Wednesday, October 31, 2018

ITC judge didn't buy testimony for which Qualcomm paid a single expert $3-4 million in Apple dispute

First, Happy Halloween!

There's a lesson for life that Qualcomm may have learned in recent weeks as an unintended side effect of its dispute with Apple: if smart judges figure out whom they can trust, the facts are not for sale.

Qualcomm's patent litigation campaign against Apple (and, by extension, chipset maker Intel) has suffered a few setbacks already. The first two judicial decisions against infringement complaints brought by Qualcomm came down in Washington, D.C. (United States International Trade Commission) in September and in Munich (Munich I Regional Court) in October. In both cases, Qualcomm's attempts to outspend Apple and Intel on expert testimony backfired:

  • The Munich I Regional Court indicated in its decision (of which the court thankfully made a redacted version available) that two expert reports Qualcomm had paid for (one by a Professor Schulze--I guessthis one--and another one by Professor Strehle) weren't persuasive in the slightest, while Apple and Intel's expert, Professor Franz Kreupl (who used to work for Infineon/Qimonda and SanDisk) "completely convinced" the court owing to the power of reason and logic.

  • This week the United States International Trade Commission published a redacted version of the late-September initial determination by Administrative Law Judge (ALJ) Thomas B. Pender in the investigation of Qualcomm's first ITC complaint against Apple. ALJ Pender was impressed with the "unequivocal, uniquely credible, and highly logical testimony of Aicha Evans, Chief of Strategic Planning for Intel and the former General Manager of the Intel [division] responsible for the baseband chipsets." The ALJ therefore found "Ms. Evans' testimony to be decisive on all matters upon which she offered her testimony concerning Intel's intentions and plans," but he also describes testimony by other witnesses called by Apple (Dr. Scott Morton, Dr. Jeffrey Eisenach, and Steven Bowers) to be "highly credible and specific."

    However, the meanwhile-retired ITC ALJ found "Qualcomm's non-fact witnesses, especially its economic witnesses, to be far less credible, for a myriad of reasons, including the likelihood of bias, the speculative and conclusory nature of their testimony, and the number of unsupported assumptions [the ALJ] found to be inherent with the testimony they offered."

    The one Qualcomm expert ALJ Pender critized most harshly is J. Gregory Sidak, who is running his own consulting firm named Criterion Economics:

    "[T]he amount of money paid to Mr. Sidak, before the current investigation, was approximately $1 million over several years and [...] the company he owns has invoiced between $3 million and $4 million just for this [ITC] investigation [of Qualcomm's first complaint against Apple] alone. [...] In my almost 39 years of practicing law, I have never seen or heard of anything even approaching this level of financial commitment by a witness to a party. Moreover, even absent this financial commitment, I was troubled by his testimony, for example his testimony about there being enough iPhones without the introduction of any new models in 2018 [...] ignored reality. From his financial relationship with Qualcomm bias may be presumed, and I find it would be an abuse of my discretion to give any material credibility to this witness or his findings. I also note the [ITC] Staff questioned his credibility twice during their discussion of the Public Interest, and that I noted it above." (emphasis added)

I heard that at the most recent Munich trial--the one leading to the decision I mentioned above--Judge Dr. Schoen ("Schön" in German), who filled in for Presiding Judge Dr. Zigann on that occasion, made an even more scathing remark about one of Qualcomm's expert witnesses in the German case. Based on hearsay, he felt but ultimately resisted the urge to call up the expert's university to complain because someone who submits, to a court of civil jurisdiction, an expert report that doesn't contain a single scientific statement "shouldn't be educating our young people." The related passage of the redacted Munich ruling doesn't contain such rhetoric, but in their purely factual it's no less dismissive.

Qualcomm must slowly but surely be concerned about its credibility. In September I mentioned that the volume of Qualcomm's share buybacks is simply irreconcilable with last year's claim (by Qualcomm's lead counsel in the Apple case in the Southern District of California, Cravath's Evan Chesler) that the house was "on fire," and in that post I quoted an analyst who complained about an unusual degree of discrepancy between what Qualcomm tells judges and what it tells financial analysts. I have many contacts in the analyst community (just yesterday I got multiple calls and messages regarding IBM's proposed acquisition of Red Hat), and recently I learned that Qualcomm told an analyst it expected to get leverage in the very near term from its Mannheim lawsuits against Apple, which makes absolutely no sense to me after attending all three trials (with more to come next year).

Those experts may be very smart people, but the facts in both cases just aren't favorable to Qualcomm's interests. The Munich case over a post-passivation layer patent and the public-interest part of the ITC investigation (the ALJ found that Qualcomm should not be granted a U.S. import ban against Intel-powered iPhones due to its anticompetitive effects) are far apart not only in geographical terms. In Munich, Qualcomm (if we believe the court) tried to make up an infringement theory out of thin air and paid two professors for testimony that was a substanceless smokescreen at best and fake science at worst. In the ITC investigation, Qualcomm faced an uphill public-interest battle since its own decision to seek an import ban exclusively against Intel-powered (but not Qualcomm-powered) iPhones made it all too easy for everyone to figure out the strategic objective: a wireless baseband chipset monopoly.

After bringing an ITC complaint that almost made Apple look like a proxy for Intel, Qualcomm then sought to limit the ITC's public-interest analysis to the implications of an iPhone ban, without thinking through the obvious effects on Intel's baseband chipset business. That was already foreseeable based on the original complaint: by not targeting Qualcomm-powered iPhones, Qualcomm wanted to be able to claim that the replacement for any banned iPhones would simply be other iPhones, plus, of course, phones made by any other manufacturer.

ALJ Pender understood the ramifications for the baseband processor industry. The doomsday scenario would be that Intel would be forced out of the market.

Just like in any antitrust case (which the ITC public-interest analysis in this case practically is, not in all but in several important regards), market segmentation is a key battle. It's the first important part of the analysis, like claim construction in patent infringement proceedings. Qualcomm failed to convince ALJ Pender that there was only one market for baseband chipsets. One segmentation vector is the usual high-end/low-end distinction. Another one, which is extremely important here, is the one between "merchant" chipset makers that sell their products to others and companies like Huawei (with its wholly-owned HiSilicon subsidiary) and Samsung (with its Exynos division) that build baseband chipsets more or less exclusively for their own devices.

Intel and Qualcomm are "merchant" suppliers: they sell to others, and that's their business. Whether a company serves customers or only builds something for its own use is one of the most obvious criteria for market segmentation one might imagine. No expert, no matter how smart, expensive or articulate, can explain away basic facts like that.

ALJ Pender's initial determination spans more than 200 pages. I'll talk about the public-interest analysis again soon (from an antitrust angle), but for now I'd just like to show you (a) the initial determination and, below that one, (b) the ITC's call for public-interest statements (which are due by November 8; by coincidence, that's also the date of the next Qualcomm v. Apple trial in Munich):

18-09-28 ITC 337-TA-1065 Qu... by on Scribd

18-10-22 ITC 337-TA-1065 Re... by on Scribd

Share with other professionals via LinkedIn:

Saturday, October 27, 2018

Google will likely ask the Supreme Court for a second time to review API copyrightability

As I wrote about six months ago when Google was preparing a petition for en banc rehearing of the Federal Circuit's Oracle v. Google Android-Java copyright decision, I'm not going to reiterate my longstanding positions (which haven't changed) on that case. However, Google filed an application for extension of time within which to file a petition for a writ of certiorari (request for Supreme Court review) about a week ago (PDF on Supreme Court website), so I wanted to update my readers in purely procedural terms.

