Re: By The Dawn's Normal Light

From: Alfredo Novoa <>
Date: 27 Oct 2004 03:46:53 -0700
Message-ID: <>

"Dawn M. Wolthuis" <dwolt_at_tincat-group.comREMOVE> wrote in message news:<clmdde$b60$>...

> I re-read Dates latest Intro... and it appears that the def of BCNF does not
> require 1NF.

1NF is assumed because all relations are in 1NF by definition.

> So, I think we could ditch 1, 2 and 3NF and just use BCNF as a
> starting point.

BCNF is a practical starting point because it is very easy to check, but we should work with 5NF designs, or 6NF if we are working with temporal databases.

Regards Received on Wed Oct 27 2004 - 12:46:53 CEST

Original text of this message