Re: Is this bad design ?

"Dawn M. Wolthuis" <dwolt_at_tincat-group.com> wrote:
>This data could be modeled as:
>Mother Table
>mother_id
>still_valid boolean
>etc_fields
>child* (* meaning can be multivalued)
> more_fields

Fabian has already suggested a solution that is very reasonable.

There are many other much more flexible such as modeling patients, using generalization/specialization for males/females and association to say that other patients are child of a specific "valid mother" patient.

>If your database permits such modeling (no matter how it actually implements
>this), then the desired constraint is built in, including knowing whether
>the mother is still_valid and also getting deleted if the mother is deleted.

This comment is completely irrelevant since we are not talking about anything that could be used to model it in this very questionable way.

Unless you want to tell everyone that "no matter how it actually implements this"
can also be something like an XML document in a text field with very clumsy way to query and update it. (ie. an absurd idea)

Ben was trying to make a point about RDBMS constraints, your suggestions to use other tools don't help the discussion, bore the audience and make Bob nervous.

>Hand data with such relationships to any secretary in the 1950's and they
>will know how to "model" it and store the data for easy retrieval in their
>paper system. Thank goodness they were not constrained by some desire to
>"normalize" the data, eh?

Now I understand where all the monster-procedural-driven-hierarchic-stores comes from.... thank God we live in year 2004 and the relational model is widely used (although in many cases without understanding the basis behind it).

>smiles. --dawn

Frankly, I'm pretty sure I'm not alone, as a long-time lurker, starting to become
very bored with you constantly flooding the newsgroup with amazingly long posts full of very personal, and much questionable, claims about theories without even being able to aknowledge very basic things.

Mr. Tuttle Received on Wed Mar 10 2004 - 12:08:30 CET