INDUSTRY VIEWPOINT - Legacy tools made of solid gold

Geoff Baker, development director at Transoft, says legacy applications should not be dismissed.

Ask anyone not connected with the IT world what image the word legacy conjures up and they will probably respond with windfall - the cash lump sum that a recently deceased distant relative would leave in their will.

Strange then that in IT the word conjures up more negative images of archaic, outmoded systems.

Legacy systems are applications developed in programming languages such as Cobal and Basic - languages that were in corporate data centres before the beginning of time. As enterprises grew around these data centres, and as the need for enterprise-wide open systems grew, it became obvious that legacy systems would not fit into programs written for open databases, for example in programming languages such as Visual Basic or C++.

Businesses with legacy applications have had to face one of two choices - either orphan these applications on the outside of their computing strategies, or completely re-engineer to integrate them into their client/server desktop systems. The latter option risks spending vast development resources and consultancy, and this does not guarantee success. Most systems integrators steer well clear of this area and, as a result, there is an enormous loss of potential income from an area which need not be as specialist or prohibited as most believe.

To date, mainstream system integration businesses have been very adept at delivering client/server networking computing strategies, but they tend to falter at integrating existing legacy systems into the overall system. This is because the legacy applications in question are structurally huge, have grown around a primitive database system, and require a complete re-design of structure. This requires a huge amount of re-coding - not the sort of thing that system integrators relish, as most legacy re-engineering projects are either cancelled before implementation or fail to meet customer expectations in the long term.

Now, software tools are available which provide integration between standalone legacy applications and PC tools such as Access, Visual Basic, Powerbuilder and the internet. The tools modernise and integrate legacy applications faster, safer and more cost effectively than re-engineering - in a way that does not require any specialist skills.

Legacy system users are able to get the benefits of modern GUIs, Web access, multi-tier client/server, and ODBC technology for their existing applications. Best of all, they can obtain these services through their preferred integrator.

Legacy systems should be seen as an opportunity to increase client relationships.

Understanding how to liberate these legacy systems is the way forward.

There are now fewer reasons than ever to steer clear of the sector in future.