Guohua Bai (Ph.D. Guest Professor)



My reading notes



Digest of Participatory Design Approach

Part 1;  Part 2; Part 3


Miller, Steven E., (1993)

From system design to democracy.

Communication of the ACM, Vol 36, No.4, P38.

PD methodologies creat better systems and product, as research and experience have demonstrated. The potential ripple effects of this new approach, however, are less well recognized.

First, PD conributes to changes in the organization in which it takes place. Front - line workers fell increased competence and confidence. Their views are more clearly heared and organization changes to give higher priority to their interest. In the best of circumstance, there is a shift in the power relations within the firm toward more democratic norms.

Second, PD is an extension of larger social forces, in scandinavia a key reason for the emergence of PD approaches is the labor movement´s social democratic orientation.

Third, PD has a potential impact on the larger society as model and source of wisdom for transforming the way government relates to its citizens and for the way the public sector relates to its clients.

PD embodies at leasst two principles that are worth making explicit:

1: Workers-and customers- are intelligent, creative, and productive contributors to organizations if they are empowered to express their insights, apply their expertise, exercise their decision making capabilites, and given responsibility for the impact of their actions.

2. PD holds that, contrary to taylorist belief, good ideas are as likely (perhaps more likely) to come from the bottom up as from the top down.

In calling for worker initiative I am not ignoring the need for leadership. But my father taught me there are two kinds of leders: those about whom, when its all over, people say, "He did it." And those about whom, when its over, people say "We did it." and they feel proud and empowered by the process.

David Osborne and Ted Gaebler describe this way: " 100 years ago the word bureaucracy meant something positive. It connotated a rational, efficient method of organization-something to take the place of the arbitrary exercise of power by authoritatian regimes. Bureacracy brought the same logic to government work that the assembly line brought to the factory. (But) The kinds of goverments that developed during the industrial era,... no longer work very well. They accomplished great things in their time, but somewhere alog the line they got away from us. They become bloated, wasteful, ineffective. And when the world began to change, they failed to change with it."

We need to help people realize that computing and informatin technology does not merely mean word processing, spreadsheets and bankmat; that it is central to the entire shape and direction of our economy, our government, our civil liberties, social services, culture, and the entire quality of life.

Bjerknes, G. (1993)

Some PD Advice.

Communication of the ACM, Vol 36, No.4, p39.

Some advice on how to avoid a failing PD project based on experience collected during the four year Florence project.

What to take care of when the project is established:

. Have Management support

. Specify in a contract how much time the users can/shall spend on the project.

. Have a steering group in which conflicts can be discussed.

. Be sure the required equipment is available for systems experts and users.

What to take care of during the project:

. Listen to the user. If they say there is something they don´t want, don´t do it, even if it seems to be the perfect computer systems for their needs from a systems expert point of view. But do not do everything the users proposed. The users are not always aware of the consequences of their proposals, and they can often be conservative in that they do not utilize the possibilities given by computer technologies. . Do not forget to inform users about any progress, otherwie the users will feel left out, and it will no longer be a PD project.

Carmel, E., Whitaker, R. D. & George, J. F. (1993)

PD and joint application design: A translatlantic comparison.

Communication of the ACM, Vol 36, No.4, PP40-47.

The PD and JAD (Joint Application Design)have established themselves in Scandinavia and North America as influential thrusts in software development. An unambigouous definition of "user" is impossible. The primary difference between various user involvement methodologies and techniques lies in the degree to which the users participate (and therefor influence and are satisfied with) the emerging system design.

PD often termed the "Scandinavian approach" to systems development- advocates a much stronger form of user involvement than that JAD. PD, as we know it today, reprsents a "second generation" of thinking aimed at developing a design methodology based on principle outlined in the "first generation" of trade union inquires into the effect of information systems on the work place. To characterize PD:

1. Workers should be given better tools instead of having their work or their skills automated.

2. Users are best qualified to " ..... determine how to improve their work and their work life."

3. Users perceptions and feeling about technology are as important as technical specifications.

4. Information technology can only be appropriately addresed within the context of the workplace.

Greenbaum lists four American perceptions that interfere with PD acceptance in North America: 1) PD is too idealistic; 2) PD is biased toward workers; 3) PD lack method or model; and 4) PD designers need to rely strictly on experience. The last two issues will be discussed in this paper.

