Our mission in the Social Computing Group is to research and develop software that contributes to compelling and effective social interactions, with a focus on user-centered design processes and rapid prototyping combined with rigorous social science.

We believe in getting our projects in the hands of users. Many of the projects listed below are available to the public.

• Hotmap: The use of Virtual Earth
  
  

  Hotmap shows where people have looked at when using Virtual Earth, the engine that powers Live Search Maps. Hotmap is a mash-up of data over the original Virtual Earth tiles. Hotmap generates new image tiles based on data stored in a database, and superposes them over Virtual Earth, using the AddTileLayer method.

  
  
• Netscan: Social network and user activity analysis of Usenet
  
  

  The Netscan System provides detailed reports on the activity of Usenet newsgroups, the authors who participate in them, and the conversation threads that emerge from their activity. Using the Netscan tool users can get reports about any newsgroup for any day, week, month, quarter, or year, since September 1999.

  
  
• SNARF: the Social Network and Relationship Finder
  
  

  SNARF was built around the notion that social network information that is already available to the computer system can be usefully reflected to the user. For example, a message from a manager might be seen differently than a message from a stranger. SNARF applies this idea to email triage, handling the flow of messages when time is short and mail is long. The SNARF UI is designed to provide a quick overview of unread mail, organized by its importance. The UI shows a series of different panes with unread mail in them and each pane shows a list of message authors. Clicking on a name shows all messages involving that person.

  
  
• Slam: Mobile group-based communication and media sharing
  
  

  Slam is a mobile device-based application that enables lightweight, group-centric real-time communication, location awareness and photo-sharing. The core concept behind Slam is a “Slam”, a group of people with whom you can exchange messages and photos. When you send a message in Slam, it is automatically sent to everyone in the group to which you sent it.

  
  
• Wallop:
  
  

  In Wallop, you can share photos, blog, and interact with your friends. Wallop is a research project that explores how people share media and build conversations in the context of social networks. We are currently conducting a small, real world trial of Wallop with small friendship groups. Therefore, membership in Wallop http://mywallop.com is limited to study participants until the trial is over.

  Images: Default view of social network, blog & profile Relationship map between 2 people showing common people, media, & comments

• Sapphire:
  
  We model the user experience after the way people (vs computers) think, feel, organize and remember. This simple but fundamental change in perspective has caused us to rethink the way the system is experienced and designed, from the user experience to the lowest levels of the system.
  The current desktop assumes you organize files by manually sorting into folders. As the amount of information increases, finding items becomes more frustrating. Our goal is to have automatic, dynamic grouping by association replace folders as the primary means of organizing. The prototypes we have developed automatically determine importance (what you care about), similarity (relationships between objects), and the current context of the user. All objects (people, communication, documents, web pages, etc.) and metadata are stored in a database and queries determining associations can be calculated in real-time.
  The Sapphire project includes the backend, data & queries, and user interface prototypes, and is a cross group effort.
  
  
  • Sapphire architecture diagram: Below are the components of the Sapphire architecture:
    
    Data collectors: Real-time monitors, email sync, etc.
    Database: Unification of all data types
    Queries and filters: Associations between objects
    User interface: Supports various visualizations and notifications, and itself is a source of data.
    
  
  
• Stacks: Family Photos
  
  

  The stacks project studies how photos can be automatically grouped by time, person, etc. Users can quickly scan through stacks of photos by mousing over an item to see a preview.

  
  
• Sapphire Toolbar: Similarity, Importance, People, Time (history) and Topic
  
  

  The Sapphire toolbar lets users quickly access important items & items similar to the current active object. It also lets users view by time, topic and personal. This working prototype is built in C# and uses the objects and associations stored in a database.

  
  
• Visual Summaries: Sapphire Data Visualization and Summarization
  
  

  The visual summary displays links between objects using the Sapphire data. In order to get an overview of all the important objects in the database, we summarize and group objects.

  
  
• Smart Previews: Dynamic previews for emails and documents
  
  

  When objects of different types are shown together in the user interface, we need consistent ways to preview/inspect items. Smart previews automatically shows salient items within mail threads, documents, etc.

  
  
• Conversation Clusters: Grouping email conversations based on text indexing.
  
  

  Using hierarchical clustering methods, we automatically group email conversations (threads) based on text indexing. Preliminary studies show that the clusters produced match the way users think about grouping their email.

