Because I know very little about Australian law, when the Kazaa decision came out, I asked Open Source Law's Brendan Scott if he would please explain what it means. He has graciously agreed. His article is available as PDF also.
I learned quite a lot from the article. I confess that I was stunned to learn of a case in Australia where a library was held guilty of infringement by authorization because it had placed a copying machine next to a stack of library books. Here's a snip from the beginning of the ruling:
In the belief that
some of the copying of copyright material which is done within university
libraries, by the use of photocopying machines, amounts to an infringement of
the copyright, it was decided to commence a test case
against a university,
and for this purpose it was arranged that one Paul Brennan, a graduate of the
University of New South Wales,
should make an infringing copy of a literary
work by the use of a photocopying machine in the library of the University of
Does common sense no longer rule? What is this excessive fear of copying? The case was about books, not digital anything, so the argument that the extreme fear and consequent extreme measures are justified based on the possibility of perfect copies in the millions doesn't apply. And you thought it was getting silly in the US. When authors decide to entrap librarians in order to criminalize them, something is really and truly going haywire.
Kazaa: Time to Rethink Authorization?
~ by Brendan Scott,
Open Source Law
“It is not for the Court to reject that guidance on the basis that the particular judge
considers the result to be unrealistic and unfair. If Parliament thinks that is, indeed, the
result of applying the Act, the remedy is in its hands.”
Universal Music v Sharman Holdings at paragraph 418
1.1 On 5 September 2005, Justice Wilcox, a single judge of the Australian Federal Court, made a finding of infringement of copyright by authorization against distributors of the peer to peer software Kazaa in a suit brought a number of record companies. The judge also made similar findings against some of the directors of companies involved, but declined to make them against some employees.
1.2 The case is 80 pages long and I do not intend in this overview to cover everything it has said, nor to cover anything said in great detail. There are a number of applicants and a number of respondents in the case and, while incorrect, for the sake of brevity I will refer to them as the “record companies” or “the applicants” and “Kazaa” or “the respondents” respectively without distinguishing the corporate from the personal parties nor the infringing from the non-infringing respondents.
2. What the Judgment is About
2.1 The key to the judgment is section 101 of the Copyright Act 1968,
2 and some recent modifications to that section. Section 101 provides that authorizing a person to infringe copyright is itself an infringement of copyright. So, if you tell someone to go make a photocopy (and they do), not only can they be in breach of the Act, you can also be in breach by the fact that you have authorized their act. The breadth of authorization is demonstrated by the UNSW v Moorhouse
3 decision, in which a university library was held to have authorized a copyright infringement by a student because it located a photocopier next to a shelf of library books. The argument being that it was effectively an open invitation to infringe and in those circumstances the library did nothing to stop infringements. The judgment applies existing case law about section 101 to the particular facts of the case.
3. The Orders
3.1 Having found that Kazaa infringed, the court made orders that Kazaa either (i) agree to a keyword filtering function with the record companies which filters out searches for music against a list of key words created by the record companies, include that filtering in new versions and apply “maximum pressure” to existing users to upgrade; or (ii) when a search is made for a work which matches “any such list” search results are limited to either licensed copies, or a warning against copyright infringement.
3.2 Exactly what the practical implication of these orders will be should give cause for concern. Does Kazaa have to just pay attention to the lists that the applicants in this case put forward, or can other copyright holders also suggest key words? If the former, these orders will give the applicants a frightening degree of market power as against their competitors. If the latter, exactly on what basis could Kazaa decline to accept the key words proposed by a copyright holder?
4 If I send them my key words, must they add them to the filter? Can they exclude my suggestions if my works are not traded on their system at the moment?
3.3 What incentive is there for a copyright holder to limit the scope of words or names? Surely they have an incentive to flood the filter with all conceivable variations on the names of their sound recordings, and even of those which they have no interest. If a competitor chooses to use Kazaa as a distribution medium, what is to stop the record companies adding key words to filter out their competitor's tracks? From a less sinister viewpoint – what of innocent collisions? This is of particular relevance to me, because there is a musician who shares my name.
5 Does this mean if I put up recordings of my law lectures they can be filtered out against my wishes?
4.1 The lawyers for Kazaa have already announced that the case will be appealed, and the judge has made orders permitting appeals from the judgment, provided that they are brought and prosecuted promptly. There are two levels of appeal available in theory – an appeal to the full court of the Federal Court, followed by an appeal to the High Court, the court of ultimate jurisdiction in Australia.
