Home/red-rock-eater/Human Values and Design of Computer Technology

Human Values and Design of Computer Technologywriting

educationmediaactivisminternet-culturedemocracysocial-computingHCI
27 min read · Edit on Pyrite

Source

Automatically imported from: http://commons.somewhere.com/rre/1997/Human.Values.and.Design..html

Content

This web service brought to you by Somewhere.Com, LLC.

Human Values and Design of Computer Technology

```

Human Values and Design of Computer Technology

edited by Batya Friedman

published by CSLI Publications and Cambridge University Press

January 1998

hardcover $64.95 ISBN 1-575-86081-3 paperback $24.95 ISBN 1-575-86080-5

Contents

Introduction Batya Friedman

PART I Conceptualizing Human Values in Design

1 Bias in Computer Systems Batya Friedman and Helen Nissenbaum

2 Accountability in a Computerized Society Helen Nissenbaum

3 Disability, Inability and Cyberspace John Perry, Elizabeth Macken, Neil Scott, and Janice L. McKinley

4 Do Categories Have Politics? The Language/Action Perspective Reconsidered Lucy Suchman

5 Categories, Disciplines, and Social Coordination Terry Winograd

6 Commentary on Suchman Article and Winograd Response Thomas W. Malone

7 Social Impact Statements: Engaging Public Participation in Information Technology Design Ben Shneiderman and Anne Rose

PART II 135 Computers as Persons? -- Implications for Design

8 Computers Are Social Actors: A Review of Current Research Clifford I. Nass, Youngme Moon, John Morkes, Eun-Young Kim, and B.J. Fogg

9 When the Interface Is a Face Lee Sproull, Mani Subramani, Sara Kiesler, Janet Walker, and Keith Waters

10 'Social' Human-Computer Interaction Sara Kiesler and Lee Sproull

11 Reasoning About Computers As Moral Agents: A Research Note Batya Friedman and Lynette I. Millett

12 Interface Agents: Metaphors with Character Brenda Laurel

13 Human Agency and Responsible Computing: Implications for Computer System Design Batya Friedman and Peter H. Kahn, Jr.

PART III Practicing Value-Sensitive Design

14 Workplace Database Systems: Difficulties of Data Collection and Presentation Harry Hochheiser

15 Eliminating a Hardware Switch: Weighing Economics and Values in a Design Decision John C. Tang

16 Steps toward Universal Access Within a Communications Company John C. Thomas

17 Social Choice About Privacy: Intelligent Vehicle-Highway Systems in the United States Philip E. Agre and Christine A. Mailloux

Introduction Batya Friedman

Many of us when we design and implement computer technologies focus on making a machine work -- reliably, efficiently, and correctly. Rarely do we focus on human values. Perhaps we believe in value- neutral technology. Perhaps we believe that issues of value belong only to social scientists, philosophers, or policy makers. Neither belief is correct. In their work, system designers necessarily impart social and moral values. Yet how? What values? Whose values? For if human values -- such as freedom of speech, rights to property, accountability, privacy, and autonomy -- are controversial, then on what basis do some values override others in the design of, say, hardware, algorithms, and databases? Moreover, how can designers working within a corporate structure and with a mandate to generate revenue bring value-sensitive design into the workplace?

This volume brings together leading researchers and system designers who take up these questions, and more. In this introduction, I situate these questions within a larger conceptual framework. I also motivate the need to embrace value-sensitive design as part of the culture of computer science. Then I offer an overview of each of the chapters. Roughly half of the chapters are new material, and the remainder are reprints of pivotal articles from recent years. For readers with specialized interests, Table 1 arranges the chapters by technical topic; Table 2 by core human value.

Does Technology Have Values?