In late August, the Federal Circuit denied Google's petition for a full-court review of a decision that (as I predicted) held the incorporation of many thousands of lines of Java API declaring code into Android did not constitute fair use.

Without the requested extension to January 25, 2019 (which was granted four days ago), Google would have had to file its cert petition in late November.

This time around, Google is trying its luck with a new Supreme Court counsel: Williams & Connolly's Kannon Shanmugam. Presumably, Oracle will continue to work with Orrick's Joshua Rosenkranz, who scored two Federal Circuit wins over Google on Oracle's behalf and fended off Google's original petition for writ of certiorari (which was about copyrightability). What he and his team have achieved in this case, and in some others, is nothing short of amazing.

As Google's application for an extension of time noted, its recent en banc petition related to both copyrightability and "fair use." I venture to guess Google's second cert petition will likewise raise both issues. One of Oracle's key arguments against the first cert petition (though we'll never know which argument bore most weight with the Supreme Court a few years ago) was prematurity: there had not been a final judgment on infringement. Chances are Google will try again.

This is now the longest-running smartphone IP dispute. It started more than eight years ago--in mid-August 2010--with Oracle's original complaint in the Northern District of California. Even Apple v. Samsung ended after "only" about seven years.

From a "justice delayed is justice denied" point of view, I don't think anyone can blame Oracle for presumably fighting very hard against Google's cert petition. However, as an app developer and IP blogger I would like the Supreme Court to grant cert--and to affirm both Federal Circuit opinions. That would provide definitive legal certainty, nationwide, on questions that many of us care about (no matter on which side we come down).

Fortunately, there is no vacancy on the Supreme Court. As some of you may have noted on Twitter and/or LinkedIn, I supported the #ConfirmKavanaugh campaign at all stages.

Share with other professionals via LinkedIn:

Thursday, October 25, 2018

Qualcomm hijacks EU Commission expert group on licensing and valuation of standard-essential patents

Last December this blog commented very favorably on the European Commission's "approach to standard-essential patents" document. Unfortunately, the implementation of one item--the creation of a group of "experts on the licensing and valuation of standard-essential patents (SEPs)"--is now going in the totally wrong direction.

On Friday (19 October) a "draft" list of experts was published on the EU Commission's website (the list should ultimately appear under the "Members" tab on this page). That list is utterly lopsided to say the least (this post continues below the image; click on the screenshot to enlarge):

The above list was taken down shortly thereafter, but it has already drawn harsh criticism from the Brussels-based Fair Standards Alliance, an industry group that promotes transparency and fairness in SEP licensing and has been mentioned here on multiple occasions:

"The Fair Standards Alliance [has] strong concerns about the unbalanced nature of this group, and fear[s] that the outcome of the discussions in the group will negatively impact incentives to innovate for the years to come, to the detriment of European industry and consumers. Furthermore, we have significant concerns about the lack of substantial experience and significant competence within the experts group in terms of the licensing and valuation of SEPs, even though the Commission decision required 'substantial experience in licensing and/or valuation of SEPs.' In our view, the group lacks a sufficient amount of industry experts that have day-to-day expertise in the negotiations of SEP licenses. We therefore urge the Commission to reconsider its selection, in order to have a group that will be able to achieve balanced views on what are the best next steps in terms of FRAND licensing for European policy makers to consider as the industry moves towards the development of 5G and IoT."

That's a pretty strong and clear statement, and I philosophically agree with it, but I'll allow myself to be much more blunt: the composition of this group is a recipe for disaster. Given the positions that many of those people have voiced on countless other occasions, the outcome is predictable. It will be a nightmare for competition and innovation, and all SEP abusers' wet dream.

This group is not the result of responsible choices made by neutral public servants desiring the best for Europe. It looks like none other than Qualcomm put it together, with other aggressive SEP monetizers like Nokia, Ericsson and InterDigital possibly having contributed to it, and some patent radicals at DG GROW (the Directorate-General for the Internal Market, which I fought hard against in the days of my NoSoftwarePatents campaign) rubberstamped it.

The Commission website says "[t]he aim of the group is to increase expertise and know-how about the determination of fair, reasonable and non-disciminatory (FRAND) licensing terms, as well as the sound valuation of intellectual property." That is fake news, and the recommendations coming out of this in the end will certainly be fake FRAND.

To be clear, there are some very smart people with lots of SEP-related expertise among them. In particular, I have the greatest respect for Damien Geradin, a lawyer and a professor (University of Tilburg, Netherlands)--but the first time I heard him speak at a Brussels event (in the spring of 2010), a large part of his presentation was simply hardcore advocacy for Qualcomm. His Tilburg Law and Economics Center (TILEC) lists Qualcomm as its number one sponsor in its annual report (page 23 of the PDF).

[Update] Professor Geradin (one of my esteemed readers) has meanwhile told me that he has not represented Qualcomm since 2010 and has, in fact, also represented clients adverse to them. While the information about TILEC being funded by Qualcomm was accurate, Professor Geradin says his work does not benefit from Qualcomm's sponsorship, and he tells me he will approach the expert group meetings with an open mind. [/Update]

Economist Justus Baron is a researched at Northwestern University's Searle Center on Law, Regulation, and Economic Growth, which announced in 2013 that it received $2 million from Qualcomm for "patent research." The press release describes Qualcomm as a "visionary company."

Another economist, Jorge Padilla, heads Compass Lexecon Europe. Qualcomm is one of his firm's major clients (possibly the primary client), and he has written various papers on SEPs, habitually denying that hold-up and patent stacking are actual issues (actually, even this blog here has provided numerous examples over the years that those concerns are well-founded).

Ruud Peters used to head the IP department of Philips from 1999 to 2013, then became a part-time adviser. If I recall correctly, I crossed paths with him at an EIF (European Internet Forum) event in 2004. In any event, Philips is very active in SEP monetization.

Bowman Heiden of the University of Gothenburg, Sweden co-authored a paper last year on "patent holdout" that was sponsored by 4iP  Council, an organization that is in turn sponsored by Qualcomm, Philips, InterDigital, and the Qualcomm's two primary allies, Nokia and Ericsson (though both at different points in time complained about Qualcomm's aggressive leveraging of SEPs).

[Update] A reader has pointed me to Mr. Heiden's PhD thesis, for which he received financial and other support from Ericsson: "First, I would like to thank Gustav Brismark at Ericsson for introducing me to the complexity of FRAND and partially funding my research while allowing me the freedom to reach my own conclusions." [/Update]

Qualcomm's allies simply outnumber potential voices of reason such as Audi's Chief Licensing Officer, Mathias Schneider.

I'm afraid DG GROW won't listen to the Fair Standards Alliance, and the set of recommendations that this group will put forward will be materially consistent with whatever Qualcomm's legal department would propose.

