Showing posts with label Google. Show all posts
Showing posts with label Google. Show all posts

Monday, April 12, 2021

Apple and Google abuse their app store monopolies to obstruct governmental COVID pandemic control efforts in the UK (not even for the first time)

For a long time I gave Apple and Google the benefit of the doubt with respect to app store policies. When my own app development company got affected by the utterly unreasonable COVID app rules those monopolists had promulgated more than a year ago, I couldn't help but conclude that the situation was unsustainable. And brought my own antitrust complaints against those companies in multiple jurisdictions (including the UK) though I continue to agree with those companies in some other areas, particularly patent policy.

It's bad enough that private companies like the Coronavirus Reporter team and mine were prevented from making our little contributions to the fight against COVID-19. But what's really unfathomable is that Apple and Google's hubris even impedes governmental pandemic control efforts.

This must be a wake-up call for lawmakers, regulators, and courts. Alternative third-party app stores for iOS and Android are absolutely needed. Even governments need such alternatives in a situation like this.

The BBC's technology desk editor Leo Kelion reported today that "[a]n update to England and Wales's contact tracing app has been blocked for breaking the terms of an agreement made with Apple and Google." Yes, this is about the official contact-tracing app provided by the National Health Service (NHS).

With UK shops, restaurants and pubs reopening today thanks to a relaxation of COVID prevention rules, it was actually a very smart idea for the NHS COVID-19 app to ask users to scan QR codes when entering such places, thereby enabling the system to inform people if they had been in a virus hotspot at a critical moment.

In the Western world, contact tracing has failed to make a noteworthy positive impact. In parts of Asia, however, those apps made a huge contribution because people were not even allowed to enter restaurants unless the contact-tracing apps on their smartphones greenlighted them (meaning they had not recently been near an infected person for a certain period). It made a whole lot of sense for the UK to adopt what worked in Asia.

But Apple and Google are not susceptible to reason in the COVID context. They shamelessly distribute (partly for free, partly for money) material that promotes bogus medications and treatments. Such material may not explicitly mention COVID-19, but stuff like Homeopathy for Epidemics discusses pandemics at a general level and certainly talks about COVID symptoms (even without mentioning COVID). Such material raises false hopes that what has been proven over and over to be fake medicine could solve the problem. Homeopathy is essentially about putting tiny white sugarballs in your ear and you'll be fine without masks, without vaccination, and don't have to see a doctor if you show symptoms of COVID-19 because your body, with the help of bogus medicine, will cope with everything all by itself.

While the U.S. Department of Justice brought enforcement action against a "chiropractor" promoting fake COVID-19 treatment, Apple and Google get away with the distribution of disinformation.

Apple and Google abuse their app store monopolies in many ways. As the BBC notes, "[u]nder the terms that all health authorities signed up to in order to use Apple and Google's privacy-centric contact-tracing tech, they had to agree not to collect any location data via the software." And on that basis, the latest update to the UK's COVID-19 app was rejected.

Pacta sunt servanda--contracts must be fulfilled--but not when abusive monopolists unilaterally impose unfair and unreasonable terms. Epic Games was absolutely right last summer to refuse to comply with Apple's and Google's in-app payment rules, and the UK government shouldn't be bound to illegal terms either.

There already is an ongoing UK investigation of Apple's suspected anticompetitive conduct in connection with the App Store. On March 30, Epic complained to the UK's Competition & Markets Authority. Competition enforcers in the UK should say "enough is enough."

About a year ago, Nature reported on contact tracing apps and mentioned that an earlier version of the NHS app was tested, "[b]ut because this app eschews Apple and Google’s protocol, it will not be able to run in the background on iPhones." An expert called this "a nail in the coffin." Obviously, contact tracing is of little use if you actually have to have the contact tracing app running in the foreground all the time.

So the UK has been hit by Apple's abusive conduct for the second time. Nature mentioned that "[t]he United Kingdom and France [we]re still pursuing centralized options." By now we know that contact tracing apps have been next to useless in those countries.

Even the contact tracing apps developed by several U.S. states, such as Utah and the two Dakotas, ran into the problem that "Apple and Google will not let apps that record location data use their APIs."

As the BBC's Rory Cellan-Jones accurately notes:

"What this underlines is that governments around the world have been forced to frame part of their response to the global pandemic according to rules set down by giant unelected corporations.

"At a time when the power of the tech giants is under the microscope as never before, that will leave many people feeling uncomfortable."

Apple uses privacy as a sword and a shield, and Google has recently discovered privacy as a means of cementing its monopoly with a technique it calls FLoC. Apple's credibility in the privacy and security contexts will be among the issues to be discussed in court next month, and Apple appears to be profoundly worried--so much so that Apple didn't event want to justify its App Store abuse at a United States Senate hearing, but after a letter from Senators Klobuchar (D-Minn.) and Lee (R-Utah) reconsidered.

Share with other professionals via LinkedIn:

Monday, April 5, 2021

Supreme Court deems Google's use of Java APIs in Android fair use, thus no infringement--doesn't reach API copyrightability

Based on how the Google v. Oracle Supreme Court hearing went in October 2020, it appeared to be a given that the Java APIs in question were copyrightable, and the fair use debate was over whether the Federal Circuit had correctly ruled against Google or whether the San Francisco jury would have had to be afforded so much deference that a judgment as a matter of law wasn't warranted. In the former case, the case would have gone back to San Francisco for a remedies determination. In the latter case, the Federal Circuit would likely have remanded for a retrial, as Oracle was disadvantaged by the district court.

Surprisingly, the Supreme Court has just declared Google's copying of thousands of lines of declaring code to be fair use, thereby substantially weakening software copyright protection in the United States as there had not previously been a case involving such a substantial amount of undisputedly original and creative program code that someone else was allowed to incorporate into a competing product and distribute billions of times.

This decision was supported by six of the nine justices. Only Justices Thomas and Alito dissented (and noted that the majority didn't want to address copyrightability because it couldn't have reached its fair use conclusion thereafter). Justice Barrett was appointed after the hearing.

The per curiam focuses only on fair use. Copyrightability didn't have to be addressed as the case has been resolved in Google's favor, more than ten years after it was brought. Last July, I already expressed concern that the court might not say much about copyrightability. At the time, I wrote:

"If the Supreme Court answered the 'fair use' question in Google's favor on the basis of jury deference [it now actually did so on the merits], it might or might not discuss the standard for software copyrightability in detail. Whether the Federal Circuit's copyrightability holding would be affirmed explicitly or (by reaching "fair use") mostly implicitly, the copyrightability of API code would continue to be a reality in the United States."

Given that the justices were pretty much unanimously leaning toward copyrightability in October, it would be quite risky for anyone to consider API declaring code uncopyrightable. However, technically the Federal Circuit's copyrightability decision hasn't been affirmed either.

Contrary to what many others will say, today's decision is bad news for software developers. We do need certain fair use rights, sure. But overreaching fair use encourages infringement. The simplest way to put it is this: if someone created a platform and later turned around on developers, alleging copyright infringement by continuing to use some API code in the apps themselves, that would raise issues--and if developers didn't have an equitable defense anyway, they should at least have fair use rights. In Oracle v. Google, however, the issue was much more narrow: it was about a new platform using another platform's API code to compete--in fact, displace--the older one.

The syllabus says: "In reaching this result, the Court does not overturn or modify its earlier cases involving fair use." That sounds like the ruling is meant to be of only a narrow scope. But it doesn't change anything about this being a major departure from what the fair use standard used to be, especially with respect to software. It definitely stretches the envelope, weakening copyright as a vehicle for protecting software.

Copyright and patents are intellectual property regimes that were created in centuries before the advent of computer programs. Without digressing into details, software patents are among the most controversial categories of patents (second only to so-called "patents on life"). With copyright, there are plenty of issues as well. For example, it is commonly accepted that object code--and not only source code--is protected by copyright. But object code is technical, binary, machine-readable, not human-readable. It's a stretch to apply copyright protection to object code, but in the alternative one would have to come up with a software-specific sui generis IPR. It has been suggested that a sui generis right--somewhere between copyright and patents--is needed, though no such initiative has gained traction to date. I wouldn't rule out that it might happen in the future, but certainly not in the near term.

There are also external factors due to which copyright protection of software as well as software licenses that rely on copyright to mandate reciprocity, which is called copyleft (and also weakened by today's ruling) are less key today than they were, say, 20 or 30 years ago: cloud computing and platforms.

  • When software actually gets distributed to end users, it's much easier to identify copyright infringements. And copyleft generally applies only to distribution. As long as software stays on a server, it may commit infringements that are never detected, and most copyleft licenses just don't apply.

  • In the platform economy, might all too often makes right. That's why Epic Games is suing Apple (the trial is less than a month away). Apple's airtight control of iOS and of what gets installed on a billion users' devices doesn't depend on whether APIs are copyrightable or whether software is patentable. Some copyright protection is needed because otherwise someone could just steal iOS and build alternative iOS devices--but they don't even need to own the copyright in their APIs as long as the operating system allows only Apple's own App Store to install apps, which in turn are "curated" by Apple and only Apple. It's all about market power, and the only remedy against that one is antitrust--or antitrust-like laws such as the upcoming EU Digital Markets Act--as fair use wouldn't open the App Store.

    There's plenty of people out there now who are celebrating today's Supreme Court decision as promoting innovation, competition, and openness. In reality, the net effect will be the opposite. When Sun created Java, they allowed everyone to make and publish apps for it. Sun adopted a dual-licensing model under which you could either get Java under the GPL free software license or take a commercial license. Sun is history--it was acquired by Oracle. The next company contemplating the development of a comparable platform will look at what happened in Oracle v. Google. Against that background, it may either be discouraged from making the investment in the first place--or it may be encouraged to pursue an Apple-like platform business model ("walled garden") and create network effects through a non-open system with cloud components, an exclusive app store, and so forth. In other words, if you can't own software, you'll try to own (access to) users.

The case appeared dead in 2012 after the district judge held thousands of lines of program code uncopyrightable, and a few years later after a second jury agreed with Google on fair use. The appellate attorney they call the Defibrillator, Orrick Herrington Sutcliffe's Joshua Rosenkranz, twice managed to revive the case. Every time he won an appeal, Google appointed a new lead counsel. Ultimately, Goldstein &, Russell's Thomas Goldstein won the case for Google. (By the way, Mr. Rosenkranz is on Apple's team against Epic, so we may soon see him in action in a high-profile software platform case.)

Lawyers are not the reason Oracle lost this. Google's network of allies and supporters, including a number of organizations funded by Google, have for more than a decade been campaigning against Oracle's case. Oracle never managed to convince large parts of various relevant communities (which are mostly just vocal minorities) that what it was trying to achieve here would ultimately be good for developers. Certain justices indicated at the October hearing that they were aware of widespread concern over an Oracle victory being harmful to software development. That was just fear, uncertainty, and doubt (FUD). But it worked.

It may also have helped Google that the Supreme Court has had to overrule the Federal Circuit in a number of patent cases, so the Fed. Cir. may have a certain reputation of being exceedingly right holder-friendly. I've seen Federal Circuit decisions that really went too far. In this case, however, the Federal Circuit was absolutely right about (un)fair use.

[Update 1] I tweeted this remark concerning the term "user interface":

[/Update 1]

[Update 2] Oracle issued the following statement attributable to Dorian Daley, Oracle's EVP and General Counsel:

"The Google platform just got bigger and market power greater — the barriers to entry higher and the ability to compete lower. They stole Java and spent a decade litigating as only a monopolist can. This behavior is exactly why regulatory authorities around the world and in the United States are examining Google's business practices."

[/Update 2]

Share with other professionals via LinkedIn:

Saturday, March 27, 2021

Has European Commissioner Thierry Breton already announced that Apple will have to allow alternative app stores? A matter of interpretation.

One of the first LinkedIn posts I read this morning was from the Coalition for App Fairness, which was founded last year by Epic Games, Spotify, Match Group and others. When the CAF started, I firstly wanted to wait and see, but at the start of this year I already predicted on this blog that it would keep growing. My own app development company may at some point apply for membership, but even in that case I'd obviously retain my independent opinion. It was high time someone founded the CAF, given that a couple of other organizations claim to represent app developers while in reality being paid and remote-controlled by Apple in one case, Google in the other. It's laughable when an entity claims to represent app developers but doesn't support Epic against Apple, for example.

