Re: help with normalization
From: mAsterdam <mAsterdam_at_vrijdag.org>
Date: Sat, 03 Sep 2005 19:51:16 +0200
Message-ID: <4319e229$0$11073$e4fe514c_at_news.xs4all.nl>
>
> Foreign keys have very little to do with normalisation. What you first
> and foremost need to do is to determine the functional dependencies.
> Without those, I don't see how it is possible to normalise in any
> systematic and formal manner.
Date: Sat, 03 Sep 2005 19:51:16 +0200
Message-ID: <4319e229$0$11073$e4fe514c_at_news.xs4all.nl>
Jon Heggland wrote:
> mAsterdam says...
>>Tonkuma wrote: >> >>>I think that you shoud consider to add forein keys and constraints, >>>before consider normalization. >> >>Some constraints, yes, but not foreign keys. >>Foreign keys are a result of the normalization steps.
>
> Foreign keys have very little to do with normalisation. What you first
> and foremost need to do is to determine the functional dependencies.
> Without those, I don't see how it is possible to normalise in any
> systematic and formal manner.
Except (obviously) for the first sentence I agree. First you should get the meaning (including the FDs) right.
A quote from Date's intro (8th ed, p 356):
"FDs are a semantic notion ... (in fact, they are part of the relvar predicate). Recognizing the FDs is part of the process of understanding what the data /means/ ... " Received on Sat Sep 03 2005 - 19:51:16 CEST