Notice that you can add all the content, with whatever non-unique attribute inside the Tags type; plus the link to access the parent's parent content and so on

Notice that you can add all the content, with whatever non-unique attribute inside the Tags type; plus the link to access the parent's parent content and so on... by reproducing the whole tree structure backwards Finally consider that getPath is not doing a lookup by id but it is really directly accessing the parent node content and up to the root, immediately. What you can't do this way is that you can't go through the other children of the parent (and all the rest of the tree structure that is not straight up in the hierarchy), but that's another story: it would need another use case to be introduced and discussed - in the first place