I proposed a radically simplified XRI resolution mechanism that relies only on HTTP, leveraging caching for efficiency, and providing a clear path for using a REST architecture to use XRIs. In other words, this resolution is RESTful, and could be used in a RESTful application.
I've gotten fairly positive feedback about it so far and I'm hoping its now within the comprehension of mere mortals. While I was really jazzed about DNS and especially DDDS, its become clear to me that the newness of using DNS for generic identifier resolution is proving to be an impediment to any serious review of the XRI resolution document.
Now, the only server-side functionality for XRI resolution you need is an HTTP server. The client stuff isn't much more complicated - you basically need an XRI parser (including escaping) and a HTTP client library. Pretty simple stuff.
So if you've been putting off looking at the XRI spec, now is the time to go back and take a second (or third) look. Start with sections 1 and 2 of draft -07 and then read through section 3.1 Skip the entire rest of section 3 and read the simplified proposal. Finally finish up the rest of draft -07.
