Thinking about a longer post on this subject. Basically: HTML has nothing to do with meaning. It’s just a set of signs that tells machines how to display or read documents. Meaning comes from human interpretation of the content. Machines cannot do meaning.

Also: there are two types of HTML signs. Those that tell machines what document role (paragraph, heading, list item etc.) a piece of content has (internal) and those that tell machines what role they serve in the world (time, small print, citation etc.) This is somewhat sketchy.

<time>1789</time> simply tells the machine that 1789 is a date. Not what it means.

And what this means? Simply that the pursuit of more meaningful markup is a waste of time. And that HTML can’t be judged or compared according to its meaningfulness.

The only considerations are how efficient, effective and manageable it is. How we can make meaningful interpretation easier.