DSL

DSL Prime News: Video on Demand

DSL Prime reminds readers that a broadband network, in the 1999 definition current in Washington, will not support video on demand, in spite of impressive technical innovation from companies like Elastic Networks.

by Dave Burstein DSL Prime [November 21, 2001]

Viva Verizon Video rides again
Hundreds of thousands of set top boxes on the way
Downloading your e-mail faster is great, but the mass deployment of DSL requires something more compelling. The Third Internet is at hand, fast enough to watch. Verizon knows this, announcing a major deal last year with Enron and Blockbuster for movies. That fell through, because Hollywood doesn't want a middleman to take a cut of the profits.

Variety reported Verizon Avenue (the basement DSLAM division) was shopping for a major film deal to go wide the past summer—a premature report because neither the equipment nor the business plan were ready.

Elastic Networks believes the problems are solved, and a very large announcement is close.

Hollywood is demanding so much for film rights that movies on demand will be a marginal contributor to profit for most DSL rollouts. It's a legitimate service with an appropriate niche, but I'd be amazed if the return is interesting.

Intertainer is now live, delivering video on demand to dozens of American cities. Sign up at Intertainer.com to see what's possible.

D.C. must watch speed and quality
Historic failure possible if Moore's law ignored
DSL Prime strongly supports the emerging compromise in DC, assuming pricing and other terms emerge that are fair to competitors. But we urge everyone to watch closely the details, ensuring that the rules specify service that is fast enough to deliver the Internet we all know is possible.

In particular, everyone from Mike Powell through the telco CEOs want the net fast enough to offer video alternatives, and the technology has already advanced to make high-speed, non-blocking remote terminals at comparable costs. Unlike PCs, replaced every few years, telco equipment is likely to stay in the field for a decade or more, so a failure now will hold the Internet, and the economy, back for many years.

Technologists know Moore's Law of improving performance, and use that to plan ahead. Lawyers know precedent, and hence often fight the last war. Video streaming capacities, bandwidth costs, PC processor speeds and the like can be reliably predicted to improve dramatically over the decade, and we need to ensure we are not creating future bottlenecks. In this case, the difference in cost is trivial, and the smart telco would choose high capacity units unless their time horizon is less than three years.

Speed alone is insufficient, because unreliable service would cripple delivery of video and especially voice over DSL. Better units avoid the "webhog" problem by reducing or eliminating oversubscription.

Otherwise, congestion would cause dropouts, unacceptable in voice calls and TV viewing. Quality service, on the other hand, would allow competitors to offer 8 or more voice lines on a single copper pair. Business plans for competitors require that voice capability, so anything less than high quality, non-blocking service is crippling to the competitive future.

Verizon's "Tauke Plan" specifically suggests that competitors connect at the central office rather than the remote terminal. The resulting regulations should require that connection be as effective as if they were connected directly to the wire. This means they get reliable service at a flat cost from lite speeds of 640K to full rate ADSL speeds of 7-10M, as they would if the wire went directly to their DSLAM.

The price should be fair, allowing the telco an assured and predictable profit, but our review of the actual hardware suggests modest pricing is practical without sacrificing quality.

The difference between unreliable slow service (oversubscribed at 500K) and video/voice quality (non-blocking at 2 meg) is about ten cents per month in the original volume/deployment estimates of Project Pronto or Verizon's similar plan.

It's stupidity to build a network for 1999 Internet traffic rates.

We are journalists, not investment advisers; invest at your own risk and do further research.

Copyright 2001 Dave Burstein.
The DSL Prime Newsletter is reprinted with permission.

"The power of the printing press belongs solely to those who own the presses"
—A.J. Leibling

The Internet is the cheapest printing press ever invented.

Related articles:
	[Nov. 7, 2001]	FCC National Broadband Policy
	[Nov. 7, 2001]	AT&T Broadband Hikes Cable Rates
	[Oct. 17, 2001]	DSL Prime Editorial: Time to Deal in D.C.