So the CAF pointed to an article published by EU Internal Market Commissioner Thierry Breton on LinkedIn, entitled DSA/DMA Myths -- What is the EU digital regulation really about?

According to CAF's interpretation of the article, Mr. Breton is "stressing the importance that all gatekeepers allow other app stores on their platforms. This would mean that for the first time, there will be real competition for the App Store." (emphasis added)

It's obvious that I would want this to happen. Competition works wonders. This isn't just about the commission on in-app payments. When they reject your app and won't let you publish it at all or force you to give up on your original concept, your focus is not on 15%, 30% or any percentage for that matters. As Epic will argue in the May trial, alternative app stores can do a better job at curation (app reviews). I don't know whether the Epic Games Store, if it already existed on iOS, would have accepted my app (we'd have to build a Windows version and submit it to them to find out), but considering that similarly-themed games are available on Steam (a pretty meaningful point of reference), the Samsung Galaxy Store, the Microsoft Store etc., I'd be reasonably optimistic. At a minimum I would know that whoever (Apple, Epic, or any third party) rejected it would have to assume that some other app store might carry it. That would discipline all of them, and rejections would become more reasonable. Some people blame the reviewers, such as the Coronavirus Reporter complaint against Apple; I prefer to focus on structural and systemic issues, but regardless of how structural or not a problem is, competitive constraints can only help.

The European Union's envisioned Digital Markets Act could become the most important piece of legislation in the technology space ever, way above such laws as the U.S. Digital Millennium Copyright Act (the substance of which I don't mean to criticize; I vocally supported its enforcement in a case involving Blizzard Entertainment).

But the question is: is Mr. Breton actually saying in that LinkedIn article that there will be alternative app stores on iOS (and Android)?

Here's the only passage in his statement that mentions apps:

"Gatekeepers will keep digital opportunities; providers of operating systems will always be able to offer all sorts of software and apps as they wish. In addition, the DMA empowers the users who do not like the preinstalled apps to switch to a different service or use a different app offered by another provider." (emphasis in original)

The narrowest interpretation would be that users must be provided with alternatives to any preinstalled apps, either by selecting different services within an app (such as by selecting a different search engine in a search app) or installing "a different app" made by another developer. In that case, one would interpret "provider" as "service provider" in the same sense that users could switch to a different service within an app.

But one doesn't even have to interpret "by another provider" as "by a different app store" in order to arrive at the CAF's desired outcome. The Apple App Store is an app itself (as is the Google Play Store). And it's a preinstalled one. So, arguably, Apple would have to offer an alternative by another service provider (such as the Epic Games Store) to the App Store. At a minimum, the CAF's interpretation is defensible, even though I'm not going to take a definitive position on whether it's the only proper interpretation (absent additional evidence).

What Mr. Breton primarily sought to accomplish with his LinkedIn article is to debunk the "myth" that Apple couldn't offer, say, a music streaming service. Instead, the DMA would impose obligations requiring "that business users and end users are not unfairly deprived of their free choice, a fundamental postulate of [the EU's] single market." An alternative app store would be as consistent with that vision as it gets. Many roads lead to a multi-app-store ecosystem, and the DMA is one of them, at least potentially.

Share with other professionals via LinkedIn:

Friday, March 19, 2021

Google left in the lurch: German government accepts Munich antitrust injunction over unfair treatment of health information portals as final decision

After the two spectacular preliminary injunctions that the Landgericht München I (Munich I Regional Court) entered on February 10 over the same issue (fair treatment of health information portals), both Google and the German federal government (specifically, the Federal Ministry of Health) had one month to appeal. What I've found out now from the regional appeals court, the Oberlandesgericht München (Munich Higher Regional Court), serves to validate the five-star rating two pre-eminent German antitrust experts gave the Munich court's decision:

While an appeal was filed in the Google case (appellate case no. 29 U 1317/21), the German government did not do so in its own case. A few days ago, Germany's leading news agency DPA reported that the Federal Ministry of Health had not responded to multiple inquiries concerning its plans. By now, about a week over the deadline, it's certain that Google has to go it alone.

I've also been able to find out in Germany that the government has formally accepted the preliminary injunction as final, meaning that the court order is now, in fact, a permanent injunction with respect to the government. There won't be a full-blown trial in the case against the government.

Google told the media a few days ago that it had merely filed a notice of appeal to buy itself more time to weigh its options. At the time of Google's notice of appeal, it wasn't clear yet what the government was going to do.

The monopolist now has to decide on whether to pursue this appeal, in which case it needs to file an opening brief soon. Otherwise its appeal will be dismissed automatically. Given that the Federal Ministry of Health is barred from this type of partnership with Google for the foreseeable future, and knowing that the government's acceptance of the preliminary injunction as the final ruling in the case lends additional credence to the lower court's ruling, Google might just abandon its appeal now. Of course, Google has the right to seek a review of the decision as a matter of principle, if only to demonstrate that it vigorously defends its conduct even though there's no more chance of an immediate commercial benefit.

For Hubert Burda Media, the owner of the NetDoktor.de health information portal, it's "mission accomplished." In my previous post on this dispute I already gave credit to the Hausfeld lawyers who made it happen. Of course, the government's decision to fold is in no small part attributable to the quality of the Munich I Regional Court's well-reasoned ruling (37th Civil Chamber; Presiding Judge: Dr. Gesa Lutz). My guess is that the government gave up because affirmance of the decision appeared rather likely and would have reflected unfavorably on Federal Minister of Health Jens Spahn.

Share with other professionals via LinkedIn:

Wednesday, March 3, 2021

Arizona House of Representatives adopts law untying in-app payment method from mobile app store monopolies: now on to the State Senate

Here's a follow-up to my very recent commentary on HB2005, a legislative proposal preventing Apple and Google from requiring developers to use only one payment system per mobile app store. Republican state lawmakers Dr. Regina Cobb and Leo Biasiucci sponsored the bill.

Today, the Arizona House of Representatives--one of the two chambers of the state legislature--PASSED the bill!

This screenshot is from the status webpage (click on the image to enlarge; the "PASSED" information may not be visible otherwise):

The result of the third reading vote was 31-29. There are 31 Republicans and 29 Democrats in the Arizona State House, and one member per party crossed the aisle, thereby canceling each other out.

A couple of proposed amendments failed, while a proposal by Dr. Cobb (enabling app developers to complain to Arizona's Attorney General about any failure by Apple or Google to comply) was adopted. (Technically, the App Store part of HB2005 was an amendment to a multi-purpose bill, which amendment then in turn got amended in the way just described.)

The Coalition for App Fairness is pleased, but notes that this is merely a first step toward a level playing field for all:

In order for this measure to be passed into law, the Arizona Senate would have to adopt it as well, and the Governor would have to sign it (as opposed to vetoing it). The (counter)lobbying onslaught by Apple and Google has been massive already, and may further intensify. There are 16 Republican and 14 Democratic senators. It is counterintuitive that Arizona Democrats have such strong reservations concerning this measure, considering that the Democratic majority in the United States House of Representatives took a clear position on tech monopolies and walled gardens in October.

This remains interesting, and meanwhile there are initiatives in various other states. Today, the Minnesota Reformer website published an opinion piece by Justin Stofferahn and Pat Garofalo, calling on the Minnesota state legislature to "curb anti-competitive tactics" in order to become, once again, "an innovation center."

And in precisely two months from today, the Epic Games v. Apple antitrust trial will start in Oakland, California.

Share with other professionals via LinkedIn:

Sunday, February 28, 2021

Could a single state legislature topple both mobile app store monopolies? At least it could make a historical contribution.

This is just my first post on legislative initiatives in multiple states concerning mobile app stores, so I've really just begun to research the topic and have a lot to learn.

A couple of weeks ago, the North Dakota state senate voted against a bill that would have required Apple's App Store and the Google Play Store to allow app developers to use other payment in-app payment systems. The fact that the state legislature decided against it doesn't represent a ringing endorsement of the status quo of mobile app stores. It's possible that many of the lawmakers who voted against the proposal simply didn't want their state, with not even a million inhabitants, to take such a fundamental decision against two of the country's largest and most powerful companies.

There definitely is broadbased political support for the fight against app store monopolies: last fall, the Democratic majority of the United States House of Representatives adopted a report on digital markets that condemns the current situation in pretty strong terms.

All three branches of U.S. government are dealing with the issue in different ways: the judiciary has various antitrust lawsuits against Apple and Google before it (with a case management conference in Epic Games v. Apple taking place tomorrow, Monday); the executive government (the DOJ's Antitrust Division and various state attorneys-general) may bring antitrust cases as well; and as far as lawmakers are concerned, there's the aforementioned House report (which is non-legislative, though it does recommend that measures be taken) as well as activities in multiple states.

As a developer who's experienced (and continues to experience) how tyrannical and harmful those app store monopolies are, I welcome any initiative that has the potential bring about change, or at least to raise the level of awareness. The situation is unsustainable. I've been in this industry for decades, and I remember the times when Microsoft was accused of expanding into other markets by abusing its control over Windows. What was alleged at the time was, seriously, negligible compared to the present setup. I remember computer book publishers who were unhappy about the creation of Microsoft Press; established office app makers like WordPerfect and Lotus complaining about Microsoft using secret Windows API calls, though I never saw any evidence for that (and, in fact, those companies initially didn't even want to make the transition from MS-DOS to Windows); and there were antitrust issues, but they affected only those competing with Microsoft at the operating system level, like Digital Research with DR-DOS, or at least at the network server level, like Samba, which got help from the European Commission. But the market was wide open then compared to what it is today. Microsoft didn't (as it couldn't) prevent anyone from publishing anything. Microsoft made itself comfortable at the top of the food chain, but a self-serving, arbitrary, or tyrannical gatekeeper it was not.

The fight against the abuse of app store monopolies is generally a marathon, not a sprint. In particular, antitrust enforcement and ligitation are time-consuming. But there can always be a sudden breakthrough somewhere that brings about change. To topple the app store monopolies thorough state legislation is rather ambitious, but my initial analysis is that the Coalition for App Fairness (CAF) has everything to gain and nothing to lose by playing that game. That's because even if a dozen state legislatures voted against such initiatives, the House report would still be far more persuasive--but if any single state enacted an app store law, app developers might find ways to benefit from it, such as by setting up shop in that state, and Congress would have a pressing reason to prioritize this subject at the federal level.

That said, it's just hyperbole that Apple claimed a measure like the North Dakota bill would be the end of the App Store as we know it. Shopping apps like Amazon or parking apps (I just used one a couple of weeks ago) are also allowed to use their own payment systems. Many users may still prefer to create just one account with Apple and to use it across all apps, but the market should decide. Apple's position is that because it made iOS, it shouldn't have to face competition in app distribution.

My own #1 (and #2, #3, #4, ...) issue with Apple and Google is about their rules relating to apps that mention COVID. I see the point, however, in some organizations' criticism of the restrictions those platforms impose on in-app payment systems. We can't solve all issues at the same time, and maybe the payment context is the one in which the cost to consumers is clearest (after all, the Supreme Court allowed a consumer class action against Apple to go forward). But it would be a misconception to believe it's just about "the 30%." In a twittersation on Thursday, the founder and CEO of Epic Games clarifies what this is fundamentally about. In response, I listed a handful of related issues:

For state legislatures, the in-app payment part is particularly intriguing because lawmakers can directly benefit consumers in their states and, potentially, attract app development companies. State-level initiatives have been reported from various states, such as Minnesota (StarTribune article) and Arizona (KAWC News).

The Coalition for App Fairness reported that the Arizona House Rules Committee "voted unanimously that [a bill including a part on app stores] is constitutional and in proper form." As a result of Apple's and Google's counterlobbying efforts (which show that they take those initiatives seriously), the question came ujp whether the proposal is allowed under the Commerce Clause, which gives U.S. Congress the power to legislate commerce. However, my research shows that the Commerce Clause doesn't prevent states from regulating commerce in their states as long as what they do doesn't run counter to federal legislation.

