User-centric identity has become a buzz word in the identity market. Eve Maler, Paul Madsen and Pete Rowley have all been talking about it. I was on a panel with Eve and Kim Cameron a couple weeks ago during Catalyst where we talked about user-centric identity. My key thoughts on it were:

• The user is in the middle of a data transaction. This does not mean the user has to approve every transaction, but that the data always flows through the user’s identity agent. This does have user control and consent advantages that others point out, but I think more importantly, it provides huge scale advantages as the Identity Provider does not have to have any prior knowledge of the Service Provider. The network of sites can build up ad-hoc, just like SMTP servers do today.
• The user has a consistent user experience. That does not mean that all users have the same user experience, but that a specific user is using the same identity agent over and over for each identity transaction, similar to the interfaces we all see for saving and printing files regardless of the application. Currently each SP provides its own user interface which means the user is learning a new interface, sometime for onetime use (eg. site registration) By separating the identity component from the rest of the application, the user also has more certainty on who the SP is which helps resolve phishing.

The scale, security and usability advantages of user-centric identity are what makes it the underpinning for Identity 2.0.