[Homunculiheaded]

I've tried many times over the years to get excited about the semantic web, but invariably always arrive at the same conclusion as you. I also think that semweb really suffers from over engineering a solution in search of a problem. If you have a problem that it seems like the semantic technologies could be helpful for, you can probably solve it faster using logic programming and automated reasoning techniques from the 90s (ask the fast majority of semantic web enthusiasts how RDF triples can be represented as prolog clauses and you'll likely be treated as though you asked how a raven is like a writing desk). Worse yet though are tools like OWL, which is over-engineered to the point where out-of-the-box it describes logic in a way that is computationally intractable. If you where designing semantic technologies to solve a real problem you would never arrive at something like OWL.

I actually think something like semantic technologies could be more useful than the tools that drive watson and google for smaller data sets, where machine/statistical learning are less useful, but even then they are an over engineered solution.