One thing that I would find cool as a follow-up to this article would be something like a decision tree (I understand it would get very complicated, very fast, but it should be fun): I, as a developer, am using AWS/Azure and I need features X,Y,Z, tell me which database I should pick.
Probably even better if it were a model trained extensively on the documentation / features of each individual database and it would also give references to the answer (see how Bing does it).
There's a big difference between wire and syntax compatibility. This compatibility matrix does a nice job of showing the differences between vendors. But yeah, purpose-built databases have a place.
Many decision trees I've seen when NoSQL purpose-built were becoming popular are obsolete now that the main SQL databases have structures, APIs and optimisations for non-relational models: JSONB, Text Search, Timeseries, Documents... Today, the differentiator on architecture goals: converged database vs. multiple services
Another useful angle to look at the PostgreSQL-compatible database landscape would probably be fully open source (e.g. Apache licensed) vs. almost open-source (e.g. AGPL or commercial licenses that convert to Apache after a few years) vs. closed source. For a variety of reasons, many large organizations that seek to build an internal multi-cloud / hybrid cloud database-as-a-service platform strongly prefer fully open-source solutions.
6 comments
[ 5.1 ms ] story [ 28.4 ms ] threadProbably even better if it were a model trained extensively on the documentation / features of each individual database and it would also give references to the answer (see how Bing does it).