5. Implications for Open Source
5.1 When determining whether infringement by authorization has occurred a court is required to have regard to: the extent of the power of a person to prevent the infringing act, the nature of the relationship with the infringing actor and whether the person has taken reasonable steps to avoid infringement.
5.2 In this case, the relevant power was the ability to provide new versions of the software and to nag people to update. The maintainers of established open source projects may well have this degree of power.
5.3 The key factor to consider for open source is whether the third criterion is met - “whether the person took other reasonable steps to prevent or avoid the doing of the [infringing] act, including whether the person complied with any relevant industry codes of practice”.
5.4 There is one word in the orders made by the judge that may make this a substantive problem for open source projects. That word is “mandatory”. The court, as part of its judgment, has required that Kazaa implement a filtering system in the software which is mandatory in the sense that a user cannot bypass it. It is almost a certainty that any litigant in the future will argue that because the court has made such an order in this case, the inclusion of a mandatory filter is such a “reasonable step”.
5.5 If this proposition is accepted, then it may well be impossible for an open source file sharing system to be legal. It is a necessary consequence of open source that any user may modify the software and, in particular, may change a “mandatory” filter to an optional one, or may remove it altogether.
6. Cui Bono? Quo Vadis?
6.1 It is far from obvious what benefit the applicants will derive from this ruling. Kazaa emerged as a result of the (successful) efforts to kill off Napster. Kazaa's design did not have some of the key elements which made Napster vulnerable to litigation. Given the incredible demand for illegal sound recordings, you would need to be very naïve to think that this judgment, or the ultimate death of Kazaa, will have anything but a transitory effect on the illegal infringement of sound recordings.
6.2 What the record companies are engaging in is a form of enforced and accelerated evolution of file sharing systems. They are implicitly wagering that there are limits on how far those systems can evolve. With each evolution they are pushing file sharing into distribution platforms with diminishing inherent controllability. To date, they have been lucky enough to have a single target operating the relevant platform. The end game of this evolution is that file sharing will be an anonymous and uncontrollable system. Today record companies have the option to work with and co-opt file sharing networks. Should they leave it too long to participate, record companies will no longer have that option. They may find in a few years that they have snookered themselves.
6.3 Meanwhile, and despite the applicants' contention that it has done so through foul rather than fair means, if their statistics are to be believed, Kazaa has created something of potentially great value to both themselves and to society. The holy grail of Microsoft, and of countless other internet companies in the 90s, was the creation of a delivery platform and associated community similar to that of Kazaa.
6.4 There is ample material in the case for record companies, should they so choose, to use in terrorizing other p2p services, or Kazaa's successor. However, this is hardly a socially valuable application of resources. In an ideal world the applicants and respondents would be a perfect match for each other. Kazaa has no interest in exactly what is traded over its network, as long as it is trading. It has expressly created a facility for the sale of DRM locked content. It provides a far more cost efficient mechanism than that currently employed by the applicants.
8 It provides a service which the applicants' customers find very valuable and, given the popularity of it, one which the applicants are apparently unable to provide themselves. It is one in which the applicants, should they choose to participate, can exercise a level of control over distribution. The applicants, on the other hand, have content which has been proven to be appropriate for the Kazaa network. Based on the public statements of the applicants around the world, they themselves are in imminent and mortal danger if they are unable to find a business response to sharing.
6.5 You don't need to be Einstein to work out why lawyers would advise record companies to spend money on litigation. Especially when, in the event of even a spectacular legal success, they have a good chance of their clients being in materially the same practical position once the next generation of peer to peer systems takes the place of their current target.
9 They can then again advise them to conduct another round of quixotic litigation. Exactly how many times do record companies need to destroy a Napster clone before they start doing something constructive about the problems they face? I don't think anyone wants to see record companies starting to treat litigation as some form of business venture. Surely it is in society's interest that these groups find a way to work together to promote economic growth rather than expending their (and the court's) time in expensive, and probably long-term futile, litigation.
6.6 The fact that they are in unnecessary conflict indicates there is something awry with the legislation. The real question is why is the legislature not looking to create incentives for these industries to work together rather than being content with legislation which thrusts them into conflict.
7. An Assault on Common Sense?
7.1 The concept of authorization in copyright law makes people responsible for the independent actions of others. It creates a positive obligation on ordinary businesses to foster the business interests of others and therefore creates a chilling atmosphere for business venturers.