Does technology have values? Consider a few examples. In the early 1900s, missionaries introduced a technological innovation -- steel ax heads -- to the Yir Yoront of Australia, a native people. The missionaries did so without regard for traditional restrictions on ownership, and indiscriminately distributed the ax heads to men, women, old people, and young adults alike. In so doing, they altered relationships of dependence among family members and reshaped conceptions of property within the culture (Sharp, 1952/1980). Another example: About four decades ago, snowmobiles were introduced into the Inuit communities of the Arctic, and have now largely replaced travel by dog sleds. This technological innovation thereby altered not only patterns of transportation, but symbols of social status, and moved the Inuit toward a dependence on a money economy (Houston, 1995; Pelto, 1973). Or a computer example. Electronic mail rarely displays the sender's status. Is the sender a curious lay person, system analyst, full professor, journalist, assistant professor, entry level programmer, senior scientist, high school student? Who knows until the e-mail is read, and maybe not even then. This design feature (and associated conventions) has thereby played a significant role in allowing electronic communication to cross traditional hierarchical boundaries and to contribute to the restructuring of organizations (Sproull & Kiesler, 1991). The point is this: In various ways, technological innovations do not stand apart from human values. But, still, what would it mean to say that technology has values?

Endogenous and exogenous views have emerged in recent years. The endogenous view holds that through the process of designing and implementing technology, the very meaning and intentions that we designers and builders bring to our task literally become a part of the technology (cf. Appadurai, 1988; Cole, 1991). But this view seems false in the sense that it imputes mental states to things which do not have the capacity to have them. Is the claim, for example, that a gun itself intends to kill someone? That proposition seems incorrect. Although arguable, it similarly seems incorrect to impute mental states to computational systems. In chapter 13, Friedman and Kahn discuss this position in more detail. In contrast, the exogenous view holds that values reside with the users, not the technology. Proponents of this view are likely to say "guns don't kill people, people do"; or similarly, "hammers don't pound nails, people do" or "computer databases don't track workers, people do." But this view ignores our everyday experience that particular tools lend themselves to particular tasks in particular ways; otherwise, one tool would serve all purposes well and we know this is not the case. Try pounding a nail with a computer database.

A third view -- which I think is correct -- holds that the way people design tools makes the tools more suitable for some uses over others. Scheffler (1991) suggests, for example, that a hammer is well suited for driving nails, poorly suited as a soup ladle, and perhaps amenable as a doorstop, bookend, or paper weight. In other words, human activity is constrained and perhaps even prodded by features of the technology, but not determined by it.

In terms of computer system design, we are not so privileged as to determine rigidly the values that will emerge from the systems we design. But neither can we abdicate responsibility. For example, let us for the moment agree with Perry et al. in chapter 3 that disabled people in the workplace should be able to access technology, just as they should be able to access a public building. As system designers we can make the choice to try to construct a technological infrastructure which disabled people can access. If we do not make this choice, then we single-handedly undermine the principle of universal access. But if we do make this choice, and are successful, disabled people would still rely, for example, on employers to hire them. Suitability (as a feature of the technology) is necessary for, but does not assure equal access (a human value).

Users, of course, are not always powerless when faced with unwelcomed value-oriented features of a technology. For example, students in one of my classes conducted an informal ethnographic investigation of computer use in a local accounting firm. The previous year, this firm had acquired an off-the-shelf packaged software program for preparing tax returns that automatically reported the number of tax returns completed by each employee. But such overt comparisons among employees resulted in office discord. Employees did not believe that the numbers fairly portrayed their overall value to the firm, let alone the real income they generated. Note that this problem arose unsolicited because of the built-in features of the software. This firm operated within a cooperative family-style work culture. Thus, in this situation, the firm members agreed to destroy the comparative data when it was produced. However, it takes courage to turn away from quantitative data that tell us how things are going even if we rationally believe the numbers are misleading and incomplete.

Human Values: A Working Definition

In some sense, we can say that any human activity reflects human values. I drink tea instead of soda. I recently attended a Cezanne exhibit instead of a ball game. I have personal values. We all do. But these are not the type of human values which this volume takes up. Rather, this volume is principally concerned with values that deal with human welfare and justice.

