Re: Modeling Question

Hi Steve,

It is a good design principle to store the same information only once in your database. Your design has this quality. It will be easier to maintain integrity for your data and you will have to do less programming. Storing the same information in more then one place, means your will have a problem keeping al those instances of the same adres consistentent if it changes. You also have to make functionality to enter, display and print addressen more than once.

But a lot of organisations do not store adresses seperately, because adresses are closely tied to a person or organisation. And from a programming point of view it is easier if the data is in the same table.

So there is no easy answer.

Your design also allows addresses to be re-used. Is this what you want? Or should an address always belang to exactly one person or company? Allowing reuse to take place creates its own problems. And your design does not allow for more then one address for each person/ company. Is this what you want. Usually companies also have a separate address for correspondence.

Greetings,

Peter

Steve Lucord wrote:

> I have several tables that contain address information in the system I
> am currently working on. Instead of putting the columns that pertain to
>
> the address in each table, I created an address table with a system
> assigned key. Each database table that needs an address contains a
> foreign key to the seperate address table.
>
> For example:
>
> CLIENT
>
> Social Security VARCHAR2(11) 555-55-5555
> Address_Id NUMBER(15) 1
> Name VARCHAR2(30) Joe Client
>
> ADDRESS
>
> Address_Id NUMBER(15) 1
> Address_Line VARCHAR(80) 123 Address Way
> City VARCHAR(30) Any City
> State VARCHAR(2) AK
> Zip Code VARCHAR(10) 77598-5339
>
> I am looking for the pros-and-cons of modeling the database in this
> manner. Philosophical as well as practical arguments are welcome.
>
> Thanks in advance for any suggestions.
>
> Steve Lucord
Received on Sun Dec 19 1999 - 10:16:17 CST