In this paper, we propose a simple and flexible execution model that (i) supports a wide spectrum of alternative optimization and execution strategies and their mixtures, (ii) provides for dynamic reconfiguration when adding/deleting queries and changing optimization goals, (iii) optimizes response time in idle-waiting prone operators, such as union, joins, and operators used in time series and temporal sequence queries. Thus, we introduce a flexible and concrete model of execution semantics for continuous DSMS queries, and demonstrate its many applications. Our tuple-oriented vetails and complements the abstract set-oriented semantics of current DSMS constructs and operators, which are often based on relational algebra and SQL enhanced with windows.