This demarcation between personal and moral values follows from the psychological literature which has investigated empirically what have been referred to as three domains of social knowledge: the moral, conventional, and personal (Nucci, 1996; Smetana, 1983; Turiel, 1983). The moral domain refers to prescriptive judgments which people justify based on considerations of justice, fairness, rights, or human welfare. Conventional issues refer to behavioral uniformities designed to promote the smooth functioning of social interactions. Personal issues refer to those which lie under the jurisdiction of the self. Over eighty-five empirical studies, in western and nonwestern countries, have demonstrated that young children, adolescents, and adults differentiate these three forms of knowledge (for reviews of the literature, see Smetana, 1995; Tisak, 1995; Turiel, 1983, in press; Turiel, Killen, & Helwig, 1987). In addition, many of these studies have analyzed how social issues can often embody all three aspects (Smetana, 1982; Turiel, Hildebrandt, & Wainryb, 1991). Abortion, for example, is a particularly controversial social issue not only because it involves a controversial assumption about when life begins, but because personal values (that a woman has personal jurisdiction over her own body) can conflict with moral values (that involve the potential welfare and rights of a fetus). It is also the case that personal and conventional issues can become moral, in specific contexts. For example, most people in western countries believe that if a man chooses to wear a bathrobe, that is his personal choice. But in certain contexts there are also conventions about appropriate dress. Thus, if a man wears his bathrobe to a dignitary's funeral, many people would judge the act not only as a conventional violation, but a moral violation -- as a sign of disrespect.

This form of analysis can bring clarity into understanding the social aspects of computing. For example, a particular feature of a computer system (the positioning of the track ball on a laptop computer) may make a system easier to use. But unless the difficulty of use systematically prevents certain groups of individuals from using the technology, ease of use per se is not a moral value, but a personal one. Or consider a voting booth for use in national elections. A person may prefer to cast his or her vote by marking an old-fashioned "X" next to a candidate's name instead of using an electronic voting system that requires punching out a dot on a voting card. But if the person can easily vote with the latter system, then preferring one system over the other reflects a personal value. On the other hand, if the system requires a short voter to pull a lever that is above his or her reach, as has actually happened (Roth, 1994), then the system prevents fair participation in the voting process. Accordingly, the human values at stake become moral, not simply personal.

When designing computer systems, conflicts also often arise between conventional and moral considerations. Think of standardization. On the one hand, standardization relies on conventional uniformities to enhance the smooth functioning among different components of a system or among different users of a system. Imagine if communication protocols on the internet were not standardized. In this light, standardization is vital. On the other hand, standardization requires individuals to give up some degree of control over the technology. Think of the frustrations some workers encounter when companies require every worker to use the identical word processor. Indeed, the debate in this volume between Suchman (chapter 4) and Winograd (chapter 5) can be understood in these terms. Suchman objects to standardized communication on the grounds that individuals are coerced by the system. Winograd counters that through the social conventional coordination of communication, more successful and beneficial communication is supported (and individual welfare increases).

The human values addressed in this volume principally refer to moral values. But I prefer to use the broader term "human values" instead of simply moral values to highlight the complexity of social life, and to provide the basis for analyses wherein personal and conventional values can become morally implicated.

Distinguishing Value-Sensitive Design from Usability

Some system designers have a tendency to conflate value-sensitive design with usability. This problem arises with good reason. Namely, usability is itself a human value, although not always a moral one. But when it is, both can be addressed by the same design. For example, systems which can be modified by users to meet the needs of specific individuals or organizations enhance their usability (cf. Adler & Winograd, 1992). At the same time, such systems can help users to realize their goals and intentions through their use of the technology -- a human value which Nissenbaum and I (Friedman, 1996) refer to as autonomy.

Sometimes, however, usability conflicts with value-sensitive design. Nielsen (1993), for example, asks us to imagine a computer system that checks for fraudulent applications of people who are applying for unemployment benefits by asking applicants numerous personal questions, and then checking for inconsistencies in their responses. Nielsen's point is that some people may not find the system socially acceptable (based on the value of privacy) even if the system receives high usability scores -- that is, it is easy to learn, easy to use, error free, efficient, and so on. Indeed, in this volume, Tang (chapter 15) provides a case study of just such a conflict. Tang describes the issues which a design team faced in deciding how to power a microphone. The solution they settled upon was to power a microphone directly from the workstation. They thereby eliminated a separate hardware on/off switch on the microphone. From a usability perspective, Tang points to benefits from this design decision insofar as a separate battery could wear down, and thereby cause an inconvenience to the user. Moreover, to preserve the battery users would no longer need to remember to turn the microphone off when it was not in use. However, despite these usability advantages, Tang also shows that the design undermined the moral values which some users placed on privacy and security.

