Property graphs have become the dominant paradigm for modeling connected data, yet they lack the formal semantic rigor that RDF/OWL brought to knowledge representation. While RDF graphs are constrained by their triple-based structure, property graphs offer richer modeling capabilities through typed edges with properties. However, this expressiveness comes at a cost: the absence of standardized formal semantics and constraint languages that can reason about graph structure and properties simultaneously.