[Update] The list is now final. Very disappointing. [/Update]

Share with other professionals via LinkedIn:

Wednesday, October 17, 2018

Why would the Federal Trade Commission snatch defeat from the jaws of victory over Qualcomm?

This is a follow-up to yesterday's post on Judge Lucy H. Koh's decision to deny a joint administrative motion by the Federal Trade Commission and Qualcomm asking her not to rule on a motion for partial summary judgment on the obligation to extend standard-essential patent (SEP) licenses on FRAND terms to rival chipset makers (such as Intel).

I didn't want to jump to conclusions from a case management decision, but in purely probabilistic terms it's a fact that Judge Koh's order increases the likelihood of summary judgment being granted. There's no harder-working judge than her, and she wouldn't have decided to cancel the hearing and take this motion (as well as several other, less important motions) under submission if there had been any questions left to ask. However, if she had been inclined to deny the actual motion, she might just have given the parties four weeks to work out a settlement--the sole remaining plausible explanation would be that she wanted to make it clear her court is nobody's tool, much less a restaurant that serves litigation à la carte where you can put one motion on hold while letting the process continue on the same schedule in all other respects.

This antitrust litigation has been going very, very well for the FTC for a long time. It wouldn't make sense to let Qualcomm off the hook now that there is a near-term opportunity (with respect to the summary judgment motion, "near-term" is actually a gross understatement) to restore fair competition in the market for baseband chipsets and with respect to cellular SEP licensing.

There's probably a lot of fighting going on in DC behind the scenes, inside and outside the FTC. Prior to the latest twist I had already tried to find out about where the current five commissioners stand on FRAND, but haven't found any information that would enable me to predict the outcome of a vote on a hypothetical settlement proposal tantamount to the agency's surrender. In the past, the positions of Commissioners Maureen Ohlhausen and Joshua D. Wright were well-known (I mentioned Mrs. Ohlhausen on several occasions, and in 2013 I dedicated a blog post to Mr. Wright's stance), but they aren't in office anymore. So I extended my search for clues to high-ranking FTC officials. It turns out that two of them--Alden Abbott (the FTC's General Counsel) and Bruce H. Kobayashi--have a certain proximity to Qualcomm and are sympathetic to Qualcomm's unFRANDly positions to a degree that is clearly a minority opinion in the legal community. While I don't have the slightest indication of any impropriety, there is a conspicuous lack of impartiality.

In Mr. Abbott's case, I didn't even have to look far. The "Speeches, Articles, and Statements" section at the bottom of his official bio page mentions a May 2018 interview conducted by two people, one of whom is "Koren Wong-Ervin (Qualcomm)." Her title, according to LinkedIn and Twitter, is Director of IP & Competition Policy. I'll mention her again further below.

Mr. Abbott was already trying hard to make a case against the case against Qualcomm before his appointment, a fact that Bloomberg Law reported on ("Critic of FTC Qualcomm Suit Named Agency General Counsel").

While at the Heritage Foundation, Mr. Abbott organized antitrust conferences at which then-Commissioner Ohlhausen explained why she dissented from the FTC's complaint against Qualcomm. In January 2018, both Mr. Abbott and Qualcomm's Mrs. Wong-Ervin both spoke at a Heritage Foundation conference.

On the Heritage Foundation's website, I googled up this article by Mr. Abbott on the Qualcomm case. He predicted that foreign competition authorities would leverage this U.S. antitrust matter against innovative U.S. companies, something which hasn't happened to date. He wanted Mrs. Ohlhausen to lead the FTC (which she did only on an interim basis).

In March 2018, Mr. Abbott spoke at the "IP Leadership" conference, whose sponsors included Qualcomm and the law firm representing it against Apple in the Southern District of California, Cravath Swaine & Moore.

The FTC's chief economist, Mr. Kobayashi, crossed paths with Qualcomm's Mrs. Wong-Ervin at George Mason University. According to the FTC's website, he had been a law professor there since 1992, and she was the Director of George Mason's Global Antitrust Institute and an adjunct law professor from November 2015 to September 2017 (according to her LinkedIn profile).

In 2016, the two teamed up with DC Circuit Jugge Douglas Ginsburg and former FTC commissioner Joshua Wright to co-author at least a couple of articles one can find with Google: an article on "extra-jurisdictional remedies involving patent licensing" (PDF; published by Competition Policy International), and a paper (SSRN) on "the Korea Fair Trade Commission's Amendment to Its Review Guidelines on Unfair Exercise of Intellectual Property Rights."

Those people's views on SEP licensing boil down to saying that patents are legal monopolies, thus there's no room for the notion of overcharging--which is misguided for various reasons, including but not limited to

  • the fact that one must separate the value of someone's technical contribution to the state of the art from the value of a standard as a whole (as so many others, such as Judge Posner, explained before), and

  • the policy consideration that, without antitrust-based restrictions, a single SEP would enable any given patent holder to prevent an entire industry from implementing a standard on a commercially viable basis (or at all), and typically there's a plurality of SEP owners, especially in the field of cellular telecommunications where hundreds and often even thousands of patents are declared to be essential to a single standard.

Qualcomm's FTC friends hold views outside the mainstream--and contrary to the public interest. Let's hope the five commissioners won't adopt those views.

Share with other professionals via LinkedIn:

Tuesday, October 16, 2018

FTC and Qualcomm trying to settle antitrust matter by November 14, but Judge Koh may rule on chipset licensing anytime

On Monday, the Federal Trade Commission and Qualcomm brought a joint administrative motion asking Judge Lucy H. Koh of the United States District Court for the Northern District of California not to adjudicate between now and November 14 the FTC's motion to require Qualcomm to extend standard-essential patent (SEP) licenses to rival chipset makers such as Intel:

"Pursuant to Local Civil Rule 7-11, Plaintiff Federal Trade Commission ('FTC') and Defendant Qualcomm Incorporated ('Qualcomm') hereby jointly move the Court to defer ruling on the FTC's Motion for Partial Summary Judgment [...] for a period of 30 days (until November 14, 2018). The FTC and Qualcomm each continues to stand behind the positions in its filings on the FTC's Motion. The Parties believe, however, that deferral of a ruling would facilitate the Parties' ongoing discussions concerning the potential settlement of this litigation.

The Parties do not request that the trial of this matter be taken off calendar and do not request any change to the schedule established by the Court's Case Management Order (ECF No. 678) or the Court's Order Continuing Further Case Management Conference (ECF No. 826). The Parties also do not request that the Court defer ruling on any other pending motion."

Judge Koh gave this administrative motion short shrift: she denied it quickly, without any further explanation.

Since my first commentary on the motion, I've always felt that nothing would make a settlement more likely at this stage than an order granting the motion and reminding Qualcomm of obligations it entered into when it made FRAND licensing declarations to two U.S. standard-development organizations, TIA and ATIS. Yesterday's administrative motion validates that assessment: while the FTC and Qualcomm have presumably talked about settlement on numerous occasions (even if just calling someone to find out whether the other party's position has changed), the current situation is unique. The importance of SEP licenses to competitors couldn't be made clearer than by a motion that says the parties don't have a problem with decisions on any other pending motion but this one. This motion unsurprisingly appears to scare the living daylights out of Qualcomm.

