Many ways to skin a cat... SQL used to have an intriguing aura of exotic knowledge back in time when SQL was still a Language to highlight on a resume and to h

Many ways to skin a cat... SQL used to have an intriguing aura of exotic knowledge back in time when SQL was still a Language to highlight on a resume and to have an RDBMS server meant spending thousands. So Venn diagrams were just adding to such aura. But it's a visual aid and is usually supplemented with examples. If it doesn't ring a bell, some hands-on or detailed explanation helps crack it. Personally, I believe JOIN should first be touched upon when introducing normalization. It just makes sense: break tables apart then join them together. Here one could also grasp a utility of an index on join column. Then progress to more advanced use of joins - outer join (needs a concept of NULL), union (basically a synthetic set) etc. Finally, a cross join - as one of those cases that may lock your terminal if you don't know what you're doing. Of course learning Cartesian product concept here, if not yet. As anything in Computer, the hands-on aspect usually clears thing much better then science-first. The science would ease in much better once the utility is grasped. Teaching means helping someone to learn, not making them understand your way of teaching. So different aids may work for different audiences.