Re: The Relational Model & Queries That Naturally Return Duplicate Rows

On Oct 11, 11:39 am, Erwin <e.sm..._at_myonline.be> wrote:
> On 11 okt, 17:27, Tegiri Nenashi <tegirinena..._at_gmail.com> wrote:
>
> > On Oct 10, 2:35 am, Erwin <e.sm..._at_myonline.be> wrote:
>
> > > ...There,
> > > "bag algebra" is investigated (and its inferiority to relational
> > > algebra exposed).
>
> > It is not that simple. Check up Green et.al. semiringshttp://www.cis.upenn.edu/~val/EDBTkeynoteLausanne.pdf
>
> Nothing about algebra is ever as simple as two lines in a response in
> a forum, or as a single chapter in a book.
>
> Does the provenance stuff convey information that cannot be
> represented relationally ?  Does the provenance stuff convey
> information that is better represented as "K-relation", rather than
> "pure-relationally" ?  The provenance stuff seems to assume that all
> relation values derive from database relvars (but maybe I
> misinterpreted this), and hence it doesn't seem to deal with things
> like TTM relation value selectors.

No, provenance is a buzzword that everybody in academia is tuned in, please ignore it. They suggested a model -- extending relations with semiring properties -- generalizing nicely 4 or so other well established models; bags/multisets among them. I used to think that nothing really happened in the last decade in database research, and Green et.al. work shattered this opinion. Received on Mon Oct 11 2010 - 21:58:20 CEST