Participatory Design

Pattern number within this pattern set: 
298
Douglas Schuler
Public Sphere Project (CPSR)
Problem: 

A large number of artifacts that people use every day are ill designed and they do not appropriately address the needs of the people for whom they are designed and produced. The problems range from the inconvenient (in setting an alarm on an unfamiliar alarm clock, for example) to the dangerous (an inadequately marked pedestrian crosswalk or scalding water from the tap when cold water was expected). And the design of groupware, software systems that facilitate group collaboration, developers may create systems that embed users in a system like cogs in a machine where a more human-centered system that was more humane — and more effective — could be developed.

Context: 

This pattern is intended to be used in any situation in which a service, policy, or other artifact is being designed. Those who will use the artifact and those who will be affected by it should be included in the design process.

Discussion: 

"The very fact of exclusion from participation is a subtle form of suppression. It gives individuals no opportunity to reflect and decide on what's good for them. Others who are supposed to be wiser and who in any case have more power decide the question for them and also decide the methods and means by which subjects may arrive at the enjoyment of what is good for them. This form of coercion and suppression is more subtle and more effective than are overt intimidation and restraint. When it is habitual and embodied in social institutions, it seems the normal and natural state of affairs." (Dewey 1937)

This "subtle form of suppression" that Dewey identified in the quotation above shows up in sociotechnological systems and in various arenas including the workplace. Without genuine participation in the design process, class, managerial, or other privileges become designed in. That is, sociotechnological systems often carry forward the perquisites and propensities of the designers, intentionally or unwittingly. Cases abound in both cases. Robert Moses, New York City's "construction coordinator," ensured that the bridges over the highways leading to the beaches from New York City were low enough to prevent buses from traveling under them (Caro 1975). This ensured that African Americans and other minorities who often had to rely on public transportation would, in large measure, be confined to the city while the more financially well-to-do could periodically escape to the seaside. There was no need to pass laws when a permanent physical structure could silently and invisibly enforce the color bar Moses preferred.

Frustrated by what they saw as unresponsiveness of software and the impending institutionalization of management prerogatives into software systems, Scandinavian researchers in the late 1970s conceived a new paradigm for software development called participatory design in which end-users worked as co-designers of the systems that they would ultimately use. They believed that adopting a participatory design approach would result in systems that better served users, initially workers in industrial settings. According to PD researchers Finn Kensing and Jeanette Blomberg (1998), "At the center of the critique was the neglect of workers’ interests — those most affected by the introduction of new technology. PD researchers argued that computers were becoming yet another tool of management to exercise control over the workforce and that these new technologies were not being introduced to improve working conditions (see e.g. Sandberg, 1979; Kyng and Mathiassen, 1982). The Scandinavian researchers and workers also worked to on the legislative front to promote "codetermination" laws in Scandinavia that ensured that workers had the right to be involved with technological decisions in the workplace were established (Sandberg et al 1992). They promoted user empowerment through education and "researchers developed courses, gave lectures, and supervised project work where technology and organizational issues were explored (see e.g. Kyng and Mathiassen, 1982)" (Kensing & Blomberg 1988)

Participatory design is an integration of three interdisciplinary concerns that span research and practice: "the politics of design; the nature of participation; and method, tools and techniques for participation" (Kensing & Blomberg 1998). In their paper, "Participatory Design: Issues and Concerns," Finn Kensing and Jeanette Blomberg discuss two primary aspects to this work:

Increasingly, ethnographically-inspired fieldwork techniques are being integrated with more traditional PD techniques (Blomberg et al., 1996; Bødker, 1996; Beyer and Holtzblatt, 1997, Kensing et al., forthcoming). The primary techniques of ethnography include open ended (contextual) interviews and (participant) observations, often supported by audio or video recordings. These techniques are employed to gain insights into unarticulated aspects of the work and to develop shared views on the work.

