FOSS Patents

Four of Qualcomm's eight Spotlight-related patent infringement actions against Apple in Munich were dismissed in January for non-infringement, while the proceedings relating to the other four cases (two lawsuits per patent targeting different Apple entities) were reopened because Qualcomm had not yet had the opportunity to argue that a panel of three examiners at the EPO's Opposition Division erred when they preliminarily agreed with Apple and Intel that those patents are invalid.

I checked again with the press office of the Landgericht München 1 (Munich I Regional Court) and was told yesterday that "for the time being" the court scheduled another trial session for June 13, 2019. The parties already argued (non-)infringement at a first hearing as well as a trial last year, so presumably the sole focus of the additional June session will be on (in)validity.

Five and a half month later, on November 26, 2019, the EPO's opposition panel will hold a hearing and make its decision (click on the image to enlarge; this post continues below the image):

In German infringement proceedings, defendants' ability to argue invalidity is limited. As opposed to a full invalidity defense, there is bifurcation: full invalidity challenges must be brought in separate fora (the EPO for "young" patents, or the Federal Patent Court of Germany), and the courts deciding on infringement will, at best, stay a case pending the resolution of a parallel invalidation proceeding. The decision on a request for a stay is based on limited analysis as opposed to a full trial. In practical terms, obviousness arguments are hard to prevail on; stays are usually based on indications of non-novelty.

However, when a competent forum deems a patent invalid, its findings bear significant weight with the infringement court. A credible invalidity opinion from a forum within the European Patent Organization (the international organization running the EPO) often has the practical effect of reversing the burden of proof, requiring the patent holder to persuade the court that the related findings are either not well-reasoned or, should a patent not be young enough that the EPO can revoke it Europe-wide, are based on a legal standard that German courts don't apply to the German parts of EPO patents.

Here, the EPO's opposition panel took a preliminary position, with the actual hearing taking place later this year. A preliminary holding like this obviously has significantly less weight than a final (but appealable) revocation decision. But it was important enough to warrant a reopening of proceedings.

Qualcomm isn't going to gain any leverage from those Spotlight (a search feature) cases anyway as iOS 12 contains a workaround. Any "win" would be purely symbolical, as would any potential damages. But Qualcomm is facing huge problems in its patent infringement actions against Apple, which is presumably the reason why it keeps spending a lot of money on cases that won't result in any serious leverage. Qualcomm is desparately trying to prove that its patents (in these cases, non-standard-essential patents) have value, and to portray Apple as an infringer--even when there's already a workaround in place.

Maybe Qualcomm will stipulate to a stay so the Munich court won't have to hold the mid-June trial session and write up an order to stay the case. Given that the EPO will decide only five and a half months later, it's hard to imagine (though I have no idea of what Qualcomm told the court as to why it believes the EPO got it wrong) that Qualcomm will be able to dissuade the court from staying the cases. It's also very, very likely that the EPO will revoke those patents, in which case Qualcomm will have to appeal the Opposition Division's decision to a Technical Board of Appeal. All of his is taking time, but those cases are pretty clearly going nowhere.

Later this month a first hearing in a later-filed Munich Qualcomm v. Apple case over an antenna patent will be held.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

This last day of January 2019 dealt another setback--technically, a set of setbacks--to Qualcomm as half of its patent infringement lawsuits against Apple's Spotlight search in Munich have been tossed and the other half is going down the tubes. On the legal front, this month as a whole has gone, for the most part, as if Qualcomm had set out to prove Murphy's law. A summary will be provided further below.

The Munich I Regional Court thankfully published a press release on its website (in German) regarding the decisions the 7th Civil Chamber under Presiding Judge Dr. Matthias Zigann announced today in Qualcomm's eight Spotlight cases.

It was already clear before today's announcement that some of those lawsuits were bound to be dismissed because there was no doubt at the hearing that a claim limitation (a "standby" feature for apps) wasn't practiced. I reported that fact after the trial and mentioned it again in my most recent post relating to this set of cases, which I wrote after the European Patent Office sided, in a preliminary opinion, with Apple and Intel (I should even say Intel and Apple because Intel was first to challenge those patents), represented by the patent attorneys of the Samson & Partner firm (which consistently delivers great work, and was on the winning side in another case today--involving totally different parties--as I'll mention at the end of this post).

