Developing a knowledge management system for storing and using the design knowledge acquired in the process of a user-centered design of the next generation information appliances Jisoo Park, Department of Emotion Engineering, Sangmyung University, 7, Hongji-Dong, Jongno-Gu, Seoul, 110-743, Republic of Korea This project developed the next generation information appliances by usercentered design, acquired and stored the knowledge which designers used in the design process, and developed a knowledge management system for supporting the knowledge users to develop their own product ideas. It applied a scenariobased design and an ethnographic study as methods for user-centered design in to developing new product ideas required by the users. It stored the designer’s knowledge in knowledge bases and linked the knowledge connected in the design process to show where it came from. The links help the knowledge users trace the design process, get the design knowledge, and develop their own product ideas. This will make the results from this project facilitate the development of the next generation information appliances. Ó 2011 Elsevier Ltd. All rights reserved. Keywords: design knowledge, design process, product design, innovation, user-centered design
T
he next generation information appliances (NGIAs) are those that can be connected with other appliances through a wired or wireless network in order to transmit and receive data between them. NGIAs can be remotely controlled from inside or outside of the home. The emergence of NGIAs makes new functions possible that were not available when they were used independently, and gives opportunities to create a new market in an already saturated home appliances market.
Corresponding author: Jisoo Park
[email protected]
Many companies have conducted user-centered design for developing NGIAs such as Vision of the Future by Philips (1998), Designing the PDA of the Future by Marcus and Chen (2002), Cooltown by HP (2002), and New Life Forms by Motorola (2000). Philips Electronics conducted research on what users think will be useful, desirable, and beneficial in the future, and made a technological road map on how to achieve these goals, while suggesting future products, making design mockups, and producing videos to show natural scenes in www.elsevier.com/locate/destud 0142-694X $ - see front matter Design Studies 32 (2011) 482e513 doi:10.1016/j.destud.2011.05.001 Ó 2011 Elsevier Ltd. All rights reserved.
482
which the proposed products are being used in realistic future situations (Philips, 1998). Based on the socio-cultural and technological research, Philips set up teams consisting of cultural anthropologists, agronomists, sociologists, engineers, product designers, interaction designers, exhibition designers, graphic designers, and video and film experts, and developed more than 300 scenarios. The scenarios (short stories describing a product concept and its use) were generated using five parameters: people, time, space, objects, and circumstances. They were refined and filtered using the four criteria: ‘Will they provide users with an obvious benefit?’, ‘Will they fit with Philips’ major areas of competence and interest?’, ‘Will they be technologically feasible?’, and ‘Will they be applicable to the socio-cultural area Philips defined?’. As a result, the original 300 scenarios were reduced to 60 well-defined concept descriptions. The 60 product concepts were then grouped into four manageable categories: personal, domestic, public, and mobile.
Marcus and Chen (2002) developed the concepts for the next generation of wireless information devices. They conducted research on market trends, emerging mobile technologies, and advanced user interface technologies. In addition, a diverse group of users, including a minister; a male college student; a female high school student; a commuting professional; and a female, single-parent Silicon Valley entrepreneur were observed closely on a daily basis. The focus was on their activities related to information use and the issue was to try to understand needs that even the users could not express. To organize the complex and multilayered habits of the users observed in this process, an analytic framework was created. The framework divided the usage categories of the wireless devices into information, self-enhancement, maintaining relationships, entertainment, and M-commerce. This framework proved useful in developing product strategies and concepts. Following the research and strategy phase, they carried out a series of structured brainstorming sessions, turning out detailed explanations, user scenarios, and sketches for approximately 100 product concepts. An interactive prototype was also made to demonstrate certain scenarios. The prototype was effective in providing solutions for complex problems, such as creating appointments, using voice commands, and taking notes during a phone call.
Similar projects were reported, called Cooltown by HP (2002) and New Life Forms by Motorola (2000). They also developed future product concepts, made them into mockups, and produced video films showing natural scenes. However, the four projects above only presented their design processes and final product ideas briefly without opening the knowledge they acquired from the processes. If Philips, for example, opens the 300 scenarios generated by a diverse group of experts using the five parameters and Marcus and Chen open the user’s activities related to information usage, the designers, who want to develop future product concepts, will not have to spend their time and efforts on generating the scenarios or activities.
Developing a knowledge management system
483
Many researchers have studied what information and knowledge the designers, engineers, or managers require in engineering design. Heisig, Caldwell, Grebibi, and Clarkson (2010) surveyed the requirements of engineers from aerospace, engineering and consulting companies in the United Kingdom, including design engineers, service engineers, software engineers, production engineers, and managers, throughout the whole product life cycle from initial concepts via manufacturing to the disposal of the product. The survey aimed to gather requirements regarding the needs to retrieve past designs and the needs to capture current designs to support future engineering tasks. The result from 139 respondents showed 69 categories of knowledge and information needs spanning the whole product life cycle covering, for example, ‘requirements’, ‘design solutions’, ‘options & choices’, ‘change/modifications’, ‘manufacturing information’, ‘service’, ‘performance’, and ‘maintenance information’. The most mentioned category relates to the ‘rationale’ (34.9%) or the ‘reasoning behind the design as it is’, ‘why choices are made’ and the ‘reasons for design decisions’. However, ‘design reuse’ was mentioned by only 5.4% of respondents. This suggests that to foster reuse in practice, more attention needs to be directed to understanding of the design process. This project can contribute to understating of the knowledge requirements by narrowing research focus on recording the real data which designers used in design process of developing new product concepts. Researches into capturing and representing design rationale for complex decision spaces can be traced back over 40 years of the pioneering work of Kunz and Rittel (1970), addressing ill-structured problems using a proposed IssueBased Information System (IBIS). IBIS consists of a tree, or directed graph, where some nodes represent issues to be solved that are linked by arcs to other nodes that represent alternative solutions. These in turn are linked to nodes representing arguments for or against. Many derivatives of the basic IBIS concept have been proposed, such as gIBIS (Conklin & Begeman, 1988), PHI (McCall, 1991), QOC (MacLean, Young, Bellotti, & Moran, 1991), EGIDE/DRAMA (Ba~ nares-Alc antara, King, & Ballinger, 1995), PROSUS (Blessing, 1995), RObjects Pepper (Ernst, 2002), IDIMS (Kato, Taketa, & Hori, 2002), DRIFT (Nomaguchi, Ohnuma, & Fujita, 2004), and Questmap (Buckingham Shum, Selvin, Sierhuis, Conklin, Haley, & Nuseibeh, 2006). However, while IBIS has strong intellectual appeal in the research community, it became apparent that such tools were rarely able to be successfully applied in industry. As a result, in the design research field at least, work on design rationale capture reduced significantly (Bracewell, Wallave, Moss, & Knott, 2009). The main areas to which IBIS have been applied include design artifact modeling (Bohma, Stone, Timothy, & Steva, 2008; Szykman, Racz, Bochenek, & Sriram, 2000), solving complex engineering problem of a turbofan engine to power civil airlines (Bracewell et al., 2009), software engineering, and sociotechnical system design (Buckingham Shum, Selvin, Sierhuis, Conklin, Haley, & Nuseibeh, 2006).
