Re: Is relational theory irrelevant?

"Paul Vernon" <paul.vernon_at_ukk.ibmm.comm> wrote in message news:bpe04s$oio$1_at_gazette.almaden.ibm.com...
> "Mikito Harakiri" <mikharakiri_at_iahu.com> wrote in message
> news:FDuub.17$wC5.84_at_news.oracle.com...
> > "Paul Vernon" <paul.vernon_at_ukk.ibmm.comm> wrote in message
> > news:bpdpuh$1v88$1_at_gazette.almaden.ibm.com...
> > > > Next, what is the fact that you describe in your query? That's a set
> of
> > > > folks that you want to target with your mail campain, right?
> Therefore,
> > > your
> > > > query fetches the result set of target emails, and your client
program
> > > opens
> > > > the cursor and iterates through it and invokes sendmail at each
step.
> > >
> > > > Where is the problem?
> > >
> > > Sniped in and bold added:
> > >
> > > your *client program* opens the *cursor* and *iterates*
> >
> > I don't understand your comment. Can my client program do better than
> that?
> > AFAIK, there is no set interface that database can connect to the
outside
> > procedural world. Or, by emphasising *client program* you suggest that
we
> > don't need it?
> >
> > Sending email is an action outside the database environment, so that we
> have
> > to leave DBMS boundary. I know, mindless throwing stuff into the
database
> is
> > pretty popular today, but there is no advantage triggering this
real-world
> > action from the inside of the database.
>
> Wrong. It is information, ergo it should be in the database.

Sending an email is not information--it is an activity. The email is information. The recipient is information. Sending is not information. Received on Tue Nov 18 2003 - 22:41:27 CET