Re: Functional Dependencies > Uniqueness Constraints

Marshall wrote:
> It is also true
> that a system that supported FDs explicitly could express
> everything that one that had only uniqueness could; in other
> words, one concept is strictly more powerful than the other,
> and I propose there is a language design principle that
> says that in that situation, you should not include the weaker
> concept.

By that reasoning, FDs shouldn't be included if you have general database constraints. And I want those.

> One side comment: I note that the use case for a means of abstracting
> constraints seems to come up repeatedly.

Cf. C. J. Date: Database In Depth (2005), p. 174--175.

-- 
Jon