Background

In the past decade, social technologies have played an increasingly important role in the maintenance of people’s day-to-day social lives. Social software, designed specifically to support people’s mainstream social practices (Davies, 2003), has become a dominant class of application in both the home and the work place. Online communication and sharing—in the form of email, mailing lists, social networks, online photo sharing, IM, SMS and blogs—appear to be an integral part of ongoing conversations, enabling increased lightweight interactions over the course of the day, and casual, last minute planning for social occasions (Farnham et al., 2004). According to a Pew Internet Report on Internet use, 63% Of Americans are online, and over 90% use email. Online posts of photos, mp3s, and journals are frequent venues for sharing around social experiences. Approximately 44% of U.S. Internet users have contributed personal content to the online world (Lenhart, Horrigon, & Fallows, 2004). Meeting people online is increasingly less stigmatized, and online dating and social networking tools, such as Match.com, MySpace.com and Facebook.com, are increasingly popular ways to meet new people (Boyd, 2004).

Similarly, in the domain of information technology, socially derived metadata has played an increasingly important role in helping people manage the overwhelming amount of content available through the Internet. This social metadata, such as who authored what content, who has looked at what content, what content do similar others like, and who is connected to whom, are important sources of information that allow content technologies to filter and prioritize information. The behavior of others around online information provides meaningful insights into the value of the information, enabling social navigation (Hook, Benyon, & Munro, 2003). For example, Google.com prioritizes search results for web pages by the extent to which others link to each page, SlashDot prioritizes articles and comments by the extent to which others positively rate the articles, and Facebook limits access to social updates depending on who knows whom.

“The ideas of social navigation build on a more general concept that interacting with computers can be seen as “navigation” in information space….Just as we use social methods to find our way through geographical spaces, so we are interested in how social methods can be used in information spaces.”

Hook, Benyaon & Munro, 2003

Although social software use has grown rapidly in the past decade, as a field of practice the development of social software is fairly young, evolving just over the past few years into a reasonably coherent set of approaches and problems. In this Handbook we argue that practitioners should adopt a “social engineering” approach in the design and development of social software, focusing on the unique opportunities afforded to users through social technologies in helping them achieve their social goals. We then review current trends in innovation in social technologies, and discuss the hard problems that need to be addressed in these areas in the next several years. In particular, with the increased a) digitization of identity and social networks, b) proliferation of social metadata, c) lightweight authoring and syndication, and d) mobility and ubiquity of social processes, we expect that the field of social software will prove to be an increasingly important arena for innovation. We expect that as social technologies continue to innovate we must increasingly envision an integrated social software experience built on the Web 2.0 platform.

Historically, much of the early work related to social technologies focused on the deficiencies found in computer-mediated social interactions. As email became an increasingly prominent communication tool, much of the early CMC (Computer-Mediated Communication) research focused on the impact the reduced nonverbal cues in text-based communication and the resulting impact on people’s abilities to maintain conversational norms, develop accurate person perceptions, or develop a sense of social presence (Walther, 1996). Much of the research on virtual environments attempted to redress the para verbal deficiencies found in such text-only communication (Smith, Farnham, & Drucker, 2000). CSCW (Computer-supported Cooperative Work) research similarly focused on the negative impact the early computer-mediated social environments had on people’s ability to collaborate, due to difficulties in achieving shared understanding and coordination around complex tasks (Preece, 2000). More recent Social Computing research has focused on how to redress the reduced levels of accountability and increased bad behavior found in online social spaces (Jensen, Davis, & Farnham, 2002; Resnick et al., 2000). 

Looking to the future, we observe an increased optimism toward the use of social software for improving and augmenting what is possible in people’s day to day lives.  when designing social software, its not just about the dificiencies in said software relative to more natural face to interactions:  it is how social software can meaningful improve the quality of the moments we share together in the real world.

The Social Engineering Approach

The goals of online social software are to engender positive, concrete social outcomes such as the development of face-to-face friendships, dating relationships, collaboration, and collective action. Thus as practitioners create social technologies, they must take on the role of “social engineers”, designing social applications that engineer the desired social experiences.

