2004. It is a Friday evening. Mom comes home after her usual hustle and bustle at work, takes her shoes off, sets her briefcase down in the foyer, calls out to Hari who is playing his computer game, slides into the recliner, picks up the remote, and turns on the TV. It is now about 8 PM and the seven o’clock evening news is just starting. She clicks a button on the remote and a menu appears. An advertisement comes on: sale on
Rajasthani jewelry at the local Fremont store. Another click and she is in a virtual video
store where she orders one of her favorite movies: To Kill a Mockingbird. The screen
confirms delivery by 9:00 PM. She then selects the music menu and creates a list of her
favorite songs, turns off the TV, and heads to the kitchen to join Dad in cooking. The stereo system comes on and Ravi Shankar’s sitar melody mixed with Alla Rakha’s heavenly tabla comes to life in the kitchen. Is this a dream?

NOT FOR LONG. SINCE THE EARLY NINETIES,
technologists and visionary business leaders from a
number of pioneering companies in the computer,
consumer electronic, television broadcast, and cable
TV businesses have been working to change the way
we watch and interact with our television. They have been
working towards an everything-on-demand future. News-on demand:
watch only the news that you care about, not what
the broadcasters force-feed you through the airwaves. Music-on
demand: listen to any tune ever composed. Education-on demand:
attend Richard Feynman’s lectures on quantum
physics from your living room. Sports-on demand: relive the
memory of Pelé’s most famous goal in the 1962 World Cup or
Sunil Gavaskar setting his century record.

It is now 9:30. Hari is off to bed. Mom and dad
sink comfortably into the living room couch. They
turn on the TV. An ad for Naz Theaters in Fremont
is displayed: Hrithik Roshan’s latest big budget
production is coming to the theater. Another ad:
Mata Amritanadamayi’s November tour in her San
Ramon ashram. The movie begins and Gregory
Peck comes to life on the screen.

In addition to providing access to any content ever
created, researchers have been working on bringing you
personalized advertising, which is a holy grail for the
advertisement industry. Targeted advertising based on the
subscriber’s needs and wants are valued very highly within the
industry.

What is going on here? A revolution in the television
industry? Actually, it is not a revolution. It is just another
evolutionary step in a process set to motion by something that
happened in a small community near Orlando, Florida in
December 1994.That was when the media giant and multiple
cable station operator,Time Warner, along with its technology
partners—such as Silicon Graphics and Scientific Atlanta—
launched the nation’s first interactive and on-demand
television system. It was called a “Full Service Network”
(FSN). FSN gave consumers access to movies-on-demand, the
ability to visit a virtual shopping mall, and to play interactive
computer games with each other. The service was very
popular, but it lasted only a year due to the high cost of
expanding and running the system. The designers went back
to the drawing board to figure out how to do it better and
cheaper.

There were many factors that contributed to the high cost
of building and running such a service. In order to provide
many channels of information, cable operators had to upgrade
their plant from analog to digital. This required building a
hybrid fiber-coax network that was highly capital-intensive.
Moreover, the early designs of the home set-top box—which
receives and decodes the pictures, controls interactivity, and
so on—cost a small fortune, at a couple thousand dollars a
pop. Imagine putting one of those boxes in each of two
million homes. There were a number of other cost issues as
well.

Fast forward to 2002. North American cable operators
have mostly upgraded their network infrastructure, which
was required to launch digital television, cable modem
services, and so on. In addition, even the good, old-fashioned
phone companies have come into the fray. They were aided
immensely by the introduction of new technologies such as
DSL, several variations of which (like ADSL and VDSL) allow
high-bandwidth data to travel to homes. Consumer
electronics companies—such as Philips and Panasonic—are
using their expertise in building mass-scale electronic devices
to build cheap set-top boxes. They no longer cost a couple
thousand dollars: these days, you can buy one for around
three hundred dollars, and the costs are dropping even

further. Indian companies are getting into
the act, too. Recently, Prasar Bharati
announced that his company will introduce
cost-effective community set-top boxes,
priced at around 1500 rupees for
Doordarshan National channel viewers.

Why the frenetic activity? One simple
reason: moving pictures (video, movies,
and so on) are the most compelling
medium out there. Mass communication is
here to stay. Just think about the success of
the Internet. The first-generation Internet
was all about text. The second-generation
Internet introduced more visually
appealing graphics, interactivity, commerce
and even audio. Now comes video.
Unfortunately, delivering high-quality
video (that has perceived “value” and, thus,
that a subscriber is likely to pay for) to a
very large number of users over today’s
networks is a tremendous challenge. First
of all, broadcast quality video, encoded
digitally and compressed down to even
100:1 of its original size using digital
compression technologies such as the
Motion Pictures Expert Groups (MPEG)
recommendations, is very large in size.
Quality video is usually encoded at four to
six megabits-per-second or higher, which
means that a typical movie would be around
three or four gibabytes. Imagine having to
transfer a four-gigabyte file through the
current infrastructure in real-time.

To get around these problems,
telecommunications and cable operators
are building what are called content
delivery networks. In such a network,
commonly used content (such as popular
movies) will be copied and stored on
servers that are located “closer” to user
communities, providing a more efficient
way of getting these programs to end users.
Not so popular content may be stored in
centralized locations, rather than copied at
all sites. In such cases, access to a program
may not be instantaneous–it may be more
akin to using as service such as Netflix
(where you order a movie and it is
delivered in a few days via snail-mail).

Fortunately, North American cable
operators have done a lot of rewiring
already and are now beginning to launch
these services. Telephone companies—not
to be outdone—have invented the next
generation of DSL, called VDSL, which is
capable of bringing up to 6 Mbps to homes
near a central exchange. They also have
plans to deploy gear in the cabinets near
our homes to reduce the distance
limitations of DSL technology. When their
financial health improves, telephone
companies will embark on a network
upgrade. On the equipment side, Moore’s
Law (which dictates that capacities double
every 18 months or so) has brought the cost
of servers and set-top boxes down
tremendously.

The possibilities of video-on-demand
technology are endless. The technology is
real, and it is coming to a television screen
in your living room.

Satish Menon, Ph.D., is the Chief Technology Officer,
Kasenna. He has a track record of technical innovation
and a long history with broadband media technologies. In
January 2000, Satish engineered the spin-off of SGI's
Broadband Media Division as a separate venture-funded
software startup company. At Kasenna, he oversees the
technology evolution of the company's ubiquitous video
delivery platform. Prior to the formation of Kasenna,
Satish was the Director of Engineering of SGI's Software
Technology Group.