Electronic voting systems should not focus only on ballot casting and recording. Instead, a user-centered perspective should be adopted for the design of a system to support information gathering and sharing, deliberation, decision making, and voting. Relevant social science literature on voting is used to develop requirements and design options.

A person's social context is an important aspect of her identity. Connections provides an interactive, web-based social map that allows users to learn more about people and their social networks within an organization. The social network information is drawn from a public domain and is thus out of the control of the person involved. We found in a user test that the social networks were reasonably accurate, however several people expressed discomfort at having their social networks inferred from public data.

In traditional human subjects research, researchers change names and other identifying information about subjects in published accounts. On the other hand, when published works are cited, researchers of course name the authors to give them credit for their work. Are the creators of material posted online vulnerable human subjects who need to be disguised for their protection or authors who deserve credit for their work? Can we include users' online pseudonyms in published accounts, or should those be treated like real names? Does it depend on where the names and pseudonyms appear online and how the data was collected? In this talk, I'll show how online, the distinctions become unclear between published and unpublished works, anonymous and identified people, and public and private spaces. Proposed practical guidelines for researchers studying computer-mediated communication will be presented.

Web sites intended for social interaction do a poor job of presenting the social context and reflecting the artifacts, patterns, and traces of social presence, activities, and organization. Consequently, communities lack mechanisms to present an outward face to visitors and visitors lack tools to enable them to acquire the social milieu of a community space. We present the design of two interactive community browsers consisting of social visualizations and dynamic query mechanisms. The visualization constructs an abstract, compact, and organically evolving visual portrait of the community based on self-description information, contributions, and/or activities of the members that is public in the community space. Each visualization presents a form of a public face of the community. The presentation of the designs and implementation the People and eTree browser is intended to illustrate possible elements of a community identity and to illustrate construction of a public face based on the collective contributions and actions of its community members.

With present and future wireless technologies, such as IEEE 802.11, BlueTooth, RF-Lite, and G3, handheld devices will frequently be in close, interactive communication. Many environments, including offices, meeting rooms, automobiles and classrooms already contain many computers and computerized appliances, and the smart homes of the future will have ubiquitous embedded computation. When the user enters one of these en-vironments carrying a handheld device, how will that device interact with the environment? We are exploring, as part of the Pebbles research project, the many ways that handheld devices such as PalmOS Organizers, PocketPC and Windows CE devices, and smart cell phones can serve as useful adjuncts to the "fixed" computers and com-puterized appliances in the user's vicinity. This brings up many interesting research questions, such as: How can the handheld device improve the user interfaces of everything else in the user's environment, rather than being just be another complex gadget that must be learned? What is the best way to provide a user interface that spans mul-tiple devices that are in use at the same time? How will users and systems decide which functions should be pre-sented and in what manner on what device? How can the user's handheld device be effectively used as a "Personal Universal Controller" to provide an easy-to-use and familiar interface to all of the complex appliances available to a user? How can communicating handheld devices enhance the effectiveness of meetings and classroom lectures? We present some preliminary observations on these issues, and discuss some of the systems that we have built to investigate them.

Georgia Tech is developing a new environment for distributed learning. The systems in the learning environment construct and maintain an understanding of a student and their interaction with other profiled objects. Though we will be able to provide many beneficial services with this information, privacy, access and persistence issues need to be considered carefully.

In this paper, we describe the identity policy decisions for a community network outside of Boston, Massachusetts. To promote trust and accountability, a member's online identity is their real-world identity; there is no anonymity. We conclude, based on analysis of the online interaction that this identity policy: 1) bridged and enriched online and face-to-face interactions, 2) promoted accountability in support of local commerce, and 3) fostered a social norm of polite conversation.

For all networked information systems' extraordinary successes at meeting communication and information exchange goals, they generally fail in equal measure at satisfying critical social and legal policy requirements such as privacy protection, reasonable intellectual property rights structures, and basic security needs. In sharp contrast to other venues of human interaction, the public policy problems we experience online are facilitated in significant part because of basic architectural design failures. User's experience of the Web today suffers from the doubly-frustrating problem that it both gives us a false sense of security and trust in many interactions, and at the same time lacks basic infrastructure to evolve more humane, socially-acceptable forms of interaction because of a fundamental lack of transparency. As these problems fall into the category of law & public policy, the general impulse is to look to the law to solve them. While law is certainly a necessary part of making the web a humane environment, it is not alone sufficient. In fact the real hope for improving the sociability of the Web and other public information systems lies with the technical architects of these systems. We must learn how to build "policy capable" infrastructure that provide users the ability to communicate about and control the policy context in which they operate along with greater transparency of policy context between system operators and users. Furthermore, we must expose this new policy layer of the infrastructure through user interfaces that provide users a real sense of control over their environment while demanding only a barely noticeable amount of their attention. Though law and social policy must continue to evolve to address the challenging questions raised by human life online, law can only achieve it goals with the support of new policy capable system architectures.

