Ok. Thank you for the clarification. I realize the whole point of the
discussion is that SQL does allow for bags in base, derived/intermediate
tables....It probably is a very good thing to understand the side affects
that occur with bag algebra since we probably use it more than we know,
though I personally try to avoid it. As others have stated, that is
probably why the Stanford trio address the issue explicitly.
"Jan Hidders" <jan.hidders_at_REMOVE.THIS.ua.ac.be> wrote in message
news:3e4d8dc3.0_at_news.ruca.ua.ac.be...
> D Guntermann wrote:
> >
> >"Jan Hidders" <hidders_at_REMOVE.THIS.ua.ac.be> wrote in message
> >news:3e4ad2b8.0_at_news.ruca.ua.ac.be...
> >> Lauri Pietarinen wrote:
> >> >
> >> >what is your take on Garcia-Molina, Ullman and Jennifer Widom
> >> >regarding their stand on duplicates?
> >> >
> >> >(see http://www.dbdebunk.com/cjddtdt.htm and cjddtdt2)
> >> >
> >[snipped for brevity]
> >>
> >> But how your algebra looks depends on how you
> >> answer question 2, because query optimization is the main raison d'etre
of
> >> the algebra, and there it is a completely different story. It can for
> >> example be more efficient to postpone duplicate elimination. If you
don't
> >> have a bag algebra you cannot express this in your algebra.
> >
> >I'll try not to sound too ignorant, but I'm afraid I will anyway, as I
> >haven't had time to read the link to Mr. Date's comments yet.
> >
> >Why does query optimization have to expose bag algebra?
>
> I don't think I said it has to.
>
> >Can't structures be manipulated at a lower internal representation that
is
> >independent of a logical data model and logical manipulative aspects
(e.g.
> >relational model with relational algebra)? Isn't this the point of
> >ANSI-SPARC architecture?
>
> Yes, it can, and yes it is.
>
> >I wonder what the ramifications are on data independence if relations,
> >whether base, derived, or intermediate, are allowed to incorporate bags
in
> >order to accomodate optimization.
>
> With "intermediate results" I meant the intermediate results of the steps
in
> the query evaluation plan. These are usually not seen by the user.
>
> -- Jan Hidders
>
>
Received on Sat Feb 15 2003 - 00:27:40 CST
