Linux Traffic Generator (LTG)”, CoRiTeL

Levels of design: from usability to experiencePatrizia MartiUniversity of SienaVia dei Termini 6, 53100 Italy marti@media.unisi.itAntonio RizzoUniversity of SienaVia dei Termini 6, 53100 Italyrizzo@unisi.itAbstractThe “user-centred design” paradigm often encompasses fundamentally different strategies and methodologies that share the common objective of generating design solutions centred on the users. In the paper we propose a framework, the “egg model”, that attempts to minimise the generality of current interaction design models and to define techniques that are most appropriate to different levels of design, from usability to interaction and experience design.1IntroductionOver the past thirty years the research in the domain of HCI has evolved from a narrow product orientated attribute, to encompass broader concerns with the experience of use. This evolution took place together with the trend of information and communication technology to move out of work situations towards everyday life activities, including learning, leisure and entertainment. In a broad spectrum of use of technology, the design of user experiences is set to be a major issue to face. Nowadays it seems evident that understanding user concerns and collecting their requirements may be not enough in case of new design challenges. The paper proposes a framework to orient the adoption and use of a set of techniques for interaction and experience design that can become as thorough as those that are currently applied to requirements engineering and usability testing. More in detail, the paper describes a model that represents three levels of design and illustrates the related methods and approaches. Moreover the model recognises the peculiarities of each level of design in relation to different contexts and objectives of interaction: the technical accomplishment of a task, the experience of being charmed and engaged with technology, the discovery of new and unexpected uses of artefacts not originally anticipated by the designer. It argues that it is possible and worth to interpret interaction design in terms of “potential of use” of an artefact rather than a mere “problem of use” to solve. The model presents an approach to interaction design that is not confined to search for solutions to usability problems, but it is open to the design for the “potentials” of the human activity, where visionary scenarios are elaborated with the aim of offering intellectually interesting solutions and stimulating creative and emerging behaviours.2The “egg model”The “egg model” is a model of design articulated on three levels, reactive, pro-active and emergent, each one consisting of a different combination of the main phases of the design process: activity analysis, concept design and evaluation (Figure1). These phases are represented by three different tones of colour. The two horizontal lines in the egg divide the model in three sections that emphasize the weight that each phase has at the single level. For example, one level can becharacterised mainly by evaluation of the interaction with an artefact, whilst another level can regard concept generation and activity analysis as the principal design activities (see below for details).The model guides the designer in understanding the context of intervention and selecting the methods that are more effective for addressing the requirements of the domain. The metaphor of the egg is functional to define the role that the fundamental phases of the design process (activity analysis, concept generation and evaluation) play in three different contexts of design, that we define reactive, pro-active and emergent levels.Figure 1: The “egg model”2.1Reactive levelThe reactive level (Figure 2) is that in which the HCI designer is called to solve problems of use of a system already existing and fully operational (for example an automatic ticketing system or a web site for on-line shopping). Her role is therefore that of redesigning and improving the existent rather that to develop a new system from scratch. The reactive level is characterized by those situations in which the artefact has been released either in final or ß version, and the user feedback should be collected to assess the system functionality. Therefore the HCI designer operates in a context where the human activity to support is well known, the user requirements have been collected and hopefully satisfied and the system has been released. Her role is therefore to assess usability and correct potential design defects. This is the reason why in the “egg model” a great emphasis is devoted to evaluation, while activity analysis (that at this level is fundamentally task analysis) and concept generation play a smaller role. The design here is intended mainly as evaluation and re-design of an existing system, typically resulting in intervention on the user interface.Figure 2: Reactive level2.2Proactive levelThe proactive level is that in which the designer is called to develop a new system to support a well known category of users and human activity (Figure 3). Take as an example the case of a new electronic tourist guide for visiting museums. During the visit, visitors need clear and complete descriptions about artworks and spatial orientation for moving in the museum without getting lost. However there is much more that the designer can offer to these visitors. Entertaining narrations, adapted contents to different typologies of visitors, situation-aware contents, environment sensitive user interface (Marti, 2001). The human activity of visiting a museum will not be dramatically modified by the new artefact. However the new system will ameliorate the activity empowering the user with new opportunities and experiences. The context of the pro-active level is different from the reactive one. The designer does not have to improve an existing system by envisaging solutions that can better satisfy user needs, but to offer new opportunities could not be express in form of requirements. The work of the interaction designer in this level is more balanced in three phases: it is necessary to proceed with a careful activity analysis that can inform concept design, that is in turn continually appraised through an intense phase of prototyping of solutions. We could say that the true problem of this level is the definition of user requirements, an activity that does not come fixed at the beginning of the design process, but remains open until concept design and prototyping do not supply all the elements and information to define the specifications of the final system. The new system, beyond to satisfy the user requirements, should offer new opportunity of use to complement what expressed in form of requirements. The pro-active level includes situations in which starting from a clear set of user requirements, the objective of design is to develop an innovative product able to satisfy those requirements, but also to