• Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint
Share this Page URL
Help

Foreword

Foreword

by Leonardo Chiariglione

Communication standards have existed for centuries. Whereas in the past standards were generated by custom or by royal or republican edicts, with the progress of technology and the formation of standards bodies such as the ITU, a new process was put in place whereby innovations leading to new forms of communication could be developed in a particular environment and ratified by a standards body. This arrangement served the needs of all concerned parties for a long time. But those were times when innovation happened slowly and industries operated independently from one another.

MPEG came to the forefront at a time when it was becoming apparent that the old arrangement would no longer work. Too many similar developments happened in too many environments wherein each of the solutions developed could be applied reasonably well to all other environments. And this was happening at such a speed that the usual ways leading to standardization not only rendered the process ineffectual, because of the time required by traditional standardization, but also could exacerbate the impact of similar standards employed by different environments for similar applications. In contrast to this, MPEG sought the involvement of all players from all relevant industries and created a new process whereby the results of research could be fed into the standardization process before the industrial development.

Retrospectively, one could venture that MPEG-1, a process driven by industry's need to have a standard for interactive video on compact disc and digital audio broadcasting, has been a failure, because the specific products that were the targets of that standard have had scant or no success. The same can hardly be said of MPEG-2, because the product that was the target of the standard—digital television—has been incredibly successful. More than the success, or lack thereof, of the standards as originally conceived, what matters is the success of products that were enabled by the standards even though they were not considered, or were thought of as being of lower priority, as targets of the standard: Video CD, MP3, DVD, and 4:2:2 profile used in the television studio are the best known cases.

In order to understand the MPEG-4 project and its prospects, we must keep in mind the lesson provided by the first two MPEG standards. While working on the MPEG-2 standard, MPEG developers began wondering what the further needs of industry in digital audiovisual coding might be, given that the 1.5 Mbit/s range had been covered by MPEG-1 and the 5 Mbit/s range had been covered by MPEG-2, even as reference to bit rate had been—correctly—removed from MPEG-2. Developers sensed that there were latent needs for very low bit-rate coding that went beyond the simple moving-pictures-on-the-telephone-wire idea which was so common at that time.

It took more than two years to develop requirements for the new standard and three years to publish Version 1 of the standard. Eventually, MPEG-4 evolved into a comprehensive set of tools capable of satisfying, through the mechanism of profiles, multiple industry needs while providing a high level of interoperability. As a result, MPEG-4 provides solutions for the mobile environment, the Web, broadcast, rich virtual environments, and even very high bit-rate applications for the studio.

Clearly, people on the Internet are going to be major users of the MPEG-4 standard, but the marriage has not necessarily been decreed in heaven. The Internet has been built on the assumption that it should be an infrastructure accessible to everybody, and that multiple applications can be developed for it. Application standards, according to this assumption, are not needed because certain algorithms are implemented in any given piece of software, so that users can download the code, allowing them to decode the particular algorithm used to produce the bits of data of interest to them. Under these conditions is there still a need to develop a standard like MPEG-4?

In the early days of MPEG-4 development, this question used to be asked often; but today, with an ever-expanding use of MP3, it is easier to understand the benefits of a standard: A playback device is not necessarily connected to the network; it may be on a broadcast channel, or it may exist as a standalone or portable device. These devices can use many different CPUs, for which it might be too costly to develop playback software, or the hardware may use an ASIC for the audiovisual decoding that is not upgradeable or may have been designed to run specifically with the amount of RAM the standard algorithm requires. Setting aside other advantages, it is simply easier to have a common standard on which business opportunities can multiply, instead of struggling with incompatibilities between applications all over the place.

The world is sufficiently complex as it is without adding further confusion by blindly worshiping flexibility, particularly when one sees the mythical world of “many applications” turning out to be a world of a single application whose access and evolution is controlled by a single company.

In opposition to this model, MPEG (and MPEG-4 in particular) offers a standard that has been developed collaboratively. Access is open to anybody who wants to be in business, and its evolution is in the hands of technical representatives of the various players involved. MPEG-4 has even produced MPEG-4 reference software, through a process similar to open-source software that can be used by any given implementer.

Ability to access does not mean that access is free, however. For the last few decades, the world of digital audio and video has been the focus of research by thousands of companies and organizations. Some have worked for the sole purpose of increasing the collective knowledge of humankind; others have worked to create for themselves the opportunity for future revenues by securing patents for their inventions. Rightly, ISO does not allow consideration of patent issues in technical groups like MPEG, but this is no reason solutions cannot be found outside. A patent pool exists today for MPEG-2, and more are likely to be generated from the activity of the MPEG-4 Industry Forum.

This book was written by some of the most authoritative individuals playing key roles in developing the MPEG-4 standard. Although the book is not a substitute for the standard itself, it provides a general overview as well as insights into the critical elements of the standard and information pertinent to understanding MPEG-4 and its practical use.

  • Creative Edge
  • Create BookmarkCreate Bookmark
  • Create Note or TagCreate Note or Tag
  • PrintPrint