Adapting to change

Proponents often repeat defensively the adage that 'legacy' indicates something that actually works reliably, having long passed the teething problems that beset some new applications for long periods. Although amusing, that is too glib an assessment for there are real problems, at operational and strategic levels, in depending on legacy technologies. One of the major associated problems is whether skills will continue to be available to meet the support and maintenance needs of organisations' legacy applications. The coding language used is the crucial factor here - in the case of many, the skills base can be seen dwindling dangerously, and inevitably the costs of those skills when needed will rise due to the laws of supply and demand. Ultimately skills shortages can prevent the organisation from considering certain avenues of change - or be a barrier when there's a need to comply with regulations or legislation. When these problems are encountered, continuing corporate life with the applications in question must be seen as constituting a real risk.Another major headache can be the platform on which an application is run becoming obsolete. Even before reaching that point, however, infrastructure platforms can introduce problems as they age, in failing to integrate with an organisation's security, application management, application integration, or reporting facilities, any of which introduce risk and extra cost.

Changing processes

Adopting a packaged solution is one option, but inevitably business processes have to be changed to adapt to what the new application supports, and any customization to support the organisation's needs will have to be paid for, and supported for the lifetime of its use. In making a decision on the approach to take, many organisations assess the value that is represented by their existing applications, and look for some way to continue to benefit from their past investments in them.
Where the coding language is the source of the problem, there are an increasing number of options to translate code to a more modern language. Applications represent part of the organisation's intellectual property (IP), and when you consider all the business rules and other logic that are bound up within them, and the fact that they are part of what makes the organisation tick, translation can be seen as a way to preserve that IP. Although translation is a specialist field, and projects can take a while, the time taken may well not compare too badly with the demanding option of defining and writing something from scratch. One characteristic of that approach is that it tends to take some of the organisation's key people out of their main roles for a while, which may not be considered viable or desirable.
Migrating an application from an obsolete platform to a more modern system is not just a lift-and-shift operation, but nevertheless it can be readily achievable in many cases. The choice of new platform might well be limited by factors such as the application code language, or the database, so in some cases there might be decisions involved about re-architecting elements of the supporting technologies - but the objective of migrating something to be near like-for-like must be borne in mind.  In the case of either translation or migration, though, it should be considered whether there might be a degree of redundancy in the existing application.

Investigating the options

During years of changes and re-design, areas of code become 'cut off' within many applications, and investigation of what code may be left behind can avoid the cost of migrating and testing code that would never be used.
Either of these approaches can enable organisations to retain the value of their IP, while taking advantages of the benefits of using modernized applications or platforms. Rationalisation of the variety of technology within the organisation should enable savings to be made, and be more efficient in terms of sustaining fewer supplier relationships, and potentially benefiting from discounts. After translation or migration, the benefits of being able to integrate applications better and use them more flexibly are substantial, as these will be essential organisational capabilities in the near future, enable the value within applications to be shared by partners and customer, and potentially to add to revenues.

BIO--------------------------------------------------------------------------------------------------------------------------------------

Alan has worked in the IT industry for over 25 years, at several organisations including IBM, Sun, CSC, ICL, General Motors, Anglian Water, Severn Trent, HSBC, and Barclaycard. Since joining Butler Group in 2002, Alan's main focus topics have been outsourcing and services, security, and infrastructure management. He has co-authored Butler Group Reports on Identity and Access Management, IT Governance, Enterprise Architectures, IT Performance Optimisation, Achieving IT Flexibility, Security Management, Application Lifecycle Management, Managed Services, Rich Web Applications, Application Delivery, and IT Systems Management.