Back in September, the EU approved a first draft of The Directive on Copyright in the Digital Single Market, which is somehow even more boring to read than you’d think. Its goal is to set standards for how European nations enforce copyright laws on the internet. Through thorough definition of terms and outlining of procedures, it establishes a basic structure for later enforcement legislation to be passed by member states. Admittedly, it doesn’t seem too unusual to catch a regulatory body writing regulations, particularly in an objectively boring fashion. However, just because the wording is obtuse, that doesn’t mean the directive’s implications are anything less than incendiary. This hot piece of jargon and legalese contains two items in particular which have kicked up quite a stir among the online community: Articles 11 and 13. It’s pretty clear upon inspection that even though these regulations appear to be inspired by legitimately benevolent aims, their execution is so flawed that they are poised to do more harm than good if and when they are actually implemented.
Before laying into the directive too harshly, it’s worth noting the noble goals behind the legislation, starting with Article 11. Many have taken to calling this provision the “link tax,” because it proposes charging sites for references to original content hosted on other sites. Ostensibly, this is intended to incentivize the production of original content without fear of others disproportionately profiting off that work. I actually think promotion of original content is an incredibly important issue, in a time of rampant fake news and misinformation. The journalists, commentators, and activists who generate such works perform a vital function in the information ecosystem. They form the basis of digital media by serving as the raw material producers, whose works get remixed, reworked, and repackaged into new forms for consumption. Without them, the internet would be a hell of a lot less interesting and exponentially less useful. Considering the number of content aggregators out there that effectively plagiarize entire articles, videos, and songs, it must certainly be frustrating for creators to see their work gain orders of magnitude more attention when it’s posted by someone else, especially when there is little to no credit given to the original author. By attempting to install a kind of royalty for this sort of use, the European Parliament aims to return the benefits of this use to the original site.
Adding onto this theme of protecting creative rights for original producers is Article 13, often referred to as the “upload filter.” Under this proposal, websites would be liable for hosting any content in violation of copyright law. Thus, sites would have a financial and legal incentive to ensure any and all user uploaded content falls well within the legal limits of use. The most practical way to implement this would be to have an automated filter which scrubs user uploads for copyrighted works and blocks any components in violation of the relevant copyright laws. Again, the aim here is to extend the same level of protection for intellectual property which has long existed in the physical world to the digital world. As information and content becomes increasingly digitized, this is a reasonable goal to promote creation and innovation. At least, it appears reasonable.
The major flaw with these articles is that they protect some types of original content while sandbagging others. Unless the item in question is essentially created in a vacuum with no reference to, or element from any other work, it is likely to be taxed or blocked. This is because the penalties for hosting works in violation of these restrictions creates a major financial incentive for companies to be extremely conservative in what they allow on their sites. It’s simply impractical for a site to have humans screen all user posts, so they will inevitably turn to automated programs, which are so far from perfect as to threaten practically all content passing through. Copyright filters, such as those in place on YouTube, have been repeatedly shown to miss large swaths of straight-up pirated content, which isn’t even hidden. I went onto YouTube just now, searched “full movie,” and got crystal clear versions of The Lion King, Olympus has Fallen, Moana, and The Life of Brian, just to name a few. I didn’t even have to try to find these, so one can only imagine how much more missed content there is, which takes a little more digging to uncover. In fact, there’s a whole subreddit dedicated design to finding pirated movies on YouTube: r/ fullmoviesonyoutube.
Not only do these filters miss illegal reproductions, they often ensnare troves of legitimate content as well. Either due to their inclusion of content under fair use or simply a flaw in the system, this creative content is routinely blocked by inept filters, threatening the creators’ freedom of expression and livelihood. Sometimes, these filters even tag videos for issues completely unrelated to copyright. For example, last year, it took a massive public outcry for YouTube administrators to adjust the site’s “restricted content” filter to stop classifying some LGBTQ+ content as inappropriate for viewers under 18. With flaws like this, it’s clear that automated filters pose a serious risk of infringing on legitimate content made by well-meaning creators. If implemented, Article 13 would result in this injustice being repeated en masse for content hosted in Europe. Although the directive claims a specific exception for parodies, there is no practical way to enforce this. Even if there was, there are many forms of fair use besides parodies which would be more likely to be excluded, such as remixes of songs and visual art which incorporates copyrighted material in an original manner. Simply put, this article protects a narrow range of content at the expense of most other types of creative content.
Sadly, Article 11 would only stand to extend this issue by applying a similar logic to online articles. The wording of the article is incredibly vague as to what qualifies as a taxable use of another source. It claims that a simple hyperlink with a few words would not be taxed, but the exact meaning of the line is nebulous at best. Because the distinction between nontaxable and taxable use is so blurry, one can assume that sites will simply err on the side of caution and restrict their use of other sites to an absolute minimum. Small creators, in particular, would be affected by these taxes since they may not have the capital necessary to make these relevant references. Larger sites would thus have an unfair advantage over newer, smaller competitors, which inhibits online growth. In short, such a practice would reduce the quality of online content and run counter to the principle that the internet should be a system of information which is as interconnected as possible. Sites would stand to become more isolated and less useful as a result, an outcome which no one in their right mind would want.
When these trickle down effects are considered collectively, it’s clear that this directive stands to create a new problem which is worse than the one it attempts to solve. While protection of original content is certainly important, the methods outlined here are simply an ineffective way to accomplish this. Legislation like this pops up all the time, due to the simple fact that the people writing these documents and voting on them often lack the technical knowledge necessary to fully understand the problem and the challenges inherent in constructing a solution. I hope that European constituents are able to convince their representatives to reconsider this directive and remove these toxic articles; if they don’t, we all stand to suffer at the hand of this tragically incompetent proposal. As for protecting content creators, my hope is that host sites simply do a better job of responding to infringement claims from their own communities. Day to day users are far more competent than any automated filter at differentiating between legitimate and illegitimate content. Simply doing a better job of engaging with these users would bring large amounts of stolen content to attention for individual review, without accidentally blocking original content in the process. Even a small number of honest users can make a big difference if they are heard by these sites. Essentially, I believe a case by case review basis is the only way to prevent collateral damage in the battle against copyright infringement. Any sort of automated filters like the EU proposes can only serve to hurt digital creators and make us all worse off. We can only hope that legislators come around to this realization as well before it’s too late.