Legacy systems will always be with us, even after Y2K. There are no panaceas to solve them - all the current 'fashions' (ERP, Java, WAP, UML, agents...) can be subject to legacy problems. This is because legacy is a fundamental property of socio-technical systems, arising because of a gap between evolving business needs and system capabilities. We should instead look to understand the nature of legacy systems (and their connection with organisational change); and hence to look both for ways to migrate existing legacy systems to new states, and to make a system less likely (for both organisational and technical reasons) to become legacy in the future.

This view of legacy systems has arisen from one of the major themes of the Systems Engineering for Business Process Change (SEBPC) programme of the UK's Engineering and Physical Sciences Research Council, which has involved 30 universities from 1997-2001. In this workshop, researchers from the programme will outline its main results in various areas in short presentations, followed by discussion. About two-thirds of the workshop time will be spent discussing the ideas and relating them to participants' own research and practice.

Workshop aims:

(1) To bring together and disseminate results concerning legacy from SEBPC

(2) To discuss whether the conclusions we have reached seem appropriate in the view of those present

(3) To look to the future of research on legacy systems

Morning session: Scoping the problem

• Introductions. Nature of change and legacy.
• Communication and language gaps.

Afternoon session: Working towards solutions

• Reactive solutions: migrating old systems.
• Proactive solutions: being ready for legacy.