Wednesday, November 30, 2011

Australian appeals court lifts Galaxy Tab ban and restores sanity in patent enforcement down under (which benefits Apple, too)

Earlier today, Samsung succeeded with its appeal against an Australian preliminary injunction banning the Galaxy Tab 10.1 over the suspected infringement of two Apple patents.

Apple might appeal this decision, but local observers appear skeptical of the possibility of the ban staying in force. Realistically, Apple will now have to focus on the main proceeding, which could still result in a permanent injunction. Today's decision only affects the preliminary injunction that was granted (and now lifted) on the fast track. The regular track -- the main proceeding -- will continue.

For Samsung, this is great news, and the Australian subsidiary of the Korean electronics giant has already said that it "is pleased with today's unanimous decision". But this decision is important way beyond the fact that the Galaxy Tab 10.1 is now likely going to become available to Australian resellers and consumers without restrictions. I think it's a decision that restores much-needed sanity in Australian patent enforcement and prevents what could otherwise have become a disaster of enormous proportions. Seriously. If the preliminary injunction had been upheld, Australia would have been the jurisdiction with the lowest hurdle of all jurisdictions I know for any plaintiff seeking to shut down an alleged patent infringer in a fast-track proceeding.

As odd as it may seem, I think that even Apple should actually welcome (even if not publicly) this ruling. Cupertino looks like a loser today, but if that injunction had been upheld, the underlying legal standard, which was just too low, would have come back to haunt Apple down under. In fact, Samsung is asserting patents against various Apple products and requested a preliminary injunction against the iPhone 4S. But there are countless other companies who might have chosen Australia as a strategic battlefield, which in each case would have affected Australian consumers and in some cases could have hurt Apple itself.

I have read today's lengthy ruling in detail and will now share some of my observations, focusing mostly on two issues: the overall standard applied to patent-related preliminary injunction requests in Australia, and the likely outcome of the main proceeding between Apple and Australia with respect to the two patents on which the preliminary injunction was based. Note that Apple asserts several more patents in the main proceeding, but it picked those two for its preliminary injunction request.

Resounding rejection of the methodology and standard applied by the first-instance judge

The appellate decision is very harsh on the primary judge. Even though I believe that it was imperative, from a policy point of view, to raise the bar for patent-based preliminary injunctions well above the standard applied by the first-instance judge, I think she didn't do quite as bad a job as today's ruling suggests. The appellate decision says that "[a]ll that her Honour did was to set out the competing contentions of the parties on infringement and invalidity without assessing those contentions as she was required to do."

In its defense of the original ruling, Apple apparently argued that the judge had in fact considered the probability of the outcome of the final case but (which is not what Apple necessarily wrote, just my interpretation) perhaps not expressed all of her probability assessments clearly in writing. That is also my impression. I think the judge gave quite a bit more thought to this but refrained from taking unambiguous positions in writing on some of this because her overall feeling was that neither party was clearly right.

The mistake she made was that she didn't require Apple to demonstrate a case strong enough to justify the substantial damage a preliminary injunction inflicts to Samsung's short-term sales and longer-term market position in Australia. But that doesn't mean that she failed to do her job of analyzing the factors determining the likely outcome.

The appeals ruling also criticizes her for basically using as a tie-breaker (again, that's just my way to put it) Samsung's initial unwillingness to proceed to an expedited trial ("early final hearing" in Australian terminology). Here's a related passage from today's ruling:

"[...] it is unfortunate that [Samsung's] refusal to adopt that suggestion was to become a significant feature in her Honour's reasons for resolving the dispute in a way which was very unfavourable to that party. [...] we fear that the disinterested bystander may well have suspected that Samsung was punished for failing to adopt the approach recommended by the primary judge."

When I saw that this issue of an expedited trial turned out to be outcome-determinative, I was also wondering for a moment to what extent the judge's disappointment with Samsung's refusal of her proposal could have played a role. However, I doubted that this was an emotional thing. It appeared rather clear to me that the judge was so torn with respect to whether or not Apple had a sufficiently strong case that she was desperately looking for indications of who may be right. By refusing to have an expedited trial, Samsung looked weak. I think the judge also saw that Samsung had previously agreed to two delays of the launch of its product in Australia. All in all, Samsung appeared -- also to me -- to be very afraid of the outcome (to a hugely greater extent, in fact, than I got the impression of the judge being miffed because Samsung didn't accept an early trial).

