My postdoc aims to understand how online groups, specifically in the context of crowdsourcing, can produce more complex, creative, and cognitively-demanding work than has been seen to date. Achieved through experiments to understand the mechanisms and processes affecting output from online groups, in the context of specific applications such as limerick writing and content analysis.

Open collaboration communities tend to thrive when participation is plentiful. Wikipedia, an open, online encyclopedia, has been highly successful in eliciting participation from internet users due, at least in part, to the extremely low entry barriers to contribute. However, low entry barriers are a double-edged sword. They make it easy for both legitimate users and malicious users to contribute. To control malicious users and preserve the quality of the encyclopedia, the Wikipedia community has raised non-technical entry barriers which recent work has implicated in causing a decline in the editing population. This paper argues for the need of an intelligent entry filter that bars the entry of deviants, but supports the entry of compliant newcomers.

This paper briefly summarizes research, service and design projects that have investigated the experiences that homeless young people have with information systems and technologies. Two current studies, one which engages social computing more directly and another with potential implications for the design of social computing, are highlighted.

Past energy consumption-related studies show that household comparisons can have an important impact on how energy is used/saved. My research explores the benefits of energy-related feedback that enables community monitoring and sharing of individual members' average daily consumption and allows community members to engage with one another to share knowledge and information.

Accurate self-assessment skills are crucial for design students and professionals. This paper describes teaching materials developed for a Stanford University design course to enable student self-assessment. Self-assessment is accurate--69% of assessments agreed with blind staff-grades-- and improved through the term. We outline our plans to extend these techniques for larger scale online design classes. PS: LISTEN TO MY BOASTER. WE HAVE NEW DATA

We investigated the potential of public digital backchannels for building feelings of community among students in university courses. We designed, deployed and evaluated a public digital backchannel tool in a 15-week field study of two undergraduate classes. We analyzed the factors that influence students' willingness to adopt public digital backchannels in classrooms. We conclude with suggestions for improving the design and deployment of course related backchannels.

We demonstrated that effective leadership in social computing system does not always come from the small set of people occupying formal leadership roles. In contrast, leadership is shared among members with different levels of engagement, from peripheral members to core members.

We are engaged in a line of research studying how people use file syncing and sharing services (e.g., Dropbox, Google Docs) to better understand how early adopters conceptualize their interactions with the cloud. After an earlier survey and interview study, new services have been launched (e.g., Google Drive, iCloud) which seem to have even more complex conceptual models. We are planning a second round of study to see how users make sense of these new services. We hope our studies will help identify users' needs in cloud services to support their activities at this crucial time in defining the cloud user experience.

The Web is a source of health information for a majority of U.S. adults. However, searchers have difficulty finding and understanding useful resources. This research utilizes dual-'process theory from social psychology to characterize and enhance exploratory health search behaviors.

We present a general-purpose implementation of a target aware pointing technique, functional across an entire desktop. Specifically, we implement Grossman and Balakrishnan's Bubble Cursor. Our implementation locates interface targets using pixel-level analysis and social annotation. We highlight how the behavior of interaction techniques can be socially defined from users.

What do people think they should be able to do with other peoples' digital stuff? We surveyed 1252 Internet-savvy respondents to discover how they feel about storing, reusing, removing, and archiving user-contributed online media. Six scenario-based surveys cover different media types, including photos, tweets, reviews, videos, podcasts, and educational recordings.

In online peer production communities such as open source software development, people often encounter unknown others who they must work with in the absence of any previous interaction or formal organizational relationship. These settings are increasingly supported by social networking systems. When connected with the work, these systems can provide detailed information about individuals' activities and behaviors. My research focuses on understanding how individuals form impressions of others in work environments supported by social media, and how these impressions influence collaborative outcomes.

For accurate non-expert visual analysis to occur, model biases (data or visual mapping error), cognitive biases (anchoring and adjustment), and social biases (social influence) must be overcome. The visual bootstrap overcomes bias by generating alternative representations of source data and presenting these to either a single or many analysts.

The purpose of our study is to explore people's motivation for seeking Internet anonymity and the strategies they take to achieve anonymity. We also examine how government policies, cultural differences, and people's computer expertise influence the use of anonymity on the Internet.

We present the Social Overlays system that enables end-users to fix common user interface problems on a web page and share their modifications to others. By having individuals improve a website's usability as a part of their individual use of that site, a community can collectively improve the websites they use.

