Mashable had sought to use one of her images in an article they were working on. However, when she turned down their licensing offer, they embedded the photo directly from her Instagram page instead.
The court, however, found in favor of Mashable and dismissed the case. The reason, the court ruled, was that Instagram’s terms of service granted it the right to give sub-licenses and that it had done so through the API.
This result caused shock waves through the internet with many wondering what this meant for publishing on Instagram. Given that Instagram offers no way to disable embedding without making an account private, many artists looked at this as a devil’s choice between sacrificing some of their rights and avoiding the most popular image-sharing service.
However, now the tide may be turning again.
Last week, a similar case that was also in New York, pits photographer Elliot McGucken against Newsweek. In an eerily familiar story, Newsweek had embedded one of his photos from Instagram onto their site after being denied a license, prompting McGucken to file a lawsuit.
Here, however, the judge found very differently, declining to dismiss the case on these grounds. Though that is not necessarily a ruling in favor of the plaintiff (it merely pushes the case closer to a potential trial), the judge’s skepticism about these arguments points to Newsweek either being forced to settle or put forth a new legal theory.
To that end, Newsweek suffered yet another setback in this case with Instagram releasing a statement that says:
While our terms allow us to grant a sub-licence, we do not grant one for our embeds API. Our platform policies require third parties to have the necessary rights from applicable rights holders.
That statement not only calls into question Newsweek’s defense, but the dismissal in the Mashable case.
However, it also raises a serious question: What is the future of embedding if the legality of it is so uncertain?
The answer, most likely, is not good news for the practice.
A Brief History of Embedding
Though we think about embedding as something that one primarily does with Instagram, Twitter and YouTube, the practice mostly got its start with images.
Photo sharing sites, such as the original Photobucket, made it easy for users to upload images and embed them on other sites. This was especially useful for people who were posting on forums that did not allow image uploads.
However, during the “Web 2.0” days of the internet, embedding became a bigger topic. At the end of the 2000s there was a virtual embedding craze with companies even launching products that aimed to make all content embeddable. Embedding was seen as a powerful tool to avoid misuse of your content by encouraging legitimate sharing that creators can still control.
Much of this gold rush was spurned on by a May 2007 ruling in the case of Perfect 10 v. Amazon, which dealt with infringing pornographic images appearing in image search results. The ruling, which was from the 9th Circuit Court of Appeals, focused on the servers the images were stored on, helping the search engines avoid liability for those infringing images appearing on their site.
This created what is known as the “server test” and it helped make embedding a very attractive solution to both easily adding content to a page, but also avoiding potential legal issues that might stem from it.
However, the New York courts are in the Second Circuit and are not bound by the server test ruling. In February 2018, a New York court rejected the server test outright in a case that dealt with an embedded tweet. That case was settled before it reached the Second Circuit Court of Appeals.
The decision of the court was directly inspired by a 2014 Supreme Court case looking at the online TV streaming service Aereo. Though the case didn’t deal with embedding directly, it looked at the technical process that Aereo used to capture and retransmit over-the-air television to its customers and found that the question of infringement “Should not hinge on invisible, technical processes imperceptible to the viewer.”
The court in the Newsweek case clearly took a similar approach and, with Instagram denying the defense in the Mashable case, it’s unclear what long-term impacts that decision will have.
Simply put, embedding is losing its luster and may, soon enough, lose its place altogether.
Changing the Way We Communicate
To be clear, the server test is far from a perfect doctrine. While it enboldened embedding, it also opened doors for pirate sites to “embed” infringing content from third parties and attempt to skirt liability. The courts are shifting their focus from how something is done to what was actually done. The is, on the whole, a good thing.
But with these recent cases, we are not talking about pirated content uploaded without permission and embedded recklessly or maliciously by news sites. We’re talking about images that were uploaded by their creators and then embedded on news websites using an established API.
In both of these cases, an implied license doesn’t really apply. Both Newsweek and Mashable had asked for a license but were denied. As such, any implied license granted by making the image embeddable was likely stripped away by the direct refusal to allow the image to be used. After all, an implied license only exists insofar as there is no direct indication of intent.
But what if the sites just hadn’t asked and instead embedded without first trying to license the work? Even with that, there might be issues. That’s because courts attempt to read such implied licenses as narrowly as possible and Instagram does not make it easy for users to either opt-in or opt-out of embedding. The only way to stop embedding is to make your account private, which greatly restricts your visibility.
YouTube, for example, makes it easy to enable embedding on a per-video basis. Twitter, however, follows Instagram’s model and only restricts embedding on protected accounts.
The more action required by the user and the more control they have over the embedding process, the stronger such an argument would be. Sadly, Instagram doesn’t give users a lot of control here and, by denying that they grant an implied sublicense through the API, make it so that one has to get explicit permission before embedding an image.
While this may well be a boon for artists and photographers, it may be the last straw for the practice of embedding.
Embedding was already on the way to becoming a legal minefield. If the server test is completely set aside, embedding will provide no legal protections should the content turn out to be infringing (as with the 2018 Twitter case) and for content that was legally-uploaded, one still needs to ask permission before embedding. At that point, it’s easier and more attractive to obtain the needed permissions and host the content directly.
If embedding can’t provide any legal assurances even when the content was legally-uploaded, it will likely further fade from consciousness. The end goal of embedding, making it easier to share content while allowing creators a measure of control over their work, has not been realized.
To that end, it’s the social media companies that deserve the most blame. They pushed for embedding to be as broadly accessible as possible, making it difficult or impossible for artists to choose what approach worked best for them. By removing that choice, they made the legality of it much more questionable.
And now those questions are going to keep people away from embedding as a practice and may ultimately limit it to rare circumstances.
Ultimately, Instagram has no one to blame here but themselves. By denying users the ability to control embeds and making user consent in this area unclear, they created this uncertainty. Combined with the ongoing dispute over the server test and there has never been a point in history where it is riskier to embed content.
There are ways social media companies could address this issue. They could make it easier to opt-out of embedding, or even better, require users to opt-in to it. But these steps would greatly reduce the effectiveness of embedding, which has largely served the goal of the social networks to keep people on their site (even when they’re reading a different one).
Right now, it looks bleak for the future of embedding as a practice. It went from the wild west days of 2010 to being in a legal retreat just a decade later.
Embedding as a social and legal norm may very well be going away in the coming years. It’s either that or it will undergo some drastic changes.
Either way, the current course is clearly not tenable and a new approach is badly needed.