|
|
*U S WEST Advanced Technologies, 4001 Discovery Drive, Boulder CO 80303 USA
**U S WEST Communications, 1801 California Street, Denver CO 80202 USA
We outline several promising areas for improvements in research and practice in the field of Human Computer Interaction (HCI). These topics show the richness and potential value of HCI work motivated by a combination of a desire to improve practice and research, and a desire to meet human needs in a responsible manner.
Future, HCI research and practice, disability access, information access, information retrieval, agents, World Wide Web, UIMSs, architectures, information poverty, communication poverty, social issues, social impact
It has become common to hear concerns regarding the alleged weakness of research in the field of Human Computer Interaction (HCI) [36, 38]. In this paper, we seek to remedy this alleged weakness by advancing a research agenda based on human needs and social responsibility. This research agenda is offered in the same spirit as Lewis [19]. We will attempt to show that research based on these motivations can lead to rich, significant opportunities for progress in both theory and practice.
In this paper, we sketch three broad HCI research areas that offer significant research challenges, and that may lead to valuable social outcomes. These areas are (1) disability ac-cess and people with disabling conditions, (2) information and communication poverty, and (3) productivity and efficiency as principal values. We close with a quick survey of other potentially productive topics, including reflexive study and the positionality of the practitioner.
Each topic, of course, could be described in greater detail in a full-length paper. Our purpose here is not to develop the ideas in detail, but to show that human needs and social responsibility may be used to develop a research agenda of richness and depth, whose topics and subtopics qualify as research projects of the first rank, worthy of solution for scientific as well as social reasons.
The research topics that we are proposing are not intended primarily as solutions to immediate business applications within the current budget cycle. Rather, we believe that these questions that can increase the field's knowledge of fundamental principles in the diverse knowledges that converge to form the research and practice area called "HCI." In keeping with the research principle of the "unity of theory and practice," we have sought practical problems with theoretical implications.
We should note that -- while we are intensely interested in the outcomes to some of these questions -- we personally do not have all of the needed knowledges[footnote 1] or disciplines to carry out much of this research. Further, the topics we des-cribe are not necessarily of interest to our employer. Our goal is to outline areas of potential growth for the fields of HCI, rather than to develop our own research agenda.
The fields that together constitute HCI have maintained a long tradition of concern for access to computer systems by persons with disabilities. However, in practice, much of the field's progress in this area has been slow. More often than not, the field's attention to this area has been an after-thought: Design and access has been based primarily on able-bodied[footnote 2] populations rather than users who are disabled.
In part, this slowness and inattention to the disabled com-munities resulted from the historical development of com-puting systems. Initially, scientific research and govern-mental needs guided the development of most computer sys-tems, and the initial, highly trained users were a very able-bodied segment of the population, who understood the sys-tem intimately. At this time, our fields gave little thought to the potential for augmenting the capabilities of people with disabilities through the capabilities of the computer.
This scene has changed dramatically in the last 20 years. As the field has matured, the number of HCI specialists and their influence on computer technology have grown. Dur-ing this time, many countries enacted laws to ensure that people with disabilities have equal access to education, work, and functional independence. It has also become ob-vious that technology offers one promising way to achieve these goals. For example, in the United States, the Tele-communication Act of 1996 now requires that telecommun-ications systems be accessible to people with disabilities.
Now, there is almost universal recognition that computer technology offers a rich and varied set of tools for accommodating people with functional limitations. There is recognition that computers can provide many people who have disabilities with a more level playing field on which to compete in today's educational, business and personal arenas. Although the potential is there, however, it has not been fully developed; and there are a number of areas in which additional research is needed to create systems which are fully accessible to all user groups. In this section, we outline several areas where we think there may be opportunities for improvement.
We believe that the field of HCI can make contributions to develop products that meet the real needs of people with diverse disabilities (e.g., [31]). People with disabilities are growing in number [26]. For example, in the United States alone, there are at least 49 million people who have a documented physical, sensory, or cognitive disability which encumbers them in performing one or more major life activities of work, self-care, communicating, or learning. In addition to those with recognized disabilities, there are many older people who develop disabling conditions, but do not consider themselves "disabled."