Of the three bills I've seen so far (Arizona, Minnesota, and North Dakota), my personal favorite is the language of the Grand Canyon State's version of the bill, which would make it illegal for Apple and Google to "require a developer that is domiciled in this state to use a particular in-application payment system as the exclusive mode of accepting payments from a user to download a software application or purchase a digital or physical product or service through a software application" (emphasis added). As Apple and Google are based in California, not Arizona, the argument is apparently made by the bill's opponents that this is interstate commerce and Arizona is just trying to favor its own companies, but again, the vote on constitutionality was unanimous and in favor of this proposal.

Assuming for the sake of the argument that Arizona (though it could also be any other state) passed such a statute into law, what would happen? Apple and Google could theoretically try to stop providing their app stores in that state, just so they would stop to meet the threshold (number of downloads, or revenue level) set forth in the bill. But after selling numerous iPhones in a given state, Apple could hardly stop serving those customers. The same applies to Google after its OEMs have sold tons of Android devcies somewhere.

Could Apple (or Google) stop doing business with app developers based in that state, especially if companies from other states set up offices in Arizona to benefit from the law? If they did so, they might be required under federal antitrust law or state UCL (unfair competition law) to make their essential facilities available to developers.

The first state legislature to enact such an app store state law could make history, and would benefit consumers and developers alike. I keep my fingers crossed, but there are so many other things going on that I'm sure it's not a question of if, but when, the app store situation will improve. When I campaigned against the EU software patent directive in 2004 and 2005, I thought it was the most fundamental threat to developers ever. It's how I became a campaigner for the first time in my life, and a little over a year after joining the fray, I received an award that went to Governor Schwarzenegger two years later, and I received more votes than fellow nominee Bono, which shows there were a number of people who thought I had made an impact on a major issue. But this app store cause is more important. I'm not going to be a full-time campaigner again, but I am determined to make my little contribution.

Share with other professionals via LinkedIn:

Tuesday, February 23, 2021

Top-notch antitrust experts award five stars to preliminary injunction against Google by Munich I Regional Court

The Munich I Regional Court (Landgericht München I) is my subjective "Technology Court of the Month" (no matter how often I disagreed with it on other issues, which may happen again anytime) as it handed down two key decisions in February that have the potential to make tech law history with repercussions throughout and potentially beyond the European Union:

This post focuses on the Google antitrust case.

I asked the press office of the Oberlandesgericht München (Munich Higher Regional Court) yesterday whether Google's appeal already had a case number and, maybe, a hearing date. They said an appeal had not been filed yet. But I'm sure that's going to happen shortly. If Google's PI appeal failed, it would theoretically still be possible for the full-blown main proceeding to have a different outcome, but it would be unlikely in this case, where the facts are pretty straightforward and the legal questions are crystal clear. However, in the main proceeding Google would be able to seek the equivalent of cert, while the regional appeals court is the final destination on the PI track.

I'm admittedly biased here because my own app development company is going to sue Google and maybe also Apple in Munich with direct references to certain holdings in NetDoktor v. Google (which I guarantee will not change a thing about how I view Apple's and Google's patent cases--in fact, I continue to keep my fingers crossed for Apple and Intel's joint antitrust action against Softbank-owned Fortress Investment). Given my own interest in antitrust action over gatekeepers using the critical nature of public health as a pretext for treating web publishers (in the NetDoktor case) and app developers (both Google and Apple) unfairly, it's all the more reassuring to see that totally impartial experts, too, heap praise on the NetDoktor PI decision:

Professors Justus Haucap (an economist the former chairman of Germany's Monopoly Commission, which advises the federal government on antitrust matters) and Rupprecht Podzsun (a legal scholar who is by now the undisputed number one among German competition law professors with respect to technology markets) awarded five stars to the Munich decision in the latest edition of their competition law podcast.

Both professors teach at the University of Dusseldorf. I met Professor Podszun at his previous university (Bayreuth) in 2014. That day I also had the honor to speak at his then-university alongside Judge Dr. Klaus Grabinski of the Federal Court of Justice about certain patent litigation topics.

In that podcast on the Munich NetDoktor v. Google decision, the professors agreed that the German federal government effectively became an economic operator by running a health information portal at taxpayers' expense, competing with private-sector offerings on the very opposite of a level playing field.

They note toward the end that even the half-hour duration of that podcast wasn't sufficient to highlight all that's remarkable about the well-reasoned ruling. But they managed to touch on various aspects, including the one I care about most: that the critical nature of health information doesn't serve to justify forcing or keeping high-quality offerings out of a market. The vertical cooperation challenged in NetDoktor was formed and launched last year, but even the COVID-19 pandemic must not be used for a pretext. The rule of law does not get infected by any virus.

Legal industry news service Juve (whose English-language spinoff Juve Patent I sometimes link to) reported on who represented the parties. The winning firm is Hausfeld, a firm with offices in the United States and Europe. In the U.S., Hausfeld focuses primarily on damages claims, and is suing Google on an indie app developer's behalf in a parallel case to Epic Games v. Google (Northern District of California). According to Juve, NetDoktor's lead counsel is Dr. Ann-Christin Richter. It was apparently the first time (but I guess not the last) for Hubert Burda Media, the media conglomerate that owns NetDoktor, to retain Hausfeld, and Hausfeld partner Professor Thomas Hoeppner ("Höppner" in German) has for years been counsel of record to a couple of publishers' associations of which the Burda group is a key member. Notably, Professor Hoeppner represented complainants against Google in the DG COMP investigation that resulted in the Google Shopping decision.

When the Supreme Court of the United States allowed the Pepper v. Apple class action over elevated App Store payment commissions to go forward, Hausfeld took a pro-developer position: Taking a Bite at the Apple: Ensuring a Level Playing Field for Competition on App Stores (which is not only about Apple's Apple Store, but also addresses the Google Play Store).

Now that I've digressed into App Store Antitrust again, I'd just like to highlight--in closing--that this has been my most popular tweet in quite a while:

Share with other professionals via LinkedIn:

Monday, February 22, 2021

UK competition court doesn't doubt the merits of Epic Games' antitrust injunction claims against Apple, Google--just forum non conveniens for Apple's U.S. corporate parent

As MLex's Lewis Crofts mentioned on Twitter, Fortnite and Unreal Engine maker Epic Games

At this stage, the UK court had to make a purely procedural decision: whether or not Epic's complaints should be served on non-UK-based entities Epic wanted to sue in the UK. The court has no problem with service on UK-based Apple and Google entities, and even some Ireland-based (not UK, but EU) Google entities as far as Epic is seeking injunctions against them. With respect to those Irish entities, what helped Epic is timing: it filed before Brexit took effect.

The ideal outcome for Apple and Google would have been if the court had held that there was no "serious issue to be tried." That would have been comparable to an outright dismissal of a case not well pled. No such deficiency was identified here with respect to the injunctions Epic is seeking--just with respect to mere declarations of breach. The court furthermore evaluated whether Epic had "gateways" (reasons for which to bring cases against non-UK entities in the UK). But what ultimately did result in the dismissal of U.S.-based Apple Inc. from the case is simply that the British court determined the Northern District of California was the forum conveniens, and that, at a minimum, London wasn't a better choice.

The court also held that some of its claims didn't entitle Epic to sue certain non-UK Google entities in the UK, but the remaining claims ("claims for breach of the Chapter I and Chapter II prohibitions under the CA 1998 as regards the alleged 'Restrictive Terms' in the DDA and the removal of Fortnite from the Google Play Store") still give Epic a potential path to victory there. The court determined that "there is no issue to be tried as regards the claims for the two declarations in the Apple and Google actions" (also noting that "it may be unfortunate that declaratory relief is not included as a remedy available in the [Competition Appeal] Tribunal").

In the Unwired Planet patent case, the UK Supreme Court actually took a very permissive approach to forum conveniens, holding that even if a smartphone maker generated only 1% of its worldwide sales in the UK, a UK court might nevertheless set a worldwide royalty rate for a standard-essential patent portfolio, and if the defendant didn't agree to a license deal on those terms, it would face a UK-wide sales ban. Here, however, a UK court exercised restraint in jurisdictional terms.

Even if Epic had received the go-ahead today to sue Apple Inc. in London, a decision in the UK would still have taken longer than the California case, in which the trial is only a few months away. The discovery cutoff date in the U.S. case was one week ago.

Today's UK decision doesn't help Google in any way, nor does it solve Apple's real problem, which is that its App Store monopoly is under pressure in multiple jurisdictions.

Share with other professionals via LinkedIn:

Wednesday, February 17, 2021

Epic Games files antitrust complaint against Apple with the European Commission's Directorate-General for Competition (DG COMP): media reports

Frankfurter Allgemeine Zeitung (FAZ) and the Financial Times report that Fortnite and Unreal Engine maker Epic Games has filed an EU antitrust complaint against Apple. Epic is being represented by Clifford Chance (the FAZ article quotes Mr. Ashwin van Rooijen), a major global firm that has been a key player in many EU tech antitrust matters over the years.

When asked why they filed this complaint against only Apple, and not simultaneously against Google, Epic's lawyers apparently said that Apple was presently the focus of EU antitrust enforcement in this context. There is an ongoing investigation of Spotify's complaint against Apple. Both Epic and Spotify are members of the Coalition for App Fairness (CAF), an organization whose positions on app distribution Microsoft has supported publicly without joining the organization.

For proper disclosure, I am a member of the informal #AppRising movement and brought my own complaints against Apple and Google in multiple jurisdictions last month (EU case numbers: AT.40747 Apple, AT.40748 Google). I consider this kind of topic to be the single most important tech antitrust issue of the 2020s. Standard-essential patent (SEP) issues continue to be very important, but are eclipsed by the #AppRising. But the concerns raised by app developers are diverse, and just like I haven't seen Epic comment on my complaint regarding COVID-related apps, I don't want to take a position on Epic's (and others') complaints over in-app payments--which doesn't mean that I necessarily disagree, just that at this point it's too early for me to speak out on a legally complex issue.

As I just mentioned my own complaints, a few hours ago iClarified mentioned me in connection with a new rule--or maybe it's more of a clarification of an existing rule--by Apple in connection with "health pass" apps (apps that enable users to show to others, such as airport security staff, that they have been vaccinated or have recently been tested negatively). So here's a bit of a clarification from my end in the context of iClarified's report on Apple's potential clarification regarding health apps: I don't envision my own app company to offer any "health pass"-like functionality and, therefore, prefer not to comment on that category of COVID-related apps other than noting that governments--rather than private entities--should regulate health passes, such as by ensuring that only health passes with a certain stamp of approval would be recognized at airports, sports venues, and so forth.

I don't mean to comment on Epic's choice of bringing a complaint against only Apple, not Google. I'm sure they had their reasons, and it's possible they previously talked to DG COMP and made their decision based on the feedback they got.

My company, however, seeks to resolve its issue with both major platforms at the same pace. Last week's decision by the Munich I Regional Court against Google, of which I've meanwhile translated the remaining 10% I just summarized a week ago, makes it clear that "gatekeepers" (the ruling already uses that key term of the upcoming Digital Markets Act) must not disadvantage high-quality health-related offerings just because they don't come from a governmental source. My company is going to bring a lawsuit in Munich on that basis pretty soon. The Munich court's holding is very clear, and Google's rule on COVID-related Android apps is fundamentally more abusive--because it results in complete foreclosure--than its preferential treatment of a government-run health information portal by its search engine. If Google doesn't take the Munich court's holdings seriously, it will just have to be enjoined again. After filing the complaint, I will publish an English translation on this blog.

Share with other professionals via LinkedIn:

Wednesday, February 10, 2021

BREAKING: Munich I Regional Court orders preliminary injunction against Google over antitrust violation in health information context

The Landgericht München I (Munich I Regional Court) has just handed down a landmark antitrust ruling against Google: the court's 37th Civil Chamber (Presiding Judge: Dr. Gesa Lutz) prohibited a partnership between Google and the German Federal Ministry of Health under which Google gives a health information portal (gesund.bund.de) prominent exposure among Google's search results, thereby disadvantaging private-sector health information providers such as the plaintiff in this case, NetDoktor.de, a subsidiary of the Hubert Burda Media conglomerate.

