Re: Pizza Example

"Anthony W. Youngman" <wol_at_thewolery.demon.co.uk> wrote in message news:<L+gV0RG4i0cAFwZf_at_thewolery.demon.co.uk>...
> In message <c0e3f26e.0404060027.55c61183_at_posting.google.com>, Tony
> <andrewst_at_onetel.net.uk> writes
> >> Anyway, who says the Pick approach is unstructured? If you go back to
> >> Dawn's original output, surely that is very similar to a relational
> >> view, ie it's structured in a manner you understand?
> >>
> >> More to the point, it's structured in a manner the database can
> >> understand, which isn't the case if the information is scattered across
> >> multiple tables :-)
> >
> >What makes you say that? The DBMS can 100% understand the data
> >"scattered" across multiple tables with appropriate constraints
> >defined. 100%.
>
> Does the database understand that all these tables refer to the same
> real-world object?

The DBMS doesn't "understand" anything in the human sense. Like all computer software it "twiddles symbols" according to rules without any deeper understanding of why it does so. But does it "know" that the toppings are part of the pizza (if that's the example we were on)? Yes, 100%. The database constraints tell it that the pizza has the toppings and the toppings are on the pizza. No information is lost. Ask the DBMS to tell you everything about a particular pizza and it can. It does not say "hey, well we have 100 pizzas and 350 toppings; who knows what goes where?" It "understands" 100% what belongs where.

Whether that means the pizza and its toppings form a single real-world object is another matter. If using a MV attribute for toppings means that the pizza and its toppings constitute a single real-world object, does that mean that a MV attribute "children" in an employee record means that the employee and all his children constitute a single real-world object also? Received on Wed Apr 07 2004 - 13:40:16 CEST