August 13, 2026
Null and void: comment war edition
Time to Move On: Querying Without Nulls and Bags
Database nerds say it’s time to ditch SQL’s messiest habits — commenters aren’t sold
TLDR: Researchers say database queries don’t need the messy habits that have confused programmers for decades, and they’re pitching a cleaner alternative. Commenters were split between intrigued and deeply cynical, with many saying the bigger obstacle isn’t the idea — it’s replacing SQL after 50 years of dominance.
A fresh academic paper just walked into one of tech’s oldest family fights and said: we can stop living like this. The authors argue that the database world should finally move past two infamous SQL quirks — duplicate-filled results and those mysterious blank-value placeholders that make queries weird, confusing, and error-prone. In plain English: they want database searches to behave more cleanly, more logically, and with fewer gotchas.
But the real fireworks were in the comments, where the mood swung hard between “finally!” and “here we go again.” One of the loudest reactions was pure battle-worn skepticism: after decades of “SQL killer” projects, some readers basically rolled their eyes and said, cute idea, but wake us up when anyone actually replaces the thing every company already uses. That vibe came through in bbkane’s instantly memeable opener: “oh great, another one.” Ouch.
Others got more specific. One commenter argued the paper skips a better fix for missing values entirely, saying the real answer is richer data types, not a grand rewrite. Another took the debate full galaxy-brain, begging everyone to stop talking about database rows like little boxes and start treating them as facts. And then came the receipts: one user dropped links to the authors’ proposed language, Rel, turning the thread from snark fest into semi-serious investigation.
So yes, the paper says the old mess is avoidable. The commenters say the bigger problem may be getting the world to care.
Key Points
- •The paper argues that SQL’s declarative design has been highly successful because it enables optimization and reduces programmer burden.
- •The authors say increasing query-language complexity requires revisiting early database language design decisions.
- •Based on their work on Rel, they claim a language built on fully normalized relations can be designed, implemented, and successfully deployed.
- •The article argues that bag semantics and null values are avoidable rather than necessary compromises.
- •The paper also states that a null-free language with set semantics offers additional opportunities beyond avoiding these features.