These are eight cases because there are two cases for each of the four patents, always one against Apple Inc.--the U.S. HQ--and one against two European Apple entities. The older ones of the four patents-in-suit (too old to be challenged before the European Patent Office post-grant, but they are being challenged in the Federal Patent Court of Germany), EP1956806 and EP1955529, are not infringed--for the aforementioned reason (no "standby"). That disposes of four of the eight cases. Qualcomm can appeal, but the "standby" question appears to be an insurmountable hurdle for Qualcomm, and the EPO opinion on the other two patents, while neither final nor binding on the Federal Patent Court of Germany, strongly suggests that the entire patent family is doomed to be invalidated.

Non-infringement was successfully argued for Apple by Freshfields Bruckhaus Deringer's Prince Wolrad of Waldeck and Pyrmont.

Instead of outright ordering a stay of the other four cases (over the two younger patents the EPO deems invalid on a preliminary basis), the court made the procedural decision to give Qualcomm a chance---by way of a reopening of the proceedings--to respond to the EPO's preliminary findings of invalidity. That's because the EPO opinion came in so shortly (just about a week) before today's ruling date in the infringement cases.

Obviously the cases will be decided by the court, not by a blog, but I really view those four cases as "zombie cases": a quartet of dead men walking. No matter what Qualcomm will say (and Quinn Emanuel's Dr. Marcus Grosch won't leave a stone unturned; after all, he even salvaged a narrowed version of Motorola's infamous push notification patent by creatively crafting a cascade of amended claims), the EPO's preliminary position on revocation will bear enormous weight with the court unless a very clear legal error were to be identified. That's because the courts of law hearing infringement cases in Germany recognize that the EPO (which also applies to the Federal Patent Court of Germany) not only spends a lot of time evaluating validity challenges but also has technical expertise on board.

Also, the EPO's opposition hearings will take place later this year, so if the Munich I Regional Court allowed enforcement (which wouldn't matter because iOS 12 contains a workaround) but the EPO issued a final (though appealable) revocation decision later, it wouldn't look good, while the EPO's preliminary opinion justifies a stay of those four cases (short of a silver bullet on Qualcomm's part), and should the EPO change mind, Qualcomm wouldn't have lost much time for enforcement.

Getting back to the reference to Murphy's law further above, this is a summary of legal developments for Qualcomm during this month (it may read as if this had been authored by those shortsellers, but it's simply the way it is:

• The FTC v. Qualcomm antitrust trial in San Jose (Northern District of California) obviously towers above all other events. Closing arguments are rarely a surprise. The FTC's case was validated by testimony from the industry at large (American companies, companies from ally nations, and others). Qualcomm only had 2.5 companies on its side--Nokia (a failure in the handset business), Ericsson (another handset failure, and they just count as 0.5 here because they attacked the value of Qualcomm's patents and its practices, siding with Qualcomm only on the question of chipset-level licensing), and InterDigiTroll.

  Qualcomm's lead counsel delivered a very strong performance on Tuesday, but Judge Koh was looking almost exclusively at her documents, the real-time transcript, and her notes, so all the rhetoric and gestures bounced off of her like a tennis ball thrown incessantly at a stone wall.

• On January 15, the Mannheim Regional Court threw out a Qualcomm suit because it held that the iPhone doesn't infringe under a proper claim construction.

• Starting that same day (January 15), the Patent Trial and Appeal Board (PTAB) of the United States Patent & Trademark Office (USPTO) initiated inter partes reviews of 21 (twenty-one!) Qualcomm patents challenged by Apple. Such reviews are initiated only if a reasonably serious issue is raised.

• Then came the EPO opinion on two patents and, a week later, the related Munich decisions discussed above.

All that bad news is not counterbalanced by the symbolical victory that is a preliminary injunction in Germany, barring Apple from saying that the iPhone 7 and 8, against which Qualcomm secured a pair of injunctions in December (with the court ordering remedies while being agnostic as to the merits), remains available at more than 4,000 points of sale. But whether or not Apple is allowed to say it, there really doesn't appear to be an availability problem. On the Internet, those devices are just one Google search away from Apple's German online shop, and in the real world, I doubt there's any Apple Store from which people couldn't find the next point of sale (where those devices continue to be sold) in walking distance--in many cases, it will be like just crossing a street.