PD techniques: To the extent that PD projects have been documented to data, there is little evidence that a standard set or ordering of practices has been decided. But two themes govern practical implementation of PD principles . The first theme is "mutual reciprocal learning, in which users and deigners teach one another about (respectively) work practices and technical possibilities through "joint experience" The second theme is design by doing, in which interactive experimentation, modeling, and testing support "hands-on design" and learning by doing."

To date, both themes have been pursued largely with "low tech" tools-those with which the users are already familiar, and which they can easily employ themselves. Modeling 1: Visualizing the current workplace:

Historical aspects focuses attention on the historical background of users´ activity to fascilitate them in discussung their individual skill, knowledge, and judgement Modeling 2: Visualizing the possible workplace:

Prototyping: Presentation and evaluation of concrete options

Cooperative prototyping involves the users more than the traditional modes of prototyping, in that they acutually work with a prototype and experience it. When a breakdown occurs, users and designers actively discuss the reason for the breakdown. Prototyping also supports mutual learning by promoting cooperative communiation. Artifacts is much more tangible than the idea.

PD faces considerable obstacles to implementation. There ia a reluctance on the part of both IS professionals and executives to increase user involvement or to experiment with new method and techniques. There are numeriou local problems in successful implementation: managerial resistance, user conservatism, lackluster workshops and poor facilitators.

With reference to the methodology´s history, we might say that JAD represents a movement toward more collaborative practices to enhance the viability of given technical goals. In contrast, PD represents a movement toward more technical practices to enhance the viability of given social goals.

PD strongly promotes a mutual learning process between members of the group: designers and workers. As the design progress, both workers and designers are transformed by learning from one another. PD specifically avoid presenting any "step- by-step" approach, urging designers to improvise and focus on the process aspects of designing. More specifically, PD does not structure the entire time span. PD approach devotes almost no guidelines to structure. Nevertheles, structure has merits; as noted , can actually enhance creativity when introduced properly.

PD References: PD first took root in Scandinavian workplace democracy movement. Strong labor unions, acting as advocates for workers, and a history of socio-technical approaches which argued for the importtance of the social dimension of work with technology, provided fertile soil for those raising concerns about the workplace and social effects of new technologies.

The value of PD was demonstrated in project such as DEMOS, FLORENCE, and UTOPIA and in international conferences such as the Information Systems Research Seminar (IRIS) series, and the IFIP Working Group on Computer and Work series (WG 9.1). A major contribution in Bjerknes, Ehn, and Kyng . Aspects on this approach see CSCW, CHI.

Much of the Sandinavian work retains an explicit commitment to workplace democracy in the context of technological growth and business development that is, direct and effective worker participation (not mere "involvement") in design activities and decisions, within the trade union context. Outside of Scandinavia, the field is more varied, with some theorists and practioners pursuing a locally adapted form of democratic decision making, and others emphasizing effective knowledge acquisition and product quality.

As part of its Workplace Project, Computer Profesionals for Social Responsibility has sponsored two participatory Design Conferences-PDC´90 in Seattle, and PDC´92 in Cambridge. 175 people from 12 nations in PDC´92 were 55% from industry; the rest were from academia, government, and non-profit organizations. Aspects concern historical and intercultural analysis (Bjerknes, Carmel, Whitaker,George...), complex interrelations of design, development, ethics, and politics (Greenbaum, Miller; and Wagner), and importance of process in working collaboratively across the boundaries of expertise, ownersgip and organization (Anderson, Blomberg, Gronbaek...)

Clement, A. and Besselaar, P.D. (1993)

A Retrospective Look At PD Projects.

Communication of The ACM. Vol.36, No.4. PP29-37.

While modern methods for information system development generally accept that users should be involved in some way, the form of the involvement differs considerably. Mostly, users are viewed as relatively passive sources of information, and the involvement is regarded as "functional" in the sense that it should yield better sytem requirements and increased acceptance by users. (not closely as on-line Feedback).

The focus of PD is not only the improvement of the information system, but also the empowerment of workers so they can codetermine the development of the information system and of their work place. Although some researchers in Scandinavia and other European countries had long experience with these approaches to participatory information systems design, it is only in the last few years that PD has recived broader attention.

