Re: Requirements for update languages?

From: Paul Vernon <paul.vernon_at_ukk.ibmm.comm>
Date: Sun, 17 Nov 2002 22:46:02 -0000
Message-ID: <ar9690$ffq$1_at_sp15at20.hursley.ibm.com>


"David Cressey" <david_at_dcressey.com> wrote in message news:euuB9.267$0I3.27200_at_petpeeve.ziplink.net...
> > Or, to put it in more usual terminology, 'additional database executed
> > updates' should be seen as a *part of * the users transaction and be
> *visible*
> > as part those transactions. As an example, see my post some time ago on
> how
> > referential actions (like 'on delete cascade') should be seen as part of
> the
> > macro expansion of statements such as DELETE.
>
> Are there any major DB systems that don't work this way?

None of them make cascaded actions directly visible to the users (although I guess users could read the calalog and parse their own statements, adding in any cascaded actions themselves, but that is hardly what I'm trying to get at)

> Incidentally, I think it's useful for the DBMS to view the application tier
> as a "user of the database". Depending on how slavish the application is to
> the user's desire, the application may behave like a more responsible
> "database user" than the "application user" would have. Still, it may be
> desirable for the database to protect itself from the application,
> depending on who is accountable for each of them.

I would go further and say that the DBMS should protect itself from *all* applications.

Regards
Paul Vernon
Business Intelligence, IBM Global Services Received on Sun Nov 17 2002 - 23:46:02 CET

Original text of this message