Re: RA with MV attributes

Marshall wrote:
> On Jan 17, 10:27 pm, "David" <davi..._at_iinet.net.au> wrote:
> > Marshall wrote:
> >
> > > > From Wikipedia : "... a binary operation on a set S is a binary
> > > > function from S and S to S"
> >
> > > Natural join is a binary operation closed over the set of all
> > > relations.
> > The very idea to put all relations of all "types" into a single set
> > suggests that the "type" has now become a property of the relation
> > value.
>
> Meh. This is the point in the conversation at which Han Solo
> pulls out his blaster and shoots the communication console.
> Without a formal definition of what it means for something
> be a property of something else, the argument is almost
> entirely a matter of word choice.

Ok, I could have said that better.

By definition, the *state* of a value equates to all the information you may gather about the value assuming you know its type. If you don't know its type then all bets are off.

If you put a value of type T1 into a set S of type set<T2> where T1 is a proper subset of type T2, you may have a problem because of loss of type information when looking at the elements of S because you only know that they are of type T2.

Given that we want to talk about operators (like natural join) on the set of all relations, it is necessary for the attribute information to be part of the relation value's state. Received on Thu Jan 18 2007 - 12:28:08 CET