Microsoft has adopted a "claims-based architecture" in its approach to managing the identities of users in its Azure cloud.

At its Professional Developers Conference in L.A. recently, it announced a Microsoft Identity Platform that invokes the architecture to establish a federated identity for users. A federated identity can be used to provide a single sign on to multiple applications, both in the enterprise and in the cloud.

A federated identity of some type is going to be necessity if there is any prospect of hybrid cloud computing coming into vogue. IT departments that ship part of their workload off to the public cloud will need to be able to allow end users of applications to follow them into the cloud and use them there as well.

Microsoft's claims-based architecture is a more flexible approach to establishing a users' identity, than a straight forward, on-premises Active Directory system. The claims-based architecture can accept digital identifiers from multiple sources, such as LDAP directories, Active Directory, Outlook or Lotus Notes directories, certificates from security services, or a Windows token, said Kim Cameron, Microsoft's chief identity architect, in an interview at the developers conference.

Once a user's identity verifier is supplied, a central brokering authority compares the "claim" to that required by a particular application. If there's a match, use of the application can proceed.

Under a claims-based architecture, retrieving some form of digital identity is not enough, said Cameron. It is just a "claim" to an end user identity until the central broker checks its authenticity and its status to see if it meets the requirements of the application. All forms of identity remain untrusted -- they're treated as claims, not proof -- until the central authority decides they meet the needs of the application, he said.