Video News Aggregator Loses Fair Use Defense–Fox v. TVEyes

TVEyes aggregates video newsclips, makes them searchable, and lets subscribers watch responsive clips. One use case is for companies’ communications departments. They can set up searches for their brands in TVEyes’ database and monitor what’s being said about them.

To create the searchable database, TVEyes copies and stores third party copyrighted video. Fox News sued for copyright infringement. The appellate court rejected TVEyes’ lack-of-volition defense because “TVEyes decides what audiovisual content to record, copies that content, and retains it for thirty‐two days.” As a result, all of TVEyes’ copying constitutes prima facie copyright infringement. TVEyes also defended on fair use grounds. Despite its recent pro-fair use rulings in HathiTrust and Google Books, the Second Circuit rejects the argument.

The Fair Use Defense

There are many facets of TVEyes’ services, and not all of them appear to be implicated in this case. Apparently Fox didn’t aappeal TVEyes’ search functionality based (primarily) on closed-captioned text–perhaps because of the Second Circuit’s strong fair use rulings in the HathiTrust and Google Books cases. With respect to search results, TVEyes allowed its subscribers to watch 10 minute clips, to watch clips sequentially (though, during the course of the litigation, TVEyes prevented sequential clip watching), download the clips and permanently archive the clips on TVEyes’ servers. TVEyes also allowed subscribers to share the clips with non-subscribers. The court explains why those functions aren’t fair use:

* Nature of the use. The court says TVEyes’ enablement of watching clips is transformative, but only slightly. It explains:

TVEyes’s copying of Fox’s content for use in the Watch function is…transformative insofar as it enables users to isolate, from an ocean of programming, material that is responsive to their interests and needs, and to access that material with targeted precision. It enables nearly instant access to a subset of material‐‐and to information about the material‐‐that would otherwise be irretrievable, or else retrievable only through prohibitively inconvenient or inefficient means. …

The Watch function certainly qualifies as technology that achieves the transformative purpose of enhancing efficiency: it enables TVEyes’s clients to view all of the Fox programming that (over the prior thirty‐two days) discussed a particular topic of interest to them, without having to monitor thirty‐two days of programming in order to catch each relevant discussion; and it eliminates the clients’ need even to view entire programs, because the ten most relevant minutes are presented to them.

In a concurrence, Judge Kaplan (sitting by designation) disagrees that these functions are transformative.

* Nature of the work. Citing Google Books, the court says “This factor ‘has rarely played a significant role in the determination of a fair use dispute,’ and it plays no significant role here.”

* Amount taken. “TVEyes makes available virtually the entirety of the Fox programming that TVEyes users want to see and hear,” and this is “radically dissimilar” to Google Books.

The court explains:

While the [Google Books] snippets allowed a user to judge whether a book was responsive to the user’s needs, they were abbreviated to ensure that it would be nearly impossible for a user to see a meaningful exposition of what the author originally intended to convey to readers.

In contrast, TVEyes’ 10 minute clips–“given the brevity of the average news segment on a particular topic‐‐likely provide TVEyes’s users with all of the Fox programming that they seek and the entirety of the message conveyed by Fox to authorized viewers of the original.”

* Market effect. “Fox argues that TVEyes undercuts Fox’s ability to profit from licensing searchable access to its copyrighted content to third parties.” The court embraces this circularity:

The success of the TVEyes business model demonstrates that deep‐pocketed consumers are willing to pay well for a service that allows them to search for and view selected television clips, and that this market is worth millions of dollars in the aggregate. Consequently, there is a plausibly exploitable market for such access to televised content, and it is proper to consider whether TVEyes displaces potential Fox revenues when TVEyes allows its clients to watch Fox’s copyrighted content without Fox’s permission.

Such displacement does occur. Since the ability to re‐distribute Fox’s content in the manner that TVEyes does is clearly of value to TVEyes, it (or a similar service) should be willing to pay Fox for the right to offer the content. By providing Fox’s content to TVEyes clients without payment to Fox, TVEyes is in effect depriving Fox of licensing revenues from TVEyes or from similar entities.


Implications for Searchable Databases. Because of the litigation posture, the court doesn’t discuss TVEyes’ aggregation of content to create a searchable database. (The opinion only analyzes TVEyes’ “watch” policy). One possible interpretation is that the Second Circuit’s fair use jurisprudence from HathiTrust and Google Books is so clear that fair use protects the aggregation of searchable databases that it’s not even worth suing over. If so, this opinion is only about snippeting, i.e., what “previews” search engines can show as part of their search results.

TVEyes’ Snippeting Policy. Google Books had a complicated and multi-layered snippeting ruleset that surely went through multiple rounds of legal review (and probably some back-and-forth with the engineers). It was an attempt to convert fair use into an algorithm, and it worked in court. In contrast, TVEyes’ snippeting policy was, at best, simplistic: categorically, every user gets 10 minutes of every video. I’d love to hear more from the lawyer(s) who blessed that as a codification of fair use. Unfortunately, it leaves open the question of whether TVEyes could have succeeded with a more nuanced policy. Indeed, the opinion doesn’t make it clear if TVEyes will be allowed to do any snippeting at all–and without snippeting previews, TVEyes’ service becomes extraordinarily more cumbersome to use because, at most, TVEyes can just provide links to source videos if available. So the failure to craft a nuanced snippeting policy might result in a worse outcome than if TVEyes had taken a more conservative approach to snippeting.

