Re: Boyce Codd Normal Form

 James Martin wrote a series of books on database design (and other topics). He was the guru everybody bowed to during the 70's and 80's. He caused the inking of several books that covered normalization (up to 5th normal as I remember) and were for the most part quite readable. The one you are looking for used Snoopy and the Happy Dog Farm for the description of why one may want to go beyond 3rd normal (TNF). If I can find the reference I'll post it here, otherwise time for you to hit the ol' library (that's a pre-internet repository of info).

If you use Myers Method to develop the data model, you don't have to worry about normalization until you are forced into a relational design. The relational entity/relationship diagram (ERD) then becomes a physical description of the implemented design. Basic flaw here. If Codd (and Martin) had their way, you could never store a dozen eggs in one carton...let alone in one refrigerator.

Meanwhile have you checked out the following? http://infolab.kub.nl/edu/databases/sheets/1998-1999/normalization/

Most authors chicken out beyond 3NF/TNF:

http://www.farmingdale.edu/~porciem/bcs260/bcs260d10.html

Ahhh, found the puppy farm example: Might save you a trip to the smelly book house.

http://www.gslis.utexas.edu/~l384k11w/normover.html

Good luck with all of those vacant stares you will receive from the students.

pete wrote:

> I am due to start teaching Normalisation to my students, and was wondering
> whether anyone can recommend an idiot's guide to BCNF for me! I am fine up
> until 3rd NF, then confusion sets in! If possible I would like to see an
> example of data normalised to BCNF.
> I would really like a simple explanation as any reference to BCNF which I
> have so far found on the Net is couched in technical terms rather than words
> of less than two syllables (what is a determinant?! what is the difference
> between a primary key and a candidate key and a super key?!)
> Any help gratefully received.
> Many thanks
> Debbie
Received on Sat Sep 28 2002 - 02:17:39 CEST