This article contributes to such historically and theoretically based understanding of PD by offering a ........ While there are by now many reports on PD projects, as well as prescriptive articles that often draw directly on just a few projects for illustration, there are no systematic surveys of .......

By examing 16 different PD projects on the ingredients of 1) Users´ access to relevant information; 2) Independent voice in decision making; 3) User-controlled development resources: time, facilities, expertise; 4) Appropriate development methods, e.g., prototyping; 5) Organizational/technical flexibility.

The earliest reported were conducted by the Norwegian Computing Center(NCC) in the 1970s. Vidal Keul worked with three unions to provide them with knowledge about how the use of new information technology could affect their working conditions and how its introduction can affect their interests. An additional aim was to encourage unions to develop and implement their own technology control activities and polices.

Probably the best knowen PD project, UTOPIA, was led by Susanne Bödker, Pelle Ehn, Morten Kyng and several other researchers from institutions in Sweden and Denmark. Working closely with unions for graphics workers, the overall objective was to contribute to the development of powerful skill-enhancing tools for graphics workers. Stress was placed on the quality of the work and product, not only in the design of technology, but also in training, work organization, and human skills.

Gro Bjerknes initiated the Florence project in a Noewegian hospital. The aim of the project was twofold: first, to develop an information system for providing daily information system about patients while decreasing the paperwork involved, and second, researchers tried to develop instruments that nurses could use when dealing with computers.

Siv Friis describes two research projects carried out recently in Sweden, in which she attempts to evalute the use of the PROTEVS (PROTotyping for EVolutionary system design) PD methodology in a local goverment adminstration seeting. The second case is in a pharmaceutical manufacturing company. A principal focus of the research was to investigate the potential of PD to affect the relationship between users and systems developers.

Only two projects UTOPIA and a later NCC project, were intended to develop a marketable product. A shift from PD in manufacturing industry in the 1970s and early 1980s to PD in offices and service industries in the late 1980s. Female occupations dominate and female researchers is significantly higher than for information systems development as a whole. (Bjerknes, Bödker, Friis).

As Kensing notes, "While the main focus in the early work was on developing the qualifications of workers/trade unions for the purpose of democratization of working life, and to some extent also on developing alternative technologies from the workers´perspective, lately the main focus has been on methods for PD in an organizational setting involving users, systems designers and management" There are contradictions to deal with unions interests in PD setting. They are often not active as supposed. PD, in one case , could get unions support, and in other could mobilized a strike. There is a general lack of commitment of central unions. Friis describes how technology was available for use but only for the duration of the project, after which it was removed from the user site. This resulted in the elimination of prototyping from the techniques available to the users. (It is owing to the lack of on-line feedback mechanism)

Knowledge and skill of workers are not always available for PD, since workers often do not know how their rutin and experience of their jobs can make contributions for the future systems. it is important to collect and prepare the knowledge of the workers, a knowledge through their jobs. (By introducing on-line feedback)

Kensing states that "Participation does not mean "holding hand" all the time." While top managers and library assistants were generally very positive about the project, some middle manager, however, had difficulty dealing with the unusually democratic practices of the project. They appeared "threatened" by the growing competence and assertiveness of those lower than them, which led to delays and frustrations in the design team. The increasing involvement of management in the later projects might indicate a more general change, the political climate is changing, and management is taking more initiative in this area.

Surprisingly, the most ambitiouss PD software development project, UTOPIA, did not achive market success with its state-of art software designed to support skilled graphics workers. Implementation of UTOPIA (in a swedish newspaper) was frustrated by opposition from both management and the union of journalists. Experiences suggest that successes according to the usual criteria of PD projects - active involvement of users, increased learning and communications, and better adapted system- are by themselves not sufficient for local self-sustaining process of participation to continue. The long-term viability of PD needs to make a main ambition of PD research. (long-term on-line feedback can make contribution here)

It appear that an animator, or a group of animators, with strong ties to the work setting is vital. To attract the interest of users, it is important that focus be on addressing their immediate needs.

Novick, D. And Wynn, E.

Participatory Conversation In PD

Communication of The ACM. Vol.36, No.4. P93.


* PD is relatively novel social situation involving people of varying professional, organizational and educational background.

