So today, senior Australian Federal Court judge Justice Wilcox handed down his decision in the trial of the Kazaa case. In this case, over 30 applicants – in essence, copyright owners – sued the companies and individuals involved in providing Kazaa software. They alleged all kinds of things, but the essence of the case is this question:

By providing P2P file-sharing software (and through all their other activities), did the respondents (Sharman companies, Altnet companies, and assorted individual directors) authorise the undoubted copyright infringement done by the users of the software?

The result?

  • The Sharman companies did authorise infringement. They did not engage in other forms of infringement/illegality alleged by the copyright owners (including direct infringement, conspiracy, misleading conduct under the TPA or unconscionable conduct);
  • The directors/head honchos in Sharman are liable for authorising infringement too;
  • Some of the other parties avoided liability.

The Australian are calling it ‘The Day the Music Died’ (a bit odd, given that the market has, of course, moved on from the Kazaa system). Below are some initial thoughts.


What is the immediate result of the case?

What Wilcox has done is make certain orders:

  • declarations of infringement;
  • an injunction against continuing authorisation of infringement, stayed for 2 months in light of the next order;
  • instructions to avoid this injunction shutting down P2P software entirely: the court thinks that modifications should be made to the technology, and has instructed the parties to agree to a protocol or will approve one himself – the court has also outlined some steps the judge thinks could be done;
  • an order for pecuniary relief (ie, damages) – the amount reserved for determination at a later hearing;
  • an order that the respondents (Sharman parties) pay 90% of the copyright owners’ costs (this would be on a party/party basis, so wouldn’t be the full amount expended by copyright owners. Still, it will be significant).

What, the judge wants to re-design the technology of Kazaa?

Yes, Justice Wilcox has decided that he is quite prepared to get into the nitty-gritty of the technology. He has ordered the parties to agree a protocol whereby filtering (of some kind) may be imposed into the software.

Thus the title of this post: this is a brave move indeed. Given the experience over in the US in the Napster litigation, where similar attempts by a trial judge led to much ongoing disputation about the form of orders that only went away when the litigation collapsed under its own weight, I’m surprised that any judge would want to get into this. All the more so given the traditional reluctance of the courts to grant injunctive relief that will involve forcing parties into an ongoing relationship, which will require ‘ongoing supervision’ by the court of the order (Patrick Stevedores v MUA (1998) 195 CLR 1 at 46-47).

Wilcox’s orders necessarily involve both – the ongoing relationship of the parties and the ongoing supervision:

  • an ongoing relationship is required because it seems that it will be the copyright owners who must provide, and update, the list of stuff to be filtered. They will have to provide that to the respondents;
  • ongoing supervision because there will be inevitable disputes (if the past relationship between these parties is anything to go by) let alone the settling that must occur.

I am reminded of a comment by Lemley and Reese in an article not that long ago:

‘there will always be disputes over what content is infringing. Copyright law is full of gray areas,and copyright owners have a history of trying to extend the law beyond its bounds. A court that decides to stop infringing content while letting the rest of the service continue will either have to enjoin all infringing content in advance (in which case no rational defendant will operate their system at all, for fear of going to jail for contempt) or will be signing up to resolve an endless series of oversight disputes about particular cases’

In effect, Wilcox wanted to split the baby. He didn’t want to absolve Kazaa in this case. He wanted to put a stop to some of the infringement. But he also wanted to make it clear that P2P file-sharing could go on. So he tried to tread a middle path. In my view, the middle path here is unlikely to be a viable one. As I mention below, I think the better approach might have been to frame the rule on liability in a way that would catch the bad actor but avoid imposing liability on all technological innovators.

One thing to note however: if you look at the judgment itself, these orders are provisional: Wilcox J has said that (at para 519):

‘I have formed some views about the appropriate form of injunctive relief and have drafted some orders. It is convenient immediately to make the orders. However, I will do so on a provisional basis, in the sense that I will be prepared to reconsider the form of the orders, if so requested by any party. I will not receive further evidence in rleation to the nature and form of the orders.’

What are the damages likely to be?

Who knows? This will be determined later, if necessary. A couple of points to note however. Australia does not have statutory damages (unlike the US) – the copyright owners can’t argue that they should just get a fixed amount ‘per infringement’.

On the other hand, Australia does have additional damages for flagrancy of infringement. It is not clear whether Wilcox J would be prepared to make such an order.

Is this case anything like the Grokster case?

People interested in digital copyright issues of course recall the recent US supreme court decision in Grokster. Like Kazaa, Grokster was about whether a provider of P2P software could be liable for the copyright infringements of its users.

There are some important differences though. In Grokster, the US Supreme Court had to decide whether the 9th Circuit was right to uphold a trial judge’s decision granting summary judgment to Grokster and other P2P software providers. Grokster has not, yet, involved a finding that the P2P software provider IS liable – but they could be.

