I am a big fan of this approach: Low-level, high-performance, immutable-by-default database engine as a building block, with bindings to multiple languages for

I am a big fan of this approach: Low-level, high-performance, immutable-by-default database engine as a building block, with bindings to multiple languages for easy application development. TrailDB (http://traildb.io http://traildb.io), which has many elements of event sourcing in it, follows this philosophy and it has proven to be pretty successful for its intended use cases. I was delighted to notice that PumpkinDB has an imperative query language inspired by Forth. We recently open-sourced a similarly imperative query language inspired by AWK, http://github.com/traildb/reel http://github.com/traildb/reel :) I will definitely follow PumpkinDB with great interest!