In recent weeks, I have had several calls from broadcasters asking if it was permissible to copy articles from other news sources and post them on the station website – with attribution to the original source. As I told them, posting content without permission of the copyright holder can lead to big problems. We have written about these issues in connection with the use of photos and video (see, for instance, our articles here, here and here), and recently even using embedded photos from a social media site have been called into question (see our article here). The copying of any substantial part of a news article raises the same issues as posting pictures or video found on the Internet onto your site. Such actions diminish the ability to of the content’s owner to profit from its own content. If someone can read a story on a broadcaster’s website, why would they need to go to the site of the originator of that content – even where attribution to the originating site (and even a link to that site) is given on the broadcaster’s site?

Years ago, there were many websites that would “aggregate” news by taking significant portions of news stories from other sites and make it available to the aggregator’s readers. There was a rash of lawsuits where content owners, including newspapers and others, claimed that aggregators using even a paragraph or two of the original story were infringing on their rights to their content. Content owners had real concerns about this aggregation sites, as a reader can usually get the gist of the story from the introductory paragraphs and, even when the aggregator provided a link to the full story, the readers would be far less likely to go to the full story when they had already been given its substance. Today, to avoid these lawsuits, most such news aggregators provide at most a headline (and sometimes even the headline can be creative enough to pose a copyright risk if run on an aggregator’s site – so just a generic paraphrase of that headline is often used), and at most a very brief description of the story on the originating site – a description that only directs the users of the aggregator site to the originating site and does not use any of the originating story’s language or original reporting, e.g. a statement that “you can find a good story about Virginia’s collapse in the NCAA tournament in this story” or “for more developments on latest in the personnel changes in the Trump Administration, check out this story in the Washington Post.” Using more than this kind of generic referral is a risk, and fair use is no often going to be available as a defense.

As we have written before, the concept of “fair use” is one that is often claimed but often rejected as a defense to potential copyright infringement. The concept of fair use is set out in Section 107 of the US Copyright Act. It basically allows for socially positive uses of copyrighted content without permission, but only where that use does not overly impinge on the financial value of the copyrighted work and effectively become a substitute for that work. In many recent cases, courts have characterized a fair use as one that is “transformative” – essentially changing the nature of the copyrighted work that is being used. So, if you use a snippet of a song or a brief quote from an article for purposes of criticizing that copyrighted work, the use is “transforming” the copyrighted material into something different (a comment or criticism), rather than merely making a reproduction of that work to provide it to an audience in the same way that the copyright holder is likely to do, in a way likely to interfere with the copyright owner’s ability to monetize his or her own work.

In conducting a fair use analysis, the Copyright Act says that you look at four factors:

  1. The nature of the use – including whether the use is commercial or for non-profit or educational purposes;
  2. The nature of the copyrighted work;
  3. The amount or substantiality of the work that is used;
  4. The effect of the use on the potential market value of the work itself.

How these factors play into any circumstance is a difficult question to analyze – and one where there are no hard and fast rules. There is no 10 second (or 30 second, or 15 second) safe harbor from potential liability that allows a user to take that amount of a copyrighted work like a song or video for use in a production (see our article here). Similarly, for written works, there is no safe harbor for using a paragraph or even a sentence. Instead, each use needs to be evaluated on its own.

The more educational or informational a use, the more that factor weighs in favor of a finding of fair use. The greater amount of the work that is used, the more likely a fair use defense is to be rejected. If the kind of use that is being made is one for which a creator could make money, the less likely fair use is to be found. But each of these factors needs to be weighed, and a court has to make a final determination of where the weight of these factors plays out. How that analysis will be resolved in most cases is hard to predict and thus makes risky all but the most limited and benign use of copyrighted content unless permission from the copyright owner is first secured.

A recent case illustrates how even a widely used, seemingly very useful service can raise significant questions about whether fair use applies. The US Court of Appeals for the Second Circuit recently analyzed the operations of a video clipping service, TVEyes, which recorded massive amounts of TV programming, indexed the video based on the words used in the programs (primarily from closed captions), and allowed clients to search the content and play back as much as 10 minutes around a search term to see if it is relevant to the user. The service was, according to the decision, widely used by journalists and government officials to track coverage of specific topics. TVEyes argued that its service was like that of Google Books (a service that provides an index of the content of a vast library of books, a service which was found by the Court of Appeals to constitute a fair use), as it was transformative – providing an index of vast amounts of content for research purposes to allow users to know when something was broadcast, not as an entertainment service as the programming was originally intended.

In weighing the various factors, the Court looked at a number of facts to conclude that the TVEyes service was not a fair use. The Court looked at factors including that 10 minutes of programming was provided with each search (not some substantially shorter excerpt), that there were no limit on the number of searches that could be conducted within the same program which could theoretically allow a user to watch complete programs, and that the content owners could potentially themselves monetize an index of their own programming to determine that this was not a fair use. The Court distinguished this case from that of Google Books where far less content was provided in any search, and where there were limits on how much content from a single book could be searched to preclude users from reading meaningful portions of any book through the service.

While most media companies are not likely to try to replicate a service like that provided by TVEyes, the case demonstrates just how difficult it is to conclude that any particular use is a “fair” one that would allow the use of copyrighted content without the permission of the copyright owner. Thus, as set forth above, in connection with the more mundane, day-to-day uses of content on the website of a broadcaster or other media company, in most cases the best advice is to get permission for the use of content before posting it to your site.