The unholy alliance in question was formed last year in connection with the COVID-19 pandemic.

Google can and presumably will appeal this preliminary injunction to the Oberlandesgericht München (Munich Higher Regional Court).

I have just obtained a copy of the written ruling and translate it here later (you'll find my translation further below).

In a recent post on Google's inexplicable approval of the "official" app of a coronavirus lockdown offender group I already mentioned the NetDoktor case. My own app development company recently brought antitrust complaints in half a dozen jurisdictions against Google's (and Apple's) rules concerning COVID-related apps. The Android situation is even worse than the one on the search engine: on Android, app developers like my company are completely barred from using COVID-related keywords, as opposed to merely giving preferential treatment to governmental entities. Yesterday Google updated its COVID app rules, but the changes Google made do nothing to alleviate my competition concerns, and today's Munich decision shows that monopolists can't just disadvantage (much less exclude from the market, as is the case on Android) private-sector offerings.

The court's press release explains that the operation of a health information portal by the Federal Ministry of Health is not merely a governmental activity, but has an economic dimension making it subject to competition law. The agreement between Google and the German government is restrictive of competition in the market for health information portals. The so-called knowledge panel (info box) that appears whenever German Google users search for any of various health-related keywords is unavailable to private-sector entities: the ministry secured this spot on an exclusive basis.

The court recognized that the plaintiff/movant in this case, NetDoktor.de, is economically dependent on its presence among Google's search results, as 90% of the people visiting that website arrive there via Google. The prominent knowledge panels divert attention away from the organic search results and satisfy many users' demand for information before private-sector competitors have any chance to offer their content.

The court rejected Google's and the German government's arguments that this anticompetitive form of collaboration was justified by "qualitative efficiency gains" (such as by reducing the amount of time it takes users to search for health information) or by the aim to improve citizens' knowledge about health. To the extent that there might be any such effects, they would be clearly outweighed by the anticompetitive implications of this cooperation.

With a view to the Android app situation, that finding against a bottom-line consumer benefit (rule of reason) is very significant. Google's official--but totally hypocritical, as the company promotes and sells material promoting what has been proven over and over to be fake science and bogus medicine--rationale for its restrictions on COVID-related Android apps is also that Google claims it doesn't want users to obtain unreliable information.

The court's press release notes that reliable private-sector health information portals should not be driven out of the market, and the diversity of media and available opinions should not be reduced. That, too, applies to COVID-related Android apps in my view.

The plaintiff/movant furthermore made a credible case for being harmed by the illegal cooperation in a way that warrants emergency relief. The court held that NetDoktor.de would not have to seek recovery of any lost advertising revenues (particularly in connection with certain illnesses) at a later stage.

NetDoktor.de also brought a claim against abusive unilateral conduct on Google's part. That part was rejected "on formal grounds." The judgment will likely shed more light on this. As far as I can see, all of the findings underlying the prohibition of the anticompetitive collaboration between Google and the German government would equally apply to an Art. 102 TFEU case (restrictive effects, competitive harm, no rule of reason; and as everyone can see, Google can't just excuse this kind of anticompetitive behavior with the special circumstances of the SARS-CoV-2 pandemic). Also, it's telling that Google has been ordered to bear 75% of the litigation expenses: NetDoktor.de overwhelmingly prevailed.

I agree with many observers who say that Google was basically just interested in currying favor with the German government, which is a key objective for Google given that it is under competition enforcement pressure in Europe.

This is a great and courageous decision, and I hope the appeals court will affirm it. I plan to attend the appellate hearing, which I guess will take place in a couple of months.

UNOFFICIAL TRANSLATION OF THE KEY PASSAGES OF THE DECISION (with my explanations in [brackets]):

In NetDoktor.de v. Google [technically, the case against the German government is a separate one, though the issues underlying this decision are the same], the Munich I Regional Court's 37th Civil Chamber--Presiding Regional Court Judge Dr. Lutz, Regional Court Judge Dr. von Merveldt, LL.M. (NYU), and Regional Court Judge Dr. Pfeiffer--hands down, on the basis of the oral hearing held on January 20, 2021 ,the following

FINAL JUDGMENT [note the difference between U.S. and German cases: in Germany, a PI motion is a separate case, and this is the final judgment with respect to the PI motion]

  1. Defendant is enjoined by way of a preliminary injunction and under the threat of a contempt fine of up to 250,000 euros or confinement of one of its general managers for contempt of court for up to six months--or a confinement in the event a contempt fine cannot be collected--

    from displaying the content of the gesund.bund.de portal in form of a Knowledge Panel showing health information if this results from previous coordination securing such display exclusively for the content of a website operated by the Federal Ministry of Justice (gesund.bund.de) and involves a link as shown, by way of example, in the following screenshots made on November 24, 2020:

    Desktop computers for end users:

    Mobile devices for end users:

    The remaining claims are rejected.

  2. Defendant bears three quarters of the litigation expenses; Plaintiff/Movant bears one quarter.

  3. This judgment is provisionally enforceable by Defendant with respect to the second item [litigation expenses]. Plaintiff/Movant can avert enforcement by Defendant [again, this relates to litigation costs] by providing collateral to the amount of 100% of the amount enforceable under this judgment, unless Defendant provides, prior to enforcement, security to the amount of 100% of the amount to be collected.

Factual Background

This dispute between the parties is about the legality of a cooperation between Defendant [Google] and the [German] Federal Government, pursuant to which content of the "National Health Portal" operated by the Federal Ministry of Health is afforded prominent exposure in special information boxes next to or, respectively, above general search results in connection with Google searches for certain illnesses.

Plaintiff/Movant, a subsidiary of [Hubert Burda Media], has been operating for more than 20 years an advertising-financed online portal for health information at the www.netdoktor.de domain, which in accordance with scientific standards, yet in a manner that is simultaneously intelligible to lay persons, provides information on illnesses, symptoms, medications, therapies, and laboratory values. It is Plaintiff/Movant's stated goal to enable its users to be an informed counterpart of their doctor or health insurance company. With a daily volume of approximately one million pageviews, Plaintiff-Movant is presently the market leader in the German or German-speaking market for health information portals ahead of the second-placed portal [named redacted out] and further competitors. Approximately 76% of all accesses to Plaintiff/Movant's portal are made via mobile end-user devices, the remainder via desktop computers for end users.

Ireland-based Defendant is a Google subsidiary and offers the Google search engine in Europe, including the offering targeting the German market ("Google Search"). That centralized service can be accessed, inter alia, under the domain www.google.de. In order to respond to search queries, the Google Search deduces the user-specific need for information from the given search terms and displays the information it considers to be most helpful and relevant from its vantage point. Such so-called generic search results, also called organic or natural search results, are traditionally presented as title equipped with an underlying link, a web address, and a short excerpt from the website's content containing the search terms ("Snippet"). The linked webpages are shown as a list ("Search [Engine] Results List") and sorted by their rank ("Ranking").

Plaintiff/Movant enjoys a high level of visibility in the general Google Search and in an overwhelming number of cases involving the relevant search terms ("Keywords") from the field of health inforation appears on the first search engine results page, frequently in the first position, which is attributable to, inter alia, considerable investment in the quality of its content and in its search engine optimization efforts. On average, 88% of the total number of visits by end users ("Traffic") to Plaintiff/Movant's health portal was generated by organic Google search results over the past two years. For visits from mobile end user devices, that share averaged 90%.

If a search query relates to short, specific, and factual information, Google Search does not merely provide a Search Results List. Instead, if a query suggests that a purely factual answer can be provided, it immediately also presents key information designed to provide as direct as possible an answer to the user's query. Such search results are highlighted and displayed in addition to the Search Results List. Such information is either displayed directly, in which case Google Search avails itself of information provided by website operators (such as the programs listed by movie theater operators) for such purpose. Alternatively, if users are not searching for individual facts, but for a group of persons, places, or things (such as the cast of a particular television show), such information is collected in the form of so-called Rich Lists. Finally, in caes where it is suitable, info boxes, so-called Knowledge Panels, are displayed and provide relevant information (such as on the height of the Eiffel Tower). The source of the content of those info boxes is typically the Knowledge Graph, an internal database used by Google Search tht contains facts relating to a wide range of topics. Other search engines offer similar kinds of search results that are separated from organic search results.

In Germany, the Google corporate group has for more than ten years continually had a share of more than 90% in the market for Internet search services.

The Federal Ministry of Health ("FMH") has been operating since September 1, 2020, a National Health Portal (NHP) at https://gesund.bund.de, which--like Plaintiff/Movant--provides health information. The stated goal of the federal government's health portal is to offer quality-assured, independent, and generally understandable health information on the Internet. Besides a directory of illness explained and illustrated in corresponding articles, the NHP contains sections such as "Healthy Living", "Care", and "Digital Health." In the "Healthy Living" section, one can find, for example, articles providing explanations and recommendations relating to "Nutrition and Exercise", "Health at the Workplace", "Growing Up Healthy", "Health for the Elderly", "Soul and Well-being", "Pregnancy and Birth", "Overcome an Addiction", and "Prophylaxis and Early Detection." The articles are illustrated with charts and photos, and some of them contain video recordings and audio files. The NHP intentionally steers clear of references to controversies relating to topcis such as vaccination and mammography or information on fields such as homeopathy and alternative medicine. [I don't doubt that this is correct with respect to the German government's health portal, but Google's hypocrisy and irresponsibility involves the distribution of homeopathy-related apps and ebooks while pretending to be concerned about users being misinformed on medical topics...]

On November 10, 2020, Defendant and the FMH announced in a joint press conference (cf. USB stick provided as an exhibit by Plaintiff/Movant) the launch of their cooperation concerning search queries on health topics. That cooperation, they said, should enable easier access to reliable information on health topics on the Internet (cf. FMH press release of November 10, 2020, Exhibit Set K29). For that purpose, the Google search engine presents a prominently exposed info box, or Knowledge Panel, with the NHP's answers when the search terms involve illness-related terminology. Via a link inside the info boxes, users can directly access the related article on the gesund.bund.de website. In response to a desktop search, the boxes appear to the right of the list containing advertisements and organic search results. On mobile devices, the info boxes appear immediately after advertisements, but ahead of general search results. Each info box has three tabs for navigation purposes: Overview, Symptoms, and Therapies. the mobile version provides links to pertinent media reports under a fourth tab ("Headlines").

To date, such info boxes are available for approximately 160 illnesses. The related text material is content provided by the FMH especially for the immediate display by search engines and comes with a markup. The info boxes retrieve this information via an open interface. That interface is, in principle, also available to other search engines.

The Google corporate group also offers info boxes on health topics in other countries, including, inter alia, the U.S. and the United Kingdom. Here, too, content is obtained from third-party sources, which are not necessarily governmental entities.

At the joint press conference [with Google], Federal Minister of Health Jens Spahn said, inter alia, the following:

"Whoever googles health information should land on our federal government's portal and find the information he needs. Gesund.bund.de should become the central go-to site for reliable health information on the Internet. Nothing is a more obvious choice than to cooperate directly with Germany's most popular search engine, i.e., Google. On Google, the {NHP's] answers to a search query involving medical keywords will be presented in a prominently exposed info box. [...] I am certain that this collaboraiton with Google gives an enormous awareness boost to the [NHP], and this coooperation will ensure that this portal can become one of the most important go-to sites for citizens looking for online health information."

He furthermore declared:

"As we are interested in conveying objective, well-founded, evidence-based information, it doesn't do me any good to be listed by Google in the 783,000th place."

The total traffic volume generated by Google Search for Plaintiff/Movant's website remained at largely the same level after the launch of the info boxes and amounts to approximately 4 million clicks per week. The traffic trend is, therefore, consistent with the same period of the year 2019 and the one in Austria, where Defendant's Google Search does not display info boxes on health topics. During the period from November 10 to December 9, 2020, gesund.bund.de received 283,256 clicks via the Google Search, while Plaintiff/Movant received 17.674 million clicks, i.e., approximately six times the volume. The number of clicks on these links accounted for 0.06% of all clicks of health portals listed on search results pages.