In short, our language and conceptualizations within the field currently provide solid means by which to pursue issues of usability (cf. Adler & Winograd, 1992; Norman, 1988; Norman & Draper, 1986; Nielsen, 1993). This work is important. At a minimum, usability represents social organizational values needed to make systems work in a functional sense. At times, usability can also support human values of moral import. But at times we need to give ground on usability to promote human values, and, conversely, at times we need to give ground on human values to promote usability. Such optimizations require judicious decisions, carefully weighing and coordinating the advantages of each.

About This Book

From a variety of perspectives, there has been increasing interest in the social implications of computer technology. Methods from anthropology, sociology and psychology, for example, have been used in understanding computing and human values from individual and organizational perspectives (Kling, 1980; 1996; Nass & Reeves, 1996; Sproull & Kiesler, 1991; Suchman, 1987; Zuboff, 1988). The literature on computer ethics has revealed the philosophical dimensions of computing (Ermann, Williams, & Gutierrez, 1990; Forester, 1989; Johnson, 1994; Johnson & Nissenbaum, 1995; Ladd, 1989; Moor, 1985). The computer-supported cooperative work (CSCW) and participatory design (PD) communities have enhanced collaboration and the democratic participation of individuals within organizations (Blomberg, Suchman, & Trigg, 1996; Bodker, 1991; Clement & Van den Besselaar, 1994; Ehn, 1988; Greenbaum & Kyng, 1991; Greif, 1988; Kuhn, 1996; Schular & Namioka, 1993). More broadly, the human-computer interaction (HCI) community has been increasingly aware of and concerned about the social dimensions of design (Baecker, Grudin, Buxton, & Greenberg, 1995; Brown & Duguid, 1994; Fernandes, 1995; Laurel, 1990; Shneiderman, 1992; Tognazzini, 1995; Winograd, 1996; Winograd & Flores, 1986). All of these perspectives have offered important if not crucial lines of inquiry. This volume adds to this growing body of literature by offering a principled and practical framework for promoting human values in system design, particularly values with moral import.

Part I: Conceptualizing Human Values in Design

The papers in this first section examine computer technology in relation to the values of accountability, autonomy, freedom from bias, privacy, and universal access. Some of these papers illustrate how value-oriented problems can be closely tied to technical decisions. For example, Friedman and Nissenbaum (chapter 1) include an analysis of Corbats's multi-level scheduling algorithm which has been implemented in a variety of timesharing systems. They suggest that when systems are placed under heavy use, the algorithm may unfairly discriminate against users with long-running computations. Other value-oriented problems are more deeply embedded within a social context. For example, Nissenbaum (chapter 2) analyzes systems which are constructed out of modules or modified from previously existing code in order that no single individual understands the entire system. In such situations, it can be difficult to identify the locus of responsibility for computer-mediated action. Similarly, Perry et al. (chapter 3) show that in order to solve the problem of access for the disabled, designers may need to find the best (or a reasonable) balance between the responsibilities of the individual user and the responsibilities of the computer industry, if not the larger society.

Over forty years ago, Norbert Wiener said that in the coming years access to information would be relatively easy, and that the hard problem for computer science would be to provide access to useful information. Today, Wiener would find little disagreement. Just note some of the millions of users as they try (often unsuccessfully) to meaningfully access the billions of accessible words on the World Wide Web. Many of the emerging solutions to this problem currently depend on filters of some sort. Filters, in turn, depend on the activity of categorization: being able to identify items which are similar along one or more dimensions, and different along others. It is here that the three chapters on categorization by Suchman, Winograd, and Malone are of great importance.

Suchman (chapter 4) begins by offering strong words of caution: that whoever determines the categories -- and how those categories can be used -- imputes their own personal values into the system and has power over the user. As noted above, Winograd (chapter 5) responds to Suchman by pointing to the frequent need for socially coordinated activity in which groups of people seek to share information and technology. In such activity, Winograd argues that we need some degree of standardization wherein designers impose categories. That is their job. The key, according to Winograd, is to cultivate regularized activity without becoming oppressive: or, in a sense, to use power wisely. In turn, Malone (chapter 6) responds to Suchman by arguing that not only are categories often useful, but to some extent they are necessary given the structure of human cognition. At the same time, Malone suggests that no category system is complete, and therefore designs need to be adaptable, often by their users.

