Re: object algebra

Spot on, Marshall!

We approach data modeling from different perspectives, but the null issue crosses most (if not all) models and the 2VL where NULL = Null Set resolves issues that arise, whatever the model.

While SQL would be hard-pressed to recover from its 3VL mistakes, I havenn't looked at the most promising of the next industry standards (such as XQuery) to see how NULLS are handled. Does anyone happen to know? If they "got it right" (from my perspective) and are applying a 2VL instead of a 3 or more VL, that should lead to better industry effectiveness in this one category (null handling), I hope.

--dawn

"Marshall Spight" <mspight_at_dnai.com> wrote in message news:4Zp0c.149745$uV3.687129_at_attbi_s51...
> "Neo" <neo55592_at_hotmail.com> wrote in message
news:4b45d3ad.0402290711.6bdb201_at_posting.google.com...
> > > Anyway, that response shows the boy falls clearly (as do you) into
> > > the group of people who are not comfortable with nullary logic.
> >
> > Isn't "nullary logic" a contradiction?
>
> The term refers to logic on empty sets.
>
> For example, we can define an operator "sum" that takes a
> set of integers and returns the "grand total" found by adding
> all the integers together. (This is called a "fold" or a "reduction"
> on addition.) What is the sum of the set of integers {1, 2, 3}?
> It is 6. What is the sum of the empty set of integers? It is 0.
>
> Cool, eh?
>
> The fold of multiply over the empty set is 1.
> The fold of OR over the empty set of booleans is false.
> The fold of AND over the empty set of booleans is true.
>
> (The last one is why I take the position that Bob and Mary's
> are all the same color.)
>
>
> Marshall
>
>
Received on Sun Feb 29 2004 - 20:18:10 CET