Of course I'm sure you ask: what IS Web 3.0?
It's the SEMANTIC WEB like all the search gurus and jargon bandwagoneers have been blogging and blagging on about for years now, right?
Sorry people, I have to disagree. Web 3.0 is going to be all about "real-time web."
Let's look at the space:
Facebook gets it! They bought FriendFeed, which is real-time-centric, has real-time-search and created SUP - Simple Update Protocol. SUP is a mechanism of faster RSS & ATOM conveyance; per Wikipedia, some implementors include YouTube, Disqus, Brightkite, Identi.ca, Backtype and 12seconds.tv. And on a separate but related note, Facebook recently launched Facebook Lite.
Twitter gets it - they've been real-time since the beginning! And they have started to analyze conversations in real-time pretty hardcore in the past year or so, beginning with their purchase of Summize. Their real-time offering is their Streaming API. Twitter's busy enough dealing with availability and scaling issues, and gaffed over OAuth too often, to offer something that I would expect to propagate across the industry as a standard protocol however.
People at Google get it, at least 20% of the time: check out PubSubHubBub, (PSHB) a real-time RSS and ATOM mechanism that runs on Google App Engine (GAE). Some big name implementors include FriendFeed, Live Journal and Six Apart. As well as Google itself of course.
Of course real-time tends to greatly increase the volume of information (if not the quality of the signal). Perhaps semantic will play a large role in helping us sift through all the feeds from our various new real-time toys, but Web 3.0's foundation is going to be real-time, period. Rich, robust semantic will perhaps be icing on the cake of real-time.