Comment Flow: Visualizing Communication along Network Path

Social networks are abstract organizational structures that help us understand the relationships among a group of interconnected individuals. Much recent research has focused on understanding the structure of these networks. Yet the network itself is a conceptual topology. The key is the activity that flows along the network paths: the support offered, the information given, the gossip exchanged. Our goal in this research is to develop ways of representing social networks that embody the individuality of the nodes and links. By incorporating this information, we can advance our understanding of the social dynamics of the network, including the roles of the individual members and the temporal changes in the network structure. We have designed and implemented a flexible tool for the content driven exploration and visualization of a social network. Building upon a traditional force-directed network layout, our system shows the activity and the information exchange between nodes, taking the sequence and age of the messages into account. This project serves both as an illustration of one approach to the general problem of individuated network visualization and as an example of the practical uses of such representations. Introduct ion In the last couple of years, online social networking platforms have become widely popular. In these services, participants create selfdescriptive profile pages that often include their tastes and preferences, and a number of links to other members of the same service, usually called “friends”. The purpose is to meet new people, make new business contacts by taking advantage of network effects. Social networks are abstract organizational structures that help us understand the relationships among a group of interconnected individuals. Much recent research has focused on understanding the structure of these networks, identifying patterns such as bridges, structural holes, etc. and on developing visualizations for these often complex entities. Typical network diagrams are detailed in their structure, but generic in their representation of the nodes and links. Yet it is the differences among the people (nodes) and their relationships (links) that create the specific structure of each network and that determine the strength and significance of the ties. The first part of the paper outlines the key issues in visualizing social networks with individuated nodes and links. The second part describes a specific research project the visualization of communication patterns in an online social network site carried out within this program. This project serves both as an illustration of one approach to the general problem of individuated network visualization and as an example of the practical uses of such representations. In the mySpace service (the networking site used for this research) network-only visualization methods are no longer sufficient to meaningfully represent the community structure. Numerous commercial profiles, fake/spam/celebrity profiles and widgets such as automatic friend adders result in a huge numbers of connections, many of which carry little information about a person’s actual social ties and behavior. The average mySpace user has around 131 friends, but there are also profiles with over a million “friends”. By going beyond the ”skeleton” of network connectivity [1] and looking at the flow of information between the individual actors we can create a far more accurate portrait of online social life. Comparing the first generation of social networking sites to the services that are currently popular, we find that today’s social software platforms are more content-centric than their earlier counterparts, which were more concerned with the structural representation of the social network. In recent social software platforms such as Flickr, Facebook, mySpace, the social network serves the purpose of providing context information for the different kinds of activities. In the last couple of years, online social networking platforms have become widely popular. In these services, participants create selfdescriptive profile pages that often include their tastes and preferences, and a number of links to other members of the same service, usually called “friends”. The purpose is to meet new people, make new business contacts by taking advantage of network effects. Social networks are abstract organizational structures that help us understand the relationships among a group of interconnected individuals. Much recent research has focused on understanding the structure of these networks, identifying patterns such as bridges, structural holes, etc. and on developing visualizations for these often complex entities. Typical network diagrams are detailed in their structure, but generic in their representation of the nodes and links. Yet it is the differences among the people (nodes) and their relationships (links) that create the specific structure of each network and that determine the strength and significance of the ties. The first part of the paper outlines the key issues in visualizing social networks with individuated nodes and links. The second part describes a specific research project the visualization of communication patterns in an online social network site carried out within this program. This project serves both as an illustration of one approach to the general problem of individuated network visualization and as an example of the practical uses of such representations. In the mySpace service (the networking site used for this research) network-only visualization methods are no longer sufficient to meaningfully represent the community structure. Numerous commercial profiles, fake/spam/celebrity profiles and widgets such as automatic friend adders result in a huge numbers of connections, many of which carry little information about a person’s actual social ties and behavior. The average mySpace user has around 131 friends, but there are also profiles with over a million “friends”. By going beyond the ”skeleton” of network connectivity (Barabási and Crandall 2002) and looking at the flow of information between the individual actors we can create a far more accurate portrait of online social life. Comparing the first generation of social networking sites to the services that are currently popular, we find that today’s social software platforms are more content-centric than their earlier counterparts, which were more concerned with the structural representation of the social network. In recent social software platforms such as Flickr, Facebook, mySpace, the social network serves the purpose of providing context information for the different kinds of activities. As Donath and Boyd described (Donath and Boyd 2004), most traditional social networking sites rely on a similar model for interpersonal connection that suffers from a problematic simplification. The links between people are mutual and public, but also unnuanced and decontextualized: no distinction is made between the connection to a close relative and a complete stranger. All friend links are equal, and usually equally visible to public. The information is not very reliable extra information is required to acknowledge the highly differentiated nature of human relationships. Feld used the term ’focus’ for the different, sometimes incompatible partitions of social relationships(Feld 1981). As a result, people browsing public profiles of strangers in order to meet new people do not know whether the interesting profile they encounter represents a real person at all it might as well be a commercial disguised as a person. There is also active spamming activity, the fake profiles don’t wait to be discovered members of sites receive numerous invitations, often accompanied by a message written in a personal tone, that are in fact automatically generated spam. The users have little awareness of the nature of the social the space they move through, no means of understanding who these other people are they meet / making sense of the space and the people they meet there. One way of solving this problem is by focusing on the communicative activities taking place in the social network. By looking at their flow of conversation, we can understand who these strangers are.