From the outside we can only ask ourselves three questions:

  1. Would a settlement between the FTC and Qualcomm be a good thing?

    The answer obviously depends on the terms. A settlement following a ruling according to which Qualcomm must license competing baseband chip makers like Intel would definitely be preferable over one at this stage since the FTC would then at least have accomplished something important and useful, and Qualcomm would be under more pressure to settle the remaining parts of the case. But if Qualcomm has simply made headway with its DC lobbying efforts and the Trump Administration wants to let it off the hook, then a settlement in all likelihood won't help and private-party litigation (such as Apple's Southern California case, a hypothetical lawsuit by Intel seeking a FRAND license to Qualcomm's SEPs, or the 250 million consumers' class action) would have to answer the most important questions. Just yesterday I published an infographic that shows all the presently-pending smartphone patent and antitrust cases, most of which involve Qualcomm in one way or another.

  2. Is it good news that Judge Koh denied the administrative motion to defer?

    Absolutely. The sooner she resolves the actual motion now, the better for industry and consumers. It's not about whether a settlement would be desirable. It's about whether the outcome will be good.

  3. Why did she deny it?

    We can just speculate. There are some possibilities that aren't even mutually exclusive. One possibility is that she's already done most of the work and a ruling is imminent. Whether or not her order is far along, she may be concerned about parties pushing a court around like they could tell a taxi driver "please wait another hour, we'll pay you for your time." Also, it may just not seem practical to her to rule on some other pending motions first: if the court finds Qualcomm has to extend licenses to rival chipset makers (as its ATIS and TIA declarations strongly suggest), some other questions (substantive as well as evidentiary) may be easier to resolve. And she, too, may feel that if a weak FTC-Qualcomm settlement doesn't really help, we'll just see more private litigation. And where would someone like Intel sue Qualcomm for a license? Presumably in her district, and the case would most likely be assigned to her.

    Judges generally like settlements, and this is particularly true of Judge Koh. But since she's a neutral arbiter--unlike the FTC, which has done great work so far but can be influenced through political lobbying that has nothing to do with the merits of the case--, she may think she has a particular responsibility here to ensure that certain problems be solved. If they can be solved without a trial, great. But if it takes a trial, or at least a decision on the motion regarding chipset licensing, so be it.

Share with other professionals via LinkedIn:

Sunday, October 14, 2018

The new smartphone patents battlemap (infographic featuring Apple, Huawei, Intel, Qualcomm, Samsung)

Eight years ago, this blog started publishing battlemaps of major smartphone patent disputes. Apple and Microsoft aren't suing Android device makers at this point, but two major disputes are ongoing: Apple and, by extension, Intel's dispute with Qualcomm, which is under fire from competition enforcers around the globe; and Huawei v. Samsung, the most massive patent clash ever between two leading Android device makers.

That's why I took the time today to draw up my first battlemap (announced in a post earlier today) in more than half a decade (click on the image to enlarge; this post continues below the document):

You can find the PDF version (a single-page document) on Scribd.

As the text in the bottom right corner explains, solid red lines indicate disputes or conflicts (competitive activity or activity that may lead to litigation, such as a regulator fining a company, which then appeals the decision to a court of law); dotted red lines show that there was a dispute, but it has been settled or at least the parties agreed on a ceasefire; and green lines denote partnerships or alliances (including joint defense).

If a line has only one arrow, it goes from the party that first went to court (or took similar action) to the target, regardless of whether the other party counterclaimed. For an example, whenever Huawei brought offensive patent infringement assertions against Samsung, Samsung countersued, but the history of that litigation and the parties' pleadings suggest that Samsung, which according to a recent filing suggested a zero-zero cross-license at some point, would have preferred to be left alone. It's like an offense actually representing a kind of defense.

Apple's patent infringement counterclaims against Qualcomm are a similar story, and that's just one of various parallels between Apple's approach to Qualcomm's conduct and Samsung's defenses against Huawei. And the way Huawei is dealing with (Pan)Optis Wireless, which is asserting former Ericsson patents against it, is not entirely dissimilar to Apple's and Samsung's defensive efforts either.

I could have added companies like Nokia and Ericsson to the list (beyond PanOptis, which is an Ericsson proxy, or so-called privateer), but they have license deals with the likes of Apple, Samsung, and Huawei in place anyway. So I focused on the ones that are presently party to major conflicts.

In the early part of this decade, Apple v. Samsung (just thinking of the presidential veto of an ITC import ban and the first Supreme Court ruling on design patents in well over 100 years), Apple v. Google/Motorola (especially the "Posner case" in Chicago!) and Microsoft v. Google/Motorola were the two disputes in which the most interesting things happened. Now we're approaching the end of the 2010s, and the disputes visualized by the battlemap above are the ones to watch.

Just one last comment on similarities between Apple's litigation with Qualcomm and Samsung's with Huawei: I've uploaded an Apple reply brief to Scribd that notes Qualcomm dropped its request for a global FRAND determination in the Southern District of California (though Qualcomm still wants the U.S. district court to hold that its demands were FRAND-compliant). This reminds me of Huawei recently stipulating to a dismissal (without prejudice) of a similar request. And Apple's brief quotes something Judge William H. Orrick said in Huawei v. Samsung:

"How am I to adjudicate whether those offers were FRAND, if that determination depends on valuation of global portfolios, and can only be made subsequent to finding each patent valid and essential to the standard?"

This insistence on firstly establishing the merits of someone's standard-essential patents (as opposed to rushing to rate-setting based on sheer portfolio size, regardless of whether those patents are infringed and valid, be it in a court of law or in arbitration, which patent holders typically want to take place under extremely unfair preconditions and parameters) is a common element of Apple's positions in its dispute with Qualcomm and Samsung's defense against Huawei.

At the same time, Qualcomm and Huawei, despite the fact that the latter is rumored not to be paying patent royalties anymore to the former, have been described as "more partners than rivals." I, too, sometimes refer to them collectively as "Qualwei," though I still hope Huawei will ultimately focus on its product business more so than on outbound patent licensing.

Finally, I'd like to explain the logic behind placing Apple and Intel in a rectangle that is the target of patent lawsuits brought by Qualcomm. Apart from China, all patent infringement actions pursued by Qualcomm against Apple (in the ITC and Germany) target Intel-powered smartphones, and I've seen opposition briefs Intel filed with the European Patent Office, seeking the revocation of certain patents asserted by Qualcomm against Apple--with the same patent attorneys making virtually identical filings on Apple's behalf as well. Obviously, some of Qualcomm's patent infringement claims are technically closer to Intel's baseband processor (especially several assertions accusing a Qorvo envelope tracker of infringement) than others, a threshold question that the German judges presiding over multiple Qualcomm v. Apple cases in Munich and Mannheim are particularly interested in.

Share with other professionals via LinkedIn:

Patent exhaustion keeps Qualcomm on the run from Apple's claims and motions

If exhaustion makes you run even further and faster, you're either above conventional physics--or you're Qualcomm.

Patent exhaustion has been an important and powerful concept for a very long time, but last year the Supreme Court provided a great deal of clarification in its Lexmark opinion. That opinion came down in the early phase of Qualcomm's disputes with the United States Federal Trade Commission (FTC) and Apple.

