Sorry to ruin the Release 2.0 party, but one of the fundamental aspects of the Web (no need for a version number there) is precisely that feeds and resources and lists and sensors and all that stuff are not different kinds of animals, as Jarvis claims. As far as the Web is concerned, they are all really just special cases of resources, identified by URIs, and part of the Web overall. We are not seeing a new architecture: We are seeing people finally make some good use of the existing one.