5 ms·
Most of the "nodejs generation" would take simple queries + aggregates (https://docs.mongodb.org/manual/aggregation/ https://docs.mongodb.org/manual/aggregation
by nnq 11y ago
Most of the "nodejs generation" would take simple queries + aggregates (https://docs.mongodb.org/manual/aggregation/ https://docs.mongodb.org/manual/aggregation/) + some flavor of map-reduce (https://docs.mongodb.org/manual/core/map-reduce/ https://docs.mongodb.org/manual/core/map-reduce/) over SQL anytime. Mostly because you only need the first for 90% of the cases, and for the rest, SQL is "brain hurting" and hard to optimize anyway. I also happen to agree that SQL should be finally killed and buried (and invent a more "in code" and closer to mathematical language notation for when you really need to do relational algebra, something like a "relational GraphQL maybe"), though most of the colleagues of my age would prefer to bury me for expressing this opinion :)
- jowiar 11y agoWorking at an organization that does both, the thing that jumps out to me is that SQL is a bit painful from the "application developer" side, but fantastic from the "ad-hoc data analysis" side. It's tough for me to say "finally killed and buried" when I have a couple dozen colleagues being wonderfully productive with SQL as one of their primary tools. Application developers and analysts don't need to use the same tool, though.