oreilly.comSafari Books Online.Conferences.
Articles Radar Books  

Stop Acting Like the Entertainment Industry!
Pages: 1, 2

Contracts and the Service Model

Service contracts usually offer the buyer only one method of recourse: walk away. If you do not like the service, then there is only one choice. Both the ReplayTV and TiVo units require that the owner subscribe to a service for them to function. Without service, both devices are doorstops. If the company goes bankrupt, or simply makes an unacceptable change to the service, the only option is to cancel the service and consider the investment in hardware to be a sizeable loss. My quarrels with the service model persist even if the service has no recurring monetary cost because the requirement for service turns ownership into rental, and I have no wish to be an indentured user. Service turns ownership into rental, and I have no wish to be an indentured user.

If a product company goes under, the downside is not necessarily all that large. Electronics tend to be fairly reliable. Although it is annoying to lose the warranty backing the device, a product can continue to function without continued support. If a service company goes away, the service stops. DVRs blend the worst of both worlds. The initial cost buys a device that stops if the associated service does, eliminating any residual value. Without service, you might as well not even own the hardware. Tying the existence of a product's value to the fortunes of a company is a mistake. The underlying model is that you are buying the right to rent the device from SONICblue or TiVo, with your usage restricted by the terms of any applicable service agreements.

Restrictions imposed by user agreements have succeeded in achieving Hollywood's dream. To use the device, you must own the service. To own the service, you must sign the contract. Therefore, your right to use a device that you have been "sold" is restricted by an agreement that gives the manufacturer the right to make changes to the device's software and the service agreement at any time. What happens if the manufacturer decides to remove a neat feature that originally was one of the main attractions of owning the device? Your only recourse is to cancel the service and consider your initial cost a loss.

The cost of the DVR starts around $200. This is well above the cost of a good VCR, though DVRs do have significant additional features that could be used to justify the higher price. VCRs, although limited compared with DVRs, do not impose ongoing additional costs or require restrictive user agreements. DVR service has a nominal monthly fee, or a large up-front fixed "lifetime" fee. The lifetime subscription costs the same as 25 months of service for ReplayTV and 19 months of service for TiVo. Given the current financial condition of both SONICblue and TiVo, it is not clear that either of them can survive for the long haul, or even long enough to break even on a lifetime service purchase.

With unregulated monopolistic service providers, it is frequently a safe assumption that service fees will rise, usually faster than inflation, once a market is established. If prices rise "too high," customers have no recourse but to cancel the service. Given that canceling the service means writing off the initial hardware cost, users are likely to be reluctant to cancel. The best analogy that comes to mind is that of cable companies, which have little competition to speak of and know that most customer threats to cancel are empty. Is the customer disservice model of the cable company the future of the DVR industry?

Furthermore, DVRs have already managed to raise the specter of privacy violations. Though both SONICblue and TiVo have privacy policies that restrict the gathering and use of personal data, they are only policies that can be changed without notice. If you disagree with the change, your only recourse is to cancel the service and lose the right to use the DVR. It is certainly possible to imagine a scenario in which the software was modified to report on your viewing habits. Major content firms won a court order to produce viewing data for ReplayTV subscribers in May 2002. Though the order was later overturned within a month, my concern will never go away. VCRs were initially established as a viable product because they enhanced the privacy of users.

Some Unsolicited Advice

As SONICblue and TiVo said in their statement burying the patent hatchet last month, their major need is to grow the DVR market. As a practical matter, then, they need to eliminate the objections people have to purchasing DVRs and get them in as many households as possible. One of the major problems the DVR has faced is that its flexibility has made it hard to explain to consumers. In an October news story, an industry analyst noted that "people really need to have hands-on experience with the device in order to understand what it can do for them." By taking steps to address concerns, more people might be willing to risk purchasing a DVR.

1. Separate the product from the service

DVRs have two main functions. They record programs for the purpose of time shifting, and they have "personalization" features that allow you to watch TV on your terms by recording shows by name, following them around the schedule, and even suggesting programs that might be of interest based on your inputs or your recording history. The two functions are complementary, but separate.

