Re: Database design

From: David Cressey <>
Date: Sat, 25 Feb 2006 13:40:26 GMT
Message-ID: <e7ZLf.3175$Gw2.710_at_trndny03>

"Mark Johnson" <> wrote in

> I wasn't sure. The only way would be to go to the local library and
> copy the actual magazine article or any follow-up paper, keeping in
> mind any change in Codd's opinions (as we all change our opinions). I
> do know it's often represented, this way. But I would also suspect
> that the same 12 Rules would read a bit differently if the source were
> Date, Darwen, etc.

Been there, done that (except for the copying part). It's easier to rely on the following link:

Given the source, I tend to believe that the copy is faithful.

You make two good points. First about the chnages in Codd's thinking as time went on. I would say the 12 rules, themselves, were a response to the onslaught of soi disant commercial relational databases in the early 1980s. Some of these were so far from following the RM that calling them "relational" was simply crass marketing. Others were pretty good, and workable approximations.

Some of the changes in Codd's thinking were motivated by his own research, rather than reacting to others.

Second, that the rules would be phrased differently by Date and Darwen. I believe you are correct. I think it's extremely unfortunate that Date chose to meddle with Codd's phrasing (e.g. redefining 1NF) rather than come up with alternative phrasing to reflect Date's discoveries of what he sees as Codd's errors.

Ideas are worth debating (sometimes). Terminology is not. Received on Sat Feb 25 2006 - 14:40:26 CET

Original text of this message