Architect Sues Pinterest Over Scraping

Last week, Estonian architecture firm Concept Engineering LLC filed a lawsuit in a Delaware federal court claiming that the image-sharing site Pinterest violated their copyright.

According to the claim (PDF), Concept is a firm that markets and licenses its architectural works worldwide. As part of their marketing efforts, they operate a website featuring numerous images of their work.

They claim that Pinterest users have been “pinning” their images via URLs. This has resulted in Pinterest scraping the content from those pages, grabbing their images and reuploading them to Pinterest’s servers without permission.

Concept alleges that it has taken several steps to prevent the misuse of its content. This includes using a no-right-click script to prevent image saving and utilizing reCAPTCHA to verify that visitors are human. They further allege that they use a specific file naming structure and that those names are changed.

The lawsuit seeks unspecified damages and injunctions preventing further violations. Pinterest has not responded to the lawsuit.

This leads to the question: Is this lawsuit likely to succeed? Although this case raises some interesting questions, it has several significant hurdles to overcome. However, there’s a much simpler solution to the perceived problem.

Understanding the Case

The basics of the case are relatively straightforward. Concept Engineering is an architecture firm that operates a website that hosts many of its images. It alleges that Pinterest, through the use of its “Save ideas from website” tool, scrapes those images unlawfully and republishes them on Pinterest’s site.

All totaled, the original claim makes four allegations:

1. Direct Copyright Infringement: Alleges Pinterest makes unauthorized copies that are displayed and distributed on the Pinterest site.
2. Circumvention of Copyright Protection Systems: Claims that Pinterest bypassed various copyright protection systems, including the anti-right-click script and reCAPTCHA.
3. Removal of Copyright Management Information: Claims that Pinterest alters the file names and removes other metadata, amounting to the removal of copyright management information (CMI).
4. Contributory Copyright Infringement: Alleges that Pinterest has enabled and aided third parties, namely its users, in committing infringements.

The first and fourth questions could raise some interesting points. Pinterest has been a source of controversy for visual artists for over a decade. Their systems do grab images from third-party sites and then republish them, full-sized, on Pinterest’s servers.

This separates Pinterest from services like Google Image Search, which only host thumbnails. However, even if it did mirror it, the server test, which Google Image Search established, is on shaky ground. Since this case is in the Third Circuit, it’s unclear how the court would apply the test.

Pinterest could also run afoul of the Grokster ruling, which found that creating a device to “induce” copyright infringement makes one liable for third-party infringement. That wouldn’t significantly help with the first point, but could help with the fourth, if it’s agreed that Pinterest’s users infringed.

However, the third point is incredibly weak. The argument that the file name amounts to CMI is, to my mind, a stretch. CMI must help in identifying the copyright holder. They use a file name system such as CH### to name their works. While that may help them search for their own content, it doesn’t help a third party find them.

Although the lawsuit mentions the removal of other metadata, it is unclear what metadata they are referring to. Neither the claim nor the exhibit references EXIF or similar metadata. The filename alone is unlikely to qualify as CMI, especially given the format’s vagueness.

The second point, however, is probably the weakest. Despite what the lawsuit claims, Pinterest’s bot didn’t circumvent anything.

First, the no right-click script only works for users with JavaScript. Bots do not typically access sites in this manner. They read the HTML and look for the relevant content. They usually ignore scripts.

The same is true for reCAPTCHA. Though the tool is designed to block “bad” bots, it likely doesn’t consider Pinterest’s bot to be one it needs to block. Since reCAPTCHA is unlikely to attempt to block Pinterest, Pinterest never bypassed it.

So, while there are interesting questions, this lawsuit was, in reality, not necessary. There is a much easier solution to this problem.

A Better Solution

To be clear, neither a no-right-click script nor reCAPTCHA was going to block Pinterest’s bot. The no-right-click script was outdated as a form of defense 15 years ago. ReCAPTCHA, though more modern, isn’t designed to solve this kind of issue.

However, there is a much simpler solution to all this. In fact, one of Concept’s exhibits references the page on which it appears. In the Pinterest help center, you can find a page about Pinterestbot. That page gives simple instructions for blocking the bot via robots.txt.

Robots.txt has been an industry standard for over 30 years. While bad bots do not follow it, Pinterestbot does. As a result, with a few lines of code, Concept could have blocked Pinterest from accessing its server. Currently, their robots.txt file allows all user-agents and only restricts access to two directories.

They also had other tools and approaches. They could have blocked Pinterest’s IP addresses, prevented direct access to their images, used watermarks or taken almost any other step to prevent Pinterest from accessing their pages.

If that didn’t work, they could have also filed DMCA notices against Pinterest for the infringing images. They might have even been able to access Pinterest’s Content Claiming Portal to automatically remove works in the future.

Instead, they relied on an outdated and easily bypassed approach that bots don’t even see and another tool not designed for that purpose.

It’s no shock that it didn’t work.

Bottom Line

When Pinterest became controversial in the early 2010s, numerous legal questions were raised about it. However, we never received satisfactory answers to those questions. The reason is that lawsuits didn’t follow the uproar.

Most decided that Pinterest either provided more benefits than harm or, at the very least, it wasn’t worth worrying about. Others decided that it was easier to block or remove themselves from the site. It never turned into a copyright battle.

That gives this story a very early 2010s feel for me. That isn’t helped by Concept’s reliance on the no-right-click script to prevent image theft. Everything about this case feels like a blast from the past.

Still, it feels as if Concept was either unaware or ignoring much easier ways to resolve this. A few lines of code could have kept Pinterestbot away from their site for good. However, instead, they opted to file a lawsuit.

Needless to say, this will be a case to watch, if nothing else, to see which issues the case ultimately addresses.