Re: A Normalization Question

> ...the whole purpose of normalization is to determine which things
> relate to which other things.

This is an incorrect understanding of normalization.

> You are overgeneralizing to the nth degree.

No, you/RM are undergeneralizing to the 5th degree.

> In normalization, one does not consider the values of the attributes...

The general form of normalization applies to all data models. Because RM is a limited data model, it is either impossible or impractical to normalize some types of data (ie values of attributes) while other data models (ie TDM) can.

> > If one chooses to ignore some things (ie 'brown', 'brown', 'brown'),
> > then one has a limited understanding/implementation of normalization.
>
> Please read SOMETHING about functional dependencies
> before you post this nonsense again.

RM incorrectly senses that 'brown', 'brown', 'brown' are not redunant, no matter how many times one rereads it, because it is a limited data model.  

> We are talking about relational databases.

I am talking about representing things in general and RM is only one limited method of doing so.

> Strings don't have attributes except ...

Sure they can as I have demonstrated earlier. It is not upto a data model to decide which data is "part of the data dictionary" or "in the normal business world". If it does, the data model is limited.

> _Entities_ have attributes.
> "Brown" is not an entity.

Because 'brown' can have attributes (ie language, typeset, font, size, color, blinking, etc) it can by your definition be an entity.

> Your view of this entire topic is completely twisted.

No, your/RM's view of this entire topic is completely limited. Received on Wed Jul 07 2004 - 03:39:00 CEST