The social engineering approach creates technological interventions: producing changes in the users’ social environment that help them achieve their social goals. Much as a match-making individual might throw a dinner party to foster a developing dating relationship between friends, so a social engineer will create an online community environment to help participants advance their community’s goals. Social engineering is by nature a cross-disciplinary activity, incorporating the knowledge and skills of social science, design, and web development, in creating social software. The following example shows how practitioners in the different discplines involved can adopt the social engineering approach to meet users’ social goals.

An Example of Social Engineering

Consider the domain of online dating. How can a team of social engineers best design a profiling and matchmaking system to increase the likelihood of two people finding each other and having a successful dating experience? Answering this question requires a basic understanding of the psychology of attraction and relationship development. For example, a primary determinant of attraction is similarity of personal characteristics. The stereotype that opposites attract is not generally supported by studies of interpersonal attraction (Bersheid & Regan, 2004). In our own studies of online profiles and matchmaking, we have found over and over again that similarity is a key determinant of liking and preference (Jensen, Davis, & Farnham, 2002). Thus an effective matchmaking system would match people based on demographic characteristics, similarity of interest, values, and so forth as provided in the users’ profiles.

Proximity is another predictor of attraction, because proximity (e.g., students living in the same dorm) engenders greater frequency of exposure, and people tend to like that which is familiar. Online systems seeking to exploit this dynamic should provide tools for frequent exposure and interaction with the same person. Evite.com, for example, provides tools through which people may observe whom in their social circle will be attending any particular event, affording increased opportunities for face-to-face exposure.

People also tend to like their friends’ friends (called Balance Theory in the field of psychology, see Heider, 1958). That is, if Jane likes John, and John like Mary, then Jane is likely to also like Mary. Many recent applications have been taking advantage of this balance effect, by placing profiles in a social context. Facebook, for example, encourages people to publicly link their own profies to their friends’ profiles. This provides a direct means for user to quickly identity friends of their friends — ideal candidates for new friendships and relationships.

Perhaps most importantly, people tend to prefer physically attractive others as dating partners. Thus it is important to encourage people to post photos in their profiles. However, other factors also predict a successful relationship in addition to perceived attractiveness. One such variable is the extent to which two people match in their desirability. While everyone would like to match with the most desirable partner, successful matches tend to occur between people who are reasonably matched in their levels of socio-economic status and physical attractiveness (Feingold, 1988). Thus, although the user may explicitly prefer to approach the most attractive person in the system, a successful dating service should filter extreme discrepancies in levels of desirability.

Finally, it is important in designing a match-making system to understand the process of relationship development. People generally achieve greater levels of intimacy with others through graduated levels of reciprocal self-disclosure.  To put it more simply, I say a little, you say a little, I say a little more, you say a little more, and so the cycle goes until we reach true intimacy.  Thus an effective online system should provide tools for gradual transitions in communication between people who have observed each other’s profiles, to people who send each other semi-anonymous messages, to people who send each other messages through their personal, completely identifiable email messages. Most systems, for example, allows people to send private messages within the system, at which time they can disclose more personal information. Match.com provides email accounts for people to email each other with, providing an initial pseudonymity which can then gracefully transition into more intimate personal exchanges.

As this example illustrates, in addition to designing and developing usable social interfaces, a social engineering team must take into consideration the underlying social processes involved in helping users meet their social goals, and map these processes on to the social software.

Principles of Social Engineering

The following is an outline of some basic principles of social engineering to keep in mind when designing, building, and deploying social technologies.

1. Define the Users’ Goals, both Implicit and Explicit

People’s social goals are largely affective in nature, relative to the more task-oriented goals of the knowledge worker. People have three primary social goals that should be taken into consideration when developing social technologies.

Liking myself. The first is that people generally want to feel good about themselves. They will avoid technologies that make them feel or appear bad, and seek out technologies that provide them with a sense of mastery or the opportunity to construct an attractive persona. It is important to note that technological features that are value expressive (that really say something about what you are like), such as a custom avatar, or a personalized skin, will prove enjoyable to the user even in the absence of any observers. People find identity construction and expression inherently enjoyable tasks. The mere presence of self is pleasing. We were particularly struck by this principle in observing the use of the Personal Map (Farnham et al., 2003), a visualization of people’s personals email social networks. Although it was superfluous to the function of the applcation, to cater to this egocentricism we placed the users’ name in the center of the map, and then organized everyone relative to that center person.

