35. Conclusion

The future is bound to bring still more proposed algorithms, targeted at specific video sources and applications. We believe that VideoScheme's flexibility makes it a useful vehicle not only for research, but also for rapid application development and prototyping. VideoScheme's high-level primitives and rapid-turnaround programming environment make programs compact and prototyping very rapid.

We have implemented a prototype of VideoScheme and have demonstrated the usefulness of our system on some real-world editing tasks. We have seen that it is possible to achieve some of the results of dedicated video authoring systems such as IMPACT in a programmable editing system with a small number of special-purpose video functions. Our VideoScheme system offers the further advantage of flexibility: it is a simple matter to experiment with new algorithms, and to build new operations by combining previously written functions. These are the classic advantages of programming, and we believe that we have shown them to be equally valid in the domain of interactive video editing.

We believe that VideoScheme is a unique attempt at achieving the best of both worlds: the ease of use of direct manipulation and the flexibility and expandability of an interpreted programming language.

After this discussion of the third of our elements of hypermedia design, content, part IV will now put together the pieces of the puzzle by presenting issues and obstacles arising when building complex hypermedia systems. We discuss our experience in publishing a series of multimedia conference proceedings over a period of four years using a variety of systems and technologies.