> There are also a lot of queries that SQL is awful at, like most things involving graphs. That's why things like graph- and time-series databases have market n

> There are also a lot of queries that SQL is awful at, like most things involving graphs. That's why things like graph- and time-series databases have market niches. I don't really understand the graph database argument. "Graph" == "Joins" in my brain. I've also seen "Graph" used in place of "we aren't sure of our schema or generally what the business is even doing yet". Time series is definitely a fair argument depending on the specifics, but I've never needed Cassandra & friends to get the job done.