Others liking me. The second social goal is that people enjoy sharing and eliciting approval and affection from others. They will seek out technologies that provide them with opportunities to find new friends or dating partners, or share with their family and friends. Again, even the mere awareness of the presence of others will generally bring a positive emotional response from users in a social system. We found in our Swarm application (Keyani & Farnham, ????), for example, a group text messaging system, that people enjoyed getting notifications of presence from their friends whether or not they ever replied, or in any way acted up the information provided in the notification. Online reputation systems take advantage of people’s desire to be liked and respected by providing numerical or visual representations of people’s status. These reputation indicators are strong incentives for people to curry the favor of others within the system (Resnick et al., 2000).

Sense of belonging. The third is that people feel an intrinsic joy in being similar to others or in being a part of social groups. Social groups are quite sticky once people develop an identification with a group. People will seek out and use technologies that increase their awareness of and identification with their social groups, even if the social group is minimally defined (e.g., blue spots vs. red spots in a game)(Billig & Tajfel, 1973). Thus technologies that list relationships or group affiliation will provide an affectively positive experience to the user even in the absence of any real functionality. It is quite remarkable, for example, the number of people in online social networks who have taken the time to provide rich profiles and develop large lists of friends who have no reason for being in the system other than “just here to help out”.

That people will seek out and use technologies that enable them to accomplish their social goals might seem an obvious point to make. However, it is often the case that social software diverts the users’ attention away from achieving their social goals, towards the software itself. In a setting where we are actually creating social software prototypes, shifting one’s approach from that of a designer, or developer, or a psychologist, to that of the Social Engineer should help us focus on enabling the users to achieve specific social outcomes. If we are not helping people achieve their goals our systems are not successful, no matter how pretty they look or how fast they run. While we might hope to delight the user upon introduction to a new system with a beautiful, usable design, ultimately the software should become invisible to the user, because they are thinking only about their desired outcome and not the system itself. When people are sharing photos with their friends, for example, they are not thinking about the quality of the picture or the communication transport; they are expressing to their friends, I like you, I miss you, I hope you will enjoy sharing this experience with me.

The Social Engineer should also take into consideration both users’ explicit and implicit goals. For example, in knowledge management systems in a corporate environment, people might explicitly seek out the highest quality answer to a particular question they are posing to the system. However, implicitly they might be hoping to develop friendly, collaborative relationships with others within the corporation. In a study of the use of social networks in a knowledge exchange situation, for example, we found surprisingly that number of reports was a strong predictor of whom people would ask for information irrespective of expertise (Farnham, Portnoy, & Turski, 2004), suggesting that people hope to develop connections with higher status persons within the corporation. A knowledge management system that simply provided the answer to a question would suffice to satisfy the users’ explicit goal, however to truly satisfy the user the system should also provide information about who answered the question, and the person’s social status.

2. Take the Perspective of the User

phe•nom•e•nol•o•gy ( P ) Pronunciation Key (f-nm-nl-j)

n.

A philosophy or method of inquiry based on the premise that reality consists of objects and events as they are perceived or understood in human consciousness and not of anything independent of human consciousness.

www.dictionary.com

People’s phenomenological, subjective interpretations of their social experiences do not necessarily map onto more objective reality. However, it is important to keep in mind that it is just this subjective interpretation that drives peoples’ beahviors, not the objective reality. We adopted such an egocentric, phenomenological approach when designing the Personal Map (Farnham et al., 2003), which implicitly infers social networks from the users’ email behavior. That is, we were much less concerned with inferring and visualizing actual email social networks than we were with matching our user interfaces to each individual’s unique expectations about his or her social space. Rather than inferring the user’s meaningful social groups from the user’s actual interaction patterns, we inferred the user’s groups from how the user grouped them (e.g., on the cc line). We found in our user feedback studies that they reported errors in cases only where important people were not extracted because they did not frequently interact with those people, but were emotionally close. A user’s husband might not appear as close to her in her social network as her dentist, for example, depending on the frequency of interaction. Thus a particularly intelligent system would infer importance not only through frequency of interaction, but also through other indicators of affective importance such as immediacy of response, longevity in the system, and the presence of emoticons in the messages. Developing intuitive, implicit models of users’ social networks requires stepping inside their minds and understanding their subjective experiences, and then looking to their behavior to explore how to operationalize these subjective experiences.

