[BiO BB] RE: Normalization WAS: database design question

[Michael Gruenberger]

I don't quite understand your reasoning: You claimed that normalization
has no practical use and that is because it's hard to understand?!

Normalization is not meant to simplify database design, but rather to
formalise it and to give guidelines as to what works and what doesn't.

I completely agree with Jordi, normalization just becomes part of your
'inner feeling' of what works and what doesn't after you've done it a
couple of times. Of course it isn't applicaple in all cases, but it
surely is better than no guidelines at all, especially for people who
have never designed a database before and who are looking for some
guidelines to point them in the right direction. And there are ways to
explain normalization in simple terms and with good examples.

So if you are saying normalization hasn't got any practical use, what
would you suggest to a newcomer to database design?

On Tue, 2004-03-02 at 15:29, Svensson, B.A.T. (HKG) wrote:
> 2nd order tensors calcus in fluid mechanics is also "easy"
> once you get used with the concept. However the difference
> we talking about here is that tensor calcus is simplifying
> the task of understanding the problem, while normalization
> theory does not, rather the opposite.
> 
> Any student of computer science who has been forced to do
> formal deduction with horn clauses know that this is the
> most rigorous and silly way to conclude the most trivial
> and obvious facts, and as a matter of fact normalization
> theory is based on this kind of logical reasoning.
> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://www.bioinformatics.org/pipermail/bbb/attachments/20040302/ed3913d1/attachment.sig>