I had pretty much the same reacon while reading the article. "BlogPosting" isn't particularily informative. The rest of the metadata looked like it could/should be put in <meta> tags, done.
A very bad example if the intention was to demonstrate how cool and useful semweb is :-)
The schema.org data is much more rich than meta tags, though. Using the latter, an author is just a string of text containing who-knows-what. The former lets you specify a name, email address, and url. And that's just for the Person type—you can specify an Organization too.
That's still just tangential Metadata. The point of a semantic web would be to annotate the semantic content of text. The vision was always that you can run a query like, say, "physics:particles: proton-mass", over the entire web, and it would retrieve parts of web pages that talk about the proton mass.
A very bad example if the intention was to demonstrate how cool and useful semweb is :-)