Commons:Copyleft trolling

"Copyleft trolling" is the use of a free license to encourage people to use one's work, only to demand money from anyone who fails to follow all the terms of the license without providing them with an opportunity to fix the error. Wikimedia Commons, which relies on freely licensed content, opposes practices that undermine the spirit of open use and sharing. However, the point at which copyright enforcement turns into copyleft trolling is not always clear. Common themes of copyleft trolling include refusing to allow violators to fix their errors rather than pay a fee, indifference to the identity or size of the violator, using an overly complicated attribution line to increase the likelihood of mistakes, demanding money for both minor and major violations, and using old Creative Commons licenses that make it easier to make a business out of enforcement. Working with companies such as Pixsy, which make it easy to pursue money for alleged infringements, may lead even well-meaning contributors to engage in copyleft trolling. Wikimedia Commons typically takes a case-by-case approach to dealing with these issues, and in extreme cases Commons has deleted or forcibly watermarked all of a user's files. This guideline explains copyleft trolling and associated concepts, provides some tips for differentiating it from other forms of copyright enforcement, and outlines the ways the Commons community has intervened in the past.

What is copyleft trolling

edit

When a copyright owner uses a Creative Commons license, they are communicating a desire for others to use their work. Copyleft trolls employ a Creative Commons license to exploit the "you can use this!" signal to take legal action against those who commit minor violations of licensing requirements in reusing the troll's copyrighted work. Copyleft trolls will often target independent or small-time reusers who cannot afford to mount a legal defense, even if the offense is trivial and even if the violation is rapidly corrected. The intent is not to actually go to court, but to intimidate the other party into a cash settlement.

The term "copyleft troll" comes from "copyright troll", companies who used automated processes to find copyrighted works online and issue legal threats and demands at scale. As an example, the website Ars Technica followed the firm Prenda's copyright trolling in detail. The best-known company associated with copyleft trolling is Pixsy, which similarly uses automated systems to search for violations and provides an interface that simplifies demanding money from people who misuse freely licensed images.

Freely licensed content is a cornerstone of Wikimedia projects. Wikimedia Commons wants people to use our content in a variety of ways, and to be confident that they will not be targeted for good faith mistakes. A business model based on suing independent media users for trivial license violations is not in line with the values of Wikimedia Commons. For example, if a personal blogger, social media user, or local charity finds an image on Wikipedia and uses it, crediting the author and citing Wikimedia Commons, but fails to link to the license, that is technically a violation of standard Creative Commons Attribution licenses. It is not, however, the sort of violation we would expect Wikimedia contributors to take legal action over, especially if the reuser offers to fix the error. Creative Commons itself published a blog post about copyleft trolling in 2022, saying "Put simply, 'license-enforcement-as-business model' is a perversion of the founding ideals of Creative Commons. We condemn this behavior."

edit

Wikimedia Commons is not solely a repository of public domain content free from all licensing restrictions. We host media with a wide range of free licenses and therefore much of our content has legal restrictions on use. In other words, media owners who upload here and use a license like CC-BY have the ability to take action against violators. If they didn't, the license would be functionally public domain. At what point enforcement begins to look like copyleft trolling will require judgment on a case-by-case basis, but will typically involve some combination of the following:

  • Working with a company that uses automated processes to find violations and issue letters asking for money.
  • Does not differentiate between independent/small-time reusers and large corporations, except perhaps in the amount of money requested.
  • Does not differentiate between audience sizes in uses, treating a personal blog with a handful of readers as equivalent to a popular national television program.
  • Does not differentiate between types of offenses, such that failing to attribute altogether is treated as equivalent to forgetting a link to the license.
  • Does not provide an opportunity to correct the error. May reduce the amount requested for a correction, but will demand money nonetheless.
  • Once the error is corrected, license is not restored and additional moneys may be demanded (see license considerations).

According to Creative Commons, there are three principles which should guide enforcement of such licenses: "the primary goal of license enforcement should be getting reusers to comply with the license", "legal action should be taken sparingly", and "enforcement may involve monetary compensation, but should not be a business model". Importantly, however, these principles communicate the values of the organization behind the licenses but do not themselves have legal bearing; only the licenses themselves do.

