There's been a bit of talk recently about getting internet service providers (ISPs) involved in the enforcement of copyright law. The federal Attorney-General and Minister for Communications recently released an Online Copyright Infringement Discussion Paper in the belief that:
even where an ISP does not have a direct power to prevent a person from doing a particular infringing act, there still may be reasonable steps that can be taken by the ISP to discourage or reduce online copyright infringement.
Exactly what might be "reasonable steps" and how they might be funded are among the subjects up for discussion. Critics fear that it means turning ISPs into copyright police.
Before evaluating any new steps, it's worth recalling digital copyright measures that have been implemented before -- to see what worked, and what didn't.
Software makers struggled with illegal copying of software long before the media industry came up against the internet. Beginning in the 1980s, software makers developed a variety of schemes seeking to prevent people from making copies of software without paying the original software maker.
By the 1990s, digital audio technology had progressed such that listeners were able to make high-quality copies of music. Like the software industry before it, the music industry turned to technology that sought to prevent CDs being used with copying devices.
Since CDs were not originally designed to prevent copying, most schemes violated the original CD specification in some way. This meant that copy-protected CDs could be unreliable and even damaging.
In the most infamous case, Sony's Extended Copy Protection system was found to install a 'rootkit' -- a hidden piece of software usually associated with viruses -- on computers.
A digital watermark is a signal inserted into a media file that cannot be heard or seen by humans, but can be recovered by a computer.
Watermarks were supposed to prevent or deter copyright infringement in various ways, but the nearest they came to commercial implementation was through the Secure Digital Music Initiative (SDMI).
SDMI went on hiatus in 2001, citing a lack of agreement on the suitability of the available technologies. Watermarking technology probably just didn't work well enough: watermarks were either too easy to remove, or too audible to listeners.
Digital rights management
Digital rights management (DRM) schemes allow sellers of media files to associate those files with a licence, which sets out what the buyer can and can't do with their purchases. In principle, DRM schemes have several advantages over simply dictating to users that they shall not make copies:
- sellers can implement business models that aren't based on the old sell-one-copy model. It is possible to write licences that support subscription models, freemium models, viral models and others
- licences can permit certain legitimate uses of copying, such as copying a file from a media collection to a portable player
- media players could be built to support licensing from the ground up instead of trying to retrofit an existing technology.
Nonetheless, DRM technology enforced restrictions that buyers found arbitrary and inconvenient. Competing DRM technologies also meant that files bought from a service supporting one technology might not play on devices that supported another technology.
Later DRM schemes attempted to improve flexibility and interoperability using ideas such as the 'authorised domain', but DRM had probably already worn out its welcome by the time these improvements became available.
A rights locker is an online database in which a buyer can store a record of his or her right to use a media file.
The idea has been around since at least the late 1990s, but obviously online lockers could only become convenient once internet access was also convenient.
Over the past few years, the movie industry has been rolling out a rights locker scheme under the name Ultraviolet. Ultraviolet claims to have around 17 million users and counts around 80 movie studios and technology companies as members (but not Disney, which went with its own locker called Disney Movies Anywhere).
All of the systems described above seem doomed to chronic unpopularity. After all, what music listener or film viewer is going to say, "I need a system telling me what I can and cannot do"?
Enter the intermediaries. For governments and copyright owners, it's a lot easier to police copyright at the level of ISPs and content-sharing sites since (a) there are fewer of them and (b) they're much bigger targets.
The most obvious way in which intermediaries could help out copyright owners is to install content-filtering technology. YouTube uses a system called Content ID through which copyright owners can ask Google to check uploaded videos against a database of copyrighted works.
Some argue that content-filtering technology might ultimately be the least expensive way of enforcing copyright law. Installing content filters might incur a cost, but this cost is less than the cost of installing rights-management technology everywhere else, or of pursuing large numbers of individual infringers through the existing court system. (But I'm not aware of any empirical studies of these costs.)
Whatever the costs of filtering, critics object to the idea that intermediaries be made into copyright police. Copyright owners might respond: we tried policing individual users, and look how popular that was.
Nicholas Sheppard was formerly employed using research grant funding for the investigation and development of copyright protection technology.