Apple asserted nine patents in its original Southern California complaint. Qualcomm surprisingly didn't bring mandatory infringement counterclaims, thereby waiving its right to assert infringement in that litigation or any other U.S. case against Apple. Apple threw in nine more patent DJs (declaratory judgment requests) in its first amended complaint, but Qualcomm persuaded Judge Gonzalo P. Curiel to throw those additional claims out. In order to do away with the original nine DJs for good (since Qualcomm would rather have everyone focus on the size of its portfolio--130K patents), Qualcomm "supersacked" (see this Lexology article for further information) Apple and the contract manufacturers by sending them a covenant not to enforce those patents against them, and in accordance with the Super Sack precedent immediately moved for dismissal of the original DJs including the patent exhaustion DJ for alleged lack of subject matter jurisdiction.

Apple and the contract manufacturers (Foxconn, Compal, Wistron, Pegatron) opposed. They basically argued that Qualcomm can't just duck patent-specific determinations while demanding billions of dollars in royalties. They pointed to a decision in the District of Massachusetts, where a judge still identified a justiciable controversy over patents despite a Super Sack-style covenant.

Qualcomm filed its reply in support of its motion for dismissal on Friday (this post continues below the document):

18-10-12 Qualcomm Reply Iso... by on Scribd

Qualcomm seeks to distinguish the Massachusetts precedent (Esoterix v. Qiagen) from its situation with Apple and the contract manufacturers. There are differences for sure, still it's true that the Massachusetts decision shows Super Sack isn't as absolute and comprehensive as some would have it.

The highlight of Qualcomm's reply is a true "gotcha" moment:

"Apple previously admitted that Qualcomm's position here is correct: “Qualcomm always has the power to remove the Nine Patents by granting Apple a covenant-not-to-sue pursuant to Super Sack.' [...] Now, faced with Qualcomm's covenant and Motion, Apple attempts to reverse course."

My guess is that Apple didn't quite expect Qualcomm to chicken out the way it ultimately did--at least not this late in the game. But this here is not about whether Apple had extended a standing invitation: a patent holder doesn't need any invitation to invoke Super Sack.

But Apple is determined to obtain judicial clarification of the all-important exhaustion question in the specific context of Qualcomm's alleged double-dipping by selling chipsets that substantially embody certain patented inventions and additionally charging patent license fees.

That's why Apple and its contract manufacturers--on Friday, just like Qualcomm's reply--brought a motion asking the court to entertain a new motion for partial summary judgment on patent exhaustion. The new motion relates to U.S. Patent No. 9,136,878 on a "file download and streaming system," one of 93 standard-essential patents (SEPs) Qualcomm listed in that San Diego case as part of its efforts to prove the value of its SEP portfolio and which its opening expert reports focus on (this post continues below the document):

18-10-12 Apple & Contr ... by on Scribd

The whole idea is to attack the '878 patent in case the pending motion for summary judgment on exhaustion would be mooted by Qualcomm's Super Sack covenant.

Qualcomm almost instantly voiced its opposition to this new motion, and asked the court for more time to respond.

In their motion, Apple and the contract manufacturers say "Qualcomm is desperate to avoid judicial review of its illegal business practices which include forced licenses of exhausted patents." Legality or illegality will have to be determined in court (antitrust authorities have already taken pretty clear positions). Any observer can already see that Qualcomm is trying to hide and and run from the question of patent exhaustion.

In this case, following high-stakes commercial litigation almost feels like watching a cartoon...

In the next post (later today) I'll show you a new infographic I created today to show the web of patent and antitrust disputes involving the likes of Apple, Samsung, Qualcomm, and Huawei.

Share with other professionals via LinkedIn:

Saturday, October 13, 2018

Qualcomm files appeal to avoid potential $5 billion payout to 250 million American smartphone buyers (1.2 billion transactions)

On Thursday I thought Munich, where Qualcomm just lost a case against Apple (and, by extension, Intel), was going to be the venue where the cellular chipset maker would file its first appeal in a litigation between private parties that is part of the current globe-spanning host of lawsuits. But it turns out that on the same day Qualcomm filed a petition to the Ninth Circuit appealing Judge Lucy H. Koh's recent class certification (this post continues below the document):

18-10-11 Qualcomm Petition ... by on Scribd

In its third question presented for review, Qualcomm describes this consumer class that was certified in the Northern District of California--up to 250 million people and, according to Qualcomm's estimate, approximately 1.2 billion claims (since people, on average, bought multiple smartphones during the roughly 8-year period the claims relate to)--as "quite likely the biggest class action in history."

That may be true with respect to membership size. It certainly isn't in economic terms since the $4.99 billion demand Qualcomm is facing ('s Scott Graham found out) is dwarfed by the $206 billion tobacco settlement in 1998 or the $20 billion Gulf of Mexico oil spill settlement in 2016. Still, $5 billion is a very significant number, which would amount to approximately $20, on average, per class member. The exact amount per member would obviously depend on the particular smartphone purchases made by each consumer. It would be the amount of each buyer's overpayment due to Qualcomm's practices (which regulators on three continents have already held to be anticompetitive), possibly enhanced by a factor of up to three. Just imagine how many people--outside of its own organization and apart from its shareholders--Qualcomm would make happy with such an involuntary gift...

One of the reasons for which Qualcomm wants the United States Court of Appeal for the Ninth Circuit to overturn Judge Koh's class certification is plain feasibility. Qualcomm argues, as it did (unsuccessfully) in San Jose, that the consumer plaintiffs "provided no plan whatsoever to process, verify, and administer more than one billion claims—a herculean task for any claims administrator." However, Qualcomm does mention that an attorney declaration submitted to the district court said that effective notice could be provided to 70% of all members of the class (approximately 175 million people).

My answer to this has two parts, neither of which Qualcomm will want to hear:

  • A $5 billion payout justifies, and effectively pays for, a whole lot of administrative effort.

  • Technology--Qualcomm knows a thing or two about it--can work wonders. In the Internet Age you can handle classes that appeared unmanageable decades ago. That's a major reason for which I'm underwhelmed by Qualcomm's citation to a 1983 Second Circuit decision. Times have changed since then. That was in the middle of Ronald Reagan's first presidential term, and most of us used to listen to vinyl records since the compact disc had just been released the year before (and nowadays music is streamed and downloaded for the most part).

    The Ninth Circuit is particularly aware of technological progress due to the cases it gets.

But manageability is not the first point Qualcomm raises. Its petition argues that Judge Koh shouldn't have applied California law to this decision since two dozen other states don't allow indirect purchaser claims on antitrust grounds; it also takes aim at the consumer plaintiffs' damages model; and argues that "large numbers of consumers" are included in the class definition but, in Qualcomm's view, "suffered no antitrust impact."

I'll wait for plaintiffs' responsive brief before looking more closely at those issues. At first sight it appears that Judge Koh had taken everything into account.

On the subject of Judge Koh's decision-making, she's just decided to take under submission (decide in writing without an oral hearing) all pretrial motions pending in FTC v. Qualcomm, including the FTC's motion for partial summary judgment regarding Qualcomm's obligation to extend FRAND licenses to rival chipset makers.

