LawFont.com

Coverage today of a judgment, handed down yesterday, in the Kazaa proceedings. For those who joined us late (are there any of you?), Kazaa (P2P Software provider) has been sued for authorising infringement of copyright by users of the P2P file-sharing software. Justice Wilcox handed down judgment last year, holding they were liable for authorisation, and an appeal from that judgment was heard in early February. Judgment in the appeal is reserved. In the meantime, however, there’s proceedings going on for contempt, because the trial judge did not stay his injunction pending the appeal. That is, Kazaa was ordered, in the meantime, to take steps to stop authorising infringement. And there’s a live question as to whether they’ve done enough. The judgment raises some really interesting questions about contempt. More, much more, over the fold.

First, some background.

At the time that Wilcox J handed down judgment, he made a rather interesting set of orders. The judge made various declarations, finding there was authorisation of infringement, but of particular interest were the orders about what Kazaa had to do about it. Justice Wilcox ordered:

So this is the main order: Kazaa must stop authorising infringement of copyright. Note the generality of this order. Any change of behaviour by Sharman would have to be assessed against the criteria of whether it constituted an authorisation or not. Such orders are not uncommon in IP cases, although there are plenty of statements in various cases about their undesirability. To avoid a conclusion that Sharman had to shut down, Wilcox made another, even more interesting order:

So Wilcox provided an ‘out’ – you are not infringing if either you put in place Filtering Type A (keyword/filename filtering) and take heavy steps to ensure existing users upgraded to that type of filtering, or Filtering Type B (polluting search results with copyright warnings) which will affect all searches. But note that the order leaves open what happens if Sharman/Kazaa do something else to remove or reduce authorisation of infringement.

Note something else about these orders. As Ed Felten noted at the time these orders were made, Order 5 is quite specific – filters must be used, but they need not be effective to comply with Order 4. What’s not clear is this: if the respondent does something else (ie, not Filtering Type A or B) to try to comply with order 4, how effective do those steps have to be?

These orders generated a fair bit of commentary at the time: I called them ‘brave’, Starkoff thought they were pretty thoughtful, Ed Felten made some very thoughtful comments on the technological feasibility of the order as well. For more commentary at the time, see this post where I collected links. They were controversial, and difficult orders because, as is obvious from a reading of them, Wilcox J was ordering changes to the technology: controversial for a judge at the best of times.

A further interesting thing about the case was that Justice Wilcox specifically chose not to suspend these particular orders pending the appeal. That is, he ordered Kazaa to take steps despite the fact that there was a pending appeal on liability (quite often, injunctions will be suspended while an appeal is pending, but Wilcox J, perhaps not surprisingly, took the view that the ongoing damage was too great to do that). All this on a punishing appeal schedule that saw hearing in February after a September judgment (a quick timetable indeed, particularly for an appeal hearing that went for 5 days).

Sharman/Kazaa were meant to do something by 5 December 2005. Kazaa did not put in filtering into its software, as required by the order. Instead, Kazaa opted to prevent downloads of its file-sharing software in Australia. From the night of Monday, 5 December, Australian users of the site were greeted with the message: “The download of the Kazaa Media Desktop by users in Australia is not permitted”.

On 9 December, the copyright owners filed a notice of motion, seeking punishment for contempt of court: that is, seeking punishment for Sharman’s failure to comply with the court’s order. Contempt can carry fines, or even jail time. The contempt matter was listed for hearing on 30 January 2006. On that day, Wilcox J heard debate on the matter, and reserved a question of law for the Full Federal Court. The question was:

‘Whether, having regard to the nature and terms of order 4 made on 5 September 2005, a determination of contempt of court may be made in respect of the contraventions of that order alleged in the statement of charge filed on 9 December 2005.’

The main argument raised by Sharman et al was that Order 4 (quoted above) was ‘so ambiguous that it is devoid of any legal meaning and incapable of giving rise to liability on a charge of contempt’.

