Re: RA with MV attributes

paul c wrote:
> David wrote:
> > paul c wrote:
> > ...
> >>I do think that once one has a definition it is important to proceed to
> >>positioning it in the scheme of things, for example is this a physical
> >>approach for storing two relations in one or is it an attempt to promote
> >>three-valued logic or is it something else?
> >
> >
> > I'm interested in both its logical and physical aspects. It seems
> > best initially to focus mainly on the logical.
> > ...
>
> I could entertain this if the motive were to formalize a physical
> storage scheme. But I don't see anything logical about presenting to a
> user the inference that fred and bill own an empty set of cars that are
> green

That's an incorrect inference. That's not what a tuple means. If you have the tuple

    Names = {Fred, Bill}, Cars = {}, CarColour = {green}

Then you are meant to read out propositions by taking the cross product of the sets. The cross product is of course empty.

> especially if the user could then project away the Cars attribute
> and conclude that they do own a green car or cars. (That seems like
> using the empty set to achieve Codd's connection trap.)

You must perform selection (to ensure there really is a car) before the projection. This is logical and AFAIK everything is self-consistent. There are some "rules to the game". Don't say it's illogical because you're not playing the game.

If you say you don't find it particularly intuitive or natural I somewhat agree with you. Received on Fri Jan 19 2007 - 18:12:53 CST