If you are wondering what the future of #databases looks like, you might be interested in this article https://doi.org/10.14778/3603581.3603604

The authors argue that despite increasing specialization, databases share the same components. By relying on standards, you can develop a composable system with:
- your preferred language interface (#SQL, #Ibis, #dplyr, ...). The query is converted into an intermediate representation (like #substrait)
- a query optimizer
- an execution engine
- an execution runtime

I wrote a short post on (experimental) usage of #Substrait to communicate logical plans between #Ibis and #DuckDB:
https://ibis-project.org/blog/ibis_substrait_to_duckdb/