[derefr]

Question: is storing full new row-tuple versions something fundamental to Postgres as a whole, or is it just a property of the default storage engine / “table access method”?

[dfox]

From the PQ protocol PoV the way how this works is pretty much irrelevant, but the actual implementation of PostgreSQL contains ridiculous amount of places that depend on the “backward” MVCC implementation of the tuple heaps.

[paulddraper]

You could store partial tuples without disrupting the interface.

Though full tuples is pretty fundamental to the underlying implementation....MVCC, VACUUM, etc. It'd be a massive change to say the least.