This paper reports on a study of factors that affect team identity in a virtual organization. 204 members of 18 different teams were studied in a large multinational organization where the majority of collaboration is conducted across geographic distance. A survey was administered and three months of team observations were conducted. The results showed that members of smaller teams judge their teams to have higher participation, better rapport, higher commitment, better knowledge of team goals, and higher awareness of their teammates. We also found that higher levels of team interdependency were associated with higher familiarity among the members and a higher level of collaboration readiness. Collaborative technology adoption also differed: smaller teams tended to adopt technologies that more directly supported collaboration, whereas larger teams tended to adopt technologies with greater reliance on coordination functionality. There was no relation of task type to team size. Putting these results together, we propose that the process of sensemaking to construct identity in a virtual organization is influenced by team size. In larger teams there is less clarity in the team boundaries, goals, membership, and less general awareness of other members.

Community network identity policy impacts user attitudes and social capital production processes. Probing these tradeoffs, we develop a design model, analyze existing interfaces, and construct a prototype allowing survey of user reaction to design elements. Test results frame value-chain tradeoff conceptualizations between virtual identity, usage motivation, and social capital production.

This paper discusses the tension between personalization and privacy in web-based applications. It outlines two avenues for reconciliation, and elaborates on one in which both users and the servers that host data about these users remain anonymous with respect to the personalized applications that utilize the data. A reference architecture for pseudonymous yet fully personalized interaction is presented which allows users to restrict data disclosure relative to their own roles and the type of application.

Since 1989 there have been over a hundred collaboratory projects, where scientists and engineers collaborate deeply on a project even though they are not collocated. Each collaboratory is a "one off," unaided by knowledge from past collaboratories, successes or failures. We at Michigan are now engaged in a meta-analysis of these collaboratory efforts, attempting to glean both the causal factors that account for success and failure and a compilation of best practices. We present our findings to date: 1) there are about seven different kinds of collaboratories, 2) success depends on the coupling of the work, common ground and the readiness of both the people and the technology, 3) in one kind of collaboratory, Community Data Systems, people have explored a variety of motivational schemes to keep people contributing high quality material.

A person's social context is an important aspect of her identity. However social information is often in a public domain and thus out of the control of the person involved. We found in a user test that several people expressed discomfort at having their social networks inferred from public data.

Although privacy is broadly recognized as a dominant concern for the development of novel interactive technologies, our ability to reason analytically about privacy in real settings is limited. A lack of conceptual interpretive frameworks makes it difficult to unpack interrelated privacy issues in settings where information technology is also present. Building on theory developed by social psychologist Irwin Altman, we outline a model of privacy as a dynamic, dialectic process. We discuss three tensions that govern interpersonal privacy management in everyday life, and use these to explore select technology case studies drawn from the research literature. These suggest new ways for thinking about privacy in socio-technical environments as a practical matter.

Human identities have always been contextual and fractured, but Web-based systems now create even more opportunities for identity construction. Should Web site designers then strive for more unified identities or should they support the user's "bricolage"? Along an example of a social interest community, the paper advocates the latter strategy based on a humanistic, constructivist perspective, favoring self-organization in identity construction as well as in community building.

Weblogs, or blogs for short, are frequently updated web sites, with relatively short time-stamped posts, most recent on top. They are highly cited with links to source material. Their clear ownership facilitates a high level of activity and a freedom of expression. Weblogs seem to be here to stay and possibly will sit next to e-mail and instant messaging as one of the three main paradigms through which people connect to each other on the Internet. HCI professionals should be cognizant of what is happening with respect to weblogs, and this paper discusses five areas in which weblogs should be of interest (privacy and online identity, community networks, social networks, research, and pedagogy).

[Version 2.1, as of 1.31.03] We analyze the co-construction of identity'of objects, collections, people and institutions'by collectors of early American glass. Focusing on the glass itself, we discuss its identity in terms of intrinsic and extrinsic features, and links between them. Finally, we discuss collectors' knowledge of glass, and some of the social and institutional mechanisms involved in establishing and vetting identity.