However, with respct to particular illnesses, both the number of accesses and the so-called click rate dropped in the second half of November. The click rate, or click-through rate, is defined as the number of pageviews per impression, i.e., per display of Plaintiff/Movant's page among the organic search results. While user traffic via Google queries generally increased during the year 2020, Plaintiff/Movant lost 5% of its desktop and 10% of its mobile traffic in the period from November 9 to 16, 2020, in connection with 19 particularly high-traffic illnesses. For select so-called Basic Search Terms, which Plaintiff/Movant describes as terms generating a high search volume and which are key to [Plaintiff/Movant's] monetization efforts (arthrosis, cystitis, bronchitis, depression, epilepsy, erysipelas, gallstones, gastritis, prostate cancer, hypothyroidism, typhoid fever, [another term meaning] erysipelas, and neurodermatitis), the click rate dropped by approximately 32.5% between November 5 and 14, 2020--notwithstanding otherwise stable visibility paraemters (i.e., continued appearance among the first organic seaerch results) and a generally positive trend. Plaintiff/Movant conducted a further analysis showing that pagreviews on its website relating to "white skin cancer" dropped by 10.9% (13.9% on mobile devices), and by 17.3% (mobile access: 19.2%) for "cystitis." Even though the average ranking of those pages remained constant oder even improved slightly, the click rate dropped by approximately 23% between November 2 and 16, 2020. A subsequent analysis performed by Plaintiff/Movant showed a decline of the average click rate relating to four commercially particularly relevant illnesses: 37.1% (cystitis), 31.1% (irritable bowel), 15.6% (hemorrhoids), and 31.8% (lactose intolerance). With respect to further studies, reference is made to Plaintiff/Movant's pleadings.

Plaintiff/Movant attributes the above-mentioned--and factually undisputed--developments to the prominently exposed info boxes containing the NHP's content, and brings antitrust and competition claims against Defendant.

Plaintiff/Movant initially bases its prayer for injunctive relief (item 1) on § 33.1 of the GWB [Act Against Restraints on Competition, "Competition Act"] in conjunction with § 1 Competition Act and Art. 101 para. 1 TFEU.

PLaintiff/Movant argues in this regard that the cooperation between Defendant and the FMH is based on a cooperation that is restrictive of competition as evidenced by Defendant's joint press conference with the Federal Minister of Health. According to Plaintiff/Movant, it expresses the joint intent by Defendant and the FMH to provide, in a coordinated fashion, the NHP's content on Google Search exclusively in a special advertising format--the info boxes--exclusively reserved for this purpose, aiming on the one hand to allegedly make Google Search more attractive for its users and on the other hand to disseminate the content of the NHP more broadly than it would be possible in the absence of this cooperation. In Plaintiff/Movant's opinion, this suffices to make a showing of an agreement. The FMH, Plaintiff/Movant argues, is acting as an enterprise as it starts to compete, through its NHP, with private-sector providers.

This cooperation, Plaintiff/Movant says, aims at and results in a significant restraint of competition. Plaintiff/Movant argues that, given the large extent to which all health information portals depend on Google Search's intermediary service, the vertical combination of search engine and info boxes is likely to reduce or even completely eliminate competition in this market. Such an effect is, according to Plaintiff/Movant, the object of theagreement, as the collaborators strive to make NHP the central go-to site for online health information to the detriment of all other providers.

Plaintiff/Movant bases its argument for the existence of a restraint of competition on the fact that Defendant draws users' attention to the prominently displayed info boxes. In this context, Plaintiff/Movant points to behavioral economics explained by the European Commission in its "Google Search (Shopping)" case. User demand for health-specific information is, according to Plaintiff/Movant, already satisfied within the search results pages. This is Defendant's intent, Plaintiff/Movant argues, as its repeated statements show that it considers this to constitute product innovation. Even if there still remained some further demand for information, Plaintiff/Movant considers it more likely that users click on the links in the info boxes than that they would search for further information among the organic search results. Plaintiff/Movant considers this effect to be particularly strong on mobile devices, as organic search results have been pushed far down due to the info boxes.

This redirection [of traffic] and the visibility loss is proven by a reduction in traffic and click rates, Plaintiff/Movant says. The only plausible reason from Plaintiff/Movant's perspective is that the visibility of the generic search result from NetDoktor.de is reduced by the NHP box. Plaintiff/Movant contradicts Defendant's position that the total traffic volume on Plaintiff/Movant's website had only changed nominally. Instead, Plaintiff/Movant attaches particular relevance to the sudden "discrepancy" between the number of impressions and the number of clicks, as reflected by the click rate.

Plaintiff/Movant furthermore claims to suffer in qualitative terms. Due to the negative trend, PLaintiff/Movant says it is forced, due to previous guarantees given to advertisers regarding the size of the audience, to compensate for the declining number of pageviews, which in turn reduces the quality of what it can offer to advertisers. As a result, PLaintiff/Movant claims to be forced to display ads on less relevant subpages. This, Plaintiff/Movant argues, distances the advertisements from the topics [of the pages], resulting in [losses resulting from a less targeted placement of advertisements].

At the same time, Plaintiff/Movant argues, the federal government's NHP has enjoyed a sudden increase in visibility, which, for instance, on November 13, 2020, amounted to 200% of the previous day's figures. In the meantime, the NHP has allegedly made further headway among the organic search results. For example, Plaintiff/Movant says that the NHP has since January 2021 already been listed in the fourth position of the first search results page for the term "white skin cancer." This now makes the NHP a direct rival for the position that the long-established, high-quality health portal [name redacted out] has, says Plaintiff/Movant.

Plaintiff/Movant attributes this to secondary effects: as a result of users clicking on the links to the NHP found within the info boxes, relevance is signaled, which the self-learning Google algorithm converts into higher rankings. The long-term effect--Plaintiff/Movant argues--would be that the NHP would sooner or later also appear as the first organic and, therefore, seemingly most relevant search result. By arguing that the mere placement of the info boxes and a click on a link contained therein does not influence the Google algorithm, Defendant omits, in Plaintiff/Movant's opinion, that an increase in the number of pageviews, irrespectively of the source, does have a positive effect on the ranking--a fact that is not in dispute.

This effect--Plaintiff/Movant argues--is not due to [Google] considering Plaintiff/Movant's content to be less relevant, but the better position and related competitive advantage of the NHP is achieved in circumvention of the general relevance determination by [Google's] algorithms. Plaintiff/Movant describes it as extremely difficult to counter a loss of a relevance rank through effective measures, which is why it would, according to Plaintiff/Movant, at some point become impossible to restore competition by deactivating the info boxes. For advertising-financed health portals, Plaintiff/Movant argues, it is not financially viable to compensate for diverted search traffic through Google Ads text advertisements or [advertisements in] mobile apps. The loss of traffic from users would--Plaintiff/Movant asserts--lead to a downbound spiral of programmatic ads and keyword-based targeting by the advertising customers of independent health portals and, therefore, endanger their financial viability. The reductions in traffic from users already affects--Plaintiff/Movant avers--all advertising contracts concluded for the year 2021, as Plaintiff/Movant takes the negative trend into account when calculating the number of placements of advertisements promised to its advertisers.

Plaintiff/Movant claims that this provides Defendant itself with a competitive advantage. There is--in Plaintiff/Movant's representation--substitutive competition between advertising on search engines and other online advertising channels; with both parties' platforms being advertising-financed, Plaintiff/Movant considers Defendant a rival competing for the same advertisers and health information-related budgets. By integrating the info boxes [into its search results pages], Defendant has allegedly gained further attractiveness as an advertising channel by comparison to health information portals. Plaintiff/Movant says some of its major advertisers already told it that they could--or might even be forced to--reallocate some of their budgets to Google Ads text advertisements.

Plaintiff/Movant does not consider the agreement [the one between Google and the federal government] to be justifiable in accordance with § 2 Competition Act, Art. 101 para. 3 AEUV. It argues that Defendant had failed to sufficiently substantiate any potential efficiency gains. In Plaintiff/Movant's opinion, such efficiency gains do not exist, as the content of the info boxes is no less reliable than quickly and easily accessible health information that is available on the Internet at any rate. Furthermore, Plaintiff/Movant argues the restraints of competition are, at a minimum, unnecessary. In order to achieve the desired effect of providing only a first reliable overview of a particular illness, Plaintiff/Movant believes a less detailed presentation of information would have been sufficient and would have made it less likely that users' appetite for information would be satisfied in its entirety. Plaintiff/Movant also notes that links to (exclusively) the NHP would not be necessary to achieve this objective.

Plaintiff/Movant furthermore brings a claim for injunctive relief under § 8 paras. 1, 3, and 4.4 of Unfair Competition Law, arguing that Defendant unfairly restricts Plaintiff/Movant['s ability to compete] through the info boxes.

Plaintiff/Movant Plaintiff/Movant furthermore bases its second prayer for injunctive relief on § 33 para. 1 Competition Act in conjunction with § 19 Competition Act, Art. 102 TFEU. According to Plaintiff/Movant, Defendant abuses its dominant position in the search engine market to cause anticompetitive effects, particularly in the market for health information portals by discriminating against Defendant and other competing portals, and by self-preferencing and an illegal technical combination of search engine and info boxes as two separate products, without an objective justification.

Additionally, Plaintiff/Movant claims to be entitled to injunctive relief under § 8 paras. 1, 3a UCL in combination with § ,94 Media Service State Treaty [a German piece of Internet regulation in the form of an agreement between the country's federal states]. Also, Plaintiff/Movant is seeking an injunction under § 8 paras. 1, 3a UCL in conjunction with Art. 5 para. 1 of the German Basic Law [the country's de facto constitution] as Defendant allegedly aids and abets through its display of the info boxes the FMH's violation of the constitutional imperative that the press be at arm's length from the government.

[The ruling then quotes the original first prayer for injunctive relief, and notes that it was withdrawn in part during the course of the oral hearing on January 20, 2021. The modified first prayer for relief is materially consistent with the injunction that has meanwhile been actually granted, see the first part of this unofficial translation. But there was also a second prayer for injunctive relief, which I will translate below; that one was not granted.]

2. Defendant is enjoined from employing, for the positioning and display of content of the FMH's website (https://gesund.bund.de) or links thereto on the general sarch engine results pages, other processes and methods than those applied to the content of Plaintiff/Movant's website (www.netdoktor.de), especially if this occurs due to the display of content exclusively afforded to the FMH's website (https://gesund.bund.de) in the form of Knowledge Panels with health information and links, as shown by way of example in the screenshots of November 24, 2020, under item 1.

Defendant requests that the {entire] motion be denied.

Defendant consents to a partial withdrawal of the motion, which could be seen in the [modified] prayers for relief stated by Plaintiff/Movant in the oral hearing, but objects to any amendment to or extension of the motion that might lie in the new version of the prayers for relief.

Defendant criticizes the indeterminate nature of the prayers for relief, including the new version of the prayers for relief, and the unclear interdependencies between the first and second prayers for relief.

Defendant is of the opinion that the FMH, as operator of the gesund.bund.de health information portal, does not act as an enterprise, but fulfills its governmental task of protecting public health. According to Defendant, the FMH does not become an economic operator and does not pursue commercial interests. The fact that the governmental action does not generate revenues is, according to Defendant, a strong indication of the absence of any entrepreneurial context. The fact that private entities could publish the same information does not--Defendant argues--alter the character of governmental efforts to inform. Defendant says that ECJ case law requires clear indications if governmental actions were to be viewed, for an exception, in an entrepreneurial context.

Defendant furthermore claims that there is no contractual agreement between itself and the FMH, and particularly no obligation to cooperate exclusively or to make payments. Instead--Defendant claims--the cooperation is limited to Defendant carrying certain content from the "Illnesses" section of the gesund.bund.de portal. Any--disputed--harm to Plaintiff/Movant would, acording to Defendant, at most result from Defendant's prior, unilateral, fundamental decision to display info boxes for certain illnesses as part of the German Google Search, but not from a decision to select a particular source. Defendant argues that its choice of a source is irrelevant to the visibility of other search results. Therefore, Defendant denies the existence of a causal nexus between the alleged agreement and the alleged harm.

