Tuesday, April 8, 2014

Apple demands 20 times more for a patent from Samsung ($12.49) than from Motorola ($.60)

We're approaching the end of Apple's case-in-chief in Apple v. Samsung II. Rutgers law professor Michael Carrier has noted on Twitter that the first week of this trial did not go as well for Apple as in 2012 because it's "not clearly winning on themes" while Samsung manages to show that the patents-in-suit are "not central". Professor Carrier appears to feel that Apple's case gets bogged down in detail: "The more time spent in the minutiae, the less a huge damages award or injunction for Apple seems appropriate." He added: "And minutiae also don't support themes from 1st trial of Apple as revolutionary innovator and Samsung as flagrant copyist."

Based on a list submitted by Apple, its damages experts MIT professor Dr. John Hauser and Dr. Christopher Vellturo will be called today (Tuesday) or, at the latest, on Friday. Therefore, I looked at the complete Daubert (pretrial motions against unreliable expert testimony) record in this case relating to those two experts, including a hearing transcript I uploaded to Scribd last month.

On page 38 of that PDF document you can find a data point that I haven't previously highlighted (but ArsTechnica's Joe Mullin, who thankfully credited me for publishing the transcript, has):

Counsel for Samsung: "Look at the '647 patent. In [Apple v. Motorola in the Northern District of Illinois, the 'Judge Posner case'], Apple says that patent is worth $.60 [per device]. In this case, they're asking for $12.49 for it."

While $.60 is less than one-twentieth of Apple's per-unit damages demand from Samsung, it was still way too high for Judge Posner, who threw this claim out (in a May 2012 Daubert ruling) and described the amount as "senseless" in light of the limited cost of a work-around. It's a very safe assumption that if Judge Posner was presiding over Apple v. Samsung II (I said that it would have been a nice April Fools' Day joke for Judge Koh to enter the courtroom and announce the transfer of the case to Chicago so that Judge Posner could handle it, sitting by designation), Apple would not have had any chance to put a $40-per-unit (or $2 billion in total) damages figure before the jury. I think he would not even have allowed $4 per unit, and rightly so.

Samsung's counsel called the $2 billion claim an insult to jurors' intelligence. I actually think it's even more of an insult to Chief Judge Rader of the Federal Circuit. The appellate hearing in the "Posner case" took place in September 2013. Chief Judge Rader called Motorola's $300 million damages demand from Apple over one wireless patent "crazy" because they were talking about "one patent in a crowded field". In that case, it was good for Apple. But Apple didn't heed the chief judge's clear guidance when the shoe was on the other foot. Apple should have noted that Chief Judge Rader's perspective on the value of one patent in this crowded field was unrelated to the fact that Motorola had made a FRAND licensing commitment and declared the patent essential to a standard. It was a general position on patent damages, not a SEP-specific position in any way, as the official recording of the hearing proves.

On average, Apple is now seeking $400 million per patent per unit from Samsung. If $300 million was "crazy" in the eyes of America's -- and probably the world's -- leading patent judge, what is $400 million? Judge Koh should have stood up to Apple, America's and Silicon Valley's richest corporation, instead of hiding behind a jury. Of course the jury could just give little or no weight to Dr. Hauser's testimony (and Dr. Vellturo's conclusions from it), but that's a gamble that could and should have been avoided. If the appeals court later decides in a manner consistent with Chief Judge Rader's perspective on reasonable per-patent per-unit damages claims in this crowded field, then whatever damages Apple may be awarded at the end of this trial will just be vacated.

I restated (in a slightly softer form) my criticism of this (non-)decision in a pre-trial interview with the Associated Press, and I have seen knowledgeable people share my views. Yankee Group Senior Analyst Boris Metodiev said he had no idea how Apple came up with that figure that is "nothing short of extraordinary" and "raising eyebrows around both the tech and legal industries". The Verge and other media quoted Santa Clara Professor Brian Love, who specializes in patent remedies and found that these patents are "stylistic glosses on top of the underlying technology" rather than breakthrough inventions. According to Professor Love, "it's hard to make the argument that those patents are worth hundreds of thousands of times their value" (which is what Dr. Hauser is nevertheless going to try today or on Friday). Derek Khanna, a Yale Law Fellow, just adopted my extrapolations of Apple's per-patent-per-unit damages claims to the totality of smartphone-related patents. I would have quoted others for the sake of a balance but I am not aware of any recognized expert who would have publicly defended Apple's damages claim in this case...

A Daubert ruling against a damages claim must be based on methodology. The absurd amount shows that something is wrong with the methodology. Judge Koh could have easily thrown out the damages claim on that basis. On this key issue (way more important than the patents video shown at the start of the trial or the question of whether Apple practices its own patent claims), she disagrees not only with Chief Judge Rader but also with Judge William Alsup, a colleague of hers in the same district who tossed an Oracle damages claim against Google because a conjoint analysis (value of particular features of multifunctional product) was used to simulate the impact of infringement on market share (which also appears to be the case in Apple v. Samsung II as the net effect of the Vellturo/Hauser testimony). And, actually, she is less than consistent with herself.

