Lost in Translation: A Critical Analysis of Actors, Artifacts, Agendas, and Arenas in Participatory Design Rog��rio DePaula Center for LifeLong Learning and Design University of Colorado UCB 430, Room 717 - Boulder, CO ��� US, 80303 +1 303 735-0223 depaula@colorado.edu ABSTRACT As computer technologies start to permeate the everyday activi- ties of a continuously growing population, social and technical as well as political and legal issues will surface. Participatory design is asked to take a more critical view of participation, design, tech- nology, and the arenas in which the network of actors and arti- facts dialectically construct the social orders. This paper has a much more modest aim of that to contribute the discussion of participation and design in part by a more in-depth understanding of the translation problem among different actors who directly participate in participatory design activities. This problem takes place when different actors come to participate in the design ac- tivities and when they are to decide whether to adopt and use a designed artifact. By analyzing a multi-year-long effort to under- stand and provide social and technical means for the use of educa- tional computer technologies in special education, this paper aims to shed new light on the understanding of this problem. The are- nas of participation framework is employed to frame the different social orders in which actors act, carry out their work practices, participate in design processes, and ultimately make use of this artifact. While fundamental to the democratization of the design of sociotechnical solutions, participatory design may not be suffi- cient to reveal all sociopolitical issues of work practices that sur- face in its adoption and use. It is necessary to take into account the different arenas in which their design and use are carried out. Categories and Subject Descriptors K.4 [Computers and Society]: Public Policy Issues ��� Privacy and Regulation Organizational Impacts ��� Computer-supported collaborative work Social Issues ��� Handicapped persons/special needs General Terms Design, Security, Human Factors, Legal Aspects. Keywords Participatory design, actor-network theory, arenas for participa- tion, translations, Web2gether, special education, social networks, adoption, privacy, collaboration. 1. INTRODUCTION Participatory Design (PD) has evolved to become a mature re- search area and design practice among computer professionals. The Scandinavian versions of PD, which were concerned primar- ily with the politics of design���namely workplace democracy and political conditions for user participation in the design and intro- duction of the computer [14]���have evolved toward an in-depth understanding of the nature of participation, and the development of methods and techniques for favoring user participation at early stages of design, namely co-design and co-evolution of computer artifacts. Computer technologies and the social contexts of use have also changed since the initial work on PD. At workplaces, for example, many of the initial claims and arguments put forth by the PD community have become intrinsic to work practices and norms. As computer technologies start to permeate the everyday activi- ties of a continuously growing population, social and technical as well as political and legal issues will surface. Indeed, experience with participation is much broader, as G��rtner and Wagner have argued [10]. In particular, in reaching out beyond the traditional office work environments, PD is asked to take a more critical view of participation, design, technology, and the arenas in which the network of actors and artifacts dialectically construct the so- cial orders. This paper has a much more modest aim, however. It aims to contribute the discussion of participation and design in part by a more in-depth understanding of the translation problem among different actors who directly participate in PD activities. Understanding translation is key to a more effective, valuable, and direct partnership between designers and users. PD can bring together field-study methods and participative de- sign activities to facilitate the creation of a common language between designers and users. The goal is to integrate ���systemic analysis, appreciative intervention, and practitioner participation��� [13] to create social-technical-political conditions that reduce the gap between design practices and users��� work practices. Whereas in-depth fieldwork, such as videotaped participant observations and in-depth interviews, helps designers understand the nuances of users��� everyday practices, participative design activities, com- prising tools and techniques such as future workshops, case-base prototyping, and cooperative prototyping [19] were developed to provide users a means to take an active part in technical design. Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, re- quires prior specific permission and/or a fee. Proceedings Participatory Design Conference 2004, Toronto, Canada. Copyright 2004 ACM 1-58113-851-2/04/07���$5.00. 162

PD is instrumental in creating a common language, and thereby a stronger relationship between designers and users [13]. These methods, tools, and techniques are not without limitations, particularly when it comes to translating different actors��� needs, goals, motivations, and values. For example, in conducting a se- ries of site visits and workshops, researchers are able to collect much information on users��� work practices, environments, and technologies, as well as preferences, values, and norms. A super- ficial analysis of these data, however, based on a literal translation (or interpretation) of users' accounts, can yield only a partial ac- count of the work practices and needs in question. Additionally, by effecting and shaping the design decisions mainly in the design arena [9], PD is likely to overlook the influence on actors��� deci- sions and choices as well as their perceptions and understandings (situated model) when networks