Query optimization has traditionally relied on the optimize-then-execute paradigm, which, while efective in static settings, faces significant limitations in dynamic and complex environments 1 such as knowledge graphs on the web. In this keynote, I will explore how early database adaptive techniques2 provided the first steps toward online learning query optimization over knowledge graphs3, allowing systems to adjust during execution. I will present results on when adaptivity enhances continuous performance4 in knowledge graphs and when it falls short.