Jonathan Simms (ccoomm) writes:
A question that does not have a clear-cut answer. For the case as presented
I might very well go for the suggested outline. But to always do like this
would lead to a mess. And many database architects would probably perfer
to duplicate the columns in each table.
When it comes to modelling data, the object-oriented paradigm is probably
superiour to the relational model. But modelling is not enough, and the
relational model has proven to work very well when it comes deal with
gigabytes and terabytes of data.
What is important is that while O-O ...