No matter how hard I try, I can't reconcile her permissive Daubert approach to Dr. Hauser's testimony with the low esteem she had for his methodology as well as its implausible results when ruling on Apple's renewed motion for a permanent injunction. What she wrote in that context is closer to Judge Alsup's concerns in Oracle v. Google than her laissez-faire in the Daubert context. In a brief and at the hearing, counsel for Samsung stressed that no judge previously allowed a conjoint analysis to form the basis of a theory regarding how market share would evolve or would have evolved. Everything can happen for the first time, but when a judge can see that the results of a study are so unbelievable that "reality distortion" would be an understatement and furthermore sees that her own concerns are not really addressed properly, she could, and in my view should, send an expert back to the drawing board.

There are two differences here that I want to point out. One, a Daubert decision to let some testimony be presented to a jury is a question of admissibility, not weight (but if something has very little weight, it's normally confusing and flawed and, therefore, inadmissible), while a judge's equitable analysis in connection with an injunction motion is about weighing factors. Two, there is a difference between Dr. Hauser's conjoint analysis in Apple v. Samsung I and the one in Apple v. Samsung II: while he focused in the first case only on what consumers would be willing to pay for certain features, according to a web-based (and misleading, as Judge Koh noted) survey, he has meanwhile allowed participants to choose not only between different (mostly hypothetical) Samsung phones with particular features but also offered an "outside option" -- and he's now trying to show not only a willingness to pay for features but also a willingness (or, conversely, unwillingness in the absence of certain features) to buy particular products.

Those differences are undeniable, but I don't think they address three (partly interrelated) fundamental problems of this kind of analysis when the question is how relevant Apple's asserted patent claims are to multifunctional smartphones and tablets:

  1. It appears to be conventional wisdom in the field of conjoint analysis that you can't conduct a survey with too many distracting features (features that are added to the survey beyond the ones that are the subject of the study). Apparently you have to basically let respondents focus on half a dozen features in total. If that is so, then it may just not be the right methodology for such extremely multifunctional products (a fact that Judge Koh did recognize in the injunction denial decision) as smartphones and tablet computers. The limitations of a methodology are not an excuse for fundamental flaws and gross distortions.

  2. The reality for (not all, but) most patents is that it's not a binay question of having a feature or not. If patents cover less than complete features (so far Apple has proven in court, in 49 months of litigation, only its ownership of rubberbanding), then the alternative to a product that practices a given patent (claim) is one that works around it in the best possible way, preserving as much as possible of the functionality. In this regard, it's unfortunate that Samsung has not provided workarounds for at least three of the patents (unlike in the ITC case, for example, where Apple won an import ban that has meanwhile turned out to be so toothless it can hang it on a wall in Cupertino but won't get any leverage whatsoever out of it). This is probably due to tactical reasons. Samsung thinks it's not infringing anyway, it points to Apple's own products as examples of workarounds for some of the asserted claims, and if it was found to infringe, it would obviously modify its products accordingly. While this is not an excuse for a fundamentally-flawed methodology employed by Apple's expert(s) and the irrational results this methodology was destined to lead to, Samsung would clearly be in a stronger position if it had delivered some evidence that even if Apple prevailed, consumers would still buy Samsung's products in the same quantities as before.

  3. The last point is closely related to the previous one. Those conjoint surveys just let end users choose between products that have certain characteristics described in simple terms. Apple and Judge Koh agree in the Daubert context that it would be too much to ask that a conjoint survey form contain a detailed claim construction guide. However, if patents don't truly cover complete features, then this may just not be the right form of analysis, period. Judge Koh actually knows very well that everything depends on what's in the claims, even in this specific context. At the Daubert hearing she said she was inclined to toss at least the part of Dr. Hauser's report that relates to the '414 synchronization patent because he based his study on the broader claim 11, not the narrower (thus, easier to work around) claim 20.

    In her Daubert ruling Judge Koh then changed mind and allowed Dr. Hauser's testimony anyway because she thought Samsung should have persuaded her that it could, contrary to Apple's representations that it wouldn't be a viable option due to Android's architecture, work around claim 20 (by having no more than two parallel synchronizations running). I don't understand this. If Apple's damages theory is based on the wrong patent claim, then how would that shift the burden of proof to Samsung?

    (By the way, the Daubert hearing transcript shows that Judge Koh considered claim 11 "likely invalid", which is why Apple withdrew it. Meanwhile Apple is lobbying against patent reform that would make it harder to enforce patents that should never have been granted in the first place. Guess why.)

So, as a result, the jury will hear highly confusing and misleading testimony that seemingly supports some out-of-this-world numbers -- numbers that are totally detached from the rates at which Apple and others in the industry license patents, and in the case of the '647 patent, 20 times higher than Apple's own damages claim against Google's Motorola Mobility.

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: