While a many-to-many DB relationship can in theory model a 1-to-many relationship via constraints and/or domain logic coding, it seems kind of anti-YAGNI to me.

While a many-to-many DB relationship can in theory model a 1-to-many relationship via constraints and/or domain logic coding, it seems kind of anti-YAGNI to me. Being that ANY 1-to-many relationship can potentially change into being many-to-many, which ones do you "pre many" and which ones are left 1-to-many? My crystal ball is not that powerful. If it were, I'd be competing with Warren Buffett instead of puzzling over database design.