Replacing the old 'research-design-make' paradigm, parallel research and
design processes iteratively diverge and converge before ultimately meeting.
This model was developed by a thinking group within i3 which includes partners of Presence and Campiello.

USER DRIVEN: observation interpretation animation "drawing out"

DESIGN DRIVEN: proposals solutions envisions simulations

Alessandra Agostini University of Milano Agostini@dsi.unimi.it

Giorgio De Michelis University of Milano Gdemich@dsi.unimi.it

Marco Susani Domus Academy Research Center, Milano Susani@domac.it

The design of computer based systems willing to support social interaction among users is not a technological problem. Technology is the basis for supporting new ways of interactions responding to user needs; but the user needs are the crucial point in conceiving any artifacts - including computer applications.

The participatory approach to system design developed in the Scandinavian countries in the last twenty years and further developed within the Computer Supported Cooperative Work (CSCW) field all over the world in the nineties - have played a major role in making the above claim widely recognized and understood. Participatory design (1) can be considered as a reaction to the typical system engineering approaches where technology remains the crucial issue and, at the very most, a mere matching between the user needs and the functionalities of the system is required. From the system engineering perspective, user interfaces assumed the role of the vehicle for maximizing the above matching as well as for optimizing the exploitation of the technology. On the contrary, participatory design - contrasting the centrality of technology - indicates user needs as the starting point of the design process and it conceives the system as a realization of the interaction patterns requested by the users.

It seems to us that participatory design, while quite effective in bringing forward the issue of user needs, has left to the engineering side of the process the responsibility of innovation and invention. Participatory design, in fact, often takes user needs as they are described by the users themselves, and in doing so disregards the potential for innovative human practices (not easily foreseeable by users) embedded in technological innovation (2). Therefore, software developers, when aiming to invent innovative systems, seem to be forced to disregard the users and to develop technological innovation per se.

It is our opinion that this weakness of participatory design can be avoided if we involve in the design process other actors bringing in the process ‘the culture of the design’. In other words, a new figure is required - namely the ‘interaction designers’ - whose focus is in the design of innovative interaction patterns and not in the development of innovative technologies (as for system ‘engineers’). We claim that (in accordance with the manifesto of the Software Design Association (3)) we can truly call software designer only those software developers that are able to cooperate with, or to act as, interaction designers.

As Terry Winograd has indicated in his recent book (1996), the design of software "that works - that really works - to be appropriate and effective for the people who use it" needs to take inspiration from other areas of design. This is particularly true for the type of software that i3 projects aim to develop; in fact, their users are not professional users and cannot be easily trained; besides, they are highly heterogeneous both from the cultural and generational point of view. Moreover, i3 projects within the connected community schema are addressing the issue of social and collective computing, going beyond any present paradigm of individual interaction with a personal computer.

In the last three years, The Cooperation Technologies Laboratory of the University of Milano (CTL) and the Domus Academy Research Center (DA) have been, and still are, involved in the design of several user oriented systems (sometimes together - within the DESARTE and CAMPIELLO Esprit Projects; sometimes not - within the COMIC and LIME Esprit Projects). During these experiences (4) we deeply discuss the differences as well as the complements between the design style characterizing the Italian School of Industrial Design (whose DA is a quite important representative) and the participatory design approach diffused in the CSCW community (to which CTL has belonged from its very beginning).

We observed that, on the one hand, the two above approaches are quite different. The former emphasizes the creative activity of the interaction designers, defending their autonomy concerning their clients and/or users. The second approach attributes to the communication and interaction with the users a central role in the design process. On the other hand, both the approaches consider the technology relevant from the point of view of the interaction patterns it offers to its users. Moreover, the ethnographic studies - built within many CSCW projects, at CTL too - as well as the user scenarios - which are the basis of the design projects of DA - allow an effective integration of design invention and user participation. Finally, user scenarios allow technological innovation to be guided towards user satisfaction, offering in the meantime to software designers a new framework within which to work.

The user scenarios are integrated representations and simulations of the system, including not only the technical solutions in place, but showing also the physical context of use, the roles of the users/actors, the content that circulates within the system, as well as the functionalities of the system. Often the user scenarios are presented in the form of storyboards, or videos, to allow easy discussion with users.

Within the above mentioned experiences - the most important of which have not yet been completed - software and interaction designers are deeply involved in collaboration processes in which they can experience the richness emerging from the cultural exchange between two cultures; of course, when they are capable to be open to each other without prejudices. One of the results of this collaboration is the skeleton of a new design methodology, trying to integrate user participation and system invention in a rather innovative way.

The methodology that we propose here is therefore the integration of two well-known design approaches, coming respectively from the industrial design and from the information systems design fields.