Defendant claims to have engaged in coordination with the FMH and the agency retained by it [to create the gesund.bund.de portal] solely for the purpose of implementing its choice of the gesund.bund.de portal [as the source of the content of its info boxes]. Defendant describes it as a technical necessity to preselect a particular source as the info boxes do not use the full text found at the source, making it a requirement that the source mark up the exceprts to be displayed in the info boxes, which enables the automatic identification and transmission of the text fragments relevant to Google Search by means of a query. Furthermore, Defendant says the preselection is necessary in order to ensure the consent of the operator of the chosen website. Also, Defendant says it had to ascertain whether the content of that portal met its linguistic and editorial criteria. However, any coordination was limited to those aspects, Defendant avers.

Defendant holds the view that the--disputed--agreement with the FMH neither has the object nor the effect of a restraint of competition. Defendant says the indispensable requirement of a binding agreement between Defendant and the FMH is not met. In any event, Defendant says the sole effect of its cooperation with the FMH is the immediate provision of reliable health information to the users of Google Search. According to Defendant, this is pro-competitive, as the Hamburg Regional Court held in its April 4, 2013 decision (case no. 408 HKO 36/13) when analyzing the provision of weather-related information on the basis of unilateral conduct. Therein, Defendant argues, lies neither an intended nor an actual restraint of competition. Furthermore, Defendant denies the presence of any abusive unilateral conduct.

Defendant disputes Plaintiff/Movant's allegations of negative effects on the order of search results.

With respect to Plaintiff/Movant's allegation of reduced traffic relating to 19 select search terms in the week from November 9 to 16, 2020, Defendant responds that in 18 of those 19 cases (the sole exception being "Alzheimer") users entered those search terms less frequently in that particular week (for example, cystitis or hypothyroidism). Defendant attributes the--minimal at any rate--decline in the numberr of clicks on Plaintiff/Movant's website in those particular cases to reduced interest as opposed to the display of info boxes. Therefore, Defendant denies that there is a correlation between the so-called click-through rate to Plaintiff/Movant's portal and the introduction of the info boxes.

Defendant also dismisses Plaintiff/Movant's claim that the display of info boxes afforded the FMH a lasting and irreversible competitive advantage. According to Defendant, this is already stablished by the low number of clicks on the links in the info boxes and the fact that Plaintiff/Movant still has a multiple of the number of users as the NHP. Defendant says any increased visibility of gesund.bund.de [presumably this refers to the NHP's ranking among organic Google search results] can be explained with public attention as a result of the FMH's press conferences.

Defendant argues that Plaintiff/Movant would have to allege a specific revenue loss. Defendant considers it insufficient to base a prayer for injunctive relief on insinuations and concerns. Even after the introduction of the info boxes, Defendant says, traffic on Plaintiff/Movant's portal as a result of clicks on search results in the search results list of Google Search did not decline. Since December 2020, Defendant considers this perspective to have been validated further, such as by way of the comparison with development in Austria, where so far no info boxes on illnesses are displayed.

Defendant denies that there are discernible negative effects of the info boxes on competition. In this context, Defendant alleges that there is no threat of Plaintiff/Movant's portal being ranked lower in the search results list. Defendant claims that the algorithm that determines the order of search results ignores clicks on the links to the gesund.bund.de site that are found inside the info boxes. Defendant denies that the display of an info boxes in response to search queries is a factor for the algorithm that determines the ranking of search results within the search results list.

Defendant claims that the info boxes presenting content of the gesund.bund.de portal are structurally and functionally disparate from health information portals such as the one operated by Plaintiff/Movant. Google's special info boxes on illness are not a substitute for the offerings of health information portals, Defendant says. They [the info boxes] represent, according to Defendant, an informational starting point with short, factual, basic information, according to Defendant. Defendant says the info boxes are meant to help users, with extremely diverse, far-reaching and more detailed information, in terms of a controlling and supporting function. Defendant says the purpose of the info boxes is not to provide exhaustive information, as shown by the prominent pointer to "further information." All that Defendant admits to import from the NHP is information on illnesses. Defendant describes such information as purely factual information as opposed to editorial content. According to Defendant, this sets the content of the info boxes apart from Plaintiff/Movant's portal.

The direct display of health information in info boxes, according to Defendant, has nothing to do with self-preferencing and, therefore, is fundamentally different from the European Commission's "Google Search (Shopping)" decision cited by Plaintiff/Movant (European Commission, June 27, 2017 decision, case AT.39740, summary in Exh. K6). According to Defendant, that case, unlike the present case, involved the allged preferencing, on the search results pages of the general Internet search, of another standalone service by Defendant. But the info boxes, do not constitute a standalone service offered by it, Defendant says.

Defendant also denies that the display of info boxes represents an additional source of (advertising) revenue for it that would result in the reallocation of advertising budgets, as Defendant says it generates Google Search revenues only through the display of advertisements. In that context, Defendant says, advertisers pay only if users click on advertisements. Defendant denies to compete directly or indirectly with Plaintiff/Movant for the same advertisers or advertising budgets. According to Defendant, the parties therefore operate in distinct online advertising markets.

Defendant argues that its agreement with the FMH, which it claims does not exist and is not restrictive of competition, would be exempt under Art. 101 para. 3 TFEU, § 2 para. 1 Competition Act. The display of info boxes represents an innovative product improvement to the benefit of users, as it saves time and is useful, Defendant claims. Defendant describes this as a new, innovative type of search results and, therefore, a qualitative efficiency gain in the form of a new feature directly beneficial to consumers. Defendant says consumers are directly, and in an organized and consistent form, shown reliable and authoritative basic information reated to their search queries on particular illnesses. Defendant claims to create added value through its search service.

Defendant claims that studies by the Bertelsmann Foundation and the AOK [a state-owned healthcare organization] (Exhibits AG 15 and AG 16) showed that there is a need on the part of German users for reliable information on the Internet. From this, Defendant says, follows that 57% of the users who know Plaintiff/Movant's portal do not consider it trustworthy. According to Defendant, no algorithm is suitable to the verification of the reliability of health information that goes into info boxes, which is why Defendant says its medical expert staff would firstly have to verify and ascertain the understandability of such material in order to meet Google's own standards, and therefore it needs to predetermine the source.

Defendant says it chose the FMH's portal as a source of reliable health information because all of its information on the topic of "Illnesses" is based on the findings of evidence-based medicine (EBM) and corresponds to current scientific standards. [I can't help but note that it is "rich" for Google to promote evidence-based medicine while distributing homeopathy crap through its Google Play Store, while disallowing COVID-related apps even if consistent with evidence-based medicine unless they come from a very few sources.] That information, Defendant says, is subsequently reviewed and validated by experts. Defendant argues the information is not affected by commercial or political interests. [It's just ridiculous that Google argues governments provide information without a political agenda.] According to its Gogole Surveys online polling service, Defendant claims 80% of users had stated that they place greater confidence in the information provided by the FMH on gesund.bund.de than information found on a portal such as NetDoktor.de. The editorially created text on Plaintiff/Movant's portal comes with spacious advertising banners, Defendant claims. The advertising also appear editorial in part, according to Defendant. Plaintiff/Movant's portal does not have to meet any standards and is not independent, and it could be sold to an acquirer anytime, Defendant argues.

Finally, filling the info boxes with content particularly from the FMH's source does not, according to Defendant, eliminate competition in the health information portal segment. As Defendant previously argued, it repeats its claim that there is not even the threat of a restraint of competition. According to Defendant, substantial traffic shifts have not occurred yet and are not expected to occur in the future as the info boxes serve different information needs than health information portals from the perspective of consumers.

Defendant argues that Plaintiff/Movant does not have any UCL claims as it does not compete with Defendant. The requirements in terms of conduct, Defendant says, are not met, nor are those under the Media Service State Treaty. Being based in Ireland, Defendant says it is not subject to the Media Service State Treaty. Furthermore, Defendant says § 94 of the Media Service State Treaty is not a rule of conduct in the marketplace, and is enforced only by state media authorities.

Defendant argues the balance of hardships, which is relevant to the analysis of the urgency [a requirement for a preliminary injunction], weighs against the motion. Plaintiff/Movant has failed, in Defendant's opinion, to make a sufficiently credible showing of negative effects (as stated above). In order to have claims under antitrust law, Defendant says those negative effects would have to border on an existential threat, which it says has not been shown. However, Defendant would be prevented from offering an important product innovation as a result of a preliminary injunction, Defendant argues. Defendant says the public interest in access to reliable health information should bear considerable weight.

The Court makes reference to the further representations by the parties in the pleadings and their exhibits as well as to the minutes of the oral hearing held on January 20, 2021. On January 22 and 26, 2021, the parties filed post-hearing briefs without having granted leave, and those briefs did not persuade the court to reopen the oral hearing (§ 156 Code of Civil Procedure). The Court took note of the parties' legal argument. [In Germany, counsel is always free to make legal argument to the court, even post-trial, but factual representations are ignored if they are out of time and were submitted without leave.]

Reasons for the decision

Plaintiff/Movant's first [the injunction as granted] and third [the standard PI sanctions] prayers for relief are admissible and meritorious. There is an entitlement to an injunction and the additional criteria for a preliminary injunction are met as well. Plaintiff/Movant is, therefore, granted an injunction as ordered, based on § 33 para. 1 Competition Act in conjunction with § 1 Competition Act, Art. 101 para. 1 TFEU.

The second prayer for relief was inadmissible and, therefore, had to be dismissed.

I.

  1. Plaintiff/Movant's first and third prayers for relief are admissible, and the first prayer for relief is, in its final form [as stated at the oral hearing], of sufficient specificity. Defendant's related criticism has been addressed in part by the modification of the prayers for relief. Defendant's doubts about which forms of collusion are meant to be barred is irrelevant, provided that--as the new prayer for relief requires--the object of collusion is the exclusive prominent display of the info boxes. Given that Plaintiff/Movant did not declare its second prayer for relief to be an auxiliary request, the Court assumes that both prayers for relief are independent of each other. [If NetDoktor.de had made its second prayer for relief an auxiliary request, it would have been treated as a fallback position in the event the first prayer for relief had failed; as the first prayer for relief succeeded, the court would then never have reached the auxiliary request.]

    The modification of the first prayer for relief was not an amendment to the complaint, but a limitation of the original prayer for relief to a specific infringing act. § 264(2) Code of Civil Procedure, in conjunction with an otherwise admissible partial withdrawal of a complaint, § 269 para. 1 Code of Civil Procedure. Defendant consented to the partial withdrawal of the complaint, which would have been admissible regardless (cf. Dusseldorf Higher Regional Court, order of July 13, 1982, case no. 2 U 54/82 - NJW 1982, 2452).

    The first prayer for relief had to be made more specific in light of the argument made, as Plaintiff/Movant seeks a prohibition of an infringing action that is based on previous coordination between Defendant and the FMH. On the one hand, Plaintiff/Movant omitted that reference in its new prayers for relief, which might indicate that it no longer seeks to capture that agreement. On the other hand, Plaintiff/Movant unambiguously and exclusively bases the argument for its first prayer for relief on Art. 101 TFEU and, respectively, § 1 Competition Act, and on the agreement between Defendant and the FMH, not on unilateral conduct on Defendant's part. Therefore, concerted behavior had to be part of the wording of the injunction, as the core of the infringing action.

  2. The second prayer for relief is inadmissible for lack of specificity. Neither does it make clear which processes and methods on Defendant's part are meant nor whether it is meant to be limited to the display of info boxes or also meant to capture organic search results. The word "particularly" suggests it is not meant to be limited. But without being limited to a specific infringing action there is no discernible boundary to the scope of such prayer for relief.

II.

Plaintiff/Movant is entitled to an injunction against Defendant's cooperation with the FMH in terms of the display of content of the gesund.bund.de portal in info boxes reserved for the FMH and which link to the NHP. § 33 para. 1 Competition Act in conjunction with § 1 Competition Act, Art. 101 para. 1 TFEU. This entitlement to an injunction also includes the elimination of a situation that has arisen from the previous agreement: the accused info boxes.

