Reengineering legacy document information systems: Challenges and solutions

Abstract

Since internet applications have reached a satisfactory level of maturity, large information systems have been developed to manage and facilitate access to documents. Simultaneously, there was an enormous international effort to digitise documents, enabling access via the internet. This endeavour facilitated the access of researchers to extensive document repositories and libraries, while also aiding companies in organising their documents. Two decades later, these vast databases are reasonably clean and well-organised, although the software used to manage and feed them is gradually becoming obsolete. Therefore, it is imperative to continuously reengineer the software to maintain optimal functionality. Furthermore, after the initial effort to digitise documents and create the initial metadata, it is reasonable to augment the metadata information pertaining to the documents. As such, two necessities are apparent: improving support for reengineering legacy document information systems and enabling data model updates and schema evolution to accommodate new information. Our goal is to automate the reengineering process as a whole.