Consider this debate in terms of this present introduction. Recall that Table 1 categorizes the chapters by technical topic, and Table 2 categorizes the chapters by human value. Suchman would have us recognize that as the editor of this volume, I have imposed my own categories onto these chapters. For example, I categorized the chapters by Suchman, Winograd, and Malone as involving the human value of autonomy. But at least three other decisions were possible. First, I could have chosen one or more other categories: following Suchman, perhaps that of politics or power. Such a choice would lead in different, and interesting, directions. Second, I could have provided no categorization. But following Winograd and Malone, categories can serve as an effective way to quickly orient readers to the subset of chapters (aka the subset of information) that might be of greatest interest to them. Third, I could have asked the contributors to categorize their own chapters, either according to a predetermined set of categories or to a category of their choosing. While such delegation was certainly possible, it would have transformed my role from one of editor, seeking to shape and provide coherence to this volume, to one of coordinator seeking to reflect the perspectives of the contributors. Since multi-authored volumes often suffer from a lack of coherence, I decided that delegating control was on balance a poor choice. My point is that the tensions encountered in using and selecting categories to characterize the chapters in this volume illustrate the same issues we face with the use of categories in system design: that appropriate, if not good, resolutions depend on finding in context the right balance between competing values.

Shneiderman and Rose (chapter 7) move this discussion on value- sensitive design to the level of codification. Based on the idea of an environmental impact statement, they propose that designers be encouraged, if not required, to provide a "social impact statement" which assesses the social impact of their designs on direct and indirect users. A key toward developing one's social impact statement would be the early identification and involvement of stakeholders in the design. Shneiderman and Rose discuss this proposal in the context of their current design for the Maryland Department of Juvenile Justice. Agre and Mailloux (chapter 17) also recommend such stakeholder involvement in their discussion of intelligent vehicle highways.

Part II: Computers as Persons? -- Implications for Design

Whether a designer chooses to present the technology to the user as an "agent," "person," "persona," or some other anthropomorphized form, might not appear to be a heavily laden value question. After all, we rarely think of the anthropomorphism of automobiles, sailing vessels, or stuffed animals as cause for alarm. The concern with computer technology arises, however, from the way in which people engage with computer technology as if it were a person to a greater extent than previously realized. Thus, as system designers, we need to understand how anthropomorphism occurs, and then decide whether or not to design features which promote it. This is the distinction I was making earlier between suitability (design features) and values (promoting computers as persons). Possibly for the first time in one place, this part brings together the growing body of work on this topic.

Typically, we think of social interaction as involving two or more people, even if people are physically distant from one another, as when talking on the telephone, reading electronic mail, or even having an internal dialogue with a friend. But can an interaction be social, or at least fully social, with computer technology? Nass et al. (chapter 8) lean toward an affirmative response. They begin by documenting that with minimal social cues (e.g., text-only), people will exhibit behavior with computers that in comparable situations parallels their behavior with other human beings (e.g., following politeness norms). Accordingly, they suggest that there may be a place to design, for example, "likable" computers which flatter their users and that adhere to human conventions of politeness. More boldly, following the title of their chapter, "Computers are Social Agents", they suggest that in system design interpersonal communication should become the metaphor and mechanism for human-computer interaction.

Sproull and Kiesler and their colleagues (chapter 9) provide evidence which support Nass et al.'s initial claim. In their study, subjects changed their personas and behavior in the presence of a simulated face (e.g., they presented themselves more positively to the simulated face than to a text interface) and attributed some personality traits to the simulated face. Other results reported by Kiesler and Sproull (chapter 10, which expands on their work in chapter 9) show that people will trust and behave cooperatively with a computer partner. Yet, in contrast to Nass and his colleagues, Kiesler and Sproull offer a more cautious interpretation of the data. It is one thing to say, as they do, that when the interface takes on a human persona through characteristics people typically associate with people -- such as human speech and physiognomy -- then it is likely to elicit what looks like social behavior in users. It is another thing to say, which they do not, that people think of machines as people.

