Semantifying Social Networks, Socializing The Semantic Web « John Wright | Software Developer | Los Angeles

(DISCLAIMER: My recollection of what was said and what was meant may be flawed. For the best record, find the sources mentioned and listen for yourself)

During the Semantic Web Meetup discussion about social networks Joe Devon made a prediction that all the current major social networks (Facebook and Twitter) will go the way of MySpace. This is because when the semantic web has matured, everyone will have control of their own (profile) data and that data will be more freely movable around the web due to common open standards. This discussion reminded of a similar discussion (Semantifying Social Networks) I listened to via a recording from the SemTech 2009 Conference. In this discussion the same vision of everyone controlling their own data and (by choice) sharing it securely via open standards was the ideal.

But another scenario was mentioned for the future of the web which was the “(Profile) Data Broker” concept. This idea is similar to the current monetary banking system. Just as you don’t store and hold your own money, or actually give the money to the one you’re making the check out to, you wouldn’t store your own profile data. Instead a profile data broker site of sorts would store all the information about you that any other service could possibly need. Then when you go to sign up for a new service on the web, you would refer them to this data broker site and the data broker would give the new service your info. Part of this vision is that your data wouldn’t be accessed without your approval and you could choose what data to give out. Also, there is the idea of incentives in exchange for your data, where you would receive something (money, a gift card, etc…) for giving your data to the new service.

To make things even more interesting, Facebook is allegedly trying to become this premier “data broker” starting with Facebook Connect and Facebooks vanity urls. This alleged roadmap for Facebook wouldn’t be using open standards as Facebook Connect is proprietary and it doesn’t appear that will change anytime soon. This is a centralized approach where Facebook is the central hub for all this data about the masses.

On the flip side we have the idea of a federated approach with the open standard FOAF which contains properties like “knows” (or “Person A” “knows” “Person B”). This provides the basics needed to socialize the semantic web, and since FOAF is already a widely used ontology social connections are already modeled on the semantic web. Although, the grander vision is that a majority of social and profile data on the web be securely available via the semantic web in RDF using FOAF or some alternative ontology. And for this social data to be completely under the control of the individuals it represents.