Unlike Grokster, however, Kazaa is a judgment after a full trial. In Grokster the lower courts decided that under existing US precedent, there was in effect no way that the software providers could be liable for copyright infringement of the users. The Supreme Court held that that was wrong, and sent the matter back to lower courts for more detailed consideration.

In Kazaa, however, we have had the full trial. In fact, we have had quite extensive evidence-gathering – despite the grumbles of the judge about the incompleteness of the information presented. This is a final judgment on liability – subject to any appeal (to the Full Federal Court, or, following that, to the High Court). The judge has indicated he would be minded to allow an appeal at this point, if the parties want to pursue it (they need leave because not all the issues are dealt with yet – damages are not yet decided).

One other point. Wilcox J specifically noted that he had thought about Grokster, but that it wasn’t relevant. According to Wilcox J,

there was a question in my mind as to whether the Supreme Court’s decision [in Grokster] provided any guidance to the resolution of this case. … I invited the parties to comment about that matter. They all did so. Their comments confirmed my impression that the differences, both factual and legal, are such as to render Grokster of little assistance to me.’

Aren’t P2P software providers ok provided their technology has non-infringing uses?

No, that isn’t the law in Australia.

As I said above, the issue in this case is whether the Kazaa parties authorised infringement by users. ‘Authorisation’ is the Australian equivalent to secondary liability under US copyright law (contributory/vicarious liability – we don’t draw the same distinction) (see my background website here). But that doesn’t mean that tests like the Betamax test, or its re-working in Grokster, are relevant here.

Under section 101 of the Copyright Act, a person can be liable for direct infringement (doing any of the acts which are part of the exclusive rights of the copyright owner) or for authorising the doing of those acts. In Australian law, the classic formulation is that a person ‘authorises’ infringement when they sanction, countenance or approve infringement (UNSW v Moorhouse). The factors a court has to consider are listed in section 101: a court must consider:

  • the extent (if any) of the person’s power to prevent the doing of the act concerned
  • the nature of any relationship existing between the person and the person who did the act concerned
  • whether the person took any other reasonable steps to prevent or avoid the doing of the act, including whether the person complied with any relevant industry codes of practice.

The question is how these factors apply to a provider of technology, like a video-recorder or P2P software.

For a long time, actually, the risk of liability posed to a provider of technology that can be used to infringe copyright has been unclear. The few cases we had (Hanimex in particular) have been unclear. No one has tried, here in Australia, to sue a provider of a video recorder. There are obiter comments by some judges in the High Court and lower down that such a person wouldn’t be liable, but things have been unclear.

Indeed, leading copyright authorities in the common law world have criticised Australian law on authorisation for being too copyright-protective. As Laddie, Prescott & Vittoria note, there are two possible meanings of ‘authorise’:

  1. The ‘wide view’: ‘sanction, approve or countenance’ infringement; and
  2. The narrow view: authorisation means a grant, or purported grant of the right to do the infringing act.

LPV say that the narrow view is better, but the wide view is the one that prevails here in Australia. The Canadian Supreme Court also criticised the Moorhouse decision in CCH.

So what happened in this case: why did the Court hold that the Kazaa parties authorised infringement?

The court offers its own little summary of why Kazaa is liable for authorising infringement. Three obvious facts are the starting point:

  1. there is widespread infringement on the Kazaa network, which
  2. The Kazaa parties undoubtedly knew about and
  3. The Kazaa parties provided the facilities (the software) for that widespread infringement.

These facts, however, would not be enough on their own. Mere knowledge is not enough. Mere provision of facilities is not enough (indeed, there is a specific exception that prevents a ‘mere’ provider of facilities from being liable – section 112E of the Act). The ‘clinchers’ were in some of the facts of this particular case:

  1. The warnings on the Kazaa website about infringement were ineffective to prevent, or even substantially to curtail copyright infringement;
  2. There were technical measures that the respondents could have used to reduce, though not to prevent, the sharing of copyright files, which the respondents did not do (and which was not in their financial interests);
  3. The financial interests of the Kazaa parties are to maximise, not minimise, music file-sharing:
  4. ‘The more shared files available through Kazaa, the greater the attraction of the Kazaa website. The more visitors to the Kazaa website, the greater its advertising value and the higher the advertising rate able to be demanded by Sharman. And what is more likely to attract large numbers of visitors to the website than music, especially currently popular ‘hits’?’

  5. The Sharman parties had included on their materials active exhortations to users to increase their file-sharing, including by a ‘Join the Revolution’ campaign that criticises record-companies for opposing P2P.
  6. ‘It seems that Kazaa users are predominately young people, the effect of [Kazaa’s] web page [with the slogan, ‘Join the Revolution’] would be to encourage visitors to think it ‘cool’ to defy the record companies by ignoring constraints’

What’s really at the heart of the finding of authorisation, then, are some very basic points: in essence, the court:

  • accepted that various bits of evidence pointed to quite a few things that indicated that the Kazaa parties had a degree of interaction with, and ‘control’ over, the conduct of their users; and
  • in particular, the court had identified to it two things that Kazaa could do to make infringement harder: keyword filtering, and ‘flooding’ of search results with copyright warning files.

