Re: Modelling Disjoint Subtypes

From: TroyK <cs_troyk_at_juno.com>
Date: 28 Mar 2007 16:15:42 -0700
Message-ID: <1175123742.694664.121030_at_p15g2000hsd.googlegroups.com>


On Mar 28, 2:08 pm, Dimitri Furman <dfur..._at_cloud99.net> wrote:
> On Mar 28 2007, 10:39 am, "V.J. Kumar" <vjkm..._at_gmail.com> wrote innews:Xns99016C61C8E14vdghher_at_194.177.96.26:
>
> > "David Portas" <REMOVE_BEFORE_REPLYING_dpor..._at_acm.org> wrote in
> >news:1175074877.506103.251950_at_e65g2000hsc.googlegroups.com:
>
> >> Yet another answer is addressed by Date in McGovern with their
> >> "Orthogonal Design" principle. Potential ambiguity is created if the
> >> same predicate is represented in multiple places in the schema
> >> because of multiple relvars with meanings that "overlap".
> >>http://www.dbdebunk.com/page/page/622331.htm
>
> > But we are discussing the disjoint case such that thanks to the
> > constraints the meanings do not overlap.
>
> But see Section 8 in Part 2 of the above article. It is not that just the
> meanings of two base relvars should not overlap - the meanings of two
> distinct projections of these relvars should not overlap either.
>
> --
> remove a 9 to reply by email

... two distinct projections from the set of nonloss decompositions, to be precise.

TroyK Received on Thu Mar 29 2007 - 01:15:42 CEST

Original text of this message