… Complementing these tools and techniques for work analysis are those focusing on system design such as scenarios, mock-ups, simulations of the relation between work and technology, future workshops, design games, case-based prototyping, and cooperative prototyping (Kensing, 1987; Ehn, 1989; Greenbaum and Kyng, 1991; Trigg et al., 1991; Mogensen, 1992, 1994; Blomberg et al., 1996; Grønbæk et al., 1997). These tools and techniques avoid the overly abstract representations of traditional design approaches and allow workers and designers to more easily experiment with various design possibilities in cost effective ways.

The nature of participatory design has changed over time. In the software world, for example, the focus has shifted from the development of site-specific software systems to the design of web applications and, perhaps more importantly, to the entirety of the information and communication infrastructure, including policy development. The ideas shows up in many guises and even open source communities could be considered a type of participatory design.

Participatory design has been advocated in a number of areas besides software. Architects Lucien Kroll (1987), John Habraken (1972), Christopher Alexander (1984), Michael Pyatok (2000) and others developed a number of techniques for allowing people to design their own working and living spaces. Artist Suzi Gablick, writing in The Reenchantment of Art (1992) describes a number of ways that the creation of art could be more participatory, while many others are advocating participatory approaches to media, policy development, citizen participation journalism (Gillmor 2004). Several books, including The Design of Work Oriented Computer Artifacts (Ehn, 1988), Participatory Design: Principles and Practices (Schuler and Namioka 1993) and Design at Work (Greenbaum and Kyng) helped provide some early guides for the use of participatory design of software and the biannual Participatory Design Conference sponsored by Computer Professionals for Social Responsibility helps to foster its continued evolution.

Participatory design is not a panacea. People may not want to participate; in many cases they quite plausibly determine that the disadvantages outweigh the advantages. Participatory design can certainly be time-consuming and higher quality of the end product cannot be guaranteed. Participatory design projects can go awry in a number of ways (as do traditional and more orthodox software development efforts.) Software users using mental models based on the software are accustomed to using may enter a design session believing that they have already fully designed the system (down to the last key-stroke short-cut). Because of this possibility (and others reasons), many PD approaches focus on fairly general high-level exercises that are fairly far removed both conceptually and physically from computers.

In some cases, a participation trap may be said to exist. This could happen when people are being brought into an effort that will ultimately make matters worse for them. In cases like this a less cooperative, more confrontational approach may be more likely to bring satisfactory results. Participation gives rise to several issues that probably must be resolved on a case-by-case basis in practice. Potential participants understand this instinctively. If, for example, the participative arena is for show only, and no idea that originates with a participant has any chance of being adopted, people can't be faulted for being dubious of the process. Genuine participation should be voluntary and honest; the relevant information, rules, constraints, and roles of all stakeholders should be well-understood by all. (A person may still decide to participate even if any and all benefits would accrue to the organizers.) Ideally, the participants would be part of any decision-making, including when to meet, how to conduct the meetings, and other processes. Kensing (1983) and Clement and Van den Besselaar (1993) describe several requirements for effective PD.

PD principles, techniques, and methodologies will continue to improve and be better known over time. PD will likely continue to involve bricolage, the ability of the participants and the people organizing the process to improvise. Unfortunately, as Kensing and Blomberg (1998) point out, building on the work of Clement and Van den Besselaar (1993), "the experimental nature of most PD projects often leads to small-scale projects which are isolated from other parts of the organization." The best way for the process to continue to improve is to build on successes and create incrementally a culture of participation on the job and in society, that is both equitable and effective at designing systems, services, tools, and technologies whose design better meets the real demands and needs of the people.

Solution: 

There should be a strong effort to include the users of any designed system (software, information and communication systems, administrative services and processes, art, city plans, architecture, education, governance, and others) into its design process in an open, authentic, and uncoerced fashion. Participatory Design, according to Finn and Blomberg, "has made no attempt to demarcate a category of work called cooperative, but instead has focused on developing cooperative strategies for system design… PD is not defined by the type of work supported, nor by the technologies developed, but instead by a commitment to worker participation in design and an effort to rebalance the power relations between users and technical experts and between workers and managers. As such PD research has an explicit organizational and political change agenda.(See Eevi Beck's "P is for Political" (2002) for more insight on this important observation.)

Pattern status: 
Released