There are many disabled and elderly people who work full time. There are also many who are underemployed, and instead received maintenance care, support from public funds, or low-paying jobs. But surveys have continually shown that these populations do not want to be dependent on others to support them, or to lose their autonomy. They wish to support themselves, maintain independent living, and contribute to society like everyone else.
Almost all of the older population and those with disabling conditions can be assisted to develop and maintain an independent and self-sustaining life through computer technology. Computer technology offers both support and augmentation possibilities to people with disabling conditions as ways to overcome many of the limitations imposed on them by the environment in which they live and work and the tasks they must perform.
Making these changes will mean that we must work in a language that is consistent with the languages of the people being served [footnote 3], something that has not traditionally been done (e.g., [11]). It requires us to develop anthropological and sociological methods for working across differences in language and abilities. Increasing the field's competence in intercultural communications and intercultural practices will aid us in working with many other groups of users, across many other dimensions of difference (e.g., [21, 24, 35]).
One area of particular concern for people with disabilities is the World Wide Web (e.g., [2]). At present, the Web is an overwhelmingly visual interface. People with certain forms of visual or motoric disabilities maybe be unable to read a graphic, or may be unable to operate a pointing device. Typically, these users prefer text-only browsers, which allow non-visual means of reading pages (or the ability to increase the display font to a needed size), and non-pointing means for selecting items. These text-only browsers are capable of ignoring a Web page's graphical content, and of presenting the textual material from a Web page in a usable alternate modality (e.g., Braille or voice) for the user.[footnote 4]
However, most Web pages appear to include text-only items as an afterthought. More ominously, there is an increasing trend for the semantics of a link to be indicated within the image that forms the clickable link item [2]. The HCI community has available to it the challenge of how to make the Web navigable by people with limited vision or limited ability to control a cursor. As a community, we might envision a set of arguably "separate but equal" interfaces (full graphics for the mainstream, and "full" text the "others"). A more challenging solution would be to implement a full-access interface within a single, unified page-design style.
In some ways, this problem is similar to the explorations over the past several years regarding transformation of windowing systems for people with visual disabilities. The Web "version" of the problem becomes potentially more important because of the growth of Web access, the diversity of people and organizations interested in the Web, and the claims of future economic importance of the Web.
We are concerned that most design strategies have called for a first design for some mainstream, reference, or "normal" group of users. The design is then altered until it appears to meet the needs of one or more non-mainstream groups. In this default approach, the initial and fundamental design is tailored for the mainstream users, and much of that tailoring enters into the core of the design. When we attempt to adapt the design for other users, we are sometimes in the position of trying to "de-mainstream" the design -- that is, first we have to remove features and capabilities that assume a mainstream constituency, and only then can we begin to adapt the design for a different constituency.
We would like to suggest two alternative approaches. What would a system or service look like if it were designed from its inception such that ...
These approaches might help us avoid the unintentional incorporation of the values and needs of the mainstream users, which may exclude other, non-mainstream users (e.g., [9]). The second of these two approaches might also help to avoid other mainstreaming issues, such as those related to gender, sexual orientation, or national culture. However, would such design efforts even be possible (see "Positionality of the Practitioner," below)? What methods would be needed to avoid unintentional mainstreaming of an allegedly "neutral" design?
The accessibility issues involved in the Web and window systems are subsets of a more challenging research project. The more general question is this: How can we design systems that permit systematic, cost-effective approaches to accommodating all users?
We might begin with the UI architectures of the 1970s and 1980s (e.g. the Seeheim model [29] or Nielsen's virtual protocol stack [28]). Nielsen's model is especially appropriate because it was based on human cognitive architectures. If we extend these models, and incorporate appropriate aspects of activity theory [24] and communications theory, it might be possible to specify the options needed to make the system usable to the broadest user base. This would allow different strategies for providing accessibility to groups with different needs (device-level options, syntax-level options, language-level options, etc.). Accommodating users with different requirements would then be a matter of interfacing the various alternative technologies to the application.
