Relational query languages have been studied and used for more than 50 years, with SQL overwhelmingly dominating, in practice. Yet that dominance is now being challenged from several directions at once: higher-level abstractions such as ER-style and functional data models, unified application languages that integrate database querying with application logic, algebraic intermediate representa- tions that blur the boundary between logical and physical query specification, and (most recently) large language models (LLMs) that both generate and explain queries. At the center of this ecosystem is SQL, a powerful but irregular language whose syntactic variants and accumulated features complicate both human understanding and automated reasoning.
This tutorial offers a systematic way to think about relational query languages. Rather than beginning from formal definitions, we instead start from a shared workload of representative queries and examine how they are expressed in SQL, classical alternatives, and several recently proposed and earlier languages. From these ex- amples, we derive a common vocabulary of recurring design dimen- sions and patterns for relational languages. Participants will leave with clearer mental models for comparing existing and proposed languages, reasoning about usability for humans and AI systems, and articulating open problems in the design and evaluation of relational query languages.
If you notice anything incorrect in the slides, please let me know via email or this anonymous feedback form. Thanks a lot!