Bar-Ziv and Elkin-Koren (2018)
|Bar-Ziv and Elkin-Koren (2018)|
|Title:||Behind the Scenes of Online Copyright Enforcement: Empirical Evidence on Notice & Takedown|
|Author(s):||Sharon Bar-Ziv, Niva Elkin-Koren|
|Citation:||Bar-Ziv, S. and Elkin-Koren, N. (2018) Behind the Scenes of Online Copyright Enforcement: Empirical Evidence on Notice & Takedown. Connecticut Law Review, Vol. 50(2).|
|Key Related Studies:|
|About the Data|
|Data Description:||The authors collected and analysed a random sample of 9,890 takedown requests received by Google Search, pertaining to content hosted on Israeli domains (.il country code). The data were obtained from the Lumen database.|
|Data Type:||Primary and Secondary data|
|Secondary Data Sources:|
|Data Collection Methods:|
|Data Analysis Methods:|
|Cross Country Study?:||No|
|Government or policy study?:||No|
|Time Period(s) of Collection:||
"Copyright enforcement was one of the early challenges to the rule of law on the internet and has shaped its development since the early 1990s. The Notice and Takedown (N&TD) regime, enacted in the Digital Millennium Copyright Act, offered online intermediaries immunity from liability in exchange for removing allegedly infringing materials upon receiving notice from rights holders. The unequivocal power of rights holders to request removal and the strong incentives for online intermediaries to remove content upon receiving a removal request have turned the N&TD regime into a robust clean-up mechanism for removing any unwarranted content. The N&TD procedure applies to private facilities, makes use of proprietary software, and is administered by private companies. This enforcement procedure is nontransparent and lacks sufficient legal or public oversight. Unlike copyright enforcement in court, where decisions are made public, we know very little about the actual implementation of the N&TD regime: Which players make use of the system? Who is targeted? What materials get removed and why? How effective is the removal of infringing materials, and does it comply with copyright law? This Article offers empirical evidence on the implementation of the N&TD regime based on the systematic coding and analysis of a large-scale data-set of removal requests sent to Google Search. The findings shed light on the major changes that have taken place in copyright enforcement following the transition to the online arena over the past decade. Analysis of the data reveals that the N&TD procedure has been extensively used to remove noninfringing materials, and most removal requests pertained to allegedly inaccurate, defamatory, or misleading content. These findings raise serious concerns that the N&TD procedure is becoming fertile ground for misuse. Moreover, online enforcement is dominated by multinational companies, which prefer to target global intermediaries rather than attempt to remove materials hosted by local platforms. This may lead to underenforcement of copyright online, as the exclusive focus on removal of links to allegedly infringing materials may limit access to these materials, yet fail to actually remove these same materials. The local hosting platforms which facilitates access to repeat infringments, are widely known within the relevant community of users. This calls into question the effectiveness of this enforcement strategy. At the same time, however, the data demonstrates instances of overenforcement, where some materials have been removed on questionable grounds. Thus, the findings raise concerns over the implications of the N&TD regime for access to knowledge and freedom of speech. Overall, the study shows that in the absence of sufficient oversight, the N&TD regime is vulnerable to misuse, carrying consequences to copyright goals, access to justice, and due process. By uncovering the invisible dynamics at work in online copyright enforcement, this Article may contribute to identifying the challenges facing policymakers in shaping online enforcement procedures and developing the appropriate measures to address them."
Main Results of the Study
The researchers found that significant proportion (66%) of the DMCA requests received during the study period did not actually pertain to copyright infringement. Many of the requests, filed using Google’s DMCA notification system, resembled ‘right to be forgotten’ requests. This may be due to the fact that Irsael is outside of the jurisdiction of the General Data Protection Regulation, and Google maintains a separate notification system for such requests from European Member States. Their overall results highlight the potential for misidentification stemming from lack of awareness or purposeful deception on the part of claimants.
Their study confirms the existence and illuminates the role of third-party enforcement services. These are companies that act on behalf of clients to monitor and remove online infringement of material from the clients’ catalogue. They are ‘repeat players’ at the takedown game, and they typically send long, automatically generated requests containing many URLs of content for removal. The authors suggest they both (1) fail to successfully identify local piracy ‘hub’ sites, because of their global range and (2) may be hazardously incentivised to engage in cat-and-mouse with pirates to generate further work for themselves.
Further results include:
• 87% of the copyright takedown requests were targeted at infringing software, with Microsoft being the most active issuer of requests during the specific study period.
• Whilst links to infringing content were removed by Google Search, in 85% of cases the original infringing content could still be found on the target page. This suggests the primacy of the Google removal tool for enforcement, with rightsholders less willing to go after specific infringements on smaller websites where those contents are available.
• Approximately 10% of the removals were questionable, with some 5% of the total sample likely non-infringing (false positive removals).
• No single counter-notice was filed during the study period.
Policy Implications as Stated By Author
The authors caution that the current system for notice-and-takedown jeopardises freedom of expression and access to knowledge by encouraging platforms to indiscriminately remove non-infringing materials without judgement. Furthermore, this may imperil the goals of copyright, by assigning the task of balancing interests between the author and user to a platform (effectively privatising copyright enforcement). Lastly, due to the lack of any substantive appeals process and presence of powerful multinational players, the authors caution that the current notice-and-takedown system may have serious implications for access to justice and due process.
Overall, the study recommends the introduction of an "oversight" mechanism or measure, as well as imposing transparency requirements on the decision-making processes of online platforms.