How are you measuring memory storage size? Are you sure you are looking as resident memory size and not just the virtual memory size?

Actual storage of the structures should be nothing. Interfaces are “fat pointers” but that should really just be an extra word which node would have at least that if not more.

My guess is that if you are looking at virtual memory that more memory/garbage is produced in PARSING and not storing and that the virtual memory size allocated is high even after garbage collection but RSS should be different.

Call me when they manage to get it up streamed into the kernel

I agree. The only one that was close for me is Just. It is just Makefiles but without all the baggage.

No harm intended but if you are reaching a conclusion that such a feature is not needed without going into more detail than “harm” can go the other way as well.

Sorry I didn’t link the discussion as I was on my phone and no partonization was intended.

Article is missing a lot of keys points. With iterator you can chain them together to provide even higher level abstractions.

There are plenty of containers (new maps with different algorithms like a BtreeMap, linked lists, etc) that now with generics could also use a generic way of iterating over them.

There was a pre-proposal discussion that went into a lot of detail of what is possible that wasn’t intended the release notes. I highly suggest the writer of this article dig much deeper into more benefits of iterators than the two trivial options that were included in the experiment description.