What brought Kazaa undone in this particular case was the way that their business model worked. You see, searches using Kazaa software returned two types of results:

  • blue files (free shared stuff) and
  • gold files (‘premium’ DRM-protected type content, supplied by Altnet people).

The mere fact that gold files could be provided in response to queries meant that something could be done to those queries.

Kazaa wanted to have their cake and eat it too. They wanted to say they weren’t controlling the activities of their users on the one hand, and yet on the other sell the ability to engage with users directly, and provide targeted content, as the ‘selling point’ of the technology. This didn’t convince the judge. And while the judge by no means accepted the extensive evidence and arguments offered by the applicants that Kazaa could engage in direct, immediate, and almost complete control of their users’ activities, the judge did accept there were certain things that Kazaa could do to make copyright infringement a whole lot more difficult.

Kazaa could put in a filter, like its adult filter, except make it the rule (ie mandated): using things like copyright owner names/performer names, to prevent files getting through. Notably, the court did not accept that filters could be so easily avoided (as they were with Napster!) that they would be pointless. As Wilcox accepted:

‘a keyword filter system that was tied to the title of the sound recording or the name of the artist would not be 100% effective. However, counsel for the applicants argued this was no reason to reject the view that the respondents could have used this technique substantially to inhibit copyright infringement’

Alternatively, Kazaa could use the ‘annoyance’ tactic ‘flood’ the results page of a Kazaa search using a copyright owners’ stuff with empty gold files. Wouldn’t this affect users of Kazaa overseas? Well, yes, but that’s immaterial:

‘The point was made by counsel for some of the respondents, and in relation to any filtering proposal, that the mechanism could not be made specific to Australian users; it would also constrain the access of non-Australian users to the copyright material included on the relevant copyright-owner’s list. That is so, but I cannot regard that as an objection to a filtering mechanism. If it is reasonable for the respondents (or any of them) to adopt a filtering mechanism in order to avoid an infringement of Australian copyright law, it is immaterial whether that step would also have been necessary in order to avoid infringement of the copyright law of some other country.

Actually, I have to say that if Kazaa did do these things, it would probably reduce copyright infringement using the Kazaa network, because they truly would drive users away. In a Wilcox ordered world:

  • Kazaa will have to put in annoying things like keyword filters and/or flood search results with empty copyright warning type files; and
  • In order to force users to accept the updates with these annoying features, Kazaa will have to apply ‘maximum pressure’ using ‘dialogue boxes’ to convince people to update. In other words, really annoying and constant pop up boxes demanding the user upgrades.

Gee, how long do you think Kazaa – already less popular – will last once it puts these in place?

So does that mean if I’m not doing what Kazaa did – if I’m not involved in users’ search results – then I’m ok?

Nope. There is no clear, bright line rule in this judgment, unfortunately. Technology innovators are given no clear rules, if you ask me, on what to do to avoid liability. No money quote. None at all.

The court points to four key facts that support a finding of infringement here:

  1. ineffective warnings;
  2. failure to take steps to reduce infringement
  3. active exhortations on people to share; and
  4. the fact that it was in the financial interest of Kazaa to encourage more sharing of copyright-protected stuff.

What is not clear is whether the absence of any of these factors would be enough to avoid liability. The judge does not make that call. Nor does the judge make a call on whether he accepts the broad view of authorisation being pushed by the copyright owners in the case. What the judge seems to be saying is that all you can do is weigh up all the factors in an individual case.

Of course, the judge did not need to offer a general ruling – he only had to consider this case. But the absence of any real ‘test’ for providers of technology puts Australia in a continuing position of uncertainty. And frankly, it will likely chill investment/innovation. Bright line rules are good for investment in innovation. We don’t have that here.

In Conclusion

In my view, it was always going to be the case that Kazaa itself went down, on the kinds of facts that we see in this case. What I was hoping was that the judge would find a way to frame a rule relating to the meaning of ‘authorisation’ in copyright law, so that it caught ‘bad actors’ without generally chilling innovation.

The court has not done that. The court has caught the bad actor but provided no guidelines for the good actor. The court has remained true to the strict letter of the legislation, which refers to ‘factors to be weighed up’. This could chill technological innovators of the good type because there aren’t any bright line rules to follow.

But most immediately striking to me is just how brave Wilcox is. In splitting the baby, and trying to get into technological design, I fear that the judge has let himself in for a helluva fight. And it’s not like he didn’t know that: he saw the litigation as it went on. I fear the dramas will continue as parties fight over orders. We are back in Napster territory again.