Implications for DVR-as-a-Service. The Second Circuit’s Cablevision ruling paved the way for DVR-as-a-service. It held that the DVR service provider avoided copyright infringement because the users, not the service provider, took the “volitional” steps of making the copies. To expedite the analysis, the Cablevision litigants waived two key arguments: the plaintiffs waived secondary infringement, and the defendant waived fair use defenses. Thus, the court never opined on whether the DVR service qualified for fair use.

A decade later, is DVR-as-a-service still legal? The Second Circuit’s ruling has not been confirmed in other circuits, so in theory it’s still litigatable elsewhere. Plus, the Supreme Court’s Aereo opinion held that Aereo’s DVR service infringed copyright when it pulled in over-the-air broadcasts.

If the issue does get litigated elsewhere and isn’t resolved on volition grounds, this opinion suggests that the DVR service provider may not qualify for fair use. After all, it’s a commercial service that gets paid to deliver the full copies of third party content. The counter-argument is that DVR subscribers have already paid licensing fees to the broadcasters through their cable bills. At the same time, broadcasters like CBS and Disney set up their own proprietary offerings to enable paid viewing, maybe the licensing market has changed as well. Subscribers may have their own independent fair use defense, but that will not help protect the DVR service provider from direct infringement claims.

What Happened to Factor 2? It’s true that the second fair use factor sometimes doesn’t provide much insight, but the factor doesn’t deserve the court’s casual denigration of it. It can make a huge difference for unpublished works, for example; indeed, it was virtually dispositive in the Harper & Row v. The Nation case until Congress expressly cabined that case’s holding. Furthermore, it plays a crucial role in cases where the copyright is “thin” because it’s largely factual.

Here, the court could have used the second factor to acknowledge that most people using TVEyes’ service are trying to mine facts from news shows. TVEyes’ viewers aren’t binge-watching a highly fictionalized work Games of Thrones; they are often monitoring fact-driven news stories to find facts that will be used to rebut someone’s claims or that need to be rebutted. (Some of the case commentary has riffed on the dubious nature of Fox News’ “facts,” but we need not go there). The court should not have breezed past this fact-driven nature of TV news, and the second factor absolutely should have helped the court acknowledge it.

The Redefinition of Factor 4. Fair use analyses are frequently riddled with circular arguments, and factor 4 especially seems to invite circularity (i.e., we can’t know if a secondary market is legitimate until we decide the fair use question, but fair use turns on whether the secondary market is legitimate). In concept, it makes sense to consider whether secondary uses will financially substitute for the copyright owner’s authorized offerings; and in cases of brazen commercial piracy, this factor helps ensure the pirates lose. But in cases like this, involving a commercial aggregator, courts routinely use factor 4 more to validate their normative conclusion, using circular arguments, than to help conduct their legal analysis.

Here, the Second Circuit makes two unfortunate moves. First, it relaxes the already-too-lax Texaco standard for determining when a copyright owner can control a market that didn’t exist. Texaco protects copyright owners’ interests in “traditional, reasonable, or likely to be developed markets”; this opinion protects “a plausibly exploitable market.” By lowering the standard from “likely to be developed” to “plausibly exploitable,” the Second Circuit grants control over many more hypothetical and unexploited markets to copyright owners than Texaco would have (and again, I think Texaco’s standard is already too broad).

Second, the Second Circuit evaluates the “plausibly exploitable” standard by looking if the defendant has been able to generate revenues from that market. But this collapses the factor, at least for commercial aggregator defendants. Any aggregator defendant that is commercially successful enough to trigger a lawsuit will have revenues, and every copyright owner can argue that those revenues demonstrate that the market was “plausibly exploitable.” Following this court’s logic, the fourth factor should always count against commercial aggregator defendants. Grr.

But here’s the real catch: in many cases, no single copyright owner can actually successfully create a market for aggregation of their works. By definition, the aggregator derives a lot of its value by assembling disparate sources for customers. Think of it this way: how many different news broadcasters are there, and what would the market look like if each broadcaster set up a siloed service that offered the same services as TVEyes, but only for its content? At minimum, the high transaction costs for potential customers would reduce the market size as customers pick-and-choose only the most important silos; and likely many customers would find the hassle wasn’t worth the services’ value and would pay none of the siloed services. So the market is “plausibly exploitable” in this case only because an independent aggregator risked the investment capital to create it. Now, this ruling raises the required level of initial investment capital to include buying licenses from all of the broadcasters (if they are even willing to grant the licenses); or it will deter any new entrant from ever creating the market, which in turn has the effect of counter-productively proving that the market never was “plausibly exploitable.” That’s an unhelpful legal interpretation.


This is not a good ruling for fair use or the Internet. The court’s mangling of fair use factors 2 and 4 now becomes the law of the Second Circuit and will be applied in cases far beyond this one.. Furthermore, as I’ve indicated, this opinion will cast a shadow over a variety of Internet aggregation efforts, ranging from TVEyes’ service to DVR-to-a-service.

Still, I think there is a lot of not-bad news here. First, the HathiTrust/Google Books ruling remains largely intact, including the legitimacy of building searchable databases. Second, it seems like the door was open for TVEyes to succeed with a narrower snippeting policy. So I think there may still be room for services like TVEyes with hobbled functionality.

Case citation: Fox News Network LLC v. TVEyes, Inc., 15‐3885(L) (2d Cir. Feb. 27, 2018)