I’m focusing on my “writes” right now using CouchDB to record my event history, which syncs to an Elm SPA, but I intend to project historical state into Neo4j for the “reads” to do deeper data analysis.
I’ve done experiments that use the CouchDB change feed to push changes into Neo4j in Elixir. Also with historical versioning of data in Neo4j. And using GRANDstack mutations to enforce a consistent schema. But my priority right now is on the CouchDB side.
But I do think that Neo4j has great potential in the arena of CQRS and event sourcing.