Re: A Normalization Question

On Mon, 12 Jul 2004 11:26:08 -0700, Neo wrote:

> Jan Hidders wrote:

>> Neo wrote:
>> >>>>You have failed to show that they represent the *same* fact.
>> >>>
>> >>>You have *failed* to define what is a fact?
>> >>
>> >>I did that elsewhere.
>> > 
>> > My apologies. Could you please refer to that post or restate?
>> 
>> A fact is a true proposition.

It's precise and correct and only weak for those who do not understand what a proposition is or when it is true. If you do not know that then you should stay away from database theory as these are fundamental notions for understanding it.

> Could you do better in context to dbs? What in
> a db is not a fact?

Nothing, of course. Some things in a database represent facts, but they are not facts themselves. How the database does that is up to the user to define and it can be formalized as a function that maps each possible database instance to a set of facts. From this follows straightforwardly the definition of redundancy that I already gave to you: if I only give you a part of the data structure and for all such functions it holds that for all instances that match the given part the associated set of propositions is the same, then the missing part is redundant. It is such redundancy that normalization attempts to remove.

• Jan Hidders