This paper discusses issues related to the integration of spatial operators into the new generation of SQL-like query languages. Starting from spatial data models, current spatial extensions of query languages are briefly reviewed and research directions are highlighted. A taxonomy of requirements to be satisfied by spatial operators is proposed with emphasis on users' needs and on the introduction of data uncertainty support. Further, spatial operators are classified into the three important categories of topological, projective, and metric operators and for each of them the state of the art is outlined.