  1. The cooperation between Defendant and the FMH constitutes an agreement between undertakings that results in a restraint of competition. Art. 101 para. 1 TEFUE, § 1 Competition Act.

    a. Defendant entered into an agreement with the FMH, the substance of which is that Defendant's envisioned info boxes on health topics are filled exclusively, and permanently, with content from the FMH's health information portal and with a link to the gesund.bund.de portal.

    (1) An agreement comes into being as soon as a principle congruence of wills has been reached between two parties. Neither is it necessary to set out all details nor is it required that the parties share an identical interest (Immenga/Mestmäcker/Zimmer, 6th edition, 2019, Art. 101 para. 1 TFEU, margin no. 68).

    The character of an agreement is easily established by contracts under civil law, i.e., in the case that two parties are bound to congruent declarations of will. But, in accordance with the jurisprudence of the European courts, it is also sufficient that the parties express their mutual will to behave in the market in a certain way, and this is so even if they do not believe to have a legal, factual, or moral obligation to do so (Immenga/Mestmäcker/Zimmer, 6th edition, 2019, Art. 101 para. 1. TFEU, margin no. 71, with further citations).

    (2) Led by such mutual will, Defendant and the Federal Minister of Health clearly showed at their joint press conference on November 10, 2020, their intent to henceforth exclusively collaborate on the creation of info boxes on health topics on Defendant's search results page. The representatives of Defendant and the Federal Minister of Health repeatedly and explicitly talked about their collaboration or cooperation. According to the mutually-corroborating representations by both, the object of this cooperation was that, if a user searches for medical keywords, the NHP's answers would be presented on Google's search results page in a prominently exposed info box. The Federal Minister of Health repeatedly and clearly exrpessed his gratitude toward Defendant for this cooperation and for the future prominent display of the NHP's information and the links.

    With those statements, Defendant's claim that there was no agreement between the FMH and Defendant beyond a coordination of technical details is absolutely irreconcilable. Even if one wanted to assume that Google's [Defendant's] executives used promotional or non-technical language, that would not explain what the cooperation, announced with significant media attention, should be based on if not an agreement between the parties [to the agreement]. The joint press conference alone presupposes that the parties [to the agreement] had coordinated that they would make an announcement, and what they would announce. The result was not the announcement of particular technical details, the format of the text material provided, or the specifics of the interface created. Instead, what was expressed was mutual excitement over the fact that the two parties [to the agreement] would, generally speaking, together see to it that the NHP's content--and exclusively that one--would be displayed in the info boxes and, thereby, link to the gesund.bund.de website.

    The exclusive nature of this cooperation was also expressed at the press conference: no announcement was made of any similar cooperation by Defendant with other health information portals. Even Defendant does not flatly deny this exclusive nature. Much to the contrary, Defendant even argues--which differs from the info boxes on health topics displayed by the Bing search engine based on content from different databases--that such cooperation would have to be generally exclusive, as it would have to preselect a content provider and verify the suitability and accuracy of the content.

    Defendant does, however, also argue that there was no term during which the Defendant and the FMH would be bound. Instead, Defendant says it did not enter into any obligation, but could at any point in time commission a different health information portal with the provision of content for the info boxes. However, this claim is diametrically opposed to the type of cooperation announced at the press conference. While it does appear correct that no specific term was agreed upon, there can be no doubt that this cooperation was intended to last.

    The clearly articulated intent and purpose of this cooperation, according to the uncontradicted statement by the Federal Minister of Health, was at least for the minister that the NHP would be established as the leading go-to site for health information on the Internet. The FMH thereby came--which was also clear to its partner--from the assumption that this cooperation would put the NHP into that position, which means it would have to last long enough for the NHP to gain a decisive advantage over other, private providers competing for visibility on the Internet. The Federal Minister of Health expected the following effects from this cooperation: "Whoever googles health information should land on our federal government's portal" and "Gesund.bund.de should become the central go-to site for reliable health information on the Internet". Clearly, the FMH did not perceive this cooperation to be in place for only a short period of time, and the statements by Defendant's representatives do not indicate otherwise. What also counsels against a short term is the non-negligible effort made by the ministry in order to create dedicated text material and digital markups in order to enable the importation of its content into the info boxes. Both parties to the agreement corroboratingly stated that they would start with 160 illness terms and intended to further expand on this cooperation. Nothing in this suggested a cooperation for only the short term or the mid term.

    b. In connection with its operation of the gesund.bund.de portal, the FMH must be categorized as an enterprise.

    (1) The term "enterprise" ["undertaking" in EU law] must be understood in a functional sense. Its meaning is to be derived from the statutory context and the objectives of the relevant competition rules. As a result, any entity carrying out an economic activity, independently of its legal form and the way it finances itself, is an enterprise for the purposes of copetition law (ECJ, decision of April 23, 1991, Rs. C-41/90, Slg. 1991, I-1979 margin no. 21 "Höfner und Elser"; ECJ, decision of September 10, 2009, Rs. C-97/08 P, Slg. 2009, I-8237, margin no. 54 "Akzo Nobel"; Immenga/Mestmäcker/Zimmer, 6th edition, 2019, Art. 101 para. 1 TFEU, margin no. 9).

    An economic activity is, in principle, any independent activity that consists of the offering of goods or services, for the purpose of revenue generation, on a particular market (cf. Dusseldorf Higher Regional Court NZKart 2017, 247 250, with further citations). The determinative aspect is not whether the enterprise in question collects a payment; what matters is whether, in the relevant market affected by a restraint of competition, the service provided is usually offered for pay--if so, the one who provides a free service is also an enterprise and, therefore, subject to Art. 101 para. 1 TFEU (Bechtold/Bosch/Brinker, 3rd edition, 2014, Art. 101 TFEU, margin no. 14).

    There is no economic character that would justify the application of competition law if activities are performed as a government discharges its public-administration duties (cf. ECJ, decisions of January 19, 1994 - C-364/92, Slg. 1994, I-43 - SAT-Fluggesellschaft; Dusseldorf Higher Regional Court NZKar 2017, 247, 250, with further citations, beck-online). However, to the extent that a governmental entity carries out an economic activity that could be separated from its public-administration duties, it acts as an enterprise with respect to such activity (cf. ECJ, decisions of March 26, 2009 - C-113/07, Slg. 2009, I-2207, margin nos. 71 et seq. - Selex Sistemi Integrati/Commission; July 12, 2012 - C-138/11, WuW/E EU-R 2472, margin no. 38, Compass-Datenbank; Dusseldorf Higher Regional Court ibd.).

    (2) The operation of health information portals is a commercial activity that consists of offering a service, particularly the provision of information relating to health questions, for Internet users. The undisputedly substantial demand for such an offering is presently and has previously been met by a multitude of private-sector providers, such as Plaintiff/Movant, through typically advertising-financed portals. The National Health Portal is such a health portal, regardless of its financing mode, which is irrelevant in this context. It does not play a role that the NHP may possibly use a simpler language or a particularly structured presentation as compared to other offerings on the Internet. Such characteristics may represent a competitive factor that sets the FMH's offering from others. Similarly, the NHP's articles may be less detailed than those of other portals. It is already questionable whether it constitutes greater neutrality to omit controversies on topics such as vaccination. In any event, those special characteristics do not outweigh the similarities between the portals and fail to make the NHP a category of its own. Instead, the portals' similarities are overwhelming: just like other health portals, the NHP also provides editorial information on health questions of all kinds, portrays illnesses, and makes recommendations for a healthy life style. Notwithstanding differences in presentation and the circumstance that Plaintiff/Movant's portal displays advertisements, there must be a presumption of the functional substitutability of the portals from the perspective of the other side of the market--consumers searchging for health information on the Internet--in light of the far-reaching overlaps in content and the portals' mission statements. That was also the FMH's assumption when it sought tenders and stated that the NHP should "close a significant gap in the traditionally commercially oriented market of digital health information on the German-language Internet" (cf. Exhibit K19, *7). Last not least, the circumstance that both portals are offered alternatively by Google Search to users in order to satisfy the same need for information speaks to their functional substitutability.

    The operation of health information was not previously and need not generally and necessarily in the future be performed by governmental entities (cf. ECJ, decision of April 23, 1991, Rs C-41/90, margin no. 22).

    In principle, it is free of charge for users to utilize health information portals if those are private-sector, advertising-financed portals. However, in those caes the acceptance of advertising must be seen as compensation, thereby constituting an economic activity.

    It is unavailing that Defendant argues that the FMH carry out public duties through the operation of its gesund.bund.de NHP, specifically, to educate about health. The participation in general commerce by a governmental entity does not lose the character of a commercial activity subject to competition law only because it is also meant to fulfill public duties or to serve the public interest. If a governmental entity--as in this case--avails itself of means that are subject to civil law, it is subjected to the same limitations as any other participant in a private-sector market and, therefore, particularly has to respect the limits to such activity prescribed by competition law (Federal Court of Justice WuW/EDE-R 289, 293 - Lottospielgemeinschaft).

    c. Through their cooperation, Defendant and the FMH cause a restraint of competition in the market for providers of health information on the Internet.

    (1) In this regard, contrary to Defendant's position, the standard is not whether a restraint of competition is intended. The ECJ applies an objective standard to the question of whether an intended restraint of competition is identified. According to that standard, it does not matter what the parties to an agreement intend--although such intent can be viewed as an indication of the suitability of an agreement to a restraint of competition--, but whether the relevant activity has an objective competition-restraining tendency. In order for an anticompetitive purpose to be found, the coordination must harm competition based on its nature alone. Therefore, it must show sufficient harm to competition in and of itself, which obviates the need for an inquiry into its effects (Immenga/Mestmäcker/Zimmer, 6th edition, 2019, Art. 101 para. 1 TFEU, margin no. 130).

    Such an objectively harmful tendency, which has been presumed to be present in horizontal price-fixing, geographic non-competes and vertical price-fixing, is no inherent to the agreement between Defendant and the FMH.

    (2) Therefore, the exclusive agreement must be judged based on its concrete effects. Based on Plaintiff/Movant's pleadings and affidavits, the potential and actual effects and the declared objective of the parties to the agreement, one must conclude that the agreement between Defendant and the FMH actually and tangibly restricts competition in the relevant market.

    (a) With respect to the restraint of competition, the focus is on the actual effects, i.e., the objective consequences of the agreement for the market. Competition-restraining effects are present if the agreement leads to a restraint of competition in the sense of a limitation of the options of third participants in the market with respect to their actions and choices (Mü-Ko WettbR/Säcker/Zorn, third edition, 2020, ARt. 101 TFEU, margin no. 274).

    For the identification of the objective consequences, one must examine the causal relationship between the agreement and the current competitive situation. At least hypothetically, the agreement must cause the existing competitive situation. In accordance with applicable case law, a counterfactual must be analyzed, i.e. a juxtaposition of the current competitive situation with the hypothetical situation that would be the case if not for the performance of the agreement in question. Not only actual, but also potential effects on competitoin must be considered, at least in situations, in which an agreement has not been implemented, and such situations in which an agreement has been implemented, but its effects, though likely to occur, are not yet identifiable. This is due to the preventive nature of the prohibition (MüKo WettbR/Säcker/Zorn, 3rd edition, 2020, Art. 101 TFEU, margin nos. 275, 277).

    (b) The options for Plaintiff/Movant's actions and choices are substantially restricted by Defendant's agreement with the FMH because, on a permanent basis, the best possible position on Google's search results page--that is, the newly created, proinently-exposed position "0" in the info box--is exclusively reserved for the NHP's content, and therefore unavailable to competitors.

