One of the most significant developments in computing technology over the past few years is the emergence of multimedia. Glossy multimedia applications are shown all over the place and on a wide range of different platforms. All major workstation hardware vendors feel the need to come to technical fairs with stunning demonstrations mixing graphics, video, imaging, and sound. Technical analysts predict that multimedia related hardware development will be one of the booming areas of electronics in the years to come.
However, most of the available multimedia environments aim at hypermedia authoring, ie, they offer means to interactively create hypermedia documents. ``Document'', as a multimedia term, means more than our traditional paper-based understanding. It should be perceived as a potentially complex composition of related media information, thus it is a multimedia document, which can be ``read'' or viewed in a non-sequential fashion by following semantic connections (or links) between the various media components, hence it is hypermedia.
Although the concept of hypermedia document is very powerful indeed, it does not cover all possible fields of applications of multimedia. The ability of combining, modifying, or even synthectically creating multimedia data is often necessary for more complex multimedia applications. For example, the user might want to extract a frame from a video sequence, modify it with standard image processing tools, combine the image with the output of synthetic graphics, and possibly exchange the original frame with the modified image. Description of such actions does not fit easily in the model of a hypermedia document, in spite of the sophisticated interaction tools which are usually provided with authoring environments. There is, therefore, a need for a programming environment which would allow for the developments of such applications, too.
The techniques to achieve combination of media are extremely disparate, and they use the results of various fields of computing technology like, for example, high quality synthetic graphics, image processing, speech synthesis, etc. Some of the techniques are also highly application dependent. It is almost impossible to define a closed programming environment which would encompass all needs. The already traditional answer to this kind of challenge is to use object-oriented techniques: services are offered in the form of objects, which are then extensible by the programmer to include any necessary application-dependent tools.
The European Communities' ESPRIT III project MADE (Multimedia Application Development Environment[16]) has set up the ambitious goal of defining and implementing such a portable object-oriented development environment for multimedia applications, based on C++[28]. The outcome of the MADE project should be a programming environment running on various UNIX platforms, as well as on MS-DOS or Windows-NT environments. This report gives a general overview of MADE. It describes its major services, and it also gives an overview of some possible ``application scenarios'', ie, major application architectures which may use these services. It is not the purpose of the paper to give a detailed technical description of the full project; this would go far beyond the scope of such a report. The interested reader should consult the ``official'' MADE documents to gain a more detailed insight (eg, [16][30][10][9][8][2]).
The MADE project is still an ongoing activity. Consequently, some problems are still open and will be solved only later in the project. For this reason this report sometimes raises issues without presenting complete solutions.