In order to better understand how people co-adapt to an intelligent system in the home, we studied the lived experience of an advanced thermostat, the Nest. Our study findings assert balancing user control and automation is critical to the desired system performance and the success of interventions to promote sustainable behaviors.

My dissertation work explores the design space of 'hybrid' social search, which transitions between active and passive states. I am using participatory design to develop a social search tool for Online Health Seeking which will explore the interaction between social search and socio-professional roles.

Social computing technologies have the potential to act as platforms for personal health'to allow people to encourage each other to share previously solitary activities and to live better lives. In my dissertation work, I study social systems that encourage middle school students to get more physical activity throughout the day. Social computing technologies may play a vital role in promoting and rewarding these activities, and it's imperative for us to understand if, how and why they work.

With advances in mobile sensors and device connectivity there is an ever-increasing amount of our lives that we can capture instantly. However, with increased ease of use and ubiquity, questions arise regarding how data collected from mobile sensors can be beneficial beyond the individual. We propose the area of Interpersonal Informatics (IPI) as a framework within which automatically collected data can be shared and utilized to foster a better understanding of one's self and the social networks in which one lives.

Sociotechnical systems can be utilized to aid individuals in their efforts to change unwanted behavior. These systems can be more effective when they leverage how individuals make decisions, especially decisions that are influenced by their friends and peers.

It is an acknowledged reality today that an increasing number of individuals are using the internet to meet their health information needs. The primary focus of my dissertation is on whether there is a difference between the health information sought by online seekers versus offline seekers and how this affects the doctor-patient relationship.

I argue that the HCI community needs to reconsider how its works reach the level of practice and the innovation of social computing, particularly in relation to how today's organization develop the competencies necessary for "architecting user experience".

This document describes my proposed dissertation, which explores information systems for supporting identity work knowledge workers in a professional community, asking the following: How can an information system be designed to support development of a professional identity across the boundary between firm-based Communities of Practice and professional communities?

Low socioeconomic status (SES) populations are prone to higher risks of acquiring chronic diseases including cardiovascular disease, diabetes, and hypertension. Among the major causes are individuals' everyday health-related decisions that affect their long term health, in particular their dietary and physical activity habits. This research proposes to design a sociotechnical intervention to improve the healthiness of snacks in the target low SES population. The intervention -- a mobile phone snack application -- will be evaluated in a six week feasibility field deployment with 10 families. This research will advance the field of human-computer interaction by providing concrete guidelines on designing mobile phone based sociotechnical intervention for low SES populations to improve their snacking.

Online health communities (OHCs) are a growing source of public medical knowledge; they facilitate several health- related tasks including searching for and acquiring new medical information, and seeking emotional support. However, many underlying attributes of OHCs, such as discussion dynamics, forum content structure, and leadership hierarchy, are not readily apparent during browsing. Knowledge of these attributes might comprise useful decision making tools for a spectrum of community participants: from leaders, who maintain OHC forums, to potential new members, who might "window shop" forums before picking a best fit. Prior work demonstrates that visualization is an effective technique for discovering and exploring underlying online community attributes. In this ?boaster?, we survey some of our work on using visualization techniques to interpret community dynamics and underlying content in MedHelp OHCs.

The death of a user challenges many of the assumptions we hold for social network sites, social media, and digital identity architecture more generally. By studying death in the context of social media, my work aims to helps us better understand how people interact with and experience digital identity systems, demonstrate limitations of current architecture, and in turn, better enable social systems to support the entirety of our lives -- including when those lives come to an end.

My dissertation work will examine how materiality practices figure in relational dynamics common to organizations involved in new product development. Early ethnographic data collection at two design firms suggest that different practitioner specialists as well as sales personnel use external artifacts and representations to manage sensitive relationships with clients and with each other.

Micro-coordination is ?the tight coupling of behaviors to possibility in the moment.? It means looking at the detailed, fine-grained interactions in situ. Micro-coordination research examines how co-located people carry out digitally augmented coordination, and aims at understanding (1) how computer mediated interactions are similar and/or different from the traditional ones, (2) what constitutes the differences in these interaction models, (3) whether and how the design of digital artifacts impacts people?s coordinative behaviors. Most importantly we study the effect of design on how we experience others and ourselves as agents, moral actors and interactors in these technology mediated settings. Upon delivering thick descriptions on critical cases which question and challenge non-discussed, often-taken-for-granted assumptions on face-to-face interactions and coordination, we tie our observations to the creation of higher level constructs which, in turn, can affect subsequent design choices.