484
Design Studies Vol 32 No. 5 September 2011
Previous studies on capturing and representing design rationale have focused mainly on recording the issues, options, arguments, and decisions acquired in the process of solving complex engineering problems based on IBIS. They have never tried in design research field to record the knowledge and information explicitly which designers use in the design process of developing future product concepts where the specialized methods for user-centered design are used. The objectives of this study are to develop the NGIAs by usercentered design, explicitly represent the tacit knowledge of designers, and facilitate expansion of the knowledge through the knowledge acquiringstoring-using cycle. In the process of developing the NGIAs, designers usually generate a great deal of valuable knowledge such as the problems the users may suffer from, user requirements and needs, new product ideas, and their design mockups and videos. The knowledge evolves through the steps from identifying user needs to generating new product ideas and producing videos. In the evolution process there are important links which say where the knowledge came from. The links help other designers trace the process. It is certain that using the knowledge will give companies strategic advantage in the development of their own new product ideas. However, the problem is that it is not easy to capture the knowledge because it is tacit one (Nonaka & Takeuchi, 1995). To overcome the problem, this study adopts the Knowledge Management System (KMS) method which is a systematic process of acquiring, accumulating, and sharing individual knowledge, and uses them to strengthen competitiveness (Davenport, Long, & Beers, 1998).
1
User-centered Design of NGIAs
The slogan at the 1933 Chicago World Fair was ‘Science finds, technology applies, people conform,’ but the slogan for user-centered product development in the 21st century is ‘People propose, science studies, technology conforms’ (Norman, 1999). The same applies in the field of developing NGIAs. People do not adapt to the functions implemented by information appliance technology; information appliance technology must adapt to functions that are required by people (Philips, 1998). This means that NGIAs should be developed based on the user needs. This section explains the process of usercentered design of NGIAs. A scenario-based design and an ethnographic study were used as methods for user-centered design. Scenario-based design, that is an analytic method, is useful to discover user needs, in the course of analyzing hypothetic scenarios of user actions, which are not obvious in actual situations. However it does not guarantee that the users really have those needs. So, an ethnographic study, an observational method, was applied in order to verify the user needs acquired in the scenario-based design, and thereby to find design implications and product suggestion by observing users.
Developing a knowledge management system
485
1.1
A scenario-based design
Scenarios play a guiding role in the process of scenario-based design for developing new product ideas. Scenarios of user actions can help designers find the problems and difficulties which the users might suffer from and suggest new product ideas. Scenarios of what new product idea does help design its style, user interface, and dynamic interaction. This project used Structured Planning (Owen, 1998) as a method for scenario-based design. Structured Planning helps a developing team discover user needs, secure new ideas, organize large amounts of information optimally for concept development, and develop solutions appropriate to the real complexity of problems. When the process of developing a new product is divided into the planning, designing and making stages, Structured Planning is used in the planning and concept designing stages. In these stages, Structured Planning provides the philosophy, framework, and the information handling formats for discovering user needs while organizing the information in the best way for designers to use it. Developing innovative and novel functions requires not only using a good method, but also coming up with a wide variety of creative ideas by forming a concept building team consisting of 6 experts in diverse fields. They are experts in the fields of cognitive engineering, user interface, computer science, cultural anthropology, interaction design, and product design from different organizations. It was expected that the experts in diverse fields would generate scenarios and suggest new product ideas from different point of views. Figure 1 shows the process of analyzing the scenarios of user actions in the home using Structured Planning. Its process was adapted to conform to the characteristics of this research; thus Figure 1 is somewhat different from the original process. In the action analysis phase the team used the same function structure that had been suggested by Owen (1984) in the House of the Future project. The function structure in Table 1 contains most of the important activities carried out by users in the home, so it was imported exactly as found. This function structure contains twelve modes with 44 activities. In the activity analysis phase, team generated scenarios of user actions related to each of the 44 activities systematically specify the functions that the system will perform in each activity. Owen (1984) specified the functions in each activity without any systematic method, such that it was difficult for the team to specify the functions. In order to generate as many scenarios as possible, each member of the team brought independent scenarios related to each of the 44 activities together; thus, they derived approximately one thousand different scenarios. By grouping the scenarios generated from each of the 44 activities, the 44 activities used by Owen (1984) were subdivided into 103
486
Design Studies Vol 32 No. 5 September 2011
Figure 1 A process of analyzing scenarios of user actions using Structured Planning
sub-activities. For example, the team subdivided the exercising activity in the health maintenance mode into the two sub-activities, preparing exercising and exercising regularly. Figure 2 shows the scenarios that were grouped into the sub-activity exercising regularly. Table 1 User activities in the home (Owen, 1984)
Mode Food consumption
Health maintenance
Hygiene
Rest Entertainment
Education
Activity Shopping Food processing Cooking Eating Dish washing Monitoring Exercising Regenerating Care giving Grooming Cleansing Eliminating Sleeping Relaxing Watching/Listening Engaging Contents Socializing Game playing Hobby working Discovering Inventing Interacting
Developing a knowledge management system
Mode Communication
Work at home
Maintenance
Protection Storage
Activity Inquiring Informing Sharing Controlling Purchasing Discovering Inventing Interacting House cleansing Clothes cleaning Monitoring Diagnosing Conditioning Repairing Replacing Detecting Reacting Storing Retrieving
Special child & Elderly care
487
Figure 2 Scenarios, user and system functions, and design factors grouped into the sub-activity exercising regularly
