Overall I like the ideas behind it and the problems with SQL that it solves, at least on a superficially observational level. It could use some real brain bendi

Overall I like the ideas behind it and the problems with SQL that it solves, at least on a superficially observational level. It could use some real brain bending examples for those of us that deal with difficult data sets. The tasks needed to actually take over where SQL has left off seem absolutely monumental though. I can't help but think it will never get there, just like every other attempted query language out there. In my opinion, there are two things that can make SQL 100% better, which wouldn't be a new language, but rather an update to the language standard: 1) algebraic data types, allowing us to get rid of terrible ternary null logic and more closely model real world data domains. 2) a really well thought out date/time API, along the lines of JSR310.