When I mention sematic markup (I never mention the sematic web when talking to clients) many people look at me like I'm speaking latin. Sematic markup is basically the HTML that surrounds your content and helps give it more meaning (either to a reader or a search engine). This ComputerWorld article indicates that there is some debate as to how far away the semantic web is:

Semantic interoperability remains a utopian vision without many practical implementations.

"...many smart people feel that automated, end-to-end, standards-based semantic interoperability (where computers exchange not just data but the data's meaning as well) is more than a pipe dream."

Semantic Web: stuck in neutral

I, for one don't think it's far at all. RSS and Atom (or any feed format that supports namespaces) already give an XML "view" of data from many different sources.

For example, Google Calendar feeds have a few "Google data namespace" items. One is gd:when. This element gives additional meaning about the feed entry that it is a part of. I would argue that namespace items like these help give the data more meaning.

When developing Interspring I used this namespace data to give different views of the feed (in this case a calendar view). GeoRSS (georgraphy namespace), mediaRSS (audio/video/photo namespace), and many others help give many feeds rich meta data to left join to some other data source on the web to present a more complete view of the feed item.