The industrial design approach has its specificity in the fact that the design and realization phases are continuously fed by a concept generation activity. In this "simulate to stimulate" phase, the designers develop and visualize scenarios of use, re-conceive the brief of the project, and specify the qualities and the attributes of the service. The concept generation phase allows a constant flow of innovation into the design process, going beyond the mere interpretation of user needs, to stimulate the demand of new functionalities that will transform the way in which the users see and understand their environment.

As we have anticipated above, the participatory approach to system design achieves a close fit between user needs and habits and the new system to be developed through the direct participation of the users in the various phases of system development. In the last ten years within the CSCW field, participatory design has been enriched both because it has been complemented by ethnographic methods of analysis and because it has been provided with a new point of view directly oriented to support the practices of the users rather than to automate parts of their work.

For the sake of clarity the participatory design approach usually misses out the concept generation activity, and the industrial design approach underestimates the social observation phase.

The approach presented here intertwines the cycles in which the User Driven and the Design Driven developments are parallel, intersecting them frequently to confront the results and re-tune the process. Toward the end of the development, the two parallel processes will merge and produce one final result that will take advantage of both contributions and manage both innovation and user adaptation.

It is important to underline that both the sides of the process, concept generation and user observation, are active, synthetic, proactive, creative processes. There is no consequence or priority among the two, nor there is a difference between the analytical and the synthetic parts of the process. Both sides have to imagine, create and propose innovation: one side does it with the competence of user understanding, the other does it with the competence of creating visions.

Our design methodology - elaborated and continously refined starting from 1996 - is well represented and summarized by the helix figure printed at the head of this article.

We like to stress two relevant aspects of our methodology. First of all, any intersection between the User Driven and the Design Driven development produces information which is collected and integrated to produce significant input for the successive phases.

Second, during the design process, artifacts at different level of granularity are created. For example, at first phases only simple demonstrations of the system are produced giving rise to the verification of the potentialities of the future system (such as, its selected innovative ideas, concepts, and technologies). As the process goes on, various demos and prototypes ever more refined are developed till the complete system is produced. This idea too is new with respect to traditional software design in which, at most, one prototype will be developed before the definitive system. We claim that this methodology will meet the objectives we want to reach; that is, "to create software that works - that really works - to be appropriate and effective for the people who use it" and in the meantime to satisfy people involved in the development of computer applications (e.g., software designers, creative designers, etc.).

The first experiences we had with the above sketched methodology show that the users’ first reaction is to object that they do not need any scenario since they know what they need (we leave to your imagination the other side's reaction to these objections). Despite the problems that this reaction can generate in principle, the conversations it opens in practice between the users and the interaction designers are quite rich: on the one hand they help the designers to listen more carefully to the user requests and ideas so that they are able to tune their scenarios as well as their designed systems, while on the other hand, the users are able to update their image of the technology in a better understanding of the new possibilities it opens to their practices.

Moreover, even software designers and engineers are forced to exploit the potential of technology in order to fulfil certain performance requirements (from the interaction point of view), recalling that system functionalities are not the only issue they have to deal with.

A final comment on the application of this methodology to design social interaction systems - as it happens with the i3 projects: when no business objective can be defined for the system to be designed, it clearly appears that the quality of the system does not depend on its capability to offer a particular type of functionality (e.g., a tridimensional representation of an interaction space; a well supported shared work space; etc.); but rather its quality depends on the way it integrates its functionalities into the users’ interaction space, enriching their capability of interacting with and/or accessing knowledge. Single applications are not interesting unless they can be embedded smoothly into the system, while the system is not useful unless it offers a variety of interfaces coupled with the behavioural habits and situations of its users.

We think that the above observation can be applied successfully also to CSCW systems (even if, of course, those functionalities allowing a simplification of work processes are relevant for the overall quality of systems to be used in working environments). In fact, such a characteristic within CSCW systems will allow those aspects of the system devoted to support a group of people participating in the same work process to join with those aimed at supporting all the work processes in which one person of the group participates.

Footnotes:

1. For an exhaustive bibliography please refer to the Proceedings of the biannual ACM Conference: PDC, Participatory Design Conference.
2. One of the major risks is to pave the calf-path (see p. 8 - Editor).
3. For more information on the Software Design Association, please see its home page at : http://www-pcd.stanford.edu/asd/home.html
4. See for an example De Michelis et al. (1997).

References:

De Michelis, Giorgio, Leiva-Lobos, Edmundo P., Covarrubias Gatica, Eliana (1997): "Augmenting and Multiplying Spaces for Creative Design." In: GROUP’97. Proceedings of the International ACM SIGGROUP Conference on Supporting Group Work, Phoenix, Arizona, November 16-19, 1997, eds. Stephen C. Hayne and Wolfgang Prinz, pp. 177-186.

Winograd, Terry, ed. (1996): Bringing Design to Software. Addison Wesley, New York.

© 1998,1999,2000 i3net	-- Search i3net --	About the i3net web pages
Revised: 17 April, 1998.		Mail to webmaster@i3net.org