Working with firms like Pixsy increases the likelihood that enforcement will veer closer to copyleft trolling. Unfortunately there is not much middle-ground between copyright owners searching and handling violations themselves and contracting with unethical firms. From the perspective of media reusers, there is little difference between well-meaning contributors who follow such a company's lead and bad faith contributors who upload with the intention of copyleft trolling.

License considerations

edit

Creative Commons licenses are periodically updated. One of the major differences from 3.0 to 4.0 is the addition of a remedy clause. In older versions of the license, once you violate the license the agreement is void -- you are no longer allowed to use the image, even if you fix the attribution later. This provides additional opportunities to copyleft trolls to seek damages at multiple points over time, even after the reuser has paid an initial sum. Version 4.0 doesn't prohibit copyright owners from seeking money from people for violations of the license, but does restore the license once the mistake has been corrected. In other words, upgrading licenses to version 4.0 does not prevent copyleft trolling but does make it a little less lucrative. It also prevents users from seeking damages with Pixsy, which will not pursue cases involving CC 4.0 licenses.

How Pixsy works

edit

Pixsy is an "image theft protection" service. Its clients provide photos which Pixsy then uses to search for uses across the internet, organizing those violations into a number of categories (country the site is hosted in, number of images used, whether the image is still online or not, commercial uses, etc.). The copyright owners then browse the uses to look for violations. Like any reverse image search, there are false positives and the burden is on the user to differentiate the images. As the service was intended primarily for copyright enforcement, and not necessarily copyleft enforcement, users are expected to verify that the terms of their free license were violated. Upon finding a violation and deciding to take action, the user has two options: pursue damages or issue a takedown notice. The latter is disincentivized -- users get a limited number of notices as part of their membership and then must pay a fee to use the feature. All it takes to pursue damages is clicking "pursue", at which point Pixsy takes over. If/when the media reuser pays a fee, Pixsy and the copyright owner split it.

  • Pixsy says it only pursues commercial users, and does not seek damages from uses on personal blogs, personal social media sites, etc., although these users can be issued takedown requests.
  • Pixsy does not issue demand letters without the copyright owner initiating the process.
  • Once Pixsy takes over the case, the role of the copyright owner is minimal.
  • Pixsy does not accept images licensed with Creative Commons version 4.0 licenses.

How Commons handles copyleft trolls

edit

Commons does not have any bright-line rule which attempts to restrict the ways copyright owners can enforce their licenses and does not attempt to prohibit users from working with companies like Pixsy. We have, however, blocked users and deleted or modified files on a case-by-case basis.

Verification

edit

Commons has limited interventions at its disposal to handle copyleft trolling, and they are typically severe remedies like deletion, blocking, or watermarking. Before seeking such remedies, it's important to verify the claims of copyleft trolling. For example, if we learn that an individual reuser received a letter from a law firm demanding money, take steps to verify the claims: was there a license violation, was there any attempt to correct it, does the law firm represent the copyright owner, have there been other complaints about the same user in the past? Some of these questions will not be answerable; the important thing is due diligence to try to substantiate the claims. Remember there have been cases when third parties like Getty Images have made monetary demands of reusers for images they didn't even have rights to. Assuming good faith is an active guideline on Commons.

Deletion

edit

A case which brought the phenomenon to the Commons community's attention was Marco Verch in 2018. Discussion of demands for money began on the German Wikipedia and led to a consensus to delete all of his contributions on Commons. Computer Weekly published an article about his practice in 2020, explaining that Verch hired low-cost contractors to create large quantities of images which he published around the internet with a Creative Commons license, then monitored use of the images to sue the users.

Forced watermarking

edit
 
A photo by Larry Philpot with forced watermark

Another early case was that of Larry Philpot, who imposed very specific licensing requirements on his photos and regularly sued over violations. He was first discussed on the administrators' noticeboard in June 2019 and his "racket" was the subject of a Bloomberg article in June 2021. A compromise was found at the deletion discussion whereby all of his photos on Commons would be modified to include a line about the credit, accompanied by the line "Larry Philpot sued users of his work for minor attribution errors. Keep the attribution intact."