This approach deliberately blurs the artificial distinction between an application and its interface. We believe that the application and the interface are interdependent at many levels where the "back-end logic" and the presentation layer interact.[footnote 5] As HCI practitioners, implementing this type of design for accommodation would require us to rethink the "separation" architectures (separation of presentation from application), that have been the predominant models of the last decade, and to develop new models which incorporate system-level (as opposed to application-level) innovations that can support a broad range of users.
We hypothesize two broad classes of models:
Implementing both of these models will require us as HCI practitioners to understand fully 1) the components and requirements of the work; 2) the capabilities and limitations associated with many disabling conditions; 3) the types of support which a computer system can provide to make it possible for a person with a disabling condition to accomplish a task; and 4) where in the system this support can be provided most effectively.
This type of architecture would reframe questions of agents and intelligent support for human activities. Conceptualizing the architecture between the human and the "back-end logic" of the application as a series of cooperating cognitive levels might allow us to conceive of AI, expert, or agent solutions not in the old monolithic (or even hegemonic) terms, but rather as piece-part tools that would be better fitted to human abilities -- better fitted from the multiple perspectives of human efficiency, human dignity, work quality, work-life quality, and workplace politics.
Modern client-server architectures provide further research challenges for such a model. We can easily imagine architectures that place certain aspects or "layers" of such a model in the client, and complementary aspects or layers in the server. What would be the "right" division of such a model in a client-server architecture? Some approaches might allow certain types of disability accommodation to operate relatively effortlessly at the level of the client-server interface. However, we noted above that we are unlikely to find a single architectural point of intervention that will meet the distinct needs of the diverse communities that desire or require accommodations. Should we then be thinking not of a dyadic client-server model, but instead of a multi-party client-intermediaries-server model?
We also briefly propose the following questions that help to bridge between our focus on accessibility, and a focus on computer-supported cooperative work: How would such an architecture support human-to-human cooperation with systems as intermediaries? Suppose we began with an interface architecture elaborated from Nielsen's cognitively-oriented virtual protocol stack model [28]. Would such a cognitively structured user interface architecture allow us to develop communication-support entities that could operate via semi-automated cooperation at one or more software levels, to support the human-to-human cooperation of the people they serve? What models of human-to-support-entity assignments and relationships would work in such an architecture -- e.g., each human with her/his own set of support entities, in cooperation with similarly supported other human? groups of human with relatively unitary, group-assigned support entities? mixed models?
The deployment of large-scale information networks such as the Global Information Infrastructure (GII) [14, 17, 32] has raised many questions about the social implications and risks of such networks. Increasingly, peoples' abilities to make choices about their lives are being determined by their ability to communicate, access and process information by electronic means. Increasingly, networked groups are working together and relying on community-based informa-tion and groupware technologies to accomplish their goals.
Information and communication poverty is a condition resulting from an inability to communicate, access and process information critical to one's life and livelihood, due either to the absence of personal or technological means, or the unavailability of human agents to perform these tasks. Barriers to using supportive information technologies are primarily economic and educational: Often, access requires ownership or computer availability, technical knowledge, and the financial resources to pay for access fees. Attitudes toward technology, and the culture of exposure to and acceptance of it, may also be significant factors.
These human and social issues lead to many new challenges and rich research topics at both a global and specific level. Potential HCI research topics of a global nature include: How can HCI workers design and develop new technologies and systems to reduce or eliminate information and communication poverty? How can we incorporate concerns about information and communication poverty into mean-ingful heuristics that can be used to guide the design, devel-opment, and evaluation of computer systems? How can we go beyond marketing definitions of "acceptance" to under-stand how persons, groups, constituencies, and cultures define concepts such as needed information, access, and usefulness of information? How can we measure persons' and groups' perceived and effective ability to access infor-mation and to communicate? In the case of distributed and non-organized groups, such as participants in the Web, how can we assess the effectiveness of their information access and communication (see below and [13])? (See also "Choosing our Criteria," below.)