Apple definitely appeared more confident than Samsung. It's a different corporate culture, different executives, different lawyers. None of that should ever have been a substitute for a stronger set of facts supporting Apple's case.

Samsung became the victim of a dangerously low standard

One of the various legal considerations on which the appellate court totally disagrees with the first-instance judge is that she was overly concerned about a possible "frustration of the Court's process" if she had denied a preliminary injunction, given that the product would have reached, more or less, the end of its shelf life by the time of a final decision (without an expedited trial).

That was the wrong approach, and a dangerous one. The fact of the matter is that Samsung could not have played a cat-and-mouse game with Apple forever. Maybe a decision would have been too late to have much impact on the Galaxy Tab 10.1, but if Apple had proven the validity and the infringement of its patents in court, it would then have been much more reasonable to grant preliminary injunctions against future products showing essentially the same infringement pattern.

The appellate court didn't take my kind of position, but it arrived at the same conclusion:

"We doubt whether the grant of interlocutory relief in the present case avoids frustration of the Court’s process. When one keeps in mind that the Court must seek to do justice to both parties, such a consequence for Samsung can hardly be described as other than a frustration of the Court’s process. In our view, the task facing her Honour was not simply to grant or refuse an injunction. The task was to protect the Court’s process from frustration, as best as could be done in the circumstances, [...]"

The last part means: the judge made it too easy for Apple to frustrate due process by getting a quick decision that was not well-considered and enabled Apple to get, with respect to the Galaxy Tab 10.1, the full benefit even on doubtful grounds.

The appellate decision leaves no doubt that in a case like this, only a "relatively strong case" can justify a preliminary injunction:

"In this case, given that the grant or refusal of interlocutory injunctive relief was going to have the practical consequence of deciding Apple's claims for final injunctions and thus deciding the commercial fate of the Galaxy Tab 10.1, Apple was required to demonstrate a relatively strong case."

Again, looking beyond this particular case, due process and a reasonably high standard for preliminary injunctions is also in Apple's own best interest.

Apple's Australian decision-makers and lawyers had the job to win, and they did what they could to win. They had to argue in favor of a low hurdle for a preliminary injunction in order to win one, just like Samsung had to argue for a high bar. But if Apple had defended the same views in a policy context, it would have made a poor choice.

Implications for impending decision on motion for US-wide preliminary injunction

I always felt that Apple pursued those international cases in no small part for the purpose of creating a more favorable climate for a preliminary injunction in other jurisdictions, especially in the United States.

Now the opposite is the case: the judge in California is likely to hear about today's Australian decision, and even though the two patents asserted in that Australian case are unrelated to the four intellectual property rights Apple asserts in its motion for a preliminary injunction in the United States, this outcome partly contradicts the story that Apple told in footnote 1 on page 1 of its reply brief in support of its motion in California.

Validity and infringement analysis

The appellate court's views on the likelihood of the asserted patent claims being both valid and infringed (anything less wouldn't do) is quite important with a view to the further process. Even though the first-instance judge will be free to arrive at whatever conclusion at the end of the main proceeding, the opinions taken by a higher court will affect her thinking for sure.

I think the course of events resulting from Apple's decision to shoot for a preliminary injunction, which was granted but lifted, has made it even harder (though definitely not impossible) for Apple to successfully enforce those two patents in the main proceeding. The judge won't like to be overruled again, especially if it can be avoided. And there's a significant probability of the appellate court later overturning any infringement finding concerning the two relevant patents.

I discussed those two patents after the injunction was granted.

The appellate court said this about the Heuristics Patent (whose inventors include Steve Jobs):

"On the present state of the evidence we are unable to see that Apple has established a prima facie case of infringement of either claim 1 or claim 55 of the Heuristics Patent."

There appears to be quite some doubt over whether Samsung indeed infringes it. Therefore, the appellate court didn't even have to elaborate on the validity of that patent.

