The Metaweb is not just the set of all Weblog posts, it is much more than that. As much as I love to blog I think many old-timers would have us view the entire Net through "blog colored glasses." But Weblog postings are just one kind of microcontent. There will be many others.
The Metaweb is the set of all microcontent on the Web. See my previous articles for a definition of microcontent.
The structure of a Weblog posting is typically defined by RSS (and soon Atom) -- it comprises of a few metatags and some text comprising a headline, a date, an author, a summary, and a few other sections of content depending on what flavor of RSS one uses. This structure is sufficient for describing basic Web resources such as a blog posting, a news article, a web page perhaps. But it is woefully inadequate for describing other types of things such as events, people, places, products. RSS and its descendants will be used for much more than just syndicating weblog postings and web site headlines -- they will be used to syndicate event listings, product listings, classified ads, reviews, pictures, audio tracks, contact records, and many other types of information. But to do that RSS must be extended to provide metadata for those types of things.
To be fair various versions of RSS provide for extensions to be created -- but what good is creating and using a bunch of custom metatags if nothing out there can understand them? The problem with all these specs is that while they allow for extensibility they provide little in the means of interoperability of extensions. Furthermore the current crop of RSS readers provide little or no support for doing anything with extended metadata. This is where I think the Semantic Web becomes increasingly important. It provides a means to rigorously define systems of metatags (using ontologies) such that they can be formally understood by any software that is enabled to use ontologies. So if I ship you some microcontent and it contains custom metadata, your software can see what ontology I am using and via that ontology it can correctly interpret my unique metatags. In any case, whether or not you recognize the value of using ontologies, the important poinit is that the Metaweb is going to contain innumerable varieties of microcontent about all sorts of things.
While weblog postings and resulting conversations will continue to be a big part of the early Metaweb, much of the future Metaweb will be comprised of non-conversational microcontent such as database records of one type or another -- most likely product catalog listings, classified ads, content abstracts, calendar events, and the like. The key here is that the concept of "microcontent" is very far reaching and ultimately much more significant than any particular application of it such as blogging. While blogging is certainly the driver of early Metaweb adoption, in the long run it will probably be corporate microcontent that creates the big business models and long-term growth and adoption of the Metaweb by the mainstream.
I should also point out that the concept of "syndication" is not essential to the Metaweb. With all due respect to my friend Doc Searls, the fact that microcontent can be syndicated is not the key contribution of microcontent, although it is useful. The greatest benefit of Microcontent will ultimately be the widespread use of metadata to frame content. That's why I use "Meta" in the term "Metaweb." As more metadata is added to the Web, the data becomes "smarter" so applications don't have to work so hard. This will result in better searches, better filtering, more targeted publishing and marketing, more productive information management, easier knowledge discovery, better decisionmaking and collaboration, and many other benefits.
There will be much metaweb content that will not necessarily be syndicated -- instead such microcontent will reside in databases, on desktops and enterprise applications, and will be embedded in Web sites. By virtue of its metadata applications such as search engines, web scrapers, RSS readers, Web browsers, intelligent agents, etc. will be able to discover this microcontent and recognize its structure. Whether or not the microcontent is pulled down automatically (e.g. "syndicated") to an RSS reader or to another site is not so important -- what is important is that the metadata exists and is useful.
At present the metaweb hasn't "crossed the chasm," but within 3 to 5 years it will. And then we will see new uses of the microcontent paradigm spreading virally through businesses, universities, communities, governments -- like wildfire -- like Web1, the original Web from which it was born.
At my company, Radar Networks, we are developing a new platform for the Metaweb that will provide the most powerful toolset available for publishing and subscribing to microcontent. We are still in stealth mode.