  
  
• Beyond Blackcomb: Storyboards of future computing experience
  
  

  These storyboards explore using simple forms to spatially display items on the desktop. Important people, conversations, documents and web pages are easy to find, and related items are appear based on the context of the user.

• Personal Map: Modeling contacts, communication groups, and social network
  
  

  The goal of the Personal Map is to help users organize their email contacts in a meaningful way, based on their email behavior, without users having to provide any additional information. The Personal Map models the users social network (who they care about and their informal groups) based on communication behavior such who they email the most and who they email together. The Personal Map provides several visualizations of the underlying user model, each with its advantages and disadvantages.

  
  
• Personal Map Outlook Integration
  
  

  We integrated the Personal Map model of the users social network into Outlook in order to assess whether people will find it useful in the context of their communication stream. Lists of important people and groups are placed to the right of the new mail message window, and can be used to populate the TO: and CC: fields. We ran a two week user study to explore how people will use the new mail message people list.

  
  
• Microsoft Research Connections: Individuals, groups, projects and topics in Microsoft Research
  
  

  Microsoft Research Connections provides an online social map that allows users to develop an awareness of the informal, dynamic groups and projects in Microsoft Research, and navigate for information using the connections between people and groups. Microsoft Research Connections uses public information (from the MSRinfo web page and active directory) to model the relationships between people and group.

• MS Connect and Point to Point
  
  

  Any knowledge or resource transfer across groups in an organization depends on peoples awareness of whos doing what, which is a challenge given the dynamic, informal nature of many groups and projects. The goal of MS Connect is to help people figure out who they should talk to learn more about a particular person or project by showing connections between people throughout the company. MS Connect uses active directory information to show both formal relationships between people and informal, dynamic relationships between people. MS Connect has a Point to Point feature, which allows users to see how they are connected to any other person or group.

• PhotoStory: Preserving emotional content when storytelling with photos and video
  
  

  People love taking and sharing photos--over 200 million photos are taken per day around the world. Recent advances in technology have combined to make the production, archiving, and storage of digital photos possible, yet most software fails to address the principle reason that people share photos in the first place: to tell personal stories. The PhotoStory application is designed to address that need. A key goal of Photostory is to allow novice end-users to quickly and easily share photo-based stories with people in a way that is simpler and more emotionally compelling than a static slide show with text.

• FastForward: Storytelling and extending/compressing time
  
  

  Home videos are notorious for being long and boring. Although often perceived as an editing problem, most professional video editors cannot make compelling movies from most home videos due to the lack of proper video coverage. People often leave out important parts (you don't see me preparing for my trip, I just show up at the Eiffel Tower) or spend too much time on something boring (endless footage of the kids). The Fast Forward project enables the user to share a story captured by a home video, by extending time (frozen/looping video) and compressing time (fast forwarding) in a way that doesn't interfere with the emotional content of the story.

• NearMe: Instant messaging and status on small devices
  
  

  The NearMe service is a location based buddy list service for GSM phones. NearMe integrates a buddy list with cell based proximity information from the cell phone. The service notifies a user when his buddies are near by using cell location information, the contact list on the cell phone, SMS messaging. Using NearMe it is very easy for a person to contact "friends" that are in their proximity. This information is accessible from web-based services in addition to the cell phone. The user can determine who sees his location information by adjusting the privacy settings.

• Family Fridge: Shared family page using "fridge door" metaphor
  
  

  Families use their refrigerator door and phone area in the kitchen to post calendars, photos, grocery lists, magnets, etc. The shared family machine is a key place to view shared family information. Using the fridge door as a metaphor, we designed a family portal/login page that the family has "up and running" by default on a shared machine in a public place (kitchen, entry) in the home. The family page works together with the personal page. The prototype was prepared for the MARs/OE6 team.

• Bridge: Location and awareness of groups and effects on learning
  
  

  Bridge is a collaboration between the MIT Architecture Schools iCampus project, the Social Computing Group and the Systems and Networking Group. Using technology developed by researcher Victor Bahl, we can determine the user's physical location on the wireless network. University students are mobile, social, and have dynamic work and social groups. The Bridge project studies the increased awareness of members in a study group (location, presence, etc.) and the effects on learning.

