Hiya Nancy ---

Egads, I don't know if it's a mistake to apply OO to data modelling but I don't think so. A complete system will always have non-OO facets but that does not preclude using UML CRC, Use Case Approach, or what not from doing OOAD and OOSE (maybe not OOP). After all, in traditional OOAD, every part of a system is an actor, INCLUDING users. Now, I don't know about you, but I can't remember the last time someone tried to apply polymorphism or inheritance to me. :-)

There have been methods for mapping relational DBMS's to object models for as long as there have been object models. Most methods have been blessed by Jacobson, Booch, et al.

>With all the hullabalOOP about object oriented design and design patterns, I'm glad to see that discussions about relational database design aren't obselete in the press.
>
>My first Enterprise Development magazine (wheeee!) arrived this week and has the first in a 4-part series on data design called "Design From First Principles: Head off curruption at the pass with good data design" by Carmine Mangione. Hurrah!
>
>Occassionaly there are posts about data designs that don't follow relational theory. It's disturbing to me if we neglect in our discussions the foundation of any system (my soapbox, I admit).
>
>One interesting (to me) point was that it's a mistake to apply object-oriented softawre design techniques to data modeling. Do you agree? Disagree? Why?