WebVisions 2007: Social Architecture

by Luke Wroblewski May 4, 2007

Sean Madden’s Social Architecture: Modeling the Next Generation talk at WebVisions 2007 outlined some current social network feature sets and proposed how we might expand our toolkits to acclerate the next iteration of social networking.

  • Passion Centric social networks: very specific focus (dogster, etc.)
  • Passion Agnostic social networks: no purpose on surface (MySpace, etc.)
  • Tagging: communicate “metadata to the masses”, easy to do, provides a forum for people to build organization systems.
  • Tagging is work. You need to be conscious about it.
  • Voting: promotes trust and vulnerability, enables community to decide what is valuable
  • Voting tends to skew toward the mean.
  • Extensions: empowers users, increases value (stickiness), establishes credibility & confidence
  • Customization: new because your customizations are visible to others, allows for self-expression, increases sense of ownership.
  • Customization is hard work.
  • All these systems (tagging, voting, customizing) detract from the really social aspects of networks. How can we automate some stuff to reduce cognitive load?
  • Ubiquitous Computing: computer functions are integrated into everyday life. Mobile phones are start of computing in the world.
  • Localized features: QR codes in Japan, lampposts that broadcast local information.
  • Calm Technology: take complexity & get it out of the mind of the user.
  • Beautiful Seams: a way to make ubiquitous computing transitions visible. Let you know when you are in a field of presence.
  • Can we enable social networks to leverage ubiquitous computing to take mental load off of users?
  • Emergence: highly complex system that arises without a master design. Built up from simple rules that build into a complex system. Needs high numbers to work.
  • Ignorance is useful: don’t want excessive thinking else simple rules break.
  • Encourage random encounters: to trigger simple rules. Look for patterns & pay attention to neighbors.
  • Is emergence is a model we can apply to social networks? Pattern matching and behavior recognition can take some of the cognitive load from users.
  • Genetic algorithms: simple steps that can be built to solve problems. Can they apply to social networks?
  • Create a base set of rules and let the community grow and cultivate them through use.
  • Emergent Organization: let the system adjust based on usage. An emergent system is a less bounded system. Admit to ourselves that we don’t know what users want.
  • To design a passion agnostic community, you need to cultivate emergent behavior.
  • Social networks have limitless potential, we need to work towards designing them that way.