The team generated user functions and system functions that the user and system would perform in each sub-activity based on the scenarios grouped into a given sub-activity. Figure 2 shows the user functions and system functions generated from the scenarios grouped into the sub-activity exercising regularly. For example, from scenario 101 in Figure 2, the team generated the user function exercise regularly and two system functions: give motivation for exercising regularly and compile statistics on my exercising. In the design factor phase, the team reviewed both the scenarios and functions grouped into each sub-activity in order to detect problems that might arise when each function would be carried out. Table 2 shows part of the dialogue that the concept building team carried on using Microsoft NetmeetingÔ. The
488
Design Studies Vol 32 No. 5 September 2011
Table 2 Part of the dialogue concerning the solution element N-Health
Who Computer scientist
Cultural anthropologist
Computer scientist
User interface designer Cultural anthropologist Interaction designer Computer scientist
User interface designer Cultural anthropologist Computer scientist Cultural anthropologist Interaction designer Cultural anthropologist
Dialogue Next topic is exercising. The main problem of exercising is that it is difficult to practice exercising regularly, so it seems to be required to give motivation to exercise regularly with fun. The scenario 206 is very funny... People buy snacks to eat during exercising... But some exercising may not be related with losing weight..so that is OK.. Well.This is a joke..but an avatar can be used to give motivation for exercising by networking it to exercising machines..and making it healthy when the user exercises a lot, but making its health deteriorated when user does not exercise regularly. The avatar for health will be marketable. Hurrah! Good idea! How about having a competition of avatars on virtual world? It will cost too much to have exercising machines at home..so if the idea of avatar is linked to health club..there will be a synergy effect of combining game with health. Fantastic idea! Men and women have different purposes on exercising. Women exercise to dress herself neat and tidy. For women, how about holding Miss Avatar Competition? Will female avatars take part in a beauty contest? Two similar ideas at the same time..Anyway the ideas are very good to give motivation for exercising. The avatar idea seems to be almost developed...Any better ideas?
dialogue shows how the team identified the design factor giving motivation to exercise regularly with fun, and suggested the n-Health solution element from the design factor. The computer scientist in Table 2 suggested an avatar for giving motivation in exercising. This shows that the computer scientist had a technology-oriented point of view, and tried to solve problems by applying new technology. The avatar idea was extended to a competition of avatars on a virtual world by the interaction designer in Table 2.
The team derived a total of 165 design factors. Figure 2 shows the design factors derived from the scenarios grouped into the sub-activity exercising regularly. There were three design factors: ‘It is required to make people feel like exercising’; ‘It is required to give motivation for exercising regularly with fun’; and ‘It is required to eliminate the restrictions of the time and place where people can exercise’.
In the solution element phase, the team suggested 127 solution elements from the 168 design factors and summarized, in the solution element document, the scenarios and design factors from which the solution elements were suggested, as well as the definition, properties (what it is), and features (what it does) of each solution. The team suggested the solution elements based on the heuristic that team members have only to improve on the other members’ ideas while not criticizing them. This encouraged all members to suggest any the ideas
Developing a knowledge management system
489
they thought of, and encouraged the formulation of these ideas into solution elements.
The team suggested the solution element, n-Health, from the design factor, give motivation for exercising regularly with fun. Most people want to be healthy, but it is often difficult to exercise regularly. The n-Health was suggested to provide a natural motivation for exercising regularly. An avatar lives in the n-Health. If a user exercises a great deal, the avatar becomes healthy; whereas if the user does not exercise, the avatar’s health deteriorates and it could eventually die. This provides the motivation for exercising regularly. An exercise machine develops specific muscles in the body. This information is transmitted from the exercise machine to the n-Health so that it can develop the same muscles of the avatar. Each user’s n-Health can also be connected to the Internet and an athletic competition for avatars can be held. The avatar that has exercised the most is the winner, and this would serve as motivation for exercising. The team summarized the scenarios and design factors from which the n-Health was suggested as well as the definition, properties, and features of the n-Health in the solution element document (Figure 3).
1.2
An ethnographic study
Ethnography is a branch of anthropology that deals with the scientific description of specific human cultures. Ethnographic approaches to HCI (HumaneComputer Interaction) and product development were pioneered by Alladi Venkatesh in the 1980s. As the home became a primary market for new technology, he argued that many technology providers, while having a sound knowledge of technological issues, paid scant regard to social organization in the home (Venkatesh, 1996). He pointed out that there may be wide gaps between what technologies can provide and what users actually want. His approach led to rich insights into the context of technology use, such as user resistance to new developments (e.g. cable and satellite TV), potential familial conflicts which might be played out through technology (e.g. who has control of the controller) and the possibilities of new kinds of social interaction facilitated by technology.
Ethnographic methodologies have also been used in conceptualizing future product requirements. Stolzoff, Chuan-Fong Shih, and Venkatesh (2000) recruited fifty American families with PC access in the home through newspaper advertisements and conducted two interviews with each family at six month intervals about smart home applications. They also employed Participant Observation to examine HCI interaction. The focus on use context illustrated that the PC is single user orientated and as a result new technologies were suggested such as remote access monitors which would allow parents to see what their children were doing on their PC, allaying their fears about inappropriate material as well as making PC use less of an isolated activity.
490
Design Studies Vol 32 No. 5 September 2011
As a method of Ethnography for observing user action processes, the author developed the Technology Biography and applied it to ten participants in York, UK. It is an eclectic procedure that modifies and combines several techniques used in a variety of qualitative studies. It was designed to generate critical and creative responses to questions of home technology development and focuses on past, present and possible future domestic products.
