by Victoria Elman and Cindy Abramson*
A pdf version of this article may be downloaded here.
American author Elaine Scott has recently filed suit against Scribd, alleging that the social publishing website “shamelessly profits” by encouraging Internet users to illegally share copyrighted books online. [FN1] Scribd enables users to upload a variety of written works much like YouTube enables the uploading of video content. [FN2] The content uploaded to the Scribd website is then accessible and downloadable by anyone who becomes a free member of the site. Scribd claims that its Copyright Management System (“CMS”) goes “beyond the requirements set forth by U.S., UK, and EU law.”[FN3] While the implementation of a copyright filter is not mandated by the Digital Millennium Copyright Act (“DMCA”), [FN4] it has been encouraged by courts in recent litigation determining Internet Service Provider (“ISP”) liability.[FN5] Despite the fact that courts encourage the use of filters, Scott’s second claim alleges that Scribd’s use of the copyrighted work to filter infringing copies itself constitutes “ongoing and permanent” infringement without permission or compensation to the author. [FN6] On one hand, the suit attacks Scribd for allowing the uploading of infringing works; on the other, it claims that Scribd’s attempts to filter the website of these infringing works without explicit permission from the copyright holders is itself infringing.
As many commentators have pointed out, the fair use doctrine is likely to hold up in a court of law with respect to this second counterintuitive claim that filtering itself is a copyright violation. [FN7] A key factor in determining fair use is the purpose and character of the alleged infringer’s use. The more transformative, or how different the purpose or function is from the original work, the more likely it is to constitute fair use. Two recent cases suggest that Scribd’s use of copyrighted works to filter is highly transformative. In Perfect 10 v. Amazon, the Ninth Circuit held that the use of thumbnail images for operating a comprehensive search engine was transformative because while the image may have been created to serve an entertaining or aesthetic function, the search engine “transforms the image into a pointer directing a user to a source of information.” [FN8] In A.V. v. iParadigms, the Fourth Circuit held that the use of archiving an essay to check whether other essays are plagiarized was transformative despite the argument that this new use did not add anything to the original work. [FN9] Because Scribd has a strong argument for transformativeness and there is currently no effect on the market for the copyrighted work—in that copyright owners do not sell or license their works for profit in exchange for inclusion in filtering databases—Scribd is likely to win on the grounds of fair use. However, in looking at the purpose of the DMCA and the courts’ encouragement of filtration to remain in compliance, it seems that a court could find instead that the use of a copyrighted work in a filtering database is really no infringement at all.
Filtration Systems: The Status Quo
Due to the concern in protecting themselves from third party liability under the DMCA, many ISPs have begun to either develop their own filters or to hire third parties to filter their sites for infringing works. Scribd’s CMS is managed in house. [FN10] This system is populated either when a copyright owner submits her work to be included in the database or when Scribd receives a DMCA takedown notice from the copyright holder. [FN11] Once the takedown notice is received by Scribd the company not only removes the infringing work but also creates a reference of that work that is included in the filtering database—the act that Scott alleges is a copyright violation. [FN12] This is a counterintuitive claim in a world where some copyright owners will actually pay for their works to be included in third-party filtering services which then charges additional fees for ISPs to use their service. [FN13]
Why Use of a Copyrighted Work in a Filtering Database is Not Infringement
In order for the fair use argument to arise, there must be an infringement in the first place. However, it seems that a court could avoid the fair use avenue of analysis altogether in tackling Scott’s counterintuitive claim. One purpose of the DMCA is to provide protection to copyright owners whose works are constantly being infringed on the ever-expanding social networks and sharing sites. Moreover, courts have time and time again looked to filtering systems as a way in which ISPs can avoid third-party liability. If filtering is the best means by which an ISP can protect itself then it should not be prevented from doing so. While not mandated, filtering is seen as a shield against liability. [FN14]Lastly, copyright owners are themselves sending takedown notices urging ISPs to prevent the infringing activity on their websites. One common sense solution to this issue is to view a copyright owner’s takedown notice as constructive authorization for an ISP to proactively prevent the infringing work from being re-posted by using the work (or a reference of the work) in a filtering system. This would not be the first time that a court had to step in to avoid a “confusing, self-contradictory catch-22 situation . . . particularly when there is a much simpler explanation.” [FN15]
The Future of Building a Filtering Database
There is a seemingly infinite number of copyrighted works that desire protection against infringement. Gaining access to (and thus the ability to filter for) 100% of these works is a challenge impossible to demand of any ISP. Currently, filtering databases are populated by free or paid submissions by copyright holders either directly to an ISP or to a third party filtering operator. In an attempt to expand its database to include works that it knew wanted protection, Scribd chose to expand its database by including in it works for which it had received DMCA takedown notices. As the need and desire for filtering increases so must the size of the database. To limit one’s ability to expand that database would to be to further complicate the growing demands that ISPs stop infringement before it starts. There should be as few hurdles as possible to an ISP’s access to these works, so long as the sole purpose is to filter out infringing content.
