MMSD’96: Workshop on Multimedia Software Development

Multimedia imposes quite a number of specific issues and requirements on software. While system support for multimedia has been seriously investigated for serveral years now, the software engineering community has not yet reached a deep understanding of the impacts of multimedia on their field. MMSD’96 brought together a representative spectrum of researchers investigating this issue. Four general observations could be retained from their contributions: 1. distributed multimedia applications exploit both more long-term potential and more persistent research problems than multimedia PC applications; 2. A reference model of multimedia software is not yet in reach; the large number of models proposed, even at MMSD, still lack better rationales, theoretic foundation, and convergence. 3. the MMSD community brings together people with a “document-background” and those with a “software background”; their views and approaches are still far from being harmonized. 4. While the multimedia community struggels for a common understanding among its members, it is in desperate need of a harmonization with approaches to distributed software development. In the remainder of this article, we will review issues to be addressed in the MMSD context, discuss the contributions made at the workshop itself, and draw conclustions for the state of the art. 1 Issues in Multimedia Software Development 1.1 General Concerns The proper coordination and handling of multiple and (in part) time-dependent media on (distributed) computer systems imposes a number of particular problems which can be gathered into five key areas. We will briefly list these key areas below and discuss their impact on software engineering. media types: the traditional computer-centered basic data types like boolean and real are complemented by human-centered media types like audio and video. These media type are complex, diverse, and evolving, as are the operations defined on them.