*edit:* ~~the more i think about it, it seems like it would be useful in the following situation:
I have a merkle-dag object for a folder on disk, `Qm...A`, I add a new file to that folder which produces `Qm...B`. 

In order to keep the structure, and contents of `Qm...A` without changing the hash, I can add a link from `Qm...A` -> `Qm...B`. This will then link the objects together, allowing me to browse through the files already in `Qm...A` before the link, but also browse `Qm...B`.

However, if this is the case, is there documentation on how exactly that works? Would `Qm...A` now be invalid because you added new content, and subsequently have an entirely new content hash?~~ -> the more i think about it, it seems like it would be useful in the following situation:
I have a merkle-dag object for a folder on disk, `Qm...A`, I add a new file to that folder which produces `Qm...B`. 

In order to keep the structure, and contents of `Qm...A` without changing the hash, I can add a link from `Qm...A` -> `Qm...B`. This will then link the objects together, allowing me to browse through the files already in `Qm...A` before the link, but also browse `Qm...B`.

However, if this is the case, is there documentation on how exactly that works? Wouldnt `Qm...A` now be invalid because you added new content, and subsequently have an entirely new content hash?