Close to the edge

In 1991, QuickTime movies were the size of a postage stamp.

Do you remember the first time you saw video playing in a postage stamp-size window on a desktop computer? Apple announced the first version of QuickTime at the American Film Institute in Hollywood 14 years ago.

Many broadcasters and video equipment manufacturers considered this to be some kind of a joke, certainly no threat to the TV in the family room or tools needed to create and distribute high-quality video. Besides, how could these new video files be delivered? Most PCs were still using floppy disks, and the ARPANET — the precursor to the Internet — was a virtual unknown research project created by the U.S. Defense Advanced Research Project Agency.

My first encounter with a video file came in the form of a floppy disk, from a company named SuperMac. The disk contained a modified version of the now famous Apple 1984 Super Bowl ad introducing the Macintosh. SuperMac was intent on changing the world of media, with plans to turn a Mac into a platform for video production and distribution. It succeeded in convincing Apple that the vision was viable, which in turn led to the announcement of QuickTime, but the company did not survive long enough to see the full realization of its dream.

Fourteen years later, the Internet was abuzz in the wake of NAB2005. Veterans of the desktop video wars were amazed once again as they played full-screen 1280×720p movie files on their Macs, enabled by QuickTime 7 and a new standards-based video compression codec developed by the Joint Video Team of the ITU and ISO. The popular acronym for this new codec is AVC, or advanced video codec. The standard is also known as the ITU standard H.264 or the ISO standard MPEG-4 part 10. (“Squeeze harder” on page 80 examines the AVC codec and other compression technologies, which drew considerable attention at this year's NAB.)

The delivery of high-quality video using broadband networking technologies has been feasible for several years, though it can be a time-consuming process, especially for program length material. The availability of affordable broadband connections to the Internet and a wide range of private data networks that offer improved performance for backhaul applications has set the stage for another revolution in the business of television.

New plumbing

As with any disruptive technology, some companies will fail to grasp the importance of what is happening — and will not survive. And some companies will embrace the changes that the new technology enables — and will thrive.

Most broadcast facilities were designed to deal with video as a continuous baseband signal. Analog routing switchers became a staple of the industry over the past two decades, allowing signals to be routed from one device to another. For example, to make a copy of a videotape, the outputs from one VTR would be routed to the inputs of another.

In recent years, computer networks have been added in facilities, but rarely to handle the routing of video files. In the computer world, routing is very different from routing baseband video. Traditional audio/video routing switchers use arrays of crosspoints to facilitate connections that look much like those of a patch panel; electronic circuits act like the pieces of wire used to make connections in those patch bays.

Routing in a digital computer network is completely different. A network node has an address and the ability to receive and send data packets. Typically, many nodes share the same physical network links. Any device on the network can route packets to any other device on the network by identifying the address to which the packets are being sent; the other nodes ignore these packets. Because multiple devices can send bits at the same time, the data network requires overhead to handle contention for access and the collisions that may occur if two or more devices try to transmit at the same time. As the data network becomes more complex, routers and switches are typically used to segment users and create connections between various sub-networks.

This all leads to a continuing debate about the plumbing requirements for the broadcast and video production facilities of the future. Should the traditional routing switcher be upgraded to add dedicated switching for Ethernet nodes, or should the data networks be provisioned to deal with the movement of large media files?

The answer should be obvious, but perhaps not for obvious reasons. Most broadcasters believe that they will still be in the same business in the future, feeding one or possibly a multiplex of TV channels. The notion that in the future broadcasters may be delivering files to their viewers via over-the-air digital broadcasts and broadband networks has yet to sink in. However, the notion that it is now possible for broadband networks to replace satellite links and the physical movement of media (e.g., tapes and discs) for backhaul applications is beginning to sink in.

Any content, anywhere, anytime

Ultimately, the survival of local broadcasting is likely to involve both traditional and new technologies, and thus, it would be wise to plan for the digital plumbing to support both. The main competitive advantage broadcasters enjoy now — one that will continue to be important in the future — is that they are close to the edge of the Internet (not the digital cliff).

Emerging digital technologies present broadcasters with all kinds of disruptions to what was once an easy-to-make-money business model, albeit, a model based on the concept of scarcity of programming choices. Cable and DBS have made inroads by addressing the issue of scarcity and, to a limited extent, by selling content directly to the consumer in lieu of all of those annoying ad interruptions.

The personal video recorder (PVR) makes it easier for consumers to time-shift their viewing and to skip ads. The delivery of video programming via DVDs and potentially broadband networks in the future, further undermines the traditional role that broadcasters have played as the gatekeepers of content for their communities.

There is a big buzz around IPTV. The term for the delivery of television programming using Internet Protocol networks is often misapplied. Several of the regional Bell companies in the United States are beginning to deploy IPTV systems. Like digital cable, these networks will support broadband data services, but the television services will not be delivered via the public Internet. These systems are being designed as walled gardens and provisioned to deal with the delivery of multiple programs simultaneously to TVs located throughout the home. A wide range of program choices will be routed to a neighborhood using IP multicast techniques, allowing viewers to select from the available multicasts. And only the programs the customer chooses to watch will be routed to the home. To do this, the system needs the ability to route between 25Mb/s and 50Mb/s to each home.

Today's broadband services are incapable of dealing with the delivery of hundreds of channels to the thousands of viewers in any community. The bottleneck comes at the edges of the network, close to the customers, where everyone is vying for access to a new world of content. To alleviate these bottlenecks, companies such as Akamai mirror bandwidth-intensive digital media content on hundreds of servers around the world. This allows customers to access servers that are close to the point of consumption, avoiding many Internet traffic jams.

Broadcasters have been close to the edge from day one. They serve local markets and in the emerging digital world, they have the ability to serve up content of interest to their communities — potentially offering a wide range of programming customized and localized to sub-markets and individual homes.

A DTV transmitter can push a virtual avalanche of bits into the homes in acommunity, and when properly integrated with broadband networks, it can become a two-way system capable of handling transactions and the localization of content to the IP addresses of individual viewers.

For today, broadcasters should be thinking about dealing with the internal movement of video files and the ability to ingest content from suppliers via public and private digital networks. With next-generation codecs, such as AVC, an SD video feed can be delivered at 700Kb/s to 2Mb/s, and an HD feed can be delivered at 8Mb/s to 15Mb/s. Already, many companies use broadband networks to deliver commercials to broadcast facilities; the delivery of program-length material is beginning as well. Cost saving relative to the physical movement of assets or the satellite time required to move them will be a major factor in the shift to broadband delivery.

Companies such as Telestream have been providing the systems necessary to create distribution networks for companies with multiple locations and vendors for several years. And virtually all of the companies involved with video compression technology have the systems needed to take content from baseband into the world of IP, where it can be transported via public and private networks.

Likewise, many STL and remote microwave links are benefiting from the ability to compress the signals. JVC is promoting HD ENG, using its MPEG-2 encoder to squeeze HD signals into the bandwidth of a single baseband microwave link. As WiFi networks are built out across the country, it may be possible to feed live radio and TV news reports using a notebook computer. We are all familiar now with live news reports delivered via broadband networks and satellite phones.

As broadcasters move into the world of Information Technology, they may discover a new world of opportunities to serve their markets. The ability to deliver bits of interest to your market and to use IT to reach sub-markets and individual homes, rather than feeding one signal to a transmitter that sends the same programming to the entire market, is likely to revolutionize our business. It is time to embrace this revolution; resistance is futile.

Craig Birkmaier is a technology consultant at Pcube labs. He hosts and moderates the OpenDTV Forum.

Send questions and comments