Re: Constraints and Functional Dependencies

Marshall wrote:
...
> Also, I'm proposing a system in which there's just one
> kind of costraint, and that requirement seems awfully arbitrary
> in that light.

I assume you're referring to being able to express any constraint in FOL, even if FOL is not the implementation language. I suppose you could call that arbitrary, but to some people, male and female are arbitrary and there's nothing the rest of us can do about that. If some of us find Codd's choice pleasing, surely the human motive trumps any arbitrariness an outsider could label it with and we all find the landing spot we like best, even if it doesn't turn out well, such as for Fabian P. In the case at hand, I would say the repeatability of Codd's logical system transcends any criticism that it is arbitrary. It goes well with digital computers, so what's the problem? (After all, there are only a handful of new-agers here, even if their posts make them seem more numerous than they are.)

p Received on Sun Feb 25 2007 - 02:56:25 CET