> I've been thinking this over and over, and it seems to me, that the way
> SETS in transactions SHOULD work is that they are all rolled back, period,
> whether the transaction successfully completes OR NOT.
Very interesting! This is a *consistant* use of SET which allows
transactions to be constructed as self-contained units without
side-effects on subsequent transactions. Beautifully powerful.
- Thomas
I've got some other thoughts on features for other aspects of schemas
and table and query properties, but this proposal for SET behavior
stands on its own so I'll hold off on muddying the discussion.