Reconsidering REST: Do we need a new ontology for the Web?

In his PhD Fielding defined properties he asserted were what made the WWW work. The recent revision of the IETF HTTP standard retains much of his terminology and worldview. In this talk I argue that although a brilliant insight into the 1995 Web, it is now so out-of-date as to be seriously misleading: it actually makes the new standard harder to understand instead of easier. I'll present empirical evidence, based on two sets of caching proxy logs totalling 160M entries, to back up this claim.