Re: Sixth normal form

"Jan Hidders" <hidders_at_gmail.com> wrote in message news:1187699487.121719.181130_at_g4g2000hsf.googlegroups.com...
> On 21 aug, 05:52, "Brian Selzer" <br..._at_selzer-software.com> wrote:
>>
>> My argument is that if there is a functional relationship from one set of
>> attributes to another in a less normalized schema, then there should
>> still
>> be a functional relationship after decomposition.
>
> This cannot be your argument because it is only a reformulation of
> your claim. You still have yet to supply any form of argumentation
> that justifies that claim.
>
> You seemed to come close to such a thing by invoking the domain
> closure property and the principle that you don't want to change the
> upper and lowerbounds of the relationships between the objects in the
> domains, but now you start special pleading because you only want to
> apply those principles for the lower and upper bounds in the direction
> of the FDs, i.e., you are only applying the principles there where you
> need them to support your claim and ignore them where they contradict
> your claim. This way you are of course proving precisely nothing.
>
> Since I don't see any progress, and suspect that there will none in
> the near future, I'm signing off from this discussion.
>

Thank you for your attention. I enjoyed the discussion and learned several things I hadn't put together before. I think that the reason I'm having trouble convincing you is because I really didn't clearly state my claim to begin with. I never really came up with a clear definition of what it means for a schema to have /at least as much/ information, nor did I provide any proof that if a relation is in 5NF, then the relationships between the dependent attributes are due to the fact that the relation is in 5NF, or that the presence of a pathological relationship between the dependent attributes indicates that the relation isn't in 5NF. One interesting consequence of the discussion that I think it should be possible to prove is that every update anomaly can be linked to the presence of a pathological many to many relationship between attributes. I'll leave that until a later date. Again, thank you.

> -- Jan Hidders
>
Received on Tue Aug 21 2007 - 17:46:13 CEST