Digital locks and ownership

The anti-circumvention, or digital lock, provisions are arguably the most controversial aspect of Bill C-32. Arguments on both sides of the debate dominate the dicussion of the proposed amendments to the Copyright Act. In this article, I suggest a different way of looking at the problem of technological protection measures and an approach to legal protection of digital locks on content that respects user expectations while leaving substantial control to copyright holders.

A lot of the legal debate about the anti-circumvention provisions tends to focus on the technicalities of access- vs. copy-controls and the effect on fair dealing. But these academic issues, while interesting, don’t explain why some of my friends and family care about digital locks. I believe the furor over the anti-circumvention provisions is a result of the shift from purchase of a particular copy to licensing as the dominant method of distribution. This shift subverts the expectations of consumers. It isn’t helped by the fact that those who sell licensed products tend to bury the fact deep within their terms of use. For example, Amazon uses precisely the same language to sell MP3s as they do to sell CDs. However, the small group of people who search out their Terms of Use will find the following:

“Upon your payment of our fees for Digital Content, we grant you a non-exclusive, non-transferable right to use the Digital Content for your personal, non-commercial, entertainment use, subject to and in accordance with the Terms of Use. You may copy, store, transfer and burn the Digital Content only for your personal, non-commercial, entertainment use, subject to and in accordance with the Terms of Use.”

Similarly, the iTunes store had a “Buy” button back when they imposed even more restrictive terms on music downloads. There is a disconnect between expectation and reality. This has led to mistrust of licenses. They are viewed as ways to take rights away from consumers, rather than grant them. Just think, have you ever heard someone say “I’m so glad this EULA grants me the right to use this software”?

It has been argued TPMs have the potential to drive innovation and reduce prices. However, the real driving force is licensing; TPMs are important only inasmuch as they are required to secure the terms of the license. For example, Rhapsody charges a flat fee for access to a huge libary of music. What makes their service particularly interesting is that you can download the music: it isn’t a stream so you don’t need a constant internet connection. The core idea of the service is a license granting access to the library. The TPM is what makes it practical. Furthermore, despite heavy reliance on TPMs, Rhapsody has not been subject to the criticism faced by other stores like iTunes and Kindle. They key–and I hope I’m merely stating the obvious–is that Rhapsody subscribers know precisely what they are paying for.

For this reason, I think that the TPM debate is really fueled by expectations regarding sale and licensing. Let’s set aside for the moment the question of identifying what transaction is a sale or a license and assume that one can be sure. Where a person has made a conscious decision to enter into a license, the TPM is not likely to cause a problem. In fact, it probably enables access that would not otherwise be possible. The TPM permits the licensor to impose creative conditions or license only particular rights which affect the price paid by the consumer. Where the person has purchased the copy, on the other hand, the likelihood of unexpected and arbitrary restrictions is greater.

By arbitrary restriction, I mean that the control exercised by the TPM prevents ordinary, non-infringing uses. This may occur by design or accidentally. For example, DVDs are protected by a TPM called the Content Scramble System (CSS). Manufacturers of DVD players (whether hardware or software) must obtain licensing keys in order to access the content of a protected DVD. This requires the manufacturer to agree to obey copy controls as well as region restrictions and playback controls. The immediate effect is that the consumer is required, as a practical matter, to obey the restrictions even if disobeying would not infringe copyright. Under Bill C-32, this would cease to be just a practical issue and become a legal requirement. This is likely to lead to future problems because it places a continuing reliance on copyright holders to support the consumer after the sale. In time, technology will change, patents on DVD technology will expire, and the consumer will still rely on the DVD Copy Control Authority (DVD-CCA) to hand out decryption keys to manufacturers. The DVD-CCA may cease to exist at some point, may change its policies, or refuse to embrace a future technology. The problems get more serious when we consider TPMs that require regular access to a central authorizing server. At some point, it will no longer be economically viable to run these servers.

The solution I suggest is that a distinction be drawn between purchase and license at least in the case of access controls. This is currently a controversial issue on its own as the distinction is not always clear. However, the act could suggest a test similar to the one in the 9th Circuit Court of Appeals case, Vernon v. Autodesk. The test should require at least that the license be clearly visible and impose notable restrictions on the licensee. Such a legal distinction would have the additional benefit of requiring licensors to be explicit in their advertising. This solution would limit legally enforceable TPMs to circumstances in which they match expectations. This does not mean that TPMs would disappear from purchased copies. A barrier to easy copying may be a useful speed bump to prevent casual piracy. However, merely circumventing that TPM would not be equated with copyright infringement. A major purpose of treating circumvention as infringement is to address devices and services that can be used to infringe copyright but do not meet the standard of authorizing in CCH. However, it comes at the cost of treating non-infringing circumvention as infringement. If copyright holders made a distinction between the TPM applied to purchased copies and licensed copies, it would be clear that there would be no non-infringing uses of devices and services that circumvented the latter kind. Thus circumvention could reasonably be treated as if it were infringement.

I acknowledge that this solution is a little far-fetched. It requires some drastic changes to the legislation and in the business practices of copyright holders. Furthermore, depending on your interpretation of our WIPO obligations, it may not provide adequate protection. However, I believe it would properly address the issues raised by the switch to licensing, digital media and TPMs. It further acknowledges the notion that sale of a particular copy means giving up control over that copy. Finally, it should combat distrust of licenses. I believe that licensing will eventually provide the best value to consumers and copyright holders but it can only do so when it loses the reputation of being a trap hidden in the Terms of Use.