And just to make sure we are on the same page here: it's not academics' job to build usable products. We will continue working on things that are novel from the academic standpoint; if people like you dismiss LD/SemWeb, those novel things will have "a terrible track record of solving real-world problems". I hope this does not come across as too personal.
And using semantic web for that is just as bad. A basic json API would be much more stable than parsing a document with navigation and similar just to get that data.