Headed for an antitrust defeat and failing to gain leverage from patent infringement actions against Apple and Intel, Qualcomm is trying to influence its fate out of court.

In other news, totally unrelated to Qualcomm, the Munich I Regional Court today rejected a complaint by Netlist against HP and SK hynix because it found no infringement. That case involves a standard-essential utility model, which is why I got interested in it. In my trial report I had already expressed my view that a rejection or stay appeared more likely than a victory for Netlist. I've already told some Qualcomm-aligned Twitter trolls that, in other words, my highest priority is to have a strong prediction record.

The utility model case was won by a Bardehle team led by Professor Peter Chrozciel and Dr. Anna Giedke as well as patent attorneys Dr. Georg Jacoby and Dr. Robert Baier from the Samson firm (mentioned further above for its accomplishments before the EPO against Qualcomm's Spotlight patents). In the United States, SK hynix is represented by Sidley Austin against Netlist. Sidley and Professor Chrocziel (at the time, with Freshfields) already worked together very successfully when representing Microsoft in the United States and Germany.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

While the Ninth Circuit's decision to let Qualcomm appeal the certification of a 250-million consumer class and Judge Koh's related announcement to stay the 5ドル billion class action (which builds upon the FTC case, but not the other way round) pending the appeal is good news for Qualcomm, Kerrisdale Capital's opinion that Qualcomm's stock may lose half its value and a whopping number of 21 (twenty-one!) inter partes reviews (post-grant proceedings that may result in the invalidation of patents) instituted over the last eight days by the Patent Trial and Appeal Board (PTAB) of the United States Patent & Trademark Office (USPTO) against Qualcomm are significant setbacks.

Now there's bad news from Munich, the very city in which a regional court (comparable to a U.S. district court, though only the nation's top courts are federal courts) recently granted Qualcomm a pair of injunctions over a chip patent against Apple that Qualcomm is enforcing despite an ongoing appeal after posting bonds over more than 1ドル.5 billion--but without getting any serious leverage over Apple as the enjoined products (iPhone 7 and 8) remain widely available through carriers and retailers. Qualcomm has obtained a preliminary injunction barring Apple from telling it like it is, so Apple must not issue press releases or give answers to journalists according to which those devices are available at more than 4,000 points-of-sale. But that preliminary injunction, which according to media reports is based on misleading promotional statements, just puts the next symbolical victory on top of another without representing actual leverage.

I asked the Munich I Regional Court, but the court declined to comment. I also asked the Munich Higher Regional Court about the status of Apple's appeal of the injunctions, but received no new information. The last thing I heard was that Qualcomm still had time to respond to Apple's motion to stay the enforcement of those injunctions.

In a week from today, the Munich I Regional Court's Seventh Civil Chamber under Presiding Judge Dr Matthias Zigann will rule on eight Qualcomm complaints over a set of four patents. The number of eight cases is due to two complaints having been filed over each of the four patents (one against Apple Inc., the U.S. parent entity, and another one against a couple of European Apple entities). The patents-in-suit are the German parts of four European patents from the same family, each covering a "method and device for communication channel selection": EP1956806, EP1955529, EP3054658, and EP3094067. The claimed "inventions" were made by SnapTrack, a GPS-focused Silicon Valley startup aquired by Qualcomm in 2000 for 1ドル billion.

Even if Qualcomm won any of those eight cases (which is highly unlikely now in light of what I'll explain below), it wouldn't get any leverage as iOS 12 contains a workaround.

But Qualcomm won't realistically win a single one of those eight cases.

Some of those cases were already dead in the water at the time of the trial because the particular patents asserted in those cases require a "standby" functionality where an app does not have to be started by the user by means of tapping on an icon.

There's a strong case for non-infringement (the absence of a filter that a correct claim construction would require), but even if the court wasn't sure about non-infringement, those cases are in all likelihood going to be stayed (they just wouldn't be stayed if they could be thrown out altogether).

As I reported in May, Apple and Intel are attacking two of those four Spotlight patents (the two youngest ones) in opposition proceedings before the European Patent Office. Yesterday (January 23, 2019), the EPO's Opposition Division scheduled a hearing in those two parallel proceedings for November 28, 2019, and tue summons to the hearing came with an annex: a preliminary opinion according to which those patents are invalid.

Here's one of the two near-identical opinions, and here's a link to the other (this post continues below the document with an explanation of the EPO's findings and a summary of Qualcomm's poor results in Germany):

19-ß1-23 Annex to EPO Commu... by on Scribd

Let me explain:

• First, the decision looks at the relevant independent claims in their granted form, and finds them to be non-novel.

• It then analyzes the relevant dependent claims and finds that what they add to the independent claims they're based on doesn't make them any more patentable.

• Earlier in the infringement proceedings, Qualcomm already realized that those patents weren't going to survive in their granted form. That's why Qualcomm's counsel, Quinn Emanuel's Dr. Marcus Grosch, altered course and henceforth sought to prevail on the basis of amended (in the sense of "narrowed") claims.

  In the EPO documents, those amended claims are called "auxiliary requests."

  The two EPO opinions find that the amended claims ("auxiliary requests") aren't patentable because the additional elements introuced aren't supported by the patent specifications.

As a result, those two patents are (unless the EPO changes its opinion before or at the late-November 2019 hearing, which I doubt very much) finished. German courts hearing infringement cases attach significant weight to such preliminary holdings by the EPO because they know that the EPO's examiners have technical knowledge and give a fair amount of thought to such matters.

Those opinions relate to only two of the four patents-in-suit: EP3054658, and EP3094067. However, they're all from the same family and largely identical. The only reason for which Apple and Intel couldn't file oppositions with the EPO is a statute of limitation: Opposition to an EPO patent must be filed within nine months of the publication of the notice that the patent has been granted. But the other two patents are being challenged before the Federal Patent Court of Germany (Bundespatentgericht), which is even stricter than the EPO.

Some other German Qualcomm v. Apple cases have yet to go to trial (or a Munich first hearing; at some point one was scheduled for late March over an antenna patent).

The ones that have gone to trial are a mix of one symbolical victory that doesn't give Qualcomm leverage and a long list of defeats:

• Defeat #1: The first German Qualcomm v. Apple patent trial, held in early June 2018 in Mannheim, involved a patent so obviously invalid that Qualcomm surrendered by stipulating to a stay.

• Defeat #2: In mid-October 2018, the Munich I Regional Court threw out a Qualcomm case over a post-passivation layer patent. Judge Dr. Schoen ("Schön" in German), who filled in for Judge Dr. Zigann in that case, said at the trial that he resisted the urge to call up one of Qualcomm's "experts'" universities given how devoid of substance his testimony was, saying that he thinks someone who submits such an expert report in a civil litigation shouldn't be allowed to teach the young generation.

• Symbolical victory: then came the envelope tracker ruling shortly before Christmas (mentioned above).

• Defeat #3: On January 15, the Mannheim Regional Court threw out a Qualcomm suit because it held that the iPhone doesn't infringe under a proper claim construction.

• Impending defeat #4 over a family of four patents: with the EPO's findings discussed further above, those Spotlight cases are going nowhere.

• Very probable defeat #5: At a Mannheim trial in October 2018, Presiding Judge Dr. Holger Kircher strongly suggested to Qualcomm to stipulate to a stay of a case over an electromigration-related patent. Like in the Spotlight cases, Qualcomm had already realized that the patent-in-suit was indefensible in its granted form, and had presented amended claims. But according to Judge Dr. Kircher, the claim limitations Qualcomm added are "devoid of technical substance." An order to stay those cases is by far the most likely outcome though one never knows what may have happened since October) when the Mannheim court announces a decision on February 19, 2019.

Qualcomm's patent infringement cases haven't delivered, apart from reports that overstated the significance of some symbolical victories. Too many gullible people write about patent cases. I know that last statement sounds arrogant, but unfortunately it's true. If everyone writing about them understood what the actual, practical, technical and economic consequences are, many misunderstandings could be avoided. Qualcomm has first-rate PR people, internally as well as externally. But I look at the substance of those cases and the actual impact, and I'm underwhelmed for the time being.

Neither has Qualcomm gained leverage nor has it had a "hit rate" that would support its claims of having an unbelievably strong patent portfolio compared to other companies in the industry. And some of the patents-in-suit aren't even original Qualcomm patents but were acquired at some point.

Follow @FOSSpatents

Share with other professionals via LinkedIn:

Yesterday the Munich I Regional Court held a six-hour (including breaks, though) trial on Qualcomm's eight lawsuits asserting four different search user interface patents against Apple's Spotlight search, with two lawsuits per patents targeting a total of three different Apple entities. A first hearing had been held in early May.

That part of the wide-ranging, earth-spanning, multifaceted Apple-Qualcomm dispute is, however, strategically so unimportant that it's not worth multiple posts or anything. That set of eight cases is a total waste of court and party resources--sort of a tempest in a teacup--as these Munich Spotlight cases have been defanged in three important ways:

• Presiding Judge Dr. Matthias Zigann made clear, and Qualcomm apparently failed to change his position, that some of those lawsuits will result in a non-infringement finding because they require a "standby" functionality where an app does not have to be started by the user by means of tapping on an icon. But even that is strategically not that important; it just has cost implications under Germany's "loser pays" rule.

• Just like in the Mannheim case tried earlier this week, at least one of Apple's contract manufacturers, Pegatron, is licensed. Qualcomm conceded this, which may have cost implications as well. It could still obtain an injunction as Pegatron is not the only Apple CM, but theoretically Apple could just have Pegatron make whatever it wishes to sell in Germany after an injunction. For the patents asserted against Spotlight, even that logistical measure won't be necessary:

• Qualcomm's lead counsel, Quinn Emanuel's Dr. Marcus Grosch, conceded that iOS 12, at least in Germany (though purportedly not in other places), "works differently" with respect to the accused feature. When asked by the court whether iOS 12 keeps clear of infringement, Dr. Grosch stopped short of admitting that much, but he did reiterate that it "works differently."

  That's a highly significant concession. Enforcement of an injunction is very similar in Germany to how it works in the U.S.: if you as a patentee identify an infringement despite an injunction being in force and effect, you have to file a contempt motion and get the infringer sanctioned. If the infringement involves new products, but is in technical terms materially the same, the old injunction applies (in the U.S. it's more important than in Germany to ensure that the language captures such future infringements). Otherwise, you need a new proceeding on the merits and a new injunction (as mentioned, for instance, in the Apple v. Samsung context). In Germany, the key term is "kerngleicher Verstoß" ("violation with identical core"). I can't see how Qualcomm could still make such a claim, should it obtain an injunction over older iOS versions, against iOS 12 with respect to the Spotlight patents.

For reasons previously explained, I don't believe there is an infringement here because the accused products lack a certain filter. But that aspect of the case wasn't rediscussed yesterday, and most of the time was spent on (in)validity issues, followed by a licensing (behind closed doors) and antitrust discussion.

In the validity part it was easy to see why iOS 12 is apparently in the clear: in connection with a search for ways to contact a person, Spotlight in iOS 12--purportedly just for Germany--displays text, but not icons, when distinguishing between a given contact's phone number, email address, and other contact data. Does this sound ridiculous to you? It sure does to me. But Qualcomm's last straw for getting at least some valid, but significantly narrowed, claims at the end of the ongoing nullity and revocation against rests on the icon/text distinction. I don't know whether one should laugh or cry. Anyway, this makes the presence of icons critical for the infringement theory as well, but iOS 12 simply doesn't show icons to German users in that particular context.

So if Qualcomm were to obtain a German injunction, it would only be symbolic, but commercially irrelevant.

Frankly, I can't even see how the patents will survive on that basis:

• Judge Dr. Zigann raised the question of whether the use of icons instead of text meets the technicity requirement under the European Patent Convention. In order to do so, it needs to have a technical effect. He tossed out the possibility of saving memory, but as a software developer I know that even on low-resolution screens in the 1980s (long before the patents' priority date), even a small monochrome (!) icon already occupied more memory than a string like "email" or "phone". Nowadays, with higher resolutions and colors, icons are more memory-consuming by a large factor. Dr. Grosch said that whether an icon saves memory "depends on details of the execution," but that's simply not true. There is no way to make it more space-saving unless you store a text string, say, 1,000 times in memory instead of once. The other possible technical effect that was discussed was that an icon might save screen space. While that is true, I believe the mere replacement of a text with an icon still isn't technical enough to meet the technicity requirement: it's at the semantic level, and then you would also save screen space by replacing "telephone" with "phone", or German texts with English or Spanish ones (less verbose languages), none of which would seriously count as a technical contribution to a solution of a technical problem. Something technically more sophisticated would have to happen in the background.

