Anything, Anytime, From Anywhere

We promised to tackle the underlying need and processes that will inevitably transform our broadcast infrastructures from static, deterministic delivery mechanisms into flexible, nondeterministic distribution engines. It may surprise you to know that such transformation is well on its way.
Author:
Updated:
Original:

In my last column, we promised to tackle the underlying need and processes that will inevitably transform our broadcast infrastructures from static, deterministic delivery mechanisms into flexible, nondeterministic distribution engines. It may surprise you to know that such transformation is well on its way.

As it is often the way with technological shifts, a landslide starts with a few wisps of dust and tumbling rocks, but then, driven by powerful market forces it unexpectedly accelerates into a full-scale landscape change. It all started with pitiful postage stamp-sized video clips on Web sites without a business plan. We all thought it cute and went back to our preparations for DTV. Next time we blinked, the news networks started using IP-based file transfers and advanced codec schemes to move proxies around and optimize their workflow. Again we looked, thought it was pretty ingenious and went back to our arguments about multicasting versus HD, progressive vs. interlaced.

PARTS OF THE PUZZLE

Next thing you know, public television is contemplating its Next Generation Interconnection System (NGIS) as a hybrid satellite and terrestrial system that distributes all nonlive content as files over an IP network, other networks are starting to distribute their full-length content as file transfers and at least one satellite service provider is leveraging Windows Media 9 as a distribution file format. All of this has happened in a mere three and a half years.

Just last month, on consecutive days no less, we saw three announcements that, if considered individually might not seem particularly significant but that, when taken as parts of a puzzle presage the enormity of the change that is upon us:

LAND SPEED RECORD

The first announcement came from Sprint and the Swedish National Research and Education Network (Sunet). It detailed how the two organizations had collaborated in setting a new world record for transporting large volumes of data over a public network.

(click thumbnail)This diagram describes the architecture that allowed Sunet to send 849 GB of data from Sweden to San Jose in a record-setting 27 minutes.
These companies managed to send 840 gigabytes of data from a computer in San Jose, Calif., to another one 10,000 miles away at the University of Lulea in northern Sweden in less than 27 minutes. It equates to approximately 140 full-length digitized movies at a rate 560 times faster than real time streaming. Mind you, much faster rates have been achieved in closed networks and specialized laboratories, but what makes this record so mind-boggling is that it was accomplished over the commodity Internet and in resource competition with everybody else's traffic on the network. While most household Internet connections pale in comparison to that amount of bandwidth, we should keep in mind that many cable companies are now providing 3 Mbps downstream speeds at a reasonably inexpensive $50 per month. In Canada, Rogers Cable is operating at 5 Mbps today and is expected to introduce 10 Mbps shortly.

WIRELESS SAFETY

The second announcement came from New York City where the municipal government announced plans to build a public safety wireless network capable of accommodating thousands of simultaneous video streams at speeds of 2 Mbps and the capacity to provide tens of thousands of mobile users connectivity even while traveling at speeds of up to 70 mph throughout the entire city. Bids from vendors are planned by this month. Most vendors are expected to put forth proposals for utilizing the latest mesh networking technology. The size of this project will no doubt foster a rapid maturation of some very exciting concepts currently in R&D. It seems like we are coming a long way from the hot spot at your local coffee shop.

NO MORE ADD-ONS

And finally the third announcement came from Intel. Its latest chip, dubbed Grantsdale and priced at $40 or less, is designed to work in tandem with the Pentium 4 to provide standard PCs with powerful sound, graphics, high speed peripherals interconnects and the ability to run a wireless data network all within a single chip, dispensing with the need for expensive add-on cards and further helping to reduce the size and price of the overall package. It was shown effortlessly displaying simultaneous high-definition video streams with accompanying surround sound audio.

If you read between the lines, you can see the picture in the puzzle starting to emerge. Nonreal-time content distribution via packet switching networks, especially using a shared medium like the commodity Internet, can generate substantial transport related savings.

If you combine cheaper content distribution with the myriad solutions that continue to mitigate the last mile problem (cable, faster DSL flavors, WiFi and mesh networks) you start to tip the balance from the deterministic, schedule-driven, traditional content distribution methodologies into the non-deterministic, on-demand, customer-driven new media landscape that is already transforming the music business and that sooner or later will change television as we know it.

Which gets us to the last step, the consumer level equipment used to consume the content. Big changes are coming in that arena and I will address them in my next column. You can count on IT!