I'm sure I'm not the first person to ask about this, but wouldn't it be cool if we could nominate two root nodes to merge/compare/de-duplicate.
The result could be a combined asset, with duplicates identified by selected attributes and handled gracefully (remove, rename, etc.), or to retain the structure, just identify and remove duplicates.
Complex and nasty, but the problem we're trying to resolve is historic, and messy user accounts, which is also complex and nasty.
I'd see such a tool being incredibly useful for removing redundant assets in document and image management.
Just daydreaming ;-)