Potential HCI research topics of a more specific nature include: How do design decisions for the GII impact the medium and the message? How can we use sophisticated information filtering techniques and intelligent agents to provide better support for user tasks? What are the challenges for networked learning communities [40]? How are special user populations' needs (e.g., students, people of different ages, and people with disabilities) enfranchised [40]? How well we can support privacy and fair information practices on the Internet (e.g., [7])? How can we measure the quality of participation in decision-making; and in the event that our systems are used in decision-making, how can we measure the quality of, for example, democratic processes that are supported (or not) by our systems?
We now shift our focus from the domain of systems to the domain of the relationship of human systems and computer systems -- conventionally analyzed as "impact models" (i.e., what is the impact of a computer innovation on human work?). Traditionally, our field has valued concep-tions of productivity and efficiency above any other criteria or concerns, in describing people's work with computers (e.g., [1]). This Tayloristic approach has been important to near-term economic goals, and has played a significant role in some of our own work. However, Taylorism has arguably had devastating impacts on the consideration of other aspects of human work, such as the quality of the work done and the quality of the human's worklife while performing the work [10]. Taylorism is particularly inappropriate when we are considering non-work human activities (see "Alternatives ... Characterizing Other Human Activities," below).
Are productivity and efficiency the only aspects of human work that we want to measure or improve? Are these attributes the only outcomes with economic implications? We hope to open a lengthy discussion by suggesting several additional attributes of human work that may also have economic importance:
We do not imagine that our list is exhaustive. We look forward to its amplification and correction by others.
More broadly, our field may be concerned with the quality of methods or practices that we bring to the design process, and especially to working with users. In this regard, we may wish to characterize our own practices in terms other than productivity, such as human-to-human communications clarity, communications accessibility, process safety (e.g., for low-status workers to contribute their ideas to a design or an evaluation), or quality of democratic processes (e.g., in the context of participatory design and of information and communication poverty).
Even more broadly, our field has participated in the design and evaluation of systems whose focus is not human work as such but rather other human activities, such as education (inside or outside of the work context), entertainment, and even spirituality. We are beginning to see forms of support for human social activities in areas of interactive television (e.g., competitive games, collaborative viewing). What are the relevant individual and social goals and values that such systems should support? How can these goals and values be translated into measurable attributes?
Pioneering work was reported at CHI 96 regarding the interface characteristic of sensuality [15]; one of the lessons of this work was how new it was, and how unprepared our field is to analyze any non-productivity concept in a systematic fashion.[footnote 6] Support of a non-work activity will require at least a superficial theory of that activity (consider attributes in the area of education -- speed of learning, retention, integration of learned materials with other knowledge, engagement, motivation, pursuit of future learning opportunities, etc.,[footnote 7] as well as the simultaneous values implied in concepts such as "edutainment"). We may have to consider entirely different theoretical frameworks, such as activity theory [24] or grounded theory [35] to be able to develop such domain-related theories.
We believe that it is fair to say that our field has little idea of how to think about these non-productivity attributes of products and practices, and still less idea of how to quantify those attributes or to make them economically meaning-ful.[footnote 8] Yet we will need to valorize these attributes if we want them to be weighted as strongly (in the decisions of our management, and even in our own practices) as the more traditional Tayloristic attributes. Developing mea-surement strategies of non-traditional product and practice attributes will require a deeper exploration of other fields, possibly including art theory, semiotics, cultural critique, and perhaps feminism, subaltern studies, sociology, liter-ature, history, and political science as well (see also "Reflexive Study," below). We will be challenged to develop (or discover, through inter-field exchanges) meaningful and non-trivial theories of broad classes of human activities. Deeper theoretical and practical knowledge of a broad range of human activities may be coupled productively (that productivity word, again!) with enhancements in user interface architectures, to provide improved solutions to a broader range of human and social needs. Accommodating computer systems to this broader conception of human activities will bring new depth to the concept of usability (see [8] for a related argument).
The brief structure of a conference paper does not allow us to detail other research and practice opportunities related to human needs and social responsibility. We will briefly list several other areas that appear fruitful to us.
