System Transformation Portal

William Ulrich

The Immigration and Naturalization Service (INS) has been taking severe heat recently. They were initially unable to reconcile data with various security agencies after the 911 incident and, most recently, issued a visa to a dead hijacker. While it would be easy to blame a manager or a clerk for these issues, it is clear that their aging computer systems are no longer up to the job of supporting the INS mission, let alone coordinating with other agencies within the federal government.

More low profile events have also been in the news. The State of Washington recently issued a slew of notices stating that people with unpaid traffic tickets should either pay up or lose their driver's license. This was no normal warning, however. The people notified were told they would lose their licenses based on tickets issued more than a decade ago. Most people claimed they had paid their tickets and one woman even produced a cancelled check. Others just paid the fine. The kicker was that the State said that people would lose their licenses by the first of March, but they could come in and challenge the State after March first.

The INS situation and the State of Washington have one thing in common: their computer systems are at the heart of their problems. One could jump to the conclusion that these are isolated circumstances found in government agencies that have neglected their applications for too long. But this is not the case. There are numerous examples of system-related failures within banks, telecommunication firms, manufacturing firms, retailers and a number of other industries.

There are three main factors that need to be considered in a strategy that addresses these issues - the management, integration and transformation of legacy application and data architectures. I have tied together legacy application and data architectures because you cannot fix one without dealing with the other.

Legacy application management involves maintaining legacy systems in a more cost and time effective manner. Such a strategy involves delivering small-scale, ongoing upgrades and enhancements to the business community as needed. There are a number of transformation tools and techniques that help IT perform legacy architecture management and maintenance tasks more effectively.

A number of organizations consider outsourcing as the ultimate solution for addressing their application management challenges. This is fine if their outsourcing strategy can be reconciled with their application integration and architecture transformation plans. If not, they should rethink their application management plans.

Architecture enterprise integration or EAI involves connecting legacy data and applications with web-based applications, using non-invasive techniques. EAI is an industry-accepted approach for addressing the near-term need to deliver enterprise data and functionality to the web while making applications and data appear to be integrated. The non-invasive approach taken by various EAI solutions provide near-term value, but can also result in a legacy architecture that stymies future business agility and continuity. In other words, uncontrolled deployment of EAI interfaces will make the management and transformation of legacy architectures that much more difficult.

The long-term strategy for legacy architectures should involve the transformation of legacy data and application architectures. This last approach differs from EAI because it is invasive and changes underlying legacy applications and data studies. Transformation options include a wide variety of data and system level activities that examine, modify and migrate legacy applications and data. Migration typically targets Web Services, component-based architectures, relational data structures, consolidated application environments, emerging computer languages and model-driven application management.

Legacy application management, integration and transformation can be incorporated into a single, multi-phased strategy that is blended with various package deployment, consolidation, data warehouse, platform migration and new technology deployment initiatives. Executives should remember to not ignore any of these issues when considering a legacy architecture strategy and be sure to consider how these three critical issues impact each other and the ability to deliver rapid value to the business community on an ongoing basis.