Judgment on the question of law was handed down by Branson J yesterday. The Full Court (in the absence of facts about the alleged contempt of course) has said that Order 4 is NOT so ambiguous that it is devoid of any legal meaning. As the Full Court pointed out, there are circumstances where this order is plenty certain. If Sharman had just sat down and said, well, we will pay no regard to this order – then that’s pretty clear contempt, and the order is clear. I think the Full Federal Court are probably right on this question of law.

So the matter must go to hearing.

But let me tell you, this hearing could be really, really interesting. Let’s assume, for the moment, that the facts outlined above, as reported in the media, are accurate. I have no idea of the full extent of facts, of course. But let’s assume that this is what happened:

• Judge makes order saying – ‘you must not authorise infringement’. He then says ‘you will not be authorising infringement if you put in Filtering Type A or Filtering Type B. Note that both these Acts are aimed at preventing new users from infringing – but are also aimed at preventing existing users from infringing (or at least, reducing infringement by existing users).
• Respondent (Sharman) doesn’t do Filtering Type A, or B. Instead it does Act C – blocking downloads. That helps to prevent new users from infringing but of course does nothing about existing users (it may eventually drive them away because they won’t get upgrades to software of course, but that’s a long term kind of effect).

Now clearly, in this scenario, Sharman/Kazaa are engaging in one, big game of legal chicken. They’ve taken a risk that what they’ve done, less onerous and difficult as it is, will not be contempt. But whether it is contempt is not answered by the Full Federal Court’s judgment. They had to make a decision in the absence of full facts about the alleged contempt.

There seems little doubt that the respondent is entitled to do something other than the filtering envisaged in Order 5, to try to comply with Order 4. But what is not clear is this:

1. Does Sharman cease to authorise infringement of copyright by blocking downloads of the software to Australia?
2. Does the answer to question (1) depend on how effective the blocking is in preventing future infringements?
3. How would you prove contempt in such a case? Could the contempt hearing have to involve a whole lot of evidence about how effective the blocking is?
4. Can any of this really be proved beyond reasonable doubt, as required in a charge of contempt?

And note: the answers to these questions – particularly question 1 – are very much dependent on what counts as authorisation. And of course guess what – that is the very matter that is on appeal.

Ick!

The Full Court were conscious of this. The leading judgment of Branson J discusses (paragraphs 40 – 49) the danger of granting injunctions in form that reproduces the statutory prohibition – but accept that authority does not prohibit such injunctions, particularly in IP cases and particularly in cases where the respondent has shown tendencies to be dishonest or untrustworthy (the Full Court make no comments on whether that is the case here). Branson J also discusses the problem that the questions raised in the contempt hearing are the same questions as are being raised in the appeal – noting that there can still be contempt for acts that occur between trial judgment and appeal. But no resolution is offered – and probably can’t be, until the facts of the alleged contempt are established.

Think about it some more. All of this has been happening on a real fast timetable. Sharman had 3 months, basically, between the judgment and the December date, to put in place some kind of filtering technology. 3 months is not all that long. Remember too that at the same time they are drafting the appeal documents, and preparing for an appeal that will be heard a bare 5 months after the trial judgment, and is a complicated appeal (5 days’ worth). Can it be contempt if you take some steps, although not fully effective and not going as far as the judge wanted? Is this sharp timetable relevant?

The confluence of judgments, actions, and issues is getting very messy here. This is a massive, highly contentious case. It is at least conceivable that we could get a judgment that says there IS contempt, being handed down around the same time as an appeal judgment gets handed down saying there was no authorisation. What then?

I can’t help but wonder whether it might not have been better to simply stay the orders, given the rapid appeal. I completely understand why it was not, but given all the technological complexity, the contentiousness between the parties (remember that putting filtering in place does require cooperation between the parties), in the event it’s only given rise to some very nasty legal complications.

Thoughts?

Update: I want to quote this comment, from the Weatherall’s Law version of this post: