It’s worth remembering that after the many hearings hauling Silicon Valley’s titans before Congress to address disinformation on social media, Congress has done nothing.  Even the most modest intervention is stalled.  This is the Honest Ads Act, introduced in the Senate in October 2017.  I wrote about it last year with Lyndsey Wajert. It would hold social media and other online platforms to the same political advertising transparency requirements that bind cable and broadcast systems. No longer would these platforms be able to deny that they know they are distributing foreign and other covert political ads (as part of a disinformation campaign), and no longer would the public be blind to advertiser identity. Above a certain threshold, the buyers of these ads would be exposed. However, paid political advertising is a relatively small part of the disinformation that infected the public sphere during the U.S. 2016 federal election. The Honest Ads Act will not reach organic social media posts, troll farms or — most fundamentally — the logic of social media sites which rewards outrage and bias confirmation. Moreover, it would not reach paid ads that are not considered “political”, and there is a risk that an overly broad definition would raise First Amendment problems. Even given these limits and caveats, even relatively weak disclosure regulations have proven to foster a culture of greater transparency in commercial advertising. They could do the same with online political advertising, without compromising free speech values. Normalizing social media to the standards of traditional media transparency is a reasonable place to start.

WINNER OF THE THIRD ANNUAL LASTOWKA SHORT-FORM WRITING COMPETITION

By Matthew Yost (Rutgers Law Student)

In July of 2016, Justin Goldman took a photograph of New England Patriots quarterback Tom Brady alongside Boston Celtics general manager Danny Ainge and posted the photo to Snapchat. Goldman’s photo went viral, working its way from Snapchat to Reddit to Twitter. A tweet displaying the photo wound up embedded on various news websites owned and operated by media outlets such as Breitbart News, the Boston Globe, Vox Media, and Yahoo.

Embedding a tweet allows one to bring content from a Twitter post to another website. Each tweet contains a code provided by Twitter in a drop-down menu. The code is then copied and pasted into the website’s code, where it is displayed as embedded into the site’s content.

Goldman sued the news outlets that embedded the tweet, claiming violations of his right to display his photo. Under the Copyright Act, to display a work publicly means to “show a copy of it….” The parties agreed to split litigation into separate phases – first, to determine whether embedding a tweet violates a copyright owner’s exclusive right to display; and second, to determine whether any defenses apply to the infringement.

The defendants relied on the so-called “Server Test” developed in the Ninth Circuit. The Server Test developed from Perfect 10 v. Google, Inc., which addressed the questions of, first, whether thumbnail images that appeared in Google search results violated display rights and, second, whether full size images that appeared after a user clicked on the thumbnail also violated the right to display. The Perfect 10 court came out differently for each question. First, the court determined that because the thumbnails were stored on Google’s server, there was infringement. However, because the full-size images were stored on third-party servers and essentially embedded into Google’s webpage, there was no infringement on the second issue. Accordingly, liability ultimately turned on whether the defendant hosted the allegedly infringing image on its own server or on a third party’s server.

The Perfect 10 ruling was affirmed by the Ninth Circuit on appeal, where the court also determined Google was likely to succeed on a fair use defense.

The Goldman defendants’ argument was simple: by embedding a tweet, the media websites claimed they simply provided a guide for the user to access to copyrighted content. In this sense, embedding a tweet is merely the same as using the Dewey Decimal System to locate a book in the library. The library’s card catalog does not infringe on a display right, the analogy dictates.

However, the Goldman court did not find the defendants’ arguments, nor Perfect 10, persuasive. The court held that Perfect 10 was out of circuit, had been repeatedly rejected by other courts within the Second Circuit, and was cabined to pertain solely to search engines. Further, the court correctly noted the flaw in the defendants’ position. Embedding a tweet displays the tweet: it actually provides the copyrighted content. The card catalog does not magically present a user with the book.

Although many quickly criticized the decision, the court was not wrong. As the opinion pointed out, there is no meaningful distinction between displaying an image from your own server or displaying an image from someone else’s. This may be relevant for the exclusive right to make a copy a work, but that presents an entirely different question from the right to display the work.

However, hope is not lost, and the internet is not necessarily going to be radically transformed. The court merely determined that embedding tweets violates the display right. It has yet to address the issue of whether some admittedly strong defenses will apply.

First, the court has conceded that there appears to be a “very serious and strong fair use defense.” Such a defense would allow media outlets and social media users to continue to embed tweets into their content without fear of liability. Second, the court also recognized that defenses under the DMCA may apply. Further, the court left open the possibility of limiting damages based on innocent infringement.

There is also the possibility that the court could find implied consent to the displaying of embedded tweets. By default, Twitter allows users to retweet content posted by other users. This is common knowledge to even novice users of the platform. It is plausible then that a court could find one posting to Twitter implicitly recognizes that their tweet could be retweeted, which is essentially embedding the tweet into a new tweet.

Twitter also offers one the ability to “lock” his or her tweets and prevent them from being retweeted. Locking a tweet does not necessarily protect it against being embedded into third party websites, but it is one step in showing that the artist did not mean for their post to become viral. Facebook offers similar technology with its “share function.”

Other social media, such as Snapchat – where Goldman originally shared his photo – and Instagram do not have share or repost functions built in. But this could be where the court effectively uses fair use and other available defenses.