We know something about how a system might fit into a user's work life. But we know much less about how a system might fit into life processes outside of work -- especially the culturally diverse lives of people who are unlike the analysts, designers, and developers involved in most hardware and software projects [21].[footnote 9] Many of us are involved in individual and corporate commitments to valuing diversity among people. These commitments encourage us to improve our field's practices with regard to working with users unlike ourselves.
Some researchers and practitioners in our field have been applying ethnographic and other anthropological methods to understanding users' world views (e.g., [5, 25]). However, there are also reports of great difficulties in making the outcomes of ethnographic studies useful in software development [16]. We may need to find a set of "discount ethnographic practices," analogous to Nielsen's "discount usability engineering" [27] to make ethnographic work tractable within the software lifecycle.
From a different perspective, we may wish to take up the question of the visibility of different communities or constituencies, as reflected in the systems we develop and in the practices that we use to develop them. Some commun-ities or constituencies have expressed a need for greater visi-bility to the mainstream world -- e.g., some of the tradi-tionally marginalized minorities, such as people of color, lesbians, gay men, and people with disabilities. However, other communities have expressed a need to shelter them-selves from too much scrutiny, or from modes of commun-ication that perpetuate their subordinate status [34]. For ex-ample, some Native American nations would like to have a Web presence, but would like to control the extent to which a Web connection can lead to acculturative mainstreaming within their communities, causing further erosion of their minority culture and language [personal communication from tribal representatives]. In formal terms, Clement and Wagner began an analysis of diverse groups' nuanced needs for both aggregation (visibility, presence, and communica-tion) and disaggregation (protected spaces, partial or asym-metric communication) [7]. Within the domain of cultural criticism, there is a strongly expressed need that minority or embattled communities be able to present their own view of themselves into shared cultural and communication spaces, rather than seeing someone else's interpretation of them-selves (for a summary from an HCI perspective, see [21]).
These community needs offer many research, practice, and product opportunities. From a research perspective, there are needs to understand each community's own needs (which may or may not be similar to those of other, similar communities). From both practice and research perspectives, there are needs to develop methods that are ethical and effective in working across dimensions of difference and power. From a product or service perspective, there are opportunities to support these embattled communities with configurable services that allow them to negotiate the extents, categories, and controls on their communications with the mainstream.
A major question for contemporary anthropologists has been phrased as the "positionality of the researcher." In the terms of our field, this phrase indicates a recognition that observation of users' work -- and evaluation of a system within the users' work context -- is not conducted by an abstract, disembodied, objective scientist. Observations are made from somewhere (i.e., from the professional and per-sonal perspectives of the observer), and evaluations presup-pose a set of criteria and a worldview that has previously validated the importance of those criteria (see "Choosing Our Criteria," above). The concern with positionality involves reflection on the role, demographics, and perspectives of the software professional, and the ways that each software professional's position inevitably introduces distortions into her or his allegedly "objective" work. According to the prevailing analysis, the distortions are neither good nor bad. What is important is the specific types and directions of distortion.
Within the area of HCI, we are aware of one workshop that has addressed this issue [30]. Much work remains to be done, to understand how our practices are based in our own situations, and how our understandings of other people's work is distorted by our unexamined assumption about their work and our own.
This relationship of anthropology and software engineering is a good example of two fields that would benefit by better working relations. One of our common themes through much of this paper has been the need to combine diverse knowledges to improve research and practice in HCI.
Our field's approach to this problem is, we believe, one of conflict and denial. In one sense, HCI is like any other scientific and engineering discipline: It intensively studies the problems that fall within its boundaries, and it excludes research questions that fall outside of its boundaries [18]. Two principal ways of excluding problems are to treat other disciplines as irrelevant (e.g., political science in most HCI treatments) or as unquestioned facts (e.g., biological science in most HCI treatments).
In another sense, HCI is better prepared than most fields to merge its knowledges with the knowledges of other disciplines. We are used to studying the work and the work practices of other people. We are used to standing back from what we study, and to reflecting on what we've found. We are used to drawing conclusions and making recommendations for improvements in what we have studied. But we have not analyzed our own work with the same enthusiasm that we have brought to the study and the "improvement" of the work of other people.