• While German courts are reluctant to stay infringement cases for doubting the validity of patents for lack of an inventive step, the use of icons instead of text is such a ridiculous thing that it's really easy to see it's common general knowledge of a person of ordinary (or even way less than ordinary) skill in the art--and would have been so not only at the time of the priority date but decades (!) before. I believe German courts just make a summary determination of the prospects of a validity challenge, but here they could bring in a court-appointed expert and resolve the question very quickly. All I can say is that I knew about icons in the early 1980s and would have known about them before if I had been born earlier.

• But the suggestion to use icons instead of text is also in the prior art reference most of the (in)validity discussion focused on. Thankfully, Apple and Intel's opposition briefs (petitions for revocation) filed with the European Patent Office are publicly available. "Haitani" is a PCT patent application for a "user-interface for computing devices" (WO2007019481). If you run a full-text search in your browser for "icon" over the description, you find multiple explanations. So there wouldn't even be a need to combine Haitani with any other prior art reference. It's all there. However, Qualcomm tried to get mileage out of some suboptimal drawings.

Due to resource constraints I'll write about in a different post where I'll express concern over the future of Munich as a patent infringement venue, the three-judge panel didn't have the chance to confer prior to the trial. They'll do so afterwards, and maybe they'll then see that a stay is warranted. Or revisit infringement. But even if Qualcomm got an injunction, it would just be toothless, pointless, useless (apart from maybe decorating someone's wall).

As for antitrust, such a weak set of patents was the least interesting context for discussing Qualcomm's "grand evil scheme" (a term that was not reiterated yesterday, but the message was consistent). I would like to mention just three things that the judge asked, all of which were very valid questions:

• How could Qualcomm enforce its intellectual property rights without violating antitrust law? Apple mentioned different aspects and criticized Qualcomm for the selectivity of its attacks. In my opinion, the single most important part is that Qualcomm would recognize the antitrust implications of its actions and would seek only monetary damages, not injunctive relief. That would not be unprecedented. It's been done in Germany (such as by IPCom against Apple, over a standard-essential patent); it's also been done in the U.S., where Nokia in its 2009 Delaware lawsuits (again against Apple) made a distinction between standard-essential patents, for which it merely sought to be compensated, and others (they called them "implementation patents"). I remember this very well because Nokia's differentiated approach made a lot of sense to me at the time. In late 2010 I compared Microsoft's FRAND lawsuit against Motorola to Apple's FRAND defense against Nokia. I would still say that's the way to go. Of course, seeking way supra-FRAND damages could also constitute a violation. But in Germany, the first step is usually just to get a ruling on the merits, with damages being addressed in a subsequent proceeding.

• For how long should Qualcomm be barred from fully enforcing its patents against Apple? The answer was: as long as it's in a dominant position in the baseband chipset market.

• Couldn't Qualcomm just work around it by means of privateering (selling patents to a troll who would then go after Apple)? This part wasn't addressed in much detail. But I've been campaigning against privateering on this blog for a few years and have looked into some of those typical patent "transfer" agreements, and I can provide the answer: it wouldn't work. In order for Qualcomm to leverage patent suits to force Apple to stop buying from Intel to and force it to return to buying baseband chips exclusively from Qualcomm again, Qualcomm would sooner or later have to tell Apple that certain troll lawsuits would also be withdrawn the moment Apple does so. So Qualcomm couldn't hide the fact. Then there are requirements for patent transfer agreements under U.S. law with a view to standing, and in the post-eBay era in the U.S., trolls hardly obtain injunctions. But even if they wanted to do it only for Germany, they'd have to sooner or later admit that they can call off the troll, and then the antitrust implications (in the event of a complaint subsequent to such admission, or to a settlement after which the troll lawsuit is magically withdrawn) would be the same.

A decision has been scheduled for December 20, 2018, which is partly attributable to leave granted to Apple for a post-trial brief (two weeks) but may mostly be due to the present resource constraints of the 7th civil chamber of the Munich I Regional Court.

Follow @FOSSpatents

Share with other professionals via LinkedIn: