Re: Modeling Address using Relational Theory

In article <1125541022.387943.240990_at_f14g2000cwb.googlegroups.com>, dawn wrote:

> Can you explain to me what the difference is between addressLine2 &
> addressLine3 as far as what the values can be and/or what they mean
> other than that one is placed in order before the other? What could
> they be named that doesn't have ordering information in the attribute
> names?

This practice (addressLine2 etc.) usually is seen in situations where you don't know exactly how the address data content will be distributed. This is often the case with foreign addresses. With domestic addresses you usually will be able to structure the content more in advance.

Offering a number of address lines, builds on the assumption that in the print-out (envelope window or label area) you have a limited space where the order can be defined individually.

In this scenario the attribute name addressLine2 will stand for "the 2nd line in the address content area".

In my experience, the problem arises when you notice that one or two of these lines are seldom or never used. Then you start thinking that it maybe would be more economic to regard the addres lines as a list. In that way you would need to save only as many address lines as neccessary.

jp

--