In online social situations when people interact outside of any social context, they experience decreased accountability and trust (Resnick et al. 2000). Social contexts provide meaningful information that facilitates users’ ability to find and trust each other. Characteristics of a person’s social network provide clues about that person’s social status, interests, and attitudes. What people in a social context have to say about a person, i.e., reputation, is especially meaningful. A person might be more likely to invite a person to a party if she knew they had a mutual friend, or if she could see he actively interacted with a number of other people she knew something about.

Many recent applications and web sites have reintroduced social contexts to online interactions through reputation systems (e.g. E-bay.com) visualizations of transaction histories (e.g. NetScan, Smith & Fiore, 2001), and explicit articulations of social networks (e.g., Facebook.com), or Friend of a Friend (FOAF) systems. Many FOAF web sites have emerged in the past few years allowing users to capitalize on their social connections to find jobs (e.g., Ryze.com, LinkedIn.com), date (e.g., Friendster.com), or post ads or event announcements (e.g., Tribe.net). Online dating is the most popular use of such sites (Donath & Boyd, 2004). Friendster.com was created to serve implicitly as a dating service, and had over 5 million registered users in 2004.

Social networking tools generally allow people to provide explicit lists of friends, so that users may browse around each other’s social spaces and add comments and recommendations, building each others’ reputations. Orkut.com does a good job of making a game of improving reputation by inspiring its members to attach reputation indicators to each person’s identity: trusty smiley faces, cool cubes, and attractiveness hearts posted by friends allows people provide feedback about each other. For some blogs have a similar social network effect, with people linking to their friends in their blog entries or blog rolls. However these links to friends only appear in about 51% of blogs (Herring et al. 2004).

While social networking sites provide valuable information about people’s social contexts and tools for navigating around each other’s social spaces, the explicit articulation of friendships does not necessarily map onto people’s real life social relationships (Donath & Boyd, 2004), which tend to be more dynamic and comprised of multiple social contexts (e.g., work vs. home life). These applications have a low barrier to the classification of a person as friend, leading to long friend lists in which it becomes difficult to meaningfully infer who is really friends with whom. Orkut.com addresses this problem by having people rate the quality of friendship relationships when accepting friendship requests, however anecdotally people report feeling uncomfortable explicitly distinguishing between friendly acquaintances and best friend. Although online dating sites face a lot of the same bad behavior problems as other social software systems, they cannot take advantage of reputation systems to reduce bad behavior in the same way. A good online dating system does not have persistent users over time, and people do not want a lot of recommendations for a date. (Imagine: “50 women recommend Jim as a good date.”)

While FOAF systems have enjoyed a measure of popularity of late, they appear to be destined for niche status as either dating sites or games. As dating sites, they have the highly valuable property of providing natural verification systems: you can find interesting people by exploring your friend’s social networks, and then you can verify the apparent desirability of your discoveries by communicating directly with intermediate – and more well-known – persons along your line of discovery.

While all FOAF systems have a game like quality (people tend to classify increasingly vague associates as friends in order to get high numbers) more recently some FOAF-like sites have embraced the gaming dimension and turned it into a core attribute of the experience. In FunHI, members participate in a trendy celebrity culture based on connectedness, gifting, and reputation. Social connections are modeled explicitly and are given different roles: you can be a fan of someone, or, if you invited them into the system, you can own their contract. Users are encouraged to initiate relationships with new people, and virtual gift objects (which cost real money) can be offered in order to entice others to accept reciprocal friendship status. Players are rated relentlessly in multiple dimensions: most fans, most received gifts, most friends, etc. In a way, this functions as an effective satire of the more self-conscious but still game-like serious social network sites: rather than awkward friend request emails based on tenuous associations, you simply pay people to like you. FunHI is reportedly a profitable and stable business.

The somewhat debased and cynical (yet fun) nature of the FunHi experience illustrates a problem with explicit social networks in general: people are often uncomfortable publicly declaring relationship status out of concern that the asserted social attributes will not be reciprocated. While you might have a set of friends with whom your status is clear, things get tricky around the edges. Is that coworker a friend or associate? Will he be offended if relegated to associate status, or will he consider you desperate and lonely if you designate him as friend? In real life, we enjoy a nuanced spectrum of relationship status, the evolution of which is enhanced specifically because it is not necessarily to explicitly rate its status.