• Shared Browsing study: Designing for remote shared computing experience
  
  

  Browsing the Internet is an activity people commonly do together from different physical locations. We compared different shared browser designs to determine what UI elements provided the most productive and the most enjoyable experience. We looked specifically at how the UI could be used to increase "social presence" or the feeling that users were participating together with their partners.

• Visual Decision Maker: Shoulder-to-shoulder movie selector
  
  

  The Visual Decision Maker project is a collaboration between the Social Computing Group and the Next Media group. We designed the user interface to promote an enjoyable shared experience for movie selection. Users use remote controls in groups of up to four people. A stream of images is continuously adjusted based on users responses and then movie recommendations are displayed using a collaborative filtering backend.

• Shoulder-to-shoulder study: Shared puzzle solving
  
  

  People commonly sit together in front of a computer and play or work together. Computers are not inherently designed to support this interaction. We created a simple puzzle game, and compared different UI elements, observed interaction styles, and watched how people share input devices.

• Online Lab:
  
  

  We use our Online Lab to user test prototypes and conduct research studies. By partnering with product groups we recruit participants, we then run controlled, online experiments and collect and analyze user data. The online lab provides informed consent and provides a secure environment for testing new ideas. Through the online lab we study individual and group behaviors online. The online lab consists of three primary components: a) the lab structure itself, comprised of pages that carry welcoming information, informed consent, registration and authentication, and links to the studies, b) the studies themselves, and c) the backend used for collecting and analyzing user data.

• KidTalk: Improving online behavior for children with Aspergers/Autism
  
  

  KidTalk is a collaboration between the University of Washington Autism Center, Dr. Felice Orlich and the Social Computing Group. Currently there are limited resources for treating individuals with high functioning autism/Aspergers. A goal of this project is to improve access to tools and services. Children who suffer from Aspergers face social isolation and improving social skills are a primary intervention target. Current treatment for Aspergers includes supervised text chat and online interaction. The KidTalk study uses scripted interactions to teach children currently undergoing group therapy at the Autism Center.

• HutchWorld: Social support for cancer patients and their caregivers
  
  

  HutchWorld is an online multi-user community environment developed in collaboration with the Fred Hutchinson Cancer Research Center (Hutch). HutchWorld provides access to social support for caregivers, families, and patients of the Hutch. The use of PC's and HutchWorld for social support was studied in a trial with Hutch patients/caregivers at a Hutch patient housing facilities.

• Peer-based Reputation and Bad Behavior: Effects of peer based reputation systems on online groups
  
  

  Achieving a critical mass of communication, and reducing bad behavior online are 2 key issues of managing groups. We ran a study of 200 people in our Online Lab http://onlinelab.org/, and found our simple peer-based, anonymous reputation system reduced bad behavior and increased good behavior.

• Peer-based Reputation: Friend-of-a-friend recommendation system
  
  

  Users who are new to an online environment often have a difficult time deciding who they should interact with. A bad first experience can scare new users away forever. This project examined different types of reputation information to determine what information about other users are most important when people decide who to interact with. We found that recommendations from friends (a "friend-of-a-friend") and similar interests were the most important criteria for selecting a chat or gaming partner. In user tests, we also found that people understand the concept of social networks (how others are related to them through mutual acquaintances) and use such information if given the opportunity.

• Allegiances and Fellowships in Asheron's Call: Studying group interaction and social features in online gaming
  
  

  Online gaming attracts huge numbers of dedicated users. We studied effects of online groups (Allegiances and Fellowships) in the multi-user game, Asheron's Call to see how group membership effected enjoyment, retention of users, and bad behavior.

• Trust Online : Using social dilemma testing to study effects on trust
  
  

  Social Dilemma Testing studies how different modes of communication and different aspects of the user interface affect trust and cooperation between users. To study the interactions, a new type of user testing adapted from quantitative sociological techniques is used to examine interactions between users.

• Online Community Analysis: What makes people come back to online communities?
  
  

  What makes people actively participate in an online community? What makes a compelling group? What are popular types of online communities? To better understand how people interact in online communities we partnered with MSN Communities and analyzed user data from 20,000 randomly sampled communities.

