Inline semantics connected to specific real-world entities have unrealized potential; they should be separated from specific applications where possible; and they are probably best kept in the source.
This blog is about structured content, semantic metadata, and writing. I'm trying to bridge the worlds of data and prose. I see content people who understand data in general, but in their writing, think in "words on a page" and can't see it as entities and types of info. On the other hand, sometimes entities and relationships — bare facts — are all that data people who work with content see. They miss the context, nuance, and narrative that writing brings.
Opinions are of course my own!