Wallop (a Flash based network that we developed as a research project) was developed to explore some of these issues: how can we exploit interaction behavior to implicitly indicate significant relationships. We expect that an implicit network, based on interaction behavior, will better represent the dynamic nature of social relationships, better indicate who’s important to whom, and reduce some of the social awkwardness found in articulating relationships in explicit social networks. Because they require minimal maintenance on the part of the user, they should also stay up to date more easily. We further expect that people will learn about each other from how they actually interact in their social networks than from what they say about how they interact.

Thus through Wallop we explored ways to use transaction history, co-occurrence information, and communication patterns, to make inferences about how people are connected to each other. Traditionally, conversations play a key role in the implicit social information people learn about each other in online situations (Cherney, 1999). The presence of each person in online environments is largely comprised of his or her messages, and the presence of each group is largely comprised of what is communicated amongst its members. Although one can gain a sense of connection to others by lurking alone (Nonnecke & Preece, 2000), a person remains invisible until he participates by committing himself to words. Similarly, most of the social information that leads to the development of community, such as identity, relationships, behavioral norms, and so forth, is transferred through text. Thus we foster communication between people in Wallop as much as possible, and have found that conversational patterns are central to determining the implicit relational ties between people.

Another issue we have encountered with Wallop is how do we maintain the advantages of the implicit network while affording people the controls they desire over access to their own content. The advantages of the implicit network are the minimal or lightweight network management, and that the network evolves dynamically as people’s interactions evolve. However when people chose to render their network ’protected’, in order to reduce the likelihood of harassment, spamming, or cross-contextual embarrassment, they lose that ability to have people fluctuate in and out of their network depending on their interactions. If everyone in a network protects their network, the system will no longer have the implicit component (you can not get into mine without explicit permission, and I can not get into yours without explicit permission).

Perhaps one of the more powerful applications of social networks is that of providing an infrastructure for knowledge management technologies (Alavi & Leidner, 1999). (e.g., TacitNetworks.com). Collaboration and knowledge sharing is a central challenge in any organization that relies heavily on the development of its intellectual capital (Stewart, 1997). However, any collaboration or knowledge transfer across work groups depends on people’s awareness of who’s doing what in the organization, which becomes exponentially more difficult for larger companies. The challenge to people in developing an awareness of activities in other groups is made more formidable by the dynamic, informal nature of many organizational project teams.

In corporations with rapid structural changes, people increasingly rely on their interpersonal connections to collaborate with others and exchange knowledge across corporate boundaries (Nardi, Whittaker, & Schwarz, 2002). Within organizations individuals are often a prime source of knowledge (Nardi, Whittaker, & Schwartz, 2002; Ackerman, 1998), and knowledge management has as much to do with locating who knows what as with managing the knowledge itself (Ehrlick & Cash, 1994). As much as people care about the information, they care about developing long-term collaborative relationships with individuals throughout the corporation.

People commonly use the Internet and email to seek out people who are knowledge experts or project contacts (Alavi & Leidner, 1999). In addition several knowledge management systems have explored how to support people’s tendency to seek out knowledge through people. Referral Web, for example, uses co-occurrence of names in documents on the web to develop connections between people and then makes referrals through a chain of such connections (NcDonald & Ackerman, 2000). The Expertise Recommender (McDonald & Ackerman, 2000) allows users to employ a social network filter which sorts recommendations based on social distance in a network. These applications are increasingly exploring how tools based on social networks may allow people to infer information about social structures, such as the centrality or importance of a person in a network and the evolution of dynamic work groups (Farnham et al., 2004; Farnham et al., 2003).

A central problem to developing digitized corporate social networks is that in order to address the privacy concerns of the corporate members, relational data must be provided voluntarily. However explicit, voluntarily provided information is difficult to keep up to date, and the network is likely to be sparse as people either neglect to provide relational data or chose not to. We have implemented several prototypes (MSR Connections, Point to Point) in the Social Computing Group to address this problem by bootstrapping the corporate social network from communication groups already public to the organization. Our studies of these prototypes have found that these communication groups are reasonably accurate (some users have declared them frightening), however they are only as accurate as the communication groups are up to date and actively used – Microsoft for example has a six month time lag between the time a group is last used and when it is deleted from the system. Furthermore many organizations do not allow its members to create communication groups so they would not reflect the more dynamic nature of project teams.