• Microsoft V-Chat:
  
  

  Microsoft V-Chat let people chat from within a 2D or 3D environment using graphical representations of themselves, known as avatars. V-Chat was originally released on the Microsoft Network (MSN) v1.0 in December of 1995-March 2001. V-Chat users select from a variety of existing avatars, or create and publish their own. Sounds, animation, and visual imagery create mood and context for these social environments.

• Microsoft Chat (Comic Chat):
  
  

  Microsoft Chat (Comic Chat) was released on Internet Explorer v3.0 in 1996-March 2001, and was developed by Microsoft Researcher David Kurlander together with the Social Computing Group and comic artist Jim Woodring. In Microsoft Chat, your online conversations are the beginning of an interactive comic strip that unfolds in real time. Comic style balloons display your conversation, and gestures generated by conversation semantics give your character a variety of emotions and movements.

• Smart Talk Tools: Creating distributed services on cellular phones and other devices: (example: NearMe/location based buddy list)
  
  

  People carry cellular phones with them in social settings. To predict how the PC and the cell phone may integrate in the future, we looked at power phone users (teens) and phone technology (Europe, Japan) and developed the SmartTalk tools for prototyping applications on the phone. By placing an interpreter on the phone, we send and receive active messages (for example, a command, application, status, or plan text) from the phone to phone or PC. Current prototypes include NearMe, location based buddy list.

• Raven: Tools for creating multi-user, peer-to-peer, distributed applications online.
  
  

  Raven simplifies the building of multi-user applications letting two client-side webpages talk to one another (peer-to-peer) without a server involved.

• Virtual World Platform
  
  

  The Virtual Worlds Platform facilitated the creation of multi-user, distributed, applications on the Internet.

Primary Contact: Lili Cheng
Affiliate Members
Photo Not Available Boyle, Claudia	Drucker, Steven	Photo Not Available Horvitz, Eric	Jacobs, Andy
Jones, David	Regan, Tim	Vronay, Dave