The phenomenological approach to understanding user behavior is particularly important in gaming contexts where the value of digital objects is becoming increasingly subjectively real to the user. Much as people now tend to perceive money has having inherent value, even though it is only a piece of paper, people are perceiving digital objects such as gaming characters as inherently valuable. The creation and exchange of meaningful digital gifts in online systems is another example of the commoditization of subjectively valued, yet ephemeral digital objects, found in applications such as www.funhi.com, or www.ecards.com. In examining these online economies, it becomes clear that recognizing the distinction between objective reality and subjective reality provides the Social Engineer with new opportunities for creating tools that meet the users’ social goals.

It is also important to take the perspective of the user when understanding how they interpret the behavior of others in online systems. For example, we found in one study of shared browsing (Farnham, Zaner, & Cheng, 2001) that people’s ability to understand the intentions of other people, what they were about to do, was as important as being able to observer what they were doing. Observing the trajectory of each others’ mouse pointers meaningfully communicated each others’ intentions, enabling smoother transitions in turn taking when mutually controlling a shared space.

3. User Behavior is a Function of the Person and the Situation

A basic principle of Social Psychology is that behavior is the function of both the person and the situation. From a phenomenological perspective, individuals behave differently according to the ways in which tensions between their perceptions of the self and of the environment are negotiated, where often social forces are experienced as much more real than any physical constraint. Group dynamics are comprised of interdependent individuals, such that a change in one member or subgroup can have a meaningful impact upon others in the system. The whole psychological field, or ’lifespace’, within which people act has to be taken into consideration, in order to understand, predict, and shape behavior (Lewin, 1951). For example, we cannot understand why the same group of teenaged student might be very talkative in one chat room, and very silent in another chat room, until we understand that one chat room is private, and the other chat room is public and accessible to their teachers.

When hoping to shape behaviors in a social system, social engineers must view the whole system as the sum of various social forces, and consider how to structure the forces experienced within the environment. Consider, for example, an online chat room where the goal is to reduce the prevalence of bad behavior in the system without employing costly, real time human moderation. One relatively obvious technical solution is to scan individual chat messages for having the characteristics of someone behaving badly (e.g. exclamation points and swear words), and then exclude badly behaving people from the system. However, an alternative solution is to recognize that most people are more motivated towards social acceptance than negative social attention from groups with which they identify, and provide tools within the system to enable users to a) easily develop a group identification with a set of people, and b) provide group members with tools for collective, public declarations of social acceptance.

We did perform one such study several years ago (Davis, Ma, & Farnham, (2002), in which we experimentally manipulated whether people in discussion groups had the ability to evaluate the quality of each others’ contributions. We found that although people did not use the tool in the experimental group, the members of the groups felt that everyone was better behaved: thus they felt the social force of possible social denunciation, even though no one actually received any negative feedback.

4. Person Characteristics Meaningfully Impact Use: Personality and Life Stage

The Social Engineer takes into consideration person characteristics that might impact their social goals. For example, in a survey study of users (Warnock & Farnham, 2004), we found that how and why people use social technologies is affected by introversion and extroversion. Extroverts reported being more likely to use asynchronous forms of communication that do not replace real time face-to-face interaction, like email, and forms of technology that are more likely to lead to face-to-face interaction, such as cell phones, than are introverts. In contrast, introverts were more likely to participate in synchronous forms of social technologies, such as online gaming, because they are more likely to use technology as a replacement for face-to-face interaction.

Life stage is also a very important predictor of how and why people use social technologies. People who are younger and single, for example, are generally in a phase of life where they are seeking out new social relationships, so applications like Friendster.com will be particularly appealing to them. People who have established careers, families, and social lives might be in a phase of life where they are trying to filter out people in their social network who are making excessive demands on their limited attention. They would probably be more inclined to use social technologies that restrict access to small groups.

5. Design Social Technologies to Augment Face-to-Face Interactions, not Take their Place

When designing social systems the social engineer should always ask the question, what social outcomes are we trying to change or improve through our social software? Just as importantly, the social engineer should consider how can we create an experience through our technology that will help the user achieve his or her social goals beyond what is already available through face-to-face interactions? A grounded perspective on social technologies states don’t fix what isn’t broken. Social technologies should not negatively impact people’s existing, healthy, face-to-face interactions.

