SQL is a pretty warty implementation of relational databases (with non-composability being its primary sin IMO), but we're stuck with it at this point. A new qu

SQL is a pretty warty implementation of relational databases (with non-composability being its primary sin IMO), but we're stuck with it at this point. A new querying DSL that fixes all of SQL's flaws is only half the story, getting enough programmers on the planet to buy into it is another half. To do that you'd need this new piece of software to at least be as fast and as battle tested as existing SQL databases. Even the new generation of massively-scalable relational databases stick with some form of SQL instead of inventing a new DSL because of the sheer momentum behind this sorry syntax.