*Victoria Elman received a B.A. from Columbia University and a J.D. from Benjamin N. Cardozo School of Law in 2009 where she was a Notes Editor of the Cardozo Law Review. She will be joining Goodwin Procter this January as an associate in the New York office.
Cindy Abramson is an associate in the Litigation Department in the New York office of Morrison & Foerster LLP. She recieved her J.D. from the Benjamin N. Cardozo School of Law in 2009 where she completed concentrations in Intellectual Property and Litigation and was Senior Notes Editor of the Cardozo Arts & Entertainment Law Journal.
The views expressed in this article are those of the authors and do not reflect the official policy or position of Morrison & Foerster, LLP.
[FN1] Complaint at 4, Scott v. Scribd, No. 4:09-cv-03039 (S.D. Tex. filed Sept. 18, 2009).
[FN2] YouTube has also been embroiled in copyright infringement litigation. In 2007, Viacom Inc. sued YouTube and its corporate parent Google Inc. for alleged copyright infringement and sought over $1 billion in damages. SeeFrank Ahrens, Viacom Sues YouTube Over Copyright, Wash. Post, Mar. 14, 2007, at D2, available at http://www.washingtonpost.com/wp-dyn/content/article/2007/03/13/AR2007031300595.html.
[FN4] The DMCA currently provides a safe harbor to ISPs who lack knowledge or apparent knowledge of infringing works and who expeditiously remove an infringing work once knowledge is obtained. 17 U.S.C. § 512(c) (2008).
[FN5] See MGM Studios, Inc. v. Grokster, Ltd., 545 U.S. 913, 939 (2005) (“This evidence of unlawful objective is given added significance by MGM’s showing that neither company attempted to develop filtering tools or other mechanisms to diminish the infringing activity using their software. While the Ninth Circuit treated the defendants’ failure to develop such tools as irrelevant because they lacked an independent duty to monitor their users’ activity, we think this evidence underscores Grokster’s and StreamCast’s intentional facilitation of their users’ infringement.”).
[FN6] Complaint at 9, Scott v. Scribd, No. 4:09-cv-03039 (S.D. Tex. filed Sept. 18, 2009). “Without permission of the authors, Scribd maintains copies of authors’ works for use in a copyright protection system . . . [O]nce a copyrighted work is uploaded to Scribd without . . . permission, the infringement is ongoing and permanent. Even if the work becomes unavailable for download by users, Scribd illegally copies the work into its copyright protection system, without permission or compensation to the author.”
[FN7] See Ashby Jones, A Copyright Head-Scratcher, Courtesy of Kiwi Camara, Wall St. J. Law Blog, Oct. 1, 2009, at http://blogs.wsj.com/law/2009/10/01/the-latest-copyright-head-scratcher-courtesy-of-kiwi-camara (last visited Oct. 25, 2009); David Kravets, Lawsuit: Copyright Filtering Technology Infringes, Wired, Sept. 21, 2009, http://www.wired.com/threatlevel/2009/09/infringingfiltering (last visited Oct. 25, 2009).
[FN8] Perfect 10, Inc. v. Amazon.com, Inc., 487 F.3d 701, 721 (9th Cir. 2007).
[FN9] A.V. v. iParadigms, 562 F.3d 630 (4th Cir. 2009).
[FN13] Audible Magic is one such service that filters copyrighted audio and video works. Audible Magic Protecting Creative Works, http://www.audiblemagic.com/products-services/registration.
[FN14] See Grokster, 545 U.S. at 939.
[FN15] UMG Recordings, Inc. v. Veoh Networks, Inc., No. CV 07-5744, 2009 U.S. Dist. LEXIS 86932, at *42-43 (C.D. Cal. Sept. 11, 2009) (citing Ellison v. Robertson, 189 F. Supp. 2d 1051, 1061 (C.D. Cal. 2001); Perfect 10, Inc. v. CCBill LLC, 488 F.3d at 1114).