Since I envision the scripts to be very small, I expect that updating them will not take long. The history of the updates will indeed be very interesting. In 10 years we can look at it and see how often each framework had breaking changes.

So instead of using debian:11-slim as I propose, one could use a Dockerfile made for Django. But that would help very little. Django even abandoned their official Dockerfile because it brings so little to the table.

In my opinion, using a higher level Dockerfile than the bare OS is a net negative. The developer won't know how much magic it hides. Even though it just hides a few lines of code. And being higher up in the stack also means stuff will break more often and the scripts need to be updated more often.