The appeals court is also unconvinced that the Touch Screen Patent is infringed:

"It is sufficient for us to express the view that, on the present state of the evidence, there is a real and substantial prospect that the importation into and supply in Australia of the Galaxy Tab 10.1 will not infringe claim 6 of the Touch Screen Patent. We have referred to a number of difficulties that confront Apple in making good its case on infringement. It may well be that, on a final hearing, Apple will meet these difficulties. But difficulties they are. Whilst we would not be prepared to say that Apple’s case on infringement is not open to be argued, the difficulties to which we have referred do affect the assessment at the present time of the probability that, if on a final hearing the evidence remains the same, Apple will be found to be entitled to final injunctive relief for infringement of that claim. If Apple has established a prima facie case at all (which we doubt), it is founded upon a construction argument which, if the evidence remains as it is, is unlikely to succeed at trial."

The above sounds like Apple has quite some work to do if it wants to prove Samsung's alleged infringement of the Touch Screen Patent.

While the appeals court didn't really have to comment on the validity of the Touch Screen Patent, it nevertheless pointed out that neither of Samsung's two prior art contentions was convincing:

"In our view it follows from that state of affairs that Samsung has not established a prima facie case that the Leeper Article anticipates the touch panel claimed in claim 6 of the Touch Screen Patent."

"Once again it follows that, in the present state of the evidence, and in light of the limited way in which Samsung's case on invalidity has been advanced, no prima facie case has been established that the Mulligan patent application contains a disclosure that anticipates the touch panel claimed in claim 6 of the Touch Screen Patent."

This means Samsung also has some more homework to do if it wants to convince the court of the invalidity of that patent. After I saw the detailed version of the original Australian decision, I also felt that Samsung's prior art wasn't necessarily good enough at this stage. But it seems that Samsung's search for prior art continued.

Permutations and probabilities

Not only did the appeals court believe that the first-instance judge didn't make the necessary assessment of the probability of Apple prevailing at the end of the process but it also disagreed with the judge's approach to the fact that Apple had, in her view, a prima facie case with respect to two entirely different patents:

"As to the question of there being two patents rather than one, we do not see any way in which that can strengthen Apple's case. That consideration says nothing about the construction of the patents for infringement purposes. Whether it has any relevance to the allegation of want of novelty depends upon the assessed strength of that allegation. In the circumstances of this case no weight should be given to the fact that there were two patents."

On this item, I actually tend to side with the first-instance judge. From a statistical point of view, a plaintiff who asserts two unrelated patents against the same product with a given probability of success is, all other things being equal, much more likely to prevail on at least one of those patents than one who asserts only one patent of that probability -- provided that the aforementioned probability is sufficient.

If the appeals court says that there's "[no] way in which that can strengthen Apple's case", it appears to defy mathematical logic at first sight. Maybe the appellate court just wanted to say that this is not a substitute for a probability assessment with respect to each patent on its own. That would make sense. Even 100 patents of a low probability of success wouldn't matter. Also, the problem is that if a preliminary injunction is based on multiple patents, it becomes even harder for the defendant to engineer around those two or more patents, but if there had been two separate lawsuits over those patents, maybe none of the patents would have been deemed valid and infringed if the argument that two or more patents might be infringed justifies an injunction.

If one considers the workaround challenge, the logic of the first-instance ruling was basically that someone who is reasonably likely to be found to infringe at least one patent deserves to be punished, even if that punishment may go too far because it relates to both patents at once. It's possible that this approach is also part of what the appellate court rejected. But I can see why the first-instance judge felt that in a balance-of-hardship context, an increased likelihood of Samsung having committed at least one infringement could be a (minor) factor.

Generally speaking, the right approach for courts in these cases would indeed be to look at this in terms of permutations and to draw up a probability tree. I don't mean to say that the rulings should include such a diagram (that would be unusual), but it would be a useful exercise for the courts to do this for their own purposes and maybe there are many judges who do this (but obviously wouldn't talk about it).

Neither the original ruling nor the appellate decision present a probability assessment clearly based on the combined probability of a given patent claim being both valid and infringed (only that combination counts), or the combined probability of multiple attacks on a given patent. For example, in the context in which Samsung presented two pieces of prior art, it's possible that neither one appeared likely to succeed on its own, but if both were somewhat likely to succeed, Samsung might be reasonably likely to prevail with just one of those, which is all it would take to defeat the patent.

If you'd like to be updated on the smartphone patent disputes and other intellectual property matters I cover, please subscribe to my RSS feed (in the right-hand column) and/or follow me on Twitter @FOSSpatents and Google+.

Share with other professionals via LinkedIn: