Getty’s high-resolution competition law complaint against Google in the EU

Getty Images has filed a competition law complaint against Google

I have some initial thoughts as follows:

Image Search vs. High-Resolution Image Search

Google’s rationale for image search in general is that displaying the image is necessary for the user to assess how well the image corresponds to their search. This practice has been litigated at least twice in the U.S. in relation to thumbnail images and has easily passed the test of fair use.

Getty’s complaint is directed more specifically to the creation of high-resolution galleries. Although Google could make a similar argument that you need to see the image in high resolution to properly evaluate it, my view is that this argument is not nearly so compelling. The high resolution display is more expressive and less informational and the potential adverse effect on the copyright owner is greater for that reason and because, as Getty points out once consumers see an image on the Web, they aren’t likely to go to the source and look at it again. Getty argues that Google Images’ creation of high-res galleries of copyrighted content is thus impacting Getty’s own image licensing business; promoting piracy and copyright infringement; and bolstering Google’s monopoly over site traffic, engagement data and ad spend.

[Update: It is worth noting that other design changes may have reduced the flow on traffic from Google Image search see this 2013 Search Engine Land story]

Basically, Getty is worried that it is far too easy to right-click and copy image from Google Image Search and that as a result people won’t click down to Getty’s now site.

Illustration: Right-Click Options On Getty’s Own Press Release

Screen Shot 2016-04-27 at 3.58.51 PM

Is there a copyright case?

Getty is an American company complaining about another American company’s treatment of its copyrighted properties and so it is less than obvious that the complaint would be dealt with in the EU under antitrust law rather than in the US under copyright law. I would need to understand the technology behind the high resolution image display to say whether Getty has a strong case for copyright infringement under U.S. law, but I think there are clear differences between the fair use status of low-resolution thumbnails and the high resolution images available on Google image search today.

Is there a competition law case?

It seems to me that providing high resolution galleries makes it at least marginally less likely that users will click through to the original site. This in turn makes it marginally more likely that users will copy and paste without authorization. But I would note that Google is under no obligation to design its information services in a way that drives traffic to a particular website, or external websites in general. The design of Google image search would make a poor antitrust case in the US but it might go further in the EU because they take a broader view of “abuse of dominance”. The main problem I see is that even if Google image search increases the unauthorized use of images, it probably does not affect the market for licensed images. I think that people looking to license stock footage will go to stock photo sites, but this is an empirical question so we would need to look at how the market actually works.


Getty Images says that, when it first raised concerns about this with Google, it was told to accept Google’s presenting of images in high-res format or opt out of image search. Since its founding, Google has relied on the fact that people can opt out of search to address complaints about the way search is run. Participation in Google image search is voluntary – on an opt out basis – but that does not give Getty, or anyone else, the right to say exactly how they would like the search engine to run. In 2013 Google agreed with the FTC to make changes to the way it scrapes the content of rivals like Yelp. Sites like Yelp can now opt out of having their content scraped without opting out of search entirely. This is probably what Getty is looking for. But the issue for Getty is that it does not compete with Google they way Yelp does, so their case is not as strong.

It’s hard to say how might EU regulators will view Getty’s complaint. It depends on some facts that we don’t have access to at the moment — I have only seen the press release, not the complaint itself. It also depends on whether the EU case against Google has more to do with politics and protectionism than it does with the merits of competition law. If Google settles with the European commission, the settlement might include modifications to the way image search works and that would be a significant win for Getty. From Google’s point of view the worst case scenario is that it would be forced to make a change to image display within the EU. I don’t think Google will change image search in the US unless it thought Getty had a case under American copyright or antitrust law.

Some measured thoughts on patent trolls

This post expands on the remarks I made today at the Chicago Tech Roundtable meeting on Patent Trolls and Chicago’s Tech Community. The meeting was attended by a number of elected officials and their representatives as well as start-ups such as Jump Rope and Options Away Travel who have had direct experiences with patent trolls. This is an important issue for Chicago’s technology sector.

Trolls and Trolling – The Nature of the Problem

Patent trolls are in the news and they have been high on the agenda of intellectual property policy makers and academics for over a decade now. I started thinking about these issues when I worked as an IP lawyer in Silicon Valley in the early 2000’s. The Federal Trade Commission sounded an important call to action on patent trolls and the balance of competition and patent law and policy in 2003 [FTC, To Promote Innovation: The Proper Balance of Competition and Patent Law and Policy, (2003)], and again in 2011 [FTC, The Evolving IP Marketplace: Aligning Patent Notice and Remedies with Competition (2011)].

I first wrote about these issues in a paper published in 2007, Patent Reform and Differential Impact (with Kurt W. Rohde who was then a student of mine and is now is a partner with McDonnell Boehnen Hulbert & Berghoff LLP), some things have changed since then, but the patent trolls problem persists, it may even be getting worse. In 2012, businesses and individuals targeted by patent aggregators and patent holding companies accounted for 58 37.8% of all patent defendants.

* An earlier version of this post contained the assertion that businesses and individuals targeted by patent aggregators and patent holding companies accounted for 58% of all patent defendants. That was a very unfortunate transcription error.

Not every non-practicing entity, patent aggregator and patent holding company is a necessarily a troll. There is obviously a role in our innovation ecosystem for people who invent but don’t have the complementary skills to commercialize. But the rough and ready correlation between patent assertion entities and trolls seems to fit. Any business that is sending out infringement notices by the hundreds or thousands can’t plausibly be doing the kind of diligence it should take to make a meritorious claim of infringement.

Debating who is and is not a troll is beside the point – it is trolling behavior that we need to address. Trolling is abusive and opportunistic behavior such as asserting bad patents or using patents to extort settlements that are only justified by the threat of legal fees. Trolling is mostly a numbers game – trolls targets hundreds or thousands of defendants, seeking quick settlements priced just low enough that it is easier for the defendant to pay the troll directly rather than pay his lawyers to defend the claim. Anyone who takes part in this kind of systematic opportunism that undermines innovation is a troll, NPE or not.

Patent trolls thrive by opportunistically taking advantage of the uncertain scope of patent claims, the poor quality of patent examination, the high cost of litigation and the asymmetry of risks and costs of litigation. There is nothing wrong with licensing your technology, or with litigating against infringers who would rather take your technology without a license. The patent system is meant to encourage investment in R&D and innovation. Businesses monetizing their technology aught to be celebrated, but using the threat of costly litigation to monetize bad or ill-fitting patent claims takes money away from R&D budgets for no social gain.


There will always be opportunists who try to exploit the system – the goal of patent reform should be to limit unproductive rent-seeking while leaving the door open to those businesses that actually contribute to the research and development that makes the U.S. a world leader in so many fields. We need reforms targeted at bad patents such as limiting continuations, rejecting highly abstract functional claims (especially in software and business method patents) and improving the quality of patent examination. But we also need reforms targeted at bad conduct. We need reforms that level the litigation playing field — it is far too easy to impose millions of dollars of defense costs based on dubious patents and tenuous theories of the scope of those patents. Some of the most useful reforms— reforms that leave the door open for well-founded claims — include: fee shifting such that the losing party pays the winning party’s fees in ordinary cases; heightened pleading standards; delaying discovery until after “claim construction,” and limiting discovery to those documents likely to be relevant to the specific litigation at hand. “Consumer stay” provisions would also do a lot to end opportunistic threats of litigation – under this proposal if, for example, a cafe was sued by offering Wi-Future interest to its customers, the manufacturer of the router could intervene and effectively consolidate the cases of all of its customers. Reforms aimed at transparency of patent ownership and taking action against misleading and deceptive language in demand letters would also be of some assistance.

House recently passed the Innovation Act to take up some of these issues and the Senate Judiciary Committee seems close to finalizing a complementary bill. Neither of these bills will put an end to opportunism, but they have the potential to make life a harder for patent trolls and a little easier for the rest of us.

Patent Troll Statistics

According to RPX Corporation PAEs initiated 62% of all patent litigation suits in 2012. [See Colleen Chien, Patent Trolls by the Numbers (] However, it is not exactly clear how RPX determines who is and is not a patent troll. Also RPX is in the business of providing “patent risk management services”, so it is not exactly a disinterested bystander in the patent troll debate.

The best empirical analysis of patent troll numbers  so far is contained by Christopher A. Cotropia, Jay P. Kesan & David L. Schwartz, Unpacking Patent Assertion Entities (PAEs) (working paper available at Figure 3 of that paper reports these numbers in terms of the number of individual defendants sued. On this metric:

  • suits by large aggregators and patent holding companies increased from 31.6% of all patent litigation in 2010 to 37.8% in 2012;
  • in contrast suits by operating companies went down from 48.9% in 2010 to 47.3% in 2012;
  • if you include the IP holding companies of operating companies, suits by operating companies went down from 51.0% in 2010 to 47.8% in 2012;

Cotropia, Kesan & Schwartz round out this picture by reporting the numbers for universities & colleges, individuals & family trusts, failed operating companies & failed start-ups, and technology development companies. Some of these suits may be troll litigation, but without case specific information it is hard to tell.


New Jersey’s anti-Tesla laws and the car dealership racket

Over 70 economists and law professors have signed a letter opposing New Jersey’s direct automobile distribution ban.

Why should you care?

This ban is aimed at keeping the Tesla electric car out of New Jersey, but this effects you even if you have no interest in an electric car. State laws protecting car dealerships add thousands of dollars to the cost of every new car.  NPR’s Planet Money program has an excellent summary of this issue.

The International Center for Law & Economics sent an open letter to New Jersey Governor Chris Christie today, urging reconsideration of the regulation.

As the letter notes:

 the regulations in question are motivated by economic protectionism that favors dealers at the expense of consumers and innovative technologies. It is discouraging to see this ban being used to block a company that is bringing dynamic and environmentally friendly products to market.

Among the letter’s signatories are some of the country’s most prominent legal scholars and economists from across the political spectrum.

Read the letter here: Open Letter to New Jersey Governor Chris Christie on the Direct Automobile Distribution Ban

Read Geoffrey Manne‘s post on Truth On The Market

x-post on running data

I posted a few comments on the ownership of running data and some implications for IP and Antitrust law a few days ago on “Reading, wRiting and Running“. The post is repeated below, but to see the comments, go to

I started running in March last year when I borrowed someone else’s ipod nano and went for a jog in Menlo Park, California. I went running because I was just visiting and had no gym work out in. I took the nano to listen to music. But the experience of having the nano tell me every kilometer that passed and give me my run stats at the end got me hooked.

Running with data is way more fun than just running! Using a website like Nikeplus to track your data over time is even more fun.

But my frustration with the nano – it kept ending my runs early when I drenched it with sweat – led to me to want to edit the data. Nike does not let you do this. They don’t even let you merge two runs together. This led me to a very cool site that takes your nikeplus data and gives you some different data visualizations and lets you manually add and edit runs.

Smashrun have worked out how to get your data from the Nike website, but they are still working on Garmin. I now use a Garmin GPS watch to track my runs and calibrate my iPod nano at the end of each run. (I find that the nano is only accurate +/- 5%) I don’t use the Garmin website much because the interface is not great and they don’t analyze the data in ways I find interesting.

Some observations

Nike and Garmin obviously want obviously want to lock their users into their platforms as a means of giving you more reason to buy their gadgets and not to switch every time something better comes along.

Nikeplus is mostly a great website, but Smashrun is better in many ways. To compete with Nike as a data platform Smashrun needs access to the my data. But when I say ‘my data’ is that legally accurate? Do I really own my data to the extent that I have the right to port it somewhere else.

In the course of writing the post I have probably violated the “Platform Use Restrictions” of Nike’s Terms of Service because my “User Generated Content” is inextricably linked with their IP. I am confident that Nike won’t mind, but given that some U.S. prosecutors interoperate the Computer Fraud and Abuse Act as being violated by a TOS violation, maybe I should think twice.

Smashrun is probably not contractually bound by Nike’s TOS (quite a debatable point), but some courts might think that it is. Smashrun might be deemed to be in violation of the Nike TOS which provide that “You agree not to use any data mining, robots, scraping or similar data gathering methods.” (I might have also violated the TOS that reads “Keep your username/password secure and do not allow anyone else to use your username/password to access the Platform.”)

In the wake of the prosecution and subsequent suicide of Aaron Swartz, Representative Zoe Lofgren has drafted a bill that would exclude terms of service violations from the 1984 Computer Fraud and Abuse Act and from the wire fraud statute. This reform is long overdue.

As a matter of consumer protection, if not competition law, data the right of users to have third parties liberate their data aught to be the norm. Then we could have device competition between Nike and Garmin and data platform between Nike, Garmin, Smashrun and host of new entrants.



Brands, Competition, and the Law

On October 19, the Institute for Consumer Antitrust Studies is co-hosting a conference on Brands, Competition, and the Law along with University College London.  This is the follow-up to a very successful program on the same theme in London in December 2011.  A book with selected papers and comments from these conferences will be forthcoming.

We have assembled an all-star lineup of economists, marketing and branding professionals, as well as antitrust and IP lawyers and professors to try to reach a common understanding of the meaning and impact of brands in the market place and the appropriate legal regime. The full details and registration information for the conference are available at

The speakers include: Deven Desai, Kirsten Edwards-Warren, Phil Evans, Warren Grimes, Greg Gundlach, James Langenfeld, Ioannis Lianos, Deborah Majoras, Mark McKenna, John D. Mittelstaedt, John Noble, Barak Orbach, Joan Phillips, Matthew Sag, Eliot Schreiber, and Spencer Weber Waller.