I can't imagine why would someone choose mongo in that case except to be slightly faster in the first few days of development, since you don't have to create a schema and update it. The problem is that you pay back that initial advantage with a lot of interests once your data becomes more complex and you need stuff like BI.
Anyway, even if your data isn't relational, JSON in postgres still works better than mongo.
The deepest question is whether it's useful to be able to model the world with first class structure in the form of embedded arrays and embedded sub structures that can be richly indexed. Does this involve certain forms of denormalization? Sure but do tables really represent how you think about the world?
That's pretty much what I asked. Still no answer.
It's not about "how you think about the world", but how you intend to use this data. Following my previous example, if you're building a blog you know you'll need to show a post along with its author, or that you'll need to filter posts by author, or maybe show only comments by verified users, etc.
Of course you can also structure your data this way on mongo, since over time they've added tools to support it, but it still feels like a feature tacked onto it rather than something "native" as it is with SQL.
Now, I understand what you're saying about my concept of relational being just a form of design and not inherent to the data itself. That's interesting indeed and I haven't thought about it this way before.
https://opentextbc.ca/dbdesign01/chapter/chapter-7-the-relat...
The problem is that nowadays universities are not teaching relational algebra as part of the software Engineering courses. And thus students dont grasp the reasons why relational databases are how they are.
"My data is relational" is generally a misunderstanding of the terms
Some new tech fades away. Some is here to stay.
SQL was the shiny thing near the start of my career. I pushed my company hard to abandon network databases in favor of SQL and that worked out quite well.
Same with object-oriented programming over classic C.
Some tech never lives up to its promise and becomes an albatross for its adopters.
There was a time when it was too early to responsibly incorporate LLM technology. There will soon be a time when the board will ask why you didn't. (Maybe for external use cases, certainly for internal ones).
One challenge for CTOs is making the call on new tech. When is boring tech the best thing for the company, when is it acceptable to introduce early stage tech, and when is new tech such an accelerator that you must adopt it?
As an example, mongodb inexplicably became a de facto DB for a bit between 2013 and 2016. It seemed like 1/3 of startup companies switched from MySQL or Postgres to Mongodb. What an absolute unmitigated shitshow clusterfuck moronic disaster