Johannes Ernst just created an excellent blog post out of an email dialog he and I had about the “big picture” of XRI resolution. It’s true that the concept of an identifier resolution network based on HTTP that parallels the DNS resolution nework based on UDP is unusual at first. But for RESTful architects like Johannes, the lightbulb goes on when look at the power of using the HTTP layer to do for digital identification what the DNS layer did for host naming — especially leveraging the simplicity and extensibility of XRDS documents (the XML format returned by an XRI resolution request, and also the format used by the Yadis URL service discovery protocol of which Johannes was the co-instigator.)

As the XRI Resolution editor’s team starts work this month on what we hope is the final working draft of XRI Resolution 2.0 (currently at working draft 10), I plan to blog more about other key features of XRDS documents — CanonicalIDs, Refs, ProviderIDs, and Service Endpoints — that all help solve higher-level identity and trust problems at a layer higher than DNS was intended to function, much like DNS solved a set of problems of logical naming and redirection at a layer higher than IP addressing was intended to function.