Table 3 shows the elements of the procedure and indicates likely areas of interest. The Personal History section of the interview focused on the experience of use,
Figure 3 Solution element document of the n-Health
Developing a knowledge management system
491
Table 3 Steps of the Technology Biography
Element
Area of likely interest
Present: Technology Tour Last Time Questions Past: Personal History Future: Guided Speculation, Three Wishes
Tension, Conflict, Humor, Pattern, Routine, Disruption Nostalgia, Loss, Celebration Hope, Fear, Skepticism, Problems addressed
and changes in domestic technologies that participants had witnessed during their lifetime. The participants were first questioned about their childhood memories of housework, who did it and how it was done. They were then questioned about how they organized housework and how it differed from the present. Technology Tours were developed by Petersen and Baillie (2001) to create a general picture of a family’s everyday use of domestic technologies. Participants showed the researcher around their home and were questioned about how particular technologies were used: by whom and how often; what trouble they had with them; whether they saved labor, time, or money; and whether they enjoyed using them. Last Time Questions were based on Critical Incident Method (Flanagan, 1954) and asked ‘When was the last time you really enjoyed a household task?’, ‘When was the last time home technology helped you out?’ and ‘When was the last time home technology did not do what you wanted it to?’ The researcher explained new technological developments to participants and discussed these with them during the visits. At the end of the visits they reflected on these in a Guided Speculation about possible future developments. What were their hopes and fears for the homes of the future? Three Wishes were adapted from Cultural Probe (Gaver, Dunne, & Pacenti, 1999). Stamped addressed letters were left with participants with three pieces of paper with the words e ‘I wish I had’ on each page. They were instructed to fill them in and send them whenever they encountered any problems that might be solved by future technological developments. This project made interview questions based on those used in Personal History, Technology Tour, Last Time Questions, Critical Incident Method, Guided Speculation, and Three Wishes. Additional questions were added to check the validity of the design factors from which the solution elements were suggested by Structured Planning. This project interviewed a single family of three households and ten individuals, and collected data using the Technology Biography procedure. The first household was that of a middle-aged couple and their youngest son, while the other two households were those of their two elder
492
Design Studies Vol 32 No. 5 September 2011
sons who were living in their own homes with their partners. In-depth interviews of between three to 4 h were conducted. A single family selected to pilot the study was an organic sample in that the roots of the immediate family were followed to include a pre-existing network of close relationships, rather than a cluster of unrelated individuals selected according to pre-set criteria. The immediate familial bifurcations of a middle-aged couple were traced to account for the entire immediate family network. In order to widen the scope of the organic sample to include demographic groups of particular interest opportunity sampling was again employed to locate international respondents. The sample was supplemented with a young Korean family including two twenty nine year olds with an 18-month-old baby. These differing demographics allowed for diverse insights and reflections. Data was coded using the qualitative data analysis package NvivoÔ. Table 4 shows the themes or Nodes used in the analysis. These nodes were native terms, i.e., they were used by the participants themselves, or else they were summary devices for areas of concern. The data was presented through the thematic schema that emerged from the coding process. After analyzing the coded data, Concerns, Processes, Design Implications, and Product Suggestions were identified separately. Concerns are summaries of what the participants noted as problems arising from the use of technology. Processes are summaries of how technology affects the participants’ daily lives. Design Implications are general statements following from these Concerns and Processes. Finally, Product Suggestions illustrate some of these design implications, which are not necessarily practical product solutions but rather thinkpieces to inspire or provoke designers. As an example of the key findings from the study, the home was regarded as a site of conflict. Domestic space is a limited resource and competing demands are made of it, in this sense, the home is a site of conflict. Participants in the study identified incompatibilities in preferences over room temperature, lighting levels, and audio volume in entertainment devices. The following shows what a participant said: Table 4 Nodes resulting from coding interview data
Space Ownership Control Exploitation Conflict Negotiation Aesthetics
Time
Desire
Routine Priorities Private Time Public Time Identities Communities
Functionality/Fun Usability/Flow Commodification/Appropriation Dissatisfaction/Resistance Role/Gender Money
Developing a knowledge management system
493
BETH: My husband always complains it’s far too hot. I on the other hand, think it’s far too cold, so yes, the minute I go out the door, off goes the central heating, and he’ll adjust the time so that it comes on about half an hour before I come in. But of course, I recognize that because I know the house isn’t at the temperature I like. The design implication here was that technology should be customizable for different individuals in the same room. Illustrative product examples included heated furniture of the kind already available in cars, so that one side of a sofa, or one chair, could be warmer than another; directional lighting and directional sound speakers were also suggested. This is shown minutely in Figure 13.
1.3
Making design mockups
A team of 12 designers in the Design Research Center of Daewoo Electronics was formed to make design mockups. The design team individually evaluated the level of marketability, user needs, user acceptance, and technological feasibility of the 150 new products, suggested from the scenario-based design and ethnographic study, using five-point scales, and made the seven products with the highest scores in total into design mockups. In the phase of scenario-based design, the concept building team described the definitions and functions of the products in the solution element documents in terms of the scenarios and design factors from which they were suggested. The design team rewrote the functions of the products in a scenario format and derived implications for the design of a user interface. Based on these implications, the team designed the styles and user interfaces of the products. The design team rewrote, in scenario format, the functions of the products which the concept building team had described in the solution element documents. This means that the solution element documents provided a communication tool between the inventors and the designers. Using the scenarios, the design team derived implications for design of the layout of the display as well as the names, sizes, and functions of the buttons. Table 5 shows the implications for design which the design team derived from the scenarios of the functions of the n-Health. The design team derived two design alternatives: a goggle with an HMD (Head Mounted Display) and a wristwatch with an LCD. The design team concluded that the goggles may not be comfortable because of the sweating and dizziness caused by the HMD; therefore, a wristwatch whose external appearance is similar with a sports equipment was decided as the design concept. The implications for design in Table 5 required an easy method of opening communication channel between the n-Health and an exercising machine for receiving information from the machine. The design team suggested that an
494
Design Studies Vol 32 No. 5 September 2011
Table 5 Scenarios of the functions and implications for design of the N-Health
Scenarios of the functions
Implications for design
While a man is exercising in a health club, the n-Health receives information, from the health machine connected by wireless network, on the exercise machine and the muscles developed by the machine.