and agendas (hidden or other- wise) are activated when new technology reaches the institutional (or industrial) arenas. This paper thus argues that for PD research- ers to have a deeper understanding of the translation problems, they ought to reach out beyond the arena of design. By analyzing our three-year-long effort to understand and provide social and technical means for the use of educational computer technologies in special education, this paper aims to shed new light on understanding the translation problem. Specifically, PD was employed to facilitate the design of a collaborative peer-to- peer technology, Web2gether, to help special education profes- sionals find educational resources as well as receive professional and social support. The challenges of promoting the widespread adoption and use of Web2gether have motivated a more in-depth analysis of the networks of actors and artifacts (or intermediaries) that were activated as we moved the design of the system through different sociopolitical arenas. To this end, this paper employs G��rtner and Wagner���s [9] arenas of participation framework to frame the different social orders in which actors act, carry out their work practices, participate in design processes, and ulti- mately make use of technology. This framework is also inspired by the actor-network theory to analyze the sociopolitical issues that have impacted the use and adoption of the system. 2. ARENAS FOR PARTICIPATION AND NETWORKS OF ACTORS This work builds on and integrates G��rtner and Wagner���s arenas of participations [9] and actor-network theory [3]. It extends the existing literature by adopting a relational (network) perspective on the study of (participatory) design and adoption of collabora- tive technologies. A relational approach looks at the actors who participate throughout the design and use of a technology and the set of types of relationships that link each other. The patterns that emerge from these relationships come then to create social struc- tures and norms of the design and use context. In this respect, this approach does not take the arenas as given������places��� in which actors interact and act���but as socially constructed by the interac- tions that take place among the actors, mediated or not by external artifacts or intermediaries. This separates the relational approach hereinafter described from the traditional structural analysis that departs from already established structured settings, or arenas, at the outset. Despite its unquestionable relevance to design, development, and deployment of new technologies in workplaces, the quest for unmediated relations and close partnerships between designers and users is not sufficient, according to G��rtner and Wagner [9]. These authors also argue for a more in-depth understanding of agenda settings, the ���political and social forces that shape the practice of design and participation��� [10, p.37]. Thus, they distin- guish three major arenas of participation: Arena A ��� Designing Work and Systems: The individual project arena where specific systems are designed and new organizational forms are created. In this arena, we find the most direct and unmediated partnership be- tween designers and users. Arena B ��� Designing Organizational Framework for Action: The institutional area where ���breakdowns��� or violations of agreements are diagnosed and stable patterns of organizational functioning are questioned and redes- igned. Arena C ��� Designing the Industrial Relations Context: The na- tional arena where the general legal and political framework, which defines the relations between the various industrial partners and sets of norms for a whole range of work-related issues, is negotiated. Such social arenas provide a grounded mapping of system design (as well as development, deployment, and use) in space and time [9]. They can be seen as different ���social spheres��� of participa- tion and interactions in which different sets of actors and artifacts come together to discuss common projects, ideas, and concerns. They are, in part, places in which PD activities and the use of the technology are realized. They are not necessarily constrained by organizational boundaries, and the debates that are held in such places reflect the networks, values, belief systems, and social worlds each participant (collectively or not) brings to bear in the arena. In this respect, participation is what defines the boundaries and, more important, the legitimate agendas for the debates. At its early stage, PD attempted to link these three arenas by ex- ploring local conditions that influence policies at the national level (Arena C) [14]. The recent focus on partnerships between designers and users emphasizes work centered on Arena A at the cost of fewer PD projects being engaged at the organizational level (Arena B), and has lost sight of the importance of participat- ing at the national level (Arena C) [12]. This paper offers a more in-depth analysis of the consequences of overlooking the influences of the organizational and national levels at the deployment phase of a technology that was initially designed in partnership with users at the individual project level (Arena A). In particular, it argues that such influences could only be felt when the networks of actors and artifacts at Arenas B and C were mobilized. Each arena ���hosts its own political framework which in turn shape what its inhabitants hold as a legitimate stan- dard, value, interest, or procedure [9, p.193]. PD activities should not be limited to the participatory activities or partnerships cre- ated at Arena A, but should also involve the new networks that were formed during the deployment phase of the project. In G��rtner and Wagner���s work [9, 10], Arena A encloses two sociopolitical processes through which systems are designed and new organizational forms are created. These are necessary and mutually constituted processes whereby (re)designing work shapes technical systems, and vice-versa, and together they create 163