Knowledge engineering is complex. Barriers to entry for building knowledge graphs are daunting. There are many standards required to master the full stack, among others RDF, SPARQL, RDFS, OWL, SHACL, JSON-LD, etc, each being specified through long boring documents. But each of these standards is necessary, and those who try to circumvent them end-up reinventing the wheel. If we add to that the many data engineering and data management skills required, we understand why knowledge graphs have found adoption mainly in large corporations with resources for dedicated teams.
In this talk we look into providing a leaner approach to knowledge engineering. We question the possibility of abstracting the underlying languages and protocols. We argue that ontology engineering, a practice which connects technical knowledge with business knowledge, should be put directly in the hands of the business. Similarly, data integration into a knowledge graph can be realized by non technical experts, by leveraging automation and low-code abstractions.