Share with other professionals via LinkedIn:

Friday, October 12, 2018

Federal Trade Commission makes important clarifications in reply to Nokia's amicus brief supporting Qualcomm

While Qualcomm's patent infringement lawsuits against Apple (and, by extension Intel) are merely a sideshow to the antitrust matters pending on three continents, let's start with a brief follow-up to yesterday's Munich judgment. The court has meanwhile, thankfully, provided a redacted copy of the decision. I've read it, and the most interesting part is that Qualcomm had submitted two expert reports in support of its claim construction, while Apple had provided only one, but the deficiencies of Qualcomm's reports were massive while Apple's expert provided instructive, helpful information. I'll publish a translation of the relevant passages soon.

Meanwhile, Judge Lucy H. Koh of the United States District Court for the Northern District of California has granted the Federal Trade Commission of the United States permission to file a reply to Nokia's recent amicus curiae brief in support of Qualcomm with respect to the FTC's motion for partial summary judgment regarding Qualcomm's obligation to extend FRAND licenses to its cellular standard-essential patents (SEPs) to rival chipset makers. In its reply brief in support of its motion (that post also contained a link to Nokia's brief), the FTC had already reserved the right to request leave to file such a reply since Nokia's brief was filed just on the eve of the FTC's reply to Qualcomm's opposition.

The FTC's reply must not be misperceived as an implicit concession that Nokia's brief contained anything that has the potential to be outcome-determinative in Qualcomm's favor. The legal question for Judge Koh to resolve is purely one of contract interpretation, and the FTC notes at the very start of its reply brief that Nokia "does not present any argument regarding the plain language of the two IPR policies that are the subject of the FTC's motion for partial summary judgment." Nokia's brief does not even present parol evidence (which would be irrelevant anyway under California law unless the whole industry had consistently taken a certain perspective on SEP licensing). Finally, Nokia's credibility is minimal in this context since Nokia itself told the European Commission in a February 13, 2006 complaint that Qualcomm breached its FRAND duty by terminating a license deal with a chipset maker (Texas Instruments).

The reason why the FTC elected to submit a reply--and which appeared significant enough to me to warrant a follow-up post--is simply that the FTC doesn't want the actual scope of its summary judgment motion to be blown out of proportion. The FTC states once again that "[its] motion [for partial summary judgment] asks the Court to interpret the plain language of Qualcomm's commitments under the IPR policies as written, pursuant to California contract law, not to determine what FRAND means in the abstract or to declare how SSOs ought to draft their IPR policies."

In the second paragraph, the FTC makes a point that is incredibly important in connection with FRAND licensing obligations (in this case and far beyond): it's one thing whether certain obligations (here, to extend licenses to chipset makers upon their request) exist, but another what the market will ultimately make of those obligations: "If Qualcomm's anticompetitive conduct is addressed and remedied, market forces will decide whether modem-chip or handset-level licensing (or some combination of the two) is most efficient."

There's a parallel here to an argument made by Huawei in its defense against (Pan)Optis Wireless in the Eastern District of Texas. In that dispute, Huawei accurately argues that a SEP holder shouldn't be able to choose one venue (here, the Eastern District) for a global FRAND rate-setting process, but in practical terms, global license agreements may very well be the most common result of free negotiations between parties. (Huawei takes the opposite position in its dispute with Samsung in the Northern District of California, but schizophrenia doesn't render its point in Texas any less well-taken.)

The FTC just wants to achieve judicial clarification that Qualcomm itself made FRAND licensing promises that also extend to rival chipset makers should they seek a license (which applies to a company like Intel). If no one made such a request, or if Qualcomm extended FRAND licenses to all device makers, then there would simply be no demand. It's a hypothetical scenario because there obviously is demand; but the bottom line is that the FTC wants to set market forces free as opposed to using an antitrust litigation in order to restrict anyone's freedom (apart from Qualcomm's freedom not extending to anticompetitive, exclusionary conduct, of course).

Again, an analogy: Huawei's opposition to a global FRAND determination in the Eastern District of Texas (a venue that has a reputation for being particularly sympathetic to patent holders, particularly non-practicing entities) doesn't mean that Huawei and (Pan)Optis might not ultimately agree on a global license. While it's always possible that parties reach a settlement in one jurisdiction (or set of jurisdictions), such as when InterDigital settled some European cases but litigation kept going in the United States or when Samsung and Apple withdrew all ex-U.S. lawsuits (but not the ones pending before Judge Koh), most negotiations will result in a global license for convenience. But it should be the result of free choice, not of undue leverage (such as from injunctive relief) or forum-shopping.

I found a couple of gems in the footnotes:

  • "Qualcomm has claimed that modem-chip-level licensing is efficient--at least when the licensee is Qualcomm itself. [...] If Qualcomm’s efficiency claims are valid, similar efficiencies may arise from other modem-chip suppliers licensing Qualcomm's cellular SEPs on FRAND terms."

  • "The FTC does not concede that construing the ATIS and TIA policies according to their plain language would create any inconsistency among relevant IPR policies."

    As I explained in my commentary on the FTC's motion, it's just that the FTC would like to streamline and simplify the Northern California case by obviating the need for an interpretation of the ETSI FRAND declaration since it's governed by French law. To me it's always been clear that the FTC's position is that Qualcomm has to extend licenses to rival chipset makers even just under ETSI's rules, but when you're litigating an issue in the United States and have two FRAND declaration texts under U.S. law (ATIS and TIA) based on which you can prevail, why make an intercontinental detour that just wastes court, federal government and private party resources?

Finally, here's the FTC's brief:

18-10-11 FTC Reply to Nokia Amicus Brief in Qualcomm Case by Florian Mueller on Scribd

Share with other professionals via LinkedIn:

Thursday, October 11, 2018

Munich I Regional Court throws out Qualcomm patent infringement lawsuit against Apple: no infringement

The Munich I Regional Court ("Landgericht München I" in German) just announced the first final judgment on a Qualcomm v. Apple patent infringement complaint anywhere in the world. Apple (and, by extension, Intel) fended off one of various Qualcomm patent infringement lawsuits.The court agreed with Apple's claim construction.

A few months after the Federal Trade Commission of the United States and, in a separate case, Apple sued Qualcomm over antitrust and contract-related matters in January 2017, the San Diego-based company that dominates the market for cellular baseband chips started a patent infringement lawsuit campaign against the iPhone maker in the U.S., Germany, and China. Qualcomm wanted to kick off the German "rulings season" with a Mannheim injunction--and got a Munich rejection.

A three-judge panel--composed of Judge Dr. Schoen ("Schön" in German), who filled in for Presiding Judge Dr. Zigann at last week's trial, and Judges Klein and Schmitz--held that Apple's iPhone 7 and iPhone 7 Plus do not infringe Qualcomm's European patent EP1199750 on a "post[-]passivation interconnection scheme on top of [an] IC chip."