go beyond them and to offer to the users new opportunities of use of the artefact.Figure 3: Proactive level2.3Emergent levelThe emergent level is that one often referred as “creative design” (Figure 4). In this context, design is viewed as completely unrelated to problem solving. It applies techniques to “envision”human activities that could not exist without the artefact or system being designed. The emergent level addresses uncertain situations but definitely the more charming and challenging for the designer. Take as an example the case of the development of a new learning environment that allows children to create stories by connecting physical and virtual worlds through the manipulation of different media (Rizzo et al. 2003). The environment is specifically devoted to sustain pedagogical objectives like the acquisition of emotional knowledge to express, represent and control emotions, the capability of meta-cognitive thinking, and the intersubjectivity to make models of external world and the others’ mind (Bruner, 1990, 1996). In this context the new system is designed almost from scratch, and the human activity and the role of intended users aredesigned together with the system. The objective here is to design for the “potentials” of the human activity to support and to transform it, to make it evolve, to offer opportunities and find new incentives in human practices for of the use of the system. This level of design cannot count on pre-defined user requirements. The observation of the user activity needs to be supported by a theoretical reflection on the nature of the human activity and the social and cultural values at stake. The activity analysis and theoretical reflection proceeds in parallel with a divergent phase of concept generation and with a preliminary analysis of enabling technology. The design process is co-evolutionary and converges on the definition of “concept scenarios”, that is with the envisioning of situations in which the designer defines the role of the users and the system in specific contexts of use. We could say that at this level the experience of use is designed rather than a system supporting a particular activity. This is the reason why in the “egg model” a large emphasis to this level comes from activity analysis intended as a theoretical reflection, the true inspiration of the design, and to concept generation. The evaluation will always play a fundamental role, but more oriented to assess high level design concepts (and not therefore specific modalities of interaction with the system).Figure 4: Emergent level3ConclusionsThe “egg model” is a framework to orient the adoption and use of a set of techniques for interaction and experience design suited to different design contexts. The model supports the designer in the definition of the context in which s/he operates and in the selection of the most suited techniques and design strategies to adopt in each particular case. The proposed framework is an original contribution with respect to other interaction design process models currently adopted. Indeed most of the available process models derive from the early work of the Xerox Star research group that proposed a user-centred view for the design of innovative products (Bewley et al. 1983, Smith et al 1982, Gould and Lewis, 1983). From that pioneering work, many other formulations of the original idea were proposed with the aim to consolidate design practices with a clear user orientation. However, all these models (Beyer and Holtzblatt, 1998, Lewis and Rieman, 1993, Preece at al. 1994, Shneiderman, 1992) do not take into account the different meanings that the “design for the users” approach assumes in different contexts. Take as an example the concept of “activity analysis”, a core concept of user-centred design. When analysing the human activity that the future system should support, a fundamental difference exists between the activity that is determined and consolidated (first and second level of design in the “egg model”) and the one that is being designed together with the system under development. In the first case it is possible to observe the activity and to evaluate, in great detail, any breakdown or interaction defect that canoccur during the use. In the second case, the observation of the activity in “similar” contexts is the only inspiration the designer can take from field studies. On the contrary, most of the insights for the design will come from theories of human cognition. It follows that whilst in the first and second design level it is possible to analyse the human activity through task analysis and ethnographic observations, in the third level, observations must necessarily be complemented by a theoretical reflection on human cognition and the main cognitive and socio-cultural processes that can inform interaction. The same difference in the interpretation of the meaning of design phases occurs in the performance of evaluation activities: at the first design level, usability evaluation techniques can be successfully applied for detecting interaction problems, at the second and third level these are inappropriate since the objective of evaluation is the appraisal of high level concepts. The “egg model” is an attempt to minimise the generality of current interaction design models and to define techniques that are most appropriate to each level of design, from usability to interaction and experience design.4ReferencesBeyer, H. Holtzblatt, K. (1998) Contextual Design. San Francisco: Morgan Kaufmann Publishers.Bewley W.L., Roberts T.L., Schroit D., Verplank W.L. (1983). Human factors testing in the design of Xerox's 8010 Star office workstation. Proceedings of CHI '83 Conference on Human Factors in Computing Systems, New York: ACM, pp.72-77.Bruner, J.S. (1990) Acts of meaning. Cambridge MA: Harvard University Press.Bruner, J.S. (1996) The culture of education. Cambridge MA: Harvard University Press.Gould J.D. and Lewis C., (1983). Designing for usability: Key principles and what designers think. Proceedings of the CHI'83 Conference on Human Factors in Computing Systems. New York:ACM, pp. 50-53.Lewis, C. & Rieman, J. (1993) Task-Centred User Interface Design: A Practical Introduction Boulder Colorado: University of Colorado. /pub/cs/distribs/clewis/HCI-Design-Book.Marti, P. (2001) Design for art and leisure. Proceedings of ICHIM 2001, Milan, Italy, pp 387-397.Preece, J. Rogers, J. Sharp, H. Benyon, D. Holland, H. Carey, T. (1994) Human-Computer Interaction. Reading, MA: Addison-Wesley.Rizzo A, Decortis, F., Marti P., Rutgers, J., Thursfield, P. (2003) Building narrative experiences for children through real time media manipulation. In Blythe, M., Monk,, A, Overbeeke, C. Wright, P. Funology: From Usability to Enjoyment, Kluwer, 2003, in press.Shneiderman B. (1992). Designing the User interface. Reading MA: Addison Wesley.Smith D.C., Irby R., Kimball W., Verplank, W.L. Harslem E. (1982). Designing the star user interface. Byte, 7, 242-282.。