For much of our history, our field has had the luxury of studying work that is of lower-status than our own. Why do we say that this has been a "luxury?" We have been able to use Tayloristic analytic and evaluative methods that most of us would not be comfortable applying to our own work -- especially if someone with a different background were to apply our own methods "on" us.[footnote 10] However, if our field is to achieve some of the benefits that we have proposed in this paper, then we will need to understand our own work practices in relation to those of our potential collaborators from other disciplines. We will need to open a number of new topics for HCI: how we do our own work; how our peers (of equal or greater status) in other fields do their work; how each party defines, develops, and critiques its own sets of knowledges; and how the work practices and disciplines of the two or more fields can be fruitfully combined. In effect, we are calling for a reflexive use of our own best practices in the study of work practices and communications among ourselves and our collaborators. The gains from this effort will include not only improvements in our own self-awareness and our own practice, but multiplied power in methods of inquiry, analysis, development, and evaluation resulting from a combination of the strengths of multiple robust disciplines.
We have outlined several areas in which a research agenda may be based in human needs and social responsibility. We have attempted to show that these motivations can lead to the identification of rich, exciting problems with important and economically significant solutions in areas of both theory and practice. We believe that further discussion will reveal other important research and practice topics that may also be based in the convergence of research and practice in areas of human needs and social responsibility. We look forward to the extension, enhancement, and transformation of this research strategy by our colleagues.
We thank Monica Marics for close readings and helpful suggestions.
[1] Bailey, R.W. (1993). Performance versus preference. In Proceedings of the Human Factors and Ergonomics Society Annual Meeting. Seattle: Human Factors and Ergonomics Society.
[2] Bergman, E., Johnson, E., Edwards, A., Kaplan, D., Lowney, G., and Raman, T.V. (1996). Universal design: Everyone has special needs. In CHI 96 Con-ference Companion. Vancouver BC: ACM, 153-154.
[3] Bikson, T., Blair, J.H., Barry, R.E., Grantham, C.E., and Winograd, T. (1988). Communication, coordin-ation, and group performance. In CSCW'88: Pro-ceedings of the Conference on Computer-Supported Cooperative Work. Portland OR: ACM, 189-190.
[4] Bjerknes, G., Ehn, P., and Kyng, M. (1987). Computers and democracy: A Scandinavian challenge. Brookfield VT: Gower.
[5] Blomberg, J., Giacomi, J., Mosher, A., and Swenton-Wall, P. (1993). Ethnographic field methods and their relation to design. In D. Schuler and A. Namioka (eds.), Participatory design: Principles and practices. Hillsdale NJ: Erlbaum.
[6] Caine, B., Grosz, E.A., and Lepervanche, M. (1988). Crossing boundaries: Feminism and the critique of knowledges. Sydney: Allen and Unwin.
[7] Clement, A., and Wagner, I. (1995). Fragmented ex-change: Disarticulation and the need for regionalized communication spaces. In H. Marmolin, Y. Sundblad, and K. Schmidt (eds.), ECSCW'95: Proceedings of the fourth European conference on computer-supported cooperative work. Dordrecht: Kluwer, 33-49.
[8] Floyd, C. (1987). Outline of a paradigm change in software engineering. In Bjerknes, G., Ehn, P., and Kyng, M. (Eds.) (1987). Computers and democracy: A Scandinavian challenge. Brookfield, VT: Gower.
[9] Friedman, B., and Nissenbaum, H. (1993). Dis-cerning bias in computer systems. In INTERCHI'93 Adjunct Proceedings. Amsterdam: ACM, 141-142.
[10] Garson, B. (1988). The electronic sweatshop: How computers are transforming the office of the future into the factory of the past. Harmondsworth, Middlesex, UK: Penguin.
[11] Gibet, S. (1994). Synthesis of sign language gestures. In CHI'94 Conference Companion. Boston: ACM, 311-312.
[12] Greenbaum, J., and Kyng, M. (1991). Design at work: Cooperative design of computer systems. Hillsdale NJ: Erlbaum.
