LEGACY SYSTEMS ENGAGING ON INVITATION TO TENDER PROCESS

Legacy Systems Engaging On Invitation to Tender Process

Legacy Systems Engaging On Invitation to Tender Process In the software maintenance research field the term “legacy systems” is usually defined using Bennett's description: large software systems that we don't know how to cope with but that are vital to our organization. However we believe such a definition is limiting as it focuses only on large software systems that are ill understood. In a recent review of the term “legacy system” Gold compares Bennett's definition with that of the Free On-Line dictionary of computing “a computer system or application program which continues to be used because of the prohibitive cost of replacing or redesigning it and despite its poor competitiveness and compatibility with modem equivalents. The implication is that the system is large, monolithic and difficult to modify, and Brodie and Stonebraker's definition “Any information system that significantly resists modification and evolution to meet new and constantly changing business requirements.

From this analysis Gold produced the simple description that, “legacy software is critical software that cannot be modified efficiently”. We believe that there are problems with accepting any one of these definitions as each makes a value judgement about “legacy” and each involves negative connotations. A workshop run under the aegis of the Systems Engineering for Business Process Change programme gathered a range of UK academics involved in “legacy systems” research. There was a general feeling in this community that the oft quoted definition of legacy system was too narrow. The critical analysis in workshop sessions moved towards a better understanding of the abstract concept of legacy, and thus what a legacy system is. However, the deliberations over meaning continue as academics debate whether the “legacy software” is the “legacy system” or whether the “legacy system” consists of the software plus some operational aspects of the organisation. Again Gold refers to commentators, such as Smith, whose work suggests that there is a wider context to a legacy system than just simply the software. That context that includes users, management, maintainers and directors. Where a sociotechnical approach is taken (as in the RAMESES project) then this wider perspective is critical.

Therefore for the purposes of our work we define a legacy as “Any existing state which impacts upon change” and thus a legacy system as “A group of interacting elements forming an entity where one ...