One of the recurring themes at Foo Camp East was the need for social networking systems that more closely accomodated the nuances of real-world social relationships. Linda Stone called for retiring the word "friend" which, as danah boyd described, does not match the many and varied form of human relationships. Chris Messina described efforts such as OpenID and DiSo to address the problem by making social networking systems (SNS) more open and interconnected. One problem is that identity itself is not well understood by the individuals who use these systems. In a social network, identity is often as simple, and as brittle, as the account one has created on a system such as Facebook or Twitter. That identity is useful to other people to the extent that it is connected by a social graph to other identities in the same system, i.e. you are who your friends say you are. The fact that a person's identiy on one such system may be difficult to match up with that same person's identity on another system is either a bug or a feature, depending on whether you are trying to preserve your privacy or whether you are trying to keep track of your friends (or customers) as they migrate from one system to another.
While we debate how easy or hard it should be to unify these personae, and while various efforts arise to allow them to be unified, some technical developments may render the issue moot. Tim O'Reilly introduced me to the concept of Information Shadows, the digital representation and associated data of a physical object, including a person. As originally described by Mike Kuniavsky, information shadows are things like Amazon's ASINs, the airlines' e-tickets and Ulla-Maaria Mutanen's Thinglink - that is the digital data that refers to, and provides information about, real-world entitites. As more people and objects acquire digital identities, the information shadows become the primary way we deal with the objects and define how we understand them, much as the shadows in Plato's Cave became the reality for the prisoners observing them.
In another development, Arvind Narayanan and Vitaly Shmatikov of the University of Texas recently showed how an analysis of one's friends in a social network could be used to unmask one's identity. In their paper, De-anonymizing Social Networks, they describe how a third of the people who have accounts on both Flickr and Twitter can be matched up with only a 12% error rate, even though the overlap in the relationships for these members is less than 15%. What this implies is that we don't need to wait for One Social Network to Rule Them All - that the computer will help us find our "friends" even if they move from one SNS to another and assume a new persona in each. Whether this is good or bad depends on whether you value privacy or connectivity, but the result is inevitable so to paraphrase Scott McNealy, you might as well get over it.