[13] Greenbaum, J., Snelling, L., Jolly, C., and Orr, J. (1994). The limits of PD? Contingent jobs and work reorganization. In PDC'94: Proceedings of the Participatory Design Conference. Chapel Hill NC: Computer Professionals for Social Responsibility, 173-174.
[14] Haywood, T. (1995). Info-rich/Info-poor: Access and exchange in the global information society. London: Bowker-Saur.
[15] Hofmeester, G.H., Kemp, J.A.M., Blankendaal, A.C.M. (1996). Sensuality in product design: A structured approach. In CHI 96 Conference Proceedings. Vancouver BC: ACM, 428-435.
[16] Hughes, J., Randall, D., and Shapiro, D. (1992). Faltering from ethnography to design. In Proceedings of CSCW'92 -- Sharing Perspectives. Toronto ONT: ACM, 115-122.
[17] Kranich, N. (1995). The Internet, access and democracy: Ensuring public places on the information superhighway. New York: Open Magazine.
[18] Laudan, L. (1977). Progress and its problems: Towards a theory of scientific growth. Berkeley: University of California Press.
[19] Lewis, C. (1990). A research agenda for the nineties in human-computer interaction. Human-Computer Interaction 5(2-3), 125-143.
[20] Messer-Davidow, E., Shumway, D.R., and Sylvan, D.J. (1993). Knowledges: Historical and critical studies of disciplinarity. Charlottesville VA: University Press of Virginia.
[21] Muller, M.J. (1995). Ethnocritical heuristics for HCI work with users and other stakeholders. In Proceedings of Computers in Context. Århus Denmark: Århus University, 10-19.
[22] Muller, M.J., Carr, R., Ashworth, C.A., Diekmann, B., Wharton, C., Eickstaedt, C., and Clonts, J. (1995). Telephone operators as knowledge workers: Consultants who meet customer needs. In Proceedings of CHI'95. Denver CO: ACM, 130-137.
[23] Muller, M.J., Ruston, L., and Cebulka, K.D. (1990). Making separation architectures serve the user. In Proceedings of IEEE International Conference on Systems Man and Cybernetics. Los Angeles CA: IEEE Press, 820-822.
[24] Nardi, B.A. (1996) (Ed.). Context and consciousness: Activity theory and human computer interaction. Cambridge MA: MIT Press.
[25] Nardi, B. (1995). Some reflections on scenarios. In J. Carroll (ed.), Scenario-based design: Envisioning work and technology in system development. New York: Wiley.
[26] National Council on Disability (1996). Achieving independence: The challenge for the 21st century. Washington DC: National Council on Disability.
[27] Nielsen, J. (1995). Scenarios in discount usability engineering. In J. Carroll (Ed.), Scenario-based design: Envisioning work and technology in system development. New York: Wiley.
[28] Nielsen, J. (1986). A virtual protocol model for computer-human interaction. International Journal of Man-Machine Studies 24, 301-312.
[29] Pfaff, G.E., (1985). User interface management systems. Berlin: Springer-Verlag.
[30] Plowman, L., Rogers, Y., and Harper, R. (1995). The 'professional stranger': the role of the fieldworker in workplace studies for CSCW . Workshop at ECSCW'95. Stockholm, September 1995.
[31] Roy, D.M., Panayi, M., Erenshteyn, R.F., and Fawcus, R. (1994). Gestural human-machine interaction for people with severe speech and motor impairment due to cerebral palsy. In CHI'94 Conference Companion. Boston: ACM, 313-314.
[32] Schiller, H.I. (1996). Information inequality. New York: Routledge.
[33] Schuler, D., and Namioka, A. (1993) (Eds.). Participatory design: Principles and practices. Hillsdale NJ: Erlbaum.
[34] Sherry, J. (1995). Cooperation and power. In H. Marmolin, Y. Sundblad, and K. Schmidt (eds.), ECSCW'95: Proceedings of the fourth European conference on computer-supported cooperative work. Dordrecht: Kluwer, 67-82.
[35] Strauss, A., and Corbin, J. (1990). Basics of qualitative research: Grounded theory procedures and techniques. Newbury Park CA: Sage.
