Re: By The Dawn's Normal Light

From: Dawn M. Wolthuis <dwolt_at_tincat-group.comREMOVE>
Date: Fri, 29 Oct 2004 15:36:17 -0500
Message-ID: <clu9kl$sgg$1_at_news.netins.net>


"Paul" <paul_at_test.com> wrote in message news:41825c43$0$4020$ed2619ec_at_ptn-nntp-reader01.plus.net...
> Dawn M. Wolthuis wrote:
> > Where ordered lists are not implemented, users place ordering attributes
> > into the data. The predicates stay the same whether a user adds in an
> > ordering attribute or the dbms software handles the ordering attribute
> > (along with inserts & deletes to an ordered list). There is no change
to
> > predicates in this case. --dawn
>
> OK, so you mean lists would still be relations underneath, but the DBMS
> would provide some "syntactic sugar" for manipulating them as lists?
>
> e.g. if you have a list done in a relation like this:
>
> (A, 1)
> (B, 2)
>
> Then you want to add a value inbetween:
>
> (A, 1)
> (X, 15)
> (B, 2)
>
> and
>
> (A, 1)
> (B, 125)
> (X, 15)
> (B, 2)
>
> then the additional operators will sort out the complicated queries
> (which are certainly possible in SQL) like "what is the 3rd member of
> the list?" or "return the list in order", "add a new value in 2nd
> place", etc. in a more user-friendly way?

Yes, just so. Cheers! --dawn Received on Fri Oct 29 2004 - 22:36:17 CEST

Original text of this message