* The primary medium of PD is conversation.

* The participant´s backgrounds set up distinct values, assumptions, and way of handling a conversation.

* The process is participatory only if there is equality in the conversation. The background differences can work subtly against this, despite good intentions. * Mismatched background assumptions show up in the language and structure of the design conversation and can be uncovered and analyzed.

Users do not challenge formal statements of work or ask for technical clarification in order to avoid seeming foolish to technical people. They are not clear about possibilities and limitations of the system-to-be.

Possible miscommunication cues:

. Disclaimers and pre-apologies

. Use of same terms in different contexts or different terms for the same thing

. Responses that don´t match a prior statement

. letting obvious misunderstandings slide

. Consistently different turn.lengths among classes of participants.

. Attempted turns or topics that never succeed.

. Patterns of interruption or utterance completion

. Not laughing at the same jokes.

Wagner, Ina (1993)

A Web of Fuzzy problems: Confronting the Ethical Issues

Communication of The ACM. Vol.36, No.4. P93.

Ethical problems can emerge in a systems design project whenever the legitimacy of the values and moral principles on which participants base their acts and decissions are contested or questioned. Such conflicts between participant´s values and norms of conduct often point to underlying basic differences between their positions in the organization, their interests, and consequently, their assessment of certain design decisions. In this regard, ethical problems have a strong political content.

Engineers´tradition of dealing with ethical issues indirectly, through the definition of "technical norms", rather than through dialogue. But Grudin points out that these norms are not automatically supportive of users´interests but often in conflict with them.

Williams, M. G. and Begg, V. (1993)

Translation between software and Users.

Communication of The ACM. Vol.36, No.4. P93.

User can not be expected to desscribe their work and needss in the language and from the point of view of an engineer. Conversely, engineers seldom have an intutive grasp of their user´s working life, or the environment in which a product will be used. Some kind of two way translation is needed between the user´s domain and the software´s. When the user and designer cannot perform this tranlation, a third role, a model, must be enlisted who acts as a bridge between one domain and another. Translation is not merely concerned with interpretation of task and terminology, but also with workplace culture and politics. Therefore, it may include negotiation or facilitation. (with combination of on-line Feedback-learning)

Harker, S. (1993)

User Participation in Prototyping.

Communication of The ACM. Vol.36, No.4. P77

A strategic program of user centered design was developed by HUSAT to deal with systems for ineractive case handling within a UK government department. The program included a prototyping exercise which not only involved user in providing feedback about the interface being designed, but also created structures which provided feedback on the impact of the proposed system on job design and organization. It is this prototyping (socio-technical prototyping) of both technical and social structures which is unusual and provides lessons for the practice of PD.

Nelson, Dave(1993)

Aspects of participatory; Response.

Communication of The ACM. Vol.36, No.10. P17-18.

PD (as defined by Muller, Kuhn, et al.) is an approach to organizing information system development projects; it putative goal is "empowering" workers. Thus it is a strategy for social change, rather than a method for information systems design. Because the focus of PD is a social engineering, rather than software engineering, it subordinates IS effectiveness to the social agenda of its proponents.

The author is negative to PD in the aspects of empowering workers, and declaims that "what is peculiar to PD is its polarized view of ISs as contested territory in the class wars(Karl Marx would applaud their misguided agenda.)"(p18). His favour to protect employees is a free market in which " producttivity, skills, knowledge and judgment are as marketable as products and services". According to his view "the scandinavian alternative is to make trade-unionist proscriptions and work rules the law of the land. As I understand PD, users should not only be defined as "workers", they can be managers, politician, etc. PDs goals are not to empower workers against management and at the expense of the quality of products or services, but take them as connecte systems as whole to improve and change. How can we bring all of the stakeholders into the design process, to contribute effectively their unique and essential expertise for the benefit of all stakeholders ? A social change can not be reached through only empowering workers.

Muller, Michael J and Kuhan Sarah (1993)

Response to Nelson.

Communication of The ACM. Vol.36, No.10. PP18

Within our respective, IS have effects upon - and are affected by - the social systems of which they are a component. As Keen pointed :" Information system development is an intensely political as well as technical proces .... Information systems increasingly alter relationships, paterns of communication and percived influence, authority, and control".

