Data portability and the home of the social graph

Phil Wilkinson wrote an interesting post on data portability today.  He is of course thinking about things from a social shopping perspective, but his points are relevant for all social applications.

I think by now most of us are sold on the idea that we should be able to extract our data from sites like Facebook and port it to new services we want to use, like say Crowdstorm.  And that should include our friends lists.  What I liked about Phil’s post is the idea that every additional service we use will enrich our social graph, and that would be for the benefit of everyone when the enriched data is ported back to the originating site, and then beyond.

In my view, the user controls the data, not the site. The process should go along the lines of:

  1. Import the social graph data from any chosen site or service
  2. Add value to it by enhancing the data within (such as adding trust between people or contextual shopping knowledge)
  3. Utilise that data as part of the service to improve the consumer offering
  4. Allow that enhanced social data set to be exported again to any other source

In this view of the world our social graph data lives not in one place, but rather in a sea of connected islands with activity on any one island potentially enriching the experience on any of the other islands.

For example if I add a friend on Facebook who is already on Crowdstorm (but not connected to me) and that friend has recently bought a new plasma then my Crowdstorm experience will be enhanced when I research my new TV there.  Furthermore, our ties on Facebook will be strengthened if it pops up in our newsfeeds or something that I have read the review of the plasma or accepted a recommendation to buy it.

This way our different social graphs would exist as different connected islands and would each enrich each other.  The principle islands would be our favourite socnet, our email social graph and our mobile phone social graph.  The smaller ones would be the other sites we use – e.g. Crowdstorm, Mybloglog, Twitter etc.

At a quite basic level connecting these islands will allow you to sync your networks (should you wish to do so) – e.g. by pointing out that the person you phone the most is on Facebook and you haven’t buddied up with them.

Where it gets more exciting is when applications that run in Facebook etc. take advantage of the extra data that becomes available.  For example games in Facebook could automatically compare your score with those of your best friends as defined by how often you call them.  Alternatively games running in Facebook could be customised based on which other services you use – adopting shopping themes for Crowdstorm fans, technology themes for people who are in the Mashable community.  Really running with this a second (it is Friday..) the data about what I read that sits in Blogfriends could be used to target ads when I am on Myspace.

Exposing our social graph data to the world and encouraging portability will unleash a lot of value, firstly by virtue of the fact that we are able to use this data for the first time, and then secondly by virtue of the fact that the very act of using the data improves it.  It is this last part that will really make a difference.