TECH Signal 506
Paper argues SQL nulls and bags are avoidable, describes null-free set-semantics approach
↗
Illustration only Photo by Christian Perner on Unsplash
A new paper by the designers of the Rel language argues that the standard justifications for SQL's nulls and bag semantics dissolve under scrutiny, and that a null-free query language based on fully normalized relations is both feasible and beneficial.
Null handling is a persistent source of bugs and semantic ambiguity in database applications. If the argument holds, it could influence how future query languages are designed, pushing toward cleaner set-based semantics rather than SQL's multiset-and-null model.
Written by elseif from the cluster below · every claim links back to a sourceThe three things worth knowing
The authors argue that common justifications for SQL's bags and nulls do not survive closer examination.
The Rel language demonstrates that a null-free, set-semantics query language can be designed, implemented, and deployed successfully.
A null-free language with set semantics opens optimization and correctness opportunities unavailable under SQL's bag-and-null model.
THE CLUSTER