Advanced Technology Corporation
Human Interface Testing Center
Kanagawa 244, Japan
Date Visited: May 26, 1995
Report Authors: J. Hollan and R. E. Kraut
J. Hollan R. E. Kraut
NTT's Human Interface Testing Center (HIT is also an acronym for Human Information Technology) is chartered to bring human interface development techniques to bear in the implementation of NTT's products and operations systems. At the most abstract level, its goal is to ensure "a harmony of people, information and technology" in NTT's products and operations systems. The Human Interface Testing Center provides a full range of HCI consulting services for NTT, ranging from the design and evaluation of multimedia communication services to the redesign of NTT work processes and the software systems to support them to consulting on collaborative software design and development. The HIT Center also provides usability testing and interface design services. Takaya Endo, Director of the HIT Center, previously was in charge of the NTT Human Interface Labs and served as the first director of those labs. NTT views its R&D activities as evolving through a series of ages: Age of Reconstruction (1945-1965), focusing on domestic production of communication technology; Age of Growth (1966-1980), featuring the introduction of new communication services; Age of Competition (1981-1990), providing world-class R&D; and Age of Cooperation (1991-present), providing R&D that leads the world.
In contrast to other HCI labs the JTEC team visited, the HIT Center has a broad view of human-computer interaction. It encompasses both organizational and social issues in the use of computing and individual usability. This idea is captured in the center's goal of constructing macro and micro interface structures. The goal is to aid in the design of both work processes and information systems that are consistent with the work process. For operations systems, the center subscribes to a process of participatory design, in which users of the systems are coparticipants in the design processes. Through interviews, questionnaires, and observation, designers develop an understanding of what their clients' main tasks are, which ones are problematic tasks and their priority, and what are the political and social constraints on the potential solutions. Rapid prototyping (often using a prototyping tool such as Macromind Director) provides a concrete image for discussion between clients and system designers, and serves as a basis for negotiation and iteration.
The HIT Center has, for example, applied this participatory design and iteration technique to the development of the design guidelines for a large (approximately 3,000 pages) on-line documentation and operations system for NTT customer service representatives. The center delivered design guidelines and a requirements document to the system developers in the form of an active macromind director document. The document embodied two overarching design principles -- "I know what I'm doing" and "correct reasoning and fast decisions." In performing a task, such as helping a customer to order a new ISDN line, the interface was configured to always show users where they were in the ordering process and to allow them to quickly, from a single screen, perform the various tests and ordering functions that are part of this routine.
The major focus for past projects has been on designing/testing interfaces and consulting about work practice and education. In addition to the customer service project mentioned above, other interface projects include testing and design of a telephone manual, development of guidelines for human interface testing, testing and designing a range of graphical user interfaces, experiments for use for supporting map operation on a display screen, behavioral analyses of virtual reality environments, and testing and designing multimedia network services.
The staff members of the HIT Center have close ties to universities and to professional organizations. There was discussion of this dual life of the key members of the HIT Center. While it creates a complex schedule, they view it as being necessary to their efforts, to future hiring, and to supporting the development of HCI within Japan. Takaya Endo, Director of the HIT Center, is a lecturer at Waseda University. He is chair of the Human Communication Engineering Group of the Institute of Electronics, Information, and Communication Engineers, and chair of the Representation and Interface Group of the Japan Cognitive Science Society. He also serves on a variety of committees: Informational Ethics Research Group of EIC, Ecological Research Group of the Information Society, and the Science and Technology Forum of the Science and Technology Agency. Shuichi Kato, who led the effort to develop human interface guidelines for NTT, serves on the Visual Interface Committee of the Institute of Japanese Standards and is a member of key interface design professional organizations. Masako Itoh is a lecturer at Keio University, serves as an editor for the Japanese Cognitive Science Society, and has membership in a variety of HCI-related organizations. Takashi Sunaga is a lecturer at the University of Tokyo and an associate professor at the Tama Art University. He leads the industrial design work of the HIT Center.
Members of the HIT Center have a very sophisticated view of the interface design process. They see three main stages in the development of a project: clarification and development of shared understanding of problems, making plans for solving the problems, and execution of the project. It is important to note that they look for technology transfer and educational opportunities in each stage. Their work appears to integrate the best work of the more cognitive approaches to interface design with participatory design methodologies to focus on supporting collaboration. Their objective is to support both micro and macro interface structure. They see the interface as embedded in a larger context of work practices and organizational structure.
Members of the HIT Center saw a need for more collaboration between the various disciplines involved in human interface design. They see four kinds of groups involved: computer scientists, human computer interface specialists, human factors specialists, and cognitive scientists. They expressed concern about how difficult it was for these groups to collaborate. Of the places the JTEC panel visited, it seems clear that the HIT Center evidences the most appreciation for the importance of such collaboration and a commitment to bringing it about within their own group.