The article dismisses XHTML with little explanation: "all the ideology surrounding XML that never came to fruition".
AFAIK the goal of XHTML was extensibility and interoperability with other data formats, by allowing arbitrary XML elements from other XML namespaces to be embedded. It seems to me that this solves exactly the same problem (as microdata) and more in a much cleaner way. Why invent yet another format?
Neither solves 'the problem', if one describes it as 'make all/most web content semantically richer'. Both just provide a mechanism, but nobody is going to invest in tagging their data without a good use case. so, it is sort of a chicken and egg problem.
Worse, marking up my data in the database from which I serve my web pages may benefit me, but currently, marking up my data in my web pages just costs bandwidth.
AFAIK the goal of XHTML was extensibility and interoperability with other data formats, by allowing arbitrary XML elements from other XML namespaces to be embedded. It seems to me that this solves exactly the same problem (as microdata) and more in a much cleaner way. Why invent yet another format?