The n-Health shows the degree of the muscles development to lead him to exercise more up to the next level. When he does not exercise regularly the n-Health makes the avatar’s health deteriorated and eventually makes it died. This gives him the motivation to exercise regularly. He connects the n-Health to the Internet to take part in an athletic competition for avatars. His avatar wins because he has exercised the most. This serves as motivation for exercising regularly.
- A wristwatch whose external appearance is similar with a sports equipment. - When user puts the n-Health near an exercising machine, the n-Health reads the RFID (Radio Frequency ID) attached to the machine, opens communication channel with the machine, and receives information from the machine through wireless network like the Bluetooth. - The n-Health needs a button to show on an LCD display the information on the exercise machine and the muscles developed by the machine. - The n-Health needs a button to show on an LCD display the information on the current state of the muscles development. - The n-Health needs a button to show on an LCD display the information on the avatar’s health index. - The n-Health needs a Menu button to set up the Internet connection and other miscellaneous things.
RFID (Radio Frequency ID) tag should be attached to the machine so that the n-Health can read the RFID and open the communication channel automatically when the user puts the n-Health near the machine. RFID technology was developed in the 1980s to track the position of a moving object. A magnetic field is created by a radio frequency emitted from an antenna attached to the RFID reader. When the RFID tag passes near the magnetic field zone, the RFID reader deciphers the data stored in the IC located in the tag.
The design team designed the layout of the display as well as the functions, sizes, and names of the buttons, based on the implications for design in Table 5. Figure 4 shows the design mockup of the n-Health.
1.4
Producing videos
Seven videos were produced to show the actual situations where the seven design mockups would be used in the home of the future. In the phase of making design mockups, the seven products were made into design mockups and the style and user interface of each was designed. However the design team considered only the static aspects of the user interface and did not clarify the interaction between the user and the product. The interaction was revealed in the course of the professional writers’ arranging the scenarios of the use of the products in the scripts of the videos. The scripts also revealed the dynamic aspects of the user interface, such as state transitions by user actions during task
Developing a knowledge management system
495
Figure 4 A Design mock-up of the n-Health
procedures, and the visual and auditory feedback at each state. The scripts were arranged so that they could show the situations where the user would use the products easily and the user’s emotion would be rich while using them. That was the reason the videos were called New Emotive Life. A team consisting of a producer, a writer, and three GUI (Graphical User Interface) designers was formed to produce the videos. The professional writer arranged the scenarios for the scripts of the videos. The producer and writer established the context for showing the functions of the product, added necessary scenarios for the story line, and arranged the most detailed scenarios for the scripts. In an example of this, the team established a context where a man is making a passionate presentation at a conference room and he seems to be exhausted. He notices that his avatar is weak, and he decides to do some exercise (Table 6). Table 6 shows the scripts the team arranged, the interaction between the user and the n-Health, the CGs (Computer Graphics) needed to make the n-Health seem to be actually working, and the screen shots of the video. The scripts also revealed dynamic aspects of the user interface. For example, when a man puts the n-Health near the running machine, the n-Health reads the RFID attached to the machine, opens a communication channel with the machine, receives information from the machine regarding the muscles developed by the machine, and finally shows the information on the LCD. The GUI designers created animations by using Macromedia FlashÔ in order to make the design mockups look like actually working. The animations were combined with real images. The videos were produced by MBC Production
496
Design Studies Vol 32 No. 5 September 2011
Developing a knowledge management system
Table 6 Scripts for producing video about the n-Health
Scene #1. In a conference room Video
Interaction
CG
Video clip
A man is making passionate presentation.
The man returns to his seat after the presentation. He seems to be exhausted. He thinks he does not have as much stamina as he used to because he has been too busy preparing the presentation. When he turns on his computer, an avatar appears feeble.
The computer automatically receives exercise history from the n-Health and show the health condition of an avatar on monitor.
Showing the degree of physical growth of the avatar on the computer and the n-Health
He sat mediation on something and decided to go exercising.
He took the n-Health on the table.
(continued on next page)
497
498
Table 6 (continued) Scene #2. In a health club He put the n-Health near a running machine
The n-Health reads the ID of the running machine, opens communication channel with the machine, and receive information on the machine.
Showing the picture and information of the running machine on the n-Health
The n-Health receives from the running machine how long and how fast he is running and shows the information on its screen.
Showing the degree of physical growth of the avatar on the n-Health while he is running.
He starts to run.
After 10 min he watches the n-Health and keeps running.
Design Studies Vol 32 No. 5 September 2011
When the n-Health gives alarms saying that he reaches a certain level of exercising, he stops running.
Inc. and KBS Media in Korea. Professional actors were filmed in scenes for six days and the animations and video footage were composed in six days.