In general, the social network is still a construct that is new to the average user, and understanding complex patterns of behavior within social networks is still the domain of the experts. Most people much more easily understand the nature of groups, even though social networks often provide a more accurate model of modern social structures (Wellman & Haythornthwaite, 2002). Developing a visual language for the end user that allows him or her to assess dynamic patterns in social networks is a hard problem. How for example, might one best gain an overall sense of activity within Wallop, and more specifically how might end users track the development and activities of emerging groups and cliques? Currently most systems address the issue of providing overviews of network activity through simple counts of overall or personal network activity (Blood, 2002), and through randomly selected highlights of people in the network.

Although social networking applications have proliferated on the Internet, there has arisen a general impression that it is not clear what they are good for in and of themselves. They serve to contextualize social interactions, providing a structure upon which to hang identity and reputation, improving trust in online social interactions. However we expect that social networks in general will recede into a secondary feature of applications that more explicitly address specific user social and information seeking goals, much as the e-bay reputation systems is a secondary feature of an online auction system.

In sum, in the future, we expect that FOAF networks will become more integrated with other applications (For example, SCG has implemented prototypes which use your social network to prioritize incoming communications) and will integrate more implicit social relatedness.

Social Web 2.o course notes:

• Social Networks
  • Defined
    • Set of pair-wise relationships vs. individual relationships or groups
    • Social network analysis
      • examine density, boundedness, size, heterogeneity
      • Study flows of influence, information, social support
    • Network vs. group social structure
• Social Networks
  • Measuring Ties
    • Ties: friendship, role sharing, common event, common property, co-occurrence
    • Tie strength: affection, frequency of interaction, trust, frequency of co-occurrence
  • Network structures
    • singletons, stars, middle region, giant
    • Tend to be reciprocal (84% in yahoo, 70% in flickr)
• Social Networks Today
  • Modern transformation in how things get done
    • communication, collaboration, information flow
  • Social networks most appropriate model
    • Lateral, not hierarchical command chains
    • Ad hoc teams, not static groups
  • Developing social capital – the value of social networks
  • NetWORK
    • It’s not what you know, it’s who you know
    • Intensional networks
      • Building
      • Maintaining
• Social Networks Online
  • Facebook
• Why Articulate Social Networks Online
  • Social networks contextualizes information and behavior
    • Increase accountability by putting people back in social context
    • Informational context, understanding author
    • Exploit transitive trust (implicit and explicit referrals)
    • Increase relevance via similarity/affiliation
    • Provide access to weak ties
    • Define access/sharing/subscription (filter out as much as increase access)
  • Navigation tool
    • Browse network vs. directed search
    • Info transfer horizontal, across hierarchical boundaries
• Social Networks Online cont’d
  • Processing outside user awareness:
    • Alternative “similarity” measures aside from explicit friend/family lists:
      • Cross-links
      • Communication patterns
      • Co-occurrence in groups
      • Co-occurrence of semantic tags
    • Prioritize match-making by distance in network
      • “We recommend you check out Jon’s story…”
      • Closer is better, # of overlap is better, etc.
    • Network/cluster analysis, use for
      • prioritizing search results
      • developing semantic hierarchies
      • Extraction of groups (dense, tightly bounded networks)
    • Isolate connectors
      • Identify people connected across network clusters, able to transfer info/trust
• Online Social Networking Issues
  • Often binary (friend/no friend) with friend list glut
  • Assume one network per person, no subnets causing role conflict
  • Social capital of “connector” lost
  • Systems do not expect social networks to be dynamic, become out of date
  • “Now what” – ok so I built my network, now what?
  • No cross-property integration, building network over and over
  • Developing critical mass
  • Visualizations often outside comprehension of average user

Social Networks and Knowledge Management

• Encourage behavior makes content prioritizable
• Rich profiles, linking, bookmarking, rating
• As much as possible social navigation based on implicit behavior
• Prioritize search results by most linked/watched profiles
• Distance in network matters, use to prioritize results
• Social Filtering
• Friend of friend path matters for scoping results, not so much after that
• Information scoped by similar others, near others, or demographic variables
• Index profiles, search by profile info
• Standing queries, saved searches.
• Promoting Healthy Users
  • Health =

    Function ((logins, content creation, commenting) * recency * longevity)

• • Characteristics of a Healthy User
  • Many active contacts in network
  • Daily posts with pictures and music
  • Multiple comments from contacts on each post
  • Rich customization of profile and blog
  • Visits lots of other people’s pages
  • Long discussion threads

Recommended Readings

Books and Papers
Boyd, D. Friendster and Publicly Articulated Social Networking. In Ext. Abstracts CHI 2004, ACM Press (2004).