• Fisher, D. (2007) Hotmap: Looking at Geographic Attention. Proceedings of Information Visualization (IEEE InfoVis) 2007, Sacramento, CA
• Counts, S. (2007) Group-Based Mobile Messaging in Support of the Social Side of Leisure. CSCW Journal.
• Riegelsberger, J., Counts, S., Farnham, S., & Philips, B. (2007). Personality Matters: Incorporating Detailed User Attributes and Preferences into the Matchmaking Process. HICCS '07.
• Grudin, J., Tallarico, S, and Counts, S. (2005). As Technophobia Disappears: Implications for Design. Group 2005.
• Farnham, S., Turski, A., Portnoy, W., & Davis, J. (2002). Microsoft Research Connections: Exploring Who Knows Whom through Social Networks. Internal paper.
• Farnham, S., & Turski, A. (2002) Social Network Project: Applications for Online Communication and Information Navigation. Internal paper.
• Farnham, S. (2002). Visualizing Discourse Architectures with Automatically Generated Person-Centric Social Networks Paper presented at CHI Workshop 2002: Discource Architectures.
• Kelly, S., Sung, C., & Farnham S. (2002). Designing for Improved Social Responsibility and Content in On-Line Communities. In Proceedings of CHI 2002, Minneapolis, April 2002.
• Jensen, C., Davis, J., & Farnham, S. (2002). Finding Others Online: Reputation Systems for Social Online Spaces. In Proceedings of CHI 2002, Minneapolis, April 2002.
• Farnham, S. (2002). Predicting Active Participation in MSN Communities. Its All in the Conversation. Microsoft Technical Report MSR-TR-2002-36.
• Davis, J., Farnham, S., Jensen, C. (2002). Decreasing Online Bad Behavior. In Extended Abstracts of CHI 2002, Minneapolis, April 2002.
• Davis, J. P. (2002). The experience of bad behavior in online social spaces: A survey of online users. Internal paper.
• Vronay, D., Farnham, S., Davis, J. (2001). PhotoStory: Preserving Emotion in Digital Photo Sharing. Internal paper.
• Farnham, S., Cheng, L., Stone, L., Zaner-Godsey, M., Hibbeln, C, Syrjala, K., Clark, A., & Abrams, J. (2002). HutchWorld: Clinical Study of Computer-Mediated Social Support for Cancer Patients and Their Caregivers. In Proceedings of CHI 2002, Minneapolis, April 2002.
• Swinth, K., Farnham, S., & Davis, J. (2002). Sharing Personal Information in Online Community Member Profiles. Internal paper.
• Cheng, L., Kimberly, G., Orlich F (2002). KidTalk: Online Therapy for Asperger's Syndrome. Internal paper.
• Turski, A., Marcjan C., Cheng, L., (2001). SmartTalk: Prototyping Social Applications for Cellular Phones. Internal paper.
• Farnham, S. D., Chesley, H. McGhee, D., & Kawal, R. (2000). Structured On-line Interactions: Improving the Decision-making of Small Discussion Groups. In Proceedings of CSCW 2000, Philadelphia, December.
• Davis, J., & Farnham, S. Sharing Whats Enjoyed and Enjoying Whats Shared: Designing for Sociability in Shared Browsing. Internal paper.
• Regan, T., Lofstrom, M. & Davis, J. P. (2002). The Effects of enhanced user feedback on shoulder-to-shoulder computing. Internal paper.
• Davis, J. P., Zaner, M., Farnham, S., Marcjan, C., & McCarthy, B. P. (2002). Wireless brainstorming: Overcoming status effects in small group decisions. Paper submitted to journal Computers in Human Interaction.
• Farnham, S., Zaner, Melora., Cheng, L. Supporting Sociability in a Shared Browser In Proceedings of Interact Conference, Tokyo, Japan, July 2001.
• Cheng, L., Stone, L., Farnham, S., Clark, A. M., & Zaner-Godsey, M. Hutchworld: Lessons Learned. A Collaborative Project: Fred Hutchsinson Cancer Research Center & Microsoft Research. In Proceedings of Virtual Worlds Conference 2000, Paris, France, June 2000.
• Farnham, S. D., Chesley, H. McGhee, D., & Kawal, R. Structured On-line Interactions: Improving the Decision-making of Small Discussion Groups. In Proceedings of CSCW 2000, Philadelphia, December 2000.
• Jensen, C., Farnham, S., Drucker, S., & Kollock, P. The Effect of Communication Modality on Cooperation in Online Environments. In Proceedings of CHI 2000, The Hague, Netherlands March 2000.
• Smith, M., Farnham, S., & Drucker S. The Social Life of Small Graphical Chat Spaces. In Proceedings of CHI 2000, The Hague, Netherlands March 2000.
• Chesley, H., Kawal, R., Landau, J., Cheng, L., Farnham, S., Seban, S. Scripting Business Social Interactions In Proceedings of SSGRR, July 2000
• Vronay, D., Farnham, S. Redesigning the Contact List. 2000
• Vronay, D., Smith, M., Drucker, S. Chat as a Streaming Media Data Type. 1999
• White, S, Gupta, A., Grudin, J., Chesley, H., Kimberly, G., Sanocki, E. Evolving Use of a System for Education at a Distance. 1999
• Vellon, M., Marple, K., Mitchell, D., Drucker, S. The Architecture of a Distributed Virtual Worlds System. 1998
• Kollock, P., Smith, M., University of California, Los Angeles. What Do People Do in Virtual Worlds? An Anlalysis of V-Chat Log File Data 1998
• Kollock, P., Smith, M., University of California, Los Angeles. Managing the Virtual Commons: Cooperation and Conflict in Computer Communities 1996
• Kollock, P., University of California, Los Angeles. Design Principles for Online Communities 1996
• Mitchell, D. From MUDs To Virtual Worlds 1995

• Integrating Diverse Research and Development Approaches to the Construction of Social Cyberspaces Workshop to be hosted at the CHI 2001 Conference, March 31-April 5, Seattle.
• Social Computing Group: An Update This presentation gives an overview of SCG's research and projects.
• HutchWorld Presentation This presentation gives an overview of the HutchWorld Project.
• Physical Cities/Digital Cities by Linda Stone
• Project Summaries This document gives a one-sheet summary for each of SCG's projects.

• Microsoft Research hosts an annual Social Computing Symposium, bringing together researchers and practioners innovating in the area of social computing. The goal for the symposium is to mind-share and stimulate collaboration and creative thinking amongst attendees.
  
  Click through for lists of speakers, agenda, and videos of talks.
• 2007
• 2006
• 2005
• 2004

To contact Social Computing Group please email us at soco@microsoft.com.

Last updated: December, 2007