The base DVR functionality of time shifting TV competes with VCRs, and that is the way that most people see the DVR at first. The cost of the DVR device should reflect its competition in the marketplace, as well as the static nature of the time-shifting task. Caveat emptor creates a vicious cycle. An "orphaned" DVR without service has no value because it lacks the ability to record anything. The basic recording function is tied to the service, which depends on the financial health of the company. I am concerned about buying a device that would be unable to do even the basic time-shifting without service, so I have not purchased one. With lots of buyers like me, there are not many subscribers to the service, which keeps the company on shaky financial ground.

To break the cycle, DVR companies could guarantee at least a VCR level of functionality for as long as the hardware runs. I currently record programs on my VCR by specifying a channel, start time, and stop time. Give the buyer VCR-like functionality out of the box, but with the DVR advantages of a longer recording time, buffering of recorded programs, the ability to watch the start of a show before it is over, tapeless operation, room-to-room streaming, and so on. The DVR advantages should enable the product to command a price premium over VCRs.

This is not to say that a subscription-based service should not exist. A service is the correct method of providing personalization features. It is labor intensive to compile the schedule that underpins personalization features. To pay for the ongoing costs of keeping the program guide and personalization up to date on an ongoing basis, the company will need an ongoing revenue stream. The key is that the premium service provides extra functionality and is not used as a means of controlling users. In essence, let the service be the "high level" programming interface to the DVR. You could have a basic low-level programming feature set ("record from 8 to 9 on Tuesday"), or you can pay extra for a high-level programming feature set ("record Buffy the Vampire Slayer, as well as anything starring John Cleese"). The former feature set can be programmed at the factory, while the latter requires an ongoing effort to maintain program guides to function.

Removing the airtight integration between purchase of a DVR and the purchase of service does leave the question of how to sell the service. Many DVR users swear by the service, even though they didn't expect to find it compelling at the time of the purchase. Try using the time-honored business tradition of giving away a sample. Let consumers buy the basic DVR, and give them a reduced cost or even free trial to the premium service. If the service is truly compelling, it should be able to stand on its own without being tied to the hardware purchase.

2. Open the protocol specification for service interaction (or license it to create a competitive listing market)

Other than being built into the business model, there is no reason to assume that the TV listing service needs to be provided by the same company that provides the hardware. In other mass markets, this is not the case. Mobile telephone service is provided by a set of voice carriers (AT&T, Sprint, Verizon, etc.) that own the end-user relationship, but the end user's hardware is provided by companies with the very different expertise of building products (Nokia, Motorola, SonyEricsson, etc.).

SONICblue and TiVo would benefit most from moving boxes into homes and establishing the market. DVRs are now under the looming threat of PCs running Microsoft Windows XP Media Center. Though PCs are more expensive, they are also more easily networked and can run without the requirement for subscriptions. In a news.com story this August, Richard Doherty of the Envisioneering Group said that "[t]he PC-DVR market can take off much quicker than the subscription-DVR market because on the PC device it would be subscription-free." Creating a competitive market for "listing services" would establish several players in the market, enable different models for supporting the personalization service, perhaps even create several different types of personalized services, and substantially reduce the perceived risk of company failure.

3. Escrow hardware specifications and software

Admittedly, I am not an expert on the ReplayTV or TiVo business models, and I can certainly understand that subscription revenue is an important component of the business. However, that very same model raises the specter that if the SONICblue or TiVo fail, I will be stuck with an expensive device that does nothing.

Another way for DVR vendors to prove to customers that the investment is protected is to demonstrate that the device will be always be useful on an ongoing basis. One obvious way to do that is to release programming specifications for the hardware and run the software as an open source project. However, this might require an extensive restructuring of the company, especially because it would necessarily remove the secrecy of the communication protocols used by the mandatory service. A middle-of-the-road alternative is to promise that if the company ever stops providing service, programming specifications and source code will be made available under standard open-source licenses, and intellectual property rights will be licensed to any projects using the code. Although many users would be unable to use such detailed technical information, it would nonetheless give me confidence that serious programmers would be able to take up the banner if the corporate entity were to fail.

