I'm not going to use a database that doesn't even have a proper date/time type and treats everything as a string.
ok, that seems like a bit of a moved goal post, but you do you.
is there any particular impact you see that having? Postgres has had it's share of data integrity issues due to things like locale differences between primary and replica, where strings would get sorted differently.
I'm not advocating for one over the other. I love both, but neither are perfect.
https://www.citusdata.com/blog/2020/12/12/dont-let-collation...
incredibly well tested for scenarios where data loss is likely, code is in the public domain, available on all major operating systems, included in language standard libraries... it's one of the most ubiquitous pieces of software out there.