No standard practice

edit

While the community has taken action in some cases, in others there has not been consensus to do anything. A 2022 blog post by Cory Doctorow highlighted photos by Nenad Stojkovic, who had a ban proposed and supported, but did not result in any action. In March 2024 a user raised concerns about Pixsy demanding money on behalf of a prolific contributor of high-quality photos. It led to a deletion request which resulted in the images being kept and a sprawling discussion of possible solutions which resulted in this guideline.

Additional steps Commons could take

edit

Commons has an obligation to protect reusers from being subject to unfair litigation for good faith efforts to use our content, but there is disagreement about how best to do so. Below are some of the options which have been proposed or discussed in the past.

Deprecate older Creative Commons licenses

edit

We could deprecate older Creative Commons licenses. Per the section on licenses above, this would not prevent copyleft trolling but could limit some of the harms in some cases and would preclude the involvement of Pixsy. As much of our media are not uploaded by the copyright owners and as a large amount of media are owned by users no longer active on Commons, this would be difficult to enforce unless it applied only to new uploads of media created after the introduction of v4.0.

Improve visibility of license terms

edit

It is easy for someone to click an image on Wikipedia and save the file from the Commons page without scrolling down to understand the licensing requirements. The interface could change to include a pop up or a large button above the image providing the necessary information. The forced watermarking intervention above is fundamentally due to licensing requirements being hidden on the file details page, below the image and out of view.

Follow the Flickr model

edit

In 2023, Flickr modified their community guidelines to add "Failure to allow a good faith reuser the opportunity to correct errors is against the intent of the license and not in line with the values of our community, and can result in your account being removed."

Commons could require uploaders to agree to something similar to Flickr's requirement. We could also adopt other statements or principles like the Creative Commons enforcement principles. These measures would only be apparent to uploaders who are also media owners, as media owners whose work was uploaded by other users would have no idea what principles apply, but they could set limits for whose files we are willing to host regardless of whether they are a Commons user.

Implications

edit

Several users are uneasy curbing the rights of media owners to enforce licenses how they see fit. In one case, a professional photographer explained that his photos are used widely by major media organizations, and that while he shares some photos on Commons for anyone to use with attribution, he charges a fee for unattributed uses such as for use in advertisements. But there is little, if any, opposition among the community for enforcing licenses when it comes to major corporations using unattributed images in advertisements. It becomes copyleft trolling when enforcement becomes indiscriminate, making a business model out of extracting money from any minor violation, regardless of whether it's fixed, and not just companies who failed to attribute altogether.

Victim of copyleft trolling?

edit

This page is primarily concerned with steps Wikimedia Commons can take to prevent or minimize the harms of copyleft trolling. If you have received a letter demanding money for a license violation:

  • Read the letter carefully to understand the specifics of the claim.
  • Verify the claim. Does the person/company asking for money actually own the rights or represent the rights holder? Search for information about the entity who sent you the letter to try to verify it isn't a scam.
  • Fix the problem immediately. That might mean removing the image (as described above, with older Creative Commons licenses if you continue to use the image after learning about the violation, you could be sued again). It could also mean fixing the way you credit the rights owner.
  • Keep records of all your interactions.
  • If you decide to pay a fee, you can often negotiate the amount demanded, which may involve explaining your situation or intentions, offering to correct the use, issuing an apology, or some other means.
  • For legal advice, consult a lawyer or seek legal aid if available.

Unfortunately, we can only provide these best practices and cannot provide legal advice on Commons. Here is what the Commons community can help with, however:

  • If you suspect a license may have changed since you initially used the media, we can help you to look through the page history to look for changes since it was initially uploaded.
  • If you suspect copyleft trolling, and not just a single enforcement action, you can share your experience in a venue like the village pump, bringing the behavior to wider attention. Such reports may not help your case, but form the basis for the community taking action against those who deploy unethical practices, thus helping to ensure others don't fall into the same trap.

Resources

edit
  • Copyright Aid - forum with information for those accused of violating copyright in the United Kingdom

See also

edit

Further reading

edit