Posted by: Jamie Winterton
Who owns a tweet? These tiny shreds of content – once 180 characters, now 260, sometimes including a photo – were originally described by Twitter founder Jack Dorsey as “short burst[s] of inconsequential information”. 12 years after Twitter’s founding, however, tweets mobilize social movements, provide a means for political organization, and occasionally perform as a role in national policy. As with any prominent technology, Twitter has played a role in a number of interesting legal battles involving libel, trademark infringement, emotional distress and, of course, copyright. The latest, Goldman v. Breitbart News et. al., is a skirmish over whether or not linking to a tweet (with a photo) counts as copyright infringement. The ruling has the potential to significantly alter online publishing, but perhaps also to provide some long-sought protections for content creators.
Twitter’s slogan is “If it’s happening anywhere, it’s happening on Twitter”. Of course, even if it is happening on Twitter, Twitter doesn’t stand alone in the online information sharing ecosystem. It’s hard to provide backstory or any meaningful analysis in snips of 260 characters – even stringing them together into a “tweetstorm”. To provide this context, online publishers will link to tweets within a story, using the tweets as examples, supporting evidence, or even as just a nice way to break up what would otherwise be a monolith of text. Inline linking is simple: for example, here’s an example from www.w3schools.com, an online tutorial for web developers.
What the web developer writes is:
And what the person visiting the website sees is:
The image in this case (“smiley.gif”) is locally stored, but a link can be created between the webpage and any publicly accessible online content. This is how much of the internet works, in fact – it’s an efficient way to share information without having to download and store it yourself.
Twitter makes it pretty easy to embed a tweet. Here’s a screen shot of a tweet I wrote for this purpose:
(Let’s face it – the copyright discussion hasn’t been that cute since Lenz v. Universal.)
With two clicks, Twitter will create the code to embed my tweet into a webpage:
Note that I didn’t have to write that code by hand – the Twitter interface will do it for me (or anyone else who has access to the tweet).
So any webpage – cyberbeartracks.com, perhaps – could include the code above, and my tweet (and adorable cat) would appear to the website visitor, as long as I don’t delete the tweet or lock my account.
In 2016, photographer Justin Goldman took a photo of Patriots quarterback Tom Brady with Celtics manager Danny Ainge, and posted it to his Snapchat account. The photo went viral, with the photo appearing on Twitter and several news outlets linking it in the manner shown above. Why? The Celtics were in talks to recruit superstar Kevin Durant, and it was widely speculated that Ainge might be leveraging Brady to help seal the deal. Goldman sued a number of these news outlets, including Time, Breitbart, and the Boston Globe, claiming that they had infringed his right to exclusively display his content.
The defendants in this case requested the Court apply the “server test” – a test established in the Perfect 10 v Amazon and Perfect 10 v Google cases that examines where the content actually “lives”. The Perfect 10 cases are copyright law classics – some of the first cases to explore what legal concepts like infringement and transformation mean in the online environment. Perfect 10 is a website that provides full-resolution photos of hot naked women to its subscribers, as well as a full docket to the LA court system. (The company’s owner, Norman Zada, has said that he spends “eight hours per day, 365 days per year” on Perfect 10 litigation, but he spends 40 to 50 hours per year-about a week-attempting to “create new artistic content.”) Google, of course, provides thumbnail images and links to content in its search capability. Perfect 10 sued, saying that Google had infringed its content by doing so. The district and 9th Circuit courts decided that Google’s linking was not infringement, since it was “highly transformative”.
In the Perfect 10 case, the courts decided that Google would only be liable for infringement if it were to host the images it displayed on its own servers. So it seems, based on this famous and well-cited precedent, that the defendants in this case were well positioned. Many were surprised, then, when the judge in this case ruled in favor of the plaintiff. “This Court concludes, for the reasons discussed below, that when defendants caused the embedded Tweets to appear on their websites, their actions violated plaintiff’s exclusive display right; the fact that the image was hosted on a server owned and operated by an unrelated third party (Twitter) does not shield them from this result.” (emphasis added)
Judge Forrest explains:
“In Perfect 10, Google’s search engine provided a service whereby the user navigated from webpage to webpage, with Google’s assistance. This is manifestly not the same as opening up a favorite blog or website to find a full color image awaiting the user, whether he or she asked for it, looked for it, clicked on it, or not. Both the nature of Google Search Engine, as compared to the defendant websites, and the volitional act taken by users of the services, provide a sharp contrast to the facts at hand.”
In this court’s opinion, it matters less where the content is located than how the content is requested and experienced by the user. In a Google search, the user initiates a request, and results are provided. Few readers of internet news sites open a webpage with the express intent to know if Tom Brady is involved in the Celtics acquisition of Kevin Durant. They may be happy to be informed, but their intention can’t be that specific. And there’s no (easy) way for the user to know where that photograph lives – whether it belongs to Breitbart, Boston Globe, or the photographer. The ownership is obfuscated by the method of display. Judge Forrest questions the congruence of the Server Test and the intent of the Copyright Act, and goes back to the definition of “display” in 17 USC 101: “To “display” a work means to show a copy of it, either directly or by means of a film, slide, television image, or any other device or process” – and she notes in detail that that inline linking, as described by the defendants, constitutes a “process”.
What happens now? The defendants in this case claim that a finding in favor of the plaintiff will have a “tremendous chilling effect on the core functionality of the web”. The court is not as certain. Judge Forrest notes that there are several other questions regarding fair use and licensing that were not addressed in this case. And given the prominence of the defendants, it’s almost certain this case will be appealed.
Who stands to benefit from this ruling? Questions about the ethics, if not the legality, of inline linking have been brewing for years, primarily in activist communities who frequently communicate via Twitter and often have their content leveraged in ways in which they don’t approve. This was a key topic in the Addressing Anti-Feminist Violence Online summit in July 2015, hosted at Arizona State University with funding from the Digital Media and Learning Competition. During the conversation, several women discussed the harassment and threats they received when their online content (primarily on Twitter) was displayed on other platforms without their consent. In another example, a BuzzFeed journalist in 2014 wrote an article about sexual assault. The piece leveraged a tweetstorm by Christina Fox (@steenfox) and her followers about their own experiences. The BuzzFeed journalist got permission of many contributors, but not Fox herself, before the story went live. The tweets were detailed and deeply personal, and even though they were posted in a public conversation, there was an outcry when these tweets were used in a different context. Anil Dash, blogger and advocate for ethical technology, tweeted in response “Billions of people know “visible on the web” doesn’t always equal “I want this published in the media with ads around it”. Respect that.” So while the Goldman v Breitbart et. al. ruling doesn’t provide copyright protection to all tweets and leaves many other questions on the table regarding fair use, this case may cause journalists to be a bit more cognizant of how they treat content, and start to include explicit permission in their journalistic processes to avoid liability.