2
Developing a knowledge management system
This project developed a web-based KMS so that the knowledge users could use it on the Internet. It was implemented using Microsoft SQL, HTML, Javascript, and Active Server Page (ASP). The knowledge users will be those who work in the field of planning, marketing, designing, and developing of the NGIAs in companies or research institutes. Figure 5 shows an overall structure of the KMS. Knowledge sources were the experts in concept building team, real users in the UK, and articles in newspapers. To acquire the knowledge from scenario-based design and knowledge from ethnographic study, Structured Planning and Technology Biography were used, respectively. Acquired knowledge was then stored in the knowledge base according to the knowledge classification scheme. The knowledge from scenario-based design consisted of scenarios of user actions, design factors, and solution elements. Table 7 shows a logical structure of a scenario of user action. Each scenario was classified by mode, activity, sub-activity, related products, user, time, and location. It was linked with the related knowledge such as design factors, solution elements, future home scenarios, knowledge from ethnographic study, and technology developments. Especially, the links connecting scenarios of user actions, design factors, and solution elements show paths from which the solution elements were suggested (the solid line A in Figure 6). These links will help the knowledge users trace the design process and generate new scenarios, derive new design factors, and suggest new solution elements. Table 8 shows a logical structure of a solution element n-Health. There are several fields such as definition, feature (what it is), and function (what it does) describeing the solution element n-Health. The solution element was classified by product, user, time, and location, and was linked directly with scenarios of user actions to show where it came from (the solid line B in Figure 6). The knowledge from ethnographic study was acquired by interviewing the real users in the UK. In the course of analyzing the interview data, the problems that the real users had while using home appliances were identified. The problems were summarized by concern, process, design implication, and product suggestion. They were classified into three groups: space, time, and desire as shown in Figure 13. The knowledge from ethnographic study was also connected with the scenarios of user actions. This help the knowledge users verify whether the scenarios, generated by the design experts, correspond with the situations where the real
Developing a knowledge management system
499
Figure 5 Overall structure of a knowledge management system
users experience (the solid line C in Figure 6). This also helps them verify the validity of solution elements by following the links connecting them with the scenarios and interview data. The knowledge from new emotive life consists of the design mockups and videos. The knowledge users can evaluate the validity of user needs and find the usability problems by watching the videos, which help them make indirect experience of using them. 430 newspaper articles, related with the NGIAs, were collected from the domestic newspapers in South Korea. 110 keywords were extracted from grouping
500
Design Studies Vol 32 No. 5 September 2011
Table 7 Logical structure of a scenario of user action stored in knowledge base
Field
Content
Scenario of User Action
Knowledge Classification Scheme
Mode Activity Sub-activity Products User Time Location Design Factor
Related Knowledge
Solution Element Future home Scenario Knowledge from Observing user actions Technology developments
It seemed to be difficult to exercise regularly. I hoped something would compile statistics on my exercising. Health maintenance Exercising Exercising regularly Exercise machines Anybody Morning, afternoon, night Health club It is required to give motivation to exercise regularly with fun. n-Health Exercising
Much demand for exercise machines for good health
them. The acquired knowledge was stored in the knowledge from surveying technology development. It was linked with both the solution elements in the knowledge from scenario-based design and the product suggestions in the knowledge from ethnographic study to help the knowledge users verify their technological feasibility (the solid line E in Figure 6).
3
Using the knowledge management system
This section explains how the KMS can help the knowledge users get the knowledge and develop new product ideas. In general the process of developing new products starts from understanding user needs. Figure 7 shows the six representative user needs in the home of the future, which included Pleasure, Privacy, Sympathy, Ease, Health, and Harmony. They were acquired by grouping 150 new product ideas, suggested from scenario-based design and ethnographic study, using card sorting according to the similarity of their functions. They can be used as keywords showing the directions of developing the NGIAs.
For example, Privacy is the desire that one feels not to have one’s privacy interrupted within a home cohabited by other family members and to use one’s own appliance in one’s own personal space. The space in a household is a limited resource and family members each have different desires, so a household could be seen as competition space. The space in the home is a private space as well as a public place. Family members share the space and the appliances but also want to have their own space and appliances. Consequently, home appliances
Developing a knowledge management system
501
Figure 6 Steps of acquiring the knowledge and links between related knowledge
are to be developed so that user can feel as if they are using their own appliances by automatically identifying where and who uses what appliances and providing an environment where the appliances are suited to each user’s preferences. This shows the direction of developing home appliances. When the knowledge users select one of the user needs, they see the design mockups grouped into the user needs. For example, when they select Privacy, they see the design mock-up of the My Watch (Figure 8). They see the functions and videos of the My Watch in Figure 8. The design mock-up and video help them experience the functions of the My Watch by showing them the real situations where it will be used in the home of the future. If they read a document explaining its functions or see a rendering drawn by a designer, the experience
502
Design Studies Vol 32 No. 5 September 2011
Table 8 Logical structure of the solution element n-Health stored in knowledge base
Field Name Definition
Feature
Function
Knowledge Classification Scheme Related Knowledge
Content n-Health The n-Health provides a natural motivation to exercise regularly by making an avatar healthy when the user exercises a lot and making its health deteriorate when the user does not exercise. - A supplementary means to provide a natural motivation to exercise regularly. - The n-Health has an avatar and controls the avatar’s health according to the amount of the user’s exercising. - The n-Health receives information about which muscles an exercise machine develops and develops the same muscles on the avatar. - When the user exercises the n-Health receives information by wireless network, from the health machines, on the amount of exercise and the muscles being developed, and it makes the muscles of the avatar in the n-Health developed. - The n-Health shows the degree of the muscle development to lead the user to exercise more. - When the user does not exercise, the n-Health makes the avatar’s health deteriorate and eventually makes it die. This gives the user the motivation to exercise regularly. Product User Time Location Scenarios of user actions
Future Home Scenarios
Exercise machines Anybody Morning, afternoon, night Health club - It seemed to be difficult to exercise regularly. I hoped something would compile statistics on my exercising. - I did not exercise regularly, even though I bought an exercise machine. - It was very difficult to find time to exercise even though I would like to learn some sports. - I had an obsession that I had to exercise, but I have not yet put the idea into practice. Exercising
will not be brought into play. Based on this experience, they can evaluate the marketability and validity of the usefulness and find the problems of the user interface design of the My Watch. This will help them suggest new product ideas. New product ideas related with the design mockup of the My Watch are listed in the lower end of Figure 8. When the knowledge users select a new product idea of the My Watch among them, they see the definition, properties, and functions of the My Watch (Figure 9). The design mock-up of the My Watch was made based on the new product idea of the My Watch suggested from the scenario-based design phase. They can suggest different kinds of design mockups with different user interfaces based on the new product ideas. For example, the new product idea of the My Watch was suggested to provide a user with a customized environment where home appliances are suited to the user’s preference by automatically identifying who and where the user is. They can suggest new design mockups which will be easier to put on at home than the My Watch. In the lower end of Figure 9 the knowledge users see the scenarios of user actions from which the new product idea of the My Watch was suggested.
Developing a knowledge management system
503
Figure 7 A snapshot of the screen showing six representative user needs in the home of the future
When they select the scenario of user action, ‘When I was watching TV in the living room, mom told me that she would like to clean the room.’, they see a window which shows detailed information about the scenario (Figure 10). It includes the activity category, scenario, design factors derived from the scenario, solution elements suggested from the design factors, future home scenarios, technology developments, and related data from observing user actions.
