Re: Relation Schemata vs. Relation Variables

Brian Selzer wrote:
>
> [...] A delete/insert combination would have to
> satisfy any delete constraints or insert constraints or both, but that
> doesn't mean that the update constraint couldn't still be circumvented.

Indeed, but the point is not whether they can be circumvented or not because in both aprroaches (your approach and a relational assignment approach) you can circumvent it. However under the relational assignment approach one combines inserts and deletes into a single assignment which would be checked as a whole, so there would at least be less opportunity to get around it. Of course the tighter control comes at a price: explicit modelling of an entity-pairing rule in the transition constraint. But given that making such a thing explicit is probably not unwise anyway and that it is more faithful to the "database is a set of propositions" paradigma, I'd say it is the better option.

• Jan Hidders