Lenhart, A., and Madden, M. Social Networking Websites and Teens: An Overview. (2007). Pew Internet and American Life Project. http://www.pewinternet.org/pdfs/PIP_SNS_Data_Memo_Jan_2007.pdf

Nardi, B., Whittaker, S., & Schwarz, H. (2002). It’s Not What You Know, It’s Who You Know: Work in the Information Age. First Monday, 5, 5. http://www.firstmonday.org/issues/issue5_5/nardi/

Coakes, E. Knowledge Management: A Sociotechnical Perspective. In Coakes, El, Willis D., and Clarke, S. (Eds) Knowledge Management in the SocioTechnical World:CSCW. Springer 2002.
Blogs and Web Sites

Judith Meskell: Home of the Social Networking Services Meta List. http://socialsoftware.weblogsinc.com/2005/02/14/home-of-the-social-networking-services-meta-list/

References

Ackerman, M. (1998). Augmenting Organizational Memory: A Field Study of Answer Garden. In ACM Transactions on Information Systems, 16, 3, 203-224.

Alavi, M. and Leidner D. (1999). Knowledge Management Systems: Issues, Challengs, and Benefits. Communications of AIS, vol 1.

Backstom, L., Huttenlocker, D., Lan, X., Kleinberg, J. (2006). Group formation in large social networks: Membership, growth, and evolution. KDD 2006.

Blood, R. The Weblog Handbook: Practical Advice on Creating and Maintaining your Blog. Cambridge, MA: Perseus Publishing, 2002.

Cherny, L. Conversation and Community: Chat in a Virtual World. Leland: CLSI Publications, 1999.

Cross, R., Borgatti, S. (2004). The ties that share: Relational characteristics that facilitate information seeking. In Huysman, M., and Wulf, V. (eds.) Social Capital and Information Technology. The MIT Press: Cambridge MA.

Donath, J., and Boyd, d. Public displays of connection. In BT Technology Journal Vol 22, No 4. October 2004, pp 71-82.

Ehrlick, K. and Cash, D. (1994). Turning Information into Knowledge: Information Finding as a Collaborative Activity. In Digital Libraries ’94, 119-125.

Farnham, S., Kelly, S.U., Portnoy, W., & Schwartz, J.L.K. Wallop: Designing Social Software for Co-located Social Networks. In Proceedings of HICSS-37, Hawaii (2004).

Farnham, S., Portnoy, W., Turski, A., Cheng, L., Vronay, D. (2003). Personal Map: Automatically Modeling the User’s Online Social Network. In Proceedings of Interact 2003, Zurich.

Herring, S., Scheidt, L., Bonus, S., & Wright, E. Bridging the Gap: A Genre Analysis of Weblogs. In Proc. HICSS-37 (2004).

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.

Kumar, R., Novak, J., Tomkins, A. (2006). Struture and evolution of online social networks. KDD 2006.

McDonald, D. and Ackerman, M. (2000). Expertise Recommender: A Flexible Recommendation System and Architecture. In Proceedings of CSCW, December 2000, Philadelphia.

Mori, J., Sugiyama, T., Matsuo, Y. (2005). Real-world oriented information sharing using social networks. GROUP 2005.

Nardi, B., Whittaker, S., & Schwarz, H. (2002). It’s Not What You Know, It’s Who You Know: Work in the Information Age. First Monday, 5, 5.

Nonnecke, B., & Preece, J. Persistence and Lurkers in Discussion Lists: A Pilot Study. In Proc. HICSS-33 (2000).

Resnick, P, Kuwabara, K., Zeckhauser, R., Friedman, E.. (2000). Reputation Systems. In Communications of the ACM. Vol. 43, 12, pp. 45-48. ACM Press.

Smith, M. and Fiore, A. (2001) Visualization Components for Persistent Conversations. In Proceedings of CHI 2001, Seattle.

Stewart, T. (1997). Intellectual Capital: The New Wealth of Organizations. Doubleday, New York, NY.

Wellman, B., Haythornthwaite, Eds.. (2002). The Internet in Everyday Life. Oxford: Blackwell.