I missed last week's trial, but attended the first hearing in May. Freshfields Bruckhaus Deringer's Prince Wolrad of Waldeck and Pyrmont was Apple's lead counsel with respect to its (now victorious) non-infringement contentions then and presumably also last week. He's a familiar face at German Apple patent hearings and trials, as are several other attorneys at law and patent attorneys. The patent attorneys also represent Intel in parallel nullity and opposition proceedings before the Federal Patent Court and European Patent Office, respectively. Qualcomm is represented in Germany by a team of Quinn Emanuel lawyers led by Dr. Marcus Grosch. He's a tremendous fighter in every sense of the word and will probably appeal this decision to the Munich Higher Regional Court ("Oberlandesgericht München" in German). However, the fact that the ruling came down just one week after trial--normally it takes several weeks, and in complex cases even months--suggests that Apple had a non-infringement argument that made this a very clear case once all the facts and theories were on the table, so chances are that Prince of Waldeck will be able to defend his victory on appeal. Also, the patent may be considerably closer to invalidation by the time of an appellate hearing.

Qualcomm is increasingly unlikely to gain leverage over Apple in the near term:

Qualcomm has meanwhile filed additional patent infringement complaints in Germany. It will keep trying, but realistically it won't be able to force Apple into a settlement before some key antitrust, contract and patent exhaustion questions are resolved in the United States.

I will soon write in a separate post about Munich as a patent litigation venue. The court is still very efficient (as evidenced by the speed with which it resolved the case decided today), but facing resource constraints stakeholders should be concerned about.

I've asked the court's press office for a redacted copy of today's judgment and may go into more detail--in a follow-up post to this one--if and when I've obtained it. I prepared only for this scenario (non-infringement) and a stay (over validity concerns) in light of the short time between trial and decision--simply too short to resolve all defenses including antitrust. That assumption was right and enabled me to publish this post right after the announcement (though without technical detail).

Share with other professionals via LinkedIn:

Sunday, October 7, 2018

Irreconcilable discrepancies between Huawei's inbound and outbound patent licensing offers

More than three years after naming and shaming Ericsson's privateering deal with Optis Wireless (among a long list of similar partnerships), I became aware of the recent verdict against Huawei in the Eastern District of Texas, which serves to show how Huawei is torn between its strategic interests as a leading Android device maker on the one hand and an aggressive patent licensor on the other hand.

A corporation of that size is party to many disputes at any given point in time, and obviously the shoe will regularly be on the other foot. Parties don't make or change laws, so litigants can merely try to undermine (to a limited and often negligible extent) their adversaries' credibility by pointing to contradictions. However, in two respects such contradictions can be very significant:

  • FRAND licensing negotiations and disputes involve bona fide questions, making inconsistencies more problematic than in such a context as patent infringement.

  • We're all among the potentially billions of judges sitting on the court of public opinion, and some companies spend huge amounts of money on PR agencies and departments--or even on advertising campaigns like Qualcomm's big-time self-promotion in San Diego--to make us believe that they're the good guys. In that respect, it's a serious issue if they blatantly and brazenly apply double standards.

For an example, while Apple's positions on standard-essential patent (SEP) licensing have been perfectly consistent over the years, I really took issue and even got emotional when Apple's lawyers in the second California case against Samsung argued that a few narrow software patents entitled the iPhone maker to a "reasonable royalty"-type of damage award of $40 per device. Those weren't standard-essential, and if that claim had simply been based on lost profits, it would have been a different story. But they said "reasonable", and that's the R--the single most important letter--in "FRAND."

Samsung used to seek remedies over SEPs that I consistently disagreed with. I viewed their litigation tactics far more favorably when they dropped the related claims against Apple and even joined pro-FRAND organizations. Same with Google, by the way, but since it divested Motorola Mobility and contributed to an LTE patent pool, it's probably not going to be active again as a SEP enforcer in the foreseeable future.

If Huawei were a person, the diagnosis would be schizophrenia. Sorry to say so about one of my two favorite Android brands. According to Wall Street rumors, Huawei is (thankfully) part of the rebellion against Qualcomm. Just a couple of days ago, the FTC pointed to a 20-year-old Qualcomm filing that took the very opposite position on the FRAND licensing of rival chipset makers as Qualcomm is taking now. In Huawei's case, we're not talking about two decades in between. What we're seeing unfold here are parallel cases (Huawei's litigation with Samsung in the U.S. and China, and in parallel, PanOptis' litigation against Huawei in the U.S. and Germany, which provoked a Huawei countersuit in China for the purpose of a FRAND determination). It's like Huawei has to wear a different hat every 20 days, and at times every 20 hours, than change positions over the course of 20 years.

In the (Pan)Optis v. Huawei case in the Eastern District of Texas, the parties filed their proposed findings of fact and conclusions of law (after the recent jury trial and with a view to an upcoming bench trial) late last month--(Pan)Optis on the left side, Huawei on the right side. And some of what Huawei says is really interesting (this post continues below the document):

18-09-28 PanOptis v. Huawei Proposed Findings of Fact by Florian Mueller on Scribd

(Pan)Optis claims to have "acquired" (though it may actually be more of a service provider to Ericsson in economic terms) 63 SEP families. I haven't seen a comparable public claim by Huawei, but this Chinese study (in English, though) concluded that Huawei had declared slightly more than 600 patent families to be essential to 4G/LTE.

According to Huawei's own proposed finding of fact #58, "Huawei sent a letter to PanOptis indicating that it was willing to enter into a license to PanOptis’ U.S. declared essential patents for a rate of 0.09%, its EU declared essential patents for 0.056%, and its China and Rest of the World declared essential patents for 0.04%."

When I see a Huawei FRAND position that has a zero to the right of the decimal point, it immediately brings back to memory the FRAND rate that a Chinese court set, at Huawei's request, for InterDigital's SEPs: 0.019%.

Samsung claims that Huawei never substantially lowered its 1.5% royalty demand. Huawei denies that allegation, but without specificity (at least in the public redacted version of the relevant filing).

The recent Texas verdict related to five patents, four of which were standard-essential (and the jury awarded a far lower per-patent rate on each of those SEPs than on the non-SEP). While Huawei can--and according to a couple of more recent filings will--seek a judgment as a matter of law on the merits and on the damages award, for the time being (Pan)Optis has established the actual infringement of four presumed-valid SEPs, while Huawei has yet to prove anything in the United States. It has prevailed on a couple of patents in China, though.

The proposed findings of fact and conclusions of law in the Texas case also deviate from Huawei's positions in the Samsung despite with respect to whether one court should set a worldwide royalty rate:

"[DCL24] As discussed above, PanOptis has never given Huawei a U.S.-only offer. But, in view of PanOptis' decision to sue Huawei on its U.S. patents in a U.S. court, and in view of the very minor share of Huawei's sales in the U.S., and that Huawei has clearly stated its willingness to agree to and desire for a FRAND license covering its US sales, FRAND requires that PanOptis offer Huawei a U.S.-only license, rather than requiring that, in order to get a U.S. license to cover its minimal U.S. sales, Huawei also take a license in other countries where its sales are much greater (e.g., more than 60% in China) under foreign patents not at issue in this litigation. Despite Huawei repeatedly seeking a U.S.-only license and offering to settle the U.S. case, PanOptis has refused to discuss such options with Huawei."

"[DCL26] Like Motorola, PanOptis violates its FRAND obligations by suing Huawei on U.S. patents and insisting on a global license, despite Huawei's comparatively low sales in the U.S. and Huawei’s offer to enter into a U.S.-only license. Admittedly, Motorola sought an injunction against Apple in Germany, and PanOptis is seeking only damages against Huawei in U.S."