The design of IS should be a part of overall design of work(i.e., the social construction of work). For us, the important questions to ask regarding social agendas are therefor: Can the design be improved so that it serves the diverse social agendas of all of its stakeholders? How can we bring all of the stakeholders into the design process, to contribute effectively their unique and essential expertise for the benefit of all stakeholders ? What social and technological processes can facilitate this mutual exchange and education ?

Kensing,F. and Munk-Madsen, A. (1993)

PD: Structure in the Toolbox.

Communication of The ACM. Vol.36, No.4. Pp78-85..

In Scandinavia user participation arose in the context of action-oriented research with trade unions in the late 1970s. We argue that system development projects fail in communication even though they use the most promising techniques.

Two projects (prototype) shows all intended functions but they are not implemented as required since a "silent feedback" from the users. Most papers and books deal specifically with techniques and tools, not with underlying theories enabling us to discuss the context and the limitations of the techniques and tools. Survys of method are usually through on details but lacking in explanatory theory.

User-developer communication in system development

Communication is of course a key issue in collective activities such as systems development. People with different backgrounds, education, training, and organizational roles exchange facts, opinions, and maybe even threaten one another.

Two communication models: Traditional and alternative model.

A traditional model view communication as a tube through which a message is transmitted to a specific receiver. The quality of communication is determined by the sender´s ability to form a rigorous message.

An alternative model focusing on the reciver´s changes triggered by the recived message. "Communication depands on not what is transmitted, but on what happens to the person who recives it."

The main domains of communication in design:

* User´s present work;

* Technological options

* New systems

These three domains knowledge must be clearified and integrated in the process of new systems design. This can provide a category for building a message catalog.

Blomberg, J., Mclaughlin, D. and Suchman, L. (1993)

Work-oriented design at Xerox

Communication of The ACM. Vol.36, No.4. P91.

For over 10 years, a network of researchers and product designers at Xerox Corporation have been exploring the relevance of work practice studies and applications codesigned for product development. We draw expertise from the fields of anthropology, sociology, computer science, human factors, graphics design, psychology and industrial design.

We are looking for new ways of getting access to the details of everyday work practices in multiactivity, technology-intensive work settings. The driving force behind our approach is the conviction that attempts to hand-off the results of field studies to product designers and developers are bound to fail. Instead, we are aiming for a situation in which we have ongoing relations with relevant work-sites, new ways of representing what we learn, and new ways of conveying those lessons to others not directly involved in our studies.(Feedback functions)

Walz, D. B., Elam, J. J. and Crrtis B. (1993)

Inside a Software Design team: Knowledge, acquisition, sharing, and integration.

Communication of The ACM. Vol.36, No.10. P70

Conflict was the mechanism for facilitating learning. It was not a debilitating factors needing to be suppresed in the software design team. In fact, we recommend consideration of formal techniques for managing conflict to help with knowledge acquisition, sharing and integration. Two techniques for programming conflict into organization decision making process have been suggested: the devil´s advocate decision program (DADP) and the dialectic method (DM). In the DADP, an individual or group plays the formal role of critic in order to help adecision maker test the assumptions and the logic of the ultimate decision. The dialectic method pits a thesis against an antithesis. Most modern legal systems today are foral dialectic processes. Two sides exist, each with champions, and case are made for each. (dialectical and contradictory approach).

Hyman, R.B. (1993

Creative Chaos in High-performance Teams: An Experience Report.

Communication of The ACM. Vol.36, No.10. Pp57-58.

Management has traditionally viewed chaos as negative, indicating an organization at risk, a group out of control, or a state out of control. But in chinese philosophy the universe began in a state called chaos, the symbol for which is a combination of two signs: the symble for confusion, and symble for light. in recent years, both science and management have come to view chaos more in line with Chinese philosophy. One of our most difficult realizations was that some talented individuals cannot flourish in a team-oriented environment. Management must recognize a mismatch and address problems before team goals are jeopardized. The key is to expect problems and to creat a culture where finding and exposing problems is rewarded, not punished. Nutual trust is a very important precondition.

Change is not only a cause of chaos, change is a law of life. We will never have enough budget, time or resources to meet all of our objectives. "The tru objective is to take the chaos as given and learn to thrive on it."

<----- Get Part 3 of Digest  Return to Reading Notes