Re: Dawn doesn't like 1NF

"Dan" <guntermann_at_verizon.net> wrote in message news:3e68f717.0410141738.5a248ae2_at_posting.google.com...

> Open any basic discrete mathematics textbook of your choosing. Next,
> turn to the chapter on relations. I challenge you to produce an
> example problem and solution where relational operators operate over
> domains of set values, lists, or "non-simple types" with internals
> exposed. I think you will find that almost all cases specify
> relations over domains of elements - nothing more, nothing less.
> Interestingly, this corresponds well with predicate and propositional
> logic.

ISTM that a textbook is going to begin with simple examples. If you are explaining relaional operators, for the first time, it seems to me that all the operands are goping to be sets of simple things. The absence of relations on domains of relations in introductory material doesn't prove that it isn't useful or powerful.

>
> That is not to say that these elements might not be something more,
> such as a set, relation, list, or non-simple type; but even in
> mathematics, at least at the basic level, structure and manipulative
> aspects of mathematical relations are 1st order (the element/object
> level). It's more than Codd.
>
> That being said, I'm also saying that it is not impossible to do what
> you advocate. It's just that I haven't seen anything thus far that
> gives as much power for so little complexity. I'm willing to be
> intellectually honest here. If you show me mathematics where
> relational operators actually operate on elements of elements, I'll
> defer.
>

I concur. IMO, I've gotten more power, with less complexity out of the RDM, even through the stimgmatic lens of SQL, than out of any other single topic in IT. I think, but I'm not sure, that Date has the answer to the question you raise.
Why haven't I read it yet? Becasue I haven't needed to. Received on Fri Oct 15 2004 - 14:30:18 CEST