[9dev]

You are correct in your entire assessment, yet, you seem to underestimate the number of boring CRUD applications serviced by mediocre programmers. Limiting the number of technologies required to have at least a little knowledge in is a benefit, even if it hampers performance, because performance doesn’t matter for the vast majority of cases. Software engineers tend to overvalue that bit; the users of line-of-business apps don’t have a say anyway, but what matters is quick adaptability to changes in business logic.

So having an ORM in place that is tightly integrated in Python and Django lets even the junior developer fresh from the bootcamp make changes to an existing application.

It’s not a pretty story, but in my opinion the reason for staggering layers of complexity is the ability to move quickly even without experts on the team.

[thruway516]

I had to scroll all the way down to find this. As a CTO who personally never understood the point of ORMs the benefits become very quickly obvious when your organization starts to scale and the prospect of dozens or hundreds of developers of unknown quality hitting your production db with raw sql becomes objectively frightening. Of course there are ways to setup and administer your db to prevent the most obvious footguns, and it is still possible to write bad queries with an ORM, but having that extra layer with limitations gives some extra peace of mind.