In the aforementioned bench trial opening brief, Huawei discourages the Texas court from setting a global FRAND rate:

"Further, as the Court recognized in limiting PanOptis' Count IX to U.S. patents [...], the Court should refrain from exercising its discretion to adjudicate FRAND issues as to foreign patents since doing so would necessitate determinations concerning the scope, infringement, and validity of non-U.S. patents governed by non-U.S. laws."

I agree--but why did Huawei ask the United States District Court for the Northern District of California to make a global FRAND determination in its dispute with Samsung? And why shouldn't Samsung firstly get the chance that Huawei actually had--even if only with respect to a handful of patents--in the Eastern of District (though it now has to get the verdict overturned, either by the district court or on appeal) to litigate its non-infringement and invalidity defenses?

Share with other professionals via LinkedIn:

Saturday, October 6, 2018

Apple and manufacturers: Qualcomm can't demand "billions in royalties" while ducking patent-specific claims

Those seeking to monetize their standard-essential patents (SEPs) very aggressively--such as "Qualwei" (Qualcomm and Huawei), Ericsson, Nokia, or InterDigital--always want to be paid on the basis of some imaginary portfolio value while avoiding concrete, patent-specific decisions altogether or asserting particular patents only for the purpose of gaining unfair leverage in the form of injunctive relief.

If Judge William H. Orrick hadn't enjoined Huawei, it might sooner or later have abused its Chinese SEP injunctions in order to force Samsung to submit to an arbitration process on unfair premises. Qualcomm isn't presently asserting SEPs, but it's trying to enforce non-SEPs, an effort that has dealt the dominant baseband chipset maker setbacks in the ITC as well as Germany.

In its original complaint against Qualcomm in the Southern District of California (filed in January 2017), Apple included declaratory judgment claims (non-infringement, invalidity, and exhaustion) relating to nine Qualcomm patents. In June 2017, Apple amended its complaint and added claims relating to nine more Qualcomm patents. In November 2017, Judge Gonzalo P. Curiel agreed with Qualcomm that there hadn't been any sufficiently clear threats (such as product-specific infringement allegations in the form of claim charts) over those second-round patents, and granted a related motion to dismiss. It was a threshold question that could also have been decided the other way, but the judge, at the time, expressed concern that countless patent-related claims could otherwise be added to that case. But, at any rate, all claims over the first round of patents remained in the case.

Three weeks ago, Qualcomm filed a motion to dismiss Apple and its contract manufacturers' (Foxconn and others) declaratory-judgment claims over the first round of nine patents. As I noted in last year's post on Apple's amended complaint, it was already highly unusual that Qualcomm didn't bring infringement counterclaims. Such counterclaims are compulsory, so by not bringing them, Qualcomm deprived itself of any future opportunity (in this case or in any other) to enforce those patents against Apple. Patent holders very rarely do so. Qualcomm argued in its motion that "resolution of claims regarding nine patents will not further resolution of the parties' broader licensing dispute concerning Qualcomm's portfolio of more than 130,000 issued patents and patent applications worldwide," and pointed to a letter it sent to Apple and the contract manufacturers right before filing the motion, according to which Qualcomm "unconditionally and irrevocably covenant[ed] to refrain from making any claim(s) or demand(s) against Apple or the CMs [...] under the patent laws of the United States, relating to any claim of the Patents-in-Suit [...]". Qualcomm described this covenant as "broad and unrestricted," but Apple and its contract manufacturers disagree: in their opposition (filed on Friday) to Qualcomm's motion, they say this covenant is to narrow to put any dispute over those patents aside, given that Qualcomm

  • is still seeking billions of dollars in royalties based on license agreements at issue in the San Diego case,

  • alleges a high portfolio value when seeking to justify its royalty demands (which Qualcomm says are FRAND, while Apple and its contract manufacturers--and, explicitly or implicitly, antitrust authorities around the globe disagree or at least have serious doubt), and

  • allegedly engages in double-dipping by charging royalties on patents that are actually exhausted due to chipset sales.

I surmise that the last point--exhaustion--is, in light of the Supreme Court's 2017 Lexmark ruling, the one Qualcomm wants to duck more than any other. And Qualcomm generally just wants to be paid on the basis of portfolio size.

Obviously, no court could practically adjudge the merits of many thousands of Qualcomm patents. But it wouldn't be a problem to rule on at least a certain sample of patents, and it just doesn't make sense to let a patent holder get away without any determination on the merits whatsoever.

The fate of Qualcomm's motion and, conversely, Apple and the contract manufacturers' opposition will now depend on whether the new situation Qualcomm has created with its covenant (actually not completely new since it had previously waived certain rights by not bringing compulsory counterclaims) renders any infringement, validity and exhaustion decisions as unnecessary as Judge Curiel thought they were with respect to the second round of nine patents or whether the judge will agree with Apple and the contract manufacturers that it's insufficient for Qualcomm to give up on those patents "under the patent laws" while effectively reserving the right to leverage those patents in antitrust, contract and other legal contexts.

Apple and the contract manufacturers' opposition brief concedes that courts have held declaratory-judgment claims over patents ripe for dismissal after a covenant by the patent holder not to sue over the relevant patents. "But, that is not always so," they write: "Where resolving patent issues remains relevant to live disputes between the parties, such as whether one party owes another payment under a license agreement, dismissal is not warranted." And then they quote from a decision the District Court of Massachusetts made in August 2016 in Esoterix v. Qiagen:

"[T]he resolution of Qiagen's contract-based Counterclaim regarding its obligation to pay royalties under the License Agreement depends on whether the underlying patents are valid or invalid. In light of this dispute, which is still very much alive, LabCorp and Esoterix's promise not to sue Qiagen for infringement does not moot the case."

That precedent, coupled with the fact that Qualcomm actually did provide infringement claim charts for the first-round (though not second-round) patents and very clearly expressed in writing that Apple, by implementing certain cellular standards, allegedly infringes on them), would spell doom for Qualcomm's motion to dismiss if the same common-sense position had been taken by the Supreme Court or the Federal Circuit, or (though patent infringement claims don't go there anymore) the Ninth Circuit. But a district court ruling from the northeast is obviously not binding in the southwest. The United States District Court for the Southern District of California may or may not adopt it. It's more than persuasive, but presumably Qualcomm will now try to find case law that supports, or at least appears to support, its motion to dismiss despite the logic and beauty of the Massachusetts opinion. That's going to be the most interesting thing to look for in Qualcomm's reply brief from a legal perspective.

Qualcomm wants a shortcut, but how could a court just award Qualcomm huge amounts of money without the slightest evidence that valid Qualcomm patents are infringed without being exhausted by chipset sales?

Sooner or later, Qualcomm will have to answer that question. And the more time goes by without Qualcomm achieving a breakthrough in Germany or in the ITC or any other jurisdiction, the harder Qualcomm will try to duck any decision on the merits. It's so much easier to just point to a Texas-size portfolio...

Finally, here's the opposition brief with all public exhibits:

18-10-05 Apple & Contr ... by on Scribd

Share with other professionals via LinkedIn: