How important is ontology engineering today?

I feel its current approaches (owl rdf etc...) are stuck somehow and there is something like a ontology Winter. Its not dead but the solutions out there do not scale well. I believe something like the gellish language approach must be developed towards a open standard and there needs to exist something like a decentralized registry of ontologies and dictionaries. Maybe like DNS.

https://en.wikipedia.org/wiki/Ontology_engineering

what exactly you refer too? Very unspecific question.