Most critical business logic and data still reside on legacy systems, according to Meta research. Today`s business demands require integration between legacy systems (anything that was installed yesterday) and new systems, yet often these attempts at `integration` are simply skimming the surface.
Three main trends underscore the move towards legacy integration. Firstly, as computing has evolved, businesses have begun to understand that standalone solutions for problems are both costly and inefficient. There is now a requirement for integrated solutions that bring together old mainframe systems and deliver them on a common user interface. Secondly, IT has moved from being technology focused to being business focused, and concomitantly, the business people within organisations have become more technology literate.
This has forced solution providers to offer a business solution rather than a technology solution, since corporate needs are framed as business problems not as technology problems. Thirdly, the small size of the South African IT industry means that competing companies are more likely to partner with one another to offer a solution.
One buzzword in this chase for integration has been Web services, which are seen as the panacea for all integration problems. And Web services certainly have this potential - if they are correctly implemented.
In the Web services arena, companies generally have a choice between two main routes of implementation: they can either go the J2E Java route, or they can go the Microsoft .Net route. Both are promising to deliver similar services, and companies are finding the choice difficult to make.
In my opinion, a company`s choice should be determined by their existing in-house skills. If you have Java skills, choose J2E. If you have Microsoft skills, choose .Net. Stick with what you know and develop on top of that, reusing as much as possible, so you don`t have to start from scratch.
The biggest mistake we see people make is shortsightedness: they implement a Web services solution for one department without determining whether the solution fits other departments as well. This can lead to yet another situation where standalone solutions are implemented, negating the benefits of true integration. Instead, companies should choose a product that accomplishes their overall goal of integration, rather than solving a single solution with a single application in a single department. The reality is that there is no silver bullet when it comes to legacy integration,but there might be a set of products from a consortium of companies that offers a close-to-perfect solution.
As companies feel the time pressure towards integration, they face the risk of making hasty decisions which are likely to cost them in the long-term. This time pressure often leads to solutions which simply make an old mainframe green-screen environment fancy - a typical case of putting lipstick on a pig. For instance, in the banking environment this could mean that while customers want a single online view of all their accounts, the lipstick of a fancy green-screen can`t provide the required consolidation of account details. Rather look for a product that can do both - provide the fancy green-screen, but also integrate Web-based technologies with old mainframe systems through adaptors.
Also, since the solutions are to have a business meaning, it pays to look for a solution provider which has a strong business background to support its technical ability, rather than a provider who has essentially a technology background. Is it a company that started up and built a product technically, and therefore has a technical solution? Or is it a company that started from a business perspective and built a product based on the business approach?
A good technical solution might not provide you with a business solution. Companies that developed technology based on business requirements, however, have a broader understanding of how businesses are run and are therefore more likely to solve a business problem rather than a technical problem. Even some of the bigger IT companies have come to understand this, and, although they started out as technology-focused companies, have bought out a number of companies which help them fit business needs.
Finally, it is worthwhile remembering that while "Web services" is probably the biggest IT buzzword currently, and this is where the decisions have to be made, it is still not an entirely established standard.
For more information go to www.commercequest.co.za.
Share
Editorial contacts