Douglas Crockford




2024 Appearances






Electric Communities


Flickr Photo Album



Pronouns: pe/per


Fall Focus 2004

I attended Fall Focus 2004: On-Demand Digital Entertainment, November 10-12, 2004, at the Loews Coronado Resort at San Diego, California. The conference was put on by Parks Associates.

On Demand

On Demand Entertainment, also known as Video on Demand, has long been predicted to be the next big thing in the delivery of entertainment. On Demand allows consumers to select programs whenever they want. In that respect, it appears to a high-value, consumer-oriented service. However, over the last ten years, nearly all On Demand services either failed or performed well below expectations.

Now it appears that On Demand is back in several new forms. Once again, the analysts are producing graphs that show rapid adoption and profitability. So On Demand has gone from 10 years of proven failure to the next big success (maybe). What has changed?

In a way, nothing has changed. Cable has been betting on On Demand all along. They are hoping that a large success will grow out of it previous limited success. Also, the analysts have never stopped believing. Periodically, they are forced to push their forecasts out a couple more years. One statement that was repeated many times was that being early and being wrong have the same outcome, suggesting that the earlier failures of On Demand were due to bad timing.

Some things have changed this time. Will these changes bring success?


One significant change is a better focus on the needs of consumers. The industries are now recognizing that ultimately the consumers are in charge, and that they will reject products and services that do not offer high-value and choice, convenience, and control.

The original On Demand systems were designed to compete directly with Blockbuster, a video rental business. The idea was to allow people to get the most popular movies electronically, saving two trips to the rental store. It offered convenience, but it reduced choice (because the number of titles available was very limited) and it reduced control (because you could not effectively pause, rewind, or watch again). Also, bad user interface design added a lot of frustration and confusion, which eliminated the convenience.

By being more mindful of consumer needs and attitudes, industry might do a better job.


There have been technological improvements. There have been big improvements in digital delivery systems, particularly in broadband to the home. The cable industry has invested heavily in infrastructure for On Demand delivery.

Even more significant are the increases in disc storage. It will soon be possible to sell a disc drive capable of holding all of the recorded music that has ever been recorded. This will radically change the way that we think about and value music. In the short term, DVRs provide interesting new ways to manage media in the home. There were some people at the conference who thought that the DVR will be replaced by a more capable network that is capable of delivering any content immediately. This is the strategy that the Telcos are pursuing. The Telcos are also motivated by the fear that VoIP will destroy their core business. I will have more information on Telcos after attended the Telco TV conference next week. Server technology now allows the offering of large libraries of programs.

The Internet might ultimately displace (or subsume) Cable and the Telcos. In the meantime, it provides an alternate form of delivery. It is currently making the Recording Industry obsolete. It is still unknown what its role will be in video. IP networks (end-to-end) still have QoS (Quality of Service) problems which interfere with video delivery.

There is also the PC and the Home Network, in which the Media PC becomes another kind of STB. Microsoft was presenting this point of view, which is contrary to other research. John Canning of Microsoft lectured us about the problems caused by complexity, when Microsoft is the worst when it comes to complex systems. He said that we should not design for nerds. He also said that "The PC is not the viewing surface," which is how a nerd would say "a PC can be attached to the TV set."

Search tools (such as Google) may have an important role in On Demand systems, since one of the most difficult aspects of them is knowing about and locating the programs. Search may become the Center of the Universe.

MoCA (Multimedia over Coax Alliance) is developing home networking standards, using coax for transport. This can be effective in homes that were built for cable, and in homes that already have multiple cable STBs. It can work with DLNA.

Business Models

There are still unknowns in the business model. Here are two dimensions to consider.

First, do consumers want to Rent, Own, or Subscribe? In magazines, people like to subscribe, but in books people like to own. In videos people like to rent and own, but Netflix is becoming so popular with subscriptions that Blockbuster and WalMart are now offering DVD subscription services.

The question is made more complicated in the presence of DRM, which could provide bundles of rights that do not map onto rent, own, or subscribe. I think this could be a problem because consumers do not like to be confused. Also, Bob Engel of HP said, "Consumers highly value control." Consumers might reject services that take too much control away from them.

