MPEG-4's ability to represent multimedia data types efficiently over networks spanning very low bit rates to very high-quality conditions, coupled with its object-based multimedia representation model, promises to open the door for a broad range of applications. The goal is that such applications, used with emerging standards like MPEG-7 (for content description) and MPEG-21 (for interoperable digital-rights management) can automatically adapt and scale to any type of network and device, to fit end users' personal preferences according to their rights to view the content.

Consumer devices likely to leverage MPEG-4 run the gamut of existing and futuristic systems: from mobile phones, PDAs and camcorders for wired and wireless video communications to digital set-top boxes or personal computers for interactive TV applications and streaming-media content delivered via the Internet. They include advanced DVD players that will play back high-definition movies with interactive content and personal video recorders storing high-quality audio/video files.

A few years from now, consumers won't notice whether what they're watching or listening to on any one of these systems has anything to do with MPEG-4. In fact, the whole point of MPEG-4 is to make invisible, at least to to consumers, the increasingly complicated job of sorting out what kind of content a system can receive, over what type of network, and when.

But for every glowing possibility MPEG-4 presents, there is a misgiving. The world is still far from embracing the premise of an MPEG-4-driven entertainment nirvana as the market evaluates each new compression format. Many companies are working on ideas to build more intelligence into a content-delivery network so that multimedia material can be transcoded or reformatted to an end user's screen.

Moreover, multiple standards are fast becoming an accepted reality in the Internet world. It is far from clear how dominant a standard MPEG-4 will prove to be several years from now.

Nonetheless, network operators, content owners, consumer electronics manufacturers and communication companies are experimenting with MPEG-4 and implementing it in their new applications and devices.

Using MPEG-4 for high-quality, lower-bit-rate streaming media or video mail for PCs and mobile phones has been widely discussed, and to some extent implemented. The content can be delivered even in error-prone wireless networks, because MPEG-4's encoding layer is resilient to residual errors for the various data types, especially under difficult channel conditions such as mobile nets.

If the technology takes off the way some industry players expect, watching a movie clip on your mobile phone before deciding on which film to see one afternoon-or receiving a video e-mail on your Dick Tracy watch-will soon be no pipe dream. Cell phone manufacturers are already building MPEG-4 decoding capabilities into advanced handsets newly designed for third-generation or even 2.5G wireless networks.

Soon, while shopping in a supermarket or department store, you'll be able to easily check with your spouse, using a camera-enabled mobile handset, to see if the fish is the sort she wants for dinner-or if that sweater that's on sale exactly matches the color of his eyes.

Some of these wireless video applications don't even require real-time streaming. Without waiting for the emergence of a 3G network, for example, one could already send an MPEG-4 compressed video file as video e-mail over 2.5G networks. Such a file can be sent in packets, when the bandwidth is available, not only to another handset but to a PC, TV or any other consumer device. Though the push is on to use MPEG-4 as a streaming format for audiovisual content delivery over the Internet, the initially proposed MPEG-4 licensing terms appear to have dampened momentum among some service providers and content owners for the moment.

Large Internet service providers such as AmericaOnline, cable/satellite operators and Webcasters are finding the so-called use fee specified by the MPEG-4 video licensing structure unacceptable. Some industry sources argue that the fee, set at 2 cents per hour based on playback/normal running time for every stream, download or other use of MPEG-4 video data, makes no sense, particularly for broadcast companies. "Imagine how much in royalties a cable company would have to pay if the use fee is set at 2 cents per hour for its 8.4 million subscribers, for example, for every video stream compressed in MPEG-4 they deliver. You do the math," said one source who spoke on the condition of anonymity. "It appears that MPEG-4 video licensors did not really consider a broadcast model very well."

Developed by the International Telecommunications Union, H.26L includes a number of tools that allow it to deliver block-based video coding much more efficiently in lower-bit-rate applications than any current MPEG standard. Moving away from the known limitations of the 8 x 8-block discrete cosine transform approach used in MPEG-2, H.26L leverages today's processing power increase to provide more choices in coding techniques, including multiple reference frames for motion compensation and residual coding based on 4 x 4 blocks and integer transform.

By selectively using some of the H.26L compression tools, one industry source said, "we know we can improve coding efficiencies by 50 percent and achieve resolutions ranging from [standard-definition] TV to HDTV and digital cinema."

The MPEG community has joined forces with ITU video-coding experts to form the Joint Video Team, which will pursue an H.26L-based MPEG-4 variant, to be called MPEG-4 part 10. Expected to completed by year's end, it is poised to give MPEG-4 a leg up over competing compression formats.

Where MPEG-4 truly shines is in its ability to do object-based multimedia representation. The underlying object-based technology is designed to give authors a standardized way to add interactivity to their content. As such, it will have the most profound impact on content developers, content owners and service providers.

More specifically, MPEG-4 makes it possible to keep a variety of elements within a scene-whether moving images derived from a video stream, graphics from a computer-aided design output or text-as separate, discrete "objects" that remain intact and behave coherently, no matter where the content is consumed. It is not necessary to re-encode the images into pixels and put them back together again in a single stream, a process that "could be devastating to the quality of the content," said Rob Koenen, president of the MPEG-4 Industry Forum.