trial and Error to Development Based on Theory
B&O considers work on usability to be a continuously developing process. We are constantly experimenting, modifying, and developing our repertoire of methods according to the need to enhance the usability aspects of new products.
The conventional usability test can be characterized as a comparative, empirical approach where the construction of improved new designs is done in a trial-and-error mode and the evaluation is post hoc. Something is constructed that seems good, conceivably better than the previous version, and is then subjected to a test by users whose use activity is monitored, and their problems of operation and verbal comments are recorded and analyzed. On the basis of these results, possible problems are amended, and a refined product is created.
Using this line of approach, the cognitive mechanisms behind the desired qualityease of operationremain implicit, and the reason why a product is "better" is not necessarily untangled. Alternatively, by taking a theoretical approach, this may be made explicit and serve as a starting point for development efforts.
B&O wishes to get to the core of what we call "intuitive user interaction" . This calls for interfaces in which functions are clearly represented and interaction is supported by feedback and indications, organized in accordance with known principles of the organization of perceptual and cognitive functions.
Users tend to explain the operation of screen-based user interfaces by using terms of handling of objects and interacting with agents in a virtual space. From the viewpoint of activity theory and ecological psychology, this is a reminder that the fundamental principles of human activity, perception, and cognition are rooted in the evolution of the human species and its cultural historical development . The metaphorical descriptions may reflect general and fundamental principles of cognition also employed in relation to the use of modern technology.
If we present information on the interface as concrete versions of various user metaphors of scenes, objects and actors, users can draw upon instinctive capacities for direct pick-up of perceptual information and intuitive cognitive functions. When the users engage a function, they are operating in a functional "space." If movements in this space and the handling of functional objects are indicated by transformations in the appearance of the interface, obeying principles defined in "ecological optics" ; then the activity and its results will be directly perceivable.
Some functions may be so complex or so abstract that they cannot be perceived directly. This necessitates the use of linguistic information and presupposes some kind of learning. In order to explain unknown functions, the linguistic information must be based on a careful consideration of the user's prior conceptual understanding of technological objects and functions, and the most appropriate way to map new functions to this knowledge.
In order to support the use of programmed functions and the learning of new functions, the normal organization of human communicative processes must also be considered. The limited possibilities of current technology to access the context of communication must also be taken into account .
We intend to investigate the most basic functions first, and to utilize available technological means to implement the knowledge we gain in suitable interaction principles. In accordance with this aim, our plan for development of intuitive user interaction has been divided into three steps: 1) Applying the principles of ecological optics to transformations of the visual interface (in order to support the direct immediate and unprocessed perception); 2) Exploring the organization of conceptual knowledge of technology in order to present information in accordance with the users' background experience in operating technological equipment; and 3) Investigating intelligent forms of communicative interaction (dialogue, search mechanisms, agents) and possible technologies for their implementation in user interfaces.
©1999 ACM 0002-0782/99/0500 $5.00
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
The Digital Library is published by the Association for Computing Machinery. Copyright © 1999 ACM, Inc.