That's what I thought too. Sounds like this guy doesn't really get web apps.
Would you really consider facebook to be a "web app"? I think facebook is, or at least should be a pretty normal web page.
Far from it, amazing complexity in how certain parts of the page load and certain page changes are coming through ajax. Very much a web app, probably as complex as any web app out there.
There's no reason for so much complexity. I'm talking of course about the core of facebook, not the third-party apps. Sure, asynchronous loads of certain content is nice, but it hardly qualifies it as a "web app"? Which parts of facebook are you referring to?
To expect the tool to have the same sort of semantic purity as the documents it interacts with is overly idealistic. It doesn't matter as much what we can systematically infer about an application, it is more important that the content be accessible in some sort of standard format.
Similarly, Scribd's HTML5 viewer, is just that, a viewer. Not a full fledged PDF to HTML converter. It's primary purpose is to allow a user to experience a document in its entirety in the browser. In a world of trade-offs, and priorities, the markup of the viewer itself is near the bottom.