504
Design Studies Vol 32 No. 5 September 2011
The knowledge users will see more scenarios acquired in the course of analyzing scenarios of user actions to select the activity category, ‘Rest-Relaxing-Enjoy contents,’ in the upper end of the window in Figure 10. Figure 11 shows scenarios grouped into the activity category, design factors, and solution elements in table format.
Figure 8 A snapshot of the screen showing the design mock-up of the My Watch
The knowledge users can browse other scenarios by selecting an activity category on the left side of Figure 11. The category has hierarchical structure so that they can fold and unfold the categories. Scenario, design factor, and solution element
Developing a knowledge management system
505
Figure 9 A snapshot of the screen showing the solution element of the My Watch
are shown in a row to show that the design factor was derived from the scenario and the solution element was suggested from the design factor. The row shows the path where the solution element was suggested. For example, the design factors, User Identification and Location Recognition, were derived from the scenario, ‘When I was watching TV, mom told me that she would like to clean the room..’ The solution element, the My Watch, was suggested from the design factors. When they select the design factor, Location Recognition, they see both the scenarios from which the design factor was derived and the solution elements
506
Design Studies Vol 32 No. 5 September 2011
suggested from the design factor (Figure 12). This means that design factors play the role of a bridge connecting scenarios with solution elements.
Figure 10 A snapshot of the screen showing scenarios of user actions
The knowledge users can suggest their own product ideas in the course of following the rows in Figure 11. For example, when they examine the scenarios and design factors in Figure 11, they can derive new design factors from the scenarios and suggest new product ideas from the existing or new design factors as the concept building team did in user-centered design phase.
Developing a knowledge management system
507
Figure 11 A snapshot of the screen showing the knowledge from scenario-based design
When the knowledge users select ‘Related Data from Observing User Actions’ in the lower end of Figure 10, they see Knowledge from Observing User Actions (Figure 13). They see concern, process, design implication, and product suggestion, acquired from applying Technology Biography to ten participants in York, UK. Figure 13 shows process, design implication, and product suggestion derived from the concern ‘Domestic space is a limited resource and individual members of the same household make different demands of it.’ The different
508
Design Studies Vol 32 No. 5 September 2011
Figure 12 A snapshot of the screen showing scenarios of user actions and solution elements related with the design factor Location Recognition
requirements raise the necessity for family members to share domestic space by personalization. The data from observing user actions supports the validity of the design factors, User Identification and Location Recognition, in Figure 11 because they are necessary to share domestic space by personalization. The knowledge users can verify the validity of design factors and solution elements by examining the related data from observing user actions. In addition they can suggest their own product ideas based on what problems the real users had
Developing a knowledge management system
509
while using home appliances and how the problems affected their daily lives which are all stored in Knowledge from Observing User Actions.
4
Summary
Figure 13 A snapshot of the screen showing the knowledge from ethnographic study
This project developed the NGIAs by user-centered design, acquired the tacit knowledge, which designers had used in the design process, represented it as the explicit one, and developed a KMS for storing and using the knowledge. A scenario-based design and an ethnographic study were used as methods for
510
Design Studies Vol 32 No. 5 September 2011
user-centered design. A scenario-based design is useful to discover user needs, in the course of analyzing hypothetic scenarios of user actions, which are not obvious in actual situations. However it does not guarantee that the users really have those needs. So an ethnographic study was conducted to verify the validity of the design factors acquired in scenario-based design because they were discovered from hypothetic scenarios of user actions generated by 6 experts, and thereby to find design implications and product suggestions by observing users. In the general process of developing a KMS, the knowledge acquisition methods have been used to acquire the tacit knowledge and represent it explicitly. The methods range from informal techniques such as user observation, through common social science methods such as interviews, questionnaires, and discourse analysis, to more formal techniques used in knowledge acquisition such as task analysis, rule-based system, case-based system, and data mining. However this study recorded the knowledge which the designers used in the process of usercentered design instead of using the methods. Structured Planning provided very useful documents for organizing large amounts of data such as 1000 scenarios generated from 44 activities, 103 sub-activities, 165 design factors, and 127 solution elements. It also linked the knowledge connected in the paths from scenarios of user actions, through design factors, to solution elements so that the knowledge users could trace the process to develop their own product ideas. Technology Biography, used as a method of an ethnographic study, also stored the 38 concerns, 35 processes, 38 design implications, and 23 product suggestions derived from coding the 6235 lines of interview data and linked the knowledge. The knowledge users can develop their own product ideas by using the knowledge and following the links stored in the KMS. By using the knowledge from scenario-based design, they can trace the paths from generating scenarios of user actions identifying design factors, to suggesting solution elements. They can also generate their own scenarios, identify new design factors, and suggest new product ideas. By using the knowledge from new emotive life, they can trace the process of making design mockups so that they can design new ones with different styles and user interfaces. They can also evaluate the validity of user needs and find the design and usability problems by using the videos stored in the knowledge from new emotive life, which help them make indirect experience of using them. Especially the videos are useful for interaction design because they can expose the dynamic aspects of the interaction which are not clear in the process of making design mockups. By using the nodes derived from coding the interview data in the knowledge from ethnographic study, they can get the knowledge of what problems the real users might suffer from and how they affect their life. They can also evaluate the validity of the scenarios of user actions, design factors, and product ideas by using the links connecting the knowledge from scenarios-based design with the knowledge from ethnographic study. They can evaluate the technological feasibility of the existing or new product ideas by using the knowledge from surveying technology developments of information
Developing a knowledge management system
511
appliances. They can add the new knowledge, which they acquire in their own design process, to the KMS. The acquiring-storing-using cycle makes this research not a case study for developing new product ideas, but rather a system for extending the knowledge and eventually promoting the development of the NGIAs. As a further study, the KMS will be opened on the web so that many designers will use the existing knowledge in it and add new one while developing their own new product ideas. In this process it will be possible to validate the usefulness and usability of the KMS. Case studies will also make it possible to validate that the KMS can extend the knowledge through the adding-and-using cycle.
