Yes, I think it's reasonable to say the NoSQL community drove the JSON columns in the relational databases. And while the JSON-ness of said columns is not necessarily the relevant bit to me, having the escape hatch of being able to just sort of splat out a "other" column has been something I've used several times, and it has relieved some non-trivial stress and strain in relational designs in my opinion. It's not even that hard to migrate something out of the JSON column if you want to raise it to a full column later, and JSON indices mean you may not even really need to.

I have a hard time justifying a document database in 2024 when I have 2024 Postgres; I don't know how large the window is where MongoDB will do something Postgres won't, but I know my systems are not in that window. (Assuming it even exists, which I wouldn't be surprised it doesn't, but for the sake of argument I will assume it does.) But a decade ago that would be a much harder decision if I truly had a document-based use case.