+1. Although it's never been about the data, it's always been about relationships between data, but the market doesn't differentiate so neither should we (when selling the idea)
I take a long multi-decade view on technology, so I believe SemWeb is worth investing in. As an industry we have cronic myopia. Emacs, Lisp, the Web and SemWeb, these bets gradually pay off.
If you don't know what your data means, it's virtually worthless.
SemWeb will likely lose to deep learning in the short term, but deep learning is a trick, and limited in application. Ultimately maths wins.