Urban, Karaganis and Schofield (2016)
Contents
Source Details
Urban, Karaganis and Schofield (2016) | |
Title: | Notice and Takedown in Everyday Practice |
Author(s): | Urban, J. M., Karaganis, J., Schofield, B. L. |
Year: | 2016 |
Citation: | Urban, J, Karaganis, J. and Schofield, B. (2016) Notice and Takedown in Everyday Practice (March 29, 2016). UC Berkeley Public Law Research Paper No. 2755628. |
Link(s): | Open Access |
Key Related Studies: | |
Discipline: | |
Linked by: | Bar-Ziv and Elkin-Koren (2018), Brøvig-Hanssen and Jones (2021), Cotropia and Gibson (2016), Elkin-Koren, Nahmias and Perel (2019), Engstrom and Feamster (2017), Erickson and Kretschmer (2018), Ibosiola et al. (2019b), Perel and Elkin-Koren (2017), Urban, Karaganis and Schofield (2017a), Urban, Schofield and Karaganis (2017b) |
About the Data | |
Data Description: | The report comprises 3 studies:
|
Data Type: | Primary and Secondary data |
Secondary Data Sources: | |
Data Collection Methods: | |
Data Analysis Methods: | |
Industry(ies): | |
Country(ies): | |
Cross Country Study?: | No |
Comparative Study?: | No |
Literature review?: | No |
Government or policy study?: | No |
Time Period(s) of Collection: |
|
Funder(s): |
|
Abstract
It has been nearly twenty years since section 512 of the Digital Millennium Copyright Act established the so-called notice and takedown process. Despite its importance to copyright holders, online service providers, and Internet speakers, very little empirical research has been done on how effective section 512 is for addressing copyright infringement, spurring online service provider development, or providing due process for notice targets.
This report includes three studies that draw back the curtain on notice and takedown:
1. using detailed surveys and interviews with more than three dozen respondents, the first study gathers information on how online service providers and rightsholders experience and practice notice and takedown on a day-to-day basis;
2. the second study examines a random sample from over 100 million notices generated during a six-month period to see who is sending notices, why, and whether they are valid takedown requests; and
3. the third study looks specifically at a subset of those notices that were sent to Google Image Search.
The findings suggest that whether notice and takedown “works” is highly dependent on who is using it and how it is practiced, though all respondents agreed that the Section 512 safe harbors remain fundamental to the online ecosystem. Perhaps surprisingly in light of large-scale online infringement, a large portion of OSPs still receive relatively few notices and process them by hand. For some major players, however, the scale of online infringement has led to automated, “bot”-based systems that leave little room for human review or discretion, and in a few cases notice and takedown has been abandoned in favor of techniques such as content filtering. The second and third studies revealed surprisingly high percentages of notices of questionable validity, with mistakes made by both “bots” and humans.
The findings strongly suggest that the notice and takedown system is important, under strain, and that there is no “one size fits all” approach to improving it. Based on the findings, we suggest a variety of reforms to law and practice.
Main Results of the Study
Study 1 reports that, notwithstanding the shifts in the nature of online copyright infringement and responses to it, the law’s notice and takedown provisions remain foundational to all the parties interviewed. OSPs consider its safe harbor provisions fundamentally important to their freedom to operate. Rightsholders considered notice and takedown central to their enforcement efforts, though they also expressed frustration with its capacity for dealing with large-scale infringement. Study 2’s quantitative analysis revealed deficiencies in notice and takedown procedures, especially automated requests, as all takedown requests in the sample appeared to be automated. Nearly 30% of takedown requests were of questionable validity. In one in twenty- five cases, targeted content did not match the identified infringed work, suggesting that 4.5 million requests in the entire six-month data set were fundamentally flawed. Another 15% of the requests raised questions about whether they had sufficiently identified the allegedly infringed work of the allegedly infringing material. Study 3’s quantitative analysis of the subset of requests sent in relation to Google Image Search revealed different characteristic issues than the Study 2 notices, which were sent in relation to Google Web Search. Study 3 requests were more likely to be sent by less professionalized claimants, including 53% by one individual, and not by automated detection systems. Seventy percent of the requests raised serious questions about their validity, including a significant number related to “improper” subject matter, fair use concerns, copyright ownership issues, and potentially inaccurate identification of the allegedly infringing material.
Policy Implications as Stated By Author
The report makes the following policy recommendations:"Specifically, policymakers should avoid requirements or effects that would have a disproportionate detrimental impact on OSPs outside the zones of heightened copyright conflict, such as the creation of rules or norms that create barriers to market entry or reduce competition in the OSP sector. Policymakers should also carefully consider the varying needs of different rightsholders, including those with fewer resources or less-sophisticated copyright knowledge. Last but not least, targets’ interests should be carefully taken into account, especially as they are unlikely to be present in policy discussions.Our recommendations for statutory reform focus on making it more difficult for senders to issue questionable notices without risk, and strengthening the ability of targets to respond. They require senders to declare under penalty of perjury that their substantive claims in a takedown notice are accurate, remove the mandatory ten-day waiting period before material goes back up, lower the standard for targets to recover damages from senders who make bogus claims, and raise the penalty for doing so. The recommendations also support reforming the current statutory damages regime to reduce OSPs’ fears of outsized liability and current bias toward takedown. The recommendations further suggest requiring notice and counter notice senders to submit notices to a centralized repository, where they can be searched and analysed.We strongly recommend avoiding statutory changes that would expand automated practices without much better control against mistake and abuse, or raise the cost of compliance for the vast majority of DMCA Classic OSPs. DMCA Plus measures should remain entirely voluntary.Beyond statutory reform, knowledge-sharing and best practices can fill gaps and improve operation. Our qualitative interviews identified a number of such practices. Accordingly, our recommendations for rightsholders emphasize both human and machine methods to help limit mistakes and misuse. We encourage rightsholders to work with OSPs to streamline processing and limit overbroad removal.Similarly, OSPs that use automated notice-processing systems should develop mechanisms to flag questionable notices for human review and reduce overbroad takedowns. Good practices include developing better filters that identify both flawed notices and questionable senders, and routine spot checks. OSPs should provide senders with educational materials and guidance about appropriate takedown requests, and provide targets with educational materials and an easy-to-use counter notice function.Stakeholders and government agencies should also develop informational resources and guidelines for senders and targets on copyright law, the scope and requirements of the notice and takedown regime, and how to send notices and counter notices. The materials could be hosted by a neutral government entity and accessible to all notice and counter notice senders."
Coverage of Study
Datasets
Sample size: | 35 |
Level of aggregation: | Individual |
Period of material under study: | 2016 |
Sample size: | 108,331,663 |
Level of aggregation: | Takedown Requests |
Period of material under study: | 2013 |
Sample size: | 1732 |
Level of aggregation: | Takedown Requests |
Period of material under study: | 2013 |
Sample size: | 1826 |
Level of aggregation: | Takedown Requests |
Period of material under study: | 2013 |
Sample size: | 100 |
Level of aggregation: | Company |
Period of material under study: | 2016 |