Rather, what unique affordances are provided through social technologies for helping people achieve their social goals that augment or enhance healthy social interactions? There are many unique affordances to new social technologies that produce increased opportunities for social interactions and engender new forms of social interactions — creating a social landscape that is very different from that of a century ago or even two decades ago.

We will discuss in greater detail these unique affordances of going online, but urge practioners to not lose their way down the path of trying to create online technologies to replace social beheviors that are better left offline.

Recommended Reading

These readings provide an introduction to the space from several perspectives.

Books and Papers

• Davies, W. You Don’t Know Me, but…Social Capitol and Social Software. ISociety(2003). http://www.theworkfoundation.com/Assets/PDFs/you_dontknowme.pdf
• Madden, M. America’s Online Pursuits: The Changing Picture of Who’s Online and What They Do. Pew Internet & American Life Project (2003).
• Wallace, P (2001). Psychology of the Internet. Cambridge University Press: Cambridge.
• Preece, J. (2000). Online Communities: Designing usability, supporting sociability.
• Rheingold, H. (2002). Smart Mobs: The Next Social Revolution.

Blogs

• Many 2 Many: A Group Weblog on Social Software. http://many.corante.com
• SmartMobs: The Next Social Revolution. http://www.smartmobs.com
• Mashable.com: Social Networking News. http://www.mashable.com
• danah Boyd’s apophenia: Making Connections Where None Previously Existed. http://www.zephoria.org/thoughts/

References

Berscheid, E. Regan, P. (2004). The Psychology of Interpersonal Relationships. Prentice Hall.

Billig, M. & Tajfel, H. (1973). Social Categorization and Similarity in Intergroup Behavior. European Journal of Social Psychology, 3, 2 7-52.

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

Davies, W. You Don’t Know Me, but…Social Capitol and Social Software. ISociety (2003).

Davis, J. P., Ma, M. and Farnham, S. D. (2002). Using a peer rating system to improve online social behavior. Unpublished paper.

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. (2004). Using email mailing lists to approximate and explore corporate social networks. Paper presented at CSCW Workshop on Social Networks.

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.

Farnham, S., Zaner, M., Cheng, L. (2001). Designing for Sociability in Shared Browsers. In Proceedings of Interact 2001, Tokyo, July, 2001.

Feingold, A. (1988) Matching for attractiveness in romantic partners and same-sex friends: a meta-analysis and theoretical critique, Psychological Bulletin, 104, 226-35.

Heider, Fritz. (1958). The Psychology of Interpersonal Relations. NY: Wiley. Chapter 7: Sentiment pp.174-217.

Hook, K., Benyon, D., and Munro, A. (Eds) Designing Information Spaces: The Social Navigation Appoach. Springer (2004).

Jensen, C., Davis, J. P. and Farnham, S. (2002). Finding others online: Reputation systems for online social spaces. In Proceedings of CHI ’02 (Minneapolis, MN, April 2002), ACM Press.

Keyani, P., & Farnham, S.(In Press). Swarm: Text Messaging Designed to Enhance Social Coordination. In Harper, R., Palen, L., Taylor, A. (Eds.) The Inside Text: Social, Cultural, and Design Perspectives on SMS.

Lenhart, A., Horrigon, J., & Fallows, D. Content Creation Online. Pew Internet & American Life Project (2004).

Lewin, K. (1951) Field theory in social science; selected theoretical papers. D. Cartwright (ed.). New York: Harper & Row.

Madden, M. America’s Online Pursuits: The Changing Picture of Who’s Online and What They Do. Pew Internet & American Life Project (2003).

Preece, J. (2000) Online communities: Designing usability, supporting sociability. Chichester, John Wiley & Sons, Ltd.

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., Farnham, S., & Drucker S. The Social Life of Small Graphical Chat Spaces . In Proceedings of CHI 2000, The Hague, Netherlands March 2000.

Walther, J. B. Computer-mediated communication: Impersonal, interpersonal, and hyperpersonal interaction. Communication Research, 23 (1996), 3-43.

Warnock, D., & Farnham, S. (2004). Different Strokes for Different Folks?: Extraversion and the effects of access to and use of social technology on relationship satisfaction. Unpublished paper.

© 2009 Waggle Labs.