|
In a perfect world, the database design for your application would never change. In the real world, most applications grow and acquire new tables, new relationships, new fields, and new features. Garry Robinson and Scott McManus show you how they keep the databases for their clients up-to-date and in sync with each other. In the early stages of database modeling and development, the users and the developers can exchange data-only (back-end) databases relatively easily. This tends to happen quite a bit early on while the developer is modifying the structure, relationships, descriptions, data types, and validation rules of the database. However, as time goes on and the number of users and the size of the back-end database increase, it may not be practical to send the data-only database over the Internet as modifications to the data structure are made. Instead, what you need is some way to modify the database design without losing the data that your users have in the tables. This article explores what you can do when you reach this stage in a project. In particular, we describe the software approach that we've taken to make these back-end modifications quick, accurate, and painless for all stakeholders. But before we get into that method, we should point out that you don't always need to implement our solution. Read More
|