> Is SQL’s dominance a sign of maturity, or is something genuinely better on the horizon? Yes to the first, and must be yes to the second or we have failed!

> Is SQL’s dominance a sign of maturity, or is something genuinely better on the horizon? Yes to the first, and must be yes to the second or we have failed! Sql is to query languages like JS is for the web. That is not a good compliment. ---- Sql was from the start a poor implementation of the relational ideas, and have accumulate absurd levels of tech debt. Enjoy: https://www.postgresql.org/docs/current/sql-keywords-appendix.html https://www.postgresql.org/docs/current/sql-keywords-appendi... Is not that hard to come with a better alternative, but just "nicer SQL" is not good enough. We need a "Rust"-like moment for this. My bet (https://tablam.org https://tablam.org) is that we need algebraic data type support, real GROUP-by, a low-level language (where for example, you encode that "a = b" is in fact calling a BTree) and high-level porcelain. The low-level language (like webassembly+relational) will unlock a necessary steps where is possible to compile "normal" SQL, that is, the thing that allows typescript to be a viable alternative to JS. There are more details, but the above is the key. Like, you can create a proxy using the postgres wire protocol that take the SQL/Nicer lang then compiles then do the magic and is the enabler for adoption IMHO. BTW: This is what I wanna do, so happy to join forces!