You are not getting the point of my comment or the original article. It's clear to me that the kind of complexity being talked about in the article is cognitive complexity not computational complexity. It's not about choosing between an O(n) copying of data and O(1) in-place mutation; it's about choosing code that's easy to comprehend and maintain.

Immutable data structures are often implemented as persistant data structures, which have the traits of immutability but does not "copy everything on every change" https://en.wikipedia.org/wiki/Persistent_data_structure

Btrfs does stuff like instantaneous snapshots for a multi terabyte size filesystem. I have no idea how they do it, but apparently it’s 100% possible.

The optimizer removes that.

To see immutability on huge state in action see git.

That is essentially what must be going on under the hood to model mutable state. Every mutation made to a variable is a commit. The only commits that are saved are the ones that have existing references.

Now you're thinking it could be a big memory leak. But you pair this with reference counting you can eliminate the leak.

Essentially when the reference of the earliest commit goes out of scope, the next earliest commit is checked out as the initial root commit and the current commit is freed.

The memory model is then like a moving git window/linked list. Mutations are committed at the tail end while memory is freed at the head.