This course of action is probably easier for TiVo to take, since TiVo's DVRs run the PowerPC Linux kernel, and TiVo has made its modifications downloadable, as required by the GPL. ReplayTV runs a custom operating system built atop VxWorks, a closed-source real-time operating system.

The important point about an "escrow" type arrangement is that it does not require any changes to the current business. If the company becomes profitable, the agreement is never invoked. It is only if the service-is-required business fails that anything happens. If I were running a business, I would much rather leave behind useful hardware and prevent ill will on the part of my customers. I hope that the executives at SONICblue and TiVo would be willing to make the same statement.

4. Give users control over the software

Part of the required service is bug fixes and upgrades, and one of the supposed high points of the service model is that your software is kept up to date without continuous user intervention. Many other consumer electronics products are supported with a different model in mind. For example, digital cameras are designed to replace film cameras. Manufacturers make money by selling the camera, and generally release firmware updates on the Web. Updates are something that consumers get for being a camera owner.

By way of comparison, DVRs replace VCRs. (While they do offer advanced features and functions, they serve the same purpose: time shifting TV programs.) But VCRs never need to "phone home." They cannot report what owners watch. Best of all, manufacturers cannot remove features from VCRs. Interestingly enough, my VCR also has the much-maligned "commercial advance" feature that has caused legal headaches for SONICblue. SONICblue's terms of service would give it the right to settle litigation by removing offending features from my DVR without my consent. They could settle litigation by deliberately removing software functionality and forcing the upgrade on me. Sadly, this scenario is not at all farfetched. SONICblue removed functionality in an April 2001 software "update".

A willful and intentional downgrade of software functionality is not a legal argument. After all, SONICblue's legal agreements give them the right to do just what they did. Customer outrage forced a quick retreat, but that does not change the fact that the company crossed a line. Service providers should not remove features without consent of the customers. The company described the software update component as a way of making life easier for its users by freeing them from the need to manage software versions. Now that the company has proven it will remove features of its own accord, no reasonable user can trust them to manage subscriber software versions responsibly. Since no legal form of redress would exist years after the end of any return or warranty period, the only way to express displeasure with such service terms is to refrain from subscribing.


I have not purchased a DVR because I do not want to "buy" (is it really buying if you need to have a subscription for it, to do anything with it?) a product that I do not have any control over, or even an ownership interest in. For a DVR to deliver the equivalent functionality to my VCR, I must depend on a continuing subscription provided by a single source that is free to change any terms or price at any time.

DVR vendors have taken something that should be freely usable after purchase, the right to time shift programming, and wrapped it into a mandatory service that gives them the right to restrict the buyer's use of the product at any point. It is an act that the movie studios and TV networks would be proud of. The product is not useful without the corresponding service. Service requires that users accept the service agreement. The service agreement gives you the right to change code, disable features, and alter the product beyond what buyers thought they were purchasing, so they don't really own it. They just rent the DVR at your sufferance. Is it any wonder that consumers are staying away?

Matthew Gast is the director of product management at Aerohive Networks responsible for the software that powers Aerohive's networking devices.

Return to OpenP2P.com.

P2P Weblogs

Richard Koman Richard Koman's Weblog
Supreme Court Decides Unanimously Against Grokster
Updating as we go. Supremes have ruled 9-0 in favor of the studios in MGM v Grokster. But does the decision have wider import? Is it a death knell for tech? It's starting to look like the answer is no. (Jun 27, 2005)

> More from O'Reilly Developer Weblogs

More Weblogs
FolderShare remote computer search: better privacy than Google Desktop? [Sid Steward]

Data Condoms: Solutions for Private, Remote Search Indexes [Sid Steward]

Behold! Google the darknet/p2p search engine! [Sid Steward]

Open Source & The Fallacy Of Composition [Spencer Critchley]