Database interoperation support in multimedia applications-architecture and methodology

This paper contributes to the definition of salient functionality for multimedia databases and proposes a way forward to tackle the anticipated problems of heterogeneous (proprietary) storage and cataloguing mechanisms already in existence. We have outlined types of requirements that need to be satisfied, and introduced an architecture and method for overcoming heterogeneity of databases. We have given examples how this is achieved for different types of software, including multimedia applications of databases. The techniques described are appropriate for knowledge-intensive metalevel representations, but are unlikely to be suitable in real-time, low-level data conversion. The contribution of our translation and re-engineering technique is that it provides ways of migrating systems as standards evolve. This can make necessary changes of a standard more palatable to the users of the standard. The examples given focus on WWW applications, but could also be extended to areas such as video-on-demand or even satellite-based telecommunications because it is foreseeable that the telecommunications infrastructures as well as the data transport and data manipulation techniques are converging to a fully digitised system. What is not yet converging at all is the plethora of data structures, communication protocols and manipulation techniques for digitised data. As a result, interoperation techniques will remain an important issue for some time to come