Part of the difficulty in interpreting the above studies arises by trying to use data on human social behavior in interaction with computers to infer human cognition. Thus Friedman and Millett (chapter 11) offer a different methodology: social-cognitive interviews. Based on this method, they assessed individuals' reasoning about computer agency and computer capabilities, and their judgments of moral responsibility in two situations that involved delegation of decision making to a complex computer system. Results showed a complex pattern of results. On the one hand, most of the individuals attributed either decision making or intentions to computers, and over one-fifth of the individuals explicitly held computers morally responsible for error. On the other hand, most of the individuals judged computer decision making and computer intentions to be different from that of humans. Reasoning focused on rule-based systems (e.g., "[the computer] can decide in a sense that somebody has programmed rules, which it follows, and in that sense it chooses a course") and algorithmic processes (e.g., "[the computer] is deciding based on a very clear strict algorithm... it's a decision but not an open-ended one"). Based on these and other data, this study does not support the stronger claims by Nass et al. that people treat computers as social actors. It does suggest that in certain respects people attribute agency to computer technology, and that such attributions go well beyond superficial use of language.

Laurel (chapter 12) would probably be neither surprised nor bothered by a user's attribution of agency to a computer. In her chapter, written in 1990 before all of the other writings in this section, Laurel distinguishes between full-bodied and unidimensional characters. By characters she means something like portrayals of people, as in characters in a play. Full-bodied characters, like real people, are complex and at times engage in unpredictable behavior. Accordingly, Laurel suggests that as a metaphor for interface design full-bodied characters lead to ineffective designs. She proposes that the appropriate metaphor should be that of an unidimensional character (or caricature): simple and predictable. Laurel suggests that such an interface would facilitate usability by minimizing ambiguity. It would also build on a user's inclination to attribute agency to a computer.

Alternatively, Friedman and Kahn (chapter 13) are reluctant to build on such user inclinations. Their reasoning follows from the philosophical position that computers, as they are known today in material and structure, are not persons capable of moral agency. They also show how serious problems arise when human beings believe to the contrary and hold computers responsible for computer-mediated action. Similar arguments are made by Nissenbaum (chapter 2) in the broader context of her discussion on maintaining accountability in a computerized society. If this line of reasoning is correct, then it follows that the use of social cues in computer systems to elicit social behavior constitutes a form of deception. Of course, deceptions may not always be wrong. A social psychologist, for example, may initially deceive his or her subjects on the nature of an experiment in order to collect certain types of social-scientific data. Thus, even if one argued that encouraging a social interaction with a computer counts as a form of deception, a decision would still need to be made on whether such deception is warranted. Yet, in most cases, Friedman and her colleagues argue it is not. This does not mean that all agent technology need be abandoned. Rather, Friedman and her colleagues argue for system design that clarifies and makes salient to the user that the computer is an "it" and not a "who" -- a machine and not a person -- and that only people can be responsible for computer-mediated action.

Part III: Practicing Value-Sensitive Design

Until now, a designer working within a corporate structure might say something like "nice theory, but so what?" How is it really possible to bring value-sensitive design into the workplace? The four chapters within Part III take up this question. The chapters report on diverse work environments, including hardware development in a medium-sized hardware company (SunSoft), research and development efforts in a communications corporation (NYNEX), in-house software development in a large urban hospital (the name is being kept anonymous), and policy analysis of intelligent vehicle highway systems at an academic institution (University of California, San Diego).

As a collection, these chapters call attention to three common questions that arise when designers explicitly address human values. First, how can one design proactively so as to minimize harm and injustice to human lives? Hochheiser (chapter 14) discusses his efforts to do so in the design of a database to aid instrument repair technicians in a large urban hospital. Part of Hochheiser's challenge was to anticipate abuses which had not yet occurred and to convince potential users and management that preventing such abuses are worthy of design time and resources. Second, is there an economic benefit to attending to human values in design? Tang (chapter 15) says that in most cases the answer is yes; and that it is the designer's responsibility to examine these benefits, and to speak the economic language of the corporate world. Toward estimating the economic benefits of value-sensitive design, Tang suggests that one can estimate lost revenue from potential buyers who would otherwise not purchase the product. In Tang's case, buyers with concerns for security, such as the federal government, might not purchase a workstation that lacked a hardware on/off switch on a microphone. Third, when in the design process is the right time to address human values? In what way? In all four chapters, the authors agree that the time to get involved is as early in the design process as possible and to continue that involvement throughout the project. Moreover, both Hochheiser (chapter 14) and Tang (chapter 15) show how important it is to address human values in the context of one's own design projects; Thomas (chapter 16) shows how to influence organizational practice and culture; and both Thomas (chapter 16) and Agre and Mailloux (chapter 17) show how to influence professional organizations, standards, and policy.