DRM is intended to create new business models, but it may also tend to destroy other business models. William O. Leszinske of Intel said the most unintentionally funny thing at the conference. He said that there will not be a single DRM system (true), that all of the DRM systems must work together (true but unlikely), and that China will have its own DRM system (funny).

The Cable Industry is very excited about Subscription Video on Demand (SVOD). This allows consumers to watch as much as they want from a selected menu for a flat monthly charge. Early indications are that this will be much more successful than the current PPV (Pay per View) products. Cable is also going to providing Free Video On Demand. This allows consumers to watch as much as they want from a selected menu for no additional fee. A possible future is that there will only be On Demand programming, that all scheduled programming will disappear.

The second dimension is granularity. Will products be sold in bundles for a flat fee, or a la carte (one at a time). A la carte allows consumers to control their spending, but bundles can provide higher value. It is not clear which way consumers will go on this. This issue came up in the context of Cable a la Carte pricing versus tiers of service. It was suggested that a la carte could be taken to the granularity of individual programs, that you would pay individually for each program you watch.

The conference did not have representatives from the Advertising Industry. Historically, advertising is the ultimate business model after all other models fail. This was the case with radio, TV, and the Web.


Bob MacIntyre of Scientific-Altanta gave us some interesting information about his business. Currently, 80% of the Cable Industry's subscriber revenue comes from people who watch 20 channels or less. Cable's biggest margins are on HSD (High Speed Data).

Two thirds of Scientific-Altanta's revenues come from the sale of STBs. The company is vertically aligned for STB production, including developing its own ASICs. It becomes clear why they are not promoting DCR (Digital Cable Ready). STBs are much cheaper to produce than PCs, and are much more reliable. Cable STBs have a more credible DRM story than PCs. DRM works today in digital STBs.

SVOD and DVRs are greatly increasing customer satisfaction, reducing churn.


Ultimately, it is futile to invent new business models and technologies for delivering video if you do not have access to high-value video to deliver. The studios were not represented at this conference (except by Peter Lee of Disney who is a reasonable man). There were many people who deal with the Studios, and they warned that the Studios are very risk averse. They will not risk their current (or imagined) revenue on unproven technology. Some ventures will fail simply because the Studios will not release their content in a timely way. Innovative new channels generally lack the buying power to make effective deals with the Studios.

The Window system was discussed at length. Windows are periods of time in which movies are placed in specific markets and delivery systems.

Window Time
Theatrical 6 months
VOD/PPV 45 days
Pay/SVOD 4.5 months

According to Hadi Partovi of Microsoft, Consumers spend $9.2B at movie theatres, $24B on DVDs, and $52B on cable and satellite. Theatres are a small part of the revenue picture for the Studios. The theatres are a marketing device for generating demand for DVD and media sales. The theatrical release establishes the brand.

If you look at the total media market, does the new technology cause the market to grow (by giving consumers more value), or does it cause the market to shrink (by giving consumers more granularity). The amount of time people spend watching TV has been declining, although people with DVRs tend more watch more. It is unlikely that people can devote significantly more time to TV watching. Can technology make people want to spend more money during that time?

If the market does not grow, what will pay for all of the investment in new technology?

Breakdown of Shared Video Files
Pornography 42%
Other 32%
Music Videos 10%
TV 9%
Feature Film 7%
Source: BigChampagne

Most of the demand in On Demand is for pornography. The discussion is so dominated by the Studios and feature films that we forget to look at where the money actually is. There was no one from the Pornography Industry to talk about how Porn makes money.

Jim Ramo of MovieLink said that they see no demand from people aged 18-24. He believes that this is due to piracy. Others suggested that that generation is not a generation of criminals, but that they have values that are not compatible with the Studios.


Computer Games have been delivered On Demand for years. New platforms and business models are being developed. Games have a much more significant piracy problem than movies have. DRM generally is not effective for games. The Games Industry has been forced to be realistic in dealing with the problem. In the face of free distribution tools like BitTorrent, game companies are finding that the best remedy is to reduce the value of pirate content. For example:

In contrast, the Studios are adopting measures that increase the value of pirate content.

The audience for games breaks into two distinct groups:

WalMart is 40% of the retail business.

A casual game costs $50K to produce. It costs $25 to download. It can sell in the millions of copies.