Individuals with cognitive disabilities are often unable to live independently due to inability to perform daily tasks. Computationally enhanced dynamic prompting systems can mitigate this. High levels of assistive technology abandonment are driven by poor user interfaces. MAPS provides an effective prompting system with an intuitive interface for configuration.

A number of research projects from both industry and academia are examining the use of intelligent systems to analyze and predict attributes such as location and availability among members of a workgroup. While often rooted in actual user data, these systems can often experience a mismatch in their representation of user behavior versus actual behavior. Insufficient knowledge in the early phases of deployment, dynamic schedules, and limits in the number of features considered are all potential causes for this disparity. When this occurs in shared groupware tools, the representation of users as perceived by the intelligent system may clash with the users' desired persona. Further, even if the system can accurately represent a user's behavior, will that representation inform a shared persona that the user wants? Cases may arise where the social environment of the workplace dictates that past behavior be ignored in favor of a more biased portrayal. In other cases, users may want to abandon intelligent assistance entirely to protect privacy or indicate some level of ambiguity. In this paper I will discuss recent work on augmenting shared electronic calendars using the Augur system at Georgia Tech's Everyday Computing Lab. I will then outline some of the current and future design decisions we are considering for Augur with respect to the aforementioned issues of intelligent systems and the shared personas they inform.

Each year children dependent on the Child Welfare System for protection get lost in the system, often as a result of overworked case managers handling too many cases. CaseWorks is a computerized tool designed to help case managers effectively organize and prioritize their caseload as conditions change in real time.

This paper describes a loss of trust in an international scientific research collaboration. A communications breakdown between research sites led to an expensive and embarrassing failure in the laboratory. Interviews with project participants suggest that cross-site collaboration may be more susceptible to losses of cognitive trust than same-site collaborations.

Awareness of the activities of one's co-workers is valuable for effective collaboration. The need for awareness is however frequently in conflict with privacy concerns of the people involved. This paper discusses various factors and principles that influence and inform a privacy-preserving design of awareness systems.

A person seeking someone else's attention is normally able to quickly assess how interruptible they are. This assessment allows for behavior we perceive as natural, socially appropriate, or simply polite. On the other hand, today's computer systems are almost entirely oblivious to the human world they operate in, and typically have no way to take into account the interruptibility of the user. This paper presents a Wizard of Oz study exploring whether, and how, robust sensor-based predictions of interruptibility might be constructed, which sensors might be most useful to such predictions, and how simple such sensors might be. The study simulates a range of possible sensors through human coding of audio and video recordings. Experience sampling is used to simultaneously collect randomly distributed self-reports of interruptibility. Based on these simulated sensors, we construct statistical models predicting human interruptibility and compare their predictions with the collected self-report data. The results of these models, although covering a demographically limited sample, are very promising, with the overall accuracy of several models reaching about 78%. Additionally, a model tuned to avoiding unwanted interruptions does so for 90% of its predictions, while retaining 75% overall accuracy.

Players of Massively-Multiplayer On-line Role Playing Games often blur the divide between virtual and real by buying and selling game items both within the game exchange mechanisms and outside of the game through on-line markets. This effectively attaches genuine economic value to software objects that are of use only within the virtual realm of the game. This paper examines the ecology of one such game, EverQuest, explicating the mechanisms of exchange and the start of a theory that explains this strange phenomenon.

Bandura's social cognitive construct 'perceived selfefficacy' has been used widely to understand individual behavior as a function of domain-specific beliefs about personal capacities. Collective efficacy is the extension of the self-efficacy construct to organizations and groups; it refers to beliefs about collective capacities in specific omains. Our research is investigating the use of collective efficacy in understanding attitudes and behaviors of members of proximal residential communities with respect to issues like attachment, engagement, and sociality, specifically as modulated by use of the Internet and community networks. This paper describes our analysis of the structure and external validity of the collective efficacy construct.

Most work carried out online is collaborative to some degree, involving activities that must be coordinated with other people. However, most user interfaces are resolutely single-user. We are exploring designs in which the single user experience is extended to incorporate the wider social frame within which work takes place. Based on the concepts of social networks and temporal rhythms, we present an infrastructure and tools for exploring the social workscape within which a user is embedded. This approach allows us to identify some key properties of social groups which can be incorporated into traditional interactive tools.