Conclusion

I have sought to convey some of the challenges and controversies which are arising in this emerging field of human values and system design. I have also suggested that as a community we need to embrace value-sensitive design as part of the culture of computer science. To do so moves us in important directions. It moves us toward the conceptualizations needed to identify shortcomings in current designs and to seek remedies which promote human well-being. It moves us toward the language needed to discuss the often immense social consequences of our work with the public at large. And it moves us toward holding out value-sensitive design as a criterion -- along with the traditional criteria of reliability, efficiency, and correctness -- by which systems may be judged poor and designers negligent. As with the traditional criteria, we need not require perfection, but commitment, which contributors to this volume demonstrate, with verve.

References

Adler, P.S., and T. Winograd, eds. 1992. Usability: Turning Technologies into Tools. Oxford: Oxford University Press.

Appadurai, A., ed. 1988. The Social Life of Things. New York: Cambridge University Press.

Baecker, R., J. Grudin, W. Buxton, and S. Greenberg, eds. 1995. Readings in Human-Computer Interaction: Toward the Year 2000 (2nd edition). San Francisco: Morgan Kaufmann.

Blomberg, J., L. Suchman, and R.H. Trigg. 1996. Reflections on a Work-Oriented Design Project. Human-Computer Interaction, 11(3):237-265.

Bodker, S. 1991. Through the Interface: A Human Activity Approach to User Interface Design. Hillsdale, NJ: Lawrence Erlbaum.

Brown, J.S., and J. Duguid. 1994. Borderline Issues: Social and Material Aspects of Design. Human-Computer Interaction, 9(1):3-36.

Clement, A., and P. Van den Besselaar. 1993. A Retrospective Look at PD Projects. Communications of the ACM, 36(4):29-37.

Cole, M. 1991 (April). Discussant to B. Friedman and P. H. Kahn, Jr. (Presenters). Who is Responsible for What? and Can What be Responsible? The Psychological Boundaries of Moral Responsibility. Paper session at the biennial meeting of the Society for Research in Child Development, Seattle.

Ehn, P. 1988. Work-Oriented Design of Computer Artifacts. Stockholm: Arbetslivscentrum. (Distributed by Lawrence Erlbaum, Hillsdale, NJ).

Ermann, M.D., M.B. Williams, and C. Gutierrez, eds. 1990. Computers, Ethics, and Society. Oxford: Oxford University Press.

Forester, T., ed. 1989. Computers in the Human Context. Cambridge, MA: The MIT Press.

Fernandes, T. 1995. Global Interface Design. Boston: AP Professional.

Friedman, B. 1996. Value-Sensitive Design. Interactions, III(6):17-23.

Greenbaum, J., and M. Kyng. 1991. Design at Work. Hillsdale, NJ: Lawrence Erlbaum.

Greif, I., ed. 1988. Computer-Supported Cooperative Work: A Book of Readings. San Mateo, CA: Morgan Kaufmann.

Houston, J. 1995. Confessions of an Igloo Dweller. New York: Houghton Mifflin.

Johnson, D.G. 1994. Computer Ethics (2nd edition). Englewood Cliffs, NJ: Prentice Hall.

Johnson, D.G., and H. Nissenbaum, eds. 1995. Computers, Ethics and Social Values. Englewood Cliffs, NJ: Prentice Hall.

Kling, R. 1980. Social Analyses of Computing: Theoretical Perspectives in Recent Empirical Research. Computing Surveys, 12(1):61-110.

Kling, R., ed. 1996. Computerization and Controversy: Value Conflicts and Social Choices (2nd edition). Boston: Academic Press.

Kuhn, S. 1996. Design for People at Work. In Bringing Design to Software, ed. T. Winograd, 273-289. Reading, MA: Addison-Wesley.

Ladd, J. 1989. Computers and Moral Responsibility: A Framework for Ethical Analysis. In The Information Web: Ethical and Social Implications of Computer Networking, ed. C. Gould, 207-227. Boulder, CO: Westview Press.