Acknowledgements Many researchers participated in this project. They devoted themselves to this project. Tae-Il Lee in the Department of Design at Korea University; HyunJung Kim, in the Graduate School of Digital Design at Kyunsung University; Young-Chul Son, in Digital Media at Samsung Electronics Corp.; and DongSeok Lee, in UX Lab at SK Telecom; analyzed scenarios of user actions and suggested new product ideas. Hyung-Rok Lee, Woo-Hoon Lee, Hee-Yun Jung, Sung-Chul Huh, Ji-Hyun Lee, Yoon-Tae Kim, Hyung-Min Kim, Sang-Jin Han, Yeon-Eui Jung, and Yoon-Jong Kang, at the Design Research Center of Daewoo Electronics Corp., made the design mockups. A large staff, directed by Producer Sung-Yeol Cho at MBC Production Inc. and by Producer SangRyong Oh at KBS Media produced the videos. Sang-Hee Lee and U-Rie Lee, at the Design Research Center of Daewoo Electronics Corp., also took part in producing the videos. Mark Blythe and Andrew Monk, in the Department of Psychology at the University of York, developed the Technology Biography and applied it to ten participants in York, UK. Dong-Hyun Baek, in the Department of Business Administration at Hanyang University; designed the logical structure of the knowledge management system (KMS). Boo-Ki Moon, in the Department of Computer Science at KAIST, developed the KMS using HTML, Java Script, and ASP (Active Server Page). This project was funded by the Ministry of Science and Technology of Korea (Project code: M1-9817-03-0006).
References Ba~ nares-Alcantara, R., King, J. M. P., & Ballinger, G. H. (1995). EGIDE: a design support system for conceptual chemical process design. In J. E. E. Sharpe (Ed.), AI System support for conceptual design (pp. 138e152). Springer-Verlag. Blessing, L.T.M. (1995). A process-based approach to computer-supported engineering design. Ph.D. thesis. University of Twente. Bohma, M. R., Stone, R. B., Timothy, W. S., & Steva, E. D. (2008). Introduction of a data schema to support a design repository. Computer-Aided Design, 40, 801e811. Bracewell, R., Wallave, K., Moss, M., & Knott, D. (2009). Capturing design rationale. Computer-Aided Design, 41, 173e186. Buckingham Shum, S. J., Selvin, A. M., Sierhuis, M., Conklin, J., Haley, C. B., & Nuseibeh, B. (2006). Hypermedia support for argumentation-based rationale: 15 years on from gIBIS and QOC. In A. Dutoit, R. McCall, I. Mistrik, &
512
Design Studies Vol 32 No. 5 September 2011
B. Paech (Eds.), Rationale management in software engineering (pp. 111e132). Springer-Verlag. Conklin, J., & Begeman, M. L. (1988). gIBIS: a hypertext tool for exploratory policy discussion. ACM Transactions on Office Information Systems, 6(4), 303e331. Davenport, T. H., Long, D., & Beers, M. C. (1998). Successful knowledge management projects. Sloan Management Review Winter. Ernst, J. (2002). Collaborative decision making and personal knowledge management with R-Objects Pepper. In Workshop WF6 - Real world RDF and semantic web applications, Proceedings of WWW2002. Flanagan, J. C. (1954). The critical Incident technique,. Psychological Bulletin, 51, 327e358. Gaver, B., Dunne, T., & Pacenti, E. (1999). Design: Cultural Probes. Interactions, 6(1), 21e29. Heisig, P., Caldwell, N. H. M., Grebibi, K., & Clarkson, P. J. (2010). Exploring knowledge and information needs in engineering from the past and for the future e results from a survey. Design Studies, 31(5), 499e532. HP. (2002). Cooltown. http://cooltown.hp.com/cooltownhome/index.asp. Kunz, W., & Rittel, H. W. J. (1970). Issues as elements of information systems. Working paper 131. Berkeley: Center for planning and development research. Kato, Y., Taketa, K., & Hori, K. (2002). Capturing design rationale by annotating e-mails. In Proceedings of 6th world multiconference on systemics, cybernetics and informatics. MacLean, A., Young, R. M., Bellotti, V., & Moran, T. P. (1991). Questions, options, and criteria: elements of design space analysis. Human-Computer Interaction, 6, 201e250. McCall, R. (1991). PHI: a conceptual foundation for design hypermedia. Design Studies, 12(1), 30e41. Marcus, A., & Chen, E. (2002). Designing the PDA of the future. Interactions, 9(1), 34e44. Motorola. (2000). New life Forms. http://www.motorola.com/mot/doc/0/ 293_MotDoc.pdf. Nomaguchi, Y., Ohnuma, A., & Fujita, K. (2004). Design rationale acquisition in conceptual design by hierarchical integration of action, model and argumentation. In Proceedings Of DETC/CIE 2004, CIE-57681. Nonaka, I., & Takeuchi, H. (1995). The knowledge creating Company. New York: Oxford University Press. Norman, D. A. (1999). The Invisible computer. London: The MIT Press. Owen, C. L. (1984). House of the future. Chicago: Institute of Design Communications Center. Illinois Institute of Technology. Owen, C.L. (1998). Design, advanced planning and product development. In Gestao do design. Uma Chave par o Sucesso. Papers of the FIESP Seminar and course. Sao Paulo, Brazil. Petersen, M. G., & Baillie, L. (2001). Methodologies for designing future household Technologies. Oikos2001. Aarhus, Denmark: Aarhus University Press. Philips. (1998). Vision of the future. Blaricum: VþK Publishing. Stolzoff, N., Chuan-Fong Shih, E., & Venkatesh, A. (2000). The home of the future: An ethnographic study of New information technologies in the home. University of California. Szykman, S., Racz, J., Bochenek, C., & Sriram, R. D. (2000). A web-based system for design artifact modeling. Design Studies, 21(2), 145e165. Venkatesh, A. (1996). Computers and other interactive technologies for the home. Communications of the ACM, 39(12), 47e54.
Developing a knowledge management system
513