[36] Strong, G.W., with Gasen, J.B., Hewett, T., Hix, D., Morris, J., Muller, M.J., and Novick, D.G. (1994). New directions in human-computer interaction education, research, and practice. Washington DC: National Science Foundation.
[37] Suchman, L. (1993). Do categories have politics? The language/action perspective reconsidered. In Proceedings of the Third European Conference on Computer-Supported Cooperative Work. Milan, September 1993.
[38] Sutcliffe, A., Carroll, J., Young, R., and Long, J. (1991). HCI theory on trial. In CHI'91 Conference Proceedings. New Orleans LA: ACM, 399-400.
[39] Vanka, S., and Klein, D. (1995). ColorTool: An information tool for cross cultural design. In Proceedings of the Human Factors and Ergonomics Society 39th Annual Meeting. Santa Monica CA: Human Factors and Ergonomics Society, 341-345.
[40] Wharton, C., Eberts, R., Angiolillo, J., Brigham, F., Givens, B., Laux, L., Neal, D., and Owens, M. (1995). Human factors issues and the internet. In Proceedings of the Human Factors and Ergonomics Society 39th Annual Meeting. Santa Monica CA: Human Factors and Ergonomics Society, 213-217.
* Muller: +1 303 541 6564, mullerm@acm.org. Wharton: +1 303 541 6292, cwharton@advtech.uswest.com or wharton.chi@xerox.com. McIver: +1 303 541 6285, mciver@advtech.uswest.com. Laux: +1 303 965 8156, llaux@notes.mnet.uswest.com.
[Footnote 1]In keeping with recent work in epistemology, feminism, and the history of science [6,20], we emphasize the plural and dialogic nature of the disciplines of HCI through plural usages of words like "knowledges," "disciplines," and "fields."
[Footnote 2]The disability communities would use the phrase "temporarily able-bodied."
[Footnote 3]One striking example was the proposed use of the teletype abbreviation of "TT" as the name for a new telecommunications device for the deaf (TDD). The well-intentioned software professionals who made this suggestion were unaware that "TT" was already a TDD abbreviation -- for "toilet."
[Footnote 4]Temporarily able-bodied users with slow network connections may also choose to work without images.
[Footnote 5]In earlier work, one of us showed that the strict assignment of application functionality to a "back-end" server, with no application semantics or functionality in a "front-end" client, was unworkable in three diverse interface-technology domains [23]. In the case of graphical user interfaces, icons and other graphical objects require some local (client) information in order to have appropriately constrained interface behaviors. In the case of a voice interface, the state-based information in a back-end database was needed at the architectural level of the speech recognition user interface technology for adequate disambiguation of users' utterances. In the case of simple form-filling interfaces, local (client) execution of field-level validation rules provided better and more tractable information for users than did post-transaction feedback from the database server.
[Footnote 6]Another lesson of this work is that gender stereotyping runs deeply through our professional culture. The product was a pager that was intended for women consumers. We wish that Hofmeester, et al., had analyzed a product charac-teristic that was more related to a women's life experiences (e.g., safety), rather than a characteristic that has been attributed to women (and the products marketed to women) by men during the long history of men essentializing and diminutizing women as colleagues and as consumers.
[Footnote 7]Many of these attributes might have relevance for assessing the quality of participation in participatory practices in the software lifecycle (e.g., participatory design and its extensions -- see [4, 12, 34]).
[Footnote 8]Consider, for example, speech act theory as applied to machine-mediated human-to-human communication. Speech act theory appeared to provide a tractable and quantifiable approach to description and evaluation of human communications. Yet the speech acts approach proved notoriously inadequate in some practice settings [3], and has been analyzed as imposing an unstated set of restrictive political perspectives upon its human users [38].
[Footnote 9]Some progress has been made in understanding cross-cultural color issues [40].
[Footnote 10]Because of the deliberately democratic nature of their work, practitioners and researchers in the area of participatory design have tended to be less likely to use methods that they would not wish applied to their own work (e.g., [4, 12, 34]).
|
|