Building on Legacy Applications

Our focus on web applications has necessarily involved retrieving data from, and interacting with, non-web enabled systems. These range from existing data stores held inside relational databases to custom applications that provide an API for external access, or frequently no interface at all, relying on data exports as simple CSV documents for example.

Wherever possible it is important to try and abstract away the details of the legacy system from the main web-application. This allows the main application to deal with the data in a way that is most suitable for the tasks it needs to perform, and not require any knowledge of the original legacy application. In many cases they web system is used in a different manner to the legacy system, and may only expose a subset of data for very specific purposes. A separate middleware component (either developed by us, or a third-party) is responsible for extracting and managing data from the legacy system and making it available to the web-application.
This separation is particularly important when any changes are made to the legacy system, as this reduces the effect of those changes to the web application.