Laurel, B., ed. 1990. The Art of Human-Computer Interaction. Reading, MA: Addison-Wesley.

Moor, J.H. 1985. What is Computer Ethics? Metaphilosophy(16):266-275.

Nass, C., and B. Reeves. 1996. The Media Equation: How People Treat Computers, Television and New Media Like Real People and Places. Stanford, CA: CSLI Publications.

Nielsen, J. 1993. Usability Engineering. Boston: AP Professional.

Norman, D.A. 1988. The Psychology of Everyday Things. New York: Basic Books.

Norman, D.A., and S. W. Draper, eds. 1986. User Centered System Design. Hillsdale, NJ: Lawrence Erlbaum.

Nucci, L. 1996. Morality and the Personal Sphere of Actions. InValues and Knowledge, eds. E.S. Reed, E. Turiel, and T. Brown, 41-60. Mahwah, NJ: Lawrence Erlbaum.

Pelto, P.J. 1973. The Snowmobile Revolution: Technology and Social Change in the Arctic. Menlo Park, CA: Cummings Publishing Company.

Roth, S.K. 1994. The Unconsidered Ballot: How Design Affects Voting Behavior. Visible Language(28):48-67.

Scheffler, I. 1991. In Praise of the Cognitive Emotions. New York: Routledge.

Schuler, D., and A. Namioka, eds. 1993. Participatory Design: Principles and Practices. Hillsdale, NJ: Lawrence Erlbaum.

Sharp, L. 1980. Steel Axes for Stone-Age Australians. In Conformity and Conflict, eds. J. P. Spradley and D. W. McCurdy, 345-359. Boston: Little, Brown, & Company. (Reprinted from Human Organization, 1952(11) 17-22.)

Shneiderman, B. 1992. Designing the User Interface: Strategies for Effective Human-Computer Interaction (2nd edition). Reading, MA: Addison-Wesley.

Smetana, J.G. 1982. Concepts of Self and Morality: Women's Reasoning about Abortion. New York: Praeger.

Smetana, J.G. 1983. Social-Cognitive Development: Domain Distinctions and Coordinations. Developmental Review(3):131-147.

Smetana, J.G. 1995. Morality in Context: Abstractions, Ambiguities and Applications. In Annals of Child Development (Vol. 10), ed. R. Vasta, 83-130. London: Jessica Kingsley.

Sproull, L., and S. Kiesler. 1991. Connections: New Ways of Working in the Networked Organization. Cambridge, MA: The MIT Press.

Suchman, L. 1987. Plans and Situated Actions. Cambridge, UK: Cambridge University Press.

Tisak, M.S. 1995. Domains of Social Reasoning and Beyond. In Annals of Child Development (Vol. 11), ed. R. Vasta, 95-130. London: Jessica Kingsley.

Tognazzini, B. 1995. Tog on Software Design. Reading, MA: Addison-Wesley.

Turiel, E. 1983. The Development of Social Knowledge. Cambridge, UK: Cambridge University Press.

Turiel, E. In press. The Development of Morality. To appear in Handbook of Child Psychology, 5th Ed., Vol. 3, ed. W. Damon : ed. N. Eisenberg, Social, Emotional, and Personality Development. New York: Wiley.

Turiel, E., M. Killen, and C.C. Helwig. 1987. Morality: Its Structure, Functions and Vagaries. In The Emergence of Morality in Young Children, eds. J. Kagan and S. Lamb, 155-244. Chicago: University of Chicago Press.

Turiel, E., C. Hildebrandt, and C. Wainryb. 1991. Judging Social Issues: Difficulties, Inconsistencies, and Consistencies. Monographs of the Society for Research in Child Development 56(2, Serial No. 224).

Winograd, T., ed. 1996. Bringing Design to Software. Reading, MA: Addison-Wesley.

Winograd, T., and F. Flores. 1987. Understanding Computers and Cognition: A New Foundation for Design. Reading, MA: Addison- Wesley. (First issued by Ablex, 1986.)

Zuboff, S. 1988. In the Age of the Smart Machine. New York: Basic Books.

end

```

This web service brought to you by Somewhere.Com, LLC.