I was surprised that it didn't mention graph databases either, which is pertinent to the conversation around the evolution of SQL in the article, because querying graph databases requires re-thinking (or evolving) the query language. I'm excited by the work Neo4J and AgensGraph is doing. I haven't played around with AgensGraph, but I hope it takes off since it builds on Postgres

Exactly my thoughts. I haven't got a chance to dive into Neo4J yet, but this looks like a very exciting field that'll really improve visualization and data analysis.

I also believe this is the first kind of ACID NoSql database, which goes along Andy Pavlo's "All programs will execute using strongly consistent ACID transactions".

Definitely going to switch over from PostGreSQL at some point.

The Use Cases for GraphDB's are surprisingly small.