7.2 An analogy to the effect of section 101 of the Copyright Act might be that if a car company (let's call them Tux Motors) happens to be aware that many people exceed the speed limit, they must take reasonable steps within their control to design their cars so that their purchasers do not speed. When they do so they should consult with every local council that sets those limits or designs carriageways. If they don't they will be liable for the speeding fines of their customers.
7.3 Such fine grained control of our actions by Tux Motors may well be possible in some future Big Brother Dystopia. At the moment, however, such a requirement would be so hard to comply with that Tux Motors might be better off declining to produce cars (or to produce them with a speed limit so low as to always be within the limits – say 15 mph). This would preserve the sanctity of speed limits, but at an enormous cost to society – the burden of compliance would drastically increase the cost of cars, would-be purchasers would be forced to buy a product which is defective by design, and purchasers would be prevented from utilization of the (legal) potential of those products. Moreover, Tux Motors would need to continually second guess what is in the interests of those third parties and what is encompassed by “reasonable”.
7.4 It is hard to see why online entrepreneurs ought to be exposed to risks which their offline counterparts are not. Indeed, in emerging industries it is more common to attempt to create a legislative environment which encourages economic activity, rather than stifles it or, as we have seen in this case, brings it into conflict with existing industries. The knowledge economy is a lot to sacrifice in the pursuit of an unattainable ideal. It may be that this aspect of the law is outdated and should be revised to bring it into line with standard commercial practice in the broader economy.
8. Some Concluding Comments on the Justice of the Case
8.1 Reading the judgment one gets the distinct impression that the court felt that there was something not quite right with Kazaa. You can notice these by the various things which are adverted to in the judgment - their Byzantine corporate structure; the incorporation of a number of companies in the corporate group in Vanuatu, a jurisdiction which prohibits disclosure of a company's details; the failure by Kazaa to lead much in the way of evidence or witnesses; where it did call witnesses, it didn't call those who the court believed had the best knowledge; of the evidence that was lead by Kazaa, the court took a dim view of some of it; Kazaa was aware its software would be used to make unauthorized copies of sound recordings; and Kazaa makes a lot of money out of advertising delivered through the software (and, therefore, indirectly out of those unauthorized sound recordings). The court also was dubious about Kazaa's expressed intention to promote the sale of legitimate recordings through its Altnet or “Gold File” system.
8.2 It may be that these all have an adequate and appropriate explanation. However, if you were so predisposed, there is ample basis on which to argue that the overall justice of the decision as between these litigants is appropriate. Whatever your view on the justice of the case, it does not turn a law which reflects poor policy into one which reflects good policy.
1 Universal Music Australia Pty Ltd & Ors v Sharman License Holdings Ltd & Ors  FCA 1242 (5 September 2005). The text of the judgment is available from: http://www.austlii.edu.au/au/cases/cth/federal_ct/2005/1242.html
2 The text of the section is available from: http://www.austlii.edu.au/au/legis/cth/consol_act/ca1968133/s101.html
3 UNIVERSITY OF NEW SOUTH WALES v. MOORHOUSE  HCA 26; (1975) 133 CLR 1 Text available from: http://www.austlii.edu.au/cgi-bin/disp.pl/au/cases/cth/HCA/1975/26.html?query=title+%28+%22+%20+moorhouse
4 That is, if it is reasonable to implement a filter in favor of one group of copyright holders, why would it not also be reasonable to implement a filter in favor of others.
6 Edward Felton has some comments on the feasibility of filtering here: http://www.freedom-to-tinker.com/?p=892
7 These words are from section 101(1A)(c) of the Copyright Act 1968, the section under which Kazaa was found liable.
8 That is, burning and shipping CDs, accepting returns, destroying or redeploying excesses. All of which are incredibly expensive.
9 I note in this respect Michael Geist's comments that Kazaa had apparently already become passe even before judgment was handed down, holding only 10% of the peer to peer market - http://www.taipeitimes.com/News/worldbiz/archives/2005/09/05/2003270490
About Open Source Law
Open Source Law is a boutique legal practice based in Sydney. The practice specializes in ICT and IP law with a special focus on open source related legal issues. Open Source Law did not act for any of the parties in this case.
(+61) 414 339 227
email: inquiries at opensourcelaw.biz
http://www.opensourcelaw.biz This article is under a Creative Commons license - Attribution-No Derivatives.