*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; and3. 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.Note: This is an updated version of the original paper. It includes two new appendices and some minor updates and corrections.Also available at SSRN: http://ssrn.com/abstract=2755628

Section 512 of the Digital Millennium Copyright Act established both "safe harbors" from liability for online service providers and the well-known "notice and takedown" process for removing online infringements of copyrighted material. In the ensuing two decades, the notice and takedown process has become a primary tool for raising and resolving copyright disputes in the United States.But despite its influence, there is little empirical research describing § 512's operation or its effectiveness. This article digests findings from a qualitative study, reported fully in the three-study report, Notice and Takedown in Everyday Practice (available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2755628) and peer-reviewed here, that helps fill this gap. Through detailed surveys and interviews with nearly three dozen respondents, we provide the first detailed account of how § 512 is implemented and experienced by online service providers and large rightsholders. All respondents agreed that the § 512 safe harbors and the ability to take down infringing material remain fundamental. But the online copyright enforcement ecosystem is also highly diverse, and all participants face challenges. The findings 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 best practices and limited legal reforms.

ACKNOWLEDGMENTS AND DISCLOSURESWe are deeply grateful to our survey respondents and interviewees for generously contributing their time and expertise, and for candidly sharing their "on the ground" experiences with notice and takedown.This work would not have been possible without both data and funding resources for the coding effort. We thank Adam Holland and Wendy Seltzer of Lumen (formerly Chilling Effects) for facilitating access to the Lumen data, which forms the basis for our quantitative work. We are grateful for funding support from Google Inc. as a gift to The American Assembly and from the Sloan Foundation for its support through the Berkeley Law Digital Library Copyright Project. Neither funder directed our approach in any way, and neither funder reviewed any methods, data, results, or reporting before public release. We are also grateful for Nash Information Services' inkind donation of the OpusData database and expert database customization advice.We are indebted to the individuals who have generously lent their time, skill, and expertise to the project: Kristoff Grospe for his assistance with the development and management of the qualitative and quantitative studies; Martyn Joyce for building and maintaining our customized database, coding and querying interfaces, and randomization and search algorithms; Bruce Nash of Nash Information Services for help with database and algorithm design; and Nora Broege of UC Berkeley's D-Lab for data preparation and statistical analysis. We are very grateful for the detailed and helpful comments on study design, findings, and drafts offered

/ OSI2017 Workgroup QuestionWhat standards, norms, best practices, exit strategies, and incentive systems does the world of scholarly communications need? What is the future ideal? What will it take (including studies or pilots) to develop a better understanding of how the scholarly communication system works now? This workgroup will also necessarily touch on norms and definitions, so will include discussions as warranted about open and impact spectrums as covered in OSI2016.

No abstract