An EU proposal would require platforms that host content to check all uploads for copyright violations, which could make software more expensive.
- A copyright proposal before the EU would require any platform that hosts a large amount of user-uploaded content to scan all that content for copyright violations.
- The proposal was designed to prevent media pirating but could have serious repercussions for developers who use services like GitHub, which would be forced to filter code under the new law.
A copyright proposal before the European Union aiming to make it harder for pirates to share programs and media is casting a wide net that may have inadvertently snagged programmers.
As GitHub points out, automatic filtering of code would be devastating for both independent and large-organization programmers. Issues faced could include false positives/negatives, the loss of dependencies, license confusion, and unnecessary burdens that slow down innovation.
The copyright proposal is still being discussed, and efforts to stop its implementation have already begun.
What the proposal says and why it’s troubling for coders
Article 13 of the proposal, which specifically concerns the implementation of automatic content filters, is the particular portion troubling GitHub and other European programmers.
Article 13 states that “providers that store and provide to the public access to large amounts of works or other subject-matter uploaded by their users” will be required to work with copyright holders to implement measures that prevent their content from being shared illegally.
“Those measures, such as the use of effective content recognition technologies, shall be appropriate and proportionate,” the proposal states. Article 13 further mentions content recognition technologies as part of best practices for catching copyright violations, making it likely that filters would be a part of the final language of any new copyright law that is passed.
What types of content are to be filtered, however, is left unspecified. Casting a net that wide is what concerns GitHub and members of Save Code Share, a petition started to stop Article 13’s implementation.
Are “censorship machines” coming for EU?
MEP Julia Reda breaks down some additional concerning portions of the copyright proposal that apply to both programming platforms and other content-sharing websites, all of which would be affected under the wide net cast by the proposal:
Platforms would be required to obtain licenses for all the copyrighted content they host, which she argues would be nearly impossible, especially for code-sharing platforms.
Platforms would be required to prevent copyrighted content from being uploaded, which she argues mandates the use of content filters.
Content filters cannot process any personal data, which she says would make filing redress for over-filtering impossible, as uploaded/blocked content would be stripped of identifiers.
Scanned files would be checked against all uploaded content, which Reda believes violates EU prohibitions on general monitoring.
Filtering doesn’t apply to online markets (like eBay), ISPs, research repositories, or cloud storage services. Reda points out that those services aren’t exempt from licensing requirements, leaving them open to lawsuits due to confusing language.
Argument about the copyright proposal is ongoing, so it’s too early to tell if some of the more generalized, harsh, or inconsistent portions of it will become law.
What may be a well-reasoned proposal for ensuring performers, authors, and publishers receive payment for their work isn’t necessarily a good fit for other communities, and anyone with an opinion is encouraged to make their voice heard by reaching out to MEPs, Council Members, and EU Commissioners.