    Here, the relevant market is the one for offerings of health information portals on the Internet, a market in which Plaintiff/Movant and the FMH are active as providers of substantially similar health information portals (as explained above). With its market share of more than 90%, Defendant is the market-dominating provider in the upstream market of the provision of general search services in Germany. Plaintiff/Movant's representation that in the vast majority of cases (88-90%) users find their way to health portals via the Google Search is undisputed. Therefore, the operators of health information portals are to a particularly high degree dependent upon achieving good visibility on Defendant's search reluts page in order to be visited by users and to generate traffic from users that can in turn be monetized through advertising contracts. So far, portal operators had the possibility at their disposal to reach the top region of the search results page by competitive means, particularly either the creation of particularly relevant content or further optimization measures with respect to their ranking in the organic search results, or--at least theoretically--through the purchase of advertising space. Now, however, besides or above the organic search results there is an info box reserved exclusively for the NHP's content, and to which the FMH's competitors in the market for health information portals will for the foreseeable future not have access.

    It is obvious that Plaintiff/Movant must fear, as a result from this, substantial competitive harm. A central marketing tool is removed from competition and exclusively granted as a "pole position" to the FMH, a competitive advantage that cannot be compensated for by other means. PLaintiff/Movement has sufficiently substantiated that users primarily take note of the topmost results on the sea4rch results page. On the one hand, Plaintiff/Movant referenced the behavioral economics-based explanations by the European Commission in its "Google Search (Shopping)" case. On the other hand, it is consistent with general experience that users are more inclined to visit the topmost results of a search query. Not least is it the stated goal of the parties to the agreement to draw the attention of users to the NHP's content through the prominently displayed info boxes. As evidenced by their statements at the press conference, Defendant and the FMH come from the assumption that those [info boxes] are more likely to be noticed by users than the remaining content of the search results page.

    To the extent that Defendant asserts that the info boxes merely provide basic information that gives a first overview without substituting the information offered by Plaintiff/Movant, that argument is at least contradictory. Defendant itself emphasizes in a different context that the info boxes provide direct answers to search queries and meet demand for health information in a time-saving and targeted fashion. Furthermore, that claim is irreconcilable with the actual content of the info boxes, which contains sufficient answers for some information seekers, and with the actual effects substantiated by Plaintiff/Movant.

    It may very well be that the users make little use of the links to the NHP, as Defendant claims. However, the reduced click rate on Plaintiff/Movant's page suggests that a large number of users abort their searchdue to the info boxes that already satisfy their need for information. What has been sufficiently credibly substantiated is that the info boxes caused (though disputed by poionting to otherwise equally occurring fluctations in the number of accesses) of the--undisputedly--reduced click rate for certain health terms. Plaintiff/Movant selected health terms for which the ranking in the organic search results during the relevant period was constant, making a reduced click rate attributable only to reduced visibility. Also, a correleation between the extent of the loss of visibility and the reduced click rate is suggested by the fact that this reduction is even greater on mobile end-user devices, where the [vertical] arrangement of ads, info boxes and organic sarch results requires scrolling further down in order to the organic search results.

    An alternative explanation for the reduced click rate has not been given by Defendant, who instead stressed over and over that--also by comparison with Austria--the (in Defendant's opinion) solely relevant and monetizable total traffic on Plaintiff/Movant's website remained constant. It is, however, unpersuasive that nominal traffic should be a better indication of [anti-]competitive effects of the accused agreement than the click rate. Nominal traffic figures do not state how that traffic would have developed if not for the accused agreement, which is, however, essential to the analysis of the [anti-]competitive effects. The click rate makes that connection at least approximatively by measuring the development of clicks on a website with a constant ranking [among organic search results] and a constant number of impressions. If Plaintiff/Movant's search results are shown at a constant frequency and with a constant rank, but clicked on less frequently, and this effect occurs simultaneously with the introduction of the info boxes, it is a clear indication that even a well-positioned organic search result has lost some attention due to the info boxes.

    It may very well be that the analyses provided by Plaintiff/Movant merely have the character of examples and that no specifically suffered revenue loss can be quantified. In this case, Plaintiff/Movant can point to the above-mentioned potential effects of the accused agreement, as the short term during which the info boxes have appeared was in all likelihood insufficient for the competition-restraining effects to materialize in full. Against this background, we do not need to consider the secondary effects Plaintiff/Movant is additionally concerned about and which one may fear as a result of a lower ranking due to more frequent visits to the NHP's website due to the info boxes and its effect on the Google search algorithm.

    The negative effects prevent Plaintiff/Movant from providing its content to as many users as it could if not for the competition-restraining agreement. Placing costly advertisements in order to maintain visibility is not an economically equivalent alternative that Plaintiff/Movant could, or would have to, accept. As a result, Plaintiff/Movant can generate lower revenues from advertising and, therefore, has less room for investment in the improvement of its service than it could if not for the existence of the eye-catching info boxes.

    To the extent that Defendant calls into question the causation of the restraints of competition by the agreement, emphasizing that the creation of the info boxes is based on a unilaterally-made decision and coordination with the FMH was, at most, necessary for its implementation, this case is no different from any other case of a vertical agreement. Whatever deal is done by a company presupposes a unilateral decision by a corporate officer to make such deal. This situation is no different. The effects on competition do not result from an enterpreneurial decision to introduce info boxes, but from the agreement, with which it is implemetned, i.e., the agreement with the BMG to cooperate exclusively to fill the info boxes with content and links, and the performance of that agreement.

    (3) An exemption of the agreement under Art. 101 para. 3 TFEU or, respectively, § 2 Competition Act is not possible. The agreement does not contribute to the improvement of the production or distribution of goods or the promotion of technical or economic progress.

    The eemption under Art. 101 para. 3 TFEU applies to agreements that generate efficiency gains. A distinction is made between qualitative and quantitative efficiency gains. The agreement must result in objective advantages capable of compensating for the negative effects of a restraint of competition. In accordance with longstanding ECJ jurisprudence, this requires that the positive effects of the competition-restraining agreement must ultimately increase public wealth (Immenga/Mestmäcker/Ellger, 6th edition, 2019, Art. 101 para. 3 TFEU, margin no. 134).

    If there is an objective benefit, it must be analyzed with a view to the [European] Union's public interest. It is not sufficient for two parties, as a result of their agreement, to merely have a more reliable basis for planning and to be able to more efficiently organize their logistics. Instead, there must be an effect on the market that is in the interest of the Union (MüKo WettbR/Wolf, third edition, 2030, Art. 101 TFEU, margin nos. 1083, 1084).

    The participating companies must plead a plausible estimate of the extent of the expected benefits that can be juxtaposed to the harm caused by the restraints of competition (cf., Immenga/Mestmäcker/Ellger, 6th edition, 2019, Art. 101 para. 3 TFEU, margin no. 138).

    (4) Defendant has not met its burden of pleading and substantiating such compensatory efficiency gains.

    (a) It is questionable whether the integration of syndicated content constitutes an improvement of a search engine, as this is ultimately less of an improvement of the search engine service than it represents an expansion of Defendant's [commercial] activity into another market, specifically the market of publishers or other content providers. Whether that content is encyclopedic or editorial, there is in any event an intentional reduction of choice and, therefore, the content is not entirely opinion-free. Defendant goes beyond its basic function of bringing together, as an Internet search engine, demand for products or services (such as information) and its providers, but also goes beyond the immediate response to factual queries (e.g. weather, height of the Eiffel Tower). Ultimately, Defendant steps outside the market of a pure search engine in the sense of an intermediary of products and users, and becomes a provider of such product by itself. On the one hand, it is consistent with the notion of performance-based competition to modify and improve products. On the other hand, if Defendant, on a permanent basis, places an info box with the NHP's content before the organic search results, it judges the various sources that would be available on the Internet to answer a search query by giving prominent exposure to one of them as the authoritative answer. In doing so, Defendant makes a preselection that is detached from the Google [page-ranking] algorithm. Not only is this less than transparent to users but it also constitutes a judgment on cntent and, ultimately, influences the public opinion-forming process. Objectively speaking, it appears at least doubtful whether this constitutes an improvement of a search engine.

    (b) Even if one subscribed to Defendant's premise that its offering becomes more attractive by incorporating info boxes with immediately visible health information (cf. Hamburg Regional Court, order of April 4, 2013, case no. 408 HKI 36/13), this would not constitute an efficiency gain within the meaning of the exemption rule of Art. 101(3) TFEU.

    Neither does making a single actor's product more attractive constitute an increase in overall public welfare nor has Defendant substantiated an effect on the marketplace that would be in the interest of the [European] Union. Such effect is not to be expected. The more attractive design of a single provider's offering cannot outweigh the downside resulting from the restraint of competition affecting all other participants in the market.

    This is particularly true when--as in this case--a company with a dominant market position affords on a permanent basis, in its role as "gatekeeper" in the upstream search engine market and to the detriment of private competitors depending on it, a tax-financed offering by the government a "pole position" in the battle for users' attention in order to improve the attractiveness or, respectively, visibility of its own products. In this case, two operators who do not have to take much of an economic risk themselves, interfere with the market and complicate the access to users of already-existing players in the market, which in the market specifically affected in this case also comes with an impact on the diversity of media offerings and opinions.

    Neither has Defendant plausibly pled how the alleged upside outweights the downside in this case, nor is it evident.

    (c) With respect to Defendant's claim that improving public education about health topics increases public welfare, there are no plausible pleadings of such benefit and the weight it should bear. In any event, if one weighs the overall circumstances described above, this benefit does not appear quite so overwhelming as to justify the threat of forcing other reliable vendors out of the market for public health information. To the contrary, there is a threat of a loss of the existing diversity of high-quality health information portals and, therefore, of the availability of medical "second opinions."

    2. Contrary to Defendant's auxiliary request, the grant or, respectively, the enforcement of the preliminary injunction does not depend on security in accordance with Articles 935, 936, 921 (second sentence) of the Code of Civil Procedure. Neither has Defendant provided reasons for that request nor has it pled and quantified potential damages nor is there an obvious reason for such order [of a requirement to post collateral].

III.

Plaintiff/Movant has also substantiated the existence of a sense of urgency.

There is urgency if there is an objective danger of an alteration of the status quo thwarting or complicating the enforcement of the moving party's rights by means of the final judgment at the end of the main proceeding and its enforcement. The moving party does not have to content itself with a damages claim (MüKoZPO/Drescher, 6th edition, 2020, Art. 935 Code of Civil Procedure, margin no. 15).

As explained above, Plaintiff/Movant has credibly pled that it loses traffic as a result of its reduced visibility on Google Search due to the info boxes, which has already materialized in the reduction of click rates relating to particular illnesses, and that user traffic is critical to the conclusion of profitable agreements with advertisers. One must fear that this reduction [of traffic] results in a specific loss of revenue until the conclusion of a [full-blown] main proceeding, especially since Plaintiff/Movant has credibly shown that, due to keyword targeting being its key marketing tool, reduced visibility in a narrowly defined range of topics is materially adversely impacted (cf. Exh. K36, page 2). Neither does Plaintiff/Movant have to wait until such revenue loss has occurred, nor is it a requirement for a preliminary injunction that the moving party credibly claims that its viability is in jeopardy, especially since Plaintiff/Movant is not seeking specific performance.

Plaintiff has not acted in contradiction to the urgency resulting from those circumstances by not having done everything in its power in order to obtain the preliminary injunction as swiftly as possible (cf. Dusseldorf Higher Regional Court, judgment of September 7, 2020, VI-U (Kart)4/20 - NZKart 2020, 545, 546). The preliminary injunction request was filed with the Court on November 27, 2020, i.e. within one month of the start of the cooperation between Defendant and the FMH.

The imminent harm [to Defendant] is not outweighted by higher-priority interests on Defendant's part. A lasting reputation loss by Defendant, resulting from the removal of a recently-launched offering, or any negative impact on the public interest in public health cannot be [realistically] expected.

IV.

The cost award is entered in accordance with Articles 92(1)(first sentence), 269(3)(second sentence) Code of Civil Procedure. The rejected second prayer for relief has materially the same thrust as the granted first prayer for relief. The extent to which the second prayer for relief and the withdrawn part of the first prayer for relief exceeded the ultimately-granted injunction cannot be afforded so much weight as to require Plaintiff/Movant to bear more than one quarter of litigation expenses.

The decision on provisional enforceability concerning Defendant's entitlement to a reimbursement of costs is based in Articles 708(6), 711 Code of Civil Procedure.

[judges' signatures]

Announced on February 10, 2021

[clerk's signature]

Share with other professionals via LinkedIn: