The Europe Union appears so desperate to punish U.S. tech giants that it’s ready to conduct radical surgery on its own online copyright laws.
The E.U. has proposed a new digital copyright directive that would force tech companies to secure rights from copyright holders to allow things like movies, shows, music, and other materials their users might upload to be shared on their sites.
That's a large departure from current copyright laws that require companies to take down copyrighted material when the copyright holders complain. It would also be prohibitively expensive for smaller European social sites, which could be a boon for American tech giants.
It’s not too late for Europe to avoid salting its own earth and ensuring the supremacy of Big Tech, as the European Parliament still must vote on these changes. But having things take such a self-destructive course offers a lesson in how abjectly tech policymaking can fail when only Big Copyright—specifically, large media companies—gets heard.
We’re getting disturbingly close to an internet split across the Atlantic—where only the likes of Facebook and Google can afford to compete on both sides.
A copyright chasm
The most dangerous part of this “Digital Single Market” copyright directive is its Article 13, “Use of protected content by online content sharing service providers,” which starts on page 65 of the text posted Wednesday.
That obligates any such online service—its definition excludes Wikipedia, online markets, messaging services and small startups less than three years old, but not social networks in general—to “obtain an authorization” from copyright holders whose work might be shared by its users.
That amounts to asking companies to license “all copyrighted content in the world,” Julia Reda, a German Pirate Party member of the European Parliament, wrote in a recent post.
Social sites that did not want to be held legally liable for their users’ posts would have to show they’d tried to secure those licenses and had made “best efforts to ensure the unavailability of specific works and other subject matter for which the rightholders have provided the service providers with the relevant and necessary information.”
That’s legalese for deploying automated filters to block the upload of files that match items in databases provided by copyright holders.
That represents an enormous departure from the “notice and take down” regime on both sides of the Atlantic, under which sites avoid liability if they promptly respond to reports of copyright infringement by taking down the content in question before letting the user challenge the takedown.
Does not scale
Few precedents exist for such a comprehensive monitoring regime aside from YouTube’s Content ID. That Google (GOOG, GOOGL) system lets copyright holders choose between having infringing copies of their content blocked or getting a share of the advertising revenue generated from them.
In a November blog post, YouTube CEO Susan Wojcicki wrote that the company paid out €800 million to EU rightsholders in the last year. Google declined to comment on the record, but in a Feb. 7 post, global affairs senior vice president Kent Walker said the proposed rules “aren’t carefully balanced.”
The GIF-sharing site Giphy also opted to license the clips behind its animated GIFs. “We went and started, day one, making licensing deals with all of the major movie producers, TV produces, studios,” founder Alex Chung said at the Collision conference in 2016.
But YouTube benefits from Google’s massive resources, and Giphy deals with a much smaller set of material.
“If anyone says Google can do it, therefore everyone else should—what a nonsensical statement to say,” said Evan Engstrom, executive director of the U.S. startup-advocacy group Engine. A 2017 study he co-wrote found that “the cost of filtering technologies far exceeds their benefit in limiting infringement.”
And although Article 13 includes fair-use rights to take copyrighted content for “quotation, criticism, review” and “caricature, parody or pastiche,” automated filters can’t sense the intent behind those legitimate reuses.
Weston Dombroski, with the legal team at the creator-crowdfunding site Patreon, echoed Engstrom’s filtering critique—“there is not a company in sight that has the filtering technology ready and available”—and noted the extra risk Article 13 poses to mash-up and remixed art.
“We have so many remixers,” he said of Patreon’s creators. “We have all sorts of people who would get caught up in a filtering system.”
How short is “very short”?
The economic and technological illiteracy continues through the directive’s Article 11, “Protection of press publications concerning online uses.” This would let news publications—but not individual bloggers—claim copyright in the “uses of individual words or very short extracts of a press publication” by a search engine.
In other words, newspapers could charge Google News for providing short, automated previews of their stories that help send readers to their own sites.
EU member states have tried this already and seen it fail. In 2014, for example, Spain mandated such a “snippet tax,” which led Google to close Google News in that country; a study by Spanish publishers then found a 13% drop in their traffic there.
Note that Google News now only shows headlines, while its much smaller rival DuckDuckGo still provides story excerpts. Article 11 would stop it from providing more information than Google. (DuckDuckGo CEO Gabriel Weinberg declined comment.)
Article 11’s foremost advocate, EU parliamentarian Axel Voss of Germany’s Christian Democratic Union, did not respond to an emailed query sent Tuesday.
More from Rob:
Email Rob at email@example.com; follow him on Twitter at @robpegoraro.