Re: Database design, Keys and some other things

"David Cressey" <david.cressey_at_earthlink.net> wrote in message news:lGw_e.4989$vw6.1351_at_newsread1.news.atl.earthlink.net...
>
> "x" <x_at_not-exists.org> wrote in message
> news:dhbjde$tru$1_at_domitilla.aioe.org...
> > Then, after the translation of this model to the relational model, the
> date
> > become the attribute of some relation (not of some entity) and the
problem
> > "which is the entity of this attribute ?" is solved.

> The problem is not so much solved as obviated. It's not quite the same
> thing.

> We can carry out a whole class of relational transformations on the data
> that give the same result, and have the same validity, regardless of the
> ultimate semantics of the data.

> However if, at some future time, we get deeply involved in the meaning of
> the data, sooner or later the question that's obviated by the relational
> model will re-emerge. That's not to say that the RM doesn't achieve
> something of value by abstracting this question out of the logical model.
> It does.

When the question will re-emerge, we have tools to decide "which is the relation of this attribute". Received on Wed Sep 28 2005 - 16:06:16 CEST