Re: Domains, types, and application engineering ....

DBMS_Plumber wrote:
> Dan wrote:
> > Do domains in predicate calculus have behavior?
>
> Insightful question. And the answer is 'no'. A domain's behavior is
> governed by the relations in which it plays a part.
>
> There's a case for a kind of 'functional' relational-DBMS where
> domains have no operators at all. Instead the domain's 'behavior' is
> expressed with a collection of relation definitions.
>
> > I believe asking
> > questions and reasoning about properites of things and relationships
> > predates behavior "in databases or DBMSs" by several centuries, no?
>
> Oh yes.
>
> [ snip ]
>
> > So maybe the question
> > could be redirected as to what "ought" to be the behavior of domains
> > mentioned on a piece of paper be? Should they dance? Back-flips?
>
> If you can describe what a 'back-flip' is on a piece of paper, you
> can model it.
>
> > A tune would be nice.
>
> Glad you asked!
>
> *ahem*
>
> "I am the very model of an application developer,
> I know 'bout databases, both SQL-Server and Oracular,
> I can whip you up a GUI with speed that's quite spectacular,
> and I know the ins and outs of every pretty-picture data modeller!"

A very good tune indeed. Thanks for being a good sport.

• Dan